Cirrhosis and pregnancy: a single centre experience

Cirrhosis is a diffuse pathology characterized by fibrosis of the liver and is the last stage of chronic liver diseases. It is a serious medical condition which seriously impacts reproduction and reproductive life span. The aim of this study is to evaluate the outcomes of pregnancies complicated with liver cirrhosis. Retrospective chart review of the fetal and maternal results of 20 pregnant women with liver cirrhosis who had undergone antenatal follow-up and delivery at a tertiary center in a 12-year period was performed. Chronic hepatitis B was found to be the leading cause of liver cirrhosis in the study group, with a rate of 25% (n: 5/20). The average MELD score was calculated as 8.8 ± 3.5. Only three patients developed hepatic decompensation during pregnancy. Fetal demise was observed in 10% of the cases (n: 2/20, MELD scores 8 and 17). MELD score was significantly higher in the patients with adverse perinatal outcomes. Even though pregnancy is rarely observed in women with liver cirrhosis, many patients are able to achieve favorable maternal and fetal results without developing hepatic decompensation with appropriate management and close follow-up. The Model for End-Stage Liver Disease (MELD) score is a clinical tool utilized to estimate the severity and survival for chronic liver disease and was previously found to be associated with unfavorable outcomes in pregnant patients. Our study confirms this finding with the current experience from a tertiary care center.


Introduction
Cirrhosis, the last stage of chronic liver diseases, is a diffuse pathology characterized by fibrosis, leaving the normal structure of the liver to abnormal non-organized regenerative nodules [1]. Pregnancy is rarely observed in women with liver cirrhosis overall. Possible causes of this epidemiological fact might be the onset of the disease at the postmenopausal period and the existence of disease-induced metabolic anomalies leading to anovulation and infertility in reproductive age women [2,3]. However, recent studies reported increasing pregnancy and live birth rates in women with cirrhosis [4,5].
The interaction between cirrhosis and the pregnancy is bilateral. As a chronic disease, the course and prognosis of cirrhosis is affected by the pregnancy, while the pregnancy is also impacted by cirrhosis. Physiologic features of pregnancy might overlap by pathologic changes caused by cirrhosis (e.g. drop in platelet count, decrease in serum albumin and sodium concentration, etc.) and distinguishing these two different physiologic mechanisms might be challenging. Previous studies evaluated obstetric and fetal outcomes in pregnant women with cirrhosis, and reported an increase in abortion, fetal growth restriction, preterm delivery, perinatal mortality and cesarean delivery rates [6][7][8][9][10][11]. In addition, complications such as esophageal hemorrhage, liver failure, ascites, encephalopathy and postpartum hemorrhage were observed more frequently in pregnant women with cirrhosis. It is also known that risk of variceal hemorrhage are increased after the second trimester due to the worsening portal hypertension caused by the increase in total blood volume [9,10,12].
In spite of the significant challenges in a pregnancy complicated by cirrhosis, a multi-disciplinary approach and well planned management by experienced providers might minimize the perinatal and maternal risks. The aim of our study is to evaluate the fetal and maternal results of pregnant women with liver cirrhosis in a tertiary health centre in a 12-year period and review the results in the light of the existing literature.

Materials and methods
Pregnant women with liver cirrhosis who had their prenatal care and were delivered at the Department of Obstetrics and Gynecology at Ege University School of Medicine Hospital between 2008 and 2020 were evaluated retrospectively. The cases were initially identified by a computerized search with the diagnosis of "liver cirrhosis" (International Classification of Diseases 10th Revision Codes; K74.0, K74.1, K74.2, K74.3, K74.4, K74.5, K74.6) among pregnant women. Prenatal records, starting from the initial visit and ending with the delivery, were investigated. Nine patients who did not attend prenatal visits regularly or with incomplete records were excluded from the study. The study was approved by the local ethics committee of Ege University School of Medicine.
The demographic features of pregnant women, age, parity, birth, delivery type, birth week, birth weight, 1st and 5th minute Apgar scores, indications were cesarean deliveries, anesthesia type, obstetric, maternal and fetal complications were obtained from antenatal follow-up files. Data regarding laboratory, imaging and endoscopy findings, pathology results and etiology of liver cirrhosis were also obtained.
Adequate prenatal care included monthly visits in the 1st and 2nd trimesters and weekly visits in the third trimester, which were recorded in detail. Maternal, obstetric and fetal complications were recorded. Preterm birth is defined as a delivery occurring at or after 20 0/7 weeks of gestation and before 37 0/7 weeks of gestation [13]. Prelabor rupture of membranes is rupture of membranes before the onset of labor and membrane rupture before labor that occurs before 37 weeks of gestation is referred to as "preterm prelabor rupture of membranes." [14].
Serum samples were obtained at least once every trimester and at certain intervals according to the recommendations of the Gastroenterology department, individualized for each patient. Data on hemoglobin, platelet count, electrolytes, liver enzymes, bilirubin, albumin, creatinine, prothrombin time, international normalized ratio for prothrombin time (INR) value were evaluated.
The Model for End-Stage Liver Disease (MELD) scoring system is a prognostic model developed to predict the severity and survival of patients with chronic liver disease [15]. In our study group, MELD scoring was performed using the first trimester bilirubin, creatinine and INR values [16]. Numerical variables are given as mean and standard deviation or median (min-max). Categorical variables are given as numbers and percentages. Multivariate analysis was not carried out due to the small sample size. Mann-Whitney U test was used to evaluate the value of MELD score in predicting adverse outcomes.

