Blood Pressure Control in Traumatic Subdural Hematomas

Background There is debate over optimal systolic blood pressure (SBP) after traumatic subdural hematoma. Increased SBP has the benefit of increasing cerebral perfusion pressure and limiting the detrimental secondary effects of traumatic brain injury but poses a risk of hematoma expansion. While prior studies have shown that SBP<90mmHg is associated with worsened morbidity and mortality in subdural hematoma patients, clinical guidelines and expert opinion have differing initial SBP goals. The aim of this study is to leverage a large database to determine the effects of two such goals, namely SBP 100-150mmHg versus SBP<180mmHg in this patient population. Methods A de-identified database network (TriNetX Research Network) was used to retrospectively query all patients with a first instance diagnosis of acute traumatic SDH, who also had a recorded GCS, with maintenance of SBP 100-150 within the first 24 hours (cohort 1) versus patients with an SBP<180 (cohort 2). Data came from 68 health care organizations (HCOs) with a total of 105,897,964 patients on 9/1/2022. The primary outcome of interest was mortality within 30 days. Secondary outcomes include gastrostomy tube placement, craniotomy/craniectomy/burr hole drainage, venous thromboembolism, ischemic stroke, myocardial infarction, seizure, falls, cardiac arrest, and acute kidney injury within 30 days. Cohorts were propensity-score matched for confounders. Results After propensity score matching, 1,243 patients were identified in each cohort. Age at index was 57.97+/-23.21 years and 58.28+/-22.35 years for cohorts 1 and 2, respectively. Mortality was seen in 243 patients (19.756%) vs. 209 (16.992%) (OR 1.203, 95% CI (0.98,1.476), p=0.0767) in cohorts 1 and 2, respectively. There was no statistical difference in secondary outcomes. Conclusion The results of this study demonstrate that the primary outcome of mortality at 30 days is not statistically different in acute traumatic SDH patients, whether their SBP is kept at 100-150 or below 180. Likewise, it shows no statistical difference in the subsequent incidence of gastrostomy tube placement, craniotomy/craniectomy/burr holes, venous thromboembolism, ischemic stroke, myocardial infarction, seizure, falls, or acute kidney injury.


Introduction
There is debate over optimal systolic blood pressure (SBP) after traumatic subdural hematoma (SDH) [1]. While hypotension, defined as SBP<90mmHg, is predictive of increased mortality in traumatic brain injury (TBI) patients [1][2][3][4][5][6][7][8], hypertension is associated with an increased risk of hematoma expansion [9][10][11], or the cause of the hematoma itself [12]. While Brain Trauma Foundation (BTF) fourth edition guidelines and expert opinion suggest SBP>100 for patients aged 50-69, SBP>110 for patients aged 15-49 or >70 years, and mean arterial pressure (MAP) >70 [2], Seattle International Brain Injury Consensus Conference recommendations are for SBP>90 and goal cerebral perfusion pressure (CPP) of 70 [5]. Attending physician preference at our institution (Penn State Milton S. Hershey Medical Center) for traumatic SDH blood pressure management in the absence of invasive intracranial monitoring has the propensity to gravitate towards one of two schools of management: keeping SBP 100-150 with MAP>65, or maintaining SBP<180 and MAP>65. The study aims to leverage an extensive database (TriNetX) to determine if there is a difference in 30-day outcomes in traumatic acute subdural hemorrhage patients who have maintained these two different SBP goals for at least 24 hours after their brain injury after propensity score matching for initial Glasgow coma score (GCS), and common comorbidities and risk factors.

Materials And Methods
A de-identified database network (TriNetX) was used to retrospectively query via International Classification of Disease (ICD-10) codes and Common Procedural Terminology (CPT) codes to evaluate all patients with a first instance diagnosis of acute traumatic SDH, who also had a recorded GCS, with the maintenance of SBP 100-150 within the first 24 hours (cohort 1) versus patients with an SBP<180 (cohort 2). Data came from 68 health care organizations (HCOs) with a total of 105,897,964 patients on 9/1/2022. Data includes demographics, diagnoses, medications, laboratory values, and procedures. The identity of the patients and HCOs is not disclosed to comply with ethical guidelines prohibiting data re-identification. An IRB waiver was granted. Previous literature informed our use of this database and its validity, and the network's exact details have been previously described [13][14][15][16].
The primary outcome of interest was mortality within 30 days. Secondary outcomes include gastrostomy tube placement, craniotomy/craniectomy/burr hole drainage, venous thromboembolism, ischemic stroke, myocardial infarction, seizure, falls, cardiac arrest, and acute kidney injury within 30 days. To adjust for hypothesized confounders on the relationship between SDH and the outcomes of interest, medical information, including age at the date of SDH, as well as sex, race, GCS scores 13-15, 9-12, and 3-8, and the comorbidities of traumatic subarachnoid hemorrhage, epidural hematoma, hypertension, obesity, ischemic heart disease, diabetes, chronic kidney disease, prior acute kidney failure, atrial fibrillation and flutter, smoking history, alcohol abuse, liver cirrhosis, and antiplatelet and anticoagulant medications was gathered. Analysis was performed using unmatched and propensity score-matched cohorts, with the greedynearest neighbor algorithm with a caliper of 0.1 pooled standard deviations. Chi-square analysis and logistic regression were performed on categorical variables.

