From the Departments of *Obstetrics and Gynecology and †Medicine, University of Washington, and the ‡Battelle Centers for Public Health Research and Evaluation, Seattle, Washington; and the §Department of Pediatrics, University of Alabama, Birmingham, Alabama
Correspondence: Carolyn Gardella, MD, MPH, Department of Obstetrics and Gynecology, Seattle, Washington. E-mail: email@example.com.
Received for publication September 19, 2007, and accepted September 21, 2007.
Neonatal herpes is the most devastating consequence of genital herpes. Even with prompt diagnosis and initiation of antiviral therapy, case fatality for disseminated disease remains 50%. Encephalitis is a component of over 50% of cases, and most survivors suffer permanent neurodevelopmental sequelae.1 Often symptoms start several days after birth, typically after hospital discharge, making rapid diagnosis and initiation of therapy difficult, especially given the lack of information regarding this disease in the general medical community, the lack of information new parents are given about the disease, and the fact that most parents are unaware that they have genital herpes. The tragedy of neonatal herpes is compounded by its impact on an otherwise healthy infant, that most cases result from a sexually transmitted disease (STD) posing an overwhelming personal strain on the parents, and the fact that neonatal herpes is preventable.
Contact with HSV in the maternal genital tract at the time of labor accounts for 90% of cases of neonatal herpes.2 HSV-1 accounts for a modest proportion of neonatal herpes. Although a few of these are acquired in the postnatal period, such as from a parent with orolabial infection, most, like HSV-2, are transmitted from maternal genital infection at the time of birth. Most HSV shedding in the genital tract is not associated with signs or symptoms so visual inspection will fail to detect most women who are infectious at the time of labor. As in nonpregnant populations, most genital HSV infections among pregnant women are subclinical—either entirely asymptomatic or causing mild or nonspecific symptoms that do not lead to clinical diagnosis—so that most cases of maternal HSV infection, whether new or established, go undetected.3
Women who are newly infected with HSV during the third trimester of pregnancy are at greatest risk for perinatal infection.4 Thus, a woman who does not have serologic evidence of HSV infection at the beginning of pregnancy is more likely to transmit HSV to her neonate than one with established infection if she acquires genital HSV infection from her partner. Up to 50% of neonates born to women with newly acquired genital HSV who are shedding virus at delivery develop neonatal herpes, compared with 1% of neonates born to women with established HSV who are shedding virus at delivery.4 However, the number of women with established infection is much higher than the number who acquire HSV in pregnancy. Therefore, at a population level, it is uncertain whether new or established maternal infection contributes more to total morbidity from neonatal herpes.
Two reports in this issue expand our understandings of the epidemiology of neonatal herpes. Morris and colleagues5 report that the incidence of neonatal herpes in California derived from hospital discharge and mortality data was unchanged between 1995 and 2003, despite an increase in cesarean delivery for herpes and a decrease of herpes complications in labor. The time period of the study included 1999–2003 when antiviral therapy during the last month of pregnancy for women with symptomatic recurrent genital herpes was endorsed by American College of Obstetricians and Gynecologists (ACOG).6 The obvious inference from these data is that current strategies to prevent neonatal herpes are not working. The report by Dinh and colleagues7 from Centers for Disease Control and Prevention (CDC) describes the infrequency of neonatal herpes case reporting in the United States and lack of national epidemiologic analysis.
For the past 20 years, neonatal herpes prevention strategies have remained essentially unchanged despite advances in the understanding of the pathogenesis of maternal-to-child transmission and the prevention of person-to-person transmission of genital herpes and the availability of reliable serologic tests for HSV-1 and HSV-2.
