Ocular Surface Outcomes in Pseudophakic Patients with and Without Dry Eye Disease in Zambia: A Cross-Sectional Study

Abstract

<title>Abstract</title> <p>Background Dry eye disease (DED) may compromise ocular comfort and surface integrity after cataract surgery, but evidence from sub-Saharan Africa is limited. This study compared symptoms and ocular-surface findings in pseudophakic patients with and without DED during the first six postoperative months. Methods This prospective cross-sectional study enrolled 54 pseudophakic adults attending two tertiary hospitals in Lusaka, Zambia. Participants were classified as having DED (n = 27) or no DED (n = 27) using a symptom-and-sign framework. Outcomes included Ocular Surface Disease Index (OSDI) score, tear break-up time (TBUT), Schirmer I test, meibomian gland dysfunction grade, and fluorescein corneal staining graded using the Baylor system. Analyses included between-group tests, multivariable linear regression adjusted for age and sex, and within-group correlations. Results Mean OSDI score was higher in the DED group than in controls (29.41 ± 12.50 vs 6.47 ± 3.74; p &lt; 0.001), although OSDI contributed to group allocation. Mean total Baylor score was also higher in the DED group in the unadjusted analysis (9.15 ± 3.34 vs 7.44 ± 2.83; p = 0.048), and nasal staining was greater (median 2.0 [IQR 2.0–3.0] vs 1.0 [IQR 1.0–2.0]; p &lt; 0.001). After adjustment, DED status was not independently associated with total Baylor score (β = −1.484 for control vs DED; 95% CI − 3.25 to 0.29; p = 0.099), whereas age was positively associated with Baylor score (β = 0.109 per year; 95% CI 0.02–0.19; p = 0.013). Schirmer I test, TBUT, and Meibomian gland Dysfunction grade did not differ significantly between groups. OSDI was not significantly correlated with objective measures within the DED group. Conclusion Pseudophakic patients classified with DED had greater symptom burden and more corneal staining in unadjusted analyses, particularly nasally. The adjusted association with overall staining was not statistically significant. Larger longitudinal studies with preoperative baseline measurements are needed.</p>