We found that hypoglycemia at admission in sepsis patients was associated with high mortality in patients with normal BMI but not in those with low and high BMI. To the best of our knowledge, this is the first study to examine the association between hypoglycemia at admission and mortality in sepsis patients according to BMI categories.
Hypoglycemia is associated with a high mortality rate in sepsis patients and critically ill patients [5–7]. The NICE-SUGAR study demonstrated that hypoglycemia was associated with an increased risk of death, especially among patients with distributive shock [6]. Patients with both septic shock and hypoglycemia on the day of enrolment had a 2.5-fold higher mortality rate than those without septic shock and hypoglycemia in a multicenter observational study of sepsis patients in Japan [5].
The distribution of BMI in the population varies widely across countries and regions. A population-based study of adults was conducted in 200 countries, and it revealed that the proportions of high BMI (BMI ≥ 25), normal BMI, and low BMI (BMI < 18.5) were 50–60%, 30%, and < 5%, respectively, in Western countries, and 30%, 50%, and 10–20%, respectively, in East Asian countries [20]. Observational studies in sepsis patients of Western countries showed that the population with high BMI accounted for more than 50% and that of low BMI accounted for less than 10%, and sepsis patients with high BMI had better survival outcomes, suggesting an “obesity paradox” [8–10]. By contrast, a study conducted in China reported that low BMI was an independent factor associated with reduced 90-day survival [11]. In addition, a retrospective observational study using two cohorts in Japan (including cohort of the FORECAST study) showed that patients with low BMI had a significantly higher 28-day mortality than those without low BMI [21]. However, previous studies did not address the association between hypoglycemia and mortality in sepsis patients according to BMI categories.
The mechanism by which a low BMI and high BMI were not associated with worse mortality in patients with hypoglycemia was not clear. Adipose tissue and lipoproteins are related to BMI [22] as well as inflammation [23, 24]. The amount of adipose tissue and lipoproteins may depend on host defense in patients with sepsis and hypoglycemia. Further studies should be conducted to elucidate the mechanisms underlying the biological reactions of adipose tissue and lipoproteins to hypoglycemia in sepsis patients.
This study has some limitations. First, because these data were collected before the sepsis-3 criteria were published, the definition of sepsis used in this study was different from the latest definition. Second, the number of patients with hypoglycemia was too small to be assessed with sufficient statistical power. Third, all of our participants were Japanese; thus, generalizability might be limited to East Asians. Fourth, we could use few explanatory variables in the multivariate analysis because not many patients died.
In conclusion, hypoglycemia at admission in sepsis patients was associated with high mortality in patients with normal BMI, but not in those with low and high BMI. This association may be used as a prognostic marker in sepsis patients. In the future, larger cohort studies are needed to confirm this finding.