Clinical Imaging xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Clinical Imaging journal homepage: http://www.clinicalimaging.org

Cerebral venous thrombosis as a rare cause of subarachnoid hemorrhage: case report and literature review☆ Neslin Sahin a,⁎, Aynur Solak a, Berhan Genc a, Nalan Bilgic b a b

Sifa University School of Medicine, Izmir; Department of Radiology Sifa University School of Medicine, Izmir; Department of Neurology

a r t i c l e

i n f o

Article history: Received 23 November 2013 Received in revised form 4 February 2014 Accepted 12 March 2014 Available online xxxx Keywords: Cerebral venous thrombosis Headache MR venography Subarachnoid hemorrhage

a b s t r a c t We report a 48-year-old woman presenting with subarachnoid hemorrhage (SAH) as the first manifestation of superior sagittal sinus thrombosis. In a literature review of 73 cases, SAH associated with cerebral venous thrombosis (CVT) was usually seen at the cerebral convexities. SAH was adjacent to thrombosed venous structures; therefore, the most possible explanation seems to be the rupture of cortical veins due to extension of thrombosis. Computed tomography (CT) was effective for diagnosis of CVT in only 32% of the cases. CVT should be considered when SAH is limited to cerebral convexities and magnetic resonance (MR) imaging with MR venography should be performed. © 2014 Elsevier Inc. All rights reserved.

1. Introduction

2. Literature review

Cerebral venous thrombosis (CVT) is a relatively uncommon cerebrovascular disease. Although CVT can occur at any age, it predominates in children and young adults, accounting for 1–2% of strokes [1]. CVT is diagnosed more commonly than previously thought with the advent of accurate noninvasive imaging methods. The estimated annual incidence of CVT is reported to be between two and seven cases per 1 million populations, but it is estimated that five to eight cases may be diagnosed at a tertiary care referral center [2]. CVT is a potentially life-threatening disease, and the clinical diagnosis can be difficult because of a wide spectrum of clinical presentations and numerous causes. There are many predisposing factors for CVT classifying as local (sinus trauma, regional infection, and neoplastic invasion or compression) or systemic (hereditary coagulopathies, peripartum state, dehydration, oral contraceptive use, and hypercoagulable states secondary to malignancy). The etiology is unknown in 25% of cases [1–3]. We report a patient presenting with subarachnoid hemorrhage (SAH) as the first manifestation of superior sagittal sinus (SSS) thrombosis. The presentation of CVT with SAH is very rare. Therefore, we performed a literature review of CVT as a cause of SAH and discussed the preferred imaging modalities in the diagnosis of this presentation.

The review of the literature included articles published in English from 1996 to 2012. We performed the literature search by using the terms subarachnoid hemorrhage, cerebral venous thrombosis and subarachnoid hemorrhage, and cortical vein thrombosis on PubMed and Medline cross-referencing pertinent articles from initial PubMed and MedLine searches. We also used the translated English abstracts of three case reports with sufficient information which were not available in English [4–6].

☆ Conflict of interest: The authors report no conflicts of interest. ⁎ Corresponding author. Sifa University School of Medicine, Department of Radiology, Fevzipasa Boulevard No. 172/2, 35240 Basmane Izmir, Turkey. Tel.: +90-232-343-44-45; fax: +90-232-343-56-56. E-mail address: [email protected] (N. Sahin).

3. Case report A 48-year-old woman with no significant medical history was admitted with a 1-week history of acute onset of severe headache and gait disturbance. Her level of consciousness was normal. She was afebrile, and her neurological examination was normal, with no evidence of meningismus. Magnetic resonance (MR) imaging; T1-weighted, T2-weighted, fluid-attenuated inversion recovery (FLAIR), diffusionweighted imaging (DWI), and susceptibility-weighted imaging (SWI) followed by MR venography (MRV) in two-dimensional (2D) time-offlight (TOF) mode (1.5 T Espree, Siemens, Erlangen, Germany) were performed on the same day. T1-weighted and FLAIR images (Fig. 1A, B) demonstrated abnormally increased signal intensity in the sulci of the bilateral frontoparietal convexity, better identified on the FLAIR sequence, and SWI (Fig. 1C) showed hypointense signal intensity most compatible with SAH. In addition, there were abnormal hyperintense signals within the SSS on T1-weighted, T2-weighted, and FLAIR images which corresponded to subacute venous thrombosis (Fig. 1A, B, D). MRV confirmed the diagnosis of SSS thrombosis (Fig. 1E). There was

http://dx.doi.org/10.1016/j.clinimag.2014.03.005 0899-7071/© 2014 Elsevier Inc. All rights reserved.

Please cite this article as: Sahin N, et al, Cerebral venous thrombosis as a rare cause of subarachnoid hemorrhage: case report and literature review, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.03.005

2

N. Sahin et al. / Clinical Imaging xxx (2014) xxx–xxx

no restricted diffusion indicative of acute infarct on DWI. MR angiography showed no evidence of an underlying vascular lesion. No predisposing event and risk factors were identified. The patient improved with anticoagulant therapy.

