The Pooled Cohort Equations 3 Years On
2016; Lippincott Williams & Wilkins; Volume: 134; Issue: 23 Linguagem: Inglês
10.1161/circulationaha.116.024246
ISSN1524-4539
AutoresPaul M. Ridker, Nancy R. Cook,
Tópico(s)Blood Pressure and Hypertension Studies
ResumoHomeCirculationVol. 134, No. 23The Pooled Cohort Equations 3 Years On Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticlePDF/EPUBThe Pooled Cohort Equations 3 Years OnBuilding a Stronger Foundation Paul M Ridker, MD and Nancy R. Cook, ScD Paul M RidkerPaul M Ridker From Center for Cardiovascular Disease Prevention, Brigham and Women's Hospital, Harvard Medical School, Boston, MA. and Nancy R. CookNancy R. Cook From Center for Cardiovascular Disease Prevention, Brigham and Women's Hospital, Harvard Medical School, Boston, MA. Originally published6 Dec 2016https://doi.org/10.1161/CIRCULATIONAHA.116.024246Circulation. 2016;134:1789–1791In November 2013, a 10-year risk prediction tool known as the Pooled Cohort Equations (PCE) was released as the foundational basis for cardiovascular prevention guidelines in the United States.1 However, the PCE calculator was soon found to overestimate cardiovascular risk in 3 major contemporary US cohorts: the Physicians' Health Study, the Women's Health Study, and the Women's Health Initiative-Observational Cohort.2 Similar overestimation of risk had already been observed in the MESA (Multi-Ethnic Study of Atherosclerosis) and the REGARDS study (Reasons for Geographic and Racial Differences in Stroke), 2 contemporary US cohorts used for external validation by the guideline creators. The poor calibration observed in these 5 contemporary external cohorts suggested that overreliance on older data occurred during the guideline modeling process (Figure). Indeed, the PCE were derived from studies that enrolled between 1968 and 1990. These older data do not reflect the lower current rates of cardiovascular disease that largely result from secular shifts in smoking, diet, exercise, and blood pressure control, issues of which the prevention community and professional societies should rightly be proud.Download figureDownload PowerPointFigure. US death rates per 100 000 from cardiovascular disease (CVD) and coronary heart disease (CHD). The Pooled Cohort Equations were developed with data from studies that enrolled between 1968 and 1990. However, cardiovascular event rates have significantly declined over the past 40 years, which may explain why overestimation of risk has been seen when these equations are evaluated in contemporary external validation cohorts.The issue of overestimation with the PCE risk calculator has now played itself out in the peer-reviewed literature. To date, ≈3 years later, overestimation has been demonstrated in 9 different external cohorts and 2 reanalyses, with partial calibration found in 3 other studies, 1 that was a reanalysis and another that substantially overlapped the derivation sample.3 In the external validation cohort most commonly cited as evidence that the equations work well,4 calibration was actually poor in the cohort as a whole. It only became acceptable after the investigators performed subgroup analyses that were not done during model derivation and then performed additional case finding through Medicare claims data linkage, a procedure also not done in the derivation cohorts. In the absence of these post hoc analyses, this study also would be considered an example of overestimation.When controversy over the new risk calculator initially presented itself, some physicians suggested that large numbers of patients would be inappropriately prescribed statin therapy because of risk of overestimation. Others expressed that increased statin prescription would nonetheless be effective because these agents have proved beneficial even in low-risk populations. Having designed and conducted 1 of the major primary prevention statin trials, we were at least partially sympathetic with this latter view.Unfortunately, the process of releasing a suboptimal risk calculator (and then failing to recalibrate it) has had inadvertent adverse consequences that go beyond statin allocation. First, the PCE are being used to allocate other preventive agents such as aspirin, where, in contrast to statins, the risk to benefit ratio in primary prevention is quite narrow. In this instance, overestimation of risk is problematic because rates of hemorrhage exceed rates of benefit for the majority of individuals treated with aspirin in primary prevention. Further, virtually