Results
After propensity score matching, 1,243 patients were identified in each cohort. Age at index was 57.97+/-23. 21

Discussion
The results of this study demonstrate that the primary outcome of mortality at 30 days is not statistically different in traumatic SDH patients, whether their SBP is kept at 100-150mmHg or below 180mmHg. Likewise, it shows no statistical difference in the subsequent incidence of gastrostomy tube placement, craniotomy/craniectomy/burr holes, venous thromboembolism, ischemic stroke, myocardial infarction, seizure, falls, or acute kidney injury.
TBI is a leading cause of death and disability, and SDH accounts for approximately 11% of TBI [5,11]. SDH management has room for improvement [11]. While the primary brain insult cannot be reversed, secondary damage can be limited [7]. Secondary brain injury can result from impaired cerebral blood flow, impairment of autoregulation, disruption of tissue oxygenation and brain metabolism, tissue inflammation and necrosis, edema, oxidative stress, and vasospasm, and from hematoma expansion due to factors such as hypertension [5,6,[9][10][11][12]. On the other hand, any episode of hypotension, defined as SBP<90, has been associated with increased mortality and worsened neurological outcomes in severe TBI [1][2][3][4][5][6][7][8]17]. This could be due to the critical importance of maintaining optimal cerebral perfusion pressure, which can mitigate the detrimental effects of cerebrovascular autoregulation impairment in TBI patients, thus reducing ischemic and hyperemic brain damage [2][3][4][17][18][19][20]. While SBP goals have been established for intraparenchymal hemorrhage, it is less clear for subdural hematomas. Some guidelines and consensus/expert opinions suggest SBP>90, and others SBP>110 or SBP>115 [2,6]. There are no clear clinical trials that address a specifically targeted SBP in SDH [11,12].
In balancing the alteration of SBP goals to limit the progression of hematoma and hemorrhagic transformation and reducing hypoperfusion and secondary cerebral ischemic insult, prior studies provide some elucidation.
In 2021, Asmar et al. performed a retrospective review of the American College of Surgeons Trauma Quality Improvement Program database looking at adult patients with TBI, their SBP, and the primary outcome of in-hospital mortality. A subgroup analysis was likewise performed based on the GCS score. The authors found that of 94,411 TBI patients, the mean SBP was 147+-28, and overall mortality was 8.6%. They found that the lowest mortality was between SBP 110-149 and the highest mortality for SBP<90 and SBP>190. SBP 110-149 also had the lowest mortality for their GCS subgroup analysis [1].
The IMPACT prospective database looked at 8,172 TBI patients and found that SBP<120 was associated with poor neurological recovery [2,8].
Erickson et al. in 2021 published a study looking at MAP and discharge outcomes for pediatric patients with severe TBI. The authors examined 166 children (aged <18 years) with GCS<9. 20.4% of the patients had a poor discharge outcome (defined as in-hospital mortality or discharge to a skilled nursing facility or longterm care facility). This poor outcome was most common in patients who were <5th percentile for low MAPs in the first 12 hours and the second most common for patients with MAPs in the 5 th -9 th percentile in the first 12 hours. Likewise, the authors found that SBP below the 5 th percentile in the first 12 hours was likewise associated with poor discharge outcomes. The authors concluded that low MAPs were more strongly associated with poor discharge outcomes for pediatric TBI patients than low SBP, noting that low SBP had a higher predictive ability [17].
In 2019, a paper by Heino et al. looked at risk factors for recurrent hematoma after surgery for acute traumatic SDH, with a total of 132 patients at one center in Finland. They retrospectively reviewed demographic, clinical, and laboratory data and found that patients with post-craniotomy hematoma (after evacuation of SDH) were more likely to have a lower maximum SBP. The mean maximal SBP in patients with post-craniotomy hematoma was 148 mmHg versus 160 mmHg in patients without post-craniotomy hematoma [21].
Kow et al. in 2020 examined the escalation of MAPs in severe TBI in 28 patients. They found that increasing MAPs to raise CPP was associated with a reduction in ICP in two-thirds of the cases, and an increase in ICP in one-third of the cases. This could be because an increase in MAP induces vasoconstriction; however, caution must be used in severe TBI patients due to dys-autoregulation. The authors found that the absence of hypotension at initial evaluation was associated with a lower ICP burden. The authors noted that prior studies have shown that in the majority of TBI patients, when MAP rises significantly, then ICP will fall, but that is not the case for all TBI patients. The authors concluded that the direction of ICP changes in the first 15 minutes after MAP increase can be useful to see if a MAP challenge is beneficial in guiding the management of intracranial hypertension [4].
A paper by Powers et al. in 2018 looked at factors associated with the progression of conservatively managed acute traumatic SDH. It followed 117 patients. The authors found that high systolic blood pressure, the presence of subarachnoid hemorrhage, and initial SDH volume were associated with changes in SDH size. Authors concluded that higher SBP was associated with increased expansion of SDH [11].
Looking at acceptable blood pressure levels in the prehospital setting for patients with traumatic brain injury, Shibahashi  This study is not without limitations. The major limitation is that this study is retrospective. Furthermore, although GCS was able to be identified at the time of SDH, information was unable to be obtained on GCS at 30 days. Likewise, radiologic data was unavailable, and thus the degree of midline shift and size and location of SDH was unable to be identified. In addition, some degree of misidentification is always inevitable in large database studies.

Conclusions
This retrospective cohort study compares acute traumatic SDH patients with initial management of SBP 100-150 mmHg versus less than 180 mmHg in the first 24 hours of brain injury. The results of this study demonstrate that the primary outcome of mortality at 30 days is not statistically different and similarly shows no statistical difference in the subsequent incidence of gastrostomy tube placement, craniotomy/craniectomy/burr holes, venous thromboembolism, ischemic stroke, myocardial infarction, seizure, falls, or acute kidney injury.

Additional Information Disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.