Association between Internal Carotid Artery Morphometry and Posterior Communicating Artery Aneurysm

The supraclinoid internal carotid artery (ICA) is defined by a pre-anterior clinoidectomy measurement of the length from the ICA bifurcation to the point where the ICA view becomes obstructed the anterior clinoid process (ACP). Occasionally, the ACP has to be partially removed to secure proximal control of the aneurysm, which is covered by the anterior clinoid process in posterior communicating artery (PCOM) aneurysm, especially when the supralinoid ICA is short. Based on intraoperative findings, we noticed that the length of the supraclinoid ICA is shorter in patients with PCOM aneurysms than in patients with aneurysms of the anterior circulation. Previously, most anatomical studies1-4 for the arteries of the circle of Willis were performed in adult cadaver heads, but there is relatively little data for patients with ruptured cerebral aneurysms.5 The following report uses direct measurement to establish the relationship between ICA morphometry and the presence of PCOM aneurysm.

Between June 2004 and December 2005, 127 patients with aneurysms were surgically treated at our hospital. For study cases, we selected 81 consecutive patients with ICA-PCOM aneurysms, anterior communicating artery (ACOM) aneurysms, or middle cerebral artery (MCA) bifurcation aneurysms (n = 27 each). We intraoperatively measured the length of the supraclinoid ICA because the proximal portion of the intradural segment of the ICA is covered by the anterior clinoid process. ICA length cannot be radiologically measured because it is impossible to localize the exact location of the ACP. Also, the length of the ICA changes according to the angle of projection on the anteroposterior view or lateral view of angiogram. The length from the most proximal part of the exposed ICA to the ICA bifurcation was measured intraoperatively with a paper ruler after complete aneurismal clipping (Fig. 1). The length of the supraclinoid ICA was compared with that of the ICA-PCOM aneurysms, ACOM aneurysms, and MCA bifurcation aneurysms.

Fig. 1
Intraoperative photograph of the measurement obtained before anterior clinoidectomy. The length of the left supraclinoid internal carotid artery (ICA) was defined from the ICA bifurcation to the point where the ICA view becomes obstructed the anterior clinoid process (ACP). (a) ICA, (b) A1 portion of the anterior cerebral artery, (c) optic nerve, (d) ACP.

• Click for larger image

Maximum-intensity pictures of the contrast-opacified arteries were obtained by three-dimensional computed tomographic angiography (3-D CTA, Siemens, Germany). We defined the angle between the skull midline and the terminal segment as the ICA angle. The ICA angle and diameter of the supraclinoid ICA were measured with an image analyzer on 3-D CT angiogram (Fig. 2). Additionally, the diameters (at the mid-portion) and lengths of the M1 segment of the MCA and dominant A1 segment of the anterior cerebral artery were measured with an image anal yzer on CT angiogram and compared for each type of aneurysms (Fig. 3). We also checked the angiographic presence of the fetal PCOM in each aneurysm. To reduce interobserver error, the neuroradiologist was blinded to the results obtained by the primary investigator who measured all the reconstructed images. A student's t-test was used to assess the differences in the estimated data. The incidence of fetal PCOM relative to each aneurysm location was compared using the X² test. A p value < 0.05 was considered statistically significant.

Fig. 2
The ICA angle and supraclinoid ICA diameter as measured with an image analyzer on CT angiogram. The ICA angle was defined as the angle between the skull midline and terminal segment.

• Click for larger image

Fig. 3
Measurement of the length and diameter of the M1 portion of the middle cerebral artery (MCA) and dominant A1 portion of the anterior cerebral artery with an image analyzer on CT angiogram.

