Chronic hepatitis B virus (HBV) infection is a continuing global public health problem, with the World Health Organization estimating that chronic carriers worldwide will number 400 million by year 2000.]1 Perinatal transmission has been recognized as the most important mode of infection in eastern Asia. Hepatitis B vaccine immunization is effective, but 5 to 10 percent of babies whose mothers were hepatitis B surface antigen (HBsAg) positive still become chronically infected with HBV although passive- active immunoprophylaxis was used. HBV nucleic acid has been detected in amniotic fluid samples, vaginal secretions from the HBsAg carrier mothers and in aspirate fluid from their neonatal stomaches.[2 ]HBV infection may occur by oral route during delivery. It has been suggested that cesarean section might eliminate the swallowing of infected maternal blood, secretions and amniotic fluid. Recently, it has also been reported that cesarean section could reduce microtransfusion from mother to fetus during delivery.[3 ]On this basis, it has been postulated that the incidence of immunoprophylaxis failure could be reduced through cesarean section. We assessed the efficacy of passive- active immunoprophylaxis on babies born by different delivery modes to clarify whether cesarean section is really effective in reducing immunoprophylaxis failure.
Between 1990 and 1998, all pregnant women who received regular antenatal examinations at ZhongShan Hospital (affiliated teaching hospital of Fudan University) were screened for HBsAg in the third trimester of pregnancy. If the test was positive, HBsAg was detected again and other HBV- markers including anti- HBs, hepatitis B e antigen and its antibody, and serum alanine transaminase (ALT) levels were examined simultaneously before labor. Their babies were inoculated with hepatitis B immunoglobulin (HBIG) 100 IU immediately after birth and then hepatitis B vaccine at 1, 2 and 7 months of age. Before 1997, plasma- derived vaccine (produced by Shanghai Institute of Biological Products following the method developed by Merck) 30 microgram per injection was used; subsequently a recombinant yeast- derived vaccine (produced by Peking Institute of Biological Products, with techniques supplied by Merck) 5 microgram per injection was used. Those babies were followed by a special clinic in our hospital to study the outcome of postexposure immunization. Sera from these babies were drawn from the femoral vein for ALT, HBsAg, and anti- HBs determinations at 1, 4, 7 and 12 months of age, and then followed up yearly. All subjects had been well informed, and consents were obtained.
Babies satisfying the following criteria were enrolled this study. First, their mothers had positive HBsAg at two occasions, without symptoms of hepatitis B and with a normal level of ALT. Second, the newborn's gestational age was greater than 37 and less than 42. Babies with an obvious abnormality, or birthweight less than 2500 grams, or Apgar scores less than 8 at 1 or 5 minutes after birth were excluded.
Maternal serum HBsAg screening was done by reverse indirect hemoagglutination test before 1995 and by enzyme immunoassay (EIA) thereafter. All the children's HBV- marker detection and mother's HBV- marker detection before labor were done by commercially available EIA kits. Serum ALT greater than 40 was considered abnormal. A case with HBsAg- positive at 12 months of age and persisting more than 6 months was defined as a chronic HBV carrier.
The differences between different delivery modes were tested with EPI INFO 6. 04 version. ANOV was used for measurement data with Gaussian distribution, Chi- square test or corrected Chi- square test or Fisher exact test for enumeration data. P <0. 05 was considered statistically significant.
301 neonates were enrolled in the study, including 144 born by normal spontaneous vaginal delivery (Vaginal group), 40 by obstetric forceps or vacuum extraction (Forceps group), and 117 by cesarean section (Cesarean group). The incidence of HBeAg- positive mothers or the baby's gender constitution was comparable among the three groups (Table 1) . 292 infants were followed to 7 months of age, and 265 to 12 months of age. The reasons for dropping out of the study were mainly: fear of puncture, long distance from the hospital or moved abroad.
Among the three groups, the positive rates of anti- HBs detected at 4, 7 and 12 months of age were similar (Table 2). There were no significant differences among the three groups.
