October 2, 2023

JAMA Netw Open. 2023 Aug 1;6(8):e2328633. doi: 10.1001/jamanetworkopen.2023.28633.


IMPORTANCE: Concussions are common in sports. Return-to-play protocols can be enhanced by objective biometrics.

OBJECTIVE: To characterize temporal changes of headpulse, a digital biometric, in athletes with sports-related concussion; to explore the association of unstructured physical activity with headpulse changes.

DESIGN, SETTING, AND PARTICIPANTS: This cohort study included headpulse measurements from players in the highest level of amateur Australian Rules Football in South Australia. Analysis included feasibility and validation phases, with the feasibility cohort recruited between August 5, 2021, and September 10, 2021, and the validation cohort recruited between May 5, 2022, and September 3, 2022. Data were analyzed October 2022 through January 2023.

INTERVENTIONS: Cranial accelerometry detected micromovements of the head following cardiac contraction (what we have described as “headpulse”). Headpulse was serially recorded for 1 month in concussed individuals.

MAIN OUTCOMES AND MEASURES: Headpulse waveforms underwent frequency transformation analysis per prespecified algorithm. Result Z scores were calculated. Headpulse Z scores exceeding 2 (2 SDs from control means) met an abnormality threshold. Headpulse sensitivity, timing, and duration of change were determined.

RESULTS: A total of 59 control and 43 concussed individuals (44 total concussions; 1 control also concussed, 1 concussed individual injured twice) provided headpulse measurements. The feasibility cohort (all male) included 17 control (median [IQR] age, 23 [19-28] years) and 15 concussed individuals (median [IQR] age, 21 [19-23] years). The validation cohort included 25 female (median [IQR] age, 21 [20-22] years) and 17 male (median [IQR] age, 26 [23-29] years) control individuals, and 8 female (median [IQR] age, 28 [20-31] years) and 20 male (median [IQR] age, 21 [19-23] years) concussed individuals. Headpulse reached abnormality threshold in 26 of 32 concussed individuals (81%; 9% on day 0, 50% by day 2, 90% by day 14). Headpulse alterations lasted 14 days longer than symptoms and were exacerbated by return-to-play or unsupervised physical activity.

CONCLUSIONS AND RELEVANCE: In this study of 101 amateur Australian Rules Football athletes, the digital headpulse biometric was evaluated in 44 sports-related concussions. Compared with controls, new headpulse changes occurred after concussion; this objective metric may complement return-to-play protocols.

PMID:37566413 | PMC:PMC10422194 | DOI:10.1001/jamanetworkopen.2023.28633