Importance: Chronic traumatic encephalopathy (CTE) is a neurodegenerative tauopathy associated with repetitive head impacts (RHIs). Prior research suggests a dose-response association between American football play duration and CTE risk and severity, but this association has not been studied for ice hockey.
Objective: To investigate associations of duration of ice hockey play with CTE diagnosis and severity, functional status, and dementia.
Design, setting, and participants: This cross-sectional study was conducted among male brain donors in the Understanding Neurological Injury and Traumatic Encephalopathy and Framingham Heart Study Brain Banks whose primary RHI exposure was from ice hockey. Donors died, brains were donated, and data were collected between July 1997 and January 2023. Data analysis was conducted from January 2023 to May 2024.
Exposures: Ice hockey years played as an RHI proxy.
Main outcomes and measures: CTE neuropathological diagnosis, cumulative phosphorylated tau (ptau) burden across 11 brain regions commonly affected in CTE, informant-reported Functional Activities Questionnaire (FAQ) score at death, and consensus dementia diagnosis were assessed.
Results: Among 77 male donors (median [IQR] age, 51 [33-73] years), 42 individuals (54.5%) had CTE, including 27 of 28 professional players (96.4%). CTE was found in 5 of 26 donors (19.2%) who played fewer than 13 years, 14 of 27 donors (51.9%) who played 13 to 23 years, and 23 of 24 donors (95.8%) who played more than 23 years of hockey. Increased years played was associated with increased odds for CTE (odds ratio [OR] per 1-year increase, 1.34; 95% CI, 1.15-1.55; P < .001) and with increased ptau burden (SD increase per 1-year increase = 0.037; 95% CI, 0.017-0.057; P < .001) after adjusting for age at death, other contact sports played, age of first hockey exposure, concussion count, and hockey position. Simulation demonstrated that years played remained associated with CTE when years played and CTE were both associated with brain bank selection across widely ranging scenarios (median [full range] OR across all simulations, 1.34 [1.29-1.40]). Increased ptau burden was associated with FAQ score (βstandardized = 0.045; 95% CI, 0.021-0.070; P < .001) and dementia (OR per SD increase, 1.12; 95% CI, 1.01-1.26; P = .04) after adjusting for age at death, other contact sports played, hockey years played, enforcer status, age of first hockey exposure, concussion count, and hockey position.
Conclusions and relevance: In this study of male former ice hockey players, a dose-response association was observed between hockey years played and risk and severity of CTE. Simulation suggested that brain bank selection may not bias the magnitude of outcomes in the association.