What does the recent literature from 2020 tell us about delivery timing?

Current Guidelines

The current guidelines for delivery of a pregnancy complicated by Intrahepatic Cholestasis of Pregnancy (ICP) are as follows:

• SMFM: Delivery is stratified by bile acid levels²
  Bile acids over 100: Delivery at 36 0/7 weeks. Delivery prior to 36 weeks (34-36 weeks) may occur with unremitting maternal pruritus, a prior history of a stillbirth before 36 weeks gestation due to ICP or preexisting or acute hepatic disease with worsening liver function.
• Bile acids 40-99: Delivery at 36 0/7-39 0/7 weeks with delivery in the earlier portion of the window recommended.
• Bile acids less than 40: Delivery at 36 0/7-39 0/7 weeks.

ACOG: Delivery as per SMFM guidelines per practice bulletin on medically indicated early deliveries.¹

What do we know about stillbirth in a pregnancy complicated by Intrahepatic Cholestasis of Pregnancy?

• Stillbirth is thought to be an acute event although the exact cause is unknown. Recently it is thought to be most likely from a fetal cardiac arrhythmia or vasospasm of the placental vessels due to elevated bile acid levels.²
• Stillbirth clusters toward the end of pregnancies and is more common after 37 weeks of gestation. It is also more common with higher bile acid levels.⁶
• Stillbirth cannot be fully prevented with antepartum monitoring as there are case reports of fetal deaths within hours of normal antepartum monitoring. There is still thought to be benefit to antepartum monitoring.^2,7

Where did 36 0/7-37 0/7 week recommendation previously endorsed by ACOG come from?

Until a recent change in January 2021, ACOG had previously recommended delivery between 36-37 weeks. The recommendation for delivery in this timeframe was based mostly upon two articles published in 2015. The first was a retrospective cohort study by Puljic et al.³ This study calculated a composite mortality risk for each week of gestation and assessed the risk of stillbirth compared to the risk of infant death after delivery. They found that among women with Intrahepatic Cholestasis of Pregnancy (ICP), the mortality risk of delivery was lower than that of expectant management at 36 weeks of gestation. There was a stillbirth rate of 0.64% assumed in this study for the women with ICP. The authors note that this recommendation for delivery at 36 weeks is based on only mortality considerations and did not take into account the risks of fetal morbidity including NICU admission and intubation. Also, the authors note that this study did not take into account the severity or the timing of the cholestasis and assumed a 0.64% stillbirth risk for all cases of cholestasis.

A second article by Lo et al used computer modeling to determine that delivery at 36 weeks in a pregnancy complicated by Intrahepatic Cholestasis of Pregnancy was the optimal delivery strategy.⁴ This study, unlike the study by Puljic et al, did take into account neonatal morbidity. Risks of IUFD, respiratory distress, necrotizing enterocolitis and cerebral palsy were all taken into account in the model. This study assumed a mean risk of stillbirth at 1.74% in their model with their modeling being applicable for a stillbirth risk between 0.7-2.26%.

What data do we have that stratifies risk by bile acids levels?

One concern with the above models is that these studies did not look at severity of individual cases in making the 36 week delivery recommendation but instead based their models on an assumed stillbirth risk.^3,4 In 2019, a study entitled “Association of adverse perinatal outcomes of intrahepatic cholestasis of pregnancy with biochemical markers” was published by Ovadia et al in the Lancet.⁵ This study was a systematic review of cholestasis pregnancies and included 5269 pregnancies complicated by ICP. This large meta-analysis allowed researchers to be able to determine stillbirth risks based on bile acid levels.⁵

The findings showed that for bile acids less than 40, the risk of stillbirth was 0.13%, for bile acids 40-99, the risk was 0.28%, and for pregnancies with bile acids over 100, this risk was 3.44%.

Overall, this study had reassuring results regarding the risks of stillbirth in ICP pregnancies. This evidence showed that the majority of fetal mortality occurs in pregnancies with bile acids over 100 μmol/L and that these pregnancies need to be more aggressively managed.

