Contraception and Diabetes Mellitus

Abstract

The aim of this narrative review was to discuss data on contraception in diabetes mellitus (DM). Women with DM rarely discuss contraception with their physicians, and healthcare providers offer advice to a very limited number of them. Overall, 1 in 8 women with DM using contraception methods was found to use an ineffective one. A further issue relates to drug-drug interactions between anti-diabetic medications and oral contraceptives. Generally, anti-diabetic agents do not alter the pharmacologic profile of hormonal contraception. However, preliminary results indicate that some novel anti-diabetic agents may even render oral contraceptive methods ineffective. Several implants can be also generally used by women with both DM types. The relationship between oral contraceptives and diabetic complications has not been clarified yet. In general, implants, intra-uterine devices or progestin-only contraceptives are considered safe options for women with DM. However, short-term use of combined hormonal contraception is also feasible for women without severe complications or risk factors.

Keywords

type 1 diabetes mellitus - type 2 diabetes mellitus - contraception - drug-drug interaction - fertility

Introduction

Women with diabetes mellitus (DM) still have the risk of suboptimal pregnancy-related outcomes [1] [2]. These include major congenital malformations [1] [2]. Stringent glycaemic control, especially in the immediate pre-pregnancy and pregnancy period, is vital to improve pregnancy-related outcomes [1] [2] [3]. Ideally, pregnancies should be planned after intensified glycaemic control and screening for chronic diabetic complications [1] [3]. Adequate planning to avoid unwanted pregnancies may be improved by the use of contraception when required, as recommended by the American Diabetes Association (ADA) [3].

In this endeavour, birth control methods and counselling gain importance. This is now reinforced by the accumulating data on the effect of some new antidiabetic agents on fertility potential [4].

However, several studies have shown a very low rate of contraception-related discussions between women with DM and their physicians, although the majority (89%) of the latter acknowledges the importance of counselling on contraception [3].

Therefore, the present narrative review aims to discuss data on contraception in DM.

Methods

We searched Scopus, PubMed, MEDLINE, and Google Scholar for articles until the 26^th of July 2024, using combinations of the following key words: “contraception”, “type 1 diabetes mellitus”, “type 2 diabetes mellitus”, “drug-drug interaction”. All types of articles (clinical trials, meta-analyses, prospective studies, retrospective studies, randomised controlled trials, cohort studies, cross-over studies, and case-control studies) from the years 2000 to date were included. Only original articles in English were considered.

Frequency of contraception in diabetes mellitus: questionnaire-based studies

Several questionnaire-based studies have assessed the beliefs of women with DM on contraception, as well as their prior or current birth control use. Data from two observational studies and one prospective descriptive study are included in this review [5] [6] [7]. An overview of the studies is presented in [Table 1].

Table 1 Questionnaire-based studies on reproduction and contraception in diabetes mellitus.

Study	Study type	Objectives	Population	Outcomes	Secondary outcomes
Feutry et al. [5]	Observational study	Assessment of contraception perceptions and use among women with DM	89 T1DM participants with mean age of 27.9 years 33 T2DM participants with mean age of 32.6 years	Effective contraception was overall reported by 66.4% 15% of participants entitled to contraception did not receive contraception 12.5 % of participants receiving contraception were found to use a non-proper method	75% of participants reported prior knowledge regarding pregnancy risks, whereas 67% were informed regarding the necessary pregnancy plan 33% of participants knew the recommended HbA1c for conception 65.9% among T1DM participants were significantly better informed on pregnancy and contraception compared with T2DM participants
Schwarz et al. [6]	Observational study	Evaluation of perceptions regarding contraception among young women with T1DM	89 T1DM participants with a mean age of 16 years	16% of participants had received contraception at least once, whereas 14% received contraception during the conduction of the study	36% of participants protested about limited contraception choices and 31% about the difficulty in receiving birth control 43% of participants considered all contraceptive methods less effective in diabetes Only 47% of participants previously discussed contraception, while 29% have never received prior information 69% of young T1DM participants were eager to discuss birth control issues
Osman et al. [7]	Prospective descriptive study	Assessment of reproductive health perceptions among women with DM	115 DM participants	10.4% of participants had never used contraception and further 33.0% of participants did not receive contraception during the study	26.1% of participants received birth control-related information, despite the fact that 88.7% visited regularly diabetes clinics Better awareness was found among T1DM participants 44.2% of participants reported unintended pregnancies

DM, diabetes mellitus; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; HbA_1c, glycated haemoglobin.

In terms of awareness, 75% of women with both DM types in the most recent study by Feutry et al. [5] reported prior knowledge of pregnancy risks, whereas 67% acknowledged the need for pregnancy planning to achieve optimal pregnancy outcomes. In addition, only 33% of participants were aware of optimal levels of glycated haemoglobin (HbA_1c). In the study by Schwarz et al. [6], contraception was discussed by 47% of women with type 1 diabetes mellitus (T1DM) in their medical reviews. Likewise, Osman et al. [7] found that only 26.1% of women with DM reported receiving birth control-related information from their physicians, although 88.7% reported regular visits to diabetes clinics [7].

Eagerness for information on birth control was observed among 7 out of 10 young women with T1DM [6]. Nevertheless, contraception was associated with major problems among women with DM. About one-third of young women with T1DM expressed concerns about contraception issues: 36% protested about limited choices among available contraceptive methods and 31% expressed difficulties in receiving contraception [6]. Apart from practical challenges, about half the number of young females with T1DM falsely believed that all methods of contraception were only partially effective or even ineffective for individuals with DM [6]. Perhaps the most striking outcome of the questionnaire-based study by Feutry et al. [5] was the differentiation between women with T1DM and type 2 diabetes mellitus (T2DM): a significant difference in favour of women with T1DM could be observed in relation to birth control and pregnancy outcomes awareness (p=0.003) [5].

No prior use of contraceptives was reported in 10.4% to 15% of women with DM potentially eligible to receiving contraception [5] [7]. Among those on contraception, 12.5% used an improper method [5]. Among young women with T1DM, 16% reported receiving contraception at least once, whereas 14% regularly received contraception during the study [6]. However, these data must be considered with caution, as the mean age of participants was only 16 years and the study aimed primarily at disclosing the stance of young women or women with T1DM towards contraception [6]. Interestingly, Osman et al. [7] observed that about half the number of participants did not receive contraception (43.4%) during the questionnaire period. In terms of efficiency, Feutry et al. [5] reported that about two-thirds of participants considered their method of contraception as overall reliable. Nonetheless, a high prevalence of unwanted pregnancies was also reported (44.2%) [7].

Drug-drug interaction studies between antidiabetic medication and oral contraceptives

Novel antidiabetic agents are frequently assessed for interaction with oral contraceptives because the latter are often being controlled for potential drug-drug interactions with other frequently used drugs, including digoxin, warfarin, simvastatin, atorvastatin, rosuvastatin, furosemide and/or lisinopril [8] [9] [10] [11] [12] [13] [14] [15]. Overall, glucagon-like peptide -1 receptor agonists (GLP-1 RAs) (semaglutide, taspoglutide, liraglutide, dulaglutide, exenatide, albiglutide) and the sodium-glucose transport proteins 2 inhibitor (SGLT-2i) canagliflozin have been examined for potential interactions with oral contraceptives [8] [9] [10] [11] [12] [13] [14] [15].

Most studies included only healthy individuals, except for the study by Kapitza et al. [8], which addressed the potential effect of semaglutide administration on the bioavailability of ethinylestradiol/levonorgestrel among 43 post-menopausal participants with T2DM [8]. The bioavailability criterion is generally confirmed if confidence intervals for the ratio of pharmacokinetic parameters between the steady and free state of the examined drug remain within predefined limits, usually ranging from 0.80 to 1.25 [8]. The bioavailability criterion was fulfilled in both assessments (area under the curve, AUC and maximum concentration, Cmax) for both drugs and consequently, the researchers did not recommend dose adjustments [8].

A more recent study examined the effect of semaglutide on the bioavailability of multiple drugs, including ethinylestradiol/levonorgestrel [9]. Jordy et al. [9] observed an AUC ratio of 1.06 for both ethinylestradiol and levonorgestrel, which did not reach statistical significance, indicating a drug-drug interaction [9]. The same was concluded for Cmax and for other minor pharmacokinetic parameters, namely t_max and t_1/2, which also remained unaffected [9].

The two studies also documented potential adverse effects. Kapitza et al. [8] mainly observed changes in laboratory parameters, including a 3-fold increase in alanine aminotransferase (ALT) and elevated lipase and amylin levels. Jordy et al. [9] found gastrointestinal untoward effects (80%), primarily nausea (64%). Albiglutide administration was also considered safe for recipients of another oral contraceptive, ethinylestradiol/norethindrone. For both norethindrone and ethinylestradiol, the AUC ratio and Cmax were both assessed within the limits of insignificant interaction in the albiglutide steady state [10]. Luteinizing hormone (LH), follicle stimulation hormone (FSH) and progesterone levels were not affected either [10].

Dulaglutide has also been studied [11]. While proper AUC ratios could be observed for both ethinylestradiol and norelgestromin in dulaglutide-free and steady state, Cmax was affected. Under dulaglutide coadministration, Cmax for ethinylestradiol decreased by 13% and by 26% for norelgestromin [11]. The most frequently described untoward effects in the study included nausea, vomiting, and diminished appetite [11].

Liraglutide has been studied by Jacobsen et al. [12]. Ethinylestradiol and levonorgestrel concentrations decreased by 12% and 13% with liraglutide co-administration, respectively [12]. However, liraglutide administration led to an increased AUC ratio by 18%, beyond the prespecified limits, despite not affecting the AUC ratio for ethinylestradiol [12]. In a time-dependent manner, bioequivalence could be observed for levonergestrel as well, 74 h after liraglutide administration [12]. Nevertheless, no special dose adjustments were generally recommended for the coadministration of liraglutide and ethinylestradiol/levonorgestrel as the effect of liraglutide administration on the major pharmacokinetic parameters of ethinylestradiol/levonorgestrel was overall insignificant [12].

Taspoglutide was evaluated, prompting similar results [13]. Although both bioequivalence parameters remained within the limits for ethinylestradiol, taspoglutide administration resulted in significant changes in both parameters (AUC geometric mean ratio: 0.93, 95% confidence interval [CI]: 0.89–0.97) and Cmax 0.83, 95% CI: 0.77–0.88) [13]. Despite these outcomes, the researchers considered dose adjustments under taspoglutide coadministration as unnecessary [13]. Taspoglutide, similarly to albiglutide, did not affect ovulation [10] [13].

In contrast to all previously evaluated GLP-1 RAs, exenatide administration resulted in noticeable changes, as indicated in the open-label, randomised crossover trial by Kothare et al. [14]. After single-dose exenatide administration, the mean Cmax decreased by 46% for ethinylestradiol and by 41% for levonorgestrel. After repeated daily administration along with exenatide, the mean Cmax was reduced by 45% for ethinylestradiol and by 27% for levonorgestrel [14]. There was a delay of 3 to 4 h until Cmax levels, suggesting that agents with a threshold-dependent action must be taken at least 1 h prior to exenatide.

Only one study has assessed canagliflozin [15], reporting an increase in Cmax by 22% and the AUC by 6% of oral contraceptive levels upon canagliflozin administration without any untoward effects [15].

