J E Dion, P C Gates, A J Fox, H J Barnett and R J Blom Clinical events ...

Clinical events following neuroangiography: a prospective study. J E Dion, P C Gates, A J Fox, H J Barnett and R J Blom Stroke. 1987;18:997-1004 doi: 10.1161/01.STR.18.6.997 Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1987 American Heart Association, Inc. All rights reserved. Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://stroke.ahajournals.org/content/18/6/997

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Stroke is online at: http://stroke.ahajournals.org//subscriptions/

Downloaded from http://stroke.ahajournals.org/ by guest on February 25, 2013

997

Clinical Events Following Neuroangiography: A Prospective Study Jacques E. Dion, MD, FRCPC, Peter C. Gates, MB, BS, FRACP, Allan J. Fox, MD, FRCPC, Henry J.M. Barnett, MD, FACP, FRCPC, and Rita J. Blom, MD, FRCPC Clinical events following cerebral angiography were prospectively evaluated in 1,002 procedures. The ischemic event rate between 0 and 24 hours was 1.3% (0.1% permanent). This incidence was higher (2.5%) in patients investigated for cerebrovascular disease, but the difference was not significant. In addition, 1.8% of the patients suffered ischemia (0.3% permanent) between 24 and 72 hours after angiography. Cerebral ischemic events occurred as a recurrence or worsening of a preexisting condition twice as often as de novo. All permanent ischemia was a worsening of a preexisting phenomenon. There was a significant increase in the incidence of neurologic events between 0 and 24 hours when the procedure lasted longer than 60 minutes and when there was systolic hypertension. Trends toward higher incidence were noted with the use of increased volume of contrast, with increased serum creatinine, when transient ischemic attacks or stroke were the indications, and when 3 or more catheters were used. The incidence of neurologic events between 24 and 72 hours increased significantly with the increase in the amount of contrast used, with age, and with diabetes. The occurrence of nonneurologic events (mostly hematomas) was significantly increased by multiple factors. This study shows that events can and do occur beyond the usual observation period of 24 hours but confirms the low risk of cerebral angiography when performed judiciously. (Stroke 1987;18:997-1004)

F

emorocerebral catheter angiography has been the accepted technique of choice in evaluating the arterial tree of the head and neck for many years. It has virtually replaced direct puncture of the axillary, brachial, carotid, or vertebral arteries except in very rare cases. The risks of this procedure have been studied by many investigators1"27 over the last 20 years and appear to have gradually decreased (Table 1). This is probably due to marked improvements in angiographic equipment, operator dexterity, and the more widespread use of small catheters and betterdesigned guidewires. The use of heparinized catheters and guidewires2*"35 may have contributed to this decrease. Systemic heparinization may offer additional protection to selected patients,3*"40 but this remains to be proven. Classically, studies evaluating the risks of cerebral angiography have been retrospective, and none included events occurring >24 hours after angiography. At the time this study was commenced, there were no detailed prospective studies in the literature, although 1 has since been published.9 We prospectively evaluated clinical events occurring up to 72 hours after cerebral angiography in 1,002 consecutive procedures performed on 724 patients. Our purpose was to identify predictible risk factors and

From the Departments of Diagnostic Radiology (J.E.D., A.J.F., R.J.B.) and Clinical Neurological Sciences (P.C.G., A.J.F., H.J.M.B.), University Hospital and University of Western Ontario, London, Canada. Address for reprints: Dr. Allan J. Fox, Department of Diagnostic Radiology, University Hospital, PO Box 5339, Station A, London, Ontario, Canada N6A 5A5. Received March 27, 1987; accepted June 30, 1987.

to compare these with risk factors previously identified by other groups. As well, we recorded events occurring at 24-72 hours after angiography in an attempt to document either delayed cerebral ischemia following angiography or a trend in the patients' natural history of those diseases that had led to the diagnostic procedure. Subjects and Methods One thousand two consecutive procedures performed on 724 patients over a 15-month period from March 1983 to May 1984 were evaluated prospectively. Each patient had a neurologic examination performed by the same neurologist prior to, at 24 hours after, and at 72 hours after angiography. Patients with events were followed until recovery or stabilization. In 25.8% of the cases, part of the clinical information had to be retrospectively gathered from the chart records. This most commonly consisted of urgent cases where the preliminary neurologic examination was done by a different physician. Since these patients were being monitored closely, it is doubtful that significant changes in status were overlooked in the postangiography follow-up period. Records were made of age, sex, indication for angiography, medical history [stroke, transient ischemic attacks (TIAs), subarachnoid hemorrhage, myocardial infarction, angina, intermittent claudication, hypertension, seizures, headache, diabetes, smoking], medication (i.e., aspirin, aminocaproic acid, etc.), relevant neurologic symptoms and dates, abnormal neurologic findings (evaluated by the same neurologist both before and after angiography), pertinent laboratory values (blood urea nitrogen, creatinine, prothrombin time, partial thromboplastin time, hemoglobin, hema-


Stroke

998

Vol 18, No 6, November-December 1987

Table 1. Prior Studies of Neuroangiographlc Complications Authors (year) 10

N

Chynn (I960) Lang (1963)" Takahashi and Kawanani (1972)13 Vitek (1973)14 Olivccrona (1977)1 Mani et al (1978)4"6 Kerberetal (1978)3 Earnest et al (1983)9* This report*

436 11,402 500 2,000 5,531 5,000 662 1,517 1,002

Reversible 0-24 hrs

Permanent

Death

Other

4.9 3.6 1.6

0 0.1 0.1

0.4 0.06 0 0.1 0.03 0.02 0 0.07 0

— — — — 17.4 — — 6.8 6.9

1.5 4.4 0.9 1.0 2.3 1.2

0.2 0.1 0 0.33 0.1

Values in percent, except N, number of procedures. •Prospective studies.

tocrit). Blood pressure was recorded in both arms, and the patients were evaluated for the presence of cervical and other bruits. Peripheral pulses were examined and graded (0-3). Blood urea nitrogen and creatinine were studied 24 hours after angiography. Parameters recorded in the neuroangiography suite included premedication (type and dosage, route of administration); type of anesthesia (local, general); vessels catheterized, volume and number of injections per vessel; type and total amount of contrast used; administration of an intra-arterial heparin bolus (2,000 units); reversal of heparinization with protamine (usually 20 mg i.v.); total volumes of intravenous and/or intra-arterial fluid infused; number, size, and type of catheters and guidewires used; technical failures; duration of procedure (defined as the time from administration of xylocaine in the groin to the last injection of contrast material); and compression time. Overall, 109 pieces of clinical and technical data were recorded for each procedure and entered into a computer for analysis. The angiographic protocol was as follows: informed consent was obtained before angiography in all cases. Where possible, solids were withheld for a few hours prior to the procedure, but patients were kept well hydrated at all times. Most patients were mildly sedated with 10 mg diazepam orally. Local anesthesia with 7-10 ml of 1% xylocaine without epinephrine was used in 98.1% and general anesthesia in 1.9% of procedures. In 7 patients, the transfemoral route could not be employed; 6 direct carotid puncture and 1 brachial angiographies were performed. The initial catheter in all cases was 5-French polyethylene, heparin-coated, and shaped by the operator over steam. When this catheter was impossible to maneuver, one of the following were used: 5- or 6.5-French Torcon Blue H1H or H3H curved catheter, 5-French polyethylene HN4 curved catheter [Cook (Canada) Inc., Toronto, Canada]. Aortic arch angiograms were performed with 6.5-French pigtail catheters. The standard guidewire was 0.038 inch in diameter and heparin-coated. Jwires were used in the aortic arch and occasionally in the subclavian system; straight wires were used in the carotid, vertebral, and subclavian systems. In more

difficult cases, variable-stiffness or moveable-core guidewires were necessary to facilitate catheterization. Unless contraindicated by recent hemorrhage, bleeding diathesis, or recent surgery, an intra-arterial 2,000-unit bolus of heparin was given on entering the arterial system and was reversed with 20 mg protamine at the end of the procedure. A combination of careful double-flush technique and constant heparinized saline infusion (3,000 units/500 ml normal saline) was used. Our standard angiographic protocol was tailored to different types of suspected pathologic processes as follows: for TIA and/or stroke — anteroposterior and lateral views of both carotid arteries (head and neck), right posterior oblique view of the aortic arch; for aneurysms — anteroposterior, lateral, and oblique views of both carotid arteries (head), anteroposterior Towne and lateral views of 1 vertebral artery; for arteriovenous malformations (AVMs) — anteroposterior and lateral views of both carotid arteries, anteroposterior Towne and lateral views of 1 vertebral artery; and for post-operative — anteroposterior and lateral views of the artery or territory that was involved. Additional projections were added as appropriate. All injections were performed with a power injector. Conray 60 (Mallinkrodt Canada, Pointe-Claire, Canada) was employed in all cases except for aortic arch angiography, in which MD-76 (Mallinkrodt) was used. Standard injection rates and volumes were as follows: for common carotid artery, 8-10 ml/sec for 12 ml; for internal carotid artery, 8 ml/sec for 10 ml; for external carotid artery, 2-3 ml/sec for 5 ml; for vertebral artery, 7 ml/sec for 9 ml; for subclavian artery, 8 ml/sec for 16 ml with inflated arm cuff; and for aortic arch, 25 ml/sec for 50 ml. During this study, virtually all filming was done by film changer. Pressure was applied at the puncture site manually. All procedures were performed by senior radiology residents or neuroradiology fellows under the close supervision of a staff neuroradiologist, who became technically involved in more difficult cases. A neurologic event was defined as any neurologic sign or symptom occurring during the procedure or in the subsequent 72 hours, whether it was considered a manifestation of the primary disease or not, and was


