ABSTRACT
PURPOSE: Laparoscopic cornuectomy, one of the most performed surgical techniques to remove Essure® implants, is associated with a risk of fragmentation of the device. We aimed to assess the diagnostic accuracy of the intraoperative radiograph of the removed Essure® microinserts compared with intraoperative inspection of the device. STUDY DESIGN: In this French retrospective cohort study including 66 patients, two experienced gynecologists and two residents evaluated the intraoperative radiograph in a blind manner. Diagnostic accuracy was calculated for each observer. RESULTS: The global sensitivity and specificity for intraoperative radiograph to detect a residual fragment for both senior gynecologists and residents were respectively 87.5% (±3.3) and 80.0% (+/12.2). Pooled positive (PPV) and negative predictive values (NPV) for intraoperative radiograph were respectively 99.2% (±0.5) and 21.6% (±6.4). There was a statistically significant difference in intraoperative radiograph analysis between senior gynecologists and residents (sensitivity: 91.4%±0.8 vs 79.6%±1.5, p=0.02; specificity: 100% vs 50.0%, p<0.0001; PPV: 100% vs 98.0% ±0.1, p=0.0007; NPV: 26.8% ±1.8 vs 7.2% ±0.5, p=0.009). Sensitivity and specificity of the intraoperative inspection of the removal specimen was respectively 95.3% and 100% (PPV and NPV were respectively 100 and 40%). There were no statistically significant differences in diagnostic accuracy between intraoperative radiographs analysis and intraoperative inspection of the device for the senior gynecologists. CONCLUSIONS: Intraoperative radiograph of the removed Essure® implants analysis is a difficult skill needing specific training. Postoperative pelvic X-ray is still recommended in clinical practice after laparoscopic cornuectomy.
