Neonatal Herpes Simplex Virus Infections: New Data, Old Conundrum

Infections with herpes simplex virus (HSV) in the neonatal period represent an ongoing challenge for practicing clinicians. Neonatal HSV has a high risk of mortality and significant neurologic disability among survivors, and timely initiation of treatment with acyclovir is needed to impact outcomes.¹ Treatment delays of even 1 day have been associated with a greater than twofold increase in odds of mortality, yet these infections can be subtle on presentation and therefore difficult to identify expeditiously.^2,–4 Furthermore, there is an added layer of complexity, which stems from increasing evidence that strategies of excessive testing and empirical treatment among infants without HSV are costly and increase the risk of adverse events, including medication error, intravenous line complications and/or infiltration, and disruption of breastfeeding (acyclovir itself is generally safe in neonates).^5,–8 

In this issue of Pediatrics, Mahant et al⁹ provide a retrospective assessment of the epidemiology of neonatal HSV in a large Medicaid population and demonstrate an increase in incidence over a 7-year period. They further demonstrate a high financial burden in these patients, including costs related to ongoing or recurrent morbidity among survivors.

The observation that the rate of neonatal HSV is increasing is important, although these initial findings should be considered with some caution. There are additional potential explanations for the observed rate increase that should be explored in future research. These include measuring the impact of more widespread use of polymerase chain reaction (PCR)–based diagnostics, which appear in preliminary studies to be more sensitive than traditional culture methods and may result in more cases being diagnosed.^10,11 Similarly, recent (2013) national recommendations for screening of asymptomatic infants exposed to HSV at the time of delivery may also have increased case numbers.¹² However, if increasing rates are indeed confirmed, the hypothesis described by the authors that this shift is mediated in part through changing HSV susceptibility patterns among women of childbearing age coupled with changes in oral sex behavior should be explored further. Such knowledge could lead to novel preventive interventions and strategies for patient education.

With their work, the authors contribute further nuance to a complicated and ongoing question: how do we correctly identify all infants with neonatal HSV in a timely manner while avoiding subjecting large numbers of children to unnecessary tests and empirical treatments? The annual rate calculated by Mahant et al⁹ is ∼3 times higher than childhood cancer, which is certainly at a level that a busy practicing physician should expect to encounter, perhaps multiple times, in a career.¹³ Therefore, it is essential that practitioners remain well educated about the risk factors and clinical signs of neonatal HSV and remain vigilant when assessing an infant with fever or other signs of illness. Concurrently, clinicians must also maintain a nuanced understanding of scenarios in which HSV is much less likely (eg, the second month of life) and the potential risks and costs of unnecessary testing and treatment.

This conundrum also represents a significant challenge for stakeholders developing guidelines and institutional protocols. There is no uniform consensus as to whether HSV testing or empirical treatment should be selective or comprehensive among febrile infants, and US hospitals employ a range of strategies.¹⁴ Current recommendations from the American Academy of Pediatrics identify and emphasize the importance of recognition of the factors associated with increased likelihood of HSV infection but do not specify a more comprehensive (eg, all febrile infants) strategy.¹⁵ 

Irrespective of the selected approach to this question, it is important to recognize that although guidelines and protocols can incorporate concrete data points (eg, age, maternal exposure, presence of vesicles, results of testing for alternate explanations of fever, etc), they cannot universally capture less tangible but nevertheless essential factors (provider experience, institutional resources, changes or stability of patient appearance over time, reliability of parents, opinion of colleagues, etc) on which clinicians rely every day in medical decision-making. Therefore, stakeholders should endeavor to provide guidance that allows flexibility for practitioners on the front lines.

Moving forward, this debate is likely to be transformed by increasing availability of rapid PCR testing for HSV, including multiplex assays with results available in a matter of 2 to 3 hours. For example, as testing becomes more efficient, it will be necessary to reassess if a delay in acyclovir initiation of a few hours confers the same degree of risk as delays >1 day; if not, rapid results may allow uncoupling testing from empirical treatment in more patients. As another example, authors from Cincinnati have instituted an approach in which febrile infants <21 days of age undergo PCR testing of the cerebral spinal fluid for HSV, but only those with further risk factors undergo a full HSV workup and receive empirical acyclovir.¹⁶ 

Identifying neonates with HSV without undue risk to infants without infection is a long-standing conundrum in pediatrics; the pathway to better clarity will depend on researchers and clinicians such as Mahant et al⁹ who continue to provide important data and ask critical questions.

HSV

herpes simplex virus

PCR

polymerase chain reaction