Variability in HSV Testing and Empiric Treatment in Febrile Infants

Abstract

Source: Treasure JD, Shah SS, Hall M, et al for the Pediatric Research in Inpatient Settings Network and the Collaborative Antiviral Study Group. Variation in diagnostic testing and empiric acyclovir use for HSV infection in febrile infants. Hospital Pediatrics. 2021;11(9):922–930; doi.10.1542/hpeds.2020-003129Investigators from multiple institutions conducted a retrospective study to describe testing for herpes simplex virus (HSV) in febrile infants <60 days old, assess variation in testing and treatment across hospitals, and evaluate the effect of HSV testing on length of hospital stay (LOS) in these patients. For the study, data in the Pediatric Health Information System (PHIS) database were reviewed. PHIS includes administrative data such as demographics, diagnostic and procedure codes, and laboratory and pharmacy orders from encounters at pediatric hospitals. For the current study, data from 44 hospitals in 26 states and the District of Columbia were reviewed. Study participants were children ≤60 days old evaluated in the ED with an ICD-9 or ICD-10 code indicative of fever. Data abstracted on these children included use of HSV testing (culture or PCR), imaging and other diagnostic evaluations, hospital, demographics, orders for acyclovir, LOS, ICD-9 or ICD-10 codes for HSV infection, and severity of illness, as estimated in All Patient Refined-Diagnosis Related Groups (APR-DRG). The primary study outcome was rate of HSV testing, measured at the patient and hospital level. The hospital-level rates of acyclovir orders were also assessed. Secondary outcomes included rate of HSV infection in study patients. LOS was compared in study patients tested for HSV and those not tested using regression analyses and accounting for confounding variables such as age, hospital and APR-DRG estimated illness severity. Analyses were stratified by age (<22 days and 22–60 days).Data were analyzed on 24,534 ED encounters, including 6,027 (24.6%) patients <22 days old and 18,508 (75.4%) 22 to 60 days old. There were 108 children (0.44%) with a diagnostic code indicative of HSV infection, yielding rates of HSV disease of 1.08% for those <22 days old, and 0.23% for those 22 to 60 days of age. Rates of HSV testing were 39.4% and 12.0%, respectively, for children <22 and 22 to 60 days old. At the hospital level, the rate of HSV testing was 35.6% (interquartile range [IQR], 28% to 53%) for the younger age cohort, and 12% (IQR, 9% to 16%) for those 22 to 60 days old. There also was substantial variation in hospital-level rates of empiric acyclovir use (IQR, 68% to 90% for children <22 days old and 44% to 73% for those 22 to 60 days of age). The prevalence of additional testing also varied by age group and hospital. After adjusting for age, severity of illness and hospital, LOS was significantly longer in patients with HSV testing than in those not tested (mean LOS, 2.6 and 1.9 days, respectively; P <0.001).The authors conclude that substantial variation in diagnostic use of HSV testing in febrile infants ≤60 days old.Dr Winer has disclosed no financial relationship relevant to this commentary. This commentary does not contain a discussion of an unapproved/investigative use of a commercial product/device.The diagnosis and care of potential meningitis/encephalitis in young infants with fever has continually evolved over time (See AAP Grand Rounds. 2010;24[2]:17). As hospitals have moved away from HSV culture and toward HSV PCR testing and the delay in obtaining results has shortened, the safety and cost-effectiveness calculations leading to optimal strategy continue to change.1Although the current investigators were limited in their ability to evaluate duration of empiric acyclovir in children for whom HSV testing is negative, this likely has shortened over time. The LOS was longer for patients who had HSV testing performed, likely related to clinical concern by the physicians that led to the decision to test in the first place.HSV testing continues to evolve, with many hospitals starting to use multiplex PCR array panels which test for multiple viral (including HSV), bacterial, and even fungal pathogens.2 Some evidence exists that initial CSF results are not sensitive enough to preclude a meningitis/encephalitis panel,3 but concerns also exist about overdiagnosis and overtreatment with universal use of these panels.Significant variability exists in the testing and empiric acyclovir usage in febrile infants. Patients tested for HSV have longer LOS than those for whom no test is performed.Fever in young infants is relatively common; neonatal HSV infection, relatively rare. Unfortunately, the results of the current study underscore the need for guidance to allow informed HSV diagnostic and therapeutic decisions. (See AAP Grand Rounds. 2018;39(5):56.)4