Dion et al

Events Following Neuroangiography

defined as transient if it lasted < 1 week and had no permanent sequelae and as permanent if it lasted > 1 week or carried permanent sequelae. A nonneurologic event was defined as any sign or symptom occurring either locally at the puncture site or systemically (i.e., angina, shortness of breath, etc.). Neurologic events were recorded according to time, occurring either immediately or up to 30 minutes following the procedure, 30 minutes to 24 hours, 24—48 hours, and 48-72 hours. Events were also categorized as recurrence of the preangiographic symptoms, as worsening of a patient's condition, or as de novo occurrences. When hematomas occurred, they were classified as small ( < 5 cm in diameter), as moderate to large but not requiring therapy (5—10 cm), or as large (> 10 cm) requiring some form of therapy (correction of hypotension or surgery). Tests of validity were performed as follows: Student's t tests were used to evaluate the associations of continuous variables; a corrected x2 analysis was applied to assess the associations of discrete variables. We defined /?50 years of age. The following histories were present: smoking (44.8%), subarachnoid hemorrhage (35.6%), hypertension (29.3%), TIAs (19%), stroke (17.4%), seizures (13.1%), headaches (10.4%), angina (7.8%), diabetes (7.1%), myocardial infarction (6.6%), and claudication (4.6%). At the ti me of the procedure, 25.7 % of the patients were taking aspirin, 2.4% persantine, 0.1% sulfinpyrazone; 0.6% were receiving anticoagulant therapy and 8.6% aminocaproic acid therapy. Evaluation of TIA and/or stroke was the most common indication, followed closely by postoperative evaluation of aneurysms and preoperative evaluation of aneurysms and AVMs (Table 2). The relatively high proportion of subarachnoid hemorrhages, AVMs, and aneurysms is due to our referral pattern. Other indications such as vasculitis, assessment of endarterectomy and extracranial-intracranial bypass, amytal testing, asymptomatic bruits, and moyamoya disease accounted for 6.4%. Systolic hypertension (blood pressure >160 mm Hg) was present in 9.9% and diastolic hypertension (blood pressure >100 mm Hg) in 3.7% of our patients at the time of angiography; the presence of a carotid bruit was recorded in 15.3%. The preangiographic creatinine level was available in 962 procedures; it was normal (30 IU), but none had any clinical manifestations. Only 2 of the 27 patients in that group

999 had received >200 ml contrast material during the procedure. Catheterizations were performed on 2,438 arteries for a total of 4,176 injections. The distribution of catheterized arteries expressed as percent of the 1,002 angiographic procedures was as follows: right common carotid artery (67.7%), left common carotid artery (67.4%), left vertebral artery (45.3%), aortic arch (21.3%), right vertebral artery (12.7%), right internal carotid artery (6.2%), left internal carotid artery (6.1%), left subclavian artery (5%), right subclavian artery (4.6%), right external carotid artery (4.2%), and left external carotid artery (3.3%). Five hundred twelve patients had 1 angiogram, 160 had 2, 40 had 3, 10 had 4, and 2 had 5. In 42% of the procedures, 3 vessels were studied, while in 24 and 22.3%, respectively, 1 and 2 vessels were studied; 4—6 vessels were studied in 11.5% of the patients. Three patients underwent angiography but vessel catheterization was unsuccessful. In all procedures, a 5-French catheter was used as the initial catheter; in 278 procedures, a 6.5French catheter was also used. The mean number of catheters for the entire population was 1.4; only 1 catheter was used in 67.8% of the procedures, 2 in 24%, 3 in 5.7%, and 4 - 6 in 1.9%. Although 6 guidewires were used in 1 procedure, only 1 guidewire was used in 30% and 2 in 57.9% of the procedures. A heparin bolus (systemic heparinization) was given in 32.3% of procedures. The mean volume of contrast material used per angiogram was 93 ml, with a maximum of 408 ml in 1 instance; in only 5.2% of the procedures was >200 ml necessary. The mean duration of the procedure for the entire group was 50.2 minutes. Two percent of the procedures lasted > 2 hours, and 94% lasted 1 catheter, by using a 6.5-French rather than a 5-French catheter, and by the presence of atherosclerosis. Tables 2, 3, 4, and 5 summarize the event rate observed in this study according to indications for angiography, age, type of event (repeat, worsening of a preexisting status, or de novo), category of event (transient neurologic, permanent neurologic, or nonneurologic), and time of onset (during the procedure to 30 minutes after, 30 minutes to 24 hours, 24—48 hours, or 48-72 hours). Neurologic Events The neurologic event rate between 0 and 24 hours was 1.3%. Only 1 patient (0.1%) suffered a permanent neurologic event in this period; he was a 44-year-old man with a Grade IV subarachnoid hemorrhage (secondary to a ruptured posterior fossa AVM) who rebled 24 hours after angiography and had a brainstem stroke. There were no deaths (Table 2). There was no significant increase in neurologic events within the first 24 hours in patients who had been maintained on aminocaproic acid therapy prior to the procedure (8.9%). It is interesting to note that in the first 24 hours, slightly over one half of the neurologic events occurred in the


