Background There is currently no classification for acute myocardial infarction (AMI) according to left ventricular ejection fraction (LVEF). We aimed to perform a retrospective analysis of patients undergoing emergency percutaneous coronary intervention (PCI), comparing the clinical characteristics, in-hospital acute heart failure and all-cause death events of AMI patients with mid-range ejection fraction (mrEF), preserved ejection fraction (pEF) and reduced ejection fraction (rEF).
Material and methods Totally 1270 patients were stratified according to their LVEF immediately after emergency PCI into pEF group (LVEF 50% or higher), mrEF group (LVEF 40%–49%) and rEF group (LVEF <40%). Kaplan-Meier curves and log rank tests were used to assess the effects of mrEF, rEF and pEF on the occurrence of acute heart failure and all-cause death during hospitalisation. The Cox proportional hazards model was used for multivariate correction.
Results Compared with mrEF, rEF was an independent risk factor for acute heart failure events during hospitalisation (HR 5.01, 95% CI 3.53 to 7.11, p<0.001), and it was also an independent risk factor for all-cause mortality during hospitalisation (HR 7.05, 95% CI 4.12 to 12.1, p<0.001); Compared with mrEF, pEF was an independent protective factor for acute heart failure during hospitalisation (HR 0.49, 95% CI 0.30 to 0.82, p=0.01), and it was also an independent protective factor for all-cause death during hospitalisation (HR 0.33, 95% CI 0.11 to 0.96, p=0.04).
Conclusions mrEF patients with AMI undergoing emergency PCI share many similarities with pEF patients in terms of clinical features, but the prognosis is significantly worse than that of pEF patients, suggesting that we need to pay attention to the management of mrEF patients with AMI.
- acute myocardial infarction
- early prognosis
- mid-range ejection fraction
Statistics from Altmetric.com
Unlike heart failure, there is no classification for acute myocardial infarction (AMI) according to left ventricular ejection fraction (LVEF). However, LVEF has been confirmed to be closely related to the prognosis of AMI patients.1 In the light of the previous classification method of heart failure, the boundary value of LVEF between heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF) was set at 50%, but more and more studies noted that most clinical characteristics of heart failure patients with ejection fraction ranging from 40% to 49% were in between HFpEF and HFrEF, some of which close to HFpEF while some close to HfrEF.2–4 This suggested that HFmrEF is a unique subtype of cardiac insufficiency clinically and pathophysiologically different from HFrEF and HFpEF.5 Furthermore, the European Society of Cardiology divided heart failure into three categories in 2016 based on the LVEF: HFpEF (LVEF ≥50%), HFrEF (LVEF <40%) and heart failure with mid-range ejection fraction (HFmrEF, LVEF 40%–49%), which initiated extensive researches on the clinical features, pathophysiology, treatment and prognosis of HFmrEF.6 At present, the evidence for the clinical features and prognosis of AMI patients with LVEF 40%–49% is obviously insufficient. It is worthwhile to study whether AMI with mid-range ejection fraction (mrEF) is significantly different from that with reduced ejection fraction (rEF) and preserved ejection fraction (pEF on clinical manifestation and prognosis. We, therefore, carried out this retrospective analysis on AMI patients undergoing emergency percutaneous coronary intervention (PCI) grouped by LVEF value within 48 hours from admission and compared mrEF patients with pEF and rEF patients in clinical features and in-hospital prognosis. This is the first study regarding to AMI patients with different EF range after emergency PCI according to the classification method announced by European Society of Cardiology in 2016. The aim of this study is to provide more evidence for AMI patients with mrEF after emergency PCI, instructing clinicians with better management and early prognostic evaluation of mrEF patients.
This study is a single-centre retrospective cohort study, including 1270 hospitalised patients with AMI undergoing emergency PCI between January 2012 and June 2017 at the First Affiliated Hospital of Soochow University (figure 1). In this study, patients with AMI after emergency PCI were divided into pEF group (LVEF ≥50%), mrEF group (LVEF 40%–49%) and rEF group (LVEF <40%) according to the value of LVEF obtained within 48 hours from admission.
