Sleep Apnea and the Risk of Pneumonia
Abstract & Commentary
By Barbara A. Phillips, MD, MSPH
Professor of Medicine, University of Kentucky; Director,
Sleep Disorders Center, Samaritan Hospital, Lexington
Dr. Phillips serves on the speakers bureau for PotomaCME.
Synopsis: In a large, population-based study, patients with obstructive sleep apnea had a 1.2-fold increased risk of incident pneumonia over a 4-year period compared with matched controls, and the risk was highest for those who used CPAP.
Source: Su VY, et al. Sleep apnea and risk of pneumonia: A nationwide population-based study. CMAJ 2014;186:415-421.
Based on plausible biologic mechanisms, these au-thors hypothesized that sleep apnea predisposes to the development of pneumonia. To address this question, they explored the Health Insurance Research Database in Taiwan, one of the largest population-based databases in the world. They identified a cohort of adult patients who were newly diagnosed with sleep apnea and a control group of patients without sleep apnea, matched for age, sex, and comorbidities. (These comorbidities included pre-existing diabetes mellitus, hypertension, coronary heart disease, heart failure, cerebrovascular disease, dementia, epilepsy, Parkinson disease, chronic kidney disease, liver cirrhosis, gastroesophageal reflux disease, cancer, asthma, chronic obstructive pulmonary disease, and tuberculosis, but did not include weight, alcohol use, or body mass index). Patients were excluded before enrollment if they had a past medical history of pulmonary infection.
The primary outcome of the study was occurrence of pneumonia, based on the presence of compatible symptoms and chest radiography. Risk of pneumonia was stratified according to the "need" for continuous positive airway pressure (CPAP) treatment, which was used as a surrogate marker for severity of sleep apnea.
Over the course of the study, 6816 patients with sleep apnea were included and were matched 4:1 with 27,284 controls without sleep apnea. The two cohorts had similar characteristics, except that the sleep apnea group had a higher prevalence of dementia, Parkinson disease, and use of inhaled steroids and statins than the control group.
During about 4.5 years of follow-up, a greater percentage of those in the sleep apnea cohort had incident pneumonia than those in the control group (9.4% vs 7.8%), and patients with sleep apnea also had a significantly higher hazard for incident pneumonia than the control group (P < 0.001). Incidences in the sleep apnea and control groups were 20.90 and 17.22 persons per 1000 person-years, respectively, a highly significant difference. Comparing patients with and without pneumonia, those with incident pneumonia were older and more likely to have diabetes mellitus, coronary heart disease, and cirrhosis. After adjustment for all comorbidities, patients with sleep apnea remained at significantly increased risk of pneumonia compared with controls.
In this study, the authors used "need" for CPAP as a surrogate marker of sleep apnea severity. Compared with the control group, the adjusted risks of pneumonia among patients with sleep apnea who "needed" and did not "need" CPAP treatment were 1.32 (95% confidence interval [CI], 1.12-1.55) and 1.15 (95% CI, 1.04-1.27), respectively, suggesting that those with more severe sleep apnea were more likely to develop pneumonia.
COMMENTARY
This study indicates that people with sleep apnea are at increased risk for pneumonia, and that those who use CPAP are at even higher risk. This robust finding persisted despite controlling for multiple confounders. This is an important consideration, given that moderate-to-severe obstructive sleep apnea affects approximately 13% of men and 6% of women aged 30-70 years in the United States1 (and probably more in Taiwan, where this study was done, since Asians appear to be at increased risk for obstructive sleep apnea2,3). Potential mechanisms to account for this association include increased risk of pulmonary aspiration of pharyngeal contents during obstructed breaths while asleep4 and immunocompromise secondary to disrupted sleep.5 The authors also speculate that the use of CPAP could enhance pneumonia risk, since the mask could reduce sputum expectoration and a humidifier could provide a potential source of bacterial contamination.6 While this issue bears considerations, CPAP use per se has not been associated with pneumonia, and pneumonia should not be considered a reason to withhold CPAP therapy from people with clinically significant sleep apnea.
The issue of CPAP use and its correlates bear a bit of scrutiny. In this study, the authors used CPAP "need" as a surrogate for sleep apnea severity. This paper lacks the usual data typically used to assess presence and severity of sleep apnea, such as apnea plus hypopnea index (AHI) or measures of oxygen desaturation. This is almost certainly because sleep apnea is a dichotomous (either you have it listed as a diagnosis or you don’t) variable in their database; these authors really had no other way to determine if sleep apnea was mild, moderate, or severe. In the discussion, they note, "The choice for CPAP therapy may not solely depend on severity of the disease; other factors, such as comorbidities or patient preference, may influence its use as well." This is a point well-made and well-taken. CPAP is more likely to be used by patients with more severe disease. But patients with more severe disease are more likely to have other comorbidities, especially a greater degree of obesity. My suspicion is that the CPAP users were likely to be heavier, and weight is a well-known risk factor for pneumonia.7 Indeed, the study is significantly flawed by the lack of data about not only weight, but also smoking and alcohol use (also risks for pneumonia).8 While this is the best study to date on the relationship between sleep apnea and obstructive apnea, it should probably be taken with a grain of salt, at least until studies can control for these very important risk factors for both sleep apnea and pneumonia.
What clinical impact does this study have? Since (for whatever reason) people with sleep apnea have increased risk for pneumonia, it behooves us to consider what happens to them when they get pneumonia. Evidence is accumulating that obstructive sleep apnea patients with pneumonia are more likely to have a variety of adverse outcomes compared with those who don’t have sleep apnea. These include an increased risk of transfer to intensive care, intubation, longer hospital stays, and higher costs (although, paradoxically, a lower mortality).9 Maybe the big take-home message here is that when people with sleep apnea develop pneumonia, they may require more intense monitoring and/or therapy.
References
- Peppard PE, et al. Am J Epidemiol 2013;177:1006-1014.
- He QY, et al. Chin Med J 2010;123:18-22.
- Chen R, et al. Sleep Breath 2011;15:129-135.
- Beal M, et al. Laryngoscope 2004;114:965-968.
- Besedovsky L, et al. Pflugers Arch 2012;463:121-137.
- Gay PC. Respir Care 2009;54:246-257.
- Phung DT, et al. Obes Rev 2013;14:839-857.
- Grau I, et al. Int J Infect Dis 2014; May 19. [Epub ahead of print.]
- Lindenauer PK, et al. Chest 2014;145:1032-1038.
