Michael J. Pencina

General Cardiovascular Risk Profile for Use in Primary Care The Framingham Heart Study

Background— Separate multivariable risk algorithms are commonly used to assess risk of specific atherosclerotic cardiovascular disease (CVD) events, ie, coronary heart disease, cerebrovascular disease, peripheral vascular disease, and heart failure. The present report presents a single multivariable risk function that predicts risk of developing all CVD and of its constituents. Methods and Results— We used Cox proportional-hazards regression to evaluate the risk of developing a first CVD event in 8491 Framingham study participants (mean age, 49 years; 4522 women) who attended a routine examination between 30 and 74 years of age and were free of CVD. Sex-specific multivariable risk functions (“general CVD” algorithms) were derived that incorporated age, total and high-density lipoprotein cholesterol, systolic blood pressure, treatment for hypertension, smoking, and diabetes status. We assessed the performance of the general CVD algorithms for predicting individual CVD events (coronary heart disease, stroke, peripheral artery disease, or heart failure). Over 12 years of follow-up, 1174 participants (456 women) developed a first CVD event. All traditional risk factors evaluated predicted CVD risk (multivariable-adjusted P<0.0001). The general CVD algorithm demonstrated good discrimination (C statistic, 0.763 [men] and 0.793 [women]) and calibration. Simple adjustments to the general CVD risk algorithms allowed estimation of the risks of each CVD component. Two simple risk scores are presented, 1 based on all traditional risk factors and the other based on non-laboratory-based predictors. Conclusions— A sex-specific multivariable risk factor algorithm can be conveniently used to assess general CVD risk and risk of individual CVD events (coronary, cerebrovascular, and peripheral arterial disease and heart failure). The estimated absolute CVD event rates can be used to quantify risk and to guide preventive care.

Vitamin D Deficiency and Risk of Cardiovascular Disease

Background— Vitamin D receptors have a broad tissue distribution that includes vascular smooth muscle, endothelium, and cardiomyocytes. A growing body of evidence suggests that vitamin D deficiency may adversely affect the cardiovascular system, but data from longitudinal studies are lacking. Methods and Results— We studied 1739 Framingham Offspring Study participants (mean age 59 years; 55% women; all white) without prior cardiovascular disease. Vitamin D status was assessed by measuring 25-dihydroxyvitamin D (25-OH D) levels. Prespecified thresholds were used to characterize varying degrees of 25-OH D deficiency (<15 ng/mL, <10 ng/mL). Multivariable Cox regression models were adjusted for conventional risk factors. Overall, 28% of individuals had levels <15 ng/mL, and 9% had levels <10 ng/mL. During a mean follow-up of 5.4 years, 120 individuals developed a first cardiovascular event. Individuals with 25-OH D <15 ng/mL had a multivariable-adjusted hazard ratio of 1.62 (95% confidence interval 1.11 to 2.36, P=0.01) for incident cardiovascular events compared with those with 25-OH D ≥15 ng/mL. This effect was evident in participants with hypertension (hazard ratio 2.13, 95% confidence interval 1.30 to 3.48) but not in those without hypertension (hazard ratio 1.04, 95% confidence interval 0.55 to 1.96). There was a graded increase in cardiovascular risk across categories of 25-OH D, with multivariable-adjusted hazard ratios of 1.53 (95% confidence interval 1.00 to 2.36) for levels 10 to <15 ng/mL and 1.80 (95% confidence interval 1.05 to 3.08) for levels <10 ng/mL (P for linear trend=0.01). Further adjustment for C-reactive protein, physical activity, or vitamin use did not affect the findings. Conclusions— Vitamin D deficiency is associated with incident cardiovascular disease. Further clinical and experimental studies may be warranted to determine whether correction of vitamin D deficiency could contribute to the prevention of cardiovascular disease.

Assessing the performance of prediction models: a framework for traditional and novel measures.

The performance of prediction models can be assessed using a variety of methods and metrics. Traditional measures for binary and survival outcomes include the Brier score to indicate overall model performance, the concordance (or c) statistic for discriminative ability (or area under the receiver operating characteristic [ROC] curve), and goodness-of-fit statistics for calibration. Several new measures have recently been proposed that can be seen as refinements of discrimination measures, including variants of the c statistic for survival, reclassification tables, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Moreover, decision–analytic measures have been proposed, including decision curves to plot the net benefit achieved by making decisions based on model predictions. We aimed to define the role of these relatively novel approaches in the evaluation of the performance of prediction models. For illustration, we present a case study of predicting the presence of residual tumor versus benign tissue in patients with testicular cancer (n = 544 for model development, n = 273 for external validation). We suggest that reporting discrimination and calibration will always be important for a prediction model. Decision-analytic measures should be reported if the predictive model is to be used for clinical decisions. Other measures of performance may be warranted in specific applications, such as reclassification metrics to gain insight into the value of adding a novel predictor to an established model.

Extensions of net reclassification improvement calculations to measure usefulness of new biomarkers

Appropriate quantification of added usefulness offered by new markers included in risk prediction algorithms is a problem of active research and debate. Standard methods, including statistical significance and c statistic are useful but not sufficient. Net reclassification improvement (NRI) offers a simple intuitive way of quantifying improvement offered by new markers and has been gaining popularity among researchers. However, several aspects of the NRI have not been studied in sufficient detail. In this paper we propose a prospective formulation for the NRI which offers immediate application to survival and competing risk data as well as allows for easy weighting with observed or perceived costs. We address the issue of the number and choice of categories and their impact on NRI. We contrast category-based NRI with one which is category-free and conclude that NRIs cannot be compared across studies unless they are defined in the same manner. We discuss the impact of differing event rates when models are applied to different samples or definitions of events and durations of follow-up vary between studies. We also show how NRI can be applied to case-control data. The concepts presented in the paper are illustrated in a Framingham Heart Study example. In conclusion, NRI can be readily calculated for survival, competing risk, and case-control data, is more objective and comparable across studies using the category-free version, and can include relative costs for classifications. We recommend that researchers clearly define and justify the choices they make when choosing NRI for their application.

Arterial Stiffness and Cardiovascular Events The Framingham Heart Study

Background— Various measures of arterial stiffness and wave reflection have been proposed as cardiovascular risk markers. Prior studies have not assessed relations of a comprehensive panel of stiffness measures to prognosis in the community. Methods and Results— We used proportional hazards models to analyze first-onset major cardiovascular disease events (myocardial infarction, unstable angina, heart failure, or stroke) in relation to arterial stiffness (pulse wave velocity [PWV]), wave reflection (augmentation index, carotid-brachial pressure amplification), and central pulse pressure in 2232 participants (mean age, 63 years; 58% women) in the Framingham Heart Study. During median follow-up of 7.8 (range, 0.2 to 8.9) years, 151 of 2232 participants (6.8%) experienced an event. In multivariable models adjusted for age, sex, systolic blood pressure, use of antihypertensive therapy, total and high-density lipoprotein cholesterol concentrations, smoking, and presence of diabetes mellitus, higher aortic PWV was associated with a 48% increase in cardiovascular disease risk (95% confidence interval, 1.16 to 1.91 per SD; P=0.002). After PWV was added to a standard risk factor model, integrated discrimination improvement was 0.7% (95% confidence interval, 0.05% to 1.3%; P<0.05). In contrast, augmentation index, central pulse pressure, and pulse pressure amplification were not related to cardiovascular disease outcomes in multivariable models. Conclusions— Higher aortic stiffness assessed by PWV is associated with increased risk for a first cardiovascular event. Aortic PWV improves risk prediction when added to standard risk factors and may represent a valuable biomarker of cardiovascular disease risk in the community.

