http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, Early Online: 1–4 ! 2015 Informa UK Ltd. DOI: 10.3109/14767058.2015.1009438
ORIGINAL ARTICLE
Outcomes of pregnancies complicated by liver cirrhosis, portal hypertension, or esophageal varices Anela Puljic1, Jennifer Salati2, Amy Doss2, and Aaron B. Caughey2 J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
1
Department of Reproductive Medicine, University of California, San Diego, CA, USA and 2Department of Obstetrics & Gynecology, Oregon Health & Science University, Portland, OR, USA Abstract
Keywords
Objective: To evaluate pregnancy outcomes in women with liver cirrhosis, portal hypertension, or esophageal varices. Study design: We analyzed a retrospective cohort of 2 284 218 pregnancies in 2005–2009 recorded in the California Birth Registry database. Utilizing ICD-9 codes we analyzed the following outcomes for liver cirrhosis, portal hypertension, or esophageal varices in pregnancy: preeclampsia (PET), preterm delivery (PTD; 537 weeks), cesarean section, low birth weight (LBW; 52500 g), small for gestational age (SGA; 510th percentile), neonatal death (NND), and postpartum hemorrhage (PPH). Results: Cirrhosis in pregnancy conferred an increased risk of PET, PTD, CS in multiparous women, LBW, and NND. Portal hypertension in pregnancy was associated with PTD, LBW, NND, and PPH. Non-bleeding esophageal varices in pregnancy were not associated with the outcomes assessed in a statistically significant manner. One case of bleeding esophageal varices was observed, resulting in PTD with a LBW infant. There were three cases of concomitant portal hypertension or concomitant esophageal varices with cirrhosis in pregnancy. Conclusion: Pregnancy in women with concomitant liver cirrhosis, portal hypertension, or esophageal varices can be successful. However, pregnancy outcomes are worse and may warrant closer antenatal monitoring and patient counseling. Cirrhosis in pregnancy with concomitant portal hypertension or esophageal varices is rare.
Esophageal varices, liver cirrhosis, portal hypertension, pregnancy outcomes
Introduction Cirrhosis of the liver is defined as irreversible chronic hepatocyte injury with extensive fibrosis and nodular regeneration that results in coagulopathy, circulatory and immunologic dysfunction [1,2]. Historically it has been uncommon in reproductive-aged women. Moreover, it is associated with metabolic derangements and anovulatory cycles resulting in infertility, making it an even more rare condition in pregnancy to the degree that the prevalence of cirrhosis in this population is yet undefined [1–5]. However, with improved therapeutic options for liver disease, more women are presenting for prenatal care with concomitant cirrhosis [5]. Additionally, the burden of conditions leading to cirrhosis of the liver are on the rise due to the increasing prevalence of alcohol abuse, non-alcoholic fatty liver disease, viral hepatitides, and hepatocellular carcinoma [6,7]. For instance, one population-based study utilizing the US Nationwide Inpatient Sample database observed an increase in the mean number of Address for correspondence: Aaron B. Caughey, MD, PhD, Department of Obstetrics and Gynecology, Oregon Health & Science University, 3181 SW Sam Jackson Park Road – Mail stop L466, Portland, OR 97239, USA. E-mail: [email protected]
History Received 7 October 2014 Revised 5 January 2015 Accepted 15 January 2015 Published online 6 February 2015
nationwide deliveries complicated by cirrhosis from 68 to 106 annually from 1993 to 1999 and 2000 to 2005 [8]. Cirrhosis inherently carries significant morbidity and mortality and may progress to also involve portal hypertension and esophageal varices, whether these conditions are associated with worse pregnancy outcomes has been debated in the literature [2,3,6,9–13]. Maternal mortality has been reported to range from 10 to 18%, primarily due to development of hepatic decompensation during pregnancy [11]. A limited number of small retrospective cohort studies and case reports have described an increase in spontaneous abortion, premature birth, small for gestational age (SMA) fetus, and perinatal death [1,11,14]. One small prospective cohort study in Egypt even observed an increased risk of preeclampsia (PET), hyperemesis gravidum, placental abruption, premature rupture of membranes, intrauterine growth restriction, postpartum hemorrhage (PPH), and cesarean section as mode of delivery [12]. The majority of literature consists of small case series and case reports with varied observations decades prior to many advances in obstetrical, neonatal, and cirrhosis care. To date, no published guidelines exist regarding the management of a pregnancy with concomitant cirrhosis, portal hypertension, or
2
A. Puljic et al.
J Matern Fetal Neonatal Med, Early Online: 1–4
The outcomes compared women with and without the hepatic conditions above included PET, preterm delivery (PTD; 537 weeks), cesarean section (nulliparous versus multiparous), low birth weight (LBW;52500 g), SMA (510th percentile of birth weight for gestational age), neonatal death (NND), and PPH. Chi-squared tests were used for statistical analysis. A p value 50.05 was considered statistically significant. The data are presented as odds ratios (OR) with 95% confidence intervals.
