Retired National Football League (NFL) linemen have an increased prevalence of risk factors for atherosclerosis and have an increased risk of cardiovascular death compared with nonlinemen and the general population. We evaluated whether playing in lineman position is independently associated with an increased risk of the presence and severity of subclinical atherosclerosis. Players were categorized as linemen if they reported playing on the offensive or defensive line during their careers. Subclinical atherosclerosis was assessed using coronary artery calcium (CAC) scores in 931 retired NFL players (310 linemen, 621 nonlinemen). CAC scores were evaluated for absence of subclinical atherosclerosis (CAC = 0), presence of mild subclinical atherosclerosis (CAC 1 to 100), and moderate to severe subclinical atherosclerosis (CAC ≥100). We performed multivariate logistic regression to determine whether the lineman position is independently associated with the presence and severity of subclinical atherosclerosis. Linemen were noted to have a lesser likelihood of absence of subclinical atherosclerosis (CAC = 0, 33.8% vs 41.7%, p = 0.02), a similar likelihood of mild subclinical atherosclerosis (CAC 1 to 100, 33.2% vs 31.8%, p = 0.7), and a greater likelihood of moderate to severe subclinical atherosclerosis (CAC >100, 32.9% vs 26.4%, p = 0.04) compared with nonlinemen. Adjusting for demographic and metabolic covariates, lineman status remained independently associated with mild subclinical atherosclerosis (CAC 1 to 100, odds ratio [OR] 1.41, 95% confidence interval [CI] 1.05 to 2.2, p = 0.04) and moderate to severe subclinical atherosclerosis (CAC ≥100, OR 1.67, 95% CI 1.05 to 2.2). The association was attenuated after adjustment for race (CAC 1 to 100, OR 1.24, 95% CI 0.82 to 1.8; CAC >100, OR 1.59, 95% CI 1.01 to 2.49). In conclusion, lineman status in retired NFL players is associated with presence and severity of subclinical atherosclerosis, which is partly explained by race.
Highlights
- 1
Retired National Football League players playing in lineman position have an increased risk of cardiovascular death compared with those playing in the nonlinemen position.
- 2
Presence and severity of subclinical atherosclerosis was assessed by coronary artery calcium scores, using a modified Agatston protocol.
- 3
Retired linemen had an independent association with development of moderate to severe subclinical atherosclerosis after adjustment for risk factors and race.
- 4
The association between linemen playing position and risk of development of mild subclinical atherosclerosis was attenuated by race.
Retired National Football League (NFL) players who played in lineman position have an increased prevalence of metabolic syndrome, a threefold increased risk of dying from heart disease compared with nonlinemen, and a 52% increased risk of cardiovascular death compared with the general population. A report comparing mortality rates among 3,850 professional football players with 2,403 professional baseball players found that football players were more than twice as likely to die before age 50 compared with baseball players. Moreover, 50% of these obese football players died before age 50 compared with 13% of those who were not obese. Risk factors for cardiovascular mortality including hypertension, sleep-disordered breathing, and metabolic syndrome are more prevalent in linemen than in nonlinemen. We hypothesize that retired football players playing in linemen position have higher prevalence of subclinical atherosclerosis evaluated by coronary artery calcium (CAC) scores compared with those playing in nonlinemen position. Such a finding will reaffirm previous data that retired linemen have indeed higher cardiovascular risk because CAC has been shown to be significantly associated with adverse coronary outcomes. We therefore evaluated the presence and severity of subclinical atherosclerosis in retired NFL players and evaluated whether playing in lineman position is independently associated with an increased risk of subclinical atherosclerosis as assessed by CAC scores.
