Changes in corneal properties and its effect on intraocular pressure measurement following phacoemulsification with intraocular lens implantation with or without trabeculectomy

Purpose: To evaluate the changes in corneal biomechanical properties and their effect on preand postoperative differences in IOP measurement by each tonometer. Design: Observational study. Methods: The study was done in subjects who underwent phacoemulsification with intraocular lens (IOL) implantation (phaco-IOL) and combined phacoemulsification with IOL implantation and trabeculectomy (phaco-trab). IOP was measured by a single trained examiner using rebound tonometer (RBT), Ocular Response Analyzer (ORA), Goldmann applanation tonometer (GAT), dynamic contour tonometer (DCT), and Tono-Pen. Corneal hysteresis (CH) and corneal resistance factor (CRF) were measured using ORA, central corneal thickness (CCT) using ultrasonic pachymeter, and corneal curvature (CR) with manual keratometry. All measurements were done one week prior to surgery and after four weeks and six weeks of the two surgeries respectively. Only the operated eye was included for analysis. Results: Twenty-nine eyes of 29 normal subjects who underwent phaco-IOL and 23 eyes of 23 glaucoma subjects who underwent phaco-trab were studied. Increase in CCT [10.2 ± 14.86 microns, p = 0.001], decrease in CH [0.82 ± 1.38 mmHg, p = 0.003] and CRF [0.97 ± 1.0 mmHg, p < 0.001] were found post-phaco-IOL, whereas post-phaco-trab decrease in CCT [16.61 ± 15.22 microns, p < 0.001], CRF [2.28 ± 1.93 mmHg, p < 0.001] with increase in CH [0.95 ± 1.89 mmHg, p = 0.03] were noted. Multiple linear regression analysis showed significant associations for change in CH and CRF with change in IOP and not with CCT and CR postoperatively. Conclusion: Alterations in CH and CRF were associated with changes in IOP measured postoperatively by different tonometers. CH and CRF changes contribute to postoperative changes in measured IOP.


Introduction
Corneal viscoelastic properties depend on the corneal collagen fibrils.The arrangement of collagen fibrils varies directionally and regionally, which results in regional differences in elasticity. 1,2Hjortdal showed that for a given intraocular pressure (IOP) radially the cornea is stronger at the center and circumferentially at the limbus.It is also known that both corneal and scleral fibrils are fused at the limbus. 3 Woo et  al. found that the corneal biomechanical characteristics are approximately similar to that of the sclera. 4The Ocular Response Analyzer (ORA; Reichert Inc., Depew, New York, USA), measures in vivo corneal biomechanical properties such as corneal hysteresis (CH) and corneal resistance factor (CRF) by analyzing corneal responses to indentation by a rapid air pulse.CH represents the corneal viscoelasticity and has been described as the ability of the tissue to absorb and dissipate energy.CRF represents overall corneal resistance and the ability to withstand the applanation force. 5,6Eyes with a higher CH tend to have increased damping capacity, which buffers the harmful effect of IOP fluctuation on the optic nerve head.[9][10][11] Several studies reported the effect of corneal factors on IOP measurements by different tonometers, such as the Goldmann applanation tonometer (GAT), dynamic contour tonometers (DCT), ORA, Tono-Pen, and rebound tonometer (RBT) among normal and glaucoma subjects.3][14][15][16][17][18] Among all the corneal parameters, the effect of CCT was studied extensively.It is estimated that every ten micron increase in CCT results in 0.35 mmHg difference between GAT and ORA corneal compensated IOP (IOPcc) measurements and 0.80 mmHg between GAT and RBT measurements. 180][21][22][23][24][25] In trabeculectomy surgery, a partial scleral flap thickness is created adjacent to the limbus.This may potentially modify the biomechanical properties of the ocular walls.To the best of our knowledge, there are no reports on alterations in corneal biomechanical properties and its effect on IOP measurement following combined phacoemulsification and trabeculectomy (phaco-trab) surgery.
The aim of this study was to evaluate the changes of corneal biomechanical properties and its effect on pre-and postoperative differences in IOP measurement by each tonometer.
The study followed the tenets of the Declaration of Helsinki.The study was approved by the Institutional Review Board of Vision Research Foundation, Chennai, India.Subjects willing to participate in the study were informed in detail about the study and its objectives.Written informed consent was obtained from subjects before enrolling them into the study.
Inclusion criteria for the phaco-IOL group were age more than 30 years, clear cornea with normal fundus, and no previous history of ocular trauma, ocular surgery or refractive surgery.Eyes with any ocular pathology other than cataract were excluded.
Inclusion criteria for the phaco-trab group were age more than 30 years, clear cornea, diagnosed as glaucoma, 26 and no previous history of ocular trauma, ocular surgery or refractive surgery.Eyes with any other ocular pathology were excluded.

