The World Health Organization-endorsed rapid assessment of avoidable blindness (RAAB) survey employs pinhole acuity to distinguish between refractive error versus conditions not correctable with eyeglasses, but few studies have validated this approach [1].

Methods

Ethical committees at the University of California, San Francisco and Narayana Nethralya Eye Hospital approved this study. A consecutive series of patients aged ≥50 years visiting the refraction clinic at Narayana Nethralaya Eye Hospital (Bangalore, India) in September 2015 had presenting vision and pinhole vision assessed using an ETDRS chart in a fully illuminated room, and then had a manifest refraction by an experienced optometrist. Analyses are reported with bootstrapped 95% confidence intervals with resampling at the participant level to account for non-independence of eyes.

Results

We assessed 204 eyes from 104 individuals (mean age 63.4 ± 7.8 years, 51.9% female). Mean spherical equivalent after manifest refraction was –0.31 (95% confidence interval [CI]: –0.07 to –0.60). Mean uncorrected visual acuity was 0.43 logMAR units (95% CI: 0.38–0.49), mean best corrected visual acuity (BCVA) was 0.16 (95% CI: 0.13–0.19) and mean pinhole acuity was 0.15 (95% CI: 0.12–0.18). Pinhole acuity had high agreement with BCVA (intraclass correlation coefficient 0.97, 95% CI: 0.96–0.98; Fig. 1). On average, pinhole acuity was less than a letter different from BCVA (mean 0.4 letters better, 95% limits of agreement by Bland–Altman method: four letters worse to five letters better). The magnitude of improvement on pinhole testing was correlated with the magnitude of spherical equivalent from refraction (Spearman’s rho = 0.68, P < 0.001; Fig. 2).

Fig. 1
figure 1

Correlation between measurements of best spectacle corrected visual acuity and pinhole acuity. Points are weighted; the area of the circles represents the number of observations at each coordinate

Full size image
Fig. 2
figure 2

Relationship between improvement with pinhole occlusion and refractive error. Eyes were stratified according to how many lines of improvement were achieved with pinhole occlusion. The distribution of spherical equivalent for each stratum is depicted as a box-and-whiskers plot

Full size image

Of 204 eyes, 21 (10.3%) had visual impairment even after subjective refraction (BCVA worse than 20/60). When treated as a diagnostic test for visual impairment not correctable with eyeglasses, pinhole acuity provided high discriminative ability, with an area under the receiver operating characteristics (ROC) curve of 0.99 (95% CI: 0.97–1.0). Pinhole acuity worse than 20/60 was 85.7% sensitive (95% CI: 59.1–100%) and 100% specific (98.0–100%) for detecting visual impairment not correctable with glasses, and had a positive predictive value of 100% (95% CI: 81.4–100%).

Conclusions

Previous studies of patients with diabetic retinopathy or low vision found that pinhole acuity was biased relative to BCVA, with pinhole acuity ~1 line worse than BCVA [2, 3]. In contrast, we found essentially no bias in this population from a refraction clinic in India. The reason for the discrepancy is unclear, but the pinhole occluder may cause more visual degradation in those with retinal pathology than in a general population like ours [2]. The estimates of sensitivity and specificity of pinhole acuity as a test for visual impairment are consistent with prior reports, and suggest that estimates of refractive error based on pinhole occlusion should not overestimate the prevalence of disease [4].

In summary, pinhole acuity agreed well with BCVA and was a specific test for visual impairment not correctable with eyeglasses. These results suggest that pinhole occlusion is a valid gauge of refractive error in the RAAB survey or other community-based surveys.