Skip Navigation

Diagnosis of Enteroviral Meningitis by Use of Polymerase Chain Reaction of Cerebrospinal Fluid, Stool, and Serum Specimens

  1. Laura Kupila1,4,
  2. Tytti Vuorinen3,
  3. Raija Vainionpää3,
  4. Reijo J. Marttila1, and
  5. Pirkko Kotilainen2
  1. 1Department of Neurology, Turku University Central Hospital, Turku
  2. 2Department of Medicine, Turku University Central Hospital, Turku
  3. 3Department of Virology, University of Turku, Turku
  4. 4Department of Neurology, Satakunta Central Hospital, Pori, Finland
  1. Reprints or correspondence: Dr. Laura Kupila, Dept. of Neurology, Turku University Central Hospital, Kiinamyllynkatu 4-8, 20520 Turku, Finland (laura.kupila{at}tyks.fi).
 
Next Section

Abstract

Background. Because enteroviruses can be detected in various clinical samples during enteroviral meningitis, we analyzed the combined diagnostic utility of polymerase chain reaction (PCR) of cerebrospinal fluid (CSF), feces, and serum for detection of enterovirus in specimens obtained from adults with aseptic meningitis or encephalitis.

Methods. PCR results were analyzed for 34 adults for whom enteroviral meningitis was diagnosed on the basis of virus isolation and antibody detection in our hospital during 1999–2003. PCR results were also analyzed for 77 adults with meningitis or encephalitis of another defined cause for whom this assay was used for diagnostic evaluation during that period.

Results. Twenty-six (76%) of 34 CSF samples and 24 (96%) of 25 fecal samples collected from patients with enteroviral meningitis had positive PCR results. The diagnostic yield of the test was lower for CSF specimens obtained >2 days after clinical onset, compared with CSF collected ⩽2 days after onset. Instead, PCR of feces was highly useful also later, because 12 of the 13 fecal specimens obtained 5–16 days after clinical onset had positive test results. None of 75 CSF samples and 2 of 48 fecal samples obtained from patients with nonenteroviral infection had positive PCR results. All serum samples were PCR negative.

Conclusions. PCR of fecal specimens obtained throughout the course of enteroviral meningitis had the highest clinical sensitivity for detecting enterovirus. It is recommended that, in addition to performance of CSF PCR, fecal samples collected from patients with suspected enteroviral meningitis should be tested by PCR, especially when the duration of symptoms is >2 days.

Enteroviruses are known worldwide as a common cause of aseptic meningitis, with both epidemic and endemic patterns, in patients of all ages [1–5]. Although enteroviral meningitis is usually benign, the clinical features and CSF findings associated with the disease can overlap those of bacterial or herpetic infections, resulting in prolonged hospital stays and initiation of presumptive treatment until the diagnosis is made [4, 6–11]. A definite diagnosis of enteroviral meningitis can be established by detection of virus directly in CSF specimens. However, enterovirus is often also detected in the gastrointestinal or respiratory tracts during enteroviral meningitis, suggesting that the virus is very likely the cause of disease [3, 5, 10, 12]. During viremia, enteroviruses can also be detected in the blood [1, 5, 13–15].

PCR methods for detecting enteroviruses have been developed during the past decade. Many studies have shown that CSF PCR is more sensitive and rapid than virus isolation for the diagnosis of enteroviral meningitis [1, 5, 12, 16–20]. Because the enterovirus load has been described as being higher in feces than in CSF during enteroviral meningitis, stool specimens are also a good target for detection of enteroviral RNA [5, 12, 21]. Although the superiority of PCR versus virus isolation for detection of enterovirus in fecal samples has been reported [3, 12, 16, 22], only 1 study has described the clinical experience with this test for patients with enteroviral meningitis [3]. Moreover, we are not aware of any study that evaluated the utility of CSF PCR and stool PCR in parallel among patients with enteroviral meningitis.

The PCR assay for enterovirus was adapted in clinical use at the Departments of Neurology and Medicine, Turku University Central Hospital (Turku, Finland) in 1999. The aim of this study was to evaluate the feasibility of PCR of CSF, stool, and serum in the etiologic diagnosis of meningitis or encephalitis in adult patients.

