Menu

  1. Home
  2. /
  3. Glaucoma–I
  4. /
  5. FP1485 : Do the Newer Tonometers Have Better Repeatability and Agreement with the Gold Standard?

Share this post

Dr. Mahalakshmi Nair V.R., N14243, Dr. Shoruba, Dr. Sathi Devi A V, Dr. Sushma Tejwani

 INTRODUCTION

Intraocular pressure (IOP) being the only identified modifiable risk factor for glaucoma development and progression requires precise estimation for proper management of these patients. The ideal tonometer is expected to be accurate, repeatable, reproducible and minimally influenced by factors such as corneal properties and different examiners. Goldmann applanation tonometer (GAT) is known to be affected by corneal properties.1Ocular response analyzer (ORA) and Corneal visualization with Scheimfplug technology(Corvis ST) are the new tonometers which take corneal biomechanical properties into consideration.2 This study compares repeatability  and agreement of GAT with the  two new tonometers in glaucoma patients.

METHODS:

This was an observational, non-randomized case series performed at a tertiary eye care center in Bengaluru, South India between November and December 2015.The study protocol was in adherence to the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of the institution. All participants provided written informed consent before enrolment. The study included 163 eyes of 140 glaucoma patients (87 men and 53 women) aged 23-83 years .All types of patients with glaucoma (primary or secondary, open or angle closure) were included. IOP less than 10 mm Hg and more than 23 mm Hg were excluded. All participants underwent a comprehensive ocular examination, which included a detailed medical history, corrected distance visual acuity measurement, slit-lamp biomicroscopy, IOP measurement by Goldmann applanation tonometry, ORA and Corvis ST  ,gonioscopy, dilated fundus examination, visual field (VF) examination. VF examination was performed using a Humphrey Field analyzer II, model 720i (Zeiss Humphrey Systems, Dublin, California, USA), with the Swedish interactive threshold algorithm standard 24-2 program.

For the intraocular pressure, a total of 2 readings were taken with each instrument and the gap between IOP measurements with different instruments was 5 minutes. Three observers one each with GAT, Corvis ST and ORA measured the IOP in every patient.

STATISTICAL ANALYSIS

Repeatability was assessed by within-subject standard deviation (Sw), within-subject coefficient of repeatability (CRw) and within-subject coefficient of variation (CVw=100x Sw / overall mean). The Sw was calculated as the square root of the within-subject mean square of error (the unbiased estimator of the component of variance due to random error) in a one-way random effects model.3CRw was calculated as 2.77 times Sw. CVw was calculated according to the procedure described by Bland and Altman 4.Standard error and confidence intervals for CVw were calculated based on the root mean square method.

The mean IOP measurements with different instruments were compared by repeated measures ANOVA. Bland Altman plots were used to assess the limits of agreement (LoA) between device pairs for IOP measurement. In a Bland and Altman plot, the difference between the IOP measurements with the two devices is plotted against their mean. 5The mean difference between the measurements on the Bland and Altman plot is an estimate of the fixed bias. Bland and Altman plot also detects the proportional bias in the measurements, which is the relationship of the difference in the measurements and the mean of the measurements. The presence of proportional bias indicates that the devices do not agree equally through the range of measurements. Proportional bias was formally evaluated by regressing the difference between the measurements with two devices on the average of the measurements with two devices.

Statistical analyses were performed using Stata version 13.1 (StataCorp, College Station, Tx) statistical software. A p value of ≤0.05 was considered statistically significant.  

RESULTS

One hundred and sixty-three eyes of 140 glaucoma patients (83 men and 53 women) with a mean age of 63.5±11.8 (range 23-83) years were analyzed.

Repeatability

Table 1 shows the Sw, CRw and CVwof the IOP measurements with GAT, Corvis and ORA. SwCRw and CVw of GAT were significantly better (smaller) than those with Corvis and ORA (95% CIs not overlapping). Sw,CRwv and CVw of Corvis were significantly smaller than those of ORA.

Agreement

Table 2 shows the mean IOP values with the three instruments. There were significant differences in IOP measurements with the three instruments. Pairwise instrument comparisons are shown as the mean difference (fixed bias) in Table 3. IOP measurement of GAT was statistically significantly lower than Corvis and ORA.IOP measurement of Corvis was statistically significantly lower than the cornea corrected IOP of ORA.

Table 3. Significant proportional bias in agreement was detected with most of the instrument pairs. This means that the agreement between instruments for IOP measurements varied significantly depending on the average IOP value .

Table 1. Repeatability estimates of intraocular pressure (IOP) measurements with different devices. Figures in the parenthesis represent 95% confidence intervals.

