Printer Friendly
The Free Library
23,383,293 articles and books


Clinical predictors of 30-day cardiac events in patients with acute coronary syndrome at a community hospital.

Objective: We sought to determine predictors of coronary events (cardiac death, acute myocardial infarction, and urgent revascularization) within 30 days after admission.

Methods: We prospectively collected data on 400 patients admitted through our emergency room for unstable angina and acute coronary syndromes. Patients with ST-segment elevation myocardial infarction and those who required thrombolysis were excluded.

Results: Of 383 patients who were eligible, 120 patients had coronary events within 30 days. Statistically significant variables associated with coronary events were advanced age, male sex, family history of premature coronary artery disease (CAD), diabetes mellitus, tobacco abuse, prior congestive heart failure, prior myocardial infarction, and history of CAD. Symptoms at presentation associated with cardiac events were typical angina and shortness of breath. Objective measures of ischemia associated with cardiac events were elevated troponin T, elevated creatine kinase MB, and ischemic electrocardiographic changes. Using forward stepwise regression analysis, we generated a model to predict 30-day major adverse cardiac events. The strongest predicting variable was serum troponin T (accounting for 33% of predicting [r.sup.2], P < 0.001) followed by typical angina ([r.sup.2] increasing to 37%), ischemic electrocardiographic changes (40%), prior CAD (42%), family history of premature CAD (44%), shortness of breath (46%), and positive creatine kinase MB (48%). The positive predictive power of the complete model was [r.sup.2] = 48%, P < 0.001.

Conclusion: Our model incorporating elements from the patient's demographic, medical history, presentation, and ischemic assessment identified 48% of patients presenting with unstable angina and acute coronary syndromes who will suffer a major adverse cardiac event within 30 days of admission. Although the strongest predictor was identified as serum troponin T, other clinical criteria offered improvement in our predictive abilities. Therefore, good initial clinical evaluation in addition to simple tests such as serum cardiac markers and electrocardiography are valuable in risk stratification of patients presenting with acute coronary syndromes and cardiac chest pain. Additional testing may be necessary to improve the positive predictive value of the model. Cardiac enzymes and electrocardiographic changes have the highest negative predictive value for occurrence of major adverse cardiac events. Identification of high-risk patients is essential to direct resources toward these patients and to avoid unnecessary costs and risk to the low-risk population.

**********

Cardiovascular disease remains the leading cause of mortality in the United States. The increased prevalence of various risk factors and comorbid illnesses in the population has led to a marked increase in the incidence of atherosclerotic heart disease and ischemic complications. In spite of the marked improvement in therapeutic modalities of ischemic heart disease, it continues to be an enormous burden on the current health care system.

Before the advent of coronary angiography and noninvasive evaluation, clinicians relied on the severity of the patient's symptoms and abnormalities on the electrocardiogram (ECG) to predict the severity of coronary artery disease (CAD). (1) In an era when technological advances have changed the methods for diagnosing and managing multiple disease states, the art of good history taking and physical examination has lost its luster. Many studies have documented the use of several objective tests, including noninvasive and invasive techniques, to predict the extent and severity of outcomes from CAD. The results of these studies have led many clinicians to rely more on the findings at cardiac catheterization and echocardiograms to determine management strategies and ignore the patient's clinical presentation. Other investigators have incorporated various clinical characteristics into models, which included other objective findings for myocardial ischemia attempting to risk stratify patients and to predict outcomes after myocardial injury such as the Thrombolysis in Myocardial Infarction (TIMI) risk score and the Braunwald classification. (2), (3)

Few data are available from individual hospitals to reflect and validate the use of risk-assessment models in primary care settings. Also, there are scarce data on the outcome of acute coronary syndromes presenting to community hospitals. Therefore, the aim of our study was to focus on the patients admitted to a community hospital from the emergency department for acute coronary syndromes, unstable angina, and non-ST-segment elevation myocardial infarction. The main aim was to identify various clinical risk factors that can predict major adverse cardiac events (MACE) for patients while in the hospital and 30 days after discharge in this cohort. A risk-assessment model using the specified factors was formulated to assess the individual risk for each patient presenting with cardiac chest pain.

