# The effect of corneal thickness and corneal curvature on pneumatonometer measurements

Post on 25-Feb-2017

212 views

TRANSCRIPT

Abstract

Purpose. The purpose of this study was to investigate theinfluence of corneal topography and thickness on intraocularpressure (IOP) and pulse amplitude (PA) as measured usingthe Ocular Blood Flow Analyzer (OBFA) pneumatonometer(Paradigm Medical Industries, Utah, USA).

Methods. 47 university students volunteered for this cross-sectional study: mean age 20.4yrs, range 18 to 28yrs; 23male, 24 female. Only the measurements from the right eyeof each participant were used. Central corneal thickness andmean corneal radius were measured using Scheimpflug biometry and corneal topographic imaging respectively. IOPand PA measurements were made with the OBFA pneu-matonometer. Axial length was measured using A-scan ultrasound, due to its known correlation with these cornealparameters. Stepwise multiple regression analysis was usedto identify those components that contributed significantvariance to the independent variables of IOP and PA.

Results. The mean IOP and PA measurements were 13.1 (SD3.3) mmHg and 3.0 (SD 1.2) mmHg respectively. IOP mea-surements made with the OBFA pneumatonometer correlatedsignificantly with central corneal thickness (r = +0.374, p =0.010), such that a 10 mm change in CCT was equivalent toa 0.30mmHg change in measured IOP. PA measurementscorrelated significantly with axial length (part correlate =-0.651, p < 0.001) and mean corneal radius (part correlate =+0.459, p < 0.001) but not corneal thickness.

Conclusions. IOP measurements taken with the OBFA pneu-matonometer are correlated with corneal thickness, but notaxial length or corneal curvature. Conversely, PA measure-ments are unaffected by corneal thickness, but correlatedwith axial length and corneal radius. These parameters

should be taken into consideration when interpreting IOP andPA measurements made with the OBFA pneumatonometer.

Keywords: pneumatonometer; intraocular pressure mea-surement; pressure pulse amplitude; corneal thickness;corneal curvature

Introduction

The measurement of intraocular pressure (IOP) is of clinicalimportance in the detection and monitoring of primary open-angle glaucoma (POAG).1,2 The influence of corneal topog-raphy on the measurement of IOP by Goldmann applanationtonometry is well documented3 and IOP measurements areknown to be influenced by corneal thickness4 and cornealcurvature.5 These influences are of clinical concern as erro-neous IOP measurements may lead to mislabelling of glaucoma patients and healthy subjects.6 Tomlinson andLeighton7 found the mean corneal radius to be flatter inpatients with normal-tension glaucoma than those withPOAG or normal patients. Copt et al.,8 on correcting IOP forcorneal thickness, calculated that 31% of their patients withnormal-tension glaucoma would be reclassified as havingPOAG and that 56% of their ocular hypertensive patientswould be classified as normal.

The principle of pneumatonometry uses elastic platetheory.9 A probe tip containing a central tube and surroundedby side exhausts is capped with a thin silastic membrane (Fig. 1). A constant flow of gas passes down the central tube,forcing open a small gap between membrane and tube edge,and is expelled through the side exhausts. When placedagainst a cornea, the cornea is flattened to the outer edges of

Received: April 11, 2002Accepted: September 4, 2002

Correspondence: Sarah L. Hosking, Neurosciences Research Institute, School of Life & Health Sciences, Aston University, Aston Triangle,Birmingham, B4 7ET, UK. Tel: +44 (0)121 359 3611 ext. 5172, Fax: +44 (0)121 333 4220, E-mail: s.l.hosking@aston.ac.uk

The effect of corneal thickness and corneal curvature on

pneumatonometer measurements

Andrew J. Morgan, Justine Harper, Sarah L. Hosking and Bernard Gilmartin

Neurosciences Research Institute, School of Life & Health Sciences, Aston University, Birmingham, UK

Current Eye Research 0271-3683/02/2502-107$16.002002, Vol. 25, No. 2, pp. 107112 Swets & Zeitlinger

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/26/

14Fo

r pe

rson

al u

se o

nly.

