cirugía refractiva láser corneal smile. resultados visuales y … · 2018. 11. 15. · cirugía...

$: Cirugía refractiva láser corneal SMILE. Resultados visuales y … · 2018. 11. 15. · Cirugía refractiva láser corneal SMILE. Resultados visuales y biomecánica corneal en miopías$
Cirugía refractiva láser corneal SMILE. Resultados visuales y biomecánica corneal en miopías bajas, medias y altas

Joaquín Fernández Pérez

www.ua.es

www.eltallerdigital.com

1

Instituto Universitario de Física Aplicada a las Ciencias y las Tecnologías

Departamento de Óptica, Farmacología y Anatomía

Facultad de Ciencias de la Salud


Joaquín Fernández Pérez

Tesis presentada para aspirar al grado de

DOCTOR POR LA UNIVERSIDAD DE ALICANTE

DOCTORADO EN FÍSICA APLICADA A LAS CIENCIAS Y LAS TECNOLOGÍAS

Dirigida por:

Dr. David P Piñero Llorens

Alicante, abril 2017

______________________________________________________________________________________________

2


3

D. DAVID PABLO PIÑERO LLORENS, Doctor por la Universidad de Alicante, Investigador

Distinguido (Acreditado para titular en el área de Óptica) del Departamento de Óptica, Farmacología y

Anatomía de la Facultad de Ciencias de la Universidad de Alicante y miembro del Instituto Universitario

de Física Aplicada a las Ciencias y a las Tecnologías:

CERTIFICA: Que la presente memoria titulada “Cirugía refractiva láser corneal SMILE.

Resultados visuales y biomecánica corneal en miopías bajas, medias y altas” ha sido realizada bajo su

dirección por Don JOAQUÍN FERNÁNDEZ PÉREZ en el Instituto Universitario de Física Aplicada a las

Ciencias y a las Tecnologías de la Universidad de Alicante y constituye su Tesis Doctoral para optar al

Grado de Doctor.

Y para que conste, y en cumplimiento de la legislación vigente, firman el presente certificado en

Alicante a de junio de dos mil diecisiete.

Fdo. David P Piñero Llorens

______________________________________________________________________________________________

4


5

Agradecimientos

A David Piñero, por el que siento una profunda admiración por su trayectoria profesional,

sincero agradecimiento por su apoyo y con el que durante todo este tiempo he forjado una

entrañable amistad basada en el afecto y respeto mutuo.

Al Departamento de I+D+i de Qvision y en especial a su Director, Manuel Rodríguez, del que

tengo el constante privilegio de enriquecerme con su búsqueda del extremo rigor metodológico

en investigación, que ha sido un pilar básico en la elaboración de esta tesis y por el que profeso

una profunda admiración por su talento.

A todo el equipo asistencial de Qvision, sin los que con su excelencia en el trabajo diario, esta

tesis hubiera sido imposible realizar.

A Virginia y Carmen por su amor y por hacer de lo pequeño y cotidiano, lo imprescindible.

______________________________________________________________________________________________

6


7

Índice

ÍNDICE DE ABREVIATURAS .............................................................................................. 9

RESUMEN ............................................................................................................................. 13

SUMMARY ........................................................................................................................... 15

CAPÍTULO 1 ....................................................................................................................................... 17

1.1. Introducción la cirugía refractiva corneal. .................................................................. 19

1.2. Cambio de paradigma en la cirugía refractiva láser. .................................................. 19

1.2.1. El láser de femtosegundo en cirugía refractiva. 19

1.2.2. La creación de un lentículo intraestromal. 20

1.2.3. Técnica de extracción del lentículo en humanos (ReLEx). 21

1.2.4. Extracción del lentículo a través de una microincisión (SMILE). 21

1.3. Controversias en torno a SMILE frente a LASIK. ..................................................... 23

1.3.1. Sensibilidad corneal y sequedad ocular. 23

1.3.2. Tiempo de recuperación visual. 24

1.3.3. Resultados en función de las características del paciente o de la cirugía. 25

1.3.4. Calidad óptica ocular. 26

1.3.5. Biomecánica corneal. 28

1.4. Evaluación clínica de la biomecánica corneal. ........................................................... 29

1.4.1. Instrumentos clínicos para la medida de la biomecánica corneal. .............................. 29

1.4.2. Medida clínica de biomecánica corneal tras cirugía refractiva................................... 31

1.5. Justificación y objetivos. ............................................................................................ 32

1.6. Estructura de la Tesis. ................................................................................................ 34

CAPÍTULO 2 ....................................................................................................................................... 37

2.1. Short-term outcomes of Small Incision Lenticule Extraction (SMILE) for low,

medium and high myopia. ...................................................................................................... 41

2.1.1. Abstract ...................................................................................................................... 41

2.1.2. Introduction ................................................................................................................ 42

2.1.3. Methods ...................................................................................................................... 42

2.1.4. Results ........................................................................................................................ 45

2.1.5. Discussion .................................................................................................................. 50

______________________________________________________________________________________________

8

2.1.6. Disclosures ................................................................................................................. 53

2.1.7. References .................................................................................................................. 53

2.2. Corneal thickness after SMILE affects Scheimpflug-based Dynamic Tonometry ..... 58

2.2.1. Abstract ...................................................................................................................... 58

2.2.2. Introduction ................................................................................................................ 59

2.2.3. Methods ...................................................................................................................... 60

2.2.4. Results ........................................................................................................................ 64

2.2.5. Discussion .................................................................................................................. 67

2.2.6. Disclosures ................................................................................................................. 70

2.2.7. References .................................................................................................................. 71

2.3. New parameters for evaluating corneal biomechanics and intraocular pressure

after SMILE by Scheimpflug-Based Dynamic Tonometry. ................................................... 75

2.3.1. Abstract ...................................................................................................................... 75

2.3.2. Introduction ................................................................................................................ 76

2.3.3. Methods ...................................................................................................................... 77

2.3.4. Results ........................................................................................................................ 80

2.3.5. Discussion .................................................................................................................. 83

2.3.6. Disclosures ................................................................................................................. 88

2.3.7. References .................................................................................................................. 89

CAPÍTULO 3 ....................................................................................................................................... 91

3.1 Discusión de los resultados. ....................................................................................... 93

CAPÍTULO 4 ....................................................................................................................................... 97

4.1 Cumplimiento de objetivos ........................................................................................ 99

4.2 Aportaciones realizadas y líneas futuras de investigación. ...................................... 100

BIBLIOGRAFÍA ................................................................................................................................ 103


9

ÍNDICE DE ABREVIATURAS

Abreviatura Inglés Español

ASCRS American Society of Cataract and

Refractive Surgery

Congreso de la Academia Americana de

Oftalmología

A1 First applanation Primera aplanación

A2 Second applanation Segunda aplanación

AL1 Applanation length at A1 Longitud de aplanación en A1

AL2 Applanation length at A2 Longitud de aplanación en A2

AT1 Time to arrive at A1 Tiempo en alcanzar A1

AT2 Time to arrive at A2 Tiempo en alcanzar A2

ATR Against the rule En contra de la regla

AV1 Velocity to arrive at A1 Velocidad de aplanación en A1

AV2 Velocity to arrive at A2 Velocidad de aplanación en A2

bIOP Biomechanically corrected

intraocular pressure

Presión intraocular corregida por

parámetros biomecánicos

C Pupil size Tamaño de pupila

CBI Corneal Biomechanical Index Índice de Biomecánica Corneal

CCT Central Corneal Thickness Espesor corneal central

CCT’ Central Corneal Thickness after

surgery Espesor corneal central tras la operación

CD Corneal densitometry Densitometría corneal

CDVA Corrected distance visual acuity Agudeza visual con corrección

CH Corneal Hysteresis Histéresis corneal

CoST Corvis ST Corvis ST

CRF Corneal Resistance Factor Factor de resistencia corneal

DA Deformation Amplitude Amplitud de deformación en HC

DA’ Deformation Amplitude after

surgery

Amplitud de deformación en HC tras

cirugía

DAc DA corrected Deformación en máxima corregida,

preoperatoria más espesor corneal

DAR Deflection Amplitude Ratio Cociente de amplitud de deformación

DAR1

Ratio of the central corneal

deflection and the average of two

points located at one millimeter at

both sides from the center

Cociente de la amplitud de deformación

en el centro de la córnea y el promedio de

dos puntos a 1 mm de ambos lados

respecto al centro.

DAR2 Ratio of the central corneal

deflection and the average of two

Cociente de la amplitud de deformación

en el centro de la córnea y el promedio de

______________________________________________________________________________________________

10

points located at two millimeters at

both sides from the center

dos puntos a 2 mm de ambos lados

respecto al centro.

DI Densitometry increase Incremento de la densitometría

DI-A1 Densitometry increase at A1 Incremento de la densitometría en A1

DI-A2 Densitometry increase at A2 Incremento de la densitometría en A2

DI-HC Densitometry increase at HC Incremento de la densitometría en HC

DIM Maximum Densitometry Increase Incremento de la densitometría máximo a

lo largo de la deformación

epi-LASIK Epithelial laser in-situ

keratomileusis

LASIK con separación del epitelio

mediante un epiqueratomo

EV Error Vector Vector de error

FLEx Femtosecond Lenticule Extraction Extracción del lentículo por

femtosegundo

FS-LASIK Femtosecond LASIK LASIK por femtosegundo

FS-LASIK WF Wavefront guided femtosecond

LASIK LASIK guiado por frente de onda

HC Highest Concavity Máxima concavidad

ICC Intraclass correlation index Índice de correlación intraclase

IOP Intraocular pressure Presión intraocular (PIO)

IR Integrated Inverse Concave Radius Integral del radio de concavidad

IRC Intended Refractive Correction Error refractivo a programado para

corrección

LASEK Laser-Assisted Subepitelial

Keratomileusis

Queratomileusis subepitelial asistida por

láser

LASIK Laser Assisted in-situ

Keratomileusis

Queratomileusis asistida mediante láser

Excímer

LRS Laser Refractive Surgery Cirugía refractiva láser

MTF Modulation Transfer Function Función de Transferencia de modulación

Nd-YAG Neodymium-doped yttrium

aluminium garnet -

NEV Normalized Error Vector Vector error normalizado

ORA Ocular Response Analyzer -

PA1 Air puff pressure at A1 Presión del pulso en A1

PA2 Air puff pressure at A2 Presión del pulso en A2

PD Peak distance Distancia entre picos

PERK Prospective Evaluation of Radial

Keratotomy -

PIO Intraocular pressure (IOP) Presión intraocular


11

PRK Photorefractive Keratotomy Queratectomía fotorrefractiva

r Pearson r Coeficiente de Pearson

ReLEx Refractive Lenticule Extraction Extracción refractiva de lentículo

RK Radial Keratotomy Queratotomía radial

Rx Preoperative spectacle refraction Refracción preoperatoria en gafa

SCI Science Citation Index Índice de citas científicas

SD Standard Deviation Desviación Estándar

SE Spherical Equivalent Equivalente esférico

SimK Simmulated Keratometry Queratometría Simulada

SMILE Small Incision Lenticule

Extraction

Extracción refractiva de lentículo por

pequeña incisión

SP-A1 Stiffness parameter at first

applanation

Parámetro de rigidez en la primera

aplanación

t1 and t2

Indexes obtained from the ratio

between the change in time and

change in thickness

Índices obtenidos mediante el cociente

entre el cambio en tiempo frente al

cambio en espesor

TEV Treatment Error Vector Vector error de tratamiento

UDVA Uncorrected distance visual acuity Agudeza visual sin corrección

VA Visual acuity Agudeza visual

WTR With the Rule A favor de la regla

ZO Optic Zone Zona óptica

α Error probability Probabilidad de error

Δ Difference preoperative value –

postoperative value

Diferencias entre valores preoperatorios

y posoperatorios

δ Means Medias

ρ Spearman rho Rho de Spearman

σ Standard deviation Desviación estándar

______________________________________________________________________________________________

12


13

RESUMEN

La técnica de cirugía refractiva láser a través de la extracción de un lentículo por una

microincisión SMILE (Small Incision Lenticule Extraction), representa una técnica novedosa

para la corrección de la miopía y un cambio de paradigma en torno a los procedimientos

anteriores de cirugía refractiva láser. Como cualquier nueva técnica que irrumpe en una

especialidad de la medicina, se ve sometida al juicio de la evidencia científica con el fin de

demostrar las hipótesis planteadas desde el marco teórico. Este hecho es si cabe más destacable

en procedimientos que podrían suponer el reemplazo de técnicas llevadas a cabo con

anterioridad, como es el caso de SMILE frente a LASIK (Laser Assisted in-situ Keratomileusis)

y PRK (Photorefractive Keratotomy).

En esta Tesis doctoral por compendio de publicaciones se ponen de manifiesto las actuales

controversias en torno a las ventajas de SMILE frente a técnicas previas. Los resultados de

eficacia, seguridad y predictibilidad para las primeras intervenciones de un cirujano

experimentado en cirugía refractiva sin experiencia previa en SMILE son evaluadas en función

del nivel del error refractivo. Estos resultados clínicos podrían estar condicionados por las

características biomecánicas de la córnea intervenida. La influencia de la biomecánica corneal

en los resultados refractivos, junto con la detección de alteraciones corneales que pudiesen

derivar en el desarrollo de una ectasia corneal, representan dos de los motivos por los cuales

medir la biomecánica de manera fiable sería de gran importancia en la toma de decisiones del

cirujano refractivo.

Además, una de las controversias más polémicas debido a su falta de evidencia en investigación

clínica es la mayor preservación de la biomecánica corneal en SMILE frente a otras técnicas. En

esta Tesis doctoral se estudian las variables de confusión en pacientes operados de SMILE y se

proponen soluciones para el diseño de estudios con el fin de minimizar el efecto de éstas y así

poder detectar cuando los cambios en los parámetros de biomecánica se deben a una

modificación de la rigidez corneal y no a un cambio en las variables de confusión. Finalmente,

se analizan los parámetros más actuales de la tonometría dinámica de Scheimpflug con el fin de

determinar cuáles deberían ser o no corregidos ante las variables de confusión y se propone una

nueva variable conocida como densitometría dinámica por su posible aportación de información

sobre la biomecánica e hidratación corneal.

______________________________________________________________________________________________

14


15

SUMMARY

The technique of laser refractive surgery consisting of the removal of a lenticule by a micro

incision SMILE (Small Incision Lenticule Extraction) represents a novel technique for the

correction of myopia and a paradigm shift around previous procedures of laser refractive

surgery. Like any new technique that breaks into a specialty of medicine, it is submitted to the

judgment of the scientific evidence in order to demonstrate the hypotheses raised from the

theoretical framework. This fact is perhaps more remarkable in procedures that could involve

the replacement of techniques carried out previously, as in the case of SMILE against LASIK

(Laser Assisted in-situ Keratomileusis) and PRK (Photorefractive Keratotomy).

In this doctoral Thesis by compendium of publications, the current controversies regarding the

advantages of SMILE in relation to previous techniques are revealed. The results of efficacy,

safety and predictability for the first interventions of a surgeon experienced in refractive surgery

without previous experience in SMILE are evaluated according to the level of refractive error.

These clinical results could be conditioned by the biomechanical characteristics of the operated

cornea. The influence of corneal biomechanics on refractive outcomes, along with the detection

of corneal alterations that could lead to the development of corneal ectasia, represent two of the

reasons why reliable biomechanical measurement would be of great importance in the decision

making process of the refractive surgeon.

Furthermore, one of the most polemical controversies due to its lack of evidence in clinical

research is the greater preservation of corneal biomechanics in SMILE compared to other

techniques. In this doctoral Thesis we study the confounding variables in patients operated on

SMILE and propose solutions for the design of studies in order to minimize the effect of these,

and in addition to be able to detect when the changes in biomechanical parameters are due to a

modification of the corneal stiffness and not to a change in the confounding variables. Finally,

the current parameters of the Scheimpflug dynamic tonometry are analyzed in order to

determine which of them should be corrected or not to the confounding variables and a new

variable known as dynamic densitometry is proposed for its possible contribution of information

on biomechanics and corneal hydration.

______________________________________________________________________________________________

16


17

CAPÍTULO 1 INTRODUCCIÓN GENERAL

______________________________________________________________________________________________

18


19

1.1. Introducción la cirugía refractiva corneal.

La mayor parte de los procedimientos de cirugía refractiva llevados a cabo durante la

década de los 80 y 90 se centraban en tres técnicas principales: la queratotomía radial (RK,

Radial Keratotomy), la queratomileusis asistida mediante láser excimer (LASIK, Laser Assisted

in-situ Keratomileusis) y la queratectomía fotorrefractiva (PRK, Photorefractive Keratotomy)

(Corcoran 2015). La primera de ellas consistía en la ejecución de una serie de 4, 6, 8 o 16

incisiones radiales distribuidas de manera simétrica alrededor de la córnea excluyendo la región

central (zona óptica) de tal manera que a menor tamaño de zona óptica y mayor número de

incisiones, mayor era el efecto alcanzado por la cirugía (Holladay 2003). Esta técnica demostró

ser efectiva a largo plazo (10 años) en el 70% de los pacientes intervenidos con un nivel de

seguridad razonable y con una regresión hacia la hipermetropía que continuaba tras la década

del procedimiento (Waring et al. 1994). La RK pasó a un segundo plano a mediados de la

década de los 90 tras los resultados obtenidos por el estudio PERK (Prospective Evaluation of

Radial Keratotomy) en los que se hacía referencia a variaciones diurnas de la visión y presencia

de halos (Kemp et al. 1999), creciendo en popularidad al mismo tiempo las técnicas PRK y

LASIK con la introducción del láser excimer de 193 nm a principios de los 90 (Sher et al. 1991;

Carr et al. 2001).

La técnica PRK consiste en utilizar luz ultravioleta de 193 nm, a través de un láser

excimer, con el fin de fotoablacionar tejido estromal sin afectar en gran medida a las regiones

circundantes a la zona de aplicación. Para ello, se elimina el epitelio completamente en la zona

óptica antes de la aplicación del láser o se retira el epitelio para ser posteriormente

reposicionado, esta variante de la PRK se conoce como queratomileusis subepitelial asistida por

láser (LASEK, Laser-Assisted Subepitelial Keratomileusis)(Ghadhfan et al. 2007). A diferencia

de la PRK y pese a utilizar el mismo láser para realizar la ablación corneal, la técnica LASIK

conlleva la creación de un flap corneal que se mantiene ligado a la córnea en forma de bisagra y

que se reposiciona tras la aplicación del láser. El LASIK posee importantes ventajas frente a la

PRK (disminución de la prevalencia de opacidades corneales (haze), eliminación del dolor

postoperatorio, etc.)(Holladay 2003) que han supuesto que esta técnica se convirtiese en la

preferida por gran parte de los cirujanos refractivos (Corcoran 2015).

1.2. Cambio de paradigma en la cirugía refractiva láser.

1.2.1. El láser de femtosegundo en cirugía refractiva.

En el apartado anterior realizamos una introducción a las técnicas principales de cirugía

______________________________________________________________________________________________

20

refractiva láser. Pese a que podemos encontrar variantes en la literatura científica sobre éstas

técnicas, como el uso de mitomicina C en la PRK para prevenir el haze en altas miopías

(Gambato et al. 2005), la separación del epitelio mediante un epiqueratomo (epi-

LASIK)(Pallikaris et al. 2005), LASIK o LASEK guiado por frente de onda (Schwiegerling

2004; Jung et al. 2015), etc., la realidad es que estas variantes continúan siendo llevadas a cabo

con láser excimer. No fue hasta principios de los 90 cuando otros láseres aparecieron con

frecuencias de pulso del nanosegundo (ns, 10-9s), picosegundo (ps, 10-12s) y femtosegundo (fs,

10-15s) para longitudes de onda del visible y del infrarrojo corto como posible alternativa al láser

excimer (Stern et al. 1989). La principal ventaja de incrementar la frecuencia del pulso desde el

ns al fs reside en una disminución del daño producido en los tejidos circundantes (Soong &

Malta 2009). De esta forma conseguimos que, a diferencia del láser Nd:YAG que utiliza una

longitud de onda en el infrarrojo corto (1064 nm) y una frecuencia de ns produciendo un daño

colateral de hasta 100 µm, empleando una longitud de onda similar (1053 nm) en fs podamos

trabajar sobre la córnea ya que el volumen colateral de tejido dañado es 103 veces menor en fs

que ps (frecuencia mayor al ns del Nd:YAG)(Stern et al. 1989).

1.2.2. La creación de un lentículo intraestromal.

Los nuevos láseres de fs permitieron actuar de distinta forma sobre el tejido corneal. En

la Tabla 1 se muestra un resumen de los principales láseres utilizados en oftalmología, su

longitud de onda y el efecto que producen sobre el tejido en el que actúan. El láser excimer

actúa de manera no localizada fotoablacionando el tejido superficial sobre el que incide, bien

sea tras retirar el epitelio (PRK) o tras realizar un flap accediendo a una capa más profunda del

estroma (LASIK). No obstante, el modo de actuación del láser de fs sobre la córnea es

totalmente diferente. Permite trabajar de manera localizada, dentro o incluso detrás de la córnea,

de tal forma que se generan pequeñas burbujas separadas un pequeño espacio (fotodisrupción)

creando un corte que alcanza alrededor de 1 µm en precisión (Soong & Malta 2009).

Tabla 1. Listado de láseres empleados en oftalmología (Soong & Malta 2009).

Láser Longitud de onda (nm) Efecto en el tejido

Dióxido de carbono

Nd:YAG

Femtosegundo

Kriptón

Argón

Excimer

10600 (infrarrojo largo)

1064 (infrarrojo corto)

1053 (infrarrojo corto)

647-531 (visible)

514-488 (visible)

193 (ultravioleta largo)

Fototermal

Fotodisrupción

Fotodisrupción

Fotoquímico (coagulación)

Fotoquímico (coagulación)

Fotoablación

Láser Nd-YAG (acrónimo del inglés neodymium-doped yttrium aluminium garnet)


21

La hipótesis de la posible cirugía refractiva intraestromal planteada a principios de los

90 por varios autores con la esperanza de mantener intactos el epitelio, la membrana de

Bowman y el endotelio era cada vez más real con la llegada del láser de fs (Kurtz et al. 1998).

Con anterioridad, varios intentos habían sido llevados a cabo con láseres de ns y ps (Vogel et al.

1994; Krueger et al. 1996). No obstante, el daño colateral producido limitaba la capacidad de

obtener una ablación eficiente y predecible incluso en el caso del ps en el que se producía un

menor daño colateral, pero se requería una disección manual que terminaba originando una

superficie irregular (Ito et al. 1996).

1.2.3. Técnica de extracción del lentículo en humanos (ReLEx).

En el año 2003 se realizó el primer estudio en humanos que implicaba el tallado de un

lentículo intra-estromal, resultando en córneas transparentes tras las primeras dos horas y

resultados refractivos estables (Ratkay-Traub et al. 2003). No obstante, no fue hasta el año 2007

cuando se introdujo el primer laser comercial (VisuMax, Carl Zeiss Meditec, AG, Jena,

Germany) para la corrección de la miopía a través de la generación de un lentículo intraestromal

(Reinstein et al. 2010). Esta técnica recibió el nombre de Refractive Lenticule Extraction

(ReLEx) por la generación del lentículo, mientras que la extracción del mismo se llevaba a cabo

previo tallado de un flap corneal similar al del LASIK, procedimiento que se pasó a llamar

Femtosecond Lenticule Extraction (FLEx) (Sekundo et al. 2008). Los resultados preliminares de

la técnica se llevaron a cabo en 10 ojos miopes durante un periodo de seguimiento de 6 meses,

los cuales demostraron ser prometedores, siendo éste el origen del cambio de paradigma en la

cirugía refractiva láser corneal (Sekundo et al. 2008). Como en cualquier inicio, la técnica no

estaba exenta de posibilidades de mejora, entre ellas el tiempo de recuperación visual que

demostró estar influenciado por la dirección en la ejecución del tallado del lentículo (Shah &

Shah 2011).

1.2.4. Extracción del lentículo a través de una microincisión (SMILE).

La técnica FLEx representaba tan solo una rápida transición hacia la que sería la cirugía

refractiva corneal mínimamente invasiva. En el año 2011, el procedimiento evolucionó a la

técnica Small Incision Lenticule Extraction (SMILE)(Sekundo et al. 2011). A diferencia de la

FLEx, la extracción del lentículo ya no se llevaría a cabo a través de la creación de un flap

corneal sino mediante la ejecución de una pequeña incisión por la que se extraería el lentículo.

La Figura 1 muestra cada uno de los pasos llevados a cabo durante la cirugía refractiva laser

SMILE para la corrección de la miopía. En la página EyeTube (https://eyetube.net/video/relex-

______________________________________________________________________________________________

22

smile-step-by-step--esweq/) puede visualizarse un video presentado en el Congreso de la

Academia Americana de Oftalmología (ASCRS) en el año 2015 por el autor de esta tesis

doctoral en el que se explica paso por paso el desarrollo de la técnica (Fernández 2015).

Los resultados de la cirugía SMILE han ido mejorando desde sus inicios. En la Tabla 2

se presenta una revisión de la literatura científica en términos de efectividad y predictibilidad

que demuestra una mejoría en los años 2013-2014 frente a los resultados obtenidos durante los

dos primeros años 2011-2012 (Reinstein et al. 2014). Algunas de las razones que podrían

explicar esta mejoría con el paso de los años podría residir en una mayor experiencia de los

cirujanos en el desarrollo de la técnica, ya que como podemos ver en la Figura 1, la técnica

SMILE posee diferentes pasos en los que la experiencia del cirujano podría ser clave. Por

ejemplo, antes de la acción del láser se requiere que el cirujano centre el cono que realizará

succión en el ojo (paso conocido como docking) sobre la pupila del paciente o el vertex corneal

(Li et al. 2014), además de ser necesaria una intervención manual durante la disección de las

superficies anterior y posterior del lentículo que podría estar relacionada con la presencia de

irregularidades en la membrana de Bowman, conocidas como microdistorsiones (Yao et al.

2013).

Figura 1. Pasos de la cirugía refractiva corneal SMILE durante la ejecución del láser: (1) tallado de la superficie posterior del lentículo o zona óptica; (2) tallado de la superficie anterior conocida como cap; (3) ejecución de la incisión en la parte inferior de la imagen correspondiente a la región superior-temporal del ojo. Durante el proceso quirúrgico: (4) delineación o reconocimiento de las superficies anterior y posterior del lentículo; (5) disección de la superficie anterior del lentículo; (6) disección de la superficie posterior del lentículo; (7) extracción del lentículo a través de la microincisión; (8) secado y regularización de la superficie.


23

Tabla 2. Resultados en términos de eficacia y predictibilidad en 2011/12 frente a 2013/14 (Tabla adaptada de (Reinstein et al. 2014).

