Cholestasis, when translated from Greek, describes the “standing still of bile.” This disturbance in bile flow can occur anywhere in the biliary system and can result from extrahepatic mechanical obstruction, intrahepatic biliary tree pathology, or individual hepatocyte dysfunction. Commonly, the disturbance does not cause complete disruption in flow and often only certain components of bile become pathologic. The clinical and biochemical presentations of cholestasis can thus vary. Conventionally, cholestatic liver diseases are first divided into those with extrahepatic versus intrahepatic etiologies. Cholestasis of pregnancy is an intrahepatic disorder unique to the pregnant state that resolves rapidly following delivery.1 It is the most common pregnancy-specific liver disorder.2 The etiology of intrahepatic cholestasis of pregnancy (ICP) is poorly understood but is likely multifactorial with genetic, environmental, and hormonal contributions to disease development and severity. Equally as unclear are the mechanisms by which ICP increases the risks of preterm delivery, meconium-stained amniotic fluid, and fetal demise. In this article, we review the epidemiology, pathophysiology, risk factors, laboratory findings, complications, treatment, management, and current evidence surrounding ICP.
ICP was first described in the literature in 1883 as “recurrent jaundice in pregnancy” that resolved with delivery.3 The disorder was commonly referred to as “icterus gravidarum, idiopathic jaundice of pregnancy, hepatosis of pregnancy, or hepatotoxemia of pregnancy.” It was linked to increased circulating levels of steroids in maternal serum and thought to occur secondary to hepatic enzyme dysfunction resulting in disturbances in liver bile function.4 It was not until 1965 that patients with this third trimester pruritic condition were noted to have abnormally increased serum bile acid levels when compared with nonpruritic controls.5 It was even later still that the association between cholestasis and adverse fetal outcomes were described. Danish investigator Friedlaender first reported in 1967 that “in [icterus gravidarum] the prognosis for the mother is good, whereas the prognosis for the child is relatively poor, partly because of the high rate of premature deliveries.”4 A decade later, Finnish investigator Laatikainen first reported significantly increased rates of stillbirth, fetal distress, and meconium-stained amniotic fluid in a small Finnish cohort.6 By the early 1990s, interest in the link between cholestasis and intrauterine fetal demise grew and British investigators Fisk and Storey7 and Californian investigators Alsulyman et al8 first began reporting on these outcomes in small cohorts of pregnancies complicated by cholestasis, though the causal mechanism remained largely unknown. It was only in the last 15 years that robust data began to emerge with larger cohort studies and the first prospective investigations into adverse perinatal outcomes related to cholestasis.
The incidence of ICP varies greatly among ethnic groups and geographic regions, likely reflecting differences in genetic susceptibility and environmental factors (Table 1). ICP occurs most frequently in South American and Northern European ethnic groups and in women with multiple gestation, advanced maternal age, conception after in vitro fertilization, hepatitis C infection, and a family history of cholestasis in pregnancy.2 ICP is also known to have seasonal patterns with increased incidence during winter months.2 In Europe the incidence of ICP ranges from 0.5% to 1.5% with the highest rates in Scandinavia. Incidence of cholestasis in the United States ranges from 0.32% in predominantly white populations to 5.6% in predominantly Latina populations.9 Worldwide, the highest incidence is found within the Mapuche Indians, a group of indigenous Chileans in which cholestasis has been reported to complicate >27% of pregnancies.10
Cholestasis in the general population can be caused by many conditions. Conventionally, cholestatic liver diseases are first divided into those with extrahepatic versus intrahepatic etiologies. Of the intrahepatic causes of cholestasis, the most well-studied conditions include primary biliary cirrhosis, primary sclerosing cholangitis, drug-induced cholestasis, hepatitis, and rare genetic conditions caused by defects in biliary transport proteins. Although ICP is the most common liver disease of pregnancy, its cause and course are still poorly understood.
The molecular structure of the bile acid hinges on its saturated tetracyclic hydrocarbon perhydrocyclopentanophenanthrene backbone known more commonly as the steroid nucleus. This common molecular structure is shared by most human hormones. Modern bile acids (often referred to as bile salts) are the 24-carbon steroid-nucleus–containing end products of hepatic cholesterol metabolism. Bile acids are highly hydrophobic and are often sulfated or aminated to increase their solubility in water. Critical to their function, human bile acids are almost always conjugated to either glycine (75%) or taurine (25%) rendering them water soluble.11 The primary function of bile acids in humans is the transport of lipids (endogenous cholesterol, dietary fat) and fat-soluble vitamins to and from the gastrointestinal tract. Bile acids are also involved in gastrointestinal calcium absorption.
Within hepatocytes, cholesterol is used to synthesize bile acids via multiple enzymatic reactions that culminate in the creation of the 2 primary human bile acids, cholic acid, and chenodeoxycholic acid. This process is highly dependent on the enzyme 7α-hydroxylase (CYP7A1), a cytochrome P450 enzyme, which is unique to the liver and is the rate-limiting step in bile acid production. The farnesoid X receptor (FXR, also known as the bile acid receptor) is an intracellular nuclear receptor found in hepatocytes and throughout the gut. It belongs to a family of ligand-activated proteins which, when activated by extracellular and intracellular steroid hormones, affect the transcription of cell proteins and enzymes. The FXR serves as a sensor that participates in multiple regulatory processes involving bile acids, cholesterol, triglyceride, and glucose metabolism. Bile acids bind to and activate the FXR with high affinity that results in suppression of CYP7A1. It is through this activation of the FXR and inhibition of CYP7A1 activity that bile acids exert negative feedback inhibition on their own synthesis (Fig. 1).
Once formed, bile acids are exported out of hepatocytes via specific bile acid transport proteins, which belong to a superfamily of transport proteins referred to as the ATP-binding cassette (ABC) transporters. These types of transport proteins are found throughout the human body and are first divided alphabetically into subfamilies and then numerically. Most bile acid transport pumps belong to subfamily B (ABC-B). Numerous protein pumps are involved in bile acid secretion; however, the 2 primary bile acid transport proteins are the ABCB-11 transport pump, known as the bile salt export pump (BSEP), and the ABCB-4 transport pump, known as the multidrug resistance transport protein 3 (MDR3). These pumps transport bile salts into the immediately adjacent bile canaliculi, a network of small tube-like spaces that converge progressively to form bile ductules, ducts, and eventually drain into the gallbladder and small intestine. Theses pumps utilize ATP to secrete bile acids into bile against the concentration gradient and characterize the rate-limiting step of bile acid transport. Once bile acids are secreted from the hepatocyte, the bile acid osmotic gradient is the predominant driving force behind the movement of bile. In addition, the bile canaliculi collecting system is sealed by tight junctions and has a contractile mechanism that allows for supplementary peristalsis-like movement of bile acids from hepatocytes to the small intestine. Once secreted into neighboring canaliculi, bile acids quickly self-associate and are incorporated along with phospholipids (primarily phosphatidylcholine) and sterols (primarily cholesterol) into small aggregates known as micelles. This solubilization of lipophilic phospholipids and cholesterol is the primary function of bile acids. Bile itself is a complex mixture of lipids, proteins, mineral salts, vitamins, and various other elements but bile acids comprise ~67% of the total composition of bile excluding water.12
After arrival to the gastrointestinal tract, bile acids maintain their emulsifying function and act as detergents to break apart and encapsulate lipids and lipid-soluble vitamins for transport. The majority of bile acids are then actively recovered by transport pumps in the ileum. They then move through the portal system, usually bound to albumin, and are actively transported back into hepatocytes via sodium-dependent taurocholate co-transport polypeptides (NaTCP). These carrier proteins have a high specificity and affinity for bile acids that results in a very high rate of bile acid recovery (up to 90%). This exceedingly efficient first-pass extraction confines bile acids to the enterohepatic circulation and results in low levels of these compounds in peripheral blood.
