The Misleading "Smoker's Paradox" in Young Stroke

Objective: Stroke in young adults is uncommon, and the etiologies and risk factors of stroke in young adults differ from those in the older populations. Smoker’s paradox is an unexpected favorable outcome, and age difference was used to explain the association between smoking and the favorable functional outcome. This study aimed to investigate the existence of this phenomenon in young stroke patients. Methods: We analyzed a total of 9,460 young stroke cases registered in the nationwide stroke registry system of Taiwan between 2006 and 2016. Smoking criteria included having a past or current history of smoking more than 1 cigarette per day for more than 6 months. After matching for sex and age, a Cox model was used to compare complications, mortality and outcomes between smokers and non-smokers. Results: Smoking was associated with older age, higher comorbidities, and higher alcohol consumption. Smoking patients with NIHSS scores of 11–15 had a worse functional outcome (adjusted OR, 0.81; 95% CI, 0.76–0.87), and smoking cessation would substantially reverse those effects. Conclusion: The smoker’s paradox denitely does not exist, and therefore we continue to strongly advocate the importance of smoking cessation.


Introduction
Stroke in young adults is relatively uncommon. The Follow-Up of Transient Ischemic Attack and Stroke Patients and Unelucidated Risk Factor Evaluation (FUTURE) study revealed that only 10% of all strokes occurred in patients aged 18-50 years [1,2]. Most strokes can be explained by hypertension, diabetes, hyperlipidemia, obesity, atrial brillation, and smoking. However, stroke is far more common in the geriatric population, whereas in the young it is generally assumed to differ in risk factors and pathogenesis [3,4]. Worldwide, more than two million young adults suffer from stroke yearly. Although the prognosis of young stroke is generally considered benign, young adults with stroke are at a higher risk of recurrent stroke and mortality than their healthy peers. Most survivors between 20 and 50 years may have emotional, social, or physical sequelae that impair their quality of life [4,5]. In addition, young stroke victims are often responsible for providing child care or generating income for their families. Therefore, young stroke is a major health and socioeconomic problem.
Smoking is one of the biggest public health threats associated with many chronic diseases such as cardiovascular disease, and cancer [6-8]. More importantly, smoking is also aggregated with other adverse behaviors such as bet-but chewing and alcohol drinking. However, several recent studies observed an improved outcome in smokers with antithrombotic therapy after an index cardiovascular event, a phenomenon called "smoker's paradox" [9][10][11][12][13]. These studies are limited by short observation periods and small sample sizes [10,11]. Moreover, no study has examined the effects of pre-stroke smoking on post-stroke complications that can prolong the patients' recovery and reduce their quality of life.
The purpose of this study was to evaluate whether smoking status was an independent prognostic factor in young stroke patients. In addition, we compared the in uence of smoking on the stroke outcome in different clinical settings, in terms of gender and stroke subtypes.

Standard protocol approvals, registrations, and patient consents
The Taiwan Stroke Registry (TSR) is the rst large nationwide stroke registry in Taiwan and is sponsored by the Taiwan Department of Health [14]. Details on the database's generation, monitoring, and maintenance are published by the Taiwan Stroke Society. [9][10][11] The data that support the ndings of this study are available from Taiwan Stroke Registry but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. All researchers can submit their proposals to the research committee of the TSR (taiwanstrokeregistry@gmail.com). After study proposals are accepted, the results will be sent back to investigators. In this study, ethical approval was granted from China Medical University and the Institutional Review Boards (IRB) of the collaborating hospitals (CMUH104-REC2-115).
The TSR enrolls patients who are presented within 10 days of symptom onset to a TSR hospital because of 1 of the 4 major stroke types (i.e., ischemic stroke, intracerebral hemorrhage, subarachnoid hemorrhage, and transient ischemic attack). The TSR program was launched in 2006, and more than 100,000 stroke events had been recorded in the TSR up to 2018 [15][16][17]. Initially, 39 hospitals across the country participated in the registration project, and study protocols were approved by the Institutional Review Board of each participating hospital. Each stroke patient who signed the informed consent was followed up by the case managers of each hospital through their medical records and/or telephone visits every 3 months for at least 12 months after discharge. The severity of stroke was assessed using the National Institutes of Health Stroke Scale (NIHSS), and the outcome was determined using the modi ed Rankin Scale (mRS) by trained neurologists. The data entry was performed by study nurses (http://stroke.cmuh.org.tw/). The Taiwan Stroke Registry Investigators are listed in Appendix S1.

