Smoking and Risk of Ischemic Stroke in Young Men

Abstract

Background and Purpose

There is a strong dose-response relationship between smoking and risk of ischemic stroke in young women, but there are few data examining this association in young men. We examined the dose-response relationship between the quantity of cigarettes smoked and the odds of developing an ischemic stroke in men under age 50.

Methods

The Stroke Prevention in Young Men Study is a population-based case-control study of risk factors for ischemic stroke in men ages 15-49. The chi-square test was used to test categorical comparisons. Logistic regression models were used to calculate the odds ratio for ischemic stroke occurrence comparing current and former smokers to never smokers. In the first model, we adjusted solely for age. In the second model, we adjusted for potential confounding factors, including age, race, education, hypertension, myocardial infarction, angina, diabetes mellitus and body mass index.

Results

The study population consisted of 615 cases and 530 controls. The odds ratio for the current smoking group compared to never smokers was 1.88. Furthermore, when the current smoking group was stratified by number of cigarettes smoked, there was a dose-response relationship for the odds ratio, ranging from 1.46 for those smoking fewer than 11 cigarettes per day to 5.66 for those smoking 40+ cigarettes per day.

Conclusions

We found a strong dose-response relationship between the number of cigarettes smoked daily and ischemic stroke among young men. While complete smoking cessation is the goal, even smoking fewer cigarettes may reduce the risk of ischemic stroke in young men.

Introduction

Incidence of ischemic stroke (IS) in young adults is increasing¹. Additionally, cigarette smoking, a modifiable risk factor for IS, has been on the rise among young adults^1–4. Our prior research among young women suggests a strong dose-response relationship between smoking and risk of IS⁵, but there are few studies examining this association in young men. Because of potential interactions between smoking and hormonal milieu, a separate examination of this issue in men is important. The extension of our study to men allows us to address this issue. In this study, we examine the dose-response relationship between the quantity of cigarettes smoked daily and the odds of developing an IS in men under age 50.

Methods

In order to minimize the possibility of unintentionally sharing information that can be used to re-identify private information, a subset of the data generated for this study is available at dbGaP and can be accessed at https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000292.v1.p1.

The Stroke Prevention in Young Men Study is a population-based case-control study of risk factors for IS in men ages 15-49. All cases were recruited within three years of stroke between 2003 and 2007. Controls were identified by random-digit dialing and were balanced with cases for geographic region of residence, age within five years, and ethnicity. Never smokers were defined as those who had not smoked greater than 100 cigarettes or 5 packs in their lifetime. Current smokers were defined as those who had smoked greater than 100 cigarettes in their lifetime and also had smoked in the 30 days preceding their stroke (for cases) or interview (for controls). Former smokers were defined as those who had smoked greater than 100 cigarettes in their lifetime, but had not smoked in the 30 days before their stroke/interview. Standardized interviews were conducted to collect data on smoking and other vascular risk factors. Further details on study methods have been published⁵.

The study was approved by the University of Maryland at Baltimore Institutional Review Board and all participants gave informed consent.

Statistical Methodology

Statistical analysis was conducted using SAS version 9.4. The chi-square test was used to test categorical comparisons. Logistic regression models were used to calculate the odds ratio for IS occurrence comparing smokers to never smokers. In the first model, we adjusted solely for age. In the second, fully-adjusted, model we adjusted for potential confounding factors, including age, race, education, hypertension, myocardial infarction, angina, diabetes mellitus and body mass index. Interaction terms were used to examine potential effect modification by other risk factors. Statistical tests were 2-tailed and P-values less than 0.05 were considered statistically significant.

Results

Among the participants, 615 out of 625 cases and 530 out of 537 controls had complete data for all covariates, leaving a final study population of 1145 subjects. Cases were less educated and were more likely to have hypertension, diabetes mellitus, myocardial infarction, angina and obesity (body mass index > 30) (all P<0.05). Among controls, current smokers were less educated and were more likely to be black than their non-smoking counterparts (all P<0.05) (Table 1).

Table 1.

