ABSTRACT
Herpes simplex virus (HSV) infections of the central nervous system (CNS) are associated with significant morbidity and mortality rates in children. This study assessed the impact of a direct HSV (dHSV) PCR assay on the time to result reporting and the duration of acyclovir therapy for children with signs and symptoms of meningitis and encephalitis. A total of 363 patients with HSV PCR results from cerebrospinal fluid (CSF) samples were included in this retrospective analysis, divided into preimplementation and postimplementation groups. For the preimplementation group, CSF testing was performed using a laboratory-developed real-time PCR assay; for the postimplementation group, CSF samples were tested using a direct sample-to-answer assay. All CSF samples were negative for HSV. Over 60% of patients from both groups were prescribed acyclovir. The average HSV PCR test turnaround time for the postimplementation group was reduced by 14.5 h (23.6 h versus 9.1 h; P < 0.001). Furthermore, 79 patients (43.6%) in the postimplementation group had dHSV PCR results reported <4 h after specimen collection. The mean time from specimen collection to acyclovir discontinuation was 17.1 h shorter in the postimplementation group (31.1 h versus 14 h; P < 0.001). The median duration of acyclovir therapy was also significantly reduced in the postimplementation group (29.2 h versus 14.3 h; P = 0.01). Our investigation suggests that implementation of rapid HSV PCR testing can decrease turnaround times and the duration of unnecessary acyclovir therapy.
INTRODUCTION
Herpes simplex viruses (HSVs) are important causes of meningitis and encephalitis, particularly among children and immunocompromised patients (1). The HSV infection rate among neonates is estimated to be 1 case per 32,000 deliveries, and approximately 1,500 cases of neonatal HSV disease occur annually in the United States (2, 3). Prompt initiation of antiviral therapy is paramount, as mortality rates for disseminated neonatal infections can be as high as 80% if the infections are left untreated; 70% of survivors are left with detrimental neurological sequelae (4–6).
Patients with HSV meningitis and encephalitis typically have initial presentations of headache, fever, and altered mental status. However, neonates with HSV central nervous system (CNS) infections may have nonspecific presentations, including lethargy, irritability, poor oral intake, and temperature instability (7). Cutaneous vesicular lesions may be indicative of HSV infections in infants, but the absence of vesicular rash does not preclude neonatal HSV CNS infection in 35% of cases (8). Other infectious etiologies associated with CNS infections can present with similar symptoms (9). Thus, laboratory testing of cerebrospinal fluid (CSF) samples is necessary to confirm the diagnoses.
Timely diagnosis of an infectious etiology is beneficial, as it can aid in patient management through the initiation of appropriate therapy or the discontinuation of unnecessary treatment. The Infectious Diseases Society of America (IDSA) recommends early treatment of all suspected HSV CNS infections with intravenous (i.v.) acyclovir and diagnostic testing by PCR (9). Pending test results, it is recommended that all patients with suspected HSV infections be treated empirically with i.v. acyclovir. Prompt initiation of acyclovir correlates with decreases in mortality rates if the drug is administered within 4 days after onset (9–12). Ruling out HSV CNS infections is also important, as it allows the discontinuation of i.v. acyclovir treatment and the avoidance of unnecessary drug costs and adverse reactions, including nephrotoxicity due to precipitation of acyclovir crystals in the renal tubules (13).
The integration of rapid laboratory diagnosis of infectious diseases and antimicrobial stewardship has been shown to directly improve patient management and overall hospital economics (14–16). Although previous studies investigated the utility of PCR in providing rapid diagnosis of HSV CNS infections for both pediatric and adult patients (17–23), to our knowledge there are currently no published data on the impact of rapid diagnosis of HSV meningitis and encephalitis on patient management. Here we sought to assess the impact of a rapid HSV PCR assay offered 24 h a day, 7 days a week, for the diagnosis of HSV meningitis and encephalitis on the time to laboratory results and the duration of acyclovir therapy among pediatric patients with presumed CNS infections.
(This work was presented in part at the 26th European Congress of Clinical Microbiology and Infectious Diseases, Amsterdam, Netherlands, 9 to 12 April 2016, and ASM Microbe 2016, Boston, MA, 16 to 20 June 2016.)