National guidelines for neonatal herpes prevention from ACOG and CDC focus on antiviral suppression in the last month of pregnancy for women with symptomatic infection and visual inspection of the maternal genital tract for herpetic lesions at the time of labor with cesarean delivery for women with lesions.8,9 The obvious problem with these strategies is that they do not account for the presence of HSV in the genital tract of asymptomatic women, the most frequent scenario that causes perinatal transmission, especially in women with newly acquired infection whose neonates are at greatest risk of infection. Despite now nearly universal availability of the necessary serologic tests, there are no authoritative recommendations to routinely identify women at risk of acquiring HSV during pregnancy by serologic testing either alone or in combination with partner testing to identify serodiscordant couples. ACOG does not recommend routine HSV testing in pregnancy, citing that cost-benefit analyses to date found discrepant results depending on the assumptions entered into the models.8
The incidence of neonatal herpes in the United States is poorly defined and varies depending on the data source. Morris and colleagues5 report an incidence of neonatal herpes of 12.1 per 100,000 live births per year in California. However, hospital discharge databases are unlikely to provide an accurate estimate of the true burden of disease because of underreporting. For example, compared with the data from Morris and colleagues, review of data from 30 managed care plans in California between 1997 and 2002 suggested a rate of neonatal herpes of 60 cases per 100,000 live births,10 and data from a prospective study of over 50,000 women in Washington state observed a neonatal HSV rate of 30.8 per 100,000.4 Without uniform data collection methods and definitions, it is difficult to determine whether there are true differences in incidence rates among populations. Despite limitations related to underreporting and imprecise ICD-9 coding, hospital discharge and mortality data are valuable to track changes in disease incidence over time and may be useful to assess new prevention strategies, assuming that data collection remains uniform.
Neonatal herpes reporting through current surveillance systems does not provide an accurate assessment of disease burden either. Dinh and colleagues7 report that neonatal herpes was reportable in only 9 states, 5 of which had no standard surveillance case definition and 4 of which lacked specific neonatal herpes case report forms. Perhaps, reflecting these flaws, few cases were reported, with an implausibly low annual rate of 4 cases per 100,000 live births. The authors provide potential alternative methods for assessing the burden of neonatal herpes, including refining current surveillance systems, emphasizing the need for a standard case definition and laboratory confirmation, and creating sentinel sites for reporting and surveillance. However, surveillance through routine case reporting clearly can work if systematically implemented by motivated public health institutions. Notably, New York recently mandated neonatal herpes reporting, and reported a rate of 17.7 cases per 100,000 live births in 2006, using mainly reflexive reporting from laboratories confirmed by case investigation, showing that case surveillance based on reporting is feasible.11
Extrapolating from the available data, if the rate calculated for California by Morris et al.5 applies to the approximately 4 million annual births in the United States, the national incidence approximates 480 cases annually; the high estimate, extrapolated from Whitley et al.,10 corresponds to 2400 cases per year. By contrast, the recent annual case rates of other serious perinatal infections for which systematic prevention programs are employed were 329 cases of congenital syphilis (most of which are asymptomatic cases by surveillance definition with little morbidity),12 67 cases of HIV infection,13 and <10 cases of congenital rubella.14 Two reviews15,16 recommended national surveillance based on routine neonatal herpes reporting predicated on the potential for prevention with readily available biomedical tools and increased awareness among health care providers and the public health establishment.
