[Malnutrition in pediatric oncology: prevalence and screening]

Algorithms for nutritional pediatric support have been proposed in a French national nutritional framework program. However, they are not specific for oncology. With the pediatric nutritional risk score (PNRS) all children with cancer have a high risk of malnutrition, but a systematic nutritional support is not possible for all of them.

Estimation of malnutrition prevalence and identification of predictive factors of major weight loss during treatment defined by a weight loss more than 5% within 1 month, 7.5% within 3 months, 10% within 6 months.

This historical study included children registered with a solid tumor in 2002 in an oncology pediatric unit. Data collected at diagnosis were weight, height, PNRS, the Lansky functional score, tumor type. Furthermore weight, height, and major weight loss were collected at each cure of chemotherapy and during evolution. Malnutrition at diagnosis was defined using the weight for height ratio. Relations between major weight loss and risks factors were estimated using logistic regression.

Seventy children were included, 16 (22.9%) were malnourished at admission. During chemotherapy, 29 (41.4%) children experienced a major weight loss. Odds ratio of those who were malnourished at diagnosis was not significantly higher in comparison to well-nourished children. Children with a high risk of malnutrition are those affected by Ewing tumor, B lymphom, head and neck localisations, osteosarcomas, metastatic cancers, or cancers treated by high dose chemotherapy with stem cell rescue. For these 29 (41.4%) children the major weight loss odds ratio was 5.9 [IC95% 2.0-16.7].

Taking into account others factors with items of PNRS allows to screen children with an higher risk of a major weight loss during treatment and to enhance nutritional care plan for them.