Advertisement

Magnetic Resonance Imaging in Aneurysmal Subarachnoid Hemorrhage: Current Evidence and Future Directions

  • Sarah E. Nelson
  • Haris I. Sair
  • Robert D. Stevens
Original Article

Abstract

Background

Aneurysmal subarachnoid hemorrhage (aSAH) is associated with an unacceptably high mortality and chronic disability in survivors, underscoring a need to validate new approaches for treatment and prognosis. The use of advanced imaging, magnetic resonance imaging (MRI) in particular, could help address this gap given its versatile capacity to quantitatively evaluate and map changes in brain anatomy, physiology and functional activation. Yet there is uncertainty about the real value of brain MRI in the clinical setting of aSAH.

Methods

In this review, we discuss current and emerging MRI research in aSAH. PubMed was searched from inception to June 2017, and additional studies were then chosen on the basis of relevance to the topics covered in this review.

Results

Available studies suggest that brain MRI is a feasible, safe, and valuable testing modality. MRI detects brain abnormalities associated with neurologic examination, outcomes, and aneurysm treatment and thus has the potential to increase knowledge of aSAH pathophysiology as well as to guide management and outcome prediction. Newer pulse sequences have the potential to reveal structural and physiological changes that could also improve management of aSAH.

Conclusion

Research is needed to confirm the value of MRI-based biomarkers in clinical practice and as endpoints in clinical trials, with the goal of improving outcome for patients with aSAH.

Keywords

Subarachnoid hemorrhage Magnetic resonance imaging Systematic review 

Introduction

Each year in the USA, up to 30,000 people experience subarachnoid hemorrhage due to rupture of an intracranial aneurysm, a condition associated with significant mortality and morbidity [1]. Despite advances in management, major gaps remain in the evaluation and treatment of brain injury due to aneurysmal subarachnoid hemorrhage (aSAH) patients, indicating an unmet need for discriminative biomarkers [2]. Brain magnetic resonance imaging (MRI) could emerge as one such biomarker because it can provide information on tissue anatomy, viability, physiology, and functional activation [3], particularly at higher-field MRI (e.g., 3 T as compared to 1.5 T) in which improved spatial resolution and temporal resolution can be achieved [4]. In comparison with computed tomography (CT), MRI does not involve ionizing radiation and has excellent soft tissue contrast [5]. In addition, various pulse sequences may be utilized to highlight specific tissue characteristics or brain physiology [5]. MRI is more widely available than other types of physiological imaging such as positron emission tomography [6] and obviates radiation exposure seen with single photon emission CT [7]. However, MRI is not routinely used for initial assessment and follow-up imaging in patients with aSAH, likely because of challenges in scanning critically ill or neurologically impaired subjects, image artifacts due to patient motion or materials used to secure aneurysms, time and cost constraints, and uncertainty regarding the clinical value of MRI in this population [8, 9, 10, 11].

Here, we provide a comprehensive overview of published research on the use of MRI in patients with aSAH and discuss new and advanced techniques that could help to inform management and predict outcomes. Despite uncertainty regarding feasibility, potential adverse events, and clinical relevance of MRI in aSAH patients [8, 9, 10, 11], studies support the use of MRI as a safe and clinically valuable test in this population, even when performed prior to aneurysm treatment [12, 13, 14, 15, 16].

Search Methodology

We searched PubMed from inception until June 2017 with keywords: (“subarachnoid hemorrhage”[tiab] OR “subarachnoid haemorrhage” OR “SAH”[tiab]) AND (“magnetic resonance”[tiab] OR “MR”[tiab] OR “MRI”[tiab]) AND (“aneurysm”[tiab] OR “aneurysmal”[tiab]). We filtered for “Humans” and for studies in the English language. Articles were selected for review if they met the following inclusion criteria:
  1. 1.

    Adult (> 16 years) patients with aSAH who underwent brain MRI after aneurysmal rupture either acutely (during initial hospitalization) and/or in the post-acute phase (following hospitalization and up to 1 year post-aSAH).

     
  2. 2.

    Initial or admission neurologic condition of patients was reported. At a minimum, this included one or more of: Hunt and Hess scale, World Federation of Neurologic Surgeons scale, Glasgow Coma Scale.

     
  3. 3.

    At least one functional outcome at hospital discharge and/or after discharge (e.g., 3, 6, or 12 months) was reported, e.g., modified Rankin Scale, Barthel Index, NIH Stroke Scale, Glasgow Outcome Scale, Glasgow Outcome Scale-Extended, Functional Independence Measure.

     
  4. 4.

    A description of MRI sequences utilized (e.g., fluid attenuation inversion recovery [FLAIR], gradient echo [GRE], diffusion-weighted imaging [DWI], diffusion tensor imaging [DTI], blood oxygen level dependent signal [BOLD], T2*WI, arterial spin labeling [ASL], magnetic resonance spectroscopy [MRS]) was included as well as a qualitative and/or quantitative account of MRI findings.

     
  5. 5.

    Description of MRI findings (criterion 4) was sufficient to evaluate their clinical impact (e.g., contained previously undocumented or insufficiently characterized findings that could change treatment or prognosis of aSAH patients).

     
Exclusion criteria were:
  1. 1.

    Non-human subjects research.

     
  2. 2.

    Non-relevant population (e.g., non-aneurysmal SAH, patients with unruptured aneurysms, mixed populations with and without aSAH, children).

     
  3. 3.

    Case reports (n < 2 patients).

     
  4. 4.

    Papers not containing original data (e.g., review articles, editorials, guidelines, letters to the editor).

     

Additional studies were then chosen on the basis of relevance to the topics covered in this review. For this type of retrospective study formal consent is not required.

MRI Can Assist with SAH Diagnosis

Diagnosing SAH is critical due to the high risk of early aneurysm rebleeding (4–13.6% in the first 24 h), which is associated with poor outcomes including death [17]. In addition, evidence of prior SAH also may influence outcome [18]. Nonetheless, diagnosis of SAH may be delayed or misdiagnosed in 5.4–21% of patients in recent studies [19, 20, 21, 22], which can lead to higher rates of poor outcomes [20, 23]. MRI is instrumental in diagnosing SAH particularly in cases where CT is ambiguous but clinical history may be compelling.

SAH was clearly detected on all FLAIR images performed within 2 days of SAH onset with sensitivity and specificity both inferred to be 100% [24] and was overall better than CT in detecting SAH when performed within 24 h of SAH onset (though no sensitivity or specificity was reported) [25]. In another study, FLAIR was much more sensitive than CT and GRE imaging when performed less than 4 days (sensitivities: 100% for FLAIR, 71.8% for CT, 37.5% for GRE) and 4–15 days (sensitivities: 100% for FLAIR, 50% for CT, 30% for GRE) after suspected onset of low-grade SAH [26].

In addition to FLAIR, GRE and susceptibility weighted imaging (SWI) may also be used to identify the presence of SAH (Fig. 1). For example, it has been shown that GRE has similar sensitivity to CT in revealing the presence of SAH acutely after ictus (sensitivities: 94% for GRE, 95% for CT); [27] further, this sequence and FLAIR were each more sensitive than CT when MRI is performed > 4 days after symptom onset (sensitivities: 100% for GRE, 87% for FLAIR, 75% for CT) [27]. SWI and CT may provide complementary information regarding SAH; for instance, in one study in which MRIs were performed approximately 1 week after traumatic brain injury, SWI did not detect hemorrhage in the basilar cisterns as well as CT but was better able to identify intraventricular hemorrhage [28]. And a recent study demonstrated that combining SWI and FLAIR resulted in a superior rate of SAH detection compared to CT when MRI was performed within 6 days of symptom onset (though sensitivity and specificity could not be calculated) [29]. Blood secondary to SAH was better identified on GRE than on FLAIR 3 months after SAH (sensitivities: 2% for FLAIR, 35% for GRE; specificities: 98% for FLAIR, 87% for GRE) [30]. Interestingly, if performed within 90 days of SAH onset, GRE may also permit calculation of the approximate day on which SAH occurred [31]. However, GRE and SWI have disadvantages, including susceptibility artifact from air–tissue interfaces [28] as well as inability to identify SAH if imaging is performed too early to allow for hemoglobin decomposition to deoxyhemoglobin [29].
Fig. 1

MRI can be more sensitive for SAH than CT. 15-year-old female who presented with likely aSAH though an aneurysm could not be found on initial vascular imaging. CT demonstrated hemorrhage in the basal cisterns. a CT at the level of the centrum semiovale does not appear to show any obvious blood products aside from hemorrhage associated with an external ventricular drain (black arrow). b At the same level, however, FLAIR shows multifocal loss of FLAIR suppression, including in the central sulci (white arrows) and in the precentral sulci (black arrows). c Consistent with FLAIR, SWI shows susceptibility in the central sulci (white arrows) as well as left precentral sulcus but not the right precentral sulcus (black arrows). The inability to identify SAH in the right precentral sulcus may be due to MRI imaging being performed too early to allow for hemoglobin decomposition to deoxyhemoglobin.

