a r c h s o c e s p o f t a l m o l . 2 0 1 3;8 8(2):45–49

ARCHIVOS DE LA SOCIEDADESPAÑOLA DE OFTALMOLOGÍA

O

Cau

Ca

b

a

A

R

A

A

K

C

F

o

U

G

O

P

E

T

d

e

s�

2

2

www .e lsev ier .es /o f ta lmologia

riginal article

omparison of central corneal thickness measured withnterior segment optical coherence tomography versusltrasonic pachymetry�,��

. Lázaroa,b,∗, E.M. Hernándeza,b, D. Martíneza, P. Redondoa

Servicio de Oftalmología, Hospital Provincial de la Misericordia, Toledo, SpainInstituto Oftalmológico de Toledo, Toledo, Spain

r t i c l e i n f o

rticle history:

eceived 12 January 2012

ccepted 23 May 2012

vailable online 10 April 2013

eywords:

entral corneal thickness

ourier-domain anterior segment

ptical coherence tomography

ltrasonic pachymetry

laucoma

cular biometry

a b s t r a c t

Purpose: To compare central corneal thickness measured with anterior segment optical

coherence tomography (AS-OCT) and ultrasonic pachymetry (UP).

Method: A prospective, observational and cross-sectional study was performed. The central

corneal thickness was measured in 112 eyes of 112 consecutive patients with AS-OCT and

UP. All examinations were performed by the same examiner. The measurements obtained

between the two instruments were compared to assess the level of agreement.

Results: The average corneal thickness value obtained by AS-OCT was 531.47 ± 26.23 �m,

while that measured by UP was 532.51 ± 26.04 �m. The intraclass correlation coefficient

was 0.957 (p < .001). The Student’s t-test did not show a statistically significant difference

(1.04 ± 7.70 �m, p = .158). Finally, the Bland–Altman analysis showed a high level of agree-

ment between two methods of measurement.

Conclusions: In the context of normal corneas, there was good agreement between the mea-

surement of central corneal thickness with AS-OCT and UP, although there was a slight

overestimation of the measurement using UP (1.04 �m).

© 2012 Sociedad Española de Oftalmología. Published by Elsevier España, S.L. All rights

reserved.

Comparación del espesor corneal central medido con tomografíade coherencia óptica de segmento anterior y paquimetría ultrasónica

alabras clave:

r e s u m e n

Objetivo: Comparar el espesor corneal central medido con tomografía de coherencia óptica

spesor corneal central

omografía de coherencia óptica

e segmento anterior de dominio

spectral

de segmento anterior (OCT-SA) y paquimetría ultrasónica (PU).

Método: Se realiza un estudio prospectivo, observacional y transversal. Se midió el espesor

corneal central en 112 ojos de 112 pacientes consecutivos mediante OCT-SA y mediante

� Please cite this article as: Lázaro C, et al. Comparación del espesor corneal central medido con tomografía de coherencia óptica deegmento anterior y paquimetría ultrasónica. Arch Soc Esp Oftalmol. 2013;88:45–9.� Presented as a spoken communication at the 87th Congress of the Spanish Ophthalmological Society, Oviedo, Spain, September 21–24,011.∗ Corresponding author.

E-mail address: carloslazarogarcia@gmail.com (C. Lázaro).

173-5794/$ – see front matter © 2012 Sociedad Española de Oftalmología. Published by Elsevier España, S.L. All rights reserved.

46 a r c h s o c e s p o f t a l m o l . 2 0 1 3;8 8(2):45–49

Paquimetría ultrasónica

Glaucoma

Biometría ocular

PU. Todas las exploraciones fueron realizadas por el mismo examinador. Se compararon las

mediciones obtenidas entre ambos instrumentos para valorar el nivel de concordancia.

Resultados: El valor de espesor corneal medio obtenido mediante OCT-SA fue

531,47 ± 26,23 �m mientras que el medido mediante PU fue 532,51 ± 26,04 �m. El coeficiente

de correlación intraclase fue 0,957 (p < 0,001). En el test «t» de Student no se observó una

diferencia estadísticamente significativa (1,04 ± 7,70 �m, p = 0,158). Finalmente, el análisis

de Bland-Altman mostró un elevado nivel de concordancia entre ambos métodos de medida.

Conclusiones: En el contexto de córneas normales, la medición del espesor corneal central

con OCTSA y con PU mostraron una elevada concordancia, existiendo una leve sobreesti-

mación de la medida con ultrasonidos (1,04 �m).

© 2012 Sociedad Española de Oftalmología. Publicado por Elsevier España, S.L. Todos los

as possible in the corneal vertex and the center of the pupil.The measure was repeated on three occasions, taking a meanvalue thereof.

Introduction

As corneal thickness measuring is a relevant factor for ponder-ing intra-ocular pressure measures, it has become a crucialtest for studying glaucoma patients. Various devices in thecurrent market provide measurements of said parameter andtherefore it is necessary to determine the precision and matchbetween them.

The technologies for measuring corneal thickness includeultrasound pachymetry (UP), non-contact mirror microscopy,the Scheimpflug camera and anterior segment optical coher-ence tomography (AS-OCT).1–11 At present, UP is the referenceinstrument for measuring corneal thickness due to itsdemonstrated validity and reliability.12 However, cornealcontact could involve contagion between different patientsin addition to producing epithelial damage over the ocu-lar surface. It can also become a source of measurementerrors due to the pressure exerted over the cornea as wellas to inadequate alignment of the terminal, which mustbe positioned absolutely perpendicular to the corneal sur-face.

The utilization of a “no contact” corneal thickness mea-suring instrument would be a magnificent option providedthat the reproducibility of the measurements is verified andan adequate correlation with UP is demonstrated.

RTVue-100 (Optovue Inc., Fremont, CA, USA) is aFourier domain optic coherence tomography that performs26,000 scans to provide high resolution images with a depthof 5 �m and transversally of 15 �m. It includes an anterior seg-ment module that analyzes corneal thickness in a diameter of6 mm in 0.32 s, minimizing the possibility of errors related toinvoluntary patient movements. The objective is to comparethis instrument with UP, the standard test at this time.

Subjects, material and methods

A prospective, observational and transversal study was car-ried out, measuring central corneal thickness in 112 eyes of112 consecutive patients by means of AS-OCT (RTVue-100,software version 4.0) (Fig. 1) and by means of UP (Corneo-

TM

Gage Plus ultrasonic pachometer, Sonogage Inc., Cleveland,OH, USA). The patients were recruited from the general prac-tices of our hospital from January to April 2011, discarding allthose having any ophthalmological disease excepting slight

derechos reservados.

refraction defects. Patient consent was obtained to carry outthe tests and the principles established in the Helsinki dec-laration were followed. Only one eye of each patient wasexamined, selecting it randomly in each case. All the measurestaken with UP and AS-OCT were taken by the same examiner.

The exclusion criteria were as follows: pregnancy, use ofcontact lenses 72 h prior to the assessment, best correctedvisual acuity under 0.8 (Snellen), intraocular pressure over21 mmHg, endothelial count under 2000 cells/mm2, pathologi-cal changes in the cornea, the conjunctiva or the free palpebraledge, and ophthalmological surgery history which could affectthe ocular surface.

The measurements taken with UP were made always afterthe AS-OCT because anesthetic drops had to be adminis-tered and contact had to be made with the corneal surface,which could distort subsequent measurements with AS-OCT.Ninety seconds after administering one drop of oxybuprocaine0.4%–tetracaine 0.1% (Colircusí Anestésico Doble; Alcon CusíSA, Barcelona, Spain), five consecutive central corneal thick-ness measurements were taken, with the final UP value beingthe mean of these five measurements. AS-OCT was made withpatients adequately positioned, requesting them to fix theirgaze in a diode placed in the axis of vision. Two additionalred diodes placed on the headrest assisted the corneal imagetakes. The image captures were adjusted centering as much

Fig. 1 – Image demonstrating the central corneal thicknessmeasurement taken with anterior segment optic coherencetomography (RTVue-100, software version 4.0, OptovueInc., Fremont, CA, USA).

a r c h s o c e s p o f t a l m o l . 2 0 1 3;8 8(2):45–49 47

Table 1 – Analysis result of individual differencesbetween ultrasound pachymetry and anterior segmentoptic coherence tomography by means of the Blandand Altman method.

