Evaluation of Enteral Versus Parenteral Feeding As Nutritional Support In Allogeneic Hematopoietic Stem Cell Transplantation

Allo-HSCT procedure is associated with a frequent and potentially severe malnutrition which could highly participate to the transplant-related morbidity (TRM). Optimal nutritional management is still poorly known while both enteral nutrition (EN) and parenteral nutrition (PN) are effective. We proposed to retrospectively evaluate the impact of EN versus PN as nutritional support on early outcome of allo-HSCT.

We retrospectively analyzed all the successive patients who needed a nutritional support during their first allo-HSCT in our center from January 2009 to October 2010, excepting whose who had a progressive disease at time of transplant. Datas were compared in an intent to treat analysis according the EN or PN initial nutritional support strategy.

We analysed early outcome of 56 successive patients. Twenty of them received a myeloablative conditioning regimen and 36 a reduced intensity one. A total of 28 agreed to receive EN via a nasogastric feeding tube and the remaining 28 received PN. No significant difference in terms of age, diagnosis, disease status at transplant, conditioning regimens, stem cell source, GVHD nor antifungal secondary prophylaxis could be observed between the EN and PN groups. We found a lower median duration of fever in EN (2[0–8] vs. 5[0–17] (days); p=0.0026) and a lower need for antifungal therapy in EN group (7/28 vs. 17/28; p=0.0069), with a lower median duration of intravenous antifungal use (0 day [0–99] in EN vs. 7 days [0–93] in PN; p=0.00034) while incidence of bacteriemiae was not different. We observed a lower rate of replacement of central veinous catheter in EN group (3/28 in EN vs. 9/28 in PN; p=0.05) and a lower rate of transfer to ICU in the EN group (2/28 in EN vs. 8/28 in PN, p=0.036) but early death rate (<100 days) was the same in each group (4/28 vs. 4/28, p=NS). Median neutropenia and thrombopenia duration and median transfusion requirements were not significantly different. Fourteen patients in EN group and 18 in PN group presented a grade 3–4 oral mucositis (p=NS). Grade III-IV GVHD incidence was comparable in both groups (4/28 vs. 8/28; p=0.19).

Compared with PN, EN directly decreases the infectious risk, particularly the fungal risk, and its complications in allo-HSCT, without influencing hematopoietic toxicity nor GVHD incidence. Based on these encouraging results, we are now conducting a prospective, multicentric and randomized trial to confirm EN benefice in allo-HSCT.

No relevant conflicts of interest to declare.