Comment on: Clinical characteristics, risk factor analysis, and outcomes of graft rejection after Descemet membrane endothelial keratoplasty

Dear Editor, We read with great interest the study by Singh et al.,[1] which evaluated graft rejection patterns following Descemet membrane endothelial keratoplasty (DMEK) in a large cohort with structured follow-up. The authors are to be commended for employing multivariable modeling and for including a robust comparison group of eyes without rejection at 2-year follow-up, thereby allowing for clearer risk stratification across real-world clinical scenarios. However, certain methodological and interpretive considerations merit further discussion. The selection of the control group, while valuable, introduces a potential survivorship bias. By limiting controls to eyes with a minimum 2-year follow-up and no rejection, the analysis may preferentially exclude early graft failures or cases lost to follow-up, both of which could distort the true background risk profile.[2] This bias may also affect the apparent strength of associations observed with modifiable factors such as corticosteroid discontinuation. Clinically, this distinction is critical because it may lead to an overestimation of protective effects or an underrepresentation of high-risk phenotypes encountered in routine care. The categorization of “rebubbling” as a discrete risk factor for rejection warrants closer examination. While the authors correctly identify rebubbling as a marker of postoperative complication, it likely functions more as a surrogate for underlying surgical complexity or graft manipulation rather than a true independent immunologic trigger. Without adjusting for baseline graft detachment severity, anterior segment anomalies, or intraoperative handling, the interpretation of rebubbling as a causative factor risks oversimplifying a multifactorial pathway.[3] From a therapeutic standpoint, this may inadvertently deter necessary interventions for graft repositioning. Additionally, the study’s approach to quantifying endothelial cell loss (ECL) post rejection appears restricted to a subset of 14 eyes, raising concerns regarding the representativeness of the reported mean loss of 72.5%. Given the heterogeneity in baseline diagnoses and severity of rejection episodes, a more granular stratification of ECL by indication, timing of presentation, and response to therapy would enhance the clinical applicability of these findings.[4] For instance, differentiating ECL in eyes with inflammatory indications from those with prior graft failures could inform prognosis and guide postrejection management. We commend the authors for generating one of the largest datasets on rejection after Descemet membrane endothelial keratoplasty in an Indian tertiary setting and for their careful alignment of surgical, clinical, and statistical domains. Future prospective studies incorporating immunologic markers, quantifiable rejection severity scores, and endothelial morphometric analysis may further refine the risk-adapted surveillance strategies that this study begins to outline. Financial support and sponsorship: Nil. Conflicts of interest: There are no conflicts of interest.