Skip Navigation
Skip to contents

KMJ : Kosin Medical Journal

OPEN ACCESS
SEARCH
Search

Articles

Page Path
HOME > Kosin Med J > Epub ahead of print > Article
Original article
Combined Hounsfield unit measurements of the intracranial cisterns and superior sagittal sinus predict delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage
Junghyun Parkorcid
> Epub ahead of print
DOI: https://doi.org/10.7180/kmj.26.167
Published online: August 31, 2026

Department of Neurosurgery, Kosin University Gospel Hospital, Kosin University College of Medicine, Busan, Korea

Corresponding Author: Junghyun Park, MD Department of Neurosurgery, Kosin University Gospel Hospital, Kosin University College of Medicine, 262 Gamcheon-ro, Seo-gu, Busan 49267, Korea Tel: +82-51-990-6465 Fax: +82-51-990-3042 E-mail: baessy12@kosin.ac.kr
• Received: April 8, 2026   • Revised: May 15, 2026   • Accepted: May 19, 2026

© 2026 Kosin University College of Medicine.

This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 527 Views
  • 6 Download
  • Background
    Delayed cerebral ischemia (DCI) is a major cause of poor outcomes after aneurysmal subarachnoid hemorrhage (aSAH). We evaluated whether Hounsfield unit (HU) measurements of the intracranial cisterns and superior sagittal sinus (SSS) on admission noncontrast computed tomography (CT) predict DCI.
  • Methods
    We retrospectively analyzed 143 consecutive patients with aSAH treated between January 2021 and December 2023. HU values were measured using region-of-interest analysis in the carotid, interpeduncular, and Sylvian cisterns and the SSS on admission CT. The primary outcome was DCI. Multivariable logistic regression and receiver operating characteristic analyses assessed predictive performance.
  • Results
    DCI developed in 36 patients (25.2%). Patients with DCI had higher cisternal HU (53.6±6.8 vs. 47.9±6.4, p<0.001) and SSS HU (66.3±8.7 vs. 55.3±9.1, p<0.001). Both cisternal HU (odds ratio [OR] per 1 HU increase, 1.07; 95% confidence interval [CI], 1.01–1.13; p=0.02) and SSS HU (OR per 1 HU increase, 1.14; 95% CI, 1.07–1.22; p<0.001) independently predicted DCI, whereas conventional clinical grades did not. The combined HU model showed good discrimination (area under the curve [AUC], 0.82), outperforming SSS HU alone (AUC, 0.78). DCI incidence increased from 9.5% to 40.6% across HU-based risk groups.
  • Conclusions
    Admission noncontrast CT-derived cisternal and SSS HU measurements independently predicted DCI after aSAH. This simple, objective approach may facilitate early risk stratification and targeted monitoring.
Aneurysmal subarachnoid hemorrhage (aSAH) remains a devastating cerebrovascular event, and delayed cerebral ischemia (DCI) is a principal contributor to secondary neurological deterioration and long-term disability. The modern definition of DCI—new focal neurological impairment or a decrease of at least two points on the Glasgow Coma Scale lasting for at least 1 hour, not apparent immediately after aneurysm occlusion, and not attributable to other causes—has enabled more consistent outcome reporting and comparability across studies [1]. Contemporary understanding also emphasizes that DCI is a multifactorial process rather than a simple downstream consequence of angiographic vasospasm; ischemic lesions may not match the distribution of spastic arteries, and therapeutic strategies that reduce vasospasm do not necessarily translate into improved functional outcomes [2]. Accordingly, there is continued clinical need for reliable early biomarkers that can identify patients at high risk of DCI and guide intensified monitoring and individualized preventive strategies.
Despite advances in critical care and aneurysm-securing techniques, DCI continues to occur in a substantial proportion of patients after aSAH. Reviews consistently report that radiographic vasospasm is common and that a meaningful subset of patients develop delayed neurological deficits with persistent neurocognitive consequences [3]. Even among good-grade patients—who otherwise have high potential for recovery—DCI remains a frequent and clinically relevant complication. In a large pooled analysis of good-grade aSAH patients, DCI occurred in approximately one-fifth of cases, and risk stratification based on baseline factors (e.g., age, thick SAH, obesity in men, and preadmission hypertension in women) was proposed to inform intensive care unit triage and surveillance intensity [4]. However, baseline clinical factors alone often lack the granularity needed for individualized prediction, and imaging markers that are objective, reproducible, and available at admission are particularly attractive.
The amount of subarachnoid blood on admission computed tomography (CT) is among the best-established predictors of vasospasm/DCI risk, but practical limitations exist for volumetric quantification in routine workflows. Hounsfield unit (HU) measurements offer a simple quantitative surrogate that reflects blood clot density (and by inference, clot concentration), potentially capturing information not fully represented by categorical grading scales. Prior work has highlighted that clot volume estimated on CT is not “true” hematoma volume because it is intermixed with cerebrospinal fluid, whereas HU may more directly represent hematoma concentration and severity of bleeding [5]. This conceptual advantage has motivated several HU-based approaches aimed at predicting delayed complications after aSAH.
