Sugar-rich foods consumed during broad-spectrum antibiotic treatment were associated with greater gut microbiome disruption among patients undergoing allogeneic hematopoietic cell transplantation. The association included lower microbial alpha-diversity and greater Enterococcus expansion, while experiments in mice provided additional evidence that sucrose could exacerbate antibiotic-induced enterococcal expansion.
Researchers conducted a prospective observational study of 173 adult patients undergoing allogeneic hematopoietic cell transplantation at Memorial Sloan Kettering Cancer Center between 2017 and 2022. Dietary intake was recorded during hospitalization, encompassing 9,419 meals, and longitudinal stool samples were collected for microbiome analysis. The primary Bayesian analysis included 1,009 samples from 158 patients. All 173 patients received at least 1 antibiotic during dietary monitoring, and 138 (80%) transitioned from prophylactic therapy to broad-spectrum antibiotics, most commonly in the setting of neutropenic fever or bloodstream infection.
The researchers examined whether dietary intake during the 2 days prior to stool collection was linked to microbiome alpha-diversity and whether those associations differed according to broad-spectrum antibiotic exposure. Secondary analyses assessed microbiome composition, including Enterococcus abundance, and mortality. Bayesian models accounted for factors including conditioning intensity, nutrition support, antibiotic exposure, repeated measurements within patients, and time relative to transplantation.
Broad-spectrum antibiotic exposure during the preceding 2 days was associated with lower microbiome alpha-diversity. Sweets intake alone was not associated with alpha-diversity, but higher sweets intake during broad-spectrum antibiotic exposure was associated with a greater decrease. For every additional 100 g of sweets consumed, measured by dehydrated food weight, the model predicted an additional 24% decrease in mean alpha-diversity compared with samples without preceding broad-spectrum antibiotic exposure.
An analysis based on macronutrients yielded a similar pattern. Each additional 100 g of sugar consumed during broad-spectrum antibiotic exposure was associated with a further 21% decrease in mean alpha-diversity. In unadjusted analyses, sweets intake was inversely linked to alpha-diversity among samples collected following broad-spectrum antibiotic exposure but not among samples without such exposure. More detailed dietary analyses implicated formulated nutritional beverages, including sports drinks and nutritional shakes, as well as added rather than naturally occurring sugars.
The association was also reflected in microbiome composition. Enterococcus showed the strongest inverse relationship with alpha-diversity among the genera examined, and higher sweets intake following broad-spectrum antibiotic exposure was linked to increased Enterococcus faecium abundance. Enterococcus is an important cause of antibiotic-resistant bloodstream infection after HCT and has been associated with adverse transplantation outcomes. The sweets-antibiotic association with lower alpha-diversity persisted in analyses accounting for factors including underlying malignant disease, comorbidities, intensive analgesic exposure, nutrition support, treatment of confirmed infections, and early transplantation samples collected prior to severe mucositis.
The researchers also examined mortality according to sugar intake. Among patients with above-median sugar intake, each additional day of broad-spectrum antibiotic exposure was associated with a 12% higher hazard of mortality; the corresponding association was not statistically significant among patients with below-median sugar intake. A dietary-pattern analysis showed a similar relationship among patients characterized by lower overall caloric intake and a greater proportion of dietary sugar. However, the mortality association could partly reflect reduced overall food intake rather than sugar consumption itself,
The researchers then tested the diet-antibiotic interaction experimentally in mice. A single dose of the carbapenem antibiotic biapenem induced an expansion of endogenous enterococci that peaked on day 3 and largely resolved by day 6. Sucrose supplementation increased enterococcal expansion 16.3-fold on day 3 and 33.4-fold on day 6 compared with antibiotic exposure without sucrose. Sucrose did not increase enterococcal burden in mice that did not receive antibiotics. The effect also persisted under fiber-free dietary conditions, indicating that the observed effect was not explained solely by reduced fiber intake.
Several limitations temper the findings. The human analysis was observational and conducted at a single center in patients undergoing allogeneic hematopoietic cell transplantation. Because prophylactic fluoroquinolone use was nearly universal, the observed interaction may more specifically reflect dietary intake during escalation to broad-spectrum antibiotics rather than exposure to antibiotics generally. Broad-spectrum therapy was initiated for clinical indications such as fever or suspected infection, raising the possibility of confounding by illness severity. Food consumed outside the hospital was incompletely captured, and treatment-related symptoms could have influenced both dietary preferences and other time-varying factors.
Taken together, the human and mouse findings suggest that simple sugars may amplify antibiotic-associated microbiome disruption, but the researchers emphasized that prospective studies are needed to determine whether reducing sugar intake during antibiotic treatment can preserve the microbiome or improve clinical outcomes. “Limiting sugar intake during antibiotic therapy may help mitigate microbiome injury,” wrote lead author Anqi Dai, of Memorial Sloan Kettering Cancer Center, and colleagues.
The study was supported by National Institutes of Health grants and institutional and philanthropic funding. Several authors reported consulting, advisory, equity, research funding, intellectual property, honoraria, or other relationships with pharmaceutical and biotechnology companies. The remaining authors reported no competing interests.
Source: Nature