Coronary Atherosclerosis in Former NFL Players

Abstract

Former professional athletes may be at paradoxically high cardiovascular risk. Retired National Football League players in particular are known to carry a high burden of atherosclerosis risk factors and to be at elevated risk of acute cardiovascular events. We present 3 cases of asymptomatic former National Football League players, each unaware of the presence of coronary artery disease despite prior noninvasive evaluation, who underwent coronary computed tomography angiography coupled with advanced artificial intelligence–enabled analysis as part of an ongoing clinical registry. All 3 former players were found to have previously undiagnosed coronary artery plaque, prompting guideline-directed medical therapy. One patient had early-stage disease burden and 2 had extensive disease, including 1 patient with multiple severe stenoses accompanied by significant physiological impairment and fractional flow reserve reduction who underwent coronary artery bypass surgery. Early patient-specific characterization of coronary artery disease via coronary computed tomography angiography and artificial intelligence–enabled analyses may enable more prompt preventive management and result in clinical risk reduction in this population.

Visual Summary

Visual Summary.

Asymptomatic Former NFL Players Underwent CCTA With AI-Enabled Analysis and Were Enrolled in the GAMEFILM Registry

Our case series describes 3 of these participants. AI = artificial intelligence; CCTA = coronary computed tomography angiography; NFL = National Football League.

National Football League (NFL) players and other professional athletes exhibit exceptional cardiac performance during their playing careers, typically with high stroke volumes and exercise capacity as well as myocardial hypertrophy.¹ Paradoxically, former NFL players have an elevated risk of cardiovascular disease, including a high prevalence of hypertension, subclinical atherosclerosis, and abnormal cardiac imaging out of proportion to self-reported risk factor burden.^2,3 Former NFL players also die from myocardial infarction at a higher rate than the general population.⁴

Take-Home Messages

• •Former National Football League players are at high risk of cardiovascular disease.
• •Preclinical coronary artery disease occurs in this population.
• •CCTA imaging and AI-enabled characterization of coronary artery disease may afford early diagnosis and preventive management.

Recent clinical studies have identified possible contributors to this paradox, including weight gain, obstructive sleep apnea, and inflammatory factors. Master athletes accumulate coronary plaque despite high fitness, exhibiting early-onset coronary artery disease (CAD).^5,6 In addition, larger population-based studies show that former NFL players and other former players of American football who experience moderate-to-severe traumatic brain injury also have up to a 2-fold excess risk of incident hypertension, CAD, myocardial infarction, and stroke.⁷

Current screening programs available to former NFL players include programs sponsored by the NFL Players Association and the NFL Player Care Foundation (PCF) to provide baseline and periodic cardiovascular screening. These typically include coronary artery calcium scoring, carotid artery ultrasound, electrocardiography, echocardiography, lipid profiling and other blood tests, and blood pressure measurement. Together, these represent an effort to screen former NFL players with assessments that have historically been the mainstay of cardiovascular evaluations to provide them with information and tools related to heart health. However, these traditional diagnostic pathways often fail to detect or fully characterize CAD risk in line with current paradigms.⁸ Furthermore, recent advancements in coronary computed tomography angiography (CCTA) coupled with artificial intelligence (AI) assessment of disease burden and physiological impact now enable a deeper patient-specific approach to CAD risk assessment.

We describe a series of 3 cases (Table 1) representative of CAD found to date in the ongoing GAMEFILM (UsinG CTA MEasures to DeFine Cardiac Risk In NFL AluMni) registry (National Clinical Trial number: NCT06831409), which characterizes CAD in asymptomatic former NFL players using AI-enabled analysis (Heartflow Inc) after CCTA.

Table 1.

