Varicella Zoster

Varicella-Zoster Virus Infections in Pregnancy

Varicella-zoster virus can cause a distinct congenital syndrome, a potentially fatal neonatal infection and life-threatening maternal illness. Physicians can reduce morbidity from these conditions by advising nonimmune pregnant women to avoid exposure to chickenpox and herpes zoster and, when indicated, by promptly administering varicella-zoster immune globulin. When prevention fails, acyclovir may be effective therapy. The concept of viral teratogenesis was introduced in 1941 after an Australian ophthalmologist noted congenital cataracts in the offspring of women who had rubella during pregnancy.[1] Of the well-known viral exanthems of childhood, rubella is still considered the most significant teratogen. Herpesviruses, particularly cytomegalovirus (CMV) and the herpes simplex viruses, are widely recognized for their ability to cause serious fetal or neonatal infections.[2] It is less well recognized that another herpesvirus, the varicella-zoster virus (VZV), is also capable of causing a distinct congenital syndrome and a fatal neonatal infection, as well as life-threatening maternal illness.[3] This article reviews the important concepts about the two VZV syndromes, varicella (chickenpox) and herpes zoster (zoster, shingles), as they relate to pregnancy.

History

The term "varicella" dates back to at least the 1700s and is a modernized Latin diminutive of variola (smallpox). The origin of the term "chickenpox" is less clear, but it may derive from the French word for chick-pea (chich-pea), a small pea. "Shingles," used as early as the 1300s, and "zoster" come from the Latin and Greek, respectively, for girdle.[3]

Von Bokay, in the 18th century, noted that varicella and herpes zoster were associated. In 1943, Garland proposed that herpes zoster was a reactivation of latent VZV, and in 1953, Weller provided evidence that varicella and zoster were indeed caused by a single virus.[4,5] LaForet and Lynch[6] described the congenital VZV syndrome in 1947, but this pattern of fetal anomalies occurring after maternal varicella was not reported again for 20 years.[1] Since then, a number of well-documented cases have been reported, and the syndrome has received considerable attention in the pediatric, obstetric, occupational medicine and general medical literature.[5,7-12]

The Virus

VZV is a member of the herpesvirus group, which includes four other human viruses, specifically Epstein-Barr virus, CMV, and herpes simplex viruses 1 and 2.[3] Based on DNA homology, the kinship between the herpes simplex viruses and VZV appears to be extremely close.[5] The only natural host for VZV is man, a fact that hinders research.[1]

Clinical Features

VARICELLA

Chickenpox ranks as one of the most communicable human diseases; the secondary household attack rate is greater than 90 percent among nonimmune contacts.[5] Immunity can be demonstrated in 85 to 95 percent of young adults living in temperate climates.[1,2,13] Despite the high attack rate, the modes of transmission are not completely understood. Virologists believe that respiratory droplets are a major source of infectivity. Airborne spread has been documented, even in the hospital setting.[3,5] Chickenpox that occurs from exposure to herpes zoster is the result of transmission is plausible because the vesicles of both varicella and zoster are full of infectious virus. Crusts and fomites are not a major source of transmission.[3]

A clear history of chickenpox reliably predicts immunity.[5,14,15] Even in the absence of a history, up to 75 percent of adults show serologic evidence of immunity.[13,16] Subclinical infection is uncommon; seropositive adults without a history of chickenpox are believed to have had a clinical infection that was forgotten.[3]

The well-known clinical features of chickenpox include an eruption of a few to thousands of pruritic, erythematous-based vesicles, which usually begin on the head and neck and progress to the trunk. In normal hosts, a two- to three-day prodrome of fever and malaise may precede the eruption.[3] Infectivity begins approximately two days before the onset of the eruption and continues until the lesions are crusted. New lesions form for about four days, and most lesions become crusted by about the sixth day after eruption.[3,5] In about 99 percent of cases, chickenpox develops between nine and 21 days after exposure to the virus,[1,5] with the usual incubation period being 13 to 17 days.[3]

HERPES ZOSTER

Ten to 20 percent of all persons have zoster at some time during their lives, with the frequency increasing with age.[5] The condition is caused by reactivation of VZV, often decades after the episode of chickenpox.[3,4,17] Recent evidence suggests that waning cellular immunity is the major factor leading to reactivation.[5,9] There is abundant evidence that zoster is not caused or triggered by exposure to an active case of chickenpox; the development of zoster in this setting represents the coincidental occurrence of two common diseases.[3,5]

In the immunocompetent host, clinical features of herpes zoster include grouped vesicles on an erythematous base, distributed unilaterally in a pattern over one or two adjacent dermatomes.[17] A prodrome of pain in the dermatome may precede the eruption.[5] New lesions may continue to erupt over four days and may last for four weeks. Pain usually accompanies the vesicles and, particularly in the elderly, may continue after resolution of the lesions. Occasionally, ten to 20 widely scattered vesicles will be found outside the involved dermatomes; this does not indicate dissemination.[17] Zoster sometimes recurs; when this appears to be the case, recurrent zoster should be differentiated from zosteriform herpes simplex by culture.[3,5]

Patients with zoster who are otherwise healthy are not more likely to have malignancies than are age-matched control subjects.[5,17] However, because of the increased incidence of herpes zoster among people infected with the human immunodeficiency virus (HIV), the possibility of HIV infection must be considered in patients with zoster.[5] Administration of passive anti-body in the form of varicella-zoster immune globulin (VZIG) has not had any impact on zoster, even in persons at high risk of complications from the disease.[5,9]

VZV Infections in Pregnancy

Because of the efficiency of VZV transmission in childhood, most pregnant women are immune to the virus. Hence, varicella is uncommon in pregnancy. Herpes zoster is also uncommon in pregnancy, since it occurs predominantly in older age groups. It is difficult to determine the true incidence of these infections during pregnancy. In one study of 30,000 pregnancies, the confirmed (minimum) rate of VZV infection was found to be five cases per 10,000 pregnancies.[3] Although exact statistics are not available, it is generally accepted that gestational zoster occurs less frequently than gestational chickenpox.[3,11]

The impact of VZV infections on the course of pregnancy is controversial. Some authors believe that these infections may lead to an increased risk of spontaneous abortion, fetal demise and premature labor.[1] Others disagree, citing prospective studies that fail to show a greater incidence of these adverse outcomes in VZV-infected mothers than in control subjects.[3]

MATERNAL INFECTIONS

Chickenpox is more serious in adults than in children; adults account for 2 percent of the cases and 25 percent of the mortality, which is primarily due to varicella pneumonia.[5,14] There have been numerous reports of fatal varicella pneumonia in previously healthy pregnant women. However, the selectivity of reports of fatal cases does not allow calculation of the case-fatality ratio. In reality, there is insufficient evidence that varicella pneumonia has any graver prognosis in pregnant women than in women who are not pregnant.[3] Many authorities, however, believe there have been enough case reports to justify considering pregnant women with varicella pneumonia a group at high risk of death.[5] Most authors concur that in the absence of pneumonia, chickenpox is not more serious in pregnant women than in other adults.[3]

Concerning chickenpox, the best advice for nonimmune pregnant women is that they avoid contact with anyone who has either varicella or herpes zoster. Since VZV infections are rarely clinically silent, exposure to unrecognized infection is probably uncommon. The exception would be exposure during the two-day, pre-eruptive, infectious stage of varicella. Fortunately, transmission does not occur from casual contact.[3]

When a patient's immunity to varicella is in question, sensitive immunoassays, such as enzyme-linked and immunofluorescent assays, can reliably detect the presence of antibody to the virus. The complement-fixation titers, although useful for detecting a rise in antibody titers in the acute and convalescent periods, wane over time and are too insensitive for prediction of immune status.[1,5]

Paryani and Arvin[10] have stated that "the maternal risk justifies the administration of varicella-zoster immune globulin to pregnant women who have no antibody to varicella and who have had a close exposure to varicella-zoster virus." Although the use of VZIG for this purpose is not among those listed by the Centers for Disease Control, a recent National Institutes of Health expert panel endorsed this additional indication.[5]

Clinical experience supports the effectiveness of VZIG in preventing or ameliorating maternal varicella,[1] although definitive studies are lacking. VZIG, 0.125 mL per kg (maximum: 6.25 mL) intramuscularly, should be given as early as possible after exposure, certainly within 72 hours.[14,18] Pregnancy is not a contraindication to the use of VZIG. Neither hepatitis B nor HIV infection has been transmitted by immune globulins, although transient, passive transfer of antibody to these viruses has occurred.[18]

If chickenpox develops during pregnancy, no special therapy is recommended in most instances. Many authors, however, recommend that antiviral therapy be started if symptomatic pneumonia develops.[19] Although vidarabine (Vira-A) is the only agent approved by the Food and Drug Administration for VZV infections, most experts now believe that acyclovir (Zovirax) is less toxic, easier to administer and perhaps more effective than vidarabine.[5,12,19,20] Acyclovir has been advocated as the drug of choice for severe varicella in pregnant women.[19] Although not labeled for use in pregnancy, acyclovir has not been implicated as a teratogen in humans, nor is it teratogenic in animals in doses comparable to those used in humans.[5,21]