Patients and Methods

The study was conducted in a 250-bed community hospital with a slightly predominant white and elderly population. The cardiologist on call decided admissions to the cardiology team. A 24-hour cardiac catheterization facility was available in the hospital. All patients admitted through the emergency department for the clinical diagnosis, as determined by the cardiologist on-call, of unstable angina and acute coronary syndromes through the period from January to June 2000 were included in the study. An acute coronary syndrome was defined as a clinical presentation that included any of the following: cardiac chest pain, electrocardiographic changes, and elevation of serum cardiac markers. Four hundred consecutive patients were identified and their medical records were collected. The major exclusion criteria were patients with acute ST-segment elevation myocardial infarction and patients requiring thrombolysis. Analyzed records included the hospital admission data, clinic notes for the 30-day period after admission, and any subsequent readmission records within this 30-day period. Patients who did not have any available hospital data for follow-up were contacted through the telephone and were questioned about occurrence of any events in the 30-day period after admission using a standard telephone survey form. Approval from the hospital institutional review board was obtained.

Demographic data included age, sex, race, and family history of premature CAD. Medical history of CAD (defined as abnormal stress test, presence of ischemic heart disease on an ECG, or coronary angiography), prior acute myocardial infarction (>2 wk), hypertension, diabetes mellitus, hyperlipidemia, tobacco abuse, chronic renal insufficiency (chronic creatinine level, >2 mg/dl), end-stage renal disease on hemodialysis, and creatinine value were recorded. Presenting symptoms included type of chest pain, shortness of breath, palpitations, nausea, vomiting, and diaphoresis. Chest pain was further subdivided into typical angina, atypical angina, and nonanginal chest pain as previously stated. (2) Typical angina was defined as discomfort thought by the cardiologist to be characteristic of myocardial ischemia that would fulfill the classic triad of substernal chest pain induced by exertion or stressful emotions and relieved by rest or sublingual nitroglycerin. Atypical angina was defined as chest discomfort that included only two of the previous criteria, and nonanginal chest pain was defined as that which did not meet more than one of these criteria. (4)

All patients had an ECG at admission that was interpreted by two different cardiologists blinded to the study outcomes. Ischemic electrocardiographic changes were defined as one or more of the following in two or more adjacent leads; >1 mV ST-segment depression or transient ST-segment elevation, and new deep T-wave inversion of at least 1-mV depth. Persistent ST-segment elevation, new left bundle branch block, and new Q waves were considered as exclusion criteria from the study. The rest of the electrocardiographic changes were labeled nonspecific electrocardiographic changes. Also, all patients had three serial sets of serum cardiac markers: at admission and at 6 and 12 hours after admission. We used cardiac troponin T (CTnT) and isoenzyme of creatine kinase with muscle and brain subunits (CK-MB) as our serum cardiac markers. Cardiac troponin T was measured by the second-generation troponin T assay using the ES 300 automated analyzer (Boehringer Mannheim, Mannheim, Germany). The lower detection limit for this assay was 0.04 ng/ml. Patients were classified as CTnT-positive if the CTnT value was >0.1 ng/ml. CK-MB was measured by immunoassay with a Stratus II instrument (Baxter Diagnostics, Miami, FL), with a limit of detection of 0.4 ng/ml. Patients were classified as CK-MB positive if their levels were >7 ng/ml. Therapeutic interventions and medications given to the patients were left to the discretion of their primary physicians.

The primary endpoint for all patients was occurrence of a composite endpoint of MACE within 30 days of admission. These events included nonfatal acute myocardial infarction, need for urgent revascularization in the form of percutaneous coronary intervention or coronary artery bypass graft surgery, and cardiac death. A new acute myocardial infarction was defined on the basis of the combination of new ischemic electrocardiographic changes and elevation of cardiac enzymes. Cardiac death was defined by death attributed directly to cardiac disease according to autopsy or by assessment of the cardiologist. The need for urgent revascularization was defined as the need for coronary revascularization in the form of angioplasty, stent, or coronary artery bypass graft surgery in a case of failed medical therapy to control angina, prevent ongoing myocardial ischemia, and restore hemodynamic stability. Elective revascularization procedures were not included in this analysis.

Statistical Analysis

Patients were first classified as having experienced occurrence of any MACE or not. Univariate analysis was performed using the [chi square] test for the categorical variables and Student's t test for continuous variables. All the baseline demographic, medical history, presenting symptoms, and electrocardiographic and serum cardiac markers were analyzed. Each factor was tested independently in a univariate logistic regression model. The significant factors were included in a multiple logistic regression analysis to determine the independent predictors for occurrence of MACE. Odds ratios (ORs) were calculated from the coefficients and the 95% confidence intervals were calculated for the significant variables. Both forward and backward stepwise regression models were used to form the final risk model using the factors that were significantly independent in the multiple logistic regression analysis. To validate the OR 95% confidence intervals for the different variables with respect to sampling variation, we used the bootstrap technique. We also used the same technique to test the validity of [r.sup.2] in the final model and to estimate the upper and lower limits for [r.sup.2]. The results of the bootstrap's 95% confidence intervals were compared with the original model's estimates to test for agreement. All statistical procedures were performed using SAS statistical software, version 8.1 (SAS Institute, Inc., Cary, NC).