Bland and Altmanmethod

Lower limit 95%confidence

interval

Higher limit 95%confidence

interval

Sample size 112Arithmetical mean 1.04 −0.41 2.48Standard deviation 7.70

ddmeriicb9rs((

R

TtT55wd(b

bbda

D

Ttutn

I1oep

30

+1.96 SD

+1.96 SD

16.1

-14.1

1.0

Mean

20

10

0

Diff

eren

ce b

etw

eeen

UP

and

AS

-OC

T

-10

-20

-30

460 480 500 520 540

Mean for UP and AS-OCT

560 580 600 620

Fig. 2 – The Bland–Altman graph compares cornealthickness measurements taken with ultrasoundpachymetry (UP) and anterior segment optic coherence

Lower limit −14.06 −16.54 −11.59Higher limit 16.14 13.66 18.61

In the statistical analysis, the quantitative variables wereescribed as mean and standard deviation. In all cases theistribution of the variable was checked against theoreticalodels, comparing with the variance homogeneity hypoth-

sis. In all of these comparisons, the null hypotheses wasejected with a type I or alpha error under 0.05. The match-ng analysis between both methods was performed with thentraclass correlation coefficient. The Bland–Altman13,14 pro-edure was carried out to graphically represent the differencesetween both measurements as well as the matching limits at5%. In addition, the Levene test for variance equality was car-ied out to subsequently perform the t for Student test. For thetatistical analysis, the SPSS statistical application was usedversion 12.0; SPSS, Chicago, USA) as well as MEDCALC 11.6.1MedCalc Software, Mariakerke, Belgium).

esults

he mean age of analyzed patients was of 60 ± 6 years andhe mean spherical equivalent was −2.75 ± 2.5 D (mean ± DE).he mean corneal thickness value obtained with AS-OCT was31.47 ± 26.23 �m while the values measured with UP were of32.51 ± 26.04 �m. Accordingly, the difference between bothas of −1.04 ± 7.70 �m (p = 0.158), a statistically not significantifference. The intraclass correlation coefficient was of 0.957

p < 0.001) and the Pearson correlation was r = 0.957 (p < 0.01),oth being statistically significant.

The Bland–Altman analysis showed an important matchetween both methods, with a difference of only 1 �m betweenoth techniques, with UP being slightly higher. Up to 95% ofifferences were found between 16.1 and −14.1 �m. The fullnalysis is shown in Table 1 and Fig. 2.

iscussion

his paper shows an important match between cornealhickness measurement with UP and AS-OCT. Even so, annderestimation of AS-OCT over UP was evidenced evenhough this difference was not statistically or clinically sig-ificant.

Said results are in contrast to those published byshibazawa et al.15 who described an underestimation of

4 �m of AS-OCT vis-à-vis UP, utilizing the same device as inur paper, i.e. RTVue-100. This difference could be due to sev-ral reasons. First, the authors utilized a different ultrasoundachymetry to that used in our case. This factor could produce

tomography (AS-OCT). SD: standard deviation.

a distortion in the differences found between both techniques.In fact, in our case the mean corneal thickness observed withRTVue-100 was 531.47 ± 26.23 �m and in the Ishibazawa et al.group it was of 530.47 ± 33 �m. These results are remarkablysimilar in a control group in both studies. Accordingly, theycould be an overestimation of the UP measurements with thepachymeter utilized by said author or an underestimation byour own pachymeter.

Likewise, the paper published by Nam et al.16 does notobserve an underestimation of the corneal thickness mea-sured with AS-OCT over UP and in addition it observes anoverestimation thereof having a value of 12.8 �m. This issurprising if we take into account that the AS-OCT device uti-lized by these authors is the same used in our work eventhough the UP device is different. This supports the ideaof different measures observed between different ultrasoundpachymeters. Said study observed an overestimation of AS-OCT over the Scheimpflug camera of about 7 �m, whereasother studies observed an underestimation of up to 22 �m.15

Accordingly, we must be cautious when analyzing the resultsobserved in these studies.

The variability of measurements made with UP hasbeen emphasized in other papers, including a second studyby Nam et al. which found corneal thickness variationsafter administering topical anesthesia. These factors couldhave contributed to the variations observed between theanalyses.17 In contrast, other authors have not observed suchdifferences after administering anesthetic eye drops, such asDíaz-Rey et al.18

On the other hand it must be taken into account that thegroup analyzed in our case was more numerous than the groupof Ishibazawa et al. (112 versus 30 cases) and this could havecaused the variation in results. Similarly, the difference inthe match limits which included 95% of observed differences

was of 30.2 �m (16.1–[−14.1]) in our study and of 31 �m in theIshibazawa et al. study. This supports the idea of a possiblesystematic error in the measurements taken with UP in oneof both studies.