Recent studies have demonstrated that cisternal HU values measured on admission noncontrast CT are associated with delayed vasospasm-related complications. Ishihara and colleagues reported that the HU value of the interpeduncular cistern on initial CT was an accurate and reliable predictor of symptomatic vasospasm (SVS); the incidence increased markedly when HU exceeded 50, and risk rose in a dose-dependent fashion with increasing HU [5]. Similarly, Park and Kang [6] and Ko et al. [7] showed that higher HU values within the basal and Sylvian cisterns were associated with SVS (and, to a lesser extent, hydrocephalus), and proposed cistern-specific HU cutoffs derived from receiver operating characteristic (ROC) analyses. More recently, Moran [8] extended the clinical relevance of initial HU measurements by demonstrating associations with DCI and overall prognosis in a surgical clipping cohort, supporting the concept that HU-derived metrics can function as practical imaging biomarkers in aSAH. Collectively, these studies support HU-based quantification as an objective approach for risk assessment, yet most prior investigations have focused primarily on cisternal clot density alone, and substantial heterogeneity persists regarding which compartments to measure and how best to standardize the measurement strategy.
A key challenge for HU-based prediction is that cisternal HU values may be influenced not only by local clot burden and concentration, but also by patient-level factors that affect blood attenuation on CT (e.g., hemoglobin/hematocrit, hydration status, and technical factors), which can complicate cross-patient comparisons. We hypothesized that incorporating a venous reference compartment measurable on the same admission noncontrast CT—specifically the superior sagittal sinus (SSS)—could provide complementary information relevant to systemic blood attenuation while remaining simple and reproducible. Therefore, we evaluated whether admission noncontrast CT-based HU measurements of both the intracranial cisterns and the SSS predict DCI after aSAH, and whether a combined HU model improves discrimination compared with cisternal HU or conventional clinical grading variables alone.
Ethical statements: This study was approved by the Institutional Review Board (IRB) of Kosin University Gospel Hospital (IRB No. 2026-01-018-003). The requirement for informed consent was waived due to the retrospective nature of the study.
1. Study design and patient population
This retrospective observational study included consecutive patients with aSAH who were admitted to a single tertiary referral center between January 2021 and December 2023. A total of 143 patients with confirmed aSAH who underwent noncontrast CT at admission were included. Patients with traumatic SAH or inadequate imaging quality were excluded. A flow diagram of patient selection is shown in Fig. 1.
2. Clinical and radiological data collection
Baseline demographic, clinical, and radiological data were extracted from electronic medical records, including age, sex, Glasgow Coma Scale (GCS) score on arrival, systolic blood pressure, vascular risk factors, aneurysm characteristics, treatment modality, and clinical grading scales (Hunt-Hess and modified Fisher grades). Baseline characteristics of the study population are summarized in Table 1.
3. Computed tomography acquisition and Hounsfield unit measurement
HU measurements on noncontrast CT have been shown to reflect subarachnoid clot density and severity of hemorrhage and have been associated with delayed cerebrovascular complications after aSAH [5-8]. All patients underwent noncontrast CT at admission using standard institutional protocols. All measurements were performed on admission noncontrast CT scans using standard brain window settings. HU measurements were performed using region-of-interest (ROI) analysis by investigators blinded to clinical outcomes.
4. Intracranial cistern Hounsfield unit measurement
HU values were measured in the intracranial cisterns, including the carotid, interpeduncular, and Sylvian cisterns. ROIs were manually placed on the most clearly visualized axial CT slice demonstrating the largest cisternal area, while carefully avoiding adjacent brain parenchyma, bone, and vascular structures (Fig. 2). ROI sizes were kept as consistent as possible across patients. Measurements were obtained separately on the left and right sides and averaged for analysis. To enhance measurement reproducibility, ROIs were placed using predefined anatomical landmarks and standardized ROI sizes whenever possible. All measurements were performed by investigators blinded to clinical outcomes, and standardized ROI placement protocols were used to enhance measurement consistency.
5. Superior sagittal sinus and lenticular nucleus Hounsfield unit measurement
HU values of the SSS were measured using a single midline ROI on the axial CT slice demonstrating the largest cross-sectional area of the sinus. Additionally, HU values of the lenticular nuclei were measured bilaterally and averaged, serving as an internal reference representing relatively stable attenuation of deep gray matter, and were not expected to differ according to DCI status (Fig. 3). ROI sizes were standardized as much as possible to minimize measurement variability.
6. Outcome definitions
The primary outcome was the occurrence of DCI, defined according to the established multidisciplinary consensus definition as a new focal neurological impairment or a decrease of at least two points on the GCS lasting for at least 1 hour, not apparent immediately after aneurysm occlusion, and not attributable to other causes [1]. SVS was recorded as a secondary outcome based on clinical deterioration attributable to cerebral vasospasm, as previously described [6,7].
7. Statistical analysis
Continuous variables are presented as mean±standard deviation or median (interquartile range), as appropriate, and categorical variables as number (%). Comparisons between groups were performed using the independent t-test or Mann-Whitney U test for continuous variables and Fisher exact test for categorical variables.
Variables with clinical relevance and/or p<0.10 in univariable analyses were entered into a multivariable logistic regression model to identify independent predictors of DCI. ROC curve analysis was performed to evaluate discriminative performance and determine optimal cutoff values using the Youden index for continuous HU-based variables. A two-sided p-value<0.05 was considered statistically significant. Statistical analyses were performed using IBM SPSS Statistics for Windows ver. 27.0 (IBM Corp.).
1. Patient characteristics
Among the 143 patients included in the analysis, 36 patients (25.2%) developed DCI. Baseline clinical and imaging characteristics of the overall cohort are shown in Table 1.
2. Comparison between delayed cerebral ischemia and non-delayed cerebral ischemia groups
Patients who developed DCI demonstrated significantly higher cisternal HU values than those who did not develop DCI (53.6±6.8 vs. 47.9±6.4, p<0.001). SSS HU values were also markedly higher in the DCI group (66.3±8.7 vs. 55.3±9.1, p<0.001). In contrast, lenticular nucleus HU values did not differ significantly between groups (p=0.74). Other clinical grading scales, including Hunt-Hess and modified Fisher grades, were not significantly different between the two groups (Table 2), highlighting the limited discriminative value of conventional categorical grading scales in this cohort.
3. Multivariable analysis for predictors of delayed cerebral ischemia
In multivariable logistic regression analysis, both cisternal HU (odds ratio [OR] per 1 HU increase, 1.07; 95% confidence interval [CI], 1.01–1.13; p=0.02) and SSS HU (OR per 1 HU increase, 1.14; 95% CI, 1.07–1.22; p<0.001) were independently associated with the development of DCI. Notably, conventional clinical grading scales, including Hunt-Hess and modified Fisher grades, did not demonstrate independent predictive value for DCI in the multivariable model, whereas both cisternal HU and SSS HU remained significant predictors (Table 3).
4. Predictive performance of Hounsfield unit-based parameters
ROC analysis demonstrated moderate discriminative ability for cisternal HU (area under the curve [AUC], 0.66) and moderate-to-good discriminative performance for SSS HU (AUC, 0.78). A combined model incorporating cisternal HU and SSS HU showed improved predictive performance (AUC, 0.82), outperforming either parameter alone as well as conventional clinical grading scales (Table 4), thereby supporting the complementary prognostic roles of local clot burden and systemic blood attenuation.
5. Hounsfield unit-based risk stratification
Patients were stratified into four risk groups based on cisternal and SSS HU cutoff values derived from ROC analysis. The incidence of DCI increased stepwise from 9.5% in the low-risk group to 40.6% in the high-risk group. After adjustment for age, Hunt-Hess grade, and modified Fisher grade, the high-risk group remained independently associated with DCI (adjusted OR 5.9; p=0.004; Supplementary Table S1). These associations remained consistent in a sensitivity analysis restricted to patients who underwent endovascular treatment only (Supplementary Table S2). Clinical and imaging variables associated with SVS are summarized in Supplementary Table S3.
In this study, we identified a simple and readily obtainable CT-based biomarker framework for early prediction of DCI after aSAH. Specifically, quantitative HU measurements of both the intracranial cisterns and the SSS on admission noncontrast CT were independently associated with the development of DCI.
Both cisternal HU and SSS HU remained significant predictors of DCI after adjustment for established clinical and radiographic grading scales. Importantly, a combined model integrating cisternal HU and SSS HU demonstrated superior discriminative performance compared with conventional clinical predictors, suggesting that simultaneous assessment of local clot burden and systemic blood attenuation may provide complementary prognostic information. These findings highlight the potential value of objective CT-derived HU measurements as practical imaging biomarkers for early DCI risk stratification in patients with aSAH.
1. Hounsfield unit-based quantification as an imaging biomarker beyond traditional grading scales
The amount of subarachnoid blood has long been recognized as a key determinant of delayed cerebrovascular complications after aSAH. Conventional grading systems such as the Fisher and modified Fisher scales were developed to stratify risk based on clot distribution and thickness; however, they remain categorical and subject to interobserver variability [9-11]. Moreover, these scales primarily reflect clot volume and location rather than concentration, a factor more closely related to the biological activity of hemoglobin degradation products implicated in vasospasm and DCI [12-14].
HU measurements provide a quantitative surrogate for clot density and concentration, potentially capturing pathophysiological information not represented by visual grading alone. Previous studies have shown that higher cisternal HU values correlate with SVS and poor outcomes, supporting the concept that clot density influences delayed ischemic injury [15-18]. Our results extend these observations by demonstrating that cisternal HU is independently associated with DCI, even when adjusted for modified Fisher grade, thereby reinforcing the limitations of categorical grading systems and supporting the value of continuous, quantitative imaging biomarkers for individualized risk prediction.
2. Incremental value of superior sagittal sinus Hounsfield unit
A novel and clinically relevant finding of the present study is the strong and independent association between SSS HU and DCI. While cisternal HU reflects local clot density, it may be influenced by patient-specific systemic factors such as hematocrit, hydration status, and CT acquisition parameters. Measurement of HU within the SSS provides a readily available venous reference that reflects systemic blood attenuation on the same noncontrast CT scan, thereby accounting for interindividual variability in blood composition and imaging conditions, particularly at the hyperacute admission stage when hemoconcentration and physiological stress responses are most pronounced.
The superior predictive performance of SSS HU observed in our study suggests that global blood attenuation characteristics may modulate the biological effects of subarachnoid clot burden. Higher venous HU may indicate higher hematocrit or hemoglobin concentration, which could amplify the effective concentration of oxyhemoglobin and other spasmogenic blood breakdown products within the subarachnoid space, even in the absence of differences in clot volume or distribution [19-21]. This hypothesis is consistent with experimental and clinical studies demonstrating that hemoglobin concentration, rather than clot volume alone, plays a critical role in triggering vasospasm, microvascular dysfunction, and secondary ischemia [22-24]. Elevated SSS HU may reflect systemic physiologic characteristics associated with DCI risk following aSAH. Therefore, SSS HU should be interpreted as a potential imaging biomarker associated with DCI risk rather than a direct mechanistic determinant. Higher venous attenuation on noncontrast CT is strongly influenced by hemoglobin and hematocrit levels; therefore, elevated SSS HU may reflect systemic blood concentration that increases the effective exposure of the brain to hemoglobin degradation products within the subarachnoid space.
Importantly, a combined model incorporating cisternal HU and SSS HU demonstrated the highest discriminative ability for DCI prediction, suggesting that local clot burden and systemic blood attenuation provide complementary rather than redundant prognostic information. This dual-compartment approach is consistent with contemporary understanding of DCI as a multifactorial process, involving both local clot-related toxicity (reflected by cisternal HU) and global systemic modifiers of cerebral perfusion and oxygen delivery (reflected by SSS HU) [25,26]. Importantly, these associations remained robust in a sensitivity analysis restricted to patients who underwent endovascular treatment only (Supplementary Table S2).
3. Lenticular nucleus Hounsfield unit as an internal negative control
In contrast to cisternal and SSS HU, HU values of the lenticular nucleus were not associated with DCI and demonstrated no meaningful discriminative performance. This finding supports the specificity of HU-based measurements within blood-containing compartments and argues against nonspecific global CT attenuation effects or scanner-related artifacts as the primary driver of our results.
The lenticular nucleus represents relatively homogeneous gray matter with stable attenuation characteristics and minimal susceptibility to acute changes in blood concentration. Its lack of association with DCI reinforces the interpretation that blood-derived HU measurements, rather than generalized brain density changes, are the relevant predictors of delayed ischemic injury. Similar internal control strategies have been used in prior neuroimaging studies to validate compartment-specific biomarkers [27,28].
4. Pathophysiological implications
DCI is increasingly recognized as a heterogeneous entity resulting from the interplay of large-vessel vasospasm, microcirculatory dysfunction, cortical spreading depolarizations, inflammation, and impaired autoregulation [29-31]. While vasospasm has historically dominated clinical attention, numerous trials have demonstrated that angiographic vasospasm reduction does not necessarily translate into improved outcomes [30-32]. Consequently, early identification of patients at high risk for DCI before clinical deterioration occurs remains a critical unmet need.
Our findings suggest that HU-based metrics may serve as early, noninvasive surrogates for the biological burden of subarachnoid blood and its downstream effects. The ability to derive these measurements from routine noncontrast CT, without additional imaging or contrast administration, enhances their feasibility for widespread clinical adoption.
5. Clinical implications
From a practical standpoint, HU-based risk stratification may inform early clinical decision-making, including intensified neurological monitoring, tailored hemodynamic management, and selection of patients for advanced neuromonitoring or investigational therapies. In particular, patients classified as high risk based on combined cisternal and SSS HU may benefit from closer surveillance during the critical window for DCI development, when preventive and therapeutic interventions are most likely to be effective. Importantly, these HU measurements can be readily obtained from routine admission noncontrast CT scans without additional imaging or contrast administration, supporting their feasibility for routine clinical use.
6. Limitations
This study has several limitations. First, its retrospective single-center design may limit generalizability. Second, HU measurements can be influenced by CT scanner calibration and acquisition parameters, although our use of standardized institutional protocols and internal venous reference mitigates this concern. Third, direct correlation analyses between SSS HU values and hematologic parameters such as hemoglobin or hematocrit were not incorporated into the present analysis. Future prospective studies integrating synchronized laboratory and imaging biomarkers are warranted to further clarify the physiologic basis and validate the proposed approach.
7. Conclusions
Quantitative HU measurements of the intracranial cisterns and SSS on admission noncontrast CT provide independent and complementary information for predicting DCI after aSAH. This simple and reproducible approach may enhance early risk stratification and support more individualized management strategies in patients with aSAH.
Supplementary materials are available at https://doi.org/10.7180/kmj.26.167

Supplementary Table S1.

HU-based risk stratification for delayed cerebral ischemia
kmj-26-167-Supplementary-Table-S1.pdf

Supplementary Table S2.

Multivariable logistic regression analysis for predictors of DCI in the endovascular-treatment cohort
kmj-26-167-Supplementary-Table-S2.pdf

Supplementary Table S3.

Comparison of clinical and imaging variables between patients with and without SVS
kmj-26-167-Supplementary-Table-S3.pdf

Conflicts of interest

No potential conflict of interest relevant to this article was reported.

Funding

None.

Author contributions

All the work was done by Junghyun Park.

Fig. 1.
Flow diagram of patient selection. The diagram illustrates the patient selection process and final study cohort classified according to the occurrence of DCI. aSAH, aneurysmal subarachnoid hemorrhage; CT, computed tomography; DCI, delayed cerebral ischemia.
kmj-26-167f1.jpg
Fig. 2.
Measurement of Hounsfield unit values in the intracranial cisterns on noncontrast computed tomography (CT). (A) Representative axial noncontrast CT image at the level of the basal cisterns. Arrows indicate the carotid cisterns, and the arrowhead indicates the interpeduncular cistern. (B) Representative axial noncontrast CT image at the level of the Sylvian fissure, with arrows indicating the bilateral Sylvian cisterns.
kmj-26-167f2.jpg
Fig. 3.
Measurement of Hounsfield unit values in the superior sagittal sinus and lenticular nucleus on noncontrast computed tomography (CT). (A) Representative axial noncontrast CT image demonstrating region-of-interest (ROI) placement within the superior sagittal sinus. (B) Representative axial noncontrast CT image at the level of the lenticular nuclei, with bilateral ROIs placed within each nucleus.
kmj-26-167f3.jpg
Table 1.
Baseline clinical and radiological characteristics of the study population (n=143)
Variable Value
Study period Jan 2021–Dec 2023
Age (yr) 58.3±10.6
Female sex 99 (69.2)
GCS on arrival 11 (10–13)
SBP on arrival (mmHg) 149.3±22.8
Hunt-Hess grade
 I 8 (5.6)
 II 21 (14.7)
 III 36 (25.2)
 IV 48 (33.6)
 V 30 (21.0)
Modified Fisher grade
 1 12 (8.4)
 2 25 (17.5)
 3 43 (30.1)
 4 63 (44.1)
Hypertension 80 (55.9)
Diabetes mellitus 30 (21.0)
Hyperlipidemia 52 (36.4)
Smoking 58 (40.6)
Aneurysm multiplicity 27 (18.9)
Aneurysm size (mm) 5.1 (2.3–10.1)
Aneurysm location
 AComA 51 (35.7)
 MCA bifurcation 49 (34.3)
 PComA 27 (18.9)
 Distal anterior cerebral artery 7 (4.9)
 Paraclinoid ICA 4 (2.8)
 Anterior temporal artery 3 (2.1)
 ICA bifurcation 2 (1.4)
Endovascular treatment 112 (78.3)
Surgical clipping 31 (21.7)
Cisternal HU 49.3±6.9
Superior sagittal sinus HU 58.1±10.2
Lenticular nucleus HU 35.4±6.1
SVS 20 (14.0)
DCI 36 (25.2)

Values are presented as mean±standard deviation, number (%) or median (interquartile range).

GCS, Glasgow Coma Scale; SBP, systolic blood pressure; AComA, anterior communicating artery; MCA, middle cerebral artery; PComA, posterior communicating artery; ICA, internal carotid artery; HU, Hounsfield unit; SVS, symptomatic vasospasm; DCI, delayed cerebral ischemia.

Table 2.
Comparison of clinical and imaging variables between patients with and without DCI
Variable DCI (n=36) Non-DCI (n=107) p-value
Age (yr) 60.2±9.7 57.7±10.9 0.31
Female sex 27 (75.0) 72 (67.3)   0.36
GCS on arrival 10 (10–13) 11 (10–13) 0.98
Hunt-Hess grade ≥III 25 (69.4) 89 (83.2)   0.09
Modified Fisher grade 3–4 27 (75.0) 79 (73.8)   1.00
Cisternal HU 53.6±6.8 47.9±6.4 <0.001
Superior sagittal sinus HU 66.3±8.7 55.3±9.1 <0.001
Lenticular nucleus HU 34.6±3.8 35.7±6.7 0.74

Values are presented as mean±standard deviation, number (%), or median (interquartile range).

DCI, delayed cerebral ischemia; GCS, Glasgow Coma Scale; HU, Hounsfield unit.

Table 3.
Multivariable logistic regression analysis for predictors of delayed cerebral ischemia
Variable Odds ratio 95% CI p-value
Age (per year increase) 1.02 0.98–1.05 0.29
Hunt-Hess grade ≥III 0.55 0.23–1.30 0.17
Modified Fisher grade 3–4 1.12 0.42–2.95 0.82
Treatment modality (endovascular vs. surgery) 1.04 0.41–2.49 0.80
Cisternal HU (per 1 HU increase) 1.07 1.01–1.13 0.02
Superior sagittal sinus HU (per 1 HU increase) 1.14 1.07–1.22 <0.001

Odds ratios for continuous variables represent risk per 1-unit increase.

CI, confidence interval; HU, Hounsfield unit.

Table 4.
Receiver operating characteristic curve analysis for predicting delayed cerebral ischemia
Predictor AUC (95% CI) Cutoff Sensitivity (%) Specificity (%)
Cisternal HU 0.66 (0.56–0.75) 51.0 69 61
SSS HU 0.78 (0.69–0.86) 62.0 78 72
Lenticular nucleus HU 0.52 (0.42–0.62) 35.5 47 55
Combined model (cisternal HU+SSS HU) 0.82 (0.74–0.89) - 81 74

Optimal cutoff values were determined using the Youden index.

AUC, area under the curve; CI, confidence interval; HU, Hounsfield unit; SSS, superior sagittal sinus.

  • 1. Vergouwen MD, Vermeulen M, van Gijn J, Rinkel GJ, Wijdicks EF, Muizelaar JP, et al. Definition of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage as an outcome event in clinical trials and observational studies: proposal of a multidisciplinary research group. Stroke 2010;41:2391–5.ArticlePubMed
  • 2. Vergouwen MD, Ilodigwe D, Macdonald RL. Cerebral infarction after subarachnoid hemorrhage contributes to poor outcome by vasospasm-dependent and -independent effects. Stroke 2011;42:924–9.ArticlePubMed
  • 3. Rowland MJ, Hadjipavlou G, Kelly M, Westbrook J, Pattinson KT. Delayed cerebral ischaemia after subarachnoid haemorrhage: looking beyond vasospasm. Br J Anaesth 2012;109:315–29.ArticlePubMed
  • 4. Frontera JA, Fernandez A, Schmidt JM, Claassen J, Wartenberg KE, Badjatia N, et al. Defining vasospasm after subarachnoid hemorrhage: what is the most clinically relevant definition? Stroke 2009;40:1963–8.ArticlePubMed
  • 5. Ishihara H, Oka F, Kawano R, Shinoyama M, Nishimoto T, Kudomi S, et al. Hounsfield unit value of interpeduncular cistern hematomas can predict symptomatic vasospasm. Stroke 2020;51:143–8.ArticlePubMed
  • 6. Park JS, Kang HG. Hounsfield unit as a predictor of symptomatic vasospasm and hydrocephalus in good-grade subarachnoid hemorrhage treated with endovascular coiling. Quant Imaging Med Surg 2023;13:6627–35.ArticlePubMedPMC
  • 7. Ko SB, Choi HA, Helbok R, Schmidt JM, Badjatia N, Claassen J, et al. Quantitative analysis of hemorrhage clearance and delayed cerebral ischemia after subarachnoid hemorrhage. J Neurointerv Surg 2016;8:923–6.ArticlePubMed
  • 8. Moran M. Cannabis use is associated with delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. Neurol Today 2022;22:8–10.Article
  • 9. Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 1980;6:1–9.ArticlePubMed
  • 10. Claassen J, Bernardini GL, Kreiter K, Bates J, Du YE, Copeland D, et al. Effect of cisternal and ventricular blood on risk of delayed cerebral ischemia after subarachnoid hemorrhage: the Fisher scale revisited. Stroke 2001;32:2012–20.ArticlePubMed
  • 11. Frontera JA, Claassen J, Schmidt JM, Wartenberg KE, Temes R, Connolly ES, et al. Prediction of symptomatic vasospasm after subarachnoid hemorrhage: the modified fisher scale. Neurosurgery 2006;59:21–7.ArticlePubMed
  • 12. Macdonald RL. Delayed neurological deterioration after subarachnoid haemorrhage. Nat Rev Neurol 2014;10:44–58.ArticlePubMedPDF
  • 13. Sehba FA, Hou J, Pluta RM, Zhang JH. The importance of early brain injury after subarachnoid hemorrhage. Prog Neurobiol 2012;97:14–37.ArticlePubMedPMC
  • 14. Pluta RM, Hansen-Schwartz J, Dreier J, Vajkoczy P, Macdonald RL, Nishizawa S, et al. Cerebral vasospasm following subarachnoid hemorrhage: time for a new world of thought. Neurol Res 2009;31:151–8.ArticlePubMedPMC
  • 15. Ko SB, Choi HA, Carpenter AM, Helbok R, Schmidt JM, Badjatia N, et al. Quantitative analysis of hemorrhage volume for predicting delayed cerebral ischemia after subarachnoid hemorrhage. Stroke 2011;42:669–74.ArticlePubMed
  • 16. de Rooij NK, Rinkel GJ, Dankbaar JW, Frijns CJ. Delayed cerebral ischemia after subarachnoid hemorrhage: a systematic review of clinical, laboratory, and radiological predictors. Stroke 2013;44:43–54.ArticlePubMed
  • 17. Vergouwen MD, Vermeulen M, Roos YB. Delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage: is angiographic vasospasm an epiphenomenon? Stroke 2009;40:e39.ArticlePubMed
  • 18. Ritzenthaler T, Gobert F, Bouchier B, Dailler F. Amount of blood during the subacute phase and clot clearance rate as prognostic factors for delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. J Clin Neurosci 2021;87:74–9.ArticlePubMed
  • 19. Caylor MM, Macdonald RL. Pharmacological prevention of delayed cerebral ischemia in aneurysmal subarachnoid hemorrhage. Neurocrit Care 2024;40:159–69.ArticlePubMedPDF
  • 20. Kramer AH, Gurka MJ, Nathan B, Dumont AS, Kassell NF, Bleck TP, et al. Complications associated with anemia and blood transfusion in patients with aneurysmal subarachnoid hemorrhage. Crit Care Med 2008;36:2070–5.ArticlePubMed
  • 21. Diringer MN, Bleck TP, Claude Hemphill J, Menon D, Shutter L, Vespa P, et al. Critical care management of patients following aneurysmal subarachnoid hemorrhage: recommendations from the Neurocritical Care Society's Multidisciplinary Consensus Conference. Neurocrit Care 2011;15:211–40.ArticlePubMed
  • 22. Pluta RM. Dysfunction of nitric oxide synthases as a cause and therapeutic target in delayed cerebral vasospasm after SAH. Acta Neurochir Suppl 2008;104:139–47.ArticlePubMedPMC
  • 23. Dreier JP, Major S, Manning A, Woitzik J, Drenckhahn C, Steinbrink J, et al. Cortical spreading ischaemia is a novel process involved in ischaemic damage in patients with aneurysmal subarachnoid haemorrhage. Brain 2009;132:1866–81.ArticlePubMedPMC
  • 24. Sehba FA, Bederson JB. Mechanisms of acute brain injury after subarachnoid hemorrhage. Neurol Res 2006;28:381–98.ArticlePubMed
  • 25. Macdonald RL, Schweizer TA. Spontaneous subarachnoid haemorrhage. Lancet 2017;389:655–66.ArticlePubMed
  • 26. van der Steen WE, Zijlstra IA, Verbaan D, Boers AM, Gathier CS, van den Berg R, et al. Association of quantified location-specific blood volumes with delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. AJNR Am J Neuroradiol 2018;39:1059–64.ArticlePubMedPMC
  • 27. Wintermark M, Ko NU, Smith WS, Liu S, Higashida RT, Dillon WP, et al. Vasospasm after subarachnoid hemorrhage: utility of perfusion CT and CT angiography on diagnosis and management. AJNR Am J Neuroradiol 2006;27:26–34.PubMedPMC
  • 28. Dankbaar JW, Rijsdijk M, van der Schaaf IC, Velthuis BK, Wermer MJ, Rinkel GJ, et al. Relationship between vasospasm, cerebral perfusion, and delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. Neuroradiology 2009;51:813–9.ArticlePubMedPMC
  • 29. Woitzik J, Dreier JP, Hecht N, Fiss I, Sandow N, Major S, et al. Delayed cerebral ischemia and spreading depolarization in absence of angiographic vasospasm after subarachnoid hemorrhage. J Cereb Blood Flow Metab 2012;32:203–12.ArticlePubMedPDF
  • 30. Macdonald RL, Pluta RM, Zhang JH. Cerebral vasospasm after subarachnoid hemorrhage: the emerging revolution. Nat Clin Pract Neurol 2007;3:256–63.ArticlePubMedPDF
  • 31. Cahill J, Zhang JH. Subarachnoid hemorrhage: is it time for a new direction? Stroke 2009;40:S86–7.ArticlePubMed
  • 32. Connolly ES, Rabinstein AA, Carhuapoma JR, Derdeyn CP, Dion J, Higashida RT, 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.ArticlePubMed