Results

During a 6-month period, 400 patients were enrolled prospectively. Three hundred eighty-three patients were eligible according to the inclusion criteria and were admitted for cardiac chest pain. Seventeen patients were excluded for various reasons such as denial to consent, evolving into an ST-segment myocardial infarction in the emergency department, and lost to follow-up. In the 30-day postadmission period, 120 patients (34%) had an event and 263 patients (66%) did not have any events (Table 1). Univariate analyses of the demographic data, elements from the family, and medical and social histories are shown in Table 2. Elder, male, and white patients were more associated with worse cardiac outcomes at 30 days (P = 0.04, P = 0.09, and P = 0.006, respectively). Family history of premature CAD was a significant factor, with an OR of 3.4 (P = 0.001). From the medical history, patients who were diabetic; had history of tobacco abuse; and with prior history of congestive heart failure, hyperlipidemia, myocardial infarction, and CAD were more associated with occurrence of major cardiac events. We did not detect any differences between both groups regarding the incidence of hypertension or the mean creatinine levels. Patients with chronic renal insufficiency and those on hemodialysis did not have worse outcomes; however, they were more associated with elevated CTnT and CK-MB (r = 0.76, P < 0.001). Analysis of presenting symptoms revealed that typical angina and shortness of breath were associated with occurrence of MACE (P = 0.001). Although weaker than typical angina, atypical angina (P = 0.02) and diaphoresis (P = 0.002) were associated with major adverse cardiac events. Analysis of the objective evidence of myocardial ischemia revealed a significant difference between both groups in both serum cardiac markers and electrocardiographic changes. Patients with MACE had elevated CTnT, elevated CK-MB, and ischemic electrocardiographic changes (P = 0.001 for all variables). The sensitivity, specificity, and positive and negative predictive values of the significant univariate variables are shown in Tables 3 and 4.

Although several risk factors were significant in the univariate analysis, seven remained as significant independent predictors of major adverse cardiac events at the 30-day point (Table 2). The independent predictors of major adverse cardiac events were (in the order of significance) positive CTnT (OR, 12.3), typical angina (OR, 7.3), ischemic electrocardiographic changes (OR, 3.7), history of CAD (OR, 3.2), elevated CK-MB (OR, 2.8), family history of premature CAD (OR, 2.6), and shortness of breath (OR, 2.4). Using forward stepwise regression, we generated the model to predict 30-day major adverse cardiac events (Tables 3-5). The strongest predicting variable was serum CTnT (accounting for 33% of predicting value [r.sup.2], P < 0.001) followed by typical angina ([r.sup.2] increasing to 37%), ischemic electrocardiographic changes (40%), prior CAD (42%), family history of premature CAD (44%), shortness of breath (46%), and positive CK-MB (48%). The positive predictive power of the model was [r.sup.2] = 48%, P < 0.001.

Using the bootstrap technique, we found the OR for all the significant variables to be in agreement with the estimate from the original model (Fig. 1). The estimated [r.sup.2] from the bootstrap was 0.481, with upper and lower limits of 0.427 and 0.537, respectively. The C statistic was 0.933.

Discussion

In our study, we generated a model that can assist primary care providers and emergency department physicians in assessing the in-hospital and short-term risk for patients presenting with cardiac chest pain and acute coronary syndromes. From the demographic characteristics, we found that advanced age, male sex, and white race were significant predictors of cardiac events only in the univariate analysis. Family history of premature CAD was significant as a univariate factor and an independent predictor in the multivariate model. In the univariate analysis of the medical history, history of prior myocardial infarction, CAD, congestive heart failure, diabetes, hyperlipidemia, and tobacco abuse was significant. In spite of the fact that patients with chronic renal failure and specifically those on hemodialysis were more associated with elevated CTnT and CK-MB, they were not significantly associated with worse short-term outcomes. We did not find any difference in the mean creatinine levels between both groups with and without MACE. Objective measures of ischemic injury were consistent with various studies in the past. Troponin T and CK-MB were significantly associated with MACE. Ischemic electrocardiographic changes, which were mostly ST-segment depression, were significantly associated with major adverse cardiac events. Cardiac troponin T was revealed to be the most significant in the univariate analysis with the highest [X.sup.2] and in the multivariate and forward stepwise regression models. Troponin T and CK-MB followed by ischemic electrocardiographic changes had the highest negative predictive values for occurrence of MACE.