4. Results We reviewed 38 articles with a total of 73 cases of SAH associated with CVT [4–41]. We essentially focused on diagnostic work-ups and radiological findings of SAH in the setting of CVT besides clinical presentations. We included the location of SAH and CVT and associated findings with respect to diagnostic imaging for this underreported phenomenon. Details regarding demographic features and radiological findings of CVT with SAH are given in Table 1. We are aware that the literature review is limited by a wide range of variability of imaging methods, subjects, interpretations, and quality of images in the reported data for each case. However, we could extract enough information to discuss the preferred imaging modalities in the diagnosis. Thirty-one (42%) patients were male and 42 (58%) female which are consistent with previous reports with female predominance. The age range was between 1 and 83 years with a mean of 42 years. The imaging modality for SAH was not available in 1 patient [6]. In 11 of a total of 72 patients, SAH was detected only on MRI, and for 5 of them, only MRI was performed [14,17,24,31]. We noted that SAH was detected on computed tomography (CT) in 61 of a total of 67 cases, whereas initial CT was normal in 4 of these 61 patients [13,28,41]. So, CT failed to detect SAH in only 6 (9%) of the cases [18,20,26,27,34,39]. Initial MRI was ineffective in 1 patient, and diagnosis was made by the subsequent MRI [17]. We could not evaluate the effectiveness of MRI for SAH because the first-line imaging test for SAH was generally CT and the findings for SAH on subsequent MRIs were not mentioned. The imaging modality for CVT was not available in three patients [21,32]. For diagnosis of CVT, MRV or digital subtraction angiography (DSA) was performed in 6 of 70 patients [27]. MRV was done for 25 of a total of 64 patients, CT venography (CTV) for 4 patients, and DSA for 38 of them alone or in combination with other modalities. DSA was the only diagnostic method used for 12 patients for CVT. According to the literature, CT was diagnostic in 21 (31%) of a total of 67 patients for CVT; however, in only 3 of them, CT alone was sufficient for clinical management [13,32]. For the other cases, further diagnostic imaging with MRI, MRV, CTV, and/or DSA was performed to characterize or confirm the extent of the thrombosis. MR imaging was usually performed with MRV and mostly used to detect parenchymal abnormalities; we could not get sufficient information about the effectiveness of MRI without MRV for CVT detection. MR imaging alone was the preferred diagnostic modality for CVT in only 5 patients [8,17,18,26,38]. However, MRI played an essential role in 5 of 10 cases of isolated cortical vein (c) thrombosis, whereas DSA or MRV was normal [14,38,39]. There were no associated findings in 40 (55%) of the 73 patients additional to SAH in the setting of CVT. Nonhemorrhagic infarct (NHI) was seen in 4 patients, venous hemorrhagic infarct (HI) in 11 (15%) patients, and edema (E) in 12 (16%) patients. There was parenchymal hemorrhage (PH) in 8 (11%) patients and subdural hemorrhage (SDH) in 2 patients. In the review of the prior reports, SSS was the most frequently thrombosed sinus as SSS was the only involved sinus in at least 14 of the 72 cases, with or without transverse sinus (T) in 6 of them [27] and with other sinuses in 30 of them. T was the second thrombosed sinus (at least 35 of the 72 cases) and was usually involved with SSS and sigmoid sinus (S) as S was never alone involved without T. Straight sinus (St) was involved in 10 cases mostly with SSS and/or transverse/Ss. Venous thrombosis of cerebral veins (CVs) including cs, galen, labbe, and trolard was seen in 23 of a total of 70 patients, while

c thrombosis was associated with dural sinus thrombosis in 6 patients and was isolated in 10 of them. The review of the previous reports showed that SAH was usually seen at cerebral convexities but never involved the skull base and basal cisterns in 62 of 65 cases. We noted that SSS thrombosis was usually associated with SAH at frontoparietal convexity, and less commonly at interhemispheric fissure (IF), and sylvian fissure. In sylvian SAH, SSS and transverse/Ss were frequently involved together. Infratentorial (I) SAH was also usually associated with the thrombosis of SSS and/or transverse and Ss. In addition, the parieto-occipital (O) and posterior temporal convexity SAH was related to T thrombosis and paramedian supra- and/or I SAH to St thrombosis. The distribution of thrombosed cerebral venous structures with the location of SAH was shown in Fig. 2. 5. Discussion In our case, we observed SAH in the sulci of the convexity and subacute venous thrombosis in the SSS as a cause of SAH on MRI and MRV. According to the literature, the distribution of SAH associated with CVT was usually seen at cerebral convexities sparing the skull base and basal cisterns. Although the location of the SAH was variable, hemorrhage was usually adjacent to thrombosed venous structures. Nonenhanced CT (NECT) was effective for the diagnosis of SAH in 91% of the patients and for the diagnosis of CVT in only 31% of the patients. CVT is a potentially fatal disease, and accurate diagnosis is crucial for prompt appropriate therapy. The main symptoms are headache, partial or generalized seizures, focal neurological deficits, and alteration in mental state [3,7]. Headache is the most frequent presenting symptom in about 90% of patients with CVT and was described as diffuse (D) and progressed in severity over days [42]. De Brugin et al. [8] described 10 patients presented with thunderclap headache mimicking SAH, which suggests a frequency of thunderclap headache of more than 10% in CVT patients. Clinical data alone are usually not sufficient for a definitive diagnosis of CVT; therefore, imaging plays an essential role to make the diagnosis. Venous thrombi can be detected on CT and MR parenchymal images or with various venographic techniques including unenhanced MRV, contrast-enhanced MRV, and CTV [2]. NECT is the initial imaging examination in most cases of sudden acute neurological symptoms. The CT imaging findings related to CVT include hyperdense thrombus in the occluded sinus, the delta sign and cord sign on NECT, and intraluminal thrombus with enhancement of the dural sinus wall, the empty delta sign on enhanced CT, which is present in 16–46% of cases [1]. These findings have low sensitivity and specificity for the diagnosis of CVT. The delta sign can also be seen in SAH which is not always reliable for diagnosis. In addition, highvelocity venous flow in children and young adults, increased venous attenuation seen in the presence of elevated hematocrit, dehydration, and beam-hardening artifacts from the skull vault can result in a falsepositive sign by mimicking a hyperdense venous sinus clot [1,2,7]. Recent reports considered MRI, including MRV, as the technique of choice for definitive diagnosis in all phases and follow-up of CVT. Parenchymal abnormalities such as cytotoxic E, vasogenic E, hemorrhage, and infarct have been reported in as many as 57% of patients with CVT and can be identified more readily on MRI than on CT [2]. In CVT, cerebral E and infarction are usually subcortical and does not conform to an arterial vascular territory as in the arterial stroke. In addition, it should be noted that diffusion abnormalities are variable and often reversible in CVT. Venous obstruction results in increased intracranial pressure and consequently decreased cerebral blood flow. So, first vasogenic E develops with elevated apparent diffusion coefficient (ADC) values; however, areas of decreased ADC, which may also be reversible, may eventually occur, and may be explained by decreased cerebral blood flow with neuronal swelling and membrane pump failure without neuronal death [1,2]. According to

