This Article Full Text Full Text (PDF) All Versions of this Article: 2/4/313    most recent CIRCOUTCOMES.108.831529v1 Submit an eLetter Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Request Permissions Google Scholar Articles by Kuch, B. Articles by Meisinger, C. PubMed Articles by Kuch, B. Articles by Meisinger, C. Related Collections Acute myocardial infarction

Original Articles

Extent of the Decrease of 28-Day Case Fatality of Hospitalized Patients With Acute Myocardial Infarction Over 22 Years

Epidemiological Versus Clinical View: The MONICA/KORA Augsburg Infarction Registry

Bernhard Kuch, MD; Wolfgang von Scheidt, MD; Anja Ehmann, MD; Birgitt Kling, MD; Claudia Greschik; Allmut Hoermann, MSc and Christa Meisinger, MD, MSc

From the I. Medizinische Klinik (B.K., W.v.S., A.E., B.K.), Klinikum Augsburg, teaching hospital of the Ludwig Maximilians University München, Augsburg, Germany; Helmholtz Zentrum München (C.G., C.M.), German Research Center for Environmental Health, Institute of Epidemiology, Neuherberg, Germany; Helmholtz Zentrum München (A.H.), German Research Center for Environmental Health, Institute of Health Economics, Neuherberg, Germany; and the Hospital of Augsburg (C.G., C.M.), MONICA/KORA Myocardial Infarction Registry, Augsburg, Germany.

Correspondence to Bernhard Kuch, MD, I. Medizinische Klinik, Klinikum Augsburg, Stenglinstr. 2, 86 156 Augsburg, Germany. E-mail bernhard.kuch{at}klinikum-augsburg.de

Received October 27, 2008; accepted March 18, 2009.

Background— No data exist regarding time trends of 28-day case fatality (CF) of patients with presumed acute myocardial infarction (AMI) using epidemiological criteria, clinical criteria, and AMI classification after validation of presumed in-hospital AMI-related deaths (gold-standard criteria).

Methods and Results— From 1985 to 2004, we prospectively examined all 9210 AMI patients consecutively hospitalized in a large teaching hospital by using a broad epidemiological AMI definition (WHO-MONICA). Twenty-eight-day CF decreased significantly from 32% in 1985–1986 to 18% in 2003–2004, mostly because of a reduction in early deaths (<24 hours). When applying the clinical AMI definition, most of the early deaths were not counted as AMI related. A retrospective validation process from a sample of all early deceased patients by the epidemiological AMI definition (388/2076) and a prospective validation of the complete cohort in 2005–2006 revealed that only about 50% of early deaths are reclassified as a real fatal AMI using newer criteria resulting in a 28-day CF of 23% in 1985–1986 and 11% in 2005–2006. The difference between the AMI 28-day CF by applying gold-standard criteria and the clinical AMI 28-day CF (18% in 1985–1986 and 7% in 2005–2006) has decreased during recent years.

Conclusions— The application of broad epidemiological criteria for AMI overestimates 28-day CF by almost 2-fold compared with gold-standard criteria (after validation of early deaths) and almost 3-fold compared to the clinical definition. The growing similarity in 28-day CF between the clinically based definition and the gold-standard criteria implies that recent clinical-based registries may represent a realistic picture of trends regarding in-hospital AMI mortality.

Key Words: myocardial infarction • epidemiology • prognosis