Patient Demographics, Cardiovascular Risk Factors, and CCTA and AI-Enabled Analysis Findings

	Patient 1	Patient 2	Patient 3
Demographics, history, and risk factors
Age, y	65	54	56
Position played	Wide receiver	Offensive tackle	Cornerback
Prior evaluation via NFL Brain and Body physical program	Yes	Yes	Yes
Prior history of CAD	No	No	No
Symptoms	Dyspnea	None	None
Lipid-lowering therapy pre-CCTA	No	Yes	No
Height, inches	75	77	72
Weight, lb	240	260	220
BMI	30.0	30.8	29.8
Diabetes	Yes	No	No
Hypertension	Yes	Yes	No
Cigarette smoking	Previous	Current	No
Family history of CAD	Yes	No	No
Hyperlipidemia	Yes	Yes	No
CCTA and Heartflow findings
Maximum stenosis (%)	70%-99%	70%-99%	30%-49%
Lowest lesion-specific FFRCT	0.65	0.93	0.95
No. of vessels with lesion-specific FFRCT ≤0.80	3	0	0
Total plaque volume, mm3	931	1,351	92
Noncalcified plaque volume, mm3	859	1,172	82
Calcified plaque volume, mm3	72	179	10
Total plaque volume (percentile for age and sexa)	91	99	31
Noncalcified plaque volume (percentile for age and sexa)	96	100	32
Calcified plaque volume (percentile for age and sexa)	58	92	35
Noncalcified: total plaque volume (%)	92%	87%	89%

AI = artificial intelligence; BMI = body mass index; CAD = coronary artery disease; CCTA = coronary computed tomography angiography; FFR_CT = fractional flow reserve derived from computed tomography; NFL = National Football League.

^aAccording to Tzimas et al.⁹

Patient 1

A 65-year-old former wide receiver who played for 2 years in the NFL and United States Football League had participated in the NFL PCF Brain and Body Screening Program previously but was unaware of having heart disease. He had diabetes, hypertension, prior smoking, hyperlipidemia, and a family history of CAD. His CCTA and Heartflow analysis (Figure 1) revealed a high total plaque burden (931 mm³; 91st percentile for men his age⁹; extensive stage¹⁰), of which 92.3% was noncalcified plaque (859 mm³; 96th percentile for men of his age⁹). He had obstructive (>70% stenosis) and physiologically flow-limiting (abnormal fractional flow reserve derived from computed tomography [FFR_CT]) CAD in all 3 vessel territories. The high proportion of noncalcified plaque highlighted a metabolically active, high-risk phenotype that would not have been fully characterized by coronary artery calcium scoring alone. In this context, the combined anatomic-physiologic data supported aggressive management including risk-factor modification and invasive evaluation. He was begun on guideline-directed medical therapy and underwent invasive coronary angiography (ICA) and then successful coronary artery bypass surgery.

Figure 1.

Patient 1: AI-Enabled Quantitative Coronary Plaque Analysis and FFR_CT

The patient has high total plaque volume and has flow-limiting stenoses in all 3 major coronary vessels. Total plaque summary is shown on the left. On the right, panels display the FFR_CT study and images of the left anterior descending coronary artery (top) and of the right coronary artery (bottom). The patient ultimately underwent coronary artery bypass graft surgery. AI = artificial intelligence; FFR_CT = fractional flow reserve derived from computed tomography.

Patient 2

A 54-year-old former offensive tackle who played for 13 years in the NFL also had participated previously in the NFL PCF Brain and Body Screening Program but was unaware of having heart disease. His risk factors included active smoking, hypertension, and hyperlipidemia. His CCTA and Heartflow analysis (Figure 2) revealed a high atherosclerotic burden, with a total plaque volume of 1,351 mm³ (99th percentile for men of his age⁹; extensive stage¹⁰) of which 86.8% was noncalcified (1,172 mm³; 100th percentile for men of his age⁹). Although his right coronary artery had a >70% stenosis, the lesion-specific FFR_CT of 0.93 indicated the stenosis to be not flow limiting. The extensive plaque in all 3 vessels and high proportion of noncalcified plaque reflected a metabolically active, high-risk plaque phenotype. These findings supported aggressive risk-factor modification including smoking cessation and intensification of lipid-lowering therapy, all of which were implemented. In addition, given the high-risk findings, ICA was recommended for confirmation of CAD extent. The patient declined ICA given his asymptomatic status, instead choosing close follow-up with a cardiologist. He was made aware of the need for prompt attention should symptoms develop.

