Clinical features and outcome of acute hepatitis B in pregnancy

It is well-known that acute hepatitis B virus (HBV) infection in adults is usually self-limiting, of which only 5%-10% develop to chronic HBV infection, and about 1% progress to acute liver failure [1]. However, in certain special populations, such as pregnant women, the clinical course of acute hepatitis B (AHB) is still largely unclear. Although a few studies have investigated the impact of pregnancy on AHB, few of them included matched controls [2–8]. Age, sex, and host immunity status are important factors influencing the clinical manifestation and outcome of acute HBV infection [9]. Clinical features and HBV serological outcome in pregnant AHB patients, such as hepatitis B surface antigen (HBsAg) loss and seroconversion, have not fully been elucidated.

The clinical presentation and natural history of HBV infection is mediated through complex interactions between the virus and the host immune response [10]. In acute HBV infection, both host innate immunity and HBV specific adaptive immunity play important roles in viral clearance [10, 11]. Impaired immunity due to various causes may influence the course of AHB [9, 12–14]. During pregnancy, the maternal immune system is altered to tolerate the genetically different fetus, and hormonal factors may also play a significant role in altering immune regulation or viral replication [15]. An early study by Mohite et al. reported that the T cell response to HBsAg was weaker in pregnant women than in adults males and non-pregnant females [16]. So, immunological response to acute HBV infection, and consequently the clinical course of AHB in pregnant women may differ from those in the general population [16].

The aim of this study was to investigate the clinical features and outcome of AHB in pregnancy.

Among a total of 618 inpatients diagnosed with AHB during the study period, 22 pregnant patients were enrolled into the pregnant group. The mean ± SD and median (min-max) age of the AHB in pregnant patients was 22.7 ± 1.9 years and 22 (21–28) years, respectively. 87 female non-pregnant patients between 21 and 28 years of age were enrolled in the control group.

HBV virological and serological outcome of AHB in pregnancy

The HBV virological and serological biochemical data of pregnant and non-pregnant AHB patients at admission, discharge, and final follow-up are summarized in Table 4. At admission, the median HBV DNA level was similar between the two groups (P = 0.69), and the HBV DNA > 1000 copies/mL rate was also similar between the two groups. The HBsAg positive (>0.05 IU/mL) rate was similar between the two groups. The HBsAg > 250 IU/mL rate and HBeAg positive rate were significantly higher in pregnant than in non-pregnant patients.

Variables	Non-pregnant AHB patients N = 87	Pregnant AHB patients N = 22	P
At admission
Days from disease onset to examination	11.8 ± 7.3	11.3 ± 4.1	0.70
HBsAg positive	6 (93.1%)	22 (100%)	0.21
HBsAg > 250 IU/mL	45 (51.7%)	17 (77.3%)	0.03
HBsAg seroconversion	3 (3.44%)	0 (0%)	0.38
HBeAg positive	52 (59.8%)	19 (86.4%)	0.02
HBeAg seroconversion	33 (37.9%)	2 (9.1%)	0.01
HBV DNA (log10 copies/mL)	3.49 (3.0-7.14)	3.49 (3.0-4.93)	0.69
HBV DNA >1000 copies/mL	55 (63.2%)	17 (77.3%)	0.21
At discharge
Days from disease onset to examination	38.6 ± 11.9	34.2 ± 7.1	0.11
HBsAg positive	54 (62.1%)	19 (86.4%)	0.03
HBsAg > 250 IU/mL	19 (21.8%)	9 (40.9%)	0.07
HBsAg seroconversion	11 (12.6%)	0 (0%)	0.08
HBeAg positive	8 (9.2%)	3 (13.6%)	0.54
HBeAg seroconversion	72 (82.8%)	15 (68.2%)	0.13
HBV DNA >1000 copies/mL	11 (12.6%)	1 (4.5%)	0.28
Final follow-up
Months from disease onset to examination	7.0 ± 0.8 (5.2-8.3)	6.9 ± 0.7 (5.8-8.2)	0.54
HBsAg positive	4 (4.6%)	4 (18.2%)	0.03
HBsAg > 250 IU/mL	1 (1.1%)	1 (4.5%)	0.29
HBsAg seroconversion	59 (67.8%)	9 (40.9%)	0.02
HBeAg positive	1 (1.1%)	1 (4.5%)	0.29
HBeAg seroconversion	80 (91.9%)	19 (86.3%)	0.42
HBV DNA > 1000 copies/mL	1 (1.1%)	1 (4.5%)	0.29

