Complications in robotic-assisted gynecologic surgery according to case type: a 6-year retrospective cohort study using Clavien-Dindo classification

Study objective: To estimate the risk of postoperative complications in robotic-assisted gynecologic surgery according to case type.

Study design: Retrospective cohort study (Canadian Task Force classification II-2).

Setting: Mayo Clinic Arizona.

Patients: All 1155 patients who underwent robotic-assisted gynecologic surgery between March 2004 and December 2009 were included. Patients were primarily white (94.3%), with a mean (SD) age of 51.5 (15.4) years, and were overweight, with body mass index (BMI) of 27.2 (6.8).

Interventions: Risk of complications, overall and according to Clavien-Dindo grade, and incidence of specific complications were analyzed. Robotic-assisted gynecologic surgical procedures were categorized postoperatively according to case type as benign simple (e.g., oophorectomy, simple hysterectomy) in 552 (47.8%) patients, benign complex (e.g., excision of invasive endometriosis) in 262 (22.7%), urogynecologic in 121 (10.5%), and oncologic in 220 (19.1%).

Measurements and main results: Intraoperative complications occurred in 3.2% of patients. Postoperative complications of any type occurred in 18.4% of patients. Conversion to laparotomy was necessary in 2.7%. Urologic complications were more common in urogynecologic cases (5.8%) as compared with benign simple (0.5%), benign complex (2.7%), and oncologic (3.2%). Bleeding complications were most common in oncologic cases (5%). Clavien-Dindo grade ≥ 3 complications occurred in 5.2% of patients overall, and were >3-fold likely to occur in benign complex, urogynecologic, and oncologic cases than in benign simple cases. When adjusted for age, BMI, estimated blood loss, operative time, length of stay, and previous pelvic surgery, complications were nearly twice as common for benign complex (odds ratio [OR] 1.7; 95% confidence interval [CI], 1.1-2.7), urogynecologic (OR 1.9; 95% CI, 1.0-3.4), and oncologic (OR 1.9; 95% CI, 1.1-3.1) cases as for benign simple cases, although weakly significant. Case type, BMI, estimated blood loss, and length of stay remained important factors in predicting postoperative complications.

Conclusion: The incidence of complications in robotic-assisted gynecologic surgery varies according to case type. Defining the role of patient and surgical variables such as case type in the occurrence of complications may help in identification of cases with increased risk, to improve patient counseling and surgical outcome.