Human Enterovirus 71 Disease in Sarawak, Malaysia: A Prospective Clinical, Virological, and Molecular Epidemiological Study
- Mong How Ooi1,3,
- See Chang Wong1,
- Yuwana Podin2,
- Winnie Akin2,
- Syvia del Sel3,
- Anand Mohan1,
- Chae Hee Chieng1,
- David Perera2,
- Daniela Clear3,
- Darin Wong3,
- Emma Blake3,
- Jane Cardosa2, and
- Tom Solomon3
- 1Department of Paediatrics, Sibu Hospital, Sibu
- 2Institute of Health and Community Medicine, Universiti Malaysia Sarawak, Kota, Samarahan, Malaysia
- 3Viral Brain Infections Group, Divisions of Neurological Science and Medical Microbiology, University of Liverpool, United Kingdom
- Reprints or correspondence: Dr. Tom Solomon, Viral Brain Infections Group, University of Liverpool, Div. of Medical Microbiology, 8th Fl., Duncan Bldg., Daulby St., Liverpool L69 3GA, UK (tsolomon{at}liv.ac.uk).
Abstract
Background. Human enterovirus (HEV)–71 causes large outbreaks of hand-foot-and-mouth disease with central nervous system (CNS) complications, but the role of HEV-71 genogroups or dual infection with other viruses in causing severe disease is unclear.
Methods. We prospectively studied children with suspected HEV-71 (i.e., hand-foot-and-mouth disease, CNS disease, or both) over 3.5 years, using detailed virological investigation and genogroup analysis of all isolates.
Results. Seven hundred seventy-three children were recruited, 277 of whom were infected with HEV-71, including 28 who were coinfected with other viruses. Risk factors for CNS disease in HEV-71 included young age, fever, vomiting, mouth ulcers, breathlessness, cold limbs, and poor urine output. Genogroup analysis for the HEV-71–infected patients revealed that 168 were infected with genogroup B4, 68 with C1, and 41 with a newly emerged genogroup, B5. Children with HEV-71 genogroup B4 were less likely to have CNS complications than those with other genogroups (26 [15%] of 168 vs. 30 [28%] of 109; odds ratio [OR], 0.48; 95% confidence interval [CI], 0.26–0.91; P = .0223) and less likely to be part of a family cluster (12 [7%] of 168 vs. 29 [27%] of 109; OR, 0.21; 95% CI, 0.10–0.46; P < .0001); children with HEV-71 genogroup B5 were more likely to be part of a family cluster (OR, 6.26; 95% CI, 2.77–14.18; P < .0001). Children with HEV-71 and coinfected with another enterovirus or adenovirus were no more likely to have CNS disease.
Conclusions. Genogroups of HEV-71 may differ with regard to the risk of causing CNS disease and the association with family clusters. Dual infections are common, and all possible causes should be excluded before accepting that the first virus identified is the causal agent.
Since 1997, countries of the Asia-Pacific rim have been affected by large outbreaks of human enterovirus (HEV)–71-associated hand-foot-and-mouth disease (HFMD), which have resulted in hundreds of thousands of cases and many deaths [1–7]. HFMD is a common exanthema of young children, characterized by fever, rash on the palms and bottoms of the feet, and ulcers in the oral cavity. Most patients have mild cases of disease, but some patients develop severe neurological complications (i.e., aseptic meningitis, acute flaccid paralysis, and encephalitis) or systemic disease (i.e., shock and cardiac dysfunction). HFMD is caused by enteroviruses (genus Enterovirus, family Picornoviridae), particularly Coxsackie A virus (CAV)–10, CAV-16, and HEV-71, which are mostly transmitted by the fecal-oral route. Phylogenetic studies have divided HEV–71 strains into genogroups A, B, and C, which have been further subdivided [4, 8–10]. The incidence of CNS disease and other severe complications appears to have varied among outbreaks of HEV-71 infection. The reason for this is unclear, but differences between genogroups [10–12] and coinfection with other viruses, such as a newly characterized adenovirus [1, 13], have been postulated. However, comparisons between outbreaks have been hampered by the retrospective nature of many studies and differences in inclusion criteria, definitions of severe disease, and viral diagnostic capabilities. We conducted a prospective, clinical, diagnostic, and molecular virological study of HEV-71–associated HFMD in 1 context for 3.5 years. We focused on risk factors for neurological disease and the putative role of different HEV-71 genogroups and dual infections in pathogenesis.
