Weight change among women using intramuscular depot medroxyprogesterone acetate, a copper intrauterine device, or a levonorgestrel implant for contraception: Findings from a randomised, multicentre, open-label trial

Abstract

Background: There is limited evidence on the impact of the use of progestin-only hormonal contraception (POC) on weight change. We conducted a secondary analysis of prospective weight change among women enrolled in the Evidence for Contraceptive options and HIV Outcomes (ECHO) trial.

Methods: The ECHO trial was conducted at 12 sites in eSwatini, Kenya, South Africa and Zambia between December 2015 and October 2018. HIV negative, women aged 16-35 years, desiring contraception, were randomised (1:1:1) to either 3-monthly intramuscular depot medroxyprogesterone acetate (DMPA-IM), levonorgestrel (LNG) implant or copper intrauterine device (IUD). Follow-up was up to 18 months. Weight (kg) was measured at baseline and study exit. Analysis was performed as intention to treat (ITT) and time on continuous contraceptive use. The primary outcome of this secondary analysis is weight change from study enrolment to the final visit at study month 12-18. The ECHO trial is registered with ClinicalTrials.gov, NCT02550067.

Findings: 7829 women were randomly assigned to DMPA-IM (n = 2609), copper IUD (n = 2607) or LNG implant (n = 2613). The ITT population included 7014 women 2293 DMPA-IM group, 2372 copper IUD group and 2349 LNG group) who were not lost to follow-up, pregnant on study, or missing weight data. The mean weight increased in all groups but was significantly different in magnitude: 3.5 kg (SD = 6.3), 2.4 kg (SD = 5.9) and 1.5 kg (SD = 5.7) in the DMPA-IM, LNG implant and copper IUD groups, respectively. Comparative differences between groups were (2.02 kg (95% CI, 1.68, 2.36, p < 0.001) for DMPA-IM versus copper IUD, 0.87 kg (0.53,1.20 p < 0.001) for LNG implant compared to copper IUD and 1.16 kg (0.82, 1.50, p < 0.001) for DMPA-IM compared with LNG implant. Results for continuous contraceptive use were similar.

Interpretation: We found differences in weight gain between POC users compared to the non-hormonal copper IUD group over 12-18 months of use. Women using POCs should be counselled about this potential side effect when choosing a contraceptive method.

Keywords: Body mass index; Contraception; Copper intrauterine device; Depot medroxyprogesterone acetate; Levonorgestrel implant; Weight.

Conflict of interest statement

Dr. Baeten reports grants from USAID, grants from BMGF, during the conduct of the study; personal fees from Gilead Sciences, outside the submitted work; Dr. Smit reports grants from FHI360, during the conduct of the study; No other authors report any conflicts of interest.

© 2021 The Author(s).

Figures

Fig. 1

Trial profile. 1. Woman who were pregnant at any time during the study. 2. Missed a baseline or exit visit weight/height or data entry error that could not be resolved.

Fig. 2

Shifts in BMI categories from baseline to final measurement. Fig. 2 displays the number of women within each randomized arm falling into each BMI category at baseline and at final measurement; the proportion of women transitioning from one category to another over time are shown as lighter-intensity paths between baseline and final visit, with the thickness of each path indicating the size of the group falling into a given transition between categories from baseline to final visit.