Sage Journals: Discover world-class research

Abstract

Angiography-based balloon test occlusion (BTO) has been empirically used to predict tolerance to permanent carotid artery occlusion. We tested the hypothesis that the laterality of the hemispheric circulation time (HCT) of the contrast medium at cerebral angiography would reflect bilateral asymmetry of the cerebral blood flow (CBF) during BTO. Thirty-one consecutive patients who underwent BTO of the internal carotid artery were retrospectively analyzed. HCT was defined as the interval between the time-to-peak in the middle cerebral artery and the cortical veins calculated using time-density curve. The difference in HCT between the occluded and nonoccluded side was calculated at the carotid or dominant vertebral angiograms obtained during BTO. We estimated the correlation between the difference in HCT and bilateral asymmetry of the CBF, which was quantitatively determined by single-photon emission computed tomography. The HCT was 5.3±1.5 seconds and regional CBF was 41.3±11.3 mL/100 g per minute in the occluded side, compared with 3.6±0.9 seconds and 48.4±14.9 mL/100 g per minute in the nonoccluded side, respectively. The difference in HCT was strongly correlated with the asymmetry ratio of the CBF (r²=0.89, P<0.0001). Angiographically based measurement of the cerebral circulation time can provide valuable information concerning cerebral hemodynamics.

Keywords

angiography balloon test occlusion carotid artery sacrifice cerebral blood flow circulation time

INTRODUCTION

Occlusion of the internal carotid artery (ICA) for treating symptomatic large or giant paraclinoid or intracavernous carotid aneurysms or complex cranial base tumors can cause ischemic complications, even in patients who successfully tolerate the balloon test occlusion (BTO) of the ICA.¹ The incidence of immediate and delayed ischemic stroke due to hypoperfusion, excluding that for embolic complications, ranges from 2% to 20%.2, 3, 4, 5 Various neuroimaging protocols such as ¹³³Xe-enhanced computed tomography,⁶ positron emission tomography,^799mTc-hexamethylpropyleneamine oxime single-photon emission computed tomography (SPECT),⁸ regional cerebral oxygen saturation monitoring,⁹ and transcranial Doppler ultrasonography¹⁰ have been reported; these protocols use modified BTO to increase the sensitivity of clinical test occlusion for predicting patients at risk for hypoperfusion-related ischemic events. However, many of these additional tests are expensive, complicated, or cumbersome to perform, and no combination of BTO and ancillary tests has significantly reduced the incidence of ischemic events compared with BTO and clinical evaluation.^{11, 12}

Several investigators have reported a simple angiographic BTO technique known as venous phase BTO.^{13, 14, 15, 16} The angiography-based BTO empirically stratifies the risk of permanent ICA occlusion based on the angiographically measured time-lag between hemispheres in the opacification of cortical veins during BTO.¹⁷ Although angiographic BTO requires additional catheter placement in the contralateral ICA or the dominant vertebral artery, this test appears to address cerebral perfusion without a requisitely protracted examination. However, although cortical vein opacification may be related to contrast medium circulation time, the physiologic meaning of the difference in circulation time between the occluded and nonoccluded hemisphere during BTO remains unknown.

Here, we aimed to establish the theoretical background of angiography-based BTO. For this purpose, the hemispheric circulation time (HCT) was evaluated by calculating the interval between the time-to-peak in the proximal middle cerebral artery and the cortical veins in each cerebral hemisphere using time-density curve derived from cerebral angiograms. We investigated the relationship between the difference in HCT and side-to-side asymmetry of the regional cerebral blood flow (CBF) as quantitatively determined with SPECT during BTO for the ICA. We also examined the types of prominent collateral circulation in relation to the asynchronicity of HCT during BTO.

MATERIALS AND METHODS

Patients

All subjects described in this study participated in a clinical neurosurgical protocol established at the Kohnan Hospital, which is affiliated with the Tohoku University Graduate School of Medicine in Sendai, Japan. The Kohnan Hospital institutional review board approved this retrospective study. Between April 2008 and March 2012, 31 consecutive patients (3 men and 28 women; mean age, 63.2 years; range, 36 to 86 years) underwent BTO of the ICA. The patients' clinical diagnoses comprise large or giant intracavernous carotid aneurysms (n=21), large paraclinoid aneurysms (n=9), or arterial dissection at the petrous portion of the ICA (n=1). One patient with right temporal lobe infarction was excluded. No other patient had abnormal findings of brain parenchyma on magnetic resonance imaging or stenoocclusive disease on cerebral angiography. All patients provided consent for participation in both BTO and SPECT imaging. Informed consent was obtained in written form as well as orally, emphasizing in particular the potential health risks associated with endovascular procedures and radiation exposure with imaging. The signed consent form contained a description of the patient's rights in agreement with the Helsinki Declaration.

