Predictors of Intracranial Carotid
Artery Atherosclerosis
Duration of Cigarette Smoking and Hypertension Are More Powerful Than Serum Lipid Levels Timothy J. Ingall, MBBS, PhD; Daniel Homer, MD; Hillier L. Baker, Jr, MD; Bruce A. Kottke, MD, PhD; W. Michael O'Fallon,
PhD; Jack P. Whisnant, MD
\s=b\ The effect of cigarette smoking on intracranial internal carotid artery atherosclerosis (ICAS) was studied by obtaining cigarette smoking histories and data on other potential predictors, including serum lipid estimations, for consecutive patients undergoing carotid arteriography. The duration of cigarette smoking was the most significant independent predictor of the presence of ICAS. Other independently significant predictors of ICAS were hypertension, diabetes mellitus, and current systolic blood pressure. The interaction of diabetes and duration of smok-
ing was patients
significant negative predictor. In for whom serum lipid values were available, lower levels of apolipoprotein A-I were associated with a higher risk of having ICAS. However, the effect of apolipoprotein A-I as a predictor of the presence of ICAS was far outweighed by the effects of duration of smoking and hypertension. (Arch Neurol. 1991 ;48:687-691) a
Tntracranial internal carotid artery atherosclerosis (ICAS) occurs less frequently than extracranial carotid atherosclerosis.1 Although these lesions Accepted for publication January31,1991. From the Departments of Health Sciences Research (Drs Ingall, O'Fallon, and Whisnant), Neurology (Drs Homer and Whisnant), Diagnostic Radiology (Dr Baker), and the Division of Cardiovascular Diseases and Internal Medicine (Dr Kottke), Mayo Clinic and Mayo Foundation, Rochester, Minn. Reprint requests to Mayo Clinic, 200 First St SW, Rochester, MN 55905 (Dr Whisnant).
related to the development of isch¬ emie cerebrovascular disease, their rel¬ ative contribution to the occurrence of stroke is unclear.2 There have been few studies of the potential risk factors for intracranial atherosclerosis3"7 and only two studies have assessed multiple risk factors in a multivariate analysis.3,4 The aim of this study was to deter¬ mine which risk factors were related to the presence of arteriographically prov¬ en ICAS by using a multivariate analy¬ sis that adjusted for possible confound¬ ing effects and interactions among the risk factors. are
PATIENTS AND METHODS Data on potential risk factors for athero¬ sclerosis were collected prospectively for 1004 patients ranging from 40 to 69 years of age who underwent carotid arteriography at the Mayo Clinic (Rochester, Minn) for any indication from 1983 through 1986. Although atherosclerotic cerebrovascular disease is the most frequent indication for carotid arte¬ riography at the Mayo Clinic, the bias toward atherosclerosis was decreased by including patients who had carotid arteriography for other reasons such as subarachnoid hemor¬ rhage or central nervous system mass lesions. A detailed history of potential vascular risk factors and conditions known to be asso¬ ciated with atherosclerosis was obtained from each patient and from the medical record at the time of the evaluation prior to arteriography. These variables are noted in Table 1. For cigarette smoking, patients were classified as never smokers, current smokers, or former smokers if they had
Downloaded From: http://archneur.jamanetwork.com/ by a University of Victoria User on 06/05/2015
ceased smoking for more than 6 months. In¬ formation on duration of smoking (years), average daily number of cigarettes, and pack-years of smoking was obtained for cur¬ rent and former smokers. The years since cessation of smoking were recorded for for¬ mer smokers. The protocol for the study was approved by the Mayo Institutional Review Board. Among the 1004 patients entered into the study, adequate views of at least one intra¬ cranial internal carotid artery were available for 1000. In the multivariate assessment, only those variables believed to be related directly to the presence or absence of athero¬ sclerosis were included—ie, age, sex, cur¬ rent smoking status, years of smoking, num¬ ber of cigarettes per day, pack-years, years since quitting, history of hypertension, sys¬ tolic and diastolic blood pressure, history of diabetes, fasting blood glucose, alcohol use, total cholesterol, triglycérides, high-density lipoprotein cholesterol, calculated low-densi¬
ty lipoprotein cholesterol, apolipoprotein -I, high-density lipoprotein cholesterohtotal cholesterol ratio, family history of stroke,
history of cancer, and physical activity. Serum lipid and lipoprotein were mea¬ sured at the time of entry of patients into the study but were performed only in the latter part of a larger study, so the patients were not selected randomly. The 302 patients who had lipid estimations and adequate views of the intracranial carotid circulation were com¬ pared with the 698 patients whose intracrani¬ al carotid arteries were visualized but who did not have lipid estimations. Comparisons were made of age, sex, current smoking sta¬ tus, years of smoking, number of cigarettes per day, pack-years, years since quitting, history of hypertension, systolic blood pres¬ sure, history of diabetes, and fasting blood
Table 2.—Univariate Logistic Regression Analyses Adjusted for Age and Sex*
Table 1.—Variables Assessed in the
Study Age Sex
Cigarette smoking History of smoking (current / former / never) Duration of smoking, y No. of cigarettes per day Pack-years No. of years since quitting Hypertension Diabetes
Alcohol
use
Physical activity Transient ischemie attack Reversible ischemie neurologic deficit History of stroke Peripheral vascular disease
Variable Duration of smoking Current smoking status
dr
!
