Physical fitness in male adolescents and atherosclerosis in middle age : a population-based cohort study

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY-NC 4.0 https://creativecommons.org/licenses/by-nc/4.0/ Physical fitness in male adolescents and atherosclerosis in middle age : a populationbased cohort study © Author(s) (or their employer(s)) 2024 Published version Herraiz-Adillo, Ángel; Ahlqvist, Viktor H; Higueras-Fresnillo, Sara; Hedman, Kristofer; Hagström, Emil; Fortuin-de Smidt, Melony; Daka, Bledar; Lenander, Cecilia; Berglind, Daniel; Östgren, Carl Johan; Rådholm, Karin; Ortega, Francisco B; Henriksson, Pontus Herraiz-Adillo, Á., Ahlqvist, V. H., Higueras-Fresnillo, S., Hedman, K., Hagström, E., Fortuin-de Smidt, M., Daka, B., Lenander, C., Berglind, D., Östgren, C. J., Rådholm, K., Ortega, F. B., & Henriksson, P. (2024). Physical fitness in male adolescents and atherosclerosis in middle age : a population-based cohort study. British Journal of Sports Medicine, Early online, Article 107663. https://doi.org/10.1136/bjsports-2023-107663 2024 1 HerraizAdilloÁ, etal. Br J Sports Med 2024;0:1–11. doi:10.1136/bjsports-2023-107663 Physical fitness in male adolescents and atherosclerosis in middle age: a populationbased cohortstudy Ángel HerraizAdillo ,1 Viktor H Ahlqvist ,2 Sara HiguerasFresnillo ,1,3 Kristofer Hedman ,4 Emil Hagström,5 Melony Fortuinde Smidt,6 Bledar Daka,7 Cecilia Lenander,8 Daniel Berglind ,2,9 Carl Johan Östgren,1,10 Karin Rådholm ,1,11 Francisco B Ortega ,12,13 Pontus Henriksson 1 Original research To cite: HerraizAdilloÁ, AhlqvistVH, HiguerasFresnilloS, etal. Br J Sports Med Epub ahead of print: [please include Day Month Year]. doi:10.1136/ bjsports-2023-107663 ►Additional supplemental material is published online only. To view, please visit the journal online (http:// dx. doi. org/ 10. 1136/ bjsports2023107663). For numbered affiliations see end of article. Correspondence to Dr Pontus Henriksson, Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Östergötland, Sweden; pontus. henriksson@ liu. se Accepted 11 January 2024 © Author(s) (or their employer(s)) 2024. Reuse permitted under CC BYNC. No commercial reuse. See rights and permissions. Published by BMJ. ABSTRACT Objectives To examine the associations between physical fitness in male adolescents and coronary and carotid atherosclerosis in middle age. Methods This populationbased cohort study linked physical fitness data from the Swedish Military Conscription Register during adolescence to atherosclerosis data from the Swedish CArdioPulmonary bioImage Study in middle age. Cardiorespiratory fitness was assessed using a maximal cycleergometer test, and knee extension muscular strength was evaluated through an isometric dynamometer. Coronary atherosclerosis was evaluated via Coronary Computed Tomography Angiography (CCTA) stenosis and Coronary Artery Calcium (CAC) scores, while carotid plaques were evaluated by ultrasound. The associations were analysed using multinomial logistic regression, adjusted (marginal) prevalences and restricted cubic splines. Results The analysis included 8986 male adolescents (mean age 18.3 years) with a mean followup of 38.2 years. Physical fitness showed a reversed Jshaped association with CCTA stenosis and CAC, but no consistent association was observed for carotid plaques. After adjustments, compared with adolescents in the lowest tertile of cardiorespiratory fitness and muscular strength, those in the highest tertile had 22% (OR 0.78; 95% CI 0.61 to 0.99) and 26% (OR 0.74; 95% CI 0.58 to 0.93) lower ORs for severe (≥50%) coronary stenosis, respectively. The highest physical fitness group (high cardiorespiratory fitness and muscular strength) had 33% (OR 0.67; 95% CI 0.52 to 0.87) lower OR for severe coronary stenosis compared with those with the lowest physical fitness. Conclusion This study supports that a combination of high cardiorespiratory fitness and high muscular strength in adolescence is associated with lower coronary atherosclerosis, particularly severe coronary stenosis, almost 40 years later. INTRODUCTION Despite positive trends in the Western world during recent decades,1 2 cardiovascular disease (CVD) remains as the leading cause of mortality worldwide.3 Atherosclerosis, an inflammatory condition affecting all arterial regions, is the principal pathway involved in CVD.3 Subclinical atherosclerosis, characterised by the presence of plaques in the arterial walls, is an early marker of CVD and an important predictor of future cardiovascular events.4 Thus, identification of early modifiable risk factors is crucial for effective prevention of CVD and mortality globally. A high level of physical fitness, including cardiorespiratory fitness and muscular strength, is considered a crucial factor in preventing CVD, cardiovascular mortality and allcause mortality. Thus, the American Heart Association recognises cardiorespiratory fitness as a vital clinical sign5 due to its strong association with positive cardiovascular WHAT IS ALREADY KNOWN ON THIS TOPIC ⇒Higher physical fitness levels, including both cardiorespiratory and muscular fitness, are associated with lower cardiovascular diseaserelated nonfatal and fatal events in adults. This association has also been observed for fitness during adolescence and later cardiovascular disease incidence and mortality. ⇒No previous study has examined physical fitness in adolescence in relation to the development of coronary atherosclerosis in middle age, which may link fitness and the risk of cardiovascular events. WHAT THIS STUDY ADDS ⇒Our study provides novel evidence supporting that the combination of high cardiorespiratory fitness and high muscular strength in adolescence is associated with lower coronary atherosclerosis, particularly severe coronary stenosis, almost 40 years later. ⇒These results suggest that coronary atherosclerosis is likely one of the mechanisms underlying the association between physical fitness and cardiovascular disease morbidity and mortality. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY ⇒Our results support the clinical value of assessing both cardiorespiratory and muscular fitness for cardiovascular risk stratification. ⇒Longterm interventions able to improve both cardiorespiratory fitness and muscular strength in adolescents could contribute to prevention of atherosclerosis in adulthood. Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from 2HerraizAdilloÁ, etal. Br J Sports Med 2024;0:1–11. doi:10.1136/bjsports-2023-107663 Original research outcomes, including improved cardiac structure and function, reduced atherosclerosis and decreased risk of CVD and allcause mortality.6–10 Additionally, while the associations are less pronounced compared with cardiorespiratory fitness, increased muscular strength also exhibits beneficial effects, including lower prevalence of atherosclerosis, decreased risk of CVD and lower mortality.11 12 Nevertheless, despite a recent emphasis on prevention of CVD in younger individuals,13 there is a lack of evidence on the impact of cardiorespiratory fitness in adolescence on the longterm development of atherosclerosis during late middle age. Such evidence could help to elucidate whether physical fitness early in life is related to atherosclerosis development decades later which may be of paramount importance for primary prevention of CVD. However, only one previous study has investigated the associations between physical fitness in adolescence and carotid atherosclerosis in middle age.14 Furthermore, to the best of our knowledge, no previous study has examined associations of physical fitness in adolescence with coronary atherosclerosis later in life. In our study, Coronary Computed Tomography Angiography (CCTA), an accurate noninvasive imaging technique, enables a comprehensive assessment of the atherosclerotic burden, since CCTA allows the characterisation and quantification of both calcified and noncalcified plaques in the coronary arteries.15 16 Therefore, the aim of this study was to examine the association between physical fitness in male adolescents with coronary and carotid atherosclerosis in middle age, using a populationbased sample and notably long followup. METHODS Study design and population This cohort study linked information on atherosclerosis in middle age using data from the Swedish CArdioPulmonary bioImage Study (SCAPIS) (n=14 646) to information on cardiorespiratory fitness and muscular strength in male adolescents, obtained from the Swedish Military Conscription Register. Linkage of both databases, which determined our sample size (n=10 802),17 was conducted through a personal identification number assigned to all Swedish residents at birth or on immigration. In our study, the Swedish Military Conscription Register comprised male adolescents born in Sweden between 1953 and 1968 who performed conscription between 1972 and 1987 (at ≈18 years of age). During this