Skip to main content

Research Repository

Advanced Search

The development and course of heart failure after a myocardial infarction

Torabi, Azam


Azam Torabi


John G. F. Cleland


Introduction: Robust epidemiological data on the incidence of myocardial infarction (MI) are hard to find, but synthesis of data from a number of sources indicates that the average hospital in the UK should admit about two patients with a first MI and one recurrent MI per 1000 population per year. Although age-adjusted incidence may be declining, this may be offset by greater longevity in the general population. The incidence of acute coronary syndromes (ACS) is much higher. The incidence and outcome of both ACS and of MI will depend greatly on how data are collected.

The cumulative incidence, persistence and resolution of heart failure (HF) after an MI in the general population are poorly described. Cardiac dysfunction subsequent to MI is a common cause for morbidity and mortality, however, there are few data on what proportion of longterm survivors of MI has important cardiac dysfunction and/or HF.

The aim of this thesis is to describe the incidence and outcome of MI in the general population and in different age groups, explain the natural history and prognosis of HF after an MI both during the index admission and long-term (6 year) follow-up in relationship to the presence of HF and also to determine the utility of amino-terminal pro-brain natriuretic peptide (NT-proBNP) alone and in conjunction with other clinical data, as a marker of left ventricular systolic dysfunction (LVSD) and subsequent prognosis in long-term survivors of MI. Subgroup data according to age and anaemia will also be reported.

Methods: Patients with a death or discharge diagnosis of MI in 1998 were identified from records of hospitals providing services to a local community of 560,000 people. Records were scrutinized to identify the development of HF, defined as symptoms and signs consistent with that diagnosis and treated with loop diuretics. HF was considered to have resolved if diuretics could be stopped without recurrent symptoms. Analyses were done on the whole population and then sub-groups by age (75 years) and anaemia status. Anaemia was defined according to WHO criteria (men haemoglobin (Hb)1 g below threshold), borderline (within 1g of threshold) and (>1g above threshold).

In 2004, surviving patients were invited to attend for clinical assessment, an echocardiogram and measurement of NT-proBNP and were subsequently followed until 31st December 2009 using medical records.

Also in 2005, another group of patients admitted with ACS to cardiology or general medical wards were identified prospectively by trained nurses from 1st January to 31st December 2005. Patients with a death or discharge code of MI were also identified by the hospital information department and from Myocardial Infarction National Audit Project (MINAP) records.

Results: For the first cohort, 896 patients were identified of whom 54% had died by December 2005. During the index admission, 199 (22%) patients died, many with HF, and a further 182 (20%) patients developed HF that persisted until discharge, of whom 121 died subsequent to discharge. Of 74 patients with transient HF that resolved before discharge, 41 had recurrent HF and 38 died during follow-up. After discharge, 145 (33%) patients developed HF for the first time, of whom 76 died during follow-up. Overall, of 281 deaths occurring after discharge, of which 235(84%) were amongst patients who first developed HF.

Of 896 patients, 311 were aged 75 years of whom, respectively, 24%, 57% and 82% had died by December 2005. During the index admission, by age group, 24 (8%), 68 (23%) and 107 (37%) patients died in each group, many with HF, and a further 37 (12%), 63 (21%) and 82 (29%) developed HF that persisted until discharge. After discharge, 53 (24%), 55 (40%) and 37 (47%) patients developed HF for the first time. Overall, of 51, 102 and 128 deaths occurring after discharge, 35 (70%), 93 (91%) and 107 (85%) were among patients who first developed HF.

Of 855 patients with an available hemoglobin during index admission, 103 were anaemic, 280 were borderline and 472 were not anaemic based on the first available haemoglobin during the index admission. 300 patients had more than one measurement of haemoglobin, of which 125 (85 unchanged status from first assessment) had definite, 289 (237 unchanged) had borderline and 441 (424 unchanged) had no anaemia on the last available measurement. During the index admission, 77 patients (75%) with definite, 130 (46%) with borderline and 196 (42%) who had no anaemia on the first available haemoglobin developed HF, of whom 41 (53%), 50 (38%) and 60 (31%) died during the admission compared, respectively, to 7 (27%), 14 (9%) and 9 (3%) deaths in patients who did not develop HF. During a six year follow-up, 543 (64%) patients developed HF and 456 (53%) died. Amongst patients with HF during the index admission, the six year mortality rates in those with definite, borderline and no anaemia (last available index admission measurement) were, 90%, 84% and 64% (P=0.0001). In patients without HF on the index admission, 6-year mortality rates were 62%, 42% and 24% (P=0.0001). Anaemia (last available index admission measurement) predicted all-cause mortality independent of the presence of HF (p=0.055).

451 had died by 2004 and only 414 were available for follow-up, of whom 175 patients attended and had NT-proBNP measured. Of these, 51 (29%) patients had LVSD, 66 (38%) had NT-proBNP >50pmol/L (423pg/ml), 86 (49%) had one or the other and 31 (18%) had both. Patients with higher NT-proBNP were more likely to have HF (and be treated with diuretics), LVSD (and therefore treatment with ACE inhibitors), a dilated atrium, substantial mitral regurgitation and atrial fibrillation (and therefore treatment with warfarin and digoxin) (p=0.0001). Thirty six patients died during follow-up; 28 (42%) with an NT-proBNP >50pmol/L (423pg/ml) (77% of all deaths). ROC curves suggested that NT-proBNP 56pmol/L (474pg/ml) had the highest sensitivity (78%) and specificity (77%) for predicting death (AUC 0.78). Echocardiography added little to the prognostic information provided by NT-proBNP alone.

In 2005, the prospective survey identified 1,731 admissions (1,439 patients) with ACS, of which 764 (704 patients) were for MI. The hospital information department reported only 552 admissions (544 patients) with MI and only 206 admissions (203 patients) were reported to MINAP. Using all three data-bases, 934 admissions (873 patients) for MI were identified, for which TnT was >1ug/L in 443, 0.04 to 1.0 in 435, 0.03ug/L but did not have ACS ascertained by any survey method. Of 873 patients with MI, 146 died during admission (17% versus 22% in the 1998 cohort) and 218 by one year.

Conclusion: The incidence of ACS/MI is highly dependent on the methodology for case-ascertainment and the method used to identify cardiac damage (for instance the sensitivity of the troponin assay used). The development of HF precedes death in most patients who die in the short- or longer-term following an MI. The risk of developing HF and dying after an MI increases progressively with age and anaemia. In patients with a remote history of MI, elevated NT-BNP identifies patients with a high prevalence of LVSD. Regardless of age, most deaths are preceded by the development of HF. Anaemia is associated with a high mortality even in the absence of HF.

Prevention of HF, by reducing the extent of myocardial damage and recurrent MI and by subsequent good management could have a substantial impact on prognosis.


Torabi, A. (2011). The development and course of heart failure after a myocardial infarction. (Thesis). University of Hull. Retrieved from

Thesis Type Thesis
Deposit Date Dec 14, 2012
Publicly Available Date Feb 22, 2023
Keywords Medicine
Public URL
Additional Information Postgraduate Medical Institute, The University of Hull
Award Date Jan 1, 2011


Thesis (1.8 Mb)

Copyright Statement
© 2011 Torabi, Azam. All rights reserved. No part of this publication may be reproduced without the written permission of the copyright holder.

You might also like

Downloadable Citations