Are adverse events during surgery for benign prostatic hyperplasia device related? A review of the MAUDE database.

PURPOSE: The Manufacturer and User Facility Device Experience database contains anonymous, voluntary medical device reports. A review of device-related adverse events associated with Benign Prostatic Hyperplasia surgeries was completed. The objective was to evaluate the occurrence and contributing factors to clinically significant complications in a cohort of patients electing to undergo surgical intervention for Benign Prostatic Hyperplasia. METHODS: The Manufacturer and User Facility Device Experience database was queried for "Aquablation, Greenlight Laser, Holmium Laser, Morcellator, Water Vapor Thermal Therapy, Loop Resection, and Prostatic Urethral Lift" from 2018 through 2021. A complication classification system (Level I-IV) based on the Clavien-Dindo system was used to categorize events. These events were then correlated with procedural technology malfunctions and classified as "device related" and "non-device related." Chi squared analysis was performed to identify associations between procedural technology and complication classification distribution. RESULTS: A total of 873 adverse events were identified. The adverse events were classified into level I (minimal harm) versus levels II-IV (clinically significant). Aquablation (p < 0.017) and Water Vapor Thermal Therapy (p < 0.012) were associated with a higher proportion of reports with Level II-IV complications compared with other procedure types. Level II-IV complications were not associated with a reported device related malfunction. CONCLUSIONS: Aquablation and water vapor thermal therapy demonstrated noteworthy clinically significant complications which were not driven by device-related malfunctions.

Relationship between operative duration and perioperative outcomes after radical cystectomy.

INTRODUCTION Prolonged operative times have been associated with an increased risk in complications in other major abdominal surgeries. This study tests the hypothesis that longer operative times will be associated with an increased risk in perioperative complications after radical cystectomy (RC). MATERIALS AND METHODS: Adult patients who underwent RC from January 1, 2012, through December 31, 2016, were identified from the National Surgical Quality Improvement Program (NSQIP) database. A natural log transformation was used to determine cutoff points for operative times at 33rd, 67th, and 90th percentiles: 272, 371, and 479 minutes, respectively. Cohorts were A (≤ 272 min), B (273-371 min), C (372-479 min), and D (> 479 min). Multivariable logistic regression analysis was performed to identify associations between operative time and perioperative complications. RESULTS: Among 5,610 patients, the distribution across cohorts was A, 1,993 patients; B, 1,818; C, 1,171; and D, 628. Cohort D had a higher incidence of pulmonary embolism (PE), deep vein thrombosis (DVT), urinary tract infection (UTI), sepsis, 30-day readmission, and blood transfusion rate and had a longer median hospital length of stay. Multivariable analysis showed that operative time (per 60 min) was associated with increased risk of DVT (OR 1.10, p = .04), PE (OR 1.15, p = .01), UTI (OR 1.08, p = .004), readmission (OR 1.04, p = .03), and blood transfusion (OR 1.23, p < .001). CONCLUSIONS: Longer operative times during RC are associated with a higher rate of perioperative complications. These findings may be confounded by disease stage, surgeon experience, variations in perioperative management protocols, or a combination of the above.

Variation in Lymph Node Yield After Radical Cystectomy.

OBJECTIVES: To test the hypothesis that lymph node yield will vary by pathology assistant (PA) in patients undergoing radical cystectomy (RC) with pelvic lymph node dissection (PLND). METHODS: This is a single-institution retrospective review that included patients who underwent an RC with PLND for bladder cancer from January 1, 2007, to January 1, 2018. Predicted mean lymph node counts were generated using multivariable regression analysis. RESULTS: In a total of 430 patients who underwent RC with PLND, the median lymph node count (interquartile range) was 15.0 (11.0-21.0). The frequency of the limits of lymphadenectomy was as follows: external iliac, internal iliac, and obturator (true pelvis) (33.3%); true pelvis plus common iliac to the level of the aortic bifurcation (47.9%); and inferior mesenteric artery (18.8%). On descriptive analysis, there were differences in lymph node yield when evaluating the following variables: level of dissection, clinical stage, neoadjuvant chemotherapy, surgical approach, surgeon, pathologist, and PA (P < .05). On multivariable analysis, adjusted lymph node counts varied between surgeons, pathologists, clinical stage, and level of dissection but not by PA (P = .18). CONCLUSIONS: Lymph node yield after RC varies on several known levels, including surgeon, extent of lymphadenectomy, clinical stage, and pathologist. This study found no significant variation in lymph node yield according to PA.