Baseline evaluation
Subjects underwent a comprehensive ophthalmic examination which included best corrected visual acuity (BCVA), IOP measured by GAT, slit-lamp bio-microscopy, gonioscopy, and stereoscopic dilated fundus evaluation on the slit lamp using a 90D lens.Subjects who had glaucomatous optic disc changes underwent standard automated visual field testing (SAP) (Humphrey Field Analyzer (HFA)) (model 750; Carl Zeiss Meditec).The axial length was measured by using ultrasound biometry (OcuScan® RxP Ophthalmic Ultrasound System, Alcon Laboratories, Inc. South Freeway, Fort Worth, Texas, U.S.A.).

Preoperative assessments
Preoperative measurements were done one week before the surgery.CR was measured using a keratometer (KMS-6; Appaswamy Associates, Chennai, India) before IOP measurements.IOP was measured using the following tonometers: RBT (Icare; Tiolat Oy, Helsinki, Finland), ORA (Reichert Ophthalmic Instruments, Buffalo, N.Y., U.S.A), Tono-Pen (Tono-Pen XL, Medtronic Solan, Southpoint Drive, North Jacksonville, U.S.A.), DCT (Swiss Microtechnology AG, Port, Switzerland), and GAT (AT-900; Haag Streit AG Gartenstadtstrasse, Koeniz, Switzerland).The RBT and ORA measurements were taken before instillation of topical anaesthesia.Tono-Pen, GAT, and DCT measurement were done after applying anaesthetic eye drops (proparacaine 0.5%, Paracain; Sunways India Pvt Ltd.Mumbai, India).The order of RBT and ORA measurements were randomized using a simple randomization method.Similarly the order of Tono-Pen, GAT, and DCT measurements were randomized for each subject.CCT was measured using an ultrasonic pachymeter (DGH-550 Pachette 2; DGH technology, Inc., PA, U.S.A.) after five minutes of the last IOP measurement.Repeated measurements were taken until a set of ten values differed by less than ten microns.

RBT
RBT was performed in the sitting position without instilling topical anaesthesia.Six measurements were acquired by lightly pressing the tonometer button.The instrument automatically averages the six measurements and the mean IOP is displayed. 27

ORA
Subjects were asked to place their head against the head rest.Four air puffs within each measurement on each eye were taken; the signal with the highest waveform was automatically selected as the best signal value (BSV) for each eye.Those with poor-quality waveforms (multiple applanation spikes or asymmetric signals) were excluded.IOPcc, IOPg, CH, and CRF were recorded. 5

Tono-Pen
The Tono-Pen was calibrated before each measurement.Tono-Pen measurements were taken after anaesthetising the eye.The cornea was applanated with the Tono-Pen tip several times until a reading was displayed.Measurements with a standard error of means (SEM) less than 5% were included.If successive measurements differed by more than 5 mmHg, the procedure were repeated.All Tono-Pen measurements were made with a disposable latex cover over the tip. 28

GAT
Calibration for GAT was performed on a daily basis.Topical anaesthesia was instilled and the tear film stained using a sodium fluorescein strip.An average of two measurements was recorded for analysis; if they differed by more than 2.0 mmHg, a third reading was taken and the mean of the values was taken for analysis. 12T DCT was performed after instilling topical anaesthesia.Each DCT IOP measurement usually requires five to eight seconds where the probe is placed to have continuous contact with the eye.The instrument gives a quality score that ranges from 1 to 5 (lower scores indicate better quality) and Ocular Pulse Amplitude (OPA) value along with IOP measurement.All measurements taken for this study had quality readings ranging from 1-3. 29

Postoperative measurements
In addition to routine assessment, pachymetry and IOP were re-evaluated four weeks after performing phaco-IOL surgery and six weeks of phaco-trab using the same devices as mentioned earlier.Both pre-and postoperative CR, CH, CRF, IOP, and CCT measurements were performed by a single observer who was blinded to actual readings and the readings were read and recorded by the second examiner.