(1) Patients who had no myocardial infarction history before and have indications of emergency PCI this time; (2) patients who are in accordance with the diagnostic criteria for AMI (type 1) and (3) available for a record of echocardiographical examination within immediately after emergency PCI.
(1) Myocardial infarction associated with interventional therapy or cardiac surgery; (2) having a history of myocarditis, congenital heart disease, cardiomyopathy, arterial inflammation or abnormal vascular development; (3) the primary diagnosis of AMI was finally ruled out by coronary angiography, echocardiography, cardiac biomarkers; (4) pre-existing severe liver and kidney diseases and dialysis patients and (5) combined with severe infection, trauma, bleeding, rheumatic or autoimmune disease and malignant tumour, which obviously affects the prognosis during hospitalisation.
Timing for reperfusion therapy
In early presenters (ie, those with ST-segment elevation myocardial infarction (STEMI) diagnosis within 3 hours from symptom onset), a primary PCI strategy is the reperfusion strategy of choice. After 3 hours (and up to 12 hours) of symptoms onset, the later the patient presents, the more consideration should be given to a emergency PCI strategy as opposed to administering fibrinolytic therapy. In evolved STEMI (12–48 hours after symptoms onset), a routine emergency PCI strategy should be considered in all patients. After 48 hours, a routine emergency PCI strategy is not recommended. Regardless of the time from symptoms onset, the presence of ongoing symptoms suggestive of ischaemia, haemodynamic instability or life-threatening arrhythmias is the indication for emergency PCI.
Measurement of clinical and laboratory parameters
The specific data to be collected are as follows: (1) the basic information of patients: name, age, gender, date of admission, date of discharge, home address, contact method; (2) previous history: hypertension, diabetes, previous history of angina; (3) history of smoking: according to WHO's definition of smoking, that is, one or more cigarettes a day, for a consecutive year or more; (4) basic conditions of admission: blood pressure, infarct location on ECG, symptom onset to admission time; (5) results of blood routine and biochemical examination within 48 hours from admission determined by the clinical laboratory of the First Affiliated Hospital of Soochow University: N-terminal pro-B-type natriuretic peptide (NT-pro-BNP), creatinekinase-MB (CK-MB), troponin I, hypersensitive C reactive protein (hs-CRP), alanine transaminase, creatinine, lipoprotein (a), platelet count, haemoglobin, white cell count, neutrophil ratio. (6) echocardiography results within 48 hours from admission: LVEF value; (7) coronary angiography: number of diseased branches, X-ray exposure time.
All-cause deaths during hospitalisation are defined as deaths of any cause during hospitalisation, and acute heart failure during hospitalisation is diagnosed according to the European Society of Cardiology (ESC) Guidelines6 (including acute onset of heart failure symptoms, such as sudden onset of dyspnoea; chest X-ray; NT-proBNP; echocardiography and Framingham criteria for heart failure) during hospitalisation.
Continuous data were expressed in the form of mean±SD or the median (IQR). Categorical variables were presented as event number (rate). The comparison between the three groups of continuous variables were performed by analysis of variance and post hoc analysis. The comparison of categorical variables was performed by the X2 test. Fisher’s exact probability method was used when event number was less than 5. The effects of mrEF, pEF and rEF on the occurrence of acute heart failure and all-cause mortality duringhospitalisation were evaluated by Kaplan-Meier method and log rank test. Cox proportional hazards model was used to adjust the age, gender, hypertension, diabetes mellitus, smoking history and the number of diseased coronary arteries. All data analyses were performed by SPSS V.24.0 software (SPSS). A p<0.05 indicated that the difference was statistically significant.