Estudio Ojos Seguimiento ±0.50D AVsc >20/20 post AVsc >20/25 post

Sekundo 2011

Shah 2011

Vestegaard 2012

Hjortdal 2012

91

51

127

670

6 meses

6 meses

3 meses

3 meses

80%

91%

77%

80%

84%

62%

37%

61%

92%

79%

73%

84%

Promedio 2011/12 82% 61% 82%

Wang 2013

Kamiya 2014

Sekundo2014

Agca 2014

Lin 2014

88

26

54

40

60

3 meses

6 meses

1 año

1 año

3 meses

-

100%

92%

95%

-

100%

96%

88%

65%

85%

-

-

98%

95%

93%

Promedio 2013/14 96% 87% 95%

1.3. Controversias en torno a SMILE frente a LASIK.

1.3.1. Sensibilidad corneal y sequedad ocular.

La córnea es uno de los tejidos humanos periféricos con mayor densidad nerviosa. Una

analogía para entender de manera sencilla como se distribuyen los nervios dentro de la córnea es

visualizar el desarrollo de ramas y hojas de una Acacia tortilis (Figura 2). Haces gruesos de

nervios penetran en la periferia de la córnea, cerca del limbo esclerocorneal, a unas 293 ± 106

µm de profundidad con respecto a la superficie corneal (Figura 1a). Justo tras penetrar en la

córnea los haces principales se dividen de manera progresiva en nervios cada vez más pequeños

(plexo en estroma medio, Figura 2b). Muchos de estos nervios continúan incrementando su

densidad hacia la superficie (plexo subepitelial, Figura 2c). Algunos de estos pequeños nervios

penetran en el epitelio (plexo subbasal y terminales intraepiteliales, Figura 2d) (Marfurt et al.

2010).

______________________________________________________________________________________________

24

Figura 2. Analogía entre el crecimiento de una Acacia Tortilis y la ramificación nerviosa dentro de la

córnea conforme nos aproximamos a su superficie. Imagen a partir de (Marfurt et al. 2010) .

La sección de la córnea para la creación de un flap con el fin de aplicar el láser excimer

en cirugía LASIK conlleva la disección de gran parte de estas fibras nerviosas reduciendo la

sensibilidad corneal y con ello la producción de lágrima basal. El LASIK y la PRK provocan

una pérdida de la densidad nerviosa en el plexo subbasal del 51% y 59%, respectivamente, que

se recupera hasta diferencias no significativas con los valores preoperatorios a los dos años en

PRK y a los cinco en LASIK (Erie et al. 2005). En cuanto a la cirugía SMILE, encontramos

ciertas discrepancias en los periodos de recuperación pese a que los resultados siempre son

favorables a SMILE (He et al. 2015). Por ejemplo, Li reportó una pérdida menos severa que en

FS-LASIK durante los tres primeros meses (Li et al. 2013), mientras que Ishii encontró

diferencias significativas al año en FLEx con respecto a los valores preoperatorios, pero no en el

caso de SMILE (Ishii et al. 2015). En resumen, los meta-análisis coinciden en que la pérdida de

sensibilidad corneal y la sequedad ocular inducida por la cirugía es menor en SMILE que FS-

LASIK (Zhang et al. 2015; Shen et al. 2016; Kobashi et al. 2016).

1.3.2. Tiempo de recuperación visual.

El tiempo de recuperación visual tras la cirugía ha sido una de las desventajas de

SMILE frente a LASIK desde el nacimiento de la técnica. Pese a que el tiempo de recuperación

visual ha disminuido aplicando ciertas modificaciones dentro de la técnica como la dirección del

láser en el tallado del lentículo (Shah & Shah 2011), existe una gran variedad en los resultados

clínicos reportados por distintos autores sobre el porcentaje de sujetos que alcanzan agudeza

visual decimal unidad a las 24 horas del procedimiento (Tabla 3).

Esta elevada discrepancia entre estudios clínicos hace pensar sobre la influencia de la

habilidad del cirujano, del instrumental o de otras variables en el tiempo de recuperación visual

(Y. Liu et al. 2016). Agca reportó una mayor dispersión lumínica en SMILE que LASIK en la

interfaz del lentículo, hipotetizando que podría ser debido a imprecisiones en el corte del

lentículo por la configuración del láser (Agca, Ozgurhan, Yildirim, et al. 2014). Estos hallazgos

están en acuerdo con los resultados que demuestran una temprana recuperación visual llevando

la energía del láser a 100 nJ (nivel 20 del láser), en lugar de 180 nJ (nivel 36 del láser)(Donate

& Thaëron 2016), pero sin diferencias en términos de calidad visual, incluyendo scattering, para

comparativa de energías de 140 nJ y 170 nJ (Kamiya et al. 2015). Otros autores han encontrado

que reposicionando el cap tras la cirugía presionando la superficie y deslizando una espátula de

Seibel unas 20 veces de arriba hacia abajo obtenían un menor número de microdistorsiones,

presentes especialmente en altos errores refractivos (Luo et al. 2015), y una mejor función de

transferencia de modulación (MTF)(Shetty et al. 2016). Este gesto ayudaría a eliminar el líquido


25

alojado en el bolsillo donde se aloja el lentículo tras la irrigación efectuada para eliminar

desechos celulares que medien la inflamación (Liu et al. 2015).

Tabla 3. Porcentaje de sujetos que alcanzan una agudeza visual decimal unidad al día, 3

meses o 6 meses según diferentes estudios.

Estudio 1 día 3 meses 6 meses

(M. Liu, Chen, et al. 2016)

Frente a FS-LASIK*

(Xu & Yang 2015)

(Kim et al. 2014)

(Zhao et al. 2015)

(Shetty et al. 2016)

(Ganesh & Gupta 2014)

Frente a FS-LASIK*

(Kim et al. 2015)

Baja miopía

Alta miopía

(Donate & Thaëron 2016)

100 nJ (nivel 20)

180 nJ (nivel 36)

(Luo et al. 2015)

> -3.00D

-3.00D a -6.00D

< -6.00 D

55%

73%*

63%

54.4%

100%

100%

96%

88%*

65.5%

63.2%

66%

23%

88.2%

69.2%

65%

93%

96%

85%

-

-

-

-

-

87.5%

78.8%

-

-

-

-

-

96%

99%

-

79.8%

-

-

96%

88%*

-

-

-

-

-

-

-

*Comparativa con los resultados de LASIK incluidos en el mismo estudio.

1.3.3. Resultados en función de las características del paciente o de la cirugía.

Los estudios clínicos en los inicios de cualquier procedimiento de cirugía refractiva se

focalizan principalmente en determinar la seguridad, eficacia y predictibilidad de la técnica,

definiendo cada uno de estos parámetros como:

Seguridad, complicaciones tras la operación y comparativa de las agudezas

visuales pre y postoperatorias con corrección.

Eficacia, como la diferencia entre la agudeza visual preoperatoria con

corrección y la postoperatoria sin corrección.

Predictibilidad, correlación entre el error corregido y el que se ha programado

corregir o el porcentaje de sujetos con un cierto residual refractivo tras la

operación.

La cirugía refractiva SMILE ha demostrado, en términos generales, ser tan segura,

eficaz y predecible, como la cirugía refractiva FS-LASIK según se recoge en los dos principales

meta-análisis comparando ambas técnicas llevados a cabo hasta la fecha (Shen et al. 2016;

Zhang et al. 2015). No obstante, una vez asentado el conocimiento general de seguridad,

______________________________________________________________________________________________

26

eficacia y predictibilidad se hace necesario estudiar los resultados de la técnica de acuerdo a las

particularidades de cada caso con el fin de establecer el origen de posibles sesgos que puedan

empeorar los resultados visuales. Por ejemplo, se ha reportado que con el incremento de la edad

los resultados de agudeza visual sin corrección 6 meses tras la cirugía empeoran a razón de una

pérdida de 0.07 logMAR por década (Chansue et al. 2015). Además, las figuras de

predictibilidad que confrontan el error tratado con el error corregido por el procedimiento

muestran, en algunos estudios, una pendiente por debajo de la unidad que sugiere una

hipocorrección con el incremento de la miopía (Lin et al. 2014; Chansue et al. 2015), hecho no

visible en autores que utilizan su propio nomograma (Pradhan et al. 2016) o inclusive en otros

que no especifican utilizar un nomograma personalizado (Hansen et al. 2016). No obstante, se

han propuesto regresiones lineales para corregir la hipocorrección especialmente por encima de

las 6 D de miopía tras manifestar este grupo de pacientes un mayor regresión miópica al año

(W. Wu et al. 2016). Además se ha reportado la influencia de otras variables como la edad en

los resultados visuales (Kim et al. 2014; Hjortdal et al. 2012), mientras que muchas otras

variantes de la cirugía no tienen afectación en los resultados visuales: espesor del cap o

profundidad del lentículo entre 120 y 140 µm (M. Liu, Zhou, et al. 2016) o entre 130 y 160 µm

(Güell et al. 2015), y localización de la incisión temporal o superior (Chan et al. 2016).

1.3.4. Calidad óptica ocular.

Los resultados de eficacia de un procedimiento de cirugía refractiva se dan en términos

de agudeza visual como vimos en el apartado anterior. No obstante, dos procedimientos pueden

resultar en eficacia similar en términos de agudeza visual pero ofrecer distintos resultados en

términos de rendimiento visual al evaluar variables más sensibles a pequeños cambios en la

calidad óptica ocular (Rodríguez-Vallejo et al. 2015). Desde el punto de vista objetivo, podemos

emplear instrumentos como aberrómetros, topógrafos corneales y sistemas de doble paso con el

fin de medir la calidad óptica tras el procedimiento ante distintas condiciones de diámetro

pupilar (Donate & Thaëron 2016; Güell et al. 2015). De manera particular, el centrado manual

por parte del cirujano y la ausencia de eye-tracking han sido valorados como debilidades de la

técnica SMILE, que podrían derivar en la incidencia de un mayor número de aberraciones

corneales, especialmente coma secundario a una ablación descentrada durante el periodo de

aprendizaje (Li et al. 2014). Si bien es cierto que desde el punto de vista teórico se ha

argumentado que un descentramiento en SMILE no posee el mismo impacto en el incremento

de las aberraciones de alto orden que un descentramiento en LASIK o PRK (Bischoff &

Strobrawa 2016). Puesto que los meta-análisis no suelen incluir amplia información sobre la

calidad óptica y las aberraciones de alto orden tras SMILE frente a LASIK, hemos recogido

dentro de la tabla 4 muchos de los estudios que incluyen información sobre las aberraciones de

alto orden, principalmente coma y aberración esférica.


27

Tabla 4. Estudios que comparan las aberraciones inducidas por el procedimiento (Postoperatorio –

Preoperatorio) entre SMILE y LASIK.

Mes Autor Aberración SMILE (A) LASIK (B) A-B ZO / Pupila

6 m (Ye et al. 2016) Esférica

|Coma H|

|Coma V|

0.17*

0.08*

0.22*

0.29*

0.01

0.18*

-0.12*

0.07*

0.04*

6.0 mm (A)

6.0 mm (B)

6.0 mm (C)

6 m (M. Liu, Chen, et al.

2016)

Esférica

Coma

0.12 ± 0.22

0.20 ± 0.21

0.28 ± 0.26

0.24 ± 0.24

-0.16*

-0.04

6.5 mm (A)

NE (B)

6 mm (C)

3 m (Gyldenkerne et al.

2015)

Esférica

Coma

0.0072 ± 0.095

0.14 ± 0.15

0.15 ± 0.084

0.23 ± 0.20

-0.14*

-0.10*

6 a 6.5 mm (A)

6 mm (B)

5 mm (C)

3 m (Ganesh & Gupta

2014)

RMS 0.061 0.174 -0.113* 6 a 6.5 mm (A)

NE (B)

5 mm (C)

3 m (Lin et al. 2014) Esférica

Coma

RMS

0.273

0.396

0.117

0.689

0.193

0.204

-0.416*

+0.203*

-0.087*

6.36 ± 0.23 (A)

6.02 ± 0.19 (B)

NE (C)

6 m (Gao et al. 2014) Esférica

Coma

0.12 ± 0.22

0.20 ± 0.21

0.28 ± 0.26

0.24 ± 0.24

-0.16*

-0.04*

NE (A)

NE (B)

6 mm (C)

6 m (Li et al. 2015) Total

Esférica

|Coma H|

|Coma V|

0.619

0.298

0.456

0.149

0,801

0,397

0,503

0,522

-0,182

-0,099

-0,047

-0,373

6 mm (A)

6 mm (B)

6 mm (C)

3 m (Wu & Wang 2015) 3er a 6º

Esférica

|Coma H|

|Coma V|

> tras cirugía*

> tras cirugía*

> tras cirugía*

> tras cirugía*

> tras cirugía*

> tras cirugía*

> LASIK*

> LASIK*

> LASIK*

> SMILE*

6 mm (A)

6 mm (B)

6 mm (C)

3 m (Wu & Wang 2016) Total

Esférica

Coma

0.45*

0.26*

0.35*

0.55*

0.33*

0.4*

-0,1

-0,07

-0,05

6 a 6.5 mm (A)

6 a 6.5 mm (B)

6 mm (C)

3 m (Yu et al. 2015) Total

Esférica

Coma

> tras cirugía*

> tras cirugía

> tras cirugía*

> tras cirugía*

> tras cirugía*

> tras cirugía*

> LASIK*

> LASIK*

> SMILE

6.5- 6.6 mm (A)

6.25-6.75mm (B)

6 mm (C)

ZO es el tamaño de zona óptica para (A) SMILE o (B) LASIK mientras que (C) representa el tamaño de pupila para el cual se han calculado las aberraciones de alto orden. * Las diferencias son significativas entre el pre y el post o para las diferentes técnicas.

De la revisión llevada a cabo en 10 estudios que incluyen las aberraciones de alto orden

(Tabla 4) podemos concluir que la aberración esférica es mayor tras un procedimiento LASIK

que en SMILE, algo en lo que coinciden todos los estudios consultados. No obstante, existe

controversia en torno a la mayor inducción de coma en SMILE con 4 estudios que resultaron en

una mayor inducción y 5 estudios en los que la inducción de coma fue menor con respecto a

LASIK, por lo que es posible que la experiencia del cirujano en el centrado esté relacionada con

la inducción de una mayor aberración comática.

______________________________________________________________________________________________

28

1.3.5. Biomecánica corneal.

Tras la aparición de la técnica SMILE se hipotetizó que ésta preservaba de mejor forma

la biomecánica corneal en comparativa con la técnica LASIK (Reinstein et al. 2013). Esta

hipótesis se fundamenta en la organización que mantienen las fibras de colágeno en distintas

profundidades del estroma. En el estroma anterior las fibras de colágeno mantienen una mayor

angulación, penetrando en la membrana de Bowman y formando un entramado similar a la

estructura que soporta la parte inferior de un puente (del inglés Bow spring-like structure)

(Abass et al. 2015). Se ha reportado que esta angulación es notablemente más marcada en las 83

µm del estroma anterior y disminuye con la profundad corneal hasta las 250 µm (Morishige et

al. 2011). La región media y posterior del estroma se caracteriza por una organización de fibras

de colágeno que se distribuyen de manera paralela a la superficie posterior en dos direcciones

principales, nasal-temporal y superior-inferior (Benoit et al. 2016). Esta organización que

mantienen las fibras de colágeno conlleva que la parte anterior del estroma posea un mayor

módulo de elasticidad que las regiones media y posterior (Quantock et al. 2015).

La hipótesis de que SMILE preserva la biomecánica corneal en mayor medida que

LASIK posee un notable fundamento teórico si consideramos que: (1) el estroma anterior posee

una mayor rigidez; (2) el lentículo se talla a partir de una profundidad mayor a las 100 m de la

superficie corneal; (3) el lentículo se extrae a través de una micro-incisión de 2 mm por lo que

se secciona un considerable número menor de fibras en comparación a la creación de un flap.

No obstante, esta hipótesis, más allá de los modelos matemáticos y biomecánicos (Reinstein et

al. 2013; Sinha Roy et al. 2014), no ha demostrado consenso en sus evidencias clínicas. De

hecho, una de las principales preocupaciones por la cual la biomecánica corneal es de gran

importancia en el campo de la cirugía refractiva, además de su relación con los resultados

refractivos (Roy & Dupps 2009), es su papel en el desarrollo de ectasias corneales (Klein et al.

2006). Y a día de hoy, córneas sospechosas de queratocono subclínico desde el punto de vista

topográfico para las cuales la cirugía LASIK está contraindicada, han terminado desarrollando

una ectasia corneal al no asumir un criterio clínico conservador y aplicar una cirugía SMILE

(Randleman 2016). Por lo que los criterios clínicos de screening preoperatorio deben ser tan

rigurosos como los que hemos aprendido de nuestra experiencia con LASIK.


29

1.4. Evaluación clínica de la biomecánica corneal.

1.4.1. Instrumentos clínicos para la medida de la biomecánica

corneal.

Pese al creciente interés en la biomecánica corneal, tan solo se encuentran disponibles

en la actualidad dos instrumentos enfocados a su medida clínica. El primero de ellos, Ocular

Response Analyzer (ORA) (Reichert Inc. USA), apareció en el año 2005 con el objetivo de

realizar una medida de la presión intraocular (PIO) que estuviese menos afectada por el espesor

y las propiedades de la córnea en comparación con otros tonómetros (Glass et al. 2008). El

funcionamiento de este instrumento es similar al de un tonómetro de aire con una duración de

pulso de unos 20 ms. La presión ejercida por el pulso de aire aplana la córnea, la cual pasa a

tener una forma cóncava en el momento de máxima presión, volviendo a su posición inicial tras

el cese del pulso. Durante el movimiento de la córnea con el pulso, un haz de luz infrarrojo

incide lateralmente sobre la misma de tal manera que cuando la córnea se aplana, un

fotodetector situado en la posición opuesta recibe el flujo máximo de fotones, reconociendo de

esta forma la presión necesaria para alcanzar la aplanación (Figura 3). Puesto que la córnea pasa

dos veces por el estado de aplanación, una primera con el incremento de la presión y otra

cuando vuelve a su posición original, el instrumento captura las presiones ejercidas en el

momento de ambas aplanaciones (Fernández & Martínez 2015).

Figura 3. Diagrama que muestra las señales máximas alcanzadas por el receptor (línea roja) en la primera

______________________________________________________________________________________________

30

y segunda aplanación frente a la curva de presión del pulso de aire (línea verde).

La diferencia entre las presiones recibe el nombre de histéresis corneal (del inglés

Corneal Hysteresis, CH), mientras que una segunda variable denominada factor de resistencia

corneal (del inglés Corneal Resistance Factor, CRF) proviene de una función lineal P1-kP2,

donde k es una constante determinada a partir de un análisis empírico de la relación entre

presiones y espesor corneal (Piñero & Alcón 2015).

El segundo de los instrumentos es el Corvis ST (Oculus, Wetzlar, Germany) cuyos

primeros resultados clínicos fueron presentados en el año 2011 (Hon & Lam 2013). Su

funcionamiento se basa también en la deformación de la córnea a través de un pulso de aire pero

se captura a través de una cámara Scheimplflug de alta velocidad (4330 imágenes por segundo)

una sección horizontal de la córnea (Hon & Lam 2013). De esta forma, se obtiene un video

durante los 30 ms que dura el pulso de aire, pasando la córnea por tres estados: de primera

aplanación (A1), máxima concavidad (del inglés Highest Concavity, HC) y segunda aplanación

(A2). La principal ventaja de este sistema frente al ORA es que en todo momento visualizamos

la dinámica de la córnea y no solo capturamos los puntos de presión en los que se producen las

aplanaciones por lo cual el número de parámetros que podemos inferir a partir de las imágenes

es considerablemente mayor.

Figura 4. Fotogramas del video capturado por Corvis ST durante la fase de resistencia de la córnea al pulso de aire (izquierda) y la fase de recuperación tras el cese (derecha). Sobre la imagen se han dibujado en amarillo algunos de los parámetros inferidos durante la primera aplanación, máxima concavidad y segunda aplanación.


31

En la literatura científica, se recogen multitud de parámetros inferidos por el software,

aunque en este apartado tan solo vamos a hacer referencia a algunos de los más utilizados:

Tiempos de aplanación: El tiempo en el que se alcanzan las aplanaciones A1 y A2. Nos

referiremos a estas variables como AT1, y AT2.

Velocidad de aplanación: La velocidad de la córnea durante A1 y A2 (AV1 y AV2).

Longitud de aplanación: Longitud horizontal de la aplanación (AL1 y AL2)(Figura 4,

parámetros 1 y 5).

Amplitud de deformación: El desplazamiento que se produce en el ápex desde la

posición original hasta la máxima concavidad (DA)(Figura 4, parámetro 2).

Radio de la máxima concavidad: El radio de una circunferencia ajustada a la córnea en

la máxima concavidad (Figura 4, parámetro 3).

Distancia entre picos: Distancia entre los picos más elevados en la máxima concavidad

(Figura 4, parámetro 4)(PD).

Presión intraocular: Presión ejercida por el pulso de aire cuando se alcanza A1 (IOP).

Espesor corneal central: Espesor corneal medido en el centro de la córnea antes de que

esta sea deformada (CCT).

Éstas son solo algunas de las variables inferidas por el software comercial, sin embargo

muchas otras no han sido descritas por su menor popularidad o porque se encuentran

disponibles solo con motivos de investigación en versiones del software enfocadas para tal

propósito. No obstante, es importante resaltar que muchas de las variables anteriormente

descritas han sido cuestionadas por su baja repetibilidad como AV1, AV2, AL1, AL2 y PD

(Hon & Lam 2013; Nemeth et al. 2013; Chen et al. 2014).

1.4.2. Medida clínica de biomecánica corneal tras cirugía refractiva.

En los últimos años ha habido un incremento del interés en la investigación clínica de la

biomecánica corneal en SMILE, tanto en referencia a variaciones dentro de la propia técnica,

como en comparativa con otras técnicas que teóricamente deberían preservar en menor medida

la integridad corneal. La mayor parte de estas investigaciones han sido llevadas a cabo con el

instrumento ORA como muestran 12 de las 16 referencias consultadas incluidas en la Tabla 5.

De entre las conclusiones principales, se encuentra una baja evidencia de que SMILE afecte en

menor medida a la biomecánica corneal con respecto a otras técnicas, aunque algunos autores

apuntan a particularidades como dependencia del error refractivo, profundidad del lentículo o

tamaño de la incisión. En torno al instrumento Corvis ST, tan solo 5 de las 16 referencias

consultadas incluyen medidas con este instrumento, las cuales apuntan de igual forma que el

ORA a la falta de evidencia clínica de mayor preservación de la integridad corneal y a

diferencias en torno a variaciones en la técnica como puede ser la profundidad del lentículo.

______________________________________________________________________________________________

32

Tabla 5. Estudios que analizan la biomecánica corneal en SMILE, tanto en relación a variaciones de la

técnica como en comparativa con otras técnicas de cirugía refractiva.

Autor Instrumentos Técnicas Conclusión

(Pedersen et al. 2014) ORA y CORVIS FS-LASIK, FLEX, SMILE Sin diferencias entre técnicas

(Shen et al. 2014) CORVIS SMILE

Diferencias pre/post debidas a la

extracción del lentículo y no al

tallado del mismo.

(Leccisotti et al. 2016) CORVIS LASIK Diferencias tras el tallado del

lentículo sin la retirada del mismo.

(Wang et al. 2014) ORA FS-LASIK, SMILE Diferencias para miopías mayores

a 6D a favor de SMILE

(Sefat et al. 2016) CORVIS FS-LASIK, SMILE Sin diferencias entre técnicas

(Y Shen et al. 2014) CORVIS FS-LASIK, LASEK,

SMILE

Sin diferencias entre SMILE con

respecto al resto de técnicas

(El-Massry et al. 2015) ORA SMILE Menor afectación biomecánica a

160 m que a 100 m

(Agca, Ozgurhan,

Demirok, et al. 2014) ORA SMILE, FS-LASIK Sin diferencias entre técnicas

(Osman et al. 2016) ORA SMILE, FS-LASIK Mayor afectación en FS-LASIK

(Wu et al. 2014) ORA SMILE, FS-LASIK Mayor afectación en FS-LASIK

(Wang et al. 2016) ORA SMILE, FS-LASIK Mayor afectación en FS-LASIK

(He et al. 2016) CORVIS SMILE Menor afectación biomecánica a

160 m que a 100 m

(Kamiya et al. 2014) ORA SMILE, FLEX La realización del flap no provoca

diferencias entre técnicas

(Zhang et al. 2016) ORA SMILE, FS-LASIK WF* Sin diferencias entre técnicas

(Z. Wu et al. 2016) ORA SMILE

Mayor tamaño de la incisión

provoca mayor variación en las

variables

(Yıldırım et al. 2016) ORA SMILE; PRK Smile induce mayores cambios en

las variables

(Chen et al. 2016) ORA SMILE; LASEK Smile induce menores cambios en

las variables

*FS-LASIK WF: Técnicas LASIK con flap por femtosegundo y láser excimer guiado por frente de onda.

1.5. Justificación y objetivos.

Durante esta introducción hemos presentado la técnica de cirugía refractiva SMILE

junto con algunas controversias actuales. Podemos afirmar, como ya expusimos en el apartado

1.3.3, que la técnica SMILE es en términos generales tan segura, eficaz y predecible como la

técnica LASIK. No obstante, es importante resaltar que la cirugía SMILE podría considerarse

más cirujano-dependiente por incluir dentro del procedimiento la disección y extracción del

lentículo, donde la experiencia del cirujano podría influir en los resultados. El primero de los

objetivos de la presente Tesis doctoral es:

1. Analizar, en función del error refractivo tratado, los resultados de eficacia, seguridad y

predictibilidad de la técnica SMILE para los primeros casos llevados a cabo por un

cirujano con experiencia en cirugía refractiva, pero no experimentado en la técnica.


33

Muchas de las hipótesis surgidas en relación a SMILE, como la menor inducción de

sequedad ocular gracias a la disección de un menor número de fibras nerviosas corneales, se

encuentran clínicamente demostradas. No obstante, la controversia más importante en la

actualidad en relación a la técnica gira en torno a la preservación de la biomecánica corneal.

Pese a que existen simulaciones basadas en modelos matemáticos y biomecánicos como

relatamos en el apartado 1.3.5 que anuncian una mayor preservación de la biomecánica corneal

en SMILE que en LASIK, esta hipótesis no ha llegado a ser confirmada por la investigación

clínica, posiblemente debido a la falta de instrumentos fiables que permitan medir, de manera

aislada, los cambios producidos en la rigidez de la córnea debido a procedimientos de cirugía

refractiva láser corneal. La aparición de nuevas tecnologías, como el sistema Corvis ST, abre

nuevas posibilidades dentro de la investigación clínica relacionada con la biomecánica corneal

dentro la cirugía refractiva laser. Esta Tesis doctoral abarca otros tres objetivos principales

relacionados con el instrumento Corvis ST y la biomecánica corneal tras SMILE:

2. Evaluar los cambios en los parámetros más reproducibles del Corvis ST en función del

error refractivo tratado y analizar las variaciones de estos parámetros en función del

espesor corneal retirado.

