Cannabis Use Linked to Higher Blood Clot Risk in Older Trauma Patients
A large registry study finds THC-positive geriatric trauma patients had roughly double the rates of DVT and PE after statistical adjustment, but the observational design cannot establish causation, and critical gaps in exposure measurement limit the strength of clinical conclusions.
Why This Matters
Cannabis use among Americans aged 65 and older has risen sharply alongside expanding legalization, yet clinicians managing geriatric trauma have had little population-level data on whether THC exposure modifies thrombotic risk in this already vulnerable group. Preclinical evidence suggests cannabinoids may influence platelet aggregation and endothelial function, providing a plausible biological pathway worth investigating. This study is among the first to examine the question at scale using a national trauma registry, arriving at a moment when emergency and surgical teams increasingly encounter THC-positive elderly patients and need evidence to guide thromboprophylaxis decisions.
Clinical Summary
Geriatric trauma patients face elevated baseline risk for thromboembolic complications due to age-related hypercoagulability, immobilization, and injury-induced inflammatory cascades. Against this backdrop, researchers used the 2017 American College of Surgeons Trauma Quality Improvement Program (TQIP) database to investigate whether THC-positive urine drug screening status was associated with higher rates of deep vein thrombosis, pulmonary embolism, stroke, and myocardial infarction. The mechanistic rationale draws on laboratory evidence that cannabinoids may promote platelet activation and alter coagulation pathways, though these mechanisms remain incompletely characterized in human subjects, particularly in the setting of acute trauma.
From 286,242 eligible patients aged 65 and older, the investigators identified 945 who tested THC-positive and propensity-matched them 1:2 with 1,890 THC-negative controls. After matching, THC-positive patients had significantly higher rates of DVT (2.2% versus 0.6%, p less than 0.01) and PE (1.5% versus 0.4%, p less than 0.01), with an overall thromboembolic complication rate of 3.0% versus 1.7%. No significant differences emerged for stroke, myocardial infarction, ICU or hospital length of stay, or mortality. Critical limitations include reliance on a single urine drug screen that cannot distinguish acute from chronic use, quantify dose, or differentiate recreational from medicinal consumption. The authors acknowledge that residual unmeasured confounding is inevitable in a retrospective registry study and call for prospective investigations before clinical protocols are altered.
Dr. Caplan’s Take
This study asks an important question at the right time. More of my geriatric patients are using cannabis, whether for pain, sleep, or general wellness, and when they present after a fall or motor vehicle accident, I want to know whether their THC status changes my risk calculus. The association with DVT and PE here is real in statistical terms, and the propensity matching is commendably thorough. But a single urine screen tells me almost nothing about what that patient’s actual cannabis exposure looks like: how much, how often, by what route, and for how long. That gap makes it impossible to translate these findings directly into dosing or prophylaxis guidance.
In practice, I already factor THC-positive status into my overall risk assessment for geriatric trauma patients, but I do not treat it as an independent indication for escalated thromboprophylaxis beyond what their injury profile and mobility status warrant. What I do is ensure that standard VTE prophylaxis is initiated promptly and that the care team is aware of the positive screen as one additional data point. If a patient asks whether their cannabis use put them at higher risk for clots after a traumatic injury, I tell them honestly that there is a signal in the data but not yet proof, and that the most important modifiable factor right now is timely, guideline-concordant prevention.
Clinical Perspective
This study sits early in the research arc for THC and thromboembolic risk in geriatric trauma. It is hypothesis-generating rather than practice-changing. The findings are consistent with a smaller body of case reports and mechanistic studies suggesting cannabinoid-mediated effects on platelet function and coagulation, but they do not confirm a causal pathway. The absence of significant differences in stroke, MI, and mortality may reflect true null effects or may simply indicate insufficient statistical power given the low absolute event rates. Clinicians should note that among patients already on chronic anticoagulation, no outcome differences by THC status were observed, which tentatively suggests that adequate anticoagulation may attenuate whatever risk THC confers.
From a safety standpoint, cannabis use in the elderly population raises additional concerns beyond thrombosis, including drug interactions with anticoagulants such as warfarin (THC and CBD can inhibit CYP enzymes involved in warfarin metabolism), increased fall risk, and cognitive effects that may complicate postoperative recovery. The finding that DVT rates remained elevated even among THC-positive patients who received standard VTE prophylaxis is notable but should not be interpreted as evidence that prophylaxis failed because of THC. The single most actionable takeaway for clinicians now is to document THC screening results systematically and ensure that VTE prophylaxis is initiated within evidence-based timeframes for all geriatric trauma patients, using THC-positive status as one element in a comprehensive risk discussion rather than as a standalone trigger for protocol changes.
