Robotic-assisted benign hysterectomy compared with laparoscopic, vaginal, and open surgery: a systematic review and meta-analysis

Abstract

The potential benefits and limitations of benign hysterectomy surgical approaches are still debated. We aimed at evaluating any differences with a systematic review and meta-analysis. PubMed, MEDLINE, and EMBASE databases were last searched on 6/2/2021 to identify English randomized controlled trials (RCTs), prospective cohort and retrospective independent database studies published between Jan 1, 2010 and Dec 31, 2020 reporting perioperative outcomes following robotic hysterectomy versus laparoscopic, open, or vaginal approach (PROSPERO #CRD42022352718). Twenty-four articles were included that reported on 110,306 robotic, 262,715 laparoscopic, 189,237 vaginal, and 554,407 open patients. The robotic approach was associated with a shorter hospital stay (p < 0.00001), less blood loss (p = 0.009), and fewer complications (OR: 0.42 [0.27, 0.66], p = 0.0001) when compared to the open approach. The main benefit compared to the laparoscopic and vaginal approaches was a shorter hospital (R/L WMD: - 0.144 [- 0.21, - 0.08], p < 0.0001; R/V WMD: - 0.39 [- 0.70, - 0.08], p = 0.01). Other benefits seen were sensitive to the inclusion of database studies. Study type differences in outcomes, a lack of RCTs for robotic vs. open comparisons, learning curve issues, and limited robotic vs. vaginal publications are limitations. While the robotic approach was mainly comparable to the laparoscopic approach, this meta-analysis confirms the classic benefits of minimally invasive surgery when comparing robotic hysterectomy to open surgery. We also reported the advantages of robotic surgery over vaginal surgery in a patient population with a higher incidence of large uterus and prior surgery.

Keywords: Benign uterine pathology; Hysterectomy; Laparoscopic hysterectomy; Open hysterectomy; Robotic surgery; Vaginal hysterectomy.

Fig. 1

Flowchart accounting for each paper found in the initial searches, from identification through screening, showing the final number of articles included in the meta-analysis and the reasons for exclusion. RCT randomized controlled trial

Fig. 2

Risk of bias assessments for randomized controlled trials using the Cochrane Handbook risk of bias tools for randomized trials (RoB-2). Domain 1 deals with bias arising from the randomization process, Domain 2 deals with bias due to deviations from the intended interventions, Domain 3 deals with bias due to missing outcome data, Domain 4 deals with bias in measurement of the outcome, and Domain 5 deals with bias in selection of the reported result

Fig. 3

Forest plots for the robotic versus laparoscopic comparison showing the results of the overall pooled results and the study type subgroup analyses for a operative (OR) time in minutes (min), b estimated blood loss (EBL) in milliliters (mL), c blood transfusions, d length of hospital stay (LOS) in days, and e the number of patients requiring a hospital stay greater than 2 days. Lap laparoscopic, SD standard deviation, IV inverse variance, CI confidence interval, calc calculation, MD mean difference, hr hour, IQR interquartile range, OR odds ratio, RR risk ratio

Fig. 4

Forest plots for the robotic versus laparoscopic comparison showing the results of the overall pooled results and the study type subgroup analyses for a conversions to open surgery, b intraoperative complications (Intraop Comps), c mortality, d postoperative complications (Postop* Comps), e reoperations, and f readmissions. L or Lap laparoscopic, M–H Mantel–Haenszel, CI confidence interval, calc calculated, OR odds ratio, adj adjusted, NS not significant, RCT randomized controlled trial, R robotic, RR risk ratio, periop = perioperative, FU follow-up, LAVH laparoscopic-assisted vaginal hysterectomy, MIS minimally invasive surgery, TLH total laparoscopic hysterectomy. *Indicates that the postoperative complication rate within 30 days was extracted preferentially, but if the paper only reported a perioperative rate, it was included so as to not exclude that paper from the analysis

Fig. 5

Forest plots for the robotic versus open surgery comparison showing the results of the overall pooled results and the study type subgroup analyses for a operative (OR) time in minutes (min), b blood transfusions, c length of hospital stay (LOS) in days, and d the number of patients requiring a hospital stay greater than 2 days, e estimated blood loss (EBL) in milliliters (mL), f mortality, g intraoperative complications (Intraop Comps), and h postoperative complications (Postop* Comps). SD standard deviation, IV inverse variance, CI confidence interval, BMI body mass index, RR risk ratio, O open, R robotic, FU follow-up. *Indicates that the postoperative complication rate within 30 days was extracted preferentially, but if the paper only reported a perioperative rate, it was included so as to not exclude that paper from the analysis

Fig. 6

Forest plots for the robotic versus open surgery comparison showing the results of the overall pooled results and the study type subgroup analyses for a readmissions and b reoperations. M–H Mantel–Haenszel, CI confidence interval, O open, R robotic, OR odds ratio

Fig. 7

Forest plots for the robotic versus vaginal surgery comparison showing the results of the overall pooled results and the study type subgroup analyses for a operative (OR) time in minutes (min), b estimated blood loss (EBL) in milliliters (mL), c length of hospital stay (LOS) in days, d the number of patients requiring a hospital stay greater than 2 days, e mortality, f blood transfusions, g conversions to open surgery, h intraoperative complications (Intraop Comps), and i postoperative complications (Postop* Comps). SD standard deviation, IV inverse variance, CI confidence interval, V vaginal, R robotic, OR odds ratio, FU follow-up. *Indicates that the postoperative complication rate within 30 days was extracted preferentially, but if the paper only reported a perioperative rate (periop), it was included so as to not exclude that paper from the analysis

Fig. 8

Forest plots for the robotic versus vaginal surgery comparison showing the results of the overall pooled results and the study type subgroup analyses for a readmissions and b reoperations. M–H Mantel–Haenszel, CI confidence interval, V vaginal, R robotic, OR odds ratio