esophageal varices. These challenging clinical scenarios raise questions regarding family planning counseling, pregnancy care, timing and mode of delivery, as well as what unique postpartum complications to expect. Our study aims to identify pregnancy outcomes associated with cirrhosis, portal hypertension, or esophageal varices in a large, more recent patient population between the years of 2005–2009.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Methods This is a retrospective cohort study including 2 284 218 singleton, non-anomalous pregnancies that occurred in California during the years of 2005 to 2009. Approval from the institutional review board at Oregon Health & Science University was obtained. The data used for outcome analysis were obtained from the California Vital Statistics Birth Certificate Data, California Patient Discharge Data, Vital Statistics Death Certificate Data, and Vital Statistics Fetal Death File [15] These data are part of the public record and de-identified, therefore, informed consent was not required. The state of California maintains linked maternal data sets starting 9 months prior to delivery and up to 1 year after delivery. The data sets also include infant birth records and all hospital admission up to 1 year of life. A unique ‘‘record linkage number’’ is assigned to the mother–infant pair by the California Office of Statewide Health Planning and Development Healthcare Information Resource Center under the state of California. International Classification of Diseases 9th Version (ICD-9) codes were used to identify the following diagnoses in our pregnancy cohort: cirrhosis (ICD-9 571.2, alcoholic; 571.5, non-alcoholic), portal hypertension (ICD-9 572.3), esophageal varices without bleeding (ICD-9 456.1 and 456.21) and esophageal varices with bleeding (ICD-9 456.0 and 456.20). Of these pregnancies, 37 were complicated by cirrhosis of the liver, 17 complicated by portal hypertension, and 10 complicated by esophageal varices without bleeding and 1 with bleeding.
Results Maternal characteristics that were associated with statistical significance in pregnant women with cirrhosis were maternal age over 35 years and chronic hypertension. Those with portal hypertension were more likely to have chronic hypertension and no prenatal care, while those with esophageal varices were more likely to be nulliparous and older than 35 years (Table 1). Maternal and fetal outcomes were observed to be worse in pregnancies complicated by liver cirrhosis, portal hypertension, or esophageal varices. (Tables 2–4) Cirrhosis confers a statistically significant increased risk of PET (13.5% versus 3.0%; p50.001), PTD (28.6% versus 8.9%; p ¼ 0.001), cesarean section in multiparous women (27.8% versus 10.1%; p ¼ 0.029), LBW (18.9% versus 4.5%; p ¼ 0.001), and NND (2.7% versus 0.1%; p ¼ 0.040). Cirrhosis was also associated with an increased risk of cesarean section in nulliparous women (41.7% versus 29.3%; p ¼ 0.351), a SMA fetus (16.7% versus 10.5%; p ¼ 0.426), and PPH (5.4% versus 2.9%; p ¼ 0.294), however, these did not reach statistical significance. Of the 37 cases of cirrhosis in pregnancy, 3 cases of concomitant portal hypertension, 3 of concomitant nonbleeding esophageal varices, and only 1 case of concomitant bleeding esophageal varices were observed. Portal hypertension in pregnancy confers a statistically significant increased risk of PTD (50.0% versus 8.9%, p50.001), LBW (41.2% versus 4.5%; p50.001), NND (5.9% versus 0.1%; p ¼ 0.019),
Table 1. Demographics of women without and with cirrhosis, portal hypertension and esophageal varices.
Demographics Ethnicity White African American Hispanic Asian Other Nulliparous Maternal age 435 Public insurance Education 412 y First PNC visit 1st trimester 2nd trimester 3rd trimester No prenatal care Chronic HTN Diabetes mellitus GDM
Control* %
Cirrhosisy %
p value
Portal HTNz %
0.173 28.5 5.0 52.4 11.8 2.3 39.4 17.1 48.7 45.0
40.5 0.0 54.1 2.7 2.7 2.4 40.5 48.7 54.1
83.2 12.6 2.4 1.8 1.1 0.8 6.7
75.7 18.9 2.7 2.7 10.8 2.7 10.8
0.384 0.001 0.998 0.266 0.661
50.001 0.169 0.316
p value
EVô %
0.300 29.4 11.8 35.3 23.5 0.0 47.1 29.4 47.1 52.9 76.5 11.8 0.0 11.8 23.5 0.0 5.9
0.520 0.193 0.894 0.508 0.021
50.001 0.720 0.894
p value
Bleeding EVx %
0.430 30.0 0.0 40.0 30.0 0.0 70.0 50.0 50.0 60.0 90.0 0.0 0.0 10.0 10.0 0.0 0.0
0.048 0.017 0.933 0.339 0.156
0.008 0.783 0.397
p value 0.923
0.0 0.0 100.0 0.0 0.0 0.0 100.0 0.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0
0.420 0.171 0.330 0.268 0.977
0.914 0.931 0.789
HTN, hypertension; EV, esophageal varices; PNC, prenatal care; HTN, hypertension; GDM, gestational diabetes mellitus. *n ¼ 2 284 218; yn ¼ 37; zn ¼ 17; ôn ¼ 10; xn ¼ 1.
