A study featured in BMJ (British Medical Journal) reveals that it is possible to identify patients with a high risk of coronary artery disease, who may require further diagnostic work, by using a new risk prediction tool, which is more accurate than existing models, and which can easily be integrated into electronic patient records or mobile applications.

Coronary artery disease is one of the leading causes of mortality worldwide. The condition occurs when fatty deposits restrict the arteries that supply the heart with oxygen and nutrients.

Due to the fact that the first sign of coronary artery disease usually consists of chest pain, current guidelines recommend using either the Diamond & Forrester model or the Duke Clinical Score as prediction tools for doctors to estimate whether they have a risk-patient who needs further tests when they assess patients with chest pain or not. Scientists have raised the questions of accuracy with regard to these tools.

A team of European of researchers designed an improved prediction model, which uses a range of variables that is known to be associated with coronary artery disease to evaluate data from 18 hospitals across Europe and the US, which included a total of 5,677 patients with chest pain, who had no previous history of heart disease. The total number of patients consisted of 3,283 men and 2,394 women.

The team developed three types of models; the ‘basic’ model predicted coronary artery disease according to symptoms, age and sex, whilst the second, ‘clinical’ model contained risk factors like hypertension, diabetes, smoking and elevated lipid levels. The third, ‘extended’ model also included the coronary calcium score, a measure of calcium in the coronary arteries, which is linked to coronary artery disease and which is a marker to determine the risk of a coronary event.

The findings indicate that the currently recommended Duke Clinical Score as per NICE guideline substantially overestimates the possibility of coronary artery disease.

The teams ‘clinical’ model proved to be superior to the Duke Clinical Score estimates, by predicting a range from between 2% probability of coronary artery disease for a 50 year old female with non-specific chest pain and no risk factors, to a 91% probability for an 80 year old, multiple-risk factor male with typical chest pain. Tests with the ‘extended’ model, i.e. including the coronary calcium score, demonstrated a further improved prediction event.

Furthermore, the new model does not require resting heart (ECG) readings, which is more appropriate for using the model in primary practice. The team also developed an online calculator, which can be implemented easily into electronic patient records or mobile applications.

The team concludes, in light of the NICE guidelines need for an accurate estimate of the probability of coronary artery disease, that their model “allows doctors to make better decisions as to which diagnostic test is best in a particular patient and to decide on further management based on the results of such tests.”

Written By Petra Rattue