Evolution of intraocular pressure after cataract surgery in nonglaucomatous patients: A post-hoc analysis of PERCEPOLIS clinical trial data

Purpose Intraocular pressure (IOP) drops after cataract surgery, including in nonglaucomatous eyes, but the mechanisms and factors determining this change remain unclear. Thus, post-hoc analysis was conducted on PERCEPOLIS, a large randomized-controlled trial showing non-inferior endothelial-cell loss between two cataract-extraction techniques in nonglaucomatous eyes. Methods Eyes with preoperative and 1-, 3-, and 12-month IOP measurements were identified. None received IOP-lowering medication before or after surgery. To identify possible mechanisms driving postoperative IOP drop, the cohort was categorized according to preoperative IOP (High: 20–29; Intermediate: 15–19; Low: 10–14 mmHg). IOP change in the whole cohort and three subgroups was determined. Multivariable analysis assessed whether age, sex, cataract density, preoperative IOP, anterior-chamber depth, axial length, lens thickness, lens position, relative lens position, cataract-extraction method, effective phaco time, or implant power predicted IOP drop at 3 months in the whole cohort and each subgroup. The literature was reviewed to determine the influence of regression-to-the-mean, a misleading statistical phenomenon, on postoperative IOP drop. Results Whole-cohort IOP dropped from 17.6 ± 3.5 mmHg by 1.9 ± 0.2 (1 month), 2.4 ± 0.2 (3 months), and 1.7 ± 0.2 (12 months) mmHg. IOP drop was greatest in the High subgroup. Preoperative IOP predicted 3-month IOP change in the whole cohort (beta = 0.53 ± 0.04, p < 0.001) and all subgroups (range: beta = 0.36–0.72 ± 0.15–0.30, p = 0.0505–0.001). Male sex also predicted larger whole-cohort IOP drop (beta = 0.79, p = 0.01). Different variables predicted greater IOP change in the three subgroups: male sex in High (beta = 1.63 ± 0.57, p = 0.005), higher implant power in Intermediate (beta = 0.19 ± 0.07, p = 0.01), and hard cataract in Low (beta = 2.14 ± 0.89, p = 0.02). Literature review suggested that regression-to-the-mean accounts for only a small proportion of IOP drop after cataract surgery. Conclusions The IOP drop after cataract surgery may largely be a real biological effect. In normotensive eyes, the strongest predictor of postoperative IOP drop was preoperative IOP. Distinct mechanisms may mediate the IOP change in normotensive eyes.