In this study, 1270 patients with AMI were treated with PCI, of whom 1048 cases were male, accounting for 82.5%, and 222 cases were female, accounting for 17.5%. The average age was 62.5±13.9 years. A total of 875 patients (68.9%) had hypertension and 297 patients (23.4%) had diabetes mellitus. A total of 800 patients (63.0%) had smoking history and 445 patients (35.0%) had previous angina pectoris issues. A total of 560 cases had multiple lesions, accounting for 44.1%. The average time from symptom to admission was 6.5±4.5 hours. A total of 149 cases were classified into rEF group (11.7%), 664 cases into mrEF group (52.3%) and 457 cases into pEF group (36.0%). During hospitalisation, there were 152 events (12.0%) of acute heart failure and 66 events (5.2%) of all-cause deaths. Comparison of patient characteristics during hospitalisation among three groups was shown in table 1.
Compared with AMI patients with rEF, patients with mrEF had younger age, higher proportion of male, higher percentage of smokers, higher proportion of previous episodes of angina, higher systolic blood pressure, higher diastolic pressure at admission, lower NT-pro-BNP, lower CK-MB, lower troponin I, lower hs-CRP, lower alanine aminotransferase, lower creatinine, lower lipoprotein (a), higher haemoglobin, lower white cell count, lower neutrophils, lower proportion of multivessel disease, lower X-ray exposure, and all of these were significantly statistically different (p<0.05).
Compared with AMI patients with pEF, patients with mrEF had higher rates of diabetes mellitus, higher proportion of anterior wall myocardial infarction, higher NT-pro-BNP, higher CK-MB, higher troponin I, higher hs-CRP, higher creatinine, higher white cell count, higher neutrophils, longer X-ray exposure time, with significant statistical differences (p<0.05).
In-hospital acute heart failure events
The difference of ‘survival’ distribution among three groups was tested by log rank method, and the results showed that the difference of ‘survival’ distribution among three groups was statistically significant (χ2=263.0, p<0.001). Figure 2 indicates the Kaplan-Meier survival curve for acute heart failure according to LVEF groups.
Multivariate Cox regression analysis showed that age (HR 1.05, 95% CI 1.04 to 1.07, p<0.001), hypertension (HR 1.66, 95% CI 1.07 to 2.56, p=0.02) and number of lesions (HR 1.88, 95% CI 1.54 to 2.31, p<0.001) were independent risk factors for acute heart failure during hospitalisation (table 2).
Compared with mrEF, rEF was an independent risk factor for acute heart failure events during hospitalisation (HR 5.01, 95% CI 3.53 to 7.11, p<0.001), pEF was an independent protective factor for acute heart failure during hospitalisation (HR 0.49, 95% CI 0.30 to 0.82, p=0.01).
In-hospital all-causes death events
The difference of survival distribution in three groups was tested by log rank method, and the results showed that the difference of survival distribution among three groups was statistically significant (χ2=167.3, p<0.001). Figure 3 indicates the Kaplan-Meier survival curve for all-causes death according to LVEF groups.
Multivariate Cox regression analysis showed that female (HR 1.07, 95% CI 1.05 to 1.09, p<0.001) and number of lesions (HR 1.68, 95% CI 1.22 to 2.31, p=0.002) were independent risk factors for all-cause death during hospitalisation (table 3).
Compared with mrEF, rEF was an independent risk factor for all-cause mortality during hospitalisation (HR 7.05, 95% CI 4.12 to 12.1, p<0.001); pEF was an independent protective factor for all-cause death during hospitalisation (HR 0.33, 95% CI 0.11 to 0.96, p=0.04).
