Background Scanning peripheral anterior chamber depth analyser (SPAC), IOLMaster and anterior segment optical coherence tomography (AS-OCT) are devices that can screen non-invasively for narrow angles (NAs) and may reduce the need for gonioscopy, especially where the availability of ophthalmologists is limited. When used individually to detect NAs, neither SPAC, AS-OCT nor IOLMaster were able to achieve both high sensitivity and specificity when using gonioscopy as a gold standard.
Aims To assess whether sequential testing improves discrimination of NAs over single testing.
Methods In a prospective community-based cross-sectional study in Singapore, 2047 phakic right eyes underwent SPAC, IOLMaster and AS-OCT. NA eyes were defined as present when Shaffer scores were 0 or 1 in ≥2 quadrants on gonioscopy. Sensitivities and specificities of single and sequential testing were compared.
Results At a fixed specificity of 95%, sensitivities in detecting NA eyes were 57.8% (95% CI 52.7% to 62.7%) for SPAC, 43.3% (95% CI 38.3% to 48.3%) for IOLMaster and 52.1% (95% CI 46.4% to 57.7%) for AS-OCT. After sequential testing using SPAC followed by AS-OCT, the sensitivity rose to 70.3% (95% CI 65.5% to 74.8%) while specificity remained at 94.3% (95% CI 93.1% to 95.4%), with sensitivity significantly higher than single testing alone (p<0.001).
Conclusion Sequential testing significantly improves the accuracy of detecting NAs and may be useful in screening settings.
- Anterior chamber
- public health
- diagnostic tests/investigation
Statistics from Altmetric.com
Funding Supported by grants from Singhealth, Singapore and National Medical Research Council, Singapore.
Competing interests KK has a Japanese patent on the SPAC (Japanese patent no. 3878164). TA has received funding, travel support and honoraria from Carl Zeiss Meditec. DSF has received an instrument loan from Carl Zeiss Meditec.
Patient consent Obtained.
Ethics approval This study was conducted with the approval of the Singapore Eye Research Institute.
Provenance and peer review Not commissioned; externally peer reviewed.