Breastmilk Hepatitis A Virus RNA in Nursing Mothers with Acute Hepatitis A Virus Infection

Abstract

Breastmilk specimens from three women with acute hepatitis A virus (HAV) infection were studied. Anti-HAV immunoglobulin M and immunoglobulin G antibodies were detected in serum and breastmilk specimens of the three women. The three women also had serum HAV RNA. However, HAV RNA was detected only in two of the three breastmilk specimens. It is interesting that none of the three infants contracted clinical HAV infection. Furthermore, mothers with HAV infection should not be encouraged to discontinue breastfeeding.

Introduction

Breastfeeding is recommended as the exclusive nutrient source for feeding preterm and term infants for the first 6 months of life and also later when solid foods are added. Breastfeeding has multiple direct and long-term benefits to the infant's nutrition, host defense, and protection against various infectious diseases.¹ However, breastmilk has also been demonstrated as a route of transmission for various viruses, including viruses that infect the liver.²

While a consensus regarding the risk of transmission of hepatitis B and hepatitis C viruses via human milk exists, very little information regarding the presence of hepatitis A virus (HAV) in human milk is available.^2,3

During a 3-year study period (July 2007–2009), three women in either the lactation or the peripartum period were hospitalized at the Hadassah Medical Center in Jerusalem, Israel because of acute HAV infection. Tests for the presence of anti-HAV antibodies and HAV RNA in the breastmilk of these women were conducted.

Subjects and Methods

Serum samples and breastmilk samples were tested for quantitative anti-HAV immunoglobulin G (IgG) (AxSYM^® HAVAB 2.0, Abbott, Abbott Park, IL) and anti-HAV immunoglobulin M (IgM) (AxSYM HAVAB-M 2.0, Abbott).

RNA was extracted from 200 μL of serum or breastmilk with the QIAamp^® MinElute virus kit (QIAgen, Hilden, Germany) according to the manufacturer's instructions. First-strand cDNA was synthesized from 10 μL of RNA using Superscript™ III (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions as described previously.⁴ Nested polymerase chain reaction (PCR) was performed on the conventional region of the HAV genome VP1-2A junction as mentioned elsewhere.⁵ For PCR amplification we used Herculase^® (Stratagene, La Jolla, CA), a Pfu/Taq polymerase enzyme with high fidelity and proofreading qualities. PCR products of the positive samples were tested for sequence reaction (Hy-Lab, Rehovot, Israel). The sequence analysis was performed using the CLUSTAL W program, in comparison with other HAV sequences obtained from the National Center for Biotechnology Information database.

Each of the above-described analytic examinations from both the breastmilk and serum was conducted two times.

Results

Patients

The three women involved in this study, 19–22 years old, were of Arab origin, residing in East Jerusalem. All presented with symptomatic acute HAV infection diagnosed by positive anti-HAV IgM serology and by the absence of other etiologies for acute liver injury. In the first two women, the HAV infection occurred 3–4 months after delivery, during which time both women were breastfeeding. The third woman contracted HAV infection while in the third trimester of pregnancy (week 36 of gestation). Eight days after the onset of HAV symptoms she had a spontaneous vaginal delivery.

The three women were previously healthy and consumed no prescribed, over-the-counter, or natural medications. All three had elevated aminotransferase activities (in the range of 30–60 times of the upper normal limit for alanine aminotransferase) and increased bilirubin (in the range of 5–9 times the upper normal limit). The three women recovered from the HAV infection without complications.

Anti-HAV IgG and IgM antibodies were detected in both the serum and either in the breastmilk or colostrum of the three women. The concentration of anti-HAV IgG antibodies in these fluids varied among the three women with no correlation to the time elapsed between the first symptoms of the HAV infection and the collection of the tested fluid. All three women were viremic during the acute stage of the HAV infection. In the two of the three women HAV RNA was also detected within the breastmilk (first patient) and within the colostrum (third patient). The second patient had the lowest titer of breastmilk anti-HAV IgG antibodies and had no breastmilk HAV RNA (Table 1).

Table 1.

Clinical and Laboratory Data of the Women with Acute Hepatitis A Virus Infection

Patient's number	Time period of HAV diagnosis	Specimen tested	Specimen collection time ^a	Anti-HAV IgM	Anti-HAV IgG (quantitative: mIU/mL)	HAV RNA
1	3 months postpartum	Serum	4	Positive	5,073	Positive
		Breastmilk	4	Positive	553	Positive
2	4 months postpartum	Serum	8	Positive	1,852	Positive
		Breastmilk	8	Positive	157	Negative
3	8 days prepartum	Serum	2	Positive	205	Positive
		Colostrum	10^b	Positive	1,330	Positive

The time period (in days) between the first symptoms of the hepatitis A virus (HAV) infection and the collection of either the serum or breastmilk.

Breastmilk was collected in this patient 10 days after the onset of HAV infection and 2 days after delivery.

IgG, immunoglobulin G; IgM, immunoglobulin M.

Nucleic acid sequence analyses of the PCR products that were detected in all of the serum and the breastmilk and colostrum samples were all compatible with genotype 1b of HAV. Nucleic acid sequences of the HAV viral particles that were isolated from both the serum and breastmilk/colostrum in each of the two women were 100% identical.

