Twenty-one studies, with approximately 4,200 participants (range 16 to 2020), were included in the review. Most included studies had an appropriate spectrum of participants and gave an adequate description of tests. Only two studies reported consecutive recruitment. Five studies reported using only minimal pelvic lymph node dissection as the reference standard or did not apply pelvic lymph node dissection in all patients (one further study did not report details of the reference standard).
The pooled estimate of sensitivity was 94% (95% CI 91 to 96%), based on 16 studies with no evidence of statistical heterogeneity. Sensitivity analysis of studies with 10 patients or more with involved pelvic lymph nodes gave similar results.
The pooled detection rate was 93.8% (95% CI 89 to 96.6%), based on 21 studies with high between-study heterogeneity (Ι² 90%). Sensitivity analysis, excluding two studies with sub-optimal labelling or low dose tracer, and subgroup analyses (laparoscopic and open) gave similar results.
Egger's test showed no evidence of publication bias. However, funnel plots were asymmetric, which indicated the presence of publication bias and Orwin's fail-safe N indicated that publication bias would be likely to have an effect, if present.