Balloon Test Occlusion Procedure

The BTO procedure was performed under local anesthesia and systemic heparinization as described previously.¹⁸ A 6F large-lumen guiding catheter, the Launcher (Medtronic, Minneapolis, MN, USA) was positioned in the cervical portion of the ICA. A 2.8 F double-lumen silicone balloon catheter, the MASAMUNE (Fuji Systems, Tokyo, Japan) was advanced into the ascending petrous portion of the ICA. The balloon was inflated under fluoroscopic visualization until the contrast injection from the guide catheter demonstrated complete flow arrest. Temporary occlusion was continued for 15 minutes, and two angiogram series (the contralateral ICA and the dominant vertebral artery) were performed with a second 5 F catheter to evaluate collateral circulation. Neurologic examination and measurement of arterial blood pressure were performed every 5 minutes. Stump pressure, which reflects postocclusion back pressure of the ICA, was measured at the distal tip of the inflated balloon through the other lumen of the MASAMUNE catheter. Patient without clinical symptoms during temporary occlusion were transferred to the radioisotope suite with the balloon deflated but still in position.

All angiographic procedures employed the Innova system (GE Healthcare, Buc, France) with the following parameters: matrix size, 750 × 750; field of view, 20 × 20 cm, 2.5 frames per second. A total of 8 mL of iodinate contrast was injected at 6 mL/second. This is the same protocol used for conventional angiography at our institution.

Cerebral Blood Flow Measurements

In the radioisotope suite, the balloon was reinflated to its previous volume, and complete occlusion was confirmed by ensuring that the stump pressure was the same as previously. SPECT with ¹³³Xe inhalation was performed before December 2010 and N-isopropyl [123I]-p-iodoamphetamine (¹²³I-IMP) was performed after January 2011 by using a three-head gamma camera system (PRISM-IRIX; Shimadzu, Kyoto, Japan). With the ¹³³Xe inhalation method, patients (patient 1 to 10) inhaled 740 MBq of ¹³³Xe gas 1 minute after balloon inflation, and CBF was quantitatively measured by dynamic SPECT (16 scans of a duration of 20 seconds each) according to the method of Kanno and Lassen.¹⁹ Eight minutes later, the balloon was deflated and removed. Data acquisition was started immediately after inflation and continued for 8 minutes. In the remaining patients (patient 11 to 30), SPECT studies were performed using the ¹²³I-IMP-ARG method.²⁰ One minute after balloon inflation, 222 MBq of ¹²³I-IMP was intravenously injected, arterial blood sampling (2 mL) was performed 10 minutes later, and the balloon was deflated and removed. Data acquisition began 1 minute after ¹²³I-IMP injection and continued for 28 minutes. ¹²³I-IMP is a radioactive tracer with high lipid solubility and a high first-pass extraction ratio in the brain, which is estimated as ∼90%. The initial uptake of ¹²³I-IMP is believed to relate largely to its high affinity for the nonspecific amine-binding sites in the brain, which reflects the slow washout of the agent during a period of more than 1 hour. Therefore, the image obtained 28 minutes postinjection is indicative of CBF distribution at the time of tracer injection, even if changes in the CBF occur after the injection. At least a 5-min occlusion may be necessary to obtain the flow during the occlusion, because the majority of IMP extraction occurs during the initial 5 minutes.²¹

A quantitative CBF map, a 128 × 128 image matrix with a slice thickness of 9 mm, consisted of 15 axial slices covering the whole brain. A total of 22 regions of interest (ROIs) (area, 600 mm²) were bilaterally and symmetrically placed on each side of the middle cerebral artery territory in four slices (Figure 1). Each ROI provided a value of CBF measured as mL/100 g per minute. The average of mean CBF values for all ROIs in each hemisphere is defined as CBF in the middle cerebral artery territory. Finally, the degree of side-to-side CBF asymmetry (asymmetry ratio, AR) was obtained using the following equation: AR=CBF_occl/CBF_non-occl

Figure 1.

N-isopropyl [123I]-p-iodoamphetamine single-photon emission computed tomography (¹²³I-IMP SPECT) images of patient 21. Four transaxial slices with regions of interest (ROIs). Twenty-two ROIs (black circles) were symmetrically placed on the cortex of the middle cerebral artery territory.

Normal CBF values were previously measured in our institute. Cerebral blood flow in the middle cerebral artery territory was 54.2±10.7 mL/100 g per minute (mean±s.d.) as determined in 12 normal individuals (7 men and 5 women) aged 40 to 70 years¹⁸ using ¹³³Xe SPECT according to the method of Kanno and Lassen.¹⁹ It was 36.6±3.0 mL/100 g per minute as determined using the ¹²³I-IMP-ARG method²⁰ in 10 normal individuals (5 men and 5 women) aged 56 to 65 years (unpublished data).

