Clinical signs as a guide for performing HSV-PCR in correct diagnosis of herpes simplex virus encephalitis
Background: Clinical criteria (symptoms) are not reliable enough to differentiate between different causes of encephalitis. The clinical presentation of herpes simplex virus encephalitis (HSVE) is not classically constant and in such a patient, therefore, it is vital to make early diagnosis. Aims: To investigate satisfactory and crucial clinical signs as guide to perform HSV-PCR in a rapid diagnosis of herpes simplex virus encephalitis. Material and Methods: A total of 156 CSF specimens from 70 patients with clinically suspected HSVE or meningoencephalitis were tested. The criteria for cases suspected of HSVE were fever >380C, altered mental status and other critical manifestations. CSF features, irregularity in brain CT scan and MRI findings were also assessed. All the specimens were collected before and after Acyclovir treatment. Polymerase chain reaction was performed using primers, which amplified DNA sequences for both HSV-1 and HSV-2. Statistical Analysis: To analyze data, two-tailed Fisher’s exact test and the X2-test with Yates’ correction were used as appropriate. The odds ratio was used to express the strength of association between the clinical factors and the PCR results. Results: HSV-DNA was detected in 18% of the specimens, belonging to 25.7% of the patients. Results indicate that the majority of the clinical symptoms are not specific to definitive clinical diagnosis of HSVE, except alteration in the level of consciousness—odds ratio [0.27 (0.07-0.96) (P=0.033)]; and lateralization sign—odds ratio [4.7 (0.98-22.6) (P=0.023)]. However, laboratory data, including total white blood cell count, especially the number of lymphocytes, and MRI findings could be suggested for HSV-PCR examination. Conclusion: At the first admission, a preliminary finding of at least two important clinical features mentioned above along with the pattern of CSF cell and differential counts could be sufficient to perform HSV-PCR which could ultimately result in a rapid and correct diagnosis of herpes simplex encephalitis.
Herpes simplex virus type 1 (HSV-1) is the major cause of devastating encephalitis,, with a predilection for the temporal lobes and a range of clinical presentations, from aseptic meningitis and fever to a severe, rapidly progressing form involving altered consciousness. Despite advances in antiviral therapy over the past two decades, herpes simplex virus encephalitis (HSVE) remains a serious illness with significant risk of morbidity and death. The classically recognized presentation of HSVE involves the relatively abrupt onset of fever, headache, focal neurological signs, seizures, and/or deteriorating level of consciousness, and in such patients, the diagnosis is usually easier and earlier. The clinical diagnosis, however, is often unreliable, as numerous neurological syndromes may mimic HSVE. For prognosis of the disease, it is important to achieve a rapid diagnosis. Examining the cerebrospinal fluid (CSF) via a spinal tap is necessary to isolate the specific cause. Since the outcome of HSV infections of the CNS can be influenced by acyclovir treatment, a correct and early diagnosis is of great importance.
The early diagnosis of HSVE is a major problem. The detection of HSV-specific intrathecal antibody is not possible early. Computer tomography (CT) scan and magnetic resonance imaging (MRI) techniques are both extremely helpful in the diagnosis of encephalitis, however, the results are still not specific for HSVE. CT scan may not reveal abnormalities until 3-5 days following symptom onset, by which time the patient may be stuporous and comatose. In the acute setting, even contrast-enhanced MRI may be negative. With the advent of polymerase chain reaction (PCR), the detection of HSVE and infections of the brain have been documented.,, PCR is a valuable test in the diagnosis of a patient with suspected HSVE. The test is positive early in the disease and has a turnaround time of approximately 24 hours. False positive reports from cross-contamination, though rare, have been reported. Generally, the test remains positive during the first week of therapy.
A combination of clinical skills and the interpretation of all the investigative data as well as PCR results are important in achieving correct diagnosis.
The purpose of the present research was to investigate the significance of the correlation between the radiological investigation alongside clinical data and the detection of herpes simplex virus DNA by the PCR in the CSF of patients suspected with HSVE.
