| |
Liver Disease in Pregnancy
|
| |
| |
The American Journal of Gastroenterology
Volume 99 Issue 12 Page 2479 - December 2004
Fabiana S. Benjaminov M.D. 1 , and Jenny Heathcote M.D., F.R.C.P., F.R.C.P. (C) 1
1Department of Gastroenterology, "Meir" Medical Center, University of Tel-Aviv, Kfar-Saba, Israel; and Department of Medicine, The University Health Network, University of Toronto, Toronto, Ontario, Canada
Liver dysfunction during pregnancy can be caused by conditions that are specific to pregnancy or by liver diseases that are not related to pregnancy itself. This review attempts to summarize the epidemiology, pathophysiology, and management of the different pregnancy-related liver diseases, and to review different liver diseases not related to pregnancy and how they may affect or be effected by pregnancy. Some of the liver diseases specific to pregnancy can cause significant morbidity and mortality both to the mother and to the fetus, while most of the liver diseases not specific to pregnancy do not have a deleterious effect on the pregnancy itself.
EXCERPTS FROM ARTICLE
NORMAL LIVER BIOCHEMISTRY DURING PREGNANCY
During pregnancy, the liver metabolic, synthetic, and excretory functions are affected by the increased serum estrogen and progesterone (1). Serum albumin concentration decreases during pregnancy and reaches a nadir towards the end of the pregnancy, secondary to increase in plasma volume (2). Alkaline phosphatase activity is increased during the third trimester both because of leakage of placental alkaline phosphatase into the maternal circulation and because of increased maternal bone turnover (3). Because of this lack of specificity, the alkaline phosphatase activity is a poor test for the diagnosis of cholestasis during the third trimester of pregnancy. Serum transaminases levels are within normal values during pregnancy (3). Increase in transaminases levels was found during labor and is, most probably, secondary to leakage from the contracting uterine muscle (4). Serum levels of Upsilon-glutamyl-transferase (UpsilonGT) and 5' nucleotidase are normally unchanged (2). Total and free bilirubin serum levels are lower in pregnancy, both because of hemodilution and the low albumin concentration (albumin being the protein that transports bilirubin) (3).
LIVER DISEASE DURING PREGNANCY
Viral Hepatitis
HEPATITIS A. The incidence of acute hepatitis A (HAV) during pregnancy is no different to the incidence of the acute infection in the specific geographic area (55). In endemic areas such as Africa, Asia, and Central America, most of the population has acquired natural immunity by the age of 10 yr. In the United States the reported incidence for HAV is 9/100,000 (56, 57).
The clinical presentation, disease course, treatment, and sequela are similar to those of nonpregnant patients except that pruritus is more common due to the high estrogen state. Perinatal transmission, as reported in case reports, is most probably secondary to horizontal transmission at the time of delivery. Both the inactivated vaccine against HAV and the postexposure immunoglobulin prophylaxis are safe in pregnancy (58).
HEPATITIS B. Hepatitis B (HBV) is a double-stranded, enveloped DNA virus, a member of the hepadena-virus group. It is highly infectious and transmitted parenterally by percutaneous or mucosal exposure, sexually, and from mother to infant. The clinical presentation and course of the disease is similar to nonpregnant infected individuals. Diagnosis of acute HBV infection is made by the detection of HBV surface antigen (HBsAg) in the serum, and confirmed by the presence of IgM antibodies to HBV core antigen (HBcAg). Hepatitis e (envelope) antigen is present in almost all acute infections and represents high infectivity. As long as HBsAg is present, the individual is infectious. As anti-HBe antibody develops, infectivity decreases. Chronic HBV infection is present in approximately 1% of pregnancies in North America (much higher than this in other parts of the world) and acute HBV complicates 1-2 per 1,000 pregnancies (58).
