PT - JOURNAL ARTICLE AU - Stefan Bösner AU - Annette Becker AU - Maren Abu Hani AU - Heidi Keller AU - Andreas C Sönnichsen AU - Jörg Haasenritter AU - Konstantinos Karatolios AU - Juergen R Schaefer AU - Erika Baum AU - Norbert Donner-Banzhoff TI - Accuracy of symptoms and signs for coronary heart disease assessed in primary care AID - 10.3399/bjgp10X502137 DP - 2010 Jun 01 TA - British Journal of General Practice PG - e246--e257 VI - 60 IP - 575 4099 - http://bjgp.org/content/60/575/e246.short 4100 - http://bjgp.org/content/60/575/e246.full SO - Br J Gen Pract2010 Jun 01; 60 AB - Background Diagnosing the aetiology of chest pain is challenging. There is still a lack of data on the diagnostic accuracy of signs and symptoms for acute coronary events in low-prevalence settings.Aim To evaluate the diagnostic accuracy of symptoms and signs in patients presenting to general practice with chest pain.Design of study Cross-sectional diagnostic study with delayed-type reference standard.Setting Seventy-four general practices in Germany.Method The study included 1249 consecutive patients presenting with chest pain. Data were reviewed by an independent reference panel, with coronary heart disease (CHD) and an indication for urgent hospital admission as reference conditions. Main outcome measures were sensitivity, specificity, likelihood ratio, predictive value, and odds ratio (OR) for non-trauma patients with a reference diagnosis.Results Several signs and symptoms showed strong associations with CHD, including known vascular disease (OR = 5.13; 95% confidence interval [CI] = 2.83 to 9.30), pain worse on exercise (OR = 4.27; 95% CI = 2.31 to 7.88), patient assumes cardiac origin of pain (OR = 3.20; 95% CI = 1.53 to 6.60), cough present (OR = 0.08; 95% CI = 0.01 to 0.77), and pain reproducible on palpation (OR = 0.27; 95% CI = 0.13 to 0.56). For urgent hospital admission, effective criteria included pain radiating to the left arm (OR = 8.81; 95% CI = 2.58 to 30.05), known clinical vascular disease (OR = 7.50; 95% CI = 2.88 to 19.55), home visit requested (OR = 7.31; 95% CI = 2.27 to 23.57), and known heart failure (OR = 3.53; 95% CI = 1.14 to 10.96).Conclusion Although individual criteria were only moderately effective, in combination they can help to decide about further management of patients with chest pain in primary care.