Progestin-only contraceptives may pose NAFLD risk among women with GDM history

The use of progestin-only contraceptives may put women with a history of gestational diabetes mellitus (GDM) at risk of nonalcoholic fatty liver disease (NAFLD), as shown in a small prospective study.

Around 1-year postpartum, median controlled attenuation parameter (CAP) on transient elastography was high at 307 dB/m among progestin-only contraceptive users and low at 213 dB/m among combined oral contraceptive users relative to 237 dB/m among participants who relied on nonhormonal contraceptive methods/abstained from contraception (p=0.035). [Contraception 2025;145:110860]

Significant difference in NAFLD prevalence was also observed across the three groups. Using a CAP threshold of ≥302 dB/m, NAFLD was found in 55.2 percent of progestin-only contraceptive users, 30.2 percent of nonhormonal users/abstainers, and 13.3 percent of combined oral contraceptive users (p=0.010).

Multivariable logistic regression analyses indicated that progestin-only contraceptives were associated with increased risk of NAFLD compared with no contraception/nonhormonal contraceptive use. This association was driven by a risk increase observed with the use of progestin-only pills (adjusted odds ratio [aOR], 8.19, 95 percent confidence interval [CI], 1.17–54.40) and depot medroxyprogesterone acetate (aOR, 6.45, 95 percent CI, 1.47–28.21) but not with implants.

On the other hand, combined oral contraceptives, which contained ethinylestradiol with either second- or third-generation progestins, had a null association with NAFLD risk.

“Our findings add to the growing body of evidence that hormonal contraceptives can have metabolic consequences for women with a history of GDM,” the investigators said.

They pointed out that the elevated NAFLD risk observed with progestin-only contraceptives may involve increases in weight or body fat mass. This was partially supported by a significantly higher median weight gain seen for progestin-only pill users than for implant users (5.2 kg vs −3.0 kg; p=0.035), with no significant difference between depot medroxyprogesterone acetate users and implant users (3.8 kg vs −3.0 kg; p=0.087). [Contraception 2013;88:503-508; EClinicalMedicine 2021;34:100800; Cochrane Database Syst Rev 2016;2016:CD008815]

“Another potential mechanism relates to triglyceride (TG) levels, which are key metabolic parameters associated with NAFLD. Similarly, but with minor differences in order, our study found the greatest decrease in median TG levels between baseline and 1 year postpartum in implant users, followed by depot medroxyprogesterone acetate users, while progestin-only pill users experienced an increase (−27, −22, and 14 mg/dL, respectively),” they added. [Biomed Rep 2014;2:633-636]

When selecting a contraceptive method for women with a history of GDM, the investigators emphasized that clinicians should carefully consider potential metabolic risks. The American Diabetes Association recommends highly effective contraceptive options, such as sterilization and long-acting reversible contraception (LARC), for this population. In line with this guideline, findings from the present study point to LARC methods (copper intrauterine devices and implants) and sterilization as more appropriate choices compared with progestin-only pills and depot medroxyprogesterone acetate, the investigators said. [Diabetes Care 2024;47:S282-S294]

“Furthermore, healthcare providers may need to regularly monitor liver health in women using hormonal contraceptives to detect early signs of NAFLD and manage its progression effectively. A tailored approach to contraceptive counselling can help mitigate long-term health risks while ensuring effective family planning,” they added.

The study included 130 participants, of which 86 (66.2 percent) used either no contraception or nonhormonal contraceptive methods, and 44 (33.8 percent) used hormonal contraceptives, including 29 progestin-only contraceptive users and 15 combined oral contraceptive users.

Among progestin-only contraceptive users, seven used progestin-only pills, 12 used depot medroxyprogesterone acetate, and 10 used implants. Among combined oral contraceptive users, five took formulations containing ethinylestradiol plus either levonorgestrel or norgestrel, and 10 used formulations with ethinylestradiol plus either desogestrel or gestodene. The median duration of contraceptive use was 8 months in the progestin-only group and 6 months in the combined oral group.