World J Gastroenterol. 2013 June 28; 19(24): 3819-3823.
Published online 2013 June 28. doi: 10.3748/wjg.v19.i24.3819.
Is alcoholic pancreatitis associated with enteroviral infection?
Jahangir Khan, Isto Nordback, Hanna Seppänen, Riitta Lappalainen-Lehto, Satu Järvinen, Sami Oikarinen, Sisko Tauriainen, Sari Räty, Heikki Hyöty and Juhani Sand.
Jahangir Khan, Isto Nordback, Hanna Seppänen, Riitta Lappalainen-Lehto, Satu Järvinen, Sari Räty, Juhani Sand, Department of Gastroenterology and Alimentary Tract Surgery, Tampere University Hospital, 33521 Tampere, Finland
Sami Oikarinen, Sisko Tauriainen, Heikki Hyöty, Department of Virology, School of Medicine, University of Tampere, 33014 Tampere, Finland
Heikki Hyöty, Department of Microbiology, Center of Laboratory Medicine, Tampere University Hospital, 33521 Tampere, Finland
Author contributions: Räty S, Seppänen H, Lappalainen-Lehto R and Järvinen S recruited the patient and control material and contributed to the design of the study; Oikarinen S and Tauriainen S performed the enteroviral analysis and contributed to the analysis of the data; Khan J analyzed the data; Khan J, Nordback I, Hyöty H and Sand J wrote the paper and designed the study.
Correspondence to: Juhani Sand, MD, PhD, Director, Department of Gastroenterology and Alimentary Tract Surgery, Tampere University Hospital, PO Box 2000, 33521 Tampere, Finland. juhani.sand@pshp.fi
Telephone: +358-3-31166375 Fax: +358-3-31164358
Received April 24, 2012; Revised September 29, 2012; Accepted October 30, 2012;
Abstract
AIM: To investigate whether enteroviral infection might trigger acute pancreatitis in patients made susceptible due to high alcohol consumption.
METHODS: Patients with alcohol-induced acute pancreatitis were analyzed for signs of simultaneous or preceding enteroviral infection. We studied the serum samples of 40 patients hospitalized for alcohol-induced acute pancreatitis and 40 controls recruited from an alcohol detoxification center. Reverse transcription-polymerase chain reaction (RT-PCR) was used to detect enterovirus RNA and diagnose acute viremia. Immunoglobulin G (IgG), immunoglobulin A (IgA) and immunoglobulin M (IgM) enteroviral antibodies were measured using enzyme immunoassay to detect subacute and previous infections. The samples were considered positive when the antibody titers were ≥ 15 IU. Furthermore, using RT-PCR, we studied pancreatic biopsy samples obtained during surgery from nine patients with chronic pancreatitis, one patient with acute pancreatitis and ten control patients with pancreatic carcinoma for evidence of persisting enteroviral RNA in the pancreatic tissue.
RESULTS: No enterovirus RNA indicating acute viremia was detected by RT-PCR in the serum samples of any patient or control. A high incidence of positive antibody titers was observed in both study groups: IgM antibodies had positive titers in 5/40 (13%) vs 4/40 (10%), P = 0.723; IgG in 15/40 (38%) vs 19/40 (48%), P = 0.366; and IgA in 25/40 (63%) vs 33/40 (83%), P = 0.045, patients and controls, respectively. Ten (25%) patients had severe pancreatitis and two (5%) required treatment in intensive care. The median length of hospitalization was 7 d (range: 3-47 d). The severity of acute pancreatitis or the length of hospitalization was not associated with enteroviral IgM, IgG or IgA antibodies. Five pancreatic biopsy samples tested positive with RT-PCR, three (8%) in the control group and two (5%) in the patient group (P = 0.64).
CONCLUSION: The rate of enteroviral infection is not increased in patients with alcohol-induced acute pancreatitis when compared to alcoholics with similar high alcohol use.
Keywords: Pancreatitis, Alcoholic, Pancreatitis, Acute necrotizing, Enterovirus
INTRODUCTION
Although heavy alcohol consumption is known to be associated with the development of acute pancreatitis, surprisingly little is known of the actual mechanism behind this association. Furthermore, only a small proportion of heavy drinkers ever develop acute pancreatitis even during long-term follow up[1]. Excessive alcohol consumption has been reported to cause 9%-70% of all cases of acute pancreatitis[2-7]; being predominant in some countries (e.g., United States, Hungary and Finland), whereas gallstones are predominant in many other countries such as China, Greece and Italy. While alcohol remains a clear risk factor for acute pancreatitis, a multitude of other factors that may be genetic or environmental could be involved in triggering or modulation of the disease.
