BiomarkerCholine and L-carnitineDiabetic retinopathyDiagnostic accuracyendothelial dysfunction

Elevated plasma trimethylamine-N-oxide levels are associated with diabetic retinopathy

  • 2020
  • Weiming Liu, Chunmin Wang, Yu Xia, et al.
Why this matters for your health

Diabetic retinopathy (DR) is a major cause of vision loss in middle-aged and older people. The paper cites a global prevalence of about 34.6% among people with diabetes. Few biomarkers are used in clinical practice, so better markers could help with early prevention and personalized care for high-risk patients. TMAO is a gut-microbiome-derived metabolite already linked to cardiovascular disease and diabetes. The authors say this is the first report linking it to DR. It suggests a possible gut-microbiome connection to eye complications of diabetes, but the findings are preliminary.

Summary

This cross-sectional study at the First Affiliated Hospital of Soochow University (Suzhou, China) enrolled 162 hospital inpatients of Chinese Han ethnicity, aged 40–85, between July and December 2019. Of these, 122 had type 2 diabetes (50 without DR, 30 with non-proliferative DR and 42 with proliferative DR). The other 40 were non-diabetic patients with idiopathic macular hole, who served as controls. Fasting plasma TMAO was measured by LC-MS/MS. Patients with antibiotic or probiotic use in the past 3 months were excluded, as were vegetarians. TMAO was lowest in controls and rose from diabetes without DR to proliferative DR. Higher TMAO was linked to higher odds and greater severity of DR after adjustment for age, sex, BMI, diabetes duration, hypertension and kidney function. Adding TMAO modestly improved a diagnostic model built from known risk factors. The authors stress that causality is unproven and call for prospective studies.

Key findings

  • Median plasma TMAO (µmol/L) rose across groups: controls 0.357, diabetes without DR 0.458, non-proliferative DR 0.687 and proliferative DR 0.793 (P < 0.001).
  • TMAO was higher in all type 2 diabetes patients than in non-diabetic controls (0.610 vs 0.357; P < 0.001) and higher in DR patients than in those without DR (0.752 vs 0.458; P = 0.001).
  • Per standard-deviation increase in ln-transformed TMAO, the adjusted odds of DR were 2.62 (95% CI 1.51–4.55; P = 0.001), and 2.31 (95% CI 1.30–4.12; P = 0.005) after also adjusting for eGFR.
  • Patients in the highest TMAO quartile had far higher odds of DR than those in the lowest (OR 8.76, 95% CI 2.21–34.76; P = 0.002). The confidence intervals are wide because the sample is small.
  • Higher TMAO also predicted worse DR severity (ordinal OR 2.05, 95% CI 1.27–3.33; P = 0.004 after adjustment).
  • The association was stronger in men (OR 4.17, 95% CI 1.73–10.04) than in women (OR 2.13, 95% CI 0.97–4.65; P-interaction = 0.006). The authors suggest dietary and gut-microbiome sex differences, which was not tested.
  • TMAO alone was a modest diagnostic marker: AUROC 0.679, and a cutoff of 0.761 µmol/L gave 50% sensitivity and 86% specificity.
  • Adding TMAO to a model of known risk factors raised the AUROC from 0.646 to 0.734 (P = 0.043), and the integrated discrimination improvement was 10.7% (P < 0.001).
  • TMAO correlated positively with age (r = 0.314), diabetes duration (r = 0.233) and serum creatinine (r = 0.393), and negatively with eGFR (r = −0.528). It did not correlate with BMI or lipids.
  • Patients with proliferative DR had lower fasting glucose and HbA1c (median HbA1c 7.3% vs 9.4% in diabetes without DR). The authors attribute this to selection bias and the timing of sampling, since proliferative DR patients were recruited at vitrectomy after glucose had been well controlled. They did not include glucose or HbA1c in their analyses.
  • Proposed mechanisms (not tested here): insulin resistance and dyslipidemia, systemic inflammation, and vascular inflammation or endothelial dysfunction. The authors also speculate that TMAO is lower in Chinese than in Western populations because of diet.
  • Limitations: the cross-sectional design cannot show cause and effect or direction. The sample is small and from one hospital. Diet, plasma choline, TMA and gut flora were not measured. TMAO in eye fluids was not measured, and lifestyle and other illnesses were not considered.
  • Registered at the Chinese Clinical Trial Registry (ChiCTR1900024138). Funding came from National Natural Science Foundation of China grants, the Jiangsu Provincial Medical Innovation Team and the Soochow Scholar Project. The authors declare no conflict of interest.

Read the full paper
VIEW PDF
Citation

Liu W, Wang C, Xia Y, Xia W, Liu G, Ren C, Gu Y, Li X, Lu P. Elevated plasma trimethylamine-N-oxide levels are associated with diabetic retinopathy. Acta Diabetologica. 2021;58:221–229. doi:10.1007/s00592-020-01610-9

Related Research Papers
n