Epidemiology and Prevention
Cardiology 1992;80:294-301
Medical Department B. Frederiksborg County Central Hospital, Hillcrod, Denmark
KeyWords Acute myocardial infarction Prognosis, myocardial infarction Angina pectoris Chest pain
Risk Factors Related to the 7- Year Prognosis for Patients Suspected of Myocardial Infarction with and without Confirmed Diagnosis
Abstract This study prospectively evaluates the long-term prognosis of patients admitted with chest pain under suspicion of acute myocardial infarction (AMI) with and without confirmed diagnosis. All patients below 76 years of age, free of other severe diseases and alive at discharge, who were admitted to a coronary care unit of a well-defined region during 1 year, con stituted the study population. In all, 275 patients with and 257 patients without confirmed AMI (non-AMI) were included. During 7 years of follow-up, 122 cardiac events (96 cardiac deaths and 26 nonfatal AMI) occurred in the AMI patients, and 69 (44 cardiac deaths and 25 nonfatal AMI) were observed in the non-AMI patients. Using univariate analysis, the following risk variables were significantly related to an impaired prognosis of non-AMI patients: age, a history of pre vious AMI, angina pectoris, clinical heart failure, diabetes and ST or T changes in the electrocardiogram (ECG) on admis sion. By multivariate analysis, the following risk factors con tained independent prognostic information for non-AMI pa tients: (1) a history of angina pectoris and (2) ST and T changes on the ECG on admission. We conclude that a subset of non-AMI patients at high risk for cardiac events even in the long term can be identified from the medical history and the ECG on admission. These patients should be carefully evalu ated prior to discharge, whereas patients without signs of isch emic heart disease have an excellent prognosis.
Received: December 12, 1991 Accepted: January 2. 1992
Jeppe Launbjerg Medical Department B Rigshospitalet, Blcgdamsvej 9 DK-2100 Copenhagen (Denmark)
©1992 S. Kargcr AG, Basel 0008-6312/92/ 0804-029452.75/0
Downloaded by: Stockholm University Library 130.237.165.40 - 12/15/2018 2:01:11 AM
Jeppe Launbjerg Per Fruergaard Henrik Loft Jacobsen Jan Kyst Madsen
Introduction Previous studies have described both the short- and the long-term prognoses of patients discharged after an acute myocardial infarc tion (AMI) [1, 2], The prognosis of patients admitted with chest pain under suspicion of AMI, but in whom the diagnosis was not con firmed (non-AMI). has been the subject of several prospective short-term studies [3-6], These studies show that non-AMI patients have an increased risk of AMI or cardiac death in the 1st year following discharge; the risk is almost as high as in AMI patients. The long-term prognosis is less well described, but the available studies indicate that non-AMI patients continue to have an increased risk of cardiac events, although with a better progno sis than AMI patients [7-10], We have previously published the short term prognosis of 257 non-AMI patients in whom the 1-vear risk of AMI or cardiac death was 9%. A multivariate analysis identified two factors with independent prognostic in formation; ( 1) a history of AMI and/or angina pectoris and (2) permanent or transient ST or T changes in the electrocardiogram (ECG) on admission [11]. The purpose of the present analysis was to describe the prognosis after 7 years of the same non-AMI and AMI patients and to iden tify long-term risk factors for cardiac events in non-AMI patients.
cardiac enzymes (LDH and SGOT): (2) development of changes in the ECG that lasted for at least 48 h and were compatible with Q-wave or non-Q-wave infarc tion, and (3) clear evidence of AMI at postmortem examination. The remaining 516 patients were classi fied as non-AMI patients [11]. The following exclusion criteria were applied to both populations: ( I ) age above 75 years on admission: (2) death during admission: (3) severe chronic disease, e.g. cancer or sequelae from stroke, and (4) admission for other reason than chest pain. In the non-AMI popu lation, patients with obvious noncardiac or cardiac noncoronary chest pain were excluded, too. In all, 257 non-AMI and 275 AMI patients were included [11]. ST and T segments on the ECGs on admission and after 24 h were coded according to the Minnesota code [12] as 1. normal (no intraventricular block or ST-T abnormalities that precluded coding): 2, intraventricu lar block (left or right bundle branch block, left hcniiblock or combined): 3, permanent changes (ST depres sion, ST elevation or negative T waves on admission and unchanged after 24 h), and 4, transient changes (any of the above-mentioned abnormalities present at the time of admission that returned to normal or to a less abnormal Minnesota code within 24 h) [II].
