Emerg Radiol DOI 10.1007/s10140-016-1377-2
ORIGINAL ARTICLE
The clinical significance of small subarachnoid hemorrhages Paul Albertine 1 & Samuel Borofsky 1 & Derek Brown 1 & Smita Patel 1 & Woojin Lee 1 & Anthony Caputy 1 & M Reza Taheri 1
Received: 13 November 2015 / Accepted: 11 January 2016 # American Society of Emergency Radiology 2016
Abstract With advancing technology, the sensitivity of computed tomography (CT) for the detection of traumatic subarachnoid hemorrhage (tSAH) continues to improve. Increased resolution has allowed for the detection of hemorrhage that is limited to one or two images of the CT exam. At our institution, all patients with a SAH require intensive care unit (ICU) admission, regardless of size. It was our hypothesis that patients with small subarachnoid hemorrhage experience favorable outcomes, and may not require the intensive monitoring offered in the ICU. This retrospective study evaluated 62 patients between 2011 and 2014 who presented to our Level I trauma center emergency room for acute traumatic injuries, and found to have subarachnoid hemorrhages on CT examination. The grade of subarachnoid hemorrhage was determined using previously utilized scoring systems, such as the Fisher, Modified Fisher, and Claassen grading systems. Electronic medical records were used to evaluate for medical decline, neurological decline, neurosurgical intervention, and overall hospital course. Admitting co-morbidities were noted, as were the presence of patient intoxication and use of anticoagulants. Patient outcomes were based on discharge summaries upon which the neurological status of the patient was assessed. Each patient was given a score based on the Glasgow outcome scale. The clinical and imaging profile of 62 patients with traumatic SAH were studied. Of the 62 patients, 0 % underwent neurosurgical intervention, 6.5 % had calvarial fractures, 25.8 % had additional intracranial
* Samuel Borofsky [email protected]
1
Department of Radiology, The George Washington University, First Floor 900 23rd St NW, Washington, DC 20037, USA
hemorrhages, 27.4 % of the patients had significant co-morbidities, and 1.6 % of the patients expired. Patients with lowgrade tSAH spent less time in the ICU, demonstrated neurological and medical stability during hospitalization. None of the patients with low-grade SAH experienced seizure during their admission. In our study, patients with low-grade tSAH demonstrated favorable clinical outcomes. This suggests that patients may not require as aggressive monitoring as is currently provided for those with tSAH. Keywords Subarachnoid hemorrhage . CT . Trauma
Introduction Traumatic subarachnoid hemorrhages (tSAH) are a common type of intracranial hemorrhage that results from stretching, tearing, and laceration of the blood vessels coursing within the subarachnoid space. The incidence of tSAH varies from 10 to 25 cases per 100,000 population per year [1]. Clinical presentation of tSAH is rather nonspecific and may often be asymptomatic. As such, patients who present to the emergency department after significant head trauma often undergo computer tomography to evaluate for the presence of a tSAH [2–4]. The use and relevance of CT scanning in predicting patient outcome has also increased as patients who are intoxicated or undergo early sedation are difficult to reliably assess clinically [5–7]. Additionally, with the increasing sensitivity of computer tomography, the detection of small intracranial hemorrhages is increasing [8]. Current guidelines based on observational data suggest that stable patients with small tSAH can be managed nonoperatively. Previous reports have suggested a 1.5 % rate for need for surgical intervention in patients with isolated subarachnoid hemorrhage [9].
Emerg Radiol
As such, recent studies have begun to question the presumed necessity of ICU monitoring of these small tSAH in patients with mild traumatic brain injury [10, 11]. At our institution, all patients with a tSAH require mandatory admission to the ICU to monitor for neurological or medical decline secondary to progression of the hemorrhage or development of secondary medical issues. The presence of an acute tSAH also places patients at risk for developing seizures, which is an additional reason that these patients may require intensive care monitoring [12, 13]. Despite the common occurrence of these injuries and the wide ranges in sizes of tSAH, little data is available which compares the extent of a tSAH to rates of neurologic decline, medical decline, and overall patient outcome. In this study, we sought to examine how the size of a tSAH may impact these parameters.
Patients and methods This IRB-approved retrospective cohort study is based on 62 patients evaluated between 2011 and 2014 who presented to a Level I trauma center emergency room for acute traumatic injuries that were found to have subarachnoid hemorrhages on CT examination. In the emergency department, Glasgow coma scale (GCS) scores, age, gender, heart rate, and systolic heart rate were performed and recorded by the trauma service. Once the patients were initially surveyed and stabilized, patients were then sent for CT examination.