Stroke

1000


Table 2. Neurologic Events According to Indication and Angiographic Technical Data Neurologic % of Transient Permanent NonDuration . No. procedures Mean age Indication (N =1,002) (years) catheters (min) 0-24 hrs 24-72 hrs 0-24 hrs 24-72 hrs neurologic 60.4 1.4%' (4) TIA/stroke 28.4 57.7 0.7% (2) 1.9 2.5% (7) 14.0% 43.1 53.4 3.7% 1.2 0 0.9%> (2) — Known aneurysm and AVM 21.8 0.4% (1) 47.9 37.6 Postoperative aneurysm 1.3% (3) 0.9%» (2) 6.6% 22.6 1.3 — 43.9 51.7 0.8% (1) 2.5%> (3) 5.0% SAH (recent) 0.9'* (1) 1.3 12.0 — 3.4%' (3) 2.3% 33.7 44.4 Postoperative AVM 8.8 0 1.1 6.4 Others 0.2% 1.6% (1) 1.6%' (1) 46.1 50.3 1.3 — Total 47.6 100 7.2% 50.2 1.2% (12) 1.5%' (15) 1.4 0.1' 7o (1) 0.3% (3) TIA, transient ischemic attack; AVM, arteriovenous malformation; SAH, subarachnoid hemorrhage.

TIA group, which represents only 28.4% of the population studied (Table 3). However, after subjecting this to x2 analysis, we could demonstrate only that patients in this group showed a trend toward having more neurologic events in the first 24 hours (p = 0.09, x2)- None of the other groups were shown statistically to be at higher risk. Patients in the TIA group were significantly older (mean age 57.7 years), more catheters and guidewires were used during angiography, which lasted longer, when compared with the rest of the patients (Table 2). One may then deduce that this group could have suffered more events on the basis of the aforementioned factors; however, when identical statistical analysis was applied to the same factors within the TIA group, none significantly increased the likelihood of an event within 24 hours. It is thus possible that merely being

part of that group is the cause for the increased number of events. Of all the parameters studied (Table 3), the only 2 factors that significantly increased neurologic risk in the first 24 hours were procedure lasting >60 minutes (p = 0.03, x2) and systolic hypertension (p = 0.04, x2)Other factors that showed a trend included volume of contrast material used (p = 0.01, t test), use of S 3 catheters (/? = 0.08, x2)> and a preangiographic creatinine level of > 120IU (p = 0.08, x2)- In this last group, we observed a 4.1 % (3 patients) incidence of neurologic events between 0 and 24 hours compared with 1.0% for the rest of the group. The fact that we could show only a trend is probably due to the relatively small number of patients in this group. Elevated creatinine level was first described in 1983 by Earnest et al9 as a significant angiographic risk factor.

Table 3. Significant Factors and Trends Significant (p==0.05)

Type of event

Probability level

Neurologic 0-24 hrs

Duration >60 min Systolic hypertension

0.03 0.04

Neurologic 24-72 hrs

Contrast volume Age Diabetes mellitus Contrast volume TIA as indication Age &3 catheters 6.5-French Duration >60 min Carotid bruit Systolic + diastolic hypertension Diastolic hypertension Systolic hypertension Any bruit Number of guidewires used

0.001 0.02 0.03 0.0000 0.0000 0.0000 0.0000 0.000 0.002 0.006

Nonneurologic

Trend (0.0550), number of catheters and guidewires used, the use of 6.5-French catheters rather than 5-French, duration of procedure (>60 minutes), the presence of carotid bruits (or of any bruit in the body), and the presence of diastolic and systolic hypertension. Hematomas occurred in 6.9% of the procedures as illustrated in Table 6; there was a constant increase in their occurrence with advancing age, with approximately half of them seen in patients > 6 0 years of age. Eighteen percent of patients > 7 0 years of age suffered hematomas, and of note, one third of these needed therapy in the form of either fluid replacement for hypotension and/or surgery. Other Events Three additional patients (0.3%) had events other than neurologic or hematomas. One had an episode of angina that responded well to sublingual nitroglycerine; 2 patients had temporary paresthesia in the distribution of the right median cutaneous nerve of the thigh.