Patients and samples
A cross-sectional study was performed on 156 CSF specimens from 70 patients with clinically suspected HSVE or meningoencephalitis admitted to the teaching hospitals of Shiraz University of Medical Sciences (SUMS), Shiraz, Iran from September 2000 to September 2002. The group consisted of 47 males and 23 females. The age of the patients under study ranged from 2 to 85 years, with an average of 38.65 years. The criteria for suspicion of HSVE in any adult or pediatric patient were fever >380C, altered mental status (low level of consciousness, agitation, lethargy) and other critical manifestations (e.g. focal neurological deficits, seizures), and CSF pleocytosis with predominant lymphocyte and/or elevated protein and negative Gram's stain. Irregularity in brain CT scan and MRI results was also evaluated. All the specimens were collected within 20 days of the onset of neurological illness. All the specimens were collected before and after Acyclovir treatment (30 mg/kg for 14 days) and stored at -700C until use. Objective data including bacterial culture, Gram's stain, serology on CSF samples, level of CSF protein and sugar, and CSF cell and differential counts were obtained from patients' clinical records. Clinical data, CT scan and MRI results were also obtained from the available clinical records. The local ethics committee granted ethical approval for the study. However, informed written consent was obtained from all participants.
Polymerase chain reaction (PCR) assay
In order to prevent any contamination we strictly adhered to the guidelines elaborated by Kowk and Higushi.9 CSF was not routinely centrifuged prior to storage but was briefly centrifuged (1000 g, 5 minutes) prior to extraction of DNA. Approximately 100ml of the sample was boiled for 20 minutes, cooled for 5 minutes and centrifuged at 14000 g at room temperature for 5 minutes. The supernatant was used as a template DNA for the PCR amplification. For PCR amplification oligonucleotid primers were selected using published10 DNA sequences for both HSV-1 and HSV-2 thymidine kinase (TK) gene whereby primer 1 and primer 2 flank 351 base-pair fragment . Five microliter of the DNA extracted was added to 45ml of the PCR reaction mixture (10mM Tris-HCl, pH 8.3; 50mM KCl, 1.5 mM MgCl2; 7% DMSO; 200mM of each dATP, dCTP, dGTP, and dTTP; 50 pmole of each oligonucleotide primer; and 2.5 units of Taq polymerase). The mixture was overlaid with 50ml of sterile mineral oil to minimize evaporative losses. Samples were heated to 940C for 5 minutes to denature DNA and were then subjected to 45 cycles of 940C for 1 minute, 570C for 1 minute (annealing for primers), 720C for 1 minute (extension of DNA), and finally 720C for 10 minutes using an automated thermal heating block (Eppendorf, Germany).
Subsequently, 10 ml of each amplified product and 1 kilobase ladder DNA were electrophoresed (1% agarose, 8 v/cm) for 45 minutes. Gels were stained with ethidium bromide and visualized by transilluminator.
Restriction Endonuclease Typing
Twenty microliter of PCR product was digested with two units of either restriction enzyme Ava I or Ava II at 370C for 2 hrs in a buffer comprising: 10 mM Tris-HCl pH 7.5; 10mM MgCl2; 50mM NaCl; and 1 mM dithioerythritol. The resulting fragments were separated by agarose gel electrophoresis (1.5% agarose in Tris-EDTA buffer run at 5 V/cm for 1 hr). After staining with ethidium bromide the gel was visualized by transilluminator.
Sensitivity and specificity of the PCR
The sensitivity of the test for HSV-PCR was determined using DNA extracted from herpes simplex virus isolated from a variety of clinical specimens and cultures on MRC-5 and Vero cell lines. The quantity of the proliferated viruses was determined according to standard procedures for virus titration and calculation of TCID50.
The specificity of the PCR was determined using DNA extracted from the herpes virus group other than HSV and, in addition, DNA extracted from uninfected MRC-5 and Vero cells.