Patient data Electronic medical records were used to evaluate for medical decline, neurological decline, neurosurgical intervention, and overall hospital course. Neurological decline was defined as any change in neurological exam that resulted in higher levels of care or a subsequent neurosurgical procedure as a result. For the purposes of this study, we included placement of an external ventricular drain or internal pressure monitoring device, craniotomy, or craniectomy in the category of neurosurgical intervention. Incidence of seizure was also noted as studies have shown that patients with intracranial hemorrhages have an increased risk of seizure activity [12]. Any significant medical decline was also noted. Medical decline was defined as any change in medical status that led to an increased level of care or prolonged hospitalization. Cases of increased medical care included cardiac decomposition, cardiac arrhythmias, pulmonary emboli, pulmonary edema, respiratory failure leading to intubation, aspiration pneumonia, and sepsis. Admitting co-morbidities were also accounted for which included patients intubated in the field prior to arrival to the ED, polytrauma requiring non-neurosurgical intervention, and acute territorial infarcts during hospitalization. The presence of patient intoxication and the current use of anticoagulants (dabigatran, ASA) were also noted. Patient outcomes were based on discharge summaries upon which the neurological status of the patient was assessed. Each patient was given a score based on the Glasgow outcome scale.
Grading scales and imaging
Results
Subarachnoid hemorrhages were evaluated based on thickness of subarachnoid blood and the presence of intraventricular hemorrhage. The tSAH were then graded based on Fisher, Modified Fisher, and Claassen classification criteria [14, 15]. Modified Fisher grading scale criteria are as follows: grade 1, thin SAH, no IVH; grade 2, thin SAH, with IVH; grade 3, thick SAH, no IVH; and grade 4, thick SAH, with IVH. Fisher scale grading scale criteria are as follows: grade 1, no SAH; grade 2, thin SAH, no IVH; grade 3, focal or diffuse thick SAH, no IVH; and grade 4, intracerebral or intraventricular blood with or without SAH. Claassen grading scale criteria are as follows: grade 1, thin SAH, no IVH; grade 2, thin SAH, with SAH; grade 3, thick SAH, no IH; and grade 4, thick SAH, with IVH. A neuroradiology attending, a neuroradiology fellow, and senior radiology resident performed and agreed on the grading measurements. Additional findings on the CT examination that could affect patient outcome were also recorded. These findings included the presence of additional intracranial hemorrhages and calvarial fractures. The number of repeat neuroimaging studies during the patient’s hospital course was also obtained.
Characteristics of the 62 patients are shown in Table 1. The majority were male (61 %) with a mean age of 50 (range, 19– 94). Of the 62 patients, no patients underwent neurosurgical intervention, 4 (6 %) had calvarial fractures, 16 (27 %) had additional intracranial hemorrhages, 17 (27 %) of the patients had significant co-morbidities, and 1 patient expired. For all patients, average length of ICU stay was 2.5 days (SD 3.9) and the average length of hospital stay was 4.6 days (SD 5.9). In an attempt to separate the patients based on tSAH severity, the patients were split into three groups based on Fisher, Modified Fisher, and Claassen criteria.
Fisher After stratifying patients based on Fisher criteria, detailed in the methods section, the results showed a significant increase in the numbers of days spent in the ICU with an increase in Fisher score (p = 1 days
No. of patients, n (%) 42 (67.74) 20 (32.26) 31 (50.00) 31 (50.00) Fisher
Modified Fisher 1, n (%) 3, n (%) 4, n (%) Claassen 1, n (%) 3, n (%) 4, n (%) Sulcus
3 (15.00)
12 (38.71) 3 (9.68)
31 (73.81) 9 (45.00) 10 (23.81) 8 (40.00)
23 (74.19) 17 (54.84) 8 (25.81) 10 (32.26)
1 (2.38)
0 (0.00)
3 (15.00)
4 (12.90)
1, n (%) 2, n (%)
16 (38.10) 8 (40.00) 12 (28.57) 2 (10.00)
13 (41.94) 11 (35.48) 7 (22.58) 7 (22.58)
3, n (%) 4, n (%)
8 (19.05) 3 (7.14)
2 (10.00) 4 (20.00)
5 (16.13) 3 (9.68)
5 (16.13) 4 (12.90)
5, n (%) 6, n (%)
1 (2.38) 2 (4.76)
2 (10.00) 2 (10.00)
2 (6.45) 1 (3.23)
1 (3.23) 3 (9.68)
Claassen After stratifying patients based on Claassen criteria, detailed in the methods section, the results showed a significant increase in the numbers of days spent in the ICU with an increase in Claassen score (p =
ORIGINAL ARTICLE
The clinical significance of small subarachnoid hemorrhages Paul Albertine 1 & Samuel Borofsky 1 & Derek Brown 1 & Smita Patel 1 & Woojin Lee 1 & Anthony Caputy 1 & M Reza Taheri 1
Received: 13 November 2015 / Accepted: 11 January 2016 # American Society of Emergency Radiology 2016
