In this MR study, we investigated the role of smoking in acne vulgaris and rosacea using summary data from the most valid available GWAS meta-analyses. To the best of our knowledge, the present study is the first to examine the causal relationship between cigarette smoking and acne and rosacea using a two-sample MR method. In this study, we used GWAS data to investigate the risk of skin disorders in individuals who are genetically predisposed to smoking compared to the control group. We performed various MR analyses using loci associated with different stages of cigarette smoking, including smoking initiation, heaviness (cigarettes per day), and age of initiation. Despite some earlier observational studies reporting no correlation between smoking and acne and rosacea, our analysis shows a causal relationship between smoking behaviors and both conditions.
Due to the harmful nature of smoking, it is not ethically possible to conduct randomized clinical trials in this regard. Therefore, our earlier knowledge about the association between smoking and skin disorders was mainly limited to observational studies with varying results.
In 2012, in a community-based study of 17345 Chinese adolescents in six cities, smoking was found to be associated with adolescent acne, which was in line with our findings (28). In terms of molecular pathways, cigarette smoking plays an essential role in lipid peroxidation of sebum in comedones, which can elevate the local levels of Interleukin-alpha1, leading to abnormal skin inflammation and keratinization (11). On the contrary, a population survey in Greece and a case-control study in Iran suggested no correlation between smoking and acne (29, 30). Results of a cohort study conducted by Klaz et al. demonstrated an inverse, dose-dependent relationship between severe acne prevalence and daily cigarette consumption in young smoker men (31, 32). Moreover, Wolkenstein and his colleagues conducted a cross-sectional population-based online survey in representative samples of individuals aged 15–24 in a few European countries. Their result showed that smoking tobacco was associated with a reduced probability of acne (32). It should be noted that there is conflicting evidence about the possible anti-inflammatory impact of nicotine on skin conditions, namely, by suppressing neutrophil-mediated inflammatory actions (33). It is noteworthy that according to previous studies, it has been well understood that smoking is responsible for causing oxidative stress, leading to the production free radicals and various proinflammatory cytokines in inflammatory skin disease, (34–36).
Like acne, the clinical data about the correlation between smoking and rosacea is currently controversial. Demonstrations of a cohort study conducted by Aldrich et al. support our results on a positive correlation between smoking and rosacea (37). Our findings confirm the observation of a previous cross-sectional study suggesting a significantly increased risk of developing rosacea among smokers (38). These findings could be due to the angiogenic effects of nicotine and the degeneration of collagen and elastic fibers caused by cigarette smoke (39). Furthermore, a few studies reported a positive association between current smoking and a decreased risk of developing rosacea (40–42). These controversial findings can be explained by the microvascular constriction caused by the nicotine in tobacco (43). To be noted, based on a systematic review and meta-analysis conducted in 2020, this protective effect of current smoking on the risk of rosacea may be temporary (44). Regarding the severity of the disease, Alinia et al. reported a significant relationship between the severity of rosacea and number of cigarettes smoked in patients who had quit smoking (45).
Our study presents several strengths. Concerning statistical analysis, we utilized two-sample MR, a method with high statistical power in assessing causality using genetic polymorphisms as instrumental variables. In this study, we used the term “lifetime smoking”, considering various elements of a person’s exposure to smoking. This term enables us to understand smoking long-term health impacts comprehensively. Moreover, the reliability and large sample size of the GWAS data used in analyses, use of multiple MR methods in assessing the causality, as well as the approaches to eliminate sources of bias including heterogeneity, pleiotropy, and sensitivity analysis can be mentioned as strength of our study.
However, some limitations of our study should be considered. Our GWAS summary statistics were mostly from European patient data. This may introduce bias and limit the generalizability of our findings. Careful interpretation is needed as our results may not apply to other universal populations. Regarding horizontal pleiotropy, despite performing different tests to detect it, the presence of horizontal pleiotropy cannot entirely be ruled out. Furthermore, the data on subtypes of acne and rosacea were unavailable. Thus, we could not perform a detailed analysis of the subtypes of these two skin disorders. To be noted, the possibility of population stratification can also be considered as a source of bias.
In conclusion, we used two sample MR methods to investigate the possible correlation between smoking and acne rosacea. We found no evidence to support the notion that an escalation in smoking initiation corresponds to an increased risk of acne but a significant causal correlation between lifetime smoking and an elevated risk of acne. Moreover, our results indicate a potential causal positive association between the initiation of smoking with rosacea and a significant correlation between lifetime smoking and an elevated risk of rosacea. Our findings do not support previous observational studies which reported no relationship between smoking and acne rosacea. Further investigations are required to validate these associations and determine the underlying mechanisms. In overall, applying these results in both clinical practice and public health decision making, due to the role of smoking behavior in enhancing the risk of acne and rosacea is of great value. Such policies can be imposed through all levels of prevention and treatment of these two inflammatory skin conditions, including educational programs providing insights to the patients with acne and rosacea into the necessary adjustments to their lifestyle.