Obesity in Survivors of Childhood Hematopoietic Stem Cell Transplantation.

Background: Obesity is an increasing problem in the United States, and a leading cause of morbidity and mortality. In recent years, survivors of certain malignancies such as childhood acute lymphoblastic leukemia (ALL) have demonstrated an increased incidence of obesity than their corresponding age group in the general population. We retrospectively evaluated the pediatric hematopoietic stem cell transplant (HSCT) population at our institution for the incidence of overweight/obese patients, as well as any factors that might be associated.

Patients and Methods: We conducted a retrospective analysis of medical records of pediatric HSCT patients treated between 1990 and 2005 at the Mattel Children’s Hospital at UCLA. Data was collected on 297 patients including body mass index (BMI), diagnosis, treatment, post-transplant disease and infection, survival. We excluded patients from our analysis that had observations from only one time point or insufficient follow-up BMI data for analysis. The remaining cohort consisted of 195 patients whose demographic and clinical data are presented below. Age, height and weight were measured at baseline and approximately at 100 day and 1year time intervals until 5 years post-transplant. The average age was 8.4 years (95% CI: 7.6, 9.3), about two-thirds of the patients were male and the predominant ethnicities were Caucasian (42.8%) and Hispanic (32.5%). The majority of patients had ALL as a diagnosis (43.6%), 1 or 2 remissions (90.4%), no prior transplant history (83.6%), a related donor (65.4%), a bone marrow stem cell source (77.3%), no antigen mismatch (80.5%), and 0 or 1 hospitalization during the 5 year post-transplant period (69.5%). BMI (measured in kg/m²) and BMI percentiles (BMI%) were calculated for all time points. Calculation of BMI% relied on the WHO growth chart for patients 0–5 yrs old, and the CDC chart for patients greater than 5 yrs old. We constructed the following BMI-related variables for all time points:

1. overweight = BMI% ≥95^th percentile for patients ≤20 yrs or BMI ≥30 kg/m²

2. overweight = BMI% ≥95^th percentile or BMI ≥30 kg/m² for patients over 20 yrs),

3. underweight = BMI% ≤5^th percentile for patients ≤20 yrs or BMI ≤18.5 kg/m² for patients over 20 yrs,

We focused on baseline and last time point measures for these variables and also determined if a patient was ever obese, ever overweight, or ever underweight at any point during the 5-year time period. We also computed the change in BMI between time points and from baseline, and categorized the patients as either having a decrease in BMI>1.5 kg/m², change in BMI ≤1.5, or increase in BMI>1.5 kg/m²

Results: At baseline and during a 5-year follow up period, our HSCT patients became substantially more overweight than the general pediatric population with initially 17.4% of patients exceeding the 95^th percentile at baseline and 34% ultimately exceeding this percentile post transplant. Length of hospital stay was moderately associated with obesity (P-value = 0.037), with a longer stay on average having a stronger association with obesity than shorter stays (where a longer stay was on average 49±30 days in comparison to 40±22 days for the shorter stay). A diagnosis of ALL and graft vs. host disease were moderately associated with whether a patient became overweight during the study (Fisher’s exact test P-values = 0.036 and 0.075 respectively). Patients who experienced stability in their BMI had significantly higher survival rates than patients whose BMI significantly changed (71.2% vs 44.9%; log rank P-value = 0.01). We also found that obesity at the last time point was significantly associated with decreased survival (log rank P-value = 0.004), with patients who were obese at the last time point having a 2-year survival rate of 38% (95% CI: 23.9, 60.4) and patients who were not obese having a 2-year survival rate of 64.2% (95% CI: 56.3, 73.3). While demographic, treatment and illness, infection, and other BMI related variables were considered, they were not significant.

Conclusion: An increased incidence of overweight and obese patients where observed in our pediatric post-transplant population, nearly doubling from baseline. Additional studies into interventions are warranted.