MON-096 Body Composition in Patients with Craniopharyngioma: Is DXA-Scanning Necessary for Evaluating the Metabolic Syndrome?

In craniopharyngioma patients, an important clinical feature is an adverse metabolic profile due to hypothalamic damage and pituitary deficiencies. In our Dutch/Swedish craniopharyngioma cohort, the occurrence of the metabolic syndrome (MetS) was higher than in the general population (46% vs.15% in the Swedish and 26% in Dutch population) (Wijnen 2017). However, in previous analysis we have used increased body mass index (BMI) to define adiposity. BMI misclassifies more than 50% of childhood cancer survivors as normal, while DXA measures increased body fat percentage (BF%) as BMI does not take an altered body composition into account (Blijdorp 2012). Our hypothesis was that MetS (defined according to the adjusted Joint Interim Statement) (Alberti 2009) and adiposity are underestimated in our previous study (Wijnen 2017). In the current retrospective cross-sectional study, craniopharyngioma patients (aged ≥ 18 yrs) were included if they had had at least one DXA-scan in the past. BMI, BF% and fat mass index (FMI) at the moment of first and, if available, last DXA-scan were gathered. These measurements or calculated standardized deviation scores (SDS) were used to evaluate obesity. The different definitions of obesity were applied as component in the definition of MetS; the laboratory results evaluating other components of MetS were gathered within 3 years of DXA-scan. We included 95 patients, 51% females and 49% with childhood-onset disease. Between 34 and 53 patients with craniopharyngioma fulfilled the criteria for the MetS (45-51% depending on the different assessment of adiposity), which was higher than the general population independent of the definition (P<0.05). Occurrence of MetS was higher if obesity was defined by DXA-measured BF% (52% vs. 42%, P=0.031) or FMI (51% vs. 43%, P=0.063) compared to BMI at last DXA-scan. A misclassification appeared in 8-9% (BF% vs. BMI) and 5-7% (FMI vs. BMI) for MetS, and respectively 25-29% and 9-13% for obesity. For MetS, an almost perfect agreement was found if increased BMI was compared with either increased BF% (Cohen’s Kappa 0.82-0.90, P<0.001) or FMI (Cohen’s Kappa 0.85-0.88, P<0.001). For obesity, agreement was only fair (Kappa 0.37, P=0.002 at last DXA-scan) to moderate (Kappa 0.53, P<0.001 at first DXA-scan) if increased BMI is compared with increased BF%. An Altman-Bland plot of BMI SDS and BF% SDS shows an upward trend with values outside limits of agreement at low and high averages. In conclusion, craniopharyngioma patients are at high risk for the MetS and obesity. DXA-scan measured BF% contributes to a more precise evaluation of obesity identification. Using BMI to estimate obesity may even underestimate the true prevalence of MetS. However, as craniopharyngioma patients have an extreme metabolic phenotype, the contribution of a better defined obesity component for identification and confirmation of MetS might be limited.