Malnutrition in paediatric patients with leukaemia and lymphoma: a retrospective cohort study

Paediatric cancer is a potentially curable disease and its prognosis has been linked to several factors, such as nutritional status. The impact of malnutrition on these patients, either by overnutrition or undernutrition, varies and its relationship with outcomes is inconsistent. This study was conducted in order to determine the frequency of malnutrition in children with haematolymphoid malignancies at the time of diagnosis, as well as during treatment and to also investigate its relationship with the development of infections and death.

A retrospective cohort study of 191 children with a recent diagnosis of a haematolymphoid malignancy. The risks and nutritional classification were determined using anthropometry, follow-ups were conducted for up to 24 months and the presentation and frequency of infections and/or death were also recorded. Bivariate and multivariate analyses were conducted using binomial logistic regressions, for death and infection outcomes during follow-up. Survival analysis was conducted for various factors and types of cancer.

83.7% of children had a sufficient nutritional classification at diagnosis, 6.8% had malnutrition by undernutrition and 9.4% by overnutrition. 83.8% had at least one infectious complication during follow-up and 47.1% had ≥ 3. This percentage increased to 69.2% when configuring it in the malnutrition by undernutrition group. 18.3% of patients died. When configuring the mortality, the percentage was greater in patients with Acute Myeloid Leukaemia (AML) (57.1%) and malnutrition by undernutrition (30.7%). The multivariate analysis for the outcome of death, only showed a statistically significant variable (AML odds ratio = 26.52; confidence interval = 1.09-643.24; p = 0.04).

No statistically significant relationship was found between the nutritional status of children with haematolymphoid neoplasms, and outcomes such as infections or death. The differences in the results obtained in these investigations may be related to the varied nutritional status definitions and the ways of measuring them, thus limiting comparisons between them.