Screening for Primary Angle Closure Disease in resource constraint region

Abstract

Purpose : Primary angle closure disease is quite often asymptomatic, reportedly more blinding than primary open angle glaucoma (POAG) and more prevalent in resource constraint regions. This study was designed to assess the performance of ocular biometry and van Herick (VH) grading assessed by teleophthalmology in identifying a gonioscopically occludable angle and to determine whether combining test results can improve the prediction of a gonioscopically occludable angle.

Methods : This observational, cross-sectional study was carried out at a primary eye care centre. Slit lamp photography and ultrasound biometry was performed by a trained vision technician. A masked ophthalmologist graded digital slit lamp photographs of the peripheral anterior chamber depth (ACD) by VH system. Eyes having VH grades 2 or less were classified to have narrow angles. Gonioscopy was performed using Sussman 4 mirror lens by one of the two experienced study optometrists. The agreement between a glaucoma specialist and each of the study optometrists for gonioscopy was good [kappa (k): 0.92 and 0.84] in a prior study.

Results : We included 1965 eyes of 1029 adult subjects. The intra-observer agreement in grading 100 randomly selected slit lamp photographs by VH grading (k: 0.76) was better than the inter-observer agreement (k: 0.56). The angle was occludable by gonioscopy in 101 (5.1%) eyes. The diagnostic accuracy of ACD at lowest quartile cut off was the highest [Sensitivity (Sn), Specificity (Sp), positive (PPV) and negative predictive value (NPV)] being 73.3%, 77.9%, 15.2 and 98.2, respectively. Similarly, Sn, Sp, PPV and NPV of VH classification were 52.5%, 92.8%, 28.2 and 97.3, respectively. Combination of biometric parameters (ACD and axial length at lowest quartile cut off; lens thickness at highest quartile cut off) and VH classification achieved Sp and PPV of 92.2% and 57.6, respectively. On the other hand, negative result of any of the biometric parameter at the above cut offs or VH classification achieved Sn and NPV of 92.1% and 99.1, respectively.

Conclusions : In isolation, van Herick test and ocular biometry demonstrated limited ability to screen for primary angle closure disease. However, test combination is a simple and inexpensive strategy to screen for angle closure disease in a resource constraint region with a downside of 42.4% false positive rate and 7.9% false negative rate at 5.1% prevalence of gonioscopically occludable angles.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.