Incidence of HBsAg positivity or chronic infection
No significant differences between the three groups of serum HBsAg positivity were found at any follow- up point (Table 3) . The incidence of chronic HBV infection at 1 year of age was 7. 3% of babies in the Vaginal group, 7. 7% in the Forceps group, and 6. 8% in the Cesarean group. There are no significant differences among the three groups(χ2=0. 04, P =0. 979). (Table 4)
HBV infection in early life often results in chronicity. The infection can be persistent, even life- long. There are no satisfactory therapies for these children. It has been estimated that 25% of them will die from HBV- related hepatocellular carcinoma or end- stage cirrhosis in the future. Mother- baby transmission is the most important mode of virus spread in eastern Asia, including China. Babies from HBV infected mothers may be infected in utero, perinatally and postnatally. It was believed that most infections in infants born to HBsAg- positive carrier mothers occurred perinatally or postnatally, before the HBV immunization program started. Hepatitis B vaccine is a hallmark in preventing the transmission of HBV. It has been demonstrated that universal vaccination also had decreased the rate of childhood hepatocellular carcinoma. Unfortunately, the protective efficacy of immunization in a baby whose mother was chronically infected with HBV was 80%-90%. Although with hepatitis B immunoglobulin combined at birth, still 5%-10% of babies become infected. It was found that HBsAg mutant virus could cause breakthrough infection in immunized babies, but this accounted for only a small proportion of immunoprophylaxis failures. How to effectively decrease immunoprophylaxis failure remains a problem.
During labor and delivery, infection with HBV may occur as a result of the microtransfusion of maternal blood into the baby's circulation across a tear in the placenta, by swallowing maternal blood, amniotic fluid, or vaginal secretions while passing through the birth canal, or by contamination of the abrasions during instrumental vaginal delivery. Because cesarean section could avoid having the baby aspirate the mother's infectious material and recent evidence suggests that cesarean section may decrease mother- baby microtransfusion, it has been advised by some clinicians that babies born to carrier mothers with high virus load should be delivered by cesarean section and then given passive- active immunization.[2,3]
The clinical efficacy of cesarean section in reducing mother- baby HBV transmission was unclear. Before hepatitis B immunization was implemented, Beasley and Stevens investigated a group of mothers with high HBsAg titer and found that cesarean section did not prevent their babies from becoming infected. After the hepatitis B vaccine was available, Lee et al observed a group of infants born to HBeAg- positive mothers with passive- active immunization, and found that cesarean section could reduce the incidence of HBV- DNA in the babies at birth and the infection rate within 6 months of age. But in Lee's study the incidence of positive HBsAg and chronic HBV infection at 1 year of age had not been followed- up, which is the most commonly used parameter to assess hepatitis B vaccine efficacy. In our research, we used the current recommended dose of HBIG: 100 IU per time. The dosage is two times that used in Lee's study and could neutralize more virulent strains. We didn't find any statistically significant differences in HBsAg positivity at 1, 4 and 7 months of age among the babies delivered by different delivery modes. A more important finding is that the incidence of chronic HBV infection in the infants from different groups is nearly the same, and there are no significant differences in abnormal serum ALT levels in babies born by different modes. This is concordant with Chen's results. Most recently, Alexander6 reported that the risk of transmission of HBV to the fetus after amniocentesis in women who are HBV carriers can be reduced by passive- active immunoprophylaxis. These results suggest that with the standard passive- active immunoprophylaxis, cesarean section does not further decrease immunization failure.
Increasing evidence suggests that immunoprophylaxis failure is due mostly to intrauterine transmission of HBV. In those intrauterine infected cases, prophylaxis with hepatitis B immunization after birth and cesarean section delivery are useless. To further reduce immunoprophylaxis failure, treatment of women before pregnancy or before labor may be useful. Mothers with chronic HBV infection shouldn't be routinely advised to have cesarean sections.
1. Malik AH, Lee WM. Hepatitis B therapy: the plot thickens. Hepatology 1999;30:579- 581.
2. Lee SD, Lo KJ, Tsai YT, et al. Role of caesarean section in prevention of mothe r- infant transmission of hepatitis B virus. Lancet 1988;2:833- 834.
3. Lin HH, Kao JH, Hsu HY, et al. Least microtransfusion from mother to fetus in e lective cesarean delivery. Obstet Gynecol 1996;87:244- 248.
4. Zhu QR, Lu Q, Xiong SD, et al. Hepatitis B virus S gene mutants in infants infe cted despite immunoprophylaxis. Chin Med J 2001;114:352- 354.
5. Chen WH, Yin CS, Chang YK, et al. Neonatal gastric aspirates as a predictor of perinatal hepatitis B virus infections. Int J Gynaecol Obstet 1998;60:15- 21.
6. Alexander JM, Ramus R, Jackson G, et al. Risk of hepatitis B transmission after amniocentesis in chronic hepatitis B carriers. Infect Dis Obstet Gynecol 1999; 7:283- 286.