Importance of T2*-weighted gradient-echo MRI for diagnosis of cortical vein thrombosis

https://doi.org/10.1016/j.ejrad.2005.05.010Get rights and content

Abstract

We examined six patients with isolated venous thrombosis (n = 2), or venous thrombosis combined with sinus thrombosis (n = 4) (CVT). The clinical symptoms were non-specific (acute cephalea, paresis, epileptic seizure, progressive speech disorder). All examinations were performed on a 1.5 T system (Magnetom Symphony, Siemens, Erlangen, Germany), maximum gradient field strenght 30 mT/m, minimal gradient rise time 450 μs, according to the following protocol:

Transverse T2-weighted turbo spin-echo (TSE), fluid attenuated inversion recovery (FLAIR), T1-weighted spin-echo (SE), before and after administration of contrast medium, T2*-weighted conventional gradient-echo (GRE), T2*-weighted spin-echo echo planar imaging (SE EPI), both without and with diffusion weighting as well as two-dimensional (2D) venous time-of-flight (TOF) MRA.

The venous thromboses were best detectable in the T2*-weighted conventional GRE sequence in all patients. In two patients, the CVT was discernible only in this sequence. The sinus thrombosis was well discernible only in the T2*-weighted GRE sequence in only one case; in the remaining cases it was detectable only with difficulty. For these cases, other sequences such as SE, diffusion-weighted, or 2D-TOF-MRA sequence were superior. The T2*-weighted conventional GRE sequence was superior to the T2*-weighted SE EPI sequence in all patients. To sum up, it can be concluded, that T2*-weighted conventional GRE sequences are possibly the best method of detection of acute cortical vein thromboses. Therefore, it seems to be of benefit to integrate a T2*-weighted conventional GRE sequence into the MR-protocol for the diagnosis of isolated cortical vein thrombosis.

Introduction

Cerebral vein thromboses (CVT) can have severe clinical outcome, some even leading to death if they are not diagnosed in time [1]. Clinical diagnosis [2] and imaging is often extremely difficult and many cases remain unrecognized [3], [4]. Moreover, isolated CVT seem to occur only very rarely. MR imaging has by now replaced catheter angiography in imaging diagnosis of isolated CVT and sinus thrombosis [5]. Intravasal clot disintegration as well as flow changes due to the thrombosis can lead to signal changes in T2- and T1-weighted spin-echo (SE) sequences. However, especially in the acute phase, T2- and T1-weighted SE sequences are not very sensitive, particularly in the case of thromboses of small veins, because the intravasal clot may not be distinguishable, or only very poorly distinguishable, from the normal blood flow signal [6]. Theoretically, T2*-weighted gradient-echo (GRE) sequences should be better suited since these exhibit considerably higher sensitivity to susceptibility differences [7]. Thus, Selim et al. could show that T2*-weighted echo planar imaging (EPI) sequences can improve the detection of intravasal clots [3]. However, in Selim's study exclusively EPI sequences were used, which may have a lower sensitivity for lesions with susceptibility effects (e.g. acute or chronic blood clots or calcifications) than according to our experience, T2*-weighted conventional GRE sequences such as T2*-weighted FLASH.

In our study, five patients with isolated or cortical vein thromboses combined with sinus thromboses were examined. To our knowledge, this is the first study to analyze the detection of isolated and combined venous thromboses in SE-sequences before and after administration of contrast medium, T2*-weighted conventional GRE and EPI (including diffusion) sequences, and 2D time-of-flight (TOF) MR angiography (MRA) sequences.

Section snippets

Materials and methods

In this study, six patients with isolated (n = 2) or venous thrombosis combined with sinus thrombosis (n = 4) were examined (four women, two men, aged 21–77 years). The clinical symptoms were non-specific (acute cephalea, paresis, epilepsy, coma, hemiplegia, progressive speech disorder). Four of these patients were in the acute stage (<7 days), two were in the subacute stage (>7 days) after the onset of clinical symptoms. After neurological therapy four patients exhibited a complete remission of

Results

The results are shown in Table 1. For detection of cortical vein thrombosis, T2*-weighted gradient-echo was the only technique, which was positive in all cases.