The multiple logistic regression analysis identified seven independent predictors after adjusting for the other factors. Serum CTnT was the most significant, with the highest OR (12.3), followed by typical angina and ischemic electrocardiographic changes as the strongest predictors. Elevated CK-MB lost most of its power in the multivariate analysis and had a marginal addition to the predictive power after troponin T and ischemic electrocardiographic changes. All the demographic factors lost their significance except family history of premature CAD and history of CAD. Shortness of breath was the other significant independent factor besides typical angina from the presenting symptoms. The findings of the forward stepwise regression analysis confirmed the previous findings. The total predictive value of the whole model allowed us to predict nearly half of the patients that would suffer from adverse cardiac outcome either in their hospitalization or within the 30-day postdischarge period.

The attempt to validate the model was geared toward confirming the consistency of the presence of the previously mentioned risk factors as independent variables. The similarity of the resulting OR and 95% confidence intervals of these factors to those generated in the original model enforces the validity and reproducibility of this model. The total [r.sup.2] generated from the bootstrap with its upper and lower limits of prediction confirms the ability of the model to assess the individual risk score of each patient within reasonable limits. The high level of C statistic of 0.933 provides confidence in the clinical value of the model. Caution must be applied in this case, however, because the C-statistic value is usually higher when assessing the predictive accuracy in the same patient data set used to develop the model rather than independent sets of patients.

Incorporating clinical factors in models to predict outcome of CAD has been well established in the medical literature. Diamond and Forrester (4) emphasized the concept of pretest and posttest likelihood of significant CAD. Using elements from the patient's history such as age, sex, and characteristics of the presenting chest pain, they devised a model, using the Bayes theorem of probability, to estimate the likelihood of significant CAD. Sox et al (5) identified age older than 60 years, male sex, typical angina, history of myocardial infarction, and 20 pack-year smoking history as significant predictors of angiographic CAD. When compared with treadmill exercise testing, evaluation by elements from history, physical examination, and ECG had similar ability to identify patients with severe CAD and those with higher mortality. (6) In this study, the authors concluded that the emergence of strategies for cost-conscious quality care should start with history, physical examination, and simple laboratory tests.

The concept of creating risk-stratifying models for ischemic heart disease has also been well documented. In many trials including patients with ST-segment elevation myocardial infarction and with the use of thrombolytics, the presence of multiple risk factors such as increased age, positive troponin T, and presence of electrocardiographic changes were associated with worse prognosis at 30-day and 1-year analysis. (7), (8) In contrast to the uniformity of the pathophysiology of ST-segment elevation myocardial infarction, acute coronary syndromes represent a gradation of severity of CAD from unstable angina through acute myocardial infarction. All share the same pathophysiology of acute plaque rupture, with various degrees of vessel occlusion and subsequent myocardial injury. Although they share this same pathophysiology on the continuum of severity, they have different outcomes and mandate different management strategies. Because of the heterogeneous nature of this disease, risk stratification has emerged as an essential tool for optimum management of the disease and better use of resources. (9) Moreover, devising a risk-stratification model for these patients was crucial to gauge the amount of benefit from specific treatment, knowing that patients with the highest risk get the utmost benefit from the treatment. The advent of improved therapies for unstable angina and non-ST-segment elevation myocardial infarction, such as platelet glycoprotein IIb/IIIa receptor inhibitors, has required the presence of risk stratification that can aid in optimum patient management and to avoid any unnecessary risks and costs. (10)

Results from recent studies that proved that the anatomic severity and extent of left ventricular function could predict outcomes better than just symptoms have dissuaded clinicians from optimal use of their clinical judgment. In spite of this, combining data from the clinical presentation with objective data may optimize our predicting capabilities. Califf et al (11) devised an angina score that included elements from the angina history such as acuteness, course, and daily frequency. In the presence of ST-segment abnormalities, these factors were significant predictors of mortality. In a model that included the ejection fraction, number of diseased vessels on cardiac catheterization, and presence of left main coronary artery stenosis, the angina score remained as a significant independent predictor of survival. The individual analysis of each factor of the score revealed that typical angina was a significant prognostic factor independent from the angina course and daily frequency. Also, the aforementioned study confirmed that the patient's symptoms could add to our ability to prognosticate even in the presence of coronary angiography findings.

