Use of Guthrie Cards for the Early Diagnosis of Neonatal Herpes Simplex Virus Disease

Neonatal herpes simplex virus (HSV) infection acquired at delivery is characterized by high morbidity and mortality. Prompt therapy with vidarabine or acyclovir improves the prognosis in terms of both mortality and psychomotor development.1, 2 In the clinical setting the doctor's principal problem is to diagnose the disease as early as possible so that treatment can be started promptly. This is mainly hampered by the ability of HSV to mimic other neonatal infections, sepsis being a frequent clinical presentation often in the absence of skin vesicles.1 Moreover the greater likelihood of a bacterial etiology, unrecognized HSV infection in the mother and a viral diagnostic facility that is either unavailable or difficult to reach can delay the correct diagnosis. On the basis of our experience acquired in the use of Guthrie cards to diagnose congenital cytomegalovirus (CMV) infection3 and on the observation that HSV is spread by the hematogenous route in neonates,2 we thought it interesting to investigate whether neonatal HSV infection could be diagnosed by seeking HSV DNA in dried blood spots (DBS) on the Guthrie cards used in screening for metabolic and genetic disorders. We therefore tested retrospectively the cards of four babies in whom a neonatal HSV infection had been suspected. In addition as a control we tested DBS obtained from 41 randomly selected neonates clinically healthy at birth and from 32 neonates without clinical signs of HSV neonatal disease in the first month of life. The latter group consisted of 6 babies with bacterial sepsis ascertained in the first week of life and cured by adequate antibiotic treatment, 7 babies born to women with HSV infection (HSV-2 genital lesions detected at delivery in 3 cases and with suspected serology during pregnancy in 4) and 19 infants with asymptomatic congenital CMV infection ascertained by virology. The relevant clinical and laboratory data of the cases are summarized in Table 1. Babies 1, 2 and 4 were delivered vaginally without complications in local hospitals to women without evidence of herpetic lesions at delivery. The mother of Case 1 was hospitalized the day before delivery with slight fever, lymphocytosis, signs of aspecific inflammation and genital hyperemia. The third baby was born by precipitate delivery to an HIV-seropositive woman receiving zidovudine. A transient positivity for HSV IgM in this mother in the fourth month of pregnancy was reported but no lesion was evident at delivery.TABLE 1: Summary of clinical and laboratory data of the patients studied with dried blood spot (DBS) All bacteriologic investigations, including those of blood, cerebrospinal fluid and urine samples, were negative. HSV-1 and HSV-2 DNA were sought in the DBS, under strict control measures, by elution of the blood from the card in minimum essential medium followed by extraction by heat treatment (55°C for 60 min, 100°C for 7 min) and amplification using a nested PCR.3 A region in the gpD gene and one in the gpG gene were amplified for HSV-1 and HSV-2, respectively, as described by Aurelius et al.4 As a control a nested PCR was performed to seek CMV DNA.3 The products of amplification were demonstrated by agarose gel electrophoresis. Each sample was tested at least three times. The DBS, which had been collected routinely before any transfusion and stored at the regional screening laboratory under standard conditions, were positive for HSV-2 in Cases 1 to 3 and for HSV-1 in Case 4 in all tests performed. All samples were negative for CMV and for the serotype not involved. The DBS of the controls were negative for HSV-1 and HSV-2 except in two babies whose Guthrie card revealed the HSV-1 genome at amplification. One of these infants was a clinically healthy child and the second had sepsis caused by group B beta-hemolytic Streptococcus. Positivity was observed in only three of six tests in each of the two children. Unfortunately it is impossible to perform a virologic examination in these two children so that the significance of the finding of HSV-1 DNA remains unclear. CMV DNA was demonstrated exclusively in the DBS from children with congenital CMV infection. Overall the DBS test for the diagnosis of neonatal HSV disease had a sensitivity of 100% (4 of 4) and a specificity of 97% (71 of 73). The finding of viral DNA in the DBS of newborns with (disseminated) HSV infection indicates that viremia may precede the appearance of symptoms by some days (Cases 1 and 2) or accompany them, as in Cases 3 and 4, although DBS in the latter were collected late and viremia could have developed earlier. Methods for the laboratory diagnosis of neonatal HSV disease range from virus isolation from pharynx or skin lesions to HSV DNA amplification by PCR in cerebrospinal fluid.2, 5 The latter method has recently been reported to be useful in evaluating the extent of disease, determining the prognosis and assessing the efficacy of antiviral therapy.5 All these methods can be used in symptomatic children to confirm the clinical diagnosis. Testing DBS for HSV DNA represents a noninvasive, rapid and widely available method that could be centralized in specialized structures. It could be applied in all cases in which HSV infection must be considered in differential diagnosis of neonatal sepsis, particularly in the absence of herpetic lesions in the mother. This would allow the prompt institution of antiviral therapy, possibly contributing to the success of the treatment. Moreover use of the DBS method to screen babies asymptomatic at birth, born to women with a history of HSV infection before or during the pregnancy, would allow the identification of children at risk of developing the disease. Thorough examination and monitoring of those babies would improve the knowledge of the natural history of neonatal HSV disease and possibly lead to definition of prophylactic antiviral treatment. Maria Barbi, D.S.C. Sandro Binda, D.S.C. Valeria Primache, B.S.C. Alberto Tettamanti, M.D. Carlo Negri, M.D. Carlo Brambilla, M.D. Istituto di Virologia; University of Milan; Milan, Italy (MB, SB, VP) Unità di Terapia Intensiva Neonatale; Ospedale Valduce; Como, Italy (AT) Divisione di Patologia Neonatale; Ospedale Multizonale; Varese, Italy (CN) Divisione di Patologia Neonatale; Ospedale Niguarda; Milan, Italy (CB)