all aspirin data in primary prevention predate the statin era, so evidence of additive effects are lacking. Yet recommendations such as those issued by the US Preventive Services Task force continue to recommend prophylactic aspirin on the basis of estimated 10-year risk.Similar concerns may be raised if proprotein convertase subtilisin-like kexin type 9 inhibitors prove effective. These expensive agents are currently being tested in outcome trials that include high-risk primary prevention patients and will be of particular interest to those who are statin-intolerant. Thus, it can be anticipated that a risk-based approach to prescription is probable for this class of agents.Risk overestimation is also problematic for clinical trial design. The federally funded REPRIEVE (Randomized Trial to Prevent Vascular Events in HIV) is addressing statin therapy in patients infected with human immunodeficiency virus, a clinical situation where inflammatory mechanisms of atherosclerosis may dominate the biology. However, because the new guidelines make it difficult to give a placebo to individuals with calculated 10-year risks >10% (even if this estimate is incorrect), enrollment of higher risk individuals into REPREIVE has been hampered. Similarly, the PCE make it problematic to consider a statin trial in the elderly despite clinical differences of opinion in this arena. This occurs in part because the PCE classify nearly all men >66 years of age and women >70 years of age as having 10-year risks high enough to justify treatment.Healthy debate regarding application of the PCE has led to broader recognition of several issues important for preventive cardiology. First, as thoughtfully recognized by the guideline creators, different risk prediction tools may be needed in different ethnic populations, emphasizing the importance of external validation.Second, genetics are not included in most prediction algorithms, yet in several studies, the addition of family history modestly improves model discrimination. Such an addition would introduce modern concepts of heritable risk into primary care practice.Third, it may be useful to align the end points used in risk prediction models with those used in the randomized trials that form the evidence base for cardiovascular care. In this respect, the guideline creators took a step forward by adding stroke to the end point predicted by the risk calculator. Yet atherosclerosis trials today almost all include the additional clinically important end point of emergent coronary revascularization.Fourth, debate over the PCE has increased recognition that prediction tools based on traditional factors and novel biomarkers all suffer from common limitations inherent to the modeling process. This disenchantment with risk calculators has led some to advocate for coronary artery calcium screening in primary prevention settings (or more limited use among those at intermediate risk). This may not seem problematic at first glance because direct measures of atherosclerosis are a strong predictor of risk, have superior reclassification characteristics compared with blood biomarkers, and modestly improve discrimination when added to standard prediction models.5 However, coronary artery calcium has adverse consequences, including cost, radiation exposure, and significant rates of inadvertent incidental findings that lead to consequent downstream inefficiencies. Further, because statins increase coronary artery calcium, repeat measurement is inappropriate. Coronary artery calcium also typically occurs late in life and is associated with stable rather than unstable plaques, yet the biology of lipid reduction indicates that starting treatment early rather than late provides the greatest relative risk reductions. Creative science addressing these challenges is ongoing.Last, with so many contemporary data now available, we can work together to rebuild the core foundation on which the American College of Cardiology/American Heart Associationguidelines are built. We do not believe it is too late to recalibrate the PCE. Prediction models can and should evolve to reflect changing patterns of risk, changing patterns of treatment, and new biologic understanding.DisclosuresDr Ridker has received investigator-initiated research grant support from the National Heart Lung and Blood Institute, Novartis, Pfizer, AstraZeneca, and Kowa. He is also listed as a coinventor on patents held by the Brigham and Women's Hospital, which relate to the use of inflammatory biomarkers in cardiovascular disease and diabetes, that have been licensed to AstraZeneca and Siemens. Dr Cook has no conflicts of interest.FootnotesThe opinions in this article are not necessarily those of the editors or of the American Heart Association.Circulation is available at http://circ.ahajournals.org.Correspondence to: Paul M Ridker, MD, Center for Cardiovascular Disease Prevention, Brigham and Women's Hospital, 900 Commonwealth Ave, Boston, MA 02215. E-mail [email protected]References1. Goff DC, Lloyd-Jones DM, Bennett G, Coady S, D'Agostino RB, Gibbons R, Greenland P, Lackland DT, Levy D, O'Donnell CJ, Robinson J, Schwartz JS, Shero ST, Smith SCJ, Sorlie P, Stone NJ, Wilson PW. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association task force on practice guidelines.Circulation. 2014; 129:S49–S73.LinkGoogle Scholar2. Ridker PM, Cook NR. Statins: new American guidelines for prevention of cardiovascular disease.Lancet. 2013; 382:1762–1765. doi: 10.1016/S0140-6736(13)62388-0.CrossrefMedlineGoogle Scholar3. Cook NR, Ridker PM. Calibration of the pooled cohort equations for atherosclerotic cardiovascular risk [published online ahead of print October 11, 2016].Ann Intern Med. 2016. doi: 10.7326/M16-1739. Accessed November 9, 2016.CrossrefMedlineGoogle Scholar4. Muntner P, Colantonio LD, Cushman M, Goff DC, Howard G, Howard VJ, Kissela B, Levitan EB, Lloyd-Jones DM, Safford MM. Validation of the atherosclerotic cardiovascular disease Pooled Cohort risk equations.JAMA. 2014; 311:1406–1415. doi: 10.1001/jama.2014.2630.CrossrefMedlineGoogle Scholar5. Yeboah J, Young R, McClelland RL, Delaney JC, Polonsky TS, Dawood FZ, Blaha MJ, Miedema MD, Sibley CT, Carr JJ, Burke GL, Goff DC, Psaty BM, Greenland P, Herrington DM. Utility of nontraditional risk markers in atherosclerotic cardiovascular disease risk assessment.J Am Coll Cardiol. 2016; 67:139–147. doi: 10.1016/j.jacc.2015.10.058.CrossrefMedlineGoogle Scholar Previous Back to top Next FiguresReferencesRelatedDetailsCited By Arora S, Qamar A, Gupta P, Hendrickson M, Singh A, Vaduganathan M, Pandey A, Bansal A, Batra V, Mukhopadhyay S, Yusuf J, Tyagi S, Girish M, Kaul P, Bhatt D and Gupta M (2022) Guideline based eligibility for primary prevention statin therapy – Insights from the North India ST-elevation myocardial infarction registry (NORIN-STEMI), Journal of Clinical Lipidology, 10.1016/j.jacl.2021.12.001, 16:2, (227-236), Online publication date: 1-Mar-2022. Brown A, Wahab R, Zhang B, Smetherman D and Mahoney M (2022) Reporting and Perceptions of Breast Arterial Calcification on Mammography: A Survey of ACR Radiologists, Academic Radiology, 10.1016/j.acra.2021.01.027, 29, (S192-S198), Online publication date: 1-Jan-2022. Takamine L, Forman J, Damschroder L, Youles B and Sussman J (2021) Understanding providers' attitudes and key concerns toward incorporating CVD risk prediction into clinical practice: a qualitative study, BMC Health Services Research, 10.1186/s12913-021-06540-y, 21:1, Online publication date: 1-Dec-2021. Moral Peláez I, Brotons Cuixart C, Fernández Valverde D, Puig Palma M, Calvo Bonacho E, Martínez Muñoz P, Catalina Romero C and Quevedo Aguado L (2021) External validation of the European and American equations for calculating cardiovascular risk in a Spanish working population, Revista Clínica Española (English Edition), 10.1016/j.rceng.2020.12.008, 221:10, (561-568), Online publication date: 1-Dec-2021. Moral Peláez I, Brotons Cuixart C, Fernández Valverde D, Puig Palma M, Calvo Bonacho E, Martínez Muñoz P, Catalina Romero C and Quevedo Aguado L (2021) Validación externa de las ecuaciones europea y americana para el cálculo del riesgo cardiovascular en población laboral española, Revista Clínica Española, 10.1016/j.rce.2020.12.008, 221:10, (561-568), Online publication date: 1-Dec-2021. Campos-Staffico A, Cordwin D, Murthy V, Dorsch M and Luzum J (2021) Comparative performance of the two pooled cohort equations for predicting atherosclerotic cardiovascular disease, Atherosclerosis, 10.1016/j.atherosclerosis.2021.08.034, 334, (23-29), Online publication date: 1-Oct-2021. Grassmann F, Yang H, Eriksson M, Azam S, Hall P and Czene K (2021) Mammographic features are associated with cardiometabolic disease risk and mortality, European Heart Journal, 10.1093/eurheartj/ehab502, 42:34, (3361-3370), Online publication date: 7-Sep-2021. Arvanitis M, Post W and Battle A (2021) Genetics Precision Medicine in Cardiovascular Disease Prevention, 10.1007/978-3-030-75055-8_3, (65-80), . (2020) Risk Predictions