Introduction

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease with heterogenous clinical presentations and disease course. SLE has a well-established female preponderance, with a female-to-male ratio of 9:1 [1]. Some characteristic features point to the role of hormones, particularly estrogen, in the pathophysiology of SLE. For example, the incidence of the disease is highest in females between age 15 and 45 years (i.e. during the ovulatory period) and disease exacerbations are often seen in important times such as during puberty and pregnancy [2]. A two- to three-fold increase in disease activity during pregnancy in some women with SLE has been well-documented in several publications [3,4,5,6], especially in women who conceived when they had active lupus [7, 8]. Furthermore, pregnant females with moderate to severe SLE have a higher incidence of disease flares and obstetrical and fetal complications [5]. These harmful effects can be attributed directly to SLE disease activity, along with the side effects of medications on fetal development [9].

Clearly, SLE is a condition whereby an unintended pregnancy can be detrimental to the health of mother and fetus. So, it is of the utmost importance that planned pregnancy in females with SLE should ideally occur during a period of disease quiescence [7, 8]. Because of the health concerns specific to pregnant SLE patients, the safe use of contraceptives has to be a subject of discussion during the disease course. Since many medications used to treat SLE have significant teratogenic potential, use of effective contraception is imperative if pregnancy is not desired.

Regardless, many patients with lupus have unprotected sex. Female patients often lean towards barrier-or-behaviour-based contraceptive methods to prevent unwanted pregnancy [10,11,12, 13••]. Previously, many women were counseled against using combined hormonal contraceptives (CHCs) based on two main concerns: (1) increased SLE flares and (2) cardiovascular and thrombosis risks [11]. In the last decade, many investigators have published reassuring data based on trials on the use of combined oral contraceptives (COCs) [14••, 15], progestin-only-pills (POPs) [14••, 16] and copper-releasing intrauterine device (IUD) [14••] on SLE disease activity during pregnancy.

This article discusses different aspects of SLE disease activity, contraception and pregnancy to present state of the art data on pregnancy planning and management in lupus patients. We present a brief overview of the types of contraceptives available, their safety profile concerning SLE disease activity in pregnancy, risk of thrombosis and interactions with concomitant drugs being taken for SLE itself.

Overview

Women with SLE seeking contraception should be counseled on the necessity of contraceptive use, as well as the full range of effective methods available. Contraceptive counseling should include a discussion of typical failure rates and the importance of using the contraceptive method consistently and correctly in order to avoid pregnancy [17•]. The choice of contraception must be individualized to each female patient with SLE and should be weighed against the potential benefits and risks taking into consideration the wishes of the patient.

Perfect versus typical use

When deciding a contraceptive method, it is vital to understand and focus on contraceptive effectiveness, which is related to “typical use” (i.e. reflecting real-world use). Contraceptive efficacy, on the other hand, demonstrates the “perfect use” of a method. The difference between failure rates of typical versus perfect use tends to decrease if a user-independent method is selected. Hence, the failure rates of typical and perfect use are nearly identical for long-acting reversible contraceptives (LARCs), which are the most effective contraceptive option, as they require no effort on the part of the patient.

Contraceptive use prevalence and underutilization of effective methods/counseling

In general, women with SLE are at risk for unplanned pregnancy and do not consistently use contraception. Even when they do, they often do not use the most effective types or use CHCs despite a potential contraindication. A recently published large multi-national inception cohort in SLE (“SLICC cohort”, n = 927) [18] reported that more than half of SLE patients using CHCs had one or more possible contraindication.

Another prospective observational study (n = 206) on use of contraceptive counseling in women with SLE [19] reported that barrier methods, which have a high failure rate with typical use (refer to Table 1), were the most common form of contraception used. This finding is consistent with a Finnish study that compared SLE patients with population controls and found that the use of effective contraception by women with SLE is relatively low [53, 54].