3. Examinar los nuevos parámetros introducidos en la versión del software 1.3r1469

(Septiembre, 2016), evaluando como cambian tras SMILE y la dependencia de estos

parámetros frente a variables de confusión como el espesor corneal.

4. Presentar una nueva hipótesis sobre la densitometría dinámica y su posible relación

con la biomecánica e hidratación corneal y, por consiguiente, su potencial aplicación

para la evaluación de los cambios biomecánicos tras SMILE.

______________________________________________________________________________________________

34

1.6. Estructura de la Tesis.

Esta Tesis ha sido llevada a cabo mediante la compilación de artículos científicos. Cada

uno de estos artículos posee una estructura de Introducción, Métodos, Resultados y Discusión

que permite su compresión de manera individual. No obstante, los tres componen un solo

trabajo con un claro hilo argumental.

La Tesis se estructura en 4 Capítulos:

1. Introducción General.

2. Publicaciones.

2.1. Short-term outcomes of Small Incision Lenticule Extraction (SMILE) for low,

medium and high myopia.

2.2. Corneal thickness after SMILE affects Scheimpflug-based Dynamic

Tonometry.

2.3. New parameters for evaluating corneal biomechanics and intraocular pressure

after SMILE by Scheimpflug-Based Dynamic Tonometry.

3. Discusión de los resultados.

4. Conclusiones.

El contenido principal de la Tesis se encuentra recogido en el Capítulo 2 que incluye 3 artículos

publicados por revistas de prestigio internacional.

El primer artículo se titula “Short-term outcomes of Small Incision Lenticule Extraction

(SMILE) for low, medium and high myopia” (Fernández, Valero, et al. 2016). Los resultados

preliminares se presentaron en el XXIV Congreso OPTOM (Hueso et al. 2016) y,

posteriormente, este artículo se ha publicado en la revista “European Journal of

Ophthalmology”. Esta revista está soportada por Thomson Reuters en el Science Citation Index

(SCI). En el año 2015 su factor de impacto ha sido de 1,007, ocupando una posición relativa de

46/56 (Q4) en la categoría “Ophthalmology” del Journal Citation Rank. Esta revista publica

artículos clínicos originales de revisión por pares que muestren observaciones clínicas e

investigaciones de laboratorio con relevancia clínica, enfocada a nuevas técnicas de diagnóstico

y quirúrgicas, actualizaciones en terapias e instrumentos, resultados de ensayos clínicos y

hallazgos de investigación. Todos los artículos de esta revista se han sometido a una rigurosa

revisión por pares, basado en el cribado inicial y arbitraje de doble ciego de dos evaluadores

anónimos internacionales.

En este primer artículo se evalúan los resultados de seguridad, eficacia y predictibilidad de la


35

técnica SMILE para los primeros 71 casos llevados a cabo por el autor de esta Tesis doctoral.

Estos resultados son de gran importancia para cualquier cirujano que se inicia en una nueva

técnica de cirugía refractiva. Pese a tener experiencia en otro tipo de cirugías refractivas

mediante láser, es importante conocer si los primeros resultados pueden verse afectados por la

experiencia del cirujano. Además, este trabajo se diferencia principalmente de otros similares en

que el análisis se lleva a cabo en función del error refractivo, lo cual aporta un valor adicional al

conocer cuáles deben ser las características refractivas de los pacientes durante la curva de

aprendizaje.

El segundo artículo se titula “Corneal thickness after SMILE affects Scheimpflug-based

Dynamic Tonometry” (Fernández, Rodríguez-Vallejo, Martínez, Tauste & Piñero 2016). Los

resultados preliminaries se presentaron en el XXXIV Congress of the ESCRS (Fernandez et al.

2016) y, posteriormente, este artículo se ha publicado en la revista “Journal of Refractive

Surgery”. Esta revista está soportada por Thomson Reuters en el Science Citation Index (SCI).

En el año 2015 su factor de impacto ha sido de 3,314, ocupando una posición relativa de 7/56

(Q1) en la categoría “Ophthalmology” del Journal Citation Rank. Esta revista publica artículos

de investigación, revisiones y evaluación de procedimientos de cirugía refractiva. Todos los

artículos de esta revista se han sometido a una rigurosa revisión por pares, basado en el cribado

inicial y arbitraje de doble ciego de dos evaluadores anónimos internacionales.

En este segundo artículo se analiza la variación de los parámetros más repetibles del sistema

Corvis ST frente al espesor corneal eliminado con la extracción del lentículo. Una de las

principales limitaciones de los instrumentos de medida clínica de la biomecánica corneal es la

gran influencia de variables de confusión como la presión intraocular o el espesor corneal. En

este trabajo, mediante la inclusión de datos preoperatorios y posoperatorios, se asume que la

presión intraocular no influye en la medida mientras que se analiza el impacto del espesor

corneal. A partir de este análisis se proponen soluciones para la corrección de los parámetros del

Corvis ST con el fin de minimizar el impacto de la variación de espesor corneal.

El tercer artículo se titula “New parameters for evaluating corneal biomechanics and

intraocular pressure after SMILE by Scheimpflug-Based Dynamic Tonometry.” (Fernández,

Rodríguez-Vallejo, Martínez, Tauste, Salvestrini, et al. 2016). Este artículo ha sido aceptado en

la revista “Journal of Cataract and Refractive Surgery”, estando aún pendiente de publicación.

Esta revista está soportada por Thomson Reuters en el Science Citation Index (SCI). En el año

2015 su factor de impacto ha sido de 3,020, ocupando una posición relativa de 12/56 (Q1) en la

categoría “Ophthalmology” del Journal Citation Rank. Esta revista publica artículos de

investigación, revisiones y evaluación de procedimientos de cirugía refractiva. Todos los

artículos de esta revista se han sometido a una rigurosa revisión por pares, basado en el cribado

______________________________________________________________________________________________

36

inicial y arbitraje de doble ciego de dos evaluadores anónimos internacionales.

Este tercer artículo supone una continuación del artículo anterior. Se incluyen nuevos

parámetros de biomecánica corneal añadidos en una importante actualización del software

llevada a cabo en Septiembre de 2016. Se analiza si estos nuevos parámetros se encuentran

sometidos a las mismas limitaciones que los descritos en el trabajo anterior. Además se

propone, por primera vez en investigación clínica en el campo de la oftalmología, la medida de

la densitometría corneal dinámica con el planteamiento de nuevas hipótesis que podrían

relacionar este parámetro con la biomecánica corneal y con la hidratación de la córnea.

En el Capítulo 3 se presenta una breve discusión acerca de los principales resultados mientras

que el Capítulo 4 muestra las conclusiones finales de la Tesis así como el cumplimiento de los

objetivos planteados. Para finalizar, se muestra la bibliografía general utilizada a lo largo de la

Tesis.


37

CAPÍTULO 2 PUBLICACIONES

______________________________________________________________________________________________

38


39

______________________________________________________________________________________________

40


41

2.1. Short-term outcomes of Small Incision Lenticule Extraction (SMILE)

for low, medium and high myopia.

Joaquín Fernández, MD;1 Almudena Valero, MD;1 Javier Martínez, OD;1 David P Piñero, PhD;2,3 Manuel Rodríguez-Vallejo, MSc *1

1Department of Ophthalmology (Qvision), Vithas Virgen del Mar Hospital, 04120, Almería, Spain 2Department of Optics, Pharmacology and Anatomy, University of Alicante, Alicante, Spain 3Department of Ophthalmology (OFTALMAR), Vithas Medimar International Hospital, Alicante, Spain *Corresponding author: [email protected] (Tel +34686500808)

2.1.1. Abstract

Purpose: To determine the safety, efficacy, and predictability of Small Incision Lenticule

Extraction (SMILE) at 6-month follow-up, depending on the level of the myopic refractive

error. The surgeries were performed by a novel surgeon in this technique.

Methods: Seventy-one subjects with a mean age of 31.86 ± 5.57 were included in this

retrospective observational study. Subjects were divided into 3 groups depending on the

preoperative Spherical Equivalent (SE): Low group from -1.00 D to -3.00 D, Medium from -

3.25 D to -5.00 D, and High from -5.25 D to -7.00 D. Manifest refraction, corrected (CDVA)

and uncorrected distance visual acuity (UDVA) were measured before surgery and at 6 months

after the treatment.

Results: In total, 1.4% of the eyes lost 1 line of CDVA after the procedure, whereas 95.8%

remained unchanged and 2.8% gained 1 line. A significant undercorrection (p=0.031) was found

in the High myopia group (median: -0.50 D), whereas Low and Medium groups remained near

to emmetropia. In terms of efficacy, no statistically significant inter-group differences for

postoperative UDVA (p = .282) were found. The vector analysis also showed undercorrection of

the preoperative cylinder, even though the standard deviations decreased from 0.9 D in the x

axis and 0.7 D in the y axis to 0.24 D and 0.27 D, respectively.

Conclusions: SMILE might be a safe, effective and predictable procedure even for non-

experienced surgeons. No differences in efficacy were found among myopia levels even though

undercorrections were found for SE and cylinder in high myopia.

Key Words: myopia, refractive surgery, small incision lenticule extraction, SMILE

______________________________________________________________________________________________

42

2.1.2. Introduction

Small-incision lenticule extraction (SMILE) has demonstrated similar or better visual and

refractive outcomes compared to traditional laser-assisted in situ keratomileusis (LASIK) in the

treatment of myopia, with the additional advantage of using only a laser system for the entire

procedure (Ang et al. 2012; Ganesh & Gupta 2014). Furthermore, some studies have shown

better results with this technique in optical quality than femtosecond LASIK (FS-LASIK), with

lower induction of higher-order aberrations (Gyldenkerne et al. 2015; Lin et al. 2014). However,

some authors suggest that SMILE has poorer visual recovery rates than LASIK (Vestergaard et

al. 2012), with considerable differences in uncorrected distance visual acuity (UDVA) achieved

between 1 day and 3 months (Hjortdal et al. 2012). It has been hypothesized that laser

parameters might affect visual recovery (Shah & Shah 2011; Kamiya et al. 2015), but other

recent studies have not found early postoperative interface scatter or delay in visual recovery

(Vestergaard et al. 2014) or have reported similar percentage for 20/20 or better at 1 day (89%)

in comparison to 1 month (91%) with SMILE (Reinstein, Carp, et al. 2014). Although longer

studies at 3, 6, and 12 months show that SMILE is a safe, effective, and predictable procedure

(Kim et al. 2014; Xu & Yang 2015), most are based on data from experienced SMILE surgeons

with personalized nomograms (Reinstein, Carp, et al. 2014) and either do not state whether the

surgeon had previous experience with SMILE before the study (Xu & Yang 2015) or analyze

the results independently of the degree of refractive error (Ivarsen et al. 2014; Li et al. 2014).

The aim of this study was to analyze the outcomes with SMILE at 6 months postoperatively

depending on the level of myopia for the first 71 consecutive patients of an inexperienced

surgeon.

2.1.3. Methods

Patients and examinations

One random eye of the first 71 consecutive patients treated with SMILE between September

2013 and January 2014 at Qvision (Department of Ophthalmology, Virgen del Mar Hospital)

were included in this retrospective observational study. Patients underwent a complete

preoperative eye examination including objective and subjective refraction performed by

optometrists, Goldmann intraocular pressure, aberrometry, pupil size, and corneal topography

with Orbscan II and Zywave systems (both from Bausch and Lomb, Rochester, NY, USA), slit-

lamp evaluation, and funduscopy. Postoperative visit at 6 months included UDVA, corrected

distance visual acuity (CDVA), manifest refraction, and slit-lamp examination to evaluate the

integrity of the anterior segment. Visual acuities (VAs) were measured with a LCD wall screen

in decimal notation scale and converted to Snellen for reporting in standardized graphs.


43

Inclusion criteria were patients undergoing 6-month follow-up after the procedure with

preoperative spherical equivalent (SE) between -1 D and -7 D and astigmatism under 3.5 D,

stable myopia for at least 1 year before surgery, and CDVA of 20/25 or better. Exclusion criteria

were pregnancy at time of surgery or follow-up, a preoperative central corneal thickness of less

than 480 μm, an expected postoperative residual stromal bed of less than 250 μm, topographic

map compatible with subclinical keratoconus or other ectatic corneal disorder, and any other

ocular disease for which laser refractive surgery procedures are not indicated (Shetty et al.

2015). The procedure was explained to all the patients who signed the preoperative informed

consent, and the study complied with the tenets of the Declaration of Helsinki.

Surgical procedure

Optical principles and general description of the SMILE procedure have been widely described

(Reinstein, Archer, et al. 2014); however, some particular laser settings or surgical maneuvers

can vary depending on the surgeon. The particular laser settings and maneuvers for this study

are detailed below. Before suction, centration was accepted when the ring of the applanation

zone was concentric with the margin of the cone and near to the pupil center (Li et al. 2014).

Suction was then applied, and a slight rotation of the applanation cone was made to compensate

for cyclotorsion in cases of high astigmatism with markings, taking as reference the horizontal

lines seen through the microscope.

The photodisruptive procedure follows the next sequence: the laser creates the lower lenticular

interface from center to periphery; this is the lenticule diameter or optical zone, which was set to

6.5 mm. A transition zone with a side cut angle of 90° follows the optical zone cut for

intersecting the upper lenticular interface. The laser creates the cap of 7.6 mm of diameter, 1.1

mm larger than the optical zone, in such a way that the lenticule is confined below the cap. The

depth or cap thickness was set to 140 μm and the laser software computes the lenticule thickness

depending on the refractive error, but a minimum thickness of 15 μm was configured. Finally,

the incision at the extreme of the cap, 2 mm of width, is created with a side cut angle of 30° for

extracting the lenticule close to 12 o’clock position. A graphical description for understanding

each one of these parameters has been detailed by other authors (Bischoff & Strobrawa 2016).

Laser configuration parameters were as follows: repetition rate of 500 kHz, spot distance of 4.50

μm for the lenticule and 2 μm for its border, and pulse energy of level 30 in the software, which

corresponds to approximately 150 nJ (Vestergaard et al. 2014). The target refractive error

correction was directly inserted in the software without applying any nomogram. After laser

treatment, the patient was moved to the surgical microscope for the second part of the

______________________________________________________________________________________________

44

procedure, which involves the following: (1) delineating front and back lenticule surfaces; (2)

surface separation using the standard lamellar corneal surgical technique of moving the

instrument back and forth using a blunt circular tip (Femto Double-Ended instrument [G-

33954], Carl Zeiss Meditec AG, Jena, Germany) starting with the complete dissection of the

front cap and following with the dissection of the posterior lenticule surface; (3) lenticule

extraction with forceps (Lenticule Forceps [G-33961], Carl Zeiss Meditec AG); (4) corneal

surface pressure from center to periphery using a dry microspear and drying the incision with

the same. Finally, all patients received 2 drops of tobramycin (3 mg) and dexamethasone (1 mg)

combination at the end of the procedure.

The same surgeon (J.F.) performed all the SMILE treatments with the VisuMax femtosecond

laser system (Carl Zeiss Meditec AG). It is important to note that this sample corresponded with

the first consecutive SMILE cases of this surgeon, including results from the early phase of the

learning curve. All patients were treated with 2 drops of topical anesthesia (oxybuprocaine

hydrochloride 0.4%) at 5 minutes and 2 further drops 1 minute before surgery. In patients

requiring astigmatism correction over 1.50 D, corneal reference marks were made before

surgery at the 3-o’clock and 9-o’clock meridians with the patient standing up.

Statistical analysis

Even though both eyes from each patient were operated and measured before SMILE surgery

and at 6 months, a random eye per subject was included in the statistical analysis because of the

high concordance shown in the preoperative SE between eyes (ICC 0.92, p<0.001; 95%

confidence interval 0.87, 0.95) (Karakosta et al. 2012). If one of both eyes of the patient

presented a complication, the patient was excluded from the randomization and the contralateral

eye was included in the statistical analysis, but the complication was included in the safety

section. The randomization was performed with a MATLAB function (The Mathworks Inc.,

Natick, MA, USA) that filtered randomly the data of one eye for each patient. Eyes were

divided into 3 groups depending on the preoperative SE. Thirty eyes (42.3%) were included in

the low group, from -1.00 D to -3.00 D, 31 (43.7%) in the medium, from -3.25 D to -5.00 D,

and 10 (14.1%) in the high, from -5.25 D to -7.00 D. Decimal VAs were converted to logMAR

for assessing the differences between groups (Yu & Afifi 2009), and were later reconverted to

Snellen for reporting results to follow the standard graphs reporting results (Waring et al. 2011).

Visual acuity of 0.9 decimal was considered as 20/25 for plotting the standard graphs (Waring

et al. 2011), but its value was maintained after conversion to logMAR for statistical purposes.

Nonparametric statistical methods were used due to nonnormal distribution of study variables.

The Wilcoxon signed rank test was used to evaluate the differences between preoperative


45

CDVA and postoperative UDVA. A Kruskal-Wallis H test was run to determine if there were

differences in some groups. Pairwise comparisons were done with a Bonferroni correction for

multiple comparisons. Statistical analyses were performed using SPSS (v20; SPSS Inc.,

Chicago, IL, USA), and significance was set at p<0.05. Standard graphics (Eydelman et al.

2006; Waring et al. 2011) were generated using Microsoft Excel 2010 (Microsoft Corporation,

Redmond, WA, USA) and our own MATLAB library was used for vector analyses.

2.1.4. Results

Seventy-one consecutive myopic eyes of 71 patients, mean age 31.86 ± 5.57 SD (range 21-43

years), were included in the sample. Table I provides the preoperative data of these patients

stratified depending on their refractive error level.

Table 1. Preoperative descriptive analysis of the sample by refractive error.

Parameters

Low (n=31) Medium (n=30) High (n=10)

Mean (SD) Median [Range]



Age 32.20 (5.25) 33.50 [25,43] 31.58 (5.64) 31.00 [22,43] 28.70 (5.38) 28.00 [21,36]

UDVA (logMAR)

0.72 (0.26) 0.7 [ 0.3, 1.3] 1.04 (0.18) 1.0 [ 0.7, 1.3] 1.33 (0.26) 1.3 [ 1.0, 2.0]

CDVA (logMAR)

-0.02 (0.04) 0.0 [-0.1, 0.0] 0.02 (0.03) 0.0 [-0.1, 0.0] 0.00 (0.05) 0.0 [-0.1, 0.0]

Manifest SE (D)

-2.07 (0.58) -2.00 [-3.00,-1.00] -3.86 (0.57) -3.75 [-5.00,-3.25] -5.86 (0.58) -5.81 [-6.88,-5.25]

Safety

In total, 95.8% (68 eyes) had unchanged CDVA, 1.4% (1 eye) lost 1 line, and 2.8% (2 eyes)

gained 1 line. The eye that lost 1 line belonged to the low myopic group, whereas eyes that

gained 1 line belonged to the high myopic group. There were no eyes with CDVA worse than

20/40 for the subgroup of eyes with CDVA of 20/20 or better preoperatively. No eye showed an

increase in manifest refractive astigmatism of 2.00 D over preoperative refraction. One eye had

a suction

loss during the surgical procedure, which was not included in the sample for reporting refractive

results. The suction happened during the lower lenticular interface creation and SMILE was

reapplied on the same day. This eye posteriorly developed epithelial ingrowth, corneal folds,

and irregular astigmatism that were solved with photorefractive keratectomy (PRK), achieving

UDVA of 20/25 and CDVA of 20/20.

______________________________________________________________________________________________

46

Fig. 2. Predictability for all eyes, and for eyes from the low, middle, and high spherical equivalent (SE) groups.

Predictability

The slope (0.9475) of the linear regression model (p<0.001) relating achieved and intended SE

confirmed the slight trend to undercorrection, especially for high myopic eyes (Fig. 1). More

detailed information about predictability is shown in Figure 2, where it can be seen that more

than

half of the subjects (52%) were close to emmetropia, and 26% achieved an SE between -0.50 D

and -0.14 D. The undercorrection was more evident in the high myopic group, with 70% of eyes

achieving SE between -1.00 D and -0.14 D.

Indeed, the median SE was 0 D for low and medium myopic groups and -0.50 D for the high

myopic group; the difference was statistically significant, p = 0.031. We also found statistically

significant differences in postoperative SE between the medium (40.95) and high myopic

groups (22.20) (p = 0.026), but not between the low (35.48) and high or medium myopic groups

(Tab. II). The percentages of eyes with SE within ±0.50 D were 87%, 92%, 80%, and 86% for

low, medium, and high myopic groups and the total sample, respectively, and 97%, 98%, 100%,

and 97% of eyes were within ±1.00 D.


47

Fig. 1. Linear regression model for predictability. The scatterplot shows undercorrection with the increase in attempted spherical equivalent.

Table 2. Postoperative analysis of eyes from the Low, Medium and High myopic refractive groups and evaluation of median differences between preoperative CDVA and postoperative UDVA.

Low (n=31) Medium (n=30) High (n=10)

Parameter Median [Range] Median [Range] Median [Range] p-value*

UDVA (logMAR) 0.0 [-0.1,0.3] 0.0 [-0.1,0.2] 0.0 [ 0.0,0.1] .282 CDVA (logMAR) 0.0 [-0.1,0.0] 0.0 [-0.1,0.0] 0.0 [-0.1,0.0] .232 Manifest SE (D) 0.00 [-1.63, 0.25] 0.00 [-1.13,0.50] -0.50 [-1.00,0.00] .031

Pre CDVA –Pos UDVA 0 [0, 0.5] 0 [-0.1, 0.3] 0 [-0.1, 0.4] .099

* Kruskal-Wallis H test

Efficacy

A postoperative UDVA of 20/20 or better was achieved by 67% and 74% of eyes in the low and

medium myopic groups, respectively. However, the percentage for this level of UDVA for the

high myopic group was 50%. We found that 100% of eyes in the high myopic group had UDVA

of 20/25, whereas for this level the percentage was 97% for the low and medium myopic groups

(Fig. 3). Median was close to 20/20 for all groups (Tab. II), and no statistically significant

differences were found among them, p = 0.282. Median differences in CDVA between groups

were not statistically significant, p = 0.232 (Tab. II).

______________________________________________________________________________________________

48

Fig. 3. Cumulative percentage of eyes with different post-small incision lenticule extraction levels of uncorrected distance visual acuity (UDVA), by refractive error group.

The comparison of results from Figure 3 (UDVA) with the percentage of subjects with

preoperative CDVA of 20/20 or better (Fig. 4) revealed that 17% of eyes achieved 1 line of

UDVA less than the preoperative VA obtained with spectacles (CDVA). This was more

remarkable in the low myopic group, where this percentage decreased from 97% to 67% (30%

of decrease), whereas in the medium and high myopic groups this percentage only decreased

from 81% to 74% (7%) and from 60% to 50% (10%), respectively. Of the 71 operated eyes, a

decline from preoperative CDVA to postoperative UDVA was found in 21 eyes, 4 eyes

presented improvement, and 46 eyes maintained the same VA. Even though the median was

20/20 for preoperative CDVA and postoperative UDVA, the Wilcoxon signed rank test showed

statistically significant differences among both VA values, p<0.0005. The mean and standard

deviation was -0.01 ± 0.04 for preoperative CDVA and 0.03 ± 0.06 for postoperative UDVA,

the difference being less than 1 line of VA. Differences in postoperative UDVA minus

preoperative CDVA among low, medium, and high myopic groups were not statistically

significant, p = 0.099 (Tab. II).


49

Fig. 4. Cumulative percentage of eyes with different pre-small-incision lenticule extraction levels of corrected distance visual acuity (CDVA), by refractive error group.

Vector analyses

Only eyes with preoperative astigmatism greater than zero were included in the vector analysis

(n = 46). Figure 5A shows that the centroid coordinates (x, y) were near to 0 (-0.13, -0.01) for

the intended refractive cylinder, and the standard deviation (radii of the ellipse) was higher in

the horizontal axis SDx = 0.9 D than in the vertical SDy = 0.7 D. Therefore, the sample was

evenly distributed between with the rule (WTR) and against the rule (ATR) astigmatisms, with

less incidence of oblique astigmatism. Figure 5B shows the error vector or manifest cylinder

after 6 months, showing how centroid coordinates were slightly close to 0 (-0.1, -0.01), although

standard

deviation decreased considerably to SDx = 0.24 D and SDy = 0.27 D. Furthermore, scatter at

the left side of Figure 5B appeared to be greater than on the right side, suggesting an

undercorrection of WTR or overcorrection of ATR. The normalized error vector in Figure 5C

and treatment error vector in Figure 5D, with overcorrections on the left side and

undercorrections on the right side, showed that an undercorrection was generally presented for

the cylinder.

______________________________________________________________________________________________

50

Fig. 5. Each radial step represents an increase of 0.50 D from the center. (A) Intended refractive correction, which is the preoperative positive cylinder. (B) Error vector represents the residual refractive cylinder or postsurgery cylinder. (C) Normalized error vector and (D) treatment error vector represent the overcorrection at the right side of the vertical axis. Some computed points appear overlapped for low dioptric values of B, C, and D.

2.1.5. Discussion

We present the early outcomes of an inexperienced SMILE surgeon with previous experience in

other laser refractive surgery techniques. One random eye from the first consecutive 71 subjects

was included and analyzed depending on the refractive error level at 6-month follow-up.

Suction loss is one of the complications that have been reported with SMILE (Wong et al.

2014), but we only found 1 case, corresponding to the consecutive second surgery. Other

complications of SMILE have been reported, such as incomplete femtosecond laser cutting

(Reinstein, Carp, et al. 2014), opaque bubble layer (Wang et al. 2014), infiltrates/keratitis or

interface inflammation (Zhao et al. 2015), abrasion at the incision, tears at the incision, cap

perforation, haze, dry surface,


51

epithelial islands at the incision, and fiber at the interface (Piñero-Llorens et al. 2016).

However, we only found 1 eye with epithelial ingrowth and corneal folds. In terms of

preoperative and postoperative CDVA, we found that SMILE was a safe procedure with only

1.4% of eyes losing 1 line of VA corresponding to the low myopic group. These results are

slightly better than those reported by other authors (Reinstein, Carp, et al. 2014; Shah et al.

2011). As the percentage of eyes gaining 1 line of VA increased with the refractive error level,

this improvement of postoperative CDVA may be due to the change in retinal image

magnification after surgery compared to the use of spectacles in high myopic eyes (Vestergaard

et al. 2014; Xu & Yang 2015).

The first predictability results for SMILE at 6 months were reported by Shah et al (Shah et al.