Contraception methods and pregnancy-related outcomes in DM: clinical trials

Among parameters associated with gynaecological history, oral contraception was the only factor associated with reduced risk of T2DM development in a recent large (262,368 women with a mean follow-up of 12.2 years) UK population-based study [16]. The hazard ratio of oral contraceptive use was 0.93 for DM (95% CI: 0.89–0.98). Conversely, all other factors examined, including early menarche, late menarche, short reproductive lifespan, gynaecological operations (hysterectomy/oophorectomy), high number of births, early age at first birth, miscarriage, stillbirth, and hormone replacement therapy, were linked with increased risk of T2DM development [16].

The protective effect of hormonal contraception among women with T1DM was further documented in the prospective cohort study by Snell-Bergeon et al. [17]. Contraception use was significantly associated with decreased coronary artery calcium; this association emerged stronger for women with T1DM (p=0.02) [17].

Gaudio et al. [18] examined parameters with disadvantageous impacts on pregnancy outcomes (suboptimal glycaemic control, microalbuminuria, hypertension, use of antidiabetic drugs, and comorbidities). At least one of these factors was seen in 50.9% of T1DM participants and 70.7% of T2DM participants. The high prevalence of comorbidities and inadequate glycaemic control highlights the crucial need for proper contraception emerges in high-risk pregnancies. Particularly, the use of antidiabetic agents apart from insulin and metformin increased from 22.3% to 27.3% among participants with T2DM [18]. This increase in the use of antidiabetic agents could highly likely exacerbate further birth control efficiency rates among participants with DM, as the risk of potential drug-drug interactions, particularly between the new hypoglycaemic medication and the combined hormonal contraception, could be aggravated.

Nowadays, there are plenty of methods available for contraception among women with DM, including oral contraceptives, intra-uterine devices or other methods [19]. An observational study of 667 DM participants with a mean age of 34.8 years by Napoli et al. [19] examined the frequency of each contraception method. Hormonal contraception was by far the most commonly used method: 30.4% preferred hormonal contraception, while only 12% utilised intra-uterine devices and 10.7% did not use any form of contraception.The cross-sectional study by Shawe et al. [20] reported the combined pill as the first option among contraceptives, particularly among participants with T1DM, followed by the progestin-only pill. The latter was used much more regularly by T1DM participants compared to the general population [20]. Depo Provera (depot medroxyprogesterone acetate, DMPA) was much more frequently used among participants with DM compared with the general population [20].

Novel devices used for contraception have also been assessed, with NuvaRing being a distinct example. NuvaRing is a vaginal ring-shaped contraceptive device, which constantly releases 120 mcg etonogestrel and 15 mcg ethinylestradiol per day and can be applied in predefined time intervals (day 1 to 5) during the menstrual cycle [21]. In a randomised controlled trial, Grigoryan et al. [22] evaluated the impact of this device on the lipid and carbohydrate metabolism in T1DM participants of late reproductive age [22]. There was no beneficial effect on lipid or carbohydrate metabolism, thus allowing the use of NuvaRing among participants with T1DM [22].

The impact of contraception on the metabolic profile of women with DM was also assessed by Vincente et al. [23]. The study considered 23 women with DM on insulin using an etonogestrel implant. No significant changes were documented in body-mass index (BMI), insulin dose or mean HbA_1c [23]. However, at 6, 12, and 24 months, a significant decrease in high-density lipoprotein cholesterol (HDL-C) and triglycerides (TGs) was observed. A significant reduction in total cholesterol was also seen at 6 and 12 months. LDL and the HDL/TC ratio remained unaffected [23]. Albumimuria also significantly decreased at 12 and 24 months. No changes were observed during funduscopy [23]. The most common menstrual patterns reported were amenorrhoea and bleeding [23].

The impact of contraception on lipid profile has been evaluated by Diab et al. [24]. The study included 80 participants with DM without diabetic complications [24]. These were allocated into four groups: 20 women with intra-uterine device, 20 women in the Norplant (subdermal levonorgestrel contraception) group, 20 women receiving oral contraceptives, and 20 women receiving DMPA [24] [25]. Oral contraceptives or DMPA resulted in increased fasting glucose levels [24]. Oral contraceptive recipients showed significantly increased TGs and HDL-C levels and decreased low-density lipoprotein cholesterol (LDL-C) levels, compared to those using intra-uterine devices [24]. In addition, the DMPA group showed significantly increased total cholesterol and HDL-C levels and decreased LDL-C levels, compared to women using intra-uterine devices [24].

Another randomised controlled trial assessed the use of a levonorgestrel-releasing device among participants with T1DM [26]. Mean HbA_1c was similar between the women using the levonorgestrel-releasing device and those using the copper device at 12 months and both mean fasting glucose and insulin units remained unaffected [26]. The open randomised controlled study by Grigoryan et al. [27] yielded similar results. HbA_1c and insulin levels remained unaffected. Furthermore, the combination of oral contraception and an intra-uterine device did not alter the lipid profile [27].

In a retrospective chart review-case series study, Lang et al. [28] evaluated the use of levonorgestrel-releasing intra-uterine device among 115 T2DM women<55 years of age with a severely high average BMI of 42.8 kg/m². This device was efficient, inasmuch as only one pregnancy was observed [28]. Pregnancies were mainly found in women experiencing expulsion or removal of the device [28]. Expulsion was seen in 3.5% of participants, abdominal or pelvic pain was reported by 13.9%, and pelvic inflammatory disease by 1.7% [28]. New diagnoses of renal disease, neuropathy, retinopathy, and vascular disease were found among 6.0%, 2.6%, 2.6%, and 2.6% of participants, but were not attributed to the device and were likely the effect of DM progression [28]. At the study end (mean follow-up of 754 days), overall compliance with the device was 77.5% [28].

