4 Herpes simplex virus infection in pregnancy DOMINIC E. DWYER ANTHONY L. CUNNINGHAM
INTRODUCTION Herpes simplex virus types 1 and 2 (HSV-1 and HSV-2) can both cause genital infection and be transmitted to the neonate, although HSV-2 is the more important. Neonatal herpes and HSV encephalitis are the two most serious diseases caused by these viruses in immune competent patients. Hence the epidemiology of HSV infections in pregnant women and strategies for prevention of vertical transmission have been intensively studied despite the difficulties in investigating a disease occurring only once in every 2000-9000 live births. The epidemiology of HSV-2 has been complicated by the fact that asymptomatic virus shedding from the genital tract is common, patients may not recognize their recurrent genital herpetic lesions, and the virus is highly sexually transmissible (often via asymptomatic partners). New serological techniques for identifying patients who are infected asymptomatically with HSV-2 have greatly contributed to a better understanding of the epidemiology of sexual transmission of this virus. These tests are also being evaluated in strategies for prevention of vertical transmission.
Virology Morphologically, HSV-1 and HSV-2 have identical structures (Figure 1). The outer lipid envelope is studded with at least eight and probably ten different viral glycoproteins which interact with the host cell and are targets for protective antibody. An amorphous protein layer (the tegmentum) separates the envelope from the core. The icosahedral core consists of seven proteins arranged in 162 subunits or capsomeres which surround the DNA. The double-stranded DNA is 150 000 base pairs long and encodes at least 71 genes and 37 different structural proteins (including the 8 envelope glycoproteins). The overall DNA homology between HSV-1 and HSV-2 is only 40%, sufficient to make them different species. However, the sequence similarities of many viral proteins results in a similar host serological response to the two viruses (Roizman and Sears, 1990). This antigenic similarity has created difficulties in obtaining an antigen entirely specific for HSV-2, but HSV-2 glycoprotein G is highly specific, and has been used in Baillibre' s Clinical Obstetrics and GynaecoIogy--
Vol. 7, No. 1, March 1993 ISBN 0-7020-1708-6
75 Copyright © 1993, by Bailli~re Tindall All rights of reproduction in any form reserved
D. E. DWYER AND A. L. CUNNINGHAM
Tegmentat proteins !-
gB (involved in virus entry)
ed in virus entry ;)otent inducer of dizing antibody) )at~gl
gG sulpna[e on cell surface) (induces type-specific antibody to HSV-1 or HSV-2) Figure 1. Diagram of the structure of herpes simplex virus.
worldwide serosurveys of HSV-2 infection (Lee et al, 1986). Herpes simplex viruses replicate in the nucleus of the host cell. They acquire their envelope from host celt membranes by budding from the inner nuclear membrane into the endoplasmic reticulum, or sometimes from the outer cell membrane itself.
Pathogenesis and immunology These viruses infect a wide range of cells and are highly destructive, quickly shutting down host cell protein production (Roizman and Sears, 1990). Both neural and epidermal tissue are infected in vivo: infection of neural tissue may result in either latency or destruction of the neurones and surrounding glial and support cells. Infection of epidermal tissue is highly destructive and productive of virus, destroying epidermal cells and resulting in the development of characteristic vesicles containing high virus concentrations. Virus is shed from these lesions and transmitted during close contact either through the genital or oral mucous membranes or through small breaks in the skin. After initial replication within epithelial tissues, virus enters the dermal twigs of sensory neurones and is transported to the sensory ganglia, where it either causes acute destruction or becomes latent within neurones through mechanisms which are still uncertain. Both humoral and cellular immune
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
responses are responsible for restricting this acute ganglionitis and preventing further spread within the nervous system. Reactivation of latent HSV within the sensory ganglia may occur spontaneously or after stimuli such as trauma, ultraviolet light exposure of the skin and fever. After reactivation, HSV is transported distally down the sensory axon to be shed asymptomatically or to infect oral or genital mucocutaneous epithelial cells, resulting in epithelial cell destruction and clinical recrudescence, i.e. recurrent herpes (Stevens and Cook, 1971). Although both viruses may cause either initial or recurrent lesions in the orolabial or genital region, HSV-1 and HSV-2 demonstrate clear preference for the trigeminal and sacral ganglia respectively. As the viruses infect sensory dendritic twigs at a dermatomal level, lesions tend to recur at the site of initial inoculation or at least in the same dermatome. For example, anal lesions occurring in the first sacral dermatome may recur both perianally and on the foot. The defensive mechanisms against any viral infection may be classified as constitutive, non-specific and specific, with the specific defences subdivided into humoral and cellular immune responses. Constitutive immunity to HSV-1 in animal models is genetic (MHC related) but has been poorly described in man (Simmons, 1989). The role of non-specific or specific humoral and cellular immune response to HSV differs in primary and recurrent attacks. In primary episodes, non-specific antiviral responses including interferon, natural killer cells and activated macrophages, are the first defensive mechanisms that restrict local viral replication. They are followed by the production of specific antibody, which is important for restricting spread and terminating viraemia. Cell-mediated immune responses involving both the helper T-lymphocyte and cytotoxic Tlymphocyte subsets are also important in controlling a primary herpetic episode (Corey et al, 1978). Animal studies suggest that probably both antibody and cell-mediated immune responses contribute to control of virus spread in the ganglion and termination of acute infection (Simmons et al, 1992). However, in recurrent infections the cell-mediated immune response appears to be much more important than humoral immunity, particularly for the control of local spread. This is confirmed by the persistence and spread of herpetic lesions in patients with lymphoma, transplantation and AIDS. The major cells infiltrating recurrent herpetic lesions are CD4+ Tlymphocytes and macrophages. The CD4+ T-lymphocytes are activated and secrete antiviral cytokines (including interferon-~,) as well as exerting a cytotoxic effect against HSV-infected epithelial cells expressing HLA-DR on their surface. Macrophages activated by interferon-~ also have antiviral effects (Cunningham et al, 1985) (see Figures 2a and 2b). Herpes lesions evolve through stages of papule, vesicle, ulcer and crust before complete healing. The formation of vesicles and ulcers is probably a consequence of both the cytolytic effects of viral replication within epidermal keratinocytes and immune factors, e.g. cytotoxic T-lymphocytes, TNF-a and other cytokines (Cunningham and Noble, 1989). Vesicle fluid contains high titres of infectious virus (103--107 viral particles/ml) and high concentrations of antiviral cyt0kines, including interferons a, 13 and ~/ (Overall et al, 1981). Maximum infectivity of the lesion usually peaks within
D. E. DWYER AND A. L, CUNNINGHAM
(b) Figure 2. (a) Recurrent cutaneous herpes simplex: intraepidermal vesicle with multinucleated cells and dermal mononuclear infiltrate (stained with haematoxylin and eosin, × 120). On the right-sided insert, there are necrotic epidermal cells at the base of a vesicle with dark stained herpes simplex glycoprotein A/B antigen within nuclei and cytoplasm. There are peroxidasenegative inflammatory cells in the underlying dermis (immunoperoxidase, ×250). (b) HLA-DR antigen-positive epidermis and cells of subepidermal infiltrate. Langerhans cells are the more intensely staining cells in the epidermis. Note the dark 'ring' (cytoplasmic) staining of infiltrate cells (immunoperoxidase, × 190). On the right-sided insert are Leu 4 positive Tlymphocytes in intradermal mononuclear infiltrates: 85-90% of cells are dark staining (immunoperoxidase, × 120). From Cunningham et al (1985). Reprinted with permission from the Journal of Clinical Investigation.