2011). They found a mean SE of +0.03 ± 0.30 D, with 91% of subjects within ±0.50 D and

100% within ±1.00 D. They reported that refractive stability was achieved within 1 month,

suggesting that predictability of SMILE would be similar in studies with longer periods of

follow-up. Subsequent studies at 3, 6, and 12 months have found that the percentage of subjects

within ±0.50 D ranged from 77% to 100%, and within ±1.00 D from 94% to 100% depending

on the study (Shah & Shah 2011; Hjortdal et al. 2012). Our results are consistent with those

reported in previous studies, with 86% of eyes with SE within ±0.50 D and 96% within ±1.00

D. Furthermore, we found a poorer predictability for refractive errors between -5.25 D and -7 D,

with a median of -0.5 D for the postoperative SE in this group, while median SE was plano for

low and medium refractive error groups. The undercorrection was associated with an increased

refractive error, in which is consistent with the slope of the attempted versus achieved SE linear

regression equation. This finding has been reported in other studies (Kamiya et al. 2012; Kim et

al. 2014).

Some concerns over the cutting accuracy of the VisuMax or difficulties handling thinner

lenticules have been pointed out (Reinstein, Carp, et al. 2014). We usually increase the diameter

of the optical zone for myopias under -1 D in order to increase the lenticule thickness to around

50 μm, but this was not done in this study because all patients were over -1 D of SE. Despite the

thinner lenticule in the low myopic group, no problems occurred in handling or extracting it.

Our efficacy results for the low myopic group contrast with those previously reported by

Reinstein et al (Reinstein, Carp, et al. 2014), who found, with their own nomogram, that 97% of

eyes achieved UDVA of 20/20 at 3 months after SMILE. In our sample, only 67% achieved a

UDVA of 20/20.

It is important to note that these differences might be due to the fact that the Reinstein et al

cohort had a cumulative percentage of preoperative CDVA (75%) of 20/16. This is considerably

higher than ours (23%); therefore it is also understandable that the percentage of subjects with

UDVA of 20/20 would be higher in their study because the preoperative CDVA was better.

______________________________________________________________________________________________

52

The median change in VA from preoperative CDVA to postoperative UDVA was significant;

nevertheless, it was less than 1 line of VA. Furthermore, no statistically significant differences

in median VA change among the 3 groups were found, even though the percentage of subjects

at 20/20 level decreased in a higher percentage (30%) in the low myopic group than in the

medium (7%) and high myopic (10%) groups. This shows that SMILE might be as effective for

low myopias as it is for medium and high myopias. It is important to note that 2 patients of the

high myopic group and one of the low myopic group were treated successfully with PRK after

this follow-up because they returned with complaints about their UDVA.

However, some other patients of the high myopic group who presented UDVA less than 20/20

were not retreated with PRK if they were satisfied with their binocular vision. Ivarsen and

Hjortdal (Ivarsen & Hjortdal 2014) reported a significant undercorrection of astigmatism as the

intended refractive correction of the cylinder increased, which was similar to or better than

FSLASIK. This undercorrection has been also reported by Kunert et al (Kunert et al. 2013),

who found the centroid moved to the right of the vertical in the normalized error vector. In our

study, we found that the correction of the cylinder is predictable with SMILE because the

standard deviation was reduced from 0.9 D and 0.7 D to 0.24 D and 0.27 D. However, as the

foregoing authors described in their studies, an undercorrection was shown in the normalized

error vector since the data are predominantly moved to the right of the vertical (Eydelman et al.

2006).

Our research may have some limitations. We have only included myopic refractive SE

refractions from -1 D to -7 D; however, SMILE has been awarded European conformity up to -

10 D at the time of our study. Therefore, for comparison purposes in future studies that include

myopias up to -10 D, the creation of a new level of very high myopia from -7.25 D to -10 D

would be recommended. Furthermore, a poorly balanced sample was used, with only 10 eyes in

the high myopia group, whereas the low and medium myopic groups had 30 and 31 eyes,

respectively. A brief analysis of astigmatism has been included in terms of magnitude;

nevertheless, it is important to note that studies centered on astigmatism results should be

performed in terms of magnitude and angle of error (Zhang et al. 2016). About cap thickness, it

is important to note that

it was set to 140 μm, which can vary between authors, but Güell et al (Güell et al. 2015)

reported no differences in refractive result for lenticule thicknesses of 130, 140, 150, and 160

μm.

In summary, short-term outcomes of the SMILE technique for a novel surgeon were as safe,

effective, and predictable as those previously reported in the literature for more experienced


53

surgeons. Furthermore, no differences in the effectiveness of the procedure were found among

low, medium, or high myopias. Future studies should include groups with myopias between -

7.25 D and -10.00 D and the development of nomograms that improve the results obtained in

this and

previous studies.

2.1.6. Disclosures

Financial support: Supported by the Sociedad Española de Cirugía Implanto Refractiva

(SECOIR) PUBLIbeCA 2014 grant.

Conflict of interest: Dr. Fernández is a consultant for Carl Zeiss Meditec (Jena, Germany). The

remaining authors have no financial or proprietary interest in the materials presented herein.

2.1.7. References

Ang, M., Tan, D. & Mehta, J.S., 2012. Small incision lenticule extraction (SMILE) versus laser in-situ keratomileusis (LASIK): study protocol for a randomized, non-inferiority trial. Trials, 13, p.75.

Bischoff, M. & Strobrawa, G., 2016. Femtosecond laser keratomes for Small Incision Lenticule Extraction (SMILE). In W. Sekundo, ed. Small Incision Lenticule Extraction (SMILE). Principles, Techniques, Complication Management, and Future Concepts. Springer, p. 8.

Eydelman, M.B. et al., 2006. Standardized analyses of correction of astigmatism by laser systems that reshape the cornea. Journal of refractive surgery (Thorofare, N.J. : 1995), 22(1), pp.81–95.

Ganesh, S. & Gupta, R., 2014. Comparison of visual and refractive outcomes following femtosecond laser-assisted lasik with smile in patients with myopia or myopic astigmatism. J Refract Surg., 30(9), pp.590–6.

Güell, J.L. et al., 2015. SMILE Procedures With Four Different Cap Thicknesses for the Correction of Myopia and Myopic Astigmatism. J Refract Surg., 31(9), pp.580–585.

Gyldenkerne, A., Ivarsen, A. & Hjortdal, J.Ø., 2015. Comparison of corneal shape changes and aberrations induced by FS-LASIK and SMILE for Myopia. J Refract Surg., 31(4), pp.160–165.

Hjortdal, J.Ø. et al., 2012. Predictors for the outcome of small-incision lenticule extraction for Myopia. J Refract Surg., 28(12), pp.865–71.

Ivarsen, A., Asp, S. & Hjortdal, J., 2014. Safety and complications of more than 1500 small-incision lenticule extraction procedures. Ophthalmology, 121(4), pp.822–828.

Ivarsen, A. & Hjortdal, J., 2014. Correction of myopic astigmatism with small incision lenticule extraction. J Refract Surg., 30(4), pp.240–7.

Kamiya, K. et al., 2012. Early clinical outcomes, including efficacy and endothelial cell loss, of refractive lenticule extraction using a 500 kHz femtosecond laser to correct myopia. J Cataract Refract Surg, 38(11), pp.1996–2002.

______________________________________________________________________________________________

54

Kamiya, K. et al., 2015. Effect of femtosecond laser setting on visual performance after small-incision lenticule extraction for myopia. Br J Ophthalmol, bjophthalm, p.Published Online First: 8 April 2015.

Karakosta, A. et al., 2012. Choice of analytic approach for eye-specific outcomes: one eye or two? American journal of ophthalmology, 153(3), p.571–579.e1.

Kim, J. et al., 2014. Efficacy, predictability, and safety of small incision lenticule extraction: 6-months prospective cohort study. BMC Ophthalmology, 14(1), p.117.

Kunert, K.S. et al., 2013. Vector analysis of myopic astigmatism corrected by femtosecond refractive lenticule extraction. J Cataract Refract Surg, 39(5), pp.759–769.

Li, M. et al., 2014. Mild decentration measured by a Scheimpflug camera and its impact on visual quality following SMILE in the early learning curve. Inves Opthal Vis Sci, 55(6), pp.3886–3892.

Lin, F., Xu, Y. & Yang, Y., 2014. Comparison of the visual results after SMILE and femtosecond laser-assisted LASIK for myopia. J Refract Surg., 30(4), pp.248–54.

Piñero-Llorens, D.P., Murueta-Goyena Larrañaga, A. & Hannekend, L., 2016. Visual outcomes and complications of small-incision lenticule extraction: a review. Expert Review of Ophthalmology, 11(1).

Reinstein, D., Carp, G.I., et al., 2014. Outcomes of small incision lenticule extraction (SMILE) in low myopia. J Refract Surg., 30(12), pp.812–818.

Reinstein, D., Archer, T.J. & Gobbe, M., 2014. Small Incision Lenticule Extraction (SMILE) history, fundamentals of a new refractive surgery technique and clinical outcomes. Eye and Vision, 1(1), p.3.

Shah, R. & Shah, S., 2011. Effect of scanning patterns on the results of femtosecond laser lenticule extraction refractive surgery. J Cataract Refract Surg, 37(9), pp.1636–1647.

Shah, R., Shah, S. & Sengupta, S., 2011. Results of small incision lenticule extraction: All-in-one femtosecond laser refractive surgery. J Cataract Refract Surg, 37(1), pp.127–137.

Shetty, R. et al., 2015. Association between corneal deformation and ease of lenticule separation from residual stroma in Small Incision Lenticule Extraction. Cornea, 34(9), pp.1067–71.

Vestergaard, A. et al., 2012. Small-incision lenticule extraction for moderate to high myopia: Predictability, safety, and patient satisfaction. J Cataract Refract Surg, 38(11), pp.2003–10.

Vestergaard, A.H. et al., 2014. Efficacy, safety, predictability, contrast sensitivity, and aberrations after femtosecond laser lenticule extraction. J Cataract Refract Surg, 40(3), pp.403–11.

Wang, Y. et al., 2014. Two millimeter micro incision lenticule extraction surgery with minimal invasion: a preliminary clinical report. Zhonghua Yan Ke Za Zhi., 50(9), pp.671–80.

Waring, G.O. et al., 2011. Standardized graphs and terms for refractive surgery results. J Refract Surg., 27(1), pp.7–9.

Wong, C.W. et al., 2014. Incidence and management of suction loss in refractive lenticule extraction. J Cataract Refract Surg, 40(12), pp.2002–2010.

Xu, Y. & Yang, Y., 2015. Small-Incision Lenticule Extraction for myopia: Results of a 12-month prospective study. Optom Vis Sci, 92(1), pp.123–131.

Yu, F. & Afifi, A., 2009. Descriptive statistics in ophthalmic research. American journal of ophthalmology, 147(3), pp.389–91.

Zhang, J., Wang, Y. & Chen, X., 2016. Comparison of moderate- to High-Astigmatism


55

corrections using WaveFront – Guided Laser In Situ Keratomileusis and Small-Incision Lenticule Extraction. Cornea, 35(4), pp.523–530.

Zhao, J. et al., 2015. Diffuse lamellar keratitis after small-incision lenticule extraction. J Cataract Refract Surg, 41(2), pp.400–407.

______________________________________________________________________________________________

56


57

______________________________________________________________________________________________

58

2.2. Corneal thickness after SMILE affects Scheimpflug-based Dynamic

Tonometry

Joaquín Fernández, MD;1 Manuel Rodríguez-Vallejo, MS;*1 Javier Martínez, OD;1 Ana Tauste, MS; 1 David P Piñero, PhD;2,3


2.2.1. Abstract

Purpose: To evaluate the corneal biomechanical changes due to small incision lenticule

extraction (SMILE) measured by Scheimpflug-based dynamic tonometry and to assess the

impact of the corneal thickness.

Methods: Sixty-eight patients measured with the Corvis ST (Oculus Optikgeräte GmbH,

Wetzlar, Germany) preoperatively and 1 month after SMILE were included in this retrospective

observational study. Patients were divided into three groups depending on the preoperative

spherical equivalent: low from -1.00 to -3.00 diopters (D), medium from -3.25 to -5.00 D, and

high from -5.25 to -7.25 D. Changes in Corvis ST parameters due to the surgery were analyzed

and new indexes for correcting the impact of corneal thickness were proposed.

Results: First and second applanation times changed after SMILE (P < .0001) but no

differences were found in the comparison between these relative changes (P = .31). First

applanation time was correlated with central corneal thickness (r = 0.368, P = .002) but not

second applanation time (r = -0.149, P = .23). The change in first applanation time due to

SMILE was different among

myopic groups (P = .007) but equal when a new index that considers the removed central

corneal thickness was used for comparison (P = .31). Deformation amplitude was also increased

after SMILE (P < .0001), but after subtracting the removed corneal thickness from the

postoperative deformation amplitude the result was equal to the preoperative deformation

amplitude (P = .26).

Conclusions: SMILE produces significant changes in the Corvis ST parameters of time and

deformation amplitude, but these changes are mainly explained by the confounding variable of

corneal thickness.

Key Words: corneal biomechanics, SMILE, Corvis ST, corneal thickness, refractive surgery


59

2.2.2. Introduction

Corneal biomechanics is of great importance in laser refractive surgery because it can affect the

prediction of clinical outcomes (Roy & Dupps 2009) and might be related to the development of

corneal ectasia after surgery (Randleman et al. 2008). Although it has still not been widely

incorporated in clinical practice for screening purposes, the fact that postoperative ectasia can

occur without apparent preoperative risk factors with current technology (Klein et al. 2006) has

led to a growing interest in new clinical methods for assessing corneal stiffness. Furthermore,

Reinstein et al. (Reinstein et al. 2013) hypothesized that the new technique small incision

lenticule extraction (SMILE) may better preserve stromal tensile strength compared to previous

laser techniques such as photorefractive keratectomy and LASIK because of the absence of a

flap and the fact that the stiffer anterior part of the cornea remains intact. Therefore, clinical

studies have

focused on comparing the cornea response after each of these procedures (Y Shen et al. 2014).

Agca et al. (Agca et al. 2014) used the Ocular Response Analyzer (Reichert, Inc., Buffalo, NY)

to measure the corneal hysteresis and corneal response factor parameters and did not obtain

significant differences between SMILE and femtosecond laser-assisted LASIK. Conversely,

Dou et al. (Dou et al. 2015) found that SMILE seemed to have less effect on corneal

biomechanics than LASEK in terms of per unit tissue removed and El-Massry et al. (El-Massry

et al. 2015) reported that postoperative corneal hysteresis and corneal response factor

parameters were dependent on the lenticule depth in patients who had SMILE.

A new instrument named Corvis ST (Oculus Optikgeräte GmbH, Wetzlar, Germany) provides

more parameters that might be related to corneal biomechanics, but comparison of techniques

with this new instrument remains controversial. Hassan et al. (Hassan et al. 2014) reported no

differences after photorefractive keratectomy and LASIK in comparison to preoperative data in

almost all of the Corvis ST parameters analyzed, but this is not in agreement with other

preoperative–postoperative studies using LASIK (Frings et al. 2015). Chen et al. (Chen et al.

2014) reported differences between photorefractive keratectomy and virgin eyes, but Pedersen

et al. (Pedersen et al. 2014) found no differences between control and LASIK, femtosecond

lenticule extraction, or SMILE groups for all parameters except AT1 deflection length. Shen et

al. (Shen et al. 2014) measured with the Corvis ST just after lenticule creation and later after

extraction, and discovered that Corvis ST parameters changed significantly only during

extraction.

The aims of the current study were to evaluate the biomechanical changes of the cornea after

SMILE and to analyze the impact that the removed corneal thickness may have on the most

repeatable Corvis ST parameters. New indexes that consider the removed corneal thickness

______________________________________________________________________________________________

60

have been proposed for future comparisons between refractive surgery techniques.

2.2.3. Methods

Patients

Patients operated on with the SMILE technique between January 2014 and January 2016 at

Qvision (Department of Ophthalmology, Virgen del Mar Hospital) were identified in this

retrospective observational study. The procedure was explained to all of the patients, who

signed the preoperative informed consent. Institutional review board approval was obtained and

the study complied with the tenets of the Declaration of Helsinki.

Inclusion criteria were myopic patients with spherical equivalent from -1.00 to -7.25 diopters

(D) and refractive astigmatism less than 3.00 D who were measured preoperatively and 1 month

after SMILE with the Corvis ST. Preoperative exclusion criteria were pregnancy at the time of

surgery or follow-up, a preoperative central corneal thickness of less than 480 μm, an expected

postoperative residual stromal bed of less than 250 μm, a topographic map compatible with

subclinical keratoconus or other ectatic corneal disorder, and any other ocular disease for which

laser refractive surgery procedures are not indicated (Shetty et al. 2015). Postoperative

exclusion criteria included intraocular pressure (IOP) greater than 19 mm Hg before or after

SMILE, central corneal thickness (CCT) greater than that before the surgery, postoperative

complications, and patients who had undergone a laser re-treatment. Measurements by the

Corvis ST with alert messages of “pressure profile” and “lost images,” which indicate poor

quality, were also excluded. However, other alerts such as “model deviation,” “lost points,” and

“alignment” were not discarded after consulting with the Corvis ST manufacturer about the

possible influence of these alerts on the results.

Surgical procedure

The same surgeon (JF) performed all SMILE treatments with the VisuMax femtosecond laser

system (Carl Zeiss Meditec AG, Jena, Germany). Two drops of topical anesthesia

(oxybuprocaine hydrochloride 0.4%) were instilled at 5 minutes and two additional drops at 1

minute before surgery. In patients requiring astigmatism correction greater than 1.50 D, corneal

reference marks were made before surgery at the 3- and 9-o’clock meridians with the patient

standing up.

The optical principles and general description of the SMILE procedure have been widely

described (Reinstein et al. 2014), but some particular laser settings or surgical maneuvers were

used. Before suction, centration was accepted when the ring of the applanation zone was


61

concentric with the margin of the cone and near to the pupil center (Li et al. 2014). Suction was

then applied, and a slight rotation of the applanation cone was made to compensate for

cyclotorsion in cases of high astigmatism with markings, taking as reference the horizontal lines

seen through the microscope. The photodisruptive procedure occurs in the following sequence:

(1) posterior lenticule creation from periphery to center (optical zone of 6.5 mm); (2) transition

zone after the peripheral optical zone greater than 1 mm of the optical zone; (3) anterior

lenticule from center to periphery with a cap diameter of 7.6 mm and cap thickness of 140 μm;

and (4) peripheral incision of 2 mm with 30° of angle for posterior lenticule extraction at 70°

(Bischoff & Strobrawa 2016). In SMILE, the cap depth is theoretically constant across the

anterior surface of the lenticule and the depth of the posterior lenticule surface is established by

the software depending on the attempted refractive correction.

Laser configuration parameters were: repetition rate of 500 kHz, spot distance of 4.5 μm for the

lenticule and 2 μm for its border, and pulse energy of level 30 in the software, which

corresponds to approximately 150 nJ (Vestergaard et al. 2014b). The target refractive error

correction was directly inserted in the software without applying any nomogram. After laser

treatment, the patient was moved to the surgical microscope for the second part of the

procedure, which involves: (1) delineating front and back lenticule surfaces; (2) surface

separation using the standard lamellar corneal surgical technique of moving the instrument back

and forth using a blunt circular tip (Femto Double-Ended instrument [G-33954]; Carl Zeiss

Meditec AG) starting with the complete dissection of the front cap and following with the

dissection of the posterior lenticule surface; (3) lenticule extraction with forceps (Lenticule

Forceps [G-33961]; Carl Zeiss Meditec AG); and (4) pressing the corneal surface from center to

periphery using a dry micro-spear and drying the incision with the same. Finally, two drops of

combined tobramycin (0.3%) and dexamethasone (0.1%) were instilled in all cases at the end of

the procedure. Postoperative treatment included ofloxacin (0.3%) for 2 days, dexamethasone

drops five, three, two, and one time per day (reducing the dosage every 7 days), and sodium

hyaluronate (0.15%) for 1 month.

Outcome measures

Corneal biomechanical parameters widely described in the literature were obtained by the

Corvis ST (Bak-Nielsen et al. 2015; Chen et al. 2014). This system is based on the concept of

dynamic corneal topography, which combines the bidirectional applanation technology, the

high-speed photography, and the corneal topography (Piñero & Alcón 2015). An air puff is

directed over the

cornea and its response is captured by a Scheimpflug camera with a frame rate of 4,330 frames

per second along an 8-mm horizontal corneal coverage (Hon & Lam 2013). Multiple data are

______________________________________________________________________________________________

62

returned during the three stages into which the process can be divided: inward applanation

(AT1),

highest concavity, and outward applanation (AT2). The CCT output from the Corvis ST was

also used for the analysis because previous studies have reported nonsignificant differences or

no trends to overestimation or underestimation compared to the values obtained with the

Pentacam system (Oculus Optikgeräte GmbH, Wetzlar, Germany)(Bak-Nielsen et al. 2015) or

ultrasound pachymeters (Smedowski et al. 2014). A good intraobserver repeatability has been

reported for only some variables, including IOP, CCT, AT1, and AT2, and maximum

deformation amplitude (DA) at the corneal apex (Nemeth et al. 2013; Chen et al. 2014). On the

other hand, poorer intraclass correlation coefficients have been reported for peak distance, time

from starting until highest concavity, and first/ second applanation lengths or velocities (Nemeth

et al. 2013; Chen et al. 2014; Hon & Lam 2013). Therefore, only the most repeatable variables

(AT1, AT2, and DA) were included in this study, for which only one measurement per eye was

taken. Other variables such as air puff pressure at both applanation times were also considered

due to their potentially high clinical relevance for the analysis even though their repeatability

has not been previously reported. Because measurements were repeated for each eye 1 month

after SMILE, these postoperative measurements are identified by an apostrophe in the text (ie,

AT1’, AT2’, and DA’). Relative time changes due to the procedure were also computed for first

(diffAT1 = AT1 – AT1’) and second (diffAT2 = AT2 – AT2’) applanation times. A new

variable named corrected DA (DAc) was computed for DA by subtracting the CCT tissue

removed by the procedure from DA’ (DAc = DA’ – (CCT – CCT’)). DAc was computed

because it is known that the Corvis ST parameters depend on corneal thickness (Ariza-Gracia et

al. 2015). For the same deformation at the posterior corneal surface, the anterior surface is going

to experience a greater depth due to the removed CCT, as described in Figure 1.

Figure 1. Schematic showing the definition of corrected deformation amplitude (DA) depending on the removed corneal thickness. CCT = central corneal thickness; CCT’ = central corneal thickness at 1 month postoperatively; DA’ = deformation amplitude at 1 month postoperatively; IOP = intraocular pressure; SMILE = small incision lenticule extraction.


63

Three indexes (t1, t2, d) were computed for comparison purposes between this study and future

Corvis ST studies with other refractive surgery techniques. These indexes were calculated by

means of the ratio of the difference for each variable before and after the procedure and the

change in CCT due to the procedure as follows:

Statistical analysis

Only the right eye of the patients was included in the statistical analysis, except if this eye

showed one of the alert messages included in the exclusion criteria. In such cases, the left eye

was included instead of the right eye. Normal data distributions were confirmed with the

Shapiro–Wilks W test for the comparison between groups and with the Kolmogorov–Smirnov

test for differences between preoperative and postoperative variables. Paired t tests were

conducted for testing differences before and after the procedure and Pearson r for evaluating

correlations. The sample was divided into three groups according to the myopia level: low from

-1.00 to -3.00 D, medium from -3.25 to -5.00 D, and high from -5.25 to -7.25 D. A one-way

analysis of variance with Bonferroni-adjusted post hoc comparisons was used to evaluate the

differences in variables among the three myopic groups. Data were analyzed using SPSS for

Windows statistical software (version 20.0; SPSS, Inc., Chicago, IL) and all of the statistical

tests were selected after checking that the required assumptions were completely accomplished.

Sample size calculation was performed to confirm whether the sample of eyes included in the

current study was of adequate size using the software PS version 3.1.2 (free availability online:

http://biostat.mc.vanderbilt.edu/wiki/Main/PowerSampleSize). This software uses the Dupont

and Plummer approach for sample size calculation (Dupont & Plummer 1990). We estimated

the number of pairs of patients needed to detect a true difference in population means d with

type I error probability a given a standard deviation s. Specifically, for a statistical power of

80%, considering d and s changes after SMILE reported for DA, AT1, and AT2 in previous

studies (Hassan et al. 2014; Frings et al. 2015; Chen et al. 2014; Pedersen et al. 2014; Shen et al.

2014), and an a error of 0.05, the sample size required was 61 eyes.

�� =��1 − ��1′

�� − ��′; �� =

��2 − ��2′

�� − ��′; � =

�� − ��′

�� − ��′

______________________________________________________________________________________________

64

2.2.4. Results

Eight eyes were excluded from the study for the following reasons: 3 eyes had an IOP greater

than 19 mm Hg before or after SMILE, 4 eyes had higher CCT after SMILE than before

surgery, and the Corvis ST pressure at AT2 was 164 mm Hg in 1 eye, which was considered as

an outlier. A total of 68 eyes of 68 patients distributed in 25 low, 32 medium, and 11 high

myopic eyes were included in the analysis (Table 1).

Table 1. Baseline demographic characteristics of all patients group

Mean ± SD (Range)

Low (n=25) Medium (n=32) High (n=11) Total (n=68)

Sex (female/male) 13/12 18/14 5/6 36/32

Age (yrs) 29.96 ± 5.81 (19 to 42)

31.97 ± 6.09 (22 to 46)

33.09 ± 6.81 (23 to 43)

31.41 ± 6.13 (19 to 46)

Intraocular Pressure (mmHg)

15.38 ± 2.28 (11.50 to 19.00)

14.63 ± 1.91 (10.00 to 18.50)

15.27 ± 2.82

(11.00 to 19.00)

15.01 ± 2.21 (10.00 to 19.00)

Spherical Equivalent Refraction (D)

-2.34 ± 0.55 (-1 to -3.13)

-4.12 ± 0.58 (-3.25 to -5.13)

-6.17 ± 0.62 (-5.75 to -7.25)

-3.80 ± 1.45 (-1 to -7.25)

Astigmatism (D) -0.78 ± 0.56 (0 to -1.75)

-0.53 ± 0.55 (0 to -2.00)

-1.09 ± 0.98

(0 to -3.00)

-0.71 ± 0.66 (0 to -3.00)

Simulated Keratometry at 3 mm (mm)

7.73 ± 0.25 (7.21 to 8.21)

7.76 ± 0.37 (7.06 to 8.86)

7.84 ± 0.27

(7.39 to 8.26)

7.76 ± 0.31 (7.06 to 8.86)

Central Corneal

Thickness (m)

553 ± 28 (494 to 604)

550 ± 22 (508 to 603)

545 ± 20

(518 to 577)

550 ± 24 (494 to 604)

Applanation times

Table 2 shows the changes in preoperative and postoperative variables. AT1 was advanced and

AT2 was delayed significantly after SMILE, whereas highest concavity did not vary. The

difference between the relative changes at first and second applanation times due to SMILE was

not significant, neither for times (|diffAT1| – |diffAT2|) nor for Corvis ST pressures (|diffPA1| –

|diffPA2|). We found a significant correlation between AT1 and CCT that increased after

SMILE (Table A, available in the online version of this article). However, no significant

correlations were found between AT2 and CCT before and after the surgery or for simulated

keratometry with AT1 and AT2 (Table A). The proposed biomechanical indexes were 5.26 ±

3.95 (range: -5.49 to 15.58) for t1 and -5.79 ± 6.32 (range: -22.04 to 13.02) for t2. These

indexes among refractive error groups are detailed in Table B.