Materials, and Methods
Setting. The study was conducted at the pediatric wards and intensive care unit at Sibu Hospital (Sarawak, Malaysia); it commenced in January 2000 and continued for 3.5 years. This 550-bed hospital serves Sibu (population, 200,000) and receives referrals from district hospitals in the Rejang basin (total population, ∼650,000). The study was approved by the Director of Health for Sarawak (Malaysia) and the Ethics Committee of the Liverpool School of Tropical Medicine (Liverpool, UK). Informed consent was obtained from each child's accompanying parent or guardian.
Case definitions. Because enteroviruses can cause HFMD, CNS infection, or both, children (age 1 month to <12 years) were screened for both of these conditions. Children with HFMD were classified according to an algorithm as having mild HFMD, severe non–CNS-HFMD, or HFMD with CNS complications (figure 1). Children with suspected CNS infection, but no HFMD, were classified as having aseptic meningitis (ASM) or viral encephalitis on the basis of clinical and CSF findings; however, there were no children with enterovirus infection in this last group; thus, they are not discussed further.
Figure 1
Algorithm showing the classification of patients in the study. Because human enterovirus (HEV)–71 can cause hand-foot-and-mouth disease (HFMD), CNS infection, or both, children with HFMD or suspected CNS infection were studied. Cases of HFMD were classified as mild HFMD, severe non-CNS HFMD, and HFMD complicated by CNS disease; children with viral CNS infection were classified as having aseptic meningitis (ASM) or viral encephalitis, but there were no children in the latter group infected with enteroviruses during this study. The final number of patients infected with HEV-71 in each diagnostic group is shown in the figure. T, temperature. *Herpangina was defined as multiple oral ulcers predominantly affecting the posterior parts of the oral cavity.
Clinical methods. A detailed history was recorded, and a clinical examination was performed by a pediatrician of the study team who looked especially for mucocutaneous lesions and cardiovascular and neurological signs; all details were recorded on standardized forms. For viral culture, specimens were swabbed from the throat and rectum of every patient, as well as from at least 1 of the vesicles on the skin and oral ulcers (if present). Blood samples were obtained for viral studies. In patients with severe disease, complete blood cell count, urea level, electrolyte level, and glucose level were determined, and electrocardiography and echocardiography were performed. CSF specimens were examined for cell count and differential, protein level, and glucose level, and Gram stain, bacterial culture, and viral studies were performed; lumbar puncture was repeated if there was a strong clinical suspicion of viral CNS infection and if the initial examination was not confirmatory. Lumbar puncture was delayed in patients who were severely unhealthy. Patients were examined daily, or more frequently (as indicated), by a member of the study team. Children with HFMD and CNS complications were treated with intravenous immunoglobulin (IVIG; Intragam P-CSL, which is derived from local Malaysian blood donors), on a presumptive basis [1, 14], at the discretion of the treating physician.
Virological methods. Viral isolation was attempted on all swab specimens, CSF specimens, and any serum samples remaining after other investigations had been completed through the inoculation of rhabdomyosarcoma (Rd) and 293 cells [13]. Isolated enteroviruses were typed by nucleotide sequencing of VP1 and VP4 genes [8, 15] and genogrouped by phylogenetic analysis [8–10]. Adenoviruses were cultured and typed in accordance with the methods described previously [13]. Paired serum samples (obtained on the day of admission, on day 7, or on the day of discharge) and CSF specimens were also tested for IgM against dengue and Japanese encephalitis virus (JEV) in parallel, using an IgM capture ELISA [16]. Seroconversion from a negative result to a positive result for a second sample or an increasing IgM optic density was considered evidence of acute infection; antibody in the CSF was diagnostic of flavivirus CNS infection [17, 18]. A decreasing IgM optic density was considered evidence of a recent infection, and low IgM optic density in the serum of children recently vaccinated against JEV was ascribed to vaccination.
Statistical analysis. Normally distributed data were compared using Student's t test; data that were not normally distributed were compared by the Mann-Whitney U test (Staview 4.02; Abacus Concepts). Differences between proportions were tested using the χ2 test with Yates's correction or Fisher's exact test (EpiInfo, version 6; Centers for Disease Control and Prevention). Variables that might relate to disease pathogenesis that were associated with CNS disease in univariate analyses were also examined in a stepwise logistic regression (SPSS software, version 9; SPSS).