Klingensmith et al. [29] conducted a randomised controlled trial examining pregnancy outcomes of 452 women with T2DM. [29]. Overall use of contraceptives was reported by 4.8%. There were 63 pregnancies from 10.2% of study participants [29]. The outcomes were suboptimal, compared with the general population: 3 pregnancies were not associated with any data, 7 were terminated, 5 led to early loss, 7 were linked with loss without further data, 2 led to stillbirth and 39 resulted in births [29]. Moreover, 15.4% of infants were born preterm, and about 1 out of 5 infants (20.5%) suffered from a major congenital malformation, thus directly indicating the harmful impact of inadequate birth control among women with DM [29].

More frequent miscarriages were noted in the observational study by Napoli et al. [19]: with the average number of 1.17 births, 1.3 miscarriages, and 0.17 induced abortions [19].

Noteworthy differences were observed between T1DM and T2DM participants in terms of contraception use. In general, hormonal contraception was more frequent among participants with T1DM: 32.2% in T1DM vs. 23.1% in T2DM [18]. By contrast, Napoli et al. [19] found that the use of oral contraception was similar between participants with T1DM and T2DM (29.4% and 27.8%, respectively) [19]. Interestingly, oral contraception was more frequently used among women with higher education level (university graduates at 37.1%, high school graduates at 32.2%, secondary school graduates at 28.2%, and primary school graduates at 15.5%) and 30% of the contraceptive recipients were smokers [19]. Studies are summarised in [Table 2].

Table 2 Studies on contraception and pregnancy-related outcomes in diabetes mellitus.

Study	Study type	Objectives	Population	Outcomes	Secondary outcomes
Fan et al. [16]	Observational UK biobank study	Evaluation of multiple reproductive factors among women with T2DM	262,368 women without T2DM (mean follow-up time at 12.2 years)	The use of oral contraceptives was the only factor linked with diminished T2DM development risk (HR: 0.93)	Factors, as early menarche, late menarche, short reproductive lifespan, gynaecological operations (hysterectomy/oophorectomy), a high number of births, early age at first birth, miscarriage, stillbirth, and hormone replacement therapy were associated with increased risk for T2DM development
Gaudio et al. [18]	Observational study	Assessment of factors affecting pregnancy outcomes among women with DM	3,218 women with T1DM or T2DM	32.2% of participants with T1DM and 23.1% of participants with T2DM received hormonal contraception Women with unfavourable factors received contraception more frequently, but the increased frequency was significant only among T1DM participants	An increased proportion of women with HbA1c<6.5% was observed, both in T1DM (from 9.0% to 19.1%) and in T2DM participants (from 27.2% to 35.4%) At least one unfavourable factor for pregnancy (obesity, suboptimal glycaemic levels, microalbuminuria, hypertension, use of antidiabetic drugs and associated comorbidities) was observed among 50.9% in T1DM participants and 70.7% of T2DM participants Prescription of antidiabetic drugs beyond insulin and metformin increased from 22.3% to 27.3% among T2DM participants
Klingensmith et al. [29]	Randomised controlled trial	Evaluation of pregnancy-related outcomes among young women with T2DM	452 young female T2DM participants	About 1 in 20 participants (4.8%) received contraception	10.2% out of 452 participants had 63 pregnancies In 63 pregnancies, the following outcomes were documented: 3 pregnancies were without data, 7 were terminated, 5 led to an early loss, 7 were linked to loss without further data, 2 led to stillbirth and 39 resulted in the birth 15.4% of infants were preterm and 20.5% suffered from a major congenital malformation
Shawe et al. [20]	Cross-sectional study	Assessment of hormonal contraception use by women with DM	947 T1DM participants with matched controls 365 T2DM participants with matched controls	Hormonal contraception is used less frequently among women with DM Women with T1DM used primarily the combined pill and secondary the progestin-only pills, although the latter was used much more frequently compared with women without DM Women with T2DM received much more rarely the combined pill	Depo Provera was much more frequently used among DM participants
Snell-Bergeon et al. [17]	Prospective cohort study	Evaluation of reproductive factors on coronary calcium progression	293 T1DM participants 319 participants without DM	T1DM participants reported less frequent contraception use compared with those without DM Contraception use was significantly associated with decreased coronary artery calcium and this association was stronger for T1DM participants	Menstrual cycle disorders were significantly more frequent among T1DM participants
Grigoryan et al. [22]	Randomised controlled trial	Assessment of the NuvaRing hormone-based contraception system in late reproductive age women with T1DM	25 T1DM participants 20 matched control participants without the use of contraception methods 20 healthy women with use of the NuvaRing	No benefit in lipid or carbohydrate metabolism	Haemostasis was unaffected
Rogovskaya et al. [26]	Randomised clinical trial	Assessment of levonorgestrel releasing intra-uterine device in women with T1DM	62 women with T1DM and without complications 29 women with levonorgestrel-releasing device 30 women with copper device	Mean HbA1c was similar between the 2 groups at 12 months Both mean fasting glucose and insulin units remained unaffected	No significant differences at 6 weeks or at 6 months
Grigoryan et al. [27]	Open randomised controlled study	Evaluation of combined oral contraception and intra-uterine device on the metabolism and haemostasis of perimenopausal women	113 women with DM 40 matched control participants	HbA1c and insulin units were unaffected by both contraceptive methods	Lipid profile remained unaffected The use of combined oral contraceptives resulted in increased platelet activity The intra-uterine device was not associated with changes in haemostasis
Napoli et al. [19]	Observational study	Assessment of contraception use among women with DM	667 fertile women with DM with mean age of 34.8 years, average BMI of 24.9 kg/m2, average diabetes duration of 12.4 years and average HbA1c of 7.4% 446 T1DM participants 201 T2DM participants	30.4% of participants used hormonal contraception 12% of participants utilised intra-uterine devices 10.7% of participants used no means of contraception The use of oral contraception was similar between T1DM and T2DM participants (29.4% and 27.8%, respectively)	30.0% of contraception users were smokers Oral contraception was more frequently used among women with higher education levels (university graduates at 37.1%, high school graduates at 32.2%, secondary school graduates at 28.2% and primary school graduates at 15.5%) The average number of births was 1.17, of miscarriages 1.3 and of induced abortions 0.17 29.4% of pregnancies were planned under the supervision of diabetologists and gynaecologists
Vincente et al. [23]	Prospective observational study	Assessment of the etonogestrel implant use among women with insulin-treated DM	23 women with DM and under insulin treatment	No changes in BMI, insulin dose or mean HbA1c were documented At 6, 12 and 24 months, a significant decrease in HDL-C and TG levels was seen, and a significant reduction in TC at 6 and 12 months was seen LDL and the HDL/TC ratio were unaffected	Albuminuria was significantly decreased at 12 and 24 months No significant retinal changes were reported Amenorrhoea and bleeding were the commonest menstrual patterns
Lang et al. [28]	Retrospective chart review-case series study	Evaluation of levonorgestrel-releasing intra-uterine device among women with T2DM	115 women with T2DM<55 years with average BMI 42.8 kg/m2	Only one pregnancy under the effect of levonorgestrel-releasing device occurred -Pregnancies occurred in women experiencing expulsion of the device or removal Expulsion of the intra-uterine device was observed among 3.5% of participants Abdominal or pelvic pain was observed in 13.9% of participants Pelvic inflammatory disease was diagnosed in 1.7% of participants	No significant change in HbA1c (mean decrease 0.17) Following the placement of the device, new diagnoses of renal disease, neuropathy, retinopathy and vascular disease were found among 6%, 2.6%, 2.6% and 2.6% of the study participants, respectively 77.5% continued using the device at the end of the study with mean follow-up of 754 days
Diab et al. [24]	Prospective comparative study	Evaluation of various contraceptive methods in women with DM	80 DM participants without diabetes-related complications 20 women with intra-uterine device 20 women in the Norplant group 20 women receiving oral contraceptives 20 women receiving DMPA	Oral contraceptive or DMPA use resulted in increased fasting glucose levels Oral contraceptive recipients showed significant increases in TG and HDL-C and a decrease in LDL-C, compared with women using intra-uterine devices The DMPA group showed significant increases in total cholesterol and HDL-C and decrease in LDL-C	Norplant use prolonged the partial thromboplastin time Total cholesterol and LDL-C were decreased TGs increased among oral contraceptive recipients HDL-C increased with oral contraceptives and decreased with Norplant and DMPA