Pregnancy outcomes in cirrhosis or its sequelae
DOI: 10.3109/14767058.2015.1009438
3
Table 2. Odds ratios for pregnancy outcomes with liver cirrhosis.
Liver cirrhosis (n ¼ 37) Preeclampsia Preterm delivery (537 weeks) Cesarean section (nulliparous) Cesarean section (multiparous, no prior) Low birth weight (52500 g) Small for gestational age Neonatal death Postpartum hemorrhage
% Control
% ICP
Odds ratio
95% Confidence interval
3.0 10 8.9 10 29.3 10.1 4.5 10.5 0.1 2.9
13.5
4.65
[1.59–12.49]
28.6
4.08
[1.83–8.88]
41.7 27.8 18.9 16.7 2.7 5.4
1.73 3.44 4.93 1.71 24.99 1.90
[0.48–6.03] [1.07–10.31] [1.98–11.74] [0.40–6.27] [1.27–169.05] [0.32–8.07]
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Table 3. Odds ratios for pregnancy outcomes with portal hypertension. % Control
% ICP
Odds ratio
95% Confidence interval
2.97 10 8.93 10 29.29 10.07 4.52 10.46 0.11 2.92
11.76
4.36
[0.69–19.81]
50.00
10.20
[3.50–29.79]
62.50 25.00 41.18 16.67 5.88 23.53
4.02 2.23 14.79 1.71 56.23 10.23
[0.84–21.16] [0.10–20.96] [5.10–42.04] [0.08–14.91] [2.78–400.01] [2.82–33.57]
Portal hypertension (n ¼ 17) Preeclampsia Preterm delivery (537 weeks) Cesarean section (nulliparous) Cesarean section (multiparous, no prior cesareans) Low birth weight (52500 g) Small for gestational age Neonatal death Postpartum hemorrhage
Table 4. Odds ratios for pregnancy outcomes with non-bleeding esophageal varices. % Control
% ICP
Odds ratio
95% Confidence interval
3.0 10 8.9 10 29.3 10.1 4.5 10.5 0.1 2.9
10.0
3.63
[0.17–27.70]
30.0
4.37
[0.90–7.24]
57.1 33.3 20.0 0.0 0.0 0.0
3.22 4.47 5.28 0.00 0.00 0.00
[0.61–18.03] [0.16–62.32] [0.78–12.35] [0.00–19.06] [0.00–473.02] [0.00–17.48]
Non-bleeding esophageal varices (n ¼ 10) Preeclampsia Preterm delivery (537 weeks) Cesarean section (nulliparous) Cesarean section (multiparous, no prior cesareans) Low birth weight (52500 g) Small for gestational age Neonatal death Postpartum hemorrhage
and PPH (23.5% versus 2.9%; p ¼ 0.001). Cesarean section in nulliparous (62.5% versus 29.3; p ¼ 0.053) and multiparous women (25.0% versus 10.1%; p ¼ 0.412) with portal hypertension was increased, but did not reach statistical significance. Likewise, the risk of PET (11.8% versus 3.0; p ¼ 0.089) and SMA (16.7% versus 10.5%; p ¼ 0.485) did not reach statistical significance. In pregnancies with nonbleeding esophageal varices, there were no statistically increased risks of the outcomes assessed. However, the risk of PTD (30.0% versus 8.9%; p ¼ 0.053) and LBW (20.0% versus 4.5%; p ¼ 0.072) were approaching statistical significance. Of the pregnancies complicated by non-bleeding esophageal varices, there were no cases of SMA, NND, or PPH. Only one case of bleeding esophageal varices was observed in our pregnancy cohort, resulting in PTD of a LBW infant. This case of bleeding esophageal varices in pregnancy was not associated with PET, cesarean section, SMA fetus, NND, or PPH.
Discussion The incidences of pregnancies complicated by liver cirrhosis, portal hypertension, or esophageal varices are on the rise and are associated with worse pregnancy outcomes. Previous studies have debated various outcomes and whether these pregnancies can be successful. The strength of our study is that it is consists of contemporary data extracted from a reliable birth registry database. Thus, we were able to make outcome observations that are relevant to recent improvements in both obstetrical care and the care of individuals with liver disease. We found that these pregnancies can be successful. Nonetheless, these high-risk patients remain prone to lifethreatening sequelae. We rarely observed cases of pregnancies complicated by liver cirrhosis with a concomitant diagnosis of portal hypertension or esophageal varices. Our findings for our cirrhosis-only pregnancy cohort are consistent with one
