Intrahepatic cholestasis of pregnancy – Diagnosis and management: A consensus statement of the Society of Obstetric Medicine of Australia and New Zealand (SOMANZ): Executive summary

INTRODUCTION

Intrahepatic cholestasis of pregnancy (ICP) is a pregnancy-specific liver disease,¹ Australian incidence 0.6%–0.7%, characterised by pruritus and increased total serum bile acids (TSBA), after excluding other hepatic causes of pruritus.² Important fetal complications include spontaneous preterm birth and stillbirth, but recent research has clarified the risk of stillbirth to increase only after 38 weeks when peak BA are 40–99 μmol/L, and from 35 weeks when peak BA are 100 μmol/L or greater.³ Maternal manifestations resolve postpartum.

EPIDEMIOLOGY

There is considerable ethnic variation in the incidence/prevalence of ICP,⁴ while a family history also increases the risk ten-fold.⁵

Rare genetic variants of BA transporters are reported in some ICP women,⁶ while recent genome-wide association studies have identified common sequence variation in liver-enriched genes and regulatory elements as contributing mechanisms to ICP susceptibility.⁷

ICP is more frequent in multiple pregnancy, with positive hepatitis C serology, and frequently accompanied by cholelithiasis,⁴ but these diagnoses do not exclude an ICP diagnosis.

CLINICAL FEATURES

Good practice points: The main symptom is pruritus, variably mild to very severe, affecting sleep and mentation.⁸ Typically distributed to palms/soles, it can become generalised.⁹ There is no rash; however, excoriations are common.

Urine may be dark. Jaundice and steatorrhea rarely occur. Commonly presenting after 28 weeks, 5% of cases occur in the first trimester.

DIAGNOSIS

Good practice point: ICP can be diagnosed in a pregnant woman with pruritus with non-fasting TSBA ≥19 μmol/L, a reliable threshold for diagnosis with good sensitivity and specificity,¹⁰ in the absence of other known active hepatic pathology.

Fasting TSBA ≥10 μmol/L can identify mild disease, with greater specificity but lower sensitivity (<30%).¹⁰Good practice points: Alanine transaminase (ALT) measurements are unnecessary for diagnosis and are poor markers of itch and pregnancy outcome. Increased TSBA should suggest considering other diagnoses (Table 1). A careful medical/obstetric history, with a full examination, are required.

Conditional recommendation: If gestational pruritus persists with normal TSBA, repeat testing is indicated 1–2 weekly to reassess the diagnosis of ICP. ¹¹

Previous childhood jaundice with pruritus (non-infective, post-neonatal) and/or oral contraceptive-related pruritus/jaundice warrants consideration of genetic causes.

Good practice points: Autoimmune liver disease rarely presents in pregnancy. A family history of autoimmune disorders (eg, thyroid, rheumatoid, type 1 diabetes mellitus), TSBA ≥ 40 μmol/L, or transaminases > 200 U/L (5× upper limit of normal) or presentation prior to 34 weeks, warrants an autoimmune screen (Table 2).

An acute viral screen can be considered if there is clinical suspicion (Table 2).

Anorexia, malaise and vomiting suggest other disorders, eg, acute fatty liver of pregnancy, occasionally seen alongside ICP.¹² Good practice points: Australian data report a threefold increase in risk of gestational diabetes mellitus (GDM), and a ten-fold increase in the risk of pre-eclampsia, with ICP.²

CONSEQUENCES OF ICP

Recommendation: The risks of preterm birth and stillbirth should be identified, depending on the peak TSBA titre,³ but a fall in TSBA from a peak value does not indicate a definite risk reduction.

Recommendation: Severe ICP should be diagnosed in women with peak TSBA ≥ 40 μmol/L, which is associated with increased spontaneous preterm labour, meconium-stained liquor and fetal asphyxial events. ¹³

Recommendation: Very severe ICP should be diagnosed in women with peak TSBA ≥ 100 μmol/L, which is associated with an increased risk of stillbirth (hazard ratio (HR) 30.50; 95%CI 8.83–105.30; P < 0.0001)³ (Figure 1).

MANAGEMENT

Good practice point: Depending on severity and gestation at presentation, pharmacological options may not be required, given their uncertain benefit.¹⁴ The risks of spontaneous preterm birth vs those of preterm intervention merit careful consideration with both increased in severe disease.Good practice point: Most women with ICP, especially mild disease, show progressive TSBA reduction as pregnancy progresses, even without treatment.¹⁵

PHARMACOLOGICAL OPTIONS

Recommendation: If treatment in pregnancy is thought worthwhile, oral ursodeoxycholic acid (UDCA) can be recommended, with its varied possible modes of action.¹⁴ Dosage is based on severity, initially 500 mg/day in mild cases, or 1500 mg/day in severe cases, using twice-daily dosing regimens, increasing weekly to a maximum 2000 mg/day (20 mg/kg).¹⁵ Good practice point: Compared with placebo, UDCA results in a small improvement in pruritus.¹⁴ While the 25% itch reduction was not thought helpful by women in the PITCHES trial,¹⁵ clinically some relief of itch is better than none, and may reduce demands for premature birth.

Good practice points: A recent systematic review and individual patient data meta-analysis reported no effect of UDCA on the prevalence of the composite outcome of stillbirth and preterm birth (odds ratio (OR) 1·28, 95% CI 0·86–1·91; P = 0·22). However, preterm birth, as well as the composite, was reduced when the analysis was restricted to the four available randomised controlled trials (RCTs) (OR 0·60, 95% CI 0·39–0·91; P = 0·016).¹⁶

With the event-rate for stillbirth alone in these studies being very low (1/439 compared with 3/429), no adequately powered study has yet assessed an effect on stillbirth rate.¹⁶

Availability and cost of UDCA may differ by country and region. In New Zealand, practitioners can apply for a subsidy by special authority for treatment of ICP. In Australia, UDCA is not listed on the Pharmaceutical Benefits Scheme for treatment of ICP, but most public hospitals will provide subsidised access for eligible patients. Alternatively, a private prescription for UDCA can be taken to any community pharmacy for dispensing (noting the considerable out of pocket cost associated with this approach).

ANTENATAL

Multidisciplinary management is essential. Good practice points: Urgent referral and consult/transfer of care should be arranged in severe/very severe cases, with continuing maternity care at a referral centre, especially for cases before 36 weeks, or where there are other morbidities (eg, GDM or pre-eclampsia) or multi-fetal gestation.

After diagnosis of ICP, TSBA should be assessed with regular clinical review (at least monthly up to 30 weeks, fortnightly in the third trimester, or weekly in severe cases) (Fig. 1). Rapidly rising TSBA at this gestation might impact decisions about early birth. Once TSBA reach ≥100 μmol/L, there is no value in further testing to identify additional risk. If UDCA is commenced, blood should be drawn for TSBA shortly before the next dose: UDCA, itself a BA, may contribute up to 60% in the assay,²⁰ although fasting is not required.

Prothrombin time/international normalised ratio (PT/INR) and/or activated partial thromboplastin time (APTT), should be measured after diagnosing severe/very severe ICP, and repeated before induction of labour. INR ≥1.4 or APTT ≥ 40 sec require treatment with intravenous (IV) vitamin K 10 mg with further weekly assessment.

Oral vitamin K may be ineffective due to poor gastro-intestinal absorption.

BIRTH PLANS

Good practice points: Birth plans should be made in consultation with the woman, including discussion of risks-vs-benefits of planned birth < 39 weeks, if:

(Some authorities recommend earlier delivery for women with very severe ICP (eg, 35–36 weeks):²¹ the evidence-base for this is not strong.)

Risk of stillbirth for TSBA 40–99 μmol/L is 0.28% (comparable to background rates). This risk of stillbirth increases to 3.4% when TSBA exceeds 100 μmol/L, most notably from 35 to 36 weeks.³

ANTENATAL ADMISSION

Good practice points: Outpatient management is recommended if TSBA < 40 μmol/L and the woman remains clinically well.Admission for assessment and planned birth may be considered if TSBA ≥ 100 μmol/L, especially after 36 weeks. For those in rural/regional areas, consultation with a referral centre will inform management. Further outpatient management can be considered if treatment reduces the TSBA to <40 μmol/L.

Further research is required to determine if timing of birth may be modified by the response to treatment.

FETAL SURVEILLANCE

Good practice point: In the absence of other maternal/fetal pregnancy complications, growth ultrasounds and cardiotocography (CTG) are not required.

Cord Doppler flow measurements have not demonstrated changes, nor prediction of fetal risk, in ICP pregnancies.²²Good practice point: Decrease/absence of fetal movements needs urgent review and is an indication for fetal assessment/CTG.

INTRAPARTUM MANAGEMENT

Good practice point: Continuous electronic fetal monitoring (EFM) intrapartum should be offered to women with severe/very severe ICP.

(Most women with severe ICP do not go into term spontaneous labour and are usually birthed electively. Thus, they are offered EFM intrapartum for reasons other than ICP.)

Good practice points: Coagulation should be checked in women with severe/very severe ICP with parenteral vitamin K (10 mg IV) administered if PT/INR or APTT are abnormal.

Active management of the third stage of labour is recommended, although reports of increased risks of postpartum haemorrhage (PPH) in ICP are conflicting.^{23, 24}

Parenteral vitamin K should be administered to all neonates of women with ICP, especially if their mothers were treated with rifampicin or cholestyramine.²⁵

POSTPARTUM MANAGEMENT

Maternal: Good practice point: cease pharmacological treatment at birth.Pruritus usually resolves 1–2 days postpartum. Maternal jaundice usually resolves in the first postnatal week.

TSBA commonly normalise within a week but should be checked, along with serum transaminases, at six weeks postpartum. Good practice point: Should biochemical abnormalities persist, investigation to exclude underlying liver disease is required.

Good practice points: Hormonal contraception is not contraindicated. If pruritus recurs in women using hormonal contraception, further investigation is required. Non-hormonal contraception should be considered if liver function remains abnormal.

Neonatal: routine care and monitoring are required.

COUNSELLING

Neonatal outcomes: once delivered, good outcomes can be expected.

Long-term outcomes: few studies have investigated long-term outcomes for children of pregnancies complicated by ICP.

Long-term maternal outcomes: the risk of ICP recurrence in a subsequent pregnancy is 40%² to 70%.²⁶ Recommendation: Women experiencing severe ICP, are at risk of chronic liver disease and should have long-term follow-up. While pruritus usually resolves postpartum, affected women have an increased risk of later hepatobiliary disease, including liver and biliary-tree cancer, and later immune-mediated diseases – diabetes mellitus, thyroid disease, psoriasis, inflammatory polyarthropathies, Crohn's disease (all with hazard ratios of 1.3–1.5).²⁷ There is no consensus on appropriate follow-up. Good practice point: An annual check of liver biochemistry seems prudent.

FUNDING

There is no financial or material support to report.