REDUCE

Published

Leuppi JD, et al. "Short-term vs conventional glucocorticoid therapy in acute exacerbations of chronic obstructive pulmonary disease: the REDUCE randomized clinical trial". JAMA. 2013. 309(21):2223-2231.
PubMed • Full text

Clinical Question

In patients with an acute COPD exacerbation, is 5 days of glucocorticoid treatment non-inferior to 14 days of glucocorticoid treatment in preventing repeat exacerbations?

Bottom Line

A 5-day course of glucocorticoids is non-inferior to a 14-day course for treatment of acute COPD exacerbations in prevention of re-exacerbations.

Major Points

The efficacy of systemic glucocorticoids in treatment of acute COPD exacerbations was demonstrated by the SCCOPE trial and a trial by Davies et al., both published in 1999. Parenteral and enteric admistration of equivalent doses of glucocorticoids were demonstrated as equivalent by de Jong and colleagues in 2007. While no large randomized trial has evaluated appropriate dosing of glucocorticoids, a 2010 prospective cohort study by Lindenauer et al. demonstrated that low dose enteral therapy had no difference in treatment failures when compared to high dose parenteral therapy in non-ICU patients. Likewise, no large randomized trial had demonstrated that a longer course was more efficacious than a short course though a 2011 Cochrane review suggested that there were no differences in treatment failures.

The 2013 Reduction in the Use of Corticosteroids in Exacerbated COPD (REDUCE) trial randomized 314 patients in Switzerland with an acute COPD exacerbation and a history of smoking to a 5- or 14-day course of glucocorticoids with additional glucocorticoids administered at the clinician's discretion. All patients received antibiotics, inhaled steroids, inhaled tiotropium, and inhaled short-acting beta-agonists. The shorter course was non-inferior to the longer course in rates of re-exacerbations at 180 days. Additionally, there was no difference in the use of open-label glucocorticoids. The 5-day group received less than half of the dose of glucocorticoids than the 14-day group. While there was no difference in glucocorticoid-related adverse events, the 5-day treatment was associated with a shorter hospitalization.

Of note, the question of corticosteroid dosing for COPD exacerbations in ICU patients was addressed in 2014 retrospective trial by Kiser and colleagues.

Guidelines

GOLD COPD (2013; before publication of REDUCE):

• Systemic corticosteroids in COPD exacerbations shortens recovery time, improves lung function, reduces risk of early relapse, and shortens hospital stay (Grade A)
• Prednisolone 30-40 mg daily are recommended for 10-14 days, though there is insufficient evidence regarding optimal duration of therapy (Grade D)

Design

• Multicenter, double-blind, randomized, non-inferiority controlled trial
• N=314
  • 5-day treatment (n=156)
  • 14-day treatment (n=155)
• Setting: 5 teaching centers in Switzerland
• Enrollment: 2006-2011
• follow-up: 6 months
• Analysis: Intention-to-treat
• Primary outcome: Rate of COPD re-exacerbation

Population

Inclusion Criteria

• Exacerbation of COPD, defined as two or more of:
  • Change in baseline dyspnea
  • Change in baseline cough
  • Change in sputum quantity or purulence
• Age >40 years
• Smoking history of ≥20 pack-years

Exclusion Criteria

• History of asthma
• FEV1/FVC ratio of >70% (evaluated by bedside post-bronchodilator spirometry immediately prior to randomization)
• Radiologic evidence of pneumonia
• Expected survival <6 months
• Pregnancy or lactation
• Inability to give written consent

Baseline Characteristics

From the 14-day treatment group. Comparisons are 14- vs. 5-day treatment.

• Demographics: Age 69.8 years, women 46.5% vs. 32.7%
• PMH: Smoker 40% (former 60%; 45 pack-years)
• Baseline health data: BP 138/80 mmHg, HR 90 BPM, O₂ saturation 90% (with nasal cannula 95%), leukocyte count 10.1x10³/uL
• Lung-specific: FEV₁ 31.3%, home O₂ 10.6%
  • GOLD:
    • Class I: 0%
    • Class II: 12.1%
    • Class III: 35.6%
    • Class IV: 52.3%
• Systemic glucocorticoid pretreatment: 18.5% (15 mg)
• Antibiotic pretreatment: 14.0%

Interventions

• Randomized to a group:
  • 5-day treatment - 5-days of glucocorticoids followed by 9 days of placebo
  • 14-day treatment - 14-days of glucocorticoids
• Glucocorticoid therapy was methylprednisolone 40 mg IV on the first day followed by prednisone 40 mg PO daily for the remaining days
• Additional treatment:
  • 7 days of antibiotics
  • Nebulized short-acting bronchodilator 4-6 times daily while in hospital
  • Tiotropium and inhaled glucocorticoid/long-acting beta-agonist combination inhalers daily
• Supplemental systemic glucocorticoids could be administered at the as seen fit by treating physician

Outcomes

Comparisons are 14-day vs. 5-day treatment. Outcomes are at 6 months unless otherwise stated.

