We would like to thank Sánchez et al for their interest in the article in which we perform outside validation of the CPU-65 index (use of aspirin, diabetes, pain type, 65 years or older)1 that we have previously described.2 These authors, after studying 1000 patients, propose an index similar to ours that includes diabetes, pain type (oppressive and retrosternal) and age (>40 years), substituting the use of aspirin with previous coronary artery disease.3 The authors obtained a sensitivity, specificity, positive predictive value and negative predictive value of 100% for the detection of acute coronary syndrome. However, when using our index in the broadened population of 4221 patients, these parameters change to 98%, 23%, 17%, and 98%, respectively. The only logical explanation for this great discrepancy is that the use of aspirin does not figure in the list of clinical variables collected by these authors.3 We believe that it is preferable to use the variable "aspirin use" since it is easy to obtain in the history and includes patients with peripheral artery disease. In addition to this, the CPU-65 index is associated with extension of coronary artery disease.4
Half of all patients who present to the emergency department with chest pain have a very low risk profile and do not require testing for detection of ischaemia. In daily practice, this is how it is done, though not according to any protocol. For example, in the Sánchez group, 480 (48%) out of 1000 patients with chest pain were initially categorised as "without acute coronary syndrome" and they did not undergo testing to detect ischaemia.5 In fact, these authors only performed stress testing on 144 patients (14%). The CPU-65 index is useful in detecting patients with a risk that is so low that testing for the detection of ischaemia would be dubious or not cost effective. Indeed, Sekhri et al, studying 4873 patients without previous coronary artery disease who were admitted to chest pain units, have shown that stress testing adds little prognostic value to clinical variables of diabetes, pain type, age, and male gender.6