Intrahepatic cholestasis of pregnancy

This is the third edition of the guideline. The first edition was published in 2006 and the second in 2011 under the title Obstetric Cholestasis. Advise women with isolated ICP and a singleton pregnancy that the risk of stillbirth only increases above population rate once their serum bile acid concentration is 100 micromol/L or more. This guideline summarises the evidence regarding the diagnosis, and the maternal and fetal risks of intrahepatic cholestasis of pregnancy (ICP), previously called obstetric cholestasis. It provides guidance regarding the different care options available. These should be considered in conjunction with the wishes of the woman, as part of shared and informed decision-making. While some high quality randomised controlled trials in ICP have now been completed, many publications do not have such a rigorous design, and this limits the ability to provide detailed evidence-based recommendations for specific aspects of care. Areas of uncertainty are highlighted along with recommendations for future research in this field. Within this document we use the terms pregnant woman and women's health. However, it is important to acknowledge that it is not only people who identify as women for whom it is necessary to access care. Obstetric and gynaecology services and delivery of care must therefore be appropriate, inclusive and sensitive to the needs of those individuals whose gender identity does not align with the sex they were assigned at birth. The care of women and pregnant people with ICP is driven by concern from women and from healthcare professionals over the potential increased risk of stillbirth. Reduction of stillbirth is a priority in maternity care in the UK.1 Prevalence is influenced by genetic and environmental aspects and varies between populations. In the UK, ICP affects 0.7% of pregnancies in multi-ethnic populations, and 1.2%–1.5% of women of Indian-Asian or Pakistani-Asian origin.2, 3 Intrahepatic cholestasis of pregnancy is a multifactorial condition. It is characterised by pruritus in the absence of a primary skin condition, with abnormal maternal bile acid concentrations. The onset of symptoms is most common in the third trimester, but can be earlier in pregnancy.4 Alternative diagnoses (such as pre-eclampsia) should always be considered before a diagnosis of ICP is made; it is also possible for other conditions to co-exist. Pruritus and raised bile acid concentrations should return to normal after birth. Ideally, all women with ICP should have liver function tests (including bile acids) checked after birth, as a proportion may have persistent abnormalities suggestive of additional or alternative comorbidities (such as non-alcoholic fatty liver). Few studies have reported comprehensive postnatal follow-up of women to assess for additional diagnoses. There are no clinical features or laboratory patterns that are unique to ICP, as other conditions can cause itching, or raised bile acid concentrations in pregnancy. Around 25% of pregnant women develop itching2, 5; the majority of these do not have and do not develop ICP. Historically, ICP has been diagnosed in women on the basis of self-reported itching together with elevation of any of a wide range of liver function tests beyond pregnancy-specific limits.6 There is now increasing evidence that in singleton pregnancies, most liver function tests do not reflect risk of fetal demise and that only maternal total bile acid concentrations results are associated with the risk of stillbirth. A meta-analysis of 23 studies involving 5557 women with ICP and 165 136 healthy controls, and the first individual patient data analysis of 5269 women with ICP from 27 studies7 has been published since this guideline was last updated. In singleton pregnancies, stillbirth was associated with maximum total bile acid concentration, especially over 100 micromol/L. In pregnancies with co-morbidities that themselves may impact on pregnancy outcome (such as multifetal, diabetic, pre-eclamptic pregnancies), these must be taken into account when considering risks and care options. Bile acid concentrations are not associated with intensity of itching.8 Other liver blood tests, such as alanine transaminase or aspartate transaminase are not associated with pregnancy outcome.7 In light of this, the consensus is now that the diagnosis of ICP requires elevated maternal bile acid concentrations, and that women and pregnant people with itching and isolated raised transaminases alone (with normal bile acid concentrations) should not be given a diagnosis of ICP.9 This is supported by the recent systematic review described above, in which there was no association between abnormal maternal transaminase concentrations and stillbirth.7 Most published studies to date have included women with ICP diagnosed on the basis of itch and elevated bile acids above the laboratory reference range. In a study of 560 women a pregnancy specific reference range for non-fasting bile acids was calculated with an upper limit of normal of 18 micromol/L.10 In light of the meta-analysis and individual patient data analysis showing that stillbirth risk is not linked with alanine transaminase levels, but is linked with peak bile acid concentration,7 the suggested terminology for pregnant women with otherwise unexplained itching is outlined in Table 1. The clinical issues for women and pregnant people with ICP may include coping with the itching, monitoring options during the pregnancy, options for controlling maternal symptoms, reducing fetal risk, preterm birth, difficulty sleeping, anxiety about the condition, and optimal timing of birth. Itching varies in nature between different women and for some women at different times, from mild to unbearable and from focal to widespread, and for some women this can have an adverse impact on their mental wellbeing. Liver failure (impaired synthetic function such as prolonged prothrombin time, or metabolic dysfunction such as hypoglycaemia) is not a typical feature of ICP. The Cochrane Library and electronic databases (DARE, EMBASE, Trip, MEDLINE and PubMed) were searched looking for the following terms in the title or abstract ‘cholestasis’, ‘intrahepatic cholestasis’, ‘obstetric cholestasis’, ‘bile acids and salts’ and ‘liver function test’. The search was restricted to articles published until August 2017. The full search strategy is available to view online as supporting information. A further search was undertaken up to February 2021 and additional articles included as appropriate. The full literature search is available to view online as supporting information (Appendices S1 and S2). This Royal College of Obstetricians and Gyneacologists (RCOG) guideline was developed in accordance with the standard methodology for producing RCOG Green-top Guidelines.11 New onset pruritus in pregnant women, if associated with rash is unlikely to be ICP. If the itchy skin looks abnormal (other than excoriations) then another cause should be considered. Liver function tests and bile acids are not required routinely. Clinicians should be aware however, that skin conditions (e.g. eczema) and ICP can co-exist. If the itchy skin looks normal, or there is only skin trauma due to scratching, the diagnosis may include gestational pruritus, or ICP (see Table 1); measurement of bile acid concentrations and liver function tests should be undertaken. Raised bile acid concentration of 19 micromol/L or more in pregnancy supports a diagnosis of ICP. [Evidence level 4] When clinically indicated, bile acid measurements should be taken at a convenient time, and do not need to be performed fasting. By taking a non-fasting upper limit of normal up to 19 micromol/L for bile acid concentrations, almost 20% of women previously considered to have ICP (as their bile acid concentrations are below 19 micromol/L but above the standard laboratory cut off) do not have this diagnosis. Published data indicate that pregnant women with raised bile acids less than 19 micromol/L are not at increased risk of stillbirth.10 In addition, as prandial readings are higher than fasting, this approach maximises the chance of detecting peak bile acid readings that are of greater clinical importance for preventing adverse pregnancy outcome. Itching of normal skin, liver dysfunction and elevated bile acid concentrations are non-specific and have a wide range of causes. A healthcare professional should carry out a structured history and examination, and consider other potential diagnoses: these may be pregnancy specific (including pre-eclampsia) or coincidental to the pregnancy (comprehensively reviewed by Walker et al).12 Drug reactions, allergic reactions, and urticaria should form part of the differential diagnosis. [Evidence level 4] In women and pregnant people with persistent itch of normal skin and normal blood results, an initial diagnosis of gestational pruritus should be considered. Women can go on to develop ICP up to 15 weeks after a diagnosis of gestational pruritus.13 If itching continues for these women, they should be offered review with repeated liver function tests and bile acid measurement as clinically indicated. The frequency and duration of review and tests should be determined on an individual basis, but might be based around scheduled care. The gestational age is also relevant in determining test frequency, particularly later in the third trimester when a diagnosis of ICP may change care around the timing of birth (aiming to reduce the risk of stillbirth) [see Section 5.2]. Use pregnancy-specific reference ranges for liver function tests.6 [Evidence level 4] Pruritus and biochemical abnormalities usually persist throughout pregnancy in women with ICP, although it is very common for them to fluctuate. However, in a few women, pruritus and biochemical abnormalities will resolve completely for the remainder of the pregnancy; clinicians may then