Table A. Pearson correlations between outcome variables.


65

AT1 and AT2: Times at first and second applanations; CCT: Central Corneal Thickness; DA: Deformation amplitude; DAc: Corrected postoperative deformation amplitude; SimK: Keratometry at 3 mm; Rx: Preoperative spectacle refraction

Table B. Outcome variables depending on the refractive error.

Mean ± SD (Range)

Low (n=25) Medium (n=32) High (n=11) F (One-way ANOVA)

P

diffAT1 0.30 ± 0.27 (-0.33 to 0.72)

0.34 ± 0.24 (-0.16 to 0.83)

0.59 ± 0.22 (0.21 to 0.81)

5.635 P = .007

Low versus Medium P>.05; Low versus High P = .005; Medium vs. High P = .019

diffAT2 -0.35 ± 0.38 (-1.08 to 0.66)

-0.36 ± 0.43 (-1.34 to 0.61)

-0.62 ± 0.28 (-1.08 to -0.21)

2.1 P = .13

t1 6.03 ± 5.36 (-5.49 to 15.58)

4.48 ± 2.95 (-3.34 to 9.04)

5.73 ± 2.28 (2.30 to 9.32)

1.181 P = .31

t2 -7.41 ± 7.79 (-22.04 to 10.75)

-4.45 ± 5.42 (-17.43 to 13.02)

-5.99 ± 2.89 (-12.47 to -2.09)

1.568 P = .22

DA (mm) 1.02 ± 0.09 (0.83 to 1.17)

1.07 ± 0.09 (0.93 to 1.26)

1.05 ± 0.1 (0.90 to 1.20)

2.431 P = .096

DA’ (mm) 1.09 ± 0.09 (0.86 to 1.28)

1.14 ± 0.1 (0.92 to 1.37)

1.18 ± 0.08 (1.06 to 1.29)

3.778 P = .02

Low versus Medium P>.05; Low versus High P = .03; Medium vs. High P>.05

DAc (mm) 1.04 ± 0.1 (0.80 to 1.23)

1.07 ± 0.1 (0.85 to 1.29)

1.07 ± 0.08 (0.96 to 1.27)

0.761 P = .47

d -1.61 ± 1.79 (-5.10 to 1.67)

-0.95 ± 1.14 (-3.53 to 1.62)

-1.18 ± 0.79 (-2.16 to 0.57)

1.605 P = .21

diffAT1 and diffAT2: Relative changes in times on first and second applanations due to SMILE; t1 and t2: Indexes obtained from the ratio between the change in time and change in thickness; DA: Preoperative deformation amplitude; DA’: Postoperative deformation amplitude; DAc: Postoperative deformation amplitude subtracting the removed corneal thickness; d: Index obtained from the ratio between the change in DA and change in thickness.

Table 2. Outcome variables before and after SMILE

Pre-SMILE Post-SMILE

AT1 (ms) vs CCT r=0.368, P=.002 r=0.42, P<.0001 AT2 (ms) vs CCT r=-0.149, P=.23 r=-0.116 P=.35

AT1 (ms) vs SimK (mm) r=-0.125, P=.31 r=-0.35, P=.003 AT2 (ms) vs SimK (mm) r=-0.027, P=.83 r=0.058, P=.64

DA (mm) vs Rx r=-0.159, P=.2 r=-0.349, P=.004 DAc (mm) vs Rx r=-0.141, P=.25

Pre – Post Variable (units) Mean ± SD (Range)

(Abbreviation) Pre-SMILE Post-SMILE Difference t (Paired t-test) P

AT1 – AT1’ (ms) (diffAT1)

7.79 ± 0.24 (7.26 to 8.25)

7.43 ± 0.23 (6.82 to 8.09)

0.37 ± 0.27 (-0.33 to 0.83)

11.4 < 0.0001

AT2 – AT2’ (ms) (diffAT2)

22.17 ± 0.44 (21.29 to 23.16)

22.57 ± 0.44 (21.57 to 23.36)

-0.4 ± 0.4 (-1.34 to 0.66)

-8.12 < 0.0001

HCT –HCT’ (ms) 17.20 ± 0.51 (15.71 to 18.02)

17.20 ± 0.56 (16.17 to 18.48)

0.000 ± 0.57 (-1.16 to 1.39)

.000 .1

PA1 – PA1’ (mmHg) (diffPA1)

52.9 ± 5.44 (40 to 63.40)

44.47 ± 5.40 (29.40 to 59.30)

8.43 ± 5.97 (-7.20 to 20.60)

11.64 < 0.0001

PA2 – PA2’ (mmHg) (diffPA2)

61.53 ± 10.81 (39.70 to 84.30)

52.02 ± 10.27 (31.80 to 80.30)

9.52 ± 10.79 (-18.90 to 31.90)

7.28 < 0.0001

CCT – CCT’ (m) 550 ± 24 (494 to 604)

478 ± 36 (415 to 550)

72 ± 25 (20 to 137)

24.14 < 0.0001

SimK – SimK’ (mm) 7.76 ± 0.31 (7.06 to 8.86)

8.35 ± 0.43 (7.32 to 9.44)

-0.58 ± 0.34 (-1.49 to 0.43)

-14.13 < 0.0001

DA – DA’ (mm) 1.05 ± 0.09 (0.83 to 1.26)

1.13 ± 0.1 (0.86 to 1.37)

-0.08 ± 0.08 (-0.27 to 0.1)

-8.1 < 0.0001

DA - DAc (mm) 1.05 ± 0.09 (0.83 to 1.26)

1.06 ± 0.09 (0.80 to 1.29)

-0.01 ± 0.08 (-0.20 to 0.18)

-1.13 .26

|diffAT1| – |diffAT2| -0.03 ± 0.23 (-0.65 to 0.52) -1.02 .31

|diffPA1| – |diffPA2| -1.09 ± 6.33 (-15.20 to 15.40) -1.42 .16

______________________________________________________________________________________________

66

AT1 and AT2: Times at first and second applanations; HCT: Time at highest concavity; PA1 and PA2: Air puff pressures at first and second applanations; CCT: Central Corneal Thickness; SimK: Simulated keratometry at 3 mm; DA: Deformation amplitude; DAc: Corrected deformation amplitude with the removed CCT; diffAT1 and diffAT2: Relative differences between preoperative and postoperative AT1 and AT2; diffPA1 and diffPA2= Relative difference between preoperative and postoperative PA1 and PA2. An apostrophe over each variable represents its postoperative value.


67

DA

Table 2 shows how DA was increased significantly after SMILE. However, when the removed

CCT was subtracted from the DA’ (DAc), this difference between preoperative and

postoperative DA disappeared. The refractive error treated was correlated with DA’ but not with

DA and DAc (Table A). Comparison between myopic groups is shown in Table B and Figure 2.

The groups showed no differences in DA before SMILE (P = .096); significant differences were

presented in DA’, but only between the low and high myopic groups, as revealed in the post hoc

comparison (P = .03). Furthermore, there were no statistically significant differences in the DA

between myopic groups for the corrected postoperative SMILE value (DAc) (P = .47). The

proposed index d was -1.23 ± 1.39 (range: -5.10 to 1.67) and no significant differences were

found among myopic

groups (Table B).

Figure 2. Maximum deformation amplitude depending on the preoperative refractive error before small incision lenticule extraction (SMILE) (DA), after SMILE (DA’), and after SMILE subtracting the removed corneal thickness (DAc). Significant differences were only found between low and high myopia for DA’ (P = .03).

2.2.5. Discussion

The cornea is a viscoelastic tissue; thus, its behavior is different during the loading and

unloading pressures of an air puff (Roberts 2014). This is shown in our study in which AT2 was

lower than AT1, indicating that the cornea needs more pressure to achieve AT1 than that

required during the recovering stage at AT2; this difference of pressures between applanation

times is called corneal hysteresis (Piñero & Alcón 2015). Corneal hysteresis has been defined as

a descriptor of the corneal viscoelastic properties, arguing that if the cornea were purely elastic,

these two pressure values would be the same (Roberts 2014). On the other hand, AT1 only

______________________________________________________________________________________________

68

describes the elastic properties because a fast applied load, as in the case of the air puff, will

result in an almost pure elastic response during the loading (Ariza-Gracia et al. 2015). Our

results show that SMILE leads

to an anticipation of AT1 and a retardation of AT2, indicating major resistance during the

inward stage and greater force of the cornea during the outward recuperation before surgery. On

the other hand, highest concavity remained constant after SMILE, which suggests that highest

concavity mainly depends on the time at which the top pressure of the pulse is achieved.

The preservation of corneal biomechanics in SMILE measured by means of corneal hysteresis

remains controversial. Some authors have not found significant differences in corneal hysteresis

measured with the Ocular Response Analyzer between SMILE and femtosecond lenticule

extraction (Vestergaard et al. 2014a; Kamiya et al. 2015) or femtosecond laser-assisted LASIK

(Agca et al. 2014). However, other studies have reported differences between SMILE and

LASIK for myopia greater than -6.00 D (Wang et al. 2014) or have suggested that the corneal

viscoelastic properties are better preserved in SMILE in comparison to LASIK (Wu et al. 2014).

In our study, we found that SMILE leads to an anticipation of AT1 and a retardation of AT2,

but the relative changes of time at both stages were not significantly different. This change

might be due to the smaller mass of the postoperative corneas that causes the anticipation of

AT1 and the retardation of AT2 (Simonini et al. 2016).

It has also been pointed out that corneal hysteresis is a meaningless parameter and the drop in

pressure observed between the two applanation times cannot be linked to viscous properties of

the stroma and can be attributed to the inertia (Simonini et al. 2016). According to the latter, our

results would describe a change only in the elastic properties of the cornea, which affect AT1

and AT2 in a similar way. However, we found that change in AT1 (diffAT1) was different

among refractive error groups but this did not happen for the change in AT2 (diffAT2).

Therefore, change in corneal thickness has a great impact on AT1, as also shown in the

correlation analysis between AT1 and CCT before and after surgery, but not in AT2. The new

indexes (t1 and t2) proposed in this study for comparison of refractive surgery procedures

consider the change in applanation times depending on the removed CCT. Thus, the differences

between refractive error groups for AT1 disappeared for the t1 index, demonstrating that this

variable should be corrected according to the CCT removed. Our results about correlations

between times and CCT are not completely in agreement with those obtained in other studies, in

which AT2 and CCT have shown significant

correlations in healthy eyes (Lee et al. 2016; Lanza et al. 2014). However, the significance in

correlations for AT2 and CCT in these studies were P = .043 and .031, considerably higher than

for AT1 and CCT (P < .005), which also demonstrate higher dependency of AT1 with CCT.

Therefore, it is possible that the discrepancy with these studies is due to the sample size or the


69

statistical analysis.

Corneal finite element model has demonstrated that corneas with different stiffness can show

the same DA depending on the IOP and CCT (Ariza-Gracia et al. 2015). In our study, we

assumed that IOP remains constant after the procedure because the same eyes were compared

before and after SMILE. Therefore, changes in DA after SMILE should be mainly due to

changes in CCT or corneal stiffness. We hypothesized that DA is increased after SMILE,

mainly due to a change of thickness, considering that for the same deformation at the posterior

corneal surface, the anterior surface is going to experience a greater depth due to the removed

CCT. Under this hypothesis, we computed a new variable DAc by means of subtracting the DA’

from the removed CCT. Thus,

although DA was significantly increased after SMILE, no significant differences were found

between the DA and DAc, which means either SMILE does not affect the corneal stiffness or

the Corvis ST cannot detect little changes in corneal stiffness by means of the air puff. The

index d did not show statistically significant differences among groups such as DAc.

Our research may have limitations. First, postoperative measures were taken only 1 month after

surgery and even though Mastropasqua et al. (Mastropasqua et al. 2014) reported no significant

differences between 30 and 90 days for variables such as AT1, AT2, or DA, we believe that

higher CCT after SMILE than before surgery might be explained by a slight increase of corneal

thickness by the postoperative dexamethasone treatment. Furthermore, the reason for other

outliers that were eliminated from the sample might be due not only to postoperative treatment

but also to a poor quality of the measurement considering that we only took one measurement

per eye. The indexes that we proposed for comparison with other techniques in future studies

should also be interpreted with caution. Although they consider the relative change in the Corvis

ST parameter according to the change in thickness that supposes an advantage for procedures

comparison, the standard deviation of these indexes is considerably high overall in the low

myopic group. It is possible that studies with longer follow-up and larger samples will reduce

the bias and the precision of these indexes will increase.

Studies that have compared the Corvis ST variables among different procedures have not

generally found differences between them (Pedersen et al. 2014; Shen et al. 2014). In our

opinion, studies comparing Corvis ST variables without information about the preoperative

values should be interpreted with caution because confounding variables such as IOP or CCT

can have a high impact on the results (Ariza-Gracia et al. 2015). Therefore, future studies

comparing techniques should include preoperative and postoperative data.

We have proposed three indexes (t1, t2, and d) for future comparisons between different

______________________________________________________________________________________________

70

refractive surgery techniques. These can only be applied in paired studies that include

preoperative and postoperative data assuming that the IOP of the patients remains constant after

the procedure. We decided to use these indexes, based on relative ratios, because they represent

the amount of change in a Corvis ST variable depending on the removed CCT. Thus, laser

refractive surgery procedures can be compared by the amount of change in a Corvis ST variable

that cannot be explained by a variation in corneal thickness. A technique with a higher index

would represent a cornea with a poorer preservation of the corneal biomechanical properties

because for the same removed thickness there would be a higher change in the Corvis ST

variable. To the best of our knowledge, no other studies have considered and evaluated the

relative change in Corvis ST variables according to the removed CCT. Therefore, our

approximation is the first to solve this problem and we believe that might improve the studies

for comparison between refractive surgery techniques until complex models for correcting

Corvis ST variables depending on CCT are developed.

We have demonstrated that if the Corvis ST parameters of time and DA represent the

biomechanical properties of the cornea, SMILE affects the corneal biomechanics because these

parameters change with surgery. However, we have also demonstrated that these changes are

mainly due to the removed corneal thickness. When this variable was corrected, differences

between myopia groups were not found and DA was equal to the preoperative values. Thus,

CCT correction might help to recognize variations in corneal biomechanics due to flap

generation or the stiffness characteristics of the tissue preserved instead of the volume of tissue

removed. For this purpose, we have proposed new indexes based on the relative change of

Corvis ST variables according to the removed CCT. These may help to improve studies of

comparison between refractive surgery techniques for answering the hypothesis of better

preservation of corneal biomechanics in SMILE that cannot be answered in this study. Similar

studies are required with other laser refractive surgery techniques. Future improvements in

Corvis ST parameters should be directed to correct the corneal thickness confounding variable

and comparison studies between

refractive surgery techniques should be based on differences between preoperative and

postoperative values to minimize the bias produced by other confounding variables such as the

IOP.

2.2.6. Disclosures

Dr. Fernández is a consultant for Carl Zeiss Meditec. The remaining authors have no financial

or proprietary interest in the materials presented herein.


71

2.2.7. References

Agca, A. et al., 2014. Comparison of corneal hysteresis and corneal resistance factor after small incision lenticule extraction and femtosecond laser-assisted LASIK: A prospective fellow eye study. Cont Lens Anterior Eye., 37(2), pp.77–80.

Ariza-Gracia, M.Á. et al., 2015. Coupled biomechanical response of the cornea assessed by non-contact tonometry. A simulation study. Plos One, 10(3), p.e0121486.

Bak-Nielsen, S. et al., 2015. Repeatability, reproducibility, and age dependency of dynamic scheimpflug-based pneumotonometer and its correlation with a dynamic bidirectional pneumotonometry device. Cornea, 34(1), pp.71–77.


Chen, X. et al., 2014. Reliability of corneal dynamic Scheimpflug analyser measurements in virgin and post-PRK eyes. PLoS ONE, 9(10), p.e109577.

Dou, R. et al., 2015. Comparison of corneal biomechanical characteristics after surface ablation refractive surgery and novel lamellar refractive surgery. Cornea, 34(11), pp.1441–1446.

Dupont, W.D. & Plummer, W.D., 1990. Power and sample size calculations. A review and computer program. Controlled clinical trials, 11(2), pp.116–28.

El-Massry, A.A. et al., 2015. Contralateral eye comparison between femtosecond small incision intrastromal lenticule extraction at depths of 100 and 160 mum. Cornea, 34(10), pp.1272–1275.

Frings, A. et al., 2015. Effects of laser in situ keratomileusis (LASIK) on corneal biomechanical measurements with the Corvis ST tonometer. Clin Ophthalmol., (9), pp.305–311.

Hassan, Z. et al., 2014. Examination of ocular biomechanics with a new Scheimpflug technology after corneal refractive surgery. Cont Lens Anterior Eye., 37(5), pp.337–341.

Hon, Y. & Lam, A.K., 2013. Corneal deformation measurement using Scheimpflug noncontact tonometry. Optom Vis Sci, 90(1), pp.e1-8.

Kamiya, K. et al., 2015. Effect of femtosecond laser setting on visual performance after small-incision lenticule extraction for myopia. Br J Ophthalmol, bjophthalm, p.Published Online First: 8 April 2015.

Klein, S.R. et al., 2006. Corneal ectasia after laser in situ keratomileusis in patients without apparent preoperative risk factors. Cornea, 25(4), pp.388–403.

Lanza, M. et al., 2014. Evaluation of corneal deformation analyzed with Scheimpflug based device in healthy eyes and diseased ones. Biomed Res Int., 2014, p.748671.

Lee, R. et al., 2016. Assessment of corneal biomechanical parameters in myopes and emmetropes using the Corvis ST. Clinical and Experimental Optometry, 99, pp.157–162.

Li, M. et al., 2014. Mild decentration measured by a Scheimpflug camera and its impact on visual quality following SMILE in the early learning curve. Inves Opthal Vis Sci, 55(6), pp.3886–3892.

Mastropasqua, L. et al., 2014. Evaluation of corneal biomechanical properties modification after Small Incision Lenticule Extraction using Scheimpflug-based noncontact tonometer. Biomed Res Int., 2014, p.290619.

Nemeth, G. et al., 2013. Repeatability of ocular biomechanical data measurements with

______________________________________________________________________________________________

72

a Scheimpflug-based noncontact device on normal corneas. J Refract Surg., 29, pp.558–63.

Pedersen, I.B. et al., 2014. Corneal biomechanical properties after LASIK, ReLEx flex, and ReLEx smile by Scheimpflug-based dynamic tonometry. Graefes Arch Clin Exp Ophthalmol., 252(8), pp.1329–35.

Piñero, D.P. & Alcón, N., 2015. Corneal biomechanics: a review. Clin Exp Optom., 98, pp.107–116.

Randleman, J.B. et al., 2008. Depth-dependent cohesive tensile strength in human donor corneas: implications for refractive surgery. J Refract Surg, 24(1), pp.S85–S89.


Reinstein, D.Z., Archer, T.J. & Randleman, J.B., 2013. Mathematical model to compare the relative tensile strength of the cornea after PRK, LASIK, and small incision lenticule extraction. J Refract Surg., 29(7), pp.454–60.

Roberts, C.J., 2014. Concepts and misconceptions in corneal biomechanics. J Cataract Refract Surg, 40(6), pp.862–9.

Roy, A.S. & Dupps, W.J., 2009. Effects of altered corneal stiffness on native and postoperative LASIK corneal biomechanical behavior: A whole-eye finite element analysis. Journal of refractive surgery (Thorofare, N.J. : 1995), 25(10), pp.875–887.

Shen, Y. et al., 2014. Changes in corneal deformation parameters after lenticule creation and extraction during Small Incision Lenticule Extraction (SMILE) procedure. PloS one, 9(8), p.e103893.


Simonini, I., Angelillo, M. & Pandolfi, A., 2016. Theoretical and numerical analysis of the corneal air puff test. Journal of the Mechanics and Physics of Solids, pp.1–17.

Smedowski, A. et al., 2014. Comparison of three intraocular pressure measurement methods including biomechanical properties of the cornea. Invest Ophthalmol Vis Sci, 55(2), pp.666–673.

Vestergaard, A.H. et al., 2014a. Central corneal sublayer pachymetry and biomechanical properties after refractive femtosecond lenticule extraction. J Refract Surg., 30(2), pp.102–8.

Vestergaard, A.H. et al., 2014b. Efficacy, safety, predictability, contrast sensitivity, and aberrations after femtosecond laser lenticule extraction. J Cataract Refract Surg, 40(3), pp.403–11.

Wang, D. et al., 2014. Differences in the corneal biomechanical changes after SMILE and LASIK. J Refract Surg, 30(10), pp.702–707.

Wu, D. et al., 2014. Corneal biomechanical effects: Small-Incision Lenticule Extraction versus Femtosecond Laser-Assisted laser in situ Keratomileusis. J Cataract Refract Surg, 40(6), pp.954–962.

Y Shen et al., 2014. Comparison of corneal deformation parameters after SMILE, LASEK, and femtosecond laser-assisted LASIK. J Refract Surg., 30(5), pp.310–8.


73

______________________________________________________________________________________________

74


75

2.3. New parameters for evaluating corneal biomechanics and intraocular

pressure after SMILE by Scheimpflug-Based Dynamic Tonometry.

Joaquín Fernández, MD;1 Manuel Rodríguez-Vallejo, PhD;*1 Javier Martínez, OD;1 Ana Tauste, MS; Patrizia Salvestrini, OD;1 David P Piñero, PhD;2,3


2.3.1. Abstract

Purpose: To evaluate the new Corvis ST (CoST) parameters and dynamic corneal densitometry

(CD) in eyes operated on with small incision lenticule extraction (SMILE).

Setting: Qvision, Vithas Virgen del Mar Hospital, Almería, Spain

Design: Retrospective observational case series.

Methods: 43 subjects/eyes from a single institution undergoing SMILE surgery were included in

the study. Preoperative and one month postoperative measures of CoST were taken.

Scheimpflug images were analyzed to calculate dynamic CD. Changes in normal (IOP) and

biomechanically corrected (bIOP) intraocular pressure and stiffness parameter at first

applanation (SP-A1) were also evaluated (CoST 1.3r1469).

Results: Mean values for the difference in IOP and bIOP before and after surgery were 2.24 ±

1.26 mmHg (p = 0.001) and 0.57 ± 1.77 mmHg (p = 0.04), respectively. All CoST parameters

changed significantly after SMILE (p<0.05). The variation of each parameter was correlated

with the removed corneal thickness (p<0.05), except SP-A1 (p=0.15). None of the four dynamic

CD parameters defined changed significantly due to the surgery (p≥0.29). A new sign described

as a brightness inclined fringe moving through corneal periphery appeared preoperatively in

eyes with higher dynamic CD. This sign was more prevalent postoperatively (48.8% vs. 72.1%,

p = 0.04).

Conclusions: The bIOP measured after SMILE with the new CoST shows better agreement

with the preoperative values than IOP. SP-A1 is not dependent on the amount of removed

corneal thickness. An interesting new sign correlated with dynamic CD that might be related

with changes in corneal hydration and biomechanics has been reported.

Key Words: corneal biomechanics, SMILE, Corvis ST, intraocular pressure, refractive surgery

______________________________________________________________________________________________

76

2.3.2. Introduction

The corneal biomechanical properties have a significant influence on the prediction of laser

refractive surgery (LRS) outcomes (Roy & Dupps 2009) and might be related with the

development of corneal ectasia after surgery (Klein et al. 2006). For this reason, the interest in

the clinical measure of corneal biomechanics has increased in the past few years not only due to

these two major concerns, but also because of the emergence of new technologies for the in-

vivo characterization of corneal biomechanics, such as the Corvis ST (CoST) system (Oculus

Optikgeräte, Inc., Wetzlar, Germany) (Hon & Lam 2013), and new techniques of LRS, such as

the small incision lenticule extraction (SMILE) technique (Carl Zeiss Meditec, Jena, Germany).

This LRS technique has been suggested to theoretically preserve the corneal strength to a

greater degree than previous procedures (Reinstein et al. 2013).

The CoST system is a Scheimpflug-based dynamic corneal tonometer with a high-speed camera

which captures 4330 frames per second during the course of an air-puff of less than 30 ms and

along 8-mm of horizontal corneal coverage.3 Multiple outcome parameters have been derived

from each one of the three stages: inward applanation (A1), highest concavity (HC), and

outward applanation (A2) (Pedersen et al. 2014). Some of these parameters have been shown to

be associated to poor repeatability, such as peak distance, time from starting until HC, and

A1/A2 lengths or velocities (Hon & Lam 2013; Chen et al. 2014; Nemeth et al. 2013).

Likewise, the most repeatable parameters, A1/A2 times and deformation amplitude, have been

also questioned as they are conditioned by confounding variables such as central corneal

thickness (CCT) or intraocular pressure (IOP) (Fernández, Rodríguez-Vallejo, et al. 2016) that

must be considered in research studies in order to avoid possible misinterpretations

(Vinciguerra, Elsheikh, et al. 2016).

The conclusions obtained from clinical research studies using the CoST system and comparing

LRS techniques still remain controversial. Chen et al (Chen et al. 2014) reported differences

between photorefractive keratectomy (PRK) and virgin eyes, but Pedersen et al (Pedersen et al.

2014) did not find differences between control eyes and eyes undergoing laser in situ

keratomileusis (LASIK), femtosecond lenticule extraction (FLEX) or SMILE groups for all

variables except for the A1 deflection length. Shen et al. (Shen et al. 2014) demonstrated that

changes in CoST parameters were due to lenticule extraction and not to lenticule creation in

eyes undergoing SMILE. Osman et al. (Osman et al. 2016) reported greater reduction of

biomechanical properties in LASIK than in SMILE, contrary to Sefat et al (Sefat et al. 2016)

who found equal differences in a good study controlling confounding variables through

subgroups.

The new CoST software (1.3r1469) offers new parameters for improving the measure of the

IOP and corneal biomechanics by reducing the influence of confounding variables and including

normative values (Vinciguerra, Elsheikh, et al. 2016). The aim of this retrospective study was to

test these new parameters in eyes before and after SMILE and also to evaluate the dynamic


77

corneal densitometry (CD) as a potential future clinical measure.

