Mortality of patients admitted with a suspected acute myocardial infarction in whom the diagnosis is not confirmed

Changing the diagnostic criteria for myocardial infarction in patients with a suspected heart attack affects the measurement of 30 day mortality but not long term survival

OBJECTIVES

To explore the effects of alternative methods of defining myocardial infarction on the numbers and survival patterns of patients identified as having sustained a confirmed myocardial infarct.

DESIGN

An inclusive historical cohort of patients admitted with a suspected heart attack. Patients were recoded from raw clinical data (collected at the index admission) to the epidemiological definitions of myocardial infarction used by the Nottingham heart attack register (NHAR), the World Health Organization (MONICA), and the UK heart attack study.

SETTING

Single health district.

PATIENTS

The NHAR identified all patients admitted in 1992 with suspected myocardial infarction.

OUTCOME MEASURES

Survival at 30 days and four year postdischarge.

RESULTS

2739 patients were identified, of whom 90% survived to discharge. Recoding increased the numbers of patients defined as having confirmed myocardial infarction from 26% under the original NHAR classification to 69%, depending on the classification system used. In confirmed myocardial infarction, subsequent 30 day survival from admission varied from 77-86% depending on the classification system; four year survival after discharge was not affected. The distribution of important prognostic variables differed significantly between groups of patients with confirmed myocardial infarction defined by different systems. Patients with suspected but unconfirmed myocardial infarction under all classification systems had a worse postdischarge mortality.

CONCLUSIONS

The classification system used had a substantial effect on the numbers of patients identified as having had a myocardial infarct, and on the 30 day survival. There were significant numbers of patients with more atypical presentations, not labelled as myocardial infarction, who did badly following discharge. More research is needed on these patients.

Use of medical resources, complications and long-term outcome in patients hospitalized with acute chest pain. A comparison between a city university hospital and a county hospital

OBJECTIVE

The primary aim was to test the hypothesis that there is a difference in long-term outcome after hospital discharge among patients hospitalized with acute chest pain in a university hospital and a county hospital. Secondary aims were to compare these two hospitals with regard to use of medical resources, occurrence of complications and risk indicators for death.

PATIENTS

All patients hospitalized at Sahlgrenska University Hospital in Göteborg (with a catchment population of 706 inhabitants/km(2)) and Uddevalla County Hospital (with a catchment population of 34 inhabitants/km(2)) due to symptoms of acute chest pain during a period of 6 months.

RESULTS

Complications, use of medical resources and mortality during the subsequent 2 years after discharge were compared among 1,592 hospitalizations in a city hospital and 822 in a county hospital due to acute chest pain. Angina pectoris after the first event, congestive heart failure and various arrhythmias were more frequently reported in the county hospital. The use of medical resources differed. Thus, the use of betablockers, heparin, antiarrhythmics, diuretics and nipride was more frequent in the county hospital, whereas the use of nitrates, digitalis, coronary angiography, percutaneous transluminal coronary angioplasty (PTCA) and coronary artery bypass grafting (CABG) was more frequent in the city hospital. Despite these differences, the mortality 2 years after hospital discharge was similar (14.7% in the city hospital and 12.8% in the county hospital, P=0.26). Two factors, intravenous digitalis in hospital and a prescription of insulin at discharge, were significantly more associated with death in the county hospital compared with the city hospital.

CONCLUSIONS

When comparing a city university hospital with a county hospital with regard to patients admitted with chest pain, major differences in terms of complications and use of medical resources were found. Thus, various complications were reported more frequently in the county hospital. The use of medical resources varied, some being used more frequently in the county hospital, whereas others were used more frequently in the university hospital. Despite these differences the mortality 2 years after hospital discharge was similar in the two cohorts.

Ten-year mortality for patients discharged after hospitalization for chest pain or other symptoms raising suspicion of acute myocardial infarction in relation to hospital discharge diagnosis

AIM

To describe the 10-year prognosis for patients discharged after hospitalization for chest pain or other symptoms giving an initial suspicion of acute myocardial infarction (AMI) in relation to the final hospital diagnosis and furthermore to compare the outcome amongst these patients with the outcome amongst a sex-, age- and community-matched control population.

METHODS

All patients who were hospitalized because of chest pain or other symptoms raising a suspicion of AMI and who were discharged alive from hospital. Patients were divided into three groups according to the final diagnosis: (1) confirmed or possible AMI, (2) confirmed or possible myocardial ischaemia and (3) other aetiology. Information on 10-year mortality was available in 3103 patients. A sex-, age- and community-matched control population (n=3221) was compared with the study population in terms of 10-year mortality. TIME OF SURVEY: 15 February 1986 to 9 November 1987.

SETTING

Sahlgrenska University Hospital.

RESULTS

Patients with confirmed or possible AMI (n=849) had a significantly higher mortality (59.4%) than patients with confirmed or possible myocardial ischaemia (n=1191) who had a mortality of 49.5% (P < 0.0001). The latter group had a higher mortality than patients with 'other aetiology' (n=1063) of whom 40.6% died (P < 0.0001). When comparing the prognosis for patients with AMI and myocardial ischaemia, there was a significant interaction with sex, with a more marked difference in women than in men. Amongst all patients, the 10-year mortality was 49.1 vs. 37.3% in the control group (P < 0.0001).

CONCLUSION

The very long term prognosis was strongly associated with diagnosis amongst patients hospitalized and discharged alive because of chest pain or other symptoms raising suspicion of AMI. The absolute mortality difference between patients who were discharged from hospital with confirmed diagnosis of AMI and those whose symptoms were considered to have other aetiology than AMI or ischaemia was nearly 20%. However, the absolute mortality difference between the patients included in the survey and a control population was only 12%.

Which factors determine the long-term outcome among patients with a very small or unconfirmed AMI

AIM

To describe various factors associated with the very long-term prognosis for patients with a very small or an unconfirmed acute myocardial infarction (AMI).

METHODS

Patients below 76 years of age, hospitalized due to suspected AMI who either developed a very small AMI (enzyme elevation<twice upper normal limit and maximum serum (S) aspartate aminotransferase (S-ASAT)<1.4 ukat/l) or an unconfirmed AMI (a suspected ischemic event with no signs of myocardial necrosis) were evaluated at our out-patient clinic. The 10-year mortality was related to the clinical history, age and sex, metabolic factors, diagnosis at hospital discharge, various psychosocial factors, use of medication, current symptoms, underlying reason to the symptoms, maximal working capacity and other observations at bicycle exercise test including signs of myocardial ischemia.

RESULTS

In all, 714 patients (33% women) with a median age of 63 years were included in the analyses. The following appeared as independent risk indicators for 10-year mortality: S-gammaglutamyl transpeptidase (GT) (P<0.0001), age (P<0.0001), current smoking (P<0.0001), a history of previous AMI (P<0.0001), maximal working capacity at bicycle exercise test (P=0.002), and current treatment with digitalis (borderline significance; P=0.022).

CONCLUSION

Among patients with a suspected acute myocardial ischemic event with no or minimal myocardial necrosis, various factors reflecting their age, history of cardiac disease and smoking, liver function, working capacity and possibly use of medication affected their very long-term prognosis.