Effect of tear osmolarity on repeatability of keratometry for cataract surgery planning.

PURPOSE To evaluate the effects of tear osmolarity on the repeatability of keratometry (K) measurements in patients presenting for cataract surgery. SETTING Three clinical practices. DESIGN Observational prospective nonrandomized study. METHODS Subjects were prospectively recruited based on tear osmolarity (Tearlab Osmolarity System); that is, osmolarity more than 316 mOsm/L in at least 1 eye (hyperosmolar) and osmolarity less than 308 mOsm/L in both eyes (normal). The baseline K value was measured, and a second measurement was taken on the same instrument (IOLMaster) within 3 weeks of the first. Variability in average K, calculated corneal astigmatism using vector analysis, and intraocular lens (IOL) sphere power calculations were compared between groups. RESULTS The hyperosmolar group (50 subjects) had a statistically significantly higher variability in the average K reading (P = .05) than the normal group (25 subjects) and a statistically significantly higher percentage of eyes with a 1.0 diopter (D) or greater difference in the measured corneal astigmatism (P = .02). A statistically significantly higher percentage of eyes in the hyperosmolar group had an IOL power difference of more than 0.5 D (P = .02). No statistically significant differences were present when the subjects were grouped by self-reported dry eye. CONCLUSIONS Significantly more variability in average K and anterior corneal astigmatism was observed in the hyperosmolar group, with significant resultant differences in IOL power calculations. Variability was not significantly different when subjects were grouped by self-reported dry eye. Measurement of tear osmolarity at the time of cataract surgery planning can effectively identify patients with a higher likelihood of high unexpected refractive error resulting from inaccurate keratometry. FINANCIAL DISCLOSURE Drs. Epitropoulos, Matossian, Berdy, and Malhotra received compensation from Tearlab for participating in the study. No author has a financial or proprietary interest in any material or method mentioned.