Primary bile acids are those that are produced in the liver, conjugated to taurine or glycine, and secreted into bile. Once primary bile acids reach the gastrointestinal tract, many will undergo bacterial metabolism resulting in the formation of over 50 different secondary bile acids. Subsequent modifications to secondary bile acids result in the formation of tertiary bile acids, of which ursodeoxycholic acid (UDCA) is the most well-known.13
Cholestasis is defined by a stagnation in bile flow, either due to impaired secretion or downstream obstruction, that results in the retention of bile components. Extrahepatic bile acids are usually bound to phospholipids but when unbound can be highly cytotoxic. Although liver biopsy is rarely needed for diagnosis, histopathologic findings in cholestasis include the abnormal presence of bile within hepatic structures (cytoplasm, canaliculi, or duct) or noninflammatory secondary changes attributed to the detergent effect of retained bile acids (hydropic pseudoxanthomatous swelling of hepatocytes, cytoplasmic yellowing or clearing, Mallory body formation, bile thrombi, dilated/hyperplastic bile ducts, biliary piece meal necrosis).14 In cholestatic disease, bile salts build up within hepatocytes that results in a compensatory accumulation of bile acids in the blood. Clinically, cholestasis results in abnormal levels of bile acids in the peripheral blood.
The etiology of cholestasis is poorly understood but is likely multifactorial with genetic, environmental, and hormonal contributions to disease development and severity. Environmental factors and genetic susceptibility unique to certain ethnic groups likely account for the extreme variation in incidence and disease severity.9,10,15,16 Overall incidence tends to be lower in Caucasian women with the exception of specific Scandinavian groups who are thought to incur genetic clustering of genes implicated in the development of cholestasis.2
Outside of pregnancy, exogeneous estrogens are known to cause intrahepatic cholestasis without hepatitis. The role of reproductive hormones in the development of cholestasis is elucidated by its natural history, by studies in which cholestasis occurred more frequently in patients receiving progesterone,17 and by animal models, which demonstrate that exogenous estrogen downregulates the expression of bile acid transport proteins and that progesterone metabolites inhibit FXR function.2,18,19 The buildup of progesterone metabolites has also been shown to overwhelm certain hepatic transport systems which could result in decreased hepatic export of bile salts. This may be the mechanism behind increased rates of cholestasis in pregnant patients receiving progesterone supplementation.17 For these reasons, the hyperestrogenic and hyperprogesteronic states of pregnancy are thought to play a role in the development of ICP. Furthermore, ICP occurs mainly in the third trimester, when serum concentrations of estrogens and progesterone are highest. In addition, ICP occurs with greater frequency in twin pregnancies, likely related to higher levels of sex hormones in multiple gestations.14 This is further supported by in vitro models which have shown that high levels of estrogen and progesterone result in decreased function of the BSEP and MDR3 hepatocellular transporters.14
That does not, however, explain why incidence varies so greatly between different populations. Familial clustering, ethnic variation, and recurrence rates as high as 90% are evidence that genetic susceptibility also plays a major role in disease development. Most data relating to genetic influences on cholestasis come from the investigation of progressive familial intrahepatic cholestasis (PFIC). PFIC is a progressive liver disorder that results from the buildup of bile in liver cells eventually leading to liver failure. While this disorder is not specific to pregnancy, the genetic underpinnings of the disease highlight genetic risk factors that may predispose to cholestasis of pregnancy. PFIC is categorized into 3 subtypes, each caused by a different gene mutation resulting in a defective bile acid transport protein. Those gene mutations occur in the ATP8B1 gene (subtype 1), the ABCB11 gene (subtype 2), and the ABCB4 gene (subtype 3). A Swedish prospective study of 693 women with severe ICP (bile acid levels ≥40 μmol/L) showed increased ABCB4 gene variants in affected patients.20 Genetic variations in the FXR, encoded by the NR1H4 gene, have also been linked to cholestasis in pregnancy.2 In addition, there are small data that suggest decreased levels of selenium and vitamin D play a potential role.2 And lastly, population-based studies have linked several chronic liver diseases to the development of intrahepatic cholestasis, namely hepatitis C infection.2
While the exact cause of cholestasis of pregnancy is complex, a genetic predisposition manifested by mutations in bile transporter proteins likely exists which, when faced with the cholestatic effects of elevated circulating sex hormones, results in clinical disease. This may be worsened further by preexisting liver disease and environmental factors yet to be identified.
SERUM BILE ACID EVALUATION
An elevated serum bile acid concentration is a requirement for diagnosis.21 Total bile acids are often reported as a summation of all bile acids including cholic, chenodeoxycholic, deoxycholic, ursodeoxycholic, lithocholic, and hyodeoxycholic acids. However, the primary human bile acids are cholic and chenodeoxycholic acid and the remaining bile acids are found in very small quantities in both normal and ICP patients.22 In healthy nonpregnant women, chenodeoxycholic acid levels in serum are higher than cholic acid levels. In women with ICP, cholic acid is greatly increased over chenodeoxycholic acid, often in a ratio of 3 to 1. Traditionally, serum bile acids were measured using liquid chromatography–tandem mass spectroscopy, which allows for reporting on individual bile acid species. This method, however, is not used by most hospitals and the need for send-out testing can be accompanied by potentially long delays in results and thus diagnosis. An alternative method involves enzymatic bile acid assays and conventional immunosorbent plate-based techniques. These assays can be performed quickly and economically in many hospitals and commercial laboratories. One limitation to this technique is the inability to report on specific bile acid species. Some clinicians have advocated for the use of a cholic to chenodeoxycholic acid ratio to increase specificity in the diagnosis of ICP. However, total serum bile acid levels have been shown to be similarly sensitive and specific in the diagnosis of cholestasis when compared with the bile acid ratio.23 It is important to know which bile acids are measured in a specific assay as certain medications that contain exogenous bile acids, for example exogenous UDCA and home remedies that contain animal bile, can increase total bile acid levels measured using an inclusive assay.