Neurologic Events According to Age Group

Age

group (yrs)

Table 5. Type of Neurologic Events Between 0 and 72 Hours Time of onset (hrs)

% Transient (n)

% (N) 0.8 (8) 4.8 (48) 11.8 (118) 14.4 (144) 15.4 (154) 26.4 (265) 20.9 (210) 5.5 (55) 100 (1,002)

% Permanent (n) 0-24 hrs 24-72 hrs 0-2'I hrs 24-72 hrs

0 (0) 0 (0) 0 (0) 1.4(2) 0.7(1) 1.9(5) 1.4(3) 1.8(1) 1.2(12)

0 (0) 0 (0) 0 (0) 2.1 (3) 1.3(2) 1.4(4) 1.9(4) 3.6 (2) 1.5(15)

0

0

0

0 (0) 0 (0)

(0) 0 (0) (0)

(0)

0.7 (1)

0

0 (0) 0 (0) 0 (0) 0.1 (1)

0.4 (1) 0 (0)

(0)

3.6 (2)

0.3 (3)

Table 6.

Age group (yrs)

0-29 30-39 40-49 50-59 60-69 70-79 Total

Hematomas

Small

(AO

n

%

(174) 4 2.3 (144) 3 2.1 (154) 4 2.6 (265) 12 4.5 (210) 21 10.0 7 12.7 (55) (1,002) 51 5.1

Moderateto-large n

Large with therapy

Total n %

%

n

0 0 1 0.7

0

0

4

0

0

4

2.8

4

2.6

I

0.6

9

5.8

3

1.1

1

0.4

5

2.4

0

0

0

0

3

5.5

1.3

5

0.5

13


%

2.3

6.0 12.4 26 10 18.2 69 6.9 16

1002

Stroke

Discussion Cerebral angiography is already established as a safe way of investigating the cerebral vascular tree, although it does carry a small, but real, risk that has been investigated by many authors.1-27 This risk is usually in the form of neurologic symptoms (transient or permanent), local complications (groin hematoma, distal limb ischemia), or systemic derangements (angina, renal failure, etc.). While the mechanisms and causes for the latter two categories are fairly limited, there exist many possible causes for the occurrence of neurologic symptoms secondary to cerebral angiography; these range from mechanical detachment of arterial wall plaques by the catheter or guidewire with distal embolization, to clot formation in or around the catheter as well as on guidewires, again with distal embolization, to platelet activation,2441-44 change in red blood cell morphology,454* and blood viscosity,47 to endothelial damage caused by contrast agents2248 and direct neurotoxic effect from contrast material49; as well, the presence of a catheter in a neck vessel is also capable of producing spasm, which may lead to neurologic symptoms. Although the custom is to assume that any event following angiography has been caused by it, the culprit may be the primary disease, which continues to manifest itself. Baum et al50 in 1966 published an interesting report on the complication of "no angiography" in which the clinical complication rate of patients scheduled for peripheral angiography but whose examination was subsequently cancelled was close to that of patients who had undergone angiography. The natural history of disease may account for a proportion of so-called complications. The present study was undertaken because of the lack of large-scale prospective studies on the risks of cerebral angiography. One has since been published.9 Prior studies consistently stopped monitoring patients for adverse effects that might have resulted from angiography 24 hours after the procedure if nothing abnormal had occurred. In our study, patients were followed for 72 hours after angiography to detect any possible delayed events. Our population contained a lesser proportion of patients with TIA and/or stroke as an indication for angiography when compared with other studies. Earnest et al9 reported a higher (but not significant) percent of reversible (3.6%) and permanent (0.6%) neurologic events at 24 hours in patients investigated for cerebrovascular disease when compared with their overall rates of 2.3 and 0.3%, respectively. In this study (Table 2), the same pattern was observed for reversible (2.5%) but not permanent (0%) events in the same type of patients when compared with our overall rates of 1.2 and 0.1%. An identical pattern persisted into the 2 4 72-hour observation period for both reversible (1.4%) and permanent (0.7%) events. Another point of difference with most other studies is that a full detailed neurologic examination was performed by a neurologist (and not a radiologist) before as well as 24 and 72 hours following angiography; it is


