Comparison of three rebound tonometers in normal and glaucomatous dogs

Abstract

Objective: The objectives of the study were to compare intraocular pressure (IOP) readings across a wide range and obtained via three rebound tonometers in ADAMTS10-mutant Beagle-derived dogs with different stages of open-angle glaucoma (OAG) and normal control dogs and to investigate the effect of central corneal thickness (CCT).

Animals studied: Measurements were performed on 99 eyes from 50 Beagle-derived dogs with variable genetics-16 non-glaucomatous and 34 with ADAMTS10-OAG. Seventeen OAG eyes were measured twice-with and without the use of IOP-lowering medications.

Procedures: IOP was measured in each eye using three tonometers with their "dog" setting-ICare® Tonovet (TV), ICare® Tonovet Plus® (TVP), and the novel Reichert® Tono-Vera® Vet (TVA)-in randomized order. CCT was measured with the Accutome® PachPen. Statistical analyses included one-way ANOVA, Tukey pairwise comparisons, and regression analyses of tonometer readings and pairwise IOP-CCT Pearson correlations (MiniTab®).

Results: A total of 116 IOP measurements were taken with each of the three tonometers. When comparing readings over a range of ~7-77 mmHg, mean IOPs from the TV were significantly lower compared with TVP (-4.6 mmHg, p < .001) and TVA (-3.7 mmHg, p = .001). We found no significant differences between TVA and TVP measurements (p = .695). There was a moderate positive correlation between CCT and IOP for TVA (r = 0.53, p < .001), TVP (r = 0.48, p < .001), and TV (r = 0.47, p < .001).

Conclusions: Our data demonstrate strong agreement between TVP and TVA, suggesting that the TVA may similarly reflect true IOP values in canines. CCT influenced IOP measurements of all three tonometers.

Keywords: ADAMTS10; canine; central corneal thickness (CCT); glaucoma; intraocular pressure (IOP); rebound tonometry.

FIGURE 1

Use of the iCare rebound tonometers TonoVet® (A) and TonoVet Plus® (B) in dogs

FIGURE 2

Use of the Reichert rebound tonometer Tono‐Vera® in dogs. (A) A ring on the display changes from red to green when the instrument is correctly aligned with the apex of the cornea. (B) Two red LED lights, one on each side of the electromagnetic probe, measure the alignment with the corneal surface

FIGURE 3

Boxplots summarizing IOPs measured with three rebound tonometers. Overall means (+/− SD) were 20.4 ± 7.5 mmHg (TV), 24.9 ± 8.9 mmHg (TVP), and 24.1 ± 7.7 mmHg (TVA). There was a significant difference between mean IOPs, with TV values being lowest (***, p < .001); there was no significant difference between TVP and TVA measurements. The boxes represent the middle 50% of data, with the centerline being the median. The bottom and top whiskers show the first and third quartile, respectively. Outliers are shown as small dots. The larger black dot inside the box represents the mean.

FIGURE 4

Tukey simultaneous tests for differences of IOP means. The black dots represent mean differences between rebound tonometers. Because the 95% confidence intervals (CI; horizontal lines) do not contain zero (vertical dashed line), the corresponding means are significantly different for TVP‐TV (4.56 ± 1.06 mmHg; CI: 2.08, 7.04) and TVA‐TV (3.7 ± 1.06 mmHg; CI: 1.22, 6.18), respectively (p ≤ .001). There was no significant difference between TVA and TVP means (−0.86 ± 1.06 mmHg; CI: −3.34, 1.62; p = .695).

FIGURE 5

Bland–Altman plots comparing three rebound tonometers. The estimated means of the differences between two tonometers are shown by the green horizontal lines. The 95% CI is marked by the red lines representing the upper (ULA) and lower (LLA) limits of agreement (mean ± 1.96 SD). Overall, the IOP measurements did not follow any trends and were mainly located within the 95% CI. The estimated means of the differences (±SD) were 4.56 ± 3.76 mmHg for TVP‐TV (A), 3.7 ± 3.03 mmHg for TVA‐TV (B), and − 0.86 ± 3.21 mmHg for TVA‐TVP (C).

FIGURE 6

IOP regression analyses between tonometers with linear equations and R² values. The red lines show strong linear correlation between TV and TVP (A), TV and TVA (B), and TVP and TVA (C) with the blue dashed lines marking the 95% prediction intervals. Compared to the black y = x line, TVP and TVA showed the best match (C); in contrast, TV underestimated most IOPs when compared to TVP (A) and TVA (B), respectively, with data points mostly located to the left of the y = x line.

FIGURE 7

Regression analyses between CCT and IOP with linear equations and R² values for TV (A), TVP (B), and TVA (C). The blue dashed lines mark the 95% prediction intervals. Even though the R² values were low, the positive linear relationships between CCT and IOP were significant for all three rebound tonometers (p < .005).