Abstract

Text-Size + –

• INTRODUCTION:

  Data on the utilization of open, laparoscopic and robotic adrenalectomy on a national level is limited.

  MATERIALS AND METHODS:

  Data on patients who underwent open, laparoscopic, or robotic adrenalectomy for benign or malignant disease in the US from 2003-2013 were extracted using ICD-9 codes from the Premier Hospital Database. Surgeon specialty, patient demographics, hospital characteristics, and complications were compared. Data were analyzed using univariate and multivariable logistic regression analyses.

  RESULTS:

  A total of 8,831 adrenalectomies were performed for benign and malignant tumors. There was no significant difference in rate of adrenalectomy with regards to comorbidities, insurance status, or hospital characteristics. Non-urologists performed adrenalectomy more often for both benign (57% versus 43%; p = 0.011) and malignant disease (66% versus 34%; p = 0.011). Across all indications, non-urologists performed open surgery most often followed by laparoscopic and robotic approaches (56.3% versus 37.4% versus 6.4%, respectively), compared to urologists (48.8% versus 38.4% versus 12.9%, respectively). Overall, urologists were more likely to use laparoscopic or robotic approaches (p = 0.001). There was no difference in complication rates or operative times between surgical specialties or by surgeon/hospital case volume. On multivariable regression analysis, the best predictor of major complication was a Charlson Comorbidity Index (CCI) ≥ 2 (OR 3.9, 95%CI 2.1-7.1; p = < 0.001). Compared to open surgery, laparoscopy had significantly reduced odds of major complication (OR 0.6, 95%CI 0.3-0.9; p = 0.03). Patients undergoing robotic procedures had the shortest length of stay.

  CONCLUSION:

  In this retrospective study, adrenalectomy was more commonly performed by non-urologists via an open approach. Patients with CCI ≥ 2 were more likely to have postoperative complications while surgeon volume, hospital volume, and surgical approach did not influence complication rates.