Effect of Sitagliptin on Cardiovascular Outcomes in Type 2 Diabetes

BACKGROUND Data are lacking on the long-term effect on cardiovascular events of adding sitagliptin, a dipeptidyl peptidase 4 inhibitor, to usual care in patients with type 2 diabetes and cardiovascular disease. METHODS In this randomized, double-blind study, we assigned 14,671 patients to add either sitagliptin or placebo to their existing therapy. Open-label use of antihyperglycemic therapy was encouraged as required, aimed at reaching individually appropriate glycemic targets in all patients. To determine whether sitagliptin was noninferior to placebo, we used a relative risk of 1.3 as the marginal upper boundary. The primary cardiovascular outcome was a composite of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or hospitalization for unstable angina. RESULTS During a median follow-up of 3.0 years, there was a small difference in glycated hemoglobin levels (least-squares mean difference for sitagliptin vs. placebo, -0.29 percentage points; 95% confidence interval [CI], -0.32 to -0.27). Overall, the primary outcome occurred in 839 patients in the sitagliptin group (11.4%; 4.06 per 100 person-years) and 851 patients in the placebo group (11.6%; 4.17 per 100 person-years). Sitagliptin was noninferior to placebo for the primary composite cardiovascular outcome (hazard ratio, 0.98; 95% CI, 0.88 to 1.09; P<0.001). Rates of hospitalization for heart failure did not differ between the two groups (hazard ratio, 1.00; 95% CI, 0.83 to 1.20; P=0.98). There were no significant between-group differences in rates of acute pancreatitis (P=0.07) or pancreatic cancer (P=0.32). CONCLUSIONS Among patients with type 2 diabetes and established cardiovascular disease, adding sitagliptin to usual care did not appear to increase the risk of major adverse cardiovascular events, hospitalization for heart failure, or other adverse events. (Funded by Merck Sharp & Dohme; TECOS ClinicalTrials.gov number, NCT00790205.).

Increasing cardiovascular disease burden due to diabetes mellitus: the Framingham Heart Study.

Background— Marked reductions in cardiovascular disease (CVD) morbidity and mortality have occurred in the United States over the last 50 years. We tested the hypothesis that the relative burden of CVD attributable to diabetes mellitus (DM) has increased over the past 5 decades. Methods and Results— Participants aged 45 to 64 years from the Framingham Heart Study, who attended examinations in an “early” time period (1952 to 1974), were compared with those who attended examinations in a later time period (1975 to 1998). The risk of CVD events (n=133 among those with and 1093 among those without DM) attributable to DM in the 2 time periods was assessed with Cox proportional hazards models; population attributable risk of DM as a CVD risk factor was calculated for each time period. The age- and sex-adjusted hazard ratio for DM as a CVD risk factor was 3.0 (95% CI, 2.3 to 3.9) in the earlier time period and 2.5 (95% CI, 1.9 to 3.2) in the later time period. The population attributable risk for DM as a CVD risk factor increased from 5.4% (95% CI, 3.8% to 6.9%) in the earlier time period to 8.7% (95% CI, 5.9% to 11.4%) in the later time period (P for attributable risk ratio=0.04), although multivariable adjustment resulted in attenuation of these findings (P=0.12); most of these observations were found among men. Conclusions— The proportion of CVD attributable to DM has increased over the past 50 years in Framingham. These findings emphasize the need for increased efforts to prevent DM and to aggressively treat and control CVD risk factors among those with DM.

Novel and conventional biomarkers for prediction of incident cardiovascular events in the community.

CONTEXT Prior studies have demonstrated conflicting results regarding how much information novel biomarkers add to cardiovascular risk assessment. OBJECTIVE To evaluate the utility of contemporary biomarkers for predicting cardiovascular risk when added to conventional risk factors. DESIGN, SETTING, AND PARTICIPANTS Cohort study of 5067 participants (mean age, 58 years; 60% women) without cardiovascular disease from Malmö, Sweden, who attended a baseline examination between 1991 and 1994. Participants underwent measurement of C-reactive protein (CRP), cystatin C, lipoprotein-associated phospholipase 2, midregional proadrenomedullin (MR-proADM), midregional proatrial natriuretic peptide, and N-terminal pro-B-type natriuretic peptide (N-BNP) and underwent follow-up until 2006 using the Swedish national hospital discharge and cause-of-death registers and the Stroke in Malmö register for first cardiovascular events (myocardial infarction, stroke, coronary death). MAIN OUTCOME MEASURES Incident cardiovascular and coronary events. RESULTS During median follow-up of 12.8 years, there were 418 cardiovascular and 230 coronary events. Models with conventional risk factors had C statistics of 0.758 (95% confidence interval [CI], 0.734 to 0.781) and 0.760 (0.730 to 0.789) for cardiovascular and coronary events, respectively. Biomarkers retained in backward-elimination models were CRP and N-BNP for cardiovascular events and MR-proADM and N-BNP for coronary events, which increased the C statistic by 0.007 (P = .04) and 0.009 (P = .08), respectively. The proportion of participants reclassified was modest (8% for cardiovascular risk, 5% for coronary risk). Net reclassification improvement was nonsignificant for cardiovascular events (0.0%; 95% CI, -4.3% to 4.3%) and coronary events (4.7%; 95% CI, -0.76% to 10.1%). Greater improvements were observed in analyses restricted to intermediate-risk individuals (cardiovascular events: 7.4%; 95% CI, 0.7% to 14.1%; P = .03; coronary events: 14.6%; 95% CI, 5.0% to 24.2%; P = .003). However, correct reclassification was almost entirely confined to down-classification of individuals without events rather than up-classification of those with events. CONCLUSIONS Selected biomarkers may be used to predict future cardiovascular events, but the gains over conventional risk factors are minimal. Risk classification improved in intermediate-risk individuals, mainly through the identification of those unlikely to develop events.

