Factors Associated with Delayed Intracranial Hemorrhage in Trauma Patients : A Retrospective Study at a Level I Trauma Center

Article information

J Korean Neurosurg Soc. 2026;69(1):135-141

Publication date (electronic) : 2025 October 31

doi : https://doi.org/10.3340/jkns.2025.0031

Juhong Park¹, Namkyu Yoo², Byung Hee Kang^,1

¹Division of Trauma Surgery, Department of Surgery, Ajou University School of Medicine, Suwon, Korea

²Department of Neurosurgery, Ajou University School of Medicine, Suwon, Korea

Address for correspondence : Byung Hee Kang Division of Trauma Surgery, Department of Surgery, Ajou University School of Medicine, 164 Worldcup-ro, Yeongtong-gu, Suwon 16499, Korea Tel : +82-31-219-7764, Fax : +82-31-219-7781, E-mail : kbhname@gmail.com

Received 2025 January 27; Revised 2025 April 17; Accepted 2025 June 11.

Abstract

INTRODUCTION

Traumatic brain injury (TBI) is associated with poor outcomes in patients with multiple traumas. Therefore, brain computed tomography (CT) is usually performed in trauma centers for initial management. However, in some cases delayed intracranial hemorrhage (DICH)—defined as the appearance of hemorrhage in areas of the brain that appeared completely or almost normal in appearance on the CT scan taken shortly after injury—can occur [6]. Because the incidence of DICH is low (about 0.3%), additional brain CT is not routinely recommended to patients, despite being prescribed anticoagulant or antiplatelet medications [5]. Moreover, following brain CT was usually recommended when deterioration was observed in mild TBI [17]. However, assessing the mental status of patients with multiple trauma in the acute phase is challenging owing to sedation from intensive care. In addition, although DICH was not common in previous studies, well-known study conducted on patients with minor injuries and discharged from the emergency room [5] revealed that the incidence of DICH might be higher in patients who were admitted to the hospital, and the result also affected the patients’ outcomes. However, it is unclear which patients require additional brain CT [3,7]. Hence, this study aimed to find the incidence and risk factors of DICH in patients with trauma admitted to trauma centers with the goal of refining criteria for follow-up imaging in this high-risk group.

MATERIALS AND METHODS

This study was approved by the Institutional Review Board of Ajou University Hospital (AJIRB-2024-0260-001) and followed ethical guidelines in the declaration of Helsinki.

Patients with trauma who were admitted to our regional trauma center between January 2018 and June 2020, were retrospectively reviewed. Patients who underwent brain CT during initial assessment were included in this study. The exclusion criteria were as follows : brain hemorrhage was detected on initial CT, subsequent CT was not taken within 48 hours, incomplete medical record and new lesions were detected but not diagnosed with hemorrhage.

Patients with minor mechanisms of injury usually visited the emergency room and were discharged if there were no significant brain injuries. However, patients with severe injuries (by field triage) visited the trauma bay and were admitted for other injuries if there was no brain hemorrhage on CT. In blunt injury, initial brain CT was usually performed in our trauma center according to the Canadian CT head rule, and subsequent brain CT should be performed in patients who present with neurologic deterioration [15]. However, subsequent brain CT was also taken for patients who had a loss of consciousness, complaints of symptoms such as severe headache or dizziness, and injury mechanism, which had a high chance of brain injury, although there was no definite protocol in the center.

DICH was defined as newly identified intracranial hemorrhage on repeat brain CT after an initial scan showed no hemorrhage. The patients were dichotomized into DICH and normal groups, and the epidemiology, injury characteristics, and outcomes were compared. Variables were selected based on epidemiological data, trauma-related factors, laboratory findings and medication. To identify the risk factors for DICH, logistic regression analysis was performed using variables with a p-value <0.05 in the univariate analysis. Injury severity score (ISS) was excluded from the regression model to avoid confounding, as the presence of DICH would elevate ISS values. Detailed clinical reviews were also conducted for each DICH case.

