Reduced stereoacuity as a predictor for clinically significant convergence insufficiency

Stereoacuity relies on accurate binocular alignment. Convergence insufficiency (CI) a binocular motor disorder, interferes with near work.

To investigate the association between convergence amplitude (CA) and stereoacuity in a large paediatric cohort.

Retrospective chart review included patients aged 6-17 years; excluded patients with amblyopia, manifest strabismus or visual acuity <20/30 in either eye. Stereoacuity, measured by Randot test was defined as normal (≤40arcsec), subnormal (50-400arcsec) and poor (>400 arcsec). CA, measured using base out prism bar was defined by fusion break point (BP) and recovery point (RP), as none (BP=0), poor (BP <20 prism diopter (PD)), borderline (BP <30 PD or RP <20 PD), good (BP ≥30 PD and RP ≥20 PD) and excellent (does not break at 40PD).

In 2200 subjects included, we found an increased prevalence of normal stereoacuity as convergence ability improves (χ² test, p<0.001) with a negative correlation between stereoacuity and BP (Pearson correlation -0.13, p<0.001).CI was significantly associated with below normal stereopsis OR 1.86 (95% CI 1.3 to 2.7, p<0.001). Conversely, prevalence of CI was similar, whether or not CI-symptoms were reported. Follow-up data of at least 2.5 years from presentation was available for a small subgroup of 21 patients treated for CI. Convergence improved in 14 (66%), rate of normal stereoacuity increased from 29% at baseline to 76% at last follow-up (p=0.006).

CA affects stereoacuity function in children. Evaluation of CA is required in all cases with poor stereoacuity, especially when other etiologies are amiss. The role of convergence improvement exercise on stereoacuity warrants further investigation.