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
4
A. Puljic et al.
previously published population-based study [8]. However, our study goes a step further by additionally assessing the pregnancy outcomes for women who have concomitant portal hypertension and esophageal varices, which are the natural sequelae of cirrhosis. We observed worse outcomes in our pregnancy cohort with cirrhosis, the strongest association being PTD followed by LBW, PET, cesarean section in multiparous women, and NND. Women with portal hypertension were more likely to deliver preterm, have a LBW infant, with an increased risk for NND and PPH. Pregnancies with liver cirrhosis, portal hypertension, or non-bleeding esophageal varices are consistently associated with an increased risk for cesarean section. However, it is unclear whether this could be due in part to provider comfort in these unique delivery scenarios. Further studies are needed to evaluate whether vaginal, assisted, or cesarean delivery is the optimal delivery mode in these conditions. An increase in PTD was observed for both liver cirrhosis and portal hypertension in pregnancy, which raises further questions regarding whether these deliveries were induced or not. These preterm deliveries may also be associated with our finding of an association of NND with both cirrhosis and portal hypertension. PPH was only observed in our portal hypertension cohort, which may be associated in part with increased portal vasculature pressure and potentially a coagulopathy. This may require pre-delivery planning for management of hemorrhage in these patients. While this study provides a large cohort in order to examine obstetric and neonatal outcomes of pregnancies complicated by severe hepatic disease, there are limitations. The conditions assessed may manifest with varying severity and sequelae in individual patients. Also, in order to obtain such a large cohort, administrative data were utilized and diagnoses were confirmed with ICD-9 codes. In general, women with such codes likely have the diseases noted, but under-coding can be a problem. Such misclassification would only lead to an underestimation of the effect difference, and because these are rare diagnoses, this underestimation should be small. Additionally, we did not have specific metrics assessing the level of disease such as length of time of disease, liver function tests, etc. Finally, even with this large sample, the study was underpowered for several outcomes and unable to stratify the conditions any further. Despite these limitations, our study denotes a number of obstetric and neonatal complications related to pregnancy with liver cirrhosis, portal hypertension, or esophageal varices. Women with these conditions can be counseled
J Matern Fetal Neonatal Med, Early Online: 1–4
both prior to pregnancy as well as during pregnancy regarding these risks. In order to reduce these risks, perhaps closer antenatal monitoring of pregnancies with these conditions is warranted. In addition, careful planning of delivery mode should take place while counseling the patient during pregnancy. As women with severe hepatic disease are maintained with their chronic disease longer due to improvements in medical care, we will see an increasing number in our practices. Careful counseling and optimization of their care are key in achieving the best outcomes possible for these truly high-risk pregnancies.
Declaration of interest The authors report no declarations of interest.
References 1. Aggarwal N, Sawnhey H, Suril V, et al. Pregnancy and Cirrhosis of the Liver. Aust NZ J Obstet Gynaecol 1999;39:503–6. 2. Lee NM, Brady CW. Liver disease in pregnancy. World J Gastroenterol 2009;15:897–906. 3. Russell MA, Craigo SD. Cirrhosis and portal hypertension in pregnancy. Semin Perinatol 1998;22:156–65. 4. Schreyer P, Caspi E, El-Hindi JM, Eshchar J. Cirrhosis-pregnancy and delivery: a review. Obstet Gynecol Surv 1982;37:304–12. 5. Tan J, Surti B, Saab S. Pregnancy and cirrhosis. Liver Transpl 2008;14:1081–91. 6. Cerqui AJ, Haran M, Brodribb R. Implications of liver cirrhosis in pregnancy. Aust NZ J Obstet Gynaecol 1998;38:93–5. 7. Kim WR, Brown Jr RS, Terrault NA, El-Serag H. Burden of liver disease in the United States: summary of a workshop. Hepatology 2002;36:227–42. 8. Shaheen AA, Myers RP. The outcomes of pregnancy in patients with cirrhosis: a population-based study. Liver Int 2010;30:275–83. 9. Borhanmanesh F, Haghighi P. Pregnancy in patients with cirrhosis of the liver. Obstet Gynecol 1970;36:315–24. 10. Cheng YS. Pregnancy in liver cirrhosis and/or portal hypertension. Am J Obstet Gynecol 1977;128:812–22. 11. Pajor A, Lehoczky D. Pregnancy in liver cirrhosis. Assessment of maternal and fetal risks in eleven patients and review of the management. Gynecol Obstet Invest 1994;38:45–50. 12. Rasheed SM, Abdel Monem AM, Abd Ellah AH, Abdel Fattah MS. Prognosis and determinants of pregnancy outcome among patients with post-hepatitis liver cirrhosis. Int J Gynaecol Obstet 2013;121: 247–51. 13. Restaino A, Campobasso C, D’Aloya A, et al. Cirrhosis and pregnancy: a case report and review of the literature. Clin Exp Obstet Gynecol 1996;23:240–7. 14. Peitsidou A, Peitsidis P, Michopoulos S, et al. Exacerbation of liver cirrhosis in pregnancy: a complex emerging clinical situation. Arch Gynecol Obstet 2009;279:911–13. 15. California Department of Health Services. Birth cohort public use file, 2005–2009. Sacramento, CA: California Department of Health Services; 2013.