Demographic Characteristics of Participants

Factor	Category	Cases n (%) (n=615)	Controls n (%) (n=530)	P†	Current Smokers* n (%) (n=160)	Former and Never Smokers* n (%) (n=370)	P†
								Age	<18	2 (0.3)	1 (0.2)	0.01	0 (0)	1 (0.3)	0.22
	18-24	11 (1.8)	10 (1.9)		6 (3.8)	4 (1.1)
	25-34	45 (7.3)	72 (13.6)		18 (11.2)	54 (14.5)
	35-49	557 (90.6)	447 (84.3)		136 (85.0)	311 (84.1)
Race	White	333 (54.2)	305 (57.6)	0.27	68 (42.5)	237 (64.1)	<0.001
	Black	258 (42.0)	199 (37.5)		86 (53.8)	113 (30.5)
	Other	24 (3.8)	26 (4.9)		6 (3.7)	20 (5.4)
Education	<12	96 (15.6)	45 (8.5)	<0.001	28 (17.5)	17 (4.6)	<0.001
	≥12	519 (84.4)	485 (91.5)		132 (82.5)	353 (95.4)
Hypertension	Yes	288 (46.8)	121 (22.8)	<0.001	41 (25.6)	80 (21.6)	0.31
Diabetes Mellitus	Yes	115 (18.7)	30 (5.7)	<0.001	10 (6.2)	20 (5.4)	0.70
Myocardial Infarction	Yes	38 (6.2)	6 (1.1)	<0.001	4 (2.5)	2 (.5)	0.05
Angina	Yes	56 (9.1)	27 (5.1)	0.01	14 (8.8)	13 (3.5)	0.12
Body Mass Index	<30	365 (59.4)	368 (69.4)	<0.001	120 (75.0)	248 (67.0)	0.07
	≥30	250 (40.6)	162 (30.6)		40 (25.0)	122 (33.0)

^ˆControls only

^†Chi-square

Table 2 shows that in the age-adjusted model, the odds ratio for the current smoking group compared to never smokers was 1.88 (95% CI-1.44-2.44). Furthermore, when the current smoking group was stratified by number of cigarettes smoked, there was a dose-response relationship for the odds ratio, ranging from 1.46 (95% CI-1.04-2.06) for those smoking less than 11 cigarettes per day to 5.66 (95% CI- 2.14-14.95) for those smoking 40+ cigarettes per day. In the fully-adjusted model, there is a similar dose response observed but with slightly lower ORs. In the age-adjusted model, the odds ratio for the former smoking group compared to never smokers was 1.42 (95% CI- 1.01-1.99), with similar results for the fully-adjusted model. (Table 2).

Table 2.

Odds ratio for Ischemic Stroke by Smoking Status

	Cases (n=615)	Controls (n=530)	Model 1*		Model 2†
			OR	95% CI	OR	95% CI
Never Smokers	239	286	…	…	…	…
Former Smokers	108	84	1.42	1.01-1.99	1.33	0.93-1.90
Current Smokers	268	160	1.88	1.44-2.44	1.63	1.21-2.19
Current Smokers by Number of Cigarettes Smoked Daily
1-10	103	81	1.46	1.041-2.06	1.21	0.83-1.77
11-20	97	64	1.74	1.21-2.49	1.64	1.10-2.43
21-39	40	10	4.29	2.09-8.80	3.51	1.65-7.45
40+	28	5	5.66	2.14-14.95	5.24	1.90-14.42

^*adjusted for age

^†adjusted for age, race, education, hypertension, myocardial infarction, angina, diabetes mellitus and body mass index

Discussion

Our study demonstrates a strong dose-response between amount of cigarettes smoked and risk of IS in young men. There is evidence for a dose-response relationship between cigarette smoking and risk of stroke in middle-aged and older adults as well, however the association is less strong⁶. Our finding is consistent with results from prior studies of women and young adults in general^2,5,7. Our earlier study among women similarly demonstrated a strong dose-response relationship between current cigarette smoking and IS risk⁵. A study of young adults in Iowa demonstrated a dose-response between amount of current cigarette smoking and IS risk but did not stratify their findings by sex⁶. These studies demonstrate that smoking amount is an important risk factor for IS, but do not characterize the dose-response relationship in young male smokers.

Our study builds on this research by focusing on young men. Our results are in line with the findings of a Swedish study of young men that analyzed smoking history among military recruits ages 18-20 as a predictor of IS before age 45⁸. Compared to the Swedish study, a strength of our study is that it includes a more ethnically diverse population and is adjusted for education.

Our study also has several limitations. Since we did not record the use of other tobacco products we cannot exclude the possibility that concurrent use of these products could have affected our results. Similarly, we did not control for factors such as alcohol consumption and physical activity in our model, which may have resulted in unmeasured or residual confounding of our risk estimates. Another limitation of our study is the case-control design, which allows for the possibility of differential recall bias by case-control status. However, the similar findings in the Swedish study, derived from a cohort design, suggest that there was not a major effect from differential recall bias.

Smoking rates among patients hospitalized for IS have been increasing¹. The clinical implications of our finding are, that while complete cessation of smoking is the goal, even reducing the number of cigarettes smoked may have beneficial health effects.