Skip to main content

05-05-2013 | General practice | Article

Primary care pneumonia diagnosis effective


Full free text

medwireNews: Study results published in the BMJ confirm that history-taking and physical examination alone allow general practitioners (GPs) to accurately identify patients at high risk for pneumonia.

Additionally, the authors show that serum C-reactive protein (CRP) concentrations can be used to exclude the infection when doubt remains after symptom assessment.

"Although the diagnostic 'symptoms and signs' model presented in this study assigned an intermediate diagnostic risk of pneumonia to most patients, history taking and physical examination alone enabled general practitioners to correctly identify a small group of patients at high risk," say Berna Broekhuizen (Julius Center for Health Sciences and Primary Care, Utrecht, the Netherlands) and colleagues.

The study included 2820 patients who presented to their GP with acute cough in winter between 2007 and 2010. Of these patients, 140 (5%) had radiographically confirmed pneumonia.

The authors found that the absence of runny nose and presence of breathlessness, crackles and diminished breath sounds on auscultation, tachycardia (>100 bpm), and fever (temperature ≥37.8°C) all had independent diagnostic value. From these findings they created a "signs and symptoms" model based on an optimum combination of these factors.

The signs and symptoms model correctly identified patients with a low (<2.5%) risk for pneumonia; only 11 (2%) of the 665 patients in this group had radiographic pneumonia. And among 63 patients deemed at high (>20%) risk for pneumonia, 24 (38%) had their diagnosis confirmed.

The average CRP concentration among patients with pneumonia was 69 mg/L in patients with pneumonia compared with 19 mg/L in those without. And, when the authors added CRP concentration, either as a continuous variable or as a threshold cut-off (30 mg/L) to the symptoms and signs model, it significantly improved its discriminatory power.

After applying a cut-off value of CRP <30 mg/L, the authors found that among the 2092 (74.2%) patients in the intermediate risk group, 957 (48.2%) were correctly re-classified as low-risk, and 64 (3.2%)were incorrectly classified as high-risk out of 1987 patients who did not have pneumonia. Conversely, 27 (25.7%) of 105 patients with pneumonia were reclassified as low-risk, and 22 (21.0%) were reclassified as high-risk.

The authors also note that while procalcitonin levels were higher in patients with pneumonia than in those without, they added little additional value to the researchers' model.

They conclude that CRP levels can be used to appropriately exclude pneumonia when there is diagnostic doubt.

"In… more severely ill patients, point of care tests, including CRP, do not seem to be useful," they write.

However, "in patients with a low risk of pneumonia based on symptoms and signs, it seems justified to withhold further diagnostic investigation and not to treat with antibiotics."

medwireNews ( is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2013

By Kirsty Oswald, medwireNews Reporter

Related topics