Before and beyond Inhaled Corticosteroids

The indications for inhaled corticosteroids (ICS) have narrowed in the past few years. One of the drivers for this change is the association between ICS and pneumonia. Prescribed for the treatment of chronic obstructive pulmonary disease (COPD) and asthma, they are only recommended as initial therapy for highly symptomatic patients with frequent exacerbations and blood eosinophil counts greater than 300 cells/dl in the former (1). In asthma, intermittent use as a rescue inhaler has been deemed noninferior to daily dosing (2).

Despite these changes, ICS continue to be among the most commonly prescribed medications in the United States. In 2018, six inhaled corticosteroid preparations made it into the top 100 list of the Center for Medicare and Medicaid services. Medicaid and Medicare Part B spent over 4 billion dollars each on ICS during the same year (3).

Pneumonia continues to be the leading infectious cause of death worldwide, and it is responsible for 1.4 million visits to the emergency department every year in the United States alone (4). Streptococcus pneumoniae is the main causative agent (5).

These staggering numbers underline the importance of the study in this issue of AnnalsATS by Torén and colleagues (pp. 1570–1575), which furthers our knowledge on the association between ICS and pneumonia on a background of ICS overutilization (6).

In this editorial, I will review some of the strengths and weaknesses of the study and the association between ICS and pneumonia. I will finish with presenting nonpharmacological interventions for the treatment of patients with airway disease and persistent symptoms or exacerbations that can prevent ICS overuse.

The authors used a case–control design, in which data were abstracted from the prospectively collected Swedish Invasive Pneumococcal Disease, National Population, Hospital Discharge, and Drug registries. Development of invasive pneumococcal disease (IPD) was the entry criteria for cases. Control subjects were obtained by matching geography, age, and sex to the general population. This approach is a strength of the study. It helped reduce unmeasured biases such as living conditions, vaccination rates, and colonization with S. pneumoniae in the community. It also provided a very robust sample size with over 4,000 cases in the final analysis, 71% of which with pneumonia. Their main conclusion was an increase in the risk of IPD with pneumonia in patients being treated with ICS. ICS did not impact the risk of IPD without pneumonia.

IPD was defined by a positive culture or detection of S. pneumoniae antigen from a sterile site. Because these diagnostic tests are usually performed only in hospitalized patients, the use of healthier control subjects from the community may have resulted in an overestimation of the risk attributable to ICS (7). Both COPD and asthma were more frequent among hospitalized cases, with 10.3% for COPD and 9.2% for asthma, than control subjects, with 0.9% for COPD and 2.3% for asthma. Smoking, the anatomical and physiological alterations of COPD, and asthma are known risk factors for IPD. In fact, in their sensitivity analysis, any use of ICS in the last 5 years remained associated with increased odds for IPD (odds ratio, 1.94; 95% confidence interval, 1.53–2.47). This finding can only be explained if ICS have very long-lasting effects or if confounding by indication is present.

One last limitation relates to the exclusion criteria. IPD affects children and the elderly disproportionately, and asthma and COPD are the main indications for ICS treatment. Asthma usually has its onset before the age of 10, and COPD increases in prevalence with age (8). The exclusion of patients aged less than 20 and greater than 65 years is problematic, as it excludes a large portion of the target population.

ICS and Risk of Pneumonia

ICS are a known risk factor for pneumonia. In a landmark population-based cohort study from the province of Québec, Canada, Ernst and colleagues (9) found that patients with an exposure to ICS had a 70% increase in their relative risk for pneumonia. Their results also substantiated a dose–response relationship, risk reduction with discontinuation of treatment, and a higher risk of mortality for patients requiring admission to the hospital who had an exposure to ICS. Their findings were independent of COPD severity (9). Similarly, secondary analysis of randomized controlled trials, including TORCH (Towards a Revolution in COPD Health), IMPACT (Informing the Pathway of COPD Treatment), INSPIRE (Investigating New Standards for Prophylaxis in Reduction of Exacerbations) (10–12), found an increased risk of pneumonia in the subgroup of patients receiving ICS.

What remains unknown is whether the increased risk of pneumonia is pathogen specific. None of the above cited studies reported the etiological agents of pneumonia. In this sense, the focus by Torén and colleagues on S. pneumoniae is a novelty. S. pneumoniae continues to be the most commonly isolated bacteria in community-acquired pneumonia. It accounts for 5–10% of the cases of community-acquired pneumonia in the United States and 25–40% of the cases in Europe (5, 13). The findings by Torén and colleagues are supported by previous studies that reported an increase in the rate of upper airway colonization by S. pneumoniae with the use of ICS and dampening of the inflammatory response required to recruit neutrophils and trigger an adaptive immune response to this particularly virulent pathogen (14, 15).

Before and beyond ICS

Overall, the available evidence supports an increase in the risk of pneumonia with the use of ICS. These findings mandate a reevaluation of the risk–benefit ratio of ICS for asthma and COPD. While we await safe and effective alternatives, maximization of available therapies and judicious use of ICS should be the norm.

Part of the problem of ICS overuse stems from a focus on inhaled pharmacotherapy. Nonpharmacological interventions are underutilized despite their known benefits. In the United States, this is exemplified by the low utilization rates of pulmonary rehabilitation in patients with COPD. A study from 2018 reported that less than 3% of Medicare beneficiaries with COPD discharged from the hospital received pulmonary rehabilitation over the following 12 months (16). Exercise training, nutritional advice to avoid the extremes of weight, and self-management education can improve symptoms and decrease the risk of exacerbations.

Attention to and mitigation of environmental and occupational exposures is another intervention often overlooked. Besides tobacco smoke, indoor and outdoor air pollution have been associated with worsening respiratory symptoms, exacerbations, and progression of asthma and COPD. Avoidance of biomass fuels, allergens, and improvements in housing and home ventilation systems can all be helpful (17). At a higher degree, COPD and asthma affect minorities and lower socioeconomic groups. Interventions to address disparities in exposures to pollutants and access to health care could also be beneficial (18).

Comorbidities are common in chronic respiratory patients and have a profound impact on quality of life. They should also be targeted for treatment. As an example, esophageal and pharyngeal dysmotility leading to aspiration can trigger exacerbations. Coronary and peripheral vascular disease can mimic exacerbations of asthma and COPD and can impact exercise tolerance and quality of life. In the face of a predictable set of comorbidities, pulmonary practices should at least ensure care pathways for their evaluation and management (19).

Finally, we need to acknowledge the limitations of diagnostic classifications created more than 100 years ago. These classifications are dichotomous; a patient can be healthy or ill, and overlap between different airway diseases is not considered. Realization of these limitations opens the door to considering overlap between COPD, asthma, bronchiectasis, interstitial lung disease, and other airway and parenchymal lung diseases. This, in turn, opens new horizons in terms of treatments available to alleviate patients’ symptoms.

In conclusion, the article by Torén and colleagues provides new epidemiological evidence of the link between ICS and IPD with pneumonia, a reminder that a new inhaler prescription might not be the right response for every patient.