Introduction: In a previous study we found that 90-day mortality after radical prostatectomy (RP) was low (39/22 344, 0.17%). However, one third of patients who died within 90 days died of prostate cancer according to Cause of Death Registry (CDR). It is unlikely that a patient would die with a disseminated prostate cancer (PC) within 90-days of curative surgery. To investigate this we conducted a validation study.
Methods: Through the National Prostate Cancer Register of Sweden (NPCR), we identified all men registered undergoing radical prostatectomy (RP) between 1st of Jan 1998 – 30th of June 2018 and died within 90 days after RP. Medical records from the surgery and from the time of death as well as autopsy reports were obtained. The information provided from the NPCR was compared with the data collected via the medical records.
Results: In NPCR, 45 148 patients were registered with a RP within the study period. 58 patients met the final criteria of death within 90 days of a RP, with a 90-day all-cause mortality rate of 0.13%. 16 (28%) patients died of PC according to CDR, however no patient had metastatic disease documented in their chart. The most common causes of death within 90 days in the chart review was heart disease (28%) and venous thromboembolic events (VTE) (19%). The 90-day all-cause mortality was 0.09% for Robotically assisted laparoscopic radical prostatectomy (RARP), and 0.18% for Retropubic Radical Prostatectomy (RRP) OR 1.97 (95%CI 1.1-3.4, p=0.002). RARP were however more performed in the later part of the study period, represented by the median year of 90-day mortality after RP was 2006 for RRP and 2013 for RARP. The risk of a VTE-related death with in 90 days was twice as high after RRP compared to RARP.
Conclusions: This study shows that there is ascertainment bias in the CDR since it is unlikely that a man would die of PC within 90-days after RP when no metastasis has been documented. The study show differences in rates and in type of mortality between RRP and RARP, however, due to the differences in time when the different surgeries were performed there are substantial risks of bias influencing these findings. The most common cause of death within 90-days of RP was related to cardiac diseases. Source of