need to reconsider the cause of the original symptoms and why resolution has occurred. There are many causes of transient liver function test abnormalities, such as drug reactions (e.g. to antibiotics) or non-specific viral illnesses. When resolution occurs during pregnancy, it is unlikely that the original diagnosis was correct. In discussion with the woman or pregnant person, ongoing care can usually return to normal, and decisions about timing of birth should be based on usual obstetric practice, although there should be greater caution if bile acid concentrations have been markedly raised (e.g. 100 micromol/L or more). [Evidence level 4] Previous RCOG guidelines14 have recommended routine laboratory and imaging investigations to exclude other causes for the clinical picture of ICP, including viral and autoimmune tests and liver ultrasound. A recent retrospective review of over 500 pregnant women with raised bile acid concentrations suggests that the likelihood of identifying a viral, autoimmune, or structural cause for the itching and liver derangement that was not suspected on other clinical grounds is extremely low as no new diagnoses were made following investigations.15 Therefore, routine use of other investigations is no longer recommended. The UK National Screening Committee does not recommend routine screening for hepatitis C in pregnancy due to lack of evidence of benefit16; the same uncertainties apply to pregnant women with ICP. Routine hepatitis C testing is therefore not currently recommended in women with suspected or proven ICP. Additional investigations (including for hepatitis C) should be considered in women and pregnant people with an atypical or uncertain picture of ICP. This may include women with markedly elevated transaminases, early onset of ICP in the first or second trimester, a rapidly progressive biochemical picture, any features of liver failure or evidence of acute infection, or if resolution does not occur after birth. [Evidence level 2+] Three cohort studies of 223 women,17 531 women15 and 745 women18 with ICP who had routine coagulation testing reported no cases of prolonged prothrombin time in women with uncomplicated clinical presentations. The small number of abnormal results were in women with alternative diagnoses (such as acute fatty liver of pregnancy). Coagulation testing is therefore not recommended routinely for women and pregnant people with uncomplicated ICP. It should be considered on an individual basis especially when failure of liver synthetic function or failure of fat absorption is suspected. [Evidence level 2+] Women who develop pruritus and abnormalities in liver function and bile acids in the first or second trimester and especially in the first trimester are more likely to have an underlying genetic predisposition or an alternative or additional diagnosis. Input from a hepatologist and/or a clinician with a special interest in cholestasis to discuss investigations and treatment options should be considered. A postnatal referral should also be considered for women and pregnant people who do not have resolution of itch and biochemical abnormalities after birth. [Evidence level 4] For many women with ICP, itching will stop very soon after birth; in the majority it stops in the first few hours or days. Liver function tests are non-specific and can become abnormal during birth. Alanine transaminase and aspartate transaminase are found in smooth muscle, breast and red blood cells and may be elevated for other reasons in the immediate post-partum period. [Evidence level 4] Women with ICP who have no other diagnoses are usually clinically well; liver function tests and bile acids should not be measured until at least 4 weeks after birth, to allow time for levels to return to a normal range.19 If the woman or pregnant person is clinically unwell, other or additional diagnoses should be suspected and liver function testing should be repeated sooner, as clinically indicated. [Evidence level 4] Itching is the main symptom of ICP. The itching is not specific to any single location; it is often generalised and may affect the palms of the hands and/or the soles of the feet; it ay vary in intensity.2 For women and pregnant people with gestational pruritus or ICP, there is poor correlation between severity of itch and level of bile acids,8 and regardless of the diagnosis, itch can be very severe for some women and may negatively impact their emotional wellbeing and mental health. The itching is often more pronounced at night, which can interfere with sleep. [Evidence level 2+] Additional symptoms of cholestasis, such as dark urine and pale stools, are infrequently reported.4 Steatorrhoea may occur,20 and women with this symptom may have malabsorption of vitamin K. Jaundice is rare, affecting less than 1% of women with ICP,15 and tends to be mild if it occurs. [Evidence level 2+] The incidence of pre-eclampsia was higher in women with ICP (odds ratio [OR] 3.7 [95% CI 3.2–4.3]): 12.2% of women with ICP had pre-eclampsia compared with 3.4% of women without ICP (228/1876 versus 3385/94 386).7 As pre-eclampsia can be diagnosed at any gestation from the mid-second trimester, healthcare professionals should ensure that women and pregnant people with ICP receive ongoing blood pressure and urinalysis screening for pre-eclampsia alongside review for ICP.21 [Evidence level 1+] In a meta-analysis of more than 5000 women with ICP, rates of gestational diabetes were higher in women with ICP (OR 2.4 [95% CI 2.1–2.8]): 13.2% of women with ICP had already been diagnosed with gestational diabetes compared with 5.9% of women without ICP (239/1806 versus 5571/94 384).7 Additional testing for gestational diabetes is not currently recommended; risk assessment and testing for gestational diabetes should follow national guidelines.22 [Evidence level 1+] A large Swedish population-based study of 11 388 women with ICP and 113 893 controls found that women who have had ICP had an increased likelihood of later being diagnosed with hepatobiliary disease (hazard ratio (HR) 2.62 [95% CI 2.47–2.77]); 15% in women with ICP versus 6.3% in controls), predominantly due to gallstone disease (HR 2.72 [2.55–2.91]; 11.6% versus 4.6%).23 However gallstones are common, affecting 5–25% of adults in high income countries,24 and it is unclear whether gallstone disease predates ICP in such women. [Evidence level 2–] The same study found an association between ICP and immune-mediated diseases later in life (HR 1.28 [1.19–1.38]; 7.2% versus 5.8%). These included diabetes (HR 1.47 [1.26–1.72]; 1.7% versus 1.2%), thyroid disease (HR 1.30 [1.14–1.47]; 2.5% versus 2.0%), psoriasis (HR 1.27 [1.07–1.51]; 1.4% versus 1.1%), inflammatory polyarthropathies (HR 1.32 [1.11–1.58]; 1.3% versus 0.9%) and Crohn's disease (HR 1.55 [1.14–2.10]; 0.4% versus 0.3%), but not ulcerative colitis (HR 1.21 [0.93–1.58]; 0.6% versus 0.5%).25 Most of these conditions remain at low absolute incidence. The benefit of routine regular screening for these conditions is not proven in women and pregnant people after ICP and is not currently recommended. [Evidence level 2–] Women with ICP have a reported small increased chance of a subsequent diagnosis of other conditions, such as hepatitis C. The UK strategy for hepatitis C detection is based on additional investigations on high-risk groups (e.g. those who have hepatitis B) and does not at present include women with current or previous ICP.26 In general, after an episode of ICP women do not require additional screening nor follow up. Stillbirth remains the major concern for women and pregnant people with ICP and for their healthcare practitioners. A large systematic review and individual patient data meta-analysis of women with ICP reported that, for singleton pregnancies, the risk of stillbirth only increased above population rate once serum bile acid concentrations were 100 micromol/L or more (Table 2).7 The national stillbirth rates from 28 weeks' gestation for 2015 for countries contributing to these ICP data varied from 0.18% to 0.72% depending on country; the UK stillbirth rate was 0.29%.7 The pathophysiology of stillbirth in ICP is uncertain, but it is thought that bile acids may cause an acute fetal anoxic event possibly due to fetal arrhythmia27 or acute placental vessel spasm.28 In singleton pregnancies, stillbirth was associated with peak total bile acid concentration but not with alanine transaminase.7 [Evidence level 1+] A 12-month UK Obstetric Surveillance System (UKOSS) study in 2010–2011 reviewed 669 cases of ICP in singleton pregnancy with bile acids 40 micromol/L or more across the UK which included 10 stillbirths.29 Of these, seven had coexistent pregnancy complications (three had gestational diabetes; two had pre-eclampsia; two had non-specified complications). These differences remained significant against national data and suggest that women with ICP and other comorbidities warrant additional surveillance. [Evidence level 2+] The aetiology of adverse perinatal outcomes, including stillbirth, in multifetal pregnancies is multifactorial. The risk of stillbirth in multifetal pregnancies is higher than in singleton pregnancies.30 [Evidence level 2–] One retrospective cohort study from China specifically evaluated ICP in twin pregnancies.31 They reviewed 129 twin pregnancies complicated with ICP and 1793 twin pregnancies without ICP (2006–2014). There was an increased risk of stillbirth in twin pregnancies with ICP compared with twin pregnancy without ICP (3.9% versus 0.8%, aOR 5.75 [95% CI 2.00–16.6]). This was further stratified as a stillbirth risk of 3.3% in women with bile acids of 10–39 micromol/L and 5.1% in women with bile acids of 40 micromol/L or more. Stillbirths with ICP in twin pregnancies occurred between 33–35 weeks' gestation, compared to 36–38 weeks' gestation among singletons. [Evidence level 2–] In the meta-analysis of more than 5000 women with ICP, women with bile acids ≥40 micromol/L had an increased overall risk of both spontaneous preterm birth (OR 3.47 [95% CI 3.06–3.95]) and iatrogenic preterm birth (OR 3.65 [1.94–6.85]),7 the latter likely reflecting the policy of ‘active management’ with planned early birth (despite a limited evidence base for this approach).32 The percentage of women with singleton pregnancies who gave birth before 37 weeks increased with increasing bile acid concentration: 16.5% of women with bile acids below 40 micromol/L (373/2264), 19.1% of women with bile acids 40–99 micromol (261/1368), and 30.5% of women with bile acids 100 micromol/L or more (157/514). The majority of multifetal pregnancies were born preterm.7 [Evidence level 1+] In the same meta-analysis of more than 5000 women with ICP, there was an increased chance of meconium stained amniotic fluid (of any grade) in women with ICP: OR 2.60 (95% CI 1.62–4.16).7 The 2010–2011 UKOSS study of 713 women with bile acids 40 micromol/L or more found that these women had meconium stained amniotic fluid at lower gestational ages, and more commonly at 35–38 weeks, when compared with women without ICP.29 The presence of meconium stained amniotic fluid in labour should be managed using national guidance.33 [Evidence level 1+] ICP is associated with a small increase in admission to the neonatal unit: OR 1.47 (95% CI 1.03–2.10).7 There was, however, no difference in the rate of a neonatal 5 min Apgar score of less than 7,7 (which is associated with increased neonatal morbidity and mortality): OR 1.41 (0.95–2.10).34, 35 [Evidence level 1+] The 2010–2011 UKOSS study on 713 UK women with ICP (bile acids 40 micromol/L or more) showed that 45% of neonatal admissions were due to preterm birth, and 30% due to respiratory problems.29 Their study had a preterm birth rate of 25%, with a 15% rate of meconium stained liquor, but no cases of meconium aspiration. The median duration of stay in the neonatal unit was 7 days (IQR 2.25–13.75 days).29 [Evidence level 1+] The frequency and content of monitoring for women and pregnant people with ICP should be determined in conjunction with the woman or pregnant person and based on the amount of discomfort or distress they experience, bile acid concentrations, gestational age and the presence of other morbidities. This might incorporate review of diagnosis, discussion of maternal and fetal wellbeing, treatment of pruritus, and need for further biochemical testing. [Evidence level 4] In ICP, there is evidence that cardiotocography (CTG) monitoring or biophysical profile do not predict stillbirth. Several studies describe fetal death despite close surveillance and previously normal ultrasound scans (including fetal Doppler measurements), biophysical profile, and/or CTG monitoring.31, 36-38 [Evidence level 3] ICP is not associated with fetal growth restriction, with no difference in birthweight centiles compared with babies born to women without ICP,7 and therefore strategies for antenatal monitoring for placental insufficiency are unlikely to be beneficial in women with isolated ICP. [Evidence level 3] All pregnant women and pregnant people should be advised to monitor the quality and quantity of their fetal movements, and report any reduction or change to their local maternity unit immediately, as recommended in national guidance. 1 Maternal detection of movements is simple and not time consuming for women or staff, but its specific role in monitoring pregnancies complicated by ICP has not been assessed. [Evidence level 4] The role of drug treatment in ICP is to try to reduce maternal itching (which may be of variable intensity and is unrelated to bile acid concentrations). There is no evidence that routine medical treatment improves maternal raised bile acid concentrations or perinatal outcomes.39 [Evidence level 1+] Although there is minimal high-quality evidence to endorse topical emollient treatment in women with ICP and it is not a disease-modifying drug, there is consensus that such treatment may relieve some of the discomfort associated with itching and has no known harmful effects. [Evidence level 4] Chlorphenamine has antihistamine properties and may have sedative side-effects in some women. The effectiveness of the treatment is uncertain in women and pregnant people with ICP, and relief may be more related to its sedative action than a direct effect. There is experience of using chlorphenamine in other conditions in pregnancy (such as hay fever) and harmful effects have not been reported. Other common antihistamine agents including loratadine and cetirizine are also used in pregnancy for other indications but do not have sedative side-effects. [Evidence level 4] Evidence from randomised controlled trials shows that there is no reduction in adverse perinatal outcomes in women prescribed ursodeoxycholic acid, compared to women in the placebo group.39, 40 No sub-group (e.g. based on maternal bile acid concentrations, or gestational age at