Diabetic foot (DF) is a common and severe chronic complication of diabetes mellitus, characterized by high incidence, high disability rate, and high recurrence rate. It severely affects the quality of life of affected people. Recent studies suggest that dysbiosis of the skin microbiota may play a critical role in the development of DF.

To investigate the differences in plantar skin microbiota composition among patients with varying risk degrees of DF using 16S rRNA gene sequencing, and to explore the association between microbiota imbalance and DF risk, thereby providing a microbial basis for early warning and intervention strategies.

A total of 64 patients with diabetes mellitus treated in the Second Affiliated Hospital of Zunyi Medical University from June 2023 to March 2024, and 16 healthy adults during the same period, were enrolled. According to the International Working Group on the Diabetic Foot (IWGDF) risk classification, participants were divided into five groups: control group (n=16), very low-risk group (VL, n=16), low-risk group (L, n=15), moderate-risk group (M, n=16), and high-risk group (H, n=17). Plantar skin swab samples were collected for DNA extraction. The V3-V4 region of the 16S rRNA gene was amplified and sequenced using the Illumina MiSeq platform. Operational taxonomic unit (OTU) clustering and annotation were performed using QIIME 2. Microbiota differences among groups were analyzed using the Linear discriminant analysis Effect Size (LEfSe) and Metastats methods.

The α-diversity indices (Chao1, ACE, Shannon, and Simpson) of plantar skin microbiota were significantly different among the five groups with varying risk levels of DF (P<0.001). Principal coordinate analysis (PCoA) based on Bray-Curtis distances revealed a significant separation of microbiota structures among groups (P=0.001). At the phylum level, the relative abundances of Bacteroidetes and Proteobacteria sequentially increased than the previous low-risk group, whereas those of Firmicutes and Actinobacteria sequentially decreased than the previous low-risk group (P<0.001). At the genus level, the abundances of Corynebacterium, Streptococcus, and Bacteroides significantly sequentially increased than the previous low-risk group, while the abundance of Staphylococcus gradually decreased than the previous low-risk group (P<0.001). LEfSe analysis identified group-specific biomarker genera, namely the Aquabacterium (VL), Bacteroides (L), Gardnerella (M), and Corynebacterium (H)(P<0.05).

The composition of plantar skin microbiota in diabetes mellitus patients is closely associated with the risk degree of DF. With the increasing DF risk, microbiota α-diversity significantly increases and microbial community structure diverges. High-risk patients exhibit elevated levels of Gram-negative bacteria like Bacteroidetes, Proteobacteria, and Bacteroides, along with reduced levels of Gram-positive bacteria like Firmicutes, Actinobacteria, and Staphylococcus, reflecting marked microbiota dysbiosis. Distinct microbial biomarkers are observed across DF risk levels, suggesting that microbial characteristics may serve as potential targets for DF risk assessment and intervention.