In pregnant women, zoster presents a different set of concerns than chickenpox does. For example, zoster may be a sign of HIV infection. Appropriate risk assessment, counseling and laboratory testing should be done in these patients.[5]

In one study,[5] 13 percent of patients thought to have herpes zoster were found to have zosteriform herpes simplex by viral. culture. Any lesion in the lumbosacral dermatomes suspected of being zoster-related should be cultured to confirm the diagnosis. Zoster should be treated conservatively with symptomatic therapy. Corticosteroid therapy is not indicated in the treatment of young patients, including pregnant women.[1,5] Administration of VZIG does not modify zoster. There is no known method of preventing reactivation of VZV, and thus avoiding zoster.

FETAL VZV INFECTIONS

Epidemiologic studies have failed to show an increased incidence of anomalies in infants born to mothers with VZV infections during pregnancy. Authorities, however, discount the ability of epidemiologic studies to detect an uncommon complication of an infection that is uncommon in pregnancy. Proof that the syndrome exists rests on reports from around the world of infants with the same pattern of anomalies following maternal VZV infection.[1,3,5] Intrauterine VZV infection was confirmed by serologic tests in many of these cases. However, culturable virus was not present after birth because the infection was acquired and resolved before delivery. Therefore, the cause of the congenital VZV syndrome is most appropriately considered to be a transient infection, rather than an active, persistent infection. As such, it is not transmissible.

The physical stigmas of congenital VZV syndrome consist of some combination of (1) scarred, segmental or dermatomal skin lesions; (2) limb deformities, almost universally ipsilateral and distal to the skin signs, and (3) central nervous system and/or ocular abnormalities.[3,9,22] The cutaneous fibrotic scarring is usually depressed and hyperpigmented, and is found overlying or just proximal to musculoskeletal abnormalities. The limb deformities are generally unilateral, with hypoplasia or the absence of digits and impaired bone development. In the lower extremity, talipes equinovarus (clubfoot) may occur. Motor and sensory deficits may also be present.

Central nervous system involvement includes cerebral cortical atrophy, mental retardation, seizures, bulbar palsy, encephalitis and intracranial calcification. Ocular abnormalities, present in most cases, include microphthalmia, corneal opacities, cataracts, chorioretinitis, optic atrophy, anisocoria and nystagmus.[3] Affected children often die in infancy of gastroesophageal reflux and recurrent aspiration pneumonia.[2]

The earlier in pregnancy a mother acquires varicella, the greater the infant's apparent risk of congenital VZV, although at least one case of the syndrome has been reported after maternal varicella at 28 weeks' gestation.[3] It must be emphasized that only a small percentage of infants of mothers who have chickenpox early in gestation will develop the syndrome. The actual risk can only be estimated.

When the results of three prospective studies[1,10,23] were combined, the total risk of congenital VZV syndrome was found to be 4.9 percent (three of 61 infants) following first-trimester varicella infection.[10] The risk of the syndrome from second- or third-trimester maternal varicella could not be estimated, but the risk can be safely presumed to be small (Table 1). In the only series identified as reporting such figures, two (15 percent) of 13 women elected to terminate their pregnancies after VZV infection.[10] Other authors concur that the majority of women choose to continue their pregnancies after VZV infection, recognizing that the overwhelming majority of infants appear normal at birth.[22]

Gershon[22] proposed, and Higa and colleagues[7] detailed, a mechanism that explains the clinical picture of congenital VZV syndrome. The appearance of infants with this syndrome suggests that both varicella and zoster develop in utero. The distribution of the skin and limb abnormalities is suggestive of zoster. The musculoskeletal and other segmental dysfunctions generally arise from the same spinal levels as the skin dermatomes involved. The segmental problems may occur secondary to involvement of the spinal cord and peripheral nerves within the dermatome--established effects of zoster. Since both acquisition and reactivation of the virus occur, it is easy to explain why only a small percentage of exposed fetuses develop the syndrome. The earlier in gestation the fetus acquires the VZV, the longer the period of risk for reactivation; hence, maternal chickenpox in the first trimester creates the greatest risk for congenital VZV syndrome.

Prevention of maternal varicella is the only known means of preventing congenital VZV syndrome. When gestational varicella occurs, the parents should be appropriately counseled about the potential fetal effects. Because of the infrequency of congenital VZV syndrome, there are no studies on whether administration of VZIG or antiviral therapy to mothers reduces the likelihood or the severity of the syndrome. The absence of an animal model approximating human varicella further hinders the development of evidence regarding possible efficacy. If maternal varicella can be prevented by counseling, postexposure VZIG administration or preconceptual immunization, the fetus will not be at risk. These strategies are preventive, and physician awareness is the key to prevention.

Fetuses exposed to maternal VZV infection are at risk of zoster in early childhood. Interestingly, these episodes of zoster occur in the absence of postnatal chickenpox. Zoster, being a reactivation of latent VZV, should not occur without previous exposure to the VZV. In children with zoster but no history of chickenpox, there is often a history of maternal VZV infection during the pregnancy.[1] These observations led to the speculation that in such instances VZV is acquired in utero.[1,3]

Zoster is rare in children under two years of age, with the exception of those exposed in utero.[3] Therefore, zoster occurring before two years of age should stimulate a search for other possible manifestations of intra-uterine varicella, such as chorioretinitis or central nervous system dysfunction.[9]

Another concern is the possibility of future malignancy in the offspring of women with varicella during pregnancy. This concern is supported by in vitro, in vivo and clinical studies. In vitro studies have shown that VZV induces a 20-fold increase in chromosomal abnormalities in infected cells as compared with control cells. In vivo, peripheral blood leukocytes show about a fourfold increase in the number of chromosomal breaks in infected patients, as compared with control subjects. In one clinical study of 270 women with varicella during pregnancy, it was noted that two of the off-spring died of leukemia, which is a statistically significant excess of leukemia-related deaths. This area requires further investigation before conclusions can be drawn about the risk of malignancy.[3]

Despite these concerns, fetal prognosis is believed to be altered only slightly by maternal varicella during the second half of pregnancy (until the perinatal period), and no therapy is indicated to improve the fetal outcome.

There are only a few reports of anomalies suggestive of congenital VZV syndrome in infants whose mothers developed zoster during pregnancy.[3, 7] However, it is estimated that thousands of pregnancies per year are complicated by maternal zoster.[1] Patients may be counseled that the risk to the fetus from maternal zoster is negligible.[1,2] Therapy aimed at reducing fetal risk is inappropriate.[2]

PERINATAL VZV INFECTIONS

Varicella in neonates can be divided into cases acquired in utero and those acquired postnatally. Varicella infections acquired transplacentally are called congenital chickenpox. A subset of infants with congenital chickenpox has a 31 percent case-fatality rate; these high-risk infants are born to mothers in whom chickenpox develops from five days before to two days after delivery.

When chickenpox is transmitted from the mother to the fetus, both the virus and the antibody to the virus may be transferred or just the virus may be transferred. If chickenpox develops more than five days before delivery, enough maternal antibody can be produced and transmitted to the fetus to protect the neonate against severe congenital chickenpox. Fetuses exposed from 20 days before delivery (the maximum incubation period) to six days before delivery may have mild postnatal chickenpox, with low morbidity and mortality. Chickenpox develops after birth in about 24 percent of infants exposed during this period.[3,11]

Infants exposed in utero during the five days preceding delivery do not receive enough maternal antibody to modify the infection. Similarly, if the eruption appears in the mother within the first 48 hours post-partum, the infant is exposed transplacentally to VZV but does not receive passive, protective antibody. Infants born to mothers who have the onset of varicella from five days before delivery to two days after delivery have a 17 percent chance of manifesting congenital varicella; the case-fatality rate is 31 percent in untreated infants, for an overall mortality rate of about 5 percent (0.17 X 0.31).[3,5]

Immediate treatment with VZIG, 1.25 mL administered intramuscularly, is mandatory for all infants born to mothers who developed chickenpox during the seven-day high-risk period.[5, 22] Despite appropriate administration of VZIG, life-threatening and fatal cases of congenital chickenpox have still occurred.[12] Therefore, if varicella develops, the infant needs to be carefully followed. If there is extensive skin involvement, liver enzyme elevation or lower respiratory tract involvement, antiviral chemotherapy should be instituted immediately.[12,19] No therapy is indicated for congenital chickenpox acquired more than five days before delivery.

