Who gets enteropathogenic E. coli infections and what else they carry

A common gut bacterium long dismissed as a threat mainly to infants in developing countries may be far more relevant to adult medicine than previously believed, according to a new study that has uncovered an unexpected link between enteropathogenic Escherichia coli (EPEC) and autoimmune disease in hospitalized adults. The research, published in the journal Gut Pathogens, analyzed 792 hospitalized adults and children who tested positive for EPEC on a rapid molecular diagnostic panel and compared them with symptomatic patients in whom no gastrointestinal pathogen could be detected. The findings suggest that the state of the gut microbiome, rather than the bacterium alone, may determine who falls ill with this poorly understood pathogen.

EPEC is a diarrheagenic strain of E. coli that attaches to the intestinal lining and disrupts the absorptive surface of the gut, producing watery diarrhea. For decades it has been regarded primarily as a cause of infantile diarrhea worldwide, and acute symptomatic infection in healthy adults was considered rare. But the diagnostic landscape has shifted dramatically in recent years. Multiplex polymerase chain reaction (PCR) panels, such as the FilmArray BioFire Gastrointestinal Panel used in this study, can detect the genetic signatures of more than twenty enteric pathogens directly from stool in about an hour, replacing slower stool cultures that often failed to identify EPEC at all. As these panels have spread through hospital laboratories, EPEC has been turning up in adult patients far more often than the old textbooks predicted, prompting clinicians to ask whether the bacterium is a genuine disease cause in adults or an incidental bystander.

To address that question, a team at the University of Florida led by Heather Lusby, Stacy G. Beal and Kenneth H. Rand conducted a retrospective case-control study covering hospitalizations from January 2016 through August 2019. They defined symptomatic diarrhea as three or more loose bowel movements within twenty-four hours and required that every patient, whether case or control, had undergone the BioFire GI panel during admission. Cases were patients positive for EPEC, including those co-infected with other organisms; controls were symptomatic patients in whom the panel detected no targets at all. Each EPEC-positive patient was matched to three controls by age within five years and by gender, producing 137 adult cases with 402 matched controls, and 73 pediatric cases with 180 controls. The researchers then mined the electronic medical record using ICD-9 and ICD-10 billing codes, supplemented by natural language processing of clinical notes, to assign each patient’s past medical history to seven categories: coronary artery disease, diabetes mellitus, chronic kidney disease, autoimmune disease, HIV infection, solid organ or bone marrow transplantation, and a history of chronic diarrhea including inflammatory bowel disease, Crohn’s disease, ulcerative colitis, celiac disease, lactose intolerance and chronic pancreatitis.

The demographic picture that emerged was itself revealing. Among children, nearly seventy percent of EPEC-positive patients were male, a marked skew compared with the roughly one-to-one ratio in adults. Co-infection was the norm rather than the exception in children: 57.5 percent of pediatric EPEC cases carried at least one additional pathogen, versus 24.8 percent of adults. Put differently, EPEC was the sole organism detected in three-quarters of adult cases but fewer than half of pediatric ones. In adults, EPEC-positive patients also stayed in the hospital about four days longer on average than controls, 13.5 days versus 9.3 days, although the difference did not reach statistical significance. Expected length of stay, a benchmark measure of how sick patients were on admission, was nearly identical between the two adult groups, hinting that EPEC itself may have delayed discharge rather than simply marking frailer patients.

The most striking result, however, concerned autoimmune disease. After applying the Benjamini-Hochberg procedure to correct for false discovery across the seven medical-history categories tested, the researchers found that 31.4 percent of adult EPEC-positive patients carried a documented autoimmune diagnosis, compared with 19.2 percent of matched controls, a difference that remained statistically significant. When one of the authors manually re-reviewed the charts of every adult classified as having autoimmune disease, confirming diagnoses through problem lists, antinuclear antibody titers of at least 1:320, other abnormal autoantibodies, positive rheumatoid factor, or repeated documentation in clinic notes, the association held and in fact sharpened: 10.9 percent of EPEC patients had verified autoimmune disease versus 3.5 percent of controls, a difference significant at p = 0.000823 before correction and still significant afterward. In the pediatric population, by contrast, no comorbidity distinguished EPEC-positive children from controls.