Abstract With advancing technology, the sensitivity of computed tomography (CT) for the detection of traumatic subarachnoid hemorrhage (tSAH) continues to improve. Increased resolution has allowed for the detection of hemorrhage that is limited to one or two images of the CT exam. At our institution, all patients with a SAH require intensive care unit (ICU) admission, regardless of size. It was our hypothesis that patients with small subarachnoid hemorrhage experience favorable outcomes, and may not require the intensive monitoring offered in the ICU. This retrospective study evaluated 62 patients between 2011 and 2014 who presented to our Level I trauma center emergency room for acute traumatic injuries, and found to have subarachnoid hemorrhages on CT examination. The grade of subarachnoid hemorrhage was determined using previously utilized scoring systems, such as the Fisher, Modified Fisher, and Claassen grading systems. Electronic medical records were used to evaluate for medical decline, neurological decline, neurosurgical intervention, and overall hospital course. Admitting co-morbidities were noted, as were the presence of patient intoxication and use of anticoagulants. Patient outcomes were based on discharge summaries upon which the neurological status of the patient was assessed. Each patient was given a score based on the Glasgow outcome scale. The clinical and imaging profile of 62 patients with traumatic SAH were studied. Of the 62 patients, 0 % underwent neurosurgical intervention, 6.5 % had calvarial fractures, 25.8 % had additional intracranial
* Samuel Borofsky [email protected]
1
Department of Radiology, The George Washington University, First Floor 900 23rd St NW, Washington, DC 20037, USA
hemorrhages, 27.4 % of the patients had significant co-morbidities, and 1.6 % of the patients expired. Patients with lowgrade tSAH spent less time in the ICU, demonstrated neurological and medical stability during hospitalization. None of the patients with low-grade SAH experienced seizure during their admission. In our study, patients with low-grade tSAH demonstrated favorable clinical outcomes. This suggests that patients may not require as aggressive monitoring as is currently provided for those with tSAH. Keywords Subarachnoid hemorrhage . CT . Trauma
Introduction Traumatic subarachnoid hemorrhages (tSAH) are a common type of intracranial hemorrhage that results from stretching, tearing, and laceration of the blood vessels coursing within the subarachnoid space. The incidence of tSAH varies from 10 to 25 cases per 100,000 population per year [1]. Clinical presentation of tSAH is rather nonspecific and may often be asymptomatic. As such, patients who present to the emergency department after significant head trauma often undergo computer tomography to evaluate for the presence of a tSAH [2–4]. The use and relevance of CT scanning in predicting patient outcome has also increased as patients who are intoxicated or undergo early sedation are difficult to reliably assess clinically [5–7]. Additionally, with the increasing sensitivity of computer tomography, the detection of small intracranial hemorrhages is increasing [8]. Current guidelines based on observational data suggest that stable patients with small tSAH can be managed nonoperatively. Previous reports have suggested a 1.5 % rate for need for surgical intervention in patients with isolated subarachnoid hemorrhage [9].
Emerg Radiol
As such, recent studies have begun to question the presumed necessity of ICU monitoring of these small tSAH in patients with mild traumatic brain injury [10, 11]. At our institution, all patients with a tSAH require mandatory admission to the ICU to monitor for neurological or medical decline secondary to progression of the hemorrhage or development of secondary medical issues. The presence of an acute tSAH also places patients at risk for developing seizures, which is an additional reason that these patients may require intensive care monitoring [12, 13]. Despite the common occurrence of these injuries and the wide ranges in sizes of tSAH, little data is available which compares the extent of a tSAH to rates of neurologic decline, medical decline, and overall patient outcome. In this study, we sought to examine how the size of a tSAH may impact these parameters.
Patients and methods This IRB-approved retrospective cohort study is based on 62 patients evaluated between 2011 and 2014 who presented to a Level I trauma center emergency room for acute traumatic injuries that were found to have subarachnoid hemorrhages on CT examination. In the emergency department, Glasgow coma scale (GCS) scores, age, gender, heart rate, and systolic heart rate were performed and recorded by the trauma service. Once the patients were initially surveyed and stabilized, patients were then sent for CT examination.