Sw (mm Hg) CRw (mm Hg) CVw (%)
GAT IOP 0.5 (0.4, 0.5) 1.3 (1.2, 1.5) 3.5 (2.9, 4.0)
Corvis IOP 1.1 (0.9, 1.2) 2.9 (2.6, 3.2) 5.2 (4.1, 6.1)
ORA IOPcc 2.1 (1.9, 2.3) 5.8 (5.2, 6.5) 9.8 (8.7, 10.8)
ORA IOPg 1.6 (1.4, 1.7) 4.3 (3.9, 4.8) 9.1 (8.0, 10.0)

Sw: within-subject Standard Deviation; CRw: coefficient of repeatability; CVw: Coefficient of variation; GAT: Goldmann applanation tonometer; ORA: ocular response analyzer; IOPcc: cornea corrected IOP; IOPg: GAT correlated IOP.

Table 2. Mean intraocular pressure (IOP) measurements with the different devices.

Mean ± SD Range
GAT IOP (mm Hg) 15.5 ± 3.6 6 to 22
Corvis IOP (mm Hg) 17.1 ± 3.5 10 to 37
ORA IOPcc (mm Hg) 20.9 ± 5.4 8 to 46
ORA IOPg (mm Hg) 17.7 ± 5.3 5 to 43
P value <0.001

GAT: Goldmann applanation tonometer; ORA: ocular response analyzer; IOPcc: cornea corrected IOP; IOPg: GAT correlated IOP.

 

Table 3. Agreement between different devices for intraocular pressure (IOP) measurements.

Device pair Mean difference P value Fixed bias Proportional bias 95% LoA
GAT-Corvis -1.6 mm Hg <0.05 Yes No -8.4 to 5.3
GAT-ORA IOPcc -5.2 mm Hg <0.05 Yes Yes -14.2 to 3.8
GAT-ORA IOPg -2.0 mm Hg <0.05 Yes Yes -10.6 to 6.5
Corvis-ORA IOPcc -3.7 mm Hg <0.05 Yes Yes -10.9 to 3.6
Corvis-ORA IOPg -0.5 mm Hg >0.05 No Yes -7.1 to 6.1
ORA IOPcc-ORA IOPg 3.2 mm Hg <0.05 Yes No -0.4 to 6.8

GAT: Goldmannapplanation tonometer; ORA: ocular response analyzer; IOPcc: cornea corrected IOP; IOPg: GAT correlated IOP; LoA: limits of agreement.

DISCUSSION

 REPEATABILTY

In our study, compared to the newer tonometers, IOP repeatability was best with GAT, followed by Corvis and then the ORA .ORA IOPg showed better repeatability than ORA IOPcc. Our study utilized Sw,CRw and CVw to report repeatability while most studies have utilized CV and ICC values to do so.

 

In our study all repeatability indices for GAT were significantly better than those of ORA and Corvis. Studies comparing GAT to Corvis have shown latter to be as good  or better than  GAT.compared to our studies. 2,7

 

In our study Corvis had better repeatability compared to ORA. Previous studies evaluating Corvis have reported excellent ICC (>0.90)2,6,7compared to good ICC values for ORA with IOPg faring better than IOP cc1similar to ours.

Previous studies comparing GAT with ORA have reported conflicting results. Our study showed better repeatability of GAT compared to ORA with CV values of 3.5%,9.8% and 9.1% respectively for GAT,IOPcc and IOPg. While one study had results similar to ours10ther studies showed ORA to be equally good if not better than GAT.1,9

In addition to Sw and CVw we have also reported CRw values as it has greater relevance for clinicians and represents test-retest variability3. In our study the repeatability co-efficient of GAT was 1.3 mm Hg which was much lesser than those reported in other studies11,12

AGREEMENT

IOP measurement of GAT was statistically significantly lower than Corvis and ORA.IOP measurement of Corvis was statistically significantly lower than the cornea corrected IOP of ORA.

The mean IOP obtained with GAT in our studies was 15.5 ± 3.6 mm Hg compared to Corvis of 17.1 ± 3.5mm Hg.Previous studies comparing GAT with Corvis have shown mixed results. While previous ones  have shown higher mean IOP values for Corvis compared to GAT ,similar to ours 6,13 others have shown  comparatively lower values.2,8

The ORA in our study overestimated GAT IOP, with IOPcc values higher than IOPg, consistent with other studies.1,9,14. However, the magnitude of difference seem to be vary between  studies.

As far as agreement between instrument pairs were concerned, poor agreement was observed between instruments pairs with wide limits of agreement in our study, the bias being least for Corvis ST and ORA IOPg of -0.5 mmHg with 95% limits of agreement of -7.1 to 6.1 mmHg .

Significant proportional bias in agreement was detected with most of the instrument pairs. This means that the agreement between instruments for IOP measurements varied significantly depending on the average IOP value, the least being for GAT-Corvis.