Infants who acquire chickenpox postnatally are at increased risk of complications and death. Even excluding neonatal infections, varicella before one year of age is associated with a mortality rate of eight per 100,000 cases, as compared with a rate of less than two per 100,000 cases in normal children over this age.[5] However, the case-fatality rate for postnatally acquired varicella is still low, and treatment recommendations are the same for this group as for older children.

The maximum incubation period of chickenpox is 20 days. If the mother develops chickenpox three weeks or more before delivery, the fetus may acquire the infection in utero. These infants have maternal antibody and therefore are not at increased risk of complications; however, they are at risk of developing zoster early in life. Any infant exposed to maternal varicella within 20 days before delivery might develop chickenpox while in the nursery and should be isolated from other infants.[3]

Table 2 summarizes the possible fetal outcomes following maternal varicella during various periods of pregnancy.

THE VARICELLA VACCINE

A potential future strategy for preventive treatment of susceptible, exposed neonates is immunization with the live, attenuated-virus vaccine. Randomized clinical trials in children have yielded a protective efficacy of 94 percent for vaccine administered within three days of exposure and an efficacy of 100 percent for protection or amelioration of disease when administered within five days of exposure. Although not addressing the subject of neonates, Arbeter and colleagues[16] have suggested that "postexposure prophylaxis could become a frequent indication for varicella vaccine before immunization of the population becomes routine." A special concern in administering the vaccine to neonates is the fear that varicella in the first year of life might lead to an increased incidence of zoster.[13]

Because of concerns about administering a live virus vaccine during pregnancy, with the possibility of viremia and fetal risk (including vaccine-induced congenital VZV syndrome), postexposure vaccination will probably not be a strategy in pregnant women. However, Paryani and Arvin[10] have suggested that "women of childbearing age who are susceptible to varicella and who are identified before pregnancy also constitute an important target population for the administration of live, attenuated varicella vaccine."

Extraordinary caution has tempered the introduction of this vaccine in the United States. Concerns that have delayed its availability are its unknown duration of protective efficacy and uncertain influence on subsequent zoster. If the protective effect were short-lived, the vaccine could delay a generally benign childhood illness until a time when it could become a higher-risk adult illness. Of the small number of patients who have been followed on a long-term basis, 97 percent have demonstrated residual antibody after ten years. Trials thus far have indicated that zoster is not a major problem with the vaccine; only one case of zoster has been identified among 8,000 healthy vaccine recipients in the United States and Japan.[4,24]

Special Situations

HEALTH CARE PROVIDERS

No precautions are needed for physicians, nurses and hospital staff who have a clear history of chickenpox. For health care providers without a history of varicella infection, serologic confirmation of immune status is recommended. Those who lack immunity should avoid contact with patients who have active VZV infection. For pregnant, varicella-immune medical personnel, there is a small risk that coincidental zoster, independent of reexposure to chickenpox, will develop. In these instances, the risk to the fetus is infinitesimal.

VARICELLA IN THE HOME

What should be done if there is a VZV infection in the home at the time of pending hospital discharge of the mother and neonate? Evidence suggests that postnatally acquired chickenpox differs little from the disease in older children,[3] except for a slightly higher case-fatality rate.[5] The conservative approach, avoiding exposure of the newborn to the infection, is recommended.[5] Either the infectious household members or the neonate and mother should stay elsewhere, until the contagious period ends. Some authorities, to avoid separating the mother and neonate, have suggested administering VZIG to the infant and allowing the mother and infant to return home together if no other options are available.[3]

LATE THIRD-TRIMESTER MATERNAL VZV

If a mother has a VZV infection one to three weeks before delivery and all lesions have healed (or progressed to crusts), the mother is no longer infectious and should be treated no differently from any other patient. However, the infant is at risk of congenital varicella, which is transmissible. Therefore, the infant should be isolated from other infants and should be cared for only by varicella-immune personnel.[3]

Final Comment

Physician awareness of congenital VZV syndrome and perinatal varicella is important. Family physicians routinely see patients with VZV infections. Although casual contact is less likely than household contact to transmit infection, it is prudent to keep nonimmune pregnant women and nonimmune medical personnel from unnecessary exposure. Failure to provide VZIG to an infant whose mother had varicella eruptions within five days before or two days after delivery could lead to infant death. The physician must also be aware that potentially fatal cases of congenital chickenpox can occur even after appropriate VZIG administration. In addition, many advocate VZIG prophylaxis for pregnant women who lack immunity to varicella and have a known, significant exposure to VZV.

