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. 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. 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). 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.
Read full abstract