One previously suggested co-factor possibly associated with the induction of acute alcohol-associated pancreatitis is enteroviral infection. Human enteroviruses typically cause mild respiratory or gastrointestinal infections, but are also associated with myocarditis and aseptic meningitis. Over 100 enterovirus serotypes have been identified, including the polio virus. Other enteroviruses are classified as coxsackie A and B viruses, enteric cytopathogenic human orphan viruses or as numbered serotypes (e.g., enterovirus 70). The evidence suggesting an association between enteroviruses and acute pancreatitis is mostly derived from case reports[8-12] and historical serological studies[13,14]. Evidence of enterovirus infection in the pancreatic beta cells has been reported by several authors[15-17]. More recently, Ozsvár et al[18] reported a significant rise in coxsackie B virus antibody titers in acute and chronic pancreatitis patients. Recent animal studies further support a possible connection between enteroviral infection and pancreatitis[19-22]. Jerrells et al[23] reported that mice on an alcohol diet and infected with a strain of coxsackie B virus developed more severe pancreatitis than control mice, and that even typically avirulent strains produced severe pancreatitis in these mice. Clemens et al[24] showed that the pancreas of mice on an alcohol diet had impaired regeneration potential compared to control mice which may be associated with the severity of acute pancreatitis and the development of chronic pancreatitis. These studies suggest that enteroviruses may play a triggering role in at least a portion of human alcoholic pancreatitis.
To the best of our knowledge, there are no studies addressing the association between enteroviral infection and alcohol-associated acute pancreatitis in humans, where the alcohol intake of the non-pancreatitis controls has been comparable. The aim of this study was to ascertain whether patients suffering from alcohol-associated acute pancreatitis show evidence of simultaneous or preceding enteroviral infection in greater numbers than control subjects with similar recent alcohol consumption, but no previous or current pancreatitis. In addition, we analyzed pancreatic biopsy samples obtained from chronic pancreatitis patients and control patients during surgery to evaluate whether chronic pancreatitis specimens showed signs of persistent enteroviral genome in the pancreas.
MATERIALS AND METHODS
This study was a retrospective analysis of previously collected serum samples from a prospective study[25]. The study patients were recruited between January 2001 and November 2005. The samples for the first group, 40 patients hospitalized due to their first alcohol-associated acute pancreatitis, were collected during the first days of hospitalization. The samples for the control group, 40 alcoholics recruited from an alcohol detoxification center, were collected during their stay in the center. The patients were diagnosed with acute pancreatitis when they met the following criteria: acute epigastric pain that led to hospitalization, clinical signs consistent with acute pancreatitis together with serum amylase activity of at least three times the upper normal range, elevated serum inflammation markers (C-reactive protein concentration and leukocyte count), and/or the findings of acute pancreatitis on imaging. Alcohol was considered the probable etiology when the patient reported high alcohol intake during the alcohol use disorders identification test (AUDIT) or in a thorough interview of the patient or the family and other etiologies were excluded by laboratory testing and imaging[26]. Heavy alcohol consumption was similarly identified in the control subjects. Previously diagnosed pancreatitis or any acute illness were exclusion criteria when recruiting the control subjects.
The length of hospitalization, the development of complications and the need for and duration of treatment in the intensive care unit in alcohol-associated acute pancreatitis patients were recorded together with basic information such as body mass index (BMI), age and gender. Acute pancreatitis was considered severe when it met the Atlanta criteria[27]. The AUDIT questionnaire, amount of alcohol consumption (g/wk) preceding hospitalization and amount of smoking were elicited by a person specialized in addiction problems. The control group was matched according to age and reported amount of alcohol consumption. Thirty-two (80%) of the patients were male with a median age of 47 years (range: 18-73 years) and median BMI of 26 kg/m2 (range: 19-34 kg/m2). In the control group, 25 (63%) patients were male with a median age of 46 years (range: 22-66 years) and median BMI of 26 kg/m2 (range: 16-34 kg/m2). The median AUDIT scores were 22 (range: 5-38) in the patient group and 29 (range: 15-36) in the control group.
