Abstract
Background
Traumatic brain injury (TBI) results in an elaborate systemic cascade of secondary injury elicited in part by an intrinsic catecholamine response, which ultimately leads to changes in inflammation and coagulopathy. Attenuation of this catecholamine response with agents such as propranolol confers a survival advantage. The related impact of propranolol on venous thromboembolism (VTE) after TBI is largely unknown.
Study Design
A single institution retrospective review was conducted of all TBI patients requiring intensive care unit (ICU) admission with an injury severity scale (ISS) ≥ 25 from January 2013 to May 2015. Patients who received at least one dose of propranolol within 24 hours of admission (PROP) were compared to patients who did not receive any doses of propranolol (NPROP) during their hospitalization.
Results
Of the 131 patients analyzed, 31 (23.7%) patients received propranolol. The PROP cohort was more severely injured overall (ISS 29 vs 26.5, P = .02). While unadjusted VTE rates were similar (16.1% vs 19.0%, P = .72), the adjusted VTE rate was lower in the PROP cohort (AOR 0.20 (95% CI 0.04-0.97), adjusted P-value < .05).
Conclusion
Propranolol use in TBI patients who have sustained critical injuries may mitigate the risk of VTE. The mechanism by which this outcome is achieved requires further investigation.
Introduction
Traumatic brain injury (TBI) is a source of significant morbidity, mortality, and hospital expenditures after trauma. 1 This may at least partially be attributed to the elaborate systemic cascade manifested during the secondary injury phase generated by an intrinsic catecholamine response. 2 The catecholamine response is well described in the literature and has been strongly associated with changes in inflammation and coagulation.3,4 Coagulation changes may manifest in either an early hypocoagulable or late hypercoagulable state.5,6 It is well know that patients with TBI have a higher rate of venous thromboembolism (VTE), which can be associated with extended hospital stays and escalation of care. 7
Attenuation of this catecholamine response by beta-adrenergic receptor blockers was associated with lower mortality in a prospective multicenter trial, with propranolol, a nonselective beta blocker which is able to cross the blood brain barrier, conferring the best survival advantage. 8 What is not clear is if mitigating the catecholamine response has any implications for alterations in coagulation after trauma.
An animal model of TBI with beta-adrenergic receptor knockout showed no increase in fibrinolysis in the beta-adrenergic deficient mice subjected to injury, implying that blunting of the catecholamine response with agents, such as propranolol, may attenuate the hypocoagulability profile after TBI acutely. 9 The impact of early propranolol administration in altering the hypercoagulable state is unknown.
The objective of this study is to determine if early propranolol is associated with a lower VTE risk in TBI patients who have sustained severe injuries. We hypothesized that propranolol is associated with a lower risk for VTE.
Methods
A retrospective review was conducted of all TBI patients 18 years and older admitted to an intensive care unit (ICU) at an academic, urban level 1 trauma center from January 1, 2013, to May 31, 2015. Patients with sealed records, preexisting VTE, those on anticoagulation prior to admission, and were in the ICU for less than 24 hours were excluded. Those with an injury severity score (ISS) of 25 or greater were selected for analysis. This analysis represents data collected prior to changes in how our institution prescribes propranolol. As propranolol is considered standard of care for TBI patients admitted to the ICU, we needed to compare cohorts with patients who were not commonly given propranolol which occurred before 2015.
Patients with polytrauma were typically admitted to the surgical ICU if polytrauma was present or to the neurosurgical ICU if injuries were limited to the head, face, or spine. Both units are staffed by intensivists, with trauma surgeons serving as critical care intensivists in the surgical ICU and neurocritical care physician leading the neurosurgical ICU. Propranolol administration was generally encouraged, but the decision to administer was at the discretion of the intensivist overseeing the care of the patient. Propranolol was discouraged if the patient was hypotensive or bradycardic. When administered, propranolol was initiated at 1 mg every 6 hours and then transitioned to orally 40 mg twice daily.
The decision to initiate VTE chemoprophylaxis was based upon input from the trauma team, the ICU team, and additional surgical teams that were involved in the patient’s care. While there was no set protocol, early chemoprophylaxis was strongly encouraged. Chemoprophylaxis was generally started 48 hours after the last stable head computed tomography unless there was concern for additional ongoing bleeding.
A selective screening protocol was used to detect VTE. Duplex ultrasonography studies of the extremities were ordered if there was a clinical suspicion for a deep vein thrombosis (DVT). Similarly, a computed tomography angiography of the chest or a ventilation-perfusion scan was ordered if a pulmonary embolism (PE) was suspected. Select patients who had a prolonged contraindication to VTE chemoprophylaxis may have undergone surveillance imaging.
Data collection included patient demographics, admission vital signs, head/neck abbreviated injury scale (AIS), injury severity score (ISS), details regarding propranolol administration, timing of VTE chemoprophylaxis, and need for mechanical ventilation. Outcomes data included VTE, ICU and hospital length of stay (LOS), and in-hospital mortality.
Patients who received at least 1 dose of propranolol within 24 hours from admission were designated to the PROP cohort; those patients were compared to those who did not receive any doses of propranolol throughout their hospital stay (NPROP). Patients receiving a dose of propranolol after 24 hours were excluded as the inflammatory cascade after TBI is thought to begin relatively early after injury, and thus, a later dose of propranolol theoretically may not have a similar response as a dose given earlier. The primary outcome of this study was the rate of VTE. Data are summarized as percentages for categorical variables and medians with interquartile ranges for continuous variables. Categorical variables were compared by utilizing a Pearson χ2 or Fisher’s exact test where appropriate, whereas comparisons of continuous variables were performed by either a Mann-Whitney U test or Student’s t-test. A P-value less than .05 was considered statistically significant. A multivariate regression analysis was performed to identify the adjusted risk for VTE. Variables with P-values < .05 on univariate analysis in addition to variables deemed clinically relevant, after excluding co-linear variables, were entered into the model. Adjusted odds ratios (AORs) and their respective 95% confidence intervals (CIs) were calculated and reported.
