Evaluation of body composition changes in children after hematopoietic stem cell transplantation using the deuterium dilution method and double-energy X-ray absorptiometry: results from a pilot study

Abstract

A comprehensive investigation of nutritional status of children with oncological diseases using reference methods of body composition analysis is of interest for optimizing nutritional support during and after cancer treatment. In this paper, we report the first clinical use of the deuterium dilution method for body composition assessment in children in the early period after hematopoietic stem cell transplantation (HSCT) in Russia. Our study is carried out as a part of the IAEA project on applying nuclear methods for the evaluation of nutritional status in childhood cancer. It was approved by the Independent Ethics Committee and the Scientific Council of the Dmitry Rogachev National Medical Research Center of Pediatric Hematology, Oncology and Immunology. From February 2021 to April 2023, the study enrolled 39 patients aged 5 to 17 years, 4 of whom were subsequently excluded. Out of the remaining 35 patients (25 boys and 10 girls with the median age of 9.7 years), 25 were diagnosed with acute lymphoblastic leukemia, 4 with acute myeloid leukemia, 3 with aplastic anemia, 2 with acute bilineal leukemia, and 1 with lymphoma. The children were assessed at the following time points: between day –30 and day –10 of HSCT (time point 1); between day –2 and day +2 of HSCT (time point 2); between day +30 and day +45 after HSCT (time point 3), between day +100 and day +125 after HSCT (time point 4), and between day +280 and day +380 after HSCT (time point 5). Two patients were assessed at all 5 time points, 21 patients – at 4 time points, 11 patients – at 3 time points, and 1 patient – at two time points. All the patients required parenteral nutrition between time points 2 and 3 (n = 35). Out of these, 7 patients also received parenteral nutrition support from time point 3 to 4. All the patients were given enteral nutrition support starting from time point 3. Evaluations at each time point included clinical status assessment, laboratory testing, a bioimpedance analysis (BIA), anthropometric measurements, the assessment of oral intake as well as the amount and quality of nutritional support. Measurements using the deuterium dilution method (DDM) and a dual-energy X-ray absorptiometry (DXA) were performed only at time points 1, 4, and 5. The DDM was applied in 8 out of 35 patients. Body composition was measured using all three methods (DDM, DXA, and BIA) at least at one time point in 7 patients and at two adjacent time points in 3 patients. The results of body composition assessment using the DDM, DXA and BIA were compared. The statistical significance of individual changes and differences between paired measurements of body composition were evaluated using the Wilcoxon signed-rank test for dependent data with a threshold significance level of p = 0.05. Before HSCT, 17 out of 35 patients were overweight or obese according to the body mass index (BMI) thresholds set by the World Health Organization, and only 2 out of 35 patients were underweight, which corresponds to the common understanding of nutritional status in children before HSCT. Nevertheless, according to BIA findings, 11 out of 18 patients with normal or reduced BMI had an increased or high body fat percentage (%BF), which may be an evidence of a catabolic state. Between time points 1 and 4, there was a sharp decrease in mean body weight and BMI due to a reduction in fat-free mass (FFM) and, to a lesser extent, in fat mass (FM), indicating the progression of catabolic changes. A comparison of body composition estimates obtained by the DDM, DXA, and BIA revealed the smallest differences between the DDM and BIA measurements (–1.0 kg difference for FFM). However, the estimates of FFM obtained by DXA were significantly lower than those obtained by the DDM and BIA (by2.3 kg and 3.3 kg, respectively), while FM and %BF estimated by DXA were significantly higher than those estimated by the DDM and BIA (by 2.6 kg and 3.8 kg, and by 7.8% and 10.9%, respectively) which could be explained by an increase in FFM hydration. Our preliminary data demonstrate good agreement between the DDM, DXA and BIA in assessing FFM, FM and %BF changes at a group level, but not at an individual level. A larger sample size is needed to clarify the obtained results and to compare the clinical significance of these methods of body composition assessment in children with cancer during and after HSCT.