The biopsy samples were collected from 20 patients who underwent pancreatic surgery: one with alcohol-associated acute pancreatitis, nine with chronic pancreatitis and ten with pancreatic carcinoma. The development of chronic pancreatitis was alcohol associated in five and idiopathic in four patients. None of the patients with pancreatic carcinoma had a history of acute pancreatitis or excessive alcohol consumption. They were operated on between December 2001 and March 2006. The biopsy samples were analyzed for the presence of enteroviral RNA using a highly sensitive reverse transcription-polymerase chain reaction (RT-PCR) method which amplifies a sequence common to all known enterovirus serotypes. The details of this method have been described earlier[28]. Frozen tissue samples were disrupted and homogenized using the TissueRuptor homogenizator (Qiagen, Hilden, Germany). RNA was extracted from the homogenized sample using the RNeasy Mini kit (Qiagen) according to the manufacturer’s instructions.
The serum samples were stored at -70 °C during the interval between their acquisition and analysis. Evidence of enteroviral infection was analyzed by detecting immunoglobulin G (IgG), immunoglobulin A (IgA) and immunoglobulin M (IgM) class antibodies by enzyme immunoassay and by detecting enteroviral-RNA using the RT-PCR method described above. IgM class enterovirus antibodies were measured against a mixture of three enterovirus antigens (coxsackie virus B3, coxsackie virus A16 and echovirus 11) using a capture enzyme immunoassay as previously described[29]. IgG and IgA class antibodies were measured against a synthetic enterovirus peptide antigen (sequence KEVPALTAVETGAT-C derived from an immunodominant region of capsid protein VP1, which is a common epitope for several enteroviruses) as described earlier[30-32]. The samples were considered positive when the antibody titers were ≥ 15 EIU.
Statistical analysis
Statistical testing was performed with SPSS statistical software using Pearson’s correlation, χ2 test, Mann-Whitney U-test and Fisher’s Exact test. P values ≤ 0.05 were considered statistically significant. This study was performed according to the Helsinki Declaration and was approved by the Ethics Committee of Tampere University Hospital.
RESULTS
Ten (25%) patients had severe pancreatitis according to the Atlanta criteria. Of these, six patients had necrotizing pancreatitis, one developed infected necrosis and three developed pseudocysts. Two patients required treatment in intensive care. The median length of hospitalization was 7 d (range: 3-47 d).
No enterovirus RNA was detected by RT-PCR in any patient or control subject. IgM antibodies had positive titers in 5/40 (13%) vs 4/40 (10%), P = 0.723; IgG in 15/40 (38%) vs 19/40 (48%), P = 0.366; and IgA in 25/40 (63%) vs 33/40 (83%), P = 0.045, patients and controls, respectively. The severity of acute pancreatitis or the length of hospitalization was not associated with enteroviral IgM, IgA or IgG antibodies.
Three pancreatic biopsy samples from patients with pancreatic carcinoma and two biopsy samples from patients with chronic pancreatitis tested positive for enteroviral RNA. The etiology of chronic pancreatitis was alcohol consumption in both patients. The tissue specimen from the patient with alcohol-induced acute pancreatitis was negative for enteroviral RNA.
DISCUSSION
In this study, we ascertained whether patients hospitalized for their first alcohol-induced acute pancreatitis had evidence of simultaneous or preceding enteroviral infection. In animal studies, enterovirus infection has been found to cause pancreatitis and, furthermore, simultaneous consumption of alcohol has been found to exacerbate the pancreatic insult. We hypothesized that enteroviral infection might be the triggering factor in at least some of the patients with their first alcohol-induced acute pancreatitis.
All the samples analyzed in this study were stored frozen. To the best of our knowledge, no studies have been reported on the possible adverse effects of prolonged storage and thawing of samples of enteroviral antibodies or on RT-PCR sensitivity. In general, repeated freezing and thawing may slightly alter the results observed, but the cycles generally do not affect samples to any clinically significant extent[33-35].