much less likely that subtle changes were missed. In most previous publications, complications were identified either by retrospectively reviewing the chart or after a radiology resident or staff member had examined the patient. Not every investigator considered worsening of the patient's status as a complication, although Earnest et al, 9 Faught et al,2 and Olivecrona' recorded events according to this method, avoiding biased judgments. Our reversible neurologic event rate of 1.2% within the first 24 hours compares favorably with that of other reports (Table 1), which range between 0.9% (Mani et al4"6) and 4.9% (Chynn10). The permanent neurologic event rate within 24 hours is surprisingly constant throughout the years and for different investigators, ranging between 0% (Chynn,10 Kerber et al3) and 0.33% (Earnest et al9); it is 0.1% in our report. The only factors that significantly increased the risk for neurologic events within 24 hours were duration of procedure >60 minutes and systolic hypertension, although we showed trends with increasing volume of contrast material, the use of ^ 3 catheters, TIA as an indication, and a preangiographic creatinine level >120 IU (Table 2). Most authors agree that patients with atherosclerosis are at increased risk. Faught et al2 and Earnest et al9 could not show an increased neurologic risk with hypertension. Mani et al4"6 had results similar to ours in that they demonstrated increased risk with duration >80 minutes, but Olivecrona1 did not find this factor significant. He showed more risk in the left carotid and right vertebral arteries compared with the right carotid and left vertebral; we could not show any pattern with vessel injected. Using smaller, softer 5-French catheters might logically decrease the rate of neurologic complications (Kerber et al ,3 Mani et al ,*-* Eisenberg et al 19M ), but we were not able to prove this relation. As well, we were unable to corroborate Faught et al2 who found women to be at significantly increased risk. Although Earnest et al9 showed that the use of > 1 catheter was a significant neurologic risk, we were not able to show an increase in risk until 3 catheters had been used. The same group was the first to identify an increased creatinine level ( > 120 IU) as a neurologic risk factor; in our study, we paralleled that finding by showing a trend (p = 0.08, x ); we suspect that, had we studied more procedures, we would have shown the same relation. Many factors (volume of contrast material, duration of procedure, TIA as an indication for procedure, age, number of catheters and guidewires used, size of catheters used, presence of a carotid or any vascular bruit, systolic and/or diastolic hypertension) were shown to significantly increase the likelihood of nonneurologic events occurring (Table 3). Adequate hydration may have played a role in reducing the number of patients with altered renal function studies (3.9%). Many of the measured parameters were intricately related; for example, a long procedure was likely to have resulted from more contrast material, the use of more catheters and guidewires, and was usually per-


Dion et al

1003

Events Following Neuroangiography

formed in older patients with atherosclerosis. The low frequency of complications precluded further multivariate analysis. Digital subtraction angiography,31"54 the use of 4-French catheters,55"37 the increased use of duplex ultrasonography,58 and new nonionic contrast media4959 may further decrease the dangers of cerebral angiography. Acknowledgments The authors wish to thank the following neurologists and neuroradiologists for their assistance: V. Hachinski, F. Vinuela, D.M. Pelz, W. Wong, T. Constantino, J.A. Rem, S. Spiegel, J.K. Chang, Y.H. Liu, and R. Ferguson, as well as Matthew Maynard for statistical analysis, and Sylvie Boulanger and Ann Hammond for secretarial assistance. References 1. Olivecrona H: Complications of cerebral angiography. Neuroradiology 1977; 14:175—181 2. Faught E, Trader SD, Hanna GR: Cerebral complications of angiography for transient ischemia and stroke: Prediction of risk. Neurology 1979;29:4-15 3. Kerber CW, Cromwell LD, Drayer BP, Bank WO: Cerebral ischemia. I. Current angiographic techniques, complications and safety. Am J Roentgenol 1978; 130:1097-1103 4. Mani RL, Eisenberg RL, McDonald EJ Jr, Pollock JA, Mani JR: Complications of catheter cerebral angiography: Analysis of 5000 procedures. I. Criteria and incidence. Am J Roentgenol 1978;131:861-865 5. Mani RL, Eisenberg RL: Complications of catheter cerebral angiography: Analysis of 5000 procedures. II. Relation of complication rates to clinical and arteriographic diagnoses. Am J Roentgenol 1978;131:867-869 6. Mani RL, Eisenberg RL: Complications of catheter cerebral angiography: Analysis of 5000 procedures. III. Assessment of arteries injected, contrast medium used, duration of procedure and age of patient. Am J Roentgenol 1978;131:871-874 7. Huclcman MS, Shenk GI, Neems RL, Tinor T: Transfemoral cerebral arteriography versus direct percutaneous carotid and brachial arteriography: A comparison of complication rates. Radiology 1979; 132:93-97 8. Miller JDR, Grace MG, Russell DB, Zacks DJ: Complications of cerebral angiography and pneumography. Radiology 1977; 124:741-744 9. Earnest F IV, Forbes G, Sandok BA, Piepgras DG, Faust RJ, Ilstrup DM, Arndt LT: Complications of cerebral angiography: Prospective assessment of risk. AJNR 1983;4:1191-1197 10. Chynn KY: Transfemoral carotid and vertebral angiography. Ada Radiol [Diagn] (Stockh) 1969;9:244-250 11. Lang EK: A survey of complications of percutaneous retrograde arteriography: Seldinger technic. Radiology 1963;81: 257-263 12. Wishart DL: Complications in vertebral angiography as compared to non vertebral cerebral angiography in 447 studies. Am J Roentgenol 1971 ;113:527-537 13. Takahashi M, Kawanani H: Complications of catheter cerebral angiography: An analysis of 500 examinations. Ada Radiol [Diagn] (Stockh) 1972;13:248-258 14. Vitek JJ: Femoro-cerebral angiography: Analysis of 2000 consecutive examinations, special emphasis on carotid arteries catheterization in older patients. Am J Roentgenol 1973; 118: 633-647 15. Muller R, Greitz I, Liliequist B, Hellstro'm L: Aortocervical angiography in occlusive cerebrovascular disease. Neurology 1964; 14:136-146 16. Sproul G, Pinto JM, Buster CD: Retrograde aortography: A screening method for cerebral vascular insufficiency. Am J Roentgenol Radium Ther Nucl Med 1965;95:837-844

