Hysteroscopic tissue removal device versus endocervical curettage for endocervical canal sampling: a retrospective...

Assessment of the endocervical canal remains a relevant limitation of colposcopic practice, particularly in women with type 2 or type 3 transformation zones. Conventional endocervical curettage is widely used for endocervical sampling, but its blind nature may result in inadequate or poorly representative histological specimens.

To compare hysteroscopic tissue removal device sampling performed under direct endocervicoscopic visualization with conventional endocervical curettage for endocervical canal assessment.

This retrospective comparative study included women referred to the Colposcopy Outpatient Clinic of the AOU Federico II, Naples, Italy, between November 2022 and June 2024. Eligible patients fulfilled at least one predefined indication for endocervical canal assessment based on cervical cytology, hrHPV persistence, and, where applicable, transformation-zone type or squamocolumnar-junction visibility. During the initial phase of the study period, conventional endocervical curettage represented the standard institutional approach. After the hysteroscopic tissue removal device became available and was integrated into the outpatient diagnostic pathway, eligible patients underwent endocervicoscopy followed by tissue removal device sampling. The primary endpoint was histological sample adequacy. Secondary endpoints included stromal representation, patient tolerability assessed by visual analogue scale, procedure duration, and concordance with endocervical cytobrush cytology.

Seventy-two patients were included (36 per group). Histological adequacy was higher with tissue removal device sampling than with curettage (34/36, 94.4 % vs 12/36, 33.3 %; absolute difference, 61.1 percentage points; 95 % CI, 40.5 to 74.8; p < 0.0001). Stromal representation among all attempted samples was also higher (94.4 % vs 22.2 %; difference, 72.2 percentage points; 95 % CI, 52.0 to 83.5; p < 0.0001). Median VAS score was lower (3.0 [IQR, 2.0-5.0] vs 6.0 [IQR, 5.0-7.0]; Hodges-Lehmann shift, -3.0; 95 % CI, -4.0 to -2.0; p < 0.001), whereas median procedure duration was longer (1.38 [IQR, 1.20-2.00] vs 1.08 [IQR, 0.82-1.25] minutes; Hodges-Lehmann shift, 0.38 min; 95 % CI, 0.18 to 0.62; p < 0.001). Crude concordance with cytobrush cytology was higher with tissue removal device sampling, but this analysis was exploratory.

In this retrospective, non-randomized comparison, hysteroscopic tissue removal device sampling under direct endocervicoscopic visualization was associated with higher specimen adequacy, greater stromal representation, and lower procedural discomfort than conventional endocervical curettage. These preliminary findings concern specimen quality and tolerability, not diagnostic accuracy. Prospective multicenter studies with contemporaneous or randomized groups and a uniform surgical reference standard are required before this technique can be recommended as a replacement for conventional curettage.