RESULTS
A total of 363 patients were enrolled in the study, including 182 patients in the preimplementation group and 181 in the postimplementation group. Patient demographic features and comorbidities were compared between the two groups and, with the exception of age, no significant differences were found (Table 1). The patient ages overall ranged from 1 day to 22 years, with a median of 1.3 years for the preimplementation group and a median of 0.1 years for the postimplementation group (P < 0.001). The postimplementation group also had a significantly greater percentage of neonates (25.8% versus 37.6%; P = 0.02). Approximately 70% of all patients had no significant medical history, with a difference between the two groups of only 5.4% (preimplementation group, 122 patients [67.0%]; postimplementation group, 131 patients [72.4%]; P = 0.30). The most common conditions among the patients with comorbidities included leukemia and lymphoma, prematurity, and cardiac disease.
Characteristics of 363 patients from the preimplementation and postimplementation groups with HSV-1 and HSV-2 PCR testing
Some of the overlapping clinical presentations seen in the preimplementation and postimplementation groups were fever, seizure/altered mental status, cough/congestion, vomiting/nausea, irritability/fussiness, and headache (Table 2). There were no significant correlations between the presence of any of those symptoms and the initiation of acyclovir therapy, and all orders were at the discretion of the primary physician. Fever was the most common symptom reported by patients in this study, i.e., 184 patients (51.7%) in total, 78 patients (42.9%) in the preimplementation group, and 106 patients (58.6%) in the postimplementation group (Table 2). Of the 184 patients, 67.4% (124 patients) were prescribed acyclovir, compared to 63.7% of patients (114/179 patients) who were without fever but were prescribed acyclovir (P = 0.51) (Table 2). Abnormal glucose and/or protein levels were detected in 193 patients (53.2%) and did not correlate with the initiation of acyclovir treatment.
Comparison of commonly reported symptoms and initiation of intravenous acyclovir therapy
All CSF specimens tested in this study were HSV negative by PCR. Seven patients (1.9%) tested HSV positive from alternative sources (three blood specimens and four swabs from vesicular lesions). Two of the 3 patients with blood specimens positive for HSV also had positive vesicular lesions. Further microbiological investigation yielded positive CSF results for 20 (5.5%) of the 363 patients, including 8 patients (4.4%) in the preimplementation group and 12 patients (6.6%) in the postimplementation group. Infectious etiologies included 17 viruses, 2 bacteria, and 1 fungus (Table 1).
Implementation of the direct HSV (dHSV) PCR assay decreased the mean time from specimen collection to result by 14.5 h (23.6 h versus 9.1 h; P < 0.001) (Table 3). A total of 157 patients (86.3%) in the preimplementation group had HSV PCR results delayed by >12 h, compared to 44 patients (24.3%) in the postimplementation group. In fact, 79 patients (43.6%) in the postimplementation group had dHSV PCR results reported <4 h after lumbar puncture. Furthermore, the 24-h testing service in the postimplementation period allowed reporting of results for 45.3% of the patients (82/181 patients) between 6 p.m. and 8 a.m.
Impact of HSV-1 and HSV-2 PCR result reporting on acyclovir usage
Intravenous acyclovir therapy was prescribed for 118 patients (64.8%) in the preimplementation group and 120 patients (66.3%) in the postimplementation group (P = 0.83) (Table 1); a total of 98.7% of the patients (235/238 patients) received at least one dose of acyclovir. The mean times from specimen collection to therapy initiation were comparable in the preimplementation and postimplementation groups (3.3 h versus 2.2 h; P = 0.13) (Table 3). Acyclovir therapy was initiated for 95 patients (39.9%), including 50 patients in the preimplementation group and 45 patients in the postimplementation group (P = 0.51), prior to CSF collection. For patients with negative HSV PCR results from CSF and other sources, the duration of acyclovir therapy ranged from 5 min to 18 days in the preimplementation group and from 1 min to 6.8 days in the postimplementation group; the median duration of i.v. acyclovir therapy was 15.4 h less in the postimplementation group (P = 0.003) (Table 1). Of note, the patients with durations of 1 and 5 min received one dose of acyclovir and had their acyclovir orders discontinued shortly thereafter. Furthermore, the duration of acyclovir treatment for patients with HSV PCR results reported within 4 h after specimen collection was significantly less than that for patients with HSV PCR results reported >4 h after specimen collection (19.2 h versus 50.3 h; P = 0.05). The mean time from CSF collection to acyclovir discontinuation was 17.1 h longer in the preimplementation group (31.1 h versus 14 h; P < 0.001) (Table 3). The median times from the availability of HSV PCR results to the discontinuation of acyclovir therapy were not significantly different between the two groups, i.e., 1.7 h (preimplementation) versus 2.6 h (postimplementation) (P = 0.6) (Table 3). Of note, acyclovir was ordered and cancelled prior to administration for three patients, due to early reporting of negative HSV results (mean time from specimen collection to result, 4.1 h). In all three cases, orders were cancelled within 40 min after the release of HSV PCR results.