Methods
Our study population consisted of 931 former NFL players (310 linemen, 621 nonlinemen) who participated in the health screening events organized by the Living Heart Foundation and the Boone Heart Institute, conducted from September 2007 to November 2009. Retired NFL players who played in the tackle, guard, center, or defensive-end positions were classified as linemen; retirees who played in the quarterback, running back, wide receiver, tight end, linebacker, cornerback, safety, kicker, or punter position were classified as nonlinemen. Details of the methods have been described previously. Briefly, retired NFL players were invited through mail or local NFL alumni chapters meetings and subsequently underwent a comprehensive cardiovascular screening examination. Key data elements recorded for the study included height, weight, waist and hip circumference, and cuff blood pressure (using CardioVision automated cuff system). We recorded 3 consecutive seated blood-pressure readings for each retired NFL player; the first reading was discarded and the mean of second and third readings was used. Body mass index (BMI) was defined as weight (in kilograms) divided by square of height (in meters). Waist circumference was measured at the top of the umbilicus in inches. Hypertension was defined as a self-reported history of hypertension or the use of antihypertensive medication. Diabetes mellitus was defined as a self-reported history of diabetes or the use of diabetes medication. Positive smoking status was defined as current or previous smoking. Metabolic syndrome was defined by the National Heart, Lung, and Blood Institute/American Heart Association guidelines as presence of ≥3 of the following criteria: waist circumference >102 cm in men, high-density lipoprotein cholesterol <40 mg/dl in men, serum triglycerides ≥150 mg/dl, blood pressure ≥130/85 mm Hg, and fasting glucose ≥110 mg/dl.
A fasting venous blood sample was collected in tubes containing ethylenediaminetetraacetic acid for plasma and in serum separator tubes for serum. The samples underwent centrifugation at the sites where blood was collected, and the samples were then shipped to Quest laboratories the same day. Measurements were performed for total cholesterol, high-density lipoprotein cholesterol, and triglycerides. Low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedewald equation. High sensitivity C-reactive protein (hs-CRP) was measured using commercially available assays.
We used CAC as a marker for the presence and severity of subclinical atherosclerosis. CAC has been shown to be associated with adverse coronary heart disease outcomes. CAC scoring was performed within 1 week of the screening visit (mostly on the same day of the screening visit), using a modified Agatston protocol with slice thickness varying between 2.5 and 3 mm, among various study sites. CAC scores were expressed in Agatston unit, which was calculated by multiplication of a hyperattenuating lesion above a threshold of 130 Hounsfield Units with an area of at least 1 mm 2 in each slice by a density factor. There was 1 designated reader for CAC scoring at most sites. CAC scores were evaluated as absence of subclinical atherosclerosis (CAC = 0), presence of mild subclinical atherosclerosis (CAC 1 to 100), and moderate to severe subclinical atherosclerosis (CAC >100).
Descriptive statistics were evaluated for each variable to assess its distribution. We performed the Wilcoxon rank sum test for continuous variables and Pearson’s chi-square test for categorical variables. We then performed multivariate logistic regression to determine whether lineman position is independently associated with the absence of subclinical atherosclerosis (CAC = 0), presence of mild subclinical atherosclerosis (CAC 1 to 100), and moderate to severe subclinical atherosclerosis (CAC >100) after adjusting for age, BMI, hs-CRP, LDL-C, diabetes mellitus, hypertension, and presence of metabolic syndrome (model I). Additionally, we also adjusted for race (in addition to variables adjusted in model I) to account for the reported differences in CAC scores with race.
All analyses were performed using IBM SPSS Version 21. All inferences were 2-tailed, and a p value <0.05 was considered statistically significant. The study was approved by the Institutional Review Board at the Baylor College of Medicine.
Results
A total of 931 participants were included in the analysis after excluding those without available CAC data (n = 92). Baseline characteristics of the cohort stratified by lineman playing position are described in Table 1 . Linemen had greater BMI, waist circumference, and a higher prevalence of hypertension, diabetes mellitus, metabolic syndrome, and higher levels of hs-CRP compared with nonlinemen. Linemen had lower levels of total cholesterol, high-density lipoprotein cholesterol, and LDL-C with no differences in age, systolic blood pressure, fasting glucose, and triglycerides compared with nonlinemen.