• Click for larger image

The average length of the measured supraclinoid ICAs from each aneurysm was 11.9 ± 2.3mm (mean ± standard deviation). The average length for PCOM aneurysms (9.7 ± 2.8mm) was shorter than that for ACOM aneurysms (13.8 ± 2.2mm, Student's t-test, p < 0.001) and also for MCA bifurcation aneurysms (12.2 ± 1.9mm, Student's t-test, p < 0.001) (Fig. 4). The average length of M1 was 20.1 ± 5.6mm in PCOM aneurysms, 21.0 ± 5.7mm in ACOM aneurysms, and 19.2 ± 4.7mm in MCA bifurcation aneurysms. The average length of A1 was 15.6 ± 3.3mm in PCOM aneurysms, 15.7 ± 3.1mm in ACOM aneurysms, and 17.6 ± 4.5mm in MCA bifurcation aneurysms. There were no significant differences of the lengths of M1 and A1 among each aneurysm. The average diameter of the supraclinoid ICA in patients with ACOM aneurysms (4.7 ± 1.3 mm) was larger than that in patients with MCA bifurcation aneurysms (3.8 ± 1.0mm) (Student's t-test, p < 0.05) and with PCOM aneurysms (4.1 ± 0.9 mm) (Student's t-test, p = 0.07) (Fig. 5). Also the average diameter of A1 in patients with ACOM aneurysms (2.4 ± 0.7mm) was larger that in patients with MCA bifurcation aneurysms (2.0 ± 0.4mm, Student's t-test, p < 0.05) and with PCOM aneurysms (2.1 ± 0.7mm) (Student's t-test, p = 0.1). The diameter of the M1 (ACOM; 3.1 ± 0.9mm, PCOM; 3.0 ± 0.8mm, MCA 2.8 ± 0.6mm) and the ICA angles (ACOM; 46.7 ± 10.7 degree, PCOM; 45.0 ± 11.1 degree, MCA; 42.3 ± 9.2 degree) were similar for each aneurysm (Fig. 6). The fetal PCOM was seen in 8 cases (29.6%) of PCOM aneurysms, 7 cases (25.9%) of MCA aneurysms, and 7 cases (25.9%) of ACOM aneurysms.

Fig. 4
The length of the supraclinoid ICA, A1, and M1 according to the aneurysm location. The lengths of the posterior communicating artery (PCOM) aneurysm (9.7 ± 2.8mm) are shorter than those of the anterior communicating artery (ACOM) aneurysms (13.8 ± 2.2mm) (Student's t-test, p < 0.001) and of MCA bifurcation aneurysms (12.2 ± 1.9mm) (Student's t-test, p < 0.001). Data are expressed as the mean ± standard deviation.

• Click for larger image

Fig. 5
The diameter of the supraclinoid ICA, A1, and M1 according to the aneurysm location. The diameters of the supraclinoid ICA in patients with ACOM aneurysm (4.7 ± 1.3mm) are larger than those of MCA bifurcation aneurysms (3.8 ± 1.0mm) (Student's t-test, p < 0.05) and of PCOM aneurysms (4.1 ± 0.9mm). Also, the diameters of the A1 in patients with ACOM aneurysms (2.4 ± 0.7mm) were larger than those of patients with MCA bifurcation aneurysms (2.0 ± 0.4mm) (Student's t-test, p < 0.05) or with Pcom aneurysms (2.1 ± 0.7mm). Data are expressed as the mean ± standard deviation.

• Click for larger image

Fig. 6
The ICA angle according to the aneurysm location. There are no significant differences among the different aneurysms. Data are expressed as the mean ± standard deviation.

• Click for larger image

Earlier hypotheses suggested that the appearance of aneurysms is exclusively due to congenital abnormalities, but scientific findings suggest that hemodynamic and degenerative factors can contribute to aneurysm development.6 It is now generally accepted that an aneurysm is an acquired lesion, resulting from a complicated interplay of anatomical, hemodynamic, and degenerative factors. Anatomically, intracranial arteries have a much thinner wall than their extracranial counterparts, contain very little elastin, and lie in the subarachnoid space, where they are poorly supported by surrounding tissue. Each of these conditions decreases the ability of an intracranial artery to resist the formation of an aneurysm.7 Hemodynamics have been implicated by clinical observations of a common association of aneurysms with high flow arteriovenous malformations and a high incidence of saccular aneurysms together with anomalies of the circle of Willis,8, 9 as well as several experimental studies on the flow patterns in aneurysms.10-12 Focal degeneration of the internal elastic lamina may result from theses hemodynamic forces, accounting for the preferential occurrence of aneurysm formation at the apex of bifurcations, as these are the sites of maximal hemodynamic stress.