Statistical analysis
The relevant information and test results were recorded in MS Office Excel 2007.Statistical analysis was carried out with SPSS statistical software (Version 14; SPSS Science Inc., Chicago, IL, USA), Med Calc Statistical software version 8.1, and MS-Excel 2007.Only data from the operated eye were considered for statistical analysis.Descriptive statistics were calculated for all parameters of all the groups.Tests for normality were carried out for each continuous variable and appropriate parametric/non-parametric analyses were performed.Type-I error was kept at 5% level.Paired t-test was done to compare between pre-and postoperative measurements.Multiple linear regression was done to find the association between postoperative changes of corneal biomechanics and IOP.No comparison was done between phaco-IOL and phaco-trab group for postoperative measurements.

Effect of axial length on corneal properties and tonometers
Simple linear regression was done using axial length (AXL) as independent variable and postoperative changes of corneal parameters (ΔCR, ΔCH, ΔCRF, and ΔCCT) and ΔIOP measurement by each tonometer as dependent variables.No significant association was found between AXL and postoperative changes of corneal parameters (p > 0.05) and postoperative changes in IOP (p > 0.05) measured by each tonometer following phaco-IOL and phaco-trab.

Discussion
Corneal properties were found to be significantly influenced by intraocular surgeries, with which the postoperative changes of IOP measured by each tonometer were explained by the changes of corneal properties.

Phaco-IOL group
In the phaco-IOL group, we found significant increase in CCT and decrease in CH and CRF after four weeks after the surgery.3][24][25] In the present study, a significant increase in CCT and decrease in CH and CRF were noted even after one month after the surgery.This decrease in CH and CRF may be due to an incision made on the cornea that could weaken the strength of the eyeball, resulting in lower corneal biomechanics.Lu et al. 30 reported that there was no significant correlation between soft contact lens-induced corneal edema, CH and CCT, yet the effect of corneal edema induced by cataract surgery or contact lens on measurements of corneal biomechanics is not clear.

Phaco-trab group
In the phaco-trab group, we found significant increase in CH and decrease in CRF and CCT six weeks after the surgery.This increase in CH may be explained by the postoperative reduction of IOP.Similar finding were reported by Sun et al. 31 where IOP was reduced due to anti-glaucoma medications and surgery in subjects with chronic angle closure glaucoma.CH values significantly increased from 6.83 ± 2.08 mmHg to 9.22 ± 1.80 mmHg at two weeks and 9.50 ±1.66 mmHg at four weeks after treatment.Neuberger et al. 32 suggested in an experimental study that ORA gives lower CH measurement in eyes with higher IOP due to limited indentation on the cornea.This reason could probably explain the increased CH with decreased IOP in post phaco-trab subjects.CH measurements were dependent on IOP level whereas CRF measurements were not influenced by IOP level.In the present study, we noted that CRF was significantly decreased both post-phaco-IOL and -phacotrab, thereby concluding that the overall resistance of the cornea, i.e., CRF, is altered following intraocular surgeries.

