NPC Archive Item: Inhaled corticosteroids and the risk of fractures in COPD

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review NPC Logo

11 November 2011

A systematic review and meta-analysis of randomised controlled trials (RCTs) and observational studies found a statistically significant increase in the risk of fractures with prolonged use of the inhaled corticosteroids (ICS), budesonide and fluticasone, in people with COPD. While there are some limitations to these data the authors calculated a number needed to harm (NNH) of 83 (95% CI 38 to 2107) over 3 years for fractures with ICS.

Level of evidence:
Level 2 (limited quality patient-oriented evidence) according to the SORT criteria.

Action
Prescribers should continue to follow the NICE guideline on COPD, which recommends that ICS should only be used in combination with a long-acting beta-agonist (LABA). See the NICE guideline for the specific circumstances when ICS may be used. The choice of drug(s) should take into account the person’s symptomatic response and preference, and the drug’s potential to reduce exacerbations, its side effects and cost.

Systemic side effects (see background section below) are of crucial importance when considering the benefits and risks for individuals with COPD or asthma for whom ICS are being considered. The NICE guideline on COPD states that healthcare professionals should be aware of these potential risks and prepared to discuss these with patients. Whilst this study suggests that patients with COPD, who are taking ICS for prolonged periods, might be at increased risk of fracture, it would seem sensible to consider this in the context of patients’ overall risk of fracture, alongside the other risks of ICS, especially when high doses are used.

What is the background to this?
ICS are commonly prescribed for both the management of asthma and COPD. Prolonged use of high doses of ICS (as with the use of oral corticosteroids) carries a risk of systemic side effects (e.g. adrenal suppression or crisis, growth retardation in children and adolescents, decrease in bone mineral density, cataract and glaucoma) and a range of psychological or behavioural effects (e.g. psychomotor hyperactivity, sleep disorders, anxiety, depression, and aggression. Healthcare professionals are reminded when treating patients with ICS in COPD to remain vigilant for the development of pneumonia and other infections of the lower respiratory tract (i.e. bronchitis). More recently, ICS have also been associated with a dose-related increased risk of both diabetes onset and diabetes progression. The Committee on Safety of Medicines (CSM) has issued warnings about the use of high dose ICS, particularly in children, and in relation to fluticasone. Because of the risk of systemic side effects, patients who require prolonged high-dose ICS should be issued with a steroid treatment card.

To date, studies showing an effect of ICS on both bone mineral density and fractures have been inconsistent. The guideline development group (GDG) for the NICE guideline on COPD did not find a difference in fractures between patients taking a LABA plus ICS and those taking a LABA alone (based on moderate-quality evidence).

This study evaluated the association between ICS (fluticasone and budesonide) and fractures in COPD through systematic review and meta-analysis of 16 RCTs (14 fluticasone, 2 budesonide; n=17,513) and 7 observational studies (n=69,000). RCTs compared ICS with control treatment for COPD over at least 24 weeks.

What does this study claim?
Meta-analysis of the RCTs found a statistically significant increase in fractures with ICS compared with control (active treatment such as LABA, or placebo) over a mean of 90 days (risk of fractures 2.0% vs. 1.7%). The Peto odds ratio (OR) for fractures was 1.27 (95% confidence interval [CI] 1.01 to 1.58, p=0.04). However, the absolute risk increase was only modest. The authors estimated an NNH for fractures with ICS of 83 (95% CI 38 to 2107) over 3 years, based on the 5.1% fracture rates in the salmeterol and placebo arms from the large TORCH trial, which dominated the results of the pooled analysis. Current or ever use of ICS was also associated with a statistically significant increased risk of fractures compared with no current exposure in the observational studies (OR 1.21, 95% CI 1.12 to 1.32, p<0.001). From six of the observational studies, each 500 microgram increase in beclometasone dose equivalents was associated with a 9% relative increase in the likelihood of fractures (OR 1.09, 95% CI 1.06 to 1.12, p<0.001).

So what?
This study adds to the ongoing debate on whether ICS increase fracture risk, the evidence for which has been largely inconsistent. For example, a Cochrane Review of ICS in COPD reported that, from the few long-term studies which assessed effects on bone, generally no major effects on fractures or bone mineral density were found over three years. This large systematic review and meta-analysis of both RCTs and observational studies suggests that prolonged used of ICS (budesonide or fluticasone) can be associated with a modest increase in the risk of fractures in people with COPD. The authors estimated an NNH for fractures with ICS of 83 (95% CI 38 to 2107) over 3 years, based on the 5.1% fracture rates in the salmeterol and placebo arms from the large TORCH trial. The TORCH trial dominated the pooled results of this study. However, in itself, it did not find a clinically or statistically significant increase in the incidence of fractures with ICS (fluticasone) either alone or in combination with a LABA (salmeterol). However, there are several other limitations to the data in this systematic review and meta-analysis. These include the following:

  • The lower limit of the confidence intervals approached one (no effect)
  • The trials included came mainly from unpublished, non peer-reviewed company reports
  • Failure to take into account the use of systemic corticosteroids
  • The trials were not specifically designed to look at fractures (so the method of determining them may have been different)
  • Variable quality of most of the study data (from company reports and observational studies; concomitant medications such as bisphosphonate use might have had an important impact on the fracture risk reported)

The strength of this study is that it includes more data than previously, potentially underpowered, studies.

In conclusion, prescribers should continue to follow the NICE guideline on COPD when considering use of ICS. Any likely benefits of ICS should be weighed carefully against the risk of all potential systemic adverse effects, which should be discussed with patients. Whilst this study suggests that patients with COPD, who are taking ICS for prolonged periods, might be at increased risk of fracture, it would seem sensible to consider this in the context of patients’ overall risk of fracture, alongside the other risks of ICS, especially when high doses are used. A NICE guideline on the risk assessment of fragility fractures in people with or at risk of osteoporosis is underway and anticipated in 2012.

Study details

Loke YK, Cavallazzi R, Singh S. Risk of fractures with inhaled corticosteroids in COPD: systematic review and meta-analysis of randomised controlled trials and observational studies. Thorax 2011;66:699–708

Design
Systematic review and meta-analysis of 16 RCTs and 7 observational studies

Patients
People with COPD of any severity. 17,513 participants in the RCTs and 69,000 in the observational studies. In the RCTs, more than 50% were male.

Intervention and comparison
RCTs of at least 24 weeks’ duration (24 to 156 weeks): Fluticasone or budesonide versus a control treatment, in which the comparison groups consisted of ICS versus placebo, or ICS in combination with a LABA versus a LABA alone

Observational studies: comparing the risk of fractures with any ICS exposure with those without ICS exposure. ICS exposure in these studies was estimated from dispensing records.

Outcomes and results
See above under ‘What does this study claim?’

Sponsorship
Supported by the John Hopkins Clinical Research Scholars Program. The trial data included in this study mainly came from unpublished, non-peer-reviewed company reports.

Further information on the use of ICS in COPD can be found on NHS Evidence and in section 2.3 of the NPC e-learning event on COPD.

Please comment on this rapid review using our feedback form.

Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you