More recently, Antman et [al.sup.2] defined the TIMI risk score that was very predictive of major adverse cardiac events and of benefit from therapy. Four of the seven identified risk factors were clinical factors: age >65, presence of at least three risk factors for CAD, severe angina, and use of aspirin in the last 7 days before presentation. In this study, the value of the risk score was apparent in the progressive increase in cardiac events with higher TIMI risk scores. Also, it was shown that the benefit from treatment was maximized in patients with the highest risk score. The only clinical limitation of the TIMI risk score is that one of the factors, significant prior coronary stenosis, required presence of coronary angiography data that may not be available to all primary care physicians at the time of the patient's presentation. Therefore, the authors concluded that similar studies should be conducted in cohorts of patients who present to emergency de partments or physicians' offices with chest pain to test for the generalizability of these risk-stratification models. Our findings are another example of the use of risk scores, but more specifically in primary care and community hospitals.

[FIGURE 1 OMITTED]

Similarly, major trials have studied adding simple objective criteria for risk stratification of CAD patients. In the multiple analyses generated from the GUSTO-IIB trial, (12) troponin T was found to be the strongest predictor of 30-day mortality, followed by electrocardiographic changes and CK-MB elevation. Other clinical factors such as hypertension, peripheral vascular disease, and smoking proved to be significant in this study. This was confirmed in the FRISC trial, (13) where CTnT was valuable in assessing risk of patients with unstable CAD followed by ischemic electrocardiographic changes. In the model generated from the PURSUIT trial, (14) the following factors were significant predictors: age, female sex, increased heart rate, lower systolic pressure, severity of prior angina, ST-segment depression, and signs of heart failure. In spite of the fact that these trials have placed the objective measures in the leading role for risk stratification, they have also incorporated strong clinical variables as independent risk factors. Our findings are in agreement with the data from those three major trials. Cardiac enzymes and ischemic electrocardiographic changes are the backbone of diagnosis and prognostication of patients with acute coronary syndromes, followed by clinical criteria.

The major limitations of our study are the number of patients in our study population and the lack of a separate validation group to test the risk-stratification model. Also, we did not include medications such as [beta]-blockers or intravenous nitrates, which were reported in a previous study as independent predictors of cardiac events in patients presenting with unstable angina. (2) Another limitation in our study was the absence of a separate analysis for the different kinds of ischemic electrocardiographic changes. The main reason was the small sample size. However, we have to take into consideration that ST-segment depression was the main ischemic electrocardiographic change in our patient population, which may have accounted for its significance. This confirms what other trials have demonstrated about the strength of ST-segment depression to predict both short-and long-term outcomes. (15), (16) Although we were not able to perform a comparative analysis between the different kinds of ischemic electrocardiographic changes, we demonstrated, with others, that the absence of ischemic electrocardiographic changes predicts a low incidence of short-term complications. (17) We excluded some objective hemodynamic measurements such as blood pressure and heart rate monitoring to maintain the simplicity of our risk model.

We excluded in the outcome analysis any elective revascularization procedure to avoid therapeutic bias. We only included emergent need for revascularization. We also did not include any data involving in-hospital therapies or interventions, as this was left to the discretion of the primary team on addition to our attempt to focus the risk-predicting model on admission data only. The fact that we did not detect the well-documented worse outcome for blacks could be because of the small sample size or because of referral bias and the skewed demographic distribution of our patient population, which included fewer blacks. Also, in spite of the higher correlation of elevated serum troponin T and CK-MB in patients with chronic renal failure, this was not translated into more MACE. This could be explained by either an increased incidence of false elevation of troponin T in patients on hemodialysis or that CTnT could be more specific to serve as a marker of chronic ischemic insult more than acute myocardial injury. We are more in favor of the latter explanation, which has been supported by various molecular studies and outcome studies that linked elevated CTnT with worse 1-year outcomes in this cohort. (18)

Our findings are important in various aspects. First, the importance of good history will allow identification of typical anginal patients. Clinical evaluation can add to the prognostic information obtained by costly tests. Using this simple riskstratification model will allow primary care providers and emergency department physicians to triage high-risk patients to either a cardiac intensive care unit or a specialized unit or referral to a tertiary care center where facilities for urgent revascularization are available. The use of all the factors in the risk model can allow prediction of MACE in approximately half of the patients presenting with acute coronary syndromes within a half hour of presentation. Additional testing may be necessary to improve the positive predictive value of the model. However, we still would like to stress the importance of starting with elements from the history and physical examination and simple laboratory studies. Objective data such as cardiac enzymes and electrocardiography remain the "gold standard" for ruling out acute coronary ischemia. Our study calls for more extensive validation of these risk factors in primary care settings of emergency departments or chest pain observation units to allow more generalizability of the results.
Table 1. Classification of major adverse cardiac events that occurred
during the 30-day follow-up period

Events No. %
Cardiac death 14 4.9
Acute myocardial infarction 70 18.2
Urgent revascularization 36 9.3

Table 2. Univariate analysis of baseline characteristics regarding
30-day MACE (a)