Hemispheric Circulation Time Measurements

To investigate the impact of BTO on HCT, ROI analysis of vessel optical density on angiograms was performed. Conventional angiographic image sequence data obtained during BTO were processed with ImageJ (NIH, Bethesda, MD, USA) to measure optical density as the contrast medium passed through the vessels at a rate of 2.5 frames per second. Regions of interest of 5.0 mm² were drawn on the most proximal zone of the middle cerebral artery and on each of 6–7 prominent cortical veins draining from the middle cerebral artery territory (Figures 2 and 3). This setting of ROIs on the arterial side enables observers to compare the HCT in the same condition among patients with dominant collateral flows through the anterior or the posterior communicating artery. On the venous side, the mean timing of cortical venous opacification could be estimated separately in each hemisphere. We specially attempted to locate ROIs that are bilaterally symmetric without superimposition of adjacent vessels. In patients in whom the anterior communicating artery was dominant, the anteroposterior view obtained by contralateral ICA injection during BTO was used for analysis (Figure 2). In patients in whom the posterior communicating artery was dominant, the anteroposterior view obtained by dominant vertebral artery injection was compared with that of the contralateral ICA injection during the ipsilateral ICA occlusion (Figure 3). The time-density curve of each ROI was given as the mean of the optical density for all the pixels in the ROI (Figure 4). All ROIs were drawn by two radiologic technicians masked to clinical condition and SPECT image.

Figure 2.

Anteroposterior angiogram views of patient 16. Arterial (A) and venous (B) phases of the left internal carotid angiograms during a right internal carotid occlusion by balloon, demonstrating collateral flow through the anterior communicating artery. White circles indicate regions of interest placed on the proximal middle cerebral arteries (A) and symmetrically on both cortical veins (B).

Figure 3.

Anteroposterior angiogram views of patient 23. Arterial (A) and venous (C) phases of the right internal carotid angiograms, and angiograms of the left vertebral artery, arterial (B) and venous (D) phases during a left internal carotid occlusion by balloon, demonstrating collateral flow through the posterior communicating artery (B, D), but not through the anterior communicating artery (A, C). White circles indicate regions of interest (ROIs) on the proximal middle cerebral arteries in panels A, B and cortical veins in panels C, D. Special effort was made to locate ROIs that are bilaterally symmetric.

Figure 4.

Illustrative case in the measurement of hemispheric circulation time (HCT) (patient 8). (A, B) Arterial (A) and venous (B) phases of the right internal carotid angiograms during a left internal carotid occlusion by balloon. White circles (RM1, LM1, RCV1–4, and LCV1–4) indicate regions of interest (ROIs) labeled corresponding to the time-density curves. Right columns: Time-density curves of the proximal middle cerebral arteries (RM1 and LM1) and cortical veins (RCV1–4 and LCV1–4). The ordinate represents the relative inverse density in ROIs. The abscissa indicates the number of frames (2.5/second). Red (RM1 and LM1) and blue (RCV1–4 and LCV1–4) curves indicate the mean inverse density in ROIs. Dotted line in each graph indicates time-to-peak (TTP). HCT_occl (the left side)=mean of (TTP_LCV1–4—TTP_LM1)=4.9 seconds. HCT_non-occl (the right side)=mean of (TTP_RCV1-4—TTP_RM1)=3.4 seconds. HCT_diff=HCT_occl—HCT_non-occl=1.5 seconds.

Hemispheric circulation time was defined as the mean of the interval between the time-to-peak in the proximal middle cerebral artery (TTP_A) and the cortical veins (TTP_V) in each hemisphere. Finally, the HCT difference (HCT_diff) between the occluded (HCT_occl) and nonoccluded (HCT_nonoccl) hemispheres of each patient was calculated with the following equations:

HCT_occl (seconds)=mean of (TTP_{V of occl}—TTP_{A of occl})

HCT_nonoccl (seconds)=mean of (TTP_{V of nonoccl}—TTP_{A of nonoccl})

HCT_diff (seconds)=HCT_occl—HCT_non-occl

Analysis of Collateral Flow During Balloon Test Occlusion

We examined biplane views to assess the absence or presence of collateral circulation during BTO of the ICA. Collaterals were angiographically recognized by contrast filling via the anterior and/or posterior communicating artery (primary collateral circulation), or leptomeningeal anastomosis after contralateral ICA or dominant vertebral artery injection. Arterial supply to the middle cerebral artery territory via retrograde filling of the ophthalmic artery fed by branches of the ipsilateral external carotid artery was examined through the contrast injection from the guiding catheter during balloon occlusion. The appearance of collaterals was then graded on a 3-point scale according to the method of Henderson et al:²² (1) Not reaching the middle cerebral artery territory; (2) partially covered in the middle cerebral artery territory; (3) fully covered in the middle cerebral artery territory. For the present study, grades 2 and 3 were combined to indicate the presence of collateral circulation. Collaterals were considered absent if they were not angiographically visualized (grade 0) or were assigned a grade of 1.

Statistical Analysis

Statistical analyses were performed using JMP software (Version 9; SAS Institute, Cary, NC, USA). Numerical data are expressed as mean±s.d., and comparisons between the occluded and the nonoccluded side were performed with paired t-tests. Correlations between HCT_diff and AR were determined using linear regression analysis. Group comparisons of categorical data were performed with Mann–Whitney U-tests. P<0.05 was considered statistically significant.

RESULTS

Table 1 presents the baseline characteristics and quantitative data of our study population. No procedural complications occurred during BTO. Among the 31 enrolled patients, 27 underwent both HCT and CBF assessment. The SPECT study could not be performed in two patients (patient 9 and 12) who exhibited new neurologic deficits after balloon occlusion. For these patients, angiographic findings and stump pressure values obtained before balloon deflation were used in subsequent studies. Complete resolution of their neurologic deficits was observed after balloon deflation. In patient 20, angiographic data were not available for the HCT measurement because of a motion artifact.

Table 1.

Baseline characteristics and quantitative data of the study population

					Collateral flow				Circulation time (second)			Regional CBF (mL/100 g per minute)			mSP
No.	Age	Sex	Diagnosis	Side	Acom	Pcom	Lept.	Oph.	O. side	N. side	Interval	O. side	N. Side	AR (× 10⁻²)	(mm Hg)
1	53	M	CA AN	L	1	3	0	1	4.2±0.2	2.6±0.2	1.6±0.3	60.5±8.9	69.4±9.2	87.4	58.0
2	73	F	CA AN	R	3	0	2	0	4.4±0.5	4.1±0.3	0.9±0.7	55.6±5.2	57.9±5.2	96.4	94.0
3	65	F	Paraclinoid AN	R	3	0	2	1	4.6±0.4	2.8±0.4	1.8±0.4	71.9±6.3	82.0±6.1	87.9	62.0
4	63	F	CA AN	R	2	2	2	1	6.6±0.5	3.2±0.0	3.4±0.5	43.9±5.4	64.3±8.0	69.8	38.0
5	66	F	Paraclinoid AN	L	2	0	2	1	5.8±0.2	1.8±1.0	4.0±0.9	42.2±5.8	79.0±10.6	53.9	33.0
6	63	F	CA AN	R	2	0	3	1	6.5±0.5	3.1±1.3	3.5±1.6	38.7±14.2	64.9±6.9	61	15.0
7	60	F	CA AN	L	2	2	0	0	4.5±0.5	3.3±0.4	1.2±0.4	57.7±5.3	65.4±7.2	88.9	68.0
8	59	F	CA AN	L	3	0	2	1	4.9±0.4	3.4±0.3	1.5±0.6	54.1±7.9	62.6±6.8	86.8	67.0
9	62	F	Petrous ICAD	L	2	0	0	2	9.6±0.5	5.2±0.0	4.4±0.5	N/A	N/A	N/A	22.5
10	55	F	Paraclinoid AN	R	2	2	0	1	4.9±0.3	4.1±0.3	0.9±0.2	40.3±3.9	43.1±6.8	94.7	54.0
11	74	F	CA AN	R	2	3	0	1	4.27±0.6	3.6±0.4	0.7±0.7	55.7±6.0	58.3±5.7	95.5	55.0
12	86	F	CA AN	R	2	0	0	2	7.5±0.2	3.9±0.6	3.9±0.6	N/A	N/A	N/A	34.0
13	84	F	CA AN	R	1	3	0	2	5.4±0.6	3.4±0.5	2.0±0.3	36.2±2.0	41.6±2.3	87.2	57.5
14	36	F	CA AN	R	3	0	0	2	5.2±0.6	3.2±0.0	2.0±0.6	29.3±2.2	33.8±2.6	86.8	30.0
15	61	F	CA AN	R	3	2	0	1	6.2±0.5	4.9±0.8	1.3±1.0	25.0±1.7	30.6±2.7	88.5	62.0
16	58	F	Paraclinoid AN	R	3	0	0	0	4.5±0.3	3.8±0.6	0.7±0.7	41.0±1.7	43.8±4.2	94.3	61.5
17	63	F	CA AN	R	2	3	0	2	6.3±0.6	3.5±0.5	2.8±0.6	33.4±3.8	45.1±3.3	74.1	41.0
18	61	F	CA AN	R	2	0	0	2	4.5±0.6	3.7±0.3	0.7±0.4	37.9±2.1	40.2±3.0	94.6	50.0
19	56	F	CA AN	R	0	3	0	1	4.1±0.4	3.7±0.5	0.4±0.4	34.5±3.1	36.5±2.2	94.5	61.0
20	56	F	CA AN	L	2	0	2	2	N/A	N/A	N/A	41.7±2.6	50.2±2.5	83	61.0
21	69	F	Paraclinoid AN	L	3	0	2	1	8.5±1.4	7.1±0.8	1.4±1.0	38.6±2.4	42.6±1.0	90.8	54.0
22	56	F	Paraclinoid AN	R	3	0	2	1	4.1±0.4	3.0±0.2	1.1±0.3	34.4±2.8	38.4±2.6	89.7	45.0
23	52	F	CA AN	L	0	3	0	1	4.6±0.3	3.6±0.4	1.0±0.5	28.1±2.5	29.4±3.0	95.7	43.0
24	60	F	Paraclinoid AN	R	1	3	0	2	4.1±0.4	2.7±0.4	1.4±0.4	35.5±3.0	39.1±2.7	90.8	117
25	83	F	CA AN	R	2	0	2	1	3.9±0.6	2.8±0.3	1.1±0.4	45.7±2.4	49.7±3.0	92.1	58.0
26	68	F	CA AN	R	2	3	0	0	3.8±0.6	3.5±0.5	0.3±0.9	36.7±2.2	38.5±2.6	95.6	69.0
27	71	F	Paraclinoid AN	L	2	2	0	1	6.0±0.0	3.9±0.7	2.1±0.7	36.5±4.1	42.4±3.1	86.2	61.0
28	63	F	CA AN	R	3	2	2	1	5.2±0.6	4.3±0.6	0.9±0.8	42.7±3.0	46.5±4.8	92.1	47.0
29	56	F	CA AN	R	2	3	0	0	3.3±1.0	3.3±0.7	0.1±0.8	35.0±3.6	36.2±3.1	96.8	58.0
30	63	M	CA AN	L	2	3	0	0	4.0±0.7	3.3±0.3	0.7±0.5	22.9±1.5	24.5±1.4	93.6	79.5