17. Marshall J: Angiography in the investigation of ischaemic episodes in the territory of the internal carotid artery. Lancet 1971;1:719—721 18. Fox AJ: Cerebral angiography in TIA and stroke, in Bamett H, Paoletti P, Flamm E, Brambilla G (eds): Cerebrovascular Disease: New Trends in Surgical and Medical Aspects. New York, Elsevier North-Holland Biomedical Press, 1981, pp 107-121 19. Eisenberg RL, Bank WO, Hedgcock MW: Neurologic complications of angiography in patients with critical stenosis of the carotid artery. Neurology 1980;30:892-895 20. Eisenberg RL, Bank WO, Hedgcock MW: Neurologic complications of angiography for cerebrovascular disease. Neurology 1980;3O:895-897 21. Fisher M, Sandier R, Weiner JM: Delayed cerebral ischemia following arteriography. Stroke 1985;16:431-434 22. Broman T, Olsson O: Experimental study of contrast media for cerebral arteriography with reference to possible injurious effects on cerebral blood vessels. Ada Radiol 1949;31:321-334 23. Studdard W, Davis D, Young S: Cortical blindness after cerebral angiography. J Neurosurg 1981;54:240-244 24. Gawel M, Burlcett M, Rose FC: Platelet activation following cerebral angiography. Acta Neurol Scand 198O;61:240-243 25. Pexman JHW, Coates RK: Amnesia after femorocerebral angiography. AJNR 1983;4:979-983 26. Cochran JW, Morrell F, Huckman MS, Cochran EJ: Transient global amnesia after cerebral angiography (abstract). Radiology 1983;147:905 27. Wales LR, Nov AA: Transient global amnesia: Complication of cerebral angiography. AJNR 1981;2:275-277 28. Cramer R, Moore R, Amplatz K: Reduction of the surgical complication rate by the use of a hypothrombogenic catheter coating. Radiology 1973;109:585-588 29. Eldh P, Jacobsson B: Heparinized vascular catheters: Clinical trial. Radiology 1974;111:289-292 30. Mani RL: Computer analysis of factors associated with thrombus formation observed on pullout angiograms. Invest Radiol 1975;10:378-384 31. Amplatz K: A simple nonthrombogenic curve coating. Invest Radiol 1977;6:28O-289 32. Hawkins IF, Kelley MJ: Benzalkonium-heparin-coated angiographic catheters, experience with 563 patients. Radiology 1973;lO9:589-591 33. Durst S, Leslie J, Moore R, Amplatz K: A comparison of the thrombogenicity of commercially available catheters. Radiology 1974;113:599-600 34. Anderson JH, Gianturco C, Wallace S, Dodd GD: A scanning electron microscopic study of angiographic catheters and guidewires. Radiology 1974;111:567-571 35. Ovitt R, Durst C, Amplatz K: Guidewire thrombogenicity and its reduction. Radiology 1974;111:43—46 36. Antonovic R, Rosch J, Dorter C: The value of systemic arterial heparinization in transfemoral angiography: A prospective study. Am J Roentgenol 1976; 127:223-225 37. Debrun GM, Vinuela FV, Fox AJ: Aspirin and systemic heparinization in diagnostic and interventional neuroradiology. AJNR 1982;3:337-34O 38. Bookstein JJ, Moser KM, Hovgic C: Coagulative interventions during angiography. Cardiovasc Intervent Radiol 1982;5:4656 39. Anderson J, Gianturco C: Anticoagulation techniques for angiography. An experimental study. Radiology 1974;111:573576 40. Wallace S, Medellin H, De Jongh D, Gianturco C: Systemic heparinization for angiography. AJR 1972; 116:204-209 41. Dougherty J, Levy D, Weksler B: Platelet activation in acute cerebral ischemia: Serial measurements of platelet function in cerebrovascular disease. Lancet 1977;l:821-824 42. Stewart M, Douglas J, Lower G, Prentice C, Forbes C: Prognostic value of beta thrombo-globulin in patients with transient cerebral ischemia. Lancet 1983;2:475-482 43. O'Connor MK, Brennan SS, Shanik DG: Indium-111 labeled platelet deposition following transfemoral angiography. Radiology 1986;158:191-194 44. Moore S, Pepper D, Cash J: The isolation and characterization of platelet-specific /3-globulin (/3-thromboglobulin) and the de-