As described prior, the efficiency of first-pass hepatic clearance of bile acids from portal blood results in low levels of circulating bile acids in normal patients. As such, elevated fasting bile acids are a sensitive marker of liver pathology. In early enzyme assay studies of bile acid levels, Barnes and colleagues found that serum bile acids concentrations were <15 μmol/L in normal subjects in both the fasting and fed states. In their study, fasting levels were elevated in 90% of patients with histologically proven hepatobiliary disease and postprandial levels were elevated in 100%. Bile acid changes in response to a standardized fatty meal were indicative of liver disease and were a more sensitive marker than other liver function tests. Their study suggests the postprandial assessment of bile acids may be a more sensitive test, whereas fasting may be more specific.24
Bile acid levels have been shown to increase markedly in the fed state in patients with various liver diseases, including cirrhosis, cholangitis, hepatitis, cholestasis, Budd-Chiari syndrome, Wilson disease, and hemochromatosis. Gall-bladder contraction in response to a meal increases the amount of bile acid entering the small intestine and subsequently returning to the liver. In nonpregnant patients with liver disease, it has been proposed that hepatic dysfunction manifested as an inability to clear bile acids from portal blood may not become apparent until biliary excretion and subsequent reabsorption of bile acids is stimulated by oral intake. However, in pregnant patients, the outcome of interest is not liver dysfunction but rather the absolute elevation in circulating bile acids.
Although fasting levels in both the pregnant and nonpregnant state are considered to be the most specific marker for the diagnosis of ICP, many providers order random (nonfasting) bile acids when clinical suspicion for ICP arises.25 Reference values used in the literature and in clinical laboratories are most often fasting values measured in nonpregnant subjects and the data for reference values in pregnancy is extremely limited.25 In the 2014 Clinical Expert Series article on cholestasis published by the American College of Obstetricians and Gynecologists (ACOG), the reference range given for bile acids in both normal pregnancy and in nonpregnant patients was 0 to 14 μmol/L, though the source of that information in unclear.2 In a 1991 longitudinal Australian study of 56 healthy pregnant patients, the mean postprandial serum bile acids increased significantly from 5.3 μmol/L at 16 weeks gestation to 6.5 μmol/L at term with an upper limit of normal in late gestation of 11 μmol/L.26 Findings were similar in another early study of 102 healthy pregnant patients in Sweden27 and in 26 healthy pregnant patients in England.28 In the latter study, fasting concentrations were noted to increase throughout pregnancy, as did the ratio of cholic to chenodeoxycholic acid. However, postprandial concentrations did not differ significantly at different gestational ages.28 In a 1996 prospective French study of 103 healthy pregnant patients matched to nonpregnant healthy controls, the mean fasting bile acids in both the pregnant and nonpregnant cohorts was ~2 with an upper limit of normal that ranged from 6.5 to 13 μmol/L. In this study, fasting bile acid concentrations did not differ significantly from the control group or between trimesters.17 And lastly, in a 2006 prospective American study of 340 predominantly Latina pregnant patients in which any patient admitted in the third trimester was eligible for enrollment, incidence of ICP defined as bile acids >20 μmol/L with a pruritus rating above 4 (on a 10-point scale) was 5.6%. Bile acids were checked at random and the mean bile acid level for the entire cohort was found to be 10.4 μmol/L.9 These 3 studies are the most widely referenced in support of using bile acids >10 μmol/L as the diagnostic threshold for ICP. In contrast, bile acid levels >10 μmol/L have been reported in up to 40% of asymptomatic women supporting the theory that normal pregnancy is a cholestatic state.25 In further support of this concept, a 1986 Swedish prospective study of 102 normal pregnant patients observed a 25% increase in total bile acids from the first to the third trimester.27 More recently, a pilot study of 10 healthy American pregnant women in the third trimester reported mean fasting bile acids levels of 3.4 μmol/L, which increased significantly to a mean of 5.6 μmol/L three hours after a 100 g glucose load.25 In another small prospective cohort study evaluating 29 healthy pregnant patients, the mean bile acid levels were found to increase significantly in the second trimester from 4.7 μmol/L in the fasting state to 12.9 μmol/L following a standardized fatty meal. Results were similarly significant in the third trimester with an increase from 9.7 μmol/L in the fasting state to 19.5 μmol/L postprandially.29
Given the very limited data regarding normal bile acid levels in pregnancy, it is not surprising that there is a wide range of diagnostic thresholds for ICP used in the literature. The Royal College of Obstetricians and Gynaecologists (RCOG) in their clinical guideline reports that bile acid levels can rise significantly after a meal but that in the majority of studies and in clinical practice, bile acids are not often checked in the fasting state. Very few guidelines require that bile acids be assessed in the fasting state.21
LIVER FUNCTION ABNORMALITIES
The tendency of bile acids to aggregate and act as emulsifying detergents is the mechanism behind their toxicity. Accumulation of bile acids in hepatocytes causes hepatotoxicity and the release of aminotransferases (also known as transaminases), bilirubin, gamma-glutamyl transpeptidase (GGT), and alkaline phosphatase into the serum. Serum aminotransferases are elevated in ~60% to 85% of cases. The transaminitis associated with ICP is not often >2 times the upper limit of normal, which is an important differentiating factor from other liver diseases that can occur in pregnancy like viral hepatitis and preeclampsia.17 In a prospective study of 84 women with ICP (pruritus with either bile acids >10 μmol/L or elevated aminotransferases), elevated aminotransferases occurred in 85% of patients, elevated alkaline phosphatase in 60%, elevated bilirubin in 14%, and elevated GGT in 11%.30 Alkaline phosphatase is not considered to be an appropriate marker of cholestasis given placental production of an alkaline phosphatase isoenzyme.