H E R P E S S I M P L E X VIRUS I N F E C T I O N IN P R E G N A N C Y
2-3 days, after which virus titres decline, although virus may still be present in ulcerated mucosal lesions and the crust phase on skin. The median duration of shedding from these lesions is longer in primary than recurrent lesions (12 versus 4 days), with higher virus titres present in primary lesions (Corey et al, 1983). Therefore, the infectivity, clinical symptoms, likelihood of virus isolation and treatment response are greater in primary than recurrent lesions. Transmission
HSV-2 infection, whether symptomatic or asymptomatic, is one of the most transmissible of sexually transmitted diseases. The disease is spread sexually by exposure of the internal and external genitalia to lesions of a partner shedding HSV. Although primary and recurrent ulcerated lesions are the most infectious, transmission is common between couples where neither is aware of ever having had recurrent lesions (Kulhanjian et al, 1992). This may occur because small lesions are unrecognized by the patient (although once educated of their nature most people recognize lesions in the future) or because the lesions are atypical or microscopic (Langenberg et al, 1989; Koutsky et al, 1992). Asymptomatic shedding in HSV-2 seropositive women has been reported to occur from cervix and vulva as often as weekly, at very low virus concentrations, and usually lasting for only a day. Asymptomatic shedding was demonstrated in 80% of all women with recurrent genital herpes cultured daily for more than 50 days (Brock et al, 1990; Mertz et al, 1992). This relative ease of transmission is reflected by the very high prevalence rates of specific antibody to HSV-2 (Johnson et al, 1989). The proportion of individuals infected (predominantly asymptomatically) by HSV-2 varies throughout the world. The best comparisons have been obtained through seroprevalence studies of women attending public antenatal clinics, and seropositivity in this group ranges from 8% in Japan, 14% in Australia, 14-19% in Sweden and up to 40% in Afro-Americans from poor socioeconomic environments in the USA (Nahmias et al, 1990). As expected, HSV-2 seroprevalence is markedly increased in homosexual men (65%), prostitutes (57%) and patients attending sexually transmitted diseases clinics (40% in Sydney, Australia), and shows some correlation with other highly transmissible sexually transmitted diseases (Cunningham et al, in press). The transmissibility of the virus is further shown by the fact that sexual partners of HSV-2 seropositive patients are usually also HSV-2 seropositive, even in the absence of lesions. Transmission of HSV to the fetus or neonate occurs most commonly during parturition but also during the pregnancy or postnatal period. Transmission during parturition is determined by a number of important factors. Primary or first episode genital herpes may result in a large area of infection on the vulva and the cervix, infecting the newborn during labour or less commonly by ascending amnionitis after rupture of membranes (Corey and Spear, 1986b). Very high titres of virus are shed during primary infection leading to transmission in 33-50% of cases of vaginal delivery. Transmission
D. E. DWYER AND A. L. CUNNINGHAM
during parturition has also been well documented with symptomatic recurrent genital herpes (approximately 3 % of cases) and asymptomatic shedding (risk not accurately quantitated but probably less than 0.01%) (Brown et al, 1991). As recurrent genital herpes usually manifests as vulval lesions with cervical lesions being relatively uncommon (8% of cases), most transmission in a recurrence occurs during passage through the birth canal (Corey et al, 1983). The development of neonatal infection and disease is also determined by the level of maternal neutralizing antibody transmitted to the neonate (Prober et al, 1987), the duration of ruptured membranes ( > 6 hours) (Nahmias et al, 1983) and delivery instrumentation resulting in epithelial defects (e.g. fetal scalp electrodes) (Parvey and Chien, 1980). Transmission to the fetus during early pregnancy from primary genital or orolabial lesions is probably associated with viraemia and is extremely rare (Whitley et al, 1991b). Some of the 33-50% transmission resulting from primary infection near term is transplacental rather than from direct genital contact. In these cases, clinical signs in the neonate are noted early and disease may occur despite caesarean section. Postnatal infection is usually caused by nosocomial HSV-1 transmission immediately after birth from orolabial or other cutaneous lesions of the mother or nursing staff. Systemic or neurological infection is more common in premature infants, and is associated with T-lymphocyte immaturity (Whitley et al, 1991b). CLINICAL MANIFESTATIONS OF HERPES SIMPLEX VIRUS INFECTIONS HSV infections in the genital tract include primary symptomatic infection, primary asymptomatic infection, non-primary symptomatic first episode, recurrent symptomatic infection and recurrent asymptomatic infection. A primary infection is one in which the patient has no serological evidence of previous HSV-1 or HSV-2 infection. The clinical manifestations vary depending on whether the patient is experiencing a first (primary or nonprimary) or recurrent episode, with first episodes usually being more severe. Genital herpes can be acquired from both symptomatic and asymptomatic partners (Rooney et al, 1986). Primary HSV infection
Primary genital HSV infection can be symptomatic or asymptomatic. Of all patients presenting with a first episode of symptomatic infection, 60% will have a primary infection with up to 90% due to HSV-2. Symptomatic primary infections usually have more severe and long-lasting local symptoms and signs than recurrent disease, and are often accompanied by systemic features. The genital lesions are initially characteristically vesicular, may be widely spaced over the labia, perineum, buttocks and thighs, and are often accompanied by tender inguinal lymphadenopathy. New lesions may continue to appear for up to 2 weeks, and take up to 3 weeks to resolve. There is cervical involvement in over 80% of primary infections in women, with increased epithelial
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
friability, ulceration and associated discharge on clinical examination. Urethritis with dysuria occurs in 83% of women (and in 44% of men) in primary disease, with virus often isolated from the urethra. Urinary retention is common. Pharyngitis occurs in 11% of women with primary genital herpes compared to 1% in women with non-primary first episode or recurrent genital HSV. Primary genital infection due to HSV-2 in men is also severe, although they are less likely to have systemic symptoms, meningitis, dysuria, and urethral discharge (Corey et al, 1983; Corey and Spear, 1986a). Primary genital infection due to HSV-1 or HSV-2 in men and women cannot be distinguished clinically. The complications of primary genital infection, more likely to be seen in females, include aseptic meningitis, extragenital lesions (e.g. buttocks, groin, thighs, fingers, eyes) and superinfection with bacteria or fungi. Disseminated primary HSV infection is rare but has been described in the third trimester of pregnancy. It is associated with a maternal and fetal mortality of 40% and fetal infection may occur despite successful treatment of the mother with acyclovir (Peacock and Sarubbi, 1983; Berger et al, 1986).
Non-primary first episode of genital HSV-2 infection Non-primary first episode infections are those that occur in people with prior exposure to HSV-1 or HSV-2 as detected by the presence of specific antibodies. In general, non-primary first episode infections are less likely to have constitutional symptoms or complications, and have a shorter duration of disease than primary infection. The clinical and virological differences between primary symptomatic HSV-2 genital infection and non-primary first episode HSV-2 infection are listed in Table 1. Approximately 30--60% of patients with a first episode genital herpes infection will have pre-existing antibodies to HSV. This is most commonly due to previous oral HSV-1 Table 1o Significant differences between symptomatic primary HSV-2 infection and symptomatic non-primary first episode HSV-2 infection.
Patients with systemic symptoms (%) Patients with meningeal symptoms (%) Mean duration of local pain (days) Mean number of lesions Mean lesion area (mm 2) Patients with bilateral lesions (%) Patients forming new lesions during disease (%) Mean duration of viral shedding from lesions (days) Mean duration of lesions (days) Patients developing extragenital lesions (%) Patients shedding virus from cervix (%, ns) Adapted from Corey et al (1983).
Primary HSV-2 Infection (n = 189)
Non-primary HSV-2 Infection (n = 76)
62 26 11.8 15.5 517 82 75 11.4 18.6 18 88
16 1 8.7 9.5 158 45 45 6.8 15.5 8 65
DVCYER A N D A . L. C U N N I N G H A M
infection, but 10-30% will have HSV-2 antibodies, suggesting previous asymptomatic HSV-2 infection (Bernstein, 1991). One study used Western blot analysis to distinguish HSV-1 and HSV-2 specific antibodies in 24 patients with non-primary first episode genital herpes. They found 7 patients with HSV-1 specific antibodies, 11 with HSV-2 specific antibodies and 6 with both HSV-1 and HSV-2 antibodies (Bernstein et al, 1984). Presumably those patients with HSV-2 antibodies presented with their first clinically apparent recurrence of a previously asymptomatic HSV-2 infection, and may not have had recent sexual contact with a HSV-2 infected partner. Previous oral HSV-1 infection may protect against genital HSV-1 infections as non-primary first episode genital infection is nearly always caused by HSV-2. Also, patients with a history of oral HSV-1 and HSV-1 specific antibodies may have less severe HSV-2 disease and less frequent asymptomatic virus shedding (Reeves et al, 1981; Bernstein, 1991; Koelle et al, 1992). However, the presence of HSV-1 specific antibodies does not necessarily protect babies born to mothers with HSV-2 infection during pregnancy nor decrease the frequency of symptomatic HSV-2 recurrences (Brown et al, 1991; Koelle et al, 1992).
Recurrent genital herpes Recurrent genital herpes can be either symptomatic or asymptomatic (viral shedding). Although there is great variability, the clinical symptoms of recurrent genital herpes are usually less severe and of shorter duration than primary or non-primary first episode infection. Genital herpes infections are far more likely to recur symptomatically than orolabial infections, and they are more likely to occur after primary HSV-2 than primary HSV-1 genital infection. Symptomatic recurrences occur after almost 90% of primary symptomatic HSV-2 infections, compared to 55% of HSV-1 genital infections. The converse happens with orolabial herpes infection, where local recurrences occur after approximately 50% of HSV-1 and only 5% of HSV-2 oral infections (Corey et al, 1983; Lafferty et al, 1987). In one study, there was a monthly rate of 0.23 genital recurrences compared to 0.045 orolabial recurrences with both viruses, and HSV-2 was more likely to recur in the genital tract than HSV-1 (HSV-2 recurred at a monthly rate of 0.33 compared to 0.02 per month for HSV-1). In orolabial infections, HSV-1 recurrences occurred at a rate of 0.12 per month compared to a HSV-2 monthly rate of 0.001 (Lafferty et al, 1987). This is presumably related to the ability of the two types of HSV to establish latency in either sacral or trigeminal ganglia, and to differences in the frequency of reactivation and the host response to reactivation of each type of virus. As with primary infection, the symptoms in recurrences are generally more marked in women than in men. Pain is more severe and lasts longer (mean 5.9 days compared to 3.9 days in men), and dysuria is more frequent (27% versus 9%). The mean length of time to crusting of lesions in recurrent disease in women was 4.7 days (range 2-13 days), the mean duration to healing was 9.3 days (range 4-29 days) and the mean duration of viral shedding from lesions was 3.9 days (range 2-14 days) (Corey et al, 1983).
H E R P E S S I M P L E X VIRUS I N F E C T I O N IN P R E G N A N C Y
Lesions may not have a typical appearance, size or location; urogenital and anorectal lesions may be common (Koutsky et al, 1992). Asymptomatic recurrences are also common (particularly after HSV-2 infection), especially in the immediate period following primary infection. Asymptomatic cervical virus shedding was three times more frequent during the first 3 months after resolution of primary HSV-2 than during later time periods. This is reflected in a higher risk of neonatal transmission in the period at and immediately after a first episode infection (Brown et al, 1991). In contrast, symptomatic recurrences appear not to change in incidence with time (Koelle et al, 1992). Asymptomatic recurrences also occur after asymptomatic primary infection. Well over half of HSV-2 seropositive women presenting to family planning or sexually transmitted diseases clinics do not give a history of clinically apparent primary or recurrent genital HSV infections. A high rate of asymptomatic HSV-2 seropositivity was associated with age, sexual activity, other genital infections, multiple sexual partners, lower family income and race (Breinig et al, 1990; Koutsky et al, 1992). Studies in pregnant women have also suggested that asymptomatic viral shedding is not uncommon, occurring in 0.20--0.35% of women at the time of labour onset. About a third of these may have recently acquired genital HSV and only about one-fifth have a history of previous genital herpes (Prober et al, 1988; Brown et al, 1991). In 414 pregnant women with a known history of recurrent genital herpes, 4.1% had HSV-2 isolated from antepartum cultures despite the absence of lesions on clinical examination. In this group, asymptomatic viral shedding occurred with the same frequency at delivery whether or not any episodes of symptomatic recurrence were noted during pregnancy (Arvin et al, 1986). Previous HSV-1 infection decreases the frequency of subsequent asymptomatic vulval shedding of HSV-2, but it does not have a significant effect on the frequency of symptomatic disease (Koelle et al, 1992). HSV infection in pregnancy
Genital HSV infection in pregnant women presents with similar clinical features to infection in non-pregnant women, although detailed information on primary infection in pregnancy is limited (Corey et al, 1983). Symptomatic primary genital infection in the third trimester of pregnancy has been associated with increased frequency of spontaneous abortion and premature delivery (Nahmias et al, 1971). Recurrent genital infection has also been associated with spontaneous abortion, especially if cervicitis is present. There are reports of disseminated primary infection, including hepatitis, in the third trimester of pregnancy in otherwise healthy pregnant women. This is an indication for specific antiviral therapy, although successful treatment of disease in the mother may not necessarily protect the neonate (Peacock and Sarubbi, 1983). The significance of asymptomatic infection (primary or" recurrent) or recurrent disease confined to the vulva to successful completion of a pregnancy is uncertain (Corey et al, 1983). Asymptomatic infection is relatively common in pregnancy given the fact that 50-70% of neonates infected with HSV are born to mothers without history of
D W Y E R A N D A . L. C U N N I N G H A M
peripartum genital herpes infection (Whitley et al, 1980; Yeager and Arvin, 1984). This was confirmed by Brown et al (1991), who reported that HSV was isolated in 0.35% of asymptomatic women in early labour. Pregnant women with primary HSV-2 infection are more likely to infect their babies than those with a non-primary first episode. Brown et al (1987a) prospectively studied 29 pregnant women who presented for the first time with genital herpes. In the 15 women with primary infection, 6 had infants with serious perinatal morbidity, compared with none out of 14 with a non-primary first episode infection. The highest danger to infants was when primary infection occurred in the mother in the third trimester. The high neonatal risk is due to more frequent asymptomatic viral shedding after a symptomatic primary episode compared to a non-primary first episode. Virus shedding is usually from the cervix after a primary infection, whereas after a non-primary episode shedding was more likely from the labia (Brown et al, 1987b). Transmission to neonates from mothers with known recurrent genital herpes, even with virus shedding at delivery, is low and this lack of transmission is associated with high titres of maternal neutralizing antibodies to HSV-2 (Prober et al, 1987). However, in pregnant women with asymptomatic virus shedding at delivery, 35% have had a recent asymptomatic primary episode. Neonatal HSV is ten times more likely to occur in cases with an asymptomatic primary episode compared with asymptomatic reactivation of HSV, and confirms the protective value to the neonate of maternal HSV-2 antibodies. Overall, 33% of babies born to mothers with a first episode of genital infection develop neonatal herpes, compared to 3% of babies born to mothers with reactivations (Brown et al, 1991). Neonatal HSV infection
Although herpes simplex virus infection of the newborn is rare, infected infants have a high morbidity and mortality. The incidence of neonatal infection varies, with reported rates of up to 1 per 1500-2000 births in the USA (Whitley et al, 1991b). Neonatal infection is acquired perinatally in at least 70% of cases from contact with infected genital secretions at delivery, congenitally from mothers who acquired primary HSV during pregnancy, and the remainder occurring postnatally from contact with people with symptomatic or asymptomatic HSV-1 infection. Nosocomial transmission has also been described. Neonatal infection includes mucocutaneous (skin, eyes, mouth) disease, encephalitis and disseminated disease. Mucocutaneous disease is the least severe form; some may go on to develop encephalitis or disseminated disease, particularly in the absence of antiviral therapy. Lesions may occur anywhere, including the skin, conjunctivae, mouth and sites of minor trauma (e.g. scalp fetal monitoring electrodes). Neonatal herpes encephalitis can occur as part of localized or disseminated disease. The clinical manifestations include fever, lethargy, poor feeding and focal or generalized seizures. The mortality from untreated encephalitis is over 50%, and there are usually long-term neurological or
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
Figure 3. Brain slice from neonate who died of HSV encephalitis, showing extensive necrosis. Courtesy of Dr R. Osborn, Department of Anatomical Pathology, ICPMR, Westmead Hospital, Westmead, Australia.
ocular sequelae in the survivors. Neonatal HSV-2 encephalitis tends to be more severe than that due to HSV-1, and is associated with a higher frequency of convulsions, more marked cerebrospinal fluid abnormalities and more severe brain damage on computed tomography (Corey et al, 1988) (Figure 3). Disseminated infection has a high mortality rate and usually involves several organs, including the liver, lung, brain, skin and adrenals. In a study of neonates with HSV-2 infection, there were no deaths in 85 with mucocutaneous disease, compared with a mortality rate of' 57% in 46 with disseminated disease and 15% in 71 with encephalitis. A higher mortality rate was associated with the presence of an impaired level of consciousness at presentation, disseminated intravascular coagulation, prematurity and pneumonitis. Morbidity in the survivors was more severe in those with encephalitis, dissemination, seizures and infection due to HSV-2 (Whitley et al, 1991b). With the advent of antiviral agents, the pattern of neonatal HSV infection has changed. From 1973 to 1980, the National Institute for Allergy and Infectious Diseases Collaborative Antiviral Study Group reported 95 cases of neonatal disease--18% had mucocutaneous disease, 32% had encephalitis and 50% had disseminated disease. However, from 1981 to 1987, when antiviral therapy was available, 195 neonates with HSV infection were reported. In this group, 43% had mucocutaneous disease, 34% had encephalitis and 23% had disseminated disease (Whitley et al, 1988). This change is attributed to earlier recognition of mucocutaneous disease and
DWYER AND A. L. CUNNINGHAM
commencement of appropriate antiviral therapy, lessening progression to encephalitis or dissemination. Neonatal HSV infection can relapse despite treatment with acyclovir. Mucocutaneous disease relapse may involve reappearance of lesions with encephalitis. The presence of three or more recurrences of vesicles has been associated with an increased risk of neurological impairment compared to those with fewer recurrences (Dankner and Spector, 1986; Brown et al, 1987a; Whitley et al, 1991a). It has been suggested that as encephalitis often presents with a longer incubation than disseminated or mucocutaneous disease, there may have been an asymptomatic (or clinically unrecognized) primary neonatal infection with a subsequent recurrence in the central nervous system (Kohl, 1990). In adults, HSV encephalitis usually occurs in people who have serological evidence of previous HSV infection. Although rare, congenital HSV infection may also occur due to transplacental transmission of virus. The clinical manifestations of intrauterine HSV infection in 13 patients studied by Hutto et al (1987) included the presence of skin lesions and scars at birth, chorioretinitis, microcephaly, hydranencephaly and microphthalmia. Two babies died in the first week of life and ten had severe neurological sequelae. All infections were caused by HSV-2. Four mothers had experienced a primary infection during the pregnancy, one had recurrent infection and the remaining eight women denied a history of genital HSV infection. Intrauterine infection may also occur late in pregnancy as there have been descriptions where presentation was within 24 hours of birth and where the babies were delivered by caesarean section, sometimes with intact membranes (Koskiniemi et al, 1989). Intrauterine infection may be the explanation for the high rate of spontaneous abortion and premature delivery in both primary and recurrent HSV infection in late pregnancy. Postnatal or nosocomially acquired neonatal HSV infection may occur. Transmission of virus, usually HSV-1, is from hospital personnel with recurrent orolabial herpes or herpetic whitlow, or from the mother and other relatives with symptomatic or asymptomatic orolabial herpes. Occasional nosocomial outbreaks have been reported, usually due to virus spread on the hands of staff. Infants with HSV infection (particularly if mucocutaneous lesions are present) should be isolated in a single room or incubator. Appropriate isolation procedures including use of gowns, gloves and frequent hand-washing should be instituted by all staff involved in patient care.
INVESTIGATION OF HERPES SIMPLEX VIRUS INFECTIONS Introduction
If HSV infection is suspected in a pregnant women or neonate, laboratory confirmation should be sought. There is a wide range of tests currently available for the detection of HSV in clinical specimens, but the current
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
'gold standard' is virus isolation and typing. There are various other methods to detect HSV. Immunoftuorescence (IF) is the most frequently used, but the availability of molecular techniques such as the polymerase chain reaction (PCR) and in situ hybridization with labelled virus-specific probes is increasing. Although less sensitive, cytological examination of lesions (including the Tzanck smear), histopathological analysis of biopsies and electron microscopy of vesicle fluid or tissue are also valuable in some clinical situations. Serological tests to detect total HSV antibody in acute and convalescent sera, and HSV IgM antibody are useful in the diagnosis of primary infection. Newer serological tests have been developed that accurately distinguish between specific antibody to HSV-1 and HSV-2, but these are not widely available (Lee et al, 1986; Ho et al, 1992). The choice of diagnostic test depends on a number of factors, particularly the sensitivity and specificity of the tests and their positive and negative predictive values in different populations. Other important considerations in determining the clinical usefulness of HSV tests include the time taken to obtain results, the ease of specimen collection and transport, the ability of a particular laboratory to perform the test and the ability of the test to distinguish between HSV-1 and HSV-2. The cost-effectiveness of a test is also an important consideration, particularly for widespread screening purposes. The diagnostic value of some tests varies with the stage of the lesion. For example, virus isolation and antigen detection are easier in primary infection where the duration of viral shedding from lesions is much longer than in recurrent episodes. Thus, the stage of clinical presentation will help determine which laboratory test is most appropriate. The various available tests for HSV detection are listed in Table 2.
Table 2. Diagnostic tests for herpes simplex virus infections. Technique Virus isolation Virus antigen detection immunofluorescence, immunoperoxidase enzyme immunoassay Virus genome detection PCR on CSF PCR on genital lesions Electron microscopy Serology (IgG, IgM):~
Time for result
~ 95 % ~ 100% 10-80% 95-99%
90-100% ~ 100% Any herpesvirus 99-100%
1-2 days 1-2 hours 4-6 hours
* Test sensitivity varies depending on the clinical stage and site of the lesions, with most tests showing a higher sensitivity when the lesions are vesicular. Although virus isolation is the 'gold standard', antigen detection methods may be positive when cultures are negative. t The time to obtain a result for virus isolation can be shortened to 48 hours by the use of immunofluorescence on 'shell vial' cultures. :~ Serology tests also refer to those that detect antibodies specific to HSV-1 or HSV-2 glycoprotein G. These tests are not yet widely available. Adapted from Moseley et al (1981), Aurelius et al (1991), Cone et al (1991), Gilbert (1991), Ho et al (1992).
D. E. DWYER AND A. L. CUNNINGHAM
Collection of specimens
Suspected genital herpes If vesicular lesions are present, vesicle fluid can be aspirated with a 1-ml syringe or capillary tube, or absorbed onto a cotton-tipped swab, placed into viral transport medium and transported rapidly to the laboratory for virus isolation. Samples should be processed on the day of collection, but if delay is unavoidable the specimen can be stored at 4°C (but not frozen) as HSV will survive in viral transport media without loss of titre for up to 72 hours (Yeager et al, 1979). Sensitivity of isolation can be improved (especially in recurrent herpes or where lesions are atypical) by taking samples from multiple sites and placing them in one vial of transport medium. To perform IF, cells are scraped from the base of a deroofed vesicle, crusted skin lesion or mucous membrane lesion using a cotton-tipped (not calcium alginate) swab. The swab is then gently smeared onto a Teflon treated glass slide, air dried for 30 minutes, acetone fixed and transported to the laboratory for IF or immunoperoxidase (IP) staining. A swab can also be placed in viral transport media for isolation or antigen detection by enzyme immunoassay (EIA). If the lesions are no longer vesicular, immunofluorescence may be easier to perform than isolation. Serological confirmation of a primary infection can be made by the detection of IgM, seroconversion or a significant increase in antibody level; these tests are of no value in recurrent genital herpes.