108 A.J. Morgan et al.

the probe tip and the membrane gap is forced closed. The gaspressure within the tube rises until it balances the forces onthe other side of the membrane (namely the IOP and themembrane mechanics of the cornea) and the gas can onceagain escape. At this balance point the gas pressure withinthe tube, in comparison to atmospheric pressure, is propor-tional to the IOP and it is from this that a measure of IOP ismade. In addition, when placed against the cornea for anumber of seconds, the high frequency response of the pneu-matonometer provides a method of measuring the small fluctuation in IOP that occurs with each heart beat.10 Theamplitude of this IOP pulsation (PA) originates from therhythmic dilation and contraction of the intraocular, primar-ily choroidal, vasculature.11 The PA measurement has shownclinical utility because it is diminished in low-tension glau-coma patients compared to healthy controls and amplified inocular hypertensive patients compared to POAG patients.12,13

It has been postulated that these variations are a manifesta-tion of anomalous ocular blood flow: low-tension glaucomapatients exhibiting an impoverished supply and ocular hyper-tensive patients an enhanced and possibly protective bloodflow.14

IOP measurements taken with pneumatonometers areused clinically15 and PA measurements are principally usedin research.16 However there have been few reports on the

influence of corneal dimensions on these measurements.Only the association between corneal thickness and IOPmeasurements by pneumatonometry has been investigatedand results have been mixed: some investigators concludethat IOP measurements are significantly influenced bycorneal thickness17 and others that they are not.18,19 Thepurpose of this study was to investigate the influence ofcorneal topography and thickness on the pneumatonometricmeasurements of IOP and PA.

Materials and methods

Subjects

47 university students volunteered for this cross-sectionalstudy: mean age 20.4yrs, range 18 to 28yrs; 23 male, 24female. In order to confirm that all eyes included in the studywere healthy, subjects underwent a full eye examination. Asone of the purposes of this study was to investigate the rela-tionship between pneumatonometer measurements and meancorneal curvature, it was felt that high degrees of cornealastigmatism or causes of corneal distortion could be a sourceof error. Subjects were therefore excluded from the study ifthey had greater than 2.00 dioptres of astigmatism, had wornhard or rigid contact lenses, or had any ocular pathology. Following the tenets of the Helsinki declaration, the studyreceived approval from Aston Universitys ethical committeeand each subject gave written informed consent before takingpart. All measurements were taken with the subject in aseated position, between the hours of 10.00 am and 4.00 pm,by the same investigator (JH).

Instruments

The pneumatonometer used in this study was the OcularBlood Flow Analyzer (OBFA; Paradigm Medical Industries,Utah, USA) which is based on the original design of Durhamet al.20 and its later modifications.21,22 After anaesthetising the cornea with 0.4% benoxinate hydrochloride (Minims,Chauvin, UK), average IOP and PA measurements were auto-matically calculated from one continuous IOP recording(approximate recording time 510 seconds) that was suffi-cient to encapsulate five similar IOP pulses. Mean cornealradius was measured with a computerised corneal topogra-pher (EyeSys 2000 Corneal Analysis System, Spectrum Ophthalmic, UK) by taking the average of three readings. Ameasure of central corneal thickness (CCT) was calculatedby taking the average of three corneal width images producedfrom a Scheimpflug camera system (CASE-S, Marcher Enter-prises, Hereford, England). In addition, because of the knowncorrelations between corneal radius with axial length23 andaxial length with pulse amplitude,24 axial length measure-ments were taken on all subjects. Axial length was calculatedautomatically as the average of 10 A-scan ultrasound read-ings (Storz Omega Compu-Scan Biometric Ruler, StorzInternational, St Louis, USA). In order to assess repeatabil-

Figure 1. Schematic of a pneumatonometer probe in contact witha cornea: P1, air pressure within probe; P2, intraocular pressure; AT,total air flow entering probe; AB, proportion of air flow escapingfrom the air bearing; AP, proportion of air flow escaping from theprobe head.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/26/

14Fo

r pe

rson

al u

se o

nly.

Corneal thickness and curvature in pneumatonometry 109

ity of measurements, coefficients of reliability for each parameter were calculated and are shown in Table 1. Thecoefficient of reliability, sometimes called the intraclass correlation coefficient, estimates the average correlationbetween all possible pairs of observations.25 For repeatedmeasurements, as in this study, the coefficient (CR) is calcu-lated via a one-way ANOVA as

where SSb and SSt are the sums of squares between subjectsand in total, respectively, and k is the number of measure-ments taken.