Internationally, there is a lack of consensus in the diagnostic criteria for ICP. Most guidelines agree on the requirement of pruritis and abnormal liver function. However, which liver function tests (bile acids, transaminases, bilirubin, GGT) should be used and their respective diagnostic thresholds can differ greatly. The South Australia Maternal and Neonatal Community of Practice (SAMNCP) guidelines require bile acids >15 μmol/L for diagnosis, whereas the threshold for the American College of Gastroenterologists (ACG) is bile acids >10 μmol/L. In the ACG guidelines bile acid concentrations in ICP are reported to be “typically >10 μmol/L with increased cholic acid levels and decreased chenodeoxycholic acid levels”; however, a reference is not cited for this statement. Currently, the Society for Maternal-Fetal Medicine (SMFM) does not offer specific guidelines but has published a review article in which it is mentioned that, “although some clinicians use a cutoff point of >10 µmol/L to define abnormal levels, proposed cutoffs range from 6 to 20 µmol/L.”31 The RCOG guideline recommends that the upper limit of pregnancy-specific ranges should be used for diagnosis and caution against assuming that laboratories will report pregnancy-specific reference ranges rather than nonpregnant ones. They explain that there are a variety of diagnostic thresholds reported in the literature and cite 3 studies for example. The first study used a diagnostic threshold of 6 μmol/L for fasting bile acids, which was the upper limit of normal for nonpregnant patients in their laboratory.17 The second used a diagnostic threshold of 8 μmol/L without mention of fasting status or reference values.32 And the third used a diagnostic threshold of 14 μmol/L without mention of fasting status for reference values.1 Lastly, though ACOG has yet to publish a clinical practice guideline, in their 2014 Clinical Expert Series article on cholestasis it is mentioned, though without reference, that most studies use an upper limit of normal of between 10 and 14 µmol/L but this may be reduced to between 6 and 10 µmol/L in women who are fasting.2
Although diagnostic thresholds may differ, severe cholestasis has been consistently defined in the literature as bile acids >40 µmol/L. This is largely due to an investigation by Glantz et al33 in which adverse perinatal outcomes increased only after bile acids reached 40 µmol/L. Interestingly, ~20% of ICP cases will develop bile acids >40 µmol/L. In a recent meta-analysis of published studies evaluating perinatal outcomes for women with intrahepatic cholestasis, extremely high serum bile acid levels were shown to markedly increase the risk of stillbirth with a hazard ratio of 30 when comparing serum levels above 100 µmol/L to levels below 40 µmol/L. Only women with bile acids above 100 µmol/L had fetal demise rates that were significantly higher than the pooled national rate (Table 2).16
ICP is classically diagnosed when a pregnant woman in the second or third trimester reports new-onset pruritus and is subsequently found to have abnormal serum bile acids. Although the majority of women will present after 30 weeks’ gestation, ~40% will present before 28 weeks, and rarely as early as 8 weeks.1,3 The onset of pruritus often precedes the elevation in bile acids and on average begins 3 weeks before diagnosis.1 Pruritus often starts on the palms and soles before becoming generalized with approximately one third of patients reporting pruritis that is worst in those locations.30 It is also worsened at night, with psychological stress, and by advancing gestational age. Pruritus is improved with cold temperatures and usually within a few days after delivery.1,30 Physical examination may show excoriation marks but the lack of primary skin lesions differentiates cholestasis from the dermatoses of pregnancy.
The mechanism behind cholestatic pruritus is hypothesized to be due to the direct pruritogenic action of certain bile acids on the skin. Classic experimental studies showed that bile acids, when injected and applied as cream to human skin, created reproducible pruritus.35 However, this hypothesis is refuted somewhat by the fact that not all patients with elevated bile acids have pruritus and, in those that do, bile acid levels do not correlate with pruritus severity.36 Others believe the pruritogen implicated in cholestasis to be an unknown substrate of bile which is released into the bloodstream after hepatocyte damage secondary to bile acid accumulation.36 There is also increasing evidence implicating the mu receptor in a central nervous system–mediated pathway of pruritus. This phenomenon is commonly encountered as itching after intrathecal opioid analgesia, for example after spinal anesthesia for cesarean delivery. In further support of this hypothesis, a reduction in cholestatic pruritus after the administration of opioid antagonists has been reported.37
OTHER (MATERNAL COMPLICATIONS)
Other symptoms of ICP may include right upper quadrant pain, nausea, poor appetite, or sleep deprivation. Mild jaundice occurs in 10% to 15% of cases, typically within 4 weeks of the onset of itching.14 Steatorrhea secondary to fat malabsorption can occur, though this is often mild. In very rare cases, steatorrhea may be dramatic enough to result in vitamin K deficiency and a prolonged partial thromboplastin time. The theoretical risk of obstetric bleeding is used by some to support a practice of empiric vitamin K supplementation in patients with ICP.3 Interestingly, higher rates of both gestational diabetes and preeclampsia have also been reported in pregnancies complicated by ICP. In their population-based cohort study published in 2013, Wikström et al38 reported a 2.8-fold increased risk of gestational diabetes and a 2.6-fold increased risk of preeclampsia when compared with controls without ICP.
ICP is considered a transient disease without long-term morbidity. Hepatocellular damage resolves after pregnancy in nearly all cases, though recurrence rates are high (up to 90% in some cases).21 Moreover, when ICP recurs it tends to be more severe and occur at an earlier gestational age in subsequent pregnancies. There is also an association with biliary disease later in life, though this relationship is not thought to be causative. Increased rates of gallstone disease, hepatitis C, fibrosis, cholangitis, hepatobiliary cancer, immune-mediated disease, and cardiovascular disease have been reported in patients with ICP. A Swedish registry-based study of over 11,000 women with ICP found increased rates of liver or biliary tract cancer, diabetes, thyroid disease, and Crohn’s disease when compared with controls without ICP.39
ICP has been clearly shown to increase the risk of spontaneous preterm birth.38 A recent meta-analysis of published studies evaluating perinatal outcomes for women with intrahepatic cholestasis found that the pooled odds ratio for spontaneous preterm birth was 3.47 (Table 2).16 Given current guidelines regarding delivery timing for ICP, it is not surprising that the risk of iatrogenic preterm birth is similarly high with an odds ratio of 3.65 (Table 2).16
Animal models have demonstrated a potential causal pathway linking elevated bile acid levels and spontaneous preterm birth. In a study by Perez et al,40 a single injection of high dose bile acid to pregnant sheep was shown to induce preterm delivery in 20% and chronic infusion of bile acids caused uterine contractions and preterm delivery in 100%.41 In a similar rodent model, Campos et al42 showed that treatment with bile acids increased myometrial response to oxytocin. In samples of human myometrium taken from patients with ICP, oxytocin induced an exaggerated contractile response when compared with control samples from patients without ICP.43 These studies inspired more recent experiments performed on human myometrium samples that demonstrated that bile acids increased the expression and sensitivity of the oxytocin receptor. In these experiments, bile acid treatment resulted in higher levels of oxytocin receptor mRNA and associated proteins. In addition, less oxytocin was required to produce uterine contractions in myometrial strips that were treated with bile acids. The investigators concluded that a bile acid–mediated increase in oxytocin-receptor expression and sensitivity may explain the increased risk of spontaneous preterm birth in patients with cholestasis. Interestingly, only cholic acid elicited this response while deoxycholic acid did not.44
