Research Article: Weak Prediction Power of the Framingham Risk Score for Coronary Artery Disease in Nonagenarians

Date Published: November 13, 2014

Publisher: Public Library of Science

Author(s): Josef Yayan, Ralf Krahe.

http://doi.org/10.1371/journal.pone.0113044

Abstract

Coronary artery disease (CAD) is caused by an acute myocardial infarction and is still feared as a life-threatening heart disease worldwide. In order to identify patients at high risk for CAD, previous studies have proposed various risk assessment scores for the prevention of CAD. The most commonly used risk assessment score for CAD worldwide is the Framingham Risk Score (FRS). The FRS is used for middle-aged people; hence, its appropriateness has not been demonstrated to predict the likelihood of CAD occurrence in very elderly people. This article examines the possible predictive value of FRS for CAD in very elderly people over 90 years of age.

Data on all patients over 90 years of age who received a cardiac catheter were collected from hospital charts from the Department of Internal Medicine, Saarland University Medical Center, and HELIOS Hospital Wuppertal, Witten/Herdecke University Medical Center, Germany, within a study period from 2004 to 2013. The FRSs and cardiovascular risk profiles of patients over 90 years of age with and without CAD after cardiac catheterization were compared.

One hundred and seventy-five (91.15%, mean age 91.51±1.80 years, 74 females [42.29%]; 95% confidence interval [CI], 0.87–0.95) of a total 192 of the very elderly patients were found to have CAD. Based on the results of our study, the FRS seems to provide weak predictive ability for CAD in very elderly people (P = 0.3792).

We found weak prediction power of FRS for CAD in nonagenarians.

Partial Text

Coronary artery disease (CAD) is the most common heart disease and hides the high risk for the cause for the development of acute myocardial infarction [1]. Numerous studies and international and national clinical practice guidelines have proven that CAD is caused by the manifestation of atherosclerosis in coronary arteries [2]–[10]. According to data from epidemiological studies, CAD has an increasingly high mortality rate around the world [11]. For this reason, the prediction of CAD risk has gained significant attention in the medical science community worldwide. The identification of risk factors for CAD is a basic requirement for establishing possible targeted medical therapy for the primary and secondary prevention of CAD. Therefore, several national and international guidelines and recommendations for preventing CAD were previously published after identifying the risk factors for CAD [12]–[15]. There are still ongoing efforts and attempts to improve the risk assessment methods for the prediction of CAD. To achieve this goal, several risk prediction scores for CAD have been developed in recent years [16]. Five or 10 risk assessments for CAD have been assumed worldwide according to the recommendations of the guidelines [17]–[18]. Currently available CAD risk prediction scores are mostly based on multivariable regression analysis deduced from the Framingham Heart Study [19] in which the traditional risk factors for CAD are taken into consideration such as age, cholesterol levels, blood pressure, smoking, and body weight [20]–[21]. The Framingham Risk Score (FRS) provides an estimation of the probability of an individual developing CAD in 10 years to detect high-risk persons and to take preventive actions [22]. Based on data obtained through the FRS calculations, high-risk patients should be treated, according to the guidelines’ recommendations, with lipid-lowering medication and aspirin in the primary prevention of CAD [23], [24]. FRS and other presently common risk estimation scores are designed for people in middle age [20], [25], [26]. The mean age in the FRS was 49 years old and people younger than 30 years and older than 74 years of age were not considered [20], [27]. Present risk prediction with the FRS might operate less effectively in elderly compared to middle-aged persons, and various traditional risk factors have a weak association with CAD risk in the elderly; for example, hypercholesterolemia is a strong cardiovascular risk factor in middle-aged individuals, but not in the elderly [27], [28]. Thus, new questions arise as to whether the FRS could be used to estimate cardiovascular risk for very elderly people over 90 years of age. We conducted the present investigation to better understand the FRS as an eligible prediction system for CAD in very elderly people over 90 years of age. Therefore, we collected data on all patients of this age group with CAD according to the International Classification of Disease from the hospital database at the Department of Internal Medicine, Saarland University Medical Center, and HELIOS Hospital Wuppertal, Witten/Herdecke University Medical Center, Germany. We used a risk assessment tool based on information from the Framingham Heart Study to calculate the FRS after confirming the presence or absence of CAD by performing cardiac catheterization to examine the FRS as an eligible scoring system for very elderly people. The variety of calculated FRS for CAD in people over 90 years of age were age, gender, systolic blood pressure, total cholesterol, high density lipoproteins (HDL), tobacco smoking, and former smoking. CAD diagnosis was made only after cardiac catheterization. The FRS for CAD was compared in patients older than 90 years of age after excluding CAD by performing cardiac catheterization. Only once we have identified the cardiovascular risk factors of CAD can we develop appropriately tailored therapies for all patients to take precautions against CAD.

In the two hospital databases, we found 126,931 patients who underwent cardiac catheterization at the Department of Internal Medicine, University Hospital of Saarland, and HELIOS Hospital Wuppertal, Witten/Herdecke University Medical Center, Germany, during the study period from 2004 to 2013. A total of 192 (0.15%, mean age 91.45±1.75 years, 97 females [50.52%]; 95% confidence interval [CI], 0.0013–0.0017) patients over 90 years of age with a cardiac catheter met the inclusion criteria for this trial. A total of 175 (91.15%, mean age 91.51±1.80 years, 74 females [42.29%]; 95% confidence interval [CI], 0.87–0.95) patients over 90 years of age had CAD (study group); in 17 patients (8.85%, mean age 90.77±0.88 years, 10 females [58.82%]; 95% CI, 0.05–0.13), CAD was excluded by means of cardiac catheter (control group). We found a higher prevalence of CAD in males, but without increased risk (1.4450 odds ratio; 95% CI, 0.5261–3.9687; P = 0.4752).

Past researchers have assumed that the incidence of acute myocardial infarction increases with advancing aging [34], [35]. According to the results of this study, after confirmation by cardiac catheterization, the FRS had an insufficient predictive value for CAD in very elderly people over 90 years of age with CAD. In this study, the assessment tool that was used to estimate 10-year risk after having a heart attack considered age, sex, total cholesterol, HDL, systolic blood pressure, smoking status, and whether patients were currently under medication for hypertension. However, age is in and of itself the strongest predictor of CAD.

We were not able to demonstrate that the FRS has sufficient predictive value in patients over 90 years of age with CAD. In addition, the scoring system with a point for each risk factor for CAD did not have sufficient predictive power for CAD in very elderly people. However, established risk factors such as hypertension, diabetes, hyperlipidemia, obesity, and smoking should be carefully considered in the therapeutic management and prevention of CAD in very elderly people, in addition to treatment for acute and chronic comorbidities.

 

Source:

http://doi.org/10.1371/journal.pone.0113044