No evidence of acute viremia was found in any of the patients. Positive IgM antibodies reflect subacute disease and 13% of our patients tested positive, with a similar rate in the control group. We also report a relatively high number of patients with positive IgA and IgG antibody titers. However, this was also the case in the control group. IgG antibodies remain elevated long after the infection, while IgA antibodies usually disappear within a few months. Therefore, we suspect that this finding reflects the fact that our patients and controls were of lower socio-economic background with a tendency to acquire such infections at an increased rate when compared to the general population. An association between lower socio-economic status and increased enteroviral infection rate has previously been reported[36,37]. Thus, our findings do not suggest a role for enteroviral infection in the pathogenesis of alcohol-induced acute pancreatitis in humans, at least to a clinically significant extent. In fact, IgG and IgA class enterovirus antibodies tended to be at lower levels in the pancreatitis group, which may reflect the general immunosuppression associated with this disease.
A surprisingly high percentage of pancreatic tissue samples obtained during surgery, from patients operated on either for chronic pancreatitis or carcinoma of the pancreas, tested positive for enteroviral RNA in RT-PCR. In an earlier study, Lászik et al[38] studied pancreatic tissue specimens obtained during surgery for acute pancreatitis using in situ hybridization and reported no evidence of enteroviral infection in any of the samples. In the present study, we did not investigate whether enteroviral genome was present in the acini or the islets of Langerhans in the pancreas. Recent studies suggest a role for enteroviral infection in the genesis of type 1 diabetes[39], and, furthermore, direct beta cell involvement[15-17]. It is therefore possible, although not certain, that the high percentage of enteroviral genome observed in the tissue samples in our study also came from the islets of Langerhans in this patient material.
In conclusion, we report no evidence of an increased rate of enteroviral infection in patients hospitalized for their first alcohol-induced acute pancreatitis when compared to alcoholics with similarly heavy alcohol consumption, but with no history or signs of acute or chronic pancreatitis. The rate of positive results in pancreatic tissue samples was clearly higher in our study than reported elsewhere, although the sample size was small.
COMMENTS
Background
Although heavy alcohol consumption is known to be associated with the development of acute pancreatitis, surprisingly little is known of the actual mechanism behind this association. Furthermore, only a small proportion of heavy drinkers ever develop acute pancreatitis even during long-term follow up.
Research frontiers
One previously suggested co-factor possibly associated with the induction of acute alcohol-associated pancreatitis is enteroviral infection. Human enteroviruses typically cause mild respiratory or gastrointestinal infections, but are also associated with myocarditis and aseptic meningitis.
Innovations and breakthroughs
There are no studies addressing the association between enteroviral infection and alcohol-associated acute pancreatitis in humans, where the alcohol intake of the non-pancreatitis controls has been comparable.
Applications
The aim of this study was to ascertain whether patients suffering from alcohol-associated acute pancreatitis show evidence of simultaneous or preceding enteroviral infection in greater numbers than control subjects with similar recent alcohol consumption, but no previous or current pancreatitis.
Peer review
This study partially answered a question in etiology of acute pancreatitis. It was well designed retrospective study.
Footnotes
P- Reviewers Muniraj T, Sezgin O S- Editor Gou SX L- Editor Webster JR E- Editor Xiong L
References
1.
Lankisch PG, Lowenfels AB, Maisonneuve P. What is the risk of alcoholic pancreatitis in heavy drinkers?. Pancreas. 2002;25:411-412.[PubMed]
2.
Frey CF, Zhou H, Harvey DJ, White RH. The incidence and case-fatality rates of acute biliary, alcoholic, and idiopathic pancreatitis in California, 1994-2001. Pancreas. 2006;33:336-344.[PubMed] [DOI]
3.
Bai Y, Liu Y, Jia L, Jiang H, Ji M, Lv N, Huang K, Zou X, Li Y, Tang C. Severe acute pancreatitis in China: etiology and mortality in 1976 patients. Pancreas. 2007;35:232-237.[PubMed] [DOI]
4.
Fagenholz PJ, Castillo CF, Harris NS, Pelletier AJ, Camargo CA. Increasing United States hospital admissions for acute pancreatitis, 1988-2003. Ann Epidemiol. 2007;17:491-497.[PubMed] [DOI]
5.
Jaakkola M, Nordback I. Pancreatitis in Finland between 1970 and 1989. Gut. 1993;34:1255-1260.[PubMed] [DOI]
6.
Gullo L, Migliori M, Oláh A, Farkas G, Levy P, Arvanitakis C, Lankisch P, Beger H. Acute pancreatitis in five European countries: etiology and mortality. Pancreas. 2002;24:223-227.[PubMed] [DOI]
7.