Previous SectionNext Section

Patients and Methods

The present work is a part of a larger prospective follow-up study in which the etiology of aseptic meningitis or encephalitis was evaluated in consecutive adults who were treated at the Turku University Central Hospital during 1999–2003. The diagnosis of aseptic meningitis was based on symptoms and signs indicative of meningeal irritation for patients with CSF pleocytosis (WBC count, >5 × 106 cells/L) and a negative CSF culture for bacteria. The diagnosis of encephalitis was based on symptoms and signs indicative of acute cerebral infection that were supported either by detection of CSF pleocytosis and/or by neuroradiological and electroencephalographic findings compatible with a diagnosis of encephalitis.

During the 5-year study period, a large array of different microbiological tests was done routinely on clinical specimens collected from patients with aseptic meningitis or encephalitis. The most common tests included CSF PCR for detection enteroviruses, herpes simplex virus (HSV), varicella-zoster virus (VZV), and Borrelia burgdorferi and stool and serum PCR for detection of enteroviruses. Also, CSF, fecal, and throat swab specimens were cultured for detection of viruses. Antibodies against enteroviruses, herpesviruses, respiratory viruses (adenovirus; influenza A and B viruses; parainfluenza virus 1, 2, and 3; and respiratory syncytial virus), tickborne encephalitis (TBE) virus, Mycoplasma pneumoniae, Chlamydia pneumoniae, and B. burgdorferi were measured in serum and CSF samples.

Study population. We included in this study 34 consecutive adults who received a diagnosis of enteroviral meningitis on the basis of virus isolation and/or antibody detection during the study period. We also included 77 consecutive adult patients who received a diagnosis of meningitis or encephalitis of another defined cause during the study period and for whom PCR for enterovirus was involved in the diagnosis.

For 19 of 34 patients with enteroviral meningitis, the microbiological diagnosis was definite, as determined by isolation of enterovirus from CSF or by EIA detection of enterovirus-specific IgM antibodies in CSF (table 1). For 15 patients, the diagnosis was probable, as determined by EIA detection of enterovirus-specific IgM antibodies in serum and/or by isolation of enterovirus from fecal or throat swab specimens, without any other identified agent for the disease (table 1).

Figure 1
View larger version:
Figure 1

Influence of sampling time on the results of CSF PCR (A) and stool PCR (B) for detection of enterovirus for 34 patients with enteroviral meningitis. Solid area, number of positive PCR results; open area, number of negative PCR results. Thirteen CSF samples were obtained 3–7 days after the onset of symptoms, and 2 samples were obtained 10–12 days after onset. Nine fecal samples were obtained 5–7 days after the onset of symptoms, and 4 samples were obtained 9–16 days after onset.

Table 1
View larger version:
Table 1

RT-PCR results for 34 patients for whom the diagnosis of enteroviral meningitis was considered to be definite or probable on the basis of viral cultures for virus and/or antibody analyses.

The causative agents for the 77 patients with meningitis or encephalitis of another defined cause included HSV (recovered from 31 patients), VZV (18), TBE virus (12), Epstein-Barr virus (3), M. pneumoniae (3), B. burgdorferi (3), parainfluenza virus (2), Toxoplasma gondii (2), paramyxovirus (1), adenovirus (1), and C. pneumoniae (1). The microbiological diagnosis of HSV infection was based on detection of HSV-specific nucleic acids in CSF. The diagnosis of VZV infection was based on detection of VZV-specific nucleic acids in CSF (for 10 patients) or on serological evidence of acute VZV infection or detection of cutaneous zoster during the meningitis or encephalitis (for 8 patients). The microbiological diagnoses of infection due to the other etiologic agents were based on antibody detection by EIA or immunoblot analysis.

PCR for enterovirus. CSF samples that underwent RT-PCR for detection of enterovirus were obtained from patients at admission to the hospital. Fecal samples and serum samples that underwent RT-PCR were obtained from patients on neurological or medical wards during the first week of hospitalization. RT-PCR for enterovirus was performed at the Department of Virology, University of Turku, as described elsewhere [23, 24].

Data analysis. Results of RT-PCR of CSF, fecal, and serum samples collected from patients with enteroviral meningitis and from patients with meningitis or encephalitis of another defined cause were analyzed for the sensitivity, specificity, and predictive values of RT-PCR. The influence of sampling time on PCR results was also analyzed.