ORIGINAL ARTICLE
Outcomes of pregnancies complicated by liver cirrhosis, portal hypertension, or esophageal varices Anela Puljic1, Jennifer Salati2, Amy Doss2, and Aaron B. Caughey2 J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
1
Department of Reproductive Medicine, University of California, San Diego, CA, USA and 2Department of Obstetrics & Gynecology, Oregon Health & Science University, Portland, OR, USA Abstract
Keywords
Objective: To evaluate pregnancy outcomes in women with liver cirrhosis, portal hypertension, or esophageal varices. Study design: We analyzed a retrospective cohort of 2 284 218 pregnancies in 2005–2009 recorded in the California Birth Registry database. Utilizing ICD-9 codes we analyzed the following outcomes for liver cirrhosis, portal hypertension, or esophageal varices in pregnancy: preeclampsia (PET), preterm delivery (PTD; 537 weeks), cesarean section, low birth weight (LBW; 52500 g), small for gestational age (SGA; 510th percentile), neonatal death (NND), and postpartum hemorrhage (PPH). Results: Cirrhosis in pregnancy conferred an increased risk of PET, PTD, CS in multiparous women, LBW, and NND. Portal hypertension in pregnancy was associated with PTD, LBW, NND, and PPH. Non-bleeding esophageal varices in pregnancy were not associated with the outcomes assessed in a statistically significant manner. One case of bleeding esophageal varices was observed, resulting in PTD with a LBW infant. There were three cases of concomitant portal hypertension or concomitant esophageal varices with cirrhosis in pregnancy. Conclusion: Pregnancy in women with concomitant liver cirrhosis, portal hypertension, or esophageal varices can be successful. However, pregnancy outcomes are worse and may warrant closer antenatal monitoring and patient counseling. Cirrhosis in pregnancy with concomitant portal hypertension or esophageal varices is rare.
Esophageal varices, liver cirrhosis, portal hypertension, pregnancy outcomes
Introduction Cirrhosis of the liver is defined as irreversible chronic hepatocyte injury with extensive fibrosis and nodular regeneration that results in coagulopathy, circulatory and immunologic dysfunction [1,2]. Historically it has been uncommon in reproductive-aged women. Moreover, it is associated with metabolic derangements and anovulatory cycles resulting in infertility, making it an even more rare condition in pregnancy to the degree that the prevalence of cirrhosis in this population is yet undefined [1–5]. However, with improved therapeutic options for liver disease, more women are presenting for prenatal care with concomitant cirrhosis [5]. Additionally, the burden of conditions leading to cirrhosis of the liver are on the rise due to the increasing prevalence of alcohol abuse, non-alcoholic fatty liver disease, viral hepatitides, and hepatocellular carcinoma [6,7]. For instance, one population-based study utilizing the US Nationwide Inpatient Sample database observed an increase in the mean number of Address for correspondence: Aaron B. Caughey, MD, PhD, Department of Obstetrics and Gynecology, Oregon Health & Science University, 3181 SW Sam Jackson Park Road – Mail stop L466, Portland, OR 97239, USA. E-mail: [email protected]
History Received 7 October 2014 Revised 5 January 2015 Accepted 15 January 2015 Published online 6 February 2015
nationwide deliveries complicated by cirrhosis from 68 to 106 annually from 1993 to 1999 and 2000 to 2005 [8]. Cirrhosis inherently carries significant morbidity and mortality and may progress to also involve portal hypertension and esophageal varices, whether these conditions are associated with worse pregnancy outcomes has been debated in the literature [2,3,6,9–13]. Maternal mortality has been reported to range from 10 to 18%, primarily due to development of hepatic decompensation during pregnancy [11]. A limited number of small retrospective cohort studies and case reports have described an increase in spontaneous abortion, premature birth, small for gestational age (SMA) fetus, and perinatal death [1,11,14]. One small prospective cohort study in Egypt even observed an increased risk of preeclampsia (PET), hyperemesis gravidum, placental abruption, premature rupture of membranes, intrauterine growth restriction, postpartum hemorrhage (PPH), and cesarean section as mode of delivery [12]. The majority of literature consists of small case series and case reports with varied observations decades prior to many advances in obstetrical, neonatal, and cirrhosis care. To date, no published guidelines exist regarding the management of a pregnancy with concomitant cirrhosis, portal hypertension, or
2
A. Puljic et al.
J Matern Fetal Neonatal Med, Early Online: 1–4
The outcomes compared women with and without the hepatic conditions above included PET, preterm delivery (PTD; 537 weeks), cesarean section (nulliparous versus multiparous), low birth weight (LBW;52500 g), SMA (510th percentile of birth weight for gestational age), neonatal death (NND), and PPH. Chi-squared tests were used for statistical analysis. A p value 50.05 was considered statistically significant. The data are presented as odds ratios (OR) with 95% confidence intervals.
