Reliability of CT Angiography in Cerebral Vasospasm: A Systematic Review of the Literature and an Inter- and Intraobserver Study

Article Figures & Data

Download figure
Open in new tab
Download powerpoint
Fig 1.
Interreader reliability coefficients (95% confidence intervals) including all subgroups (senior, junior, interventional, and diagnostic neuroradiologists) for the 3 main questions: detection of moderate-severe vasospasm in any arterial segment (≥ 50% narrowing), recommendations of medical treatment augmentation, and DSA ± angioplasty based on imaging findings. The horizontal dashed line indicates the threshold for substantial agreement (κ > 0.6). INR indicates interventional neuroradiology; DNR, diagnostic neuroradiology.
Download figure
Open in new tab
Download powerpoint
Fig 2.
Intrareader reliability coefficients for the 3 main questions (95% confidence intervals) among 4 observers: detection of moderate-severe vasospasm in any arterial segment (≥ 50% narrowing), recommendations for medical treatment augmentation, and DSA ± angioplasty based on imaging findings. The horizontal dashed line indicates the threshold for substantial agreement (κ > 0.6).

Table 1:

Baseline demographics, clinical, and radiologic characteristics of the patients^{^a}

Characteristics
Age (median) (range, IQR) (yr)	60 (38–88, 56–69)
Female sex	27 (54)
Hunt and Hess scale
1	5 (10)
2	12 (24)
3	8 (16)
4	17 (34)
5	8 (16)
Hydrocephalus	47 (94)
EVD placement	38 (76)
Craniectomy	4 (8)
Modified Fisher scale
1	1 (2)
2	8 (16)
3	2 (4)
4	39 (78)
Treatment of culprit vascular lesion
Endovascular	39 (78)
Surgical	9 (18)
None^{^b}	2 (4)
>1 Lesion treated^{^c}	3 (6)
Ruptured aneurysm/lesion location^{^d}
Anterior communicating artery	19 (38)
Posterior communicating artery	8 (16)
Middle cerebral artery bifurcation	6 (12)
PICA	6 (12)
Carotid ophthalmic	3 (6)
Pericallosal	2 (4)
Basilar tip	1 (2)
Carotid terminus	1 (2)
Posterior cerebral artery (P1)	1 (2)
Vertebral artery dissection	1 (2)
None^{^b}	2 (4)

Note:—IQR indicates interquartile range; PICA, posterior inferior cerebellar artery; EVD, external ventricular drain.
↵a Data in parentheses are percentages except where indicated.
↵b No culprit lesion identified on repeat DSA examinations.
↵c Three aneurysms, 2 treated by endovascular coiling, and 1, by surgical clipping.
↵d In case of >1 treated aneurysm, this reflects the most likely source of hemorrhage.

Table 2:

Vasospasm-related characteristics^{^a}

Characteristics
Reason to perform follow-up CTA (vasospasm assessment)
Altered level of consciousness	14 (28)
New neurologic deficit	12 (24)
Raised intracranial pressure	4 (8)
New ischemic lesion on CT	2 (4)
None other than rule out vasospasm	18 (36)
Vasospasm severity according to original CTA reports
None	10 (20)
Mild	17 (34)
Moderate	18 (36)
Severe	5 (10)
Invasive vasospasm management performed following CTA
DSA	4 (8)
Intra-arterial milrinone	1 (2)
Angioplasty	0 (0)