Previous SectionNext Section

Results

Patients with enteroviral meningitis. CSF PCR for enterovirus was done on samples obtained from 34 patients with enteroviral meningitis; 26 of 34 samples tested positive for enterovirus (sensitivity, 76%) (table 2). Results were positive for 14 of 17 CSF samples from which an enterovirus was isolated and for 10 of 15 CSF samples from which an enterovirus was not isolated (table 1). In addition, 2 CSF samples that were PCR positive were not sent for isolation of virus.

Table 2
View larger version:
Table 2

Accuracy of CSF PCR for detection of enterovirus in the etiologic diagnosis of meningitis or encephalitis.

Twenty-four of 25 stool samples obtained from patients with enteroviral meningitis were PCR positive (sensitivity, 96%) (table 3). These included 22 stool samples from which an enterovirus was isolated, as well as 2 stool samples that were not sent for virus isolation (table 1). Stool PCR was not performed on 2 samples from which an enterovirus was isolated. The only fecal sample that remained PCR negative for enterovirus also remained culture negative.

Table 3
View larger version:
Table 3

Accuracy of stool PCR for detection of enterovirus in the etiologic diagnosis of meningitis or encephalitis.

CSF and fecal specimens obtained from 25 of 34 patients with enteroviral meningitis were analyzed by PCR (table 1). Results of CSF PCR were positive for 20 patients, and results of stool PCR were positive for 24 patients. For the 4 patients who had enteroviral RNA in feces but not in CSF, enterovirus was isolated in CSF for 1 patient, and enterovirus-specific IgM antibodies were detected in serum specimens for 3 patients. For 1 patient, results of both PCR tests remained negative, although enterovirus was isolated from CSF. All 23 serum samples collected from patients with enteroviral meningitis were PCR negative for enterovirus.

PCR-positive CSF samples were obtained a median of 1.5 days (range, 0.5–7 days) after the onset of symptoms, and PCR-negative CSF samples were obtained a median of 4.5 days (range, 1–12 days) after symptom onset. PCR-positive fecal specimens were obtained a median of 4.5 days (range, 2–16 days) after symptom onset; 12 were collected 5-16 days after onset. The single fecal sample that remained PCR negative for enterovirus was obtained 5 days after the onset of symptoms. The influence of sampling time on the results of CSF and stool PCR is illustrated in figure 1. Serum samples that underwent PCR for detection of enterovirus were obtained a median of 4 days (range, 1-14 days) after symptom onset; 8 were obtained ⩽3 days after onset.

Patients with meningitis or encephalitis of another defined cause. None of the 75 CSF samples collected from patients with meningitis or encephalitis of another defined cause were PCR positive for enterovirus (specificity, 100%) (table 2). Two of 48 fecal samples were PCR positive for enterovirus (table 3). Enterovirus was also cultured from 1 of these 2 stool specimens. CSF, serum, and throat swab specimens did not indicate signs of an acute enteroviral infection in the 2 patients from whom these stool specimens were obtained, and for both patients, TBE virus was diagnosed as the etiologic agent of meningitis. Thus, the specificity of stool PCR for enterovirus was 96% (table 3). All 42 serum samples obtained from patients with meningitis or encephalitis of nonenteroviral cause were PCR negative for enterovirus.

CSF samples that underwent PCR analysis were collected a median of 2.5 days (range, 0.5–21 days) after the onset of symptoms; stool samples, a median of 6.5 days (range, 2–18 days); and serum samples, a median of 5 days (range, 1–29 days).

Predictive values of CSF and stool PCR. Results of CSF PCR were positive only for patients with enteroviral meningitis, indicating a positive predictive value of 100% (table 2). Results of PCR were positive for 24 of 25 fecal samples obtained from patients with enteroviral meningitis and for 2 of 48 fecal samples collected from patients with nonenteroviral infection, indicating a positive predictive value of 92% (table 3).

CSF PCR was negative for enterovirus for all 75 patients with meningitis or encephalitis of another defined cause and for 8 of 34 patients with enteroviral meningitis, indicating a negative predictive value of 90% (table 2). Stool PCR for enterovirus was negative for enterovirus for 46 of 48 fecal samples obtained from patients with nonenteroviral disease and for 1 of 25 fecal samples collected from patients with enteroviral meningitis, indicating a negative predictive value of 98% (table 3).