As with any research, there are some limitations of this study. There was a large prevalence of 25.3% preterm birth in the study and this might have contributed to the prevention of later stillbirth. The majority of pregnancies were also completed before 39 weeks gestation, therefore conclusions cannot be made past this gestation. Many of the pregnancies were delivered prior to 38 weeks of gestation with only about 30% of the pregnancies continuing to this gestational age.

The main question is whether these reassuring stillbirth rates can be applied to pregnancies that are not actively managed. This study in its discussion states that “although our data cannot confirm that the risk of stillbirth is not increased for women with bile acids of less than 100 μmol/L compared with the background population if not actively managed, these women were unlikely to be managed more proactively than those with bile acids of 100 μmol/L or more.”⁵

Therefore, even though reassuring stillbirth rates were reported in this meta-analysis, there needs to be consideration that some of the stillbirth risk might have been mitigated by early delivery. The study was not meant to prove whether active management improves outcomes. This would need to be studied with a randomized controlled trial which is unlikely to be performed.

What bile acid levels should be used to guide delivery timing?

SMFM states that bile acids do not need to be performed in a fasting state and can be drawn at any point throughout the day.² A recent study showed that drawing bile acids in a fasting state would have missed diagnosis in all women with mild disease. Also the majority of women with severe disease would not have been classified as severe had their levels been drawn fasting.⁶ If levels were only measured in a fasting state in these pregnancies, the true risk to the fetus would have been missed. Therefore, it could be argued that non-fasting levels are a more true representation of the risks throughout the day to the fetus.⁶

The highest ever bile acid throughout pregnancy should be used. As the full mechanism for stillbirth is unknown, there is no clear indication that if bile acids are lowered with treatment that fetal risk is lessened.

Where are we in 2020 in putting the evidence together?

A comprehensive management guide for cholestasis of pregnancy entitled “SMFM Consult Series #53: Intrahepatic Cholestasis of Pregnancy” was published in AJOG on November 13, 2020.² This guide outlined the changes to the delivery recommendations as discussed above and extended the recommended delivery window to include 39 0/7 weeks of gestation for mild cases of Intrahepatic Cholestasis of Pregnancy. This extension of the delivery window was largely based on the study by Ovadia et al showing that the risk of stillbirth is lower in these patients with low bile acids than previously thought.⁵

In both the articles by Lo and Puljic that established the prior 36 week recommendations, stillbirth risk was assumed to be higher than the true risks found by Ovadia for patients with bile acids remaining below 100 μmol/L. In their calculations these studies did not separate cases based on bile acid levels.^3,4 The question then arises whether these calculations can accurately be applied to pregnancies with lower stillbirth rates (with bile acids below 100 μmol/L).

The meta-analysis by Ovadia et al has shown us that some pregnancies need to be more aggressively managed and that some pregnancies may not need to be as aggressively managed.⁵ It must be noted that the data from Ovadia et al, while reassuring, cannot confirm that without active management that the stillbirth risks in these women would still be at these low and reassuring levels as this was not the main outcome of the study. While there is not a statistically significant increase in stillbirth risk in pregnancies with bile acids below 100 μmol/L based on the meta-analysis data, this data needs to be interpreted on a case by case basis since there was such a high preterm delivery rate. In deciding the optimal delivery timeline in a pregnancy complicated by Intrahepatic Cholestasis of Pregnancy, there is always a balance between prematurity and stillbirth and therefore this should be a shared decision between patients and providers.

In conclusion, the newest evidence points to the fact that the levels of bile acids are probably the most important factor in determining a delivery timing window. Bile acids do not need to be tested in a fasting state and the highest ever bile acid in pregnancy should be used to guide management. The SMFM recommendations published in 2020 expand this window to allow for a later delivery for mild cases but need to be interpreted with the above cautions. Bile acid levels over 100 μmol/L also need especially aggressive management. SMFM also acknowledges that each case is individual and treatment should be tailored to each patient with careful counseling and a shared decision making model. Treatment has shifted in recent years from a one-size fits all treatment to a more tailored treatment based on risk stratified by bile acid levels. While there is now evidence to point that in mild cases it is safe to prolong the pregnancy to 39 0/7 weeks gestation, earlier delivery is still an option in a well counseled patient who is aware of all of the risks.