Most morphometric studies measuring arteries in the circle of Willis were done during autopsy.1-4 The present analyzed data were obtained in the intraoperative field and by CT angiography, and were compared with the other data series. Actually, the supraclinoid portion of the ICA begins where the artery passes above the ACP to enter the subarachnoid space and terminate at the ICA bifurcation, and its average length is 14.8 ±3.0mm.1 We defined the supraclinoid ICA as the distance from the ICA bifurcation to the point where the ICA view becomes obstructed by the ACP, as measured by pre-anterior clinoidectomy. In the cadaveric morphometric study of Evans et al.,1 the left and right combined mean values (10.5 ± 2.4mm) of the ICA length before removal of the ACP were shorter than the mean values (11.9 ± 2.3mm) of the supraclinoid ICA in our intraoperative observations. In particular, the lengths in PCOM aneurysms (9.7 ± 2.8mm) were statistically significantly shorter than those in MCA bifurcation aneurysms (12.2 ± 1.9mm) or in ACOM aneurysms (13.8 ± 2.2mm). Marinković et al.3 reported the lengths of A1 as ranging from 8-18.5mm (average 13.5mm), which are shorter than those in our study (16.3 ± 3.6mm). Tanriover et al.4 reported lengths of M1 ranging from 10.1-29.3mm (average 17.8mm), which are also shorter than our data (20.1 ± 5.3mm). Rhoton Jr., et al.13 reported the diameter of the supraclinoid ICA as ranging from 2.5-7.0mm (average 4.3mm), which is similar to those (4.2 ± 1.0mm) in our study. Marinković et al.3 reported diameters of A1 ranging from 0.7-3.75mm (average, 2.1mm) and of the M1 segment, ranging from 1.5-3.5mm (average, 2.7mm), which is similar to those (2.2 ± 0.6mm, 2.9 ± 0.8mm, respectively) in our study. The differences between the results obtained from the cadaveric data and our intraoperative and radiological observations may be caused by the shrunken state of the major artery at autopsy. Chrzanowski14 reported ICA angles ranging from 20-43 degrees (average 31.5 degrees), which was also smaller than the ones measured in our study (44.7± 0.3 degree).

Based on the results of our study, there were no differences in the length of A1 and M1 according to the aneurysm location. The ICA angles did not have any significant relationship with the type of aneurysm. There was no significant difference between the presence of the fetal PCOM and aneurysm location. Thus, we did not analyze the characteristics of the PCOM, including the size of the PCOM and the angle between the ICA and the PCOM origin. The lengths of the supraclinoid ICA in ICA-PCOM aneurysms, however, were shorter than those in ACOM aneurysms and in MCA bifurcation aneurysms. Of 11 circles of Willis where an aneurysm arose from one of two unequal ICAs, Stehbens9 observed that the lesion developed on the larger vessel seven times, on the smaller vessel twice, and bilaterally twice, and suggested a likelihood at which an aneurysm would occur on the larger vessel having high flow rates. The geometry of the ICA is composed of abrupt bendings and several branchings that finally make the ICA bifurcations. In such conditions, a remarkably disturbed flow develops in the abrupt bending and branching points.15, 16 The hemodynamic forces in these areas are expected to be maximal because of larger flow volume and velocity compared to that in other intracranial vessels. These results may be explained by the observation that the momentum (momentum = mv, where m is the mass and v is the velocity) per unit volume of flow is greatest in the central streams because of the relatively high flow velocities.17 The peak force of the impulse by central streams will be transmitted to the branching points. Moreover, this pulsatile impulse is great because of the brief impact time at the relatively short supraclinoid ICA (impulse equals the change in momentum, and may be expressed as ∫^t1_t2 F dt where F is the net force producing the change in momentum, and dt is the time of impact).7 For a given change in momentum, a short impact time will result in the transmission of a high force. Only a small increase in impulse (caused by the relatively short supraclinoid ICA) may be sufficient to initiate aneurysm formation once a disruption of the internal elastic lamina has occurred. Although this association does not necessarily imply direct causality, the authors speculate that more hemodynamic stress may be produced to the proximally located ICA PCOM junction at higher flow rates than at other locations, resulting in aneurysm formation from developing degenerative change. The diameters of dominant A1 and supraclinoid ICA were relatively larger in ACOM aneurysm than in other aneurysms. Several authors6, 8, 9 have drawn attention to the size inequality of the proximal segments of the anterior cerebral arteries in the presence of an ACOM aneurysm. ACOM may be exposed to high hemodynamic stresses by a considerable shunt of blood across the ACOM to the contralateral vessel.

The pathophysioloy of ICA-PCOM aneurysm formation is probably multifactorial. Indirect conclusions can be drawn from our results, providing the first evidence that the length of the supraclinoid ICA may be a novel risk factor for the development of ICA-PCOM aneurysm. Further supplemental investigation is necessary to confirm our findings.