48 a r c h s o c e s p o f t a l m o l . 2 0 1 3;8 8(2):45–49

Table 2 – Published studies comparing different corneal thickness measuring instruments.

Author # of eyes Devices Mean ± SD �m Match limitsat 95% in �m

Pearsoncorrelation

Li et al.19 42 AS-OCT −6.4 ± 8.6 −23.2 to 10.4 0.969Zhao et al.8 285 Visante −16.5 ± 11.7 −6.1 to 39.1 0.930Li et al.9 70 Visante −14.74 ± 10.84 −36.0 to 6.5 0.936Kim et al.20 155 Slit lamp OCT −26.3 ± 14.2 −54.7 to 2.1 r2 = 0.82Fukuda et al.22 40 3D CAS-OCT −2.39 ± 8.63 −14.51 to 14.30 0.977Ishibazawa et al.15 30 RTVue-100 −14 ± 8 −29 to 2 0.97Nam et al.16 104 RTVue-100 12.8 ± 1.5 2.8 to 22.7 –Lázaro et al. 112 RTVue-100 −1.04 ± 7.70 −14.1 to 16.1 0.957

mogr

r

1

1

1

1

1

1

SD: standard deviation; AS-OCT: anterior segment optic coherence to

It is very important to determine the possible matchesbetween both techniques because this can be a crucial fac-tor to assess the ocular hypertension grade being observed.Given the results that have been obtained in this study, thedifferences are not clinically significant and therefore, eventhough it cannot be said that both techniques have obtainedinterchangeable values, it can be stated that the results arecomparable and therefore both allow us to make an objectiveanalysis of corneal thickness.

As can be seen in Table 2,8,9,15,19–22 other studies exhib-ited a high degree of variability between both measurementtechniques. It is not surprising to observe differences evenbetween authors who used the same AS-OCT device becausethey can be explained by the same reasons discussed above.Thus, Prospero Ponce et al.21 observed a difference of −7.5 �mutilizing the Visante device (Carl Zeiss Meditec, Dublin, CA,USA), while Zhao et al. Found −16.5 �m with the same device.In any case, all the analyzed studies demonstrate a significantmatch between both instruments.Even though our study hasnot determined reproducibility and intra- and inter-observermatching with this diagnostic technique, other studies havefound a significant reproducibility and matching, includingintra-examiner matches between 0.948 and 0.999 and inter-examiner matches between 0.767 and 0.996.10,16,17,22–24

Accordingly, even though it would be convenient to haveprospective studies with larger numbers of patients, the dataavailable to date allow us to state that AS-OCT is a valid tech-nique for corneal thickness studies.

Conflict of interest

No conflict of interest has been declared by the authors.

e f e r e n c e s

1. Choma M, Sarunic M, Yang C, Izatt J. Sensitivity advantageof swept source and Fourier domain optical coherencetomography. Opt Express. 2003;11:2183–9.

2. Amano S, Honda N, Amano Y, Yamagami S, Miyai T,

Samejima T, et al. Comparison of central corneal thicknessmeasurements by rotating Scheimpflug camera, ultrasonicpachymetry, and scanning-slit corneal topography.Ophthalmology. 2006;113:937–41.

1

aphy.

3. Buehl W, Stojanac D, Sacu S, Drexler W, Findl O. Comparisonof three methods of measuring corneal thickness andanterior chamber depth. Am J Ophthalmol. 2006;141:7–12.

4. Kawana K, Tokunaga T, Miyata K, Okamoto F, Kiuchi T, OshikaT. Comparison of corneal thickness measurements usingOrbscan II, non-contact specular microscopy, and ultrasonicpachymetry in eyes after laser in situ keratomileusis. Br JOphthalmol. 2004;88:466–8.

5. Suzuki S, Oshika T, Oki K, Sakabe I, Iwase A, Amano S, et al.Corneal thickness measurements: scanning-slit cornealtopography and noncon-tact specular microscopy versusultrasonic pachymetry. J Cataract Refract Surg. 2003;29:1313–8.

6. Wong AC, Wong CC, Yuen NS, Hui SP. Correlational study ofcentral corneal thickness measurements on Hong KongChinese using optical coherence tomography, Orbscan andultrasound pachymetry. Eye. 2002;16:715–21.

7. Bechmann M, Thiel MJ, Neubauer AS, Ullrich S, Ludwig K,Kenyon KR, et al. Central corneal thickness measurementwith a retinal optical coherence tomography device versusstandard ultrasonic pachymetry. Cornea. 2001;20:50–4.