Acom, anterior communicating artery; AR, asymmetry ratio; CA AN, intracavernous carotid aneurysm; CBF, cerebral blood flow; lept., leptomeningeal anastomosis; mSP, mean stump pressure; N/A, data not available; n side, nonoccluded side; oph., ophthalmic artery; o.side, occluded side; paraclinoid AN, paraclinoid aneurysm; pcom, posterior communicating artery; petrous ICAD, internal arterial dissection in the petrous portion.

Numerical data are expressed as mean±s.d. Cerebral blood flow studies of patients 9 and 12 are not available, because a cortical symptom occurred after balloon occlusion. Analysis of circulation time in patient 20 is not available because of poor angiographic quality. Stratification of the collateral circulation is determined according to the method of Henderson et al.:²¹ Grade 0, no collateral vessels; Grade 1, collateral flow not reaching the middle cerebral artery territory; Grade 2, collateral flow partially covered in the middle cerebral artery territory; Grade 3, collateral flow fully covered in the middle cerebral artery territory.

The mean HCT in the occluded side was significantly more extended than that in the nonoccluded side (5.2±1.5 versus 3.6±1.0 seconds, P<0.001). The mean CBF in the occluded side was significantly lower than that in the nonoccluded side (41.3±11.3 versus 48.4±14.9 mL/100 g per minute, P<0.001). Asymmetry ratios (ARs) were 0.81±0.15 with ¹³³Xe SPECT, and 0.91±0.06 with ¹²³I-IMP SPECT, and there was no significant difference between radiotracers used (P=0.250).

Comparison Between HCT _diff and Asymmetry Ratio

Figure 5 shows the comparison of HCT_diff and AR in each patient. The fit of the values to the linear regression line was significant (r²=0.89, P<0.0001). According to this line, the AR values of patients who could not undergo the SPECT study (patient 9 and 12) could be estimated as 0.59 and 0.64, respectively (y=1.03–0.1x; y, AR; x, HCT_diff).

Figure 5.

Linear regression analysis results. Regression line (solid line) and two-sided 95% confidence interval (dotted line). The abscissa and ordinate are the difference of hemispheric circulation time (HCT) and asymmetry ratio (AR), respectively (r²=0.89, P<0.0001).

Comparison of Collateral Circulation with HCT _diff and Asymmetry Ratio

Table 2 shows the comparison of collateral circulation with HCT_diff and AR. HCT_diff was 1.8±1.3 seconds and AR was 0.87±0.11 in the 25 patients with a dominant anterior communicating artery (grade 2 or 3), and there was no significant difference compared with the 16 patients who had a dominant posterior communicating artery (HCT_diff, 1.3±0.9 seconds; AR, 0.90±0.08; P=0.389, P=0.489, respectively). In the 11 patients with both dominant communicating arteries, the HCT_occl obtained by collateral circulation via the anterior communicating artery was not significantly different compared with the HCT_occl obtained by collateral circulation via the posterior communicating artery. There was no significant difference in HCT_diff and AR between patients with (grade 2 or 3) and without (grade 0 or 1) opacification of leptomeningeal anastomosis during occlusion of the ipsilateral ICA (HCT_diff, 2.0±1.2 second versus 1.5±1.2 second, P=0.770; AR, 0.82±0.14 versus 0.91±0.06, P=0.240). However, HCT_diff was significantly more extended, and AR was significantly lower in patients with collateral circulation via the ophthalmic artery supplied by branches of the external internal artery during ipsilateral ICA occlusion than in patients without it (HCT_diff, 2.5±1.3 seconds versus 1.4±1.0 seconds, P=0.031; AR, 0.86±0.07 versus 0.88±0.11, P=0.031).

Table 2.

