Even low-dose steroid treatments substantially increase CVD risk

While high doses of steroids are known to increase the risk of cardiovascular disease (CVD), the impact of lower doses is unknown, a Leeds University study in PLOS Medicine suggests that even low doses of glucocorticoid may increase the risk of cardiovascular diseases.

Why was this study done?
• Glucocorticoids (steroids) are widely used to reduce disease activity and inflammation in patients with a range of immune-mediated inflammatory diseases, such as rheumatoid arthritis, polymyalgia rheumatica, giant cell arteritis, and inflammatory bowel disease.
• Adequate assessment of cost-effectiveness of new steroid-sparing treatments for immune and inflammatory diseases require modelling of estimates of risk and cost of the main treatment complications of steroids.
• It is widely recognised that high-dose steroids may increase the risk of cardiovascular disease (CVD; heart disease, stroke, or other vascular diseases), but it is debated whether this increase also applies to lower steroid doses.
• Earlier studies of CVD risk associated with glucocorticoid therapy failed to account for changes in dose over time and for use of non-oral steroids and other potentially confounding therapies.

What did the researchers do and find?
• In 87,794 adults with immune-mediated inflammatory diseases and no prior CVD (5-year median follow-up), we studied the risk of 6 common CVDs associated with the steroid dose prescribed, quantified either as current or as cumulative dose.
• We found strong dose-dependent risks of all CVDs, including myocardial infarction, heart failure, atrial fibrillation, and cerebrovascular disease, in patients diagnosed with the 6 inflammatory diseases studied.
• After 1 year, the overall absolute risk of CVD doubled for individuals using less than 5 mg prednisolone per day and was 6 times higher for users of 25 mg or greater.
• Many individuals had known modifiable cardiovascular risk factors, including current smoking (24%), obesity (25%), or hypertension (25%).

What do these findings mean?
• We have provided evidence that individuals receiving steroids have an increased risk of developing a broad spectrum of fatal and nonfatal CVDs and that this risk increases with the dose of steroids and with the duration of steroid treatment.
• It was previously believed that less than 5 mg of prednisolone was safe long term, but even at this “low dose” patients with immune-mediated inflammatory diseases have a doubling of their underlying risk of CVD.
• New treatment approaches that avoid the need for long-term steroid treatment and have better cardiovascular safety profile are required for immune-mediated inflammatory diseases.
• All patients requiring long-term steroid treatment should be prescribed the lowest effective steroid dose and have a personalised CVD risk prevention plan that takes into account current and prior steroid use.

Dose-dependent oral glucocorticoid cardiovascular risks in people with immune- mediated inflammatory diseases: A population-based cohort study

Authors: Mar Pujades-Rodriguez, Ann W. Morgan, Richard M. Cubbon, Jianhua Wu.

Published 3 December, PLOS Medicine

Abstract

Methods and findings
We conducted a population-based cohort analysis of medical records from 389 primary care practices contributing data to the United Kingdom Clinical Practice Research Datalink (CPRD), linked to hospital admissions and deaths in 1998–2017. We estimated time-variant daily and cumulative glucocorticoid prednisolone-equivalent dose-related risks and hazard ratios (HRs) of first all-cause and type-specific cardiovascular diseases (CVDs). There were 87,794 patients with giant cell arteritis and/or polymyalgia rheumatica (n = 25,581), inflammatory bowel disease (n = 27,739), rheumatoid arthritis (n = 25,324), systemic lupus erythematosus (n = 3,951), and/or vasculitis (n = 5,199), and no prior CVD. Mean age was 56 years and 34.1% were men. The median follow-up time was 5.0 years, and the proportions of person–years spent at each level of glucocorticoid daily exposure were 80% for non-use, 6.0% for <5 mg, 11.2% for 5.0–14.9 mg, 1.6% for 15.0–24.9 mg, and 1.2% for ≥25.0 mg. Incident CVD occurred in 13,426 (15.3%) people, including 6,013 atrial fibrillation, 7,727 heart failure, and 2,809 acute myocardial infarction events. One-year cumulative risks of all- cause CVD increased from 1.4% in periods of non-use to 8.9% for a daily prednisolone-equivalent dose of �25.0 mg. Five-year cumulative risks increased from 7.1% to 28.0%, respectively. Compared to periods of non-glucocorticoid use, those with <5.0 mg daily prednisolone-equivalent dose had increased all-cause CVD risk (HR = 1.74; 95% confi- dence interval [CI] 1.64–1.84; range 1.52 for polymyalgia rheumatica and/or giant cell arteritis to 2.82 for systemic lupus erythematosus). Increased dose-dependent risk ratios were found regardless of disease activity level and for all type-specific CVDs. HRs for type-spe- cific CVDs and <5.0-mg daily dose use were: 1.69 (95% CI 1.54–1.85) for atrial fibrillation, 1.75 (95% CI 1.56–1.97) for heart failure, 1.76 (95% CI 1.51–2.05) for acute myocardial infarction, 1.78 (95% CI 1.53–2.07) for peripheral arterial disease, 1.32 (95% CI 1.15–1.50) for cerebrovascular disease, and 1.93 (95% CI 1.47–2.53) for abdominal aortic aneurysm. The lack of hospital medication records and drug adherence data might have led to underestimation of the dose prescribed when specialists provided care and overestimation of the dose taken during periods of low disease activity. The resulting dose misclassification in some patients is likely to have reduced the size of dose–response estimates. Conclusions
In this study, we observed an increased risk of CVDs associated with glucocorticoid dose intake even at lower doses (<5 mg) in 6 immune-mediated diseases. These results highlight the importance of prompt and regular monitoring of cardiovascular risk and use of primary prevention treatment at all glucocorticoid uses.

[link url="https://doi.org/10.1371/journal.pmed.1003432"]Full PLOS Medicine study (Open access)[/link]

https://doi.org/10.1371/journal.pmed.1003432