If the acute maternal infection occurs in the first trimester and resolves, the risk for neonatal infection is minimal. In contrast, an infection during the second and third trimesters poses a threat of 10% and 90%, respectively, for vertical transmission. Most perinatal transmission occurs intrapartum (95%). Intrauterine infection is rare. Recently, a study by Xu et al. (59) described the risk factors for intrauterine transmission, which were: maternal serum HBeAg positivity, history of threatened preterm labor, and HBV in the placenta. A child born to a HBeAg positive mother has 70-90% chance of being infected (60). In the Mediterranean basin, Middle East, and Asia, where a variant HBV with a precore/base core promotion mutation, which causes loss of HBeAg synthesis is common (61), infants, when infected, do not develop chronic HBV infection (62) but rather an acute hepatitis (63). A child infected with HbeAg at birth has a 90% chance to progress to chronic infection, while an infection between 7 and 12 months will progress to chronicity in 40% of cases and only 10-20% of cases if the infection occurred between 1 to 3 yr of age (64). The presence of HBsAg in pregnant women does not pose additional risk for the pregnancy and its outcome (65). Since 1991, there is a recommendation to vaccinate individuals in high risk groups such as children born to HBsAg negative mothers (with an addition of immunoprophylaxis with HBV immunoglobulin for children of HBsAg positive mothers) (66). The vaccine is safe to use in pregnancy. Passive and active immunization, given together, are very effective in preventing neonatal transmission, reducing the carrier state of infants born to HBeAg/HBsAg positive women, from 70-90% to almost zero.
With this proper immunoprophylaxis, breastfeeding of infants to chronic HBV carrier women poses no additional risk factor for the transmission of the HBV virus (67).
HEPATITIS D. Hepatitis D (HDV) is a single-stranded circular RNA virus, which depends for its replication on the presence of HBV virus. One of the main routes of transmission is mother to infant. Coinfection with HBV and HDV causes a more severe disease with much higher chronicity rate (70-80% of infected patients will develop cirrhosis), and a much more rapid course of disease. Measures to prevent HBV infection are effective in preventing HDV (58).
HEPATITIS C. Hepatitis C (HCV) is an enveloped single-stranded RNA virus, related to the flavivirus family. It is less common than HBV, but HCV infection becomes chronic in a larger proportion of cases (80%) (58). Percutaneous transmission accounts for at least 60% of cases; others may have no identifiable risk factor. To date, childhood acquisition of HCV infection through maternal-infant transmission has become the most important mode of spread (68). The prevalence of detectable HCV antibodies in pregnant women is 1% overall (range 0.1-2.4%,) not different than that in the general age-matched population. Approximately two-thirds may be expected to have active infection with detectable serum HCV RNA (69). Different studies have shown different conclusions regarding the effect of pregnancy on chronic infection with HCV. The data from these studies is inconsistent and currently no definitive conclusion can be drawn. To date, HCV is not considered a contraindication for pregnancy. Since HCV RNA levels fluctuate during pregnancy, HCV RNA levels should be measured in the third trimester, since the first trimester levels may not be representative or predictive of HCV-RNA concentrations at the time of delivery. Chronic infection with HCV does not have an adverse effect on the pregnancy and its outcome (70). Since passive transmission of IgG antibodies to HCV is present through the placenta, detectable anti-HCV antibodies in the newborn do not mean that a perinatal infection has occurred. Detectable anti-HCV antibodies in infants more than 18 months old or a detectable HCV RNA in infants 3-6 months old can define mother-to-infant transmission of the infection (69). If the mother is viremic with a detectable HCV RNA, the transmission rate reaches 4.3%. The higher the viral load, the higher likelihood of transmission. Geographic location also plays a role, with higher transmission rates reported in Italian and Japanese studies than in studies from other locations (5.6%, 6.9%, and 3.1%, respectively). Coinfection with HIV increases significantly the risk for HCV transmission, being 19.4% in coinfected mothers, compared to a much lower risk of transmission in mothers with HCV alone. Antiretroviral treatment for HIV during pregnancy decreases the rate of HCV transmission to levels of women not coinfected with HIV (71). The HCV genotype has no effect on transmission. Amniocentesis or fetal blood monitoring via scalp vein catheter is considered a relative contraindication during pregnancies of mothers with HCV infection. The type of delivery (cesarean vs vaginal) does not seems to influence the rate of mother to infant transmission (69). Although HCV virus can be detected in breast milk and colostrum, breastfeeding is not considered a risk factor for HCV transmission (72) (note: I question that statement). Treatment of HCV infection with alpha-interferon and ribavirin is contraindicated during pregnancy. alpha-interferon is contraindicated in children under the age of 2 yr because of severe neurotoxicity and because ribavirin is teratogenic. Screening of all prospective mothers is still not indicated. However screening of high risk groups, such as women positive for HIV, previous or current use of injection drugs, current or previous sex partners known to use injection drugs, women who received blood transfusions before 1992, and people from certain geographic areas, can have its advantages (69). The outcome of the children infected perinataly is still not clear. Several studies have demonstrated spontaneous clearance of HCV by 6-24 months of age (73, 74). It is still to be investigated if HCV acquired from mother-to-infant transmission is different than transfusion related HCV and from adult onset HCV.