esophageal varices. These challenging clinical scenarios raise questions regarding family planning counseling, pregnancy care, timing and mode of delivery, as well as what unique postpartum complications to expect. Our study aims to identify pregnancy outcomes associated with cirrhosis, portal hypertension, or esophageal varices in a large, more recent patient population between the years of 2005–2009.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Methods This is a retrospective cohort study including 2 284 218 singleton, non-anomalous pregnancies that occurred in California during the years of 2005 to 2009. Approval from the institutional review board at Oregon Health & Science University was obtained. The data used for outcome analysis were obtained from the California Vital Statistics Birth Certificate Data, California Patient Discharge Data, Vital Statistics Death Certificate Data, and Vital Statistics Fetal Death File [15] These data are part of the public record and de-identified, therefore, informed consent was not required. The state of California maintains linked maternal data sets starting 9 months prior to delivery and up to 1 year after delivery. The data sets also include infant birth records and all hospital admission up to 1 year of life. A unique ‘‘record linkage number’’ is assigned to the mother–infant pair by the California Office of Statewide Health Planning and Development Healthcare Information Resource Center under the state of California. International Classification of Diseases 9th Version (ICD-9) codes were used to identify the following diagnoses in our pregnancy cohort: cirrhosis (ICD-9 571.2, alcoholic; 571.5, non-alcoholic), portal hypertension (ICD-9 572.3), esophageal varices without bleeding (ICD-9 456.1 and 456.21) and esophageal varices with bleeding (ICD-9 456.0 and 456.20). Of these pregnancies, 37 were complicated by cirrhosis of the liver, 17 complicated by portal hypertension, and 10 complicated by esophageal varices without bleeding and 1 with bleeding.
Results Maternal characteristics that were associated with statistical significance in pregnant women with cirrhosis were maternal age over 35 years and chronic hypertension. Those with portal hypertension were more likely to have chronic hypertension and no prenatal care, while those with esophageal varices were more likely to be nulliparous and older than 35 years (Table 1). Maternal and fetal outcomes were observed to be worse in pregnancies complicated by liver cirrhosis, portal hypertension, or esophageal varices. (Tables 2–4) Cirrhosis confers a statistically significant increased risk of PET (13.5% versus 3.0%; p50.001), PTD (28.6% versus 8.9%; p ¼ 0.001), cesarean section in multiparous women (27.8% versus 10.1%; p ¼ 0.029), LBW (18.9% versus 4.5%; p ¼ 0.001), and NND (2.7% versus 0.1%; p ¼ 0.040). Cirrhosis was also associated with an increased risk of cesarean section in nulliparous women (41.7% versus 29.3%; p ¼ 0.351), a SMA fetus (16.7% versus 10.5%; p ¼ 0.426), and PPH (5.4% versus 2.9%; p ¼ 0.294), however, these did not reach statistical significance. Of the 37 cases of cirrhosis in pregnancy, 3 cases of concomitant portal hypertension, 3 of concomitant nonbleeding esophageal varices, and only 1 case of concomitant bleeding esophageal varices were observed. Portal hypertension in pregnancy confers a statistically significant increased risk of PTD (50.0% versus 8.9%, p50.001), LBW (41.2% versus 4.5%; p50.001), NND (5.9% versus 0.1%; p ¼ 0.019),
Table 1. Demographics of women without and with cirrhosis, portal hypertension and esophageal varices.
Demographics Ethnicity White African American Hispanic Asian Other Nulliparous Maternal age 435 Public insurance Education 412 y First PNC visit 1st trimester 2nd trimester 3rd trimester No prenatal care Chronic HTN Diabetes mellitus GDM
Control* %
Cirrhosisy %
p value
Portal HTNz %
0.173 28.5 5.0 52.4 11.8 2.3 39.4 17.1 48.7 45.0
40.5 0.0 54.1 2.7 2.7 2.4 40.5 48.7 54.1
83.2 12.6 2.4 1.8 1.1 0.8 6.7
75.7 18.9 2.7 2.7 10.8 2.7 10.8
0.384 0.001 0.998 0.266 0.661
50.001 0.169 0.316
p value
EVô %
0.300 29.4 11.8 35.3 23.5 0.0 47.1 29.4 47.1 52.9 76.5 11.8 0.0 11.8 23.5 0.0 5.9
0.520 0.193 0.894 0.508 0.021
50.001 0.720 0.894
p value
Bleeding EVx %
0.430 30.0 0.0 40.0 30.0 0.0 70.0 50.0 50.0 60.0 90.0 0.0 0.0 10.0 10.0 0.0 0.0
0.048 0.017 0.933 0.339 0.156
0.008 0.783 0.397
p value 0.923
0.0 0.0 100.0 0.0 0.0 0.0 100.0 0.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0
0.420 0.171 0.330 0.268 0.977
0.914 0.931 0.789
HTN, hypertension; EV, esophageal varices; PNC, prenatal care; HTN, hypertension; GDM, gestational diabetes mellitus. *n ¼ 2 284 218; yn ¼ 37; zn ¼ 17; ôn ¼ 10; xn ¼ 1.
