The Use of 3D FS T1 Sequence to Diagnose Cervical Arterial Dissections in Peripartum Populations: A Case Series

Introduction

Cervical artery dissections (CAD) are the leading cause (20-25%) of ischemic stroke in young and middle age adults. CADs often occur due to the formation of mural hematomas in the arterial walls and most commonly affect the internal carotid and vertebral arteries [1]. CAD generally happen after cervical trauma, but can also occur spontaneously in the absence of other associated risk factors such as recent infections, hypertension, and connective tissue disorders (Ehlers Danlos and Marfan syndrome) [2]. Although they can have benign clinical manifestations such as neck pain, cranial nerve palsy and Horner’s syndrome, a majority of patients with CAD eventually develop TIA or stroke like symptoms [3].

One particular population that is at increased risk is the peripartum woman. It is still unclear about what factors cause increased risk within this population but multiple cases have been documented in the academic literature [4]. Patients can present at any time during the pregnancy or recently after and often have symptoms similar to those described above however there have been reports of stroke, subarachnoid hemorrhage (SAH), and even death [5]. Some of the mechanisms of action that have been proposed to cause CAD in the pregnant population include hormonal and mechanical factors, hypertensive disorders such as preeclampsia, and other pregnancy related etiologies [6, 7].

Imaging plays a crucial role for the initial diagnosis and follows up of CAD. Although conventional angiography is the gold standard for evaluating the arterial lumen, it doesn’t offer information about the vascular wall and is unable to detect mural hematomas, thus increasing false negative rates. On the contrary, MRI imaging provides direct visualization of the mural hematoma with greater sensitivity and specificity than CT [1]. MRI also facilitates the detection of luminal abnormalities and cerebral ischemic events resulting from CAD. Thus it is the imaging of choice when suspecting CAD [3].

Currently, the 2D fat saturated (FS) T1 weighted sequence spin echo (SE) is one of the main tools in the protocol to study cervical artery dissections, yet this sequence has many limitations. The implementation of a 3D FS T1 sequence using different flip angle evolutions (SPACE), has proven to have increased and possibly better value than the 2D FS T1 SE sequence on the optimal diagnosis of cervical artery dissection [8]. We present 3 cases of recently postpartum females who presented with neck pain, headache, and several other neurologic symptoms including stroke like symptoms and were found to have CADs on 3D FS T1 sequence MRI.

Discussion

Stroke is an important factor when evaluating maternal morbidity and mortality during pregnancy. The incidence of stroke in this population is currently, 4-40 strokes per 100,000 pregnancies, and it keeps uptrending when associated with comorbid conditions such as hypertension, obesity, hypercoagulable state and cardiac disease [6]. CAD is recognized as a rare etiology for ischemic stroke in pregnancy (6%) with only about 30 cases reported in literature, mainly in the postpartum period [9].

The higher risk for CAD during pregnancy may be related to the different cardiovascular changes that take place during this period such as increase in cardiac output, increased blood flow through the carotids, and vascular compliance, which can increase shearing forces across the endothelial lining making the vessels more susceptible to intimal tear and dissection [6]. Additionally, the hypercoagulable state associated with pregnancy, can lead to the formation of thrombus at the dissection site and embolism [9].

One of the challenges, that cause delays in the clinical diagnosis of CAD during pregnancy, is the fact that the main initial symptom in nonpregnant as well as pregnant population is headache and neck pain [7]. There is a high prevalence of headaches during pregnancy associated with other conditions such as preeclampsia, eclampsia, withdrawal from caffeine, post lumbar puncture headache, and increased stress [10]. Thus the use of noninvasive imagines is crucial for prompt diagnosis under a strong clinical suspicion for CAD, since immediate therapy with anticoagulation should be initiated to prevent/minimize damage, particularly with ischemic lesions [10].

Currently, the 2D fat saturated (FS) T1 weighted sequence spin echo (SE) is one of the main tools in the protocol to study cervical artery dissections. Fat suppression is crucial to be able to differentiate subacute intramural hemorrhage from surrounding fat tissue since both produce hyperintense signal. Although this technique is effective, it has many limitations: it is time consuming to cover all neck and head area and is suboptimal on arteries with tortuous course [6]. The implementation of a 3D FS T1 sequence using different flip angle evolutions (SPACE) has proven to have added and possibly better value to the 2D FS T1 SE sequence on the optimal diagnosis of cervical artery dissection [6]. The 3D T1 FS SPACE main parameters consists of 56 slices with 256 x 256 matrix (with a 512 x 512 interpolation), echo time 22 ms, repetition time 750, variable flip angle, a field of view of 23 x 23 cm and acquisition time of 3 min 25 sec [11].

Some research studies have already demonstrated the advantages of the 3D T1 FS SPACE sequence when compared to the 2D FS T1 SE sequence. This sequence is able to cover all arterial portions (cervical and intracranial) in a single acquisition and with specific parameters being SPAIR fat saturation mode, averages 1.4 shorter acquisition time (3mins) when compared to the 2S FS T1 SE (11min) [12]. This provides the advantage to detect all lesions in just one acquisition in a shorter MR protocol time [12]. Additionally this sequence allows multiplanar reformatting that clearly delineate the hyper intensity of the intramural hematoma which facilitates the assessment of the extension of the dissection, particularly when looking at tortuous arterial segments such as the carotid petrous entry or the atlas loop of the vertebral artery that have been previously missed by 2D FS T1 SE [13]. It also makes it easier to analyze vascular loops, particularly in the V3 segment of the vertebral artery. In addition it facilitates reproducible comparisons with gadolinium enhanced MRA [8].

The T1 FS SE sequence has noted limitations in terms of quality in the presence of magnetic susceptibility artifacts. Previous case reports have demonstrated the ability of 3D T1 FS SPACE sequence to overcome magnetic susceptibility artifacts from dental implants, surgical metallic hardwares as well as movement artifacts [14, 15]. This enhances optimal analysis of the extension of the dissections. Moreover, it is also known that the quality of fat suppression is suboptimal at lower cervical levels, a limitation that is also overcome by 3D T1 FS SPACE sequence [14]. Other advantages of the 3D T1 FS SPACE sequence reported in literature are its higher sensitivity for detecting tumors in the cervical region, MS brain lesions and leptomeningeal abnormalities when compared to other sequences such as the T1 MPRAGE while allowing reformatting in any plane [11, 16].

Conclusion

Our cases highlight the advantages of adding the 3D T1 FS SPACE sequence to the protocol for the diagnosis of cervical artery dissection in women during pregnancy and postpartum period under the appropriate clinical setting.

We encourage physicians to consider ordering the SPACE MRI study for any woman during terminal pregnancy or puerperium period that presents with persistent headaches, neck pain and associated neurological symptoms such as ipsilateral Horner syndrome or TIA, particularly with underlying risk factors such as preeclampsia, hypertension and obesity, since they would be at high risk of having cervical artery dissection. With its many advantages (such as faster speed of acquisition, greater field of coverage and superior spatial resolution), 3D SPACE MRI will speed early and accurate diagnosis of this entity, facilitate prompt treatment, prevent associated ischemic events and subsequently decrease the rate of morbidity and mortality in this high risk population.

Disclosure: The authors report no conflict of interest concerning the materials of methods used in this manuscript. This research was supported (in whole or in part) by HCA and/or an HCA affiliated entity. The views expressed in this publication represent those of the author(s) and do not necessarily represent the official views of HCA or any of its affiliated entities. There were no external sources of financial support for this manuscript.