Carotid-Wall Intima-Media Thickness and Cardiovascular Events

BACKGROUND Intima-media thickness of the walls of the common carotid artery and internal carotid artery may add to the Framingham risk score for predicting cardiovascular events. METHODS We measured the mean intima-media thickness of the common carotid artery and the maximum intima-media thickness of the internal carotid artery in 2965 members of the Framingham Offspring Study cohort. Cardiovascular-disease outcomes were evaluated for an average follow-up of 7.2 years. Multivariable Cox proportional-hazards models were generated for intima-media thickness and risk factors. We evaluated the reclassification of cardiovascular disease on the basis of the 8-year Framingham risk score category (low, intermediate, or high) after adding intima-media thickness values. RESULTS A total of 296 participants had a cardiovascular event. The risk factors of the Framingham risk score predicted these events, with a C statistic of 0.748 (95% confidence interval [CI], 0.719 to 0.776). The adjusted hazard ratio for cardiovascular disease with a 1-SD increase in the mean intima-media thickness of the common carotid artery was 1.13 (95% CI, 1.02 to 1.24), with a nonsignificant change in the C statistic of 0.003 (95% CI, 0.000 to 0.007); the corresponding hazard ratio for the maximum intima-media thickness of the internal carotid artery was 1.21 (95% CI, 1.13 to 1.29), with a modest increase in the C statistic of 0.009 (95% CI, 0.003 to 0.016). The net reclassification index increased significantly after addition of intima-media thickness of the internal carotid artery (7.6%, P<0.001) but not intima-media thickness of the common carotid artery (0.0%, P=0.99). With the presence of plaque, defined as intima-media thickness of the internal carotid artery of more than 1.5 mm, the net reclassification index was 7.3% (P=0.01), with an increase in the C statistic of 0.014 (95% CI, 0.003 to 0.025). CONCLUSIONS The maximum internal and mean common carotid-artery intima-media thicknesses both predict cardiovascular outcomes, but only the maximum intima-media thickness of (and presence of plaque in) the internal carotid artery significantly (albeit modestly) improves the classification of risk of cardiovascular disease in the Framingham Offspring Study cohort. (Funded by the National Heart, Lung, and Blood Institute.).

Predicting the 30-Year Risk of Cardiovascular Disease. The Framingham Heart Study

Background— Present cardiovascular disease (CVD) risk prediction algorithms were developed for a ≤10-year follow up period. Clustering of risk factors at younger ages and increasing life expectancy suggest the need for longer-term risk prediction tools. Methods and Results— We prospectively followed 4506 participants (2333 women) of the Framingham Offspring cohort aged 20 to 59 years and free of CVD and cancer at baseline examination in 1971–1974 for the development of “hard” CVD events (coronary death, myocardial infarction, stroke). We used a modified Cox model that allows adjustment for competing risk of noncardiovascular death to construct a prediction algorithm for 30-year risk of hard CVD. Cross-validated survival C statistic and calibration &khgr;2 were used to assess model performance. The 30-year hard CVD event rates adjusted for the competing risk of death were 7.6% for women and 18.3% for men. Standard risk factors (male sex, systolic blood pressure, antihypertensive treatment, total and high-density lipoprotein cholesterol, smoking, and diabetes mellitus), measured at baseline, were significantly related to the incidence of hard CVD and remained significant when updated regularly on follow-up. Body mass index was associated positively with 30-year risk of hard CVD only in models that did not update risk factors. Model performance was excellent as indicated by cross-validated discrimination C=0.803 and calibration &khgr;2=4.25 (P=0.894). In contrast, 30-year risk predictions based on different applications of 10-year functions proved inadequate. Conclusions— Standard risk factors remain strong predictors of hard CVD over extended follow-up. Thirty-year risk prediction functions offer additional risk burden information that complements that of 10-year functions.