2.3.3. Methods

Patients

This study was approved by the Ethics Committee of Research from Almería Center

(Torrecardenas Hospital Complex) and performed in adherence to the tenets of the Declaration

of Helsinki. Data from fifty-four subjects operated on with the SMILE technique at Qvision

(Department of Ophthalmology, Virgen del Mar Hospital) were extracted from our historic

CoST database. Inclusion criteria included myopic patients with spherical equivalent from -1.00

D to -7.00 D and refractive astigmatism below 2.25 D. Manifest refraction was measured

preoperatively and one month after SMILE with CoST. Preoperative exclusion criteria were

pregnancy at time of surgery or follow-up, a preoperative central corneal thickness of less than

480 µm, an expected postoperative residual stromal bed of less than 250 µm, topographic map

compatible with subclinical keratoconus or other ectatic corneal disorder, and any other ocular

disease for which LRS procedures are not indicated (Shetty et al. 2015). Those patients with

alert messages in the CoST measurement of “Pressure Profile” and “Lost Images” indicating a

poor quality of the measure were excluded. However, measurements with other alerts were not

discarded in case the manual inspection resulted in a good delimitation of the corneal profile.

Surgical procedure

The same surgeon (JF) performed all the SMILE treatments with the VisuMax femtosecond

laser system (Carl Zeiss Meditec AG, Jena, Germany). Two drops of topical anesthesia

(Oxybuprocaine Hydrochloride 0.4%) were instilled at 5 minutes and 2 further drops at 1

minute before surgery. Optical principles and general description of the SMILE procedure have

been widely described (Reinstein et al. 2014), and the particular laser settings or surgical

maneuvers used in this study have been detailed in a previous work of our research group

(Fernández, Valero, et al. 2016). However, some additional important information about the

surgery include: the use of a cap diameter of 7.6 mm and cap thickness of 140 µm, the

performance of a peripheral incision of 2 mm with 30º of angle for posterior lenticule extraction

at 70º and the direct insertion of the target refractive error correction in the software without

applying any nomogram. After laser treatment, the patient was moved to the surgical

microscope for the second part of the procedure. Finally, 2 drops of tobramycin (3 mg) and

dexamethasone (1 mg) combination were instilled in all cases at the end of the procedure.

Postoperative treatment included ofloxacin (3 mg) during 2 days, dexamethasone drops 5, 3, 2

and 1 per day reducing the dosage each seven days and sodium hyaluronate (0.15%) during a

month.

______________________________________________________________________________________________

78

New CoST parameters

The CoST obtains the parameters among three stages during the air puff course: (1) the air puff

starts and the cornea is flattened in the center, this corresponds to A1 during the inward stage of

the air-puff; (2) the pressure continues up to the peak pressure, instant when the HC is achieved;

(3) the air puff pressure decreases and the cornea comes back to the baseline state achieving the

A2 just before to recover its original shape (Pedersen et al. 2014). New indexes have been

developed considering the corneal response at these three stages, including:

The biomechanically corrected intraocular pressure (bIOP): An estimate of the IOP for

minimizing the influence of different variables such as CCT and age over the

conventional IOP (Joda et al. 2016). The final bIOP included in the new CorvisST is a

modified algorithm of the previous published formula (Joda et al. 2016), which has

confirmed the previous results (Vinciguerra, Elsheikh, et al. 2016). The new version

takes the dynamic corneal response into consideration and it also corrects based on

radius at HC using a proprietary algorithm.

Deflection Amplitude Ratio (DAR): The ratio of the central corneal deflection and the

average of two points located at one (DAR1) or two (DAR2) millimeters at both sides

from the center. The deflection length is obtained by means of correcting the whole eye

movement and overlapping the peripheral cornea at the baseline state with the cornea at

the highest concavity (Cynthia Roberts 2016). Stiffer corneas would have lower DARs

as the corneal center and the cornea at 1 or 2 mm deflect at same time, whereas higher

DARs indicate that central cornea deflects more than the average of the other two other

points, corresponding to softer tissue.

Stiffness Parameter at First Applanation (SP-A1): Defined as the resultant pressure at

A1 from the difference between the air puff pressure at the corneal surface and the

bIOP, divided by the deflection amplitude (Cynthia Roberts 2016). Higher values

indicate stiffer corneas.

Integrated Inverse Concave Radius (IR): The integrated area under the curve of the

inverse concave radius, which is the radius of curvature during the concave phase of

deformation. Higher IR indicates softer tissue.

Corneal Biomechanical Index (CBI): Normalized index from 0 (normal) to 1

(abnormal) obtained by means of logistic regression with the combination of different

CoST parameters that enhances the sensitivity between keratoconic and healthy eyes

using a proprietary algorithm. These parameters include: DAR1, DAR2, velocity at A1,

standard deviation of deformation amplitude at HC, the Ambrosio relational thickness


79

to the horizontal profile, and the SP-A1 (Vinciguerra, Ambrósio, et al. 2016).

Dynamic Corneal Densitometry

CD is a measure of the corneal backscattered light by the analysis of the Scheimpflug image

brightness represented in percentage of grey levels (Ní Dhubhghaill et al. 2014). Therefore, the

dynamic CD can be defined as the increase in the densitometry during the corneal deformation

by the air-puff. Four variables are distinguished: the maximum densitometry increase (DIM)

which is the maximum increase of densitometry achieved during the deformation, and the

densitometry increase (DI) at each one of the stages A1 (DI-A1), HC (DI-HC) and A2 (DI-A2).

CD variables are only included in the research software and not in the commercial version.

Statistical Analysis

Only right eyes were included in the sample, but the left eye was included instead of the

right if exclusion messages appeared in the measurement of such eye. Outliers from the

difference between preoperative and postoperative measures were identified by plotting

and analyzing the boxplots. Those cases at more than 1.5 box-lengths of the edge of the

box were considered as outliers. Eleven out of 54 cases were identified as outliers and

an individual data inspection was performed in order to verify the reliability of the data.

These cases were finally excluded after verifying that were erroneous and not reliable

measurements: higher corneal thickness after surgery than before (4 cases), higher

intraocular pressure after surgery (3 cases), and alert messages of model deviation (4

cases). Paired t-tests were conducted for testing differences before and after the

procedure and Pearson r for evaluating correlations. For all the variables which were not

normally distributed, the Wilcoxon signed rank test was used for testing paired

differences and spearman rho for correlations. Bland-Altman plots were used to

evaluate the agreement between preoperative and postoperative IOPs. Measures were

classified in two time slots, from 9:00 to 15:00 and from 15:00 to 21:00, in order to

compute mean differences for measures taken during the same time slot.

______________________________________________________________________________________________

80

2.3.4. Results

A total of 43 eyes from 18 men and 23 women of mean age 31.63 ± 6.55 years old

(range, 22 to 45 years) were included in the analysis. The mean preoperative spherical

equivalent was -3.87 ± 1.43 D (range, -1.00 to -6.88 D).

New CoST parameters

The IOP and bIOP parameters correlated significantly with the subjects’ age before

SMILE (Table 1). Significant mean differences were obtained, either for IOP or bIOP,

between preoperative and postoperative values, although differences were lower for

bIOP (Table 2). The removed CCT was significantly correlated with the change in IOP,

but not with the variation in bIOP (Table 3). The mean difference in hours throughout

the day between the preoperative measure and the postoperative measure was

significantly correlated with the mean bias due to surgery, either for IOP and bIOP (r = -

0.42, p = 0.006). A total of 15 subjects were measured at the first time slot

preoperatively and in the second post-operatively. A total of 21 subjects were measured

at the same time slot resulting in a mean bias of 2.24 ± 1.26 mmHg for IOP (z=-3.32, p

= 0.001) and 0.57 ± 1.77 mmHg for bIOP (z=-2.09, p = 0.04).

Table 1. Spearman rho correlations between age and the preoperative and postoperative parameters. Age

Preoperative Postoperative Difference ()

IOP (mmHg) 0.39, p=0.02* 0.08, p=0.62 0.33, p=0.03* bIOP (mmHg) 0.32, p=0.04* -0.01, p=0.93 0.25, p=0.10 SP-A1 (mmHg/mm) 0.23, p=0.13 0.27, p=0.08 -0.14, p=0.37 DAR1 -0.42, p=0.005* -0.28, p=0.07 -0.04, P=0.81 DAR2 -0.39, p=0.01* -0.29, p=0.06 -0.001, p=0.99 IR (mm-1) -0.22, p=0.15 -0.18, p=0.25 0.006, p=0.97 CBI -0.34, p=0.03* -0.26, p=0.09 0.24, p=0.12 DIM (%) -0.52, p<0.001* -0.15, p=0.33 -0.14, P=0.36 DI-A1 (%) -0.42, p=0.005* -0.29, p=0.06 -0.14, P=0.36 DI-HC (%) -0.50, p<0.001* -0.13, p=0.41 -0.17, P=0.29 DI-A2 (%) -0.24 p=0.12 -0.01, p=0.95 -0.16, p=0.31

IOP= Non-contact intraocular pressure; bIOP = Biomechanical corrected IOP; SP-A1 = Stiffness parameter; DAR = Deflection amplitude ratio at 1 mm (DAR1) and 2 mm (DAR2); IR = Integrated inverse curvature radius; CBI = Corneal biomechanical index; DIM = Densitometry increase maximum; DI = Densitometry increase at A1 (DI-A1), HC (DI-HC) and A2 (DI-A2). * p <0.05


81

Table 2. Differences between preoperative and postoperative measures, excluding the outliers.

Variable Pre-SMILE mean ± SD [range] median, IR [range]

Post-SMILE mean ± SD [range] median, IR [range]

t-test

wilcoxona

IOP (mmHg) 15 ± 2.53 [11 to 23]

11.80 ± 1.92 [8 to 17.50]

t= 7.69, p<0.001*

bIOP (mmHg) 14.68 ± 1.99 [11 to 21.60]

13.30 ± 1.74 [10.80 to 17.70]

t= 3.99, p<0.001*

SP-A1 (mmHg/mm) 148.95 ± 12.94 [126.61 to 183.87]

142.48 ± 17.31 [101.67 to 176.79]

t= 3.05, p=0.004*

DAR1 1.56 ± 0.55 [1.43 to 1.71]

1.67 ± 0.06 [1.52 to 1.79]

t= -13.96, p<0.001*

DAR2 4.18 ± 0.40 [3.23 to 4.92]

5.16 ± 0.56 [3.65 to 6.26]

t= -18.07, p<0.001*

IR (ms*mm-1) 6.87 ± 0.98 [4.91 to 9.07]

9.17 ± 1.25 [6.54 to 12.62]

t= -22.75, p<0.001*

CBI 0.002, 0.16 [0 to 0.09]

0.42, 0.86 [0 to 1]

z= 5.58, p<0.001a*

DIM (%) 31.05 ± 4.47 [22.93 to 44.26]

29.68 ± 5.92 [15.23 to 40.81]

t= 1.49, p=0.14

DI-A1 (%) 6.25 ± 1.85 [3.04 to 11.74]

6.26 ± 1.44 [3.97 to 9.75]

t= -0.06, p=0.95

DI-HC (%) 30.55 ± 4.43

[22.70 to 43.78] 28.99 ± 6.02

[14.04 to 40.63] t= 1.67, p=0.10

DI-A2 (%) 10.99, 2.94 [6.66 to 17.67]

11.72, 4.18 [-2.91 to 42.86]

z= 1.04, p=0.30a

t-test for paired samples. Mean ± Standard Deviation (SD) is shown. a wilcoxon signed-rank test. Median, Interquartile Range (IR) is shown. IOP= Non-contact intraocular pressure; bIOP = Biomechanical corrected IOP; SP-A1 = Stiffness parameter; DAR = Deflection amplitude ratio at 1 mm (DAR1) and 2 mm (DAR2); IR = Integrated inverse curvature radius; CBI = Corneal biomechanical index; DIM = Densitometry increase maximum; DI = Densitometry increase at A1 (DI-A1), HC (DI-HC) and A2 (DI-A2). * p <0.05

Table 3. Correlations between the variation of variables from preoperative to postoperative and the

removed corneal thickness.

CCT DIM DI-A1 DI-HC DI-A2

IOP (mmHg) r= 0.42*, p=0.005

r= -0.58*, p<0.001

r= -0.001, p=0.99

r= -0.55*, p<0.001

ρ = -0.31*, p=0.04

bIOP (mmHg) r= 0.24, p=0.12

r= -0.58*, p<0.001

r= -0.03, p=0.85

r= -0.55*, p<0.001

ρ = -0.31*, p=0.04

SP-A1 (mmHg/mm) r= 0.23, p=0.15

r= -0.09, p=0.56

r= -0.04, p=0.79

r= -0.05, p=0.76

ρ = 0.02, p=0.92

DAR1 r= -0.48*, p=0.001

r= 0.09, p=0.58

r= -0.34*, p=0.03

r= 0.09, p=0.58

ρ = -0.05, p=0.73

DAR2 r= -0.50*, p=0.001

r= 0.17, p=0.28

r= 0.06, p=0.71

r= -0.13, p=0.39

ρ = -0.04, p=0.82

IR (ms*mm-1) r= -0.70*, p<0.001

r= 0.38*, p=0.01

r= 0.09, p=0.54

r= 0.35*, p=0.02

ρ = 0.12, p=0.44

CBI ρ = -0.47*, p=0.002

ρ = -0.12, p=0.45

ρ = 0.04, p=0.83

ρ = -0.08, p=0.59

ρ = -0.29, p=0.06

Pearson correlations (r) and spearman rho (ρ).

= Difference preoperative value – postoperative value; ρ = Spearman rho; r = Pearson r; IOP= Non-contact intraocular pressure; bIOP = Biomechanical corrected IOP; SP-A1 = Stiffness parameter; DAR = Deflection amplitude ratio at 1 mm (DAR1) and 2 mm (DAR2); IR = Integrated inverse curvature radius; CBI = Corneal biomechanical index; DIM = Densitometry increase maximum; DI = Densitometry increase at A1 (DI-A1), HC (DI-HC) and A2 (DI-A2). * p <0.05

______________________________________________________________________________________________

82

All new CoST parameters changed significantly after SMILE (Table 2). The increase in

DAR1 and DAR2 was 0.11 and 0.98, respectively, after surgery. Negative significant

correlations of both changes with the removed CCT (CCT) were found (Table 3). This

means that the more CCT was removed, the higher was the increment in DARs

(DARs). Similar behavior was obtained for IR, with a stronger correlation with CCT

than DARs (Table 3). The stiffness parameter SP-A1 decreased significantly after

SMILE and was not correlated with CCT (Table 3). None of variations of these

parameters described in Table 3 were significantly correlated with the age of the

subjects (p > 0.05)(See Difference in Table 1).

Dynamic Corneal Densitometry

Negative significant correlations were found between preoperative CD variables and

age but these significances disappeared after surgery (Table 1). No significant changes

in any of the densitometry parameters defined and evaluated were found after surgery

(Table 2). Correlations were also manifested for the variation of some of the previous

new CoST parameters and the modification of CD due to surgery (Table 3). The mean

percentage of DIM before SMILE was close to the mean achieved for DI-HC (Table 1).

Mean differences between DI-A1 and DI-A2 for the preoperative dataset were -5.11 ±

1.84 (paired t-test, t=-18.24, p<0.001). These tendencies were similar after SMILE, with

DIM also close to DI-HC and DI-A1 lower than DI-A2. An interesting sign was

observed at the dynamic response of the cornea described as a brightness inclined fringe

that appears in the corneal peaks at HC, moving to the corneal periphery during the

outward stage (Figure 1). The visualization of the 43 preoperative and postoperative

videos resulted in the recognition of this sign in 48.8% (n=21) of the preoperative

records and in 72.1% (n=31) of the postoperative records for the same eyes. This higher

prevalence of this sign postoperatively was statistically significant (McNemar's test, p =

.04).

A comparison of mean densitometry for eyes which presented the sign and those for

which the sign was not visible was performed either for preoperative and postoperative

records (Table 4). Eyes which showed the sign in the preoperative records had a higher

densitometry percentage for all the densitometry variables, but this difference was only

statistically significant for DI-A1.


83

Table 4. Corneal densitometry in groups for which of the bright fringe in movement was shown (Sign) for

the preoperative and postoperative records.

Figure 1. Sign observed at the dynamic response of the cornea after SMILE surgery which is a brightness

inclined fringe that appears in the peripheral corneal peaks at the highest concavity stage (HC) (bottom

image), moving to the corneal periphery until its disappearance during the outward stage. This sign is not

commonly observed preoperatively (top image).

2.3.5. Discussion

In this study, we present the changes of new CoST parameters due to SMILE surgery

and we analyze the dynamic CD increase as a new potential variable for explaining

corneal biomechanics. We found better agreement between preoperative and

postoperative intraocular pressure for bIOP than for conventional IOP measured with

Preoperative Records; mean ± SD [range] or median, IR [range] Postoperative Records; mean ± SD [range] or median,

IR [range]

No Sign (n = 21)

Sign (n = 22)

t-test U Mann-Whitneyb

No Sign (n = 12)

Sign (n = 31)

t-test U Mann-Whitneyb

DIM (%) 29.45 ± 3.05

[22.93 to 35.12] 32.58 ± 5.11

[24.39 to 44.26] t=-2.43, p = 0.02*

28.40 ± 7.02 [15.23 to 39.95]

30.18 ± 5.48 [19.02 to 40.81]

t=-0.88, p = 0.38

DI-A1 (%) 5.49 ± 1.14

[3.04 to 7.46] 6.97 ± 2.12

[3.81 to 11.74] t=-2.86, p = 0.007*

5.44 ± 0.95 [3.97 to 6.97]

6.58 ± 1.48 [4.42 to 9.75]

t=-2.48, p = 0.02*

DI-HC (%) 28.91 ± 3.10

[22.70 to 34.28] 32.12 ± 4.99

[24.05 to 43.78] t=-2.52, p = 0.02*

27.88 ± 7.23 [14.04 to 39.95]

29.42 ± 5.55 [18.96 to 40.63]

t=-0.74, p = 0.46

DI-A2 (%) 10.67, 2.51

[6.66 to 14.28 ] 12.08, 3.08

[8.77 to 17.67] z=2.70, p = 0.007b*

11.85 ± 5.83 [1 to 22.86 ]

11.90 ± 2.89 [6.74 to 18.05 ]

z=-0.80 p = 0.42b

t-test for independent samples. Mean ± Standard Deviation (SD) is shown. b U Mann-Whitney. Median, Interquartile Range (IR) is shown. DIM = Densitometry increase maximum; DI = Densitometry increase at A1 (DI-A1), HC (DI-HC) and A2 (DI-A2). * p <0.05

______________________________________________________________________________________________

84

CoST. Despite of the fact that significant differences were found between preoperative

and postoperative means, these results should be interpreted with caution since some

measures were taken at different time slots. Considering that IOP decreases during the

day (David et al. 1992) and up to 15 patients were measured in the first time slot

preoperatively and in the second postoperatively, some differences can be explained by

this study limitation. In fact, we found a correlation between the difference in hours

along a day and mean IOP differences due to surgery in such a way that higher

differences in IOP were found for those patients measured early in the morning

preoperatively and in the evening postoperatively. After considering only the 21

subjects measured in the same time slot, the bias was reduced from 1.39 ± 2.28 mmHg

to 0.57 ± 1.77 mmHg for bIOP. The preoperative bIOP was slightly correlated with age,

even though this new parameter considers age in the correction algorithm (Vinciguerra,

Elsheikh, et al. 2016). On the other hand, the change of bIOP due to surgery was not

correlated with the amount of removed corneal thickness which suggests an important

improvement in the measure of intraocular pressure with the instrument.

The variation of some parameters, such as applanation times or deformation amplitude,

due to SMILE can be explained mainly by the removed corneal thickness (Fernández,

Rodríguez-Vallejo, et al. 2016). It is well known that CCT and IOP are confounding

variables that may influence on other CoST parameters (Y Shen et al. 2014). Therefore,

comparison studies of LRS techniques should be well designed with uniform samples

with same removal of CCT and preoperative IOP in order to avoid misinterpretations of

the results (Vinciguerra, Elsheikh, et al. 2016).

We found significant changes in all the new CoST parameters due to SMILE, but

interestingly all the parameters were significantly correlated with the removed CCT

except the stiffness parameter SP-A1. This suggests that this new parameter would be a

promising indicator to predict the biomechanical properties of the cornea with

independence of the tissue volume. We also found significant negative correlations

between DARs and CBI parameters with age indicating a slight decrease in both with

age; this is in agreement with the increase of stiffness with age (Elsheikh et al. 2007).

Conversely, the stiffness parameter SP-A1 was not correlated with age. All these

correlations became poorer after surgery which indicates that the normal age-related

corneal response might be altered due to surgery. However, this should be interpreted

with caution because the non-uniform correction of refractive errors for the different

ages.

Concerning densitometry, it is important to differentiate between static CD measured


85

with Pentacam and dynamic CD measured with CoST. The first one represents the

natural state of the corneal fibrils and corneal hydration whereas the latter

hypothetically would represent the modification of collagen fibers order and fluidics

movement along the cornea during the air-puff course. In our series, at the baseline

state, the densitometry measured with Pentacam was maximal in the anterior 120 µm of

the cornea in comparison with the center and posterior layers (Ní Dhubhghaill et al.

2014). Considering that the normal epithelial thickness is around 53.4±4.6 μm (Elsheikh

et al. 2007) and that the lamellar angles relative to the stromal surface are highest in the

anterior-most 83 µm of the corneal stroma, our hypothesis is that the increase of the

anterior densitometry is not only due to epithelial thickness but also to the angle of the

collagen lamellae (Abass et al. 2015) at this part of the cornea and possibly due to the

fact that the anterior stroma tends to be less hydrated.

In the study of dynamic CD, we found that CD was increased during the inward stage

achieving the maximum close to the HC, whereas during outward stage the CD at A2

was higher than the obtained at A1. Ariza et at. (Ariza-Gracia et al. 2015) pointed out

that during the loading pressure the anterior stroma goes from a tension state to a

compression whereas the posterior stroma experiences greater tensional stress. Our

explanation about the course of dynamic CD is that the stromal fluid goes from the

anterior to the posterior stroma with the air puff pressure, whilst the anterior fibers are

compressed or reordered. Therefore, both corneal hydration (Wang et al. 2004) and

fibers arrangement (Leonard & Meek 1997) would play an important role in light

scattering. This should be confirmed in future studies.

Finally, we found in dynamic CD a clinical sign consisting of an inclined brightness

fringe which is moved through the corneal periphery. The origin of this sign has not

been previously reported and it is completely unknown. Our hypothesis is that this sign

might be explained by the reorder/compression of the fibers and the stromal fluid

repositioning through peripheral areas of less pressure in the stroma. Thus, the sign

would appear in corneas with higher hydration and greater dynamic CD. For a better

understanding, we attach a video that shows the dynamic CD profile preoperatively, at 1

month and 24 months after surgery in one eye from our series undergoing SMILE. As

displayed, the sign was not present preoperatively, appeared at 1 month and at 24

months after surgery is almost eliminated. Considering that the increase of corneal

backscattered light in Pentacam has been also correlated with the serum concentration

of the active metabolite N-desethylamiodarone during the amiodarone therapy

(Alnawaiseh et al. 2016), future research studies in treated versus untreated corneas

achieving different degrees of hydration would help to confirm our hypothesis. We also

______________________________________________________________________________________________

86

found that the dynamic CD decreases with the increase of the age whereas the static CD

measured with Pentacam have been reported to have the contrary effect (Ní Dhubhghaill

et al. 2014). This might be due to the formation of cross-links, decrease in interfibrillar

spacing and increase of collagen fibrils diameter (Quantock et al. 2015) which difficult

fibers reorder and fluidics transition with age (Whitford et al. 2015).

We are aware that our research has two potential limitations. First, postoperative

measures were taken only 1 month after surgery and some corneas might not be

stabilized by the recent end of post-operative ophthalmologic treatment. This may be

one reason explaining the presence of some atypical results (outliers) that were detected

and removed from the statistical sample. In any case, a reanalysis including the outliers

was also performed, but the results did not change the conclusions obtained in the study

(Table 5). Nevertheless, future studies should be conducted for longer periods of

follow-up because as it is shown in the attached video, the detected sign might appear

for reasons that are not covered in this work and this should be specifically investigated

in connection with corneal biomechanics. Second, an early study was not available in

order to compute the required sample for each one of the hypothesis managed in this

study. However, we conducted a posterior power analysis using G Power version 3.1

(available at http://www.gpower.hhu.de/) to confirm whether the sample of eyes

included was of adequate size to detect a true difference in population means () with

type I error probability () given a standard deviation (). Specifically, the sample was

enough for achieving a statistical power of 80%, considering and changes after

SMILE for all the new parameters. However, for detecting differences in CD among the

groups that presented or not the sign, the statistical power decreased to 76% for DIM,

and 70% for DI-HC in the preoperative records suggesting the need of a higher sample

in future studies involving CD.


87

Table 5. Differences between preoperative and postoperative measures, including the outliers. Variable Pre-SMILE

mean ± SD [range] median, IR [range]

Post-SMILE mean ± SD [range] median, IR [range]

t-test

wilcoxona

IOP (mmHg) 14.50, 3.50 [10 to 26]

11.50, 3.00 [8 to 17.50]

z= -5.51, p<0.001a*

bIOP (mmHg) 14.74 ± 2.39 [10.10 to 23.90]

13.38 ± 1.85 [10.60 to 17.70]

t= 4.11, p<0.001*

SP-A1 (mmHg/mm) 148.48 ± 13.89 [123.31 to 183.87]

142.30 ± 17.03 [101.67 to 176.79]

t= 3.22, p=0.002*

DAR1 1.56, 0.06 [1.43 to 2.01]

1.66, 0.1 [1.50 to 1.79]

z= 5.45 p<0.001 a*

DAR2 4.22, 0.58 [3.23 to 8.04]

5.13, 0.82 [3.65 to 6.26]

z= 5.77, p<0.001a*

IR (ms*mm-1) 6.86, 1.24 [4.91 to 10.03]

9.10, 1.81 [6.54 to 12.62]

z= 6.15, p<0.001a*

CBI 0.002, 0.01 [0 to 1]

0.27, 0.85 [0 to 1]

z= 5.67, p<0.001a*

DIM (%) 30.60, 5.17 [7.74 to 44.26]

29.05, 9.11 [1.11 to 40.81]

z= -1.38, p=0.17a

DI-A1 (%) 5.91, 2.41 [1.14 to 37.92]

6.16, 2.29 [-11.38 to 41.48]

z= 0.82, p=0.41a

DI-HC (%) 29.90, 5.47

[-36.32 to 43.78] 27.77, 9.26

[-52.48 to 40.63] z= -1.51, p=0.13a

DI-A2 (%) 11.06, 3.69 [-30.13 to 54.00]

12.03, 4.43 [-44.60 to 57.00]

z= 1.88, p=0.06a

t-test for paired samples. Mean ± Standard Deviation (SD) is shown. a wilcoxon signed-rank test. Median, Interquartile Range (IR) is shown. IOP= Non-contact intraocular pressure; bIOP = Biomechanical corrected IOP; SP-A1 = Stiffness parameter; DAR = Deflection amplitude ratio at 1 mm (DAR1) and 2 mm (DAR2); IR = Integrated inverse curvature radius; CBI = Corneal biomechanical index; DIM = Densitometry increase maximum; DI = Densitometry increase at A1 (DI-A1), HC (DI-HC) and A2 (DI-A2). * p <0.05

In conclusion, biomechanical changes characterized using the new parameters provided

by the CoST system occur after SMILE. SP-A1 has not shown correlations with the

removed CCT which supposes an advance for avoiding the effect of confounding

variables, and preoperative DARs and CBI are able to predict the increase of corneal

stiffness with age. Likewise, the postoperative bIOP measured with the improved

version of the CoST shows better agreement with the preoperative values than IOP with

independence of CCT. For bIOPs measured during the same time slot, the preoperative

bIOP can be estimated adding 0.57 mmHg in subjects operated on SMILE. Considering

the standard deviation of 1.77 mmHg, the maximum bias after applying this correction

would be of 2x1.77 = 3.54 mmHg with a 95% of confidence. The dynamic CD have not

shown significant differences due to surgery and an interesting new sign that might be

related with changes in corneal hydration and biomechanics has been reported even

though some caution should be taken before to assume this relationship because it is a

hypothesis that must be tested in controlled studies in vitro or in vivo. Future research

including this new parameter would help to understand the structure of the cornea in

terms of collagen fibers order and hydration after SMILE.