Evidence-Based Diagnosis, 10.1017/9781108500111.007, (144-174) Mansoor H, Jo A, Beau De Rochars V, Pepine C and Mainous A (2019) Novel Self‐Report Tool for Cardiovascular Risk Assessment, Journal of the American Heart Association, 8:24, Online publication date: 17-Dec-2019. Kuragaichi T, Kataoka Y, Miyakoshi C, Miyamoto T and Sato Y (2019) External validation of pooled cohort equations using systolic blood pressure intervention trial data, BMC Research Notes, 10.1186/s13104-019-4293-1, 12:1, Online publication date: 1-Dec-2019. Damen J, Pajouheshnia R, Heus P, Moons K, Reitsma J, Scholten R, Hooft L and Debray T (2019) Performance of the Framingham risk models and pooled cohort equations for predicting 10-year risk of cardiovascular disease: a systematic review and meta-analysis, BMC Medicine, 10.1186/s12916-019-1340-7, 17:1, Online publication date: 1-Dec-2019. Lloyd-Jones D, Braun L, Ndumele C, Smith S, Sperling L, Virani S and Blumenthal R (2018) Use of Risk Assessment Tools to Guide Decision-Making in the Primary Prevention of Atherosclerotic Cardiovascular Disease: A Special Report From the American Heart Association and American College of Cardiology, Circulation, 139:25, (e1162-e1177), Online publication date: 18-Jun-2019. Lloyd-Jones D, Braun L, Ndumele C, Smith S, Sperling L, Virani S and Blumenthal R (2019) Use of Risk Assessment Tools to Guide Decision-Making in the Primary Prevention of Atherosclerotic Cardiovascular Disease, Journal of the American College of Cardiology, 10.1016/j.jacc.2018.11.005, 73:24, (3153-3167), Online publication date: 1-Jun-2019. Cardoso R, Blumenthal R, Kopecky S, Lopez-Jimenez F and Martin S (2019) How Low to Go With Lipid-Lowering Therapies in a Cost-effective and Prudent Manner, Mayo Clinic Proceedings, 10.1016/j.mayocp.2018.08.011, 94:4, (660-669), Online publication date: 1-Apr-2019. Bui Q and Daniels L (2019) A Review of the Role of Breast Arterial Calcification for Cardiovascular Risk Stratification in Women, Circulation, 139:8, (1094-1101), Online publication date: 19-Feb-2019.Musunuru K, Arora P, Cooke J, Ferguson J, Hershberger R, Hickey K, Lee J, Lima J, Loscalzo J, Pereira N, Russell M, Shah S, Sheikh F, Wang T and MacRae C (2018) Interdisciplinary Models for Research and Clinical Endeavors in Genomic Medicine: A Scientific Statement From the American Heart Association, Circulation: Genomic and Precision Medicine, 11:6, (e000046), Online publication date: 1-Jun-2018. Rodríguez-Roca G, Segura-Fragoso A, Villarín-Castro A, Alonso-Moreno F, Rodríguez-Padial L, Rodríguez-García M, Fernández-Conde J, Rojas-Martelo G, Menchén-Herreros A, Escobar-Cervantes C, Fernández-Martín J, Artigao-Rodenas L, Carbayo-Herencia J and Hernández-Moreno J (2018) Características de la población general incluida en el estudio RICARTO (RIesgo CARdiovascular y eventos cardiovasculares en la población general del área sanitaria de TOledo): resultados con los primeros 1.500 individuos incluidos en el estudio, Medicina de Familia. SEMERGEN, 10.1016/j.semerg.2017.07.002, 44:3, (180-191), Online publication date: 1-Apr-2018. Rodríguez-Roca G, Rodríguez-Padial L, Alonso-Moreno F, Segura-Fragoso A, Villarín-Castro A, Rodríguez-García M, Menchén-Herreros A, Rojas-Martelo G, Fernández-Conde J, Artigao-Rodenas L, Carbayo-Herencia J, Escobar-Cervantes C, Hernández-Moreno J and Fernández-Martín J (2018) Justificación, objetivos y diseño metodológico del estudio RICARTO (RIesgo CARdiovascular y eventos cardiovasculares en la población general del área sanitaria de TOledo)., Medicina de Familia. SEMERGEN, 10.1016/j.semerg.2017.04.007, 44:2, (107-113), Online publication date: 1-Mar-2018. Shaw L, Pepine C, Xie J, Mehta P, Morris A, Dickert N, Ferdinand K, Gulati M, Reynolds H, Hayes S, Itchhaporia D, Mieres J, Ofili E, Wenger N and Bairey Merz C (2017) Quality and Equitable Health Care Gaps for Women, Journal of the American College of Cardiology, 10.1016/j.jacc.2017.05.051, 70:3, (373-388), Online publication date: 1-Jul-2017. Shaw L, Chandrashekhar Y and Narula J (2017) Fixing the Prevention Gap Using Imaging-Guided Risk Estimates, JACC: Cardiovascular Imaging, 10.1016/j.jcmg.2017.01.001, 10:2, (214-216), Online publication date: 1-Feb-2017. December 6, 2016Vol 134, Issue 23 Advertisement Article InformationMetrics © 2016 American Heart Association, Inc.https://doi.org/10.1161/CIRCULATIONAHA.116.024246PMID: 27920070 Originally publishedDecember 6, 2016 Keywordsrisk factors/global assessmentrisk assessmentrisk predictionPDF download Advertisement
Referência(s)