Results
Epidemiology
During the study, 773 children (485 boys [63%]) were recruited. Five hundred forty children (70%) had mild HFMD; 83 (11%) had severe HFMD with no CNS complications; 102 (13%) had HFMD with CNS disease, 3 of whom died; and 48 (6%) had ASM, 1 of whom died. Most of the children were recruited during 2 large outbreaks of HEV-71–associated HFMD during 2000, 2001, and 2003 (figure 2). Genogroup analysis showed that in the outbreak that occurred during 2000–2001, most HEV-71 isolates belonged to genogroup B4 [9, 10], and some belonged to C1. In 2003, in addition to genogroup C1 strains, a previously undescribed genogroup, which has been named B5, emerged during the outbreak.
Figure 2
Epidemic curve of 773 children with hand-foot-and-mouth disease (HFMD) and/or aseptic meningitis (ASM) during the period from 1 January 2000 through 31 July 2003. The continuous line describes the total number of patients recruited into the study each month; the histogram below shows the number of patients with a positive virus isolate, who are grouped as follows: human enterovirus (HEV)–71, HEV-71 isolated; HEV-71/Coxsackie A virus (CAV)–16, HEV-71 and CAV-16 isolated; CAV-16, CAV-16 isolated; CAV, other CAVs isolated, but not CAV-16 or HEV-71; CBV, other CBVs isolated but not CAV-16 or HEV-71; and Echo, echoviruses isolated, but not CAV-16 or HEV-71. The inset boxes show the number of patients with different genogroups of HEV-71 isolates during the 2000 and 2003 outbreaks.
Most of the patients entered the study once, but 4 boys were admitted to the study twice. For the purposes of analysis, each hospitalization was considered to be a separate entity. Thirteen other children had also been hospitalized in the past with HFMD (outside the study). For 211 children (27%), there was a history of contact with another person infected with HFMD. One hundred three children (13%) had at least 1 other child in the family (sibling or cousin; maximum, 4) admitted into the study and were considered to be part of a family cluster. Seventy-seven of the children of these 44 family clusters had mild disease, 9 had severe non-CNS HFMD, and 17 had HFMD with CNS complications. When the index case of a family cluster could be identified clearly, they were no more or less likely to have CNS disease than subsequent cases. Forty-one of the children in family clusters had HEV-71 isolated (with only 1 genogroup isolated from each family); 3 of these children also had CAV-16 isolated; thus, in 2 families, both CAV-16 and HEV-71 were isolated. Twelve other children had CAV-16, and 3 had other viruses.
Virology
Dual infections. We attempted virus isolation on 3006 samples (622 throat swab specimens, 631 rectal swab specimens, 768 mouth ulcer swab specimens, 706 vesicle swab specimens, 91 serum samples, and 188 CSF samples) from 672 (87%) of the 773 patients seen during the study period. There were no important demographic differences between those who were investigated and those who were not investigated. Two hundred seventy-seven patients (41%) had HEV-71 isolated. Twenty-eight of these patients with HEV-71 had a second virus isolated, including 14 patients coinfected with CAV-16 and 9 coinfected with other enteroviruses (CAV-4, CAV-5, CA-16, CAV-10, Echo 25, CBV-1, and CAV-24); 1 of each was coinfected with an oral polio vaccine virus, an untyped enterovirus, adenovirus (Ad)–1 (isolated from the rectum), Ad7 (rectum), or an unidentified virus (isolated from the CSF). CAV-16 was isolated from an additional 85 patients, 4 of whom had coinfection; 2 had CBV-1 isolated from the throat, and 2 had adenoviruses isolated from the rectum (Ad2 in one and Ad4 in the other). Fifty-eight patients were infected with other enteroviruses, adenoviruses, or unidentified viruses (12 of whom had multiple viruses isolated). Most of the presenting clinical features of the HEV-71–infected and virus-isolation–negative patients were similar, with the exception of 26 (10%) of the 252 patients who had no virus isolated and had seizures, compared with only 6 (2%) of the 263 patients infected with HEV-71 (OR, 0.20; 95% CI, 0.07–0.53; P < .001).