Comparison of collateral circulations with HCT_diff and AR

	HCT _diff (s)	AR (× 10⁻²)
Anterior communicating artery
Yes (n=25)	1.8±1.3	86.7±11.4
No (n=5)	1.3±0.6	91.1±3.9
P value	0.125	0.313
Posterior communicating artery
Yes (n=16)	1.3±0.9	89.5±7.7
No (n=14)	2.1±1.4	84.8±13.4
P value	0.106	0.175
Ophthalmic artery
Yes (n=8)	2.5±±1.3	86.1±7.1
No (n=22)	1.4±1.0	87.8±11.4
P value	0.031	0.031
Leptomeningeal anastomosis
Yes (n=11)	2.0±1.2	82.1±14.1
No (n=19)	1.5±1.2	90.9±5.7
P value	0.770	0.240

AR, asymmetry ratio of the regional cerebral blood flow between hemispheres; HCT_diff, difference in hemispheric circulation time between occluded and nonoccluded side.

Numerical data are expressed as mean±s.d. Statistical differences were tested by the Mann–Whitney U-test. Stratification of the collateral circulation is determined according to the method of Henderson et al.:²¹ Grade 0, no collateral vessels; Grade 1, collateral flow not reaching the middle cerebral artery territory; Grade 2, collateral flow partially covered in the middle cerebral artery territory; Grade 3, collateral flow fully covered in the middle cerebral artery territory. Yes, grade 2 or 3; No, grade 0 or 1.

DISCUSSION

Angiographic-based BTO has been used to evaluate tolerance against carotid occlusion by measuring the time-lag in the opacification of cortical veins between the tested hemisphere and contralateral hemisphere during ipsilateral carotid occlusion by balloon catheter. Although thresholds for predicting hypoperfusion remain controversial among published protocols, this method has been widely used because of its simplicity, and has been proven safe when performed under general anesthesia, a beneficial option that permits intraoperative assessment for either planned therapeutic carotid occlusion or a bail-out strategy during complicated or deteriorating reconstructive aneurysm surgery.^{13, 14, 15, 16, 17} However, assessing synchronicity with venous filling by sight is subjective and obscure. Especially when collateral filling is via the posterior communicating artery, one has to assess the synchronicity between the venous phase of the tested hemisphere and that of the contralateral posterior circulation area on vertebral angiography. To what extent and which collaterals can modify the timing of the opacification of cortical veins during BTO are unknown. Perhaps surprisingly, the correlation of the HCT of the contrast medium with cerebral perfusion during BTO has not been proven.

In this study, we quantitatively evaluated cerebral circulation time in each hemisphere, namely HCT, by calculating the interval between time-to-peak in the proximal middle cerebral artery and the cortical veins on angiograms during BTO. The HCT difference between the occluded and nonoccluded hemisphere strongly correlated with side-to-side CBF asymmetry in the middle cerebral artery territory. The HCT difference was not associated with a dominant primary collateral pathway, but it was significantly extended in patients with secondary collateral circulation via the ophthalmic artery. These results suggest that the HCT difference between hemispheres could indicate asymmetric change of cerebral perfusion during BTO.

Although there are no strict criteria for perfusion abnormalities in BTO with SPECT, it is generally accepted that areas of diminished perfusion are abnormal when the radioactivity is less than 90% of normal.²³ When judging tolerance against carotid occlusion based on angiographic BTO, one should consider that 90% of contralateral CBF (AR=0.9) corresponds to 1.3 seconds in HCT difference according to this study (y=1.03–0.1x; y, AR; x, HCT_diff).

Greitz first described the method for evaluating cerebral circulation time by using cerebral angiography.²⁴ The time-lag from the maximum depiction time of the carotid siphon to that of the Rolandic vein was reported to be 3.5±0.5 seconds in 1,500 patients without neurologic deficit. Several previous studies demonstrated that the cerebral circulation time could reflect the state of cerebral perfusion and be one of the relevant parameters for cerebral hemodynamics.^{24, 25, 26, 27, 28} However, absolute values of cerebral circulation time cannot be used to define critical thresholds for cerebral perfusion because of great interpersonal variability and a host of multifactorial influences, such as the nature of contrast material, condition of injection, and the location of ROIs.²⁶ Cerebral circulation time is generally thought to be prolonged in the following scenarios: dilatation of the cerebral vascular bed, decrease of cerebral vascular resistance, and the decline in cerebral perfusion pressure.^{29, 30} Cerebral autoregulation is characterized by the physiologic mechanism that maintain CBF at an appropriate level during changes in blood pressure. This regulation is achieved primarily by small arteries and arterioles, which either dilate or contract under multiple complex physiologic control system, and can affect the cerebral circulation time.^{31, 32} Namely, cerebral circulation time may be prolonged despite a stable CBF level when cerebral autoregulation operates against a low perfusion pressure. This may be the main reason for the AR plateau that can be observed for HCT difference up to 1 s on Figure 5. With an HCT difference of >1 second, AR may start to decline because the autoregulation phenomenon may have reached a maximum level.

Approximately 55% of the cerebrovascular bed is thought to be composed of capillary vessels and veins, and 60% of cerebrovascular resistance is controlled by leptomeningeal arteries.³³ These data suggest that prolonged HCT due to reduced cerebral perfusion pressure mainly depends on a delay of the venous phase. Therefore, our results support the validity of angiographic BTO for measuring time-lag in the opacification of cortical veins.