DM, diabetes mellitus; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; HbA_1c, glycated haemoglobin; HR, hazard ratio; HDL-C, high-density lipoprotein-cholesterol; TG, triglycerides; TC, total cholesterol; LDL-C, low-density lipoprotein-cholesterol; DMPA, depot-medroxyprogesterone acetate

Use of contraception in diabetes mellitus: what to choose

Birth control includes nowadays a wide variety of methods. In general, women with DM, optimal glucose regulation and the absence of complications or major comorbidities can use the contraceptive method of their wish [30] [31]. Even the absence of optimal glucoregulation, should not be considered as an absolute contraindication for the combined oral contraceptive [30]. According to the ADA [1], unplanned pregnancies in DM are more dangerous than current contraception modalities.

Salinas et al. [31] emphasised that subcutaneous progestin implants (either containing etonogestrel or levonorgestrel) and intra-uterine devices (copper, levonorgestrel-releasing) are the most efficient methods for contraception, followed by other forms, including injections, pills, rings, patches or diaphragms. In general, the choice of contraception among women with DM relies on multiple parameters. These include current antidiabetic therapy, DM duration, diabetic complications, comorbidities, compliance, and efficiency [31] [32]. Due to the known prothrombotic action of estrogens in the combined hormonal contraception, women with a DM duration >20 years and/or major micro- or macrovascular complications are generally advised to prefer either non-hormonal ways of contraception or to use only progestin implants, if necessary [31] [33] [34].

A Cochrane systematic review including four studies showed that hormonal and non-hormonal contraception could be considered equally effective in terms of glucoregulation [33]. It must be taken into consideration that combined hormonal contraception, but not progestin-based contraception, affects glucose metabolism [33]. In a prospective, longitudinal, population-based, cohort study of 1879 women by Mosorin et al. [34], three alternative birth control methods were compared. Combined hormonal contraception was significantly associated with the development of both prediabetes (odds ratio [OR]: 2.0, 95% CI: 1.3–3.2) and T2DM (OR: 3.3, 95% CI: 1.1–9.7) [34]. The risk was further aggravated after 5 years of contraception use: the prediabetes risk increased 2.2 times (95% CI: 1.3–3.7) and the T2DM risk increased 4.5 times (95% CI: 1.5–13.5) [34]. Therefore, long-term use of combined hormonal contraception must be cautious [34].

A major difference between individuals with T1DM and T2DM is associated with the increased prevalence of obesity among T2DM individuals, although there has been a tendency to an increased prevalence of obesity among T1DM individuals as well [30] [35]. Due to the increased risk for deep vein thrombosis or even thromboembolism, BMI > 35 kg/m² must be considered as a contraindication for combined oral contraception, and alternative methods should be preferred [30].

Proper management of contraception among women with DM has not been extensively clarified and generalised yet, primarily due to the lack of recent studies. There is a general recommendation that women facing multiple risk factors related to the cardiovascular system should use non-hormonal- or progestin-only-based contraceptive methods [32]. Also, common comorbidities, such as dyslipidaemia, obesity or detrimental habits, such as smoking related to thromboembolism, a major adverse effect of oral contraceptives, are to be identified and treated [32].

