Prebiotics preserve the microbial diversity and maintain the population of butyrate-producing bacteria and butyrate concentration after all-HSCT. (A) The SI in the gut microbiota during the pretransplantation phase and on day 28. The SI during the pretransplantation phase was significantly higher in the historical control group than in the prebiotics group (P = .011), whereas there was no significant difference between the groups on day 28 (P = .444) (left). Exclusion of patients who did not receive any antibiotics at the time of pretransplantation feces collection did not affect the outcome (right). (B) The microbial diversity before and on day 28 after allo-HSCT were evaluated by SI. Patients were classified into 3 groups according to the dynamics of the SI (deteriorated [≤−0.5], invariant [from −0.5 to 0.5], and improved [≥0.5]). Compared with the historical control group, the microbial diversity in the prebiotics group was well maintained or even improved in some cases (deteriorated, 57.7% vs 76.4%; invariant, 23.3% vs 20.8%; improved, 20% vs 2.8%; P = .004). (C) Butyrate-producing bacteria were quantified at preconditioning and on day 28 after allo-HSCT. Compared with the historical control group, the butyrate-producing bacteria counts in the prebiotics group were sustained (P = .027; prebiotics, n = 30; historical control, n = 72). Fecal butyrate concentration was quantified at preconditioning and on day 28 after allo-HSCT. Butyrate levels at preconditioning in the historical control group were significantly higher than those in the prebiotics group (P = .013). No difference was observed between the 2 groups on day 28 (P = .331), and butyrate levels were below the detection limit in many samples in both groups. (D) Changes in fecal butyrate concentration in the 2 groups. Butyrate concentration was maintained or increased (ie, the ratio of concentrations at preconditioning to those on day 28 was ≥0.8, as indicated by the red lines) in some patients. This was more frequently observed in the prebiotics group (6 of 30; 20%) than in the historical control group (3 of 68; 4.4%). (E) Patients in the prebiotics group were divided into 2 subgroups. Subgroup 1 (n = 14) included patients with SI before allo-HSCT of >2 and whose prebiotic intake was >50%. The other patients were included in subgroup 2 (n = 16). The cumulative incidence of grade 2 to 4 aGVHD on day 100 tended to be lower in subgroup 1 than in subgroup 2 (14.3% and 43.8%, respectively; P = .093).