Prognostic Models for Stable Coronary Artery Disease
Prognostic Models for Stable Coronary Artery Disease
Aims: The population with stable coronary artery disease (SCAD) is growing but validated models to guide their clinical management are lacking. We developed and validated prognostic models for all-cause mortality and non-fatal myocardial infarction (MI) or coronary death in SCAD.
Methods and results: Models were developed in a linked electronic health records cohort of 102 023 SCAD patients from the CALIBER programme, with mean follow-up of 4.4 (SD 2.8) years during which 20 817 deaths and 8856 coronary outcomes were observed. The Kaplan–Meier 5-year risk was 20.6% (95% CI, 20.3, 20.9) for mortality and 9.7% (95% CI, 9.4, 9.9) for non-fatal MI or coronary death. The predictors in the models were age, sex, CAD diagnosis, deprivation, smoking, hypertension, diabetes, lipids, heart failure, peripheral arterial disease, atrial fibrillation, stroke, chronic kidney disease, chronic pulmonary disease, liver disease, cancer, depression, anxiety, heart rate, creatinine, white cell count, and haemoglobin. The models had good calibration and discrimination in internal (external) validation with C-index 0.811 (0.735) for all-cause mortality and 0.778 (0.718) for non-fatal MI or coronary death. Using these models to identify patients at high risk (defined by guidelines as 3% annual mortality) and support a management decision associated with hazard ratio 0.8 could save an additional 13–16 life years or 15–18 coronary event-free years per 1000 patients screened, compared with models with just age, sex, and deprivation.
Conclusion: These validated prognostic models could be used in clinical practice to support risk stratification as recommended in clinical guidelines.
Population ageing and improvements in survival after acute coronary syndromes (ACS) have contributed to worldwide increases in the number of patients with stable coronary artery disease (SCAD). Stable coronary artery disease encompasses a heterogeneous spectrum of syndromes including patients with stable angina and those who have become stable after ACS. In the USA alone, over 16 million people (7% of the US population) suffer from coronary disease with 500 000 new stable angina cases being reported each year. The 2012 ACCF/AHA guidelines for prognostication in SCAD recommend that patients are stratified into high (>3%), intermediate (1–3%), and low (<1%) annual mortality risk groups, which then define different investigation and treatment pathways.
So far, prognostic models proposed for SCAD (compared in Supplementary material online, Table S1) have been based on data collected for research purposes rather than the information that clinicians record in real-world practice, and none has been recommended for clinical use. Among their limitations is the failure to incorporate the 'broad range of relevant data' identified as important in the guidelines, such as sociodemographic characteristics, cardiovascular (CVD) and non-CVD comorbidities, mental health, symptom severity, and clinically available biomarkers. These data are routinely recorded for most patients before more costly information becomes available from further, often invasive, investigations. Other limitations of previous models include the use of selected samples from trials or voluntary registries, and covering a narrow range of SCAD such as excluding or being confined to post-ACS patients. Furthermore, the ACCF/AHA guidelines emphasize the importance of both all-cause mortality and coronary events as outcomes, but no previous studies have assessed both. Importantly, none of the previous models has been validated in an external data set.
We sought to address these limitations by analysing large-scale, population-based, linked electronic health records data. Our objective was to develop and validate the performance of prognostic models that incorporate clinical measures recommended in guidelines and are commonly available in patients with SCAD. Following recent methodological guidance, we assessed the accuracy of the predictions from the prognostic models based on their calibration, discrimination (C-index), and reclassification improvement. To further evaluate potential clinical benefits, we estimated life years saved when the models are used to guide management decisions. Based on these analyses, we developed prognostic models to predict all-cause mortality and non-fatal myocardial infarction (MI) or coronary death in SCAD and evaluated their performance in an independent data set.
Abstract and Introduction
Abstract
Aims: The population with stable coronary artery disease (SCAD) is growing but validated models to guide their clinical management are lacking. We developed and validated prognostic models for all-cause mortality and non-fatal myocardial infarction (MI) or coronary death in SCAD.
Methods and results: Models were developed in a linked electronic health records cohort of 102 023 SCAD patients from the CALIBER programme, with mean follow-up of 4.4 (SD 2.8) years during which 20 817 deaths and 8856 coronary outcomes were observed. The Kaplan–Meier 5-year risk was 20.6% (95% CI, 20.3, 20.9) for mortality and 9.7% (95% CI, 9.4, 9.9) for non-fatal MI or coronary death. The predictors in the models were age, sex, CAD diagnosis, deprivation, smoking, hypertension, diabetes, lipids, heart failure, peripheral arterial disease, atrial fibrillation, stroke, chronic kidney disease, chronic pulmonary disease, liver disease, cancer, depression, anxiety, heart rate, creatinine, white cell count, and haemoglobin. The models had good calibration and discrimination in internal (external) validation with C-index 0.811 (0.735) for all-cause mortality and 0.778 (0.718) for non-fatal MI or coronary death. Using these models to identify patients at high risk (defined by guidelines as 3% annual mortality) and support a management decision associated with hazard ratio 0.8 could save an additional 13–16 life years or 15–18 coronary event-free years per 1000 patients screened, compared with models with just age, sex, and deprivation.
Conclusion: These validated prognostic models could be used in clinical practice to support risk stratification as recommended in clinical guidelines.
Introduction
Population ageing and improvements in survival after acute coronary syndromes (ACS) have contributed to worldwide increases in the number of patients with stable coronary artery disease (SCAD). Stable coronary artery disease encompasses a heterogeneous spectrum of syndromes including patients with stable angina and those who have become stable after ACS. In the USA alone, over 16 million people (7% of the US population) suffer from coronary disease with 500 000 new stable angina cases being reported each year. The 2012 ACCF/AHA guidelines for prognostication in SCAD recommend that patients are stratified into high (>3%), intermediate (1–3%), and low (<1%) annual mortality risk groups, which then define different investigation and treatment pathways.
So far, prognostic models proposed for SCAD (compared in Supplementary material online, Table S1) have been based on data collected for research purposes rather than the information that clinicians record in real-world practice, and none has been recommended for clinical use. Among their limitations is the failure to incorporate the 'broad range of relevant data' identified as important in the guidelines, such as sociodemographic characteristics, cardiovascular (CVD) and non-CVD comorbidities, mental health, symptom severity, and clinically available biomarkers. These data are routinely recorded for most patients before more costly information becomes available from further, often invasive, investigations. Other limitations of previous models include the use of selected samples from trials or voluntary registries, and covering a narrow range of SCAD such as excluding or being confined to post-ACS patients. Furthermore, the ACCF/AHA guidelines emphasize the importance of both all-cause mortality and coronary events as outcomes, but no previous studies have assessed both. Importantly, none of the previous models has been validated in an external data set.
We sought to address these limitations by analysing large-scale, population-based, linked electronic health records data. Our objective was to develop and validate the performance of prognostic models that incorporate clinical measures recommended in guidelines and are commonly available in patients with SCAD. Following recent methodological guidance, we assessed the accuracy of the predictions from the prognostic models based on their calibration, discrimination (C-index), and reclassification improvement. To further evaluate potential clinical benefits, we estimated life years saved when the models are used to guide management decisions. Based on these analyses, we developed prognostic models to predict all-cause mortality and non-fatal myocardial infarction (MI) or coronary death in SCAD and evaluated their performance in an independent data set.
