Advertisement

A comparison of contrast-free MRA at 3.0T in cases of intracranial aneurysms with or without subarachnoid hemorrhage

Open AccessPublished:October 25, 2017DOI:https://doi.org/10.1016/j.clinimag.2017.10.012

      Highlights

      • VR 3D-TOF-MRA is a non-invasive approach with high accuracy in the diagnosis of intracranial aneurysms.
      • 3D-TOF-MRA had a high diagnostic accuracy, sensitivity, specificity, PPV and NPV for detection of intracranial aneurysms.
      • 33D-TOF-MRA could be a preferred contrast-free and noninvasive modality for the evaluation of intracranial aneurysms.

      Abstract

      Purpose

      To determine the diagnostic accuracy of contrast-free MRA at 3.0 T for detection of intracranial aneurysms in patients with subarachnoid hemorrhage.

      Methods

      411 patients (183 with SAH and 228 with non-SAH) underwent MRA. Accuracy, sensitivity, specificity, positive predictive values (PPV), and negative predictive values (NPV) were measured and compared with DSA.

      Results

      Except for a slight difference in sensitivity in patient-based and aneurysm-based evaluations (P = 0.037), there were no other significant differences in accuracy, specificity, PPV, and NPV.

      Conclusion

      VR 3D-TOF-MRA is a non-invasive approach with high accuracy in the diagnosis of intracranial aneurysms.

      Keywords

      1. Introduction

      Approximately 80% of subarachnoid hemorrhages (SAH) are caused by ruptured intracranial aneurysms, carry a morbidity rate of 10% to 20% and a mortality rate of 40% to 50% [
      • Broderick J.P.
      • Brott T.G.
      • Duldner J.E.
      • Tomsick T.
      • Leach A.
      Initial and recurrent bleeding are the major causes of death following subarachnoid hemorrhage.
      ], and accurate imaging information for preoperative evaluation is necessary to perform safe and proper surgical clipping or endovascular treatment in patients with ruptured or unruptured cerebral aneurysms [
      • Hirai T.
      • Korogi Y.
      • Ono K.
      • Murata Y.
      • Suginohara K.
      • Omori T.
      • et al.
      Preoperative evaluation of intracranial aneurysms: usefulness of intraarterial 3D CT angiography and conventional angiography with a combined unit—initial experience.
      ]. Until now, digital subtraction angiography (DSA) is still considered to be the gold standard for the detection and endovascular treatment planning of intracranial aneurysms [
      • Anxionnat R.
      • Bracard S.
      • Ducrocq X.
      • Trousset Y.
      • Launay L.
      • Kerrien E.
      • et al.
      Intracranial aneurysms: clinical value of 3D digital subtraction angiography in the therapeutic decision and endovascular treatment.
      ]. However, it is invasive, radiation-associated, time-consuming, relatively expensive, and hospitalization-based. Moreover, it not only needs to be performed by well experienced experts but also carries a 1–2% complication risk with a 0.5% rate of permanent neurological deficit [
      • Cloft H.J.
      • Joseph G.J.
      • Dion J.E.
      Risk of cerebral angiography in patients with subarachnoid hemorrhage, cerebral aneurysm, and arteriovenous malformation: a meta-analysis.
      ,
      • Willinsky R.A.
      • Taylor S.M.
      • TerBrugge K.
      • Farb R.I.
      • Tomlinson G.
      • Montanera W.
      Neurologic complications of cerebral angiography: prospective analysis of 2899 procedures and review of the literature.
      ]. Although, computed tomographic angiography (CTA) is the most frequently used diagnostic tool for the detection of intracranial aneurysms [
      • Hirai T.
      • Korogi Y.
      • Ono K.
      • Murata Y.
      • Suginohara K.
      • Omori T.
      • et al.
      Preoperative evaluation of intracranial aneurysms: usefulness of intraarterial 3D CT angiography and conventional angiography with a combined unit—initial experience.
      ,
      • Papke K.
      • Kuhl C.K.
      • Fruth M.
      • Haupt C.
      • Schlunz-Hendann M.
      • Sauner D.
      • et al.
      Intracranial aneurysms: role of multidetector CT angiography in diagnosis and endovascular therapy planning.
      ], it is often limited by over-projecting bone at the skull base, radiation exposure and injecting of iodine contrast media [
      • Wardlaw J.M.
      • White P.M.
      The detection and management of unruptured intracranial aneurysms.
      ,
      • Hiratsuka Y.
      • Miki H.
      • Kiriyama I.
      • Kikuchi K.
      • Takahashi S.
      • Matsubara I.
      • et al.
      Diagnosis of unruptured intracranial aneurysms: 3 T MR angiography versus 64-channel multi-detector row CT angiography.
      ].
      In recent years, Magnetic resonance angiography (MRA) and Three-dimensional Time-Of-Flight MRA (3D-TOF-MRA) have been widely used as noninvasive and contrast-free methods for screening of cerebral aneurysms with an overall accurate diagnosis of approximately 90%, as compared with conventional three-dimensional (3D) DSA [
      • Wardlaw J.M.
      • White P.M.
      The detection and management of unruptured intracranial aneurysms.
      ,
      • Hiratsuka Y.
      • Miki H.
      • Kiriyama I.
      • Kikuchi K.
      • Takahashi S.
      • Matsubara I.
      • et al.
      Diagnosis of unruptured intracranial aneurysms: 3 T MR angiography versus 64-channel multi-detector row CT angiography.
      ,
      • Atlas S.W.
      • Sheppard L.
      • Goldberg H.I.
      • Hurst R.W.
      • Listerud J.
      • Flamm E.
      Intracranial aneurysms: detection and characterization with MR angiography with use of an advanced postprocessing technique in a blinded-reader study.
      ,
      • Schwab K.E.
      • Gailloud P.
      • Wyse G.
      • Tamargo R.J.
      Limitations of magnetic resonance imaging and magnetic resonance angiography in the diagnosis of intracranial aneurysms.
      ,
      • Okahara M.
      • Kiyosue H.
      • Yamashita M.
      • Nagatomi H.
      • Hata H.
      • Saginoya T.
      • et al.
      Diagnostic accuracy of magnetic resonance angiography for cerebral aneurysms in correlation with 3D-digital subtraction angiographic images: a study of 133 aneurysms.
      ,
      • White P.M.
      • Teasdale E.M.
      • Wardlaw J.M.
      • Easton V.
      Intracranial aneurysms: CT angiography and MR angiography for detection prospective blinded comparison in a large patient cohort.
      ]. However, the value of 3D–TOF-MRA in the detection of ruptured intracranial aneurysms is controversial because MRA and TOF are prone to artifacts that significantly limit reliable identification of aneurysms, especially for very small aneurysm (<3 mm) [
      • Korogi Y.
      • Takahashi M.
      • Mabuchi N.
      • Miki H.
      • Fujiwara S.
      • Horikawa Y.
      • et al.
      Intracranial aneurysms: diagnostic accuracy of three-dimensional, Fourier transform, time-of-flight MR angiography.
      ]. Therefore, the purpose of this study was to determine the diagnostic accuracy of contrast-free MRA at 3.0 T for detection of intracranial aneurysms in a large patient population with clinically SAH [Glasgow coma scale (GCS) of 15′] and compared the results with Non-SAH with 3D DSA as reference standards.

