Article Text
Abstract
Objective The medium-term outcome and cause of death in patients with myocardial infarction with non-obstructive coronary arteries (MINOCA) is not well characterised. The aim of this study was to compare mortality and rates of recurrent events in post myocardial infarction (MI) patients with obstructive coronary artery disease (CAD) and in patients with MINOCA compared with an age and sex-matched cohort without cardiovascular disease (CVD).
Methods We performed a national cohort study of consecutive patients undergoing coronary angiography for MI during 2 years between 2013 and 2015 from the All New Zealand Acute Coronary Syndrome—Quality Improvement (ANZACS QI) registry. MI patient registry data were linked anonymously to national hospitalisation and mortality records. Age and sex matched patients without known CVD formed the comparison group.
Results Of the 8305 patients with MI, 897 (10.8%) were classified as MINOCA. Compared with those without known CVD, the adjusted HRs for the primary outcome (all-cause death or recurrent non-fatal MI) were 7.81 (95% CI 6.64 to 9.19, p<0.0001) in those with obstructive CAD and 4.64 (95% CI 3.54 to 6.10, p<0.0001) in those with MINOCA. Kaplan-Meier all-cause mortality at 2 years was 7.9% for those with obstructive CAD, with nearly half being CVD deaths (3.6% CVD deaths and 4.5% non-CVD deaths, respectively). In contrast, MINOCA all-cause mortality was 4.9% with non-CVD death (4.5%) predominating.
Conclusions MINOCA is common and has an adverse outcome rate approximately half than that of those with obstructive CAD. The predominant contributor to mortality is non-CVD death. The rate of events in MINOCA is significantly greater than the population without CVD.
- coronary artery disease
- acute coronary syndromes
- cardiac catheterization and angiography
- acute myocardial infarction
Statistics from Altmetric.com
- coronary artery disease
- acute coronary syndromes
- cardiac catheterization and angiography
- acute myocardial infarction
Introduction
Patients with non-obstructive coronary artery disease (CAD) constitute approximately 10% of patients presenting with ST-segment elevation myocardial infarction (STEMI) or non–ST-segment elevation myocardial infarction (NSTEMI) and this presentation is more common in women.1–3 Early mortality for NSTEMI patients without obstructive CAD is reported at 0.6%,2 4 significantly lower than the 2.8% early mortality in those with obstructive CAD.4 However, these reported outcomes do not include patients with STEMI.
Series including patients presenting with either STEMI or NSTEMI without significant coronary artery obstruction report a 1-year to 2-year mortality of 3.3%5 to 4.4%.6 There is limited information regarding medium-term mortality and cause of death in real-world patients presenting with myocardial infarction with non-obstructive coronary arteries (MINOCA),7 8 and their risk in relation to the normal age and sex-matched population is unknown.
The aim of this study was to determine medium-term mortality, cause of death and rates of recurrent myocardial infarction (MI) in patients presenting with acute MI stratified by the presence of obstructive CAD, by MINOCA and compared with an age and sex-matched cohort without prior cardiovascular disease (CVD).
Methods
We performed a national analysis of consecutive patients undergoing coronary angiography for MI between 1 November 2013 and 31 October 2015. Individual patient demographic, risk factor, diagnostic, treatment and outcome data were collected prospectively using regional data from the All New Zealand Acute Coronary Syndrome—Quality Improvement programme (ANZACS-QI).9 Individual patient registry data were linked anonymously to national hospitalisation and mortality records ensuring complete ascertainment of adverse outcomes.9 This study complies with the Declaration of Helsinki and is part of the VIEW research programme which was approved by the Northern Region Ethics Committee Y in 2003, with subsequent amendments to include the ANZACS-QI registries and annual reapproval by the national Multi-Region Ethics Committee since 2007. Individual patient consent was not required as all data are anonymised.
Data and definitions
MI was defined according to the ‘Third universal definition’.10 The study population included both those with STEMI and NSTEMI. Cardiac biomarker levels were measured with high sensitivity troponin assays. Only patients with type 1 MI as classified by the admitting cardiologist were included in this analysis. Patients diagnosed with non-acute coronary syndrome (ACS) conditions, including stress cardiomyopathy and myocarditis, were excluded. Classification as MINOCA required no ≥50% diameter stenosis on coronary angiography in all major epicardial coronary arteries as visually estimated by the reporting cardiologist.11 Those classified as MINOCA were subdivided into those with completely normal coronary arteries and those with mild disease (>0 to <50% diameter stenosis).