Previous SectionNext Section

Discussion

To our knowledge, the present study is the first in which CSF, stool, and serum PCR for detection of enterovirus have been analyzed in parallel for adult patients with enteroviral meningitis. In this study, the clinical sensitivity of CSF PCR was high during the early course of enteroviral meningitis; 17 of the 19 CSF samples obtained during the first 2 days of illness were PCR positive. Moreover, during the entire disease course, enteroviral RNA was detected by PCR in 10 CSF samples from which virus could not be isolated. These results show that CSF PCR is of great value for establishing a definite diagnosis of enteroviral meningitis, corroborating previous studies [1, 5, 12, 17–20]. Yet a new finding was that the diagnostic yield by PCR was clearly lower for CSF obtained >2 days after the clinical onset of disease (figure 1), possibly indicating that the utility of CSF for detection of enteroviral RNA decreases a few days after the onset of meningitis.

The most important new finding here was that the fecal PCR test was clinically the most sensitive to detect enterovirus during meningitis, because the sensitivity of PCR was 96% for fecal samples and 76% for CSF, and none of the serum samples tested positive by this test. Furthermore, our results show that the clinical sensitivity of stool PCR remained very high throughout the course of the disease; 12 of 13 fecal samples obtained 5–16 days from the clinical onset were PCR positive. Glimåker et al. [3] have reported that stool PCR is highly valuable for diagnosing enteroviral meningitis (sensitivity, 93%), but the influence of sampling time on the PCR results was not evaluated in that study. They did not compare the utility of stool versus CSF PCR, because the diagnoses of enteroviral meningitis were based on virus isolation or on antibody analysis, and PCR analyses of CSF were not done.

Enteroviral replication in the gastrointestinal tract may persist for several weeks after a clinical infection, even without any clinical disease [25]. Therefore, detection of enteroviral RNA in fecal samples does not confirm that enterovirus is the causative agent of meningitis, although the finding suggests that this etiology is highly probable. In the present study, enteroviral genome was detected by PCR or enterovirus was isolated from CSF specimens obtained from 21 (88%) of 24 patients who had enteroviral RNA in fecal samples, confirming the etiological role of this virus in their meningitis.

None of our patients with meningitis or encephalitis of another defined cause had enteroviral RNA in CSF, indicating that the specificity of CSF PCR was 100%. Stool PCR also proved to be highly specific (96%); results were positive for only 2 of 48 fecal samples obtained from patients with meningitis or encephalitis of another defined cause. In the 2 patients from whom these PCR-positive samples were obtained, the causative agent was TBE virus. In one patient, enterovirus was also isolated from the fecal specimen, confirming the existence of this virus in the gastrointestinal tract, presumably as an “innocent bystander.” In the other patient, because results of stool culture were negative, the positive stool PCR result was the only sign of the existence of enterovirus during meningitis; therefore, the relevance of this positive PCR result remains undefined.

We also determined the predictive values of CSF and stool PCR in the etiologic diagnostics of meningitis or encephalitis. In our study, a positive CSF PCR result reliably identified enteroviral meningitis (positive predictive value, 100%), but a negative test result did not reliably rule out this possibility (negative predictive value, 90%), especially if the CSF sample was collected >2 days after the onset of clinical symptoms. On the other hand, a negative stool PCR result practically excluded an enteroviral etiology of meningitis (negative predictive value, 98%), even if the stool sample was obtained several days after the onset of clinical symptoms.

Several reports based on virus isolation have shown that a viremic stage occurs in systemic enteroviral infections [13, 14, 26]. In our study, however, all 23 serum samples obtained from patients with enteroviral meningitis yielded negative PCR results. This finding is consistent with results of a previous study showing a low diagnostic value of serum PCR; PCR results were positive for only 1 of 7 adults for whom enteroviral RNA was detected in CSF [1]. In that same study, results of serum PCR were positive for 5 of 10 children who had positive CSF PCR results, suggesting that detectable viremia is more common in pediatric patients. However, 2 other studies reported good diagnostic yield of serum PCR for adults with aseptic meningitis, even if samples were collected 4 or 5 days after clinical onset [5, 15]. Although the median time at which serum samples were obtained was 4 days for our patients, 8 samples were obtained ⩽3 days after the onset of meningitis. On this basis, our negative findings cannot be explained merely by the delay in sampling time. Thus, additional studies are needed to evaluate the utility of serum PCR for adults with aseptic meningitis.