NOTE. Data are expressed as mean ± SD,median (min-max), or n (%) of patients; HBV, hepatitis B virus; AHB, acute hepatitis B; HBsAg,hepatitis B surface antigen; HBeAg, hepatitis e antigen.

At discharge, the HBsAg positive rate was significantly higher in pregnant patients than in non-pregnant patients, whereas virological and other serological parameters were similar between the two groups.At final follow-up, the HBsAg positive rate was significantly higher in pregnant patients than in non-pregnant patients. 18.2% pregnant patients and 4.6% non-pregnant patients developed chronic HBV infection. For pregnant patients, the relative risk (95% confidence interval) of HBsAg positive at the end of follow-up was 4.6 (1.1-20.2). The Kaplan–Meier estimates showed the median (95% confidence interval) days of HBsAg seroclearance form disease onset in pregnant and non-pregnant patients were 145.0 (110.5-179.5) and 80.0 (62.6-97.4), respectively (Figure 1).

Just as it is generally accepted that pregnancy does not alter clinical recovery and mortality of AHB [25], the present study also did not find any significance difference on AHB mortality and occurrence of flulminant hepatitis between pregnant and non-pregnant patients. Clinical recovery, defined as relief of signs and symptoms, and normalization of ALT, was also similar between pregnant and non-pregnant females. However, the present study did reveal certain clinical and serological features of AHB during pregnancy.

Most of the patients in the present study were symptomatic (83.8%), whose signs and symptoms included prodromal fever, fatigue, and loss of appetite. Fever was significantly less common (0% vs. 20.7%, P = 0.02) in pregnant than in non-pregnant AHB patients, whereas the presence of other symptoms, such as fatigue and loss of appetite, were similar between the two groups. This finding may be partly explained by the “febrile hyporesponsiveness” phenomenon during pregnancy [26]. The febrile response is a significant contributor to the pathogenesis, clinical presentation and outcome of many illnesses and diseases [21]. During pregnancy, febrile response is suppressed by the following potential mechanisms: suppression of pro-inflammatory processes, augmentation of anti-inflammatory processes, and alterations in steroid hormones [21]. The suppressed febrile response during pregnancy can lead to a compromised ability to fight infection while it could be beneficial to the mother and the fetus. Whether the altered immunity during pregnancy affects HBV DNA replication and eradication is still inconclusive.

Several studies investigated the effect of pregnancy on serum HBV DNA levels of pregnant women with chronic HBV infection [27–29]. The study by Soderstrom et al. observed increased serum HBV DNA levels in late pregnancy or early post-partum [28]. But, Nguyen et al. reported that the levels of HBV DNA and ALT during pregnancy were highly variable [27]. Little information is available regarding the virological and serological changes during acute HBV infection in pregnant patients.

In the present study, the median HBV DNA levels and HBV DNA undetectable rate were similar between pregnant and non-pregnant patients. HBV DNA and ALT levels were highly variable in both groups.

We noted significant lower ALT levels in pregnant women. A study on acute liver failure during pregnancy by Bhatia et al. also observed significantly lower ALT levels in pregnant women than in men and non-pregnant women [30]. Elevated serum ALT levels or symptoms reflect the T-cell-mediated HBV-specific immune response [10]. The mild symptoms and relatively low ALT levels in pregnant AHB patients indicate that HBV specific immunity in pregnancy may differ from those in the general population.