No prior CVD comparison cohort
The comparison population were patients without prior CVD assessed using the PREDICT software between 1 November 2013 and 12 October 2015, sourced mainly from participating primary health organisations (PHOs) in the Auckland and Northland regions of New Zealand.12 These PHOs provide primary healthcare for approximately 80% of the region’s population and the PREDICT-CVD cohort comprises 4 93 993 individuals, about 90% of those eligible for CVD risk assessment in the region. Exclusion criteria for the ‘no prior CVD’ cohort were the following: any prior CVD (including angina, hospitalisation for ischaemic heart disease, transient ischaemic attack, cerebrovascular disease or peripheral vascular disease); previous angina treatment (including dispensed glyceryl trinitrate, isosorbide dinitrate, isosorbide mononitrate, nicorandil or perhexiline); prior history of high CVD risk condition (including atrial fibrillation, diabetes with confirmed microalbuminuria or overt nephropathy, estimated glomerular filtration rate ≤30 mL/min/1.73 m2 and those admitted for heart failure and/or dispensed loop diuretics as a proxy for congestive heart failure). These additional risk factors are obtained from the PREDICT-CVD risk assessment or through linkage to national hospitalisation and drug dispensing databases and a regional laboratory results database. An age and sex-matched ‘no prior CVD’ cohort in a ratio of one PREDICT patient to one MI patient was obtained from patients having their first PREDICT-CVD risk assessment between 1 November 2013 and 12 October 2015. A simple random sample was taken without replacement from each stratum in the PREDICT cohort using PROC SURVEYSELECT. In total, a sample of 8305 subjects matched on sex and age group (<40, 40–<50, 50–<60, 60–<70, 70+years) were selected.
Outcomes
Outcome data are derived from the ANZACS-QI in-hospital outcomes and national mortality and hospitalisation databases. The primary outcome reported is the composite of all-cause mortality or non-fatal MI. Deaths were categorised as CVD or non-CVD from International Classification of Diseases, 10th Revision, Australian Modification coded national hospital and mortality datasets,13 with CVD death defined as deaths associated with atherosclerotic cardiac, cerebral or peripheral vascular disease (online supplementary table 2). To identify readmissions with further MI, a primary MI code on a subsequent admission was required. The process of anonymised linkage of the ANZACS-QI registry with the national datasets has been previously described.9
Supplementary file 1
Statistical analysis
Continuous variables are summarised as mean with SD and compared using Student’s t-test. Categorical data are reported by frequency and percentage and compared using Pearson’s χ2test or Fisher’s exact test as appropriate.
Survival analyses included Kaplan-Meier curves in MI patients with and without obstructive CAD and in the ‘no prior CVD’ cohort for the composite outcome of all-cause mortality or non-fatal MI. To quantify the relative risk of outcomes in the three patient groups, Cox regression models were used to derive adjusted HRs for all-cause mortality, CVD death, non-CVD death, non-fatal MI and the composite outcome. Covariates in the models were age, sex, systolic blood pressure (SBP), smoking, diabetes and total cholesterol to high-density lipoprotein (TC/HDL) ratio.
An additional set of models were derived adjusting for covariates obtained during the hospital admission which included age, sex, heart rate, SBP at admission, Killip score,14 cardiac arrest at presentation, ST-segment depression, elevated cardiac enzymes, creatinine, treatment with: aspirin, HMG-CoA (5-hydroxy-3-methylglutaryl-coenzyme A) reductase inhibitor (statin), second antiplatelet agent, ACE inhibitor/angiotensin receptor blocker (ARB) or beta-blocker. For each outcome, the assumption of proportional hazards was assessed by the interaction between the independent variable and the logarithm of time. All p-values reported are two tailed and a p-value <0.05 was considered statistically significant. Data were analysed using SAS statistical package, version 9.4.
Results
There were 8305 patients with MI, 897 (10.8%) of these with MINOCA, 7.1% of men and 19.3% of women (table 1). Compared with patients with obstructive CAD, those with MINOCA were younger (mean 63.1±12.5 vs 65.1±12.0 years, p<0.001), more likely to be: female (54.3% vs 27.5%, p<0.001), lifelong non-smokers (46.0% vs 42.1%, p=0.04), non-diabetic (84.3% vs 77.3%, p<0.01); had lower TC/HDL cholesterol (3.81±1.43 vs 4.42±2.17, p<0.001), lower LDL cholesterol (2.53±1.15 vs 2.72±1.25 mmol/L, p<0.001) and higher SBP (143.3±27.2 vs 141.0±27.2, p=0.021). They were also more likely to present with NSTEMI (90.0% vs 64.2%, p<0.001), have lower rates of heart failure (Killip classes II, III, IV; 5.8% vs 9.4%), in-hospital death (0.2% vs 1.5%, p=0.002) and a higher prevalence of normal left ventricular ejection fraction (56.6% vs 43.7%, p<0.001). The only significant difference in the prevalence of other comorbidities in patients with MINOCA was higher rates of pulmonary disease (7.3% vs 4.1%, p<0.001). The characteristics of the age and sex-matched subjects with no prior CVD are detailed in table 1.
The characteristics of the MINOCA subgroups, MI with normal coronary arteries and MI with mild CAD (>0 to <50% stenosis) are presented in online supplementary table 1. Compared with those with mild CAD, those with normal coronary arteries were: younger (mean 60.1±12.9 vs 64.7±12.0 years, p<0.001), more likely to be female (63.3% vs 49.6%, p<0.001), non-diabetic (89.3% vs 81.7%, p=0.003) and had lower rates of prior CVD (14.3% vs 30.2%, p<0.001).
Patients with MINOCA were less likely to be treated with (table 2) aspirin (90.4% vs 97.2%, p<0.001); dual antiplatelet therapy (65.1% vs 82.3%, p<0.001); statin (89.8% vs 95.8%, p<0.001); and triple therapy with statin, antiplatelet agent and ACE inhibitor and/or beta-blocker (74.9% vs 89.7%, p<0.001). In comparison to those with mild CAD, those with normal coronary arteries were also less likely to receive aspirin (82.7% vs 94.4%, p<0.001), dual antiplatelet therapy (50.2% vs 73.0%, p<0.001), statin (82.7% vs 93.5%, p<0.001) and triple therapy (64.2% vs 80.4%, p<0.001).
Outcomes
The cumulative 2-year Kaplan-Meier incidence of death or non-fatal recurrent MI in patients with MI and obstructive CAD was 19.1% (mean follow-up 1.01±0.62 years) compared with 3.0% in those without prior CVD (mean follow-up 1.44±0.53 years, table 3 and figure 1). Patients with normal coronary arteries and patients with mild obstructive CAD had similar 2-year incidences of the composite outcome, 11.7% (mean follow-up 1.13±0.60 years) and 10.6% (mean follow-up 1.09±0.57 years), respectively. Patients with obstructive CAD had a 3.6% incidence of CVD death and a 4.5% incidence of non-CVD death. Patients with MINOCA had a similar low incidence of CVD death compared with those with no prior CVD (0.39% vs 0.41%) and no difference after adjustment for age, sex, SBP, smoking, diabetes and TC/HDL cholesterol (table 4; HR 1.68 (95% CI 0.50 to 5.68, p=0.4)). The predominant cause of all-cause mortality (4.9%) in patients with MINOCA was non-CVD death (4.5%). The adjusted HR of non-CVD death was similarly increased in patients with obstructive CAD: HR 2.23 (95% CI 1.76 to 2.83, p<0.001) and with MINOCA: 2.66 (95% CI 1.72 to 4.11, p<0.001) compared with those without CVD.
Compared with those without CVD, age, sex, SBP, smoking, diabetes and TC/HDL cholesterol adjusted HRs for the composite end point were 7.81 (95% CI 6.64 to 9.19, p<0.0001) and 4.64 (95% CI 3.54 to 6.10, p<0.0001) for MI patients with CAD and MINOCA, respectively. There was no difference in HRs between the subgroups of MINOCA patients with normal coronary arteries or mild CAD.
Patients with MINOCA (normal and mild CAD) had an unadjusted HR of 0.54 (95% CI 0.43 to 0.68, p<0.001) for the composite outcome compared with subjects with obstructive CAD (table 5). After adjustment for age, sex, mortality risk factors and guideline-based medical therapy, the HR for those without obstructive CAD was 0.48 (95% CI 0.37 to 0.61, p<0.001) compared with those with obstructive CAD.
Discussion
In this analysis of 8305 patients presenting with type 1 MI, one in nine patients had no obstructive CAD at coronary angiography. Although compared with those with obstructive CAD, patients with MINOCA had about half the risk of recurrent MI or death at 2 years, their risk was not low: 7.0% suffered recurrent MI and 4.9% died, with the greatest proportion of these (4.5%) non-CVD deaths. The rate of all-cause mortality was substantially higher than age and sex-matched subjects without CVD suggesting predisposing factors for MINOCA remain a potent risk factor for non-CVD death.
Prevalence of MINOCA
In 7656 patients enrolled in non–ST-segment elevation acute coronary syndrome (NSTE-ACS) trials, the prevalence of non-obstructive CAD was 9.1%.15 Gehrie et al evaluated 55 514 patients with NSTE-ACS and reported 10.0% had non-obstructive CAD—15.1% in women and 6.8% in men.2 In patients presenting with STEMI, Hochman et al reported a prevalence of non-obstructive CAD of 10.2% in women and 6.8% in men.1 Registries of unselected patients presenting with acute MI report the prevalence of non-obstructive CAD ranges from 7.4%5 to 13.0%.6
We report a similar incidence in a real-world series of patients presenting with acute MI of 10.8%–7.1% in men and 19.3% in women. In addition to a predominance of women, and consistent with previous reports, we found that patients with MINOCA were younger than those with obstructive CAD4 and that those with normal coronary arteries were younger than those with mild CAD.5 6
Prognosis of MINOCA
Previous single-centre studies have reported a low risk of recurrent MI and of death in this group of patients.16 17 Several large series of NSTEMI patients without obstructive CAD have reported an in-hospital and 30-day mortality of 0.6%2 4 and 6-month mortality of 1.3%.4 Larsen et al evaluated a consecutive series of 9796 patients who underwent angiography for acute MI.5 They reported a 1-year mortality of 3.3% in patients without obstructive CAD, approximately 50% of the overall study mortality of 6.6%. A recent meta-analysis has reported a 1-year mortality of 4.7%,18 and Nordenskjold et al report a 14% mortality after a mean 4.5 years of follow-up.8
In the present study of a real-world consecutive multicentre series of patients with acute MI, we report a 1-year mortality of 3.2% and a 2-year mortality of 4.9% in patients with MINOCA. In another series reporting outcomes of unselected patients with ACS, those without obstructive CAD had a similar 2-year mortality of 4.4%.6 We also found a significant 7% risk of recurrent non-fatal MI at 2 years in those with MINOCA, approximately 50% of the risk of recurrent non-fatal MI in patients with obstructive CAD.
Cause of death in MINOCA
This study demonstrated that the predominant cause of death in patients with MINOCA was due to non-CVD causes with a similar rate of non-CVD death to those with obstructive CAD. There were no differences in the baseline rates of cancer and metastatic cancer in patients with obstructive CAD and those with MINOCA. Both groups had higher rates of non-CVD death than the no prior CVD group raising the intriguing possibility that the occurrence of MI itself is a marker for an increased risk of non-CVD death independent of the presence of obstructive CAD. Nordenskjold et al recently reported 57.3% of deaths 30 days after MINOCA were non-CVD.8 The difference in all-cause mortality between patients with obstructive CAD and those with MINOCA is largely accounted for by the increased risk of CVD death in those with obstructive CAD, implying approximately 50% of all-cause mortality relates to the presence of angiographically determined significant obstructive CAD. Interestingly, the increased risk of recurrent MI in the MINOCA group did not translate into an increased risk of CVD death compared with the no prior CVD group.
Comparison to subjects with no prior CVD
This study contributes important new information in that we have established the risk of all-cause death and recurrent non-fatal MI in real-world subjects presenting with acute MI without obstructive CAD in comparison to their age and sex-matched peers without CVD. This risk is substantial with a 2-year incidence of 11.1% compared with 3.0% in an age and sex-matched control group from the same population over a similar time period. This finding suggests that the predisposing characteristics of the patients leading to the index MI remain a potent risk factor for all-cause death or recurrent MI despite the absence of angiographically significant obstructive CAD. It is possible that MI in these subjects is a marker for the presence of other non-CAD-related risk factors such as an increased prevalence of malignancy and associated coagulation disorders and/or inflammation.5
Comparison of patients with and without obstructive CAD
Patients with MINOCA had an unadjusted HR for mortality or recurrent non-fatal MI of 0.54 compared with those with obstructive CAD with similar risks for those with normal coronary arteries or mild CAD. After adjustment for age, sex and baseline MI risk factors, the HRs remained similar in both these groups indicating similar rates of the primary outcome after accounting for differences in presentation of MI risk factors. Previous studies have similarly reported lower risks of mortality7 and the combined end point of mortality and recurrent MI19 in patients with MINOCA compared with those with obstructive CAD.