 Positive Negative
Factor MACE MACE(%)

Demographic

 Male 52.5 43.4
 Female 47.5 56.6
 Mean age 76 [+ or -] 13.3 63
 White race 75.8 61.6
 Black race 24.2 38.4

Family history 39.1 15.9

Medical history

 Myocardial infarction 32.77 15.6
 CAD 67.2 41.1
 CHF 41.03 26.24
 HTN 73.1 64.3
 CRI 28.3 25.1
 ESRD 11.1 8.75
 CRF 38.33 34.6
 Creatinine 2 1.92
 Hyperlipidemia 25.4 16.8
 DM 43.7 30.9
 Tobacco abuse 66.1 48.9

Symptoms

 Typical angina 34.75 3.1
 Atypical angina 29.66 19.5
 Shortness of breath 77.3 50.76
 Diaphoresis 22.1 15.3

Ischemic electrocardiographic changes 70 22.5

Serum cardiac markers

 TnT 92.5 28.9
 CK-MB 73.3 15.97

Factor Wald [X.sup.2] OR (95% CI)

Demographic

 Male 2.76 1.4 (0.9-2.2)
 Female
 Mean age 6.42 1.86 (1.3-3.1)
 White race 7.31 1.95 (1.2-3.2)
 Black race

Family history 22.1 3.4 (2-5.7)

Medical history

 Myocardial infarction 14.05 2.64 (1.6-4.4)
 CAD 21.64 2.94 (1.9-4.6)
 CHF 7.56 1.9 (1.2-3)
 HTN 2.9 1.5 (0.95-2.4)
 CRI 0.44 1.18 (0.72-1.91)
 ESRD 0.06 1.08 (0.57-2.04)
 CRF 0.49 1.17 (0.75-1.83)
 Creatinine 0.12 0.98 (0.89-1.08)
 Hyperlipidemia 4.07 1.7 (1.01-2.8)
 DM 5.8 1.7 (1.1-2.7)
 Tobacco abuse 9.2 2 (1.3-3.2)

Symptoms

 Typical angina 48.07 16.9 (7.6-37.6)
 Atypical angina 4.76 1.74 (1.05-2.9)
 Shortness of breath 22.6 3.3 (2.01-5.4)
 Diaphoresis 11.9 2.1 (1.7-4.7)

Ischemic electrocardiographic changes 70.48 8.02 (4.9-13.1)

Serum cardiac markers

 TnT 84.01 30.3 (14.6-62.9)
 CK-MB 100.62 14.46 (8.6-24.4)

Factor P value

Demographic

 Male 0.09
 Female
 Mean age 0.04
 White race 0.006
 Black race

Family history 0.001

Medical history

 Myocardial infarction 0.002
 CAD 0.001
 CHF 0.006
 HTN 0.08
 CRI 0.5
 ESRD 0.8
 CRF 0.47
 Creatinine 0.7
 Hyperlipidemia 0.04
 DM 0.01
 Tobacco abuse 0.002

Symptoms

 Typical angina 0.001
 Atypical angina 0.02
 Shortness of breath 0.001
 Diaphoresis 0.002

Ischemic electrocardiographic changes 0.001

Serum cardiac markers

 TnT 0.001
 CK-MB 0.001

(a) OR, odds ratio; CI, confidence interval; MACE, major adverse cardiac
events; CAD, coronary artery disease; CHF, congestive heart failure;
CK-MB, creatine kinase MB fraction; CRI, chronic renal
insufficiency; CRF, chronic renal failure; DM, diabetes
mellitus; ESRD, end-stage renal disease; HTN,
hypertension; TnT, troponin T.

Table 3. Sensitivity, specificity, and positive and negative predictive
values of the significant univariate variables (a)

 Sensitivity Specificity PPV NPV
 (%) (%) (%) (%)

Troponin T 92 71 72 94

Typical angina 34 97 85 74

Ischemic 68 77 61 83

 electrocardiographic
 changes

History of CAD 67 59 45 78

Family history of 38 84 54 74
 premature CAD

Shortness of breath 35 85 54 72

CK-MB 72 84 70 85

White race 76 38 39 75

(a) CAD, coronary artery disease; CK-MB, creatine kinase MB fraction;
PPV, positive predictive value; NPV, negative predictive value.

