Venous Transcranial Doppler Ultrasound Monitoring in Acute Dural Sinus Thrombosis
Report of Two Cases
Background and Purpose We sought to establish the efficacy of the Doppler technique in the evaluation of the intracranial venous system and to assess its usefulness in the monitoring of venous collateral pathways in superior sagittal sinus thrombosis.
Methods Venous Doppler ultrasound was performed with a range-gated 2-MHz transducer in 10 healthy volunteers and in two patients with superior sagittal sinus thrombosis confirmed by cerebral angiography.
Results In normal control subjects, a venous signal was found at a depth ranging from 40 to 72 mm, which was considered to correspond to the deep middle cerebral vein and the basal vein of Rosenthal. Mean blood flow velocities ranged from 9 to 20 cm/s. In both patients with superior sagittal sinus thrombosis, Doppler studies detected elevated mean blood flow velocities (146 and 33 cm/s), which normalized after 16 weeks and 1 week, respectively.
Conclusions Venous transcranial Doppler ultrasonography provides a reliable, noninvasive, and rapid technique for intracranial venous examination. It was performed without difficulty in young healthy volunteers, and it can be applied as a monitoring tool in the evaluation of collateral venous flow in superior sagittal sinus thrombosis.
In dural sinus thrombosis (DST), it is a well-known fact that topography of venous occlusion and clinical symptoms correlate poorly. The severity of the clinical course seems to depend on the acuteness of thrombotic formation and the capacity for compensation by major venous vessels. Little is known about the time course of recanalization and the quantitative significance of collateral venous pathways. Cerebral angiography, the most reliable technique for diagnostic purposes, is not practical for monitoring the changes in venous cerebral blood flow because of its invasiveness. MR angiography (MRA) not only seems to offer important advantages as a noninvasive tool in diagnostic procedures but also seems useful as a follow-up instrument, allowing the documentation of thrombus regression, venous collateralization, and recanalization.1 Limits exist, however, as a result of the costs and logistic problems found in intensive care patients. Since its introduction in 1982, transcranial Doppler ultrasonography (TCD) has been shown to be a reliable, noninvasive, and inexpensive method in the examination of the large basal cerebral arteries.2 3 Its usefulness in the analysis of venous blood flow has not yet been established systematically. Normal venous blood flow velocities in the straight sinus have been reported in a study dealing with aspects of cerebral autoregulation.4 Regarding DST, one anecdotal report showing venous TCD signals has recently been published.5
The detection of a prominent venous signal, considered as originating from the basal vein of Rosenthal in two of our patients suffering from acute thrombosis of the superior sagittal sinus (SSST), led to the hypothesis that TCD may facilitate quantification of venous collateral pathways and the recognition of the time course of venous recanalization. This study presents the results of the venous Doppler monitoring of two patients with acute SSST and the examination of the venous system in 10 healthy subjects; we aimed to test the reliability of venous TCD recording and to establish normal values for venous mean blood flow velocity (mBFV).
Subjects and Methods
Venous Doppler ultrasonography was performed in 10 healthy volunteers (4 men, 6 women; age range, 20 to 36 years; mean age, 24.6 years) and in 2 female patients aged 34 and 39 years.
Doppler ultrasound was performed with patients in the supine position with a range-gated 2-MHz transducer (Multidop-X, Firma Elektronische Systeme GmbH). After detection of the P2 segment of the posterior cerebral artery through the posterior temporal window at a depth of approximately 60 mm, a search was conducted for a venous signal. After detection of the venous signal with a direction flow away from the probe, which was considered to correspond to the basal vein of Rosenthal, the signal boundaries were assessed by increasing and then decreasing the sample volume depth by 2-mm increments. A venous signal, paralleling the middle cerebral artery, was considered to belong to the deep middle cerebral vein. Valsalva tests confirmed the venous origin. We recorded mBFV in centimeters per second. The pulsatility index (PI) was calculated as systolic velocity minus diastolic velocity divided by mBFV.
In all 10 healthy volunteers, a venous signal characterized by a low-amplitude pulsatile flow away from the probe (Fig 1⇓) was found at a depth ranging from 40 to 72 mm (mean±SD, 50.5±4.9 to 64.2±4.7 mm) in either one side or bilaterally. The mBFV ranged from 9 to 20 cm/s (mean±SD, 12.1±3.5 cm/s).
A 34-year-old woman was admitted with a 3-day history of severe headache and vomiting. She revealed fluctuating consciousness and moderate tetraparesis. A CT scan on admission showed bilateral parietal hemorrhages, diffuse swelling, and hyperdense venous sinuses. Cerebral angiography confirmed SSST. A marked collateral flow was seen in the sphenoparietal sinus and in the basal vein of Rosenthal (Fig 2⇓). Contraceptive medication was the only risk factor. The symptoms completely resolved during high-dose heparin treatment.
TCD examination was started on admission and repeated 23 times in a follow-up of 230 days (Fig 3⇓). Initially, a venous Doppler signal confirmed by Valsalva maneuver away from the probe was detected at a depth of 48 to 66 mm at the proximity of the left P2 segment of the posterior cerebral artery, characterized by a high mBFV (146 cm/s) and a low PI of 0.36 (Fig 4A⇓ and 4B⇓). In the following weeks, the mBFV decreased to 70 cm/s and reached a steady state of approximately 100 cm/s. After 4 months, mBFV dropped to normal values of 11 cm/s (Fig 4C⇓). Oral anticoagulation was discontinued at this time without any negative influence on venous mBFV or the clinical state during a follow-up of 4 months.
This 39-year-old woman was admitted with a 3-day history of generalized headache, left-sided paresis, and secondary generalized epilepsy. Contraceptive medication was the only known risk factor. A CT scan demonstrated venous infarction in the right frontotemporal area with mild diffuse edema. Angiography revealed SSST, affliction of the right transverse sinus, and a venous collateralization via the left vein of Labbé, right cavernous sinus, and both veins of Rosenthal (Fig 5⇓). High-dose heparin was administered, and barbiturate coma was initiated to control epilepsy during epidural intracranial pressure monitoring. She worsened acutely on day 17 as a result of hemorrhagic enlargement of the known infarction area. Despite surgical treatment, she died 1 week later. No autopsy was performed.
Venous Doppler examination was performed 10 times during a follow-up of 17 days (Fig 6⇓). A venous signal was detected at a depth of 48 to 64 mm bilaterally with an increased venous mBFV of 33 cm/s at the first examination (Fig 7A⇓ and 7B⇓). During barbiturate coma, the values dropped to 14 cm/s and reached the previous values a week later after the cessation of barbiturate administration. Venous mBFVs of 20 cm/s and less were registered in the second week. On day 17, a marked elevation to 50 cm/s was observed 12 hours before the patient died.
The superior sagittal sinus is the most commonly affected site in DST.6 In cases of complete SSST including the torcular Herophili, the collateral flow proceeds through the sphenoparietal sinus, cavernous sinus, vein of Labbé, and basal vein of Rosenthal. It is well established that an acute evolution of thrombosis, coma, and an involvement of the deep venous system is associated with increased mortality and morbidity.7 The capacity to open venous collateral pathways and the velocity of recanalization are most probably important prognostic factors. However, the quantification of the venous collateral capacity and the time course of recanalization are difficult to assess because repeated angiographic studies have not been performed regularly in these patients. A few MRA studies have shown that this method is capable of evaluating venous collateral pathways, thrombus regression,1 and recanalization between 10 and 30 days.8 9 Information on the success of recanalization may be important not only for prognostic statements but also for delineating therapeutic measurements. High-dose heparin followed by oral anticoagulation is currently standard treatment. The duration of oral anticoagulation is still open to discussion since no clinical or technical parameter is known that can determine the point at which oral anticoagulation can be safely discontinued.10
Before our study, TCD had not been performed to establish its ability to monitor venous parameters in DST or to investigate adult venous cerebral anatomy and blood flow velocities.
Blood Flow Aspects
In newborns, normal mBFVs were evaluated with the use of color Doppler ultrasound by Taylor11 in 1992. In his study, which was facilitated by the transparency of newborn skulls, mBFV rose, corresponding to vessel size, ranging from 4.3 cm/s in the vein of Galen to 9.2 cm/s in the superior sagittal sinus. In adults, normal venous Doppler findings were reported in the straight sinus by Aaslid et al4 in 1991. Using the occipital window, he found the straight sinus in 9 of 12 healthy subjects. No information on mBFV was given. However, in one figure a Doppler spectrum less than 20 cm/s was documented. Our group pioneered TCD examinations in healthy volunteers to establish normal venous mBFV. The values were quite similar in all 10 examined subjects, revealing low pulsatility signals ranging from 9 to 20 cm/s (mean mBFV, 12.1±3.5 cm/s). Similar mBFVs have been found in the major sinus by MRI techniques. With the use of a phase-sensitive, limited-flip-angle, gradient-recalled method, the dural sinus mBFV found in normal subjects ranged from 9.9 to 14.4 cm/s in the transverse and superior sagittal sinuses.12
TCD recordings in DST have not been previously published with the exception of Wardlaw et al.5 They demonstrated a prominent venous signal adjacent to the middle cerebral artery away from the probe, which disappeared as the patient improved clinically. In this case, angiography was not performed to confirm the diagnosis.
In our two patients with acute SSST, we were able to detect a marked elevation in venous mBFV of 146 and 33 cm/s, which normalized within 16 weeks and 1 week, respectively. The interpretation of these findings is not easy and has to remain in part hypothetical because of a lack of basic pathophysiological knowledge. The elevated venous mBFV in the basal veins corresponds to the angiographically documented enlarged venous collateral pathways, which we partially consider to belong to the basal vein of Rosenthal. The elevated velocity may designate the grade of venous overloading and probably reflects the risk of venous infarction and cerebral bleeding. It may also present an indicator of the severity and acuteness of the thrombotic formation in the superior sagittal sinus. On the other hand, it may indicate a good collateral capacity and therefore reflect a more favorable prognosis. Whether the normalized mBFV only corresponds to sufficient collateralization with permanent thrombosis or ensures complete recanalization must be analyzed by future MRA studies.
Doppler ultrasound investigation of the major basal arteries has become a well-established method. In contrast, no attention has been paid to the examination of the basal veins, which form a venous circuit at the base of the brain, corresponding to that of the circle of Willis (Fig 8⇓). The basal vein of Rosenthal, which is usually formed by the junction of the middle deep cerebral vein and the anterior cerebral vein, reveals a regular anatomy in 80% of cases.13 14 It could be recorded in all of our subjects at a depth of approximately 60 mm. At this point, the vein lies slightly superior to the posterior cerebral artery and passes parallel to it around the midbrain. In our opinion, venous signals, followed to a depth of 46 mm and sometimes recorded in conjunction with the middle cerebral artery, correspond to the deep middle cerebral vein. No prominent differences in mBFV between these two venous vessels were observed in our series.
In conclusion, we have shown the capacity of TCD to detect and monitor basal venous flow velocities in two patients with acute and severe SSST and in 10 healthy subjects to establish normal venous Doppler values. The clinical relevance of these findings remains open to discussion. Further investigations will be required in normal subjects to investigate the use of venous TCD in more detail. More patients with acute SSST and cerebral thrombosis at other sites, as well as those in chronic evolution, must be followed up to establish the reliability of this method.
We wish to express our appreciation to Dr Thomas Davenport for linguistic assistance and to Gerd Dombrowsky for technical support.
- Received February 28, 1995.
- Revision received April 13, 1995.
- Accepted April 14, 1995.
- Copyright © 1995 by American Heart Association
Vogl TJ, Bergmann C, Villringer A, Einhäupl K, Lissner J, Felix R. Dural sinus thrombosis: value of venous MR angiography for diagnosis and follow-up. AJNR Am J Neuroradiol. 1994;162:1191-1198.
Aaslid R, Newell DW, Stooss R, Sorteberg W, Lindegaard KF. Assessment of cerebral autoregulation dynamics from simultaneous arterial and venous transcranial Doppler recordings in humans. Stroke. 1991;22:1148-1154.
Bousser MG, Chiras J, Bories J, Castaigne P. Cerebral venous thrombosis: a review of 38 cases. Stroke. 1985;16:199-213.
Ameri A, Bousser MG. Cerebral venous thrombosis. Neurol Clin. 1992;19:87-111.
Tsuruda JS, Shimakawa A, Pelc NJ, Saloner D. Dural sinus occlusion: evaluation with phase-sensitive gradient-echo MR imaging. AJNR Am J Neuroradiol. 1991;12:481-488.
Huber P. Zerebrale Angiographie für Klinik und Praxis. 3rd ed. Stuttgart, Germany: Thieme Verlag; 1979:554.