Collateral circulation is the main mechanism of clinical tolerance of carotid occlusion. Primary collaterals include the arterial segments of the Circle of Willis, whereas the ophthalmic artery and leptomeningeal vessels constitute secondary collaterals. Secondary collateral pathways are presumed to be recruited once primary collaterals at the Circle of Willis have failed.³⁴ Although longitudinal studies have not chronicled this sequence of collateral failure, the presence of secondary collateral pathways is considered a marker of impaired cerebral hemodynamics.³⁵ This could be the main reason an extended HCT difference and the lower CBF AR were seen in patients with collateral flow via the ophthalmic artery, and likely but not statistically, in patients with leptomeningeal anastomosis.

The limitations of this study include the small number of patients and retrospective approach. We did not compare CBF and HCT between the before and after occlusion in the affected hemisphere. Cerebral hemodynamics may have already impaired at baseline, although no patient included in these analyses had abnormal cerebral parenchyma or stenoocclusive diseases on image studies. Measurement of HCT may involve large interobserver and interprocedural variability. More advanced approaches such as a stereotactic coordinates would be better suited for clinical applications. Because we were unable to simultaneously measure CBF and perform cerebral angiography, the variability may be due to measurements taken during different balloon inflations.

In conclusion, the present study can support the validity of angiography-based BTO and our findings are consistent with the notion that the difference of cerebral circulation time between hemispheres reflects asymmetric CBF distribution in the situation of abrupt carotid occlusion. These data can yield a meticulous assessment of the tolerance for carotid occlusion in an emergent or intraoperative condition, which is unsuitable for the use of quantitative CBF measurement modalities. A prospective randomized controlled study is warranted for testing the ability of HCT measurement during BTO to predict complications from permanent ICA occlusion.

Footnotes

The authors declare no conflict of interest.

References

Linskey

Jungreis

Yonas

Hirsch

Jr Sekhar

Horton

Stroke risk after abrupt internal carotid artery sacrifice: Accuracy of preoperative assessment with balloon test occlusion and stable xenon-enhanced CT. AJNR Am J Neuroradiol 1994; 15: 829–843.

De Vries

Sekhar

Horton

Eibling

Janecka

Schramm

A new method to predict safe resection of the internal carotid artery. Laryngoscope 1990; 100:85–88.

Gonzalez

Moret

. Balloon occlusion of the carotid artery prior to surgery for neck tumors. AJNR Am J Neuroradiol 1990; 11: 649–652.

Larson

Tew

Jr Tomsick

Van Loveren

. Treatment of aneurysms of the internal carotid artery by intravascular balloon occlusion: Long-term follow-up of 58 patients. Neurosurgery 1995; 36: 26–30.

Standard

Ahuja

Guterman

Chavis

Gibbons

Barth

Balloon test occlusion of the internal carotid artery with hypotensive challenge. AJNR Am J Neuroradiol 1995; 16: 1453–1458.

Johnson

Stringer

Marks

Yonas

Good

Gur

. Stable xenon CT cerebral blood flow imaging: Rationale for and role in clinical decision making. AJNR Am J Neuroradiol 1991; 12: 201–213.

Brunberg

Frey

Horton

Deveikis

Ross

Koeppe

. [150]H20 positron emission tomography determination of cerebral blood flow during balloon test occlusion of the internal carotid artery. AJNR Am J Neuroradiol 1994; 15: 725–732.

Peterman

Taylor

Jr Hoffman Jr.

JC. Improved detection of cerebral hypoperfusion with internal carotid balloon test occlusion and 99mTc-HMPAO cerebral perfusion SPECT imaging. AJNR Am J Neuroradiol 1991; 12: 1035–1041.

Kaminogo

Ochi

Onizuka

Takahata

Shibata

. An additional monitoring of regional cerebral oxygen saturation to HMPAO SPECT study during balloon test occlusion. Stroke 1999; 30: 407–413.

10.

Kofke

Brauer

Policare

Penthany

Barker

Horton

. Middle cerebral artery blood flow velocity and stable xenon-enhanced computed tomographic blood flow during balloon test occlusion of the internal carotid artery. Stroke 1995; 26: 1603–1606.

11.

Allen

Alastra

AJG

Nelson

. Proximal intracranial internal carotid artery branches: Prevalence and importance for balloon occlusion test. J Neurosurg 2005; 102:45–52.

12.

Chen

Ortiz

Page

Siddiqui

Veznedaroglu

Rosenwasser

. Spontaneous systolic blood pressure elevation during temporary balloon occlusion increases the risk of ischemic events after carotid artery occlusion. Neurosurgery 2008; 63: 256–265.

13.

Abud

Spelle

Piotin

Mounayer

Vanzin

Moret

. Venous phase timing during balloon test occlusion as a criterion for permanent internal carotid artery sacrifice. AJNR Am J Neuroradiol 2005; 26: 2602–2609.

14.

Sanna

Piazza

Ditrapani

Agarwal

. Management of the internal carotid artery in tumors of the lateral skull base: Preoperative permanent balloon occlusion without reconstruction. Otol Neurotol 2004; 25: 998–1005.

15.