HEPATITIS E. Hepatitis E (HEV) is a nonenveloped, single-stranded RNA virus. It is endemic in developing countries and shares the route of transmission, risk factors, and chronicity rate with HAV. During pregnancy, HEV can cause fulminant hepatitis indistinguishable from AFLP. There is a significant mortality rate of 16% in pregnant women with acute HEV infection. Transmission occurs intrapartum and peripartum through close contact of mother and neonate. Evidence suggests significant vertical transmission among HEV-RNA positive mothers of up to 50%. Among women with symptomatic infection the rate of transmission is up to 100%, with significant perinatal morbidity and mortality (75, 76).
Autoimmune Hepatitis
Autoimmune hepatitis (AIH) is a chronic liver disease of unknown etiology. It is characterized by a progressive destruction of liver parenchyma leading frequently to cirrhosis. This disease often affects women in their childbearing years, but since cirrhosis is often present at the time of diagnosis, pregnancy is not expected in the untreated patient. Oligo-amenorrhea, which at times is the presenting symptom, is frequent in AIH and is probably due to hypothalamic pituitary dysfunction (77). When disease activity regresses under adequate immunosupression, normal menstruation returns, and a pregnancy is achievable. Two recent studies (78, 79) have described the natural history of AIH during pregnancy. Both studies found a significant decrease in disease activity during pregnancy. Heneghan and colleagues (78) found 11.5% incidence of flare-up during pregnancy, with 11.3% occurring during the 3 months period following delivery. Buchel and colleagues (79) found no exacerbation during pregnancy but relapse was observed in 86% of pregnancies in the postpartum period. In both series the perinatal morbidity and mortality was significantly reduced compared to previous studies. No perinatal death was recorded. Prematurity was observed in 6% of newborns and the caesarian section rate was 3%, compared to 30% and 26%, respectively, in previous studies (80). This improvement is attributed to better immunosupression leading to better disease control. Pregnancy induces a state of immunotolerance, accommodating the fetus, and causes a shift of a TH1 to TH2 immune response (81). This shift causes an amelioration in disease activity which is TH1 dependent, like rheumatoid arthritis, and an exacerbation in disease post partum when the immune status is back to the TH1 predominance (79). Since the pathogenesis of AIH is still not clear, the relevance of a shift from TH1 to TH2 immune response remains speculative. The high hormone levels in pregnancy also play a part in the immunotolerance. Estrogen, in high levels, may inhibit immune response and progesterone and testosterone promote TH2 cells and have antiinflammatory property (82). The use of prednisone is considered safe during pregnancy and lactation but is associated with a small but significant risk of cleft palate in babies to women using the drug in the first trimester of pregnancy (83). Azathioprine seems a safe drug to use during pregnancy, with expanding experience in women with rheumatoid arthritis, inflammatory bowel disease, and after renal transplantation.
Although only 1.2% of the absorbed amount of azathioprine is excreted in breast milk, it is only classified as "probably safe" for use during breastfeeding. Since exacerbation was observed in high frequency postpartum, it is advisable to closely monitor these patients and even enhance the immuno-suppressive therapy to prevent serious flare-ups (79).
Cirrhosis and Portal Hypertension
A woman with decompensated cirrhosis has a low chance of becoming pregnant, mostly because of hypothalamic pituitary dysfunction. Becoming pregnant, her chances to rear her child to adulthood without liver transplantation are 10% (99). Cirrhosis is not a contraindication for pregnancy since the pregnancy does not have a deleterious effect on well-compensated cirrhosis with mild portal hypertension (100). Pregnancy should be planned to a period of time when the liver disease is well compensated. All medications should be reviewed for potential teratogenicity. Early termination of the pregnancy should be considered when hepatic decompensation is present. If an infectious disease is the cause, than proper prophylactic measures should be taken, as detailed previously. Maternal complications arise in nearly half of cirrhosis-affected pregnancies with significant portal hypertension (101). These complications include variceal hemorrhage, hepatic failure, encephalopathy, splenic artery aneurysm, and rupture and malnutrition.
