Assessment of narrow angles by gonioscopy, Van Herick method and anterior segment optical coherence tomography

Abstract

To evaluate anterior chamber (AC) angles using gonioscopy, Van Herick technique and anterior segment optical coherence tomography (AS-OCT). One hundred forty-eight consecutive subjects were enrolled. The agreement between any two of three diagnostic methods, gonioscopy, AS-OCT and Van Herick, was calculated in narrow-angle patients. The area under receiver-operating characteristic curves (AUC) for discriminating between narrow and open angles determined by gonioscopy was calculated in all participants for AS-OCT parameter angle opening distance (AOD), angle recess area, trabecular iris surface area and anterior chamber depth (ACD). As a subgroup analysis, capability of AS-OCT parameters for detecting angle closure defined by AS-OCT was assessed in narrow-angle patients. The agreement between the Van Herick method and gonioscopy in detecting angle closure was excellent in narrow angles (κ = 0.80, temporal; κ = 0.82, nasal). However, agreement between gonioscopy and AS-OCT and between the Van Herick method and AS-OCT was poor (κ = 0.11-0.16). Discrimination capability of AS-OCT parameters between open and narrow angles determined by gonioscopy was excellent for all AS-OCT parameters (AUC, temporal: AOD500 = 0.96, nasal: AOD500 = 0.99). The AUCs for detecting angle closure defined by AS-OCT image in narrow angle subjects was good for all AS-OCT parameters (AUC, 0.80-0.94) except for ACD (temporal: ACD = 0.70, nasal: ACD = 0.63). Assessment of narrow angles by gonioscopy and the Van Herick technique showed good agreement, but both measurements revealed poor agreement with AS-OCT. The angle closure detection capability of AS-OCT parameters was excellent; however, it was slightly lower in ACD.