A population-based analysis of predictors of penile surgical intervention among inpatients with acute priapism.

Abstract

While consensus exists regarding risk factors for priapism, predictors of operative intervention are less well established. We assessed patient and hospital-level predictors associated with penile surgical intervention (PSI) for patients admitted with acute priapism, as well as length of stay (LOS) and total hospital charges using the National Inpatient Sample (2010-2015). Inpatients with acute priapism were stratified by PSI, defined as penile shunts, incisions, and placement of penile prostheses, exclusive of irrigation procedures. Survey-weighted logistic regression models were utilized to assess predictors of PSI. Negative binomial regression and generalized linear models with logarithmic transformation were used to compare PSI to LOS and total hospital charges, respectively. Among 14,529 weighted hospitalizations, 4,953 underwent PSI. Non-Medicare insurances, substance abuse, and ≥3 Elixhauser comorbidities had increased odds of PSI. Conversely, Black patients, sickle cell disease, alcohol abuse, neurologic diseases, malignancies, and teaching hospitals had lower odds. PSI coincided with shorter median LOS (adjusted IRR: 0.62; p < 0.001) and lower ratio of the mean hospital charges (adjusted Ratio: 0.49; p < 0.001). Additional subgroup analysis revealed penile incisions and shunts primarily associated with reduced LOS (adjusted IRR: 0.66; p < 0.001) and total hospital charges (adjusted Ratio: 0.49; p < 0.001). Further work is required to understand predictors of poor outcomes in these populations.