Source: Johns Hopkins Hospital

The double inversion recovery sequence, which utilizes 2 inversion recovery pulses to suppress white matter and cerebrospinal fluid, may also play a role in SAH diagnosis; it was found to have a higher sensitivity than CT, 2D FLAIR, 3D FLAIR, GRE, and SWI sequences for diagnosing SAH in patients undergoing MRI 14–16 days after symptom onset (sensitivity and specificity of the double inversion recovery sequence both inferred to be 100%) [32].

MRI Changes are Associated with Neurologic Severity

The relationship between severity of initial neurologic examination and MRI findings is detailed in Table 1 [12, 13, 14, 15, 16, 33, 34, 35, 36]. Presence and/or size of MRI lesions (e.g., DWI, FLAIR, T1, T2, T2*) was associated with more severe neurologic presentations in nearly all studies [12, 13, 14, 15, 16, 33, 36]. These associations were found both during acute hospitalization [12, 13, 14, 15, 16] (and prior to aneurysm securement except for 32 of 61 patients in one study [16]) and following discharge including up to 1 year post-aSAH [33, 35, 36]. For example, Fig. 2 provides an example of an aSAH patient who suffered multiple ischemic infarcts with an associated poor neurologic exam. The association between acute neurologic severity and DWI lesions was also noted in a recent systematic review [37]. Prior studies have also demonstrated an association between neurologic exam findings in the acute stage and the presence of cerebral infarcts on head CT [38, 39]. However, not all MRI lesions seem to be associated with initial neurologic examination: an MRI study evaluating cerebral volumes 1.5 years after aSAH did not find an association between clinical condition on admission and parenchymal volumes or lateral ventricular volumes [40]. Thus, MRI can help identify patients with, or at risk for, higher burdens of neurologic injury and therefore patients who may benefit from targeted therapeutic intervention in clinical trials.
Table 1

Studies evaluating the association between MRI findings and neurologic severity

Study

MRI timing

Sample size

Finding(s)

Hadeishi et al. [12]

Early

32

Patients with low-grade SAH [World Federation of Neurological Surgeons (WFNS) 1–2] had no DWI abnormalities

   

5 of 7 patients (71%) with high-grade SAH (WFNS 4–5) had DWI lesions

Wani et al. [13]

Early

16

DWI lesions were noted in 8 of 10 (80%) poor grade patients (HH3–4) and 1 of 6 (16.6%) good grade patients (HH1–2) (p < 0.05)

   

Focal neurologic deficits were seen in 8 of 9 (88.9%) patients with DWI lesions and in 0 patients with no DWI lesions (p < 0.05)

Sato et al. [14]

Early

38

Among patients grouped according to DWI lesion area [none (N), spotty (S, ≤ 10 mm2), and areal (A, > 10 mm2), within about 12 h of SAH onset, recovery to Hunt and Kosnik grade III was achieved in more patients in group N than in group A (57.1 vs. 0%; p = 0.023)]

De Marchis et al. [15]

Early

27

DWI lesion volume was 18-fold greater in Hunt–Hess (HH) 4–5 than in HH1–3 patients (p = 0.02)

   

Each 10 mL increase in DWI lesion volume was associated with 1 additional HH grade [OR 2.01 (95% (CI) 1.10–3.68; p = 0.02)]

   

FLAIR lesion volume was > 25-fold greater in HH4–5 than in HH1–3 patients (p = 0.02)

   

For each 10 mL increase in FLAIR lesion volume, the OR of a 1 HH grade increase was 1.34 (95% CI 1.06–1.68; p = 0.01)

Frontera et al. [16]

Early

61

The following associations were significant:

   

 HH and presence of DWI/ADC lesion (p = 0.039)

   

 HH and DWI/ADC lesion volume (p = 0.002)

   

 GCS and DWI/ADC lesion volume (p = 0.032)

Bendel et al. [34]

Late

77

Amygdalae and hippocampal volumes were similar in both the HH1–2 group and in the HH3–5 group (amygdalae: right 17.8 ± 4.6 vs. 20.2 ± 5.1, p = 0.04; left 18.2 ± 4.3 vs. 20.2 ± 2.5, p = 0.04; hippocampi: right 22.9 ± 4.2 vs. 22.7 ± 4.0, p = 0.79; left 21.1 ± 3.6 vs. 21.1 ± 3.9, p = 0.78). Volumes recorded as [volume of structure/intracranial area in reference slice] × 100

Bendel et al. [33]

Late

138

Parenchymal high-signal intensity lesions on T2- and intermediate-weighted MRI lesions were more often seen in HH4–5 as compared to HH1–3 (100 vs. 65.1%, p = 0.013)

Bendel et al. [35]

Late

76

Patients with HH3–4 and patients with Fisher 3–4 had greater maximal ventricular body width divided by maximal intracranial width compared to HH1–2 (0.29 ± 0.05 vs. 0.21 ± 0.06; p < 0.001) and Fisher 1–2 (0.25 ± 0.06 vs. 0.20 ± 0.06; p = 0.001), respectively

   

CSF/ICV was also greater for patients with Fisher 3–4 than for patients with Fisher 1–2 (37.13 ± 6.10 vs. 32.75 ± 7.75; p = 0.014) but was similar in HH3–4 and HH1–2 patients (37.15 ± 5.18 vs. 34.98 ± 7.54; p = 0.293)

Falter et al. [36]

Late

90

Greater WFNS scores (3.18 ± 1.21 vs. 2.30 ± 0.98; p = 0.0008) were associated with hemosiderin deposition on gradient echo sequence

MRI timing: early, during acute hospitalization. Late, following discharge from acute hospitalization and up to 1 year post-aSAH

ADC apparent diffusion coefficient, CI confidence interval, CSF cerebrospinal fluid, DWI diffusion weighted imaging, FLAIR fluid attenuation inversion recovery, GCS Glasgow Coma Scale, HH Hunt and Hess Scale, ICV total intracranial volume, MRI magnetic resonance imaging, OR odds ratio, SAH subarachnoid hemorrhage, WFNS World Federation of Neurologic Surgeons scale

Fig. 2

Infarcts on MRI can be associated with poor neurologic exam in aSAH. 39-year-old male who presented with Hunt–Hess grade 4 SAH status post coiling of anterior communicating aneurysm with course complicated by multiple infarcts as seen on DWI (a) and FLAIR (b) likely secondary to increased intracranial pressure and herniation with best exam showing no movement on the right and localization on the left.

Source: Johns Hopkins Hospital

MRI Changes are Associated with Post-SAH Outcomes

Studies reporting on functional outcomes and their relationship to MRI findings [12, 13, 14, 15, 16, 18, 33, 34, 35, 41, 42, 43, 44, 45] (Table 2) indicate that pathological changes identified on brain MRI were consistently associated with worse functional outcomes [12, 13, 14, 15, 16, 18, 33, 34, 35, 41, 42, 43, 44, 45]. Notably, these MRI findings were very diverse and included lesions, atrophic changes, and alterations in connectivity and were noted on many different sequences (DWI, FLAIR, T1, T2, T2*, and fMRI). Infarcts and atrophy identified on CT have been linked with poor outcome in other studies of aSAH patients [38, 39, 46, 47, 48, 49]. Research is needed to confirm the types and locations of lesions that are associated with specific outcomes following aSAH and to determine how MRI data can contribute to prediction models and especially to optimization of management of aSAH patients.
Table 2

Studies evaluating the association between MRI findings and outcome

Study

MRI timing

Sample size

Outcome measured

Finding(s)

Hadeishi et al. [12]

Early

32

Not clearly described

In 5 of 7 WFNS 4–5 patients with DWI lesions on admission, patients had outcomes described as favorable or associated with memory disturbance or disorientation. The other 2 high-grade SAH patients did not have DWI abnormalities but died within 2 days of SAH onset (no p value provided)

Mayer et al. [45]

Early

2

Neurologic examination

One patient in vasospasm had improved left-sided weakness after IA nimodipine along with resolved findings on PWI and no changes on DWI

    

A second patient became unconscious and tetraplegic in the setting of vasospasm despite IA nimodipine with bilateral occipital and right paramedian pontine infarcts on MRI and deficits notable for a quadrantanopia and cranial nerve VI palsy 6 months later

Wani et al. [13]

Early

16

Dependence for daily needs

7 of 9 (77.8%) patients with DWI lesions and only 2 of 7 (28.6%) patients with no DWI lesions had a poor clinical outcome (poor clinical outcome = dependent on others) (no p value provided)

