Background and purpose: Seven percent of patients with prostate cancer (PCa) who are exposed to androgen deprivation therapy (ADT) may need transurethral resection of the prostate (TURP). Our objective was to examine the rate and the predictors of 30-day mortality (30dM) after TURP in patients who were exposed to ADT in a large, contemporary Canadian cohort.
Patients and methods: We assessed the 30dM rate after TURP in 853 men with the diagnosis of PCa who were treated with primary ADT or radiation therapy followed by ADT. The effect of age, comorbidity (coded according to the Charlson Comorbidity Index [CCI]), number of previous TURP procedures, history of radiation therapy, exposure to antiandrogens, and the type and the duration of ADT before TURP were all tested in univariable and multivariable logistic regression models that predicted 30dM after TURP.
Results: During the initial 30 days after TURP, 38 deaths occurred (4.5%, 95% confidence interval: 3.2%-6.2%). Of all variables, the CCI was the only statistically significant (P = 0.001) predictor of 30dM after TURP. The accuracy of CCI in predicting 30dM after TURP in individual patients was 65.1%. Lack of consideration of clinical variables that could predict the 30dM rate after TURP, such as prostate size or prostate-specific antigen level, represents a limitation of this study.
Conclusions: A substantial risk of 30dM is associated with TURP that is performed in patients who are exposed to ADT. Unfortunately, the predictors used in this analysis could not define the individual risk of 30dM with sufficient accuracy. Nonetheless, the average 4.5% risk should be considered at the time of informed consent.