Study design and eligibility criteria
The study comprised all consecutive patients aged between 20 and 50 years who were diagnosed with any type of stroke from August 1, 2006 to May 20, 2016. Patient characteristics relevant to acute stroke, including stroke type, neurological de cit severity de ned by the National Institute of Health Stroke Scale (NIHSS), medical history, pre-existing comorbidities, imaging, in-hospital management and complications, and functional outcomes, were collected according to a prede ned system. Smokers were de ned as those with a past or current history of smoking more than one cigarette per day for more than six months. Meanwhile, former smokers were those who stopped smoking over six months before stroke. In total, 9,460 young stroke patients were included in this cohort study.

Main outcome measures and statistical analysis
The primary outcomes included in-hospital mortality and the functional outcome measured by the modi ed Rankin Scale (mRS). Follow-up evaluations were conducted at 3 months, and outcome events were classi ed by using information from interviews (directly during follow-up visits or via telephone) with patients, or from hospital records. All descriptive data are expressed as numbers (N) and percentages (%) of patients. Hazard ratios (HRs) and 95% con dence intervals (CIs) were assessed by Cox proportional hazards models in univariate analyses to compare demographic variables and risk factor prevalence at baseline, and in strati ed multivariable analyses to detect the independent predictors of mortality. Then, we used logistic regression analysis to calculate the odds ratios (ORs) for the evaluation of potential factors associated with discharging stroke patients for the poor functional outcome (mRS score of [3][4][5] in the derivation group. Statistical signi cance was considered at a p value of < 0.05.

Baseline characteristics of young stroke patients
In the TSR, 9,460 young stroke patients were identi ed, including 4,784 smokers and 4,676 non-smokers.
Detailed demographic and clinical characteristics for this cohort are presented in Table 1. The smoker group had an older age and a higher stroke severity (NIHSS on admission) than the non-smoker group (Table 1). Compared with the non-smoker group, the smoker group had higher proportions of certain stroke risk factors, like hypertension, diabetes, total cholesterol, previous stroke, and alcohol consumption. Smoking and mortality in young stroke patients  (Table 4). However, no signi cant association was observed between smoking categories and mortality due to stroke in the multivariate Cox regression model.   Smoking and the functional outcome in young stroke patients Table 5 shows the crude and adjusted odds ratios (ORs) for the poor functional outcome (with the modi ed Rankin Scale score of 3-5) and the NIHSS score between the two cohorts. For NIHSS scores ranging between 16 and 20, adjusted ORs were only signi cant in current smokers (OR, 2.75, 95% CI, 1.12-6.77, respectively).