The characteristics of the patients and the results obtained by the PCR method were analyzed by two-tailed Fisher's exact test and the X2-test with Yates' correlation as appropriate. The odds ratio was used to express the strength of association between the clinical factors and the PCR results.
Sensitivity and specificity of the HSV-PCR
When evaluated by gel electrophoresis and ethidium bromide staining, the detection limit of the PCR was found to be 50 HSV particles equal to 1 TCID50. No cross-reaction was found between selected primes and the DNA extracted from the other herpes virus group. Based on clinical features, responses to Acyclovir, and MRI and CT scan findings, the sensitivity and specificity of the HSV-PCR were 85% and 100% respectively.
Clinical features of the patients with suspected HSVE
In our study patients with suspected HSVE usually manifested six major clinical features including fever (81.4%), headache (71.4%), altered level of consciousness (71.4%), lateralization sign (62.9%), gustatory hallucination (58.6%) and cranial nerve defect (50%). Laboratory data indicated red blood cells (5-1500/mm3) in CSF (75.7% patients), white blood cells (5-450/mm3) with mononuclear cells predominance (75.7% patients), and with mild increase in protein (78% patients). However, CSF glucose was in the normal range (40-80 mg/dl) in 90% of the patients.
CT scan and MRI results were available for only 30 out of the 70 (42.8%) patients. Both CT scan and MRI results were available for 12 patients. Twelve patients had only MRI results and 6 patients had only CT scan results available respectively. In 30 patients abnormalities were diagnosed in different parts of the brain.
PCR results and its correlation with clinical symptoms
HSV-DNA was detected in 28 out of 156 CSF specimens (18%) from 18 out of 70 patients (25.7%) with suspected HSVE. The final diagnosis of the remaining 52 patients negative for HSV-PCR included: brucella encephalitis (n=2); TB meningitis (3); pnumococcal meningitis (n=1); meningococcal meningitis (n=1); hemaphilus influenza (n=1); non-HSV viral encephalitis (n=12); non-PCR-clinically diagnosed HSVE (n=4); unknown encephalitis (10); abscess (n=1), toxic encephalitis (n=1), vascular disorder (n=3), tumor (n=1); subdural hemorrhage (n=3); cerebritis (n=2); stroke hemorrhage, (n=2); stroke ischemic (n=2); epidural and subdural inflamation (n=3). Using restriction enzyme technique, HSV-2 was found in 1 out of 18 patients (5.5%) and the remaining (94.5%) were infected with HSV-1. Ava II will only cut an HSV-1 amplicon and Ava I cuts an HSV-2 amplicon. The patient with HSV-2-PCR positive was a 2-year-old girl with a diagnosis of transient meningitis. The patient recovered without complication.
CT scan and MRI results were available for 13 of the 18 (72.2%) patients with HSV-DNA in their CSF samples. MRI of the brain showed bilateral pathologic changes in the medial temporal and inferior areas in 11 out of the 13 (85%) patients (P=0.02), which is highly suggestive of HSVE. Conversely, the CT scan was abnormal in 3 out of the 13 (23%) HSVE patients with positive HSV-1 DNA (P=0.21).
The correlation between the different clinical presentations of patients with suspected HSVE and the results obtained from the HSV-PCR on CSF samples has been shown in [Table - 1]. As the results indicate, most clinical symptoms are not specific to definitive clinical diagnosis of HSVE, except the lateralization sign (Fisher's exact test P=0.023) and alteration in the level of consciousness (Fisher's exact test P=0.033). However, laboratory data, especially bloody CSF (P=0.02), number of lymphocytes (P=0.01), and MRI findings could be suggested for HSV-PCR examination. HSV-PCR positive changed to negative 4-7 days after a one-dose treatment of Acyclovir in the majority of the patients with confirmed HSVE. However, HSV-PCR remained positive in 1 patient 16 days after Acyclovir treatment. The patient was treated with one more dosage of Acyclovir and then HSV-PCR was repeated. This time HSV-PCR was negative.
