[Literature Interpretation] Accuracy Validation Study of iCare TONOVET Plus and TONOVET Pro Tonometers in an Ex Vivo Cynomolgus Monkey Model: Comparison Between Canine and Rabbit Calibration Modes

Release Date:

Nov 18,2026

Author:

Source:


Abstract

 

Background

Accurate measurement of intraocular pressure (IOP) is fundamental to glaucoma diagnosis and ocular toxicity evaluation in preclinical studies. Nonhuman primates (NHPs), especially cynomolgus monkeys (Macaca fascicularis), are a core animal model in translational ophthalmology research because their eyes are highly homologous to human eyes in anatomical structure and physiological characteristics. Critically, NHPs are the only species among widely used preclinical animal models that possess a true macula with a fovea, making them indispensable for evaluating the pharmacokinetics, efficacy, and safety of novel retinal therapeutics. For example, the landmark preclinical development of anti-vascular endothelial growth factor (VEGF) agents for age-related macular degeneration (e.g., ranibizumab and aflibercept) relied heavily on NHP models to evaluate ocular toxicity and pharmacological profiles. In addition, NHPs play a key role in preclinical safety and biodistribution assessment of subretinal gene therapy, including voretigene neparvovec, which has been approved by the FDA for inherited retinal dystrophy. With the recent surge in development of ocular drugs administered by intravitreal and subretinal injection, accurate IOP monitoring in NHPs has become increasingly critical in safety evaluation, because these routes often cause transient or persistent IOP elevation. Rebound tonometry is widely used in laboratory animals; it enables rapid, noninvasive IOP measurement without topical anesthesia. The iCare TONOVET (TV; Icare Finland Oy, Helsinki, Finland) animal tonometer has been validated in multiple species and is widely used. This study aimed to evaluate the accuracy of iCare TONOVET Plus (TVP) and TONOVET Pro (TVPR) for IOP measurement in ex vivo cynomolgus monkey eyes, using invasive gold-standard hydraulic manometry as the reference for true IOP, and to compare the measurement accuracy of the canine and rabbit calibration modes.

Methods

The eyes used in this ex vivo study were obtained from 3 male cynomolgus monkeys (Macaca fascicularis), weighing 3.4–4.7 kg and aged 4 years. The animals were euthanized for reasons unrelated to this study, and the animal protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of Keyprime Research (approval number: IACUC-AP-24-018). All 6 eyes were enucleated immediately after euthanasia. All eyes showed no abnormalities on clinical examination before euthanasia, but 1 eye was excluded due to marked corneal opacity during tissue preparation. Therefore, 5 eyes were included in the final statistical analysis. After enucleation, the eyes were stored in 0.9% saline at 4°C and equilibrated to room temperature before measurement. All hydraulic manometry experiments were completed within 6 hours after enucleation to minimize postmortem histomorphological changes.

Anterior chamber cannulation was performed, and IOP was controlled by hydraulic manometry over a range of 5–80 mmHg. IOP was increased stepwise from 5 to 80 mmHg; data were collected in 5 mmHg increments from 5 to 40 mmHg and in 10 mmHg increments from 40 to 80 mmHg. Linear mixed-effects models (LMM), Lin's concordance correlation coefficient (CCC), and repeated-measures Bland–Altman analysis were used to analyze agreement between rebound tonometer readings and manometric readings. To avoid artifactual bias from the device's maximum display limit, data points at the manometry set pressure of 80 mmHg were excluded.

Two rebound tonometers were used: iCare TONOVET Plus (TVP) and TONOVET Pro (TVPR) (Icare Finland Oy, Helsinki, Finland). Each device was used to collect data in two different calibration modes: canine (dog) mode and lapine (rabbit) mode. At each manometric pressure step, IOP was allowed to stabilize for approximately 1 minute. All measurements were performed by the same experienced operator, who was blinded to the manometric pressure values. The tonometer probe was positioned perpendicular to the center of the cornea, with a probe-to-cornea distance of 4–8 mm. At each pressure level, three consecutive measurements were obtained for each device mode, and the average was used for subsequent analysis. During the experiments, the corneal surface was periodically moistened with balanced salt solution to prevent corneal drying.

Statistical analyses were performed using Python (v3.10) with the statsmodels and matplotlib libraries. At a manometric set pressure of 80 mmHg, the device reached its maximum display limit (99 mmHg), and data points at this pressure were excluded from all relevant statistical analyses to avoid artificial distortion. LMM was used to analyze the correlation between tonometer measurements and manometric IOP values, with Eye ID as a random effect to account for repeated measurements. Marginal R² was used to assess model goodness of fit. Lin's CCC (ρc) was used to assess measurement agreement. To estimate systematic and proportional bias and to correct for within-subject variance, repeated-measures Bland–Altman analysis was performed, and mean bias and 95% limits of agreement (LoA) were calculated. P < 0.05 was considered statistically significant.

Results

A total of 720 independent IOP readings were collected (three repeated measurements per pressure step). The results are summarized in Table 1, which lists descriptive statistics (mean and standard deviation [SD]) for each manometric set pressure under all device and calibration mode combinations. However, data points at the manometric set pressure of 80 mmHg were excluded from subsequent statistical modeling because the device reached its maximum display limit of 99 mmHg at this pressure, to mitigate a potential ceiling effect and avoid artificial distortion of the results. Therefore, the final inferential statistical analyses (including LMM, CCC, and repeated-measures Bland–Altman analysis) were based on 220 data points (5 eyes × 11 pressure steps × 4 device/mode combinations). All data included in the statistical models were normally distributed (Shapiro–Wilk test, p > 0.05).

Across all calibration modes, manometric IOP values showed a strong linear correlation with rebound tonometer measurements. However, the rabbit mode showed proportional overestimation, with a regression slope higher than that of the canine mode. The canine mode showed excellent agreement (CCC = 0.967 for TVP and CCC = 0.971 for TVPR), with minimal mean bias. In contrast, the rabbit mode had a clear positive bias and wider limits of agreement, especially in the high IOP range.

Conclusion

When the canine calibration mode was used, both iCare TVP (Plus) and TVPR (Pro) provided accurate and reliable IOP measurements in cynomolgus monkeys. Although the central corneal thickness of cynomolgus monkeys is closer to that of rabbits, the rabbit mode caused significant IOP overestimation and is not recommended. Under the settings tested in this study, the canine mode yielded high IOP measurement accuracy with minimal bias, whereas the rabbit mode markedly overestimated IOP values.

Ethical Statement

The monkey eyes used in this study were provided by Keyprime Research and were obtained after euthanasia of animals for another unrelated study. That independent study protocol was approved by the Institutional Animal Care and Use Committee of Keyprime Research (IACUC approval number: AP-24-018). Keyprime Research has maintained full accreditation from AAALAC International since 2024.

Conflict of Interest Statement

The authors declare no conflicts of interest.

Publication Information

Authors: Wontae Kim, Seonmi Kang, Jiwon Lee, Jun Young Lee, Kangmoon Seo

Journal: Veterinary Ophthalmology

07 September 2026

doi: https://doi.org/10.1111/vop.70265

Original article link:

https://onlinelibrary.wiley.com/doi/epdf/10.1111/vop.70265

Keywords:

PREVIOUS

None

NEXT

Congratulations to the School of Continuing Education of Shanghai Jiao Tong University School of Medicine for successfully organizing the “Laboratory Animal Practitioner Professional and Technical Grade Training” and “Small Animal Cardiovascular Model and

YEERAN

GET QUOTE

Please check your spam folder in the coming days to ensure our response is not missed.