presentation) was identified that might benefit. [Evidence level 1+] There is a small (around 5 mm, on a linear 0 to 100 mm itch scale with 0 ‘no itch’ and 100 ‘worst imaginable itch’) reduction in maternal itch in women taking ursodeoxycholic acid.39 Women and clinicians considered that a reduction of at least 30 mm on the itch scale would be clinically relevant and worthwhile;40 the majority of women would therefore not consider this a useful treatment. It remains possible that some women's and pregnant people's itching may reduce with ursodeoxycholic acid, but it is not clear how such women might be identified. A recent secondary analysis of the largest trial could not identify a cohort based on bile acid concentration or itch score who would benefit.8 [Evidence level 1+] A systematic review and individual participant meta-analysis of ursodeoxycholic acid in ICP included four randomised controlled trials of over 800 women (of whom 183 had bile acid of 40 micromol/L or more) with a primary outcome of stillbirth and a composite secondary outcome of stillbirth and preterm birth.41 Ursodeoxycholic acid had no impact on the primary endpoint. Spontaneous preterm birth under 34 weeks' gestation was not reduced (5/387 women taking ursodeoxycholic acid versus 6/366 women taking placebo, aOR 0.75, 95% CI 0.23–2.51, p = 0.65). Spontaneous preterm birth under 37 weeks' gestation was reduced in women taking ursodeoxycholic acid compared with placebo (18/387 versus 32/366 aOR 0.46, 95% CI 0.25–0.86, p = 0.015). In women with bile acid concentrations 40 micromol/L or more who are 34–36 weeks' gestation, ursodeoxycholic acid may offer some benefit in reducing late preterm birth. However, as with other circumstances where preterm birth occurs42 it is not clear that this reduction results in any benefit to the baby. The optimal starting gestation and dosing regimen are unclear. Some women and pregnant people with bile acid concentrations of 40 micromol/L or more may wish to take ursodeoxycholic acid with a view to prolonging gestation, but as this does not prevent stillbirth, the advantage of doing so may be less clear, especially for those with bile acids over 100 micromol/L. In the largest trial, maternal bile acid concentrations were found to be higher in the group treated with ursodeoxycholic acid,39 possibly as a result of standard laboratory assays being unable to distinguish between endogenous and exogenous sources. Ursodeoxycholic acid cannot therefore be recommended for the purpose of reducing this biochemical marker of disease. In the same trial, women treated with ursodeoxycholic acid had lower alanine transaminase levels than those taking placebo, but the clinical implications of this are uncertain, as alanine transaminase levels have no association with stillbirth.7 [Evidence level 1+] Use of rifampicin has been reported largely in single cases43 and by questionnaire survey of affected women,44 but there is no evidence from randomised controlled trials to support its routine use in ICP. Further research is underway to evaluate its use in women with ICP.45 In women and pregnant people with early-onset severe disease, an opinion from a specialist in ICP should be sought before considering rifampicin treatment. [Evidence level 3] The experience of experts is that the large majority of women with ICP will not have evidence of reduced fat absorption9 and routine use of vitamin K treatment is not indicated. If women have symptoms such as steatorrhoea,20 coagulation assessment should be performed and use of vitamin K treatment considered (prescribed as a water-soluble formulation such as menadiol sodium phosphate at a dose of 10 mg daily). [Evidence level 4] Active care in ICP, usually referring to planned birth around 38 weeks' gestation, came into practice in many settings, including the UK, despite inadequate evaluation of its benefit or an understanding of which women with ICP might be at increased risk of adverse perinatal outcomes. Previous studies had reported on cohorts of women with ICP, often after introduction of active care, and speculated that low stillbirth risk is related to such a policy,46, 47 but few had evaluated prognostic factors to allow better stratification to tailor timing of birth. [Evidence level 1+] A large systematic review and individual patient data meta-analysis of women with ICP has reported that the risk of stillbirth is 0.13% in women with peak bile acids less than 40 micromol/L, which is not higher than background population risk.7 Although the risk of stillbirth remains low throughout gestation for these women, the benefits of continuing the pregnancy after 40 weeks' gestation may be outweighed by the risk, therefore it is reasonable to discuss with the woman or pregnant person whether they wish to continue the pregnancy or have a planned birth. [Evidence level 1+] The risk of stillbirth in women with peak bile acids of 40–99 micromol/L was 0.28%. This was not higher than overall back