Contraception: is there any effect on diabetic complications?

Apart from a case report examining the impact of oral contraceptives on peripheral arterial disease in a woman without DM, there are no major studies on the effect of oral contraceptives on women with DM and the risk of peripheral arterial disease [36]. With the exception of diabetic neuropathy, other microvascular complications (diabetic nephropathy and retinopathy) have been studied more extensively in relation to the effect of oral contraceptives [37] [38] [39] [40] [41].

Ahmed et al. [37] assessed renal function through renal plasma flow response to captopril, an inhibitor of the angiotensin-converting enzyme, among oral contraceptive users, non-users and users with DM. They observed an increased response in oral contraceptive users compared to non-recipients, which also emerged to be statistically significant (p=0.02) [37]. In the case of oral contraceptive recipients with DM, the anticipated increased response could be observed and the response was found to be the greatest compared to oral contraceptive recipients with DM (p=0.04) [37]. Furthermore, 18% of oral contraceptive recipients developed macroalbuminuria, compared with only 2% of non-recipients and the correlation was statistically significant (p=0.003) [37]. Cox regression analysis showed that the use of oral contraceptives is a significant risk factor for macroalbuminuria (p=0.008) [37].

Garg et al. [38] assessed renal function among women with DM and concluded that the use of oral contraceptives by women with insulin-dependent DM was not related to early diabetic nephropathy development [38]. The same outcome was determined for diabetic retinopathy as well, as the mean eye grades did not vary significantly between the study arms [38].

Another study also suggested that oral contraceptive use did not affect the progression of diabetic nephropathy, retinopathy or even cardiovascular disease [39]. The effect of oral contraceptives on diabetic retinopathy has been ambiguous. Klein et al. [40] observed a tendency to even less severe retinopathy under simultaneous oral contraceptive use, although no major effects were noted. The Wisconsin Epidemiologic Study of Diabetic Retinopathy assessed women with DM with a follow-up period up to 14 years and found no effect of oral contraceptives on the severity of retinopathy or macular oedema [41].

The limited number of studies evaluating the effect of oral contraceptive use on diabetes-related complications might be attributed to the small number of participants exhibiting diabetes-related complications in the studies assessing the use of contraceptives among participants with DM. In the questionnaire-based study by Feutry et al. [5] from 2022, among 122 participants with DM, retinopathy was reported by 12 (9.8%), nephropathy by 8 (6.6%) and neuropathy or myocardial infarction by only 1 participant (0.8%). Thus, many researchers in the field recommend studies with preferably a prospective design and larger populations [37].

Gourdy [30] argued that non-proliferative retinopathy and non-significant microalbuminuria should not be considered as contraindications for combined hormonal contraception. However, in the presence of severe complications, including nephropathy, macroalbuminuria, retinopathy, cardiovascular disease and/or peripheral/autonomic neuropathy, the combined oral contraception is absolutely contra-indicated, primarily among women with T1DM [30].

New antidiabetic medication: any effect on fertility?

As aforementioned, almost all GLP-1 RAs (except for exenatide) and canagliflozin did not significantly affect pharmacokinetic or pharmacodynamic parameters of the widely used hormonal contraception [8] [9] [10] [11] [12] [13] [14] [15]. However, clinical trials relied only on clinical pharmacology parameters. Nowadays, after longer clinical experience with GLP-1 RAs, new considerations have arisen, particularly in relation to semaglutide [4]. Indeed, unintended pregnancies were documented among semaglutide recipients, even under simultaneous use of contraceptives, leading to the striking term “Ozempic babies” [4]. Experimental data from 2015 showed that GLP-1 affected the menstrual cycle in mice by increasing the pre-ovulatory LH peak, thus resulting in an increased number of viable offspring [42]. More recent data showed that GLP-1 levels are regulated by a bacterial species, Bacteroides vulgates, which suppressed ovarian function in mice [43]. Ovarian function was restored with GLP-1 administration [43].

New considerations have come to light with the dual GLP-1 and glucose-dependent insulinotropic polypeptide (GIP) agonist tirzepatide [44]. Preliminary data from an open-label, study, available from the National Library of Medicine and conducted by Eli Lilly, the pharmaceutical company which developed the drug, assessed the effect of tirzepatide administration on pharmacokinetic parameters of norelgestromin and ethinylestradiol in 40 women (NCT04172987) [4] [45]. AUC ratios decreased by 20% for ethinylestradiol and 21% for norelgestromin, whereas the impact on the Cmax was even more profound [45]. The Cmax of ethinylestradiol decreased by 59%, while an even higher decrease in norelgestromin by 66% was reported [4] [45]. Tirzepatide administration resulted in a remarkable delay in the maximum plasma concentration time by 2.5 to 4.5 h [45]. Based on these outcomes, Eli Lilly advised tirzepatide recipients either to use simultaneously various contraceptive methods or to increase the dose of their oral contraceptives [4].

Discussion

The present review summarises data on contraception in DM. Its strengths include the broad spectrum of studies on birth control and pregnancy-related outcomes, as well as the data on the point of view of women with DM and drug-drug interactions. It also has certain limitations. The first relates to the scarcity of data on the choice of contraception for individuals with microvascular complications. Moreover, data synthesis to determine the general benefit of specific hormonal combinations or other ways of contraception was not feasible due to the heterogeneity in the study design. This heterogeneity was pertinent to research questions, study populations (women with T1DM and/or T2DM), methods used (evaluation of single or multiple contraceptive methods) and outcomes. Accordingly, it was not possible to use a population/intervention/comparison/outcome (PICO)-based approach.

