Comment on ‘Adverse in‐hospital outcomes in patients with paraplegia who undergo radical prostatectomy’

Marmiroli et al. 1 report a nationwide, hospital-based analysis of radical prostatectomy (RP) among men with paraplegia using the National Inpatient Sample (2000–2019). After 1:10 propensity score matching and hospital-clustered multivariable modelling, paraplegia remained associated with materially higher in-hospital mortality, multiple complication domains, transfusion, and prolonged length of stay compared with matched non-paraplegic patients, thereby quantifying perioperative hazard in a small but clinically vulnerable population. While the study provides valuable evidence, several issues merit further consideration. First, outcome ascertainment is restricted to events during the index admission. Because clinically important complications, re-interventions, and deaths frequently occur after discharge, an analytical horizon limited to the inpatient window may underestimate early postoperative burden. Where feasible, linkage to datasets with 30-day follow-up—or a complementary 30-day composite in a prospective registry—would align with contemporary surgical-quality convention and enable comparison with external series 2. Second, refinement of the exposure definition could reduce misclassification. The International Statistical Classification of Diseases and Related Health Problems (ICD) cluster used to identify ‘paraplegia’ includes codes denoting neurological entities that are distinct from spinal cord injury (SCI)-related paraplegia (e.g., cerebral palsy, hemiplegia) in addition to SCI-specific categories. A sensitivity analysis restricting to SCI-specific codes (e.g., G82.* and related subcodes), with re-estimation of associations under this narrower definition, would test robustness and improve specificity for the patient group most relevant to surgical decision-making. Third, presentation of effect measures could be augmented to improve clinical interpretability. Several endpoints are common in the matched cohorts (e.g., prolonged length of stay), a setting in which adjusted odds ratios can diverge from risk ratios and be misinterpreted at the bedside. In parallel with existing logistic models, estimating adjusted risk ratios via a modified Poisson approach with robust variance, and presenting adjusted absolute risks or risk differences via marginal standardisation, would quantify ‘how much more likely’ and ‘how many more per 100 patients’ without altering study design or covariates 3. In conclusion, Marmiroli et al. 1 provide robust, contemporary evidence from a nationwide cohort showing that RP in men with paraplegia—although uncommon—is associated with substantially higher in-hospital morbidity, mortality, and prolonged length of stay despite rigorous matching and hospital-level adjustment. By delineating risk across major complication domains and quantifying effect sizes, the study offers clear benchmarks for preoperative counselling, perioperative planning, and institutional quality monitoring in this high-risk population, thereby making a substantive contribution to the evidence base guiding clinical decision-making. The authors declare no conflicts of interest.