Cortical vein thrombosis in adult patients of cerebral venous sinus thrombosis correlates with poor outcome and brain lesions: a retrospective study

Thrombosis of the cerebral venous system is an important cerebrovascular disease [1] that can lead to severe brain injuries (such as venous infarction and parenchymal hemorrhage) and disability. Cerebral venous thrombosis has been reported to usually involve the venous sinuses [2, 3]. Only a few cases with isolated cortical venous thrombosis (ICVT) in which the major sinuses are not involved have been reported [4, 5]. The International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT), a large prospective multi-central observational study, showed that 17.1% of patients with cerebral venous sinus thrombosis (CVST) had concomitant cortical venous thrombosis (CVT) accompanied by occlusion of the major sinuses [3]. In our study, we defined CVT as cortical venous thrombosis with concomitant occlusion of the major sinuses and ICVT as isolated cortical venous thrombosis without occlusion of the major sinuses. The diagnosis of CVT is challenging due to the relatively smaller sizes of the cortical veins compared with that of the cerebral venous sinuses. In early studies, CVT was usually identified using percutaneous puncture cerebral angiography [1], which was not a conventional vascular assessment method. Thus, we speculated that the incidence of concomitant CVT in CVST patients might have been underestimated in the past. With the development of non-invasive neuroimaging techniques (e.g., magnetic resonance imaging [MRI] and computed tomography [CT]) and improved recognition of CVT, a greater incidence of CVT has been diagnosed with contemporary neuroimaging techniques compared to the past decade. In a recent study that included pediatric CVST patients [6], concomitant CVT was found to be not rare (24%) and was associated with seizures and brain lesions. However, the significance of concomitant CVT in adult CVST patients is not clear. Thus, we used contemporary neuroimaging techniques to identify CVT, to investigate the associations of concomitant CVT with the clinico-radiological characteristics of adult CVST patients, and to further explore the pathophysiological mechanisms of cerebral venous thrombosis in adult patients.

According to the aforementioned inclusion and exclusion criteria, 44 CVST patients (age: 33.8 ± 14.0 years, range from 19 to 62 years; 28 males) were enrolled into this study. The median duration from the onset of symptoms to hospital arrival was 9 days (range from 1 to 30 days). The CVT group included 27 patients (61.4%), and the non-CVT group included 17 patients (38.6%) (Table 1). Seven patients were excluded due to insufficient radiological assessments (neither CTV nor CE-3D-MPRAGE was performed).

In our study, a higher incidence (61.4%) of concomitant CVT was found among adult CVST patients than pediatric CVST patients (24%) and the rate reported in a previous study (17%) [3, 6], which might be due to the integration of multiple imaging techniques for the detection of CVTs in our study (see Additional file 1: Tables S1 and S2).

In a recent study of pediatric CVST patients [6], the frequencies of consciousness disorder (58% vs. 21%) and seizure (58% vs. 21%) rather than focal neurological deficits (41% vs. 31%) were significantly higher among patients with concomitant CVT than patients without concomitant CVT. Similarly, seizure and focal neurological deficits were common (67% and 62%, respectively) among isolated CVT patients according to a systemic review [4]. However, in contrast to this result for pediatric CVST patients with concomitant CVT, consciousness disorders are rare among isolated CVT patients (2%) [4]. Seizure and focal neurological deficits were also common among the adult CVST patients with concomitant CVT in our study (44.4% and 63%, respectively). These results were similar to the results from the pediatric CVST patients and isolated CVT patients, but the rate of consciousness disorder (33.3%) was lower than the rate among the pediatric CVST patients. Moreover, in our study, focal neurological deficit, seizure, and consciousness disorder were more common among the adult CVST patients with concomitant CVT than among those without concomitant CVT. In summary, concomitant CVT was associated with severe clinical manifestations in both pediatric and adult CVST patients, which suggested a similar pathophysiological mechanism. In our adult patients, concomitant CVT was also related to a poor short-term outcome.

Parenchymal hemorrhage and venous infarction (50% and 75%, respectively) are common findings among pediatric CVST patients with concomitant CVT [6]. Similarly, our study also demonstrated a high frequency of brain lesions (venous infarction: 63% and parenchymal hemorrhage: 40.7%) among adult CVST patients with concomitant CVT. In addition to the analogous clinical manifestations of pediatric and adult CVST patients with concomitant CVT, the similarities of both conditions in terms of radiological profiles suggest the same pathophysiological mechanisms. We speculated that the mechanisms were as follows. (1) The blood flows of the cortex and its adjacent white matter are drained to cortical veins via the superficial medullary veins, which have poor collaterals [12]. Blockage of the outlets of the superficial medullary veins by thrombi in the cortical veins results in severe congestion of the brain tissues within the drainage areas of the cortical veins, leading to venous infarction or parenchymal hemorrhage. (2) In the condition of CVST without concomitant CVT, thrombosis in the venous sinus that blocks the outlets of the cortical veins may not cause severe congestion due to the good collaterals between the cortical veins [13]. Therefore, in practice, we should focus on the early diagnosis and treatment of CVST to prevent thrombi from extending from the cerebral venous sinuses to the cortical veins and to reduce the risk of venous infarction or hemorrhage and improve the prognosis.

In our study, we have compared the frequency of thrombosis in each venous sinus between CVT patients and non-CVT patients. We discovered that the superior sagittal sinus was more frequently involved in the CVT patients than in the non-CVT patients. However, the frequency of thrombosis in each venous sinus does not indicate the severity of venous sinus thrombosis; it denotes the spatial distribution of venous sinus thrombosis. When evaluating the severity of venous sinus thrombosis, we compared the total number of cerebral venous sinuses involved in thrombosis, and we did not find a significant difference between the CVT patients and non-CVT patients. Furthermore, the duration from onset of symptoms to hospital arrival is not significantly different between CVT group and non-CVT group. Thus, it seems that the severity of the involvement of sinuses and the duration from onset of symptoms to hospital arrival do not contribute to CVT. Further studies are needed to explore the factors predicting CVT in CVST patients.

Anticoagulation is a cornerstone for the treatment of thrombosis of the cerebral venous system [1]. Other treatments include endovascular recanalization and surgical decompression [1, 14, 15]. Our results suggest that different pathophysiological mechanisms are operating in CVST patients with and without concomitant CVT. However, with regard to therapeutic strategies, the significance of these differences remains unclear and should be investigated in future studies. Additionally, the CVST patients with concomitant CVT in our study exhibited poor short-term outcomes at discharge, and the majority were administered anticoagulation therapy. Anticoagulation therapy alone may not be sufficient for these patients, and other treatments should be developed. A case series study that included 6 CVST patients with concomitant CVT and severe clinical manifestations demonstrated that combined intra-arterial and intravenous thrombolysis could markedly improve the outcomes (i.e., the mRS scores at 3 months were zero) [16]. Thus, this therapeutic strategy seems promising, but future prospective cohort studies with large sample sizes are needed to verify its safety and efficiency.

This study had some limitations. First, the study was a retrospective research study with a relatively small sample size. Our findings need to be verified by future prospective studies with large sample sizes. Second, the methods used to detect CVT were not entirely uniform. CE-3D-MPRAGE was performed in the majority of our patients and exhibited high efficacy for the detection of CVT in this study. T2^* gradient echo (T2* GE) imaging and susceptibility-weighted imaging (SWI) utilize the magnetic susceptibility effect (MSE) of hemoglobin products to detect thrombi [17]. A previous study [18] demonstrated that T2* GE and SWI were more capable of detecting ICVT in the early and late phases than standard MRI protocols. In contrast, MSEs of the skull and intracranial hemorrhages also present as hypotensive, which may lead to confusion in the identification of CVT if the thrombus is adjacent to the skull or to a hemorrhagic lesion. Additionally, among the CVT patients with concomitant occlusions of the major sinuses, we found a high rate (40.7%) of intracranial hemorrhaging. Unfortunately, not all of the included patients underwent T2* or SWI in their MRI protocols. Thirdly, the ICVT patients were not included in this study. Thus, we cannot compare with CVST patients. Moreover, future studies should be conducted to compare the diagnostic efficacies of CE-3D-MPRAGE and T2^*GE or SWI for CVTs; the latter method is recommended for CVST patients according to a recent guideline [7]. Third, because our hospital is a regional tertiary referral center, we risk selection bias in which the severity and the incidence of concomitant cortical vein thrombus may be overestimated.

In conclusion, a concomitant CVT was a common finding in adult CVST patients and was associated with severe clinical manifestations, a poor short-term outcome, and brain lesions. The differences in the clinico-radiological features of adult CVST patients with and without a concomitant CVT are suggestive of different pathophysiological mechanisms. Future studies are needed to explore the significance of our findings for the treatment and prognosis of adult patients with thrombosis of the cerebral venous system.