Print this post Print this post  




Corticosteroids may reduce length of hospital stay but do not reduce mortality in community-acquired pneumonia

MeReC Rapid Review NPC Logo

18 July 2011

This randomised controlled trial (RCT) found that treatment with 5mg intravenous (IV) dexamethasone daily for four days on admission to hospital with community-acquired pneumonia reduced the median length stay by one day compared with placebo. However, it did not have any significant effect on in-hospital mortality or mortality at 30 days.

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

The role of corticosteroids in the management of community-acquired pneumonia remains unclear, and they aren’t generally recommended. Clinicians should continue to follow appropriate clinical guidelines, such as the BTS guidelines for the management of community-acquired pneumonia in adults, summarised in MeReC Rapid Review No. 674.

What is the background to this?
Community-acquired pneumonia is associated with significant morbidity, and has a mortality rate of 5–35%, depending on severity of disease.1 The aim of treatment with corticosteroids in pneumonia is to reduce systemic inflammation, without adversely affecting the local pulmonary cytokine response required to clear the infection. Trials have produced variable results and there is no consensus on the use of corticosteroids in community-acquired pneumonia, in terms of whether they are of benefit, whether their use should be targeted at patients with severe illness, or the optimum length and dose of treatment if they are prescribed.

What does this study claim?
This study assessed the effect of dexamethasone 5mg IV versus placebo once daily for four days in 304 patients hospitalised with community-acquired pneumonia. All patients received IV antibiotics prior to receiving the study drug. The first dose of study drug was administered within 12 hours of admission.

The primary outcome was length of hospital stay. Dexamethasone was associated with a statistically significant one day reduction in median hospital stay (6.5 days vs. 7.5 days, (95%CI of difference in medians 0 to 2 days; P=0.0480). However, no significant differences were found in mortality during hospital stay (8 [5%] vs. 8 [5%], P=0.98) mortality at 30 days (9 [6%] vs. 11 [7%], P=0.68), or intensive care admissions (7 [5%] vs. 10 [7%], P=0.47).

Severe adverse events were infrequent and rates did not differ between the groups. However, there was a significant increase in the risk of hyperglycaemia in the dexamethasone group (67 [44%] vs. 35 [23%], P< 0.0001, NNH 4), although the number of patients in each group requiring treatment for hyperglycaemia was not significantly different (7 [5%] vs. 5 [3%], P=0.57). One patient who received dexamethasone developed a gastric perforation that could be attributed to use of this drug.

How does this relate to other studies?
A recent Cochrane review (including 4 trials in 235 adults with pneumonia) found no significant mortality benefit with either hydrocortisone or prednisolone as adjunctive treatment in pneumonia. The authors of the review concluded that the evidence from the included studies was not strong enough to make any recommendations.

In a hospital-based RCT of 213 adults with community–acquired pneumonia, oral prednisolone 40mg once daily for one week did not improve clinical outcomes when compared with placebo, with no significant differences in mortality or in 7 day clinical cure rates between the two groups. Corticosteroid treatment was associated with increased late treatment failure (defined as a recurrence of signs and symptoms after 72 hours admission, following an initial beneficial response to treatment; 20 [19.2%] vs. 10 [9.2%], P=0.04, NNH 10), and a non-significant lower 30 day clinical cure rate (69 [66.3%] vs. 84 [77.1%], P=0.08).

So what?
Reduction of length of hospital stay, symptom resolution and reduction in mortality following infection are all important clinical goals in the treatment of patients with community-acquired pneumonia. This is the largest published trial of corticosteroids in community-acquired pneumonia; many of the previous trials conducted were small and underpowered. It is difficult to compare trial results because different corticosteroid drugs and regimens have been used, and different clinical outcomes have been measured. However, this study, like the earlier studies discussed above, found no reduction in mortality associated with corticosteroid use.

This study did not detect the reduction in length of stay it was powered to achieve (two days), but the authors argue that a one day reduction in length of stay is clinically relevant. This may also be seen as important in health economic terms. However, any benefits gained from corticosteroid use must be balanced against their potential harms (such as superinfections and gastric disturbances).

The findings of this study may not be applicable to a large number of patients with community-acquired pneumonia, including those who have chronic obstructive airways disease (COPD). Around 170 (21% of all patients screened) patients with COPD were assessed for study eligibility. However, only 34 (4% of all patients screened) were included in the study, as patients taking or requiring systemic corticosteroids were excluded, and these are used in patients with COPD to reduce bronchial obstruction in pneumonia.

More studies are required to establish if corticosteroids are of benefit in community-acquired pneumonia, as well as which steroid, in which patients, and for how long.

Study details
Meijvis SCA, et al. Dexamethasone and length of hospital stay in patients with community-acquired pneumonia: a randomised, double-blind, placebo-controlled trial. Lancet 2011;377:2023–30

A double-blind, placebo-controlled trial of adults aged 18 years or older with confirmed community-acquired pneumonia who presented to emergency departments of two teaching hospitals in the Netherlands. The study was powered to detect a two day reduction in median length of hospital stay.

304 adults with community-acquired pneumonia.143 (47%) patients had pneumonia of pneumonia severity index class 4–5; 79 (52%) patients in the dexamethasone group and 64 (42%) controls. Patients were ineligible if they were immunocompromised, needed immediate transfer to an intensive-care unit (ICU), or were already receiving corticosteroids or immunosuppressive drugs.

Intervention and comparison
Patients were randomised to receive either 5mg IV dexamethasone (151 patients) or placebo (153 patients) once a day for 4 days from admission. Admission to ICU was defined as an end-point of the study (as all patients admitted to ICU were given corticosteroids), and so the study could not assess the affect of dexamethasone on mechanical ventilation.

Outcomes and results

The primary outcome was length of hospital stay. Dexamethasone was associated with a statistically significant one day reduction in median hospital stay (6.5 days vs. 7.5 days, (95%CI of difference in medians 0 to 2 days; P=0.0480). However, no significant differences were found in mortality during hospital stay (8 [5%] vs. 8 [5%], P=0.98) mortality at 30 days (9 [6%] vs. 11 [7%], P=0.68), intensive care admissions (7 [5%] vs. 10 [7%], P=0.47), hospital readmissions (7 [5%] vs. 7 [5%], P=0.98), empyema or pleural effusion (7 [5%] vs. 5 [3%], P=0.54).

The authors assessed quality of life at day 3 (n=209) and day 30 (n=95) using the RAND-36 survey. Quality of life was similar in the two groups at day 3, with significant improvement seen only in social functioning in the dexamethasone group at day 30 (P=0.0091), which represents 2 of the 36 items of the questionnaire.



  1. 1. Loeb M. Community-acquired pneumonia. Clin Evid (Online); 2010. pii: 1503.

Further information can be found on NHS Evidence.

Please comment on this rapid review in the NPC discussion rooms, or 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