period, conscription was mandatory by law, except in rare circumstances, and the Swedish Military Conscription Register therefore includes 82%–92% of all Swedish men at the time of conscription.18 SCAPIS is a collaborative project comprising six different universities in Sweden (Gothenburg, Linköping, Malmö/Lund, Stockholm, Umeå and Uppsala) aiming to predict and prevent cardiovascular and pulmonary disease. The participants included in SCAPIS were between 50 and 64 years old. Details about the SCAPIS protocol have been published elsewhere.19 In this study, the inclusion criteria were: (1) men <20 years old at conscription with available data on cardiorespiratory fitness, muscular strength and covariates (age, site, body mass index (BMI), duration of smoking and conscription year) and (2) available data on coronary or carotid atherosclerosis and covariates (age, site and educational status) in SCAPIS. Online supplemental figure 1 depicts a flow chart for the study. In brief, of the 14 646 male participants included in SCAPIS, 8986 male adolescents had data on exposures, covariates and at least one of the atherosclerosis outcomes. Thus, the final sample sizes consisted of 8006, 7849 and 8934 participants for the analysis of coronary stenosis, Coronary Artery Calcium (CAC) score and carotid plaques, respectively. Exposures at conscription Details about cardiorespiratory fitness and muscular strength protocols have been published elsewhere.20–22 Briefly, cardiorespiratory fitness was assessed with a maximal exercise test using an electrically braked cycleergometer test, provided participants had a normal ECG at rest. The conscription protocol commenced with a 5min warmup, during which the workload was determined based on the individual’s weight. Subsequently, the workload was stepwise increased by 25 W every minute until exhaustion or incapacity to maintain the intended pedal cadence (60–70 revolutions/min). Cardiorespiratory fitness was defined as the maximal work rate achieved (in W).23 Three different measures were considered for muscular strength: knee extension, handgrip and elbow flexion strength (in N). Knee extension strength was considered as the main exposure since previous studies have suggested it to be the most powerful indicator of healthrelated muscular strength in the Swedish Military Conscription Register.22 Strength variables were measured with an isometric dynamometer test performed at maximal contraction capacity. Knee extension and elbow flexion strength were evaluated in a sitting position with 90° flexion over the knee and elbow joint, respectively, while handgrip strength was measured by positioning the hand vertically, with 90° flexion over the elbow joint. Atherosclerosis outcomes at SCAPIS Coronary atherosclerosis The detailed imaging protocol for SCAPIS has been published elsewhere.19 Participants with a technical failure in any of the four proximal segments on the CCTA images were excluded for the analysis of coronary plaques and CAC score.19 24 Coronary plaques were studied through two different levels of characterisation: grade of lumen stenosis and composition of the plaques from an arterial tree level. In our study, regarding the grade of lumen stenosis, the participants were finally categorised considering the segment with the greatest amount of stenosis within the 11 clinically most relevant segments (1–3, 5–7, 9, 11–13, 17)25 as follows: no stenosis, 1%–49% stenosis and severe (≥50%) stenosis.19 24 The presence of a ‘calcium blooming’ artefact and stent were considered as 1%–49% stenosis and ≥50% stenosis, respectively. A segment involvement score was calculated as the total number of relevant coronary segments with atherosclerosis irrespective of the degree of stenosis (range 0–11).26 Regarding composition of the plaques from an arterial tree level, coronary atherosclerosis was further characterised as: no plaque, only noncalcified plaque/s (all identified plaque/s are noncalcified), only calcified plaque/s (all identified plaque/s are calcified) and mixed composition (presence of both calcified and noncalcified segments in the arterial tree). In addition to CCTA images, a total CAC score was obtained according to an international standard protocol27 by adding the calcium content in each coronary artery,28 29 and the total CAC score was divided into three categories commonly used in clinical practice as follows: 0, 1–99 and ≥100 Agatston units. Subjects with implanted stent or post coronary artery bypass grafting were not evaluated for CAC. Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from 3 HerraizAdilloÁ, etal. Br J Sports Med 2024;0:1–11. doi:10.1136/bjsports-2023-107663 Original research Carotid atherosclerosis Carotid artery two‐dimensional grey scale images were examined using a standardised protocol with a Siemens Acuson S2000 ultrasound scanner equipped with a 9L4 linear transducer (Siemens, Forchheim, Germany) and interpreted by regularly trained operators.19 Carotid plaque was defined in accordance with the Mannheim consensus.30 Common carotid, bulb and internal carotid arteries were examined, and participants without valid readings in both right and left carotid arteries were excluded for the analysis. Participants were classified as having either no plaque, unilateral plaque/s or bilateral carotid plaques.31 For splines analysis, a carotid plaque score was calculated as follows: no plaque=0, unilateral plaque/s=1 and bilateral plaques=2. Covariates BMI at conscription was calculated as weight (kg)/height squared (m2) obtained by standardised procedures. Years of smoking at conscription were calculated based on selfreported age of smoking initiation in SCAPIS (for those who reported previous or ongoing smoking). To account for temporal trend differences at conscription, the year of conscription (ranging from 1972 to 1987) was categorised into four distinct periods, each spanning 4 years. Educational status at SCAPIS was categorised as unfinished primary school, primary school, secondary school and university degree. Statistical analysis We performed a complete case analysis excluding participants without complete data on exposures (0.7% in muscular strength and 13.2% in cardiorespiratory fitness), outcomes (0.8% in carotid plaque, 4.3% in CAC and 6.1% in coronary stenosis) and any covariates (14.3%). In total, 16.8% of participants had missing values in any exposure or covariate. Three types of analyses were performed. First, the adjusted nonlinear associations between quantitative exposures and atherosclerosis outcomes (summarised as scores) were evaluated trough linear regression models incorporating restricted cubic splines with four knots located at percentiles 5th, 35th, 65th and 95th.32 33 Second, the associations between tertiles of physical fitness in adolescence and atherosclerosis outcomes (CCTA coronary stenosis, CAC score and carotid plaque) in middle age were examined through multinomial logistic regression models and adjusted (obtained by marginalisation/parametric gformula) prevalences.34 Third, we performed restricted cubic splines (four knots located at percentiles 5th, 35th, 65th and 95th) within multinomial logistic regression models. The analyses had increasing level of covariate control: (1) unadjusted model; (2) adjusted model (by age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription and years of smoking at conscription). We created a directed acyclic graph to illustrate the hypothesised associations of physical fitness with atherosclerosis (online supplemental figure 2). Adjusted models were further adjusted for knee extension strength in cardiorespiratory fitness and for cardiorespiratory fitness in muscular strength outcomes. Adjusted models were selected as the main analysis in splines and multinomial regressions for a better understanding of the isolated contribution of exposures on atherosclerosis outcomes without the influence of known confounders. The reference category for fitness tertiles was selected as the lowest tertile, while in multinomial models, the absence of atherosclerosis was chosen as the reference. Cutoffs for the tertiles of the different exposures are shown in table 1. Combined associations of cardiorespiratory fitness and knee extension strength were performed considering the first tertiles at the low categories, and the second and third tertiles as the high categories. To examine the robustness of our main findings, we conducted a series of sensitivity analyses in coronary stenosis as follows: (1) including participants with data on all 18 coronary segments of the arterial tree (instead of including participants with data on the 11 most relevant segments), (2) recategorising calcium blooming as ≥50% stenosis (instead of analysing calcium blooming as 1%–49% stenosis), (3) excluding coronary segments with a stent (instead of considering stents as ≥50% stenosis), (4) excluding participants with selfreported CVD (myocardial infarction, coronary artery bypass grafting, percutaneous coronary intervention, stroke or peripheral arterial disease intervention) in SCAPIS, (5) excluding presumably submaximal exercise tests (either ≤85% or ≤90% of the predicted maximal heart rate calculated as 208–(0.7×age)),35 36 (6) further adjusting for height at conscription, (7) without adjusting for BMI at conscription and (8) without adjusting for muscular strength in cardiorespiratory fitness and without adjusting for cardiorespiratory fitness in muscular strength. Furthermore, to assess potential selection bias, we conducted multinomial logistic models that integrated inverse probability weighting to account for missing data in exposures, outcomes and the covariates used in the analysis.37 Finally, a sensitivity analysis was conducted, incorporating quadratic and cubic terms for quantitative covariates to evaluate the presence of nonlinearity in these covariates. All statistical tests were twosided and p<0.05 was considered statistically significant. Analyses were conducted using IBMSPSS28 (IBM Corp) and Stata V.18 (StataCorp 2021). Equity, diversity and inclusion statement This study uses data from the Swedish Military Conscription Register, which includes only male participants, a limitation we acknowledge in the limitations section. The SCAPIS is a populationbased study that includes men and women from various birth regions. Thus, 15.6% of the male participants were born outside of Sweden. We did not impose additional restrictions related to race, ethnicity, culture, socioeconomic status or representation from marginalised groups during the study’s design or data analysis. The research team comprises a diverse group of clinical and academic researchers from different countries including both women and men (4 women and 9 men). RESULTS Overall, included participants had a more favourable profile in smoking status, educational status, physical fitness and atherosclerosis compared with excluded participants (online supplemental table 1). The characteristics of the study population by tertiles of cardiorespiratory fitness and knee extension strength are presented in table 1. At conscription, the mean age of participants was 18.3 years, whereas the mean cardiorespiratory fitness and knee extension strength were 259 W and 557 N, respectively. In SCAPIS, the mean age of participants was 56.5 years (mean followup 38.2 years), and 52.6% and 58.8% of participants had coronary stenosis and carotid plaques, respectively. Cardiorespiratory fitness in adolescence and atherosclerosis in middle age The continuous (left panel) and categorical associations (right panel) of cardiorespiratory fitness in adolescence with coronary Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from 4HerraizAdilloÁ, etal. Br J Sports Med 2024;0:1–11. doi:10.1136/bjsports-2023-107663 Original research and carotid atherosclerosis in middle age are shown in figure 1 (online supplemental tables 2 and 3 depict the ORs and adjusted prevalences for such associations). In general, splines (left panel) showed a trend towards inverse associations between cardiorespiratory fitness related to segment involvement scores and CAC scores that were more pronounced for the low range of cardiorespiratory fitness values. In adjusted models, compared with adolescents in the lowest tertile of cardiorespiratory fitness, those in the medium and highest tertiles had respectively 18% (OR 0.82; 95% CI 0.66 to 1.02) and 22% (OR 0.78; 95% CI 0.61 to 0.99) lower ORs for severe (≥50%) coronary stenosis, right panel. However, there was no clear association between tertiles of cardiorespiratory fitness and 1%–49% coronary stenosis or CAC scores. Online supplemental figure 3 depicts the multinomial logistic splines for the association between cardiorespiratory fitness (as continuous variable) and atherosclerosis indicators. Regarding carotid atherosclerosis, a different pattern arises contrasting with coronary atherosclerosis. Compared with adolescents in the lowest tertile of cardiorespiratory fitness, those in the medium and highest tertiles had 18% (OR 1.18, 95% CI 1.04 to 1.34) and 17% (OR 1.17, 95% CI 1.02 to 1.35) higher ORs for unilateral carotid plaque/s, respectively, while there were no clear associations between cardiorespiratory fitness and bilateral carotid plaques. Considering composition of the coronary plaques, individuals in the highest tertile of cardiorespiratory fitness had 22% (OR 0.78, 95% CI 0.61 to 0.99) lower odds of mixed composition in the arterial tree (online supplemental figure 4 and online supplemental table 4). Table 1 Descriptive characteristics of the participants in the study by tertiles of cardiorespiratory fitness and knee extension strength in adolescence Entire sample Low CRF Medium CRF High CRF Low strength Medium strength High strength n=8986 n=3008 n=2999 n=2979 n=3035 n=2867 n=3084 Baseline, conscription Age, years 18.3±0.5 18.3±0.5 18.3±0.5 18.3±0.5 18.3±0.5 18.3±0.5 18.3±0.5 Height, cm 179.7±6.5 178.0±6.5 179.8±6.2 181.4±6.3 179.1±6.6 179.8±6.4 180.3±6.5 Weight, kg 68.9±9.1 64.8±8.9 69.5±8.5 72.5±8.1 64.9±8.1 69.1±8.3 72.8±9.0 BMI, kg/m221.3±2.4 20.5±2.5 21.5±2.3 22.0±2.1 20.2±2.2 21.4±2.2 22.4±2.3 Smoking at conscription No smoker 6451 (71.8) 1795 (59.7) 2144 (71.5) 2512 (84.3) 2059 (67.8) 2044 (71.3) 2348 (76.1) Exsmoker 60 (0.7) 12 (0.4) 33 (1.1) 15 (0.5) 17 (0.6) 23 (0.8) 20 (0.6) Current 2475 (27.5) 1201 (39.9) 822 (27.4) 452 (15.2) 959 (31.6) 800 (27.9) 716 (23.2) Cardiorespiratory fitness,* W 259.1±42.6 215.9±15.5 254.1±10.7 307.8±29.1 242.4±38.2 258.6±39.4 276.1±43.0 Extension knee strength,† N 557.0±114.1 509.5±104.1 560.5±106.7 601.5±112.2 438.7±52.7 549.8±25.3 680.2±74.9 Handgrip strength,‡ N 613.9±97.2 585.1±92.1 616.7±94.7 640.1±96.8 566.8±85.6 616.9±87.0 657.3±95.9 Elbow flexion strength,§ N 373.1±81.0 344.0±73.3 378.1±78.7 397.5±81.7 327.1±63.4 373.7±69.9 417.9±80.9 Followup, SCAPIS Followup, years 38.2±3.8 39.4±3.6 38.2±3.6 37.0±3.8 38.8±3.7 38.3±3.8 37.6±3.8 Age, years 56.5±3.9 57.8±3.8 56.5±3.8 55.3±3.8 57.1±3.9 56.6±3.9 55.9±3.9 BMI, kg/m227.4±4.0 27.1±4.0 27.5±4.0 27.5±4.0 26.6±3.7 27.5±4.0 28.0±4.1 Educational status Unfinished primary school 30 (0.3) 20 (0.7) 7 (0.2) 3 (0.1) 13 (0.4) 9 (0.3) 8 (0.3) Primary school 781 (8.7) 398 (13.2) 262 (8.7) 121 (4.1) 295 (9.7) 249 (8.7) 237 (7.7) Secondary school 4511 (50.2) 1589 (52.8) 1531 (51.1) 1391 (46.7) 1530 (50.4) 1442 (50.3) 1539 (49.9) University degree 3664 (40.8) 1001 (33.3) 1199 (40.0) 1464 (49.1) 1197 (39.4) 1167 (40.7) 1300 (42.2) Coronary stenosis, n=8514 No 4039 (47.4) 1253 (44.3) 1340 (47.3) 1446 (50.6) 1341 (47.0) 1282 (47.3) 1416 (48.0) 1%–49% 3729 (43.8) 1267 (44.8) 1249 (44.1) 1213 (42.5) 1246 (43.7) 1174 (43.3) 1309 (44.4) ≥50% 746 (8.8) 307 (10.9) 241 (8.5) 198 (6.9) 265 (9.3) 255 (9.4) 226 (7.7) CAC score, Agatston units, n=8642 0 4231 (49.0) 1319 (45.8) 1392 (48.4) 1520 (52.6) 1424 (48.6) 1337 (48.9) 1470 (49.4) 1–99 2999 (34.7) 1034 (35.9) 1002 (34.9) 963 (33.4) 1008 (34.4) 976 (35.7) 1015 (34.1) ≥100 1412 (16.3) 527 (18.3) 481 (16.7) 404 (14) 501 (17.1) 423 (15.5) 488 (16.4) Carotid plaques, n=8934 No 3690 (41.3) 1172 (39.2) 1207 (40.5) 1311 (44.2) 1182 (39.2) 1200 (42.1) 1308 (42.6) Unilateral 2725 (30.5) 863 (28.9) 940 (31.6) 922 (31.1) 949 (31.5) 845 (29.6) 931 (30.3) Bilateral 2519 (28.2) 953 (31.9) 831 (27.9) 735 (24.8) 882 (29.3) 808 (28.3) 829 (27.0) Low, medium and high strength refer to knee extension strength. All data refer to mean±SD or frequency (%). *Tertile 1: <237 W; Tertile 2: 237–274 W; Tertile 3: ≥275 W. †Tertile 1: <502 N; Tertile 2: 502–599 N; Tertile 3: ≥600 N. ‡Tertile 1: <570 N; Tertile 2: 570–648 N; Tertile 3: ≥649 N. §Tertile 1: <331 N; Tertile 2: 331–399 N; Tertile 3: ≥400 N. BMI, body mass index; CAC, coronary artery calcium; CRF, cardiorespiratory fitness; N, Newtons; SCAPIS, Swedish CArdioPulmonary bioImage Study; W, Watts. Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from 5 HerraizAdilloÁ, etal. Br J Sports Med 2024;0:1–11. doi:10.1136/bjsports-2023-107663 Original research Figure 1 Associations of cardiorespiratory fitness in adolescence with coronary and carotid atherosclerosis in middle age. Left panel depicts adjusted restricted cubic splines with 95% confidence bands for the association of cardiorespiratory fitness in adolescence with segment involvement score (0–11), CAC score and carotid plaque score (0–2) in middle age. Xaxes are trimmed to depict the associations for the 1st to 99th percentile of cardiorespiratory fitness values. Right panel depicts adjusted multinomial regression models with 95% CIs for the association of cardiorespiratory fitness in adolescence with coronary stenosis, CAC score and carotid plaques in middle age. Both splines and multinomial models are adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription and knee extension strength. BMI, body mass index; CAC, coronary artery calcium; SCAPIS, Swedish CArdioPulmonary bioImage Study. Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from 6HerraizAdilloÁ, etal. Br J Sports Med 2024;0:1–11. doi:10.1136/bjsports-2023-107663 Original research Figure 2 Associations of knee extension strength in adolescence with coronary and carotid atherosclerosis in middle age. Left panel depicts adjusted restricted cubic splines with 95% confidence bands for the association of knee extension strength in adolescence with segment involvement score (0–11), CAC score and carotid plaque score (0–2) in middle age. Xaxes are trimmed to depict the associations for the 1st to 99th percentile of knee extension strength values. Right panel depicts adjusted multinomial regression models with 95% CIs for the association of knee extension strength in adolescence with coronary stenosis, CAC score and carotid plaques in middle age. Both splines and multinomial models are adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription and cardiorespiratory fitness. BMI, body mass index; CAC, coronary artery calcium; SCAPIS, Swedish CArdioPulmonary bioImage Study. Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from 7 HerraizAdilloÁ, etal. Br J Sports Med 2024;0:1–11. doi:10.1136/bjsports-2023-107663 Original research Muscular strength in adolescence and atherosclerosis in middle age Figure 2 depicts the continuous (left panel) and categorical (right panel) associations of knee extension strength in adolescence with coronary and carotid atherosclerosis in middle age (online supplemental tables 5,6 depict the ORs and adjusted prevalences for such associations). Overall, splines showed inverse associations between knee extension strength and atherosclerosis outcomes. In consonance with cardiorespiratory fitness, there was a negative association between knee extension strength and severe (≥50%) coronary stenosis in the adjusted model. When compared with the lowest tertile of knee extension strength, those in the medium and highest tertiles had 11% (OR 0.89; 95% CI 0.72 to 1.10) and 26% (OR 0.74; 95% CI 0.58 to 0.93) lower ORs for severe coronary stenosis, respectively. Regarding CAC score, adolescents in the medium and highest tertiles had 19% (OR 0.81; 95% CI 0.68 to 0.96) and 16% (OR 0.84; 95% CI 0.70 to 1.01) lower ORs for a CAC score ≥100. No clear associations were observed between knee extension strength and carotid plaques. Online supplemental figure 5 depicts the multinomial logistic splines for the association between knee muscular strength (as continuous variable) and atherosclerosis indicators. Considering composition of the plaques, knee extension strength did not show clear associations with any types of coronary plaques (online supplemental figure 4 and online supplemental table 7). Handgrip strength and elbow flexion strength exhibited somewhat similar, although weaker, patterns of association with coronary and carotid atherosclerosis compared with knee extension strength (online supplemental figure 6 and online supplemental tables 8,9). Combined associations of cardiorespiratory fitness and knee extension strength in adolescence with atherosclerosis in middle age Figure 3 depicts the combined associations of cardiorespiratory fitness and knee extension strength in adolescence with atherosclerosis in middle age, while online supplemental tables 10,11 depict the ORs and adjusted prevalences for these associations. There was a trend towards less severe (≥50%) coronary stenosis with higher levels of cardiorespiratory fitness and strength, with those in the highest physical fitness group having 33% (OR 0.67; 95% CI 0.52 to 0.87) lower OR compared with those with the lowest physical fitness. Regarding CAC score, participants in the highest physical fitness group had 24% (OR 0.76; 95% CI 0.62 to 0.93) lower OR for a CAC score ≥100 compared with those in the lowest physical fitness group. However, those in the highest physical fitness group did not have lower ORs for carotid plaques. Sensitivity analyses In coronary atherosclerosis, the inclusion of all coronary segments, the recategorised definition of ‘calcium blooming’ artefact or stent, as well as the exclusion of participants with CVD, did not significantly alter the associations between cardiorespiratory fitness and knee extension strength in relation to coronary stenosis (online supplemental table 12). In a second sensitivity analysis, the exclusion of presumably nonmaximal tests generally strengthened the associations with coronary stenosis (online supplemental table 13). Further adjustment for height in adolescence generally attenuated the associations between cardiorespiratory fitness and coronary stenosis but did not influence corresponding associations with knee extension strength (online supplemental table 14). As shown in online supplemental table 15, removing the adjustment for BMI attenuated the associations of cardiorespiratory fitness and knee extension strength with coronary stenosis. Associations of cardiorespiratory fitness and knee extension strength were generally unaffected when they were not mutually adjusted for each other (online supplemental table 16). Finally, associations between cardiorespiratory fitness and atherosclerosis outcomes remained robust in the inverse probability weighting analysis (online supplemental table 17) and when incorporating quadratic and cubic terms for quantitative covariates in multinomial logistic (online supplemental tables 18,19) and linear models (data not shown). DISCUSSION This large population basedstudy showed inverse associations between cardiorespiratory fitness during adolescence and coronary atherosclerosis, particularly severe (≥50%) coronary stenosis, almost 40 years later. Furthermore, knee extension strength in adolescence showed inverse associations not only Figure 3 Combined associations of cardiorespiratory fitness and knee extension strength in adolescence with coronary stenosis, CAC score and carotid plaques in middle age. All models depict multinomial regression models with 95% CIs adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription and years of smoking at conscription. Low categories refer to the first tertile, while high categories refer to the second and third tertiles. BMI, body mass index; CAC, coronary artery calcium; CRF, cardiorespiratory fitness; Strength, knee extension muscular strength; SCAPIS: Swedish CArdioPulmonary bioImage Study. Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from 8HerraizAdilloÁ, etal. Br J Sports Med 2024;0:1–11. doi:10.1136/bjsports-2023-107663 Original research with severe coronary stenosis but also with high CAC scores in middle age. However, neither cardiorespiratory fitness nor knee extension strength was robustly associated with the presence of bilateral carotid plaques. Finally, the combination of high cardiorespiratory fitness and knee extension strength levels was strongly associated with a lower presence of severe coronary stenosis and high CAC scores. To the best of our knowledge, this is the first study analysing the associations between cardiorespiratory fitness in adolescence and coronary atherosclerosis in middle age measured with CCTA, an accurate noninvasive imaging technique that allows characterisation and quantification not only of calcified but also noncalcified plaques. In our study, the splines linear models suggested a somewhat reverse Jshape pattern for the association between cardiorespiratory fitness and segment involvement score, with values below 240 W (≈first tertile) associated with a worse segment involvement score. In consonance with this, in our study, after adjustments, individuals in the highest tertile of cardiorespiratory fitness had 22% lower odds of having severe coronary stenosis. Interestingly, the splines in multinomial models also supported that higher cardiorespiratory fitness associates with decreased coronary atherosclerosis. However, very high fitness levels may not confer similar protection as moderately high levels, even suggesting a potential negative effect at very high levels of cardiorespiratory fitness (around the 95th percentile). Nevertheless, wide CIs in extreme fitness values preclude definitive associations. Regarding this, it should be acknowledged that while better levels of cardiorespiratory fitness have been crosssectionally associated with a lower risk of coronary calcification,38 certain populations with very high levels of cardiorespiratory fitness such as endurance athletes seem to have an increased burden of coronary atherosclerosis, suggesting a Ushape relationship.38–41 Yet, the clinical significance of accelerated coronary artery atherosclerosis in athletes engaged in very high volumeintensity exercise remains unclear.42 43 Further studies are needed in this context. Interestingly, high cardiorespiratory fitness was associated with less prevalence of a mixed composition (presence of both calcified and noncalcified segments) in the arterial tree, which