Figure & Data

References

    Citations

    Citations to this article as recorded by  

      • PubReader PubReader
      • ePub LinkePub Link
      • Cite
        CITE
        export Copy
        Close
      • Download Citation
        Download Citation
        Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

        Format:
        • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
        • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
        Include:
        • Citation for the content below
        Combined Hounsfield unit measurements of the intracranial cisterns and superior sagittal sinus predict delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage
        Close
      • XML DownloadXML Download
      Figure
      • 0
      • 1
      • 2
      Combined Hounsfield unit measurements of the intracranial cisterns and superior sagittal sinus predict delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage
      Image Image Image
      Fig. 1. Flow diagram of patient selection. The diagram illustrates the patient selection process and final study cohort classified according to the occurrence of DCI. aSAH, aneurysmal subarachnoid hemorrhage; CT, computed tomography; DCI, delayed cerebral ischemia.
      Fig. 2. Measurement of Hounsfield unit values in the intracranial cisterns on noncontrast computed tomography (CT). (A) Representative axial noncontrast CT image at the level of the basal cisterns. Arrows indicate the carotid cisterns, and the arrowhead indicates the interpeduncular cistern. (B) Representative axial noncontrast CT image at the level of the Sylvian fissure, with arrows indicating the bilateral Sylvian cisterns.
      Fig. 3. Measurement of Hounsfield unit values in the superior sagittal sinus and lenticular nucleus on noncontrast computed tomography (CT). (A) Representative axial noncontrast CT image demonstrating region-of-interest (ROI) placement within the superior sagittal sinus. (B) Representative axial noncontrast CT image at the level of the lenticular nuclei, with bilateral ROIs placed within each nucleus.
      Combined Hounsfield unit measurements of the intracranial cisterns and superior sagittal sinus predict delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage
      Variable Value
      Study period Jan 2021–Dec 2023
      Age (yr) 58.3±10.6
      Female sex 99 (69.2)
      GCS on arrival 11 (10–13)
      SBP on arrival (mmHg) 149.3±22.8
      Hunt-Hess grade
       I 8 (5.6)
       II 21 (14.7)
       III 36 (25.2)
       IV 48 (33.6)
       V 30 (21.0)
      Modified Fisher grade
       1 12 (8.4)
       2 25 (17.5)
       3 43 (30.1)
       4 63 (44.1)
      Hypertension 80 (55.9)
      Diabetes mellitus 30 (21.0)
      Hyperlipidemia 52 (36.4)
      Smoking 58 (40.6)
      Aneurysm multiplicity 27 (18.9)
      Aneurysm size (mm) 5.1 (2.3–10.1)
      Aneurysm location
       AComA 51 (35.7)
       MCA bifurcation 49 (34.3)
       PComA 27 (18.9)
       Distal anterior cerebral artery 7 (4.9)
       Paraclinoid ICA 4 (2.8)
       Anterior temporal artery 3 (2.1)
       ICA bifurcation 2 (1.4)
      Endovascular treatment 112 (78.3)
      Surgical clipping 31 (21.7)
      Cisternal HU 49.3±6.9
      Superior sagittal sinus HU 58.1±10.2
      Lenticular nucleus HU 35.4±6.1
      SVS 20 (14.0)
      DCI 36 (25.2)
      Variable DCI (n=36) Non-DCI (n=107) p-value
      Age (yr) 60.2±9.7 57.7±10.9 0.31
      Female sex 27 (75.0) 72 (67.3)   0.36
      GCS on arrival 10 (10–13) 11 (10–13) 0.98
      Hunt-Hess grade ≥III 25 (69.4) 89 (83.2)   0.09
      Modified Fisher grade 3–4 27 (75.0) 79 (73.8)   1.00
      Cisternal HU 53.6±6.8 47.9±6.4 <0.001
      Superior sagittal sinus HU 66.3±8.7 55.3±9.1 <0.001
      Lenticular nucleus HU 34.6±3.8 35.7±6.7 0.74
      Variable Odds ratio 95% CI p-value
      Age (per year increase) 1.02 0.98–1.05 0.29
      Hunt-Hess grade ≥III 0.55 0.23–1.30 0.17
      Modified Fisher grade 3–4 1.12 0.42–2.95 0.82
      Treatment modality (endovascular vs. surgery) 1.04 0.41–2.49 0.80
      Cisternal HU (per 1 HU increase) 1.07 1.01–1.13 0.02
      Superior sagittal sinus HU (per 1 HU increase) 1.14 1.07–1.22 <0.001
      Predictor AUC (95% CI) Cutoff Sensitivity (%) Specificity (%)
      Cisternal HU 0.66 (0.56–0.75) 51.0 69 61
      SSS HU 0.78 (0.69–0.86) 62.0 78 72
      Lenticular nucleus HU 0.52 (0.42–0.62) 35.5 47 55
      Combined model (cisternal HU+SSS HU) 0.82 (0.74–0.89) - 81 74
      Table 1. Baseline clinical and radiological characteristics of the study population (n=143)

      Values are presented as mean±standard deviation, number (%) or median (interquartile range).

      GCS, Glasgow Coma Scale; SBP, systolic blood pressure; AComA, anterior communicating artery; MCA, middle cerebral artery; PComA, posterior communicating artery; ICA, internal carotid artery; HU, Hounsfield unit; SVS, symptomatic vasospasm; DCI, delayed cerebral ischemia.

      Table 2. Comparison of clinical and imaging variables between patients with and without DCI

      Values are presented as mean±standard deviation, number (%), or median (interquartile range).

      DCI, delayed cerebral ischemia; GCS, Glasgow Coma Scale; HU, Hounsfield unit.

      Table 3. Multivariable logistic regression analysis for predictors of delayed cerebral ischemia

      Odds ratios for continuous variables represent risk per 1-unit increase.

      CI, confidence interval; HU, Hounsfield unit.

      Table 4. Receiver operating characteristic curve analysis for predicting delayed cerebral ischemia

      Optimal cutoff values were determined using the Youden index.

      AUC, area under the curve; CI, confidence interval; HU, Hounsfield unit; SSS, superior sagittal sinus.


      KMJ : Kosin Medical Journal
      TOP