O’Brien RC, Ishwaran H, Szczotka-Flynn LB, et al. - In this investigation involving 1,090 participants (1,330 eyes), researchers sought to reanalyze types of intraoperative complications associated with Descemet stripping automated endothelial keratoplasty (DSAEK) graft failure in the Cornea Preservation Time Study using random survival forests. In the final model with 5 predictors, random survival forests rated DSAEK intraoperative complication as the third most predictive factor of graft failure, after surgeon and eye bank. Such findings, while post hoc, support the assumption that random survival forests allow for an improved analytic approach for recognising factors predictive of graft failure and for obtaining adjusted graft survival estimates. Random survival forests provide an opportunity to direct the development of future population-based cohort ophthalmic surgical studies, identifying definitive factors for procedural success.