Sato et al. [14]

Early

38

GOS

When DWI findings were grouped into none (N), spotty (S, ≤ 10 mm2), and areal (A, > 10 mm2), good outcomes (classified as good recovery and moderate disability) were greater in group N than in group A (100 vs. 0%; p < 0.01) but not in group S (53.8%; no p value provided for this comparison)

De Marchis et al. [15]

Early

27

mRS

After adjusting for Hunt–Hess grade, age, and modified Fisher score, DWI lesion volume was not significantly associated with mRS at 3 months (no p value provided)

    

After adjusting for Hunt–Hess grade, age, and modified Fisher score, each 10 mL increase in FLAIR lesion volume was associated with 139% greater odds of an increase of 1 point on the mRS (OR 2.39, 95% CI 1.13–5.07; p = 0.02)

Frontera et al. [16]

Early

61

mRS, BI, Telephone Interview for Cognitive Status

After adjusting for Hunt–Hess grade, age, and aneurysm size, DWI/ADC lesion volume was not associated with mRS 4–6 (p = 0.066) but final infarct volume on FLAIR was (p = 0.019)

    

DWI/ADC lesion presence, volume, and number were associated with worse BI at 3 months (p = 0.029, p = 0.004, p = 0.046, respectively)

    

DWI/ADC lesion presence, volume, and number were not associated with the Telephone Interview for Cognitive Status at 3 months (all p > 0.05)

Maher et al. [41]

Early

14

CEFS

Patients were classified as executively impaired or unimpaired based on the CEFS, and impaired patients had more DWI lesions than unimpaired patients (15 vs. 1, p = 0.037)

Koivisto et al. [43]

Late

87

GOS

After also adjusting for presence of non-parental ischemic lesion, size of parental artery ischemic lesion, size of non-parental territory ischemic lesion, and greater ventricular-intracranial width ration, presence of parental artery territory ischemic lesion (OR 6.20 95% CI 1.67–23.05, p = 0.006) and deficit secondary to intracerebral hematoma noted preoperatively (OR 4.23 95% CI 1.16–15.39, p = 0.029) were independent predictors of poor outcome at 12 months

Imaizumi et al. [18]

Late

58

Karnofsky scale

Each hemisphere was divided into 5 regions: frontal, parietal, temporal, occipital, and sylvian fissure:

    

 Karnofsky scale score of patients with < 4 hemosiderin regions was greater than that of those with ≥ 4 hemosiderin regions (92 ± 15 vs. 73 ± 26%, p = 0.0004)

    

 In multivariable analysis that also included age ≥ 54 years, Fisher grade 3, female sex, GCS ≤ 14, hydrocephalus, cisternal drainage, walking SAH (patient walked to hospital), and interval between SAH onset and MRI < 4 months, Karnofsky scale score ≤ 80% was significantly associated with hemosiderin deposition extent (≥ 4 vs. < 4 regions; OR 12.8 95% CI 1.97–83.3, p = 0.0077)

Bendel et al. [34]

Late

77

Cognitive battery

Smaller hippocampal volumes correlated with neuropsychological impairments on several tests [e.g., verbal IQ, performance IQ, Boston Naming, Stroop A, Stroop B, Stroop C, Trail-Making A, Trail-Making B (all p < 0.01)]

    

Amygdaloid volumes did not correlate with neuropsychological test results (all p > 0.025, the level of significance used in this paper)

Bendel et al. [33]

Late

138

Cognitive battery

Several neuropsychological tests were significantly associated with parenchymal, retraction, and ischemic lesion volumes [e.g., performance intelligence quotient, delayed story recall, delay visual reproduction, Boston naming, Stroop C, Trail-making A, and Trail-making B were significantly correlated with all 3 types of lesion volumes (all p < 0.05)]

Bendel et al. [44]

Late

37

Cognitive battery

In patients with GOS 4–5, patients with executive function impairment had lower GM/ICV ratios (0.342 ± 0.015) and higher CSF/ICV ratios (0.433 ± 0.032) than patients with no executive function deficit (GM/ICV 0.366 ± 0.028, p = 0.014 and CSF/ICV 0.403 ± 0.033, p = 0.013)

Bendel et al. [35]

Late

76

GOS, cognitive battery

As compared to patients with GOS 5, patients with GOS 2–4 had larger maximal ventricular body width divided by maximal intracranial width (0.28 ± 0.07 vs. 0.23 ± 0.06; p < 0.028)

    

Patients with neuropsychological deficits had decreased GM/ICV ratios (36.00 ± 4.29 vs. 39.59 ± 5.19; p = 0.003) than in patients without neuropsychological deficits

    

Greater CSF/ICV ratios were found in patients (31.39 ± 7.52 vs. 26.56 ± 4.05; p = 0.027) compared with healthy controls

de Bresser et al. [42]

Late

55

mRS

After adjusting for age and sex, greater lateral ventricular volume and smaller parenchymal volume were associated with worse outcome by mRS ≥ 2 at 6 months (OR 7.4, 95% CI 1.6–33.5 and OR 38.8, 95% CI 4.6–329.0, respectively)

    

After also adjusting for infarct volume, only parenchymal volume remained associated with worse outcome (mRS ≥ 2) at 6 months (OR 36.8, 95% CI 3.9–346.1)

Maher et al. [41]

Late

14

CEFS

In 14 patients and 14 controls in which functional connectivity of 6 regions-of-interest in the frontoparietal network on resting-state fMRI was examined and who were classified as executively impaired or unimpaired based on the CEFS:

    

 Compared to controls, impaired patients had greater connectivity between left dorsolateral prefrontal cortex and precuneus and also between right dorsolateral prefrontal cortex and left superior frontal gyrus (both p < 0.05)

    

 Compared to unimpaired patients, impaired patients had greater connectivity between right dorsolateral prefrontal cortex and left superior frontal gyrus (p < 0.05)

    

 No significant differences were found between unimpaired patients and controls (p > 0.05)

MRI timing: early, during acute hospitalization. Late, following discharge from acute hospitalization and up to 1 year post-aSAH. All associations are unadjusted unless otherwise indicated

ADC apparent diffusion coefficient, BI Barthel Index, CEFS Composite Executive Function Score, CI confidence interval, CSF cerebrospinal fluid, DWI diffusion weighted imaging, FLAIR fluid attenuation inversion recovery, GCS Glasgow Coma Scale, GM gray matter, GOS Glasgow Outcome Scale, ICV total intracranial volume, IQ intelligence quotient, MRI magnetic resonance imaging, mRS modified Rankin Scale, OR odds ratio, PWI perfusion weighted imaging, SAH subarachnoid hemorrhage, WFNS World Federation of Neurologic Surgeons Scale, WM white matter

Greater Incidence of MRI Lesions in Surgical Versus Endovascular Aneurysm Treatment

MRI generally revealed a higher incidence of lesions in patients who underwent surgical, as opposed to endovascular, aneurysm treatment [33, 34, 35, 43, 44, 50] (Table 3). In particular, when compared to patients treated with endovascular therapy, patients who underwent surgical aneurysm treatment had a higher burden of cerebral infarction and greater likelihood of developing cerebral atrophy [33, 34, 35, 43, 44, 50]. However, it possible that other differences between groups such as earlier treatment in endovascular cases (since risk of rebleeding is highest immediately after aSAH onset) and complications such as vasospasm could also contribute to this finding [33, 43, 50]. Prior studies using CT in subarachnoid hemorrhage did not find a difference in infarcts between endovascular and surgical treatment [51, 52], possibly due to greater sensitivity of MRI in detecting brain lesions in general. Overall, findings shown on MRI seem consistent with the achievement of similar or higher levels of functional independence seen in aSAH patients receiving endovascular therapy compared to conventional surgical repair in clinical trials [53, 54].
Table 3

Studies evaluating MRI findings in surgical versus endovascular aneurysm treatment groups

Study

MRI timing

Sample size

Finding(s)

Koivisto et al. [43]

Late

87

Parental artery territory ischemic deficits [8/40 (20%) endovascular vs. 21/47 (45%) surgical, p = 0.018] and superficial brain retraction deficits [4/40 (10%) endovascular vs. 21/47 (45%) surgical, p < 0.001] were different between the endovascular and surgery groups

Hadjivassiliou et al. [50]

Late

46

Focal encephalomalacia due to the surgical approach was found only in the clipped group (19/23 patients in the surgical group vs. 0/23 in the endovascular group; p < 0.001)

   

Number of patients with infarcts [all of which occurred in the vascular distribution of the aneurysm] differed [20 (87%) surgical patients vs. 13 (57%) endovascular patients, p < 0.05]

   

Large vessel infarcts occurred similarly in both groups (8 in surgical group vs. 9 in endovascular group, not significant but no p value provided)

Bendel et al. [34]