Acyclovir therapy was continued for four (57.1%) of the seven patients who tested positive for HSV from alternative sources, including a patient with recurrent neutropenia, a high-risk patient with acute myelogenous leukemia, a patient with lupus, intractable seizures, and autism, and a patient with no documented comorbidity. The other three patients had no or very short exposure to acyclovir. In addition, two patients (one in the preimplementation group and one in the postimplementation group) were positive for Epstein-Barr virus (EBV) and one patient in the postimplementation group was positive for varicella-zoster virus (VZV) from CSF samples. In all three cases, acyclovir therapy was continued despite negative HSV PCR results.
The creatinine levels of patients treated with acyclovir were assessed for potential renal toxicity after antiviral therapy. Despite the significant decrease in mean turnaround time (23.6 h versus 9.1 h; P < 0.001) after implementation of the rapid HSV PCR test (Table 3), no difference in the numbers of patients who developed renal toxicity from potential exposure to acyclovir was observed. Of the 238 patients prescribed acyclovir, creatinine levels were monitored for 211 (103 in the preimplementation group and 108 in the postimplementation group) at the time of the acyclovir order. A total of 4 patients (2 in each group) experienced potential acyclovir-related toxicity (Fig. 1). In the preimplementation group, creatinine levels in patients 1 and 2 were elevated within 38 to 63 h after acyclovir initiation and declined within 48 h after acyclovir discontinuation (Fig. 1). In the postimplementation group, in contrast, creatinine levels were elevated within 14 h for patient 3 and doubled within 6 h of acyclovir administration for patient 4. The mean time to PCR results was 15.6 h shorter for the two postimplementation patients (19.3 h versus 3.8 h), allowing a significantly shorter duration of acyclovir treatment (means of 78.1 h versus 26.8 h).
Creatinine levels of four patients with elevated creatinine levels after acyclovir initiation. The timeline shows the approximate time after the initiation of intravenous acyclovir treatment. Patients 1 and 2 were from the preimplementation group, and patients 3 and 4 were from the postimplementation group. The downward arrow on each line indicates when the HSV PCR results were reported. The large plus sign on each line indicates when acyclovir therapy was discontinued. The solid lines indicate patients from the preimplementation group, and the dashed lines indicate patients from the postimplementation group.
DISCUSSION
Molecular diagnostic applications that are conducive to rapid and continuous testing are important in clinical laboratories, to provide actionable results to clinicians in a timely manner. In this retrospective study, we evaluated the impact of implementing continuous 24-h/7-day HSV PCR testing in reducing unnecessary acyclovir exposure. As expected, the time from specimen collection to reporting of results was significantly decreased in the postimplementation period, when testing was performed using a direct sample-to-answer assay 24 h a day, 7 days a week.
It has been shown that rapid diagnosis of etiological agents can result in the reduction or optimization of antimicrobial therapy (24–26). In one study of 143 children who tested positive for adenovirus in direct fluorescent assays or viral cultures, 36% demonstrated a change in management, including discontinuation of antimicrobial agents and/or discharge from the hospital (24). Patients who were positive for respiratory viruses were also less likely to be discharged with oral antimicrobial therapy (26). We previously described two cases in which rapid diagnosis of bacterial meningitis, attributable to Listeria monocytogenes and Haemophilus influenzae, with a molecular panel assay allowed definitive diagnosis and optimization of antimicrobial therapy (27–29). In this study, the availability of the HSV PCR results on a 24-h/7-day basis resulted in a significant reduction in the duration of acyclovir treatment in the postimplementation arm. We also demonstrated a downward trend, approaching statistical significance, in the duration of acyclovir treatment for patients with HSV PCR results reported <4 h from the time of collection, compared to those with turnaround times of >4 h.
The mean time from specimen collection to result reporting was significantly decreased, by 14.5 h, postimplementation. Furthermore, 45.3% of the HSV PCR results were reported during the evening and night shifts in the postimplementation group. This finding confirms the importance of rapid HSV PCR testing in clinical laboratories and agrees with previous studies showing the impact of timely results on patient care. In a study of 134 patients, the closure of microbiology services after 11 p.m. significantly affected the time to the reporting of Gram stain results from positive blood cultures (7.5 h to 2.2 h; P = <0.001), due to the large number of blood cultures that were flagged positive during off hours (30). Another study reported a negative impact on patient survival of 9.8% (18.8% versus 9.0%; P = 0.624) when blood cultures were flagged positive during off hours (31).
