Repeatability and reproducibility of TonoLab rebound tonometer in DBA/2J mice / 中华实验眼科杂志

ABSTRACT

Background Reliable intraocular pressure (IOP) measurement is essential in glaucomatous study.TonoLab is a newly designed tonometer for small animals,however,its repeatability and reproducibility remains unknown.Objective This study aimed to evaluate the repeatability and reproducibility of TonoLab rebound tonometer in DBA/2J mice.Methods Intraocular pressure was measured in both eyes of 63 DBA/2J mice aged 7 months under systemic anesthesia with TonoLab rebound tonometer.IOP of each eye was measured for 3 times by two observers independently and consecutively using TonoLab rebound tonometer,and the data were obtained in single blind method.Within standard deviation (Sw),interclass correlation coefficient (ICC),within coefficient of variation (CVw) and corresponding 95％ confidence interval (CI) were calculated to evaluate the repeatability of measuring result from single observer,and Bland-Altman method was used to assess the agreement between two observers.The abnormal data which affected repeatability was analyzed.Results Among the 126 eyes,105 eyes showed the normal IOP with mean value (12.98 ± 1.05) mmHg from observer 1 and (12.41 ±0.92) mmHg from observer 2.Ocular hypertension was found in 21 eyes with mean value (27.76 ± 1.95) mmHg and (27.49 ± 1.28) mmHg for observer 1 and observer 2 respectively.Sw,ICC and CVw for observer 1/observer 2 were 1.34/1.11 mmHg,0.90/0.92,8.24％/7.42％ in normal eyes and 2.21/1.49 mmHg,0.89/0.95,6.97％/4.82％ in hypertensive eyes.The average bias between two measuring values from different observers was 0.60 mmHg in the normal eyes (95％ CI-5.20-6.30 mmHg) and 0.30 mmHg in the hypertensive eyes (95％ CI-7.30-7.80 mmHg).Conclusions TonoLab rebound tonometer can offer a good intraobserver repeatability in both normal and hypertensive eyes in mice,but the agreement between two observers is outside an acceptable level.So we consider IOP in mice obtained by TonoLab rebound tonometer is reliable and accurate from a single observer.