Gastrointestinal hemorrhage are reported to occur in up to 24% of pregnancies in cirrhotic patients with significant portal hypertension and in patients with known varices, up to 78% of cases are reported to have bled prior to the introduction of prophylactic beta-blockers or banding (102). The bleeding occurs mostly in the second or third trimester and is related to the time of maximal blood volume expansion and increased compression on the inferior vena cava by the gravid uterus. Maternal mortality when acute variceal bleeding occurs is high, ranging from 20% to 50% (99). Thus, all pregnant patients with cirrhosis should be screened for varices during the second trimester and treated with selective beta-blockers, when required. It is recommended to keep a short second stage of the labor, to avoid too much valsalva maneuver, which may also promote variceal hemorrhage. The treatment of variceal bleeding in pregnancy is similar to the treatment in nonpregnant patients and consists of endoscopic and pharmacologic treatment. Of note, there is still little experience in pregnancy with the use of octreotide. Vasopressin causes placental ischemia, necrosis, and amputation of digits in the fetus. Twenty-four percent of pregnant cirrhotic patients will experience hepatic decompensation that can lead to rapid deterioration (99). Successful liver transplantation has been described in pregnant women with good results to both woman and fetus (103). Cirrhotic pregnant patients have a 2.6% chance of rupturing a splenic artery aneurysm. There is a female predominance in rupturing a splenic artery aneurysm in all cirrhotic patients. Twenty percent of the ruptures occur during pregnancy. When associated with pregnancy, 70% of ruptures occur in the third trimester. Maternal and fetal mortality is very high, 70% and 80%, respectively (104). Ascites rarely occurs during pregnancy but may suggest Budd-Chiari syndrome.
Aside from the potential complications of variceal hemorrhage, pregnancy does not have a deleterious effect on the cirrhotic liver with mild portal hypertension. The high estrogen state may induce pruritus particularly in those with background cholestatic liver disease, and repeated plasmaphoresis may be required to control this symptom. Similarly, any drug reaction occurring during pregnancy is more likely to be cholestatic.
Liver Transplantation
Nearly 90% of patients resume normal menstruation within 7 months of transplantation (109). Conception should be postponed for at least 6 months posttransplant. This period coincides with the period of maximal immunosuppression. Infections with herpesviruses, especially cytomegalovirus (CMV), are very common at this period and can cause very high maternal and fetal morbidity and mortality. Many of the important considerations about pregnancy after liver transplantation are directly related to the use of immunosuppressant medications during pregnancy.
Corticosteroids are considered safe during pregnancy although they have been associated with fetal growth restriction, suppression of fetal adrenal axis, and premature rupture of membranes (110). As mentioned above, there is a small risk of cleft palate in babies to mothers using steroids in the first trimester (83). Although azathioprine (AZA) crosses the placenta, teratogenicity has not been reported (79). Neonatal bone marrow suppression correlates with maternal leukopenia, which should be avoided. AZA does not cross into breast milk, but breastfeeding experts should be consulted in these situations (111). Cyclosporine readily crosses the placenta and is secreted in high concentrations in breast milk, thus its use is not recommended during pregnancy and lactation. Cyclosporine is highly bound to erythrocytes. In pregnancy, the plasma volume expansion exceeds red cell volume expansion so whole plasma drug levels will be lower than in plasma. In addition, metabolism and clearance are unpredictable during pregnancy, thus prudent monitoring of drug level should be performed (112).
Tacrolimus (FK506) crosses the placenta and is found in significant concentration in the fetal blood and in breast milk. Teratogenicity has not been reported in a study of 27 pregnancies (113). Regarding mycophenolate mofetil (MMF), data remain limited and are insufficient to determine a specific malformation incidence (114). Pregnant women after liver transplantation are at higher risk for hypertension, preeclampsia, premature rupture of membranes, infection, and first trimester abortions. Pregnancy does not seem to alter the function of the grafted liver. A mild-to-moderate elevation in liver enzymes is noted periconception, but does not require aggressive evaluation. Neonatal and perinatal outcomes are favorable (115).
REFERENCES
55. Rustgi VK, Hoofnagle JH. Viral hepatitis during pregnancy. Semin Liver Dis 1987;7: 40-6.
56. Melnick J. History and epidemiology of hepatitis A virus. J Infect Dis 1995;171(Suppl 1):S2.
57. Shapiro C, Coleman P, McQuillan G, et al. Epidemiology of hepatitis A: Seroepidemiology and risk groups in the USA. Vaccine 1992;10(Suppl 1):S59.
58. Magriples U. Hepatitis in pregnancy. Seminars in perinatology 1998;22(2):112-7.
59. Xu DZ, Yan YP, Choi BC, et al. Risk factors and mechanism of transplacental transmission of hepatitis B virus: A case-control study. J Med Virol 2002;671(1):20-6.