None of the 363 patients in this study had CSF specimens that were positive by HSV PCR. This finding supports previous studies in which CNS infections caused by HSV were uncommon and not easily detectable, despite upward trends in the empirical treatment of patients for HSV meningitis or encephalitis (32, 33). In fact, data have suggested that recovery of any etiological agent from CSF is rare. The California Encephalitis Project was unable to detect infectious agents in about 71% of their cases from 1998 to 2000 (34). Similarly, even with the increased sensitivity of PCR, HSV was detected in only 6.2% of 6,607 CSF samples at the Mayo Clinic from 1993 to 1997 (35). Cutaneous manifestations of HSV in patients with HSV CNS involvement have been reported to be present in <10% of cases (36–38). Therefore, detection of HSV at other sites is not necessarily predictive of CNS involvement. On the other hand, it is important to note that a negative HSV PCR result from CSF does not absolutely exclude HSV meningoencephalitis, particularly if the CSF sample was obtained early in the course of the illness, and correlation with the clinical presentation (i.e., temporal lobe involvement) or HSV positivity from other sites is essential. Seven patients in this study demonstrated HSV positivity from sources other than CSF. Four of the seven patients continued acyclovir therapy. The other three patients either were never prescribed acyclovir or had acyclovir treatment discontinued prior to result reporting or within 1 h after the HSV PCR results. A total of 46 patients (12.7%) had acyclovir treatment discontinued prior to the availability of HSV PCR results, likely due to clinical presentations and assessment findings that did not warrant acyclovir therapy.
The complications associated with exposure to antimicrobial agents are not benign, and judicial use of antimicrobial agents is necessary. Adverse effects, including neurotoxicity and renal dysfunction, have been reported after i.v. acyclovir therapy (39, 40). The development of renal dysfunction due to acyclovir was dose dependent, and most cases occurred within 48 h after treatment initiation (39). Only four patients in total might have had acyclovir-related renal toxicity and, despite a significant decrease in the mean duration of acyclovir (by 17.1 h) in the postimplementation group, there was no observed impact on renal toxicity.
There are a number of limitations in this study. This is a single-site study and was retrospective in nature. The absence of HSV-positive results from CSF samples within the time frame can be considered a limitation but does not affect the assessment of unnecessary acyclovir usage in this study. We assumed that HSV meningitis was prominent in the differential diagnosis for these patients and HSV PCR results were considered to determine changes in patient management. For patients who continued acyclovir therapy after results were reported, we assume that other contributing factors, such as comorbidities, potential exposure, and alternative infectious and noninfectious diagnoses, contributed to the duration of therapy. Another limitation was the restriction to pediatric patients, and the findings for our patient population and clinical practices may not be representative of other institutions. For instance, since we were already performing molecular assays in the preimplementation period, our baseline mean turnaround time was 23.6 h, which may be considered much shorter than the times for laboratories that send out samples for HSV PCR assays or perform non-molecular-assay-based testing for HSV. Thus, more significant changes in acyclovir usage and toxicity may be found at other institutions. Lastly, measurement of creatinine levels was the only method used to assess renal function in this study, and neurotoxicity was not assessed. We assumed that any elevated creatinine levels were related to acyclovir therapy, and the small sample size did not allow direct correlations between rapid HSV testing and the avoidance of renal toxicity. Other potential contributing factors for nephrotoxicity were exposure to other antimicrobials and preexisting conditions such as hypoperfusion of the kidney due to placental abruption (patient 3).
Laboratory diagnosis of HSV CNS infections is critical for the treatment of pediatric patients. This study demonstrated that timely exclusion of HSV infection, through the availability of a 24-h/7-day dHSV PCR test, allowed prompt discontinuation of acyclovir therapy for HSV-negative patients.
MATERIALS AND METHODS
Study design.This was a retrospective study conducted at Children's Hospital Los Angeles (CHLA) (a freestanding tertiary care pediatric institution) and approved by the CHLA Institutional Review Board. A total of 363 patients who presented to the CHLA emergency department (ED) between 1 September 2013 and 30 April 2015 with signs and symptoms of meningitis or encephalitis (e.g., fever, headaches, seizures, altered mental status, fussiness, and vomiting) and for whom a HSV PCR test of a CSF specimen was ordered were enrolled in the study. Duplicate CSF samples submitted from the same patient were omitted from the study. Patients with CSF specimens tested for HSV between 1 September 2013 and 30 June 2014 were included in the preimplementation group, and those with CSF specimens tested between 1 July 2014 and 30 April 2015 were included in the postimplementation group. Patients in the preimplementation group had CSF specimens tested for HSV-1/2 using a laboratory-developed test (LDT) that required a separate nucleic acid extraction step. In contrast, patients in the postimplementation group had CSF specimens tested for HSV-1/2 using a rapid HSV PCR test that is a direct sample-to-answer assay, without a separate nucleic acid extraction step. Abstractions of medical charts were performed for all enrolled patients.