In conclusion, contraception is suboptimally used by women with DM. Physicians should improve patient counselling on contraception. Potential drug-drug interactions between antidiabetic medication and oral contraceptives merit further study. Generally, GLP-1 RAs and canagliflozin did not exhibit major changes in the pharmacological profile of combined hormonal contraception. However, preliminary results indicate that some novel antidiabetic agents may even render oral contraceptive methods ineffective. Several implants can also be used in women with both DM types. The relationship between oral contraceptives and diabetic complications has not been clarified yet. Therefore, large prospective studies are needed. Although women with DM could use combined hormonal contraception, long-term use, pronounced hyperglycaemia and the presence of severe complications or risk factors must be taken into consideration. Alternative ways of contraception, including implants, intra-uterine devices or progestin-only birth control, are generally considered suitable for women with DM.

Conflict of Interest

Evanthia Gouveri has attended conferences sponsored by Berlin-Chemie, Sanofi, AstraZeneca, Novo Nordisk, Lilly and Boehringer Ingelheim; received speaker honoraria by Boehringer-Ingelheim and Menarini. Dimitrios Papazoglou declares associations: with Menarini, Novo Nordisk, Astra-Zeneca, Boehringer Ingelheim and Sanofi-Aventis. Nikolaos Papanas has been an advisory board member of Astra-Zeneca, Bayer, Boehringer Ingelheim, Menarini, MSD, Novo Nordisk, Pfizer, Takeda and TrigoCare International; has participated in sponsored studies by Astra-Zeneca, Eli-Lilly, GSK, MSD, Novo Nordisk, Novartis and Sanofi-Aventis; has received honoraria as a speaker for Astra-Zeneca, Bayer, Boehringer Ingelheim, Eli-Lilly, Elpen, Menarini, MSD, Mylan, Novo Nordisk, Pfizer, Sanofi-Aventis and Vianex; and has attended conferences sponsored by TrigoCare International, Eli-Lilly, Bayer, Galenica, Novo Nordisk, Pfizer and Sanofi-Aventis. The other authors have no conflicts.