Postoperative changes in IOP
A study by Kucumen et al. showed no significant difference pre-and post-1month IOPcc and IOPg measurements following phaco-IOL. 22In the present study, we found no significant difference in pre-and postoperative (one month) IOP measurements obtained by all tonometers in the phaco-IOL group except DCT (p = 0.04).Several studies reported that DCT underestimates IOP in eyes with high IOP compared to GAT and overestimates IOP in eyes with low IOP. 33,34 n the present study, we noted that DCT underestimated IOP (0.1 ± 4.4 mmHg, p < 0.05) in preoperative eyes (phaco-trab) as compared to GAT, similar to Tonnu et al. 34 However, postoperatively (phaco-trab) mean DCT-IOP was 2.2 mmHg (± 2.9) (p > 0.05) higher than GAT-IOP.Hamilton et al. 35 and Oh et al. 36 have also reported that DCT underestimates IOP in eyes with contact lens-induced corneal edema.The reason for this difference in IOP measured by DCT following post-one-month phaco-IOL was not clear.However, the least mean difference was noted [6.12 mmHg (± 5.85), p < 0.001] between pre-and postoperative IOP obtained by DCT following phaco-trab among all the tonometers.In our pilot study, we found that DCT overestimated IOP measurement as compared to GAT in eyes with IOP less than 20 mmHg, whereas this pattern reverses in eyes with IOP greater than 20 mmHg.

Effect of corneal properties on tonometers
CH is significantly lower in glaucomatous eye as compared to non-glaucomatous eyes.Similar findings were noted in the present study. 7-9Luce 5 reported that CH and CRF are direct measurements of corneal biomechanical properties and more completely describe the effect of corneal biomechanics on IOP measurements than CCT.Hence, we studied the postoperative changes in corneal biomechanics and its effect on IOP measured by all the tonometers.
CH: In the present study, ΔCH was negatively associated with ΔIOPs both in the phaco-IOL and phaco-trab groups.Multiple linear regression analysis showed postoperative increase in CH results decrease IOP following phaco-IOL and phaco-trab.
CRF: ΔCRF was positively associated with ΔIOPs.Multiple linear regression analysis showed postoperative decrease in CRF results decrease IOP following phaco-IOL and phaco-trab.
Our findings indicate that corneal biomechanics (CH and CRF) independently influenced all tonometers.DCT measurement was less influenced by corneal biomechanics compared to other tonometers.DCT measures IOP by a sensor tip which does not applanate the corneal surface, so theoretically it is not affected by any force-to-pressure translations.The variability between GAT and DCT decreased by 0.7 mmHg for every ten micron increase in CCT. 37Doyle et al. 38 reported that the mean difference between GAT-IOP and DCT-IOP was -2.6 mmHg in thin corneas and -0.06 mmHg in thick corneas.According to Kotecha et al. 16 GAT significantly underestimates IOP compared to DCT and ORA IOPcc in eyes with low CRF, whereas Wang et al. 17 reported that GAT underestimates IOP compare to DCT eyes with low CH.Those two previous studies did not consider both CH and CRF as influential factors for IOP measurement.Hence, the variability in IOP measurement among various tonometers can be explained in a better manner by considering all the corneal properties.
In the present study, OPA was significantly higher in glaucomatous eye as compared to non-glaucomatous eyes, which significantly decreased in the phacotrab group but not in the phaco-IOL group.This suggests that OPA is dependent on the level of IOP, a finding similar to that reported by Knecht et al. 39 In the present study, from the simple and multiple linear regression model we did not find any association between postoperative changes of OPA and ΔIOP measured by all tonometers (Table 4).Even after removing OPA from the multiple regression models, we found no difference in the strength of association.
IOP reduction after cataract surgery ranged from 1.3 to 2.05 mmHg at one year follow-up in a few studies. 40Issa et al. 41 reported that the preoperative anterior chamber depth was inversely related to the postoperative IOP reduction, whereas Bhallil et al. 42 and Huang et al. 43 reported that postoperative IOP reduction was associated with the lens thickness and axial length, not with preoperative and postoperative anterior chamber depth.In the present study we did not find association between axial length and the reduction of IOP following phaco-IOL and phaco-trab.There was also no association between postoperative changes in corneal parameters (ΔCR, ΔCH, ΔCRF, and ΔCCT) with axial length in our study.
This study was done on limited subjects and may not reflect population data.We did not look into the variability of surgeon factor, which could be a confounder of the outcome.Hence, we need quantitative and longitudinal assessment of the corneal biomechanical parameters in eyes that undergo intraocular surgery and its effects on IOP measurements.