In conclusion, stool PCR was clinically the most sensitive for detecting enterovirus during enteroviral meningitis and could give a presumptive diagnosis throughout the disease course. A definite diagnosis was obtained by CSF PCR, but its utility was clearly lower for samples obtained >2 days after clinical onset. Therefore, it is recommended that, in addition to performance of CSF PCR, fecal samples obtained from adults with suspected enteroviral meningitis should be tested by PCR, especially when the duration of the symptoms is >2 days.

Previous SectionNext Section

Acknowledgments

We thank Katja Rannikko for database management.

Financial support. Research Foundation of Orion Corporation and Finnish Neurological Foundation.

Potential conflicts of interest. All authors: no conflicts.

  • Received September 27, 2004.
  • Accepted October 22, 2004.
Previous Section
 

References

  1. ↵
    1. Andreoletti L,
    2. Blassel-Damman N,
    3. Dewilde A,
    4. et al
    . Comparison of use of cerebrospinal fluid, serum, and throat swab specimens in diagnosis of enteroviral acute neurological infection by a rapid RNA detection PCR assay. J Clin Microbiol 1998;36:589-91.
    Abstract/FREE Full Text
    1. Rotbart HA,
    2. Brennan PJ,
    3. Fife KH,
    4. et al
    . Enterovirus meningitis in adults. Clin Infect Dis 1998;27:896-8.
    FREE Full Text
  3. ↵
    1. Glimåker M,
    2. Abebe A,
    3. Johansson B,
    4. Ehrnst A,
    5. Olcén P,
    6. Strannegård Ö
    . Detection of enteroviral RNA by polymerase chain reaction in faecal samples from patients with aseptic meningitis. J Med Virol 1992;38:54-61.
    MedlineWeb of Science
  4. ↵
    1. Chambon M,
    2. Archimbaud C,
    3. Bailly JL,
    4. et al
    . Circulation of enteroviruses and persistence of meningitis cases in the winter of 1999–2000. J Med Virol 2001;65:340-7.
    CrossRefMedlineWeb of Science
  5. ↵
    1. Thorén A,
    2. Widell A
    . PCR for the diagnosis of enteroviral meningitis. Scand J Infect Dis 1994;26:249-54.
    MedlineWeb of Science
  6. ↵
    1. de Freixo MF
    . Aseptic meningitis by echovirus associated with herpes progenitalis. Arch Neurol 1980;37:787.
    Abstract/FREE Full Text
    1. Rice SK,
    2. Heinl RE,
    3. Thornton LL,
    4. Opal SM
    . Clinical characteristics, management strategies, and cost implications of a statewide outbreak of enterovirus meningitis. Clin Infect Dis 1995;20:931-7.
    Abstract/FREE Full Text
    1. Hallam NF,
    2. Eglin RP,
    3. Holland P,
    4. Bell EJ,
    5. Squier MV
    . Fatal coxsackie B meningoencephalitis diagnosed by serology and in-situ nucleic acid hybridisation. Lancet 1986;2:1213-4.
    MedlineWeb of Science
    1. Cree BC,
    2. Bernardini GL,
    3. Hays AP,
    4. Lowe G
    . A fatal case of coxsackievirus B4 meningoencephalitis. Arch Neurol 2003;60:107-12.
    Abstract/FREE Full Text
  10. ↵
    1. Peigue-Lafeuille H,
    2. Croquez N,
    3. Laurichesse H,
    4. et al
    . Enterovirus meningitis in adults in 1999–2000 and evaluation of clinical management. J Med Virol 2002;67:47-53.
    CrossRefMedlineWeb of Science
  11. ↵
    1. Ramers C,
    2. Billman G,
    3. Hartin M,
    4. Ho S,
    5. Sawyer MH
    . Impact of a diagnostic cerebrospinal fluid enterovirus polymerase chain reaction test on patient management. JAMA 2000;283:2680-5.
    Abstract/FREE Full Text
  12. ↵
    1. Riding MH,
    2. Stewart J,
    3. Clements GB,
    4. Galbraith DN
    . Enteroviral polymerase chain reaction in the investigation of aseptic meningitis. J Med Virol 1996;50:204-6.