Previous studies reporting the negative correlation between LVEF and mortality were mainly studied in patients with left ventricular systolic dysfunction (usually EF <40%), showing a significant selection bias7–10 as many AMI patients with pEF could also develop into heart failure. In addition, because of the relatively older age of the patients with HFpEF, some of the patients with high risk of cardiovascular adverse events were excluded, which might lead to bias. Moreover, HFpEF was more likely to occur in female patients, which would affect the results to some extent. The CHARM study had overcome some of these limitations by including the normal ejection fraction patients and elderly female patients.11 The Candesartan in Heart failure-Assessment of moRtality and Morbidity (CHARM) study showed that in patients with symptomatic heart failure, LVEF was an important predictor of fatal and non-fatal cardiovascular outcomes, including death and rehospitalisation, sudden death and fatal and non-fatal myocardial infarction, but this was only seen in those with moderate to severe left ventricular systolic dysfunction. In patients with LVEF higher than 45%, LVEF was not associated with cardiovascular prognosis, suggesting that patients with LVEF higher than 45% had similar prognostic characteristics.12–14 At present, the prognosis of AMI patients with LVEF 40%–49% after emergency PCI remains unknown. However, more and more studies have noted that most of the clinical features of 40%–49% in patients with congestive heart failure are in between HFpEF and HFrEF, with some indicators close to HFpEF, and some close to HFrEF, suggesting that HFmrEF is a unique subtype of cardiac insufficiency clinically and pathophysiologically different from HFrEF and HFpEF.6 Therefore, the purpose of this study is to provide more evidence on the clinical features and prognosis of AMI patients with LVEF 40%–49% after PCI.
Our findings indicated that the clinical features of AMI patients with mrEF after emergency PCI were very close to those with pEF (age, gender ratio, history of hypertension, smoking history, previous angina pectoris episode, symptom onset to admission time, systolic pressure, diastolic pressure, alanine transaminase, lipoprotein (a), platelet count, haemoglobin and multiple lesion proportions had no statistically significant differences), but were significantly different from those with rEF (hypertension history, history of diabetes, onset to admission time, anterior wall infarction ratio, platelet count had no statistically significant differences, but the rest all had), However, during hospitalisation, the prognosis of AMI patients with mrEF was significantly worse than pEF patients. When AMI patients with pEF were compared with mrEF, the HR for incidence of acute heart failure events during hospitalisation was 0.49 (95% CI 0.30 to 0.82, p=0.01); the HR of the all-cause death event was 0.33 (95% CI 0.11 to 0.96, p=0.04). The prognosis of AMI patients with rEF was significantly worse than those with mrEF, with the HR 5.01 (95% CI 3.53 to 7.11, p<0.001) for acute heart failure and 7.05 (95% CI 4.12 to 12.1, p<0.001) for all-cause death during hospitalisation.
These findings suggested that after emergency PCI in AMI, AMI patients with mrEF were independent subgroup with different characteristics compared with those with rEF or pEF, and occupied a large proportion of the patients population. Another noteworthy issue was that in patients with AMI, the LVEF was in a dynamic process after emergency PCI, and there were two main trends. One is that the myocardial stunning was relieved gradually after the timely reperfusion, and some patients would obtain an improvement on the LVEF.15 On the other hand, irreversible myocardial necrosis could contribute to impaired LVEF, chronic cardiac dysfunction and ventricular pathological remodelling, resulting in heart failure and death.16 In this study, the LVEF value of echocardiography obtained within 48 hours from admission was used for the reference of grouping, and the effect of dynamic changes of LVEF on the results was reduced. Previous studies have shown that for patients with AMI undergoing emergency PCI, the LVEF value at 3 days after operation was an important predictor of heart failure and death during hospitalisation,17 which is consistent with the conclusion of our study. Burns et al have found that baseline LVEF <40% was the strongest predictor of the 1-year mortality rate after adjustment, and LVEF can be used for risk stratification in short-term and long-term cardiovascular outcomes.18 In addition, previous study showed that the improvement in LVEF was more obvious in patients with lower baseline LVEF.19 The European Society of Cardiology suggested that patients with HFmrEF should be an independent subgroup. Moreover, recent evidence suggests that heart failure patients with mrEF or rEF are more likely to be complicated with ischaemic heart disease than patients with pEF,20 indicating that some AMI patients with 40%–49% ejection fraction developed into HFmrEF after emergency PCI, which is consistent with the conclusion of our study.