Table 1 Contraceptive options
Full size table

It is argued that the main approach to tackle these issues and to increase the use of effective methods is contraceptive counseling [54]. Moreover, interdisciplinary collaboration, by involving an obstetrician/gynecologist, is an important predictor of both contraceptive counseling and use [19].

Contraceptive methods

Many contraceptive options are available to physicians. Options discussed are based on the Clinical Practice Guidelines: Canadian Contraception Consensus [17•]. A detailed summary has been provided in Table 1. Each contraceptive option falls into different categories based on the three tiers of typical use effectiveness [55]. Specific names and choice of contraception options are also mentioned.

Impact on disease activity

The main precaution against using hormonal contraception in patients with SLE relates to SLE disease activity. The high female predominance in SLE during childbearing years has implicated estrogen in the development, and perhaps the worsening of the disease. In the past, many women were counseled against using CHC based on two primary outcomes of concern: a rise in SLE flares and risk of thrombosis [11]. These concerns were thoroughly studied in the last two decades, and of particular interest are two randomized controlled therapeutic trials published in 2005 [13••, 56••]. These trials evaluated whether the use of CHCs was associated with worsening SLE disease.

The first trial [13••] was a single-blind, randomized control trial (RCT) from Mexico which divided 162 patients equally in 3 groups (combined oral contraceptives (COCs), progestin-only-pills (POPs) and copper IUD) and found that overall, there was no difference in SLE global disease activity (measured by the Systemic Lupus Erythematosus Disease Activity Index [SLEDAI]), irrespective of the type of contraceptive they received. The second RCT [56••] was a double-blind, placebo-controlled trial in which 183 women with inactive (76%) or stable active (24%) lupus were randomly assigned to either COCs (n = 91) or placebo (n = 92). The authors of this study concluded that the occurrence of a severe flare was not different between the groups (7.7% in the COC group versus 7.6% placebo group). Furthermore, the 12-month severe flare rate was 0.084 for COCs and 0.087 for placebo, a difference of − 0.0028 (p = 0.95). Even though both clinical trials excluded women with severe disease and high risk of thrombosis, the results indicate that there is generally excellent tolerance of COCs concerning disease activity for women with inactive or stable active lupus.

These findings are further validated by two systematic reviews [14••, 57••]. The authors reported that based on available evidence, the use of COCs does not lead to increased flares of disease or worsening disease activity in women with inactive or stable active SLE. Hence, the benefits of use outweigh the potential risks.

Thromboembolism and contraceptives

Thrombosis and SLE

Venous thrombosis causes substantial morbidity and mortality in patients with SLE. In a review of studies relating to mortality in lupus, the three most common causes of death include infections, cardiovascular disease, and active SLE or associated organ failure [61]. According to a large cohort [62] based on multiethnic patients (n = 1930) from the Lupus Genetics Project [63], several risk factors significantly associated with venous thrombosis in SLE include positive antiphospholipid antibodies (aPL) (odds ratio [OR] 3.22, p < 10−9), immunomodulatory medication use (OR 1.40, p = 0.011), nephritis (OR 1.35, p = 0.036), smoking (OR 1.26, p = 0.011), and longer disease duration (OR 1.26 per 5 years p = 0.027 × 10−7). Among these, aPL positivity is shown to be the most critical liability for venous thrombosis in SLE. Risk is highest among patients with lupus anticoagulant (LAC) and high titer immunoglobulin G anticardiolipin (aCL) [64]. Moreover, a large prospective study published in Canada [65] demonstrated that among patients with SLE who have LAC, as much as 42% will develop a venous thrombosis within 20 years of SLE diagnosis (95% confidence interval [CI] 21% to 63%, p < 0.0001). The authors of the study also concluded that LAC is a better predictor of risk for venous thrombosis than aCL.