60. Beasley RP, Trepo C, Stevens CE, et al. e anigen and vertical transmission of hepatitis B antigen. Am J Epidemiol 1977;105: 94-8.
61. Carman WF, Jacyna MR, Hadziyannis S, et al. Mutations preventing formation of hepatitis B e antigen in patients with chronic hepatitis B infection. Lancet 1989;2: 588-91.
62. Hadziyannis SJ. Hepatitis Be antigen negative chronic hepatitis B: From clinical recognition to pathogenesis and treatment. Viral Hepat Rev 1995;1: 7-36.
63. The Incident Investigation Teams. Transmission of hepatitis B to patients from four surgeons without hepatitis Be antigen. N Engl J Med 1997;336: 178-84.
64. Beasley RP, Hwang LY, Lin CC, et al. Incidence of hepatitis B virus infections in preschool children in Taiwan. J Infect Dis 1982;146: 198-204.
65. Wong S, Chan LY, Yu V, et al. Hepatitis B carrier and perinatal outcome in singleton pregnancy. Am J Perinatol 1999;16(9):485-8.
66. CDC. Update: Recommendations to prevent Hepatitis B virus transmission-United States. MMWR 1995;44: 574-5.
67. Hill JB, Sheffield JS, Kim MJ, et al. Risk of hepatitis B transmission in breast-fed infants of chronic hepatitis B carriers. Obstet Gynecol 2002;99(6):1049-52.
68. Bortolotti F, Resti M, Giacchino R, et al. Changing epidemiologic pattern of chronic hepatitis C virus infection in Italian children. J Pediatr 1998;133: 378-81.
69. Roberts EA, Yeung L. Maternal-infant transmission of hepatitis C virus infection. Hepatology 2002;36: S106-13.
70. Jabeen T, Cannon B, Hogan J, et al. Pregnancy and pregnancy outcome in hepatitis C type 1b. Q J Med 2000;93: 597-601.
71. Conte D, Franquelli M, Prati D, et al. Prevalence and clinical course of chronic hepatitis C virus infection and rate of HCV vertical transmission in a cohort of 12,250 pregnant women. Hepatology 2000;31: 751-5.
72. Polywka S, Schroter M, Feucht HH, et al. Low risk of vertical transmission of hepatitis C by breast milk. Clin Infect Dis 1999;29: 1327-9.
73. Ceci O, Margiotta M, Marello F, et al. High rate of spontaneous viral clearance in a cohort of vertical infected hepatitis C virus infants: What lies behind? J Hepatol 2001;35: 687-8.
74. Ketzinel-Gilad M, Colodner SL, Hadary R, et al. Transient transmission of hepatitis C virus from mothers to newborns. Eur J Clin Microbiol Infect Dis 2000;19: 267-74.
75. Kumar KM, Udumans S, Rana S, et al. Seroprevalence and mother-to-infant transmission of hepatitis E virus among pregnant women in the United Arab Emirates. Eur J Obstet Gynecol Reprod Biol 2001;100(1):9-15.
76. Singh S, Mohanty A, Joshi YK, et al. Mother-to-child transmission of hepatitis E virus infection. Indian J Pediatr 2003;70(1):37-9.
109. Laifer SA, Guido R. Reproductive function and outcome of pregnancy after liver transplantation in women. Mayo Clin Proc 1995;70: 388-94.
110. Beitinis IZ, Bayard F, Ances IG, et al. The transplacental passage of prednisone in pregnancy near term. J Perinatol 1972;81;936-45.
111. Casele HL, Laifer SA. Pregnancy after liver transplantation. Semin Perinatol 1998;22(2):149-55.
112. Ross WB, Richards T, Williams GL, et al. Cyclosporine and pregnancy. Transplantation 1988;45(6):1142.
113. Jain A, Venkateramanan R, Lever J, et al. FK 506 in pregnancy in liver transplant patients. Transplantation 1993;56: 1588-9.
114. Armenti VT, Radomski JS, Moriz MJ, et al. Report from the national transplantation pregnancy registry (NTPR): Outcomes of pregnancy after transplantation. Clin Transpl 2001;97-105.
115. Casele HL, Woelkers DA, Laifer SA. Pregnancy after liver transplantation. Am J Obstet Gynecol 1997;176: S23.
|
|
| |
| |
|
|
|