Patient data Electronic medical records were used to evaluate for medical decline, neurological decline, neurosurgical intervention, and overall hospital course. Neurological decline was defined as any change in neurological exam that resulted in higher levels of care or a subsequent neurosurgical procedure as a result. For the purposes of this study, we included placement of an external ventricular drain or internal pressure monitoring device, craniotomy, or craniectomy in the category of neurosurgical intervention. Incidence of seizure was also noted as studies have shown that patients with intracranial hemorrhages have an increased risk of seizure activity [12]. Any significant medical decline was also noted. Medical decline was defined as any change in medical status that led to an increased level of care or prolonged hospitalization. Cases of increased medical care included cardiac decomposition, cardiac arrhythmias, pulmonary emboli, pulmonary edema, respiratory failure leading to intubation, aspiration pneumonia, and sepsis. Admitting co-morbidities were also accounted for which included patients intubated in the field prior to arrival to the ED, polytrauma requiring non-neurosurgical intervention, and acute territorial infarcts during hospitalization. The presence of patient intoxication and the current use of anticoagulants (dabigatran, ASA) were also noted. Patient outcomes were based on discharge summaries upon which the neurological status of the patient was assessed. Each patient was given a score based on the Glasgow outcome scale.
Grading scales and imaging
Results
Subarachnoid hemorrhages were evaluated based on thickness of subarachnoid blood and the presence of intraventricular hemorrhage. The tSAH were then graded based on Fisher, Modified Fisher, and Claassen classification criteria [14, 15]. Modified Fisher grading scale criteria are as follows: grade 1, thin SAH, no IVH; grade 2, thin SAH, with IVH; grade 3, thick SAH, no IVH; and grade 4, thick SAH, with IVH. Fisher scale grading scale criteria are as follows: grade 1, no SAH; grade 2, thin SAH, no IVH; grade 3, focal or diffuse thick SAH, no IVH; and grade 4, intracerebral or intraventricular blood with or without SAH. Claassen grading scale criteria are as follows: grade 1, thin SAH, no IVH; grade 2, thin SAH, with SAH; grade 3, thick SAH, no IH; and grade 4, thick SAH, with IVH. A neuroradiology attending, a neuroradiology fellow, and senior radiology resident performed and agreed on the grading measurements. Additional findings on the CT examination that could affect patient outcome were also recorded. These findings included the presence of additional intracranial hemorrhages and calvarial fractures. The number of repeat neuroimaging studies during the patient’s hospital course was also obtained.
Characteristics of the 62 patients are shown in Table 1. The majority were male (61 %) with a mean age of 50 (range, 19– 94). Of the 62 patients, no patients underwent neurosurgical intervention, 4 (6 %) had calvarial fractures, 16 (27 %) had additional intracranial hemorrhages, 17 (27 %) of the patients had significant co-morbidities, and 1 patient expired. For all patients, average length of ICU stay was 2.5 days (SD 3.9) and the average length of hospital stay was 4.6 days (SD 5.9). In an attempt to separate the patients based on tSAH severity, the patients were split into three groups based on Fisher, Modified Fisher, and Claassen criteria.
Fisher After stratifying patients based on Fisher criteria, detailed in the methods section, the results showed a significant increase in the numbers of days spent in the ICU with an increase in Fisher score (p = 1 days
No. of patients, n (%) 42 (67.74) 20 (32.26) 31 (50.00) 31 (50.00) Fisher
Modified Fisher 1, n (%) 3, n (%) 4, n (%) Claassen 1, n (%) 3, n (%) 4, n (%) Sulcus
3 (15.00)
12 (38.71) 3 (9.68)
31 (73.81) 9 (45.00) 10 (23.81) 8 (40.00)
23 (74.19) 17 (54.84) 8 (25.81) 10 (32.26)
1 (2.38)
0 (0.00)
3 (15.00)
4 (12.90)
1, n (%) 2, n (%)
16 (38.10) 8 (40.00) 12 (28.57) 2 (10.00)
13 (41.94) 11 (35.48) 7 (22.58) 7 (22.58)
3, n (%) 4, n (%)
8 (19.05) 3 (7.14)
2 (10.00) 4 (20.00)
5 (16.13) 3 (9.68)
5 (16.13) 4 (12.90)
5, n (%) 6, n (%)
1 (2.38) 2 (4.76)
2 (10.00) 2 (10.00)
2 (6.45) 1 (3.23)
1 (3.23) 3 (9.68)
Claassen After stratifying patients based on Claassen criteria, detailed in the methods section, the results showed a significant increase in the numbers of days spent in the ICU with an increase in Claassen score (p =