Suspected neonatal herpes Isolation of HSV can be performed on most tissue including cerebrospinal fluid (CSF), urine, nasopharyngeal aspirates or pharyngeal swabs, conjunctival swabs or biopsies. Specimens should be transported quickly to the laboratory for analysis. Immunofluorescence or IP can be done rapidly on impression smears (or frozen section) of biopsy tissue, conjunctival swabs and nasopharyngeal aspirates. Electron microscopy, in situ hybridization and PCR can be performed on certain specimens such as CSF and biopsy tissue, but these tests may be limited in their availability. Sera should also be collected for IgM and total antibody detection. Other specimens may also be necessary to exclude other pathogens responsible for severe neonatal disease. If a mother has recurrent genital herpes at the time of vaginal delivery, swabs should be collected from the baby (conjunctiva, umbilicus, nasopharynx, rectum and any suspicious skin lesions) for viral isolation and/or antigen detection at 24-48 hours and then twice a week for at least 3 weeks.
Screening of asymptomatic women in pregnancy If no genital lesions are present, swabs can be used to obtain epithelial cells and secretions from the cervix or posterior vaginal wall, or can be rubbed on the labia or other sites where recurrent herpetic lesions have occurred.
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
These swabs can be placed in viral transport media and sent for virus isolation. Routine swabbing of asymptomatic women in the antenatal period is not indicated. Currently there is no rapid bedside test for the detection of asymptomatic HSV infection at the time of labour.
Laboratory tests Virus isolation
Virus isolation is the most sensitive method to diagnose HSV infection as it will grow readily in a number of different cell culture systems, including monkey kidney and human fibroblasts. Cytopathic effect (CPE) is often detectable within 12-24 hours and in the majority of cultures within 3--4 days. The rapidity of culture can be increased by using IF or EIA on inoculated shell vial cultures within 48 hours. After CPE is seen, the isolate is typed by neutralization, monoclonal antibody IF or EIA. Molecular techniques including restriction endonuclease analysis, PCR and sequencing can also be used to distinguish HSV-1 and HSV-2, and to show strain-specific D N A patterns within serotypes. As genital HSV-2 infection is more likely than HSV-1 to recur symptomatically or asymptomatically (especially in the first few months after a primary infection), it is important to type all HSV isolates from a first episode of infection. The sensitivity of virus isolation will vary with the clinical stage of the lesion. Isolation rates of 94% from vesicular lesions, 87% from pustular lesions, 70% from ulcers and 27% from crusted lesions have been reported (Moseley et al, 1981). Also, the mean duration of viral shedding is longer in primary infection (11.4 days) compared to recurrent episodes (3.9 days). HSV can be isolated in 82% of ulcerative lesions in first episodes compared with 42% of ulcerative lesions in recurrent episodes (Corey and Holmes, 1983). Detection of H S V antigens or nucleic acid
Techniques such as IF, IP or EIA can be used as rapid tests for HSV antigen detection. Immunofluorescence is based on the use of fluorescein-labelled monoclonal or polyclonal antibodies specific for HSV-1 or HSV-2. Although the technique is simple, some skill is required in the collection of specimens and interpretation of results. Immunofluorescence is used more widely than EIA although EIA may be slightly more sensitive. In the study by Moseley et al (198i), direct IF was positive in 71% of lesions in the vesiculopustular stage, compared with IP in 76% and virus isolation in 90%. In ulcerative lesions, the positive rates were lower--IF 38%, IP 55% and isolation 72%. The overall concordance between virus isolation and direct IF was 74%. In this study where patients had clinically suspected genital herpes but virus was not isolated, 35% had HSV antigen detected by IF or IP, with HSV later isolated from subsequent clinical episodes. The overall predictive value of HSV being isolated from a genital or cervical lesion which was IF positive was 88%, and the predictive value of a positive direct
DWYER AND A. L. CUNNINGHAM
IF from a genital lesion was 92% (Moseley et al, 1981). The sensitivity of IF can be increased if only adequate specimens (i.e. slides containing > 25 exfoliated cells) are accepted for testing. Occasionally IF or EIA (and more frequently PCR) are positive when the culture is negative, and this is usually due to non-infectious viral antigen or nucleic acid, the presence of antibody in secretions or loss of virus viability during transport. Newer methods for rapid diagnosis include the use of PCR and in situ hybridization. In patients with suspected herpes simplex encephalitis, PCR performed on CSF is rapid, non-invasive, highly sensitive and specific. In one study, PCR detected HSV DNA in 42 of 43 patients with proven herpes simplex encephalitis, and all but one were positive in the first CSF sample taken (Aurelius et al, 1991). A positive PCR result may negate the need for invasive brain biopsy. In situ hybridization and PCR can also be used on brain or other infected tissue and are more specific than histopathological staining or electron microscopy. PCR performed on genital lesions is more sensitive and is positive for longer than virus isolation, but is not yet available as a routine diagnostic test (Cone et al, 1991).
Serology A serological diagnosis of primary HSV infection can be made by detecting virus-specific IgM in a single serum sample, or seroconversion in acute and convalescent sera. This may be particularly useful during pregnancy because of the prognostic significance of primary infection. Non-type-specific antibodies can be detected by complement fixation, IF and EIA. Type-specific antibody can be detected by neutralization, but is time consuming to perform. IgM can be detected by IF, EIA and Western blot techniques but is usually not detected in the first 10 days of disease (Ho et al, 1992). Demonstration of HSV-specific IgM or IgG in serum or CSF can be useful in neonatal HSV infection if virus isolation is unsuccessful. Retrospective diagnosis of neonatal HSV infection can be made by demonstrating HSV antibodies in convalescent CSF, a rising titre of HSV-specific IgG or its persistence in serum beyond 4-6 months of age when maternal IgG usually disappears. Antibody titres in the CSF to HSV usually rise too late to be helpful in diagnosis of encephalitis although occasionally specific IgM may appear early (Dwyer et al, 1986). There has been increased pressure to develop serological tests that distinguish between HSV-1 and HSV-2 infection. For example, a typespecific assay would allow the differentiation between HSV-1 and HSV-2 primary genital infection, help in the identification of asymptomatic infection, and allow seroepidemiological surveys to be performed in various population groups. A number of techniques have been developed which claim to differentiate type specific antibodies accurately, but problems have arisen with these tests because of antigen and antibody cross-reactivity between HSV-1 and HSV-2, and because of the high incidence of HSV-1 antibodies already present in HSV-2 infected patients. This makes it difficult to distinguish between dual infection and non-specific cross-reaction. Clinicians need to be aware that, despite their claims, many commercially
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
available antibody tests do not accurately distinguish HSV-1 and HSV-2 antibodies (Parkes et al, 1991; Field et al, 1992). The 92 kDa glycoprotein G-2 on the virus surface is the only type-specific protein present in HSV-2. This protein has been purified or produced by recombinant technology and used in immunodot enzyme assays, Western blots and EIA to detect HSV-2 specific IgG and IgM (Lee et al, 1986; Ashley et al, 1988; Ho et al, 1992). Recent studies have used HSV-2 specific tests to determine more accurately the serological status of different population groups, including pregnant women (Johnson et al, 1989; Brown et al, 1991; Kuthanjian et al, 1992). Currently these tests are only performed in research institutions, but development of commercially available EIA tests for widespread testing is in progress. The role of virus isolation in pregnancy
In pregnant women it is particularly important to diagnose genital HSV accurately, and to establish whether it is primary or recurrent. Virus isolation and typing is the best test if genital ulceration is present, but the sensitivity will depend on the clinical stage of the lesions. Isolation can be complemented with antigen detection using IF or other techniques. In the absence of a known history of genital herpes but a positive culture, serological testing should be done to determine whether it is a primary, first episode non-primary or recurrent infection. This is necessary to plan appropriate management of the delivery and infant surveillance. The situation becomes more difficult when assessing the role of virus isolation in asymptomatic pregnant women. Maternal history is a poor predictor of (risk of) neonatal infection, as approximately 70% of mothers whose babies develop HSV infection will have no history of genital herpes. This includes mothers with serological and culture evidence of primary infection prior to delivery. However, not all infants exposed to HSV in the maternal genital tract will become infected. The ideal test would be rapid enough to allow early identification of those women shedding virus at delivery so that appropriate intervention could occur, and to identify those with a primary infection. It had been hoped that screening with antepartum cultures might identify those at risk of transmitting,HSV. It has been reported that of asymptomatic women with a history of previous genital infection, 4.1% will have positive antepartum cultures, 1.4% will have positive cultures at delivery even in the absence of symptoms, and the rate of asymptomatic viral shedding is similar whether or not the mother had symptomatic recurrences during pregnancy. If serial specimens are obtained from the same subject during pregnancy, at least one episode of asymptomatic reactivation can be detected in 2.3--14% .of pregnant women with known previous genital infection (Vontver et al, 1982; Wittek et al, 1984). To further examine the value of delivery cultures, Brown et al (1991) prospectively studied 15 923 asymptomatic women in early labour and cultured both the cervix and external genitalia in most cases. Herpes simplex virus was isolated from 56 women (0.35%), of whom 18 (35%) had serological evidence of a recently acquired but subclinical
D . E. D W Y E R A N D A. L. C U N N I N G H A M
primary episode, and 34 (65%) had reactivation. Neonatal infection developed in 6/18 infants (33%) born to mothers with a first episode of genital HSV and 1/34 (3%) infants born to mothers with a reactivation. Neonatal HSV also occurred in 3 of 15 867 women with negative cultures, a transmission rate of 0.02%. Of the 56 women who shed virus asymptomatically at labour, 7 transmitted HSV to their infants, a transmission rate of 12.5 %. The overall rate of neonatal herpes in this study was approximately 1 in 2000 liveborn infants (Brown et al, 1991). Others have confirmed that women with a history of previous genital infection and virus excretion at delivery have a low transmission rate to their children (Prober et al, 1987, 1988). Therefore, given the facts that a maternal history of genital herpes is often inaccurate in predicting neonatal disease and virus shedding, that antepartum cultures do not predict viral shedding at delivery in either asymptomatic or symptomatic women, and that neonatal HSV infection can occur even in the absence of positive cultures at delivery, antepartum cultures are not useful in predicting babies at risk. There are also practical problems with cultures. Both the cervix and the labia would need to be cultured as virus excretion can occur from both or just one of these sites. Results take some days, thus offering limited value in the management of labour (Gibbs et al, 1988; Lissauer and Jefferies, 1989). The role of serology in pregnancy