Preimplementation period.A total of 182 patients for whom HSV PCR tests with CSF specimens were ordered between 1 September 2013 and 30 June 2014 were included in the preimplementation group. Testing of patients enrolled in the preimplementation period was by an indirect HSV (inHSV) PCR assay that was available only from 9 a.m. to 5 p.m., Monday through Saturday. The inHSV assay was a multiplex real-time PCR assay (RealStar alpha herpesvirus PCR kit 1.0; Altona Diagnostics, Hamburg, Germany), performed on an Applied Biosystems 7500 cycler (Thermo Fisher Scientific, Waltham, MA), for the detection and differentiation of HSV-1, HSV-2, and VZV DNA. Samples required a nucleic acid extraction step and were batched and tested once a day. Briefly, 200 μl of a CSF sample, spiked with 5 μl of the Altona internal control (IC), was extracted using the NucliSENS easyMAG system (bioMérieux, Durham, NC) and eluted in 90 μl. PCR was performed using 5 μl nucleic acid combined with 15 μl master mix, with the following cycling conditions: 95°C for 10 min followed by 45 cycles of 95°C for 15 s and 58°C for 1 min. All results were reported in the hospital's electronic medical records. Positive results were considered critical values and were called directly to the physician.
Postimplementation period.A total of 181 patients who presented to the CHLA ED and for whom HSV PCR tests with CSF specimens were ordered between 1 July 2014 and 30 April 2015 were enrolled in the postimplementation arm. The Simplexa HSV 1 & 2 Direct assay (DiaSorin Molecular, Cypress, CA), performed on a 3M Integrated Cycler (41–43), was implemented at CHLA on 1 July 2014 and was performed for all CSF specimens collected from patients enrolled in the postimplementation period. This dHSV PCR assay does not require a separate nucleic acid extraction step, which is conducive to continuous testing. Testing by dHSV PCR was performed upon receipt of each CSF specimen in the laboratory, 24 h a day, 7 days a week. All results were reported in the hospital's electronic medical records. Positive results were considered critical values and were called directly to the physician.
Data collection.The following demographic data were collected for each patient enrolled in the study: age at the time of presentation, gender, admission and discharge dates, comorbidities, and antimicrobials prescribed. Additional results from clinical tests ordered as part of standard care, including CSF chemical analyses, microbiological cultures, and other viral studies, were also collected. Analysis was performed for all 363 patients, regardless of other viral or bacterial pathogens detected in the CSF specimens. Creatinine levels measured over the course of the patient's admission were also extracted, to evaluate kidney functions during i.v. acyclovir therapy.
Only one CSF specimen per patient was included in the study; additional CSF specimens collected from the same patient during the study period were excluded from the study. None of the excluded duplicate CSF specimens were positive for HSV. Patients who were receiving acyclovir therapy prior to presentation and continued to receive acyclovir therapy after the clinical event specific to the study period were excluded from the study. A total of 354 CSF samples (97.5%) were obtained via lumbar puncture, and 9 samples (2.5%, including 5 samples in the preimplementation period and 4 samples in the postimplementation period) were collected from ventriculoperitoneal shunts.
Statistical analyses.Statistical analyses were performed using Student's t tests, 2-by-2 contingency tables, and Fisher's exact tests, with GraphPad QuickCalcs software (Web-based version). The Mann-Whitney test, using XLSTAT version 18.07 (Addinsoft, New York, NY), was used to compare median values. Comparisons of symptoms between the preimplementation and postimplementation groups were performed using comparisons of proportions (MedCalc Software, Ostend, Belgium). For all calculations, P values of ≤0.05 were considered statistically significant.
ACKNOWLEDGMENTS
We thank the staff in the clinical virology and clinical microbiology laboratories at Children's Hospital Los Angeles for technical support.
FOOTNOTES
- Received 29 December 2016.
- Returned for modification 9 January 2017.
- Accepted 6 March 2017.
- Accepted manuscript posted online 8 March 2017.
- Copyright © 2017 American Society for Microbiology.
