Smoking is associated with an increased risk of acute chest syndrome and pain among adults with sickle cell disease

To the editor:

Acute chest syndrome (ACS) is a leading cause of hospitalization and death for people with sickle cell disease (SCD).¹  Cigarette smoking has been inconsistently related to an increased risk of ACS in prior studies^2,3  and may be one of the few potentially modifiable risk factors for ACS. We tested the hypothesis that adults with SCD with exposure to tobacco smoke have a higher rate of ACS episodes compared with adults who are not exposed.

We examined a cohort of adults with SCD who had received all of their SCD care at Washington University continuously since 2004, were only hospitalized at Barnes-Jewish Hospital or affiliated hospitals, and completed questionnaires about smoke exposure.⁴  Electronic medical records from January 1, 2004, to March 1, 2009, were reviewed for ACS and pain episodes using standardized data extraction forms. After examining clinic records and reviewing study participants with SCD clinical personnel, patients known to have been admitted to hospitals outside of the Barnes-Jewish Hospital system were excluded from analyses. Enrollment into this study began in August 2006 and questionnaires were completed at the time of study consent. This study was approved by the Human Research Protection Office at Washington University and informed consent was obtained from participants. ACS was defined as a new infiltrate in the context of a respiratory illness with or without fever.⁵  Pain episodes were defined as hospitalizations with a primary diagnosis of pain related to SCD.⁶  The relationships between smoke exposure and rates of ACS and pain were examined using multivariate negative binomial regression models.

Our cohort was composed of 106 adults with SCD (median age 32.5 years, range 18-72 years). The prevalence of smoke exposure was high, with 38 (36%) active smokers and 18 (17%) nonsmokers who reported environmental tobacco smoke (ETS) exposure. Among the smokers, all reported they started smoking before 2004 when morbidity data collection began. In separate multivariate models, active smoking and ETS exposure were each associated with more than twice the rate of ACS episodes compared with no smoke exposure. There was also a significant association between active smoking and pain rate in the multivariate model (Table 1).^3,7,8  To determine whether there was a dose effect associated with cigarette consumption, we limited the multivariate analysis to smokers. There was no dose effect for number of cigarettes smoked per day and rate of ACS (P = .816) or pain episodes (P = .903).

We can only postulate about potential mechanisms for the association between smoking and SCD-related morbidity. First, data from the general population indicates that both active cigarette smoking^9-11  and ETS exposure¹²  increase vascular inflammation, platelet aggregation, and expression of endothelial adhesion molecules, processes that would increase vaso-occlusion. Second, adults with SCD who smoke may be increasing their risk of community acquired pneumonia (CAP). In a prospective cohort study of 26 429 men from the Health Professionals Follow-Up Study and 78 062 women from the Nurses' Health Study II, current smoking was associated with an increased risk of CAP in both men (odds ratio [OR] 1.46, 95% confidence interval [CI] 1.0-2.1) and women (OR 1.55, 95% CI 1.2-2.1).¹³  CAP is a known risk factor for ACS.³ 

The association between smoking and ACS in the present study was not shown in Cooperative Study of Sickle Cell Disease (CSSCD).³  In the CSSCD, however, patients with ETS exposure were not explicitly differentiated from active smokers or nonsmokers, potentially biasing results toward the null.

This study demonstrates a significant, clinically meaningful association between smoking and increased rate of ACS and pain events among adults with SCD. Although self-reported smoking status and use of a convenience clinic sample limit our results, these data provide additional rationale for SCD providers to emphasize smoking cessation to patients and their families. Lack of a dose effect for number of cigarettes smoked per day may be due to underreporting of amount smoked.¹⁴  Further systematic studies are needed to confirm these findings and define the mechanism by which cigarette smoking worsens SCD-related morbidity.