• References

• 1 American Diabetes Association. 15. Management of diabetes in pregnancy: Standards of care in diabetes—2024. Diabetes Care 2024; 47: S282-S294
• 2 GBD 2021 Diabetes Collaborators. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: A systematic analysis for the Global Burden of Disease Study 2021. Lancet 2023; 402: 203-234
• 3 Abushamat LA, Sayres L, Jeffers R. et al. Unmasking barriers in the delivery of preconception counseling and contraception provision for patients with type 1 or type 2 diabetes. Clin Diabetes 2023; 41: 567-572
• 4 Dohrn G. Does Ozempic boost fertility? What the science says. Nature 2024; Jun 26. Online ahead of print
• 5 Feutry L, Barbe C, Marquet-Dupont A. et al. Contraception use and knowledge related to pregnancy in diabetic women. Ann Endocrinol (Paris) 2022; 83: 88-94
• 6 Schwarz EB, Sobota M, Charron-Prochownik D. Perceived access to contraception among adolescents with diabetes: Barriers to preventing pregnancy complications. Diabetes Educ 2010; 36: 489-494
• 7 Osman A, Hoffman A, Moore S. et al. Reproductive knowledge and use of contraception among women with diabetes. S Afr Med J 2015; 105: 760-764
• 8 Kapitza C, Nosek L, Jensen L. et al. Semaglutide, a once-weekly human GLP-1 analog, does not reduce the bioavailability of the combined oral contraceptive, ethinylestradiol/levonorgestrel. J Clin Pharmacol 2015; 55: 497-504
• 9 Jordy AB, Albayaty M, Breitschaft A. et al. Effect of oral semaglutide on the pharmacokinetics of levonorgestrel and ethinylestradiol in healthy postmenopausal women and furosemide and rosuvastatin in healthy subjects. Clin Pharmacokinet 2021; 60: 1171-1185
• 10 Bush M, Scott R, Watanalumlerd P. et al. Effects of multiple doses of albiglutide on the pharmacokinetics, pharmacodynamics, and safety of digoxin, warfarin, or a low-dose oral contraceptive. Postgrad Med 2012; 124: 55-72
• 11 de la Peña A, Cui X, Geiser J. et al. No dose adjustment is recommended for digoxin, warfarin, atorvastatin or a combination oral contraceptive when coadministered with dulaglutide. Clin Pharmacokinet 2017; 56: 1415-1427
• 12 Jacobsen LV, Vouis J, Hindsberger C. et al. Treatment with liraglutide--a once-daily GLP-1 analog--does not reduce the bioavailability of ethinyl estradiol/levonorgestrel taken as an oral combination contraceptive drug. J Clin Pharmacol 2011; 51: 1696-1703
• 13 Bogman K, Brumm J, Hofmann C. et al. assessment of drug-drug interactions between taspoglutide, a glucagon-like peptide-1 agonist, and drugs commonly used in type 2 diabetes mellitus: Results of five phase I trials. Clin Pharmacokinet 2019; 58: 1205-1214
• 14 Kothare PA, Seger ME, Northrup J. et al. Effect of exenatide on the pharmacokinetics of a combination oral contraceptive in healthy women: An open-label, randomised, crossover trial. BMC Clin Pharmacol 2012; 12: 8
• 15 Devineni D, Manitpisitkul P, Vaccaro N. et al. Effect of canagliflozin, a sodium glucose co-transporter 2 inhibitor, on the pharmacokinetics of oral contraceptives, warfarin, and digoxin in healthy participants. Int J Clin Pharmacol Ther 2015; 53: 41-53
• 16 Fan G, Liu Q, Bi J. et al. Reproductive factors, genetic susceptibility and risk of type 2 diabetes: A prospective cohort study. Diabetes Metab 2024; 50: 101560
• 17 Snell-Bergeon JK, Dabelea D, Ogden LG. et al. Reproductive history and hormonal birth control use are associated with coronary calcium progression in women with type 1 diabetes mellitus. J Clin Endocrinol Metab 2008; 93: 2142-2148
• 18 Gaudio M, Dozio N, Feher M. et al. Trends in factors affecting pregnancy outcomes among women with type 1 or type 2 diabetes of childbearing age (2004-2017). Front Endocrinol (Lausanne) 2021; 11: 596633
• 19 Napoli A, Colatrella A, Botta R. et al. Contraception in diabetic women: An Italian study. Diabetes Res Clin Pract 2005; 67: 267-272
• 20 Shawe J, Mulnier H, Nicholls P. et al. Use of hormonal contraceptive methods by women with diabetes. Prim Care Diabetes 2008; 2: 195-199
• 21 Sarkar NN. The combined contraceptive vaginal device (NuvaRing): A comprehensive review. Eur J Contracept Reprod Health Care 2005; 10: 73-78
• 22 Grigoryan OR, Grodnitskaya EE, Andreeva EN. et al. Use of the NuvaRing hormone-releasing system in late reproductive-age women with type 1 diabetes mellitus. Gynecol Endocrinol 2008; 24: 99-104
• 23 Vicente L, Mendonça D, Dingle M. et al. Etonogestrel implant in women with diabetes mellitus. Eur J Contracept Reprod Health Care 2008; 13: 387-395
• 24 Diab KM, Zaki MM. Contraception in diabetic women: comparative metabolic study of Norplant, depot medroxyprogesterone acetate, low dose oral contraceptive pill and CuT380A. J Obstet Gynaecol Res 2000; 26: 17-26
• 25 Cooper M. Norplant. Aust N Z J Obstet Gynaecol 1991; 31: 265-272
• 26 Rogovskaya S, Rivera R, Grimes DA. et al. Effect of a levonorgestrel intrauterine system on women with type 1 diabetes: A randomized trial. Obstet Gynecol 2005; 105: 811-815
• 27 Grigoryan OR, Grodnitskaya EE, Andreeva EN. et al. Contraception in perimenopausal women with diabetes mellitus. Gynecol Endocrinol 2006; 22: 198-206
• 28 Lang B, Josephy T, Micks E. et al. Use of the levonorgestrel intrauterine device in women with type 2 diabetes. Clin Diabetes 2018; 36: 251-256
• 29 Klingensmith GJ, Pyle L, Nadeau KJ. et al. Pregnancy outcomes in youth with type 2 diabetes: The TODAY study experience. Diabetes Care 2016; 39: 122-129
• 30 Gourdy P. Diabetes and oral contraception. Best Pract Res Clin Endocrinol Metab 2013; 27: 67-76
• 31 Salinas A, Merino PM, Giraudo F. et al. Long-acting contraception in adolescents and young women with type 1 and type 2 diabetes. Pediatr Diabetes 2020; 21: 1074-1082
• 32 Merino PM, Codner E. Contraception for adolescents and young women with type 2 diabetes-specific considerations. Curr Diab Rep 2022; 22: 77-84
• 33 Visser J, Snel M, Van VHA. Hormonal versus non-hormonal contraceptives in women with diabetes mellitus type 1 and 2. Cochrane Database Syst Rev 2013; 2013: CD003990
• 34 Mosorin ME, Haverinen A, Ollila MM. et al. Current use of combined hormonal contraception is associated with glucose metabolism disorders in perimenopausal women. Eur J Endocrinol 2020; 183: 619-626
• 35 Koufakis T, Patoulias D, Zografou I. et al. Drawing lines in the sand: The growing threat of obesity in type 1 diabetes. World J Diabetes 2024; 15: 823-827
• 36 Pallavee P, Samal S, Samal R. Peripheral arterial disease in a female using high-dose combined oral contraceptive pills. Indian J Pharmacol 2013; 45: 303-304
• 37 Ahmed SB, Hovind P, Parving HH. et al. Oral contraceptives, angiotensin-dependent renal vasoconstriction, and risk of diabetic nephropathy. Diabetes Care 2005; 28: 1988-1994
• 38 Garg SK, Chase HP, Marshall G. et al. Oral contraceptives and renal and retinal complications in young women with insulin-dependent diabetes mellitus. JAMA 1994; 271: 1099-1102
• 39 Metabolic effects of oral contraceptives: Fact vs. fiction. Contracept Rep. 1996 6. 4-14
• 40 Klein BE, Moss SE, Klein R. Oral contraceptives in women with diabetes. Diabetes Care 1990; 13: 895-898
• 41 Klein BE, Klein R, Moss SE. Exogenous estrogen exposures and changes in diabetic retinopathy. The Wisconsin epidemiologic study of diabetic retinopathy. Diabetes Care 1999; 22: 1984-1987
• 42 Outeiriño-Iglesias V, Romaní-Pérez M, González-Matías LC. et al. GLP-1 increases preovulatory LH source and the number of mature follicles, as well as synchronizing the onset of puberty in female rats. Endocrinology 2015; 156: 4226-4237
• 43 Yun C, Yan S, Liao B. et al. The microbial metabolite agmatine acts as an FXR agonist to promote polycystic ovary syndrome in female mice. Nat Metab 2024; 6: 947-962
• 44 Corrao S, Pollicino C, Maggio D. et al. Tirzepatide against obesity and insulin-resistance: Pathophysiological aspects and clinical evidence. Front Endocrinol (Lausanne) 2024; 15: 1402583
• 45 Skelley JW, Swearengin K, York AL. et al. The impact of tirzepatide and glucagon-like peptide 1 receptor agonists on oral hormonal contraception. J Am Pharm Assoc (2003) 2024; 64: 204-211.e4

Publication History

Received: 14 October 2024

Accepted after revision: 18 March 2025

Accepted Manuscript online:
18 March 2025

Article published online:
29 April 2025