Pubertal Body Mass Index Change Is Associated With Adult Coronary Atherosclerosis and Acute Coronary Events in Men

Supplemental Digital Content is available in the text.


The BMI Epidemiology Study Gothenburg
The BMI Epidemiology Study Gothenburg (BEST Gothenburg) was initiated with the overall aim to study the impact of BMI during childhood and puberty on adult diseases. To that end we collected data on birthweight as well as directly measured height and weight throughout childhood from centrally archived School Health Care (SHC) records for all men born 1945 to 1961 in Gothenburg, Sweden. We also collected height and weight at young adult age from military conscription tests, kept by the Swedish Defense Recruitment Agency. Conscription was mandatory until 2010 for all Swedish men. The study cohort was linked to national disease registers using the Personal Identity Numbers (PIN) from the included participants.
Eligible individuals were those with a SHC record in the central archive and a ten-digit PIN.
Participants with data available for calculation of both childhood and young adult BMI were included in the present study (n=37,672).

The SCAPIS study
The Swedish CArdio Pulmonary bioImage Study (SCAPIS) cohort includes a randomly selected population sample representing Swedish adults aged 50 to 64 years, recruited from the census register (around 50% participation rate). The study participants underwent a thorough cardiovascular characterization during two days, including extensive imaging and functional studies of the heart, lungs, and metabolism. They also completed an extensive questionnaire regarding lifestyle and living conditions 16 . The SCAPIS study has been approved by the Ethical Review Board of Umeå, Sweden.

Exposures
Pre-pubertal childhood BMI at 8 years of age and young adult BMI at 20 years of age were calculated using all paired height and weight measurements in the period between 6.5 and 9.5 years of age for pre-pubertal childhood BMI, and in the period 17.5 to 22 years of age for young adult BMI. All paired measurements within these intervals were used to construct a linear regression model and the data for individual subjects were then corrected on this regression to obtain BMI at 8 and 20 years of age. After this correction, the BMIs were used to classify subjects as overweight and obese at 8 years of age (using the Centers for Disease Control and Prevention's cutoffs 24 at 8 years of age for overweight: BMI≥17.9 kg/m 2 , and for obesity: BMI≥20.0 kg/m 2 ) and at 20 years of age (based on BMI ≥25 or 30 kg/m 2 , respectively). Overweight refers to the population with a BMI above the overweight cutoff and includes both overweight and obese subjects at either 8 or 20 years of age, respectively.
Pubertal BMI change was defined as the difference between young adult BMI and childhood BMI, and young adult BMI thereby is the sum of childhood BMI and pubertal BMI change.

Measurement of Coronary Artery Calcification
Coronary artery calcification (CAC) was assessed using a state-of-the-art multi-slice computed tomography scanner (Siemens, Somatom Definition Flash, Siemens Medical Solution, Forchheim, Germany). Imaging and analyses were performed using a calcium scoring protocol according to the standardization suggested by McCollough et al. and Agatston et al. 17,25 and as previously described 26 .

Linkage to national registers
Linkage to registers held by the National Board of Health and Welfare and Statistics Sweden was performed with the participants' ten-digit PIN. Dates and diagnoses for the first appearance of a diagnosis of Myocardial Infarction (MI) as main diagnosis from the National Patient Register or Coronary Heart Disease (CHD) as underlying cause of death from the Cause of Death Register were retrieved (Table SII). The MI diagnosis has been shown to have a high validity in the National Patient Register 27 . Diagnoses were coded according to the International Classification of Diseases system.
We retrieved the study subjects´ highest achieved education level during the years 1990-2012, categorized as low (elementary school), medium (secondary school) or high (university) from the Longitudinal integration database for health insurance and labor market studies held by Statistics Sweden.

Representativeness of BEST-SCAPIS sub-cohort with BEST Gothenburg
In order to investigate if those included in the BEST-SCAPIS sub-cohort are different from those included in the complete BEST Gothenburg cohort, we evaluated childhood and young adult BMI. Neither young adult BMI nor childhood BMI differed significantly between the two groups (Table 1), demonstrating that the BEST Gothenburg-SCAPIS sub-cohort is representative of the complete BEST Gothenburg cohort. Moreover, we also compared height, weight and BMI from conscription for individuals included in the present study, with individuals not included in the present study who had a measurement from conscription available. For BMI, no significant difference was seen between the included and non-included individuals, and for height and weight, marginal differences (<0.5%) were seen (Table SI).

Statistical Analyses
Birthweight, childhood BMI, and young adult BMI were log-transformed when used in the Cox regression models. We evaluated a potential non-linear association by inclusion of a quadratic term and if significant, further evaluations were undertaken using a restricted cubic spline-approach in the Cox regression analysis for a flexible non-linear assessment of the hazard ratio (HR) in relation to pubertal BMI change 28 . Four knots placed at the pubertal BMI

Adjustment for birthweight, country of birth, and education level
Adjustment for birthweight (n=35,662) did not alter the observed results for the overweight groups compared with men who were never overweight, and the risk of adult acute coronary events (Table SIII). We categorized education level in three categories (low (elementary school), medium, or high (university level)) and used it as a proxy for socioeconomic status.
Adjustment for the individuals´ highest achieved education level did not alter the association between childhood and young adult overweight and acute coronary events (Table SIV). The association between overweight groups and the risk of adult acute coronary events was also evaluated for the subgroup with Sweden as country of birth for the subject and both parents (n=31,407). These analyses showed similar results as in the entire cohort (Table SV).

BMI change
In order to allow validation and replication of experiments, all essential research materials listed in the Methods should be included in the Major Resources Table below. Authors are encouraged to use public repositories for protocols, data, code, and other materials and provide persistent identifiers and/or links to repositories when available. Authors may add or delete rows as needed.

Description
Source / Repository Persistent ID / URL The data that support the findings of this study are available from the corresponding author upon reasonable request and upon approval from the University of Gothenburg according to mandatory national law, but are not publicly available due to privacy and ethical restrictions.

Other Description
Source / Repository Persistent ID / URL NA