Impact of FilmArray Meningitis Encephalitis Panel on HSV Testing and Empiric Acyclovir Use in Children Beyond the Neonatal Period

Abstract

Following implementation of the FilmArray meningitis and encephalitis panel, which enables rapid syndromic cerebrospinal fluid testing, HSV testing doubled in children >60 days with suspected central nervous system infection at Children’s Hospital Colorado. Acyclovir initiation was unchanged, but duration decreased. Diagnostic and antimicrobial stewardship is needed for MEP optimization.

Introduction

Empiric acyclovir for herpes simplex virus (HSV) in children beyond the neonatal period undergoing lumbar puncture for suspected central nervous system (CNS) infection doubled from 7.5% in 1999 to 15.6% in 2012 in US pediatric hospitals, while HSV incidence remained unchanged (0.22%)¹. Acyclovir initiation within a day improves outcomes for confirmed HSV CNS infection², but widespread empiric use while awaiting testing results carries significant nephrotoxicity and costs^1,3.

Duration of empiric acyclovir therapy in low-risk children is determined by turnaround time of cerebrospinal fluid (CSF) HSV testing results, which varies widely (hours-days)⁴. A new syndromic multiplex PCR panel (FilmArray Meningitis Encephalitis Panel [MEP]; BioFire Diagnostics/Biomerieux, Salt Lake City, UT) rapidly detects 14 pathogens, including HSV-1 and HSV-2, directly from 200μL of cerebrospinal fluid (CSF) in 1–2 hours⁵. A single retrospective observational study in a combined adult and pediatric patient population who had a lumbar puncture and received acyclovir (n=208) demonstrated a median 10.8 hour decrease in acyclovir duration after MEP implementation⁶. However, the impact of MEP implementation on CSF HSV testing practices and acyclovir initiation in children beyond the neonatal period with suspected CNS infection is unknown.

Methods

We retrospectively compared CSF testing and acyclovir use in a pre-MEP era (1/1/2007–1/22/2017) to a post-implementation era of MEP (1/23/2017–12/31/17) amongst children >60 days with a CSF specimen sent to the Children’s Hospital Colorado (CHCO) microbiology laboratory for bacterial culture or viral PCR testing. A subanalysis compared children with CSF testing HSV negative by single-plex PCR through both eras with those testing HSV negative on MEP. CSF HSV singleplex-PCR testing was available throughout both eras, conducted following extraction using Virus Minikits 2.0 (Qiagen, Valencia, CA) on the BioRobot EZ1, using the MultiCode-RTx HSV 1&2 Kit (Luminex, Austin, TX). Following implementation, MEP was performed on-demand 24 hours/day, 7 days/week. Electronic medical record clinical decision support recommended ordering MEP if providers had concern for more than one pathogen, and discouraged ordering MEP in immunocompetent children >2 months without encephalitis with <5 white blood cells (WBCs)/μL in CSF.

Demographic and clinical characteristics were compared using Wilcoxon Rank-Sum tests, Fisher’s exact tests and chi-squared tests for continuous and categorical variables. Significance was set at 0.05. R version 3.4.1 software (R Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org/) was used for analysis.

Results

5,296 CSF specimens from children >60 days of age were included with similar demographics in both eras (Table 1). The proportion of CSF specimens from children with suspected CNS infection tested for HSV in CSF (MEP or HSV PCR) doubled from 25% (1225/4888) in the pre-MEP era to 54% (219/408) in the MEP era (p<0.001; Figure 1A). In the MEP era, HSV testing was conducted by MEP in 96% and HSV PCR in 8% of cases and by both methods in 4% of cases. In both eras, most HSV-tested CSF specimens had less than 5 white blood cells/μL (63% vs 59%, p=0.27).

Table 1.

Clinical characteristics of children >60 days with cerebrospinal fluid tested for suspected central nervous system infection in the Children’s Hospital Colorado microbiology laboratory.

Characteristic	Pre-MEP era (n = 4888)	MEP Era (n = 408)	P
Age (median years: IQR)	5 (1. 13)	6(1. 13)	0.134
2–6 months	997 (20%)	75 (18%)	0.354
6 months-1 year	430 (9%)	28 (7%)
1-4 years	938 (19%)	81 (20%)
>=4 years	2523 (52%)	224 (55%)
Male	2479 (51%)	199 (49%)	0.35
White Race	3159 (67%)	264 (66%)	0.933
CSF Pleocytosis (> 5 WBCs)	1048 (24%)	106 (28%)	0.097
Positive CSF Bacterial Culture*	70 (1%)	6 (2%)	1
Positive CSF Enterovirus Test*	153 (10%)	11 (5%)	0.033
CSF Tested for HSV (MEP or HSV PCR)	1225 (25%)	219 (54%)	<0.0001
Tested by HSV PCR	1225 (25%)	18 (4%)	<0.0001
Tested by MEP	0 (0%)	211 (52%)	<0.0001
Positive CSF HSV Test*	8 (0.6%)	0 (0%)	0.23
Tested for HSV with CSF WBC <5/μL	720 (63%)	121 (59%)	0.27
Received acyclovir**	648 (13%)	48 (12%)	0.43
Median doses of acyclovir (IQR)**	5 (3.8)	3(1.9)	0.05

^*Denominator includes only those tested.

^**Excludes the 8 children with positive CSF HSV tests in the pre-MEP era.

Fig. 1.

A. Percentage of CSF specimens sent to CHCO microbiology laboratory from children >60 days tested for HSV from 2007–2017 by year with delineation betvwen the pre-MEP era from 2007–2016 and post-MEP era in 2017. The percentage of CSF specimens tested by single-plex HSV PCR is represented with a dashed line, tested by MEP with a dotted line, and tested for HSV by any method (single-ptex HSV PCR or MEP) by the solid line B. percentage of CSF specimens sent to CHCO microbiology laboratory from children >60 days in the pre-MEP era (above) and MEP era (below) undergoing HSV testing (bracketed) and started on acyclovir (+ACV. yellow) or not started on acyclovir (no ACV. orange), or not tested for HSV (No HSV test, blue).

Overall, a similar proportion of children with suspected CNS infection received acyclovir in the pre-MEP and MEP eras (13% vs. 12%; Figure 1B), despite a low rate of HSV positivity (0.6% vs. 0%). In subanalysis throughout both eras amongst children with negative CSF testing for HSV, those tested HSV negative by MEP were less likely to be started on acyclovir than those with tested HSV negative by singleplex PCR (18% vs. 50%, p<0.001) and, amongst those started, acyclovir was discontinued sooner, after a median 2 (IQR: 1–5) vs 5 (IQR: 3–8) doses (p=0.001).

Discussion

Syndromic testing by MEP was rapidly adopted and transformed CSF diagnostic testing at our institution, doubling testing for HSV amongst children >60 days with suspected CNS infection. MEP implementation decreased empiric acyclovir duration amongst those started on acyclovir similar to Evans et al.⁶ likely due to decreased turnaround time. MEP failed to impact empiric acyclovir initiation rates likely due to more widespread syndromic testing with MEP outside of those at high-risk for HSV. This study suggests that optimized use of newer rapid syndromic tests has potential to improve antimicrobial use, but must overcome challenges with over-testing, clinical uncertainty, and creeping empiricism.

Diagnostic and antimicrobial stewardship efforts may help optimize MEP impact. In high-risk neonates and children with suspected encephalitis, rapid CSF HSV testing should be routinely conducted⁷. However, low-risk children beyond the neonatal period without suspected encephalitis and without CSF pleocytosis on lumbar puncture to suggest meningitis are unlikely to have a CNS infection and may provide an opportunity for diagnostic stewardship of MEP. Screening criteria for deferring viral PCR testing in immunocompetent adults with normal CSF indices decreased testing by 46% without missing clinically significant results⁸, and, applied to HSV testing alone, could save $127–158 million annually in the United States⁹. In high-risk neonates and children with suspected encephalitis, acyclovir should be initiated immediately and in some cases continued in spite of initial negative CSF HSV results⁷ due to decreased sensitivity of early testing¹⁰ (regardless of type of HSV test). However, low-risk children without suspicion for encephalitis who have negative HSV test results within 2 hours may provide an opportunity for antiviral stewardship of acyclovir enabled by MEP. A recent study in a population of non-high-risk neonates found that initiating acyclovir only after a positive rapid HSV PCR test result or change in clinical status decreased acyclovir use more than three-fold without delays in acyclovir initiation for confirmed HSV¹¹. More directive diagnostic stewardship targeting MEP testing to high-risk patient populations, including those with suspected encephalitis or CSF pleocytosis, and antiviral stewardship targeting acyclovir use to those with positive HSV testing on MEP or with a high clinical suspicion of HSV encephalitis could further improve clinical impact and cost-effectiveness.

Abbreviations:

CNS

central nervous system

CSF

cerebrospinal fluid

MEP

meningitis encephalitis panel

CHCO

Children’s Hospital Colorado