is consistent with the lack of association observed between tertiles of cardiorespiratory fitness and CAC. These findings align with previous studies that have reported lower prevalence of mixed plaques in the coronary artery among athletes44 or individuals with high exercise volume,41 which is of relevance given the clear association between cardiorespiratory fitness and exercise.45 This observation may be of importance since individuals with noncalcified or mixed plaques have been associated with a worse prognosis compared with those with predominantly calcified plaques.46 47 In our sensitivity analyses, associations between cardiorespiratory fitness and coronary stenosis were attenuated when estimates were not adjusted for BMI at conscription. This is intriguing and could be attributed to the selected cardiorespiratory fitness test (ie, nonweightbearing). Notably, BMI and performance in cycleergometer tests (measured in W) often exhibit a positive correlation,48 possibly because higher body mass can generate more power. However, BMI is also strongly linked to atherosclerosis risk, which might account for the observed attenuation in our sensitivity analyses. Additional research on this subject is needed. Although no previous study has explored the associations of cardiorespiratory fitness in adolescence with later coronary atherosclerosis, our findings may be compared with previous studies that have linked cardiorespiratory fitness in adulthood to CAC later in life. The CARDIA study found that high levels of cardiorespiratory fitness in young adults were associated with 41% lower odds of coronary calcification after 15 years of followup.7 However, another study also based on the CARDIA cohort found that although cardiorespiratory fitness was favourably associated with cardiac structure and function, it was not associated with CAC scores approximately 27 years later.8 Despite different levels of covariate adjustment or followup could partially explain these differences, the baseline level of cardiorespiratory fitness (and physical activity) could also influence the associations between cardiorespiratory fitness and CAC. In our study, despite a lack of clear association for tertiles of cardiorespiratory fitness and CAC, the observed pattern in linear and multinomial logistic regression splines in CAC was concordant with that observed for coronary stenosis, suggesting that being unfit (cardiorespiratory fitness levels below first tertile, ≈240 W) is associated with greater risk. In previous studies, the associations between muscular strength and CVD have generally been weaker compared with those observed for cardiorespiratory fitness.49 However, in our study, the associations with coronary stenosis for muscular strength were similar or even slightly stronger than those for cardiorespiratory fitness. In fact, knee extension strength (more than handgrip strength or elbow flexion strength) was inversely associated not only with the presence of severe coronary stenosis, but also with a high CAC score, which was not clearly associated with tertiles of cardiorespiratory fitness. These findings are consistent with our results regarding the combined associations of cardiorespiratory fitness and knee extension strength. They indicated that achieving lower odds of coronary atherosclerosis requires the simultaneous presence of acceptable levels of cardiorespiratory fitness and knee extension strength, underscoring the integrated nature of physical fitness. Our results regarding carotid plaques are intriguing: we did not observe consistent associations between cardiorespiratory fitness and bilateral plaques, but we found an inverted Ushaped association with unilateral plaques. This is in contrast with a previous study also analysing conscripted Swedish men, which found that cardiorespiratory fitness was associated with 19% lower odds of carotid plaques at 60 years of age.14 However, this study analysed a sample size 10 times smaller, and considered carotid plaques as a dichotomous variable (no plaque, plaque/s) instead as continuous and multinomial ones (no plaque, unilateral plaque/s, bilateral plaques) as in our study. In addition, the Cooper Center Longitudinal Study found that midlife cardiorespiratory fitness was inversely associated with carotid artery disease measured almost two decades later.50 However, this study characterised low cardiorespiratory fitness as the first quintile and used a different definition of carotid artery disease than our study. Further studies are therefore needed to elucidate the associations of cardiorespiratory fitness in adolescence with the development of carotid plaques later in life. Strengths and limitations The main strength of this study was the utilisation of CCTA on a populationbased scale, enabling the characterisation of calcified and noncalcified coronary plaques within a sizeable sample of the population. Furthermore, the study benefits from a young cohort that was followed up for nearly 40 years, minimising the possibility of reverse causation, as it is highly unlikely that disease in adolescence caused low physical fitness. The study is also informative of the verylong term prognostic value of cardiorespiratory fitness and muscular strength. Additionally, physical Kirjasto/Kausijulkaisut. Protected by copyright. on February 21, 2024 at Jyvaskylan Yliopostohttp://bjsm.bmj.com/Br J Sports Med: first published as 10.1136/bjsports-2023-107663 on 14 February 2024. Downloaded from Supplementary Table 11. Adjusted (marginal) prevalences for the combined associations of cardiorespiratory fitness and knee extension strength in adolescence and coronary stenosis, CAC scores, and carotid plaques in middle-age. Supplementary Table 12. Associations of cardiorespiratory fitness and knee extension strength in adolescence and coronary stenosis in middle-age, sensitivity analysis. Supplementary Table 13. Multinomial logistic regression, associations of cardiorespiratory fitness in adolescence and coronary stenosis, CAC scores, and carotid plaques in middle-age, sensitivity analysis of potentially submaximal tests. Supplementary Table 14. Multinomial logistic regression, associations of cardiorespiratory fitness and knee extension strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis with and without adjustment for height at conscription. Supplementary Table 15. Multinomial logistic regression, associations of cardiorespiratory fitness and knee extension strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis with and without adjustment for BMI at conscription. Supplementary Table 16. Multinomial logistic regression, associations of cardiorespiratory fitness and knee extension strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis with and without adjustment for knee extension strength and cardiorespiratory fitness. Supplementary Table 17. Multinomial logistic regression, associations of cardiorespiratory fitness in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age by different degrees of covariate control, sensitivity analysis by inverse probability weighting. Supplementary Table 18. Multinomial logistic regression, associations of cardiorespiratory fitness in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis considering quadratic and cubic terms in quantitative covariates. Supplementary Table 19. Multinomial logistic regression, associations of knee extension strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis considering quadratic and cubic terms in quantitative covariates. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Figures Supplementary Figure 1. Flow chart of the study. Supplementary Figure 2. Directed acyclic graph for the association between cardiorespiratory fitness and muscular strength and atherosclerosis outcomes. Supplementary Figure 3. Multinomial logistic regression with restricted cubic splines for the associations of cardiorespiratory fitness in adolescence and atherosclerosis in middle age. Supplementary Figure 4. Associations of cardiorespiratory fitness and knee extension strength in adolescence with composition of coronary plaques in middle age. Supplementary Figure 5. Multinomial logistic regression with restricted cubic splines for the associations of knee extension strength in adolescence and atherosclerosis in middle age. Supplementary Figure 6. Associations of handgrip strength and elbow flexion strength in adolescence with coronary stenosis, CAC score and carotid plaques in middle age. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 1. Descriptive characteristics of the included versus excluded participants in the study. Included, n=8986 Excluded, n=5660 n n Baseline, conscription Age, y 8986 18.3 ± 0.5 2127 19.1 ± 1.5 Height, cm 8986 179.7 ± 6.5 2075 179.1 ± 6.8 Weight, kg 8986 68.9 ± 9.1 2070 67.5 ± 8.7 BMI, kg/m2 8986 21.3 ± 2.4 2070 21.1 ± 3.1 Smoking at conscription 8986 1785 No smoker 6451 (71.8) 952 (53.3) Ex-smoker 60 (0.7) 22 (1.2) Current 2475 (27.5) 811 (45.4) Cardiorespiratory fitness, Watts 8986 259.1 ± 42.6 658 254.3 ± 43.2 Extension knee strength, Newtons 8986 557.0 ± 114.1 2074 544.9 ± 109.5 Handgrip strength, Newtons 8986 613.9 ± 97.2 2074 605.6 ± 96.2 Elbow flexion strength, Newtons 8986 373.1 ± 81.0 2074 371.7 ± 86.0 Follow up, SCAPIS Follow up, y 8986 38.2 ± 3.8 2127 42.3 ± 4.6 Age, y 8986 56.5 ± 3.9 5660 59.1 ± 4.6 BMI, kg/m2 8986 27.4 ± 4.0 5658 27.6 ± 4.0 Educational status 8986 5136 Unfinished primary school 30 (0.3) 70 (1.4) Primary school 781 (8.7) 612 (11.9) Secondary school 4511 (50.2) 2360 (46.0) University degree 3664 (40.8) 2094 (40.8) Coronary stenosis 8514 1991 No 4039 (47.4) 626 (31.4) 1-49% 3729 (43.8) 1058 (53.1) ≥50% 746 (8.8) 307 (15.4) CAC score, Agatston units 8642 2121 0 4231 (49.0) 715 (33.7) 1-99 2999 (34.7) 784 (37.0) ≥100 1412 (16.3) 622 (29.3) Carotid plaques 8934 2264 No 3690 (41.3) 692 (30.6) Unilateral 2725 (30.5) 691 (30.5) Bilateral 2519 (28.2) 881 (38.9) BMI: body mass index, CAC: coronary artery calcium, SCAPIS: Swedish CArdioPulmonary bioImage Study. All data refer to mean ± standard deviation or frequency (%). BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 2. Multinomial logistic regression, associations of cardiorespiratory fitness in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age by different degrees of covariate control. Unadjusted Adjusted Cardiorespiratory fitness OR (95% CI) P OR (95% CI) P CCTA coronary stenosis 1-49% Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.91 (0.81-1.01) 0.087 0.97 (0.86-1.10) 0.677 Highest tertile 0.82 (0.74-0.92) <0.001 0.99 (0.87-1.13) 0.871 ≥50% Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.75 (0.61-0.91) 0.004 0.82 (0.66-1.02) 0.073 Highest tertile 0.58 (0.47-0.71) <0.001 0.78 (0.61-0.99) 0.040 CAC score 1-99 Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.88 (0.78-1.00) 0.046 0.92 (0.81-1.05) 0.223 Highest tertile 0.81 (0.72-0.91) <0.001 0.93 (0.81-1.07) 0.314 ≥100 Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.85 (0.73-0.99) 0.038 0.92 (0.78-1.09) 0.357 Highest tertile 0.69 (0.59-0.81) <0.001 0.90 (0.75-1.09) 0.281 Ultrasound carotid plaque Unilateral Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 1.06 (0.94-1.19) 0.370 1.18 (1.04-1.34) 0.012 Highest tertile 0.95 (0.85-1.08) 0.460 1.17 (1.02-1.35) 0.026 Bilateral Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.85 (0.75-0.96) 0.008 1.04 (0.92-1.19) 0.510 Highest tertile 0.69 (0.61-0.78) <0.001 1.04 (0.90-1.20) 0.576 BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomographic angiography, CI: confidence interval, OR: odds ratio, ref.: reference, SCAPIS: Swedish CArdioPulmonary bioImage Study. Reference category for CCTA stenosis, CAC score and carotid plaques: no stenosis, CAC score=0 Agatston units and no plaque, respectively, Adjusted model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and knee extension strength. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 3. Adjusted (marginal) prevalences for the multinomial logistic regression model of cardiorespiratory fitness in adolescence and coronary and carotid atherosclerosis in middle-age. Cardiorespiratory fitness Adjusted prevalence (95% CI) Adjusted prevalence (95% CI) Adjusted prevalence (95% CI) No coronary stenosis 1-49% coronary stenosis ≥50% coronary stenosis Lowest tertile 47.37 (45.38-49.35) 43.84 (41.85-45.82) 8.80 (7.70-9.89) Medium tertile 48.55 (46.70-50.39) 43.98 (42.12-45.84) 7.47 (6.49-8.45) Highest tertile 48.42 (46.47-50.37) 44.53 (42.55-46.51) 7.05 (5.99-8.12) CAC score=0 CAC score 1-99 CAC score ≥100 Lowest tertile 49.05 (47.04-51.06) 35.01 (33.05-36.96) 15.94 (14.53-17.36) Medium tertile 50.91 (49.05-52.78) 33.68 (31.86-35.49) 15.41 (14.07-16.75) Highest tertile 50.93 (48.96-52.91) 33.98 (32.05-35.92) 15.08 (13.62-16.55) No carotid plaque Unilateral carotid plaque/s Bilateral carotid plaques Lowest tertile 43.00 (41.11-44.90) 28.47 (26.74-30.20) 28.53 (26.88-30.17) Medium tertile 40.46 (38.73-42.20) 31.56 (29.89-33.24) 27.97 (26.39-29.56) Highest tertile 40.57 (38.74-42.40) 31.46 (29.66-33.26) 27.97 (26.21-29.73) BMI: body mass index, CAC: coronary artery calcium, CI: confidence interval, SCAPIS: Swedish CArdioPulmonary bioImage Study. Marginal prevalences are adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and knee extension strength. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 4. Multinomial logistic regression, associations of cardiorespiratory fitness in adolescence and composition of coronary segments in middle-age by different degrees of covariate control. Unadjusted Adjusted Cardiorespiratory fitness OR (95% CI) P OR (95% CI) P Composition of coronary segments at an arterial tree level Only noncalcified segments Lowest tertile. ref. 1.00 - - 1.00 - - Medium tertile 1.17 (0.83-1.65) 0.357 1.22 (0.85-1.75) 0.274 Highest tertile 1.02 (0.72-1.44) 0.916 1.07 (0.72-1.59) 0.742 Only calcified segments Lowest tertile. ref. 1.00 - - 1.00 - - Medium tertile 0.88 (0.78-0.98) 0.026 0.96 (0.84-1.08) 0.486 Highest tertile 0.80 (0.71-0.89) <0.001 0.98 (0.86-1.13) 0.831 Mixed composition Lowest tertile. ref. 1.00 - - 1.00 - - Medium tertile 0.81 (0.66-0.99) 0.042 0.84 (0.68-1.05) 0.121 Highest tertile 0.63 (0.51-0.78) <0.001 0.78 (0.61-0.99) 0.040 BMI: body mass index, CI: confidence interval, OR: odds ratio, ref.: reference, SCAPIS: Swedish CArdioPulmonary bioImage Study. Reference categories for composition of coronary segments was “no plaque”. Mixed composition: presence of both calcified and noncalcified segments in the arterial tree. Adjusted model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and knee extension strength. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 5. Multinomial logistic regression, associations of knee extension strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age by different degrees of covariate control. Unadjusted Adjusted Knee extension strength OR (95% CI) P OR (95% CI) P CCTA coronary stenosis 1-49% Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.97 (0.87-1.09) 0.592 0.94 (0.83-1.06) 0.324 Highest tertile 0.99 (0.88-1.10) 0.797 0.95 (0.84-1.08) 0.443 ≥50% Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.93 (0.76-1.14) 0.491 0.89 (0.72-1.10) 0.284 Highest tertile 0.77 (0.62-0.94) 0.012 0.74 (0.58-0.93) 0.012 CAC score 1-99 Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 1.04 (0.92-1.17) 0.520 1.02 (0.89-1.15) 0.797 Highest tertile 0.98 (0.87-1.10) 0.732 0.95 (0.83-1.09) 0.502 ≥100 Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.87 (0.74-1.02) 0.091 0.81 (0.68-0.96) 0.016 Highest tertile 0.93 (0.80-1.09) 0.384 0.84 (0.70-1.01) 0.060 Ultrasound carotid plaque Unilateral Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.88 (0.78-0.99) 0.036 0.89 (0.79-1.02) 0.086 Highest tertile 0.89 (0.79-1.00) 0.049 0.93 (0.81-1.06) 0.268 Bilateral Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.90 (0.80-1.02) 0.106 0.97 (0.85-1.11) 0.687 Highest tertile 0.85 (0.75-0.96) 0.009 1.02 (0.89-1.18) 0.748 BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomographic angiography, CI: confidence interval, OR: odds ratio, ref.: reference, SCAPIS: Swedish CArdioPulmonary bioImage Study. Reference category for CCTA stenosis, CAC score and carotid plaques: no stenosis, CAC score=0 Agatston units and no plaque, respectively. Adjusted model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and cardiorespiratory fitness. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 6. Adjusted (marginal) prevalences for the multinomial logistic regression model of knee extension strength in adolescence and coronary and carotid atherosclerosis in middle-age. Knee extension strength Adjusted prevalence (95% CI) Adjusted prevalence (95% CI) Adjusted prevalence (95% CI) No coronary stenosis 1-49% coronary stenosis ≥50% coronary stenosis Lowest tertile 46.92 (44.99-48.85) 44.40 (42.44-46.36) 8.68 (7.56-9.79) Medium tertile 48.52 (46.65-50.39) 43.43 (41.55-45.32) 8.05 (7.02-9.08) Highest tertile 48.89 (46.97-50.80) 44.34 (42.42-46.27) 6.77 (5.79-7.74) CAC score=0 CAC score 1-99 CAC score ≥100 Lowest tertile 49.26 (47.31-51.21) 33.71 (31.80-35.62) 17.03 (15.54-18.53) Medium tertile 50.39 (48.49-52.29) 35.28 (33.42-37.15) 14.33 (12.99-15.66) Highest tertile 51.22 (49.29-53.15) 33.67 (31.80-35.54) 15.11 (13.75-16.48) No carotid plaque Unilateral carotid plaque/s Bilateral carotid plaques Lowest tertile 40.50 (38.66-42.34) 31.81 (30.04-33.58) 27.69 (26.06-29.32) Medium tertile 42.20 (40.42-43.98) 29.67 (28.00-31.35) 28.12 (26.51-29.74) Highest tertile 41.21 (39.41-43.02) 29.98 (28.27-31.70) 28.81 (27.11-30.51) BMI: body mass index, CAC: coronary artery calcium, CI: confidence interval, SCAPIS: Swedish CArdioPulmonary bioImage Study. Marginal prevalences are adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and cardiorespiratory fitness. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 7. Multinomial logistic regression, associations of knee extension strength in adolescence and composition of coronary segments in middle-age by different degrees of covariate control. Unadjusted Adjusted Knee extension strength OR (95% CI) P OR (95% CI) P Composition of coronary segments at an arterial tree level Only noncalcified plaques Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.95 (0.67-1.34) 0.759 0.94 (0.66-1.34) 0.740 Highest tertile 1.07 (0.77-1.48) 0.698 1.09 (0.75-1.57) 0.658 Only calcified plaques Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.98 (0.87-1.09) 0.676 0.95 (0.84-1.07) 0.375 Highest tertile 0.96 (0.85-1.07) 0.440 0.92 (0.81-1.05) 0.228 Mixed composition Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.92 (0.75-1.13) 0.423 0.89 (0.71-1.10) 0.276 Highest tertile 0.88 (0.72-1.08) 0.234 0.86 (0.68-1.08) 0.191 BMI: body mass index, CI: confidence interval, OR: odds ratio, ref.: reference, SCAPIS: Swedish CArdioPulmonary bioImage Study. Reference categories for composition of coronary segments was “no plaque”. Mixed composition: presence of both calcified and noncalcified segments in the arterial tree. Adjusted model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and cardiorespiratory fitness. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 8. Multinomial logistic regression, associations of handgrip strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age by different degrees of covariate control. Unadjusted Adjusted Handgrip strength OR (95% CI) P OR (95% CI) P CCTA coronary stenosis 1-49% Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 1.06 (0.95-1.19) 0.283 1.01 (0.89-1.13) 0.905 Highest tertile 1.08 (0.97-1.21) 0.154 0.99 (0.88-1.12) 0.923 ≥50% Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.96 (0.78-1.18) 0.730 0.90 (0.72-1.11) 0.319 Highest tertile 0.91 (0.74-1.12) 0.393 0.81 (0.64-1.02) 0.069 CAC score 1-99 Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 1.06 (0.94-1.20) 0.330 1.01 (0.89-1.14) 0.885 Highest tertile 1.08 (0.96-1.22) 0.221 1.00 (0.87-1.14) 0.982 ≥100 Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 1.02 (0.87-1.20) 0.781 0.93 (0.78-1.10) 0.404 Highest tertile 1.08 (0.93-1.27) 0.313 0.92 (0.77-1.10) 0.382 Ultrasound carotid plaque Unilateral Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 1.13 (1.00-1.27) 0.053 1.12 (0.99-1.27) 0.076 Highest tertile 1.04 (0.92-1.17) 0.571 1.03 (0.90-1.17) 0.704 Bilateral Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.98 (0.87-1.12) 0.809 1.00 (0.87-1.14) 0.952 Highest tertile 1.02 (0.90-1.15) 0.760 1.05 (0.92-1.21) 0.451 BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomographic angiography, CI: confidence interval, OR: odds ratio, ref.: reference, SCAPIS: Swedish CArdioPulmonary bioImage Study. Reference category for CCTA stenosis, CAC score and carotid plaques: no stenosis, CAC score=0 Agatston units and no plaque, respectively. Adjusted model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and cardiorespiratory fitness. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 15. Multinomial logistic regression, associations of cardiorespiratory fitness and knee extension strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis with and without adjustment for BMI at conscription. Cardiorespiratory fitness Knee extension strength Adjusted Adjusted minus BMI at conscription Adjusted Adjusted minus BMI at conscription OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P CCTA coronary stenosis n=8006 n=8006 1-49% Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.97 (0.86-1.10) 0.677 1.03 (0.92-1.16) 0.603 0.94 (0.83-1.06) 0.324 1.02 (0.91-1.15) 0.730 Highest tertile 0.99 (0.87-1.13) 0.871 1.08 (0.95-1.23) 0.215 0.95 (0.84-1.08) 0.443 1.10 (0.98-1.24) 0.111 ≥50% Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.82 (0.66-1.02) 0.073 0.90 (0.72-1.11) 0.324 0.89 (0.72-1.10) 0.284 1.02 (0.82-1.25) 0.882 Highest tertile 0.78 (0.61-0.99) 0.040 0.90 (0.70-1.14) 0.370 0.74 (0.58-0.93) 0.012 0.94 (0.75-1.18) 0.607 CAC score. Agatston units n=7849 n=7849 1-99 Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.92 (0.81-1.05) 0.223 0.97 (0.86-1.11) 0.690 1.02 (0.89-1.15) 0.797 1.09 (0.97-1.24) 0.156 Highest tertile 0.93 (0.81-1.07) 0.314 1.01 (0.88-1.16) 0.844 0.95 (0.83-1.09) 0.502 1.09 (0.96-1.24) 0.189 ≥100 Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.92 (0.78-1.09) 0.357 1.03 (0.87-1.21) 0.766 0.81 (0.68-0.96) 0.016 0.95 (0.80-1.13) 0.564 Highest tertile 0.90 (0.75-1.09) 0.281 1.06 (0.88-1.28) 0.516 0.84 (0.70-1.01) 0.060 1.13 (0.95-1.34) 0.160 Ultrasound carotid plaque n=8934 n=8934 Unilateral Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 1.18 (1.04-1.34) 0.012 1.18 (1.04-1.34) 0.011 0.89 (0.79-1.02) 0.086 0.90 (0.79-1.02) 0.090 Highest tertile 1.17 (1.02-1.35) 0.026 1.17 (1.02-1.35) 0.023 0.93 (0.81-1.06) 0.268 0.93 (0.82-1.06) 0.285 Bilateral Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 1.04 (0.92-1.19) 0.510 1.05 (0.92-1.19) 0.487 0.97 (0.85-1.11) 0.687 0.97 (0.85-1.11) 0.690 Highest tertile 1.04 (0.90-1.20) 0.576 1.04 (0.91-1.21) 0.543 1.02 (0.89-1.18) 0.748 1.02 (0.89-1.17) 0.722 BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomographic angiography, CI: confidence interval, OR: odds ratio, SCAPIS: Swedish Cardiopulmonary Bioimage Study. Reference category for CCTA stenosis, CAC score and carotid plaques: no stenosis, CAC score=0 Agatston units and no plaque, respectively. Adjusted model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and knee extension strength (for cardiorespiratory fitness) and cardiorespiratory fitness (for knee extension strength). Adjusted model minus BMI at conscription: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, years of smoking at conscription, and knee extension strength (for cardiorespiratory fitness) and cardiorespiratory fitness (for knee extension strength). BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 16. Multinomial logistic regression, associations of cardiorespiratory fitness and knee extension strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis with and without adjustment for knee extension strength and cardiorespiratory fitness. Cardiorespiratory fitness Knee extension strength Adjusted Adjusted minus Knee extension strength Adjusted Adjusted minus Cardiorespiratory fitness OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P CCTA coronary stenosis n=8006 n=8006 1-49% Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.97 (0.86-1.10) 0.677 0.97 (0.86-1.10) 0.669 0.94 (0.83-1.06) 0.324 0.94 (0.84-1.06) 0.339 Highest tertile 0.99 (0.87-1.13) 0.871 0.99 (0.87-1.12) 0.859 0.95 (0.84-1.08) 0.443 0.96 (0.84-1.08) 0.480 ≥50% Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.82 (0.66-1.02) 0.073 0.80 (0.65-0.99) 0.044 0.89 (0.72-1.10) 0.284 0.88 (0.71-1.09) 0.244 Highest tertile 0.78 (0.61-0.99) 0.040 0.74 (0.59-0.94) 0.014 0.74 (0.58-0.93) 0.012 0.72 (0.57-0.91) 0.006 CAC score. Agatston units n=7849 n=7849 1-99 Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.92 (0.81-1.05) 0.223 0.92 (0.81-1.05) 0.217 1.02 (0.89-1.15) 0.797 1.02 (0.90-1.15) 0.795 Highest tertile 0.93 (0.81-1.07) 0.314 0.93 (0.81-1.07) 0.297 0.95 (0.83-1.09) 0.502 0.95 (0.83-1.09) 0.494 ≥100 Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.92 (0.78-1.09) 0.357 0.92 (0.77-1.08) 0.306 0.81 (0.68-0.96) 0.016 0.81 (0.68-0.96) 0.015 Highest tertile 0.90 (0.75-1.09) 0.281 0.89 (0.74-1.07) 0.206 0.84 (0.70-1.01) 0.060 0.84 (0.70-1.00) 0.049 Ultrasound carotid plaque n=8934 n=8934 Unilateral Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 1.18 (1.04-1.34) 0.012 1.17 (1.03-1.33) 0.015 0.89 (0.79-1.02) 0.086 0.90 (0.80-1.03) 0.122 Highest tertile 1.17 (1.02-1.35) 0.026 1.16 (1.01-1.33) 0.035 0.93 (0.81-1.06) 0.268 0.95 (0.83-1.08) 0.456 Bilateral Lowest tertile. ref. 1.00 - - 1.00 - - 1.00 - - 1.00 - - Medium tertile 1.04 (0.92-1.19) 0.510 1.04 (0.91-1.19) 0.529 0.97 (0.85-1.11) 0.687 0.98 (0.86-1.11) 0.713 Highest tertile 1.04 (0.90-1.20) 0.576 1.04 (0.90-1.20) 0.609 1.02 (0.89-1.18) 0.748 1.03 (0.90-1.18) 0.685 BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomographic angiography, CI: confidence interval, OR: odds ratio, SCAPIS: Swedish Cardiopulmonary Bioimage Study. Reference category for CCTA stenosis, CAC score and carotid plaques: no stenosis, CAC score=0 Agatston units and no plaque, respectively. Adjusted model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and knee extension strength (for cardiorespiratory fitness) and cardiorespiratory fitness (for knee extension strength). Adjusted minus Knee extension strength model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, and years of smoking at conscription. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Adjusted minus Cardiorespiratory fitness model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, and years of smoking at conscription. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 17. Multinomial logistic regression, associations of cardiorespiratory fitness in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age by different degrees of covariate control, sensitivity analysis by inverse probability weighting. Adjusted Adjusted, IPW1 Cardiorespiratory fitness OR (95% CI) P OR (95% CI) P CCTA coronary stenosis N=8006 N=8514 1-49% Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.97 (0.86-1.10) 0.677 0.99 (0.88-1.11) 0.819 Highest tertile 0.99 (0.87-1.13) 0.871 0.99 (0.87-1.13) 0.891 ≥50% Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.82 (0.66-1.02) 0.073 0.80 (0.65-0.98) 0.033 Highest tertile 0.78 (0.61-0.99) 0.040 0.75 (0.60-0.94) 0.013 CAC score N=7849 N=8642 1-99 Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.92 (0.81-1.05) 0.223 0.97 (0.85-1.10) 0.632 Highest tertile 0.93 (0.81-1.07) 0.314 0.91 (0.79-1.05) 0.192 ≥100 Lowest tertile, ref. 1.00 - - 1.00 - - Medium tertile 0.92 (0.78-1.09) 0.357 0.93 (0.79-1.10) 0.421 Highest tertile 0.90 (0.75-1.09) 0.281 0.86 (0.71-1.03) 0.097 BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomographic angiography, CI: confidence interval, OR: odds ratio, ref.: reference, SCAPIS: Swedish CArdioPulmonary bioImage Study. Reference category for CCTA stenosis, CAC score and carotid plaques: no stenosis, CAC score=0 Agatston units and no plaque, respectively, Adjusted model: adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and knee extension strength. 1 Multinomial logistic model incorporates inverse probability weighting to account for missing data in exposures, outcomes and main analysis covariates. We used the exposure, outcome and completely observed covariates to predict missingness. Later, we used the inverse of these predictions to reweigh our main analysis (Seaman, S. R. et al. Stat Methods Med Res, 2013: 22(3), 278–295). BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 18. Multinomial logistic regression, associations of cardiorespiratory fitness in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis considering quadratic and cubic terms in quantitative covariates. Main model1 Quadratic2 Cubic3 Cardiorespiratory fitness OR (95% CI) P OR (95% CI) P OR (95% CI) P CCTA coronary stenosis 1-49% Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.97 (0.86-1.10) 0.677 0.98 (0.87-1.11) 0.737 0.98 (0.87-1.11) 0.740 Highest tertile 0.99 (0.87-1.13) 0.871 0.99 (0.87-1.14) 0.919 0.99 (0.87-1.14) 0.923 ≥50% Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.82 (0.66-1.02) 0.073 0.81 (0.65-1.00) 0.054 0.81 (0.65-1.00) 0.055 Highest tertile 0.78 (0.61-0.99) 0.040 0.76 (0.59-0.97) 0.028 0.76 (0.59-0.97) 0.028 CAC score 1-99 Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.92 (0.81-1.05) 0.223 0.94 (0.82-1.07) 0.336 0.94 (0.82-1.07) 0.339 Highest tertile 0.93 (0.81-1.07) 0.314 0.95 (0.82-1.10) 0.475 0.95 (0.82-1.10) 0.487 ≥100 Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.92 (0.78-1.09) 0.357 0.91 (0.77-1.08) 0.271 0.91 (0.77-1.08) 0.277 Highest tertile 0.90 (0.75-1.09) 0.281 0.88 (0.73-1.06) 0.181 0.88 (0.73-1.06) 0.184 Ultrasound carotid plaque Unilateral Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 1.18 (1.04-1.34) 0.012 1.16 (1.02-1.33) 0.023 1.17 (1.02-1.33) 0.022 Highest tertile 1.17 (1.02-1.35) 0.026 1.16 (1.01-1.34) 0.043 1.16 (1.00-1.34) 0.044 Bilateral Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 1.04 (0.92-1.19) 0.510 1.04 (0.91-1.18) 0.603 1.03 (0.91-1.18) 0.623 Highest tertile 1.04 (0.90-1.20) 0.576 1.04 (0.89-1.20) 0.630 1.03 (0.89-1.20) 0.663 BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomographic angiography, CI: confidence interval, OR: odds ratio, ref.: reference, SCAPIS: Swedish CArdioPulmonary bioImage Study. Reference category for CCTA stenosis, CAC score and carotid plaques: no stenosis, CAC score=0 Agatston units and no plaque, respectively. 1 Adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription and knee extension strength. 2 Adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, knee extension strength, and quadratic terms for age at conscription, age at SCAPIS, BMI at conscription, years of smoking at conscription and knee extension strength. 3 Adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, knee extension strength, and cubic terms for age at conscription, age at SCAPIS, BMI at conscription, years of smoking at conscription and knee extension strength. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Table 19. Multinomial logistic regression, associations of knee extension strength in adolescence and coronary stenosis, CAC scores and carotid plaques in middle-age, sensitivity analysis considering quadratic and cubic terms in quantitative covariates. Main model1 Quadratic2 Cubic3 Knee extension strength OR (95% CI) P OR (95% CI) P OR (95% CI) P CCTA coronary stenosis 1-49% Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.94 (0.83-1.06) 0.324 0.90 (0.78-1.03) 0.131 0.91 (0.80-1.03) 0.135 Highest tertile 0.95 (0.84-1.08) 0.443 0.85 (0.69-1.06) 0.146 0.86 (0.71-1.03) 0.107 ≥50% Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.89 (0.72-1.10) 0.284 0.83 (0.65-1.07) 0.153 0.84 (0.67-1.06) 0.148 Highest tertile 0.74 (0.58-0.93) 0.012 0.64 (0.42-0.97) 0.035 0.65 (0.45-0.94) 0.020 CAC score 1-99 Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 1.02 (0.89-1.15) 0.797 0.97 (0.84-1.12) 0.681 0.97 (0.85-1.12) 0.708 Highest tertile 0.95 (0.83-1.09) 0.502 0.84 (0.67-1.06) 0.141 0.83 (0.68-1.02) 0.075 ≥100 Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.81 (0.68-0.96) 0.016 0.75 (0.62-0.92) 0.005 0.77 (0.64-0.92) 0.005 Highest tertile 0.84 (0.70-1.01) 0.060 0.72 (0.53-0.98) 0.036 0.74 (0.57-0.97) 0.032 Ultrasound carotid plaque Unilateral Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.89 (0.79-1.02) 0.086 0.89 (0.77-1.03) 0.110 0.89 (0.78-1.02) 0.103 Highest tertile 0.93 (0.81-1.06) 0.268 0.93 (0.74-1.17) 0.528 0.95 (0.78-1.16) 0.603 Bilateral Lowest tertile, ref. 1.00 - - 1.00 - - 1.00 - - Medium tertile 0.97 (0.85-1.11) 0.687 1.02 (0.87-1.18) 0.851 1.00 (0.87-1.15) 0.995 Highest tertile 1.02 (0.89-1.18) 0.748 1.15 (0.90-1.47) 0.256 1.13 (0.91-1.40) 0.257 BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomographic angiography, CI: confidence interval, OR: odds ratio, ref.: reference, SCAPIS: Swedish CArdioPulmonary bioImage Study. Reference category for CCTA stenosis, CAC score and carotid plaques: no stenosis, CAC score=0 Agatston units and no plaque, respectively. 1 Adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription and cardiorespiratory fitness. 2 Adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, cardiorespiratory fitness, and quadratic terms for age at conscription, age at SCAPIS, BMI at conscription, years of smoking at conscription and cardiorespiratory fitness. 3 Adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, conscription year, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, cardiorespiratory fitness, and cubic terms for age at conscription, age at SCAPIS, BMI at conscription, years of smoking at conscription and cardiorespiratory fitness. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Figure 1. Flow chart of the study. BMI: body mass index, CAC: coronary artery calcium, CCTA: coronary computed tomography angiography, SCAPIS: Swedish CArdioPulmonary bioImage Study. en in CA ro ect n 1 A olescent men in CA an conscri tion atabases n 10 0 art c pants not n conscr pt on 2584 art c pants 20 years old at conscr pt on 1260 art c pants w th m ss ng data n co ar ates at conscr pt on 5 mo ng at conscr pt on 331 Educat onal status at A 258 A olescent men in CA an conscri tion atabases it e os res n 1 A olescent men in CA an conscri tion atabases it e os res an co ariates n art c pants w th m ss ng data n exposures ard oresp ratory f tness 1429 trength ar a les 72 Caroti la e ata n art c pants w th m ss ng data n any four prox mal rele ant coronary segments 980 CC A stenosis ata n 00 CAC score ata n BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á Supplementary Figure 2. Directed acyclic graph for the association between cardiorespiratory fitness and muscular strength and atherosclerosis outcomes. BMI: body mass index, CRF: cardiorespiratory fitness, SCAPIS: Swedish CArdioPulmonary bioImage Study, SES: socioeconomic status. Exposure, Outcome, Ancestor of exposure, Ancestor of outcome, Ancestor of exposure and outcome, Causal path, Biasing path. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á All models depict multinomial regression models adjusted for age at conscription, age at SCAPIS, site in conscription, site in SCAPIS, trend year at conscription, educational status at SCAPIS, BMI at conscription, years of smoking at conscription, and cardiorespiratory fitness. BMI: body mass index, CAC: coronary artery calcium, OR: odds ratio, SCAPIS: Swedish CArdioPulmonary bioImage Study. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Br J Sports Med doi: 10.1136/bjsports-2023-107663–11.:10 2024;Br J Sports Med, et al. Herraiz-Adillo Á