Although chest radiographs can confirm physical examination findings, little is known about the utility of the physical examination alone in predicting pulmonary disease or in distinguishing among pulmonary conditions. Increasing time constraints on patient visits and the immediate availability of radiography have led physicians to question the need for a detailed lung examination when lower respiratory infection is suspected. Wipf and associates conducted a prospective study to assess the interobserver reliability of physical examination alone in diagnosing pneumonia in patients with suspected lower respiratory infection.

Patients who presented to the emergency department with signs of lower respiratory infection, including an acute cough and copious or dark-colored sputum, were eligible for the study. Each patient was examined sequentially by at least two or three board-certified physicians who had no knowledge of the patient's history, vital signs or radiographic findings. The physicians were not allowed to ask the patient questions. Patients were examined in the sitting and right- and left-lateral decubitus positions. In the sitting position, patients were examined for rales, rhonchi, wheezing, tactile fremitus and other signs consistent with pneumonia. In the left and right lateral decubitus positions, patients were examined for rales. Chest radiographs were obtained for all patients and were considered the gold standard for confirming the diagnosis; physical findings were considered correct if they were present in the same general areas as infiltrates on the chest radiographs.

Fifty-two male patients, who were generally elderly with a history of smoking and asthma or chronic obstructive pulmonary disease, were enrolled in the study. Of these, 24 had pneumonia confirmed by chest radiographs, and 28 did not have pneumonia. The latter group was believed to have some type of bronchitis. In four of the patients with pneumonia, the chest radiographs were interpreted as “possible pneumonia.” Most cases of pneumonia were located in the right lower lobe of the lung.

Physical examination of the chest took about 10 minutes. The two most frequent abnormal findings in all patients were rales in the sitting position (22 to 65 percent) and bronchial breath sounds (8 to 43 percent). Other chest findings were uncommon. Physician consistency was highest for rales. Overall, sensitivity and specificity of physical findings varied considerably among physicians, as well as for a given physician in eliciting findings between the right and left lungs. The most consistently helpful maneuvers were auscultating for rales with the patient in the sitting position (highest sensitivity) and performing auscultatory percussion for egophony and rales in the left-lateral decubitus position (highest specificity).

The authors conclude that the physical examination had moderate sensitivity and specificity in determining the presence of pneumonia and identifying the affected site. Physical examination alone was not sufficiently accurate to confirm or exclude the diagnosis of pneumonia. Chest radiographs remain the best way to confirm the diagnosis.