Previous studies comparing Corvis to GAT have also confirmed poor agreement between the devices similar to ours2,6 Comparison of GAT and ORA in our studies have shown poor agreement with  both IOPcc and IOPg values  with 95% limits of agreement of 14.2 to 3.8mm Hg and 10.6 to 6.5 mm Hg respectively, with significant proportional bias between the two. These findings are consistent with previous literature comparing the 2 methods.9,14

For a tonometer to be acceptable and interchangeable with GAT, the limits of agreement is considered to be ±3mm Hg.Both the newer tonometers had limits of agreement outside this range,and there was proportional bias,suggesting they are not interchangeable with GAT.15

Overall poor agreement was observed between instrument pairs in our studies with wide limits of agreement, the bias being least for Corvis and ORA IOPg .Although ORA and Corvis ST are the latest tonometers taking corneal biomechanical properties into consideration, GAT remains the gold standard with better IOP repeatability and better agreement with Corvis ST than ORA depending on average IOP values.

REFERENCES 

  1. Wang AS, Alencar LM, Weinreb RN, Tafreshi A et al. Repeatability and reproducibility of Goldmannapplanation, dynamic contour,and ocular response analyzer tonometry. J Glaucoma. 2013 Feb;22(2):127-32.
  2. Hong J, Xu J, Wei A, Deng SX et al. A new tonometer–the Corvis ST tonometer: clinical comparison with noncontact and Goldmannapplanationtonometers. Invest Ophthalmol Vis Sci. 2013 Jan 23;54(1):659-65. = 410
  3. Bland JM, Altman DG. Measurement error. BMJ. 1996;313:744.
  1. Bland JM, Altman DG. Statistics Notes: Measurement error proportional to the mean. BMJ 1996;313:106.
  2. Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res 1999;8:135–60.
  1. Reznicek L, Muth D, Kampik A, Neubauer AS, Hirneiss C. Evaluation of a novel Scheimpflug-based non-contact tonometer in healthy subjects and patients with ocular hypertension and glaucoma. The British journal of ophthalmology. 2013; 97(11):1410–4. doi: 10.1136/bjophthalmol-2013-303400 PMID: 23969314
  2. Salvetat ML, Zeppieri M, Tosoni C, Felletti M, Grasso L, Brusini P. Corneal Deformation Parameters Provided by the Corvis-ST Pachy-Tonometer in Healthy Subjects and Glaucoma Patients. Journal of glaucoma. 2015; 24(8):568–74. Epub 2014/10/17. doi: 10.1097/IJG.0000000000000133 PMID: 25318572
  3. Matsuura M, Hirasawa K, Murata H, Yanagisawa M, Nakao Y, Nakakura S, et al. (2016) The Relationship between Corvis ST Tonometry and Ocular Response Analyzer Measurements in Eyes with Glaucoma. PLoS ONE 11(8): e0161742. doi:10.1371/journal.pone.0161742
  4. Carbonaro F, Andrew T, Mackey D, Spector T, Hammond C. Comparison of three methods of intraocular pressure measurement and their relation to central corneal thickness. Eye. 2010;24(7):1165–70. pmid:20150923
  5. Sullivan-Mee M, Gerhardt G, Halverson KD, Qualls C. Repeatability and reproducibility for intraocular pressure measurement by dynamic contour, ocular response analyzer, and goldmannapplanation tonometry. J Glaucoma. 2009; 18(9):666–673. [PubMed: 20010245
  6. Tonnu PA, Ho T, Sharma K, et al. A comparison of four methods of tonometry: method agreement and interobserver variability. Br J Ophthalmol 2005;89:847–50.
  7. Pandav SS, Sharma A, Gupta A, et al. Reliability of proton and Goldmannapplanationtonometers in normal and postk- eratoplasty eyes. Ophthalmology 2002;109:979–84.
  8. Lanza M, Iaccarino S, Cennamo M, Irregolare C, Romano V, Carnevale UA. Comparison between Cor- vis and other tonometers in healthy eyes. Contact lens & anterior eye: the journal of the British Contact Lens Association. 2015; 38(2):94–8. Epub 2014/12/04.
  9. Kaushik S Pandav SS Banger A Aggarwal K Gupta A. Relationship between corneal biomechanical properties, central corneal thickness, and intraocular pressure across the spectrum of glaucoma. Am J Ophthalmol . 2012; 153: 840–849. [CrossRef][PubMed]
  10. Lam A, Chen D, Chiu R, Chui WS. Comparison of IOP Measurements between ORA and GAT in Normal Chinese. Optom Vis Sci 2007; 84: 909–914.

 

 

Leave a Comment