      2. Materials and methods

      2.1 Patients

      The institutional review board (IRB) approved the study protocol, and patients or qualifying family members provided informed consent before participation. From January 2010 to December 2015, 183 consecutive patients with SAH (GCS = 15′) and 228 consecutive patients with non-SAH, confirmed by a non-contrast head computed tomographic scan, underwent MRA for the detection of intracranial aneurysms at our institution. DSA was performed after MRA to confirm the diagnosis of intracranial aneurysm. VR-DSA was obtained from the rotational DSA (RDSA) data, and regarded as the gold standards.

      2.2 Image acquisition

      2.2.1 MRA

      All MRA examinations were performed on a 3.0 T system (Achieva X-Series, Philips Medical Systems) with a Sense-Head-8 receiver head coil, and details of the MRA examinations and the specialized post-processing 3-D techniques described previously [
      • Li M.H.
      • Li Y.D.
      • Tan H.Q.
      • Gu B.X.
      • Chen Y.C.
      • Wang W.
      • et al.
      Contrast-free MRA at 3.0 T for the detection of intracranial aneurysms.
      ,
      • Li M.H.
      • Li Y.D.
      • Gu B.X.
      • Cheng Y.S.
      • Wang W.
      • Tan H.Q.
      • et al.
      Accurate diagnosis of small cerebral aneurysms ≤5 mm in diameter with 3.0-T MR angiography.
      ]. Briefly, the 3D-TOF-MRA was obtained using 3D-T1-weighted fast field (T1-FFE) sequences with TR/TE, 35/7; flip angle, 20°; field of view (FOV) 250 × 190 × 108; four slabs (180 slices), slice thickness, 0.8 mm; matrix, 732 × 1024; and an acquisition time of 8 min and 56 s. The acquired image data sets were then transferred to a workstation (EWS, Philips Medical) for 3D-volume inspection (Philips Medical).
      To reduce artery overlay and to effectively identify intracranial aneurysms, the single-artery highlighting method was applied, referenced as catheter cerebral angiography [
      • Li M.H.
      • Li Y.D.
      • Tan H.Q.
      • Gu B.X.
      • Chen Y.C.
      • Wang W.
      • et al.
      Contrast-free MRA at 3.0 T for the detection of intracranial aneurysms.
      ,
      • Li M.H.
      • Li Y.D.
      • Gu B.X.
      • Cheng Y.S.
      • Wang W.
      • Tan H.Q.
      • et al.
      Accurate diagnosis of small cerebral aneurysms ≤5 mm in diameter with 3.0-T MR angiography.
      ]. For the imaging of left or right internal carotid artery (ICA) system, we removed the right or left internal carotid artery system respectively. And, we defined the middle of the two posterior communicating arteries (PComAs) as the starting points of posterior circulation. For the imaging of posterior circulation system, we removed the anterior circulation system from the middle of the two PComAs. We analyzed three vessels systems in each patient from six basic views (anterior, posterior, bilateral, superior, and inferior) and from arbitrary angles to clearly depict the aneurysm origins and courses.