Seven hundred sixteen patients had serum samples, and, when available, CSF specimens (233 patients) tested for antibodies to JEV and dengue. Three patients had serological evidence of acute JEV infection, including a girl with mild HFMD and HEV-71 isolated from vesicles and the throat and a boy with mild HFMD and fever who had HEV-71 isolated from vesicles and CAV-16 isolated from the throat that had acute peripheral JEV infection (seroconversion in the serum). In addition, an 18-month-old boy with HFMD and CNS disease (manifested by a generalized seizure and a CSF pleocyte level of 10 cells/µL) had CAV-16 isolated from a vesicle, which seroconverted to JEV in both the serum and CSF; he made a full recovery. Three patients had acute dengue. All had HFMD and CNS disease (defined as lethargy, irritability, and CSF pleocytosis) and HEV-71 isolated (1 from the CSF, 1 from the throat and vesicles, and 1 from vesicles and the rectum). Dengue was diagnosed by IgM seroconversion, but there was no dengue antibody in the CSF. Two patients with HFMD had serological evidence of a recent JEV infection or recent vaccination, and 6 patients had serological evidence of recent dengue. Only 2 of the children with CAV-16 and HFMD had CNS disease; 1 of these was the child coinfected with JEV described above and for the other child, CAV-16 was isolated from rectal and vesicle swab specimens, and CBV-1 was isolated from the throat.
HEV-71 genogroups. Phylogenetic analysis of all 277 HEV-71 isolates showed that 168 (61%) of the isolates belonged to genogroup B4 (165 of these were isolated in 2000, 1 was isolated in 2002, and 2 were isolated in 2003), and 68 (25%) of the isolates belonged to C1 (4 of these were isolated in 2000, and 64 were isolated in the 2003 outbreak); in addition, 41 HEV-71 strains (15%) isolated in 2003 were from a new genogroup (B5) that emerged during the course of the study [9, 19, 20]. Neighbor-joining trees of the VP1 (figure 3A) and VP4 (figure 3B) genes revealed that representative HEV-71 strains from all 3 genogroups (B4, B5, and C1), described at Sibu, clustered tightly with bootstrap support values >90% and >70%, respectively.
Figure 3
Neighbor-joining phylogenetic trees of human enterovirus (HEV)–71 strains isolated in this study. A, Tree based on the alignment of a partial VP1 (nucleotide positions 2442–3281) or complete VP1 (nucleotide positions 2442–3332) gene sequences. The tree shows 14 representative strains isolated in this study (boxed) and representative HEV-71 strains for other genogroups. B, Tree based on the alignment of a partial VP4 (nucleotide positions 744–950). The tree shows 7 representative strains isolated in this study (boxed) and representative HEV-71 strains for other genogroups. For both trees, sequence from the prototype Coxsackie A virus (CVA)–16-G10 was used as an outgroup. The year of isolates is given as a suffix, and Genbank accession numbers are in parentheses. Horizontal branch lengths are proportional to the number of nucleotide changes between viruses, and the scale and the bootstrap values in 1000 pseudoreplicates for major lineages within the dendrogram are shown as percentages. PCR products were sequenced using Big Dye, version 3.0, and run on an ABI377 automated sequencer (Applied Biosciences), and nucleotide sequences were aligned (DNASTAR).
Clinical Features of HEV-71 Infection
Nearly two-thirds of the patients with HEV-71 infection had mild HFMD; 12% had severe non-CNS disease, 19% had HFMD with CNS disease, and only 1% had ASM (table 1). For the purposes of analysis, patients with CNS infection, defined by a CSF pleocytosis (with or without HFMD), were compared with patients with no CNS disease (including mild and severe non-CNS HFMD). Patients with CNS disease were younger and were more likely to give a history of fever, vomiting, mouth ulcers, breathlessness, cold limbs, poor urine output, and neurological features, including altered sensorium and seizures (table 1). During examination, patients with CNS disease were more likely to look toxic, be dehydrated and febrile, have altered consciousness, and experience neck stiffness, but they were less likely to have mouth ulcers (table 2). Excluding the 4 patients with ASM and no rash did not significantly alter these findings. The CSF was examined in 90 children (34%) infected with HEV-71, of whom 56 (62%) had a CSF WBC count >5 cells/µL. For these patients, the median WBC count was 75 cells/µL (range, 6–1090 cells/µL). Typically, this was a lymphocytic CSF with a normal glucose ratio, but 12 (21%) had neutrophil predominance, and 10 (18%) had a CSF to plasma glucose ratio <50%.
