Study objective: To evaluate the diagnostic accuracy of Hoover sign for detecting obstructive airway disease (OAD), compared with wheezes, rhonchi, reduced breath sounds, and clinical impression, and to analyze the observer agreement on these signs.
Design: Prospective, blind comparison, with a reference standard (spirometry) among a consecutive series of patients.
Setting: Outpatient pulmonary clinic.
Patients: One hundred seventy-two patients (117 men [68%] and 55 women [32%]), who were > 40 years of age, had not been known previously by the participating physicians, and met at least one of the following criteria: smokers of > 20 pack-years; patients who had received a diagnosis of or had self-reported COPD (or chronic bronchitis or emphysema); patients who had received inhaler bronchodilator treatment for > 6 months; or patients with any degree of dyspnea.
Measurements: Patients were examined by a first-year resident in family medicine and by a pulmonologist. Spirometry was performed by a blinded trained technician. OAD was defined as an FEV1/FVC ratio of < 0.70.
Results: OAD was present in 64 patients (37%). Observer agreement (κ statistic) was 0.74 for Hoover sign and was lower for the rest of the signs. Hoover sign had a sensitivity of 58% and a specificity of 86% for detecting OAD, and it had a positive likelihood ratio of 4.16, which was higher than that of the other signs.
Conclusions: Hoover sign, a frequently forgotten sign, is easy to recognize and is useful for detecting OAD.