Follow-Up Complete information on deaths was obtained from the National Person Registration Office. The causes of death were revealed from death certificates in all cases and from medical records when available. Data concerning later admissions with confirmed AMI were gathered for all patients at any of the four hospi tals of the county. Furthermore, all non-AMI patients were sent a questionnaire and one reminder about pos sible admissions for AMI outside the region. Ninetyfive percent answered, only I case of AMI outside the county was registered from the questionnaires. The patients were followed from the point o f time of dis charge. and the follow-up was terminated on the 31st of December, 1989. The median observation time was 88 months (range 86-98).
Patients and Methods Statistical Methods Cardiac events were defined as nonfatal AMI or cardiac death (fatal AMI. sudden cardiac death or other cardiac death). Only the first event of each patient was considered in the analyses. In the univar iate analysis, the percentage without cardiac events was estimated using the Kaplan-Meier method, and comparison was performed by means of the log-rank test [13]. Noncardiac deaths were treated as censored observations.
295
Downloaded by: Stockholm University Library 130.237.165.40 - 12/15/2018 2:01:11 AM
.
From November 1. 198 1. to October 31,1 982 975 patients were admitted under suspicion of AMI to one of the four coronary care units in Frederiksborg Coun ty. Denmark (population: 332.000). Detailed defini tion of AMI. medical history, exclusion criteria and the number of excluded patients have been reported [11]. AMI was confirmed in 459 patients because at least one of the following criteria in addition to the clinical symptoms was fulfilled: (1) characteristic elevation of
Table 1. Univariate analysis o f various risk factors regarding cardiac events in the AMI and non-AMI population
Non-AMI
Total
AMI n
without CE. %
275
57
216 59
56 53
0.26
Cardiology 1992;80:294-301
Medical Department B. Frederiksborg County Central Hospital, Hillcrod, Denmark
KeyWords Acute myocardial infarction Prognosis, myocardial infarction Angina pectoris Chest pain
Risk Factors Related to the 7- Year Prognosis for Patients Suspected of Myocardial Infarction with and without Confirmed Diagnosis
Abstract This study prospectively evaluates the long-term prognosis of patients admitted with chest pain under suspicion of acute myocardial infarction (AMI) with and without confirmed diagnosis. All patients below 76 years of age, free of other severe diseases and alive at discharge, who were admitted to a coronary care unit of a well-defined region during 1 year, con stituted the study population. In all, 275 patients with and 257 patients without confirmed AMI (non-AMI) were included. During 7 years of follow-up, 122 cardiac events (96 cardiac deaths and 26 nonfatal AMI) occurred in the AMI patients, and 69 (44 cardiac deaths and 25 nonfatal AMI) were observed in the non-AMI patients. Using univariate analysis, the following risk variables were significantly related to an impaired prognosis of non-AMI patients: age, a history of pre vious AMI, angina pectoris, clinical heart failure, diabetes and ST or T changes in the electrocardiogram (ECG) on admis sion. By multivariate analysis, the following risk factors con tained independent prognostic information for non-AMI pa tients: (1) a history of angina pectoris and (2) ST and T changes on the ECG on admission. We conclude that a subset of non-AMI patients at high risk for cardiac events even in the long term can be identified from the medical history and the ECG on admission. These patients should be carefully evalu ated prior to discharge, whereas patients without signs of isch emic heart disease have an excellent prognosis.