In March 2007, the Council of State and Territorial Epidemiologists (CSTE) and CDC convened an expert panel of representatives of ACOG, the American Academy of Pediatrics, the National Coalition of STD Directors, the Association of Public Health Laboratories, selected state and local health departments, and clinicians with expertise in pediatric infectious diseases, neonatology, and obstetrics and gynecology to make recommendations regarding neonatal herpes surveillance. Almost unanimously, the panel concluded that neonatal herpes should be nationally reportable using a standard case definition. The goals for surveillance are to describe the incidence and risk factors for neonatal herpes, to recognize and respond to emerging disease patterns, to identify missed opportunities for prevention and early treatment of affected neonates, and to use that knowledge to educate providers and the public about neonatal herpes and to establish a baseline measure of disease burden from which to monitor the effectiveness of prevention strategies.17
Despite the panel’s recommendation, the CSTE membership voted not to make neonatal herpes nationally reportable at this time, but moved toward helping states to improve surveillance by using a standard case definition and encouraging states to voluntarily make neonatal herpes reportable. Sentinel surveillance was proposed as an alternative to meet the national prevention needs. Indeed, there are advantages to sentinel surveillance, notably the potential for higher quality data than passive case reporting, and the ability to focus resources in willing states. However, devoting sufficient national resources (e.g., from CDC) to such an effort may be problematic, and many jurisdictions may not have the resources or motivation to participate. Further, from the perspective of individual states, data from other geographic areas, no matter how compelling, may not penetrate the barrier of denial harbored by many providers, some public health agencies, and most of the public.15,16
In the rich medical and social history of STDs, at times advocates for improved prevention have stressed the tragedy of the innocent victim. For the most part, that emphasis was misguided because it implied that the “not innocent”—sexually active persons who place themselves at risk—were less deserving of prevention efforts. However, if a case can be made for emphasizing the truly innocent victim, it lies in the realm of perinatal transmission of STDs, of which neonatal herpes is now by far the dominant problem in the United States. We believe that lack of nationwide incidence data is critically hindering the development effective prevention strategies. We call on the STD prevention community, with our colleagues in obstetrics and gynecology and pediatrics, to continue to press for effective nationwide prevention of neonatal herpes, and we call on CSTE and CDC to carefully reconsider the central role of national reporting and surveillance in that goal.
1. Kimberlin D, Lin C-Y, Jacobs R, et al. Natural history of neonatal herpes simplex virus infections in the acyclovir era. Pediatrics 2001;108:223–229.
2. Kimberlin DW. Neonatal herpes simplex infection. Clin Microbiol Rev 2004; 17:1–13.
3. Brown ZA, Selke SA, Zeh J, et al. Acquisition of herpes simplex virus during pregnancy. N Engl J Med 1997; 337:509–515.
4. Brown ZA, Wald A, Morrow RA, et al. Effect of serologic status and cesarean delivery on transmission rates of herpes simplex virus from mother to infant. JAMA 2003; 289:203–209.
5. Morris S, Bauer H, Samuel MC, Gallagher D, et al. Neonatal herpes morbidity and mortality in California. Sex Transm Dis 2007; 35:14–18.
6. ACOG Practice Bulletin. Management of herpes in pregnancy. Number 8 October 1999. Clinical management guidelines for obstetrician-gynecologists. Int J Gynaecol Obstet 2000; 68:165–173.
7. Dinh T-H, Dunne E, Markowitz LE, Weinstock H, Berman S. Assessing Neonatal Herpes Reporting in the United States, 2000–2005. Sex Transm Dis 2007; 35:19–21.
8. ACOG Practice Bulletin. Clinical management guidelines for obstetrician-gynecologists. No. 82 June 2007. Management of herpes in pregnancy. Obstet Gynecol 2007; 109:1489–1498.
9. Workowski KA, Berman SM. Sexually transmitted diseases treatment guidelines, 2006. MMWR Recomm Rep 2006; 55(RR-11):1–94.
10. Whitley R, Davis EA, Suppapanya N. Incidence of neonatal herpes simplex virus infections in a managed-care population. Sex Transm Dis 2007; 34:704–708.
11. Handel S, Schillinger J, Hennessy R, Cox A, et al. Population-based surveillance of neonatal herpes in New York City: Findings from the first year. Presented at: International Society for STD Research; 2007; Seattle, WA; 2007.
15. Handsfield HH, Waldo AB, Brown ZA, et al. Neonatal herpes should be a reportable disease. Sex Transm Dis 2005; 32:521–525.
16. Donoval BA, Passaro DJ, Klausner JD. The public health imperative for a neonatal herpes simplex virus infection surveillance system. Sex Transm Dis 2006; 33:170–174.