ABSTRACT
OBJECTIVE: The first-line approach for childhood obesity is lifestyle intervention (LI); however, success varies. This study aimed first to identify distinct subgroups of response in children living with overweight and obesity and second to elucidate predictors for subclusters. METHODS: Based on the obesity patient follow-up registry the APV (Adipositas-Patienten-Verlaufsdokumentation) initiative, a total of 12,453 children and adolescents (median age: 11.5 [IQR: 9.7-13.2] years; BMI z score [BMIz]: 2.06 [IQR: 1.79-2.34]; 52.6% girls) living with overweight/obesity and participating in outpatient LI were studied. Longitudinal k-means clustering was used to identify individual BMIz response curve for up to 2 years after treatment initiation. Multinomial logistic regression was used to elucidate predictors for cluster membership. RESULTS: A total of 36.3% of children and adolescents experienced "no BMIz loss." The largest subcluster (44.8%) achieved "moderate BMIz loss," with an average delta-BMIz of -0.23 (IQR: -0.33 to -0.14) at study end. A total of 18.9% had a "pronounced BMIz loss" up to -0.61 (IQR: -0.76 to -0.49). Younger age and lower BMIz at LI initiation, larger initial BMIz loss, and less social deprivation were linked with higher likelihood for moderate or pronounced BMIz loss compared with the no BMIz loss cluster (all p < 0.05). CONCLUSIONS: These results support the importance of patient-tailored intervention and earlier treatment escalation in high-risk individuals who have little chance of success.
Subject(s)
Overweight , Pediatric Obesity , Female , Adolescent , Humans , Child , Male , Overweight/therapy , Pediatric Obesity/therapy , Body Mass Index , Outpatients , AdiposityABSTRACT
OBJECTIVES: Associations between body mass index (BMI)- standard deviation score (SDS)/waist-to-height ratio (WHtR) were studied with (i) serum uric acid (sUA)/gamma-glutamyl-transferase (GGT) and (ii) cardiometabolic risk markers in children with obesity, considering sex, pubertal development, and degree of weight loss/type of patient care. METHODS: 102 936 children from the Adiposity-Follow-up registry (APV; 47% boys) were included. Associations were analysed between sUA/GGT and anthropometrics, transaminases, lipids, fasting insulin (FI), homeostasis model assessment of insulin resistance (HOMA-IR), triglycerides to HDL-cholesterol (TG/HDL)-ratio. Follow-up analyses (3-24 months after baseline) considered a BMI-SDS reduction ≥0.2 (n = 11 096) or ≥0.5 (n = 3728). Partialized correlation analyses for sex and BMI-SDS were performed, taking pubertal development into consideration. RESULTS: At baseline, BMI-SDS showed the strongest correlations to sUA (r = 0.35; n = 26 529), HOMA-IR/FI (r = 0.30; n = 5513 /n = 5880), TG/HDL-ratio (r = 0.23; n = 24 501), and WHtR to sUA (r = 0.32; n = 10 805), GGT (r = 0.34; n = 11 862) and Alanine-aminotransferase (ALAT) (r = 0.33; n = 11 821), with stronger correlations in boys (WHtR and GGT: r = 0.36, n = 5793) and prepubertal children (r = 0.36; n = 2216). GGT and sUA (after partializing effects of age, sex, BMI-SDS) showed a correlation to TG/HDL-ratio (r = 0.27; n = 24 501). Following a BMI-SDS reduction ≥0.2 or ≥0.5, GGT was most strongly related to Aspartate-aminotransferase (ASAT)/ ALAT, most evident in prepuberty and with increasing weight loss, and also to TG/HDL-ratio (r = 0.22; n = 1528). Prepubertal children showed strongest correlations between BMI-SDS/WHtR and GGT. ΔBMI-SDS was strongly correlated to ΔsUA (r = 0.30; n = 4160) and ΔGGT (r = 0.28; n = 3562), and ΔWHtR to ΔGGT (r = 0.28; n = 3562) (all p < 0.0001). CONCLUSION: Abdominal obesity may trigger hyperuricemia and hepatic involvement already in prepuberty. This may be stronger in infancy than anticipated to date. Even moderate weight loss has favourable effects on cardiometabolic risk profile and glucose homeostasis.