The most concerning fetal complication associated with cholestasis of pregnancy is fetal demise. The link between elevated bile acid levels and stillbirth is reported consistently in the literature; however, most studies to date are not powered to report on stillbirth. In 2004, Glantz and colleagues were the first to report on a cohort large enough to demonstrate an association between cholestasis and adverse outcomes. In their study of 505 Swedish women with mild or severe cholestasis, they found that spontaneous preterm delivery, fetal asphyxia events, and meconium staining increased by 1% to 2% per additional 1 µmol/L of serum bile acids above 40 µmol/L.33 In 2014, Geenes and colleagues were the first to report on a cohort large enough to demonstrate an association between cholestasis and stillbirth specifically. In their study of 713 British women with severe cholestasis (bile acids ≥40 µmol/L) they reported an odds ratio of 2.6 for stillbirth (10/664, 1.5% vs. 11/2205, 0.5%; unadjusted odds ratio, 3.05; adjusted odds ratio, 2.58; 95% confidence interval, 1.03- 6.49) when compared with controls.2 Authors Ovadia and colleagues, in their recent meta-analysis evaluating perinatal outcomes for women with intrahepatic cholestasis, found that the pooled odds ratio from aggregate data for fetal demise was 1.46. In this study, extremely high serum bile acid levels were shown to markedly increase the risk of stillbirth with a hazard ratio of 30 when comparing bile acid levels above 100 µmol/L to levels below 40 µmol/L. The pooled rate of stillbirth for women with bile acid levels above 100 µmol/L was 3.44%, 12 times what was considered to be the background risk of stillbirth in the population. However, contrary to common belief, only women with bile acids above 100 µmol/L had fetal demise rates significantly higher than the pooled national rate (Table 2).16
The mechanism of fetal demise in the setting of ICP is poorly understood. Adverse fetal outcomes are thought to be secondary to increased bile acid levels in the fetal compartment, evidenced by increased measurements of bile acids in amniotic fluid, cord serum, and meconium in cholestasis-affected pregnancies.3 Fetal death is thought to be directly related to bile acid effects on the fetal heart and not to chronic placental insufficiency.45 Clinical studies in support of the arrhythmogenic effect of bile acids on the fetal heart have reported fetal atrial flutter and supraventricular tachycardia in cholestatic pregnancies.45 Furthermore, in vitro animal and human stem-cell cardiomyocte models have shown that bile acid administration decreases contractile function and induces arrhythmia. In those same studies, administration of UDCA prevented this effect.2,15 In comparison to adult cardiomyocytes, neonatal rat cardiomyocytes are more sensitive to bile acid and show greater changes in calcium release and contraction amplitude when exposed.45 There is also small evidence that the pathologic effect of bile acids on immature cardiomyocytes is mediated by the partial agonist effect of bile acids on muscarinic receptors.45 In addition, there is evidence that bile acids have a pathologic effect on the placenta causing marked vasoconstriction of chorionic vessels. This has been hypothesized as a contributor to the risk of sudden fetal death in ICP.2
Increased rates of respiratory distress syndrome (RDS) have been demonstrated in pregnancies complicated by ICP.2,46 Although increased rates of RDS are likely compounded by both spontaneous and iatrogenic preterm delivery, animal models have demonstrated causative relationships. In a rabbit model, direct tracheal injection of bile acids resulted in atelectasis, eosinophilic infiltration, and formation of hyaline membrane, which was reversed by the administration of surfactant. In a porcine model, bile acids were shown to cause severe chemical pneumonitis and pulmonary edema.3 In a 2006 study of human neonates born from cholestatic pregnancies, Zecca et al46 demonstrated an ~2.5-fold higher risk of RDS when compared with controls. They hypothesized that elevated bile acid levels affect alveolar enzyme function that results in decreased surfactant levels and subsequent RDS.
Meconium-stained amniotic fluid is another complication associated with ICP. A recent meta-analysis evaluating perinatal outcomes for women with intrahepatic cholestasis found that the incidence of meconium staining increased from a baseline of ~11% to ~19% in pregnancies complicated by ICP with a pooled odds ratio of 2.60 (Table 2).16 In a 1996 study of women in Southern California, meconium passage was noted in 44% of patients with cholestasis.8
Meconium staining is traditionally thought to be secondary to vagal stimulation of the mature fetal gastrointestinal tract. Often in the setting of hypoxic stress, vagal nerve stimulation results in gut peristalsis and anal sphincter relaxation leading to meconium passage. Discussion of the effects of meconium on perinatal outcomes is beyond the scope of this review. However, the mechanism behind meconium staining in ICP is less understood and is likely not related to fetal hypoxic stress. In animal studies, bile acids have been shown to directly stimulate gut motility. In a rabbit model, colonic smooth muscle contractility measurement demonstrated a dose-dependent increase with the addition of bile acids.47 In a lamb model, meconium-stained amniotic fluid was observed in 100% of lambs that received cholic acid infusion without other signs of fetal distress.2,41 Although data are limited, meconium passage in cholestasis may be better explained by an increase in colonic motility secondary to bile acids rather than neural stimulation secondary to fetal hypoxia.
There have been numerous trials evaluating medications used in the treatment of cholestasis. Most studies have primarily evaluated drug efficacy in decreasing pruritis and improving biochemical abnormalities, as perinatal outcomes are often too rare to study without large sample sizes. There are, however, a select few trials that have reported on adverse perinatal outcomes.2 Notable medications and the trials evaluating their efficacy will be reviewed here.
ICP is most often treated with UDCA, commonly known as Ursodiol. UDCA is a naturally occurring tertiary bile acid which is formed in the gastrointestinal tract as a result of bacterial metabolism of primary bile acids. UDCA acts in the intestine to disrupt micelles and decrease the rate at which cholesterol is absorbed. When administered exogenously, it is thought that UDCA acid concentrates in hepatocytes and bile and results in decreased hepatic cholesterol synthesis, hepatic cholesterol secretion, and intestinal cholesterol reabsorption. UDCA administration also increases the overall volume of bile secreted and is considered both an anticholelithic (cholesterol decreasing) and choleretic (bile flow increasing) medication. The cholerectic effects of UDCA are potentially mediated by its activity at the BSEP and MRP3 receptors.48 It is for this reason that UCDA is used in the nonsurgical treatment of gallstones. Reduced cholesterol concentration in bile results in the gradual solubilization of cholesterol from gallstones and their eventual breakdown.
As discussed previously, excreted bile acids undergo very efficient enterohepatic recycling. When administered chronically, UDCA is actively recycled and concentrates in the liver and bile. Because of the efficiency of enterohepatic recycling, clearance of exogenous UCDA is low. Eventually, UDCA becomes the primary circulating and biliary bile acid, changing the overall composition of bile. Although endogenous UDCA constitutes <3% of total bile acid stores in the body, chronic exogenous UDCA can ultimately constitute up to 50% of the total bile acid pool.3 In essence, UDCA is a hydrophilic nontoxic bile acid which replaces cholic acid, the hydrophobic toxic bile acid thought to be implicated in adverse outcomes. UDCA both decreases the overall levels of primary bile acids and normalizes the ratio of cholic acid to deoxycholic acid. UDCA has also been shown to reduce bile acid levels in other compartments, namely amniotic fluid and cord blood.3 It does not decrease bile acid level in meconium, possibly due to the temporal spacing between bile acid concentration in meconium and treatment initiation.3
UDCA administration results in improvement in pruritus in ~60% of women and complete cessation of pruritus in ~40%.49 Symptom improvement is usually observed within 1 to 2 weeks after initiation and a decrease in serum bile acids 2 weeks thereafter. UCDA can be titrated to a maximum dose of 21 mg/kg/d (~1500 mg for a 72 kg patient), usually split into multiple doses.