Please cite this article as: Sahin N, et al, Cerebral venous thrombosis as a rare cause of subarachnoid hemorrhage: case report and literature review, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.03.005

N. Sahin et al. / Clinical Imaging xxx (2014) xxx–xxx

3

Fig. 1. Axial T1W (A) and FLAIR (B) images demonstrate hyperintense signal in the sulci, better identified on the FLAIR sequence, and SWI shows (C) hypointense signal most compatible with hemorrhage (small white arrows). Axial T1W (A), axial FLAIR (B), and sagittal T1W (D) images reveal hyperintense signal within superior sagittal sinus consisting with thrombosis (big white arrows). Brain MRV (E) confirms the occlusion of superior sagittal sinus.

Please cite this article as: Sahin N, et al, Cerebral venous thrombosis as a rare cause of subarachnoid hemorrhage: case report and literature review, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.03.005

4

N. Sahin et al. / Clinical Imaging xxx (2014) xxx–xxx

Table 1 Summary of literature review of the presentation of CVT with SAH including demographic features and radiological findings Authors

Age, sex

SAH modality

CVT modality

CT for CVT

Location of CVT

Location of SAH

Associated findings

de Brugin et al. 1 (8) de Brugin et al. 2 Ohta et al. (5) Ciccone et al. (9) Ra et al. (10) Sztajzel et al. (11) Selim et al. (12) Widjaja et al. 1 (13) Widjaja et al. 2 Widjaja et al. 3 Oshiro et al. (6) Chang and Friedman 1 (14) Chang and Friedman 2 Chang and Friedman 3 Zare and Mirabdolbaghi (15) Tidahy et al. (16) Oppenheim et al. 1 (17) Oppenheim et al. 2 Oppenheim et al. 3 Oppenheim et al. 4 Spitzer et al. 1 (18) Spitzer et al. 2 Spitzer et al. 3 Adaletli et al. (19) Rice and Tang (20) Senel et al. 1 (21) Senel et al. 2 Senel et al. 3 Senel et al. 4 Senel et al. 5 Senel et al. 6 Lin et al. (22) Shukla et al. (23) Kasuga et al. (7) Matthew et al. (24) Ko et al. (25) Wang et al. (26) Tang et al. 1 (27) Tang et al. 2 Tang et al. 3 Tang et al. 4 Tang et al. 5 Tang et al. 6 Jaiser et al. (28) Lai et al. (29) Glikstein et al. (4) Lee et al. 1 (30) Lee et al. 2 Bittencourt et al. (31) Panda et al. 1 (32) Panda et al. 2 panda et al. 3 Pandaet al. 4 Panda et al. 5 Panda et al. 6 Panda et al. 7 Panda et al 8 Panda et al. 9 Panda et al. 10 Hegazi et al. (33) Sharma S. et al. (34) Field and Kleinig (35) Kato et al. (36) Oz et al. (37) Oda et al. 1 (38) Oda et al. 2 Oda et al. 3 Oda et al. 4 Renou et al. 1 (39) Renou et al. 2 Sharma B. et al (40) Sayadnasari et al. 1 (41) Sayadnasari et al. 2 Sahin et al.

