Robert G. Kowalski
A conversation with Robert G. Kowalski, MD, MS, Principal Investigator, Craig Hospital, and Assistant Clinical Professor of Neurology and PM&R, University of Colorado School of Medicine, about stroke following traumatic brain injury.
Interviewed by José G. Merino, MD, Associate Professor of Neurology, University of Maryland School of Medicine.
They will be discussing the paper, “Acute Ischemic Stroke After Moderate to Severe Traumatic Brain Injury: Incidence and Impact on Outcome,” published in the July issue of Stroke.
Dr. Merino: Thank you for agreeing to the interview. Can you first briefly describe the methods and main findings of the analysis published in Stroke?
Dr. Kowalski: The study was a research collaboration between the Centers for Disease Control and Prevention (CDC) and the Traumatic Brain Injury Model Systems (TBIMS) program. It was led by researchers at Craig Hospital in Englewood, CO. Investigators studied more than 6,400 traumatic brain injury (TBI) patients over a 7.5-year period to evaluate risk factors for onset, incidence, and predictors of outcome in ischemic stroke occurring acutely after TBI. We found that 2.5% of individuals who experience a moderate to severe TBI also suffer an acute ischemic stroke (AIS) at the time of the injury. In half of these cases, the individuals experiencing stroke concurrent with brain trauma were age 40 or younger. Additionally, the study found the risk of acute ischemic stroke immediately following traumatic brain injury was 10 times the risk of ischemic stroke in the general population.
Individuals with AIS were more likely to have early radiographic evidence of intracranial mass effect and were more likely to have a carotid or vertebral artery dissection. High-velocity accidents were more common as the cause of TBI with cervical dissection. In multivariable analyses controlling for age, sex, initial clinical status, and neuroanatomic injury characteristics, AIS predicted poorer outcome by all measures assessed, including functional outcome and duration of post-traumatic amnesia.
This research project team was comprised of investigators with expertise in the treatment, outcome, and epidemiology of ischemic stroke and traumatic brain injury. These included Juliet Haarbauer-Krupa, PhD, a Senior Health Scientist in the Division of Unintentional Injury Prevention, Centers for Disease Control and Prevention; Michel Torbey, MD, a stroke neurologist and current President of the Neurocritical Care Society; John D. Corrigan, PhD, Professor, Department of Physical Medicine and Rehabilitation at Ohio State University, and Chairman of the Executive Committee of the TBI Model Systems Project; and Gale Whiteneck, PhD, a Principal Investigator at Craig Hospital who previously served for more than 30 years as the Director of the hospital’s Research Department.
Dr. Merino: In your patients, what was the timing of the stroke in relation to TBI?
Dr. Kowalski: All episodes of ischemic stroke were diagnosed during the acute phase of care for the TBI survivors included in the study.
Dr. Merino: How was stroke diagnosed in your study? Did all patients have a comprehensive stroke evaluation to determine the etiology?
Dr. Kowalski: Ischemic stroke was identified in the study data sample by ICD-9 code (International Classification of Diseases, Ninth Revision, Clinical Modification codes 433.01 to 433.91 and 434.01 to 434.91). The reliability of this coding was assessed by a review of medical records and radiology reports for all TBI patients in the sample with acute ischemic stroke from the center with the largest number of patients with concurrent TBI and AIS. This sensitivity analysis found 100% of these patients had new-onset ischemic stroke. The extent to which individuals with concurrent TBI and ischemic stroke in the overall study sample underwent a comprehensive stroke evaluation during acute care is unknown, and is a limitation of the study.
Dr. Merino: You found that patients with TBI and stroke most often had an arterial dissection. Do you have information on possible mechanisms for the dissection (type of injury, velocity, etc.)?
Dr. Kowalski: Of those patients with concurrent TBI and ischemic stroke, 6% had carotid artery dissection, 4% had vertebral artery dissection, and 9% had carotid or vertebral artery dissection. Patients with dissection were more likely to have a higher velocity (motorcycle, automobile, and other motor vehicle collision) cause of traumatic injury (64% higher velocity cause in dissection vs. 42% higher velocity cause with no dissection, P<0.001; OR, 2.491; 95% CI, 1.671–3.712).
Dr. Merino: What are the clinical implications of your findings? What are the barriers for acute stroke interventions (or preventive therapies) in the setting of acute TBI?