Table 1
Features in the history for 277 patients who had human enterovirus (HEV)–71 isolated.
Table 2
Examination and investigation findings for 277 patients who had human enterovirus (HEV)–71 isolated.
Three children had HFMD complicated by acute flaccid paralysis; 1 had HEV-71–B4 genogroup isolated from his throat, and the 2 others had negative culture results. Sixty-two children with severe HFMD had cardiac enzyme levels measured (27 children with HEV-71). Seventy-eight of the 233 patients with more severe HFMD were treated with IVIG on a presumptive basis, including 42 patients who were infected with HEV-71. Five patients with HEV-71 required ionotropic support, and 3 required ventilation. Four children died, including 2 who had HEV-71 isolated (1 genogroup B4 and 1 genogroup C1).
Pathogenesis of Neurological HEV-71 Disease
Children from whom HEV-71 and another virus were isolated were no more likely than those with HEV-71 alone to have neurological involvement (2 [7%] of 28 vs. 54 [22%] of 249; OR, 0.28; 95% CI, 0.04–1.26; P = .1167). Repeating this analysis excluding children with only rectal isolates (which could represent continued carriage rather than acute pathogenic infection) did not alter these findings (2 [7%] of 28 vs. 47 [21%] of 226; OR, 0.29; 95% CI, 0.05–1.34; P = .1407). All 3 of the children with HEV-71 and serological evidence of acute dengue infection had HFMD with CNS disease, compared with 52 (20%) of 264 HEV-71–infected children with no dengue infection (P = .008). In contrast, the 2 HEV-71–infected children with serological evidence of JEV infection (as shown by seroconversion in the serum) had HFMD with no CNS disease, compared with 55 (21%) of 266 children with no JEV infection (P = 1).
To examine the possible role of different HEV-71 genogroups in the clinical phenotype of HEV-71 infection, we determined the likelihood of CNS disease for each genogroup. Children with B4 genogroup were less likely to have CNS infection than were those with other genogroups; 26 (15%) of 168 children infected with genogroup B4 had CNS infection, compared with 30 (28%) of 109 with other genotypes (OR, 0.48; 95% CI, 0.26–0.91; P = .0223) (table 2). Pair-wise comparisons of the genogroups showed that this occurred because genogroup B4 was associated with less risk of CNS disease than genogroup C1 (26 [15%] of 168 vs. 19 [28%] of 68; OR, 0.47; 95% CI, 0.23–0.98; P = .0429), whereas there was no altered risk for genogroup B4 versus B5 (26 [15%] of 168 vs. 11 [27%] of 41; OR, 0.50; 95% CI, 0.21–1.21; P = .139) or for B5 versus C1 (11 [27%] of 41 vs. 19 [28%] of 68 [28%]; OR, 0.95; 95% CI, 0.36–2.45; P = .9240). Children with HEV-71 genogroup B4 were also less likely to be part of a family cluster; 12 (7%) of 168 children with genogroup B4 positive were in family clusters versus 29 (27%) of 109 children with other genogroups (OR, 0.21; 95% CI, 0.10–0.46; P < .0001). In contrast, those with B5 genogroup were more likely to be part of a family cluster; 17 (41%) of 41 children with genogroup B5 were in a cluster versus 24 (10%) of 236 children with other genogroups (OR, 6.26; 95% CI, 2.77–14.18; P < .0001). Genogroup, age, and being part of a family cluster were initially entered into a multiple logistic regression model; increasing age and genogroup B4 versus C1 were associated with a reduced risk of CNS infection (table 3), whereas being in a family cluster did not affect risk.
Table 3
Multivariate logistic regression analysis of variables associated with CNS disease in human enterovirus (HEV)–71 infection.