An epidemiologic investigation revealed that none of the three infants (two males and one female) contracted clinical HAV infection.

Discussion

HAV infection during pregnancy is an infrequent clinical event.⁶ Moreover, worldwide transmission of HAV from the infected mother to her child has been rarely reported.^3,6 Perinatal mother-to-child transmission of any infectious particles may occur via several routes: Parenteral (intrauterine, before or at birth), by the ingestion of infected breastmilk, or by the fecal–oral route (reviewed by Selander et al.³).

Intrauterine mother-to-child transmission of HAV RNA was not detected. Despite the presence of maternal viremia in women with acute HAV infection during the late stages of pregnancy and delivery, serum HAV RNA was not detected in the neonates born to these mothers.⁶

All of the three women presented in this study were viremic. In the two of these patients HAV RNA was also detected in either the breastmilk or colostrum that was collected early during the clinical course of their HAV infection. Another unsuccessful attempt to recover HAV RNA from the breastmilk, collected 3 days after delivery, was reported from Sweden.³

As far as we know, none of the infants born to the HAV-infected mothers described in this study and in our previous study⁶ suffered from clinical HAV infection. Because neither biochemical nor serological blood tests were performed in the infants born to these mothers, an asymptomatic HAV infection that occurred in the infants in the postexposure period cannot be ruled out. Alternatively, it may be speculated that breastmilk HAV RNA (as was detected in the two of the women in the present study) has a low infectivity potential. Several possible explanations may be pertinent to this speculation:

1. The shedding of HAV RNA in breastmilk is either very low or fluctuating.

2. Mothers who suffer from acute symptomatic HAV infection are either not willing or are too sick to breastfeed their infants.

3. As shown in our study, breastmilk contains anti-HAV (IgG and IgM) antibodies. In one of our patients the concentration of anti-HAV IgG antibodies in the colostrum even exceeded the concentration of the serum anti-HAV IgG. These antibodies and other anti-infective components in breastmilk may be capable of neutralizing the HAV, thus reducing its infectivity.^1,7

Mother-to child transmission of HAV RNA may also occur via the fecal–oral route. As HAV viremia and fecal excretion of HAV may persist for several weeks after the onset of symptomatic HAV infection, the infant of an HAV-infected mother should be protected. Although most newborn infants with HAV infection are asymptomatic and anicteric and no cases of fulminant HAV infection in newborns have been reported, protection is still needed because symptomatic cases have been reported.⁸ In order to protect the infant against this mode of transmission, administration of immunoglobulin had been suggested.^9,10 Alternatively, in the case of unavailability of immunoglobulin, administration of inactivated HAV vaccine, although not yet licensed for children<12 months of age, may be considered.^10–12 This should be done irrespective to the mode of feeding: breastfeeding or formula feeding.

In summary, although HAV RNA can be detected in breastmilk in lactating mothers with acute HAV infection, there is no indication that breastfeeding contributes to transmission of HAV from an infected mother to her child. Mothers with HAV infection should not be discouraged from continuing to breastfeed. Protection of the infant from HAV infection should be considered by the administration of either immunoglobulin or inactivated HAV vaccine.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Chirico

, Marzollo

, Cortinovis

et al. Anti-infective properties of human milk. J Nutr, 2008; 138:1801S–1806S.

Transmission of infectious agents via human milk. Pickering

. 2009 Red Book: Report of the Committee on Infectious Diseases, 28th. American Academy of Pediatrics: Elk Grove Village, IL, 2009; 119–124.

Selander

, Bläckberg

, Widell

et al. No evidence of intrauterine transmission of hepatitis A virus from a mother to a premature infant. Acta Paediatr, 2009; 98:1603–1606.

Rousseau

, Birditt

, McKay

et al. Large-scale amplification, cloning and sequencing of near full-length HIV-1 subtype C genomes. J Virol Methods, 2006; 136:118–125.

Bower

, Nainan

, Han

et al. Duration of viremia in hepatitis A virus infection. J Infect Dis, 2000; 182:12–17.

Elinav

, Ben-Dov

, Shapira

et al. Acute hepatitis A infection in pregnancy is associated with high rates of gestational complications and preterm labor. Gastroenterology, 2006; 130:1129–1134.

Zhang

, Rao

, Li

et al. Functional recombinant human anti-HAV antibody expressed in milk of transgenic mice. Transgenic Res, 2009; 18:445–453.

Linder

, Karetnyi

, Kuint

et al. Symptomatic hepatitis A virus infection during the first year of life. Pediatr Infect Dis J, 1995; 14:628–629.

Lawrence

, Lawrence

. Breast milk and infection. Clin Perinatol, 2004; 31:501–528.

10.

Hepatitis

. Pickering

. 2009 Red Book: Report of the Committee on Infectious Diseases, 28th. American Academy of Pediatrics: Elk Grove Village, IL, 2009; 329–337.

11.

Victor

, Monto

, Surdina

et al. Hepatitis A vaccine versus immune globulin for post exposure prophylaxis. N Engl J Med, 2007; 357:1685–1694.

12.

De Silvestri

, Zara

, Terulla

et al. Immunogenicity of hepatitis A-inactivated vaccine administered to seronegative infants, and serological follow-up 12 months after second dose. Acta Paediatr, 2006; 95:1582–1585.