Discussion
In our community-based longitudinal study in young stroke, a total of 9,460 (10.63%) strokes were identi ed among patients aged ≤ 50 years. We observed a gender difference between smokers and nonsmokers.
Smokers were associated with the higher severity of the initial stroke (NIHSS score upon admission) and more ischemic stroke. In addition, modi able risk factors are prevalent in smokers, including hypertension (66.41%), diabetes (23.59%), hypercholesterolemia (31.52%), previous stroke (11.50%), and alcohol consumption (42.35%). According to the strati ed analysis, smoking was an independent predictor of the functional outcome recorded at 3 months after stroke. However, only lower crude ratios for mortality, not adjusted hazard ratios, were observed in the smoker group. Due to the poorer outcome in the smoker group, the smoking paradox does not exist.
Approximately 10% of strokes occur at ages ≤ 50 years, and the proportion of strokes in young adults has increased over time [3,4]. Despite their more favorable stroke outcome, younger adult patients still have an obvious socioeconomic consequence because a large proportion of them are at a higher risk of future cardiovascular events and labor productivity loss. Besides, the burden of disease is heavier in the case of recurrent events. Stroke is far more common in the geriatric population, and associations between risk factors for acute stroke and the clinical outcomes are stronger among older adults [1][2][3][4]. However, the knowledge gleaned from research on older adults cannot always be applied to younger adults. Lutski et al. showed that young adults have a high prevalence of modi able vascular risk factors and especially a high rate of smoking [18]. In our study, we found that most strokes in young smokers were related to the existence of traditional stroke risk factors. There are several reasons why smokers may have a higher risk of developing stroke.
Smoking increases the risk of complex atherosclerotic cardiovascular events, including deleterious effects on the endothelial function, in ammation, lipids, and thrombosis [9,10]. Nicotine exposure could also induce a reduction in insulin release, and negatively affect insulin action, suggesting that nicotine could be a cause for the development of insulin resistance. Therefore, young stroke patients who smoked were more likely to have those comorbidities.
Initially, the term "smoker's paradox" was an observational phenomenon of an unexpected favorable outcome in smokers who experienced acute myocardial infarction and had a smoking duration of over 25 years [9]. Recent studies in stroke have indicated a strong positive correlation between recanalization and smokers, indicating that thrombolytic therapy acts more effectively in smokers [11,12]. Then smokers who experienced subarachnoid hemorrhage (SAH) were also reported with many paradoxically superior outcome measures, including reduced odds of a poor outcome in poor-grade patients [11,12]. In other words, these studies stated that smoking predicted a better outcome following various reperfusion strategies. However, they had some shortcomings, including short observation durations, or their patients' low clinical risk pro les and low NIHSS scores on admission. In addition, those studies showed that ignore smoking was associated with a younger age of stroke. The risk of recurrent stroke after the rst stroke is about 10% at 1 year, 25% at 5 years, and 40% at 10 years, and about 40% of stroke survivors become disabled after stroke. Recurrent strokes often have a higher rate of death and disability. In our present study, we only investigated young stroke patients, and the data suggested that smoking was associated with the higher severity of the initial stroke, more modi able risk factors and the poor functional outcome recorded at 3 months after stroke. Therefore, smoker's paradox is misleading.
The difference in the incidence of stroke between two sexes have been well established. Several epidemiological studies reported higher age-speci c stroke rates in men [19][20][21]. However, women still experience more frequent stroke events because of their increased longevity and high stroke incidence at older ages. The higher prevalence of traditional vascular risk factors, including hypertension, diabetes mellitus, and hyperlipidemia in middle-aged men may contribute to this result [22,23]. Our ndings are consistent with those of a meta-analysis and certain community-based reports on the relationship between stroke and sex. More interestingly, age-speci c stroke rates are higher in men, and the higher prevalence of traditional vascular risk factors is associated with smoking.
Patients who cease smoking are expected to live longer and less likely to develop tobacco-related diseases, including coronary heart disease, stroke, cancer, and pulmonary disease [24][25][26][27]. In the present study, smoking cessation is particularly bene cial as it was shown to reverse the post stroke function outcome of young stroke patients. In 1997, Taiwan established its rst comprehensive policy package for tobacco control, the Tobacco Hazards Prevention Act (THPA), to include pictorial health warnings, smoke-free worksites and restaurants, and a ban on most tobacco advertising [24,25]. In 2018, Taiwan recorded an adult smoking rate of 13%, its lowest rate since 1990. This resulted from the efforts of the government to discourage smoking since the enactment of the Tobacco Hazards Prevention Act in 1997. Taiwan was the second Asian country after Bhutan to institute an indoor smoking ban. However, Taiwan's tobacco price still remains too low, and now, Taiwan government only levies one package of 20 cigarettes at NT$ 10 (US$ 0.3). According to a World Bank study, a 10-percent rise in the cigarette price leads to a 4-to 8-percent reduction in cigarette consumption.
Therefore, it is imperative to raise the price of tobacco.
The TSR provides the opportunity to investigate these issues owing to its large sample size, the homogeneous demographic characteristics and clinical phenotypes of the study subjects, and the standard diagnostic workup [17,28].
However, the present study has a number of limitations that need to be addressed. First, the baseline characteristics of the study groups were only obtained in the hospitals; hence, certain unadjusted potential confounders may still exist. Second, the sample sizes was su ciently large, representing 3% of the total adult population in Taiwan. However, our cohort consisted of participants with an above-average socioeconomic status engaging in a medical screening program. This may affect the generalizability of our ndings. Third, our data included only stroke survivors as the TSR does not include data on patients who died before reaching the hospital, leading to the possible underestimation of the mortality rate.

Conclusion
In conclusion, this study uncovers an intriguing relation between smoking and young stroke. High smoking rates are found among male young stroke patients, compared to much lower rates among females. Smoking is associated with higher comorbidities and poor long-term function outcomes. Smoking cessation would substantially reduce those effects. This implies that physicians should put more efforts into reducing cigarette smoking serves in Taiwan.

Declarations
Ethics approval and consent to participate Ethics approval of this study was obtained from China Medical University and the Institutional Review Boards (IRB) of the collaborating hospitals (CMUH104-REC2-115). The written informed consent was obtained from all individual participants included in the study.

Consent to publish
Not applicable Availability of data and materials The datasets generated and/or analyzed during the current study are not publicly available due to the restrictions of the local ethics committee and institutional data security and privacy policies. The data access request needs institutional and ethics committee's approval.

Competing interests
The author(s) con rm that this article content has no con icts of interest. Taichung, Taiwan. The funders had no role in the study design, data collection, analysis, or interpretation.