Modified ocular surface disease index as a screening criteria for dry eye syndrome presenting after successful dacryocystorhinostomy

OBJECTIVE: To find preoperative screening criteria for dry eye syndrome (DES) that present after successful endoscopic dacryocystorhinostomy (EDCR). METHODS: We retrospectively analyzed medical records of 110 patients who underwent EDCR for nasolacrimal duct obstruction. DES diagnostic criteria were defined as tear break-up time (TBUT) less than 10 seconds, and ocular surface disease index (OSDI) score greater than 13 points. After EDCR, patients were divided into DES group and control group according to the DES diagnostic criteria. Preoperative OSDI score alone or in combination of preoperative TBUT and OSDI score were used to find screening criteria, which could discriminate the two groups preoperatively with a high positive predictive value (PPV). Criteria A was set same as the diagnostic criteria of DES, and Criteria B and C were set to improve PPV by increasing specificity while maintaining similar sensitivity to Criteria A. RESULTS: Thirty patients (27.3%) were diagnosed with DES after EDCR, while 80 patients (72.7%) were normal. In patients with DES, preoperative TBUT was not different (p = 0.851), but OSDI score was significantly higher (p<0.001). Criteria A showed a sensitivity of 73.3%, specificity of 55.0%, and PPV of 38.0%. Criteria B and C excluded preoperative TBUT, which had no difference between the two groups, and set screening criteria with preoperative OSDI score alone. Criteria B (preoperative OSDI score of 19.6 points or more) showed a sensitivity of 75.0%, specificity of 60.6%, PPV of 41.7% and AUC of 0.739 (p < 0.001). Criteria C was consisted of 5 out of 12 OSDI items that showed significant differences between the two groups; blurred vision, reading, working with a computer, low humidity, and air conditioning. Criteria C (preoperative 5-item OSDI score of 24.4 points or more) was a better predictability, with a sensitivity of 75.0%, specificity of 71.3%, PPV of 49.5%, and AUC of 0.804 (p < 0.001). The AUC of Criteria C was significantly higher than that of Criteria B (p = 0.0037). CONCLUSION: DES occurred after successful EDCR in 27.3% of patients, and an OSDI questionnaire helped to screen DES. The predictability could improve using the modified OSDI score which showed noticeable difference in five OSDI items before and after EDCR.