Association of Dose of Inhaled Corticosteroids and Frequency of Adverse Events

Abstract

Background: Inhaled corticosteroids (ICS) are the cornerstone of asthma treatment and significantly improve morbidity and mortality. Adverse effects of oral corticosteroids are well documented, but less is known about ICS.

Methods: We conducted observational studies in adults with asthma using two different UK nationwide datasets: Clinical Practice Research Datalink (CPRD) Aurum and CPRD GOLD. The exposure was incident ICS; the outcomes were major adverse cardiac events (MACE), arrhythmia, pulmonary embolism (PE) and pneumonia over 12-months. Our main analyses used a cohort method with stabilized inverse probability treatment weighting to balance confounding between exposed and unexposed patients. Secondary analyses included nested case-control studies, and self-controlled case series. ICS was treated both as a categorical and continuous variable. Absolute risk was estimated using weighted flexible parametric models.

Findings: From 162,202 patients in our main cohort, there was an association with all outcomes at medium daily ICS dose or higher (HR, 95%CI at 201-599mcg: MACE=2.63, 1.66-4.15, arrhythmia=2.21, 1.60-3.04, PE=2.10, 1.37-3.22, pneumonia=2.25, 1.77-2.85; at ≥600mcg: MACE=4.63, 2.62-8.17, arrhythmia=2.91, 1.72-4.91, PE=3.32, 1.69-6.50, pneumonia=4.09, 2.98-5.60). There were no associations with lower doses of ICS. Secondary analyses produced similar results. The number needed to harm (95%CI) using 12-months of ICS 201-599mcg: MACE=473 (344-754), arrhythmia=567 (395-1006), PE=1221 (744-3388) and pneumonia=230 (177-327) and using ICS ≥600mcg: MACE=224 (148-461), arrhythmia=396 (228-1523), PE=577 (309-4311), pneumonia=93 (69-141).

Interpretation: Short-term use of low dose ICS was not associated with adverse effects. Moderate-high daily ICS doses were associated with an increased risk, but low-frequency, of cardiovascular events, pulmonary embolism and pneumonia. It is important for clinicians to adhere to guideline recommendations to use the lowest effective ICS dose. This article is open access and distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).

Keywords: asthma; cardiovascular disease; corticosteroids; pneumonia; pulmonary embolism.

Figure 1.

Association between ICS use and outcomes, categorized by average daily ICS dose (main analysis: Clinical Practice Research Datalink Aurum inverse probability treatment weighting cohort). ICS = inhaled corticosteroid; MACE = major adverse cardiovascular events; PE = pulmonary embolism; UTI = urinary tract infection.

Figure 2.

Association between average daily ICS dose as a continuous outcome and outcomes (main analysis: Clinical Practice Research Datalink Aurum inverse probability treatment weighting cohort). ICS = inhaled corticosteroid; MACE = major adverse cardiovascular events; UTI = urinary tract infection.

Figure 3.

Association between inhaled corticosteroid (ICS) use and outcomes by time since last ICS prescription (secondary analysis: Clinical Practice Research Datalink Aurum nested case–control). BMI = body mass index; cap = community-acquired pneumonia; cva = cerebrovascular accident; CVD = cardiovascular disease; IMD = Index of Multiple Deprivation; LABA = long-acting β-agonist; OCS = oral corticosteroid; PE = pulmonary embolism; SABA = short-acting β-agonist.

Figure 4.

Association between ICS use (during exposure and after exposure stopped) and outcomes: CVD and the negative controls of UTI, abdominal pain, and cellulitis (secondary analysis: Clinical Practice Research Datalink Aurum self-controlled case series). CVD = cardiovascular disease; ICS = inhaled corticosteroid; UTI = urinary tract infection.

Figure 5.

Absolute risk of each outcome associated with categorized average daily dose of inhaled corticosteroid (ICS) (main analysis: Clinical Practice Research Datalink Aurum inverse probability treatment weighting cohort). Red line denotes high-dose ICS, dashed green line denotes medium-dose ICS, and red dotted line denotes low-dose ICS. Gray shading denotes the 95% confidence interval around the estimate. MACE = major adverse cardiovascular events.