A retrospective analysis of factors associated with deep contraceptive implant removals compared to superficial removals.

Abstract

ObjectiveTo identify factors associated with the need for a deep etonogestrel contraceptive implant removal as compared to superficial removal. Study DesignWe conducted a retrospective cohort study of patients undergoing contraceptive implant removal from January 2014 to January 2023. We extracted key patient characteristics from electronic health record review and compared patients requiring deep removal versus routine superficial removal using Chi-squared, Fischer’s exact, and Mann-Whitney U test. A multivariate logistic regression identified variables associated with increased odds of requiring a deep implant removal. ResultsThe deep and superficial removal groups included 162 and 585 patients, respectively. Deep removal was associated with younger age at removal (median 25.0 vs 26.0 years, p=0.005), lower body mass index (BMI) at insertion (median 23.2kg/m2 vs 26.6kg/m2, p = 0.024), BMI >40kg/m2 at removal (15.2% vs 7.0%, p = 0.007), weight gain during implant use (median 6.6 vs 1.8 kg, p= <0.001), longer duration of use (median 36.0 vs 27.5 months, p <0.001), implant exchange (37.3% vs 17.4%, p <0.001), and insertion by non-physician (43.3% vs 19.3%, p <0.001) or non-OB/GYN (31.4% vs 11.8%, p <0.001). Lower BMI at insertion (aOR 0.92, [95% CI 0.87-0.98]), weight gain during use (aOR 1.06 [95% CI 1.02-1.10]), and longer duration of use (aOR 1.05 [95% CI 1.02-1.07]) remained significantly associated with deep removal in regression analysis. Conclusion(s)We identified lower BMI at insertion, weight gain during use, and longer duration of use as independent factors associated with increased likelihood of needing a deep contraceptive implant removal. IMPLICATIONSClinicians should utilize proper technique when inserting contraceptive implants, especially in patients at risk for deep insertion, and ensure immediate referral to Centers of Experience for patients with non-palpable implants.