Survival After Minimally Invasive vs Open Radical Hysterectomy for Early-Stage Cervical Cancer: A Systematic Review and Meta-analysis

Abstract

Importance: Minimally invasive techniques are increasingly common in cancer surgery. A recent randomized clinical trial has brought into question the safety of minimally invasive radical hysterectomy for cervical cancer.

Objective: To quantify the risk of recurrence and death associated with minimally invasive vs open radical hysterectomy for early-stage cervical cancer reported in observational studies optimized to control for confounding.

Data sources: Ovid MEDLINE, Ovid Embase, PubMed, Scopus, and Web of Science (inception to March 26, 2020) performed in an academic medical setting.

Study selection: In this systematic review and meta-analysis, observational studies were abstracted that used survival analyses to compare outcomes after minimally invasive (laparoscopic or robot-assisted) and open radical hysterectomy in patients with early-stage (International Federation of Gynecology and Obstetrics 2009 stage IA1-IIA) cervical cancer. Study quality was assessed with the Newcastle-Ottawa Scale and included studies with scores of at least 7 points that controlled for confounding by tumor size or stage.

Data extraction and synthesis: The Meta-analysis of Observational Studies in Epidemiology (MOOSE) checklist was used to abstract data independently by multiple observers. Random-effects models were used to pool associations and to analyze the association between surgical approach and oncologic outcomes.

Main outcomes and measures: Risk of recurrence or death and risk of all-cause mortality.

Results: Forty-nine studies were identified, of which 15 were included in the meta-analysis. Of 9499 patients who underwent radical hysterectomy, 49% (n = 4684) received minimally invasive surgery; of these, 57% (n = 2675) received robot-assisted laparoscopy. There were 530 recurrences and 451 deaths reported. The pooled hazard of recurrence or death was 71% higher among patients who underwent minimally invasive radical hysterectomy compared with those who underwent open surgery (hazard ratio [HR], 1.71; 95% CI, 1.36-2.15; P < .001), and the hazard of death was 56% higher (HR, 1.56; 95% CI, 1.16-2.11; P = .004). Heterogeneity of associations was low to moderate. No association was found between the prevalence of robot-assisted surgery and the magnitude of association between minimally invasive radical hysterectomy and hazard of recurrence or death (2.0% increase in the HR for each 10-percentage point increase in prevalence of robot-assisted surgery [95% CI, -3.4% to 7.7%]) or all-cause mortality (3.7% increase in the HR for each 10-percentage point increase in prevalence of robot-assisted surgery [95% CI, -4.5% to 12.6%]).

Conclusions and relevance: This systematic review and meta-analysis of observational studies found that among patients undergoing radical hysterectomy for early-stage cervical cancer, minimally invasive radical hysterectomy was associated with an elevated risk of recurrence and death compared with open surgery.

Figure 1.. PRISMA Flow Diagram

PRISMA indicates Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Figure 2.. Meta-analysis Results

A and B, Confounder-adjusted hazard ratio (HR) (95% CI) for recurrence or death^,,,,,,,,,,,, (A) and for all-cause mortality^,,,,,,,,,,,, (B) among patients who underwent minimally invasive radical hysterectomy (compared with open surgery) is shown for individual studies and pooled results from meta-analysis. The box size corresponds to the weight of the study in the meta-analysis. The diamond depicts the point estimate (95% CI) of the pooled estimate. The vertical dotted black line is centered at the null, whereas the dashed black line is centered at the pooled HR estimate. Summing weights may not equal 100 due to rounding. NCRAS indicates National Cancer Registration and Analysis Service.

Figure 3.. Funnel Plots Assessing the Distribution of Study Results Against Study Uncertainty

A and B, The relative symmetry of association estimates (blue dots) around the pooled estimates (solid vertical line) suggests the absence of publication bias. Dashed lines depict the pseudo–95% CI around the pooled estimate, and the dotted light blue line corresponds to a null association. HR indicates hazard ratio; SE, standard error.