After further adjustment for ACS evidence-based treatments, the adjusted HR for those with MINOCA was 0.48 compared with those with obstructive CAD. This supports the hypothesis that intrinsic patient factors, predisposing the patient to present with an index MI account for approximately 50% of the risk of future death or non-fatal MI independent of the presence of what has been regarded as clinically important or angiographically significant (>50% diameter stenosis) CAD.
ACS therapy in those without obstructive CAD
Previous studies have indicated that subjects presenting with MINOCA have lower rates of prescription of ACS evidence-based therapies which reduce morbidity and mortality.6 19 20 We observed similar findings in the present study with all major classes of ACS evidence-based agents prescribed less frequently to those without obstructive CAD compared with those with obstructive CAD, and lowest treatment rates in those with normal coronary arteries.
A recent Swedish observational study, however, has reported the most important component of ACS evidence-based treatment reducing recurrent major adverse cardiac events in patients with MINOCA were statins and ACE inhibitors/ARBs with a neutral effect of dual antiplatelet therapy.21 This finding suggests that the treatment-related reduction in all-cause mortality and adverse cardiac events in patients with MINOCA is predominantly influenced by the pleiotropic effects of statins22 and ACE inhibitors/ARBs23 rather than antithrombotic mechanisms.
Limitations of the study
Coronary stenosis was visually estimated by experienced angiographers rather than by quantitative measurement. This was a coronary angiographic study and did not include information from coronary intravascular ultrasound which commonly reveals underlying coronary plaque below plaque disruption in angiographically normal vessels.24 There is a possibility that some patients characterised as having MI had myocarditis which can be detected using cardiac MRI in a limited proportion of those presenting with NSTE-ACS.25 Patients with known ‘no obstructive CAD’ may be more likely to be coded as a non-CVD death leading to an underestimate of the CVD death rate in this group.
Conclusion
MINOCA occurs in one in nine patients and has an adverse outcome in 11% within 2 years. Patients with MINOCA had an HR for recurrent MI of 9.8 and for all-cause mortality of 2.5 (largely non-CVD death) compared with age and sex-matched subjects without CVD. Patients with MINOCA had lower rates of ACS evidence-based treatments potentially resulting in increased event rates.
Key messages
What is already known on this subject?
Myocardial infarction with non-obstructive coronary arteries (MINOCA) occurs in approximately 10% of patients presenting with acute myocardial infarction. Short-term mortality is lower in those with MINOCA than those with myocardial infarction and obstructive coronary artery disease.
Medium-term outcomes and cause of death in this group of patients are lacking.
What might this study add?
Patients with MINOCA have an adjusted HR for all-cause death or recurrent myocardial infarction of 4.64 compared with age-matched controls without cardiovascular disease.
Kaplan-Meier all-cause mortality at 2 years in patient with MINOCA was 4.9% with non-cardiovascular death (4.5%) predominating.
Outcomes were similar in MINOCA patients with non-obstructive coronary artery disease and those with normal coronary arteries.
How might this impact on clinical practice?
These findings suggest further research on interventions to decrease events in patients with MINOCA is required.
Acknowledgments
The authors thank the National Institute for Health Innovation (NIHI) at the University of Auckland for ANZACS-QI programme implementation, coordination and analysis.
References
Footnotes
Contributors MJAW, PRB and AJK: Study concept, design and implementation. ML and KKP: Analyses and model construction. MJAW and AJK: Manuscript drafting. All authors: critical revision of the manuscript for important intellectual content and approval of the final version.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests The VIEW and PREDICT programmes were developed and are maintained by the VIEW team in the Section of Epidemiology and Biostatistics at the University of Auckland. The ANZACS-QI and PREDICT software were developed and supported by Enigma Solutions. AJK and KKP have support from the Health Research Council of New Zealand for the submitted work.
Patient consent Not required.
Ethics approval Northern Region Ethics Committee Y (AKY/03/12/314) and national Multi-Region Ethics Committee (MEC/01/19/EXP).
Provenance and peer review Not commissioned; externally peer reviewed.