Results
During the study period, 20 pregnancies complicated by liver cirrhosis were identified. Demographic data, obstetric features, maternal and fetal results of the patients are shown in detail in Table 1. Chronic hepatitis B was found to be the leading cause of liver cirrhosis in the study group, with a rate of 25% (n: 5/20). In 20% of cases, despite the comprehensive clinical, serological and pathological evaluation, etiology was not fully elucidated.
Medical outcomes are depicted in Table 2. None of the patients in the study received formal periconceptional counseling prior to conception. Only three patients developed hepatic decompensation during pregnancy which presented with significant abdominal distention due to ascites. MELD score was found to be 9, 17 and 11, respectively. In two cases with hepatic decompensation which presented with new onset ascites and significant distention, pregnancy was terminated with the recommendation of Gastroenterology and in compliance with the family request. The third case decided to pursue the pregnancy until 31 weeks and was delivered with cesarean section due to fetal distress. Six patients had a history of variceal hemorrhage before pregnancy and none of the cases had variceal bleeding or any other morbidity during their pregnancies. These patients were managed with propranolol only, which was continued during pregnancy and none of these patients received any prophylactic interventions before pregnancy as well. No mortality was observed until July 2020 and liver transplantation from cadaver was performed in two cases independent of the pregnancy status.
Mean serum parameters obtained in first trimester and at delivery time are presented in Table 3. Anemia (hemoglobin < 11 g/dL or hematocrit < 33 percent) was observed in 55% of the cases (n: 11/20), and thrombocytopenia (platelet count < 150,000/microL) in 65% (n: 13/20). Model for End-Stage Liver Disease score was calculated using the first trimester serum bilirubin, creatinine and INR values. The average MELD score was calculated as 8.8 ± 3.5. Six patients with a platelet count below the critical value (< 50,000/ microL) received prophylactic platelet transfusion.
MELD score was compared between patients with live birth and with fetal demise and terminations due to complications developed later (n: 14 vs n: 4). MELD score was significantly higher in the group with adverse outcomes (p < 0.05).
Fetal demise was observed in 10% of the cases (n: 2/20, MELD scores 8 and 17). Fetal autopsy was performed and no significant anomalies were found in both fetuses. Termination was performed in a total of four (20%, MELD score, respectively, 9, 10, 6, 11) cases. Two of these patients were with hepatic decompensation and the other two patients were offered termination during first trimester in accordance with the consultation with the Gastroenterology team. The remaining 14 cases resulted in live birth. Among these, the mean gestational week at delivery was 35.7 ± 2.7 and the premature delivery rate was found to be 42.8% (n: 6/14). The cesarean delivery rate was 78.5% (n: 11/14). Vaginal delivery was performed in three cases and neither induction nor neuroaxial blockade was applied. 36% of cesarean section deliveries were performed due to obstetric (Malpresentation, placenta previa, history of previous cesarean section) indications, whereas 27% were due to maternal (Deteriorating liver enzymes) and 18% were due to fetal indications (Non reassuring fetal status). One patient which underwent a trial of labor after cesarean section was delivered with a repeat cesarean section due to patient's desire during labor. However, chart review also revealed non reassuring status, therefore the indication was left as unknown. In cases delivered by cesarean section, regional anesthesia was mostly preferred. (63.6% n: 7/11). Postpartum hemorrhage was observed in only three cases which were successfully managed medically and with blood transfusion. No fetal anomaly was found in any case resulted in live birth, and only one case was complicated with fetal growth restriction.