Interestingly, we noted a significantly higher HBsAg levels and lower HBsAg severconversion rate in pregnant patients than in non-pregnant patients. At admission and discharge, the percent of HBsAg > 250 IU/mL in pregnant patients was higher than that in non-pregnant patients. Moreover, at final follow-up, about 7 months from disease onset, 18.2% of the pregnant patients were still HBsAg positive, a chronic HBsAg carrier state, whereas only 4.6% of the non-pregnant patients were HBsAg positive. Also, at final follow-up, the HBsAg seroconversion rate was significantly lower in pregnant than in non-pregnant patients (40.9% vs. 67.8%, P = 0.02). HBsAg levels reflect the host immune control state and cccDNA levels in hepatocytes [31, 32]. T-cells are markedly reduced during early pregnancy up to the 20th week of gestation to reduce the level of immunity [15]. Increased hormone levels during pregnancy, including progesterone, estrogen and human chorionic gonadotropin, have been shown to have a clear suppressive effect on cell-mediated immunity [26]. This status may indicate that the altered immunity status may influence the eradication of HBV infected cells and the production of HBsAg antibodies. This notion is supported by the study of Mohite et al. which reported that the T cell response to HBsAg in pregnant women with AHB was absent throughout the illness, whereas that in adult males and non-pregnant females increased gradually after AHB onset [16].

We also noted a significantly more frequent and severe jaundice in pregnant women. At admission, hyperbilirubinemia was seen in all pregnant patients, and the total bilirubin level was significantly higher in pregnant than in non-pregnant patients. In acute hepatitis, serum total bilirubin level usually correlates positively with severity of liver injury [33]. However, we observed an inconsistence of total bilirubin level with peak ALT levels, a sensitive indicator of liver injury, in pregnant AHB patients. This indicate hyperbilirubinemia in pregnant AHB patients might be caused not only by intrahepatic inflammation, but possibly also by other factors influencing bilirubin clearance. Serum albumin and UDP-glucuronosyltransferase(UGP) are involved in bilirubin metabolism [34, 35]. Significant lower serum albumin level, a protein that transports bilirubin, were observed in our and other studies [34, 36]. An animal study revealed decreased expression of UGP1A1 during pregnancy in rat liver [37]. Recent genome-wide association studies indicated the UGP1A1 were important genes in controlling serum bilirubin [35]. These results might partially explained the inconsistence between bilirubin concentration and ALT level in pregnant AHB patients.

The serum GGT levels were also significantly lower in pregnant patients. The serum GGT activity are lower in normal pregnant women than in non-pregnant women [36]. Hepatic synthesis of GGT could be inhibited by hormone secretion during pregnancy [36].

The strong point of the present study is the use of controls matched in age, sex, alcohol consumption, and other factors which may influence the course of AHB, with pregnant patients. Days from disease onset were also comparable between the two groups. To our knowledge, this is the first study that investigated the course of AHB during pregnancy with well matched controls.

There is a concern that 7 month follow up is too short. However, according to current acute hepatitis B investigative guidelines, those still HBV DNA-positive or HBsAg-positive are considered confirmed chronic carriers, and should be counseled accordingly [17]. In fact, we followed the patients for at least 1 year. For the pregnant AHB patients, none of them used antiviral therapy during follow-up, 4 patients were still HBsAg positive after at least 12 month follow-up. However, among the 618 AHB patients, for those without pregnancy, a significant portion of those still HBV DNA positive after 6 months of infection were treated with interferon or nucleotide analogues. Also, this was a hospital-based study, most patients in this study were symptomatic and relatively severe. The clinical features and outcome of asymptomatic AHB in pregnancy is not clear. To answer these questions, a large scaled population based study is needed.

Although the clinical recovery of AHB differs little in pregnant and non-pregnant patients, the HBsAg loss and seroconversion is delayed and lower in pregnant patients. Pregnancy might be a possible risk of chronicity following acute HBV infection. Further clinical and basic studies are necessary to confirm the findings and to elucidate the underlying mechanisms.