Gut microbiota predict retinopathy in patients with diabetes: A longitudinal cohort studyOriginal paper
Karen Pendergrass is a microbiome researcher specializing in microbiome-targeted interventions (MBTIs). She systematically analyzes scientific literature to identify microbial patterns, develop hypotheses, and validate interventions. As the founder of the Microbiome Signatures Database, she bridges microbiome research with clinical practice. In 2012, based on her own investigative research, she became the first documented case of FMT for Celiac Disease, four years before the first published case study.
Read MoreWhat was studied?
This study examined whether gut microbiota predict diabetic retinopathy (DR) using both cross-sectional and prospective designs. It sought early microbial markers of DR risk. Stool underwent 16S rRNA sequencing, with plasma short-chain fatty acids measured by gas chromatography-mass spectrometry and blood cell mRNA sequenced. Random forest classifiers built microbial panels. Dietary questionnaires and causal mediation analysis linked diet, microbes, and DR. A 2-year follow-up tested prediction.
Who was studied?
The cross-sectional cohort was 99 people with type 2 diabetes from Shanghai communities, aged 36 to 92 (median 70 years). Of these, 46 had DR and 53 had diabetes without retinopathy. Median diabetes duration was 13 years. DR patients had higher systolic blood pressure, longer diabetes duration, and worse glycemic control. The prospective cohort followed 62 patients initially free of retinopathy over 2 years, during which 16 developed DR.
What were the most important findings?
DR patients showed decreased Butyricicoccus and Ruminococcus torques group, plus upregulated methanogenesis pathways and lower plasma short-chain fatty acids. Alpha-diversity did not differ significantly. A combined panel of 50 microbial genes and 10 genera discriminated DR with an AUC of 0.951. In the prospective cohort, 30 genera and genes predicted incident DR with an AUC of 0.864. High baseline Butyricicoccus tracked with lower DR incidence (12.1 percent versus 41.4 percent, p equals 0.018). Low baseline plasma acetate tracked with higher DR incidence (52.9 percent versus 13.9 percent, p equals 0.003).
What are the greatest implications of this study?
The work suggests gut microbial signatures, especially Butyricicoccus and plasma acetate, could serve as early indicators of diabetic retinopathy risk. Prospective validation strengthens their predictive value. Dietary factors like higher cholesterol intake correlated with the altered flora, hinting at modifiable upstream drivers. This points toward possible microbiome-targeted prevention. The cohort is small and used 16S rather than metagenomic sequencing, risking overfitting and lower resolution. The authors urge replication in larger cohorts.