Randomized Clinical Trial; Medical vs Bariatric Surgery for Adolescents (13-16 y) With Severe Obesity

A multicentre randomised clinical trial where three tertiary referral hospitals recruit patients.

Patients and their parents receive information about the study at their paediatric clinic and at the trial web page (www.amos2.se). Standardised procedure, including signed informed consent.

Interventions

At the end of day of baseline examination patients are randomised to either of two arms:

• Bariatric surgery with regular follow up
• Optimised conservative treatment starting with an 8-week Low Calorie Diet period followed by tailored support and treatment by the multidisciplinary team with an intensity of at least one visits a month over at least 2 years

Patients in the conservative arm will be reassessed regarding interest of undergoing bariatric surgery two years after inclusion.

Sample Size A sample size of 50 (25+25) leaves a power of more than 95% to evaluate a possible superiority of >10% weight difference between the two groups, assuming a 15% standard deviation in weight loss over follow-up at 2.5% significance level.

This number also allows assessment for differences in cardiovascular risk factors with sufficient power and acceptable power for demonstrating differences in quality of life and cognitive functions.

3.6 Randomization The computerized random allocation is performed during the day of baseline examination and patients were informed of their assignment. Forty-nine patients have been included until May 2017, and the last patient will be operated on June 14th. Stratification has been performed according to the recruitment centre

3.8 Statistical Methods & Additional Analyses Safety & Efficacy Outcomes: Analyses will be by intention to treat, including all randomised patients. Difference in treatment effect will be evaluated with hazard ratio between the treatment groups, and the corresponding confidence interval will be calculated. For secondary outcomes, the incidence of comorbidities during follow-up will be evaluated using time-to-event models, and for continuous variables mixed models will be used to evaluate differences between the treatment groups.

Incremental cost-effectiveness analysis will be conducted by calculating the incremental cost-effectiveness ratio (ICER; dividing the between-group difference in costs with the between-group difference in quality-adjusted life-years (QALYs) as well as life-years). Probabilistic sensitivity analysis will be conducted and results presented in an ICER scatterplot and cost-effectiveness acceptability curve.

Follow-Up Clinical Data Collection: Study point are 6 weeks, 1 year, 2 years, and 5 years after the start of intervention. Weight, quality of life, and adverse events, as well as clinical data regarding treatments for co-morbidities. Blood sampling for assessment of nutritional deficiencies, glucose control, blood lipids and inflammation will be collected at baseline and 1, 2 and 5 years.

Cognitive functions: Assessments will be performed at baseline and 1 and 2 years after start of intervention.

Register-Based Data Collection and Health-Economic Outcomes: Collection and analysis of Swedish national health care and other official registries

We decided in the Study Steering Committee to change time points for long term follow up to be 5, 10 and 15 years after treatment initiation to better comply with standard schedules for clinical follow up in registers (instead of 7, 12 and 17y after treatment initiation).

Inclusion criteria

• Age 13-16 years
• BMI >35
• Failed comprehensive treatment for obesity > 1 year
• Passing assessment of psychologist
• Tanner 3 or more

Exclusion criteria

• Monogenic obesity (for example Prader Willis, Laurence Moon-Bardet-Biedl)
• Obesity secondary to brain injury
• Severely mentally disabled
• Not eligible for general anesthesia
• Psychotic or other major psychiatric illness
• Previous major gastrointestinal surgery