• 
• 

Three Part Question

Does the [adjuvant use of steroids with antibiotics] improve [mortality or reduce hospital length of stay] in [adult patients with Community Acquired Pneumonia]

Clinical Scenario

A 38 year-old teacher presents to the ED with bilateral pneumonia. On arrival he is confused, SpO2 is 91% on air, RR is 38/min and BP is 105/59mmHg. You start treatment for severe CAP with a CURB-65 score of ≥3 and inform ITU.

Would the addition of steroids to antibiotics improve this patient’s mortality and shorten his length of stay in hospital?

Search Strategy

Interface used: www.evidence.nhs.uk
Database searched:
CINAHL, Medline, EMBASE, Cochrane Database of Clinical Trials
(((pneumonia OR bronchopneumonia OR (lower AND respiratoryAND tract AND infection)) AND (steroids OR prednisolone OR hydrocortisone OR dexamethasone OR glucocorticoids)))

Search Outcome

8 Clinical Trials identified

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Wagner, H. N. Jr, Bennett, I. L. Jr, Lasagna, L. Cluff, L. E. Rosenthal Mb, Mirick G. S. 1956 United States of America	113 patients with proven pneumococcal pneumonia	52 patients received Penicillin G and Hydrocortisone 61 patients received Penicillin G and Placebo	Steroids provided symptomatic relief but it did not reduce mortality or length of stay		Small number of patients Flawed randomisation and blinding No intention to treat No power calculation and statistical tests
Kirby, K. I. R. B. Y. J. C., Jr. Polis, P. O. L. I. S. G. Romansky, R. O. M. A. N. S. K. Y. M. J. 1960 USA	Patient admitted to a medical ward with Community Acquired Pneumonia	42 patients, 21 patients received Tetracycline 6-Methylprednisolone in identical capsules, 21 patients had Tetracycline alone in identical capsules	Steroids provided symptomatic relief but did not reduce mortality or length of stay in hospital		Small numbers, non-randomized single blinded trial, no statistical tests prior to conducting trial. One patient in steroids group developed empyema and died at 8 weeks after the trial.
McHardy, V. U. Schonell, M. E. 1972 UK	126 patients with Community Acquired Pneumonia admitted to a respiratory ward in a teaching hospital in UK	43 patients received Ampicillin 2g alone and 20 patients received Ampicillin 2g and Prednisolone 20mg/od for 5 days 43 patients received Ampicillin 1g and 20 patients received Ampicillin 1g with Prednisolone 20mg/od for 5 days	Steroids did not improve length of stay or mortality rate in patients with Community Acquired Pneumonia		Complex design increases susceptibility to multiple hypothesis bias. It has poor internal validity and is prone to selection, assessment and attrition bias. It has small numbers. It also has poor external validity as the exclusion criteria were extensive making it difficult to generalise to emergency department patients.
Marik, P. Kraus, P. Sribante, J. Havlik, I. Lipman, J. Johnson, D. W. 1993 United States of America	30 patients with severe CAP admitted to ICU	16 patients received hydrocortisone 10mg/kg 30 min prior to antibiotics 14 patients received placebo 30 min prior to antibiotics	Hydrocortisone given prior to antibiotic treatment had no effect on the clinical course of pneumonia		This was an explanatory randomised placebo-controlled trial. The exclusion criteria were quite extensive making it difficult to generalise to Emergency Department practice. Small number of patients randomised makes the study underpowered.
Mikami, K. Suzuki, M. Kitagawa, H. Kawakami, M. Hirota, N. Yamaguchi, H. Narumoto, O. Kichikawa, Y. 2007 Japan	31 patients with CAP requiring hospital admission in a medical ward in a General District Hospital	16 patients received IV antibiotics plus 40mg of iv prednisolone for 3 days 15 patients received IV antibiotics alone	No deaths observed during the trial Steroids did not reduce length of stay	Reduction in length of stay p=0.182	Small numbers, it is a cooperative trial therefore no blinding. Nearly 50% of eligible patients were not recruited in the trial. Also there were no deaths observed. This indicates that authors choose less unwell patients to participate in the trial.
Snijders, D. Daniels, J. M. de, C. S. Cs van, T. S. Ts Boersma, W. G. 2010 The Netherlands	213 patients age 18 or over with CAP requiring admission to hospital were recruited	104 patients received antibiotics and 40mg of prednisolone for 7 days 109 patients received antibiotics and placebo for 7 days	Prednisolone did not improve Mortality or length of stay in hospitalised patients with Community Acquired Pneumonia	30 day mortality p=0.93, Length of stay p=0.16	Small numbers. Patients with COPD were included in analysis. There is lack of clarity created by the use of a Kaplan-Meyer curve to analyse length of stay. It is unclear how the investigators dealt with deaths or transfers to Intensive Care Unit as this would have resulted in a drop in the graph potentially skewing the final result.
Fernandez-Serrano, Silvia Dorca, Jordi Garcia-Vidal, Carolina Fernandez-Sabe, Nuria Carratala, Jordi 2011 Spain	56 patients were recruited in the trial if they had severe pneumonia, ie at least two lobe involvement and respiratory failure	28 patients received Ceftriaxone + Levofloxacin and 200mg of Methyl-prednisolone 30 min prior to antibiotics followed by a reducing dose maintenance steroid regime 28 patients received Cetriaxone+ Levofloxacin and Placebo 30 min prior to antibiotics followed by a maintenance dose mirroring the steroid regime	Methyl-prednisolone treatment in combination with antibiotics does not improve mortality or shorten length of stay in patients with community acquired pneumonia		Small numbers of patient recruited. Extensive exclusion criteria makes the result difficult to apply to everyday practice.
Meijvis, S. C. A. Hardeman, H. Remmelts, H. H. F. Heijligenberg, R. Rijkers, G. T. Van, H. Voorn, G. 2011 The Netherlands	304 patients age 18 years or older with CAP requiring admission	151 patients received antibiotics and dexamethasone 5mg od for 4 days 153 patients received antibiotics and placebo od for 4 days	Dexamethasone reduces the length of stay by 1 day but does not improve the mortality in patients with community acquired pneumonia	30 day mortality p=0.68 length of stay p=0.048	Significant loss to follow up, nearly 50% of patients did not have a review at 30 days. Not a multi-centre or multi-country study.