Pregnancy outcomes in cirrhosis or its sequelae
DOI: 10.3109/14767058.2015.1009438
3
Table 2. Odds ratios for pregnancy outcomes with liver cirrhosis.
Liver cirrhosis (n ¼ 37) Preeclampsia Preterm delivery (537 weeks) Cesarean section (nulliparous) Cesarean section (multiparous, no prior) Low birth weight (52500 g) Small for gestational age Neonatal death Postpartum hemorrhage
% Control
% ICP
Odds ratio
95% Confidence interval
3.0 10 8.9 10 29.3 10.1 4.5 10.5 0.1 2.9
13.5
4.65
[1.59–12.49]
28.6
4.08
[1.83–8.88]
41.7 27.8 18.9 16.7 2.7 5.4
1.73 3.44 4.93 1.71 24.99 1.90
[0.48–6.03] [1.07–10.31] [1.98–11.74] [0.40–6.27] [1.27–169.05] [0.32–8.07]
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Table 3. Odds ratios for pregnancy outcomes with portal hypertension. % Control
% ICP
Odds ratio
95% Confidence interval
2.97 10 8.93 10 29.29 10.07 4.52 10.46 0.11 2.92
11.76
4.36
[0.69–19.81]
50.00
10.20
[3.50–29.79]
62.50 25.00 41.18 16.67 5.88 23.53
4.02 2.23 14.79 1.71 56.23 10.23
[0.84–21.16] [0.10–20.96] [5.10–42.04] [0.08–14.91] [2.78–400.01] [2.82–33.57]
Portal hypertension (n ¼ 17) Preeclampsia Preterm delivery (537 weeks) Cesarean section (nulliparous) Cesarean section (multiparous, no prior cesareans) Low birth weight (52500 g) Small for gestational age Neonatal death Postpartum hemorrhage
Table 4. Odds ratios for pregnancy outcomes with non-bleeding esophageal varices. % Control
% ICP
Odds ratio
95% Confidence interval
3.0 10 8.9 10 29.3 10.1 4.5 10.5 0.1 2.9
10.0
3.63
[0.17–27.70]
30.0
4.37
[0.90–7.24]
57.1 33.3 20.0 0.0 0.0 0.0
3.22 4.47 5.28 0.00 0.00 0.00
[0.61–18.03] [0.16–62.32] [0.78–12.35] [0.00–19.06] [0.00–473.02] [0.00–17.48]
Non-bleeding esophageal varices (n ¼ 10) Preeclampsia Preterm delivery (537 weeks) Cesarean section (nulliparous) Cesarean section (multiparous, no prior cesareans) Low birth weight (52500 g) Small for gestational age Neonatal death Postpartum hemorrhage
and PPH (23.5% versus 2.9%; p ¼ 0.001). Cesarean section in nulliparous (62.5% versus 29.3; p ¼ 0.053) and multiparous women (25.0% versus 10.1%; p ¼ 0.412) with portal hypertension was increased, but did not reach statistical significance. Likewise, the risk of PET (11.8% versus 3.0; p ¼ 0.089) and SMA (16.7% versus 10.5%; p ¼ 0.485) did not reach statistical significance. In pregnancies with nonbleeding esophageal varices, there were no statistically increased risks of the outcomes assessed. However, the risk of PTD (30.0% versus 8.9%; p ¼ 0.053) and LBW (20.0% versus 4.5%; p ¼ 0.072) were approaching statistical significance. Of the pregnancies complicated by non-bleeding esophageal varices, there were no cases of SMA, NND, or PPH. Only one case of bleeding esophageal varices was observed in our pregnancy cohort, resulting in PTD of a LBW infant. This case of bleeding esophageal varices in pregnancy was not associated with PET, cesarean section, SMA fetus, NND, or PPH.
Discussion The incidences of pregnancies complicated by liver cirrhosis, portal hypertension, or esophageal varices are on the rise and are associated with worse pregnancy outcomes. Previous studies have debated various outcomes and whether these pregnancies can be successful. The strength of our study is that it is consists of contemporary data extracted from a reliable birth registry database. Thus, we were able to make outcome observations that are relevant to recent improvements in both obstetrical care and the care of individuals with liver disease. We found that these pregnancies can be successful. Nonetheless, these high-risk patients remain prone to lifethreatening sequelae. We rarely observed cases of pregnancies complicated by liver cirrhosis with a concomitant diagnosis of portal hypertension or esophageal varices. Our findings for our cirrhosis-only pregnancy cohort are consistent with one