The authors emphasize that this connection between EPEC and autoimmune disease has not been reported before, and they propose a biological explanation rooted in gut dysbiosis, the disruption of the normal intestinal microbial community. A healthy microbiome defends its territory through what microbiologists call colonization resistance, a layered defense system. Commensal bacteria produce bacteriocins, antimicrobial peptides with both antibacterial and anti-inflammatory activities, and generate short-chain fatty acids such as butyric acid that can suppress pathogen growth and dampen the expression of virulence genes. Resident microbes also compete with incoming pathogens for nutrients and for binding sites on the intestinal wall, reinforce tight junctions between epithelial cells, maintain the protective mucus layer, and stimulate innate anti-inflammatory cytokine production. If autoimmune diseases erode this system, the argument goes, patients with those conditions may become uniquely vulnerable to EPEC colonization and infection, explaining why the bacterium appeared disproportionately in their charts.

The dysbiosis hypothesis has considerable independent support. Patients with early rheumatoid arthritis show significant alterations in their fecal microbiota compared with fibromyalgia controls, and the abnormal microbiome associated with rheumatoid arthritis partially normalizes during treatment. In systemic lupus erythematosus, disordered intestinal microbes correlate with disease activity and display a signature distinct from that of rheumatoid arthritis. Gut microbial disruption has likewise been documented in type 1 diabetes, where dysbiosis and increased intestinal permeability have been observed in children with islet autoimmunity, and in autoimmune hepatitis. A randomized controlled trial has even suggested that fecal microbiota transplantation can halt the progression of new-onset type 1 diabetes in humans. Against this backdrop, the Florida finding reframes EPEC less as an aggressive invader and more as an opportunist that exploits weakened microbial defenses.

The pediatric results add an intriguing counterpoint. Children positive for EPEC actually had shorter hospital stays than controls, 12.7 days versus 25.8 days, a seemingly counterintuitive result that the investigators attribute largely to extremely long admissions among control infants under twelve months, including newborns whose stays sometimes exceeded one hundred days. When the analysis excluded these prolonged hospitalizations, or restricted attention to children between one and seventeen years of age, the difference in length of stay largely disappeared, and the expected stay was actually longer in the control group, suggesting that those children were sicker overall for reasons unrelated to EPEC. The male predominance among pediatric EPEC cases echoes earlier observations that atypical EPEC strains, which lack some of the classic virulence genes, occur more frequently in boys, for reasons that remain unknown.

The study has limitations the authors are careful to acknowledge. It was conducted at a single academic center, and past medical history was reconstructed from billing codes and chart review, methods whose consistency within an institution may not translate across institutions. Medication data were not collected, leaving open the possibility that immunosuppressive drugs used to treat autoimmune conditions, rather than the diseases themselves, could account for the increased susceptibility to EPEC. The case-control design also cannot establish whether autoimmune disease predisposes to EPEC infection, whether EPEC infection somehow contributes to autoimmunity, or whether both share a common cause such as an underlying immune imbalance.

Even so, the implications are significant. If the association is confirmed in larger, multicenter cohorts, clinicians may come to view a positive EPEC result on a GI panel not as a laboratory curiosity but as a potential marker of an altered gut ecosystem, and possibly of unrecognized immune dysregulation in adults. The finding also strengthens the broader case that the microbiome is a gatekeeper for enteric infection, and it points toward follow-up studies that could directly measure microbial community structure, immune markers and EPEC burden in autoimmune patients. For a bacterium that clinicians have often struggled to interpret when it appears on a multiplex panel, EPEC may finally be revealing its clinical context, and that context appears to lie in the complex ecology of the gut.