Continuous variables were compared between the two groups using Mann Whitney test (median [interquartile]) after normality test, while categorical variables were compared using chi-squared test of Fisher’s exact test. All analyses were performed using SPSS (version 23.0; IBM Corp., Armonk, NY, USA).

RESULTS

General characteristics and outcome

Of the 784 patients, 37 (4.7%) were included in the DICH group, and 747 were included in the normal group (Fig. 1). Within DICH group, subdural hemorrhage was the most common type occurring in 59.5% of cases (Table 1).

Fig. 1.

Study scheme. CT : computed tomography, DICH : delayed intracranial hemorrhage.

Table 1.

Types of delayed intracranial hemorrhage

DICH group showed a higher portion of ≥65 years old (29.7% vs. 14.2%; p=0.010) shock during resuscitation in the trauma bay (lowest systolic blood pressure [SBP] ≤90 mmHg) (35.1% vs. 11.8%; p<0.001), massive transfusion rate (13.5% vs. 3.8%; p=0.004), portion of Abbreviated injury scale (AIS) ≥3 in abdomen or pelvis (29.7% vs. 11.7%; p=0.001), and ISS (26 [22–37] vs. 22 [19–30]; p<0.001). Among the associated head and neck injuries, cervical spine fractures were significantly more common (27.0% vs. 11.1%; p=0.008) in the DICH group. However, the use of antithrombotic medication did not differ significantly between the two groups (Table 2).

Table 2.

Demographics and clinical characteristics

DICH group had longer intensive care unit stay (18 [10–29] vs. 8 [4–15] days; p<0.001) and hospital stay (43 [24–66] vs. 31 [20–51]; p<0.001); however, no mortality was observed in DICH group (Table 3).

Table 3.

Treatment outcome according to DICH

Risk factors for DICH

Multivariable logistic regression revealed that age ≥65 years (odds ratio [OR], 2.681; 95% confident interval [CI], 1.250–5.753; p=0.011), lowest SBP ≤90 mmHg during resuscitation in trauma bay (OR, 2.678; 95% CI, 1.247–5.750; p=0.012), AIS abdomen or pelvis ≥3 (OR, 2.667; 95% CI, 1.213–5.864; p=0.015), and cervical spine fracture (OR, 2.408; 95% CI, 1.084–5.351; p=0.031) were associated with DICH (Table 4).

Table 4.

Multivariate analysis for risk factors of delayed intracranial hemorrhage

Invasive procedure for DICH

Surgery was not performed in any of the 37 patients with DICH; only a 17-year-old male patient underwent intracranial pressure (ICP) monitoring, after being admitted to the trauma bay after a motorcycle accident. Initial SBP was 88 mmHg, and a combined skull and pelvic fracture was diagnosed. No brain hemorrhage was found on the initial CT scan, but follow-up CT showed multiple intracerebral hemorrhages and brain swelling. ICP monitoring was performed in ICU, and the patient was discharged without further intervention (Fig. 2).

Fig. 2.

Delayed intracranial hemorrhage case for invasive procedure (initial CT and follow up CT after 7 hours). CT : computed tomography.

DISCUSSION

In this study, the incidence of DICH was 4.7% and risk factors of DICH were age ≥65 years, SBP ≤90 mmHg during initial resuscitation, severe abdomen or pelvic injury, and cervical spine fracture.

The incidence of DICH was much higher than that reported in the previous studies (approximately 0.3%) [2,5]. Because patients who did not undergo CT within 48 hours were excluded, the sample size was smaller than those of previous studies. In the current study, a total of 1101 patients were excluded, and they were assumed to not have DICH. However, if those patients were included, the incidence would be higher by approximately 1.9%. This might be because the included patients were more severely injured than the previous studies which included those with isolated head trauma. Trauma-induced coagulopathy (TIC) is related to injury severity and could cause delayed bleeding [4]. Hence, it seems that DICH was higher in this study because included patients usually had polytrauma, which included mean ISS of 14.17±9.73. In addition, following brain CT performed selectively for symptomatic patients, including those with neurologic deterioration; hence, the incidence might be higher. The rate of 4.6% is quite high; therefore, DICH should be evaluated in patients with polytrauma as well.