Late

77

Right hippocampal (22.50 ± 3.73 vs. 24.71 ± 3.71, p = 0.024), left hippocampal (20.81 ± 3.70 vs. 23.67 ± 3.64, p = 0.004), and right amygdaloid (16.31 ± 4.82 vs. 20.96 ± 3.91, p < 0.001) volumes were smaller in surgical patients as compared to controls; left amygdaloid volumes were not different between these groups

   

Volumes of right hippocampus (23.56 ± 4.43 vs. 24.71 ± 3.71, p = 0.281), left hippocampus (21.79 ± 3.76 vs. 23.67 ± 3.64, p = 0.055), right amygdala (20.7 ± 4.38 vs. 20.96 ± 3.91, p = 0.790), and left amygdala (19.66 ± 2.90 vs. 20.52 ± 3.67, p = 0.281) were similar between age- and sex-matched controls and endovascular patients

Bendel et al. [33]

Late

138

High-signal intensity parenchymal lesions on T2- and intermediate-weighted images were more frequent in the surgical group (78.9% surgical vs. 56.7% endovascular, p = 0.005)

   

Infarction in the vascular territory of the ruptured aneurysm was more frequent in the surgical group (46.5 vs. 22.4%, p = 0.003)

   

Of patients with infarcts, mean infarct volume was not different between the 2 groups (17.6 cm3 endovascular vs. 20.9 cm3 surgical, p = 0.209)

   

Including all patients, mean infarct volume in the distribution of the parental artery was greater in the surgical group (9.7 cm3 surgical vs. 3.9 cm3 endovascular, p = 0.002)

Bendel et al. [44]

Late

37

Of patients with a ruptured anterior cerebral artery aneurysm and Glasgow Outcome Scale of 4 or 5:

   

 Total parenchymal lesion volume was greater after surgical than after endovascular treatment (15.3 ± 10.1 cc after surgical treatment vs. 4.2 ± 11.8 cc after endovascular treatment, p < 0.001)

   

 GM/ICV ratio was decreased in the surgical patients as compared to age- and sex-matched controls (0.356 ± 0.029 vs. 0.378 ± 0.027, p = 0.016) but CSF/ICV ratio was increased in the surgical patients as compared to controls (0.417 ± 0.032 vs. 0.384 ± 0.033, p = 0.003)

   

 There were no differences in GM/ICV (0.367 ± 0.025 vs. 0.378 ± 0.027, p = 0.155) and CSF/ICV ratios (0.402 ± 0.035 vs. 0.384 ± 0.033, p = 0.135) between the endovascular treated patients and controls

   

 Executive function impairment was more common in surgically treated patients than in endovascularly treated patients (72.7 vs. 27.3%, p = 0.033); otherwise there were no differences in tests for intelligence, memory, verbal skills (p = 0.266, 0.900, and 0.763, respectively)

Bendel et al. [35]

Late

76

Maximal ventricular body width divided by maximal intracranial width was comparable between treatment groups: 0.23 ± 0.1 after endovascular therapy vs. 0.23 ± 0.1 after surgery (no p value provided)

   

CSF/ICV was 34.49 ± 7.7 after endovascular treatment and 36.78 ± 6.0 after surgical treatment (p = 0.289)

   

Total parenchymal lesion volume (13.12 ± 15.38 cm3 vs. 4.25 ± 12.82 cm3; p < 0.001) and ischemic brain volume (6.86 ± 13.62 cm3 vs. 3.48 ± 12.15 cm3; p = 0.004) were larger in patients who underwent surgery than those undergoing endovascular therapy

   

Cognitive impairment was similar in patients who underwent surgery vs. endovascular treatment (69.4 vs. 52.0%; p = 0.069)

MRI timing: early, during acute hospitalization. Late, following discharge from acute hospitalization and up to 1 year post-aSAH

CSF cerebrospinal fluid, GM gray matter, ICV total intracranial volume, MRI magnetic resonance imaging

Inconsistent Association Between MRI Lesions and Vasospasm and Delayed Cerebral Ischemia

Studies examining the relationship between vasospasm or delayed cerebral ischemia and MRI changes are in Table 4 [13, 14, 16, 18, 33, 34, 43, 55, 56, 57]. In general, while some studies reported a higher incidence of MRI lesions (including perfusion weighted imaging (PWI)/DWI mismatch, infarcts on DWI, hemosiderin deposition on T2*, lesions on T2) in patients with symptomatic vasospasm [16, 18, 43, 55, 56, 58, 59, 60, 61, 62], others did not (including infarcts on DWI and volumes on T1) [13, 14, 34]. Thus, DWI lesions do not always appear to be associated with vasospasm and delayed cerebral ischemia. Of particular note in assessing vasospasm and delayed cerebral ischemia is the use of using PWI [56, 58] (Fig. 3) and PWI/DWI mismatch [55, 59, 60, 61, 62], which are often used clinically and have some support in the literature [55, 58, 59, 60, 61, 62]. It should also be acknowledged that magnetic resonance angiography (MRA) may also be used to detect areas of stenosis seen in vasospasm due to aSAH and in one study was shown to have a sensitivity of 92% and specificity of 97% in detecting vasospasm [63].
Table 4

Studies evaluating the association between MRI findings and vasospasm/delayed cerebral ischemia

Study

MRI timing

Sample size

Finding(s)

Rordorf et al. [59]

Early

8

All 6 patients with symptomatic and angiographic vasospasm had an area of abnormal MTT that was much larger than the area of DWI lesion

   

MRI images were normal in an asymptomatic patient with angiographic vasospasm and in a patient with no symptomatic vasospasm and no angiographic vasospasm (no p values provided)

Leclerc et al. [58]

Early

11

PWI showed evidence of hypoperfusion in 7 of 11 aSAH patients, though single photon emission computed tomography showed hypoperfusion in 2 additional patients

Hertel et al. [60]

Early

20

Areas of PWI changes correlated with neurologic deficits and were greater than areas of DWI changes

Beck et al. [61]

Early

10

In all cases with a PWI/DWI mismatch, angiography confirmed severe vasospasm

   

Complete prevention of infarction resulted from balloon angioplasty-induced reduction in the perfusion delay from 6.2 ± 1 s (mean ± standard error of the mean) to 1.5 ± 0.45 s

   

Reduction in the delay from 6.2 ± 2.7 to 4.1 ± 1.9 s resulted in only small infarcts in vessel territories

   

Without angioplasty, perfusion delay remained or increased, and infarction of a territory occurred

Ohtonari et al. [62]

Early

17

Of 17 aSAH patients who underwent MRI, 3 developed symptomatic vasospasm and all 3 were noted to have PWI/DWI mismatch

   

Following therapy including induced hypertension, the mismatch resolved and T2 sequences showed no new infarction

Wani et al. [13]

Early

16

No difference in angiographic vasospasm between patients with DWI lesions and without (p > 0.05)

Sato et al. [14]

Early

38

No significant difference in symptomatic vasospasm among patients divided into 3 groups based on DWI findings [none (N), spotty (S, ≤ 10 mm2), and areal (A, > 10 mm2)] (no p value provided)

Vatter et al. [55]

Early

25

Compared to brain segments without risk, those with risk (risk = PWI/DWI mismatch) had a significantly greater infarct rate (37 vs. 4%) (p < 0.01)

Frontera et al. [16]

Early

61

New DWI/ADC lesions were associated with symptomatic vasospasm that was confirmed angiographically (p = 0.020)

Kamran et al. [56]

Early

3

In 3 aSAH patients, the region of PBV abnormality as measured on C-arm flat detector CT was similar to areas of decreased perfusion on MRI. These abnormalities also seemed to correspond to angiographic vasospasm except no angiographic vasospasm was noted in one case

Koivisto et al. [43]

Late

87

Clinical symptoms of vasospasm were associated with ischemic lesions in the parental artery territory (p = 0.0001)

   

Clinical symptoms of vasospasm were associated with ischemic lesions in other locations (p < 0.001)

Imaizumi et al. [18]

Late

58

Symptomatic vasospasm was significantly associated with hemosiderin deposition extent (≥ 4 vs. < 4 regions; p = 0.0041) (in each hemisphere, the subarachnoid space was divided into 5 regions: frontal, parietal, temporal, occipital, and sylvian fissure)

Bendel et al. [34]

Late

77

Amygdaloid and hippocampal volumes on T1 were not associated with symptomatic vasospasm (no p value provided)

Bendel et al. [33]

Late

138

T2- and intermediate-weighted parenchymal lesions at 1 year were more common in Fisher III-IV patients than Fisher 0-II patients (77.3 vs. 52.0%, p = 0.002; Fisher scale can predict risk of symptomatic vasospasm [96])

Rodriguez-Regent et al. [57]