Failure to consider the possibility of HSVE can result in delayed diagnosis and inefficient treatment, with subsequently increased risks of morbidity and mortality. According to our results, no pathogonomonic clinical findings are directly associated with HSVE. Furthermore, the diagnosis can be confirmed by PCR and MRI findings. A significant correlation between HSV-PCR positive results and MRI temporal lesions was found in such cases. These findings are in agreement with previous studies describing MRI lesions in patients with HSVE., The majority of HSVE (94.5%) cases in adults are caused by HSV-1. Some of the HSVE patients treated with Acyclovir continued to suffer from neurological disability after treatment. In our investigation, PCR-positive signals could be determined up to 10-12 days after the onset of the neurological symptoms. However, our recommendation for the diagnosis of HSVE is that CSF must be taken for HSV-PCR as early as possible after the onset of symptoms.
The study population comprised 47 male subjects and 23 female subjects. Although HSVE was diagnosed by HSV-PCR in 34.7% of female subjects and in 21.2% of male subjects respectively, there was no significant relationship between gender and PCR status (P=0.22).
In the present study the MRI finding was compared with PCR results in 13 out of 18 HSVE patients whose MRI results were available. PCR was positive in 11 out of 13 HSVE patients (84.6%) with brain MRI abnormalities reflecting lesions at the infero-medial region of one or both temporal lobes. There was a significant correlation between the PCR-positive result and MRI temporal lesions. Since the CT scan was abnormal in just 3 out of the 13 HSV-PCR positive (23%), patients, statistically no significant correlation was demonstrated between CT scan findings and PCR results (P=0.2).
HSV-2 PCR was positive in the CSF of 1 child (2-year-old) with meningitis. However, in immunocompetent adults, HSV-2 may induce meningoencephalitis and meningitis as well as rare cases of HSV-2 brainstem encephalitis and recurrent myelitis.
Comparison between CT scan and MRI results in patients with HSVE, confirmed by HSV-PCR, revealed that CT scan is not highly sensitive for early diagnosis of HSVE, as compared to MRI findings. Occasionally, MRI can also show hemorrhage. This finding is association with the presence of red blood cells in CSF could be a signal and an expressive factor to perform HSV-PCR.
In the present study we used some statistical analyses to compare the clinical features of the patients having positive PCR results with those having negative PCR results. No association was found between the clinical features of patients with positive PCR and patients with negative PCR results. However, MRI findings, CSF red blood cell count and white blood cell count were each independently associated with positive PCR results.
In a study to evaluate the utility of an assay based on a PCR of CSF in the management of patients with suspected HSVE, Teba et al suggested that empiric therapy with Acyclovir be widely used without laboratory confirmation. And, accordingly, in their model, they compared a PCR-based approach with empiric therapy that corresponds to current practice. Under very conservative assumptions, the PCR-based approach not only provided better outcomes but also resulted in extensively less Acyclovir.
In this study, the proper execution of a systematic experimental design and the exact application of the current codes of practice reduced the potential for false-positive reactivity.
To conclude, the results manifested in this study illustrated the worth of the PCR for the diagnosis and follow-up treatment of HSVE patients. Except for 2 important clinical features (i.e. alteration in the level of consciousness and lateralization sign), no significant correlation was established between the other clinical characteristics of HSVE on the one hand and PCR results on the other. PCR analysis allows the antiviral therapy to be performed efficiently and helps to detect the appearance of a resistant strain of the virus. However, at the first admission, preliminary findings of at least 2 important clinical features mentioned above along with the pattern of CSF cell and differential counts could be sufficient to perform HSV-PCR, which could ultimately result in a rapid diagnosis of herpes simplex virus encephalitis.
The authors would like to thank the authorities in the Shiraz University of Medical Sciences (SUMS) Shiraz, Iran for financially supporting the project by offering a grant to cover the expenses involved. The authors would also like to thank Dr. S. Sajjadi for reading the manuscript and providing useful comments and Dr. Mohagheghzadeh for his great help in statistical analysis of the data.