| Linemen (n=310) | Non-linemen (n=621) | p value | |
|---|---|---|---|
| Age (years), median (IQR) | 54 (44-63) | 54 (45-62) | 0.59 |
| Coronary Artery Calcium score, median (IQR) | 14.2 (0-214.5) | 5.9 (0-121) | 0.04 |
| Years since retirement, median (IQR) | 24 (14-35) | 25 (18-33) | 0.40 |
| Body mass index (Kg/m 2 ), median (IQR) | 33.6 (30.5-37.9)) | 30.3 (27.7-33)) | <0.001 |
| African American race | 32.8% | 45.9% | 0.001 |
| Hypertension | 38.8% | 28.5% | 0.003 |
| Diabetes Mellitus | 10.4% | 5.5% | 0.007 |
| Waist circumference (cm), median (IQR) | 109.2 (99.1-119.4) | 99.1(91.4-106.6)) | <0.001 |
| Systolic blood pressure (mmHg), median (IQR) | 131 (122-144) | 130 (120-143) | 0.54 |
| High sensitivity-C Reactive Protein (mg/L), median (IQR) | 1.5 (0.7-3.7) | 1.1(0.5-2.2) | 0.01 |
| Total cholesterol (mg/dL), median (IQR) | 190 (167.5-214) | 198 (173-227) | 0.02 |
| High Density Lipoprotein cholesterol (mg/dL), median (IQR) | 45 (39-55.8) | 48 (40-57) | 0.006 |
| Low Density Lipoprotein cholesterol (mg/dL), median (IQR) | 117.5 (98-143) | 127 (104-151.8) | 0.001 |
| Triglycerides (mg/dL), median (IQR) | 88 (62-141) | 91 (66-140.5) | 0.86 |
| Fasting glucose (mg/dL), median (IQR) | 92 (84-100) | 90 (83-97.3) | 0.19 |
| Metabolic syndrome | 25.8% | 16.5% | <0.001 |
Linemen had higher median CAC scores compared with nonlinemen ( Table 1 ). Linemen were also noted to have a lesser likelihood of absence of subclinical atherosclerosis, a similar likelihood of mild subclinical atherosclerosis, and a greater likelihood of moderate to severe subclinical atherosclerosis compared with nonlinemen ( Table 2 ).
| Coronary Artery Calcium Score | Linemen (N=310) | Non-linemen (N=621) | p value |
|---|---|---|---|
| CAC = 0 (N=364) | 105 (33.88%) | 259 (41.7%) | 0.02 |
| CAC 1- 100 (N=301) | 103 (33.22%) | 198 (31.88%) | 0.71 |
| CAC > 100 (N=266) | 102 (32.90%) | 164 (26.41%) | 0.04 |
In unadjusted models ( Table 3 ), lineman playing position was associated with 12% lesser odds of absence of subclinical atherosclerosis, 25% greater odds of presence of mild subclinical atherosclerosis, and 36% greater odds for the presence of moderate to severe subclinical atherosclerosis. After adjusting for age, BMI, hs-CRP, LDL-C, diabetes mellitus, hypertension, and presence of metabolic syndrome in a multivariate model, lineman playing position was associated with 41% greater odds of presence of mild subclinical atherosclerosis. However, after adjusting for race in addition to the previously mentioned covariates, the association between lineman playing position and presence of mild subclinical atherosclerosis was greatly attenuated and became nonsignificant.
| Coronary Artery Calcium Score | Unadjusted OR (95% CI), p | (Model 1) ∗ OR (95% CI), p | Model 2 OR (95% CI), p |
|---|---|---|---|
| CAC = 0 (N=364) | 0.88 (0.79-0.97), p= 0.02 | 0.63 (0.43-0.93), p=0.02 | 0.79 (0.53-1.19), p=0.25 |
| CAC 1 – 100 (N=301) | 1.25 (1.04-1.72), p=0.04 | 1.41 (1.05-2.2), p=0.04 | 1.24 (0.82-1.8), p=0.28 |
| CAC > 100 (N=266) | 1.36 (1.06-1.83), p=0.03 | 1.67 (1.10-2.51), p=0.03 | 1.59 (1.01-2.49), p=0.04 |
∗ Adjusted for age, body mass index, hs-CRP, LDL cholesterol, diabetes mellitus, hypertension, and metabolic syndrome (Model 1).
Importantly, lineman playing position was associated with a 67% increased odds of the presence of moderate to severe subclinical atherosclerosis after adjusting for age, BMI, hs-CRP, LDL-C, diabetes mellitus, hypertension, and metabolic syndrome ( Table 3 ). After further adjustment for race, lineman position remained independently associated with 59% greater odds of severe subclinical atherosclerosis. In a predictive model, we found that age at screening and Caucasian race were also significantly associated with mild subclinical atherosclerosis and moderate to severe subclinical atherosclerosis ( Tables 4 and 5 ).