Late

47

In a study to investigate whether CT perfusion parameter variation between days 0 and 4 after aSAH predicts DCI (defined as FLAIR MRI evidence of infarct at 3 months):

   

 10 of 47 aSAH patients had DCI+

   

 Between DCI + and DCI- patients, change in MTT (percent variation 14.0 + 19.5% and − 6.2 + 16.4%, respectively, p < 0.0001) and CBF (percent variation − 31.4 + 12.8% and 22.7 + 63.7%, respectively, p < 0.0001) between days 0 and 4 were significantly different

   

 Change in MTT and CBF were each independent predictors of DCI in multivariate analysis (per 20% decrease in change in CBF: OR 1.91 95% CI 1.13–3.23, p = 0.02; per 20% increase in change in MTT: OR 14.70 95% CI 4.85–44.52, p < 0.0001)

MRI timing: early, during acute hospitalization. Late, following discharge from acute hospitalization and up to 1 year post-aSAH

ADC apparent diffusion coefficient, CBF cerebral blood flow, CI confidence interval, DCI delayed cerebral ischemia, DWI diffusion weighted imaging, FLAIR fluid-attenuated inversion recovery, MRI magnetic resonance imaging, MTT mean transit time, OR odds ratio, PBV parenchymal blood volume, PWI perfusion weighted imaging, SAH subarachnoid hemorrhage

Fig. 3

PWI sequence can be useful in aSAH. 67-year-old female with likely aSAH though an aneurysm could not be found on initial digital subtraction angiography. MRI obtained 3 days after ictus showed decreased perfusion on the left on time-to-peak sequence with both expressive and receptive aphasia noted on exam. Interestingly, transcranial Doppler was unremarkable; no further vascular imaging was obtained that day.

Source: Johns Hopkins Hospital

Emerging Evidence that MRI Can Identify Aneurysm Occlusion and Possibly Rupture

Detection of aneurysm occlusion after intervention presents another potential use for MRI. While cerebral angiography remains the reference standard and magnetic resonance angiography has demonstrated high sensitivity to residual aneurysm flow in several studies [64, 65] including new techniques such as zTE MRA that are less sensitive to artifacts from coils [66], case reports and case series have shown that T1-weighted sequences such as BRAVO (Brain Volume, which allows for multiplanar reconstructions) and black blood (which has been shown to effectively image vessel walls [67]) may be useful in demonstrating aneurysm occlusion [68, 69, 70, 71]. Furthermore, according to preliminary studies that use vessel wall imaging, sequences such as black blood and turbo spin-echo may have utility in identifying ruptured aneurysms [72, 73, 74]. The black blood sequence has even been shown to detected ruptured aneurysms in patients who have multiple aneurysms [72].

Newer MRI Sequences Can Reveal Physiological and Structural Changes in aSAH

Newer pulse sequences not routinely included in clinical brain MRI scans have the potential to shed light on important changes in the brain that can be seen in aSAH. These include ASL, vascular space occupancy (VASO), BOLD, DTI, and MRS.

ASL relies on magnetically labeled arterial blood water protons as an endogenous tracer to map cerebral blood flow (CBF), can be obtained quickly, may allow quantification of absolute CBF, and has high inter-rater reliability [75, 76, 77, 78, 79, 80, 81]. Because ASL does not require an exogenous tracer, unlike other types of perfusion imaging, repeated measurements may be obtained over time and it can be used in cases where kidney dysfunction and lack of intravenous access are concerns [78, 81]. Several studies have shown that decreases in CBF as shown on ASL correlate with either clinical symptoms or with symptomatic and angiographic vasospasm [82, 83, 84]. On the other hand, in a study evaluating whether decreased CBF on ASL may be correlated to neurologic exam in 12 aSAH patients, no association was found possibly due to a small sample size and variable acquisition time of ASL after SAH onset [79]. In addition, given that hypoperfusion (in the interictal phase) or hyperperfusion (in the ictal phase) seen on ASL can denote seizures [85, 86], it is possible that this sequence could help identify aSAH-related seizures.

VASO is a relatively new MRI sequence that uses T1 differences between blood and tissue. With nulling of the intravascular signal, MRI signal will decrease if blood volume increases and vice versa. While it has not yet been used in aSAH, it has the potential to provide quantitative mapping of cerebral blood volume in aSAH patients [87].

The BOLD signal, which is proportional to tissue changes in deoxyhemoglobin, is widely used to infer regional functional activation [88, 89]. One study suggested that impaired cerebrovascular reserve derived from BOLD and delayed cerebral ischemia can occur in the same vascular territory in aSAH patients [90]. A BOLD study performed in SAH patients (not all of whom had SAH due to aneurysmal rupture) found impairment in working memory task performance along with increased BOLD activity in many areas of the cortex [91]. Further, data from BOLD plus additional sequences such as VASO can be combined to compute oxygen extraction fraction and cerebral metabolic rate of oxygen—variables of interest in identifying regional or global ischemia [92, 93, 94].

DTI utilizes the diffusivity vectors of water molecules to determine the integrity and architecture of white matter tracts [95]. DTI has been acquired in SAH patients thus demonstrating its feasibility in this patient population [96, 97]. For example, a recent study in which non-traumatic SAH patients underwent DTI in the first 72 h suggested that this sequence could assist with prognosis since lower fractional anisotropy values in the cerebellum increased the odds of delayed cerebral ischemia occurrence, and greater apparent diffusion coefficient values in the frontal centrum semiovale demonstrated a trend toward worse functional outcome at 3 months [98].

MRS, which analyzes regional metabolite concentrations in the brain, has been used in patients with aSAH, in whom it may detect ischemia [99] although a more recent study appeared to cast doubt on this conclusion [100]. MRS has also revealed global metabolic changes in SAH patients with or without perfusion deficits or ischemia [101]. Overall, multiple lines of evidence indicate that multimodal MRI is emerging as a highly versatile tool in the evaluation of patients with SAH that could plausibly be used to guide management and outcome prediction.

Conclusion

In patients with aSAH, MRI identifies brain abnormalities associated with neurologic presentation, outcomes, and aneurysm treatment, and thus has the potential to strengthen knowledge of aSAH pathophysiology and to guide treatment and outcome prediction. Moreover, MRI can be used to assist with diagnosis of SAH and may help with detecting aneurysm occlusion and rupture. Newer pulse sequences have the potential to reveal structural and physiological changes in aSAH that could lead to significant improvements in aSAH management. Finally, it should be acknowledged that MRI may not always be possible in this population given their critically ill nature or imaging artifacts such as patient motion. Additional research is needed to validate pragmatic MRI-based biomarkers that could be used in clinical practice and as endpoints in clinical trials, with the goal of improving outcome for patients with aSAH.

Notes

Compliance with Ethical Standards

Funding

This study received no funding.

Conflict of interest

The authors declare that they have no conflicts of interest.