Table 4. Multiple logistic regression model for predicting 30-day major
adverse cardiac events (a)

Factor Coefficient Standard error [X.sup.2]

Constant -5.8 0.8 50
Positive CTnT 2.5 0.59 17.7
Typical angina 2.56 0.69 13.6
Electrocardiographic changes 1.23 0.36 11.2
History of CAD 1.16 0.44 6.98
Positive CK-MB 1.05 0.45 5.35
Family history 0.95 0.47 4.05
SOB 0.91 0.48 3.5
Race 0.66 0.4 2.68
Age 0.65 0.41 2.43
Atypical angina 0.64 0.42 2.3
History of MI 0.48 0.46 1.08
Diaphoresis 0.3 0.4 0.84
History of CHF -0.28 0.41 0.47
Hyperlipidemia -0.301 0.46 0.413
CRF 0.1 0.54 0.4
Tobacco abuse -0.235 0.408 0.33
Sex 0.22 0.39 0.3
Hypertension -0.1156 0.42 0.07
Diabetes 0.04 0.39 0.01

Factor OR (95% CI) P value

Constant <0.001
Positive CTnT 12.3 (3.8-39.6) <0.001
Typical angina 7.3 (3.3-24.5) 0.001
Electrocardiographic changes 3.4 (1.6-7.02) 0.008
History of CAD 3.2 (1.35-7.5) 0.008
Positive CK-MB 2.8 (1.17-7.04) 0.02
Family history 2.6 (1.02-6.6) 0.04
SOB 2.4 (1.1-6.4) 0.05
Race 1.9 (0.8-4.3) 0.1
Age 1.9 (0.8-4.4) 0.1
Atypical angina 1.9 (0.83-4.3) 0.12
History of MI 1.62 (0.65-4.02) 0.29
Diaphoresis 1.4 (0.6-3.2) 0.35
History of CHF 0.75 (0.33-1.68) 0.489
Hyperlipidemia 0.74 (0.29-1.85) 0.52
CRF 1.2 (0.45-1.9) 0.7
Tobacco abuse 0.79 (0.35-1.75) 0.56
Sex 1.2 (0.57-2.67) 0.579
Hypertension 0.89 (0.38-2.04) 0.78
Diabetes 1.04 (0.47-2.28) 0.9

(a) OR, odds ratio; CI, confidence interval; CAD, coronary artery
disease; CHF, congestive heart failure: CK-MB, creatine kinase MB
fraction; CRF, chronic renal failure; SOB, shortness of breath;
TnT, troponin T; MI, myocardial infarction.

Table 5. Forward stepwise regression model for predicting 30-day major
adverse cardiac events (a)

Factor Coefficient Standard error

Constant -5.66 0.67
Serum TnT 2.6566 0.5425
Typical angina 2.2571 0.6340
Ischemic electrocardiographic changes 1.3317 0.3513
History of CAD 1.27 0.3734
Family history 1.05 0.423
SOB 1.05 0.4645
CK-MB 0.89 0.4233

Factor [X.sup.2] OR (95% CI)

Constant
Serum TnT 23.98 14.24 (4.9-41.26)
Typical angina 12.67 9.55 (2.75-33.1)
Ischemic electrocardiographic changes 14.36 3.7 (1.9-7.5)
History of CAD 11.59 3.56 (1.71-7.41)
Family history 6.18 2.86 (1.25-6.57)
SOB 5.13 2.86 (1.15-7.11)
CK-MB 4.49 2.45 (1.07-5.62)

Factor [R.sup.2] P value

Constant
Serum TnT 0.33 <0.001
Typical angina 0.37 <0.001
Ischemic electrocardiographic changes 0.40 <0.001
History of CAD 0.42 0.01
Family history 0.44 0.01
SOB 0.46 0.02
CK-MB 0.478 <0.001

(a) OR, odds ratio; CI, confidence interval; CAD, coronary artery
disease; CK-MB, creatine kinase MB fraction; SOB, shortness of breath;
TnT, troponin T.


From the Departments of Internal Medicine, Cardiology, and Clinical Research, Geisinger Medical Center, Danville, PA.

Reprint requests to George M. Tadros, MD, Geisinger Medical Center, 100 N. Academy Avenue, Danville, PA 17822-0139. Email: gmtadros@netscape.net

Accepted September 19, 2002.

Copyright [c] 2003 by The Southern Medical Association

0038-4348/03/9611-1113

References

1. White PD. Heart Disease. New York, Macmillan, 1944, ed 3, pp 828-832.

2. Antman EM, Cohen M, Bernink PJ, et al. The TIMI risk score for unstable angina/non-ST elevation MI: A method for prognostication and therapeutic decision making. JAMA 2000;284:835-842.

3. Calvin JE, Klein LW, VandenBerg BJ, et al. Risk stratification in unstable angina: Prospective validation of the Braunwald classification. JAMA 1995;273:136-141.

