Objective: To analyze the early changes in the gut microbiota of elderly patients with severe burns using high-throughput sequencing of 16S ribosomal RNA (16S rRNA). Methods: Thirteen patients with severe burns (Burn group) and twelve healthy volunteers (Control group) were enrolled according to the inclusion and exclusion criteria. Clinical data and fecal samples were collected from both groups, and 16S rRNA V4 region gene sequencing was performed to assess the relative abundance of various bacterial taxa. The Rarefy method was employed to generate operational taxonomic units (OTUs), and Z-score normalization was applied to identify differentially abundant bacteria. A heatmap for differential bacterial communities was constructed. The number of fecal microbiota OTUs and diversity indices were analyzed using QIIME (version 1.9.1). Linear discriminant analysis effect size (LEfSe) was used to identify dominant bacterial groups. The functional abundance of the microbiota was predicted using PICRUSt2 software. The correlations among differential bacterial taxa at the genus level were visualized using the igraph package in R language. Data were analyzed using independent sample t-tests, Wilcoxon rank-sum tests, chi-square tests, and Kruskal-Wallis tests, with a significance threshold of P < 0.05. Results: The early gut microbiota of severe burn patients was predominantly composed of Firmicutes and Bacteroidota. Twenty-three bacterial genera exhibited significant differences between the two groups (P < 0.05). The heatmap of differential bacterial communities indicated that the Burn group had a slightly lower abundance of microbial taxa than the Control group. LEfSe analysis indicated that the LDA scores for Clostridia in the Control group and Bacilli in the Burn group were both greater than 4. Compared to the Control group, the Burn group exhibited a significant increase in bacterial species in one phylum, two classes, six orders, six families, and ten genera. The microbial distribution dendrogram suggested that Bacilli were the primary distinguishing marker for the Burn group, while Clostridia were the main marker for the Control group. KEGG functional prediction analysis indicated no significant differences between the two groups (P > 0.05). Combined network visualization analysis revealed that Firmicutes exerted the greatest influence, with 136 bacterial genera showing significant positive or negative correlations (| cor | > 0.3; P < 0.05), primarily through synergistic interactions. Conclusion: There are significant differences in the early gut microbiota of elderly patients with severe burns compared to their healthy peers, characterized by a reduction in beneficial bacterial species and abundance, an increase in the proportion of anaerobic bacteria, and decreased microbial diversity, along with certain synergistic and antagonistic interactions among different microbiota.