The current coronavirus disease 2019 (COVID-19) has spread worldwide since starting in China in December 2019, involving over 46 million people by November 2020. Additionally, more than 1.2 million deaths have been reported so far all over the world [1]. The information about the impact of COVID-19 among people living with human immunodeficiency virus (PLWH) is limited, however preliminary data have shown an incidence of SARS-CoV-2 infection in this subgroup similar as that found among the general population [2].
The impact of smoking on the risk of developing COVID-19 is unclear and has been a matter of controversy. While some studies have observed an association with an increased severity of COVID-19 [3–5], other reports have suggested that the incidence of SARS-CoV-2 infection among current smokers could be lower [6–8].
In the present study, we assessed the incidence of SARS-CoV-2 infection in a cohort of HIV-infected patients followed up in a hospital in Spain. Our aim was to evaluate the association of current smoking on the probability of developing SARS-CoV-2 infection in PLWH.
We performed a retrospective nested case-control study within a cohort of HIV infected patients followed at the Unit of Infectious Diseases of a University Hospital in Spain. All individuals were evaluated at least every six months until 31st October 2020 and routinely asked about symptoms consistent with COVID-19. A diagnosis procedure was conducted when any symptom was present. Cases were patients from this cohort with a confirmed diagnosis of COVID-19 between February and October 2020. Control group included the remaining HIV-infected individuals from the cohort without a diagnosis of COVID-19. Diagnosis of COVID-19 was established by the detection of SARS-CoV-2 RNA or antigen in nasopharynx exudate or bronchial-alveolar lavage, by PCR or EIA, respectively, or by presence of plasma SARS-CoV-2 antibodies by ECLIA. Overall, 397 (56.6%) individuals of this population were tested for plasma SARS-CoV-2 antibodies as a part of a seroincidence study [9]. Variables that were significantly associated with the occurrence of COVID-19 in the bivariate analysis, along with sex and Charlson index, were entered in a logistic regression model, using the enter method. All patients gave informed consent for using their data.
Seven hundred and two individuals were included at the cohort. All patients were receiving antiretroviral therapy by the time of the study. A diagnosis of COVID-19 was established in 18 (2.56%) individuals. Three hundred and sixty-nine (52.6%) individuals were current smokers. A higher proportion of non-smokers vs. smokers acquired the HIV infection through sexual intercourse [213 (64.9%) vs. 170 (46.1%), p= 0.001]. Comparative characteristics of COVID-19 cases vs. no COVID-19 controls are shown in table 1. COVID-19 diagnosis was based on PCR in 11 (61.1%) patients, on antigen in 4 (22.2%) subjects and on serology in the remining 3 (16.7%) individuals. Regarding the clinical spectrum, 2 (11.1) patients were asymptomatic, 12 (66.6%) had upper-respiratory tract infections and 2 (11.1%) had viral pneumonia which required hospital admission with non-invasive mechanical ventilation. One (5.6%) individual showed extra respiratory symptoms which also required hospital admission. One (5.6%) patient developed severe pneumonia and died. The relationship between several potential risk factors and COVID-19 development are shown in table 1. After multivariate analysis, adjusting by sex, tobacco smoking and Charlson index, only tobacco smoking was associated with the risk of SARS-CoV2 infection [aOR=9.42 (95% confidence interval 2.15-41.33), p = 0.003].
This study shows that tobacco smoking is associated with a lower incidence of SARS-CoV-2 infection among PLWH. Despite of higher risk of infection that would be expected among smokers, because of the frequent face touching during smoking act, the probability of being active smoking among participants diagnosed with COVID-19 was 9-fold lower than that of subjects without such a diagnosis.
The role that tobacco may be playing on the risk of SARS-CoV-2 infection is not entirely understood. On one hand, smoking has been reported to be associated with an increased COVID-19 severity [3]. Tobacco smoke exposure results in inflammatory processes in the lung, increased mucosal inflammation, expression of inflammatory cytokines and tumor necrosis factor α, increased permeability in epithelial cells, mucus overproduction, and impaired mucociliary clearance [10], all of which may prompt a more severe lung damage. On the other hand, there are some data suggesting the inferior risk for COVID-19 among smokers. Thus, in a study conducted in England a lower incidence of SARS-CoV-2 infection was observed among smokers, as a secondary result. Similarly, in a recent meta-analysis (10), whose aim was to assess the impact of tobacco smoking on the risk of developing SARS-CoV-2 infection, current smoking also appeared to be associated with a lower risk of COVID-19.
The underlying mechanisms involved in the protective effect of tobacco are unclear. Nicotine has a marked impact on regulation of the activity of angiotensin-converting enzyme receptors (ACE) [11], which may interfere SARS-CoV-2 entrance into respiratory epithelial cells, thus reducing the risk of infection. However, Smith et al showed a correlation between higher ACE2 mRNA levels and SARS-CoV-2 infection in a study conducted in human lung tissue from current smokers. In addition, the frequent habit mouth-hands of smokers could also boost the risk of infection. Because of all these reasons, a higher risk of infection would be expected [4]. Nevertheless, a recent study performed in mammalian cell culture models has shown that other host molecular factors and pathways can be essential for SARS-CoV-2 infection [12]. Whether the expression of these additional factors is deregulated in current smokers is something that needs to be analyzed, but it might explain the reduced SARS-CoV-2 infection in this population with high levels of ACE2.
This study has some limitations. Given its retrospective design, there might be hidden conditions associated with tobacco use that result in less likelihood of SARS-CoV-2 infection. Thus, greater proportion non-smokers individuals acquired HIV infection through sexual transmission, particularly MSM (64.9% vs. 46.1%) and COVID-19 might be more common in this specific subset. Nevertheless, the risk factors for HIV-infection were not associated with a higher risk SARS-CoV-2 infection in our study. Similarly, asymptomatic or mild COVID-19 cases may have gone unnoticed among control group. However, in most COVID-19 cases symptoms are present [13]. Moreover all subjects included in this study were questioned about COVID-19 symptoms and close contact to patients with confirmed COVID-19. Those reporting symptoms or close contact to a COVID-19 confirmed case were tested for SARS-CoV-2 infection. In any case, further prospective cohort studies based on serology determinations should clarify this problem.
The molecular and physiological mechanisms by which tobacco smoking might protect against SARS-CoV-2 infection should be further investigated in order to achieve potential targets for the prevention and treatment of COVID-19. In the meanwhile, public-facing messages about a possible protective effect of tobacco smoking on the risk of COVID-19 must be avoided, taking into account the high risk for a wide spectrum of extremely severe heath problems associated with tobacco smoking, which counterbalance any protective role of this habit against SARS-CoV-2 infection.