Dr. Kowalski: The findings of this investigation, we believe, are important for patients in the United States and elsewhere in the world who present initially with brain trauma but quickly develop ischemic stroke that may not immediately be recognized. This is particularly true in younger patients, for whom a stroke suspicion may be low in the emergency department setting, but for whom a stroke may cause lifelong deficits. The findings of this research, we believe, highlight the importance of early identification of ischemic stroke following TBI, given the narrow available therapeutic window, and the consequences of unrecognized stroke.
Thrombolysis with tissue plasminogen activator is contraindicated for severe TBI under 2013 guidelines of the American Heart Association and American Stroke Association. However, those guidelines note that the recommendation is based on sparse clinical evidence. Furthermore, the recommendations specify that a physician with expertise in treatment of acute stroke may modify the contraindication list. Enactment of acute stroke protocols after TBI, with rapid magnetic resonance imaging (MRI) acquisition and other specialized neuroimaging, including angiography, to better identify AIS and dissection, may be warranted when clinical evidence suggests onset of acute ischemic stroke. Similarly, antiplatelet or anticoagulation therapy may be an appropriate consideration if dissection is observed in these patients.
Dr. Merino: Can you tell us more about the Traumatic Brain Injury Model Systems (TBIMS) National Database (NDB). How is it funded? What hospitals participate? What are the criteria for entry?
Dr. Kowalski: The TBIMS NDB is a longitudinal study of outcomes following moderate to severe traumatic brain injury, which currently includes data on more than 15,000 TBI survivors with outcomes recorded as long as 25 years after injury for some subjects. The TBIMS NDB currently is funded by the U.S. Department of Health and Human Services, National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR). The study included patients who were diagnosed with TBI and AIS at acute care hospitals and eventually admitted for treatment at 22 inpatient rehabilitation centers in the U.S., with a disperse geographical distribution. Moderate to severe TBI is defined for the TBIMS NDB as brain tissue damage caused by an external mechanical force. Inclusion criteria include age ≥ 16 years at injury, presentation to a Model System designated acute hospital within 72 hours of injury, and at least one of the following: post-traumatic amnesia (PTA) >24 hours, loss of consciousness >30 minutes, a Glasgow Coma Scale (GCS) Total score <13 on initial presentation to a health care facility, or trauma-related intracranial abnormalities identified on neuroimaging.
Dr. Merino: Do you have advice for early career physicians interested in doing research using large databases such as the TBIMS NDB (training, resources, etc.)?
Dr. Kowalski: The TBIMS NDB is available for analysis by researchers. A protocol for use of the database, with a process for notification of research intent, is described here: https://www.tbindsc.org/Researchers.aspx
The TBIMS NDB includes a rich, multi-year, multi-center set of data that has been assembled for survivors of moderate to severe TBI. This and other similar large databases provide a significant opportunity for early career investigators to explore novel hypotheses related to patient and injury characteristics, as well as outcomes, for this injury type. These databases also may provide pilot data for grant funding applications for expanded research proposals, for both early career and veteran research investigators.
Dr. Merino: Can you identify specific areas where TBI and stroke research converge?
Dr. Kowalski: This is an insightful question. In clinical practice, head trauma, particularly when accompanied by polytrauma, is often treated primarily by trauma teams in the emergency department and acute phases of hospital care. Regarding stroke, significant advances have been made in acute care of ischemic stroke since the advent of tPA for this indication. The efficacy of urgency in such care has been potentially strengthened with recent encouraging results of endovascular thrombectomy trials in ischemic stroke. Findings that TBI creates an increased risk of ischemic stroke, particularly in younger individuals, may lend impetus to greater involvement of neurology specialties in acute clinical care and research related to TBI.
Dr. Merino: Is there anything else I should have asked you about your study or about the topic and the field in general?
Dr. Kowalski: There has been a recent observation of an increase in ischemic stroke incidence in a younger population. While the mechanism of this change is unclear, it has been attributed partly to lifestyle factors such as lack of physical activity and poor diet. This research on concurrent TBI and acute ischemic stroke may suggest an additional explanation for some of this trend. This study’s findings, we think, suggest that vigilance is warranted in the emergency department setting when patients present with traumatic brain injury and symptoms suspicious for concurrent stroke.