Van Rooij

Sluzewski

Metz

Nijssen

Wijnalda

Rinkel

Carotid balloon occlusion for large and giant aneurysms: Evaluation of a new test occlusion protocol. Neurosurgery 2000; 47: 116–121.

16.

Van Rooij

Sluzewski

Slob

Rinkel

. Predictive value of angiographic testing for tolerance to therapeutic occlusion of the carotid artery. AJNR Am J Neuroradiol 2005; 26: 175–178.

17.

Vazquez Añon

Aymard

Gobin

Casasco

Ruffenacht

Khayata

Balloon occlusion of the internal carotid artery in 40 cases of giant intracavernous aneurysm: Technical aspects, cerebral monitoring, and results. Neuroradiology 1992; 34: 245–251.

18.

Shimizu

Matsumoto

Tominaga

. Parent artery occlusion with bypass surgery for the treatment of internal carotid artery aneurysms: Clinical and hemodynamic results. Clin Neurol Neurosurg 2010; 112: 32–39.

19.

Kanno

Lassen

. Two methods for calculating regional cerebral blood flow from emission computed tomography of inert gas concentrations. J Comput Assist Tomogr 1979; 3: 71–76.

20.

Iida

Ito

Nakazawa

Hatazawa

Nishimura

Onishi

Quantitative mapping of regional cerebral blood flow using iodine-123-IMP SPECT. J Nucl Med 1994; 35: 2019–2030.

21.

Furuhata

Kubo

Kawase

Ibata

Toya

. A simple technique to measure regional cerebral blood flow during intravascular balloon clamping. Radioisotopes 1988; 37: 226–338.

22.

Henderson

Eliasziw

Fox

Rothwell

Barnett

. Angiographically defined collateral circulation and risk of stroke in patients with severe carotid artery stenosis. North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. Stroke 2000; 31: 128–132.

23.

Monsein

Jeffery

Van Heerden

Szabo

Schwartz

Camargo

Assessing adequacy of collateral circulation during balloon test occlusion of the internal carotid artery with 99mTc-HMPAO SPECT. AJNR Am J Neuroradiol 1991; 12: 1045–1051.

24.

Greitz

. Normal cerebral circulation time as determined angiography with sodium and methylglucamine diatrizoate (urografin). Acta Radiol Diagn (Stockh) 1967; 7: 331–336.

25.

Cronqvist

. Regional cerebral blood flow and angiography in apoplexy. Acta radiol Diagn (Stockh) 1968; 7: 521–535.

26.

Cronqvist

Greitz

. Cerebral circulation time and cerebral bloods flow. Acta radiol Diagn (Stockh) 1968; 8: 296–304.

27.

Okada

Shima

Nishida

Yamane

Hatayama

Yamanaka

Comparison of transcranial Doppler investigation of aneurysmal vasospasm with digital subtraction angiographic and clinical findings. Neurosurgery 1999; 45: 443–449.

28.

Touho

Karasawa

Shishido

Yamada

Shibamoto

. Hemodynamic evaluation before and after the STA-MCA anastomosis—with special reference to measurement of regional transit time with intra-arterial digital subtraction angiography. Neurol Med Chir (Tokyo) 1990; 30: 663–669.

29.

Hamilton

Remington

. Comparison of the time concentration curves in arterial blood of diffusible and non-diffusible substance when injected at a contrast rate and when injected instantaneously. Am J Physiol 1947; 148: 35–39.

30.

Zierler

. Equations for measuring blood flow by external monitoring of radioisotopes. Circ Res 1965; 16: 309–321.

31.

Yoshioka

Matsukado

Kodama

Hirata

Fuwa

Takada

Analysis of cerebral blood circulation using interventional digital subtraction angiography. Neuro Med Chir (Tokyo) 1987; 27: 1053–1060.

32.

Okudaira

Arai

Sato

. Cerebral blood flow alteration by acetazolamide during carotid balloon occlusion. Parameters cerebral perfusion pressure in the acetazolamide test. Stroke 1996; 27: 617–621.

33.

Nylin

Hedlund

Regnstrom

. Studies of the cerebral circulation with labeled erythrocytes in healthy man. Circ Res 1961; 9: 664–674.

34.

Coyle

Heistad

. Development of collaterals in the cerebral circulation. Blood Vessels 1991; 28: 183–189.

35.

Liebeskind

. Collateral circulation. Stroke 2003; 34: 2279–2284.

Angiographic Circulation Time and Cerebral Blood Flow during Balloon Test Occlusion of the Internal Carotid Artery

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Patients

Balloon Test Occlusion Procedure

Cerebral Blood Flow Measurements

Hemispheric Circulation Time Measurements

Analysis of Collateral Flow During Balloon Test Occlusion

Statistical Analysis

RESULTS

Comparison Between HCT diff and Asymmetry Ratio

Comparison of Collateral Circulation with HCT diff and Asymmetry Ratio

DISCUSSION

Footnotes

References

Comparison Between HCT _diff and Asymmetry Ratio

Comparison of Collateral Circulation with HCT _diff and Asymmetry Ratio