The following is a summary of “Endocervical sampling in women with suspected cervical neoplasia: a systematic review and meta-analysis of diagnostic test accuracy studies” published in the December 2022 issue of Obstetrics & Gynecology by Damkjær et al.


Either endocervical brushing or endocervical curettage can be used to take an endocervical sample from a woman with cervical neoplasia. For a study, researchers sought to calculate the diagnostic precision, pain, and the number of insufficient samples for each test.

On June 9, 2022, 4 bibliographic databases were searched without regard to time or language. For women with a colposcopy indication, they compared endocervical curettage with endocervical brushing in all diagnostic investigations and randomized clinical trials. The grading of the Recommendations, Evaluation, Development, and Evaluation (GRADE) technique was used by two authors to independently screen studies, extract data, execute the risk-of-bias assessment (Quality Assessment of Diagnostic Accuracy Studies-2), and grade the confidence of the evidence. An investigation of diagnostic test accuracy using a bivariate random-effects model was conducted.

They included 7 research: 3 randomized clinical trials and 4 diagnostic cohort studies. Hysterectomy or conization was the gold standard. There was a substantial risk of bias and worry regarding the applicability of several studies on patient selection, flow, and timing. Endocervical brushing had an overall pooled sensitivity of 81% (95% CI, 48-95; 799 women; 7 studies; low quality of evidence), and endocervical curettage had an overall pooled sensitivity of 70% (95% CI, 42-89; 761 women; 7 studies; low quality of evidence). Endocervical brushing had an overall pooled specificity of 73% (95% CI, 36-93; 799 women; 7 studies; low quality of evidence), while endocervical curettage had an overall pooled specificity of 81% (95% CI, 56-94; 761 women; 7 studies; low quality of evidence). When comparing endocervical curettage to endocervical brush, the risk ratio for insufficient samples was 2.53 (95% CI, 0.58-11.0; P=.215; low-certainty evidence). Two trials examined patient discomfort; one indicated that the endocervical brush group experienced reduced discomfort, and the other found no difference.

There was no discernible difference between endocervical curettage and brushing in terms of side effects, insufficient sampling rates, or diagnostic accuracy. The external validity was constrained by the diversity of women’s traits and the ensuing diagnostic pathways.

Reference: ajog.org/article/S0002-9378(22)00592-0/fulltext