Report by Amy Au-Yong

Checked by Tim Coats, Professor of Emergency Medicine

Institution: Emergency Department, Leicester Royal Infirmary, Leicester, UK

A short-cut review was carried out to establish whether C-reactive protein (CRP) has any utility in the differential diagnosis of heart failure and pneumonia. A total of 601 papers was found using the reported searches, of which three presented the best evidence to answer the clinical question. The author, date and country of publication, patient group studied, study type, relevant outcomes, results and study weaknesses of these best papers are tabulated. It is concluded that there is no evidence for the use of CRP to discriminate between a diagnosis of pneumonia and heart failure in patients who have acute shortness of breath.

CLINICAL SCENARIO

A 70-year-old man presents to the emergency department with acute shortness of breath. You wonder whether sending a sample of his blood for CRP levels will help differentiate between the diagnoses of pneumonia or heart failure.

THREE-PART QUESTION

In [patients presenting as an emergency with acute shortness of breath] do [C Reactive Protein levels] help in [differentiating between the diagnoses of chest infection or heart failure]?

SEARCH STRATEGY

Medline using the OVID interface 1966 to September week 4 2008: {[c reactive protein.mp. or exp C-Reactive Protein or CRP]} AND {[exp respiratory tract infections OR exp respiratory tract diseases OR exp pneumonia OR pneumonia.mp]} OR {[exp dyspnea OR dyspnoea.mp]} OR {[exp heart failure, congestive OR exp ventricular dysfunction, right OR exp ventricular dysfunction, left OR heart failure.mp OR exp heart failure/]} LIMIT TO {(humans and English language and (“diagnosis (sensitivity)” “all adult (19 plus years)” and abstracts.

OUTCOME

A total of 601 papers was found of which three were found to be relevant, one of which was a systematic review that contained data from other relevant papers (see table 1).

COMMENTS

Acute shortness of breath is a common presentation to the emergency department (ED) and has a long list of differential diagnoses. Two of the common diagnoses are pneumonia and heart failure. If a simple blood test for CRP levels could differentiate between the two this would aid clinical practice both for doctors in the ED and acute physicians. Following a search of the evidence no papers were found that looked at this relationship directly, therefore the search was widened and evidence taken from studies examining different inclusion criteria and different outcomes to the one we specifically wanted. This evidence was examined to see whether it could be extrapolated to provide the best answer to our question. Many more papers were found on the use of CRP in the diagnosis of pneumonia than there were for that of heart failure (although the role of CRP in myocardial infarction is being extensively studied). The best level of evidence found for the use of CRP in the diagnosis of pneumonia was from a systematic review, which found that the area under the receiver operator characteristic curve was 0.80, which shows that it is neither a particularly good or bad test. The best evidence found for the use of CRP in the diagnosis of heart failure was a cross-sectional diagnostic study, which found that CRP had limited diagnostic value. Unfortunately, this study was performed in primary care and it is difficult to extrapolate these findings to the ED. Our findings do not allow us to give a definitive answer to our three-part question.