The multivariate Cox regression analysis found that age was an independent risk factor for the occurrence of acute heart failure during hospitalisation (HR 1.05, 95% CI 1.04 to 1.07, p<0.001), possibly because the elderly patients tended to be complicated with fibrosis of the myocardium. The ventricular wall elasticity decreased and ventricular filling capacity was impaired in varying degrees.21 In addition, women (HR 1.07, 95% CI 1.05 to 1.09, p<0.001), number of diseased artery (HR 1.68,95% CI 1.22 to 2.31, p=0.002) are independent risk factors for all-cause deaths during hospitalisation. The effect of sex on the prognosis of patients with myocardial infarction has been a hot topic in previous years, and most studies have reported higher short-term mortality among women. Possible reasons for the poorer outcomes for women include: women tend to have a higher average age of onset than men, more likely to be complicated with diabetes, heart failure and other complications, and some have prolonged ischaemic time due to atypical chest pain delays.22 The number of coronary lesions suggests the area of myocardial necrosis. The larger the infarct area, the greater the likelihood of in-hospital death.
To sum up, the clinical characteristics of AMI patients with mrEF after emergency PCI are very close to those with pEF, but the prognosis is significantly worse than that with pEF. The clinical features and prognosis between mrEF and rEF patients are significantly different. This suggests that we should pay more attention to the management of AMI patients with mrEF after PCI, and more research is needed to optimise the treatment strategies for patients with mrEF after emergency PCI.
This study is a single-centre retrospective cohort study. Although the sample size is relatively large, we only collected data on prognosis during hospitalisation. More large-scale, multicentre, long-term follow-up studies are needed in the future to further reveal the characteristics and prognosis of AMI patients with mrEF, in order to provide better recognition and early prognosis evaluation of patients with mrEF.
The clinical characteristics of AMI patients with mrEF after emergency PCI were similar to those of pEF patients, but the prognosis was significantly worse than that of pEF patients, suggesting that we should pay more attention to the management of AMI patients with mrEF. The clinical features and prognosis of patients with mrEF and rEF were significantly different. In addition, age, hypertension and number of diseased branches were independent risk factors for the occurrence of acute heart failure during hospitalisation, and females and the number of diseased branches were independent risk factors for all-cause deaths during hospitalisation.
This was a single real world study.
We set up a brand-new classification method for acute myocardial infarction (AMI) patients after percutaneous coronary intervention (PCI).
Mid-range ejection fraction (mrEF) patients share similar clinical features with preserved ejection fraction patients but poorer prognosis.
Heart failure patients with mrEF or reduced ejection fraction are more likely to be complicated with coronary heart disease (CHD).
Current research questions
Patients’ characteristics in different groups of ejection fraction after AMI.
Prognostic significance of mrEF in patients with AMI after emergency PCI.
The outcome of patients with AMI after revascularisation in China.
What is already known on the subject
The European Society of Cardiology divided heart failure patients with ejection fraction (EF) of 40% to 49% into an independent subgroup in 2016.
However, there is currently no classification for acute myocardial infarction according to left ventricular ejection fraction. The evidence on clinical characteristics and prognosis of patients with acute myocardial infarction accompanied by EF of 40% to 49% is obviously insufficient.
Whether AMI with mid-range EF is significantly different in clinical presentation and prognosis from that of reduced EF and preserved EF is a pending clinical issue to besolved.
YJ and SH contributed equally.
Contributors YJ and SH: main authors of the study, established the idea to study the mid-range ejection fraction patients of AMI in Chinese population. Writing main ideas for this research, main results and discussion of the findings. MC, XL, JZ and BD: interpreted statistical analysis and conducted multivariate analysis to prove the main findings of this project. YZ: corresponding author of the study, contributed on editing this manuscript and giving advice for the main authors to organise the manuscript and ideas of the project. FZ and TC: contributed with study methodology.
Funding This work was supported by grants from National Natural Science Foundation of China (81873486), Natural Scientific Fund of Jiangsu province (BK20161226), Jiangsu Province’s Key Provincial Talents Program (ZDRCA2016043), Jiangsu Province’s 333 High-Level Talents Project (BRA2017539).
Disclaimer The funders had no roles in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval Our study was approved by the ethics committee of the First Affiliated Hospital of Soochow University (IRB No.2018096).
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available on reasonable request.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.