Perioperative outcomes and cost of robotic vs open simple prostatectomy in the modern robotic era: results from the National Inpatient Sample.

Abstract

To perform a comparative analysis of perioperative outcomes and hospitalisation cost between open (OSP) and robot-assisted simple prostatectomy (RASP) for treatment of benign prostatic hyperplasia (BPH) using the National Inpatient Sample (NIS) in the contemporary robotic era. The NIS was queried for cases of OSP and RASP for the treatment of BPH between 2013 and 2016. Perioperative complications, unadjusted hospital cost and length of stay (LOS) were compared between RASP and OSP. Smoothed linear regression curves comparing hospitalisation cost by increasing LOS was stratified by surgical approach to identify point of cost equivalency between RASP and OSP. Multivariable linear regression analysis was used to construct a hospitalisation cost model to examine the contribution of the robotic approach and LOS to hospitalisation cost. The total analytical cohort included 2551 OSP and 704 RASP procedures. Patients undergoing RASP were younger, at a median (interquartile range [IQR]) age of 68(63-73) vs 71(65-77)years, and with less comorbidity (76.8% vs 86.5%, P < 0.01). RASP was associated with fewer total complications (11.1% vs 29.2%, P < 0.01) and a greater likelihood of routine discharge to home rather than another facility (88.9% vs 76.7%, P < 0.01). While LOS was shorter with RASP (median [IQR], 2[1-3] vs 4[3-6]days, P < 0.01), total unadjusted hospitalisation cost (in United States dollars) was greater (median [IQR], $10855 [$7965-$15675] vs $13467 [$10572-$17722], P < 0.01). Presence of any complication increased both LOS and hospitalisation cost (P < 0.01). Linear regression modelling determined the point of cost equivalence between RASP staying a median of 2days was an OSP case staying between 5 and 6days. On multivariable regression analysis, the robotic approach contributed an additional $6175 (P < 0.01) to the cost model, whereas each additional day of hospitalisation contributed $1687 (P < 0.01), suggesting LOS would need to be 3-4days shorter with RASP to offset surgical costs of the robot. While RASP appears to have significantly better perioperative complication rates with shorter LOS and likely discharge to home, total hospitalisation cost remained greater, likely related to upfront operative costs. While this retrospective study is limited by selection bias for patients undergoing RASP, the benefits of improved convalescence, discharge to home, and lower rate of perioperative complications appear to justify performance of RASP in an experienced pelvic robotic centre despite relatively greater hospitalisation cost if referral to an experienced holmium laser enucleation of the prostate centre is not feasible.