References

  1. 1.
    Zacharia BE, Hickman ZL, Grobelny BT, et al. Epidemiology of aneurysmal subarachnoid hemorrhage. Neurosurg Clin N Am. 2010;21:221–33.CrossRefPubMedGoogle Scholar
  2. 2.
    Carpenter KLH, Czosnyka M, Jalloh I, et al. Systemic, local, and imaging biomarkers of brain injury: more needed, and better use of those already established? Front Neurol. 2015;6:1–20.CrossRefGoogle Scholar
  3. 3.
    de Oliveira Manoel AL, Mansur A, Murphy A, et al. Aneurysmal subarachnoid haemorrhage from a neuroimaging perspective. Crit Care. 2014;18:557.CrossRefPubMedPubMedCentralGoogle Scholar
  4. 4.
    Soher BJ, Dale BM, Merkle EM. A review of MR physics: 3T versus 1.5T. Magn Reson Imaging Clin N Am. 2007;15:277–90.CrossRefPubMedGoogle Scholar
  5. 5.
    Bitar R, Leung G, Perng R, et al. MR pulse sequences: what every radiologist wants to know but is afraid to ask. RadioGraphics. 2006;26:513–37.CrossRefPubMedGoogle Scholar
  6. 6.
    Hirano T. Searching for salvageable brain: the detection of ischemic penumbra using various imaging modalities? J Stroke Cerebrovasc Dis. 2014;23:795–8.CrossRefPubMedGoogle Scholar
  7. 7.
    Brix G, Nekolla EA, Borowski M, et al. Radiation risk and protection of patients in clinical SPECT/CT. Eur J Nucl Med Mol Imaging. 2014;41(Suppl 1):S125–36.CrossRefPubMedGoogle Scholar
  8. 8.
    Marshall SA, Kathuria S, Nyquist P, et al. Noninvasive imaging techniques in the diagnosis and management of aneurysmal subarachnoid hemorrhage. Neurosurg Clin N Am. 2010;21:305–23.CrossRefPubMedGoogle Scholar
  9. 9.
    Expert Panel on MR Safety, Kanal E, Barkovich AJ, et al. ACR guidance document on MR safe practices: 2013. J Magn Reson Imaging. 2013;37:501–30.CrossRefGoogle Scholar
  10. 10.
    Hennemeyer CT, Wicklow K, Feinberg DA, et al. In vitro evaluation of platinum Guglielmi detachable coils at 3 T with a porcine model: safety issues and artifacts. Radiology. 2001;219:732–7.CrossRefPubMedGoogle Scholar
  11. 11.
    Khursheed F, Rohlffs F, Suzuki S, et al. Artifact quantification and tractography from 3T MRI after placement of aneurysm clips in subarachnoid hemorrhage patients. BMC Med Imaging. 2011;11:19.CrossRefPubMedPubMedCentralGoogle Scholar
  12. 12.
    Hadeishi H, Suzuki A, Yasui N, et al. Diffusion-weighted magnetic resonance imaging in patients with subarachnoid hemorrhage. Neurosurgery. 2002;50:741–8.CrossRefPubMedGoogle Scholar
  13. 13.
    Wani AA, Phadke R, Behari S, et al. Role of diffusion-weighted MRG in predicting outcome in subarachnoid hemorrhage due to anterior communicating artery aneurysms. Turk Neurosurg. 2008;18:10–6.PubMedGoogle Scholar
  14. 14.
    Sato K, Shimizu H, Fujimura M, et al. Acute-stage diffusion-weighted magnetic resonance imaging for predicting outcome of poor-grade aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow Metab. 2010;30:1110–20.CrossRefPubMedPubMedCentralGoogle Scholar
  15. 15.
    De Marchis GM, Filippi CG, Guo X, et al. Brain injury visible on early MRI after subarachnoid hemorrhage might predict neurological impairment and functional outcome. Neurocrit Care. 2015;22:74–81.CrossRefPubMedGoogle Scholar
  16. 16.
    Frontera JA, Ahmed W, Zach V, et al. Acute ischaemia after subarachnoid haemorrhage, relationship with early brain injury and impact on outcome: a prospective quantitative MRI study. J Neurol Neurosurg Psychiatry. 2015;86:71–8.CrossRefPubMedGoogle Scholar
  17. 17.
    Connolly ES, Rabinstein AA, Carhuapoma JR, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/american Stroke Association. Stroke. 2012;43:1711–37.CrossRefPubMedGoogle Scholar
  18. 18.
    Imaizumi T, Chiba M, Honma T, et al. Detection of hemosiderin deposition by T2*-weighted MRI after subarachnoid hemorrhage. Stroke. 2003;34:1693–8.CrossRefPubMedGoogle Scholar
  19. 19.
    Vannemreddy P, Nanda A, Kelley R, et al. Delayed diagnosis of intracranial aneurysms: confounding factors in clinical presentation and the influence of misdiagnosis on outcome. South Med J. 2001;94:1108–11.CrossRefPubMedGoogle Scholar
  20. 20.
    Kowalski RG, Claassen J, Kreiter KT, et al. Initial misdiagnosis and outcome after subarachnoid hemorrhage. JAMA. 2004;291:866–9.CrossRefPubMedGoogle Scholar
  21. 21.
    Vermeulen MJ, Schull MJ. Missed diagnosis of subarachnoid hemorrhage in the emergency department. Stroke. 2007;38:1216–21.CrossRefPubMedGoogle Scholar
  22. 22.
    Inagawa T. Delayed diagnosis of aneurysmal subarachnoid hemorrhage in patients: a community-based study. J Neurosurg. 2011;115:707–14.CrossRefPubMedGoogle Scholar
  23. 23.
    Siddiq F, Chaudhry SA, Tummala RP, et al. Factors and outcomes associated with early and delayed aneurysm treatment in subarachnoid hemorrhage patients in the United States. Neurosurgery. 2012;71:670–7.CrossRefPubMedGoogle Scholar
  24. 24.
    Noguchi K, Ogawa T, Inugami A, et al. Acute subarachnoid hemorrhage: MR imaging with fluid-attenuated inversion recovery pulse sequences. Radiology. 1995;196:773–7.CrossRefPubMedGoogle Scholar
  25. 25.
    Shimoda M, Hoshikawa K, Shiramizu H, et al. Problems with diagnosis by fluid-attenuated inversion recovery magnetic resonance imaging in patients with acute aneurysmal subarachnoid hemorrhage. Neurol Med Chir (Tokyo). 2010;50:530–7.CrossRefGoogle Scholar
  26. 26.
    da Rocha AJ, da Silva CJ, Gama HPP, et al. Comparison of magnetic resonance imaging sequences with computed tomography to detect low-grade subarachnoid hemorrhage. J Comput Assist Tomogr. 2006;30:295–303.CrossRefPubMedGoogle Scholar
  27. 27.
    Mitchell P, Wilkinson ID, Hoggard N, et al. Detection of subarachnoid haemorrhage with magnetic resonance imaging. J Neurol Neurosurg Psychiatry. 2001;70:205–11.CrossRefPubMedPubMedCentralGoogle Scholar
  28. 28.
    Wu Z, Li S, Lei J, An D, Haacke EM. Evaluation of traumatic subarachnoid hemorrhage using susceptibility-weighted imaging. AJNR Am J Neuroradiol. 2010;31:1302–10.CrossRefPubMedPubMedCentralGoogle Scholar
  29. 29.
    Verma RK, Kottke R, Andereggen L, et al. Detecting subarachnoid hemorrhage: comparison of combined FLAIR/SWI versus CT. Eur J Radiol. 2013;82:1539–45.CrossRefPubMedGoogle Scholar
  30. 30.
    Mulé S, Soize S, Benaissa A, et al. Detection of aneurysmal subarachnoid hemorrhage 3 months after initial bleeding: evaluation of T2* and FLAIR MR sequences at 3 T in comparison with initial non-enhanced CT as a gold standard. J Neurointerv Surg. 2016;8:813–8.CrossRefPubMedGoogle Scholar
  31. 31.
    Inoue T, Takada S, Shimizu H, et al. Signal changes on T2*-weighted magnetic resonance imaging from the acute to chronic phases in patients with subarachnoid hemorrhage. Cerebrovasc Dis. 2013;36:421–9.CrossRefPubMedGoogle Scholar
  32. 32.
    Hodel J, Aboukais R, Dutouquet B, et al. Double inversion recovery MR sequence for the detection of subacute subarachnoid hemorrhage. Am J Neuroradiol. 2015;36:251–8.CrossRefPubMedGoogle Scholar
  33. 33.
    Bendel P, Koivisto T, Könönen M. MR imaging of the brain 1 year after aneurysmal subarachnoid hemorrhage: randomized study comparing surgical with endovascular treatment1. Radiology. 2008;246:543–52.CrossRefPubMedGoogle Scholar
  34. 34.
    Bendel P, Koivisto T, Hänninen T, et al. Subarachnoid hemorrhage is followed by temporomesial volume loss: MRI volumetric study. Neurology. 2006;67:575–82.CrossRefPubMedGoogle Scholar
  35. 35.
    Bendel P, Koivisto T, Äikiä M, et al. Atrophic enlargement of CSF volume after subarachnoid hemorrhage: correlation with neuropsychological outcome. Am J Neuroradiol. 2010;31:370–6.CrossRefPubMedGoogle Scholar
  36. 36.
    Falter B, Wiesmann M, Freiherr J, et al. Frequency and appearance of hemosiderin depositions after aneurysmal subarachnoid hemorrhage treated by endovascular therapy. Neuroradiology. 2015;57:999–1006.CrossRefPubMedGoogle Scholar