Sand J, Lankisch PG, Nordback I. Alcohol consumption in patients with acute or chronic pancreatitis. Pancreatology. 2007;7:147-156.[PubMed] [DOI]
8.
Lal SM, Fowler D, Losasso CJ, Berg GG. Coxsackie virus-induced acute pancreatitis in a long-term dialysis patient. Am J Kidney Dis. 1988;11:434-436.[PubMed]
9.
Dettmeyer RB, Padosch SA, Madea B. Lethal enterovirus-induced myocarditis and pancreatitis in a 4-month-old boy. Forensic Sci Int. 2006;156:51-54.[PubMed] [DOI]
10.
Fechner RE, Smith MG, Middlekamp JN. Coxsackie B virus infection of the newborn. Am J Pathol. 1963;42:493-505.[PubMed]
11.
Coplan NL, Atallah V, Mediratta S, Bruno MS, DePasquale NP. Cardiac, pancreatic, and liver abnormalities in a patient with coxsackie-B infection. Am J Med. 1996;101:325-326.[PubMed] [DOI]
12.
Chrysos G, Kokkoris S, Protopsaltis J, Korantzopoulos P, Giannoulis G. Coxsackievirus infection associated with acute pancreatitis. JOP. 2004;5:384-387.[PubMed]
13.
Arnesjö B, Edén T, Ihse I, Nordenfelt E, Ursing B. Enterovirus infections in acute pancreatitis - a possible etiological connection. Scand J Gastroenterol. 1976;11:645-649.[PubMed]
14.
Capner P, Lendrum R, Jeffries DJ, Walker G. Viral antibody studies in pancreatic disease. Gut. 1975;16:866-870.[PubMed] [DOI]
15.
Ylipaasto P, Klingel K, Lindberg AM, Otonkoski T, Kandolf R, Hovi T, Roivainen M. Enterovirus infection in human pancreatic islet cells, islet tropism in vivo and receptor involvement in cultured islet beta cells. Diabetologia. 2004;47:225-239.[PubMed] [DOI]
16.
Dotta F, Censini S, van Halteren AG, Marselli L, Masini M, Dionisi S, Mosca F, Boggi U, Muda AO, Del Prato S. Coxsackie B4 virus infection of beta cells and natural killer cell insulitis in recent-onset type 1 diabetic patients. Proc Natl Acad Sci USA. 2007;104:5115-5120.[PubMed] [DOI]
17.
Richardson SJ, Willcox A, Bone AJ, Foulis AK, Morgan NG. The prevalence of enteroviral capsid protein vp1 immunostaining in pancreatic islets in human type 1 diabetes. Diabetologia. 2009;52:1143-1151.[PubMed]
18.
Ozsvár Z, Deák J, Pap A. Possible role of Coxsackie-B virus infection in pancreatitis. Int J Pancreatol. 1992;11:105-108.[PubMed]
19.
Huber S, Ramsingh AI. Coxsackievirus-induced pancreatitis. Viral Immunol. 2004;17:358-369.[PubMed] [DOI]
20.
Minnich LL, Ray CG. Variable susceptibility of mice to group B coxsackievirus infections. J Clin Microbiol. 1980;11:73-75.[PubMed]
21.
Ostrowski SE, Reilly AA, Collins DN, Ramsingh AI. Progression or resolution of coxsackievirus B4-induced pancreatitis: a genomic analysis. J Virol. 2004;78:8229-8237.[PubMed] [DOI]
22.
Tracy S, Höfling K, Pirruccello S, Lane PH, Reyna SM, Gauntt CJ. Group B coxsackievirus myocarditis and pancreatitis: connection between viral virulence phenotypes in mice. J Med Virol. 2000;62:70-81.[PubMed] [DOI]
23.
Jerrells TR, Chapman N, Clemens DL. Animal model of alcoholic pancreatitis: role of viral infections. Pancreas. 2003;27:301-304.[PubMed]
24.
Clemens DL, Jerrells TR. Ethanol consumption potentiates viral pancreatitis and may inhibit pancreas regeneration: preliminary findings. Alcohol. 2004;33:183-189.[PubMed]
25.
Pelli H, Sand J, Laippala P, Nordback I. Long-term follow-up after the first episode of acute alcoholic pancreatitis: time course and risk factors for recurrence. Scand J Gastroenterol. 2000;35:552-555.[PubMed] [DOI]
26.