    CrossRefMedlineWeb of Science
  13. ↵
    1. Dagan R,
    2. Jenista JA,
    3. Prather SL,
    4. Powell KR,
    5. Menegus MA
    . Viremia in hospitalized children with enterovirus infections. J Pediatr 1985;106:397-401.
    CrossRefMedlineWeb of Science
  14. ↵
    1. Prather SL,
    2. Dagan R,
    3. Jenista JA,
    4. Menegus MA
    . The isolation of enteroviruses from blood: a comparison of four processing methods. J Med Virol 1984;14:221-7.
    MedlineWeb of Science
  15. ↵
    1. Thorén A,
    2. Robinson AJ,
    3. Maguire T,
    4. Jenkins R
    . Two-step PCR in the retrospective diagnosis of enteroviral viraemia. Scand J Infect Dis 1992;24:137-41.
    CrossRefMedlineWeb of Science
  16. ↵
    1. Vuorinen T,
    2. Vainionpaa R,
    3. Hyypia T
    . Five years' experience of reverse-transcriptase polymerase chain reaction in daily diagnosis of enterovirus and rhinovirus infections. Clin Infect Dis 2003;37:452-5.
    Abstract/FREE Full Text
  17. ↵
    1. Gorgievski-Hrisoho M,
    2. Schumacher JD,
    3. Vilimonovic N,
    4. Germann D,
    5. Matter L
    . Detection by PCR of enteroviruses in cerebrospinal fluid during a summer outbreak of aseptic meningitis in Switzerland. J Clin Microbiol 1998;36:2408-12.
    Abstract/FREE Full Text
    1. Hukkanen V,
    2. Vuorinen T
    . Herpesviruses and enteroviruses in infections of the central nervous system: a study using time-resolved fluorometry PCR. J Clin Virol 2002;25(Suppl):87-94.
    1. Yerly S,
    2. Gervaix A,
    3. Simonet V,
    4. Caflisch M,
    5. Perrin L,
    6. Wunderli W
    . Rapid and sensitive detection of enteroviruses in specimens from patients with aseptic meningitis. J Clin Microbiol 1996;34:199-201.
    Abstract/FREE Full Text
  20. ↵
    1. Carroll KC,
    2. Taggart B,
    3. Robison J,
    4. Byington C,
    5. Hillyard D
    . Evaluation of the Roche AMPLICOR enterovirus PCR assay in the diagnosis of enteroviral central nervous system infections. J Clin Virol 2000;19:149-56.
    CrossRefMedlineWeb of Science
  21. ↵
    1. Kinnunen E,
    2. Hovi T,
    3. Stenvik M,
    4. et al
    . Localized outbreak of enteroviral meningitis in adults. Acta Neurol Scand 1987;75:346-51.
    CrossRefMedlineWeb of Science
  22. ↵
    1. Abebe A,
    2. Johansson B,
    3. Abens J,
    4. Strannegård Ö
    . Detection of enteroviruses in faeces by polymerase chain reaction. Scand J Infect Dis 1992;24:265-73.
    MedlineWeb of Science
  23. ↵
    1. Hukkanen V,
    2. Rehn T,
    3. Kajander R,
    4. Sjöroos M,
    5. Waris M
    . Time-resolved fluorometry PCR assay for rapid detection of herpes simplex virus in cerebrospinal fluid. J Clin Microbiol 2000;38:3214-8.
    Abstract/FREE Full Text
  24. ↵
    1. Lönnrot M,
    2. Sjöroos M,
    3. Salminen K,
    4. Maaronen M,
    5. Hyypiä T,
    6. Hyöty H
    . Diagnosis of enterovirus and rhinovirus infections by RT-PCR and time-resolved fluorometry with lanthanide chelate labeled probes. J Med Virol 1999;59:378-84.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Chung PW,
    2. Huang YC,
    3. Chang LY,
    4. Lin TY,
    5. Ning HC
    . Duration of enterovirus shedding in stool. J Microbiol Immunol Infect 2001;34:167-70.
    Medline
  26. ↵
    1. Rotbart HA
    . Enteroviral infections of the central nervous system. Clin Infect Dis 1995;20:971-81.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. Clin Infect Dis. (2005) 40 (7): 982-987. doi: 10.1086/428581
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most