  37. 37.
    van der Kleij LA, De Vis JB, Olivot J, et al. Magnetic resonance imaging and cerebral ischemia after aneurysmal subarachnoid hemorrhage. Stroke. 2017;48:239–45.CrossRefPubMedGoogle Scholar
  38. 38.
    Jabbarli R, Reinhard M, Niesen W-D, et al. Predictors and impact of early cerebral infarction after aneurysmal subarachnoid hemorrhage. Eur J Neurol. 2015;22:941–7.CrossRefPubMedGoogle Scholar
  39. 39.
    Schmidt JM, Rincon F, Fernandez A, et al. Cerebral infarction associated with acute subarachnoid hemorrhage. Neurocrit Care. 2007;7:10–7.CrossRefPubMedGoogle Scholar
  40. 40.
    de Bresser J, Schaafsma JD, Luitse MJA, et al. Quantification of structural cerebral abnormalities on MRI 18 months after aneurysmal subarachnoid hemorrhage in patients who received endovascular treatment. Neuroradiology. 2015;57:269–74.CrossRefPubMedGoogle Scholar
  41. 41.
    Maher M, Churchill NW, De Oliveira Manoel AL, et al. Altered resting-state connectivity within executive networks after aneurysmal subarachnoid hemorrhage. PLoS ONE. 2015;10:1–18.CrossRefGoogle Scholar
  42. 42.
    De Bresser J, Vincken KL, Kaspers AJ, et al. Quantification of cerebral volumes on MRI 6 months after aneurysmal subarachnoid hemorrhage. Stroke. 2012;43:2782–4.CrossRefPubMedGoogle Scholar
  43. 43.
    Koivisto T, Vanninen R, Hurskainen H, et al. Outcomes of early endovascular versus surgical treatment of ruptured cerebral aneurysms. A prospective randomized study. Stroke. 2000;31:2369–77.CrossRefPubMedGoogle Scholar
  44. 44.
    Bendel P, Koivisto T, Niskanen E, et al. Brain atrophy and neuropsychological outcome after treatment of ruptured anterior cerebral artery aneurysms: a voxel-based morphometric study. Neuroradiology. 2009;51:711–22.CrossRefPubMedGoogle Scholar
  45. 45.
    Mayer TE, Dichgans M, Straube A, et al. Continuous intra-arterial nimodipine for the treatment of cerebral vasospasm. Cardiovasc Intervent Radiol. 2008;31:1200–4.CrossRefPubMedGoogle Scholar
  46. 46.
    Ferguson S, Macdonald RL. Predictors of cerebral infarction in patients with aneurysmal subarachnoid hemorrhage. Neurosurgery. 2007;60:658–67.CrossRefPubMedGoogle Scholar
  47. 47.
    Juvela S, Siironen J. Early cerebral infarction as a risk factor for poor outcome after aneurysmal subarachnoid haemorrhage. Eur J Neurol. 2012;19:332–9.CrossRefPubMedGoogle Scholar
  48. 48.
    Rosenberg NF, Liebling SM, Kosteva AR, et al. Infarct volume predicts delayed recovery in patients with subarachnoid hemorrhage and severe neurological deficits. Neurocrit Care. 2013;19:293–8.CrossRefPubMedGoogle Scholar
  49. 49.
    Tam AKH, Kapadia A, Ilodigwe D, et al. Impact of global cerebral atrophy on clinical outcome after subarachnoid hemorrhage. J Neurosurg. 2013;119:198–206.CrossRefPubMedGoogle Scholar
  50. 50.
    Hadjivassiliou M, Tooth CL, Romanowski CA, et al. Aneurysmal SAH: cognitive outcome and structural damage after clipping or coiling. Neurology. 2001;56:1672–7.CrossRefPubMedGoogle Scholar
  51. 51.
    Dumont AS, Crowley RW, Monteith SJ, et al. Endovascular treatment or neurosurgical clipping of ruptured intracranial aneurysms: effect on angiographic vasospasm, delayed ischemic neurological deficit, cerebral infarction, and clinical outcome. Stroke. 2010;41:2519–24.CrossRefPubMedGoogle Scholar
  52. 52.
    Hohlrieder M, Spiegel M, Hinterhoelzl J, et al. Cerebral vasospasm and ischaemic infarction in clipped and coiled intracranial aneurysm patients. Eur J Neurol. 2002;9:389–99.CrossRefPubMedGoogle Scholar
  53. 53.
    Molyneux AJ, Birks J, Clarke A, et al. The durability of endovascular coiling versus neurosurgical clipping of ruptured cerebral aneurysms: 18 year follow-up of the UK cohort of the International Subarachnoid Aneurysm Trial (ISAT). Lancet (London, England). 2015;385:691–7.CrossRefGoogle Scholar
  54. 54.
    Spetzler RF, McDougall CG, Zabramski JM, et al. The barrow ruptured aneurysm trial: 6-year results. J Neurosurg. 2015;123:609–17.CrossRefPubMedGoogle Scholar
  55. 55.
    Vatter H, Guresir E, Berkefeld J, et al. Perfusion-diffusion mismatch in MRI to indicate endovascular treatment of cerebral vasospasm after subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry. 2011;82:876–83.CrossRefPubMedGoogle Scholar
  56. 56.
    Kamran M, Downer J, Corkill R, et al. Non-invasive assessment of vasospasm following aneurysmal SAH using C-arm FDCT parenchymal blood volume measurement in the neuro-interventional suite: technical feasibility. Interv Neuroradiol. 2015;21:479–89.CrossRefPubMedPubMedCentralGoogle Scholar
  57. 57.
    Rodriguez-Régent C, Hafsa M, Turc G, et al. Early quantitative CT perfusion parameters variation for prediction of delayed cerebral ischemia following aneurysmal subarachnoid hemorrhage. Eur Radiol. 2016;26:2956–63.CrossRefPubMedGoogle Scholar
  58. 58.
    Leclerc X, Fichten A, Gauvrit JY, et al. Symptomatic vasospasm after subarachnoid haemorrhage: assessment of brain damage by diffusion and perfusion-weighted MRI and single-photon emission computed tomography. Neuroradiology. 2002;44:610–6.CrossRefPubMedGoogle Scholar
  59. 59.
    Rordorf G, Koroshetz WJ, Copen WA, et al. Diffusion- and perfusion-weighted imaging in vasospasm after subarachnoid hemorrhage. Stroke. 1999;30:599–605.CrossRefPubMedGoogle Scholar
  60. 60.
    Hertel F, Walter C, Bettag M, et al. Perfusion-weighted magnetic resonance imaging in patients with vasospasm: a useful new tool in the management of patients with subarachnoid hemorrhage. Neurosurgery. 2005;56:28–35.CrossRefPubMedGoogle Scholar
  61. 61.
    Beck J, Raabe A, Lanfermann H, et al. Effects of balloon angioplasty on perfusion- and diffusion-weighted magnetic resonance imaging results and outcome in patients with cerebral vasospasm. J Neurosurg. 2006;105:220–7.CrossRefPubMedGoogle Scholar
  62. 62.
    Ohtonari T, Kakinuma K, Kito T, et al. Diffusion-perfusion mismatch in symptomatic vasospasm after subarachnoid hemorrhage. Neurol Med Chir (Tokyo). 2008;48:331–6.CrossRefGoogle Scholar
  63. 63.
    Grandin CB, Cosnard G, Hammer F, et al. Vasospasm after subarachnoid hemorrhage: diagnosis with MR angiography. AJNR Am J Neuroradiol. 2000;21:1611–7.PubMedGoogle Scholar
  64. 64.
    Hacein-Bey L, Provenzale JM. Current imaging assessment and treatment of intracranial aneurysms. Am J Roentgenol. 2011;196:32–44.CrossRefGoogle Scholar
  65. 65.
    Lavoie P, Gariepy JL, Milot G, et al. Residual flow after cerebral aneurysm coil occlusion: diagnostic accuracy of MR angiography. Stroke. 2012;43:740–6.CrossRefPubMedGoogle Scholar
  66. 66.
    Shang S, Ye J, Luo X, et al. Follow-up assessment of coiled intracranial aneurysms using zTE MRA as compared with TOF MRA: a preliminary image quality study. Eur Radiol. 2017;27:4271–80.CrossRefPubMedGoogle Scholar
  67. 67.
    Qiao Y, Steinman DA, Qin Q, et al. Intracranial arterial wall imaging using three-dimensional high isotropic resolution black blood MRI at 3.0 Tesla. J Magn Reson Imaging. 2011;34:22–30.CrossRefPubMedGoogle Scholar
  68. 68.
    Guan J, Karsy M, McNally S, et al. High-resolution magnetic resonance imaging of intracranial aneurysms treated by flow diversion. Interdiscip Neurosurg Adv Tech Case Manag. 2017;10:69–74.CrossRefGoogle Scholar
  69. 69.
    Jungreis CA, Jannetta PJ, Yonas H. Timing treatment of a giant intracranial aneurysm by the use of magnetic resonance imaging for the determination of intraluminal clot stability. Skull Base Surg. 1993;3:32–6.CrossRefPubMedPubMedCentralGoogle Scholar
  70. 70.
