Contemporary rates and predictors of prolonged hospital stay after radical cystectomy: a population-based analysis.

Objective: To test contemporary rates and predictors of prolonged hospital stay after radical cystectomy (RC) and ileal conduit in non-metastatic bladder cancer patients.

Methods: Within the National Inpatient Sample database (NIS, 2008-2019), we identified ileal conduit RC patients and tabulated length of stay (LOS) ≥ 75th percentile vs. others. Temporal trends and multivariable logistic regression models (LRM) were fitted.

Results: Of 10,934 patients, 3,116 (28%) exhibited LOS ≥ 75th percentile (≥ 10 days), with rates decreasing from 35.4% in 2008 to 20.0% in 2019 (p < 0.001). In multivariable LRM, independent predictors of LOS ≥ 75th percentile were age ≥ 80 years (OR 1.37), CCI ≥ 2 (OR 1.55), coagulopathy (OR 1.55), obesity (OR 1.15), African American race/ethnicity (OR 1.57), female sex (OR 1.17), Medicare (OR 1.24) or Medicaid (OR 1.41) insurance, and treatment at low- (OR 1.68) or medium-volume hospitals (OR 1.33). Conversely, minimally invasive surgery (OR 0.59) exhibited the opposite protective effect (all p ≤ 0.03). Interestingly, the combined effect of age (≥ 80 vs. < 80 years) and surgical approach (minimally invasive vs. open) also achieved independent predictive status. Specifically, patients aged ≥ 80 undergoing minimally invasive, < 80 undergoing open, and ≥ 80 undergoing open surgery were, respectively, 1.35, 1.65, and 2.29 times more likely to require LOS ≥ 75th percentile compared to those aged < 80 undergoing minimally invasive surgery (all p ≤ 0.004).

Conclusion: The proportion requiring LOS ≥ 75th percentile decreased over time. Multivariable LRMs showed that age, race, insurance, hospital characteristics, and surgical approach were significantly associated with LOS ≥ 75th percentile. Combinations of variables, such as age ≥ 80 and open surgery, identified higher-risk subgroups.