Discussion
Since 1997, HEV-71 infection has become a major public health problem in developed and developing countries in the Asia-Pacific region, and it has the potential to spread further. Molecular epidemiological studies have documented remarkable changes in the circulating HEV-71 genogroups in the Asia-Pacific region during this time [9, 10, 21–23]. However, whether such genetic differences explain the differing clinical presentations is not clear [4, 24]. Other postulated factors include differences in pre-existing immunity of the pediatric population [25], differences in host genetic susceptibility [12, 26, 27], and coinfection with other viruses [1]. Our detailed investigation of 773 patients, which included 277 patients with HEV-71 infection, has produced intriguing clinical epidemiological evidence and suggests that the behavior of genogroups of HEV-71 may differ. Children infected with genogroup B4 were less likely to present with CNS infection than those infected with C1 or B5 and were also less likely to be part of a family cluster. In contrast, children infected with B5 were more likely to be part of a family cluster, and there was a trend towards a greater incidence of CNS disease in these patients.
However, our findings must be interpreted with caution. Although we attempted to minimize confounding variables, there were factors beyond our control. In particular, the likelihood of children presenting to the hospital with HFMD may have varied during the study, although altered referral rates would be unlikely to explain the different risk of family clustering for the different genogroups. There may also have been other inherent features that differed between the cohort studied in 2000 and 2003, although the basic demographic features were similar (data not shown).
Our study has also confirmed findings from earlier reports that young age, fever, vomiting, and hyperglycemia are associated with severe HEV-71 infection [28, 29]; in addition, we found tachycardia, breathlessness, and the absence of mouth ulcers were associated with CNS disease. Overall, the incidence of severe disease was milder in our study than in previous studies, because only 4 fatalities occurred in our study, all of which were a result of pulmonary edema. In addition, our series was unusual in having only 4 patients with pure aseptic meningitis.
Our study has also made an important contribution to the debate about the role of dual infections in viral CNS disease. As others have shown previously [30], we found several enteroviruses were cocirculating during the HFMD outbreaks. We also documented instances of coinfection of HEV-71 with other enteroviruses in individual patients. We found no evidence of an increased likelihood of CNS disease in such patients, although we did find an intriguing association of coinfection with dengue viruses and neurological presentation. Previously, we postulated that a new adenovirus type 21 might make an important contribution to fatal disease during an HEV-71–associated HFMD outbreak [1, 13]; in this new study, we isolated adenoviruses from only 4 patients, and none was adenovirus type 21. Our rigorous investigations for multiple viral agents showed the importance of pursuing all possible avenues before attributing causation to the first pathogen encountered. For example in 2 children, CNS disease would have been attributed to CAV-16, which is an unusual cause of CNS disease, had our thorough diagnostic examination not revealed alternative, more likely viral causes (JEV in 1 child and CBV-1 in the other). Whether unsuspected dual infections lie behind other unusual manifestations of common, relatively benign enteroviruses, such as acute flaccid paralysis recently attributed to CAV-24 [31], is not known .
In summary, our detailed study provides clinical epidemiological evidence for different biological behavior of HEV-71 genogroups, with regard to transmissibility within families and risk of CNS disease. It also highlights the importance of detailed investigations for multiple pathogens during HFMD outbreaks. Only with such thorough investigations will the pathophysiology of HEV-71 be fully determined.
Acknowledgments
We thank Tan Sri Dr. Mohamad Taha Arif, Director-General of Health, (Ministry of Health of Malaysia, Kuala Lumpur), and Dr. Yao Sik Chi, State Director, for granting us permission to publish this article and for their support; the Director, Abdul Rahim Abdullah, and the doctors, nurses, and medical records officers of Sibu Hospital, Flora Ong, Martin Wong, Peng Chin Pek, Guloi Selingau, Teresa Preston, Tio Phaik Hooi, Apandi Mohammed, and Peter McMinn for administrative, clinical, and laboratory assistance; and D.W. Chadwick and C.A. Hart for their support.
Financial support. The Ministry of Science, Technology, and Innovation, Government of Malaysia (06-02-09-002 BTK/ER/003); the Walton Centre for Neurology and Neurosurgery Research Fund; and the Wellcome Trust of Great Britain. M.H.W. is a Wellcome Trust Clinical Training Fellow, and T.S. is a United Kingdom Medical Research Council Senior Clinical Fellow.
Potential conflicts of interest. All authors: no conflicts.
- Received May 23, 2006.
- Accepted October 23, 2006.
- © 2007 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)
- Full Text (PDF)
Classifications
-
- Articles and Commentaries
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