As has been stated, in pregnant women who shed HSV-2 asymptomatically in early labour about one-third have recently acquired genital herpes, and their infants are about ten times more likely to develop HSV than those born to women with asymptomatic reactivation. This suggests that maternal antibodies specific to HSV-2 (but not to HSV-1) will reduce the transmission of HSV-2. The next step is to identify women who may be at risk of unsuspected infection with HSV-2 during pregnancy. A recent study addressed this problem by prospectively examining pregnant women presenting to private obstetric practices in California, USA. They found that 32% of 277 women and 25% of 190 partners studied were seropositive for HSV-2 at the initial antenatal visit. Most couples (139/190 or 73%) were serologically concordant for HSV-2, with 57% being both seronegative and 16% both seropositive. There were 18 seronegative women (9.5%) whose partners were seropositive, and 10 of these partners had no history of genital herpes. Therefore, approximately 10% of pregnant women were at risk of contracting a primary HSV-2 infection from their seropositive husbands and in approximately half, this risk was unsuspected. In this situation, antepartum serological testing identified HSV-2 seronegative mothers in a high seroprevalence population who could be unsuspectingly exposed to HSV-2 (Kulhanjian et al, 1992). The frequency of seroconversion during pregnancy in this study was 6% (1/18). The annualized incidence of seroconversion was estimated at 11% in seronegative women with seropositive partners and 20% in such couples who engaged in unprotected intercourse during the pregnancy. It would appear that serology testing of women at the first
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
antenatal visit and their partners would be the only way theoretically to assess those at risk of acquiring primary HSV-2 in pregnancy. Possible advantages are that appropriate precautions to prevent transmission could be suggested to those at risk (e.g. condom use or abstinence, especially in the third trimester), and HSV-2 seropositive females could be identified (including those with no history of genital herpes) and advised about early recognition of genital lesions. 'Surveillance' cultures could be taken from seropositive mothers and their babies at delivery to aid the early identification of HSV-infected infants. However, routine screening of partners may not be practical. Moreover, at present these serological tests are only available in research institutions, and there has been no analysis of the cost of testing versus the number of neonatal infections that may be prevented. As up to 35% of neonatal herpes infections are due to HSV-1, HSV-1 testing would also have to be performed. Testing at the first antenatal visit may also miss primary infections that occur later in pregnancy and obviously will not identify the approximately 10% of neonatal herpes infections that are acquired postnatally. Thus routine antenatal screening for HSV antibodies is unlikely to be cost-effective. MANAGEMENT OF HERPES SIMPLEX INFECTIONS IN THE PREGNANT WOMAN AND NEONATE Antiviral agents
Acyclovir Acyclovir (ACV) has been used for the treatment of herpes simplex infections for over a decade and is still the safest, most effective drug available. It is highly effective in vitro and in vivo against both HSV-1 and HSV-2 (IC50=0.02-0.2 and 0.03--0.5 p.g/ml respectively). However, it is ineffective against latent infection and does not eradicate HSV from the sacral or trigeminal sensory ganglia. The drug owes its efficacy to monophosphorylation and facilitation of cell entry by a viral enzyme, thymidine kinase, found only within infected cells. Acyclovir monophosphate is then converted to the active triphosphate form by cellular kinases. This inhibits viral D N A polymerase (the main enzyme involved in virus replication) and is also incorporated into the growing viral DNA chain, causing inhibition of DNA chain lengthening. The combination of its specific entry mechanism into infected cells and its specific action on viral DNA polymerase results in a 300-fold difference in inhibitory effects of the drug on viral and cellular D N A polymerase, resulting in a very high ratio of efficacy to cellular toxicity. Virus mutants resistant to ACV have been described in vitro and in vivo, especially in severely immunocompromised patients (particularly AIDS patients) receiving multiple courses of antiviral treatment. The mechanisms of ACV resistance are loss or mutation of the viral thymidine kinase, or mutation of the target viral DNA polymerase gene. Most ACV
D W Y E R A N D A . L. C U N N I N G H A M
resistant HSV clinical isolates are thymidine kinase deficient mutants; the highly resistant DNA polymerase mutants occur rarely in vivo (Erlich et al, 1989; Whitley and Gnann, 1992). Acyclovir may be administered intravenously (i.v.), orally or topically, with oral administration resulting in blood levels approximately 20% those of i.v. administration. Acyclovir is mainly eliminated by the kidneys (90%), with some metabolized by the liver. Side-effects are uncommon with i.v., and rare with oral, ACV. Local extravasation of i.v. ACV may result in .severe phlebitis, and rapid i.v. infusions administered to dehydrated patients may result in renal deposition of ACV crystals and renal impairment. Neurotoxicity has been documented in patients with renal and hepatic failure if dosage is not modified. In animal models, ACV is not teratogenic or oncogenic and has no effect on fertility or fetal development. Exposure to ACV during pregnancy has not so far been shown to increase birth defects or maternal risk, but the number of monitored pregnancies is too small to detect any low-frequency events (Andrews et al, 1992). Although ACV has shown mutagenic potential in a minority of in vitro systems, it is unknown whether this has implications for long-term human therapy (Moore et al, 1983). Vidarabine
Vidarabine (adenine arabinoside, Ara-A) was the first antiviral agent used successfully for the treatment of systemic herpes simplex infections, including encephalitis (Whitley et al, 1980). It is effective against both HSV-1 and HSV-2 (IDs0 = 3 txg/ml). Its mechanism of action is not completely understood although it is known to be phosphorylated by cellular enzymes, and competitively inhibits viral DNA polymerase to a greater extent than cellular D N A polymerase. It is also incorporated into, and inhibits, the growing viral DNA chain. Viral DNA polymerase mutants may become resistant to vidarabine in vitro but this has not been confirmed clinically. As ACV, vidarabine and foscarnet do not share a common mechanism of resistance, vidarabine and foscarnet can be used in ACV-r~sistant infections (Field et al, 1981). Vidarabine must be administered by i.v. infusion and has variable penetration into the CSF, with the CSF/plasma ratio ranging from 35% in adults to 90% in infants. Vidarabine and its active metabolite, hypoxanthine arabinoside, are mainly excreted by the kidneys, and dose modification is necessary in renal failure. Vidarabine has been shown to be teratogenic, mutagenic and oncogenic in animals. In addition, gastrointestinal toxicity is common and the drug is not very soluble, requiring administration of large fluid volumes (a problem in neonates or in patients with cerebral oedema complicating encephalitis). Neurotoxicity may be encountered at high doses or with concurrent interferon or allopurinol therapy. Haematological toxicity has also been documented at high doses. Nevertheless, in neonates the discrepancy between the toxicities of ACV and vidarabine appears to be much less than in adults (Whitley et ai, 1991a). As ACV is generally more effective and less toxic, vidarabine is now restricted to second-line treatment of neonatal
HERPES SIMPLEXVIRUSINFECTIONIN PREGNANCY
herpes and serious infections with resistant HSV in immune compromised hosts.
Foscarnet Foscarnet (phosphonoformate) is a pyrophosphate analogue that binds to the pyrophosphate binding site on the viral DNA polymerase. It is effective against both herpes simplex viruses and cytomegalovirus. It is given as an i.v. infusion because oral absorption is low. It penetrates into the CSF and the eye, and up to 20% of the administered dose accumulates in bone. It is mainly excreted by the kidneys and requires dosage adjustment in renal impairment. The main toxicity of the drug is renal, causing tubular atrophy and polyuria. Like vidarabine, the mechanisms of foscarnet resistance do not overlap with ACV. Its main use is the treatment of severe cytomegalovirus infection in immunocompromised patients, but it is also used as a second-line drug for severe infections caused by resistant HSV (Chatis et al, 1989). Topical foscarnet therapy for genital herpes has not proven to be consistently effective.
Interferons and other drugs The supply of purified interferons et and 13has been markedly increased by recombinant DNA technology. However, weak efficacy combined with considerable toxicity (fever, fatigue, gastrointestinal side-effects and haematological abnormalities) in controlled trials has reduced its role to that of an occasional partner in combination therapy for ACV-resistant HSV infections (Mendelson et al, 1986; Lassus et al, 1987; Pazin et al, 1987). Recent trials have shown isoprinosine (inosine pranobex), an immunomodulator, to be relatively ineffective in the treatment or prophylaxis of genital herpes simplex infections compared with ACV (Mindel et at, 1988a). Many other drugs, including topical treatment (e.g. idoxuridine, trifluorothymidine, nonoxynol 9), have been used for the treatment of genital herpes, but none has been effective in controlled trials. In view of the clinical variability of recurrent genital herpes, drugs with anecdotal benefit can only be recommended when efficacy is proven in controlled trials.
Treatment of genital herpes Acyclovir is very effective in shortening the duration of virus shedding and new lesion formation in initial genital herpes (Bryson et al, 1983). It has less effect in recurrent genital herpes because of differences in duration of virus shedding from initial and recurrent herpes lesions. In primary genital herpes, viral shedding is reduced from a median of 12 to 6 and 9 days with i.v. and oral ACV respectively. In recurrent genital herpes viral shedding is reduced from a median of 4 to 3 days by oral ACV. The narrow window for efficacy in recurrent genital herpes may be further reduced if the patient presents late after onset of lesions. In primary genital herpes, total healing time is reduced from 21 to 9 days and in recurrent genital herpes from 10 to 7
D . E. D W Y E R A N D A . L. C U N N I N G H A M
days by ACV (Corey and Holmes, 1983). Topical ACV therapy also reduces viral shedding and hastens healing of initial genital herpes, but efficacy in recurrent lesions has not been consistently demonstrated (Corey et al, 1982; Fiddian et al, 1983). Therapy with oral ACV (i.v. in severe cases) should be used in all patients with initial genital herpes and associated cervical, urethral and pharyngeal involvement to allow more rapid healing. Aseptic meningitis and acute urinary retention associated with primary infection may require i.v. therapy. Universal treatment of recurrent genital herpes is more difficult to justify, particularly as the severity, duration and frequency of genital herpes varies considerably. In some cases shortening episodes by even one day may be of benefit to some people with frequent severe episodes (Corey and Spear, 1986b).
