Extremely old people have such high mortality rates that studies such as this one here become practical, answering the question of how the gut microbiome changes in the final decline into death. It is well established that the gut microbiome is influential on health, and undergoes detrimental changes across the course of adult life, although it remains to be determined as to which of the possible mechanisms are most important. In particular, it is unclear as to whether gut microbiome changes provoke inflammatory immune dysfunction or whether age-related immune dysfunction allows more inflammatory microbes to prosper. Or whether both directions of causation are relevant.
Several studies have revealed certain unique characteristics of gut microbiome in centenarians. We established a prospective cohort of fecal microbiota and conducted the first metagenomics-based study among centenarians. The objective was to explore the dynamic changes of gut microbiota in healthy centenarians and centenarians approaching end of life and to unravel the characteristics of aging-associated microbiome. Seventy-five healthy centenarians participated in follow-up surveys and collection of fecal samples at intervals of 3 months. Data pertaining to dietary status, health status scores, cause of disease and death, and fecal specimens were collected for 15 months.
Twenty participants died within 20 months during the follow-up period. The median survival time was 8-9 months and the mortality rate was 14.7% per year. The health status scores before death were significantly lower than those at 3 months before the end of the follow-up period. At this time, the participants mainly exhibited symptoms of anorexia and reduced dietary intake and physical activity. Metagenomics sequencing and analysis were carried out to characterize the gut microbiota changes in the centenarians during their transition from healthy status to death.
Analysis showed a significant change in gut microbiota from 7 months prior to death. All participants were grouped with 7 months before death as cut-off; no significant difference in α diversity was found between the two groups. Analysis revealed significant changes in the abundance of ten bacterial species before death; of these, eight species were significantly reduced (Akkermansia muciniphila, Alistipes finegoldii, Alistipes shahii, Bacteroides faecis, Bacteroides intestinalis, Butyrivibrio crossotus, Bacteroides stercoris, and Prevotella stercorea) while two were significantly increased before death (Bifidobacterium longum and Ruminococcus bromii). We speculate that these changes might occur before the clinical symptoms of deterioration in health status.