Primary Outcomes

Rate of re-exacerbation

Defined as an acute clinical deterioration beyond usual day-to-day variation requiring interaction with a clinician. Analysis is non-inferiority.

36.8% vs. 35.9% (HR 0.95; 90% CI 0.70-1.29; P=0.006)

Per-protocol: 38.3% vs. 36.7% (HR 0.93; 90% CI 0.68-1.26; P=0.005)

Secondary Outcomes

Analyses are superiority.

All-cause mortality

8.4% vs. 7.7% (HR 0.93; 95% CI 0.40-2.20; P=0.87)

Need for mechanical ventilation

13.6% vs. 11.0% (OR 0.78; 95% CI 0.37-1.63; P=0.49)

Median time to open-label glucocorticoid treatment

9 vs. 6 days (P=0.52)

Any open-label treatment: 8.4% vs. 5.8% (OR 0.67; 95% CI 0.28-1.61; P=0.39)

Median cumulative prednisone dose

560 vs. 200 mg (P<0.001)

Mean: 793 vs. 379 mg (Difference in means -414; 95% CI -521 to -307; P<0.001)

Additional glucocorticoids during follow-up

Of the 144 and 148 patients with follow-up data.

44.4% vs. 36.5% (OR 0.72; 95% CI 0.45-1.15; P=0.19)

Median total dose: 250 vs. 258 mg (P=0.85)

Any infection

28.4% vs. 28.2% (OR 0.99; 95% CI 0.59-1.67; P>0.99)

New or worsening condition at discharge

Hyperglycemia: 57.4% vs. 56.9% (OR 0.98; 95% CI 0.58-1.66; P>0.99)

Hypertension: 17.8% vs. 11.6% (OR 0.61; 95% CI 0.28-1.29; P=0.22)

Other potential glucocorticoid adverse event

Defined as GI bleeding, symptomatic GERD, symptoms of HF or ischemic heart disease, sleep disturbance, fractures, or depression.

11.6% vs. 11.5% (OR 0.99; 95% CI 0.47-2.12; P>0.99)

Median hospital stay

9 vs. 8 days (HR 1.25; 95% CI 0.99-1.59; P=0.04)

Subgroup Analysis

For the primary outcome. Analysis are non-inferiority.

GOLD Classification

GOLD Class I and II: 33.3% vs. 26.1% (HR 0.73; 90% CI 0.28-1.88; P=0.10)

GOLD Class III: 35.9% vs. 33.3% (HR 0.93; 90% CI 0.52-1.67; P=0.08)

GOLD Class IV: 39.7% vs. 40.5% (HR 0.99; 90% CI 0.66-1.49; P=0.04)

Prior treatment with glucocorticoids

Yes: 46.4% vs. 45.7% (HR 0.93; 90% CI 0.50-1.72; P=0.09)

No: 33.3% vs. 35.8% (HR 0.88; 90% CI 0.61-1.26; P=0.006)

Criticisms

• Unclear significance of higher rate of women in the 14 day group
• High rate of severe or very severe COPD limits extrapolation to milder disease
• Non-smokers excluded, unclear if this group would have a similar outcome
• The duration of the follow-up period precludes long-term safety data in patients with multiple exacerbations
• Not powered to detect differences in the subgroups
• Aggressive treatment with adjunctive treatments like antibiotics, inhaled corticosteroids, inhaled long-acting anticholinergics, and inhaled beta-agonists limits generalizability

Funding

University Hospital Basel, Hospital Center of Biel-Bienne, Freie Akademische Gesellschaft, AstraZeneca, Viollier Laboratory, Fond fur Lehre und Forschung, and Gottfried und Julia Bangerter-Rhyner-Stiftung fur Medizinische Forschung.