Prophylaxis of recurrent genital herpes In some people with frequent (4-12 recurrences per year) or severe recurrent genital herpes, suppressive therapy may be advisable. Controlled trials with oral ACV (200 mg two to five times daily for up to 2 years) prevented recurrences in 65-85% of patients and reduced the number of recurrences in 90%. However, breakthroughs with either ACV sensitive or resistant HSV occur in approximately a quarter of these patients every 3 months, and transmission may occur during these breakthroughs. Early studies used a variety of dosages but recent studies have suggested that 400 mg b.d. or 200 mg q.i.d, of ACV are the best starting regimens, followed by dosage reduction after several months if breakthroughs do not occur (Douglas et al, 1984; Straus et al, 1984; Mindel et al, 1988b). The long-term safety of suppressive oral ACV has been the subject of speculation and controversy although side-effects are rare. In a recent trial of over 1000 people taking oral ACV for 1 year, no significant clinical or laboratory toxicity, or decreased spermatic function in males, was observed (Mertz et al, 1988): In choosing between intermittent or suppressive therapy of recurrences, clinicians must take into account the frequency and severity of lesions, the social and psychological setting and the cost.
Acyclovir treatment or prophylaxis of genital herpes in pregnancy
Pharmacokinetics and toxicity Despite increased renal blood flow in pregnancy, blood levels of ACV after oral therapy are not reduced. Acyclovir crosses the placenta and is excreted by the less efficient fetal kidney. Nevertheless, ACV does not accumulate in the fetus although renal excretion does lead to its accumulation in amniotic fluid. Acyclovir also accumulates in breast milk (Englund et al, 1991; Frenkel et al, 1991). Acyclovir has been closely monitored for potential teratogenic effects because of its wide usage in sexually active women of child-bearing age. The best data on the lack of teratogenic effects of ACV in humans comes from the 'Acyclovir in Pregnancy Registry' managed by
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
B u r r o u g h s - W e l l c o m e C o m p a n y in collaboration with the Centers for Disease C o n t r o l , A t l a n t a , U S A . A l t h o u g h A C V is not r e c o m m e n d e d in p r e g n a n c y except in severe infections, the registry has followed 312 A C V e x p o s e d pregnancies, including 39 exposed during the first trimester. A m o n g the 241 cases followed prospectively to delivery, there were 10 congenital abnormalities, a rate of 4.1% (95% confidence interval: 1.6--6.6%). This is similar to the 3% rate in the general p o p u l a t i o n o f liveborn infants. In addition, there was no consistent p a t t e r n in the a b n o r malities to suggest that A C V is teratogenic. T h e r e were also 24 s p o n t a n e o u s abortions a m o n g s t 265 prospective cases at a rate of 9 . 1 % , similar to the 1 0 - 2 0 % rate in the general population. T h e authors c a u t i o n e d that the current sample size is still quite small and only capable of detecting a 20-fold increase in risk for a birth defect with a baseline rate of 1 per 1000 ( A n d r e w s et al, 1992). R e c o m m e n d e d dosages for H S V infection in p r e g n a n c y and the n e o n a t e are listed in Table 3. Table
3. Recommended dosages for herpes simplex infections in pregnancy and the neonate.
Pregnancy Initial genital herpes Acyclovir 200 mg orally 5 times daily for 7-10 days Acyclovir 5 mg/kg i.v. 8-hourly for 7-10 days if severe or complicated infection (oral therapy can be substituted depending on clinical response) Acyclovir ointment (5%) topically 6-hourly is a less satisfactory alternative Encephalitis Acyclovir 10 mg/kg i.v. 8-hourly for 14 days Hepatitis Acyclovir 5 mg/kg i.v. 8-hourly for 7 days ACV-resistant HSV infections (immune compromised, e.g. AIDS) Foscarnet 60 mg/kg i.v. 8-hourly, ganciclovir 5 mg/kg i.v. 8-hourly, or vidarabine 10 mg/kg i.v. 8-hourly for 10-14 days
Neonatal disease Acyclovir 10 mg/kg i.v. 8-hourly for 10-14 days Vidarabine 30 mg/kg per day (concentration no greater than 0.7 mg/ml in standard i.v. fluid) as a continuous i.v. infusion over a 12-hour period as an alternative to acyclovir Doses need to be adjusted in renal failure. Acyclovir in recurrent genital herpes (200 mg orally 5 times daily for 3-5 days) and as suppressive therapy (200-400 mg orally 8-12 hourly) is rarely indicated in pregnancy. Longer term acyclovir therapy (e.g. 21 days) may be required in some cases of severe neonatal infection. Prophylactic acyclovir (10mg/kg 8-hourly for 5-7 days) may be indicated in babies delivered vaginally from mothers with primary genital herpes. In children from 1 to 12 years of age, acyclovir dosage for severe infections can also be expressed as 250-500 mg per square metre of body surface area i.v. 8-hourly.
When should acyclovir be used in pregnancy? D e s p i t e the reassuring toxicity data, the advisory c o m m i t t e e to the ' A c y c l o v i r in P r e g n a n c y Registry' advised adhering to the Centers for Disease C o n t r o l Sexually T r a n s m i t t e d Disease T r e a t m e n t Guidelines: ' I n the p r e s e n c e of life threatening m a t e r n a l H S V infections (e.g. disseminated infection that
DWYER AND A. L. CUNNINGHAM
includes encephalitis, pneumonitis, and/or hepatitis), acyclovir administered intravenously is probably of value. Among pregnant women without life threatening disease, systemic acyclovir treatment should not be used for recurrent genital herpes episodes or as suppressive therapy to prevent reactivation near term' (Centers for Disease Control, 1989). The role of ACV in pregnant women with uncomplicated primary genital herpes or as prophylaxis for recurrences is uncertain (Brown and Baker, 1989). One proposed strategy for prevention of vertical transmission of HSV is the use of ACV in patients near term with a past history of recurrent genital herpes (Gibbs and Mead, 1992). However, ACV has not been shown to prevent asymptomatic virus shedding, and the majority of neonatal herpes is not associated with a maternal history of recurrent herpes (Straus et al, 1989; Yeager and Arvin, 1984). Treatment of primary genital herpes (transmission rate 30-50%) if present before or during labour is warranted, but the benefits of treatment of recurrent genital herpes present at parturition (transmission rate 3%), are doubtful and data from controlled trials are awaited. Treatment of neonatal herpes
In a large multicentre randomized controlled trial of 202 babies less than 1 month of age with virologically confirmed HSV infection, i.v. vidarabine and ACV were found to be equally effective in reducing morbidity and mortality. The outcome varied significantly according to the extent of disease. There was no mortality among babies with skin, eye or mouth involvement and over 90% appeared to develop normally on follow-up. Severe neurological morbidity or long-term sequelae (as distinct from acute morbidity) was observed in some of the remaining 10%, with a widely distributed skin rash (suggesting viraemia and dissemination) being a poor prognostic indicator. Mortality and morbidity increased progressively for babies with encephalitis (t4% and 43% respectively) and disseminated disease (50% and 42% respectively). Vidarabine therapy produced significantly more haematological abnormalities, although these were not associated with clinical complications. Of the surviving babies with encephalitis or disseminated disease, 8% had a recurrence of the neurological disease within a month of completing therapy. Recurrent skin lesions developed within a month of therapy in 19% of vidarabine recipients and 35% of ACV recipients. Six months after therapy, the incidence of recurrences had increased to 46% in both treatment groups (Whitley et al, 1991a). In conclusion, there was no significant difference in the efficacy of vidarabine or ACV, but as ACV is the easier and tess toxic drug to administer, it is the drug of choice for neonatal herpes. However, the results of therapy are still unsatisfactory in view of the high morbidity and mortality in systemic HSV infection. Strategies for improving the outcome include earlier detection of cases of neurological and disseminated disease, higher doses of i.v. therapy, prolonging therapy to 21 days and perhaps continuing therapy with oral ACV to prevent recurrences. No controlled trials have yet addressed the need for prophylactic ACV treatment of a neonate where a vaginal birth inadvertently occurs in a
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
mother with active primary or recurrent genital herpes. In the absence of controlled trial data, it would seem that treatment should be administered to babies born of mothers with primary genital herpes as the transmission rate is approximately 30% (Brown et al, 1991). However, this is less certain in babies born to mothers with active recurrent genital herpes where the low transmission rate (approximately 3%) must be balanced against the risks of i.v. ACV. Counselling of the mother, surveillance cultures from the child's mouth, eyes, groin, rectum and any suspicious skin lesions, and close observation over the first 3 weeks for any clinical evidence of localized or disseminated herpes infection will guide the need for ACV therapy. Strategies for prevention of vertical transmission of HSV The rarity of neonatal HSV infection, combined with its severity, makes its management one of the more difficult problems in obstetrics and neonatal medicine. The guiding principle of prevention has been to identify mothers likely to be infectious at delivery (i.e. with primary or non-primary first episodes, or clinically apparent recurrences), and then deliver their infants by caesarean section. Caesarean section does not prevent all cases of neonatal disease in mothers with primary genital herpes at delivery. Practically, the identification of asymptomatic mothers likely to be infectious at term has proved extremely difficult (Gibbs and Mead, 1992). This is because most cases of neonatal herpes occur without an identifiable history of past recurrent genital herpes, examination of the vulva at term may be inadequate to detect virus shedding, and surveillance cultures prior to the onset of labour are not predictive of virus shedding at birth (Arvin et al, 1986). It has been suggested that women at risk may be identified by a HSV-2 specific antibody test (Ho et al, 1992; Kulhanjian et al, 1992), but this is unlikely to be cost-effective with present techniques. Alternatively, the use of a rapid diagnostic test for HSV antigen in all pregnant women has been proposed, but this would require exquisite specificity and sensitivity to be useful. The sensitivity and specificity of the best HSV antigen EIA has been reported as 92-95% (Dascal et al, 1989). In addition, caesarean section does not prevent neonatal herpes acquired congenitally (5-8% of all cases), perinatally by transplacental spread during primary maternal viraemia, or postnatally (8-10%) (Hutto et al, 1987; Chuang, 1988). Libman et al (1991) used decision analysis to decide between eight possible strategies for the prevention of vertical transmission (Table 4). These ranged from physical examination of the vulva and cervix at labour to antigen testing of all women to viral culture or antigen testing of genital secretions of seropositive women in labour. In these strategies there was an inverse relationship between the reduction in numbers of cases of neonatal herpes and the number of excess caesarean sections. The most dramatic reduction in the number of cases occurred with the use of the rapid diagnostic test for HSV shedding at delivery. However, this resulted in a tenfold increase in the number of caesarean sections. The greatest reduction in cases of neonatal herpes for the smallest increase in caesarean sections was examination of the vulva and cervix for
