Endocrine Abstracts

Background: One in four children in England are now obese by school year 6. Childhood obesity is associated with significant co-morbidity including type 2 diabetes (T2DM), non-alcoholic fatty liver disease (NAFLD), obstructive sleep apnoea (OSA) and depression. NICE guidance suggests consideration of co-morbidity screening in children with body mass index (BMI) >98th centile but BMI does not accurately reflect adiposity in children owing to confounding effects of gender, height, ethnicity and puberty. We hypothesised that percentage excess weight (%EW) better predicts co-morbidity than BMI standard deviation score (SDS).

Method: We reviewed data from the last 100 patients seen in our tertiary paediatric weight management clinic. Ideal weight was determined by the weight on the same centile as the height centile. %EW was calculated by: actual weight minus ideal weight, divided by the ideal weight and multiplied by 100 (all in kilograms). Statistical analysis was performed using SPSS version 26.

Results: Patients were aged range 2-18 years, (65% were male). %EW was 11-181 (median 79). BMISDS was 1.84-7.75 (median 3.82). Weight-related co-morbidity was detected in 85% (85/100) patients: 55% (40/73) had insulin resistance (Homeostatic Model Assessment for Insulin Resistance (HOMA-R) >4), 6.3% (2/32) had T2DM on oral glucose tolerance test, 39.7% (31/78) had dyslipideamia, 42.6% (40/94) had NAFLD (raised alanine aminotransferase +/- ultrasound), 45% (14/31) had OSA, 25.6% (11/43) were hypertensive (systolic BP>90th centile) and 25% (25/100) had formal treatment for anxiety or depression. Forty percent of all patients had a co-morbidity diagnosed prior to referral (30.7% (8/26) of patients with >100%EW). %EW positively correlated to the number of co-morbidities (rs=0.23, P=0.02) and to HOMA-R (rs=0.34, P=0.004). BMISDS did not correlate with co-morbidities or HOMA-R (p 0.73 and 0.79 respectively). A co-morbidity was present in all children with ≥100%EW weight and 85% of those with >50%EW.

Conclusion: Unlike BMISDS, %EW correlates to the number of weight-related co-morbidities and HOMA-R. All patients with ≥100%EW had at least one weight-related co-morbidity, with no comparable BMISDS cut off. Co-morbidity was likely underrepresented due to high was-not-brought rates and limited availability of testing. We suggest %EW is a useful, accessible and understandable predictor of weight-related co-morbidity.