Study at a Glance
- Study Type
- Retrospective cohort study with propensity score matching (1:2 ratio)
- Population
- Geriatric trauma patients aged 65 and older from over 700 US trauma centers
- Intervention
- No intervention; THC exposure defined by positive urine drug screen within 24 hours of admission
- Comparator
- THC-negative geriatric trauma patients matched on demographics, injury severity, comorbidities, anticoagulant use, and thromboprophylaxis
- Primary Outcomes
- Thromboembolic complications (DVT, PE, stroke, MI), ICU and hospital length of stay, mortality
- Sample Size
- 286,242 eligible; 2,835 in matched cohort (945 THC+, 1,890 THC-)
- Journal
- Published 2022; TQIP registry data from 2017
- Year
- 2022
- Database
- ACS Trauma Quality Improvement Program (TQIP), 2017
- Funding Source
- Not specified in available data
What Kind of Evidence Is This
This is a retrospective observational cohort study using national trauma registry data, with propensity score matching employed to reduce measured confounding. It sits in the lower-to-middle tier of the evidence hierarchy, above case series and cross-sectional analyses but well below randomized controlled trials. The single most important inference constraint is that no amount of propensity matching can eliminate unmeasured confounding, meaning the observed association between THC-positive status and thromboembolic events cannot be interpreted as causal.
How This Fits With the Broader Literature
A growing but still limited body of literature has linked cannabis use to cardiovascular events including arterial thrombosis, stroke, and myocardial infarction, primarily through case reports and small retrospective studies. Preclinical research has demonstrated that THC can increase expression of tissue factor and glycoprotein IIb/IIIa on platelet surfaces, providing a plausible prothrombotic mechanism. However, most prior clinical work has focused on younger populations, and the findings have been inconsistent across study designs. This study extends the literature by focusing specifically on the geriatric trauma population and leveraging a large, multisite registry, though it shares the exposure measurement limitations common to nearly all cannabis epidemiology research.
The observation that chronic anticoagulation appeared to mitigate THC-associated risk is novel and mechanistically interesting, though it emerged from a small subgroup and requires replication. The broader literature has not yet produced a prospective study capable of resolving the question of whether THC independently promotes venous thromboembolism in any population.
Common Misreadings
The most likely overinterpretation is concluding that marijuana use causes blood clots in elderly trauma patients. The study’s own abstract uses causal language (“THC exposure increases the risk”), which slightly overstates what a retrospective observational design can establish. Association is not causation, and the reliance on a single qualitative urine screen means the “exposure” is poorly characterized. A patient who used cannabis once in the prior month and a daily heavy user would both appear as THC-positive. Additionally, the 0.33% THC-positive rate in eligible patients likely reflects inconsistent screening across trauma centers rather than true prevalence, introducing selection bias that could systematically differ between patients who were and were not screened.
Bottom Line
This large registry study identifies a statistically significant association between THC-positive urine screens and higher rates of DVT and PE in geriatric trauma patients, but it cannot establish that cannabis caused those events. The findings are hypothesis-generating and support including THC status in clinical risk discussions. They do not, however, justify changing thromboprophylaxis protocols based on THC status alone. Prospective studies with quantified, longitudinal cannabis exposure data are needed before clinical recommendations can be made with confidence.
References
- Hernandez I, He M, Brooks C, et al. THC exposure increases the risk of thromboembolic complications in geriatric trauma patients. ACS TQIP registry study, 2022.
- American College of Surgeons. Trauma Quality Improvement Program (TQIP). Available at: https://www.facs.org/quality-programs/trauma/quality/trauma-quality-improvement-program/
- Wolff V, Armspach JP, Lauer V, et al. Cannabis-related stroke: myth or reality? Stroke. 2013;44(2):558-563. doi:10.1161/STROKEAHA.112.671347
- Nascimento CJ, Bhatt DL. Cardiovascular effects of cannabis: an evolving evidence base. J Am Coll Cardiol. 2020;75(7):831-834.
- Rezkalla S, Kloner RA. Cardiovascular effects of marijuana. Trends Cardiovasc Med. 2019;29(7):403-407. doi:10.1016/j.tcm.2018.11.004
Have thoughts on this? Share it:
