Background and Purpose The aim of this study was to investigate flow velocity and flow direction in the posterior communicating artery (PcomA) by means of transcranial color-coded duplex sonography (TCCD) and to compare the results with angiographic findings.
Methods Thirty patients with unilateral occlusion of the internal carotid artery (ICA) due to atherosclerosis (n=15) or balloon occlusion (n=15) and 50 normal subjects were included. The circle of Willis was insonated through the temporal bone window. In 24 patients with unilateral ICA occlusion, angiograms were available and were compared with the results of TCCD.
Results The PcomA could be detected unilaterally in 70% of normal subjects and bilaterally in 30%. A retrograde flow direction in the PcomA from the posterior cerebral artery to the ICA was found in 75% of the normal control subjects. The mean peak flow velocity in normal PcomAs was 36±15 cm/s (±SD). No significant differences in flow velocity were found between unilaterally and bilaterally detectable PcomAs or between retrograde and orthograde PcomAs. In patients with unilateral ICA occlusion we observed ipsilaterally a retrograde flow direction, with an elevation of flow velocity (64±10 cm/s) compared with the contralateral side (27±14 cm/s; P<.001).
Conclusions TCCD appears to be a valuable method to determine flow velocity and flow direction not only in the large intracranial vessels but also in the smaller communicating arteries. In the future this method could be useful for the planning of ICA balloon occlusions and in deciding whether to perform extracranial/intracranial bypass surgery. It could furthermore show intracranial collaterals in patients with cerebrovascular disease and help to estimate the risk of watershed infarctions in patients with asymptomatic high-grade ICA stenosis and in patients undergoing carotid endarterectomy.
The PcomA has its embryonic origin in the ICA and connects with the PCA. The vessel sends 5 to 12 branches to the thalamus, the reticular nucleus, the mammillothalamic tract, the diencephalon, and the caudate nucleus. The PcomA varies extensively in length, diameter, and course. In 33% of cases the vessel exists only unilaterally, and in more than 50% abnormalities such as aplasia, hypoplasia, or duplication are demonstrable.1 2 An autopsy study has yielded a mean outer diameter of 1.2 mm and a mean length of 12.6 mm.3
In addition to collateral flow through the anterior communicating artery, the ophthalmic artery, and leptomeningeal vessels, the PcomA is an important collateral vessel in patients with unilateral ICA occlusion.4 5 6 Data concerning the variation of flow direction and flow velocity in the PcomA and the significance of this vessel for the blood supply under physiological and pathophysiological circumstances are rare. This is largely because the detection of the PcomA with the use of TCD is very difficult and often necessitates the use of compression maneuvers.5 6 7 8 9 The aim of the present study was to estimate flow velocity and direction in the PcomA in normal subjects and in patients with unilateral ICA occlusion with the use of TCCD.
Subjects and Methods
The first group involved 50 normal subjects (20 men, 30 women) with a mean age of 38±11 years (±SD). Extracranial color-coded duplex sonography and TCD showed no vascular disease in this subset of patients. The second group consisted of 15 patients (mean age, 49±10 years) who had undergone unilateral balloon occlusion of the ICA for the treatment of intracavernous aneurysms and 15 other patients (mean age, 60±12 years) with atherosclerotic ICA occlusion. The patients with intracavernous carotid aneurysms were selected for balloon occlusion without previous extracranial/intracranial bypass surgery because of sufficient collateral pathways shown in a previous arterial digital subtraction angiography. The involved patients with atherosclerotic ICA occlusion did not necessarily have symptoms of cerebral ischemia.
A TCCD machine with a 2- to 2.5-MHz transducer (Acuson XP 128/10v ) was used to insonate the circle of Willis through the temporal bone window in transverse planes (Fig 1⇓). Patients with inadequate bone windows, in whom the PcomA could not be assessed on either side, were excluded from the study as well as patients with a fetal origin of the PcomA. To visualize the PcomA, we used a small insonation sector combined with a high frame rate (56 Hz), high persistence of the color-coded pixels, and high color Doppler gain settings. The specific TCCD criteria for identification of the PcomA were (1) the color-coded signal of the vessel between the ICA and the precommunicating segment of the PCA and (2) evidence of a pulsatile Doppler signal within the supposed vessel. A misdiagnosis of the anterior choroidal artery for the PcomA was less likely because this vessel, which originates close to the PcomA just below the ICA bifurcation, is very small (mean OD, 0.78 mm).3 The peak flow velocity and flow direction were measured on both sides for the PcomA, the MCA, and the two segments of the PCA with the use of integrated pulsed-wave Doppler (Fig 2⇓) The mean time interval between balloon occlusion of the ICA and the TCCD examination was 22±6 months. In all 15 patients with ICA balloon occlusion and in 9 patients with atherosclerotic ICA occlusion, cerebral angiograms were available and were compared with the TCCD findings. Differences between subgroups were estimated with the use of paired and unpaired t tests. The overall accuracy of TCCD was calculated on the basis of arterial digital subtraction angiography as the “gold standard.”
Hemodynamics in Normal Subjects
The PcomA could be detected bilaterally in 30% and unilaterally in 70% of the normal control subjects (Table 1⇓). In 74% of the unilaterally detectable PcomAs and 77% of the bilaterally detectable PcomAs, the flow direction was retrograde, from the PCA to the ICA. Differences of peak flow velocities between retrograde (36±15 cm/s) and orthograde (34±16 cm/s) were not significant. Flow velocities in unilaterally (37±16 cm/s) and bilaterally (34±14 cm/s) demonstrable PcomAs were likewise not significantly different.
Hemodynamics in ICA Occlusion
In the subgroup of patients with unilateral ICA occlusion, the PcomA was found bilaterally in 83% and unilaterally in 17%. All 30 patients of this group showed a retrograde flow direction in the PcomAs (n=30) on the side of the ICA occlusion and in 76% of 25 detectable vessels on the contralateral side.
In 24 patients of the second group, the results of cerebral angiography were available and confirmed TCCD findings concerning the existence and flow direction of the PcomA, with an overall accuracy of 89.6% (Table 2⇓). Two PcomAs could be demonstrated by angiography but not by TCCD. They were small and short and could therefore not be reliably differentiated from artifacts. In three other patients the PcomA contralateral to the ICA occlusion was detectable by means of TCCD and was not visible on angiography; these three patients were reconsidered with TCCD. To identify the PcomA clearly, a short compression maneuver of the ipsilateral ICA was performed and showed increased retrograde flow in this vessel.
As expected, elevated peak flow velocities (64±10 cm/s) were observed in the ipsilateral PcomA in patients with ICA occlusion (Table 3⇓), while the contralateral PcomA revealed normal velocities (27±14 cm/s; P<.001). The peak flow velocities in the corresponding precommunicating segment of the PCA were also elevated (83±9 cm/s) compared with the contralateral side (59±7 cm/s; P<.001). In the ipsilateral MCA (84±12 cm/s) the flow velocities were only slightly reduced compared with the contralateral side (93±16 cm/s; P<.01). No significant differences were seen between flow velocities in the ipsilateral PcomA in patients with balloon occlusion (65±11 cm/s) and atherosclerotic occlusion (63±9 cm/s).
The aim of the present study was to investigate flow velocity and flow direction in the PcomA by means of TCCD. To our knowledge this is the first study using TCCD to investigate the hemodynamic situation of the PcomA directly. It was not planned as a validation study because no reference method is available for the estimation of PcomA flow velocity.
Detection of the PcomA
While anatomic studies1 3 have reported unilateral PcomAs in one third of normal subjects and hypoplasia in 24%,9 TCCD showed an apparently unilateral PcomA in 70% of the control subjects. The collateral flow in retrograde PcomAs of patients with unilateral ICA occlusion was detectable in all patients. Thus, we must suppose that TCCD is not able to demonstrate hypoplastic PcomAs but is able to show hemodynamically relevant PcomAs.
Flow Direction in the PcomA
The observations in normal subjects in the present study concerning the ratio of flow directions in the PcomA are at variance with the results of phase-contrast MR angiography studies. Ross and coworkers2 found retrograde flow in the PcomA in 8% of 39 normal subjects. A previous angiographic study reported this situation in 27%,10 while we observed this in 75% of cases. Our result may be biased by the low mean age of the normal subjects and the exclusion of patients with inadequate insonation conditions. However, the circle of Willis is a system of communicating tubes, and therefore the large MCA territory may also steal from the posterior circulation through the PcomA in normal subjects. Since the data from angiography and MR angiography studies are not representative, it is difficult to estimate the real percentage of normal control subjects with retrograde flow in the PcomA.
We found a retrograde flow direction and significantly elevated flow velocities in the ipsilateral PcomA in all patients with a unilateral ICA occlusion. These data are important since Schomer and coworkers,11 using MR angiography, have observed that an ipsilateral small or absent PcomA is a significant risk factor for watershed infarctions12 in patients with ICA occlusion.
Clinical Value and Limitations of the TCCD Findings
TCCD appears to be a valuable method to determine flow velocity and flow direction not only in the large intracranial vessels but also in the smaller communicating arteries. In the future this method could be useful for the planning of ICA balloon occlusions and in decision making for the performance of extracranial/intracranial bypass surgery. It could furthermore show intracranial collaterals13 14 in patients with cerebrovascular disease and help to estimate the risk of watershed infarctions in patients with asymptomatic high-grade ICA stenosis and in patients undergoing carotid endarterectomy.15
TCD is only able to indirectly assess collateral flow through the PcomA if elevated flow velocity is found in the precommunicating segment of the PCA and normal or reduced values in the postcommunicating segment.6 The direct insonation of the PcomA with TCD failed because of the absence of spatial information and the small dimensions of the vessel.7 8
TCCD has to compete with other noninvasive methods, such as MRI and MR angiography, but there are considerable advantages of TCCD in terms of cost, the duration of the examination, and the possibility of repeating the investigation of the PcomA as a bedside test. The application is limited by inadequate temporal bone windows, especially in older patients, but the availability of ultrasonic contrast agents,16 which improve the color Doppler signal, and the use of power Doppler options17 will further enhance the clinical value of TCCD.
Selected Abbreviations and Acronyms
|ICA||=||internal carotid artery|
|MCA||=||middle cerebral artery|
|PCA||=||posterior cerebral artery|
|PcomA||=||posterior communicating artery|
|TCCD||=||transcranial color-coded duplex sonography|
|TCD||=||transcranial Doppler sonography|
- Received July 7, 1995.
- Revision received November 13, 1995.
- Accepted November 13, 1995.
- Copyright © 1996 by American Heart Association
Wells CE. The cerebral circulation: the clinical significance of current concepts. Arch Neurol. 1960;3:319-323.
Ross MR, Pelc NJ, Enzmann DR. Qualitative phase contrast MRA in the normal and abnormal circle of Willis. AJNR Am J Neuroradiol. 1993,14:19-25.
Lang J, Schaffrath H, Fischer G. Weitere Befunde zu den Rami diencephali. Neurochirurgie.. 1987;30:103-108.
Harrison MJ, Marshall J. The variable clinical and CT findings after carotid occlusion: the role of collateral blood supply. J Neurol Neurosurg Psychiatry. 1988;51:269-272.
Hedera P, Traubner P, Bujdakova J. Short-term prognosis of stroke due to occlusion of internal carotid artery based on transcranial Doppler ultrasonography. Stroke. 1992;23:1069-1072.
Schneider PA, Rossman ME, Bernstein EF, Torem S, Ringelstein EB, Otis SM. Effect of internal carotid artery occlusion on intracranial hemodynamics: transcranial Doppler evaluation and clinical correlation. Stroke. 1988;19:589-593.
Bass A, Krupski WC, Dilley RB, Bernstein EF, Otis SM. Comparison of transcranial and cervical continuous-wave Doppler in the evaluation of intracranial collateral circulation. Stroke. 1990;21:1584-1588.
Krayenbühl H, Yasargil G, Huber P. Zerebrale Angiographie für Klinik und Praxis. Stuttgart, Germany: Thieme Medical Publishers, Inc; 1978.
Krayenbühl H, Yasargil MG. Cerebral arteries: radioanatomy. In: Vinken PJ, Bruyn GW, eds. Handbook of Clinical Neurology. Amsterdam, Netherlands: North Holland Publishing Co; 1985;11:72.
Anzola GP, Gasparotti R, Magoni M, Prandini F. Transcranial Doppler sonography and magnetic resonance angiography in the assessment of collateral hemispheric flow in patients with carotid artery disease. Stroke. 1995;26:214-217.
Bogdahn U, Becker G, Schlief R, Reddig J, Hassel W. Contrast-enhanced transcranial color-coded real-time sonography: results of a phase-two study. Stroke. 1993;24:676-684.
Griewing B, Doherty C, Kessler C. Power Doppler: a new tool for transcranial duplex assessment of intracranial vasculature. Stroke. 1995;26:721. Abstract.