Antiplatelet therapy is a risk factor for brain hemorrhage. Many guidelines recommend brain CT scans in blunt trauma and observation within 24 hours for DICH in patients receiving antiplatelet medication [1,9,16]. In contrast, the incidence of DICH was not higher in patients taking antiplatelet therapy [2,10,14]. Consistent with previous findings, antiplatelet or anticoagulant agents were not risk factors for DICH in this study. However, the number of patients taking antithrombotic drugs was relatively small, and more studies are required to draw definitive conclusions. Several DICH case reports related to skull fracture were published [13,18,19]. Although facial bone fracture was not associated with DICH in this study, skull fracture could mean the significant damage to the head. In addition, cervical spine fracture is associated with brain hemorrhage for a similar reason [8]. Nevertheless, facial bone fractures were not associated, which might be because severe facial bone fractures were not included. Only three cases with a face AIS ≥3 were included in this study. Severe facial bone fractures usually present with brain hemorrhage in the initial CT, hence they might have been excluded. However, brain hemorrhage might not be detected immediately after trauma; therefore, a subsequent brain CT might be recommended for patients with skull or spine fractures.

SBP ≤90 during resuscitation and severe abdominopelvic injury were also associated with DICH in this study. Lee and Lui [11] suggested that hypotension may have acted as protective mechanism obscuring the ICH. Similar to previous study, we identified other risk factors related to extracranial injury. Severe pelvic fracture could cause exsanguinous bleeding with hypotension; hence, it could easily result in TIC [12]. TIC might be a more significant contributor to hemorrhage than are antithrombotic drugs because medication usage was not identified as a risk factor in the current study. However, there is no standard marker for TIC and the initial lab findings were not significantly different between two groups, further studies are needed to investigate the relationship between TIC and DICH.

Despite the high incidence of DICH, good clinical outcomes were observed in the study. Only one patient required an invasive procedure, and there was no patient death. Therefore, it is still inconclusive whether repeated brain CT should be routinely performed because the treatment strategy would not be changed. Nevertheless, the diagnosis of DICH is crucial because it can affect the use of anticoagulants for deep vein thrombosis prophylaxis and injury score coding. Moreover, the patients need to be informed whether a brain hemorrhage exists in view of insurance, compensation, or legal aspects. Therefore, repeat brain CT scans may be helpful in patients with cervical spine fracture, and severe injuries, especially with abdominopelvic injury with shock.

Our study has several limitations. First, it was conducted retrospectively; however, brain CT imaging was performed prospectively. Second, there was no standardized protocol for repeat brain CT, resulting in variation in the number of included patients. Although not all patients underwent follow-up imaging, if we had included patients who did not receive repeat CT, the true incidence of DICH might have been even higher than reported. Third, the ISS was higher in the DICH group, likely because the presence of brain hemorrhage itself contributed to an increased ISS, potentially biasing the analysis.

Despite these limitations, we believe our findings provide preliminary insights into risk patterns associated with DICH. Although the identified risk factors—such as shock, abdominopelvic injury, and cervical spine fractures—may largely reflect general trauma severity, they can help guide clinical suspicion. While our conclusions should be interpreted with caution, we cautiously suggest that repeat brain CT may be considered in patients with severe trauma or hemodynamic instability, even when the initial scan is normal. These results highlight the need for future prospective studies with standardized imaging protocols to better define the indications for follow-up CT in trauma patients.

CONCLUSION

In conclusion, our findings suggest that the incidence of DICH may be higher in patients with severe trauma. Although a change in treatment strategy is not currently supported, repeat brain CT could be considered in selected blunt trauma patients—particularly those who experienced shock during initial resuscitation, had severe abdominopelvic injury, or sustained cervical spine fractures—even when initial brain CT findings are normal. Further prospective studies are needed to establish clearer guidelines.

Article information Continued