Although still not approved by the Food and Drug Administration for use in pregnancy, UDCA was first reported as a treatment for cholestasis in 1992.50 Since then, multiple studies have attempted to evaluate the efficacy of UDCA in treating pruritis, normalizing liver function tests, and decreasing risk of adverse perinatal outcomes. These studies have included multiple meta-analyses, including those of Bacq et al,49 Gurung et al51 for the Cochrane Database, Grand’Maison et al,52 and Kong et al53 Each of these analyses use different inclusion criteria and methodologies but report overall similar findings that UCDA significantly improves itching and serum bile acid levels and may potentially reduce adverse perinatal outcomes, though data surrounding the latter is limited. Furthermore, while some of the aforementioned meta-analyses strongly suggest that UDCA improves adverse perinatal outcomes, before 2019 there was no single study large enough to properly evaluate this effect.49 In 2019, Chappell and colleagues published the results of the PITCHES trial, a multicenter, randomized, placebo-controlled trial of UCDA in 605 women from the United Kingdom with ICP. A diagnostic bile acid threshold of 10 to 14 μmol/L was used and ~75% of the cohort had bile acid levels <40 μmol/L. Treatment recipients received UCDA 500 mg twice daily and 3 fetal deaths occurred overall (1 in the UCDA group and 2 in the placebo group). They found that treatment with UCDA did not significantly reduce the risk of their primary outcome, which was a composite of perinatal death, preterm delivery, or neonatal unit admission.54
Dexamethasone has also been evaluated for treatment of ICP. In an early observational study of a small cohort of Finnish women, dexamethasone was shown to improve symptoms and total bile levels.2 The proposed mechanism was inhibition of placental estrogen synthesis evidenced by decreased circulating levels of estriol and estradiol. However, these findings were not supported by subsequent studies and there exists concern regarding repeat fetal exposure to antenatal corticosteroids.3
Cholestyramine is a strong ion exchange resin, which binds to bile acids in the intestine, sequestering them from enterohepatic recycling and committing them to fecal excretion. Though evidence that cholestyramine affects total bile acid levels is lacking, there have been studies to suggest that it can reduce symptoms of pruritis.3 Cholestyramine is not considered a first-line treatment of ICP and there exist concerns surrounding the associated decrease in vitamin K levels.
Rifampin is a semisynthetic derivative of one of the rifamycins, a group of macrocyclic antibiotics produced by Streptomyces mediterranei. Rifampin inhibits bacterial RNA production and is most commonly used in the treatment of tuberculosis. Though not yet studied for use in ICP, there is evidence to suggest that rifampicin is an effective second-line treatment for primary biliary cirrhosis.3 In these studies, rifampin therapy was shown to significantly decrease pruritus and liver function enzymes, including total bile acids. A proposed mechanism of action is that rifampin inhibits bile acid reuptake by hepatocytes and enhances bile acid detoxification, potentially augmenting the efficacy of UDCA.3,34 Descriptions of its use in the obstetric population are limited to case reports though findings suggest utility.55
Vitamin K is sometimes prescribed prophylactically to protect against the theoretical risk of bleeding related to vitamin K deficiency secondary to steatorrhea. This practice is not evidence based.3 Menthol, lotions, and antihistamines (oral and topical) have all been shown to be safe and effective in relieving pruritus. These medications do not affect liver function or bile acid levels.
Investigations of guar gum, activated charcoal, and S-Adenosyl-L-methionine have demonstrated limited success in the treatment of ICP3 and a 2013 Cochrane Review concluded that there was insufficient evidence to determine their efficacy in the treatment ICP.21
Recommendations for bile acid surveillance differ. Many societies, including the RCOG, advocate for drawing weekly liver function tests as disease progression can alter management strategies.21 In their 2014 Clinical Expert Series article on cholestasis the ACOG recommended regular monitoring of bile acids throughout pregnancy.2 The SMFM does not offer guidance regarding surveillance of bile acids.31 In a 2019 Lancet meta-analysis evaluating perinatal outcomes, authors Ovadia et al16 advocate for monitoring bile acids weekly as levels can change rapidly with advancing gestational age. Lastly, some guidelines advocate evaluating for other potential causes of liver dysfunction, including viral, autoimmune, and obstructive etiologies.21
The role of antenatal testing in ICP remains controversial. While increased rates of fetal demise are well-documented in the literature, the mechanism of stillbirth is not thought to be secondary to placental insufficiency. And while fetal heart tracing abnormalities have been reported in the setting of ICP, including decreased variability, tachycardia, and bradycardia, these abnormalities were not related to fetal demise.3 It is, therefore, unclear whether antenatal testing can identify fetuses at risk of demise. In an early study by Fisk and Storey7 of 83 patients with ICP, nonstress testing did not show abnormalities in the 2 patients with ICP complicated by fetal demise. In another early study by Alsulyman and colleagues comparing cases of ICP followed with antenatal testing to controls who had a history of fetal demise who were also followed with antenatal testing, the 2 fetal demise cases both occurred within 5 days of normal antenatal testing.8 In addition, there are numerous case reports in the literature of fetal demise occurring in the setting of recently normal antenatal testing.
Although general consensus agrees that antenatal testing likely does not prevent fetal demise, it is still often used by clinicians. In an UK-based survey of providers, 84% of obstetricians and 93% of midwives reported antenatal testing as part of their management of ICP.56 Antenatal testing, though not associated with improved outcomes and potentially associated with unnecessary intervention, can offer reassurance to both patients and providers. Furthermore, it is used in many of the major investigations in which outcomes are reported which leads to a perpetuation of common practice when evidence is lacking.
To date, the SMFM is the only governing society that recommends antenatal testing though they comment that the type, duration, and frequency of testing has not yet been identified.31 In contrast, ACOG guidelines do not list ICP as an indication for antenatal testing. Other societies, including the RCOG, do not make clear recommendations but comment that antenatal testing is a consideration that lacks efficacy data.21
Elective early delivery of ICP has also become widely implemented in the management of ICP due to increasing rates of fetal demise with advancing gestational age, particularly after the 36th week. In the most robust evaluation of fetal demise risk to date, Puljic and colleagues aimed to characterize the risk of infant and fetal death by each additional week of expectant management versus immediate delivery in pregnancies complicated by cholestasis. In their retrospective cohort study published in 2015 of over 1.6 million pregnancies, they assessed the risk of fetal demise, the risk of infant death, and the composite risk of expectant management for 1 additional week for each week of gestation from 34 to 40 weeks. They found that the balance between the risk of perinatal mortality associated with delivery and the risk of fetal demise associated with expectant management began to shift at 36 weeks gestation (perinatal mortality risk associated with delivery 4.7 vs. expectant management 19.2 at 36 wk) with the risk of expectant management increasing thereafter (perinatal mortality risk associated with delivery 18.3 vs. expectant management 33.6 at 39 wk).57 These data support previously reported literature which demonstrated increasing rates of fetal demise after 36 weeks.3,58 However, a major limitation of the study is the lack of information on bile acid levels and treatment as well as the small absolute number of deaths. As mentioned previously, a recent meta-analysis evaluating perinatal outcomes found that the pooled odds ratio for fetal demise was 1.46 (Table 2). In this study, extremely high serum bile acid levels were shown to markedly increase the risk of stillbirth with a hazard ratio of 30 when comparing serum levels above 100 µmol/L to levels below 40 µmol/L. However, only women with bile acids above 100 µmol/L had fetal demise rates that were significantly higher than the pooled national rate. The authors acknowledge that there was a high proportion of iatrogenic preterm birth for women irrespective of peak bile acid concentration, which may account for the lower rates of fetal demise observed in women with total bile acid levels <100 µmol/L.16
Although the SMFM has not published official guidelines, their 2011 review publication states that, while an evidence-based recommendation is not available for the timing of delivery, most management strategies would advocate for delivery between 37 and 38 weeks and that prior obstetrical history, antenatal testing, and gestational age should be considered. They do not recommend using bile acid levels to inform delivery timing.31 The ACOG, in their committee opinion detailing medically indicated late-preterm and early-term deliveries, recommends delivery at 36 to 37 weeks gestation. They also state that delivery before 36 weeks may be indicated depending on laboratory and clinical circumstances.59 In contrast, the RCOG remains neutral, stating that the widely adopted practice of offering delivery at 37 weeks gestation is not evidence based.21 Of note, there is data indicating that early induction of labor for ICP does not result in higher rates of cesarean delivery or operative vaginal delivery.2 In summary, there are some data to suggest that delivery at 37 weeks, especially in patients with severely elevated bile acids, may improve outcomes. However, there are no randomized trials evaluating elective early delivery and thus optimal delivery timing remains unknown for pregnancies complicated by ICP.