      2.2.2 DSA

      DSA was performed by an interventional neuroradiologist within 14 days after the MRA, and details of the DSA were described previously [
      • Li M.H.
      • Li Y.D.
      • Tan H.Q.
      • Gu B.X.
      • Chen Y.C.
      • Wang W.
      • et al.
      Contrast-free MRA at 3.0 T for the detection of intracranial aneurysms.
      ,
      • Li M.H.
      • Li Y.D.
      • Gu B.X.
      • Cheng Y.S.
      • Wang W.
      • Tan H.Q.
      • et al.
      Accurate diagnosis of small cerebral aneurysms ≤5 mm in diameter with 3.0-T MR angiography.
      ].
      All patients with possible intracranial aneurysms underwent 2D-DSA and VR-DSA of the affected and contralateral arteries, obtained in 2–4 projections. 2D-DSA was performed for the remaining arteries. A complete DSA was consisted of at least a 3-vessel 2D-DSA and a 2-vessel VR-DSA for each patient. Two observers (M. H. L. and W. W., with 17 and 10 years of experience in neurointerventional radiology respectively), highly experienced in neurointerventional radiology and blinded to all clinical and previous imaging results, identified and analyzed all intracranial aneurysms together.

      2.3 Image review

      Three observers were blinded to all clinical and VR-DSA results. They analyzed independently all 3D-TOF-MRA datasets on an offline-workstation from multiple on-screen viewing angles by using the single-artery highlighting approach. The source images and MIPs were presented on-screen, thus allowing for adjusting the appropriate threshold of the window width and level to diagnose or differentiate small aneurysms with infundibula. For interobserver discrepancies in detection of intracranial aneurysms, consensus was achieved or majority decision obtained.
      Confidence in diagnosing aneurysms was assessed using a previously described five-point scale [
      • Atlas S.W.
      • Sheppard L.
      • Goldberg H.I.
      • Hurst R.W.
      • Listerud J.
      • Flamm E.
      Intracranial aneurysms: detection and characterization with MR angiography with use of an advanced postprocessing technique in a blinded-reader study.
      ]: 5, aneurysm definitely absent; 4, aneurysm probably absent; 3, uncertain; 2, aneurysm probably present; and 1, aneurysm definitely present. Studies with one or more aneurysms identified as probably or definitely present were considered positive; all others were negative. Aneurysm size was recorded as the maximum 2D angiographic dimension: (1) <3 mm, (2) ≥3 mm.

      2.4 Statistical analyses

      The categorical demographic and basic characteristic variables, expressed as numbers and percentages, were compared using the chi-square test. Continuous variables were expressed as mean ± standard deviation (SD) and compared by an unpaired t-test, if normally distributed. Descriptive statistics were performed on 3 levels: patient-by-patient (no or any intracranial aneurysm per patient), aneurysm-by-aneurysm, and size-by-size. The diagnostic performance parameters of VR 3D-TOF-MRA at 3.0 T for the diagnosis of intracranial aneurysms compared with those of DSA (namely, accuracy, sensitivity, specificity, PPV, and NPV) were expressed as percentage (95% CI).

      3. Results

      3.1 Patient population

      A total of 411 consecutive patients were enrolled in the study. A flow chart of patient enrollment is shown in Fig. 1. The demographic and clinical characteristics of the 411 patients with suspected intracranial aneurysms are summarized in Table 1. The patients included 177 men and 234 women with a mean age of 54.41 ± 12.95 years (range: 19–84 years).
      Fig. 1
      Fig. 1Flow chart of patients with SAH through the study. IA = intracranial aneurysm.
      Table 1Baseline characteristics of 411 patients between the two groups.
      CharacteristicAll patients

      (n = 411)
      SAH

      (n = 183)
      Non-SAH

      (n = 228)
      P value
      Age—year54.41 ± 12.9552.49 ± 12.9055.48 ± 12.880.012
      Female sex—no. (%)234 (48.2)98 (48.2)136 (59.6)0.301
      Aneurysm location
       ICA255 (60.8%)81 (43.8%)174 (74.3)<0.001
       ACA87 (20.7%)62 (33.5%)25 (10.6)<0.001
       MCA53 (12.5%)24 (13.0%)29 (12.3)0.859
       VAB34 (8.0%)18 (9.7%)16 (6.8)0.282
      Number of aneurysms
       Single aneurysm281 (67.1%)123 (66.5%)158 (67.5%)0.823
       Multiple aneurysms138 (32.9%)62 (33.5%)76 (32.5%)
      Aneurysm size (mm)
       <3 mm142 (33.9%)48 (25.9%)94 (40.1%)0.002
       3 to <5 mm137 (32.7%)61 (32.9%)76 (32.5%)0.915
       5 to <10 mm103 (24.6%)63 (34.1%)40 (17.1%)<0.001
       ≥10 mm37 (15.8%)13 (7.0%)24 (10.3%)0.247
      Average of aneurysm size (mean ± SD) — mm5.30 ± 4.67 (1.0–42)5.11 ± 2.79 (1.5–18)5.45 ± 5.73 (1.0–42)0.466

      3.2 DSA results

      According to the reference standard, 185 aneurysms were detected in 151 patients with SAH: In 6 patients, three aneurysms at a time were detected; 22 patients had two aneurysms detected at the same time; and a single aneurysm only was detected in 123 patients, and 234 aneurysms were detected in 193 patients with non-SAH: In 1 patient, four aneurysms at a time were detected; 4 patients had three aneurysms detected at the same time; 30 patients had two aneurysms detected at the same time; and a single aneurysm only was detected in 158 patients.
      The location and size of the aneurysms in the two groups are shown in Table 1. There were more intracranial aneurysms in SAH group located at the ACA (including AComA), and more intracranial aneurysms in non-SAH group situated in the ICA. In addition, more intracranial aneurysms were in size <3 mm in non-SAH group, and more intracranial aneurysms were in size 5 to 10 mm in SAH group. There was no significant difference between the two groups in the mean diameter of the maximal aneurysm sac (P > 0.05, Table 2).
      Table 2Diagnostic performance of VR 3D-TOF-MRA at 3.0 T for the detection of intracranial aneurysms on DSA in patient-, aneurysm, and size-based evaluation [95% confidence interval (CI)].
      NTPTNFPFNκSensitivity, % (95% CI)Specificity, % (95% CI)PPV, % (95% CI)NPV, % (95% CI)Accuracy, % (95% CI)
      Patient-based evaluation
      SAHAll patients18314731140.89–1.097.4 (94.8–99.9)96.9 (96.5–103.2)99.3 (98.0–100.7)88.6 (77.5–99.7)97.3 (94.9–99.7)
      Non-SAHAll patients22819131600.87–0.9410083.8 (71.3–96.2)97.0 (94.5–99.4)10097.4 (95.3–99.5)
      P value0.0370.1130.2460.1161.000
      Aneurysm-based evaluation
      SAHAll aneurysms21818131240.89–1.097.8 (95.7–100)93.9 (85.3–102.5)98.9 (97.4–100.4)88.6 (77.5–99.7)97.2 (95.1–99.4)
      Non-SAHAll aneurysms27123431600.88–0.9410083.8 (71.3–96.2)97.5 (95.5–99.5)10097.8 (96.0–99.5)
      P value0.0370.2670.4750.1160.773
      Aneurysm-size-based evaluation
      SAH<3 mm814631220.87–1.095.8 (90.0–101.7)93.9 (85.3–102.5)95.8 (90.0–101.7)93.9 (85.3–102.5)95.1 (90.2–99.9)
      Non-SAH<3 mm1319432500.85–0.9310086.5 (74.9–98.0)94.9 (90.6–99.3)10096.2 (92.9–99.5)
      P value0.1130.4341.0000.4920.734
      SAH3 mm16913532021.098.5 (96.5–100.6)10010094.1 (85.8–1025)98.8 (97.2–100.5)
      Non-SAH3 mm17514034100.93–1.010097.1 (91.3–102.9)99.3 (97.9–100.3)10099.4 (98.3–100.6)
      P value0.2441.0001.0000.4930.617
      Note: TP = true-positive; TN = true-negative; FP = false-positive; FN = false-negative; PPV = positive predictive values; NPV = negative predictive values; SAH = subarachnoid hemorrhage.

      3.3 Diagnostic performance between the two groups

      The diagnostic accuracy, sensitivity, specificity, PPV and NPV of VR 3D-TOF-MRA on patient-, aneurysm-, and aneurysm-size-based evaluation at 3.0 T between the two groups are detailed in Table 2.

      3.3.1 Patient-based evaluation

      In patients with SAH, at least one intracranial aneurysm was detected by VR 3D-TOF-MRA in 150 patients, and 151 patients at least one aneurysm was identified by VR-DSA. Single-patient sample images performed by 3D-TOF-MRA, DSA and VR-DSA are shown in Fig. 2, Fig. 3. One of the two VR-DSA-detected aneurysms in a vessel and one out of the three VR-DSA-detected aneurysms in a vessel were false-negatives on VR 3D-TOF-MRA. The intracranial aneurysms in two patients were false-negatives on VR 3D-TOF-MRA because they could be confirmed by VR-DSA. The intracranial aneurysm in one patient was false-positives on VR 3D-TOF-MRA because they could not be confirmed by VR-DSA.
      Fig. 2
      Fig. 2Aneurysm in the left anterior communicating artery in a 61-year-old male patient with SAH and a GCS score of 15. (A) VR 3D-TOF-MRA reveals an aneurysm (arrow) located at the left anterior communicating artery. (B) DSA and (C) VR-DSA demonstrate the aneurysm (arrow) located at the left anterior communicating artery.
      Fig. 3
      Fig. 3Aneurysm in the left anterior communicating artery and C5 segment of left ICA in a 73-year-old female patient with SAH and a GCS score of 15. (A) VR 3D-TOF-MRA reveals an aneurysm (arrow) located at the left anterior communicating artery, and a large aneurysm situated in the left C5 segment (arrowhead). (B) DSA and (C) VR-DSA demonstrate the aneurysms located at the left anterior communicating artery (arrow) and the left C5 segment (arrowhead).
      In patients with Non-SAH, at least one aneurysm was identified by VR 3D–TOF-MRA, and 193 patients at least one aneurysm was identified by VR-DSA. One of two aneurysms in two patients was a false-positive on VR 3D-TOF-MRA, because only one could be confirmed by VR-DSA. The intracranial aneurysms in four patients were false-positives on VR 3D-TOF-MRA because they could be confirmed by VR-DSA. Except for a slight difference in sensitivity, there were no significantly differences in accuracy, specificity, PPV, and NPV.

      3.3.2 Aneurysm-based evaluation

      In patients with SAH, VR-DSA revealed 185 aneurysms in 151 of the 183 patients, and a total of 183 intracranial aneurysms were visualized by VR 3D-TOF-MRA. Four patients with four intracranial aneurysms were false-negatives, and one patient with two micro-aneurysms at left and right M1-2 segment were false-positives on VR 3D-TOF-MRA, because the vessel was excessively tortuous and overlaid by the parent arteries.
      In patients with non-SAH, VR-DSA revealed 234 aneurysms in 193 of the 228 patients, and a total of 240 intracranial aneurysms were visualized by VR 3D–TOF-MRA. Six intracranial aneurysms were false-positives, and two were false-negatives due to an acute aneurysm rupture. Except for a slight difference in sensitivity, there were no significantly differences in accuracy, specificity, PPV, and NPV.

      3.3.3 Aneurysm-size-based evaluation

      In patients with SAH, A total of 48 of 185 aneurysms identified by VR-DSA were <3 mm, 61 were 3–5 mm, 63 were 5–10 mm, and 13 were >10 mm in maximum diameter. A total of 48 of the 183 intracranial aneurysms identified by VR 3D–TOF-MRA were <3 mm, 60 were 3–5 mm, 62 were 5–10 mm, and 13 were >10 mm in maximum diameter. VR 3D–TOF-MRA identified two false-positive aneurysms <3 mm that could not be confirmed by VR-DSA. In two patients, one out of two RICAS aneurysms <3 mm and one out of three RICAS aneurysms <3 mm were false-negatives on VR 3D–TOF-MRA. There was one false-negative 3–5 mm aneurysm in one patient and one false-negative 5–10 mm aneurysm in one patient that could not be confirmed by VR-DSA.
      In patients with non-SAH, A total of 94 of 234 aneurysms identified by VR-DSA were <3 mm, 76 were 3–5 mm, 40 were 5–10 mm, and 24 were >10 mm in maximum diameter. A total of 99 of the 240 intracranial aneurysms identified by VR 3D-TOF-MRA were <3 mm, 77 were 3–5 mm, 40 were 5–10 mm, and 24 were >10 mm in maximum diameter. VR 3D–TOF-MRA identified five false-positive aneurysms <3 mm and one false-positive aneurysms 3–5 mm that could not be confirmed by VR-DSA. There were no significantly differences in accuracy, sensitivity, specificity, PPV, and NPV between SAH and Non-SAH in <3 mm group and >3 mm group.

      4. Discussion

      A detective rate for ruptured intracranial aneurysms equal to or better than DSA is essential if 3D-TOF-MRA is to serve as a noninvasive replacement for DSA and ensure a central role for 3D-TOF-MRA in making a high quality treatment decisions for patients with SAH prior to coiling or surgery. In this single-institute study, we found a high diagnostic accuracy, sensitivity, specificity, PPV and NPV of 3D-TOF-MRA at 3.0 T for detection of intracranial aneurysms in patient with SAH (GCS = 15′), even in depicting aneurysms <3 mm in diameter. Moreover, the diagnostic accuracy, sensitivity, specificity, PPV and NPV of 3D–TOF-MRA at 3.0 T were equal or not inferior to patients with non-SAH. These results indicated that the current technology for 3D-TOF-MRA image acquisition and post-processing was effective in accurately diagnosis of intracranial aneurysm in patients with SAH (GCS = 15′) and capable of accurate aneurysm characterization.
      The findings of this study were better than other big-scale studies [
      • Hiratsuka Y.
      • Miki H.
      • Kiriyama I.
      • Kikuchi K.
      • Takahashi S.
      • Matsubara I.
      • et al.
      Diagnosis of unruptured intracranial aneurysms: 3 T MR angiography versus 64-channel multi-detector row CT angiography.
      ,
      • Atlas S.W.
      • Sheppard L.
      • Goldberg H.I.
      • Hurst R.W.
      • Listerud J.
      • Flamm E.
      Intracranial aneurysms: detection and characterization with MR angiography with use of an advanced postprocessing technique in a blinded-reader study.
      ,
      • Schwab K.E.
      • Gailloud P.
      • Wyse G.
      • Tamargo R.J.
      Limitations of magnetic resonance imaging and magnetic resonance angiography in the diagnosis of intracranial aneurysms.
      ,
      • Okahara M.
      • Kiyosue H.
      • Yamashita M.
      • Nagatomi H.
      • Hata H.
      • Saginoya T.
      • et al.
      Diagnostic accuracy of magnetic resonance angiography for cerebral aneurysms in correlation with 3D-digital subtraction angiographic images: a study of 133 aneurysms.
      ,
      • White P.M.
      • Teasdale E.M.
      • Wardlaw J.M.
      • Easton V.
      Intracranial aneurysms: CT angiography and MR angiography for detection prospective blinded comparison in a large patient cohort.
      ,
      • Korogi Y.
      • Takahashi M.
      • Mabuchi N.
      • Miki H.
      • Fujiwara S.
      • Horikawa Y.
      • et al.
      Intracranial aneurysms: diagnostic accuracy of three-dimensional, Fourier transform, time-of-flight MR angiography.
      ,
      • Raaymakers T.W.
      • Buys P.C.
      • Verbeeten Jr., B.
      • Ramos L.M.
      • Witkamp T.D.
      • Hulsmans F.J.
      • et al.
      MR angiography as a screening tool for intracranial aneurysms: feasibility, test characteristics, and interobserver agreement.
      ,
      • Grandin C.B.
      • Mathurin P.
      • Duprez T.
      • Stroobandt G.
      • Hammer F.
      • Goffette P.
      • et al.
      Diagnosis of intracranial aneurysms: accuracy of MR angiography at 0.5 T.
      ,
      • Tang P.H.
      • Hui F.
      • Sitoh Y.Y.
      Intracranial aneurysm detection with 3 T magnetic resonance angiography.
      ]. Okahara et al. [
      • Okahara M.
      • Kiyosue H.
      • Yamashita M.
      • Nagatomi H.
      • Hata H.
      • Saginoya T.
      • et al.
      Diagnostic accuracy of magnetic resonance angiography for cerebral aneurysms in correlation with 3D-digital subtraction angiographic images: a study of 133 aneurysms.
      ] found that the overall sensitivity range of MRA on a per-aneurysm basis were 60%–79%. In the same series, the sensitivity of all readers (38% to 55%) was significantly lower for small aneurysms (<3 mm) than for large aneurysms (68% to 89%). Moreover, the sensitivity was also lower in the detection of aneurysms with SAH (54% to 79%) than in the detection of aneurysms without SAH (65% to 79%). White et al. [
      • White P.M.
      • Teasdale E.M.
      • Wardlaw J.M.
      • Easton V.
      Intracranial aneurysms: CT angiography and MR angiography for detection prospective blinded comparison in a large patient cohort.
      ] reported the accuracy and sensitivity of 3D-TOF-MRA at 1.5 T as 85% and 67%, respectively, and the sensitivity for detection of aneurysms smaller than 5 mm was 35% compared with 86% for detection of aneurysms 5 mm or larger.
      The higher accuracy, and sensitivity reported herein are the direct result of technical evolutions in image acquisition and post-processing algorithms that have become available since the earliest reports [
      • Sevick R.J.
      • Tsuruda J.S.
      • Schmalbrock P.
      Three-dimensional time-of-flight MR angiography in the evaluation of cerebral aneurysms.
      ]. We used 3D–T1-weighted fast field (T1-FFE) sequences with field of view (FOV) 250 × 190 × 108; four slabs (180 slices), slice thickness, 0.8 mm; matrix, 732 × 1024; and an acquisition time of 8 min and 56 s to obtain the source image. This sequence allows a high spatial resolution of images for diagnosis of intracranial aneurysms. Analysis of the methods of previous investigators with poor detection rates for intracranial aneurysms reveals suboptimal MRA scan parameters [
      • Schwab K.E.
      • Gailloud P.
      • Wyse G.
      • Tamargo R.J.
      Limitations of magnetic resonance imaging and magnetic resonance angiography in the diagnosis of intracranial aneurysms.
      ]. Many authors used a comparatively low matrix and shorter time for aneurysm detection, which may lead to their lower reported sensitivities for aneurysm detection. The application of the advanced 3.0 T MRI systems which have an increased signal-to-noise ratio and improved background suppression may be another important factor for the accurate diagnosis of the intracranial aneurysms because it allows better delineation of vessel walls and better visualization of the intracranial aneurysms and the relative position between the aneurysms and the adjacent. In addition, a trained and experienced neuroradiologist or neurosurgeon may greatly enhance the diagnostic accuracy of the intracranial aneurysms.
      In this study, some false-positive and negative results observed on the basis of 3D DSA findings. In the final VR-DSA, 3D-TOF-MRA missed two aneurysms <3 mm in diameter (false-negatives) and two aneurysms >3 mm in diameter (false-negatives), and misdiagnosed two aneurysms <3 mm in diameter (false-positives). Two aneurysms >3 mm at anterior communicating artery and left C7 were missed due to motion artifacts, and the other two aneurysms <3 mm at right C5 and C6 were missed due to misdiagnose with infundibula. The two false-positives at the left and right M1-2 were misdiagnosed because of the excessively tortuous and overlaid by the parent arteries which made it looked like small aneurysms.
      There were limitations to this study. Firstly, this was a single-centre study. Secondly, the neck size of the aneurysm was overestimated on VR 3D-TOF-MRA, and the outline of large aneurysms cannot be ascertained by VR 3D-TOF-MRA due to the loss of blood-flow-related signals. Thirdly, for a few patients with SAH and a GCS <15, MRA may misdiagnose or may miss aneurysms as a result of severe movement. In this situation, CTA may be a better diagnostic method due to the short acquisition time.

      5. Conclusions

      VR 3D–TOF-MRA is a non-invasive approach with high accuracy in the diagnosis of intracranial aneurysms in patients with SAH (GCS = 15′) or without SAH. And, it could be a preferred contrast-free and noninvasive modality for the evaluation of intracranial aneurysms for patients with SAH, even when the aneurysm's diameter is <3 mm.

      Acknowledgment

      This study was supported by the National Health and Family Planning Commission of the People's Republic of China (Contract No.: 201301006 ).

      Conflict of interest statement

      We declared that we had no conflict of interests.

      References

        • Broderick J.P.
        • Brott T.G.
        • Duldner J.E.
        • Tomsick T.
        • Leach A.
        Initial and recurrent bleeding are the major causes of death following subarachnoid hemorrhage.
        Stroke. 1994; 25: 1342-1347
        • Hirai T.
        • Korogi Y.
        • Ono K.
        • Murata Y.
        • Suginohara K.
        • Omori T.
        • et al.
        Preoperative evaluation of intracranial aneurysms: usefulness of intraarterial 3D CT angiography and conventional angiography with a combined unit—initial experience.
        Radiology. 2001; 220: 499-505
        • Anxionnat R.
        • Bracard S.
        • Ducrocq X.
        • Trousset Y.
        • Launay L.
        • Kerrien E.
        • et al.
        Intracranial aneurysms: clinical value of 3D digital subtraction angiography in the therapeutic decision and endovascular treatment.
        Radiology. 2001; 218: 799-808
        • Cloft H.J.
        • Joseph G.J.
        • Dion J.E.
        Risk of cerebral angiography in patients with subarachnoid hemorrhage, cerebral aneurysm, and arteriovenous malformation: a meta-analysis.
        Stroke. 1999; 30: 317-320
        • Willinsky R.A.
        • Taylor S.M.
        • TerBrugge K.
        • Farb R.I.
        • Tomlinson G.
        • Montanera W.
        Neurologic complications of cerebral angiography: prospective analysis of 2899 procedures and review of the literature.
        Radiology. 2003; 227: 522-528
        • Papke K.
        • Kuhl C.K.
        • Fruth M.
        • Haupt C.
        • Schlunz-Hendann M.
        • Sauner D.
        • et al.
        Intracranial aneurysms: role of multidetector CT angiography in diagnosis and endovascular therapy planning.
        Radiology. 2007; 244: 532-540
        • Wardlaw J.M.
        • White P.M.
        The detection and management of unruptured intracranial aneurysms.
        Brain. 2000; 123: 205-221
        • Hiratsuka Y.
        • Miki H.
        • Kiriyama I.
        • Kikuchi K.
        • Takahashi S.
        • Matsubara I.
        • et al.
        Diagnosis of unruptured intracranial aneurysms: 3 T MR angiography versus 64-channel multi-detector row CT angiography.
        Magn Reson Med Sci. 2008; 7: 169-178
        • Atlas S.W.
        • Sheppard L.
        • Goldberg H.I.
        • Hurst R.W.
        • Listerud J.
        • Flamm E.
        Intracranial aneurysms: detection and characterization with MR angiography with use of an advanced postprocessing technique in a blinded-reader study.
        Radiology. 1997; 203: 807-814
        • Schwab K.E.
        • Gailloud P.
        • Wyse G.
        • Tamargo R.J.
        Limitations of magnetic resonance imaging and magnetic resonance angiography in the diagnosis of intracranial aneurysms.
        Neurosurgery. 2008; 63 (discussion -5): 29-34
        • Okahara M.
        • Kiyosue H.
        • Yamashita M.
        • Nagatomi H.
        • Hata H.
        • Saginoya T.
        • et al.
        Diagnostic accuracy of magnetic resonance angiography for cerebral aneurysms in correlation with 3D-digital subtraction angiographic images: a study of 133 aneurysms.
        Stroke. 2002; 33: 1803-1808
        • White P.M.
        • Teasdale E.M.
        • Wardlaw J.M.
        • Easton V.
        Intracranial aneurysms: CT angiography and MR angiography for detection prospective blinded comparison in a large patient cohort.
        Radiology. 2001; 219: 739-749
        • Korogi Y.
        • Takahashi M.
        • Mabuchi N.
        • Miki H.
        • Fujiwara S.
        • Horikawa Y.
        • et al.
        Intracranial aneurysms: diagnostic accuracy of three-dimensional, Fourier transform, time-of-flight MR angiography.
        Radiology. 1994; 193: 181-186
        • Li M.H.
        • Li Y.D.
        • Tan H.Q.
        • Gu B.X.
        • Chen Y.C.
        • Wang W.
        • et al.
        Contrast-free MRA at 3.0 T for the detection of intracranial aneurysms.
        Neurology. 2011; 77: 667-676
        • Li M.H.
        • Li Y.D.
        • Gu B.X.
        • Cheng Y.S.
        • Wang W.
        • Tan H.Q.
        • et al.
        Accurate diagnosis of small cerebral aneurysms ≤5 mm in diameter with 3.0-T MR angiography.
        Radiology. 2014; 271: 553-560
        • Raaymakers T.W.
        • Buys P.C.
        • Verbeeten Jr., B.
        • Ramos L.M.
        • Witkamp T.D.
        • Hulsmans F.J.
        • et al.
        MR angiography as a screening tool for intracranial aneurysms: feasibility, test characteristics, and interobserver agreement.
        AJR Am J Roentgenol. 1999; 173: 1469-1475
        • Grandin C.B.
        • Mathurin P.
        • Duprez T.
        • Stroobandt G.
        • Hammer F.
        • Goffette P.
        • et al.
        Diagnosis of intracranial aneurysms: accuracy of MR angiography at 0.5 T.
        AJNR Am J Neuroradiol. 1998; 19: 245-252
        • Tang P.H.
        • Hui F.
        • Sitoh Y.Y.
        Intracranial aneurysm detection with 3 T magnetic resonance angiography.
        Ann Acad Med Singapore. 2007; 36: 388-393
        • Sevick R.J.
        • Tsuruda J.S.
        • Schmalbrock P.
        Three-dimensional time-of-flight MR angiography in the evaluation of cerebral aneurysms.
        J Comput Assist Tomogr. 1990; 14: 874-881