In the future, varicella vaccine may be offered routinely to children or to women who lack evidence of varicella immunity before pregnancy, similar to the current usage of rubella vaccination. A safe, effective vaccine would virtually eliminate the risks associated with VZV infections during pregnancy. [Tabular Data 1 and 2 Omitted]

REFERENCES [1]Enders G. Varicella-zoster virus infection in pregnancy. Prog Med Virol 1984;29:166-96. [2]Prober CG, Arvin AM. Perinatal viral infections. Eur J Clin Microbiol 1987;6:245-61. [3]Young NA, Gershon AA. Chickenpox, measles and mumps. In: Remington JS, Klein JO, eds. Infectious diseases of the fetus and newborn infant. 2d ed. Philadelphia: Saunders, 1983;375-427. [4]Plotkin SA. Hell's fire and varicella-vaccine safety [Editorial]. N Engl J Med 1988;318:573-5. [5]Straus SE, Ostrove JM, Inchauspe G, et al. NIH conference. Varicella-zoster virus infections. Biology, natural history, treatment, and prevention. Ann Intern Med 1988;108:221-37. [6]LaForet EG, Lynch C. Multiple congenital defects following maternal varicella. N Engl J Med 1947;236:534-7. [7]Higa K, Dan K, Manabe H. Varicella-zoster virus infections during pregnancy: hypothesis concerning the mechanisms of congenital malformations. Obstet Gynecol 1987;69:214-22. [8]Gestal JJ. Occupational hazards in hospitals: risk of infection. Br J Ind Med 1987;44:435-42. [9]Weller TH. Varicella and herpes zoster. Changing concepts of the natural history, control, and importance of a not-so-benign virus. N Engl J Med 1983;309:1434-40. [10]Paryani SG, Arvin AM. Intrauterine infection with varicella-zoster virus after maternal varicella. N Engl J Med 1986;314:1542-6. [11]Stagno S, Whitley RJ. Herpesvirus infections of pregnancy. II. Herpes simplex virus and varicella-zoster virus infections. N Engl J Med 1985;313:1327-30. [12]Bakshi SS, Miller TC, Kaplan M, Hammerschlag MR, Prince A, Gershon AA. Failure of varicella-zoster immunoglobulin in modification of severe congenital varicella. Pediatr Infect Dis 1986;5:699-702. [13]Gershon AA. Live attenuated varicella vaccine. Annu Rev Med 1987;38:41-50. [14]Conte JE, Barriere SL. Varicella-zoster immune globulin. In: Manual of antibiotics and infectious diseases. 6th ed. Philadelphia: Lea & Febiger, 1988:263-73. [15]Epilogue. Pediatrics 1986;78(4 Pt 2):763-5. [16]Arbeter AM, Baker L, Starr SE, Levine BL, Books E, Plotkin SA. Combination measles, mumps, rubella and varicella vaccine. Pediatrics 1986;78(4 Pt 2):742-7. [17]Glaser RB. Clinical aspects of herpes zoster. West J Med 1983;139:718-20. [18]Weintrub PS. Uses of immune globulins in the prophylaxis and treatment of viral infections. Clin Lab Med 1987;7:897-910. [19]Bean B, Englund JA. Treatment of varicella-zoster virus infections. Clin Lab Med 1987;7:853-68. [20]Nahata MC. Clinical use of antiviral drugs. Drug Intell Clin Pharm 1987;21:399-405. [21]Dorsky DI, Crumpacker CS. Drugs five years later: acyclovir. Ann Intern Med 1987;107:859-74. [22]Gershon AA. Viral infections of infancy. Compr Ther 1983;9:9-14. [23]Siegel M. Congenital malformations following chickenpox, measles, mumps, and hepatitis. Results of a cohort study. JAMA 1973;226:1521-4. [24]Gershon AA. Live attenuated varicella vaccine. J Pediatr 1987;110:154-7.

GARY N. FOX, M.D. is clinical assistant professor of family and community medicine at Pennsylvania State University College of Medicine, Hershey. He is also associate director of the family practice residency program at Reading (Pa.) Hospital and Medical Centre. Dr. Fox graduated from Duke University of Medicine, Durham, N.C., served a family practice residency in Charlotte, N.C., and completed a fellowship in family medicine at Temple University School of Medicine, Philadelphia. JOSEPH W. STRANGARITY, M.D. is in the private practice of family medicine in Bowmansville, Pa. A graduate of Temple University School of Medicine, he completed a family practice residency at Reading (Pa.) Hospital and Medical Center.

COPYRIGHT 1989 American Academy of Family Physicians
COPYRIGHT 2004 Gale Group