Comment(s)

8 clinical trials were identified involving a total of 915 patients spanning nearly 60 years. The cumulative findings of these trials are as follows: Mortality Six studies showed no difference in mortality, one study showed a trend in increasing mortality and one study showed a trend in mortality reduction. This clearly confirms the conclusion that steroids in addition to antibiotics do not reduce mortality in patients with Community Acquired Pneumonia. Length Of Stay One study showed no difference, six studies showed a trend in reducing Length Of Stay that was not statistically significant. One study of excellent quality that had >150 patient in study group showed a reduction by one day that was statistically significant. However it also demonstrated that patients on steroids had more side effects such as hyperglycaemia and potentially life-threatening gastric perforation. There is therefore a consistent finding of a trend for reduction in Length Of Stay with use of steroids. As the risk of type 2 error reduces with increasing sample size, it becomes statistically significant that steroids do reduce Length Of Stay, albeit by a modest reduction of one day. However it is questionable whether the benefit of using steroids to reduce Length Of Stay by one day outweighs the risk of serious side effects.

Clinical Bottom Line

Current evidence does not support the routine use of steroids in addition to antibiotics for patients presenting to the Emergency Department with Community Acquired Pneumonia.

References

1. Wagner, H. N. The effect of hydrocortisone upon the course of pneumococcal pneumonia treated with penicillin Bulletin of the Johns Hopkins Hospital Issue 3, Pages 197-215
2. Kirby, K A controlled blind study of pneumococcal pneumonia treated with tetracycline and tetracycline plus 6-methyl prednisolone American Journal of the Medical Sciences Vol 240, Pages 64-9
3. McHardy, V Ampicillin dosage and use of prednisolone in treatment of pneumonia: co-operative controlled trial British medical journal Volume 4, Issue 5840, Pages 569-73
4. Marik, P. Hydrocortisone and tumor necrosis factor in severe community-acquired pneumonia: A randomized controlled study Chest 1993, Vol 104, Issue 2, Pages 389-392
5. Mikami, K Efficacy of corticosteroids in the treatment of community-acquired pneumonia requiring hospitalization Lung Year 2007; Vol 185, Issue 5, Pages 249-255
6. Snijders, D Efficacy of corticosteroids in community-acquired pneumonia: a randomized double-blinded clinical trial American Journal of Respiratory & Critical Care Medicine Year 2010; Vol 181, Issue 9, Pages 975-983
7. Fernandez-Serrano, Silvia Effect of corticosteroids on the clinical course of community-acquired pneumonia: a randomized controlled trial Critical Care 2011; Vol 12, Issue 2, Pages R96
8. Meijvis, S Dexamethasone and length of hospital stay in patients with community-acquired pneumonia: A randomised, double-blind, placebo-controlled trial The Lancet 2011; Vol 377, Issue 9782, Pages 2023-2030