Nordback I, Sand J, Andrén-Sandberg A. Criteria for alcoholic pancreatitis. Results of an international workshop in Tampere, Finland, June 2006. Pancreatology. 2007;7:100-104.[PubMed] [DOI]
27.
Bradley EL. A clinically based classification system for acute pancreatitis. Summary of the International Symposium on Acute Pancreatitis, Atlanta, Ga, September 11 through 13, 1992. Arch Surg. 1993;128:586-590.[PubMed] [DOI]
28.
Lönnrot M, Sjöroos M, Salminen K, Maaronen M, Hyypiä T, Hyöty H. Diagnosis of enterovirus and rhinovirus infections by RT-PCR and time-resolved fluorometry with lanthanide chelate labeled probes. J Med Virol. 1999;59:378-384.[PubMed] [DOI]
29.
Viskari HR, Roivainen M, Reunanen A, Pitkäniemi J, Sadeharju K, Koskela P, Hovi T, Leinikki P, Vilja P, Tuomilehto J. Maternal first-trimester enterovirus infection and future risk of type 1 diabetes in the exposed fetus. Diabetes. 2002;51:2568-2571.[PubMed] [DOI]
30.
Sadeharju K, Knip M, Virtanen SM, Savilahti E, Tauriainen S, Koskela P, Akerblom HK, Hyöty H. Maternal antibodies in breast milk protect the child from enterovirus infections. Pediatrics. 2007;119:941-946.[PubMed] [DOI]
31.
Samuelson A, Glimåker M, Skoog E, Cello J, Forsgren M. Diagnosis of enteroviral meningitis with IgG-EIA using heat-treated virions and synthetic peptides as antigens. J Med Virol. 1993;40:271-277.[PubMed] [DOI]
32.
Hyöty H, Hiltunen M, Knip M, Laakkonen M, Vähäsalo P, Karjalainen J, Koskela P, Roivainen M, Leinikki P, Hovi T. A prospective study of the role of coxsackie B and other enterovirus infections in the pathogenesis of IDDM. Childhood Diabetes in Finland (DiMe) Study Group. Diabetes. 1995;44:652-657.[PubMed] [DOI]
33.
Kueltzo LA, Wang W, Randolph TW, Carpenter JF. Effects of solution conditions, processing parameters, and container materials on aggregation of a monoclonal antibody during freeze-thawing. J Pharm Sci. 2008;97:1801-1812.[PubMed] [DOI]
34.
Männistö T, Surcel HM, Bloigu A, Ruokonen A, Hartikainen AL, Järvelin MR, Pouta A, Vääräsmäki M, Suvanto-Luukkonen E. The effect of freezing, thawing, and short- and long-term storage on serum thyrotropin, thyroid hormones, and thyroid autoantibodies: implications for analyzing samples stored in serum banks. Clin Chem. 2007;53:1986-1987.[PubMed] [DOI]
35.
Pinsky NA, Huddleston JM, Jacobson RM, Wollan PC, Poland GA. Effect of multiple freeze-thaw cycles on detection of measles, mumps, and rubella virus antibodies. Clin Diagn Lab Immunol. 2003;10:19-21.[PubMed]
36.
Honig EI, Melnick JL, Isacson P, Parr R, Myers IL, Walton M. An endemiological study of enteric virus infections: poliomyelitis, coxsackie, and orphan (ECHO) viruses isolated from normal children in two socioeconomic groups. J Exp Med. 1956;103:247-262.[PubMed] [DOI]
37.
Seiskari T, Kondrashova A, Viskari H, Kaila M, Haapala AM, Aittoniemi J, Virta M, Hurme M, Uibo R, Knip M. Allergic sensitization and microbial load--a comparison between Finland and Russian Karelia. Clin Exp Immunol. 2007;148:47-52.[PubMed] [DOI]
38.
Lászik ZG, Kallajoki M, Hyypiä T, Rima B, Aho HJ, Nevalainen TJ. Mumps, enteroviruses, and human acute pancreatitis. Scand J Gastroenterol. 1990;25:906-910.[PubMed]
39.
Yeung WC, Rawlinson WD, Craig ME. Enterovirus infection and type 1 diabetes mellitus: systematic review and meta-analysis of observational molecular studies. BMJ. 2011;342:d35.[PubMed] [DOI]