Thrombosis and pregnancy

In women of reproductive age, over half of all venous thrombotic events are related to pregnancy [66]. Pregnancy is considered to be a hypercoagulable state since fibrin generation and levels of coagulation factors II, VII, VIII, and X are all increased, while fibrinolytic activity and free protein S levels are decreased [67]. The risk is considered to be highest during the 6-week post-partum period followed by the third trimester [68]. This finding can be attributed to a reduction in venous flow velocity of approximately 50% occurring in the legs by 25 to 29 weeks of gestation and lasting until approximately 6 weeks after delivery, after which it returns to normal non-pregnancy flow velocity rate [69, 70]. Moreover, the increase thrombotic risk during pregnancy/post-partum period (73 in 10,000) is significantly higher than that associated with the use of CHCs (5 in 10,000) [71]. Therefore, it is essential to balance the risks associated with CHCs with the risks of unintended pregnancy.

Thrombosis and contraception

Combined hormonal contraception

Risk of venous thromboembolism (VTE)

Combined hormonal contraceptives (CHCs) have been shown to increase the risk of venous thromboembolism (VTE) [72]. This finding is well known in the general population; a meta-analysis in 2013 concluded that combined oral contraceptive (COC) use increased the risk of VTE fourfold [73]. The risk further increases in the presence of associated risk factors such as smoking, obesity, polycystic ovary syndrome, older age, immobilization [74], and thrombophilia [71]. Some women with SLE have hypercoagulability associated with antiphospholipid antibodies. Disease activity also increases the risk of VTE in SLE.

The progestin component of CHCs may affect thrombotic risk. Some studies [75, 76] have evaluated the differential thrombotic risk of third-generation (containing desogestrel, gestodene, or norgestimate) versus second-generation (containing norgestrel or levonorgestrel) oral contraceptives. The overall findings have substantiated that the thrombotic risk appears to be lower with second-generation progestins as compared to third-generation or unclassified COCs (drospirenone) [77, 78].

Myocardial infarction (MI) and stroke

A recently published meta-analysis reported a significantly higher risk of developing MI and stroke among COC users compared to nonusers [79]. Further evidence comes from a large multicenter population-based case-control study [80] (RATIO trial: Risk of Arterial Thrombosis in Relation to Oral Contraceptives), evaluating myocardial infarction (MI) and ischaemic stroke in women less than 50 years of age. For MI, the OR increased from 5.3 (95% CI 1.4–20.8) in the presence of lupus anticoagulant (LAC) to 21.6 (1.9–242.0) in the presence of LAC plus COCs. Similarly, the OR for ischaemic stroke was 43.1 (12.2–152.0) in the presence of LAC, which increased to 201.0 (22.1–1828.0) if there was a positive LAC plus COCs.

Although prospective studies dedicated to the evaluation of the thrombotic risk of CHC use in patients with aPL are lacking, the results linking CHC and venous thrombosis in the general population are in line with the safety data from the two RCTs in 2005 [13••, 56••].

Progestin-only contraceptives (POCs): progestin-only-pills (POPs), depot medroxyprogesterone acetate (DMPA) and levonorgestrel intrauterine system (LNG-IUS)

POCs are widely accepted as a lower-risk alternative to COCs for the general population. A meta-analysis [81] showed that collectively, they are not associated with increased risk of thrombosis compared with non-users (RR = 1.03, 95% CI 0.76 to 1.39). Moreover, in subgroup analysis, patients using POPs (RR 0.90, 95% CI 0.57 to 1.45) and LNG-IUS (RR 0.61, 95% CI 0.24 to 1.53) showed no significant increase in the risk of venous thromboembolism compared with non-users. On the other hand, DMPA was found to significantly increase the risk of developing thrombosis (RR = 2.67, 95% CI 1.29–5.53) when compared to non-users. However, due to the low number of DMPA users, the significance of this finding is uncertain.

Copper intrauterine device

The use of copper IUD in SLE has been evaluated in a RCT by Sanchez-Guerrero. Copper IUD (n = 54) was compared with COCs (n = 54) and POPs (n = 54) [13••], and although the risk of developing thrombosis was not studied as the primary outcome, it was reported as a secondary outcome. Patients in the copper IUD group did not develop any thrombotic event although the number of patients was small. However, a similar conclusion was reported in a follow-up review of this RCT [82].