1004

45. 46. 47. 48. 49.

50. 51. 52. 53.

Stroke

tection of anti-urokinase and anti-plasmin released from thrombin-aggregated washed human platelets. Biochim BiophysActa 1975;379:360-369 Aspelin P, Schmid-Schonbein H: Effect of radio-contrast media on the red cell. Blood Cells 1977;3:397-407 Aspelin P: Effect of ionic and non-ionic contrast media on morphology of human erythrocytes. Ada Radiol [Diagn] (Stockh) 1978;19:675-687 Aspelin P: Effect of ionic and non-ionic contrast media on whole blood viscosity, plasma viscosity, and hematocrit in vitro. Acta Radiol (Diagn] (Stockh) 1978;19:977-989 Laerum F: Injurious effects of contrast media on vascular endothelium. Invest Radiol 1985;20(suppl):598-599 Drayer BP, Velaj R, Bird R, Albright R, Roberts L, Allen S, Triolo P, Yeates A, Ross M, Bates M: Comparative safety of intracarotid iopamidol, iothalamate meglumine, and diatroate meglumine for cerebral angiography. Invest Radiol 1984; 19(suppl):212-218 Baum S, Stein GN, Kuroda KK: Complications of "no arteriography." Radiology 1966;86:835-838 Hoffman MG, Gomes AS, Pais SO: Limitations in the interpretation of intravenous digital subtraction angiography. AJNR 1983;4:1167-1170 Eggers FM, Price AC, Allen JH, James AE: Neuroradiologic applications of intraarterial digital subtraction angiography. AJNR 1983;4:854-856 Brant-Zawadzki M, Gould R, Norman D, Newton TH, Lane B: Digital subtraction cerebral angiography by intraarterial in-

54.

55. 56. 57. 58.

59.


jection: Comparison with conventional angiography. AJNR 1982;3:593-599 Zimmerman RD, Goldman JM, Auster M, Chen C, Leeds N: Aortic arch digital arteriography: An alternative technique to digital venous angiography and routine arteriography in the evaluation of cerebrovascular insufficiency. AJNR 1983;4: 266-270 Wolfel DA, Lovett BP, Ortenburger AI, Johnson LS, Sommerville DL: Outpatient arteriography: Its safety and cost effectiveness. Radiology 1984;153:363-364 Saint-Georges G, Aubd M: Safety of outpatient angiography: Prospective study. AJR 1985; 144:235-236 McCreary JA, Schellhas KP, Brant-Zawadzki M, Norman D, Newton TH: Outpatient DSA in cerebrovascular disease using transbrachial arch injections. AJNR 1985;6:795-801 Wolverson MK, Heiberg E, Tantana S, Pilla TJ: Intravenous DSA and duplex sonography as screening examinations for carotid disease: Comparison in 102 vessels. AJNR 1985;6:569574 Kido DK, Potts DG, Bryan RN, Norman D, PelzDM, Fox AJ, Shaw DD: Iohexol cerebral angiography: Multicenter clinical trial. Invest Radiol 1985;20(suppl):55-57

KEY WORDS • cerebral angiography cerebral angiography


• complications of