    De Gast AN, Sprengers ME, Van Rooij WJ, et al. Midterm clinical and magnetic resonance imaging follow-up of large and giant carotid artery aneurysms after therapeutic carotid artery occlusion. Neurosurgery. 2007;60:1025–9.CrossRefPubMedGoogle Scholar
  71. 71.
    Toni F, Marliani AF, Cirillo L, et al. 3T MRI in the evaluation of brain aneurysms treated with flow-diverting stents: preliminary experience. Neuroradiol J. 2009;22:588–99.CrossRefPubMedGoogle Scholar
  72. 72.
    Matouk CC, Mandell DM, Günel M, et al. Vessel wall magnetic resonance imaging identifies the site of rupture in patients with multiple intracranial aneurysms: proof of principle. Neurosurgery. 2013;72:492–6.CrossRefPubMedGoogle Scholar
  73. 73.
    Nagahata S, Nagahata M, Obara M, et al. Wall enhancement of the intracranial aneurysms revealed by magnetic resonance vessel wall imaging using three-dimensional turbo spin-echo sequence with motion-sensitized driven-equilibrium: a sign of ruptured aneurysm? Clin Neuroradiol. 2016;26:277–83.CrossRefPubMedGoogle Scholar
  74. 74.
    Wang G, Wen L, Lei S, et al. Wall enhancement ratio and partial wall enhancement on MRI associated with the rupture of intracranial aneurysms. J Neurointerv Surg. Epub 2017.:neurintsurg-2017-013308.Google Scholar
  75. 75.
    Fernández-Seara MA, Edlow BL, Hoang A, et al. Minimizing acquisition time of arterial spin labeling at 3T. Magn Reson Med. 2008;59:1467–71.CrossRefPubMedGoogle Scholar
  76. 76.
    Chen Y, Wang DJJ, Detre JA. Test-retest reliability of arterial spin labeling with common labeling strategies. J Magn Reson Imaging. 2011;33:940–9.CrossRefPubMedPubMedCentralGoogle Scholar
  77. 77.
    Liu AA, Voss HU, Dyke JP, et al. Arterial spin labeling and altered cerebral blood flow patterns in the minimally conscious state. Neurology. 2011;77:1518–23.CrossRefPubMedPubMedCentralGoogle Scholar
  78. 78.
    Günther M. Perfusion imaging. J Magn Reson Imaging. 2014;40:269–79.CrossRefPubMedGoogle Scholar
  79. 79.
    Nelson S, Edlow BL, Wu O, et al. Default mode network perfusion in aneurysmal subarachnoid hemorrhage. Neurocrit Care. 2016;25:237–42.CrossRefPubMedPubMedCentralGoogle Scholar
  80. 80.
    Detre JA, Leigh JS, Williams DS, et al. Perfusion imaging. Magn Reson Med. 1992;23:37–45.CrossRefPubMedGoogle Scholar
  81. 81.
    Essig M, Shiroishi MS, Nguyen TB, et al. Perfusion MRI: the five most frequently asked technical questions. Am J Roentgenol. 2013;200:24–34.CrossRefGoogle Scholar
  82. 82.
    Aoyama K, Fushimi Y, Okada T, et al. Detection of symptomatic vasospasm after subarachnoid haemorrhage: initial findings from single time-point and serial measurements with arterial spin labelling. Eur Radiol. 2012;22:2382–91.CrossRefPubMedGoogle Scholar
  83. 83.
    Kelly ME, Rowland MJ, Okell TW, et al. Pseudo-continuous arterial spin labelling MRI for non-invasive, whole-brain, serial quantification of cerebral blood flow following aneurysmal subarachnoid haemorrhage. Transl Stroke Res. 2013;4:710–8.CrossRefPubMedGoogle Scholar
  84. 84.
    Labriffe M, Ter Minassian A, Pasco-Papon A, et al. Feasibility and validity of monitoring subarachnoid hemorrhage by a noninvasive MRI imaging perfusion technique: pulsed arterial spin labeling (PASL). J Neuroradiol. 2015;42:358–67.CrossRefPubMedGoogle Scholar
  85. 85.
    Deibler AR, Pollock JM, Kraft RA, et al. Arterial spin-labeling in routine clinical practice, part 2: hypoperfusion patterns. Am J Neuroradiol. 2008;29:1235–41.CrossRefPubMedPubMedCentralGoogle Scholar
  86. 86.
    Deibler AR, Pollock JM, Kraft RA, et al. Arterial spin-labeling in routine clinical practice, part 3: hyperperfusion patterns. Am J Neuroradiol. 2008;29:1428–35.CrossRefPubMedPubMedCentralGoogle Scholar
  87. 87.
    Lu H, Hua J, van Zijl PCM. Noninvasive functional imaging of cerebral blood volume with vascular-space-occupancy (VASO) MRI. NMR Biomed. 2013;26:932–48.CrossRefPubMedPubMedCentralGoogle Scholar
  88. 88.
    Forster BB, MacKay AL, Whittall KP, et al. Functional magnetic resonance imaging: the basics of blood-oxygen-level dependent (BOLD) imaging. Can Assoc Radiol J. 1998;49:320–9.PubMedGoogle Scholar
  89. 89.
    Ogawa S, Lee TM, Kay AR, et al. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci USA. 1990;87:9868–72.CrossRefPubMedPubMedCentralGoogle Scholar
  90. 90.
    da Costa L, Fierstra J, Fisher JA, et al. BOLD MRI and early impairment of cerebrovascular reserve after aneurysmal subarachnoid hemorrhage. J Magn Reson Imaging. 2014;40:972–9.CrossRefPubMedGoogle Scholar
  91. 91.
    Ellmore TM, Rohlffs F, Khursheed F. fMRI of working memory impairment after recovery from subarachnoid hemorrhage. Front Neurol. 2013;4:179.CrossRefPubMedPubMedCentralGoogle Scholar
  92. 92.
    Lu H, van Zijl PCM. Experimental measurement of extravascular parenchymal BOLD effects and tissue oxygen extraction fractions using multi-echo VASO fMRI at 1.5 and 3.0 T. Magn Reson Med. 2005;53:808–16.CrossRefPubMedGoogle Scholar
  93. 93.
    Cheng Y, van Zijl PCM, Hua J. Measurement of parenchymal extravascular R2* and tissue oxygen extraction fraction using multi-echo vascular space occupancy MRI at 7 T. NMR Biomed. 2015;28:264–71.CrossRefPubMedGoogle Scholar
  94. 94.
    Hua J, Stevens RD, Huang AJ, et al. Physiological origin for the BOLD poststimulus undershoot in human brain: vascular compliance versus oxygen metabolism. J Cereb Blood Flow Metab. 2011;31:1599–611.CrossRefPubMedPubMedCentralGoogle Scholar
  95. 95.
    Soares JM, Marques P, Alves V, et al. A hitchhiker’s guide to diffusion tensor imaging. Front Neurosci. 2013;7:31.CrossRefPubMedPubMedCentralGoogle Scholar
  96. 96.
    Yeo SS, Choi BY, Chang CH, et al. Evidence of corticospinal tract injury at midbrain in patients with subarachnoid hemorrhage. Stroke. 2012;43:2239–41.CrossRefPubMedGoogle Scholar
  97. 97.
    Jang SH, Choi BY, Kim SH, et al. Injury of the mammillothalamic tract in patients with subarachnoid haemorrhage: a retrospective diffusion tensor imaging study. BMJ Open. 2014;4:e005613.CrossRefPubMedPubMedCentralGoogle Scholar
  98. 98.
    Fragata I, Alves M, Papoila AL, et al. Early prediction of delayed ischemia and functional outcome in acute subarachnoid hemorrhage: role of diffusion tensor imaging. Stroke. 2017;48:2091–7.CrossRefPubMedGoogle Scholar
  99. 99.
    Brooke NS, Ouwerkerk R, Adams CB, et al. Phosphorus-31 magnetic resonance spectra reveal prolonged intracellular acidosis in the brain following subarachnoid hemorrhage. Proc Natl Acad Sci USA. 1994;91:1903–7.CrossRefPubMedPubMedCentralGoogle Scholar
  100. 100.
    Macmillan CSA, Wild JM, Wardlaw JM, et al. Traumatic brain injury and subarachnoid hemorrhage: in vivo occult pathology demonstrated by magnetic resonance spectroscopy may not be “ischaemic”. A primary study and review of the literature. Acta Neurochir (Wien). 2002;144:853–62.CrossRefGoogle Scholar
  101. 101.
    Wagner M, Jurcoane A, Hildebrand C, et al. Metabolic changes in patients with aneurysmal subarachnoid hemorrhage apart from perfusion deficits: neuronal mitochondrial injury? Am J Neuroradiol. 2013;34:1535–41.CrossRefPubMedGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature and Neurocritical Care Society 2018

Authors and Affiliations

  • Sarah E. Nelson
    • 1
    • 2
  • Haris I. Sair
    • 4
  • Robert D. Stevens
    • 1
    • 2
    • 3
    • 4
  1. 1.Department of NeurologyJohns Hopkins University School of MedicineBaltimoreUSA
  2. 2.Department of Anesthesiology and Critical Care MedicineJohns Hopkins University School of MedicineBaltimoreUSA
  3. 3.Department of NeurosurgeryJohns Hopkins University School of MedicineBaltimoreUSA
  4. 4.Departments of RadiologyJohns Hopkins University School of MedicineBaltimoreUSA

Personalised recommendations