D.E. DWYERAND A. L. CUNNINGHAM
Table 4. Summaryof possible strategies for the prevention of neonatal infectionwith HSV (Libmanet al, 1991). Strategy
Description Physicalexaminationat labour of vulva and cervix Culture of women clinicallyat high risk Culture of HSV-2 seropositivewomen Antigen testing of all womenplus physicalexamination Antigen testing of all women, regardlessof physicalfindings Antigen testing of HSV-2 seropositivewomenplus physicalexamination Antigen testingof HSV-2 seropositivewomen regardlessof physicalfindings Antigen testingof HSV-2 seronegativewomenonly No test
herpes lesions in all women coming to labour, or HSV antigen testing at term on pregnant women found seronegative for HSV-2. In the latter setting, such women could be tested for an asymptomatic HSV-2 (or HSV-1) primary genital herpes (Libman et al, 1991). However, this strategy presumes a diagnostic test which could produce results in less than one hour and retain a high specificity and sensitivity. Such a test is currently not available. Until such a test is available, the most useful strategy is the examination of the vulva and cervix of all pregnant women at term, and performance of a caesarean section on those with clinically apparent herpes lesions. Unfortunately even here it is not possible to focus solely on the small group of women already known to be HSV-2 seropositive as this would exclude patients with recurrent genital HSV-1 infection and those who develop asymptomatic primary infection close to term. The latter patients provide a significant proportion of the cases of vertical transmission to infants (Prober et al, 1987). Orogenital contact in late pregnancy in HSV-1 seronegative women is also a likely risk factor for primary genital HSV-1 and transmission close to parturition. SUMMARY
Genital infections in pregnancy caused by herpes simplex virus types 1 and 2 are a difficult management problem. Both primary and recurrent genital herpes can range from severe, extensive ulceration to asymptomatic virus shedding. Although neonatal herpes is a well recognized complication of symptomatic maternal primary genital infection at the time of delivery, most cases are associated with asymptomatic virus shedding and absence of a history of genital herpes. Neonatal herpes may also be acquired in utero and in the postnatal period. The diagnosis of herpes simplex infection is made most reliably by virus isolation or antigen detection from samples obtained from clinically apparent lesions. Serology is useful for diagnosing primary herpes, and newer serological techniques allow the detection of HSV-2 specific antibodies. Strategies to prevent neonatal herpes are limited by the failure of currently available diagnostic tests to rapidly detect women in labour who
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
are at risk of t r a n s m i t t i n g h e r p e s , by the a b s e n c e o f p r o v e n a n t e n a t a l s c r e e n i n g tests for H S V , a n d by t r a n s m i s s i o n of h e r p e s to n e o n a t e s f r o m a s y m p t o m a t i c m o t h e r s . T h e most useful c u r r e n t strategy is careful e x a m i n a t i o n of the v u l v a a n d cervix for h e r p e s lesions in w o m e n c o m i n g to l a b o u r . C a e s a r e a n section is i n d i c a t e d in w o m e n with clinically a p p a r e n t g e n i t a l h e r p e s at delivery, Effective a n d safe antiviral agents are available for t r e a t m e n t of m a t e r n a l a n d n e o n a t a l herpes.
Acknowledgements Our thanks go to Ms Claire Greenall for preparation of the manuscript, and Dr David Holland and Dr Eva Hochstein for reviewing the manuscript.
REFERENCES Andrews EB, Yankaskas BC, Cordero JF et al (1992) Acyclovirin pregnancy registry: six years experience. Obstetrics and Gynecology 79: 7-13. Arvin AM, Hensleigh PA, Prober CG et al (1986) Failure of antepartum maternal cultures to predict the infant's risk of exposure to herpes simplex virus at delivery. New England Journal of Medicine 315: 796-800. Ashley RL, Miltoni J, Lee F, Nahmias A & Corey L (1988) Comparison of Western blot (immunoblot) and glycoprotein-G-specific immunodot enzyme assay for detecting antibodies to herpes simplex virus types 1 and 2 in human sera. Journal of Clinical Microbiology 26: 662-700. Aurelius E, Johansson B, Skoldenberg B, Staland A & Forsgren M (1991) Rapid diagnosis of herpes simplex encephalitis by nested polymerase chain reaction assay of cerebrospinal fluid. Lancet 337: 189--192. Berger SA, Weinberg M, Treves T et al (1986) Herpes encephalitis during pregnancy: failure of acyclovir and adenine arabinoside to prevent neonatal herpes. Israeli Journal of Medical Science 22: 41-44. Bernstein DI (1991) Effects of prior HSV-1 infection on genital HSV-2 infection. Progress in Medical Virology 38: 109-127. Bernstein DI, Lovett MA & Bryson YJ (1984) Serologic analysis of first-episode nonprimary genital herpes simplex virus infection. American Journal of Medic,ine 77: 1055-1060. Breinig MK, Kingsley LA, Armstrong JA, Freeman DJ & Ho M (1990) Epidemiology of genital herpes in Pittsburgh: serologic, sexual, and racial correlates of apparent and inapparent herpes simplex infection. Journal oflnfectious Diseases 162: 299--305. Brock BV, Selke S, Benedetti J, Douglas JM Jr & Corey L (1990) Frequency of asymptomatic shedding of herpes simplex virus in women with genital herpes. Journal of the American Medical Association 263: 418-421. Brown ZA & Baker DA (1989) Acyclovir therapy during pregnancy. Obstetrics and Gynecology 73: 526-531. Brown ZA, Ashley R, Douglas J, Keilly M & Corey L (1987a) Neonatal herpes simplex virus infection: relapse after initial therapy and transmission from a mother with an asymptomatic genital herpes infection and erythema multiforme. Pediatric Infectious Diseases Journal 6: 1057-1061. Brown ZA, Vontver LA, Benedetti J et al (1987b)Effects on infants of a first episode of genital herpes during pregnancy. New England Journal of Medicine 317: 1246-1251. Brown ZA, Benedetti J, Ashley R et al (1991) Neonatal herpes simplex virus infection in relation to asymptomatic maternal infection at the time of labor. New England Journal of Medicine 324: 1247-1252. Bryson YJ, Dillon M, Lovett M et al (i983) Treatment of first episodes of genital herpes simplex virus infection with oral acyclovir.New England Journal of Medicine 308: 916-921.
D. E. DWYER AND A. L. CUNNINGHAM
Centers for Disease Control (1989) Sexually transmitted diseases: treatment guidelines. Morbidity and Mortality Weekly Report 38: 15-16. Chatis PA, Miller CH, Schrager L E e t al (1989) Successful treatment with foscarnet of an acyclovir-resistant mucocutaneous infection with herpes simplex virus in a patient with acquired immunodeficiency syndrome. New England Journal of Medicine 320: 297-300. Chuang TY (1988) Neonatal herpes: incidence, prevention and consequences. American Journal of Preventative Medicine 4: 47-53. Cone RW, Hobson AC, Palmer J, Remington M & Corey L (1991) Extended duration of herpes simplex virus DNA in genital lesions detected by polymerase chain reaction. Journal of lnfectious Diseases 164: 757-760. Corey L & Holmes KK (1983) Genital herpes simplex virus infections: current concepts in diagnosis, therapy and prevention. Annals of Internal Medicine 98: 973-983. Corey L & Spear PG (1986a) Infections with herpes simplex viruses (first of two parts). New England Journal of Medicine 314: 686~591. Corey L & Spear PG (1986b) Infections with herpes simplex viruses (second of two parts). New England Journal of Medicine 314: 749-757. Corey L, Reeves WC & Holmes KK (1978) Cellular immune response in genital herpes simplex virus infection. New England Journal of Medicine 299: 986-991. Corey L, Nahmias A J, Guinan ME, Benedetti JK, Critchlow CW & Holmes KK (1982) Double blind placebo controlled trial of topical acyclovir in first and recurrent episodes of genital herpes simplex virus infection. New England Journal of Medicine 306: 1313-1319. Corey L, Adams HG, Brown ZA & Holmes KK (1983) Genital herpes simplex virus infections: clinical manifestations, course and complications. Annals of Internal Medicine 98: 958-972. Corey L, Whitley RJ, Stone EF & Mohan K (1988) Difference between herpes simplex virus type 1 and type 2 neonatal encephalitis in neurological outcome. Lancet i: 1--4. Cunningham AL & Noble JR (1989) Role of keratinocytes in human recurrent herpetic lesions. Journal of Clinical Investigation 83: 490--496. Cunningham AL, Turner RR, Miller AC, Para MF & Merigan TC (1985) Evolution of recurrent herpes simplex lesions. Journal of Clinical Investigation 75: 226-233. Cunningham AL, Lee FK, Ho DWT et al (in press) Prevalence of antibody to Herpes simplex virus type 2 in patients.attending antenatal and sexually transmitted diseases clinics in Sydney. Medical Journal of Australia in press. Dankner WM & Spector SA (1986) Recurrent herpes simplex in a neonate. Pediatric Infectious Diseases 5: 582-586. Dascal A, Chan-Thim J, Morahan Met al (1989) Diagnosis of herpes simplex virus infection in a clinical setting by a direct antigen detection enzyme immunoassay kit. Journal of Clinical Microbiology 27: 700-704. Douglas JM, Critchlow C, Benedetti J e t al (1984) A double-blind study of oral acyclovir for suppression of recurrences of genital herpes simplex virus infection. New England Journal of Medicine 310: 1551-1556. Dwyer DE, O'Flaherty S, Packham D & Cunningham AL (1986) Herpes simplex encephalitis in infants. Medical Journal of Australia 144: 714--715. Englund JA, Fletcher CV & Balfour HH (1991) Acyclovir therapy in neonates. Journal of Pediatrics 119: 129--135. Erlich KS, Mills J, Chatis P et al (1989) Acyclovir-resistant herpes simplex virus infections in patients with the acquired immunodeficiencysyndrome. New England Journal of Medicine 320: 293-296. Field H, McMillan A & Darby G (1981) The sensitivity of acyclovir-resistantmutants of herpes simplex virus to other antiviral drugs. Journal oflnfectious Diseases 143: 281-284. Field PR, Ho DWT, Irving WL, Isaacs D & Cunningham AL (1992) The reliability of serological tests for the diagnosis of genital herpes: a critique. Pathology in press. Fiddian AP, Kinghorn GR, Goldmeier D et al (1983) Topical acyclovir in the treatment of genital herpes: a comparison with systemic therapy. Journal of Antirnicrobial Chemotherapy 12: 67-77. Frenkel LM, Brown ZA, Bryson YJ et al (1991) Pharmacokinetics of acyclovir in the term human pregnancy and neonate. American Journal of Obstetrics and Gynecology 164: 569-576. Gibbs RS & Mead PB (1992) Preventing neonatal herpes--current strategies. New England Journal of Medicine 326: 946-947.