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
4
A. Puljic et al.
previously published population-based study [8]. However, our study goes a step further by additionally assessing the pregnancy outcomes for women who have concomitant portal hypertension and esophageal varices, which are the natural sequelae of cirrhosis. We observed worse outcomes in our pregnancy cohort with cirrhosis, the strongest association being PTD followed by LBW, PET, cesarean section in multiparous women, and NND. Women with portal hypertension were more likely to deliver preterm, have a LBW infant, with an increased risk for NND and PPH. Pregnancies with liver cirrhosis, portal hypertension, or non-bleeding esophageal varices are consistently associated with an increased risk for cesarean section. However, it is unclear whether this could be due in part to provider comfort in these unique delivery scenarios. Further studies are needed to evaluate whether vaginal, assisted, or cesarean delivery is the optimal delivery mode in these conditions. An increase in PTD was observed for both liver cirrhosis and portal hypertension in pregnancy, which raises further questions regarding whether these deliveries were induced or not. These preterm deliveries may also be associated with our finding of an association of NND with both cirrhosis and portal hypertension. PPH was only observed in our portal hypertension cohort, which may be associated in part with increased portal vasculature pressure and potentially a coagulopathy. This may require pre-delivery planning for management of hemorrhage in these patients. While this study provides a large cohort in order to examine obstetric and neonatal outcomes of pregnancies complicated by severe hepatic disease, there are limitations. The conditions assessed may manifest with varying severity and sequelae in individual patients. Also, in order to obtain such a large cohort, administrative data were utilized and diagnoses were confirmed with ICD-9 codes. In general, women with such codes likely have the diseases noted, but under-coding can be a problem. Such misclassification would only lead to an underestimation of the effect difference, and because these are rare diagnoses, this underestimation should be small. Additionally, we did not have specific metrics assessing the level of disease such as length of time of disease, liver function tests, etc. Finally, even with this large sample, the study was underpowered for several outcomes and unable to stratify the conditions any further. Despite these limitations, our study denotes a number of obstetric and neonatal complications related to pregnancy with liver cirrhosis, portal hypertension, or esophageal varices. Women with these conditions can be counseled
J Matern Fetal Neonatal Med, Early Online: 1–4
both prior to pregnancy as well as during pregnancy regarding these risks. In order to reduce these risks, perhaps closer antenatal monitoring of pregnancies with these conditions is warranted. In addition, careful planning of delivery mode should take place while counseling the patient during pregnancy. As women with severe hepatic disease are maintained with their chronic disease longer due to improvements in medical care, we will see an increasing number in our practices. Careful counseling and optimization of their care are key in achieving the best outcomes possible for these truly high-risk pregnancies.
Declaration of interest The authors report no declarations of interest.
References 1. Aggarwal N, Sawnhey H, Suril V, et al. Pregnancy and Cirrhosis of the Liver. Aust NZ J Obstet Gynaecol 1999;39:503–6. 2. Lee NM, Brady CW. Liver disease in pregnancy. World J Gastroenterol 2009;15:897–906. 3. Russell MA, Craigo SD. Cirrhosis and portal hypertension in pregnancy. Semin Perinatol 1998;22:156–65. 4. Schreyer P, Caspi E, El-Hindi JM, Eshchar J. Cirrhosis-pregnancy and delivery: a review. Obstet Gynecol Surv 1982;37:304–12. 5. Tan J, Surti B, Saab S. Pregnancy and cirrhosis. Liver Transpl 2008;14:1081–91. 6. Cerqui AJ, Haran M, Brodribb R. Implications of liver cirrhosis in pregnancy. Aust NZ J Obstet Gynaecol 1998;38:93–5. 7. Kim WR, Brown Jr RS, Terrault NA, El-Serag H. Burden of liver disease in the United States: summary of a workshop. Hepatology 2002;36:227–42. 8. Shaheen AA, Myers RP. The outcomes of pregnancy in patients with cirrhosis: a population-based study. Liver Int 2010;30:275–83. 9. Borhanmanesh F, Haghighi P. Pregnancy in patients with cirrhosis of the liver. Obstet Gynecol 1970;36:315–24. 10. Cheng YS. Pregnancy in liver cirrhosis and/or portal hypertension. Am J Obstet Gynecol 1977;128:812–22. 11. Pajor A, Lehoczky D. Pregnancy in liver cirrhosis. Assessment of maternal and fetal risks in eleven patients and review of the management. Gynecol Obstet Invest 1994;38:45–50. 12. Rasheed SM, Abdel Monem AM, Abd Ellah AH, Abdel Fattah MS. Prognosis and determinants of pregnancy outcome among patients with post-hepatitis liver cirrhosis. Int J Gynaecol Obstet 2013;121: 247–51. 13. Restaino A, Campobasso C, D’Aloya A, et al. Cirrhosis and pregnancy: a case report and review of the literature. Clin Exp Obstet Gynecol 1996;23:240–7. 14. Peitsidou A, Peitsidis P, Michopoulos S, et al. Exacerbation of liver cirrhosis in pregnancy: a complex emerging clinical situation. Arch Gynecol Obstet 2009;279:911–13. 15. California Department of Health Services. Birth cohort public use file, 2005–2009. Sacramento, CA: California Department of Health Services; 2013.