1
75.74
Artery Atherosclerosis
Duration of Cigarette Smoking and Hypertension Are More Powerful Than Serum Lipid Levels Timothy J. Ingall, MBBS, PhD; Daniel Homer, MD; Hillier L. Baker, Jr, MD; Bruce A. Kottke, MD, PhD; W. Michael O'Fallon,
PhD; Jack P. Whisnant, MD
\s=b\ The effect of cigarette smoking on intracranial internal carotid artery atherosclerosis (ICAS) was studied by obtaining cigarette smoking histories and data on other potential predictors, including serum lipid estimations, for consecutive patients undergoing carotid arteriography. The duration of cigarette smoking was the most significant independent predictor of the presence of ICAS. Other independently significant predictors of ICAS were hypertension, diabetes mellitus, and current systolic blood pressure. The interaction of diabetes and duration of smok-
ing was patients
significant negative predictor. In for whom serum lipid values were available, lower levels of apolipoprotein A-I were associated with a higher risk of having ICAS. However, the effect of apolipoprotein A-I as a predictor of the presence of ICAS was far outweighed by the effects of duration of smoking and hypertension. (Arch Neurol. 1991 ;48:687-691) a
Tntracranial internal carotid artery atherosclerosis (ICAS) occurs less frequently than extracranial carotid atherosclerosis.1 Although these lesions Accepted for publication January31,1991. From the Departments of Health Sciences Research (Drs Ingall, O'Fallon, and Whisnant), Neurology (Drs Homer and Whisnant), Diagnostic Radiology (Dr Baker), and the Division of Cardiovascular Diseases and Internal Medicine (Dr Kottke), Mayo Clinic and Mayo Foundation, Rochester, Minn. Reprint requests to Mayo Clinic, 200 First St SW, Rochester, MN 55905 (Dr Whisnant).
related to the development of isch¬ emie cerebrovascular disease, their rel¬ ative contribution to the occurrence of stroke is unclear.2 There have been few studies of the potential risk factors for intracranial atherosclerosis3"7 and only two studies have assessed multiple risk factors in a multivariate analysis.3,4 The aim of this study was to deter¬ mine which risk factors were related to the presence of arteriographically prov¬ en ICAS by using a multivariate analy¬ sis that adjusted for possible confound¬ ing effects and interactions among the risk factors. are
PATIENTS AND METHODS Data on potential risk factors for athero¬ sclerosis were collected prospectively for 1004 patients ranging from 40 to 69 years of age who underwent carotid arteriography at the Mayo Clinic (Rochester, Minn) for any indication from 1983 through 1986. Although atherosclerotic cerebrovascular disease is the most frequent indication for carotid arte¬ riography at the Mayo Clinic, the bias toward atherosclerosis was decreased by including patients who had carotid arteriography for other reasons such as subarachnoid hemor¬ rhage or central nervous system mass lesions. A detailed history of potential vascular risk factors and conditions known to be asso¬ ciated with atherosclerosis was obtained from each patient and from the medical record at the time of the evaluation prior to arteriography. These variables are noted in Table 1. For cigarette smoking, patients were classified as never smokers, current smokers, or former smokers if they had
Downloaded From: http://archneur.jamanetwork.com/ by a University of Victoria User on 06/05/2015
ceased smoking for more than 6 months. In¬ formation on duration of smoking (years), average daily number of cigarettes, and pack-years of smoking was obtained for cur¬ rent and former smokers. The years since cessation of smoking were recorded for for¬ mer smokers. The protocol for the study was approved by the Mayo Institutional Review Board. Among the 1004 patients entered into the study, adequate views of at least one intra¬ cranial internal carotid artery were available for 1000. In the multivariate assessment, only those variables believed to be related directly to the presence or absence of athero¬ sclerosis were included—ie, age, sex, cur¬ rent smoking status, years of smoking, num¬ ber of cigarettes per day, pack-years, years since quitting, history of hypertension, sys¬ tolic and diastolic blood pressure, history of diabetes, fasting blood glucose, alcohol use, total cholesterol, triglycérides, high-density lipoprotein cholesterol, calculated low-densi¬
ty lipoprotein cholesterol, apolipoprotein -I, high-density lipoprotein cholesterohtotal cholesterol ratio, family history of stroke,
history of cancer, and physical activity. Serum lipid and lipoprotein were mea¬ sured at the time of entry of patients into the study but were performed only in the latter part of a larger study, so the patients were not selected randomly. The 302 patients who had lipid estimations and adequate views of the intracranial carotid circulation were com¬ pared with the 698 patients whose intracrani¬ al carotid arteries were visualized but who did not have lipid estimations. Comparisons were made of age, sex, current smoking sta¬ tus, years of smoking, number of cigarettes per day, pack-years, years since quitting, history of hypertension, systolic blood pres¬ sure, history of diabetes, and fasting blood
Table 2.—Univariate Logistic Regression Analyses Adjusted for Age and Sex*
Table 1.—Variables Assessed in the
Study Age Sex
Cigarette smoking History of smoking (current / former / never) Duration of smoking, y No. of cigarettes per day Pack-years No. of years since quitting Hypertension Diabetes
Alcohol
use
Physical activity Transient ischemie attack Reversible ischemie neurologic deficit History of stroke Peripheral vascular disease
Variable Duration of smoking Current smoking status
dr
!
1
75.74