Received: December 12, 1991 Accepted: January 2. 1992
Jeppe Launbjerg Medical Department B Rigshospitalet, Blcgdamsvej 9 DK-2100 Copenhagen (Denmark)
©1992 S. Kargcr AG, Basel 0008-6312/92/ 0804-029452.75/0
Downloaded by: Stockholm University Library 130.237.165.40 - 12/15/2018 2:01:11 AM
Jeppe Launbjerg Per Fruergaard Henrik Loft Jacobsen Jan Kyst Madsen
Introduction Previous studies have described both the short- and the long-term prognoses of patients discharged after an acute myocardial infarc tion (AMI) [1, 2], The prognosis of patients admitted with chest pain under suspicion of AMI, but in whom the diagnosis was not con firmed (non-AMI). has been the subject of several prospective short-term studies [3-6], These studies show that non-AMI patients have an increased risk of AMI or cardiac death in the 1st year following discharge; the risk is almost as high as in AMI patients. The long-term prognosis is less well described, but the available studies indicate that non-AMI patients continue to have an increased risk of cardiac events, although with a better progno sis than AMI patients [7-10], We have previously published the short term prognosis of 257 non-AMI patients in whom the 1-vear risk of AMI or cardiac death was 9%. A multivariate analysis identified two factors with independent prognostic in formation; ( 1) a history of AMI and/or angina pectoris and (2) permanent or transient ST or T changes in the electrocardiogram (ECG) on admission [11]. The purpose of the present analysis was to describe the prognosis after 7 years of the same non-AMI and AMI patients and to iden tify long-term risk factors for cardiac events in non-AMI patients.
cardiac enzymes (LDH and SGOT): (2) development of changes in the ECG that lasted for at least 48 h and were compatible with Q-wave or non-Q-wave infarc tion, and (3) clear evidence of AMI at postmortem examination. The remaining 516 patients were classi fied as non-AMI patients [11]. The following exclusion criteria were applied to both populations: ( I ) age above 75 years on admission: (2) death during admission: (3) severe chronic disease, e.g. cancer or sequelae from stroke, and (4) admission for other reason than chest pain. In the non-AMI popu lation, patients with obvious noncardiac or cardiac noncoronary chest pain were excluded, too. In all, 257 non-AMI and 275 AMI patients were included [11]. ST and T segments on the ECGs on admission and after 24 h were coded according to the Minnesota code [12] as 1. normal (no intraventricular block or ST-T abnormalities that precluded coding): 2, intraventricu lar block (left or right bundle branch block, left hcniiblock or combined): 3, permanent changes (ST depres sion, ST elevation or negative T waves on admission and unchanged after 24 h), and 4, transient changes (any of the above-mentioned abnormalities present at the time of admission that returned to normal or to a less abnormal Minnesota code within 24 h) [II].
Follow-Up Complete information on deaths was obtained from the National Person Registration Office. The causes of death were revealed from death certificates in all cases and from medical records when available. Data concerning later admissions with confirmed AMI were gathered for all patients at any of the four hospi tals of the county. Furthermore, all non-AMI patients were sent a questionnaire and one reminder about pos sible admissions for AMI outside the region. Ninetyfive percent answered, only I case of AMI outside the county was registered from the questionnaires. The patients were followed from the point o f time of dis charge. and the follow-up was terminated on the 31st of December, 1989. The median observation time was 88 months (range 86-98).
Patients and Methods Statistical Methods Cardiac events were defined as nonfatal AMI or cardiac death (fatal AMI. sudden cardiac death or other cardiac death). Only the first event of each patient was considered in the analyses. In the univar iate analysis, the percentage without cardiac events was estimated using the Kaplan-Meier method, and comparison was performed by means of the log-rank test [13]. Noncardiac deaths were treated as censored observations.
295
Downloaded by: Stockholm University Library 130.237.165.40 - 12/15/2018 2:01:11 AM
.
From November 1. 198 1. to October 31,1 982 975 patients were admitted under suspicion of AMI to one of the four coronary care units in Frederiksborg Coun ty. Denmark (population: 332.000). Detailed defini tion of AMI. medical history, exclusion criteria and the number of excluded patients have been reported [11]. AMI was confirmed in 459 patients because at least one of the following criteria in addition to the clinical symptoms was fulfilled: (1) characteristic elevation of
Table 1. Univariate analysis o f various risk factors regarding cardiac events in the AMI and non-AMI population
Non-AMI
Total
AMI n
without CE. %
275
57
216 59
56 53
0.26
