Short communication Transient suppression of heart rate complexity in concussed athletes Michael F. La Fountaine a, ⁎, Kevin S. Heffernan c , James D. Gossett e , William A. Bauman a,d , Ronald E. De Meersman b a Center of Excellence on the Medical Consequences of Spinal Cord Injury, James J. Peters VA Medical Center, Bronx, NY, USA b Department of Biobehavioral Sciences, Teachers College, Columbia University, New York, NY, USA c Department of Medicine — Division of Cardiology and the Molecular Cardiology Research Institute, Tufts Medical Center, Boston, MA, USA d Department of Medicine and Rehabilitation Medicine, Mount Sinai School of Medicine, New York, NY, USA e Department of Sports Medicine, Columbia University, New York, NY, USA abstract article info Article history: Received 13 January 2009 Received in revised form 3 March 2009 Accepted 4 March 2009 Keywords: Mild traumatic brain injury Concussion Autonomic nervous system Approximate entropy Heart rate complexity Heart rate variability (HRV) and complexity (HRC) were calculated at rest and during an isometric hand grip test (IHGT) within 48-hours (48 h) and two weeks (Week Two) of a concussion in athletes (CG) and control subjects. No differences were present at rest or in HRV during IGHT. HRC was significantly lower in the CG compared to controls at 48 h during IHGT. In CG at Week Two during IHGT, HRC was significantly greater than 48 h observations and not significantly different than controls. The findings suggest that HRC may have utility in detecting efferent cardiac autonomic anomalies within two weeks of concussion. Published by Elsevier B.V. The calculation of heart rate variability (HRV), a measure of the periodicity around a mean heart rate due to continuous alterations of sympathovagal balance (Van Ravenswaaij-Arts et al., 1993), has proven useful in quantifying autonomic dysfunction across a broad spectrum of traumatic brain injury (TBI) severity, including concus- sion (King et al., 1997, Goldstein et al., 1998, Gall et al., 2004). Autonomic dysfunction after TBI is the purported result of an uncoupling between the autonomic centers and cardiovascular system (Goldstein et al., 1998). Following head trauma, HRV has been observed to be significantly lower than non-injured control subjects and inversely associated with long-term outcomes (Goldstein et al., 1998, Meglic et al., 2001, Keren et al., 2005). More importantly, a trend toward normalization of HRV was observed during the first 3 months post-injury, suggesting a recovery process (Meglic et al., 2001, Keren et al., 2005). Use of an exercise provocation in a recent report of concussed athletes and matched teammates, unmasked significant decreases in HRV upon repeated evaluation that were not present at rest (Gall et al., 2004). The measure of HRV employed in these studies may not capture the full breadth of the disorder due to the linear nature of the calculation. HRV is calculated from and depends upon the harmonic periodicity in a signal (Kuusela et al., 2002). Biological signals possess a degree of randomness and irregularity that may not be captured by the linear methods employed (Godin and Buchman, 1996, Kuusela et al., 2002). To account for the irregularity in a dynamic signal, or complexity, non-linear methods quantify the uncertainty of re-occurring patterns within a time series event (Lipsitz, 1995). One such calculation, approximate entropy (ApEn), a unitless measure of HRC, quantifies the likelihood of regularity (similar and predictable patterns) in the signal with more regularity yielding smaller values and less regularity yielding larger values representing a more complex structure (Pincus, 1995). A recent study demonstrated that healthy, young adults (mean age 29 years) have a daytime ApEn value of 0.99 (±0.11) (Cysarz et al., 2007). The objective of this investigation was to determine whether HRC, as calculated by ApEn, varied in persons with a recent concussion at rest and during an isometric hand grip test (IHGT) when compared to control subjects within 48-hours and two weeks following a concussive injury. In addition, linear analysis methods of HRV were calculated in parallel for comparative purposes. Three athletes with concussion and three matched control subjects participated in this study. An athlete suspected of sustaining a head injury was identified by the sports medicine staff and removed from athletic activities for further evaluation. Clinical evaluation of the injury and in the determination of return to play included a review of symptoms and completion of a commercially available computer- based neurocognitive test battery which was compared to baseline pre-season data. Control subjects were recruited from in-season athletes of similar age, gender, height, weight, and when possible, sport and position (assumption of similar fitness). The study Autonomic Neuroscience: Basic and Clinical 148 (2009) 101–103 ⁎ Corresponding author. Center of Excellence: Medical Consequences of Spinal Cord Injury, James J. Peters VA Medical Center; Room 1E-02, 130 West Kingsbridge Road, Bronx, NY 10468, USA. Tel.: +1718 584 9000x3121; fax: +1718 741 4675. E-mail address: Michael.LaFountaine@va.gov (M.F. La Fountaine). 1566-0702/$ – see front matter. Published by Elsevier B.V. doi:10.1016/j.autneu.2009.03.001 Contents lists available at ScienceDirect Autonomic Neuroscience: Basic and Clinical journal homepage: www.elsevier.com/locate/autneu