______________________________________________________________________________________________

88

2.3.6. Disclosures

a. Funding/Support: Authors have not received financial support to conduct this

research.

b. Financial Disclosures: J.F. reports personal fees from OCULUS and personal fees

from ZEISS, outside the submitted work. The remaining authors have nothing to

disclose.


89

2.3.7. References

Abass, A. et al., 2015. Transverse depth-dependent changes in corneal collagen lamellar orientation and distribution. J. R. Soc. Interface, 12, p.20140717.

Alnawaiseh, M. et al., 2016. Corneal densitometry as a novel technique for monitoring amiodarone therapy. Ophthalmology, 123(11), pp.2294–2299.



Cynthia Roberts, 2016. Two novel stiffness parameters for the Corvis® ST. In XXXIV Congress of the ESCRS. Copenhagen. Available at: http://www.corneal-biomechanics.de/.

David, R. et al., 1992. Diurnal intraocular pressure variations: an analysis of 690 diurnal curves. British Journal of Ophthalmology, 76(5), pp.280–283.

Elsheikh, A. et al., 2007. Assessment of corneal biomechanical properties and their variation with age. Current eye research, 32(1), pp.11–19.

Fernández, J., Rodríguez-Vallejo, M., et al., 2016. Corneal thickness after SMILE affects scheimpflug-based dynamic tonometry. Journal of Refractive Surgery, 32(12), pp.821–828.

Fernández, J., Valero, A., et al., 2016. Short-term outcomes of small-incision lenticule extraction (SMILE) for low, medium, and high myopia. European Journal of Ophthalmology, Jul 18(0), pp.0–0.

Hon, Y. & Lam, A.K., 2013. Corneal deformation measurement using Scheimpflug noncontact tonometry. Optom Vis Sci, 90(1), pp.e1-8.

Joda, A.A. et al., 2016. Development and validation of a correction equation for Corvis tonometry. Computer methods in biomechanics and biomedical engineering, 19(9), pp.943–53.


Leonard, D.W. & Meek, K.M., 1997. Refractive indices of the collagen fibrils and extrafibrillar material of the corneal stroma. Biophysical Journal, 72(3), pp.1382–1387.

Nemeth, G. et al., 2013. Repeatability of ocular biomechanical data measurements with a Scheimpflug-based noncontact device on normal corneas. J Refract Surg., 29, pp.558–63.

Ní Dhubhghaill, S. et al., 2014. Normative values for corneal densitometry analysis by scheimpflug optical assessment. Inves Opthal Vis Sci, 55(1), p.162.

Osman, I.M. et al., 2016. Corneal biomechanical changes in eyes with small incision lenticule extraction and laser assisted in situ keratomileusis. BMC Ophthalmology, 16(1), p.123.

Pedersen, I.B. et al., 2014. Corneal biomechanical properties after LASIK, ReLEx flex, and ReLEx smile by Scheimpflug-based dynamic tonometry. Graefes Arch Clin Exp Ophthalmol., 252(8), pp.1329–35.

Quantock, A.J. et al., 2015. From nano to macro: Studying the hierarchical structure of the corneal extracellular matrix. Experimental Eye Research, 133, pp.81–99.

Reinstein, D., Archer, T.J. & Gobbe, M., 2014. Small Incision Lenticule Extraction (SMILE) history, fundamentals of a new refractive surgery technique and clinical

______________________________________________________________________________________________

90

outcomes. Eye and Vision, 1(1), p.3. Reinstein, D.Z., Archer, T.J. & Randleman, J.B., 2013. Mathematical model to compare

the relative tensile strength of the cornea after PRK, LASIK, and small incision lenticule extraction. J Refract Surg., 29(7), pp.454–60.

Roy, A.S. & Dupps, W.J., 2009. Effects of altered corneal stiffness on native and postoperative LASIK corneal biomechanical behavior: A whole-eye finite element analysis. Journal of refractive surgery (Thorofare, N.J. : 1995), 25(10), pp.875–887.

Sefat, S.M.M. et al., 2016. Evaluation of changes in human corneas after femtosecond laser-assisted LASIK and Small-Incision Lenticule Extraction (SMILE) using non-contact tonometry and ultra-high-speed camera (Corvis ST). Current Eye Research, 41(7), pp.917–922.



Vinciguerra, R., Ambrósio, R., et al., 2016. Detection of keratoconus with a new biomechanical index. Journal of refractive surgery (Thorofare, N.J. : 1995), 32(12), pp.803–810.

Vinciguerra, R., Elsheikh, A., et al., 2016. Influence of pachymetry and intraocular pressure on dynamic corneal response parameters in healthy patients. Journal of Refractive Surgery, 32(8), pp.550–561.

Wang, J., Simpson, T.L. & Fonn, D., 2004. Objective measurements of corneal light-backscatter during corneal swelling, by optical coherence tomography. Investigative Ophthalmology and Visual Science, 45(10), pp.3493–3498.

Whitford, C. et al., 2015. Biomechanical model of the human cornea: Considering shear stiffness and regional variation of collagen anisotropy and density. Journal of the Mechanical Behavior of Biomedical Materials, 42, pp.76–87.

Y Shen et al., 2014. Comparison of corneal deformation parameters after SMILE, LASEK, and femtosecond laser-assisted LASIK. J Refract Surg., 30(5), pp.310–8.


91

CAPÍTULO 3 DISCUSIÓN DE LOS RESULTADOS

______________________________________________________________________________________________

92


93

3.1 Discusión de los resultados.

La inclusión de cualquier nuevo procedimiento de cirugía refractiva comienza

con la evaluación de los resultados de eficacia, seguridad y predictibilidad. Estos

adquieren especial importancia durante los primeros sujetos sometidos a la nueva

cirugía, en los cuales entra en juego la curva de aprendizaje del cirujano. El primero de

los artículos incluidos en esta Tesis por compendio de artículos se centra en la

evaluación de los resultados de seguridad, eficacia y predictibilidad para los 71

primeros sujetos intervenidos con SMILE por un cirujano experimentado en cirugía

refractiva, pero sin experiencia previa en la técnica. Para las condiciones metodológicas

particulares de este estudio, encontramos que la agudeza visual con corrección tras el

procedimiento se mantenía estable en un 95.8% de los casos, con tan solo un 1.4% (1

ojo) perdiendo una línea de agudeza visual con la mejor corrección. A estos resultados

de seguridad se suma la presencia tan solo de una pérdida de succión con las

complicaciones secundarias de endocrecimiento epitelial, pliegues corneales y

astigmatismo irregular.

Una de las ventajas de nuestra metodología frente a estudios similares es la

estratificación de la muestra en tres niveles de error refractivo, bajo, medio y alto. Esta

estratificación permite evaluar en mejor medida la predictibilidad de la técnica.

Considerando todo el rango de error refractivo, nos encontramos una pendiente para la

curva que relaciona error refractivo planificado y tratado realmente de 0.95 (R2 = 0.99).

Esto describe una hipocorrección de la miopía con el incremento del error de refracción.

La comparativa del error de refracción posoperatorio en los tres grupos resultó en

diferencias significativas con medianas sobre la emetropía (0 D) para errores de

refracción entre -1.00 D y -5.00 D, mientras que la mediana para el grupo de alta miopía

(entre -5.25 D y -7.00 D) se encontraba en -0.50 D, sugiriendo la necesidad de una

corrección del nomograma de tratamiento en aquellos casos en los que la miopía supere

las -5.00 D.

La tendencia a la hipocorrección del grupo de alta miopía se encuentra

relacionada con los resultados de eficacia. Mientras que el 67% y 74% de los grupos de

baja y media miopía alcanzaban una agudeza visual sin corrección de 20/20, tan solo el

50% alcanzó dicho valor en el grupo de alta miopía. Pese a los peores resultados en el

grupo de alta miopía, la mediana se encontró sobre 20/20 en los tres grupos sin

diferencias significativas entre ellos. La comparación entre los resultados de agudeza

visual preoperatoria con gafas y postoperatoria sin gafa mostró un menor porcentaje de

sujetos (17% frente a 4%) alcanzando valores por encima (20/16) de la agudeza

considerada como normal (20/20). Esta disminución fue notablemente más marcada en

el grupo de baja miopía, en el que 23% de los sujetos que tenían 20/16 antes de la

______________________________________________________________________________________________

94

operación con su mejor corrección pasaron a ser 7% sin gafa tras la operación y del 97%

con 20/20 pasaron a 67%. La explicación de este mayor salto en el caso de la baja

miopía se debe principalmente al efecto de magnificación de la gafa que reduce el

tamaño de las imágenes en mayor medida para los grupos de media y alta miopía y, por

tanto, reduce la agudeza visual preoperatoria con gafa. De hecho, tan solo el 81% y 61%

de los sujetos con miopías media y alta alcanzaban una agudeza visual de 20/20 antes de

la operación con su mejor corrección.

Los astigmatismos tratados en la muestra de estudio se encontraban en su gran

mayoría por debajo de 1 D bien sea a favor o en contra de la regla, tal y como muestran

las desviaciones estándar de 0.9 D en la horizontal y 0.7 D en la vertical de la figura de

doble ángulo. Éstas desviaciones se redujeron a 0.24 D y 0.27 D respectivamente,

mostrando los resultados del astigmatismo una ligera tendencia a la hipocorrección del

astigmatismo. Es importante resaltar que pese a incluir el análisis de astigmatismo

dentro del artículo, tan solo 46 ojos presentaban astigmatismos bajos por lo que estos

resultados deben ser interpretados con precaución antes de tomar cualquier tipo de

decisión clínica como la creación de un nomograma para la corrección del

astigmatismo.

El segundo de los artículos se centra en la evaluación de la biomecánica corneal

en ojos operados de SMILE. La biomecánica corneal se encuentra relacionada con los

posibles resultados refractivos obtenidos tras la cirugía, de tal manera que la inclusión

de parámetros biomecánicos en los nomogramas de ajuste del error refractivo a tratar

podría ayudar a mejorar la predictibilidad y eficacia del tratamiento. No obstante, la

medida clínica de la biomecánica corneal no se encuentra lo suficientemente madura

como para llevar a cabo este tipo de trabajo de investigación, por lo que el siguiente

paso en esta Tesis doctoral se centró en el estudio del cambio de parámetros del Corvis

ST tras SMILE.

Los resultados mostraron un menor tiempo en alcanzar la primera aplanación

(AT1) y un mayor tiempo en obtener la segunda (AT2), lo que significa que tras la

cirugía SMILE la córnea se aplana con una mayor velocidad y tarda más en volver a su

estado natural. No obstante, no se encontraron diferencias significativas entre los

cambios relativos de tiempos en la primera y segunda aplanación (AT1 – AT1’ vs AT2

– AT2’). De igual forma, encontramos que la amplitud máxima de deformación se

incrementa tras SMILE (DA’ > DA). Cabe destacar en la comparativa de grupos de

baja, media y alta miopía que para los tres parámetros de Corvis ST incluidos en el

análisis (AT1, AT2 y DA) se produce un mayor cambio debido a la cirugía con el

incremento del error refractivo tratado. Este hallazgo derivó en la necesidad de realizar

una comparativa entre grupos considerando el posible efecto que podría tener sobre

cada uno de los parámetros el cambio de espesor corneal que se produce tras la retirada


95

del lentículo. En esta segunda comparativa entre los tres grupos de miopía,

considerando el cambio relativo en cada uno de los parámetros en función del cambio

relativo de espesor corneal, no se encontraron diferencias significativas lo cual podría

sugerir que no existen cambios en la rigidez corneal para los diferentes niveles de error

refractivo tratado más allá de los debidos a la presencia de un mayor o menor volumen

de tejido.

Basándonos en estos hallazgos, otra de las cuestiones importantes a resolver fue

si, al igual que ocurría con la ausencia de diferencias entre grupos de refracción más allá

de las explicadas por el espesor corneal, podría darse el caso que las diferencias entre

valores preoperatorios y posoperatorios fuesen debidas exclusivamente al cambio de

espesor corneal. Para resolver esta cuestión se planteó un modelo basado en la amplitud

de deformación máxima cuya hipótesis nula era que la amplitud de deformación

máxima tras la operación (DA’) podía considerarse como la suma entre la amplitud de

deformación máxima antes de la operación más la cantidad de espesor corneal

eliminado (DA+CCT-CCT’). Tras evaluar las diferencias en las medias de amplitud de

deformación posoperatoria y preoperatoria más el espesor eliminado, encontramos que

no podíamos rechazar la hipótesis nula, siendo la amplitud de deformación tras la

cirugía totalmente explicable por la eliminación del espesor corneal y que tras sumar ese

espesor corneal eliminado a la amplitud de deformación preoperatoria, ésta se igualaba

a los valores posoperatorios (DA’ = DA + CCT – CCT’).

El artículo anterior denota la necesidad de emplear nuevos parámetros cuya

variación tras cirugía no se encuentre correlacionada con el cambio de espesor corneal o

que las comparativas entre técnicas se realicen con la inclusión de los cambios relativos

en los parámetros anteriores en función del espesor eliminado. En Septiembre de 2016,

el software comercial del Corvis ST sufrió una importante actualización con la inclusión

de hasta cinco nuevos parámetros que no habían sido considerados en nuestro estudio

anterior. Entre estos parámetros se encuentran la presión intraocular biomecanicamente

corregida (bIOP), el ratio de amplitud de deformación (DAR), el parámetros de rigidez

en la primera aplanación (SP-A1), la inversa del radio integrado en la máxima

concavidad (IR), y el índice de biomecánica corneal (CBI). Dentro de la versión del

software exclusiva con motivos de investigación, recurrimos de igual forma a la medida

de la densitometría dinámica, siendo éste el primer trabajo publicado que recoge

resultados acerca de este parámetro.

Los resultados demostraron que se producían cambios significativos en la bIOP

debido a la cirugía, aunque estos cambios eran menores que en el caso de la medida

convencional de la presión intraocular (IOP) de versiones anteriores del software. Para

sujetos cuya presión intraocular fue medida en la misma franja horaria, las diferencias

preoperatorias y postoperatorias para la IOP fueron de 2.24 ± 1.26 mm Hg, mientras que

______________________________________________________________________________________________

96

para la IOPb fueron de 0.57 ± 1.27 mmHg, demostrando este último ser un mejor

estimador de la presión intraocular en pacientes operados de SMILE. Además, mientras

que el cambio relativo IOP por la cirugía se correlacionó con el espesor eliminado, esta

correlación no se presentó para el parámetro bIOP.

Todos los parámetros del software comercial incluidos en el estudio cambiaron

de manera significativa tras SMILE. No obstante, cabe destacar que tan solo el

parámetro de rigidez en la primera aplanación (SP-A1) no mostró correlaciones

significativas con la variación del espesor corneal, siendo éste el primer parámetro que

presenta este comportamiento de entre los evaluados de manera previa en nuestras

investigaciones. El parámetro SP-A1 disminuyó de 148.95 ± 12.94 a 142.48 ± 17.31

mmHg/mm indicando que la rigidez de la córnea disminuye tras SMILE con

independencia del espesor eliminado.

Este tercer trabajo recoge el análisis, por primera vez en la literatura científica,

del parámetro de densitometría corneal dinámica, tan solo disponible con motivos de

investigación. Nuestros resultados mostraron que la densitometría máxima alcanzada

durante el pulso de aire y la correspondiente a cada una de las etapas medidas con el

Corvis ST (primera y segunda aplanación, y en el momento de la máxima concavidad)

no diferían entre los valores preoperatorios y posoperatorios con respecto a la media de

los ojos incluidos en el análisis. Se encontraron diferencias entre la densitometría media

alcanzada en cada una de estas etapas, siendo por ejemplo mayor en la segunda que en

la primera aplanación (5.11 ± 1.84). Además, un nuevo signo visible a través del video

de movimiento de la córnea con el pulso de aire fue correlacionado con la cirugía

SMILE. Este signo se corresponde con una línea inclinada brillante que aparece en los

extremos de los dos picos de la córnea cuando esta se encuentra en la fase de máxima

concavidad, esta línea se desplaza desde ambos máximos hacia la periferia conforme la

córnea vuelve a su posición inicial. Especialmente interesante fue que esta línea

apareció en un mayor número de ojos tras la cirugía SMILE. Mientras que 48.8% de los

ojos mostraron este signo antes de la operación, tras la operación el signo apareció en el

72.1% de los casos, siendo el incremento significativo.


97

CAPÍTULO 4 CONCLUSIONES Y LÍNEAS FUTURAS

______________________________________________________________________________________________

98


99

4.1 Cumplimiento de objetivos

La presente Tesis doctoral finaliza con un repaso al cumplimiento de los objetivos

principales marcados al principio del proyecto. En esta sección haremos un repaso a los

objetivos y describiremos como cada uno de ellos ha sido cubierto a través de cada una de las

tres publicaciones incluidas en el compendio.

Los objetivos planteados al inicio de la Tesis doctoral incluyen:

1. Analizar, en función del error refractivo tratado, los resultados de eficacia, seguridad y

predictibilidad de la técnica SMILE para los primeros casos llevados a cabo por un

cirujano con experiencia en cirugía refractiva, pero no experimentado en la técnica.

2. Evaluar los cambios en los parámetros más reproducibles del Corvis ST en función del

error refractivo tratado y analizar las variaciones de estos parámetros en función del

espesor corneal retirado.

3. Examinar los nuevos parámetros introducidos en la versión del software 1.3r1469

(Septiembre, 2016), los cambios que se producen tras SMILE y la dependencia de

estos parámetros frente a variables de confusión como el espesor corneal.

4. Presentar una nueva hipótesis sobre la densitometría dinámica y su posible relación

con la biomecánica e hidratación corneal.

El primer artículo “Short-term outcomes of small incision lenticule extraction (SMILE)

for low, médium, and high myopia” cubre el primer objetivo de la Tesis. En dicho trabajo se

evaluaron los resultados de seguridad, eficacia y predictibilidad de SMILE a los 6 meses tras la

operación para los primeros 71 casos abordados por el cirujano, autor de la presente Tesis

doctoral. La principal conclusión obtenida de este trabajo es que SMILE cumple los tres

criterios de seguridad, eficacia y predictibilidad inclusive durante la curva de aprendizaje del

cirujano. La novedad de este trabajo frente a trabajos similares previamente publicados es el

análisis en función del error refractivo a tratar, concluyendo que para altas miopías por encima

de -5.00 D es necesario el reajuste del nomograma con el fin de evitar una ligera hipocorrección

de la miopía de -0.50 D.

El segundo artículo “Corneal thickness after SMILE afects Scheimpflug-based dynamic

tonometry” justifica el cumplimiento del objetivo 2. En este trabajo se analizan los cambios en

los parámetros más reproducibles, según la literatura científica previa, del Corvis ST debidos a

la cirugía refractiva SMILE. Todos los parámetros mostraron cambios significativos tras la

cirugía. No obstante, en nuestro trabajo demostramos que todos estos cambios son explicados

______________________________________________________________________________________________

100

por el espesor de tejido eliminado para corregir la miopía, e incluso que si se adiciona este

espesor a la amplitud de deformación máxima preoperatoria, los resultados ofrecidos por el

instrumento en torno a este parámetro son iguales a los posoperatorios.

Los objetivos 3 y 4 se cumplen mediante el tercer y último artículo, “New parameters

for evaluating corneal biomechanics and intraocular pressure after SMILE by Scheimpflug-

based dynamic tonometry”. Este último trabajo supone una continuación del segundo artículo,

centrándonos en la problemática del espesor corneal como factor de confusión a la hora de

establecer los cambios de rigidez de la córnea debidos al procedimiento. El nuevo índice de la

medida de la presión intraocular biomecánicamente corregida demostró un mejor acuerdo con la

presión intraocular preoperatoria que la medida convencional ofrecida por el instrumento. Todos

los nuevos parámetros biomecánicos mostraron correlación con el cambio de espesor debido al

procedimiento, menos el nuevo índice de rigidez corneal que no mostró dicha correlación. El

nuevo parámetro de densitometría dinámica ofreció interesantes resultados en torno a la

diferencia de densitometría en las tres etapas principales (primera y segunda aplanación, y

máxima deformación) y un nuevo signo que podría estar relacionado con la biomecánica o

hidratación corneal, el cual incrementó su prevalencia tras la cirugía.

4.2 Aportaciones realizadas y líneas futuras de investigación.

Los resultados obtenidos en la presente Tesis indican que un cirujano experimentado en

cirugía refractiva podría alcanzar buenos resultados de seguridad, eficacia y predictibilidad,

incluso durante la curva de aprendizaje. Sin embargo, no podemos afirmar que esto sea

extrapolable para todos los cirujanos, ya que los resultados pertenecen únicamente a los

alcanzados por el autor de esta Tesis. Este trabajo también pone de manifiesto que el

nomograma puede necesitar de un reajuste en altas miopías. En relación a esta necesidad, se

plantea como posible línea futura de investigación el desarrollo de dicho nomograma cubriendo

variables de personalización como la edad, error refractivo, e inclusive datos de biomecánica

corneal, una vez que los parámetros del Corvis ST demuestren su aplicabilidad clínica en este

aspecto.

Esta Tesis doctoral ha servido para poner de manifiesto la necesidad de optimizar los

parámetros del Corvis ST. El propósito de esta optimización es aislar los posibles cambios en la

biomecánica corneal debido a variaciones en la rigidez del tejido, de aquellos producidos por la

modificación en el espesor corneal. La inclusión de los parámetros del Corvis ST en un

nomograma de corrección del error refractivo con la diferenciación anterior supondría que las

variables “error refractivo” y el “parámetro de biomecánica” no estarían correlacionadas entre si


101

posibilitando el desarrollo de nomogramas a través de regresiones múltiples. Además,

planteamos como necesarios para la comparativa entre técnicas de cirugía refractiva, los

siguientes dos requerimientos: (1) La inclusión de la diferencia entre valores preoperatorios y

posoperatorios para estudios comparativos y (2) el empleo de índices basados en ratios que

muestren el cambio en cualquiera de los parámetros del Corvis ST en función del espesor

corneal eliminado. El primero de los requerimientos se hace necesario para descartar la presión

intraocular como posible variable de confusión, siempre y cuando, el paciente haya cesado

cualquier tipo de tratamiento que pueda influir en la presión intraocular y los datos sean

tomados durante la misma franja horaria durante el examen preoperatorio y posoperatorio. El

segundo de los puntos es necesario para descartar el espesor corneal como variable de confusión

ya que las diferentes técnicas de cirugía refractiva láser pueden consumir distinto volumen de

tejido para una misma corrección del error refractivo.

Dentro de los nuevos parámetros del Corvis ST, el nuevo índice de rigidez corneal en la

primera aplanación es el único cuya variación no ha demostrado correlación con la variación de

espesor por lo que podría ser utilizado sin necesidad de aplicar el ratio anteriormente

mencionado. No obstante, nuestro estudio es el primero en incluir información relevante a este

parámetro en cirugía refractiva por lo que se requieren más estudios para determinar su validez

cínica. Además, hemos presentado hallazgos especialmente interesantes en relación a la

densitometría corneal dinámica, un parámetro nunca antes reportado en la literatura científica.

Las diferencias entre su valor para las distintas fases de aplanación, el signo de la franja

luminosa que se desplaza hacia la periferia con el pulso de aire, y el incremento de la

prevalencia de este signo tras SMILE deben ser estudiados en profundidad en trabajos futuros y

en condiciones experimentalmente controladas. Esto confirmaría si nuestra hipótesis que

relaciona este signo con la hidratación corneal podría ser cierta.

______________________________________________________________________________________________

102


103

BIBLIOGRAFÍA

______________________________________________________________________________________________

104


105

Abass, A. et al., 2015. Transverse depth-dependent changes in corneal collagen lamellar orientation and distribution. Journal of The Royal Society Interface, 12, p.20140717.

Agca, A., Ozgurhan, E.B., Demirok, A., et al., 2014. Comparison of corneal hysteresis and corneal resistance factor after small incision lenticule extraction and femtosecond laser-assisted LASIK: A prospective fellow eye study. Contact Lens and Anterior Eye., 37(2), pp.77–80.

Agca, A., Ozgurhan, E.B., Yildirim, Y., et al., 2014. Corneal backscatter analysis by in vivo confocal microscopy: Fellow eye comparison of small incision lenticule extraction and femtosecond laser-assisted LASIK. Journal of Ophthalmology, 2014(Article ID 265012), p.2-8.

Alnawaiseh, M. et al., 2016. Corneal densitometry as a novel technique for monitoring amiodarone therapy. Ophthalmology, 123(11), pp.2294–2299.

Ang, M., Tan, D. & Mehta, J.S., 2012. Small incision lenticule extraction (SMILE) versus laser in-situ keratomileusis (LASIK): study protocol for a randomized, non-inferiority trial. Trials, 13, p.75.


Bak-Nielsen, S. et al., 2015. Repeatability, reproducibility, and age dependency of dynamic scheimpflug-based pneumotonometer and its correlation with a dynamic bidirectional pneumotonometry device. Cornea, 34(1), pp.71–77.

Benoit, A., Latour, G. & Allain, J., 2016. Simultaneous microstructural and mechanical characterization of human corneas at increasing pressure. Journal of the Mechanical Behavior of Biomedical Materials, 60, p.93-105


Carr, J.D. et al., 2001. Laser in situ keratomileusis: surgical technique. Ophthalmology clinics of North America, 14(2), p.285–94, vii.