Outcome prediction following radical nephroureterectomy for upper tract urothelial carcinoma.

OBJECTIVE: To predict overall survival, cancer, and metastasis specific survival in upper tract urothelial carcinoma (UTUC) following radical nephroureterectomy (RNU). MATERIALS AND METHODS: All nonmetastatic UTUC patients who underwent RNU with a curative intent at 1 institution between December 1998 and January 2017 were included.  Detailed data were collected. End points for this study included OS, CCS, and MFS. Univariate and multivariate analysis were conducted. Log Rank tests and Kaplan-Meier curves were generated. Backward elimination and boot strapping was used to identify the most parsimonious model with the smallest number of variables in order to predict the outcomes of interest. A separate second institution data base was used for external validation. RESULTS: There were 218 patients in the development cohort. Mean follow-up was 42 months (±39.6). There was 99 (45.4%) deaths, 28 (12.8%) cancer related deaths, 72 (33%) recurrences, and 54 (24.8%) metastases. The c-index for our model was 0.71 for OS, 0.72 for MFS and 0.74 for CSS. The nomograms did not show significant deviation from actual observations using our calibration plots. We divided the patient into 3 different groups (low, intermediate and high risk) based on their final total score for each outcome and compared them. On external validation our accuracy was 78.4%, 71.4%, and 75.3% for OS, CSS, and MFS survival respectively. CONCLUSION: We designed a predictive model for survival outcomes following RNU in UTUC. This model uses simple, readily available data for patients without the need for expensive or additional testing.

How the Beneficial Effects of Alvimopan Differ With Surgical Approach for Radical Cystectomy.

OBJECTIVE: To assess whether the beneficial perioperative effects of alvimopan differ with surgical approach for patients who undergo open radical cystectomy (ORC) vs robot-assisted radical cystectomy (RARC). METHODS: This retrospective study reviewed all patients who underwent cystectomy with urinary diversion at our institution between January 1, 2007, and January 1, 2018. Data were collected on demographic characteristics, comorbidities, surgical approach, alvimopan therapy, hospital length of stay (LOS), days until return of bowel function (ROBF), and complications. Outcomes and interactions were evaluated through regression analysis. RESULTS: Among 573 patients, 236 (41.2%) underwent RARC, 337 (58.8%) underwent ORC, and 205 (35.8%) received alvimopan. Comparison of 4 cohorts (ORC with alvimopan, ORC without alvimopan, RARC with alvimopan, and RARC without alvimopan) showed that patients who underwent ORC without alvimopan had the highest rate of postoperative ileus (25.6%, P = .02), longest median hospital LOS (7 days, P < .001), and longest time until ROBF (4 days, P < .001). On multivariable analysis, the interaction between surgical approach and alvimopan use was significant for the outcome of ROBF (estimate, 1.109; 95% confidence interval, 0.418-1.800; P = .002). In the RARC cohort, multivariable analysis showed no benefit of alvimopan with respect to ileus (P = .27), LOS (P = .09), or ROBF (P = .36). Regarding joint effects of robotic approach and alvimopan, RARC had no effect on gastrointestinal tract outcomes. CONCLUSION: We observed a diminished beneficial effect of alvimopan among patients undergoing RARC and a statistically significant benefit of alvimopan among patients undergoing ORC. The implications of these findings may permit more selective medication use for patients who would benefit the most from this drug.