All statistical analysis was performed using IBM SPSS statistics for Windows, version 25 (IBM Corp., Armonk, N.Y., USA). The study was approved by the Cedars-Sinai Medical Center’s Institutional Review Board with the need for consent formally waived.
Results
Comparison of Patients Who Received Propranolol (PROP) to Those Who Did Not Receive Propranolol (NPROP).
Abbreviations: AIS, abbreviated injury scale; GCS, Glasgow Coma Scale; IQR, interquartile range; ISS, injury severity score; LOS, length of stay; SD, standard deviation; VTE, venous thromboembolism.
Patients in the PROP cohort were more likely to be placed on pharmacologic prophylaxis at some point during their hospitalization (83.9% vs 46.0%, P < .01); however, there were no differences in the proportion of patients started on a chemoprophylactic agent within 2 days after admission (3.2% vs 7.0%, P = .68) or 5 days after admission (41.9% vs 35.0%, P = .48).
Most patients in both cohorts required mechanical ventilation. Intensive care unit (11.9 ± 12.1 vs 7.4 ± 9.8 days, P < .01) and hospital LOS (21.6 ± 20.7 vs 11.7 ± 11.2 days, P < .01) were significantly longer in the PROP cohort. The mortality rate was significantly lower among those patients receiving early propranolol (12.9% vs 32.0%, P = .02).
The unadjusted VTE rate was similar (16.1% vs 19.0%, P = .72). After adjusting for confounders including head AIS, ISS, receiving VTE chemoprophylaxis within 2 days of admission, ventilator days, and hospital LOS, patients receiving early propranolol had a lower risk of VTE (AOR 0.20 (95% CI 0.04-0.97), adjusted P-value < .046). The area under the receiver operating characteristic curve was 0.79 (95% CI 0.67-0.90).
Discussion
Traumatic brain injury induces multiple changes in central and systemic physiology, including dynamic alterations in coagulopathy, which in part stems from the innate catecholamine response. 10 While there is an immediate hypocoagulable phase observed soon after trauma, there is a delayed hypercoagulable phase observed later. 11 It is unclear if propranolol, a nonselective beta blocker which is able to cross the blood brain barrier, has the potential of altering the resulting hypercoagulable state. We found that although the unadjusted VTE rate was similar between those patients receiving an early dose of propranolol compared to those who did not, the adjusted risk for VTE was substantially lower for the patients who received propranolol. This suggests that propranolol may potentially mitigate hypercoagulability after TBI.
The neuroinflammatory cascade results in secondary brain injury as manifested by neuronal cell death, apoptosis, and neurodegeneration that has the potential for long-term detrimental effects. 12 This is mediated by both alpha- and beta-adrenergic receptors. Partially blunting this response with beta blockers can alleviate oxidative stress, increase oxygenation, and lessen the risk of cardiac dysfunction.13,14 These various implications may culminate in end organ damage and explain why beta blockers appear to confer a survival advantage.8,13
The mechanism by which TBI results in a hypercoagulable state is inherently complex and can manifest overtly as a VTE event or innocuously as microthrombosis. 15 The breakdown of the blood brain barrier after injury and the resulting dysfunction coupled with acute lymphocytic infiltration is thought to contribute to cerebral thromboinflammation.16,17 Systemic thromboinflammation, in contrast, is secondary to a number of factors including platelet and leukocyte aggregation mediated by multimers of von Willebrand factor. 18 Release of brain-derived extracellular vesicles and additional proteins, such as tissue factor, into the bloodstream can further yield a hypercoagulable state. 19 In addition, delays in initiating VTE chemoprophylaxis in TBI patients due to the concern for bleeding are a contributing factor. In fact, early initiation of these agents has been associated with lower VTE rates without increasing morbidity.20,21
While this study established that propranolol is associated with a lower VTE rate, there are also studies that concluded propranolol may reduce the factors associated with thromboembolic events. Hoppener et al have shown that patients with a documented DVT had higher factor VIII:C levels at baseline and those treated with propranolol for 14 days subsequently had decreased factor VIII:C levels compared to patients that were not treated. 22 Similar findings were noted in a study conducted by Schönauer et al. 23 As higher plasma factor VIII:C levels are a known risk factor for VTE, 24 these studies provide a mechanistic pathway for how propranolol reduces VTE.
There are several limitations to our conclusions that need to be acknowledged. First, our findings were based on a retrospective study limited to a single center. As all patients did not receive routine surveillance, the VTE rate may have been underestimated. There was an overall low rate of VTE chemoprophylaxis, and for those that were started on such an agent, possible delays. This late initiation and low rate likely stems in variation in practice among providers regarding the use of chemoprophylaxis after TBI and may have influenced our results. Due to the limited sample size, a propensity-matched study was not possible. Instead, the study was limited to patients with an ISS ≥25 in order to allow for an adequate number of VTE events with relatively comparable cohorts.
Despite these limitations, there appears to be a lower risk of VTE in the subset of TBI patients with critical injuries who received propranolol. In addition to improving survival rates, propranolol may mitigate morbidity in select trauma patients. The mechanism by which this outcome is achieved requires further investigation. A larger, prospective study may be warranted to understand the optimal timing and application of propranolol in TBI patients.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Author Note
This manuscript was selected for plenary presentation at the American College of Surgeons Southern California Chapter Annual Scientified Meeting. Sant Barbara, California, January 2021 (Canceled due to COVID-19 pandemic).