Intrahepatic cholestasis is a condition defined by pruritis and elevated bile acids. The etiology of cholestasis is poorly understood though likely related to genetic, environmental, and hormonal contributions to disease development and severity. It is most notably associated with spontaneous preterm labor and intrauterine fetal demise. ICP is a difficult condition to manage, largely related to the paucity of data surrounding its diagnosis, management, and adverse outcomes. Recent data have confirmed an increased risk of fetal demise, possibly due to bile acid toxicity in fetal cardiac myocytes, but potentially only when bile acid levels are severely elevated. UDCA may improve pruritus, bile acid levels, and liver function though has not yet been shown to statistically improve fetal outcomes possibly due to the rarity of those outcomes. Although antenatal testing is recommended by certain societies and commonly used by providers, a proven benefit has not yet been shown in the literature. And while some literature supports a practice of elective early delivery, optimal delivery timing has yet to be established. ICP remains a common, dangerous, and poorly understood complication of pregnancy for which continued investigation is certainly warranted.
1. Kenyon AP, Piercy CN, Girling J, et al. Obstetric cholestasis
, outcome with active management: a series of 70 cases. BJOG An Int J Obstet Gynaecol. 2002;109:282–288.
2. Williamson C, Geenes V. Intrahepatic cholestasis
. Obstet Gynecol. 2014;124:120–133.
3. Geenes V, Williamson C. Intrahepatic cholestasis
. World J Gastroenterol. 2009;15:2049–2066.
4. Friedlaender P, Osler M. Icterus and pregnancy
. Am J Obstet Gynecol. 1967;97:894–900.
5. Sjövall K, Sjövall J. Serum bile acid
levels in pregnancy
with pruritus (bile acids and steroids 158). Clin Chim Acta. 1966;13:207–211.
6. Laatikainen T, Ikonen E. Serum bile acids in cholestasis
. Obstet Gynecol. 1977;50:313–318.
7. Fisk NM, Storey GN. Fetal outcome in obstetric cholestasis
. Br J Obstet Gynaecol. 1988;95:1137–1143.
8. Alsulyman OM, Ouzounian JG, Ames-Castro M, et al. Intrahepatic cholestasis
: perinatal outcome associated with expectant management. Am J Obstet Gynecol. 1996;175 (pt 1):957–960.
9. Lee RH, Goodwin TM, Greenspoon J, et al. The prevalence of intrahepatic cholestasis
in a primarily Latina Los Angeles population. J Perinatol. 2006;26:527–532.
10. Reyes H, Gonzalez MC, Ribalta J, et al. Prevalence of intrahepatic cholestasis
in Chile. Ann Intern Med. 1978;88:487–493.
11. Monte MJ, Marin JJG, Antelo A, et al. Bile acids: Chemistry, physiology, and pathophysiology. World J Gastroenterol. 2009;15:804–816.
12. Boyer JL. Bile formation and secretion. Compr Physiol. 2013;3:1035–1078.
13. Said HM. Physiology of the Gastrointestinal Tract, Sixth Edition. 2018:1–2.
14. Kondrackiene J, Kupcinskas L. Intrahepatic cholestasis
-current achievements and unsolved problems. World J Gastroenterol. 2008;14:5781–5788.
15. Geenes V, Chappell LC, Seed PT, et al. Association of severe intrahepatic cholestasis
with adverse pregnancy
outcomes: a prospective population-based case-control study. Hepatology. 2014;59:1482–1491.
16. Ovadia C, Seed PT, Sklavounos A, et al. Association of adverse perinatal outcomes of intrahepatic cholestasis
with biochemical markers: results of aggregate and individual patient data meta-analyses. Lancet. 2019;393:899–909.
17. Bacq Y, Sapey T, Bréchot MC, et al. Intrahepatic cholestasis
: a French prospective study. Hepatology. 1997;26:358–364.
18. Abu-Hayyeh S, Papacleovoulou G, Lövgren-Sandblom A, et al. Intrahepatic cholestasis
levels of sulfated progesterone metabolites inhibit farnesoid X receptor resulting in a cholestatic phenotype. Hepatology. 2013;57:716–726.
19. Chen Y, Vasilenko A, Song X, et al. Estrogen and estrogen receptor-α-mediated transrepression of bile salt export pump. Mol Endocrinol. 2015;29:613–626.
20. Wasmuth HE, Glantz A, Keppeler H, et al. Intrahepatic cholestasis
: the severe form is associated with common variants of the hepatobiliary phospholipid transporter ABCB4 gene. Gut. 2007;56:265–270.
21. Bicocca MJ, Sperling JD, Chauhan SP. Intrahepatic cholestasis
: review of six national and regional guidelines. Eur J Obstet Gynecol. 2018;231:180–187.
22. Ye L, Liu S, Wang M, et al. High-performance liquid chromatography-tandem mass spectrometry for the analysis of bile acid
profiles in serum of women with intrahepatic cholestasis
. J Chromatogr B Anal Technol Biomed Life Sci. 2007;860:10–17.
23. Huang WM, Gowda M, Donnelly JG. Bile acid
ratio in diagnosis of intrahepatic cholestasis
. Am J Perinatol. 2009;26:291–294.
24. Barnes S, Gallo GA, Trash DB, et al. Diagnostic value of serum bile acid
estimations in liver disease. J Clin Pathol. 1975;28:506–509.
25. Adams A, Jacobs K, Vogel R, et al. Bile acid
determination after standardized glucose load in pregnant women. Am J Perinatol Reports. 2015;05:e168–e171.
26. Carter J. Serum bile acids in normal pregnancy
. BJOG An Int J Obstet Gynaecol. 1991;98:540–543.
27. Järnfelt-Samsioe A, Bremme K, Eneroth P. Steroid hormones in emetic and non-emetic pregnancy
. Eur J Obstet Gynecol Reprod Biol. 1986;21:87–99.
28. Fulton IC, Douglas JG, Hutchon DJR, et al. Is normal pregnancy
cholestatic? Clin Chim Acta. 1983;130:171–176.
29. Smith DD, Kiefer MK, Lee AJ, et al. Effect of fasting versus postprandial state on total bile acid levels in pregnancy.
Accepted for poster presentation at the Society for Maternal-Fetal Medicine 39th Annual Pregnancy
30. Kondrackiene J, Beuers U, Kupcinskas L. Efficacy and safety of ursodeoxycholic acid versus cholestyramine in intrahepatic cholestasis
. Gastroenterology. 2005. Available at: https://doi.org/10.1053/j.gastro.2005.06.019
31. Intrahepatic cholestasis
explained. 2011. Available at: www.smfm.org/publications/96-understanding-intrahepatic-chole-stasis-of-pregnancy
. Accessed July 30, 2019.
32. Heinonen S, Kirkinen P. Pregnancy
outcome with intrahepatic cholestasis
. Obstet Gynecol. 1999. Available at: https://doi.org/10.1016/S0029-7844(99)00254-9
33. Glantz A, Marschall HU, Mattsson LÅ. Intrahepatic cholestasis
: relationships between bile acid
levels and fetal complication rates. Hepatology. 2004;40:467–474.
34. Prince MI, Burt AD, Jones DEJ. Hepatitis and liver dysfunction with rifampicin therapy for pruritus in primary biliary cirrhosis. Gut. 2002;50:436–439.
35. Varadi DP. Pruritus induced by crude bile and purified bile acids: experimental production of pruritus in human skin. Arch Dermatol. 1974;109:678–681.
36. Ghent CN, Bloomer JR, Klatskin G. Elevations in skin tissue levels of bile acids in human cholestasis
: relation to serum levels and to pruritus. Gastroenterology. 1977. Available at: https://doi.org/10.1016/s0016-5085(19)31870-0
37. Bergasa NV. The pruritus of cholestasis
. Semin Cutan Med Surg. Available at: https://doi.org/10.1002/hep.27582
38. Wikström Shemer E, Marschall HU, Ludvigsson JF, et al. Intrahepatic cholestasis
and associated adverse pregnancy
and fetal outcomes: a 12-year population-based cohort study. BJOG An Int J Obstet Gynaecol. 2013;120:717–723.
39. Marschall HU, Wikström Shemer E, Ludvigsson JF, et al. Intrahepatic cholestasis
and associated hepatobiliary disease: a population-based cohort study. Hepatology. 2013. Available at: https://doi.org/10.1002/hep.26444
40. Perez R, Garcia M, Ulloa N, et al. A single intravenous high dose of cholic acid to a pregnant ewe does not affect fetal well-being. Res Exp Med. 1994. Available at: https://doi.org/10.1007/BF02576367
41. Campos GA, Guerra FA, Israel EJ. Effects of cholic acid infusion in fetal lambs. Acta Obstet Gynecol Scand. 1986;65:23–26.
42. Campos GA, Castillo RJ, Toro FG. Effect of bile acids on the myometral contractility of the isolated pregnant uterus. Rev Chil Obstet Ginecol. 1988;53:229–233.
43. Israel EJ, Guzman ML, Campos GA. Maximal response to oxytocin of the isolated myometrium from pregnant patients with intrahepatic cholestasis
. Acta Obstet Gynecol Scand. 1986;65:581–582.
44. Germain AM, Kato S, Carvajal JA, et al. Bile acids increase response and expression of human myometrial oxytocin receptor. Am J Obstet Gynecol. 2003;189:577–582.
45. Williamson C, Miragoli M, Sheikh Abdul Kadir S, et al. Bile acid
signaling in fetal tissues: Implications for intrahepatic cholestasis
. Dig Dis. 2011;29:58–61.
46. Zecca E, De Luca D, Marras M, et al. Intrahepatic cholestasis
and neonatal respiratory distress syndrome. Pediatrics. 2006;117:1669–1672.
47. Snape WJ, Shiff S, Cohen S. Effect of deoxycholic acid on colonic motility in the rabbit. Am J Physiol Liver Physiol. 1980;1:321–325.
48. Lazaridis KN, Gores GJ, Lindor KD. Ursodeoxycholic acid “mechanisms of action and clinical use in hepatobiliary disorders”. J Hepatol. 2001. Available at: https://doi.org/10.1016/S0168-8278(01)00092-7
49. Bacq Y, Sentilhes L, Reyes HB, et al. Efficacy of ursodeoxycholic acid in treating intrahepatic cholestasis
: a meta-analysis. Gastroenterology. 2012;143:1492–1501.
50. Palma J, Reyes H, Ribalta J, et al. Effects of ursodeoxycholic acid in patients with intrahepatic cholestasis
. Hepatology. 1992;15:1043–1047.
51. Gurung V, Stokes M, Middleton P, et al. Interventions for treating cholestasis
. Cochrane Database Syst Rev. 2013. Available at: https://doi.org/10.1002/14651858.CD000493.pub2
52. Grand’Maison S, Durand M, Mahone M. The effects of ursodeoxycholic acid treatment for intrahepatic cholestasis
on maternal and fetal outcomes: a meta-analysis including non-randomized studies. J Obstet Gynaecol Canada. 2014;36:632–641.
53. Kong X, Kong Y, Zhang F, et al. Evaluating the effectiveness and safety of ursodeoxycholic acid in treatment of intrahepatic cholestasis
: a meta-analysis (a prisma-compliant study). Med (United States). 2016. Available at: https://doi.org/10.1097/MD.0000000000004949
54. Chappell LC, Bell JL, Smith A, et al. Ursodeoxycholic acid versus placebo in women with intrahepatic cholestasis
(PITCHES): a randomised controlled trial. Lancet. 2019;394:849–860.
55. Liu J, Murray AM, Mankus EB, et al. Adjuvant use of rifampin for refractory intrahepatic cholestasis
. Obstet Gynecol. 2018;132:678–681.
56. Saleh MM, Abdo KR. Consensus on the management of obstetric cholestasis
: National UK survey. BJOG An Int J Obstet Gynaecol. 2007;114:99–103.
57. Puljic A, Kim E, Page J, et al. The risk of infant and fetal death by each additional week of expectant management in intrahepatic cholestasis
by gestational age. Am J Obstet Gynecol. 2015;212:667.e1–667.e5.
58. Lo JO, Shaffer BL, Allen AJ, et al. Intrahepatic cholestasis
and timing of delivery. J Matern Neonatal Med. 2015;28:2254–2258.
59. ACOG Committee Opinion. Medically indicated late-preterm
and early-term deliveries. Obstet Gynecol. 2019. Available at: https://doi.org/10.1097/AOG.0000000000003083