Association between Malnutrition and Inpatient Mortality in Hematopoietic Stem Cell Transplant (HSCT) Population: A Nationwide Inpatient Sample (NIS) Database Analysis

Introduction: Patients undergoing Hematopoietic Stem Cell Transplant (HSCT), particularly allogeneic HSCT, are at a high risk of malnutrition, which in turn, may contribute to other serious complications. We hypothesize that adult recipients of HSCT who have concomitant malnutrition, as compared to those without malnutrition, are at a higher risk of inpatient mortality and other complications. To test our hypothesis, we used a large national database and performed a multivariate analysis comparing outcomes of HSCT patients with versus without malnutrition.

Methods: The United States (Nationwide Inpatient Sample) NIS database was used to extract hospitalization data of adults admitted between 2019 to 2021. Patient data were identified using the International Classification of Diseases for oncology 10th version (ICD-10) codes for HSCT (autologous and allogeneic), malnutrition, mucositis, Graft Versus Host Disease (GVHD - for allogeneic HSCT patients), length of hospital stay (LOS), inpatient mortality, and infection in immunocompromised hosts. We compared the two groups of patients: those with versus without malnutrition. The primary endpoint was inpatient-related mortality. Secondary outcomes included LOS, and other HSCT complications. Several risk factors were analyzed to estimate likelihood of increased inpatient mortality. Risk factors included but are not limited to age, hypertension, complicated and uncomplicated diabetes, and heart failure. Multivariate logistic regression analysis was used to adjust for relevant variables.

Results: 1,116 (8.9%) patients out of a total of 12,541 patients with HSCT had a concomitant diagnosis of malnutrition. Patients with versus without malnutrition had no statistically significant difference in mean age (52 vs 50 years), sex, race, comorbid medical conditions, and other baseline characteristics. The likelihood of inpatient mortality was higher among patients with vs. without malnutrition (11.29% vs. 3.63%). Also, mean LOS was significantly higher among patients with versus without malnutrition (13.6 vs. 6.7 days). In multivariate regression analysis, malnutrition was the only comorbid conditions that is associated with a statistically significant increased likelihood of inpatient mortality (Odds Ratio (OR) 3.29, p-value <0.01). Age was the only other risk factor associated with increased likelihood of inpatient mortality. Malnutrition was also associated with increased hospital stay (OR 2.72, p <0.01) and higher likelihood of other HSCT complications in general (OR 1.66, p = 0.02), and specifically higher likelihood of sepsis (OR 1.67, p <0.01), acute kidney injury (OR 1.76, p < 0.01), and GVHD (OR 1.84, p <0.01).

Conclusion: This is one of the first large-scale studies that demonstrate an independent association between malnutrition and inpatient mortality as well as other HSCT complications. These findings highlight the importance of designing interventional nutritional trials in the HSCT population to reduce the risks associated with malnutrition.