Chan, T.C.Y. et al., 2016. Effect of location of opening incision on astigmatic correction after small-incision lenticule extraction. Scientific Reports, 6(September), p.24 October 2016, Article number 35881.

Chansue, E. et al., 2015. Efficacy, predictability and safety of small incision lenticule extraction (SMILE). Eye and vision, 2, p.14.

Chen, M., Yu, M. & Dai, J., 2016. Comparison of biomechanical effects of small incision lenticule extraction and laser-assisted subepithelial keratomileusis. Acta Ophthalmologica, 94(7), pp.e586–e591.


Corcoran, K.J., 2015. Macroeconomic landscape of refractive surgery in the United States. Current Opinion Ophthalmology, 26(4), pp.249–54.

Cynthia Roberts, 2016. Two novel stiffness parameters for the Corvis® ST. In XXXIV Congress of the ESCRS. Copenhagen. Available at: http://www.corneal-biomechanics.de

______________________________________________________________________________________________

106

David, R. et al., 1992. Diurnal intraocular pressure variations: an analysis of 690 diurnal curves. British Journal of Ophthalmology, 76(5), pp.280–283.

Donate, D. & Thaëron, R., 2016. Lower Energy Levels Improve Visual Recovery in Small Incision Lenticule Extraction (SMILE). Journal of Refractive Surgery, 32(9), pp.636–642.

Dou, R. et al., 2015. Comparison of corneal biomechanical characteristics after surface ablation refractive surgery and novel lamellar refractive surgery. Cornea, 34(11), pp.1441–1446.

Dupont, W.D. & Plummer, W.D., 1990. Power and sample size calculations. A review and computer program. Controlled clinical trials, 11(2), pp.116–28.

El-Massry, A.A. et al., 2015. Contralateral eye comparison between femtosecond small incision intrastromal lenticule extraction at depths of 100 and 160 mum. Cornea, 34(10), pp.1272–1275.

Elsheikh, A. et al., 2007. Assessment of corneal biomechanical properties and their variation with age. Current Eye Research, 32(1), pp.11–19.

Erie, J.C. et al., 2005. Recovery of corneal subbasal nerve density after PRK and LASIK. American journal of ophthalmology, 140(6), pp.1059–1064.

Eydelman, M.B. et al., 2006. Standardized analyses of correction of astigmatism by laser systems that reshape the cornea. Journal of Refractive Surgery, 22(1), pp.81–95.

Fernández, J. et al., 2016. Assessment of corneal biomechanical changes after one month of SMILE using Scheimpflug-based dynamic tonometry. In XXXIV Congress ESCRS.

Fernández, J., Rodríguez-Vallejo, M., Martínez, J., Tauste, A. & Piñero, D.P., 2016. Corneal thickness after SMILE affects scheimpflug-based dynamic tonometry. Journal of Refractive Surgery, 32(12), pp.821–828.

Fernández, J., Rodríguez-Vallejo, M., Martínez, J., Tauste, A., Salvestrini, P., et al., 2016. New parameters for evaluating corneal biomechanics and intraocular pressure after SMILE by Scheimpflug-Based Dynamic Tonometry. Journal of Cataract & Refractive Surgery, p.[Accepted for publish].

Fernández, J., 2015. New Technique of Refractive Surgery: ReLEx SMILE. In ASCRS. Film Category: Refractive/Cornea Surgery. San Diego (EEUU). Available at: https://ascrs.confex.com/ascrs/15am/webprogram/Paper16063.html

Fernández, J., Valero, A., et al., 2016. Short-term outcomes of small-incision lenticule extraction (SMILE) for low, medium, and high myopia. European Journal of Ophthalmology, Jul 18(0).

Fernández, J. & Martínez, J., 2015. Entendiendo las medidas de biomecánica corneal. In Óptica para el cirujano faco-refractivo. Editores Ruiz Mesa & Tañá Rivero. Elservier. p. 141.

Frings, A. et al., 2015. Effects of laser in situ keratomileusis (LASIK) on corneal biomechanical measurements with the Corvis ST tonometer. Clinical Ophthalmology, (9), pp.305–311.

Gambato, C. et al., 2005. Mitomycin C modulation of corneal wound healing after photorefractive keratectomy in highly myopic eyes. Ophthalmology, 112(2), pp.208–218.

Ganesh, S. & Gupta, R., 2014. Comparison of visual and refractive outcomes following femtosecond laser-assisted lasik with smile in patients with myopia or myopic


107

astigmatism. Journal of Refractive Surgery, 30(9), pp.590–6. Gao, S. et al., 2014. Early changes in ocular surface and tear inflammatory mediators

after small-incision lenticule extraction and femtosecond laser-assisted laser in situ keratomileusis. PloS one, 9(9), p.e107370.

Ghadhfan, F., Al-Rajhi, A. & Wagoner, M.D., 2007. Laser in situ keratomileusis versus surface ablation: visual outcomes and complications. Journal of Cataract and Refractive Surgery, 33(12), pp.2041–8.

Glass, D.H. et al., 2008. A viscoelastic biomechanical model of the cornea describing the effect of viscosity and elasticity on hysteresis. Investigative Ophthalmology and Visual Science, 49(9), pp.3919–3926.

Güell, J.L. et al., 2015. SMILE Procedures With Four Different Cap Thicknesses for the Correction of Myopia and Myopic Astigmatism. Journal of Refractive Surgery, 31(9), pp.580–585.

Gyldenkerne, A., Ivarsen, A. & Hjortdal, J.Ø., 2015. Comparison of corneal shape changes and aberrations induced by FS-LASIK and SMILE for Myopia. Journal of Refractive Surgery, 31(4), pp.160–165.

Hansen, R.S. et al., 2016. Small-incision lenticule extraction (SMILE): outcomes of 722 eyes treated for myopia and myopic astigmatism. Graefe’s Archive for Clinical and Experimental Ophthalmology, 254(2), pp.399–405.

Hassan, Z. et al., 2014. Examination of ocular biomechanics with a new Scheimpflug technology after corneal refractive surgery. Contact Lens and Anterior Eye., 37(5), pp.337–341.

He, M. et al., 2016. Comparison of Two Cap Thickness in Small Incision Lenticule Extraction: 100μm versus 160μm. Plos One, 11(9), p.e0163259.

He, M., Huang, W. & Zhong, X., 2015. Central corneal sensitivity after small incision lenticule extraction versus femtosecond laser-assisted LASIK for myopia: a meta-analysis of comparative studies. BMC Ophthalmology, 15(1), p.141.

Hjortdal, J.Ø. et al., 2012. Predictors for the outcome of small-incision lenticule extraction for Myopia. Journal of Refractive Surgery, 28(12), pp.865–71.

Holladay, J.T., 2003. Optical quality and refractive surgery. International ophthalmology clinics, 43(2), pp.119–136.

Hon, Y. & Lam, A.K., 2013. Corneal deformation measurement using Scheimpflug noncontact tonometry. Optometry and Vision Science, 90(1), pp.e1-8.

Hueso, E. et al., 2016. Seguridad, eficacia y predictibilidad de SMILE para distintos grados de miopía. In XXIV Congreso Internacinonal de Optometría, Contactología y Óptica Oftálmica.

Ishii, R. et al., 2015. Influence of femtosecond lenticule extraction and small incision lenticule extraction on corneal nerve density and ocular surface: a 1-year prospective, confocal, microscopic study. J Refract Surg, 31(1), pp.10–15.

Ito, M. et al., 1996. Picosecond laser in situ keratomileusis with a 1053-nm Nd:YLF laser. Journal of Refractive Surgery, 12(6), pp.721–8.

Ivarsen, A., Asp, S. & Hjortdal, J., 2014. Safety and complications of more than 1500 small-incision lenticule extraction procedures. Ophthalmology, 121(4), pp.822–828.

Ivarsen, A. & Hjortdal, J., 2014. Correction of myopic astigmatism with small incision lenticule extraction. Journal of Refractive Surgery, 30(4), pp.240–7.

Joda, A.A. et al., 2016. Development and validation of a correction equation for Corvis

______________________________________________________________________________________________

108

tonometry. Computer methods in biomechanics and biomedical engineering, 19(9), pp.943–53.

Jung, J.W. et al., 2015. Comparison of measurements and clinical outcomes after wavefront-guided LASEK between iDesign and WaveScan. Journal of Refractive Surgery, 31(6), pp.398–405.

Kamiya, K. et al., 2012. Early clinical outcomes, including efficacy and endothelial cell loss, of refractive lenticule extraction using a 500 kHz femtosecond laser to correct myopia. Journal of Cataract and Refractive Surgery, 38(11), pp.1996–2002.

Kamiya, K. et al., 2015. Effect of femtosecond laser setting on visual performance after small-incision lenticule extraction for myopia. British Journal of Ophthalmology, 99(10):1381-7

Kamiya, K. et al., 2014. Intraindividual comparison of changes in corneal biomechanical parameters after femtosecond lenticule extraction and small-incision lenticule extraction. J Cataract Refract Surg, 40(6), pp.963–970.

Karakosta, A. et al., 2012. Choice of analytic approach for eye-specific outcomes: one eye or two? American Journal of Ophthalmology, 153(3), p.571–579.e1.

Kemp, J.R. et al., 1999. Diurnal fluctuations in corneal topography 10 years after radial keratotomy in the Prospective Evaluation of Radial Keratotomy Study. Journal of Cataract and Refractive Surgery, 25(7), pp.904–10.

Kim, J. et al., 2014. Efficacy, predictability, and safety of small incision lenticule extraction: 6-months prospective cohort study. BMC Ophthalmology, 14(1), p.117.

Kim, J.R. et al., 2015. One-year outcomes of small-incision lenticule extraction (SMILE): mild to moderate myopia vs. high myopia. BMC Ophthalmology, 15(1), p.59.


Kobashi, H., Kamiya, K. & Shimizu, K., 2016. Dry eye after Small Incision Lenticule Extraction and Femtosecond Laser-Assisted LASIK: Meta-Analysis. Cornea, 36(1), pp.85–91.

Krueger, R.R. et al., 1996. Ultrastructure of picosecond laser intrastromal photodisruption. Journal of Refractive Surgery, 12(5), pp.607–12.

Kunert, K.S. et al., 2013. Vector analysis of myopic astigmatism corrected by femtosecond refractive lenticule extraction. Journal of Cataract and Refractive Surgery, 39(5), pp.759–769.

Kurtz, R.M. et al., 1998. Lamellar refractive surgery with scanned intrastromal picosecond and femtosecond laser pulses in animal eyes. Journal of Refractive Surgery, 14(5), pp.541–8.

Lanza, M. et al., 2014. Evaluation of corneal deformation analyzed with Scheimpflug based device in healthy eyes and diseased ones. Biomed Research International, 2014, p.748671.

Leccisotti, A. et al., 2016. Changes in ocular biomechanics after femtosecond laser creation of a laser in situ keratomileusis flap. Journal of Cataract and Refractive Surgery,, 42(1), pp.127–131.

Lee, R. et al., 2016. Assessment of corneal biomechanical parameters in myopes and emmetropes using the Corvis ST. Clinical and Experimental Optometry, 99, pp.157–162.

Leonard, D.W. & Meek, K.M., 1997. Refractive indices of the collagen fibrils and extrafibrillar material of the corneal stroma. Biophysical Journal, 72(3), pp.1382–


109

1387. Li, M. et al., 2013. Confocal comparison of corneal reinnervation after small incision

lenticule extraction (SMILE) and femtosecond laser in situ keratomileusis (FS-LASIK). PloS one, 8(12), p.e81435.

Li, M. et al., 2014. Mild decentration measured by a Scheimpflug camera and its impact on visual quality following SMILE in the early learning curve. Investigative Opthalmology and Vision Science, 55(6), pp.3886–3892.

Li, X., Wang, Y. & Dou, R., 2015. Aberration compensation between anterior and posterior corneal surfaces after Small incision lenticule extraction and Femtosecond laser-assisted laser in-situ keratomileusis. Ophthalmic and Physiological Optics, 35, pp.540–551.

Lin, F., Xu, Y. & Yang, Y., 2014. Comparison of the visual results after SMILE and femtosecond laser-assisted LASIK for myopia. Journal of Refractive Surgery, 30(4), pp.248–54.

Liu, M., Chen, Y., et al., 2016. Clinical outcomes after SMILE and Femtosecond Laser-Assisted LASIK for myopia and myopic astigmatism. Cornea, 35(2), pp.210–216.

Liu, M., Zhou, Y., et al., 2016. Comparison of 120- and 140-microns SMILE cap thickness results in eyes with thick corneas. Cornea, 35(10), pp.1308–1314.

Liu, Y.-C. et al., 2015. Effect of Intraoperative Corneal Stromal Pocket Irrigation in Small Incision Lenticule Extraction. BioMed Research International, 928608, pp.1–9.

Liu, Y., Pujara, T. & Mehta, J.S., 2016. Surgical instruments for small incision lenticule extraction (SMILE). Expert Review of Ophthalmology, 11(3), pp.171–172.

Luo, J. et al., 2015. Quantitative Analysis of Microdistortions in Bowman’s Layer Using Optical Coherence Tomography After SMILE Among Different Myopic Corrections. Journal of Refractive Surgery, 31(2), pp.104–109.

Marfurt, C.F. et al., 2010. Anatomy of the human corneal innervation. Experimental Eye Research, 90(4), pp.478–492.

Mastropasqua, L. et al., 2014. Evaluation of corneal biomechanical properties modification after Small Incision Lenticule Extraction using Scheimpflug-based noncontact tonometer. Biomed Research International, 2014, p.290619.

Morishige, N. et al., 2011. Three-dimensional analysis of collagen lamellae in the anterior stroma of the human cornea visualized by second harmonic generation imaging microscopy. Investigative Ophthalmology and Vision Science, 52(2), pp.911–915.

Nemeth, G. et al., 2013. Repeatability of ocular biomechanical data measurements with a Scheimpflug-based noncontact device on normal corneas. Journal of Refractive Surgery, 29, pp.558–63.

Ní Dhubhghaill, S. et al., 2014. Normative values for corneal densitometry analysis by scheimpflug optical assessment. Investigative Ophthalmology and Vision Science, 55(1), p.162.

Osman, I.M. et al., 2016. Corneal biomechanical changes in eyes with small incision lenticule extraction and laser assisted in situ keratomileusis. BMC Ophthalmology, 16(1), p.123.

Pallikaris, I.G. et al., 2005. Epi-LASIK: Preliminary clinical results of an alternative surface ablation procedure. Journal of Cataract and Refractive Surgery, 31(5), pp.879–885.

Pedersen, I.B. et al., 2014. Corneal biomechanical properties after LASIK, ReLEx flex,

______________________________________________________________________________________________

110

and ReLEx smile by Scheimpflug-based dynamic tonometry. Graefes Archives Clinical and Experimental Ophthalmology, 252(8), pp.1329–35.

Piñero-Llorens, D.P., Murueta-Goyena Larrañaga, A. & Hannekend, L., 2016. Visual outcomes and complications of small-incision lenticule extraction: a review. Expert Review of Ophthalmology, 11(1), pp. 59-75.

Piñero, D.P. & Alcón, N., 2015. Corneal biomechanics: a review. Clinical and Experimental Optometry, 98, pp.107–116.

Pradhan, K.R. et al., 2016. Quality control outcomes analysis of small-incision lenticule extraction for myopia by a novice surgeon at the first refractive surgery unit in Nepal during the first 2 years of operation. Journal of Cataract and Refractive Surgery, 42(2), pp.267–274.

Quantock, A.J. et al., 2015. From nano to macro: Studying the hierarchical structure of the corneal extracellular matrix. Experimental Eye Research, 133, pp.81–99.

Randleman, J.B. et al., 2008. Depth-dependent cohesive tensile strength in human donor corneas: implications for refractive surgery. Journal of Refractive Surgery, 24(1), pp.S85–S89.

Randleman, J.B., 2016. Ectasia after corneal refractive surgery: Nothing to SMILE about. Journal of Refractive Surgery, 32(7), pp.434–435.

Ratkay-Traub, I. et al., 2003. First clinical results with the femtosecond neodynium-glass laser in refractive surgery. Journal of Refractive Surgery, 19(2), pp.94–103.

Reinstein, D., Carp, G.I., et al., 2014. Outcomes of small incision lenticule extraction (SMILE) in low myopia. Journal of Refractive Surgery, 30(12), pp.812–818.


Reinstein, D.Z. et al., 2010. Accuracy and reproducibility of artemis central flap thickness and visual outcomes of LASIK with the Carl Zeiss Meditec VisuMax femtosecond laser and MEL 80 excimer laser platforms. Journal of Refractive Surgery, 26(2), pp.107–19.

Reinstein, D.Z., Archer, T.J. & Randleman, J.B., 2013. Mathematical model to compare the relative tensile strength of the cornea after PRK, LASIK, and small incision lenticule extraction. Journal of Refractive Surgery, 29(7), pp.454–60.

Roberts, C.J., 2014. Concepts and misconceptions in corneal biomechanics. Journal of Cataract and Refractive Surgery, 40(6), pp.862–9.

Rodríguez-Vallejo, M. et al., 2015. Designing a new test for contrast sensitivity function measurement with iPad. Journal of Optometry, 8, pp.101–108.

Roy, A.S. & Dupps, W.J., 2009. Effects of altered corneal stiffness on native and postoperative LASIK corneal biomechanical behavior: A whole-eye finite element analysis. Journal of Refractive Surgery, 25(10), pp.875–887.

Schwiegerling, J., 2004. Wavefront-guided LASIK. Optics and Photonics News, 15(2), p.26.

Sefat, S.M.M. et al., 2016. Evaluation of changes in human corneas after femtosecond laser-assisted LASIK and Small-Incision Lenticule Extraction (SMILE) using non-contact tonometry and ultra-high-speed camera (Corvis ST). Current Eye Research, 41(7), pp.917–922.

Sekundo, W. et al., 2008. First efficacy and safety study of femtosecond lenticule extraction for the correction of myopia. Six-month results. Journal of Cataract and


111

Refractive Surgery, 34, pp.1513–1520. Sekundo, W., Kunert, K.S. & Blum, M., 2011. Small incision corneal refractive surgery

using the small incision lenticule extraction (SMILE) procedure for the correction of myopia and myopic astigmatism: results of a 6 month prospective study. British Journal of Ophthalmology, 95(3), pp.335–339.

Shah, R. & Shah, S., 2011. Effect of scanning patterns on the results of femtosecond laser lenticule extraction refractive surgery. Journal of Cataract and Refractive Surgery, 37(9), pp.1636–1647.

Shah, R., Shah, S. & Sengupta, S., 2011. Results of small incision lenticule extraction: All-in-one femtosecond laser refractive surgery. Journal of Cataract and Refractive Surgery, 37(1), pp.127–137.


Shen, Z. et al., 2016. Small Incision Lenticule Extraction (SMILE) versus Femtosecond Laser-Assisted In Situ Keratomileusis (FS-LASIK) for Myopia: A Systematic Review and Meta-Analysis. Plos One, 11(7), p.e0158176.

Sher, N.A. et al., 1991. The use of the 193-nm excimer laser for myopic photorefractive keratectomy in sighted eyes. A multicenter study. Archives of Ophthalmology, 109(11), pp.1525–30.


Shetty, R. et al., 2016. Intra-operative cap repositioning in Small Incision Lenticule Extraction (SMILE) for enhanced visual recovery. Current Eye Research, 3683(May), pp.1–7.

Simonini, I., Angelillo, M. & Pandolfi, A., 2016. Theoretical and numerical analysis of the corneal air puff test. Journal of the Mechanics and Physics of Solids, pp.1–17.

Sinha Roy, A., Dupps, W.J. & Roberts, C.J., 2014. Comparison of biomechanical effects of small-incision lenticule extraction and laser in situ keratomileusis: Finite-element analysis. Journal of Cataract and Refractive Surgery, 40(6), pp.971–980.

Smedowski, A. et al., 2014. Comparison of three intraocular pressure measurement methods including biomechanical properties of the cornea. Investigative Ophthalmology and Vision Science, 55(2), pp.666–673.

Soong, H.K. & Malta, J.B., 2009. Femtosecond lasers in ophthalmology. American Journal of Ophthalmology, 147(2), p.189–197.e2.

Stern, D. et al., 1989. Corneal ablation by nanosecond, picosecond, and femtosecond lasers at 532 and 625 nm. Archives of Ophthalmology, 107(4), pp.587–92.

Vestergaard, A. et al., 2012. Small-incision lenticule extraction for moderate to high myopia: Predictability, safety, and patient satisfaction. Journal of Cataract and Refractive Surgery, 38(11), pp.2003–10.

Vestergaard, A.H. et al., 2014a. Central corneal sublayer pachymetry and biomechanical properties after refractive femtosecond lenticule extraction. Journal of Refractive Surgery, 30(2), pp.102–8.

Vestergaard, A.H. et al., 2014b. Efficacy, safety, predictability, contrast sensitivity, and aberrations after femtosecond laser lenticule extraction. Journal of Cataract and Refractive Surgery, 40(3), pp.403–11.

______________________________________________________________________________________________

112

Vinciguerra, R., Ambrósio, R., et al., 2016. Detection of keratoconus with a new biomechanical index. Journal of Refractive Surgery, 32(12), pp.803–810.

Vinciguerra, R., Elsheikh, A., et al., 2016. Influence of pachymetry and intraocular pressure on dynamic corneal response parameters in healthy patients. Journal of Refractive Surgery, 32(8), pp.550–561.

Vogel, A. et al., 1994. Intraocular photodisruption with picosecond and nanosecond laser pulses: Tissue effects in cornea, lens, and retina. Investigative Ophthalmology and Visual Science, 35(7), pp.3032–3044.

Wang, B. et al., 2016. Comparison of the change in posterior corneal elevation and corneal biomechanical parameters after small incision lenticule extraction and femtosecond laser-assisted LASIK for high myopia correction. Contact Lens and Anterior Eye, 39(3), pp.191-196.

Wang, D. et al., 2014. Differences in the corneal biomechanical changes after SMILE and LASIK. Journal of Refractive Surgery, 30(10), pp.702–707.

Wang, J., Simpson, T.L. & Fonn, D., 2004. Objective measurements of corneal light-backscatter during corneal swelling, by optical coherence tomography. Investigative Ophthalmology and Visual Science, 45(10), pp.3493–3498.

Wang, Y. et al., 2014. Two millimeter micro incision lenticule extraction surgery with minimal invasion: a preliminary clinical report. [Zhonghua yan ke za Zhi] Chinese Journal of Ophthalmology, 50(9):671-680.

Waring, G.O. et al., 2011. Standardized graphs and terms for refractive surgery results. Journal of Refractive Surgery, 27(1), pp.7–9.

Waring, G.O., Lynn, M.J. & McDonnell, P.J., 1994. Results of the prospective evaluation of radial keratotomy (PERK) study 10 years after surgery. Archives of Ophthalmology, 112(10), pp.1298–308.

Whitford, C. et al., 2015. Biomechanical model of the human cornea: Considering shear stiffness and regional variation of collagen anisotropy and density. Journal of the Mechanical Behavior of Biomedical Materials, 42, pp.76–87.

Wong, C.W. et al., 2014. Incidence and management of suction loss in refractive lenticule extraction. Journal of Cataract and Refractive Surgery, 40(12), pp.2002–2010.

Wu, D. et al., 2014. Corneal biomechanical effects: Small-Incision Lenticule Extraction versus Femtosecond Laser-Assisted laser in situ Keratomileusis. Journal of Cataract and Refractive Surgery, 40(6), pp.954–962.

Wu, W. et al., 2016. One-year visual outcome of small incision lenticule extraction (SMILE) surgery in high myopic eyes: retrospective cohort study. BMJ open, 6(9), p.e010993.

Wu, W. & Wang, Y., 2016. Corneal Higher-Order Aberrations of the Anterior Surface, Posterior Surface, and Total Cornea After SMILE, FS-LASIK, and FLEx Surgeries. Eye & Contact Lens: Science & Clinical Practice, pp.1–8.

Wu, W. & Wang, Y., 2015. The correlation analysis between corneal biomechanical properties and the surgically induced corneal high-order aberrations after small incision lenticule extraction and femtosecond laser in situ keratomileusis. Journal of Ophthalmology, 758196, p.10.

Wu, Z. et al., 2016. Comparison of corneal biomechanics after microincision lenticule extraction and small incision lenticule extraction. British Journal of Ophthalmology, pp.1–5.


113

Xu, Y. & Yang, Y., 2015. Small-Incision Lenticule Extraction for myopia: Results of a 12-month prospective study. Optometry and Vision Science, 92(1), pp.123–131.

Yao, P. et al., 2013. Microdistortions in Bowman’s Layer Following Femtosecond Laser Small Incision Lenticule Extraction Observed by Fourier-Domain OCT. Journal of Refractive Surgery, 29(10), pp.669–675.

Y Shen et al., 2014. Comparison of corneal deformation parameters after SMILE, LASEK, and femtosecond laser-assisted LASIK. Journal of Refractive Surgery, 30(5), pp.310–8.

Ye, M. et al., 2016. SMILE and Wavefront-Guided LASIK Out-Compete Other Refractive Surgeries in Ameliorating the Induction of High-Order Aberrations in Anterior Corneal Surface. Journal of Ophthalmology, 2016, pp.1–7.

Yıldırım, Y. et al., 2016. Comparison of Changes in Corneal Biomechanical Properties after Photorefractive Keratectomy and Small Incision Lenticule Extraction. Türk Oftalmoloji Dergisi, 46(2), pp.47–51.

Yu, F. & Afifi, A., 2009. Descriptive statistics in ophthalmic research. American journal of ophthalmology, 147(3), pp.389–91.

Yu, M. et al., 2015. Comparison of Visual Quality After SMILE and LASEK for Mild to Moderate Myopia. Journal of Refractive Surgery, 31(12), pp.795–800.

Zhang, J. et al., 2016. Corneal biomechanics after small-incision lenticule extraction versus Q-value–guided femtosecond laser-assisted in situ keratomileusis. Journal of Current Ophthalmology, 28(4), pp.181–187.

Zhang, J., Wang, Y. & Chen, X., 2016. Comparison of moderate- to High-Astigmatism corrections using WaveFront – Guided Laser In Situ Keratomileusis and Small-Incision Lenticule Extraction. Cornea, 35(4), pp.523–530.

Zhang, Y. et al., 2015. Clinical Outcomes of SMILE and FS-LASIK Used to Treat Myopia: A Meta-analysis. Journal of Refractive Surgery, 32(4), pp.256–265.

Zhao, J. et al., 2015. Diffuse lamellar keratitis after small-incision lenticule extraction. Journal of Cataract and Refractive Surgery, 41(2), pp.400–407.

Zhao, Y. et al., 2015. Development of the Continuous Curvilinear Lenticulerrhexis Technique for Small Incision Lenticule Extraction. Journal of Refractive Surgery, 31(1), pp.16–21.

cirugía refractiva láser corneal smile. resultados visuales y … · 2018. 11. 15. · cirugía...

Documents