37, F 32, F 54, F 36, F 60, M 58, F 36, M 33, F 22, F 43, F 43, F 46, F 29, F 64, F 45, F 27, F 69, M 55, F 32, F 51, F 41, F 65, M 42, M 14, M 56, F 27, F 44, F 43, F 1, F 32, M 41, F 44, M 40, M 39, M 40, M 38, F 33, F 43, F 37, F 46, F 48, M 30, M 23, M 53, F 34, M 56, M 72, F 39, F 31, F 32, M 50, M 27, F 33, M 32, F 25, M 38, M 30, M 38, M 25, M 38, F 59, M 70, M 52, F 22, M 83, M 67, F 57, M 31, F 74, F 36, F 40, M 42, F 36, M 49, F

NECT NECT NECT NECT NECT NECT NECT/MRI NECT(in) NECT NECT NA MRI MRI MRI/NECT NECT NECT/MRI MRI(in) NECT/MRI NECT NECT/MRI NECT NECT/MRI NECT(n)/MRI NECT/MRI NECT(n)/MRI NECT NECT NECT NECT NECT NECT NECT/MRI NECT NECT MRI NECT NECT(n)/MRI NECT CECT/MRI NECT/MRI CECT(n)/MRI NECT NECT NECT(in) NECT NECT/MRI NECT NECT MRI NECT NECT NECT NECT NECT NECT NECT NECT NECT NECT NECT NECT(n)/MRI NECT/MRI NECT NECT NECT/MRI NECT/MRI NECT/MRI NECT/MRI NECT/MRI CECT(n)/MRI NECT/MRI NECT(in) NECT(in) MRI

NECT/DSA MRI DSA DSA/MRV NECT/DSA CECT/DSA/MRV MRV/DSA NECT/DSA NECT NECT/CEMRI DSA MRI/MRV(n) MRI/MRV MRI/MRV(n) MRV/DSA NECT/CEMRI/MRV MRI/DSA MRI/NECT/DSA MRI/DSA CEMRI CTV MRV/DSA MRI DSA/MRI MRI/MRV NA NA DSA MRV DSA DSA CECT/DSA/CEMRI DSA/MRV CECT/DSA MRI/MRV DSA MRI/DSA(n) NECT and MRV or DSA CECT/MRI and MRV or DSA NECT/MRI and MRV or DSA CECT/MRI and MRV or DSA NECT and MRV or DSA MRV or DSA CTV/MRV DSA/MRI NECT/MRI/MRV DSA DSA/MRI MRI/MRV DSA MRV DSA CTV NECT CTV/MRV NECT DSA DSA Perioperative NECT/MRV MRV/DSA NECT/MRI/MRV DSA NECT/CEMRI/DSA MRI DSA/MRI MRI/DSA(n) MRI/MRV/DSA(n) MRI/DSA(n) MRI/DSA(n) MRV/DSA MRI/MRV MRI/MRV MRI/MRV

1 0 0 0 1 1 0 1 1 1 0 2 2 0 0 1 2 1 0 0 0 0 0 0 0 NA NA 0 0 0 0 1 0 1 2 0 0 1 1 1 1 1 0 0 0 1 0 0 2 0 0 0 0 1 0 1 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 0 2

SSS + T + S T+S SSS T + S + CV(c) SSS + T + S T+S SSS + T + St + CV + IJV SSS + CV(c) SSS SSS + T + St + CV(c) CV(c) CV CV CV SSS + T + S + St SSS + T SSS + T SSS + T SSS + T SSS + CV(c) SSS CV(c) SSS + T SSS + St + CV SSS + T T+S SSS SSS T SSS SSS SSS + T SSS + T + S + ISS SSS + T SSS + T CV(c) CV(c) SSS ± T SSS ± T SSS ± T SSS ± T SSS ± T SSS ± T SSS SSS SSS + T T + S + St + CV St + ISS + CV CV(c) SSS + T SSS + T + St SSS + T + St SSS SSS SSS SSS + T SSS ALL NA SSS + T SSS + T SSS SSS + T + S + St SSS + T T + S + CV(c) CV(c) CV(c) CV(c) CV(c) CV(c) SSS + T + S T+S SSS + T + S SSS

I I S T+S F+P I T+P+O S IF T+S NA CS F+P F+P T+S T+P F D D D F CS F D F+P IF F + IF I + IF P F IF P S S P F+P F+P NA NA NA NA NA NA F F+P P PM PM P F + P + IF + S IF + S F + IF F + P + IF + S S F F I+S S F NA F+P CS I+T F+T+S O S T+P T+P F P F+P P S F+P

NONE NONE HI NONE PH HI + SDH PH NONE PH NONE NHI NONE NONE NONE HI NONE NONE NONE NONE E E NONE NONE HI NHI HI NONE NONE NHI NONE NONE NONE NONE NONE PH + SDH NONE NONE PH + E E PH + E NONE NONE E NONE NONE NONE HI NHI E NONE NONE NONE NONE NONE HI HI NONE NONE HI NONE E NONE HI PH E E E E NONE NONE NONE HI PH NONE

Please cite this article as: Sahin N, et al, Cerebral venous thrombosis as a rare cause of subarachnoid hemorrhage: case report and literature review, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.03.005

N. Sahin et al. / Clinical Imaging xxx (2014) xxx–xxx

the literature, 45% of the patients presented with associated findings additional to SAH in the setting of CVT. Venous HI, E, and PH were the most commonly seen associated findings. It should be noted that these findings would be more prominent if the imaging modality was MRI instead of CT in some of the cases. The MR imaging findings of sinus thrombosis are due to the absence of a flow void and the presence of altered signal intensity in the sinus which is dependent on the age of the clot. In the acute stage of CVT, the signal of thrombus is predominantly isointense on T1W and strongly hypointense on T2W images that may mimic a normal flow state. Therefore, gradient-echo T2* sequence seems to be an important diagnostic aid by showing directly the thrombus with an abnormal low-signal intensity due to the magnetic susceptibility effect of deoxyhemoglobin in this stage. In the subacute phase, the signal becomes hyperintense on T1W and T2W images due to methemoglobin. The signal intensity of chronic thrombosis with incomplete recanalization of the sinus may vary, which is typically isointense or hyperintense on T2W images and isointense on T1W images [1,2]. In the review of previous reports, we noted that MRI with gradient-echo images were also helpful in the detection of isolated cortical venous thrombosis with blooming artifacts within the thrombosed veins, whereas other techniques fail to make the diagnosis. MRV, especially contrast-enhanced techniques, allows a confident diagnosis for CVT. The two most common techniques are based on TOF effects [2D and three-dimensional (3D) TOF techniques] of moving spins and on motion-induced phase shifts (the 3D phase-contrast technique). Two-dimensional TOF MRV is the most frequently used method because of excellent sensitivity to slow flow and diminished sensitivity to signal loss from saturation effects compared with the 3D TOF techniques. But venous flow in the plane of image acquisition may produce signal loss from saturation effect and cause false-positive results for interpretation, particularly in nondominant Ts. The other technical drawback is the increased signal intensity with the inclusion of substances of subacute thrombus, such as methemoglobin, mimicking normal flow signal intensity. Phase-contrast MRV is less often used because of the longer imaging time that makes the technique more susceptible to motion artifacts. The other disadvantages of this technique are aliasing artifacts and intravoxel phase dispersion due to turbulent flow. The variants of normal venous anatomy that may be seen as a flow gap are also potential pitfalls in image interpretation by mimicking sinus thrombosis for both techniques. Contrast-enhanced MRV is a relatively new venographic method which provides improved small-vessel visualization compared with that of TOF MRV. This technique is also considered to be superior in identification of the dural sinuses because of a decrease in flow-related artifacts [1,2]. In the current review, DSA was the most commonly preferred method to confirm the diagnosis of CVT, but we determined that the use of MRV was increased over the years due to improved imaging techniques. CTV is a rapid and accurate technique which is at least equivalent to MRV for the detection of CVT and a potential alternative to MRV particularly for imaging uncooperative patients. Drawbacks of this technique include the difficulty of reconstructing maximum intensity projection images from the source image data sets by the subtraction of all the bone adjacent to the dural sinuses. However, CTV may also have limited use because of the required use of ionizing radiation and iodinated contrast material [1,2,42,43]. In the current review, CTV was the least preferred method for the diagnosis of CVT. CVT may have similar clinical and radiological features to acute SAH and may also be associated with nonaneurysmal SAH. Therefore, the diagnosis of CVT can be more difficult, particularly when patients initially present with acute SAH which is a rare condition.

5

Fig. 2. MR venogram shows the distribution of thrombosed cerebral venous structures with the location of SAH.

Nontraumatic SAH results in 85% of ruptured aneurysms and in 10% of nonaneurysmal perimesencephalic (PM) hemorrhages, which manifest themselves with SAH at the skull base. The remaining 5% are related to various vascular and nonvascular pathologies such as arterial dissection, cerebral arteriovenous malformation, dural arteriovenous fistula, septic aneurysm, pituitary apoplexy, cocaine abuse, tumors, vasculitis, amyloid angiopathy, coagulopathies, posterior reversible encephalopathy syndrome, isolated angiitis of the central nervous system, amphetamine-induced vasculopathy, cavernoma bleeding, intracerebral abscesses and, finally and usually not mentioned, CVT [7,39]. The exact cause of SAH in the setting of CVT is unknown. Different theories have been proposed to explain SAH associated with CVT: (a) CVT may cause a local inflammatory response that allows extravasation of blood into the subarachnoid space due to increase in vascular permeability; (b) venous HI may be responsible for secondary rupture into subarachnoid spaces; however, HI is not commonly described in the literature that Panda et al. [32] detected in only 1 of 10 patients with SAH due to CVT, and there was HI in only 11 of a total of 73 cases in the literature; (c) the superficial veins bridging the subarachnoid and subdural spaces have thin walls, no smooth muscle fibers, and no valves, which enable an important capacitance of CVs and reversal of the direction of blood flow if drainage is occluded. CVT associated with secondary venous hypertension may trigger rupture of these fragile superficial veins. Furthermore, dural sinus thrombosis may extend into the superficial veins and cause dilation of fragile cs with localized venous hypertension and subsequently bleed into subarachnoid (SA) space. The specific anatomical features of the cs explain the mechanism of rupture in subarachnoid space and rare manifestation of SAH in CVT. This mechanism seems more likely to be responsible because SAH is usually seen adjacent to thrombosed venous structures as in our case [7,17,19,32]. The possible causes of SAH in the setting of CVT are shown in Fig. 3. We found a total of 73 reviewed cases of SAH associated with CVT in the literature. Although this presentation has rarely been reported, the cases seem to be increasingly diagnosed in recent years most

Modality: NECT; (in) = initial normal; NA = not available; (n) = normal; CECT = contrast-enhanced CT; CEMRI = contrast-enhanced MRI. CT for CVT: 0 = ineffective; 1 = effective; 2 = not performed. Location of CVT: SSS; T; S; CV (including c, galen, labbe, and trolard); St; IJV = internal jugular vein; ISS = inferior sagittal sinus. Location of SAH: I; Sy = sylvian; Te = temporal; F = frontal; P = parietal; O; IF; CS = central sulcus; D; PM. Associated findings: HI; PH; SDH; NHI; E. In addition, the findings of our case are presented in the last paragraph.

Please cite this article as: Sahin N, et al, Cerebral venous thrombosis as a rare cause of subarachnoid hemorrhage: case report and literature review, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.03.005

6

N. Sahin et al. / Clinical Imaging xxx (2014) xxx–xxx

Fig. 3. Possible causes of SAH associated with CVT. The most possible explanation for this presentation (written in bold) seems to be the rupture of dilated cs due to extension of thrombosis and subsequent venous hypertension.

likely due to improved imaging techniques. Panda et al. [32] found evidence of SAH in 10 (4.3%) of 233 patients with CVT on CT in a retrospective study and observed SAH early in the course of disease. The frequency of SAH caused by CVT was 3% (4 of 145 cases) in a retrospective review [38]. Although this presentation is rare, it appears to be underestimated that the total number of reviewed cases is not large. The current capacity to diagnose SAH in the setting of CVT still seems to be suboptimal probably due to a low index of clinical suspicion and nonspecific CT findings for CVT. In addition, the amount of blood in the subarachnoid space may be so small in the setting of CVT that this presentation may be underestimated when the initial imaging test is CT. SAH associated with CVT is of venous origin, and the clinical management in these cases is different from arterial SAH. The delay in the diagnosis could be potentially life-threatening with a high mortality rate in untreated patients. Therefore, diagnostic imaging has an essential role in the early diagnosis. Although NECT is the first-line imaging study in suspected SAH, MR imaging is the most valuable method for diagnosis of SAH and the identification of the underlying etiology, particularly in patients with a negative CT scan. CT is effective for acute SAH with a sensitivity of about 90%, but the sensitivity decreases with time and approaches 0% at 3 weeks. For the detection of SAH, the sensitivity of FLAIR was 100% versus 67% for CT and 36% for T2*-weighted images. FLAIR is known to be extremely sensitive in the diagnosis of acute or subacute low-grade SAH [44]. However, subarachnoid hyperintensity on FLAIR images is not specific for SAH and may also be seen in other pathologies such as meningitis and meningeal carcinomatosis [39]. SWI is a relatively new MR imaging sequence that is based on susceptibility differences between tissues and is extremely sensitive to paramagnetic substances and venous blood vessels. This sequence has been demonstrated to be very sensitive to small amounts of SAH and superior to CT in detecting intraventricular hemorrhage. Therefore, SWI may provide complementary information to CT in imaging SAH [45]. Computed tomography may offer some clues to the diagnosis of CVT when presenting with SAH. Benabu et al. [7] reviewed a case with the literature of 16 documented reports including 26 cases of SAH

associated with CVT. SAH was detected in 86% of 26 cases, and CVT was confirmed in only 36% of total cases on NECT. In our review, we found that CT detected SAH in 61 of a total of 67 cases whereas initial CT was normal in 4 of them. CT failed to detect SAH in 6 (9%) of them, and diagnosis was made by the subsequent MRI. On the other hand, CT was not very effective for the diagnosis of CVT and was able to make the diagnosis in only 32% of the patients. In the literature, the distribution of SAH associated with CVT was described to be usually different from that of SAH of arterial origin, which has a characteristic pattern. SSS thrombosis was usually associated with SAH involving frontoparietal convexity, sylvian fissure, and IF. The parieto-occipital and posterior temporal convexity SAH was related to T thrombosis and paramedian supra- and/or I SAH to St thrombosis, and both of these sinuses were usually involved with SSS thrombosis. SAH was usually seen at cerebral convexities but never involved the skull base and basal cisterns as in our case. Adaletli et al. [19] reported CVT presenting with excessive SAH including basal cisterns possibly because of the delay in diagnosis. Lee et al. [30] reviewed two cases of PM SAH in association with CVT that could be explained by the underlying abnormalities in the venous circulation. The dural sinuses most commonly involved were SSS and transverse/ Ss alone or in combination that were consistent with the distribution of CVT in the literature [2]. Although c thrombosis was mentioned to be associated with dural sinus thrombosis, it seems to be underestimated in the presence of dural sinus thrombosis that only six cases were reported. 6. Conclusions According to the data from the review of the literature, CVT presenting with SAH is rare. The diagnosis of CVT can be difficult and is further complicated when patients initially present with SAH. A review of prior reports showed that in the setting of CVT, SAH was usually localized to the region of the thrombosed venous structures, hence suggesting thrombosis and subsequent rupture of the adjacent cs or superficial venous sinuses. Because the management of this presentation is quite different from that of arterial SAH, CVT should be considered, particularly when SAH is limited to cerebral convexities sparing the basal cisterns and imaging of the cerebral venous system,

Please cite this article as: Sahin N, et al, Cerebral venous thrombosis as a rare cause of subarachnoid hemorrhage: case report and literature review, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.03.005

N. Sahin et al. / Clinical Imaging xxx (2014) xxx–xxx

including MRI and MRV, should be performed. Imaging will allow early diagnosis and initiation of appropriate therapy to reduce the risk of acute complications and long-term sequelae of this presentation.

References [1] Renowden S. Cerebral venous sinus thrombosis. Eur Radiol 2004;14:215–26. [2] Leach JL, Fortuna RB, Jones BV, Gaskill-Shipley MF. Imaging of cerebral venous thrombosis: current techniques, spectrum of findings, and diagnostic pitfalls. Radiographics 2006;26(Suppl. 1):S19–41. [3] Bousser MG, Chiras J, Bories J, Castaigne P. Cerebral venous thrombosis – a review of 38 cases. Stroke 1985;16:199–213. [4] Ohta H, Kinoshita Y, Hashimoto M, Yamada H, Urasaki E, Yokota A. Superior sagittal sinus thrombosis presenting with subarachnoid hemorrhage in a patient with aplastic anemia. No To Shinkei 1998;50:739–43. [5] Oshiro S, Motomura K, Fukushima T. Systemic lupus erythematosus manifesting as subarachnoid hemorrhage induced by cortical venous thrombosis and followed by medial medullary infarction. No To Shinkei 2003;55:791–5. [6] Kasuga K, Naruse S, Umeda M, Tanaka M, Fujita N. Case of cerebral venous thrombosis due to Graves' disease with increased factor VIII activity. Rinsho Shinkeigaku 2006;46:270–3. [7] Benabu Y, Mark L, Daniel S, Glikstein R. Cerebral venous thrombosis presenting with subarachnoid hemorrhage. Case report and review. Am J Emerg Med 2009;27:96–106. [8] de Brugin SF, Stam J, Kappelle LJ. Thunderclap headache as first symptom of cerebral venous thrombosis. CVST Study Group. Lancet 1996;348:1623–5. [9] Ciccone A, Citterio A, Santilli I, Sterzi R. Subarachnoid haemorrhage treated with anticoagulants. Lancet 2000;356:1818. [10] Ra CS, Lui CC, Liang CL, Chen HJ, Kuo YL, Chen WF. Superior sagittal sinus thrombosis induced by thyrotoxicosis. Case report. J Neurosurg 2001;94:130–2. [11] Sztajzel R, Coeytaux A, Dehdashti AR, Delavelle J, Sinnreich M. Subarachnoid hemorrhage: a rare presentation of cerebral venous thrombosis. Headache 2001;41:889–92. [12] Selim M, Fink J, Linfante I, Kumar S, Schlaug G, Caplan LR. Diagnosis of cerebral venous thrombosis with echo-planar T2*-weighted magnetic resonance imaging. Arch Neurol 2002;59:1021–6. [13] Widjaja E, Romanowski CA, Sinanan AR, Hodgson TJ, Griffiths PD. Thunderclap headache: presentation of intracranial sinus thrombosis? Clin Radiol 2003;58:648–52. [14] Chang R, Friedman DP. Isolated cortical venous thrombosis presenting as subarachnoid hemorrhage: a report of three cases. AJNR Am J Neuroradiol 2004;25:1676–9. [15] Zare M, Mirabdolbaghi P. Cerebral venous thrombosis presented as subarachnoid hemorrhage and treated with anticoagulants. J Res Med Sci 2005;10:251–4. [16] Tidahy E, Derex L, Belo M, Dardel P, Robert R, Honnorat J, Nighoghossian N, Trouillas P. Cerebral venous thrombosis presenting as subarachnoid hemorrhage. Rev Neurol (Paris) 2004;160:459–61. [17] Oppenheim C, Domigo V, Gauvrit JY, Lamy C, Mackowiak-Cordoliani MA, Pruvo JP, Méder JF. Subarachnoid hemorrhage as the initial presentation of dural sinus thrombosis. AJNR Am J Neuroradiol 2005;26:614–7. [18] Spitzer C, Mull M, Rohde V, Kosinski CM. Non-traumatic cortical subarachnoid haemorrhage: diagnostic work-up and aetiological background. Neuroradiology 2005;47:525–31. [19] Adaletli I, Sirikci A, Kara B, Kurugoglu S, Ozer H, Bayram M. Cerebral venous sinus thrombosis presenting with excessive subarachnoid hemorrhage in a 14-year-old boy. Emerg Radiol 2005;12:57–9. [20] Rice H, Tang YM. Acute subarachnoid haemorrhage: a rare presentation of cerebral dural sinus thrombosis. Australas Radiol 2006;50:241–5. [21] Senel A, Dagcinar A, Kaya AH, Kuruoglu E. Localized spontaneous subarachnoid hemorrhage as first manifestation of cerebral venous thrombosis. Neurosurg Q 2006;16:117–20. [22] Lin JH, Kwan SY, Wu D. Cerebral venous thrombosis initially presenting with acute subarachnoid hemorrhage. J Chin Med Assoc 2006;69:282–5.

7

[23] Shukla R, Vinod P, Prakash S, Phadke RV, Gupta RK. Subarachnoid haemorrhage as a presentation of cerebral venous sinus thrombosis. J Assoc Physicians India 2006;54:42–4. [24] Mathew T, Sarma GR, Kamath V, Roy AK. Subdural hematoma, subarachnoid hemorrhage and intracerebral parenchymal hemorrhage secondary to cerebral sinovenous thrombosis: a rare combination. Neurol India 2007;55:438–9. [25] Ko YP, Hsu CY, Yang CL, Cheng SJ, Tsai HJ, Chen CP, Hsu YW. Postcesarean subarachnoid hemorrhage as the initial presentation of cerebral venous thrombosis. Taiwan J Obstet Gynecol 2007;46:317–9. [26] Wang YF, Fuh JL, Lirng JF, Chang FC, Wang SJ. Spontaneous intracranial hypotension with isolated cortical vein thrombosis and subarachnoid haemorrhage. Cephalalgia 2007;27:1413–7. [27] Tang PH, Chai J, Chan YH, Chng SM, Lim CC. Superior sagittal sinus thrombosis: subtle signs on neuroimaging. Ann Acad Med Singapore 2008;37:397–401. [28] Jaiser SR, Raman A, Maddison P. Cerebral venous sinus thrombosis as a rare cause of thunderclap headache and nonaneurysmal subarachnoid haemorrhage. J Neurol 2008;255:448–9. [29] Lai NK, Hui JW, Wong GK, Yu SC, Sun DT, Poon WS. Cerebral venous thrombosis presenting as subarachnoid haemorrhage. Hong Kong Med J 2008;14:499–500. [30] Lee J, Koh EM, Chung CS, Hong SC, Kim YB, Chung PW, Suh BC, Moon HS. Underlying venous pathology causing perimesencephalic subarachnoid hemorrhage. Can J Neurol Sci 2009;36:638–42. [31] Bittencourt LK, Palma-Filho F, Domingues RC, Gasparetto EL. Subarachnoid hemorrhage in isolated cortical vein thrombosis: are presentation of an unusual condition. Arq Neuropsiquiatr 2009;67:1106–8. [32] Panda S, Prashantha DK, Shankar SR, Nagaraja D. Localized convexity subarachnoid haemorrhage – a sign of early cerebral venous sinus thrombosis. Eur J Neurol 2010;17:1249–58. [33] Hegazi MO, Ahmed S, Sakr MG, Hassanien OA. Anticoagulation for cerebral venous thrombosis with subarachnoid hemorrhage: a case report. Med Princ Pract 2010;19:73–5. [34] Sharma S, Sharma N, Yeolekar ME. Acute subarachnoid hemorrhage as initial presentation of dural sinus thrombosis. J Neurosci Rural Pract 2010;1:23–5. [35] Field DK, Kleinig TJ. Aura attacks from acute convexity subarachnoid haemorrhage not due to cerebral amyloid angiopathy. Cephalalgia 2011;31:368–71. [36] Kato Y, Takeda H, Furuya D, Nagoya H, Deguchi I, Fukuoka T, Tanahashi N. Subarachnoid hemorrhage as the initial presentation of cerebral venous thrombosis. Intern Med 2010;49:467–70. [37] Oz O, Akgun H, Yücel M, Battal B, Ipekdal HI, Ulaş UH, Demírkaya S, Odabaşi Z. Cerebral venous thrombosis presenting with subarachnoid hemorrhage after spinal anesthesia. Acta Neurol Belg 2011;111:237–40. [38] Oda S, Shimoda M, Hoshikawa K, Osada T, Yoshiyama M, Matsumae M. Cortical subarachnoid hemorrhage caused by cerebral venous thrombosis. Neurol Med Chir (Tokyo) 2011;51:30–6. [39] Renou P, Tourdias T, Fleury O, Debruxelles S, Rouanet F, Sibon I. Atraumatic nonaneurysmal sulcal subarachnoid hemorrhages: a diagnostic workup based on a case series. Cerebrovasc Dis 2012;34:147–52. [40] Sharma B, Satija V, Dubey P, Panagariya A. Subarachnoid hemorrhage with transient ischemic attack: another masquerader in cerebral venous thrombosis. Indian J Med Sci 2010;64:85–9. [41] Sayadnasiri M, Taheraghdam AA, Talebi M. Cerebral venous thrombosis presenting as subarachnoid hemorrhage: report of two cases. Clin Neurol Neurosurg 2012;114:1099–101. [42] Saposnik G, Barinagarrementeria F, Brown RD, Bushnell CD, Cucchiara B, Cushman M, deVeber G, Ferro JM, Tsai FY. American Heart Association Stroke Council and the Council on Epidemiology and Prevention. Diagnosis and management of cerebral venous thrombosis: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2011;42:1158–92. [43] Ozsvath RR, Casey SO, Lustrin ES, Alberico RA, Hassankhani A, Patel M. Cerebral venography: comparison of CT and MR projection venography. AJR Am J Roentgenol 1997;169:1699–707. [44] da Rocha AJ, da Silva CJ, Gama HP, Baccin CE, Braga FT, Cesare Fde A, Veiga JC. Comparison of magnetic resonance imaging sequences with computed tomography to detect low-grade subarachnoid hemorrhage: role of fluid-attenuated inversion recovery sequence. J Comput Assist Tomogr 2006;30:295–303. [45] Wu Z, Li S, Lei J, An D, Haacke EM. Evaluation of traumatic subarachnoid hemorrhage using susceptibility-weighted imaging. AJNR Am J Neuroradiol 2010;31:1302–10.

Please cite this article as: Sahin N, et al, Cerebral venous thrombosis as a rare cause of subarachnoid hemorrhage: case report and literature review, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.03.005