Association Between Traumatic Brain Injury and Increased Risk of Stroke: A Systematic Review and Meta-analysis.
Dmitry Esterov, Michael A Sperl, Emily A Hines and 2 others
PMID 36594863WHAT IT FOUND
Adults who had a traumatic brain injury were about twice as likely to have a stroke as adults who did not.
All eight studies were retrospective; the authors rate the overall evidence moderate, but low for mild injury and for risk beyond a year.
Key findings
01Stroke was significantly more common in adults who had a traumatic brain injury than in adults who did not. The primary pooled estimate was a hazard ratio of 2.06 (95% CI, 1.28-3.32), and pooling all six studies that reported hazard ratios gave 2.09 (95% CI, 1.54-2.82).
02Where studies reported time points, the association was strongest soon after injury and smaller later: one study reported hazard ratios of 10.21 (95% CI, 8.71-11.96) at 3 months, 4.61 (95% CI, 4.16-5.11) at 1 year and 2.32 (95% CI, 2.17-2.47) at 5 years, and the authors note that ICD-9 coding can misclassify bleeding from the injury itself as a haemorrhagic stroke.
03The authors rate the evidence for an overall association as moderate, but low for the subgroups of stroke more than a year after injury, mild traumatic brain injury and stroke, and haemorrhagic stroke.
STILL TO COME
How it was doneWhat they foundWhat it means for PTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Five of the six studies reporting hazard ratios used the same Taiwanese national insurance database, so the samples were not independent; the authors state that pooling non-independent samples can narrow confidence intervals artificially and produce misleadingly significant results. Every included study was retrospective and identified both the brain injury and the stroke from ICD-9 codes. The authors note these codes can mistake bleeding from the original injury for a haemorrhagic stroke, which may explain the unusually high estimates for stroke in the first weeks after injury. Statistical heterogeneity was extreme (I2 = 100%), and subgroup analyses by stroke type, injury severity and time since injury did not resolve it. Studies without a comparison group were excluded, so much of the work on stroke in the weeks after hospitalisation for moderate-severe traumatic brain injury was not reviewed systematically. Because the studies used different age ranges and different definitions of injury severity, analyses by age and comparisons between severity groups could not be made reliably. No study independently reported short-term versus long-term risk, so the timing of the risk increase rests on a few studies reporting separate time points.
Declared interests
The authors declare that they have no conflicts of interest to disclose. The article is indexed as receiving extramural NIH research support; no other funder is named in the text.
The easy way to misread this
Do not read the very high early risk figures, such as the tenfold increase in stroke in the first three months after injury reported in one study, as a settled effect of head injury. The authors point out that ICD-9 codes can misclassify bleeding caused by the injury itself as a haemorrhagic stroke, and most of the pooled data came from a single country's insurance database. This paper reports an association only. It tested no treatment, so it cannot tell you that anything done in rehabilitation lowers stroke risk.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →