Ex-vivo forces associated with intrauterine device placement and perforation: a biomechanical evaluation of hysterectomy specimens

BACKGROUND: This biomechanical analysis of hysterectomy specimens assesses the forces associated with intrauterine device placement. These include compressive forces required to cause uterine perforation with two commonly available commercial intrauterine device placement instruments and a metal uterine sound. METHODS: We obtained hysterectomy specimens at a single academic center. All specimens resulted from excision for benign conditions in premenopausal women by any operative method. Within one hour of excision, we stabilized uterine specimens in an apparatus specifically designed for this analysis. A single, experienced clinician performed all experimental maneuvers and measured forces with a Wagner FDIX-25 force gauge. The investigator applied traction on a tenaculum to approximate force used during an intrauterine device placement. The maximum compressive force to the uterine fundus was determined by using manufacturers’ placement instruments for two commercially available products and a metal sound. RESULTS: Sixteen individuals provided hysterectomy specimens. No complete perforations occurred while using loaded intrauterine devices; in a single observation the LNG IUS entered the myometrium. The plastic intrauterine device placement rod bowed in all attempts and did not perforate the uterine serosa at the fundus. A metal uterine sound created a complete perforation in all specimens (p < .001). The lowest mean maximum force generated occurred with the levonorgestrel intrauterine system placement instrument 12.3 N (SD ± 3.8 N), followed by the copper T380A intrauterine device placement instrument 14.1 N (SD ± 4.0 N), and highest for the metal sound 17.9 N (SD ± 7.6 N) (p < 0.01). CONCLUSIONS: In this ex-vivo model, metal uterine sounds caused complete perforation and intrauterine device placement instruments did not. This study received Institutional Review Board (IRB0059096) approval.