Diabetes mellitusDiabetic autonomic neuropathyHydrogen Breath TestMeta-analysisPrevalence

The prevalence of small intestinal bacterial overgrowth in diabetes mellitus: a systematic review and meta-analysis

  • 2022
  • Xin Feng, Xiao-Qing Li
Why this matters for your health

Diabetes affects hundreds of millions of people (382 million in 2013, projected to reach 592 million by 2035), and gut complications are common. Small intestinal bacterial overgrowth (SIBO) can cause diarrhea, bloating, malabsorption and weight loss. Earlier studies disagreed on whether diabetes raises the risk, and the authors say this is the first meta-analysis to pool the evidence. The pooled results suggest roughly three in ten people with diabetes test positive for SIBO. They also point to diabetic gut-nerve damage as a likely driver, and to SIBO possibly worsening blood-sugar control in return. The link to diabetes was not statistically significant in the main analysis, so it remains unproven.

Summary

The authors searched PubMed, EMBASE and the Cochrane Library (inception to June 2021) following PRISMA recommendations. They included 14 studies with 1,417 diabetic patients and 649 controls, rated high quality on the Newcastle–Ottawa scale. They pooled SIBO prevalence in diabetes with a random-effects model because heterogeneity was very high (I² = 92%). They ran subgroup analyses by diagnostic test, region and diabetes type, and compared diabetic patients with controls. In the discussion they propose that diabetic autonomic neuropathy slows gut movement and allows bacteria to overgrow. They also suggest that oxidative stress and inflammation may contribute, and that SIBO may in turn worsen blood-sugar control, insulin release and nutrition. These mechanisms are not tested in the paper. They conclude that diabetes “could be a predisposing factor” for SIBO, which is stronger wording than their main result supports.

Key findings

  • Pooled SIBO prevalence in diabetes was 29% (95% CI 20–39%), with individual studies ranging from 8% to 75% and very high heterogeneity (I² = 92%). Egger’s test showed no publication bias for prevalence (p = 0.6137).
  • By diagnostic test: jejunal aspirate culture (2 studies) 39% (95% CI 12–66%), lactulose breath test (6 studies) 31% (95% CI 18–43%) and glucose breath test (7 studies) 29% (95% CI 14–43%).
  • By region: Western countries 35% (95% CI 21–49%) versus Eastern countries 24% (95% CI 14–34%).
  • By diabetes type: type 2 30% (95% CI 13–47%), type 1 25% (95% CI 14–36%), and studies including both types 40% (95% CI 33–46%). The authors say type 1 and type 2 did not differ significantly.
  • Compared with controls (nine studies, 1,105 diabetic patients and 649 controls), the odds ratio for SIBO in diabetes was 2.91 (95% CI 0.82–10.32; p = 0.10), which is not statistically significant. Heterogeneity was high (I² = 89%), and the funnel plot hinted at possible publication bias.
  • The abstract and conclusion say the risk is “2.91 times higher” in diabetes. That overstates the result, because the confidence interval includes no difference.
  • In a sensitivity analysis excluding one Adamska study, the odds ratio rose to 4.18 (95% CI 1.34–13.05; p = 0.01), with I² falling to 81%. The authors explain that nearly half of that study’s controls were hospital staff and relatives. Both Adamska studies also drew on the same laboratory database during overlapping periods, so some participants may overlap.
  • The authors note that breath tests are less invasive than aspiration but can give false positives (lactulose) or false negatives (glucose). Aspiration has its own problems, including contamination by oral bacteria and missing strict anaerobes. They suggest the higher culture-based prevalence may reflect contamination.
  • Proposed explanations (not tested here): diabetic autonomic neuropathy, oxidative stress, and inflammation slowing gut movement. They also cite one study in which type 2 diabetic patients with SIBO had worse glycaemic control, lower insulin release and lower BMI. The authors speculate that SIBO may harm beta-cell function and nutrition.
  • Limitations: small studies, possible publication bias for the odds ratio, and differences in diagnostic tests and geography causing heterogeneity.
  • No funding was received, and the authors declare no conflicts of interest.

Read the full paper
VIEW PDF
Citation

Feng X, Li XQ. The prevalence of small intestinal bacterial overgrowth in diabetes mellitus: a systematic review and meta-analysis. Aging. 2022;14(2):975–988. doi:10.18632/aging.203854

Related Research Papers
n