Discussion
In our study, the maternal and fetal outcomes of 20 pregnancies complicated by liver cirrhosis performed in a single healthcare institution between 2008 and 2020 were evaluated. Liver cirrhosis is a life threatening, chronic and progressive condition which is defined by histopathologic evaluation. The etiology of liver cirrhosis also varies according to the economic status of the countries. Alcoholic liver disease, chronic hepatitis C and non-alcoholic fatty liver disease make up 80% of patients waiting in line for a liver transplant in the United States. [17]. Chronic hepatitis B is the most prevalent diagnosis among our patient population. While many women with liver cirrhosis may be able to pursue their pregnancies without hepatic compromise, some women may show signs of hepatic decompensation, such as ascites, variceal bleeding, encephalopathy, and hepatorenal syndrome [18]. Previous studies reported that the rate of decompensation in pregnant women with liver cirrhosis was 24% and variceal bleeding rate was 32%, while Shaheen et al. reported a decompensation rate as 15%, ascites 11% and variceal bleeding as 5% [9,19,20]. In our study, decompensation was observed in 15% of the patient group, and the ascites was seen in all of these cases. The increase in the total blood volume associated with pregnancy might worsen portal hypertension, and an increased risk of variceal bleeding is also observed, especially after the second trimester [19,21]. Without timely  diagnosis and management, it is inevitable to encounter catastrophic consequences such as increased rates of morbidity and mortality In the literature, maternal mortality rates due to liver cirrhosis during pregnancy vary between 10 and 61%, with the inclusion of earlier studies [10,20,22]. In our study, six patients had a history of variceal bleeding before pregnancy and none of the patients in the study encountered variceal bleeding and maternal mortality during pregnancy. Various abnormalities in serum parameters can be seen in patients with liver cirrhosis. In addition, the disease may even manifest with these abnormalities. Some of the alterations include elevated serum bilirubin, albumin, aminotransferase, alkaline phosphatase, gamma-glutamyl transpeptidase, prolonged prothrombin time and INR values, as well as hyponatremia, anemia, and thrombocytopenia [23][24][25]. Among these, serum albumin levels may lose their prognostic value due to the overlapping with physiologic drop in serum albumin values in pregnancy. Mean albumin levels were found to be decreased at delivery time. However, it is unclear if this drop has contributed to the development of ascites in study patients. Increasing aminotransferase levels can be confusing during preeclampsia workup. However none of the patients in the study developed elevated blood pressures. In our study, mean AST and ALT levels were found to be increased. Thrombocytopenia is the most common hematological disorder, leukopenia and anemia may occur later in the disease [26], 26]. In our study, anemia was noted in 55% of patients, whereas thrombocytopenia was found in 65%. All patients with anemia were evaluated with iron studies, oral and IV iron supplementation was provided if indicated. Three patients require packed red blood cell transfusion before cesarean delivery. Preoperative prophylactic platelet transfusion was needed in approximately half of the cases with thrombocytopenia. While thrombocytopenia in cirrhosis patients is mostly due to congestive splenomegaly and portal hypertension, the development of anemia can be related to a variety of factors such as acute and chronic gastrointestinal blood loss, folate deficiency, alcohol-related toxicity, hypersplenism, anemia of chronic disease and hemolysis [26].
The Model for End-Stage Liver Disease (MELD) score is a clinical tool utilized to estimate the severity and survival for chronic liver disease [15]. In the study of Westbrook et al., the MELD score of 10 and above was associated with poor maternal results [12]. In our study, even though MELD score of 6 patients was 10 and above, the mean MELD score was 8.8. Patients with high MELD scores were associated with poor maternal and fetal outcomes.
The incidence of abortion, stillbirth and preterm delivery appears to be increased in pregnancies complicated by liver cirrhosis [9,12]. Spontaneous abortion occurred in 10% of the cases and pregnancy was terminated in 20% of the patients. Preterm delivery rate was 42.8% and the indication for preterm deliveries were obstetric and fetal in majority of patients. There is still not enough evidence regarding the ideal delivery route in pregnancies complicated with cirrhosis. A theoretical increase in the risk of bleeding that may result from injuries to abdominal collaterals during the entry into the abdomen in cesarean section is likely. However, an increase in the risk of variceal bleeding due to increased abdominal pressure during vaginal deliveries is another concern. Due to the lack of consensus in the method of delivery, there are obvious differences among the studies reported in terms of cesarean delivery rates. In the study conducted by Rasheed et al. cesarean rate in pregnant women with cirrhosis was 91.5%, while it was reported as 26% in the study conducted by Palatnik et al. [18,27]. In our study, while cesarean delivery rate was 78.5%, most of the cesarean sections were performed due to obstetric indications. In addition, regional anesthesia was mostly preferred in cesarean deliveries due to its advantages on maternal cardiopulmonary and hemodynamic system.
It is important to emphasize that the limitations of our study include the number of the patients enrolled in the study and retrospective nature of the study. Statistical analysis may be underpowered due to the sample size. In addition, the end of observation period in the study protocol which was specified as the delivery is a major limitation of the study.
According to the studies in the literature, many patients with chronic liver disease are able to achieve favorable maternal and fetal results without developing hepatic decompensation with appropriate management and close follow-up. Despite all these favorable advancements, termination of the pregnancy should be strongly considered in pregnant women with high risk of variceal bleeding or in the presence of symptoms concerning for hepatic decompensation. Therefore, risk of variceal bleeding should be thoroughly assessed before pregnancy and counseling should be provided with respect to patient's autonomy. Pregnancy in patients with liver cirrhosis should be managed in a multi-disciplinary approach, including a maternal-fetal medicine specialist, neonatologist and hepatologist.
Author contribution All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by MI, HE, FO and ME. The first draft of the manuscript was written by MI, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.