In one of the two patients with isolated CVT, the thrombosis was only discernible in the T2*-weighted conventional GRE sequence. In the second patient the venous thrombosis was best visualized in the T2*-weighted conventional GRE sequence. The remaining sequences were considerably inferior in the delineation of CVT.

The other four

Discussion

Various studies have shown that MR imaging is very useful in the diagnosis of CVT [8], [9]. Up to now, the indicator of CVT was the missing signal void in standard T1- and T2-weighted SE sequences, due to the presence of intravasal clots. In addition, this is mainly the fact for larger vessels only, such as the sinus, whose presence at the given localizations is clear on the basis of anatomical landmarks. Small vein thromboses are often not detectable with these techniques (Fig. 1), because

References (13)

There are more references available in the full text version of this article.

Cited by (53)

  • Cerebral Venous Thrombosis

    2021, Stroke: Pathophysiology, Diagnosis, and Management
  • Current imaging modalities for diagnosing cerebral vein thrombosis – A critical review

    2020, Thrombosis Research
    Citation Excerpt :

    However, this technique was not sensitive for assessing smaller veins (i.e. in branches of cortical veins) [55,57,58]. Most studies evaluated the diagnostic accuracy of non-contrast-enhanced flow related MRI techniques compared to the combination of multiple imaging modalities and final clinical outcome or contrast-enhanced MRV (Appendix 5) [21,32,59–68], were adequate sensitivity and specificity for CVT were found too. Non-contrast-enhanced TOF MRV and PC MRV had a sensitivity of 64–100% and 48–100%, respectively, although with wide 95% confidence intervals.

  • Cerebral venous thrombosis: A case series and a neuroimaging review of the literature

    2018, Journal of Clinical Neuroscience
    Citation Excerpt :

    However the T2* gradient-recalled echo (GRE) sequences have recently been considered useful for the early diagnosis of CVT due to the sensitivity of these sequences to the paramagnetic product of hemoglobin; they allow to show the blooming artifacts in the thrombosis venous segments as an hypointense signal [28]. Selim et al. and Fellner et al. studies [29,30] has shown the high diagnostic sensitivity of T2* GRE sequences in detecting CVT, even in the narrower brain venous districts due to the physical characteristics of this sequence that is highly susceptible to magnetic susceptibility artifacts; the use of GRE sequences help the recognition of the acute stage of thrombosis, showing it as an hypointense area due to the marked susceptibility to artifacts by magnetic field irregular signal, induced by hemoglobin degradation products. The non linear magnetic susceptibility of the ferromagnetic substances causes signal alteration also on the adjacent tissues, so in case of CVT, the products of blood degradation produce alteration of signal also at the surrounding water molecules.

  • Diagnostic value of T2*-weighted gradient-echo MRI for segmental evaluation in cerebral venous sinus thrombosis

    2015, Clinical Imaging
    Citation Excerpt :

    In gradient-echo (GRE) T2*-weighted sequences, exaggerated signal loss is often seen because of the increased susceptibility effect of certain paramagnetic breakdown products of hemoglobin, which can be particularly useful in the early diagnosis of CVST [2,3]. Few reports have evaluated the appearance of CVT on GRE T2*-weighted sequences [3,5–7]. Moreover, only one report shows the GRE appearance of cerebral venous thrombi in different stages of evolution [8].

  • The value of T2*-weighted gradient-echo MRI for the diagnosis of cerebral venous sinus thrombosis

    2013, Clinical Imaging
    Citation Excerpt :

    Otherwise, an intermediary phase can also occur related to intracellular methemoglobin and where there is hyperintense signal intensity on T1 weighting and hypointense signal intensity on T2 weighting [7]. T2*-weighted GRE sequences are more sensitive to paramagnetic effects than those based on spin echo techniques [3,8]. A paramagnetic compound (such as deoxyhemoglobin, intracellular methemoglobin, and hemosiderin) can produce significant signal intensity loss on GRE sequences due to its magnetic susceptibility effect.

View all citing articles on Scopus
View full text