8. Zhao PS, Wong TY, Wong WL, Saw SM, Aung T. Comparison ofcentral corneal thickness measurements by Visante anteriorsegment optical coherence tomography with ultrasoundpachymetry. Am J Ophthalmol. 2007;143:1047–9.

9. Li EY, Mohamed S, Leung CK, Rao SK, Cheng AC, Cheung CY,et al. Agreement among 3 methods to measure cornealthickness: ultrasound pachymetry, Orbscan II, and Visanteanterior segment optical coherence tomography.Ophthalmology. 2007;114:1842–7.

0. Mohamed S, Lee GK, Rao SK, Wong AL, Cheng AC, Li EY, et al.Repeatability and reproducibility of pachymetric mappingwith Visante anterior segment-optical coherencetomography. Invest Ophthalmol Vis Sci. 2007;48:5499–504.

1. Muscat S, McKay N, Parks S, Kemp E, Keating D. Repeatabilityand reproducibility of corneal thickness measurements byoptical coherence tomography. Invest Ophthalmol Vis Sci.2002;43:1791–5.

2. Marsich MW, Bullimore MA. The repeatability of cornealthickness measures. Cornea. 2000;19:792–5.

3. Bland JM, Altman DG. Statistical methods for assessingagreement between 2 methods of clinical measurement.Lancet. 1986;1:307–10.

4. Bland JM, Altman DG. Comparing methods of measurement:why plotting difference against standard method ismisleading. Lancet. 1995;346:1085–7.

5. Ishibazawa A, Igarashi S, Hanada K, Nagaoka T, Ishiko S, ItoH, et al. Central corneal thickness measurements withFourier-domain optical coherence tomography versus

ultrasonic pachymetry and rotating Scheimpflug camera.Cornea. 2011;30:615–9.

6. Nam SM, Im CY, Lee HK, Kim EK, Kim TI, Seo KY. Accuracyof RTVue optical coherence tomography. Pentacam, and

m o l

1

1

1

2

2

2

2

2

and reproducibility of slit-lamp-adapted optical coherence

a r c h s o c e s p o f t a l

ultrasonic pachymetry for the measurement of centralcorneal thickness. Ophthalmology. 2010;117:2096–103.

7. Nam SM, Lee HK, Kim EK, Seo KY. Comparison of cornealthickness after the instillation of topical anesthetics:proparacaine versus oxybuprocaine. Cornea. 2006;25:51–4.

8. Díaz-Rey JA, Giráldez MJ, García-Resúa C, Yebra-Pimentel E.Efecto de la instilación del colicursi® anestésico doble(tetracaína 0,1% y oxibuprocaína 0,4%) sobre el espesorcorneal central y paracentral. Arch Soc Esp Oftalmol.2009;84:23–30.

9. Li Y, Shekhar R, Huang D. Corneal pachymetry mapping withhigh-speed optical coherence tomography. Ophthalmology.2006;113:792–9.

0. Kim HY, Budenz DL, Lee PS, Feuer WJ, Barton K. Comparisonof central corneal thickness using anterior segment optical

coherence tomography vs. ultrasound pachymetry. Am JOphthalmol. 2008;145:228–32.

1. Prospero Ponce CM, Rocha KM, Smith SD, Krueger RR. Centraland peripheral corneal thickness measured with optical

. 2 0 1 3;8 8(2):45–49 49

coherence tomography, Scheimpflug imaging, and ultrasoundpachymetry in normal, keratoconus-suspect, and post-laserin situ keratomileusis eyes. J Cataract Refract Surg.2009;35:1055–62.

2. Fukuda S, Kawana K, Yasuno Y, Oshika T. Anterior ocularbiometry using 3-dimensional optical coherence tomography.Ophthalmology. 2009;116:882–9.

3. Rosas Salaroli CH, Li Y, Zhang X, Tang M, Branco Ramos JL,Allemann N, et al. Repeatability of laser in situ keratomileusisflap thickness measurement by Fourier-domain opticalcoherence tomography. J Cataract Refract Surg.2011;37:649–54.

4. Sandler SF, Zelefsky JR, Dorairaj S, Arthur SN, Ritch R,Liebmann JM. Intra-observer and inter-observer reliability

tomography for evaluation of anterior chamber depth andcentral corneal thickness. Ophthalmic Surg Lasers Imaging.2008;39:299–303.