Analysis

As physiological data from the two eyes of a single subjectcorrelate highly, combining right and left eye measurementscan lead to erroneous statistical significance.26 To ensureindependence of the data, and in convention with otherstudies,27 measurements were analysed for the right eye onlyof each subject. Due to the possible intercorrelations of theindependent variables (corneal radius with corneal thick-ness28 and corneal radius with axial length23), a stepwise mul-tiple regression analysis was performed. Multiple regressionanalysis determines which predictor variables make a signif-icant contribution to the variance in the dependent variableand provides a figure for the unique contribution each vari-able makes after the other variables have been taken intoaccount (i.e., part correlation coefficient).29

Results

The mean IOP and PA measurements were 13.1 (SD 3.3;range 7.0 to 20.7) mmHg and 3.0 (SD 1.2; range 1.1 to 6.2)mmHg respectively. The axial length ranged from 22.31mmto 26.40mm (mean 24.05mm, SD 1.00mm) and CCT from398mm to 589mm (mean 507 mm, SD 40mm). The meanspherical refractive error, measured by subjective refraction,averaged -0.86 DS with a negative skew (range + 0.75 DSto -6.50 DS). Table 2 shows the results of the stepwise mul-tiple regression analysis for IOP and PA measurements. IOPmeasurements correlated significantly with CCT measure-ments (r = +0.374, p = 0.010) and accounted for 14% of thevariance (R2) in the data. Figure 2 shows the plot of IOP mea-

CRkSSb SSt

k SSt=

-( )-( )1

surements versus CCT measurements. The linear regressionequation was IOP (mmHg) = 0.030 * CCT (mm) 2.2: 95%confidence intervals for the slope being 0.007 to 0.053. A 10mm change in CCT was equivalent to a 0.30mmHg changein measured IOP. There was no significant correlationbetween the measurement of IOP and those of corneal radiusand axial length (p > 0.05). PA measurements correlated neg-atively with axial length and positively with mean cornealradius: part correlation coefficients of -0.651 and +0.459respectively (both p < 0.001). The model calculated by step-wise multiple regression attributed 46% of the overall vari-ance in PA measurements to axial length and mean cornealcurvature: the individual contributions of variance being 25%for axial length and 21% for mean corneal curvature. Therewas no significant correlation between PA and CCT mea-surements (p > 0.05).

Discussion

Intraocular pressure measurements

This study found that IOP measurements made with theOBFA pneumatonometer were significantly influenced bycorneal thickness and accounted for 14% of the variance inthe measurements. Our results support those of a similarstudy17 who, using the same pneumatonometer, reported achange in measured IOP of 0.20mmHg per 10 mm change inCCT. The standard deviation of normal CCT measurements,using a Scheimpflug camera, has recently been reported as40mm.30 This would indicate that the influence of CCT onIOP measurements made with the OBFA pneumatonometeris only of clinical significance in those corneas at the extremelimits of their range in thickness.

The influence of corneal thickness on IOP measurementswith Goldmann applanation tonometry has long been recog-nised.31 A number of studies however have concluded thatpneumatonometry is less influenced by corneal thicknessthan Goldmann applanation tonometry.18,19,32 Walker andLitovitz,9 in their theoretical paper on pneumatonometry, alsoclaimed that corneal thickness could be ignored. They statedthat it is the bending forces necessary to flatten the corneathat are dependent on corneal thickness and not the tensionforces that are used in the balancing of pressures by the pneu-matonometer; as the bending forces used in flattening thecornea are taken up by the outer edges of the 5mm diame-ter probe, they are independent of the balancing forces occur-

Table 1. Measurement parameters coefficients of reliability.

Intraocular Pressure Mean Central Axialpressure pulse corneal corneal length

measurement amplitude radius thickness

Coefficient 0.98 0.94 0.99 0.89 0.96of reliability

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/26/

14Fo

r pe

rson

al u

se o

nly.

110 A.J. Morgan et al.

ring under the central (approximately 2.5mm in diameter)portion of the probe. However, since the introduction of anair-suspension bearing to hold the probe against the corneafor continuous IOP recordings,21 a proportion of the pneu-matonometers air flow is required to produce this initialcorneal flattening and the influence of corneal thickness maypossibly arise here (Fig. 1). It would be of interest to inves-tigate whether corneal thickness influences the force requiredto flatten the cornea more than the balancing force createdby the gas pressure in the centre of the tube.

A possible explanation for the discrepancy in the conclu-sions on whether the OBFA pneumatonometer is influencedby CCT, is in the respective probe diameters of the pneumaticand Goldmann tonometers. The studies that found IOP mea-surements varied less by pneumatonometry than Goldmanntonometry with a change in corneal thickness were based on measurements taken before and after laser refractivesurgery.18,19,32 As the ablation diameters in these studies were

56mm, it would not be unexpected that the force requiredto applanate a 3.06mm corneal diameter, as in the case ofthe Goldmann probe, would be influenced by the reductionin corneal thickness. The larger 5.0mm diameter pneu-matonometer probe however would exert its bending forcewhere the corneal thickness has had no or minimal reductionand therefore may be less influenced by the central cornealthinning. This may also explain why IOP measurements inGoldmann and pneum...

Recommended