4. Diamond GA, Forrester JS. Analysis of probability as an aid in the clinical diagnosis of coronary-artery disease. N Engl J Med 1979;300:1350-1358.

5. Sox HC Jr, Hickam DH, Marton KI, et al. Using the patient's history to estimate the probability of coronary artery disease: A comparison of primary care and referral practices. Am J Med 1990;89:7-14.

6. Pryor DB, Shaw L, McCants CB, et al. Value of the history and physical in identifying patients at increased risk for coronary artery disease. Ann Intern Med 1993;118:81-90.

7. Lee KL, Woodlief LH, Topol EJ, et al. Predictors of 30-day mortality in the era of reperfusion for acute myocardial infarction: Results from an international trial of 41,021 patients--GUSTO-I Investigators. Circulation 1995;91:1659-1668.

8. Califf RM, Pieper KS, Lee KL, et al. Prediction of 1-year survival after thrombolysis for acute myocardial infarction in the global utilization of streptokinase and TPA for occluded coronary arteries trial. Circulation 2000;101:2231-2238.

9. Newby LK, Christenson RH, Ohman EM, et al. Value of serial troponin T measures for early and late risk stratification in patients with acute coronary syndromes: The GUSTO-IIa Investigators. Circulation 1998;98:1853-1859.

10. Roberts R, Fromm RE. Management of acute coronary syndromes based on risk stratification by biochemical markers: An idea whose time has come. Circulation 1998;98:1831-1833.

11. Califf RM, Mark DB, Harrell FE Jr, et al. Importance of clinical measures of ischemia in the prognosis of patients with documented coronary artery disease. J Am Coll Cardiol 1988;11:20-26.

12. Ohman EM, Armstrong PW, Christenson RH, et al. Cardiac troponin T levels for risk stratification in acute myocardial ischemia: GUSTO IIA Investigators. N Engl J Med 1996;335:1333-1341.

13. Lindahl B, Venge P, Wallentin L. Relation between troponin T and the risk of subsequent cardiac events in unstable coronary artery disease: The FRISC study group. Circulation 1996;93:1651-1657.

14. Boersma E, Pieper KS, Steyerberg EW, et al. Predictors of outcome in patients with acute coronary syndromes without persistent ST-segment elevation: Results from an international trial of 9461 patients--The PURSUIT Investigators. Circulation 2000;101:2557-2567.

15. Savonitto S, Ardissino D, Granger CB, et al. Prognostic value of the admission electrocardiogram in acute coronary syndromes. JAMA 1999;281:707-713.

16. Hyde TA, French JK, Wong CK, et al. Four-year survival of patients with acute coronary syndromes without ST-segment elevation and prognostic significance of 0.5-mm ST-segment depression. Am J Cardiol 1999;84:379-385.

17. Stark ME, Vacek JL. The initial electrocardiogram during admission for myocardial infarction: Use as a predictor of clinical course and facility utilization. Arch Intern Med 1987;147:843-846.

18. Ishii J, Toriyama T, Nomura M, et al. Circulating concentration of cardiac troponin T is useful for risk stratification in patients with chronic hemodialysis: One-year outcome analysis in 394 hemodialyzed patients. J Am Coll Cardiol 2001;37(Suppl A):313A (Abstract 1050-76).

George M. Tadros, MD, Timothy R. McConnell, PHD, G. Craig Wood, MS, John M. Costello, MD, FACC, and Elias A. Iliadis, MD, FACC
COPYRIGHT 2003 Southern Medical Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

 Reader Opinion

Title:

Comment:



 

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Original Article
Author:Iliadis, Elias A.
Publication:Southern Medical Journal
Date:Nov 1, 2003
Words:6118
Previous Article:Evaluation of lactate and C-reactive protein in the assessment of acute myocardial infarction.
Next Article:Bioimpedance-derived differences in cardiac physiology during exercise stress testing in low-risk chest pain patients.
Topics:



Related Articles
Plasma elimination of cardiac troponin i in end-stage renal disease.
Electroconvulsive therapy-associated acute coronary syndrome in the absence of coronary artery disease.
Understanding heart disease in female patients: the standard model for diagnosis of cardiovascular disease often doesn't fit women. Here are some...
Clinical use of C-reactive protein for cardiovascular disease.
CAR-4. Chest pain and increased Troponin: acute coronary syndrome?
A rare case of tako-tsubo cardiomyopathy complicated by a left ventricular thrombus.
Non-ST segment elevation acute coronary syndromes.
Ode to the hospitalist.
How to improve the management of chest pain: hospitalists and use of prediction rules.

Terms of use | Copyright © 2014 Farlex, Inc. | Feedback | For webmasters