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
Gibbs RS, Amstey MS, Sweet RL, Mead PB & Sever J L (1988) Management of genital herpes infection in pregnancy. Obstetrics and Gynecology 71: 779-780. Gilbert GL (1991) Herpes simplex virus infections. In Gilbert GL (ed.) Infectious Diseases in Pregnancy and the Newborn Infant, pp 103-125. Chur: Harwood Academic Press. Ho DWT, Field PR, Sjogren-Jansson E, Jeansson S & Cunningham AL (1992) Use of an indirect ELISA for the detection of HSV-2 specific IgG and lgM antibodies with glycoprotein G (gG-2). Journal of Virological Methods 36: 249-264. Hutto C, Arvin A, Jacobs R et al (1987) Intrauterine herpes simplex virus infections. Journalof Pediatrics l l 0 : 97-101. Johnson RE, Nahmias AJ, Magder LS et al (1989) A seroepidemiologic survey of the prevalence of herpes simplex virus type 2 infection in the United States. New England Journal of Medicine 321: 7-12. Koelle DM, Benedetti J, Langenberg A & Corey L (1992) Asymptomatic reactivation of herpes simplex virus in women after the first episode of genital herpes. Annals of Internal Medicine 116: 433-437. Kohl S (1990) A hypothesis on the pathophysiology of neonatal herpes simplex virus encephalitis: clinical recurrence after asymptomatic primary infection. Pediatric Infectious Diseases Journal 9: 307-308: Koutsky LA, Stevens CE, Holmes KK et al (1992) Underdiagnosis of genital herpes by current clinical and viral isolation procedures. New England Journal of Medicine 326: 1533-1539. Koskiniemi M, Happoneu JM, Jarvenpaa AL, Pettay O & Vaheri A (1989) Neonatal herpes simplex virus infection: a report of 43 patients. Pediatric Infectious Diseases Journal 8: 30-35. Kulhanjian JA, Souroush V, Au DS et al (1992) Identification of women at unsuspected risk of primary infection with herpes simplex virus type 2 during pregnancy. New England Journal of Medicine 326: 916-920. Laffery WE, Coombs RW, Benedetti J, Critchlow C & Corey L (1987) Recurrences after oral and genital herpes simplex virus infection. New England Journal of Medicine 316: 14441449. Langenberg A, Benedetti J, Jenkins J e t al (1989) Development of clinically recognizable genital lesions among women previously identified as having 'asymptomatic' herpes simplex virus type 2 infection. Annals of Internal Medicine 110: 882-887, Lassus A, Bergelin I, Paloranta A e t al (1987) Efficacy of interferon and placebo in the treatment of recurrent genital herpes: a double-blind trial. Sexually Transmissable Diseases 14: 185-190. Lee FK, Pereira L, Griffin C et al (1986) A novel glycoprotein for detection of herpes simplex virus type 1-specific antibodies. Journal of Virological Methods 14:111-1 t 8. Libman MD, Dascal\A, Kramer MS & Mendelson J (1991) Strategies for the prevention of neonatal infection with herpes simplex virus: a decision analysis. Reviews of Infectious Diseases 13: 1093-1104. Lissauer T & Jefferies D (1989) Preventing neonatal herpes infection. British Journal of Obstetrics and Gynaecology 96: 1015-1023. Mendelson J, Clecner B & Eiley S (1986) Effect of recombinant interferon alpha 2 on clinical course of first episode genital herpes infection and subsequent recurrences. Genitourinary Medicine 62: 97-101. Mertz GJ, Jones CC, Mills J e t al (1988) Long-term acyclovir suppression of frequently recurring genital herpes simplex virus infection: a mulficenter double-blind trial. Journal of the American Medical Association 260: 201-206. Mertz GJ, Benedetti J, Ashley R, Selke SA & Corey L (1992) Risk factors for the sexual transmission of genital herpes. Annals of Internal Medicine 116: 197-202. Mindel A, Carney O, Sonnerc C e t a l (1988a) Suppression of frequently recurring genital herpes: acyclovir versus inosine pranobex. Genitourinary Medicine 65: 103-105. Mindel A, Faherty A, Carney O e t al (1988b) Dosage and safety of long term suppressive therapy for recurrent genital herpes. Lancet i: 926--928. Mindel A, Kinghorn G, Allason-Jones E et al (1987) Treatment of first attack genital herpes, acyclovir versus inosine pranObex. Lancet i: 1171-1173. Moore HL, Szczech GM, Rodwell DE et al (1983) Preclinieal toxicology studies with acyclovir: teratologic, reproductive and neonatal tests. Fundamentals of Applied Toxicology 3: 560--568.
D. E. DWYER AND A. L. CUNNINGHAM
Moseley RC, Corey L, Benjamin D, Winter C & Remington ML (1981) Comparison of viral isolation, direct immunofluorescence, and direct immunoperoxidase techniques for detection of genital herpes simplex virus infection. Journal of Clinical Microbiology 13: 913-918. Nahmias AJ, Josey WB, Naib ZM, Freeman MG, Fernandez RJ & Wheeler JH (1971) Perinatal risk associated with maternal genital herpes simplex virus infection. American Journal of Obstetrics and Gynecology 110: 825-837. Nahmias A J, Keyserling HL & Kerrick GM (1983) Herpes simplex. In Remington JS & Klein JO (eds) Infectious Diseases of the Fetus and Newborn Infant, pp 6364538. Philadelphia: WB Saunders. Nahmias AJ, Lee FK & Beckman-Nahmias S (1990) Sero-epidemiological and sociological patterns of herpes simplex virus infection in the world. Scandinavian Journal of Infectious Diseases 69: 19-36. Overall JC, Spruance SL & Green JA (1981) Viral induced leukocyte interferon in vesicle fluid from lesions of recurrent herpes labialis. Journal oflnfectious Diseases 143: 543-547. Parkes DL, Smith CM, Rose JM, Brandis J & Coates SR (1991) Seroreactive recombinant herpes simplex virus type 2 specific glycoprotein G. Journal of Clinical Microbiology 29: 778-781. Parvey LS & Chien LT (1980) Neonatal herpes simplex virus infection introduced by fetal monitor scalp electrodes. Pediatrics 65: 1150-1153. Pazin GJ, Harger JH, Armstrong JA et al (1987) Leukocyte interferon for treating first episodes of genital herpes in women. Journal of Infectious Diseases 156: 891-898. Peacock JE & Sarubbi MD (1983) Disseminated herpes simplex virus infection during pregnancy. Obstetrics and Gynecology 61: 13-18. Prober CG, Sullender WM, Yasukawa LL et al (1987) Low risk of herpes simplex virus infections in neonates exposed to the virus at the time of vaginal delivery to mothers with recurrent genital herpes simplex virus infections. New England Journal of Medicine 316: 240-244. Prober CG, Hensleigh PA, Boucher FD et al (1988) Use of routine viral cultures at delivery to identify neonates exposed to herpes simplex virus. New England Journal of Medicine 318: 887-891. Reeves WC, Corey L, Adams HG et al (1981) Risk of recurrence after first episode of genital herpes. New England Journal of Medicine 305: 315-319. Roizman B & Sears AE (1990) Herpes simplex viruses and their replication. In Fields BN & Knipe DM (eds) Virology, pp 1795-1841. New York: Raven Press. Rooney JF, Felser JM, Ostrove JM & Straus SE (1986) Acquisition of genital herpes from an asymptomatic sexual partner. New England Journal of Medicine 314: 1561-1564. Simmons A (1989) H-2-1inked genes influence the severity of herpes simplex virus infection of the peripheral nervous system. Journal of Experimental Medicine 169: 1503-1507. Simmons A, Tscharke D & Speck P (1992) The role of immune mechanisms in control of herpes simplex virus infection of the peripheral nervous system. In Rouse B (ed.) Herpes Simplex Virus: Pathogenesis, lmmunobiology and Control, pp 31-56. Berlin: Springer-Verlag. Stevens JG & Cook ML (1971) Latent herpesvirus in spinal ganglia of mice. Science 173: 843-845. Straus SE, Takiff HE, Seidlin Met al (1984) Suppression of frequently recurring genital herpes. New England Journal of Medicine 310: 1545-1550. Straus SE, Seidlin M, Takiff HE et al (1989) Effect of oral acyclovir treatment on symptomatic and asymptomatic virus shedding in recurrent genital herpes. Sexually Transmitted Diseases 16: 107-113. Vontver LA, Hickok DE, Brown Z, Reid L & Corey L (1982) Recurrent genital herpes simplex virus infection in pregnancy: infant outcome and frequency of asymptomatic recurrences. American Journal of Obstetrics and Gynecology 143: 75-84. Whitley RJ & Gnann JW (1992) Acyclovir: a decade later. New England Journal of Medicine 327: 782-789. Whitley RJ, Nahmias AJ, Visintine AM, Fleming CL & Alford CE (1980) The natural history of herpes simplex virus infection of mother and newborn. Pediatrics 73: 489-494. Whitley RJ, Corey L, Arvin A e t al (1988) Changing presentation of herpes simplex virus infection in neonates. Journal oflnfectious Diseases 158: 109-116. Whitley RJ, Arvin AT Prober C et al (1991a) A controlled trial comparing vidarabine with
HERPES SIMPLEX VIRUS INFECTION IN PREGNANCY
acyclovir in neonatal herpes simplex virus infection. New England Journal of Medicine 324: 444-449. Whitley ILl, Arvin A, Prober C et al (1991b) Predictors of morbidity and mortality in neonates with herpes simplex virus infections. New England Journal of Medicine 324: 450--454. Wittek AE, Yeager AS, Au DS & Hensleigh PA (1984) Asymptomatic shedding of herpes simplex virus from the cervix and lesion site during pregnancy: correlation of antepartum shedding with shedding at delivery. American Journal of Diseases in Childhood 138: 43%442. Yeager AS & Arvin AM (1984) Reasons for the absence of a history of recurrent genital infections in mothers of neonates infected with herpes simplex virus. Pediatrics 73: 188-193. Yeager AS, Moris JE & Prober CG (1979) Storage and transport of cultures for herpes simplex virus type 2. American Journal of Clinical Pathology 72: 977-979.