Recommendations

Table 2 presents the recommendations for contraceptive use for women with SLE. These are based on the evidence-based medical eligibility criteria (MEC) put together by the World Health Organization (WHO) [59••] and United States Centres for Disease Control and Prevention (CDC) [60••]. They have assigned four categories to each medical condition: no restriction of use (level 1), expected benefit higher than risk (level 2), risks usually outweigh contraceptive benefits (level 3), unacceptable health risk if used (level 4).

Table 2 Summary of recommendations based on medical eligibility criteria (MEC) by the World Health Organization (WHO) [59••] and US Centres for Disease Control and Prevention (CDC) [60••]
Full size table

Moreover, the United States Medical Eligibilty Criteria (US-MEC) [60••] has separated recommendation for depot medroxyprogesterone acetate (DMPA) and copper IUD into initiation (I) and continuation (C), signifying different risk and benefit in each scenario. Regarding initiation (I) of contraception, WHO [83••] and US-MEC [84••] recommend certain precautionary evaluations prior to beginning contraception (see Fig. 1).

Fig. 1
figure 1

Based on selected practice recommendation for contraceptive use (SPR)-US-MEC [84••].

Full size image

Interactions with concomitant medication

Since a majority of patients with SLE are on multiple medications, it is important to identify and avoid any pharmacological interactions. These drug interactions are responsible for unplanned pregnancies as some result in decreased contraceptive efficacy.

Hormonal contraceptives are affected more by pharmacokinetic rather than pharmacodynamic interactions. The hepatic metabolism of ethinyl estradiol (EE) and progestins and their narrow therapeutic index makes them very sensitive to hepatic enzyme induction, the major risk being the decrease in contraceptive efficacy. Some medications may have potential interactions with CHC. For example, mycophenolate mofetil may reduce the efficacy of CHCs and preference should be given to progesterone eluting IUDs instead. CHCs may increase cyclosporine concentrations due to reduced metabolism requiring more careful side effect monitoring and possible dose reductions of cycloposporine. In the case of prednisone, CHCs may increase prednisone concentrations suggesting heightened awareness of possible steroid side effects and possible steroid dose reductions [85••, 86, 87].

Conclusion

Despite improvements in SLE including pre-conception and pregnancy and post-partum care, women with SLE are at most risk from harmful effects of pregnancy, especially if they conceive during active disease. Such pregnancies are associated with higher maternal morbidity and mortality, as well as poor fetal outcomes. Moreover, some medications used for the management of SLE are teratogenic or require careful use in the peripartum period. Considering these potential side effects, physicians should not be deterred from providing effective contraception in SLE.

When choosing between the various available contraceptive options, it is critical to recognize the contraindications, side effects, and potential for any drug interactions with concomitant SLE medications. Long-acting reversible contraceptives are considered to be the most effective and safest contraception available and should be encouraged even for patients with a history of thrombosis. Long-acting reversible contraceptives have the additional benefit of avoiding issues with compliance. Combined hormonal contraceptives are considered safe in stable-low disease activity and documented negative antiphospholipid antibodies, but are associated with a significant risk of venous thromboembolism and should be used after thorough assessment and evaluation for any contraindications. Progestin-only contraceptives provide a reliable alternative option for patients who cannot take combined hormonal contraceptives and who are not willing to use a long-acting reversible contraceptive. Despite a high rate of discontinuation due to irregular menstrual bleeding, progestin-only contraceptives are an effective option, and they are commonly used. Barrier methods are the least effective contraceptive method, and they should be reserved for situations when hormone-containing contraceptives or intrauterine devices must be avoided or are unacceptable to the patient.

Regardless, many women rely on contraceptive methods with relatively high failure rates. Therefore, it is recommended that patients with SLE are provided guidance regarding the effective and safe methods of contraception. Physicians should also facilitate access to interdisciplinary contraceptive counseling, which has been shown to increase compliance and use of effective methods.