Figure 2.

Patient 2: AI-Enabled Quantitative Coronary Plaque Analysis and FFR_CT

The patient has very high total plaque volume, but FFR_CT does not show evidence of flow-limiting stenoses. Total plaque summary is shown on the left. On the right, panels display the FFR_CT study and images of the left anterior descending coronary artery (top) and of the right coronary artery (bottom). Although invasive coronary angiography was recommended, the patient declined given his asymptomatic status. AI = artificial intelligence; FFR_CT = fractional flow reserve derived from computed tomography.

Patient 3

A 56-year-old former cornerback who played for 7 years in the NFL also had participated in the NFL PCF Brain and Body Screening Program previously and was asymptomatic. Although he did not have conventional risk factors for CAD, his CCTA and Heartflow analysis (Figure 3) revealed coronary artery plaque in his left main and left anterior descending coronary arteries, with maximum stenosis of 30% to 49% and total plaque volume of 92 mm³ (31st percentile for men his age⁹; mild stage¹⁰), with a large proportion (89%) being noncalcified plaque. His maximum coronary stenosis was 30% to 49% with preserved flow (FFR_CT: 0.95). After the evaluation, he was initiated on intensive lipid-lowering therapy.

Figure 3.

Patient 3: AI-Enabled Quantitative Coronary Plaque Analysis and FFR_CT

The patient has below-average total plaque burden but significant plaque in the left main and left anterior descending coronary arteries. FFR_CT reveals no flow-limiting stenoses. Total plaque summary is shown on the left. On the right, the panel displays the FFR_CT study and images of the left anterior descending coronary artery. The patient was treated with intensive lipid-lowering therapy. AI = artificial intelligence; FFR_CT = fractional flow reserve derived from computed tomography.

Discussion

Former NFL players typically have hypertension, weight cycling, chronic inflammation, and high rates of sleep apnea. Given the long asymptomatic phase of CAD despite substantial plaque burden, the cases presented demonstrate the utility of advanced imaging to guide preventive management. CCTA and AI-enabled analysis provide direct visualization and quantification of coronary pathology including stenosis severity, plaque quantity, location, composition, and flow impairment. As evidenced by the 3 cases described, this diagnostic pathway provides comprehensive, early, and actionable insight into CAD ranging from preventive medical management for nonobstructive disease to invasive management and revascularization when indicated for more advanced flow-limiting stenoses. All 3 of the patients had undergone prior noninvasive stress testing–based evaluations for CAD, all of which had been negative. This contrast underscores the utility of CCTA coupled with additional AI-enabled analyses for informing and guiding management by clinicians and promoting disease awareness in patients.

CCTA with AI-enabled analysis in high-risk asymptomatic populations such as retired NFL players can establish the diagnosis of CAD and prompt initiation of guideline-directed management.

Funding Support and Author Disclosures

Funding was received from Heartflow Inc (Mountain View, California). Dr Rogers and Mr Farquhar are employees of and receive salary and equity from Heartflow. Dr Maron reports research funding to his institution from the National Heart, Lung, and Blood Institute, Cleerly Inc, and Omada Health; advisory board membership with New Amsterdam Pharma and Bayer; equity in Ablative Solutions Inc, PreemptiveAI, and JVMP Labs; consultant fees from HeartFlow, Innomed Inc, Johnson & Johnson, Regeneron, and Scilex Holding Company; DSMB Astra Zeneca AZURE trials, and NHLBI PREEMPT trial. Dr Kovach reports consultant fees from Heartflow. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.