Causes of Death after Type 2 Myocardial Infarction and Myocardial Injury

While Type 2 MI (T2MI) and myocardial injury are encountered frequently in clinical practice, there are no evidence-based therapies to reduce subsequent mortality. A more detailed understanding of cause-specific mortality may help identify therapeutic targets. Accordingly, we performed a community cohort study to evaluate cause-specific mortality after T2MI and myocardial injury, using type 1 MI (T1MI) as the comparator.

We prospectively collected all cases with a serum cardiac troponin-T (cTnT) concentration above the 99th percentile upper reference limit (≥0.01ng/mL) from any clinic or hospital location within Olmsted County, Minnesota, between 01/01/2003 and 12/31/2012. We previously reported trends in incidence and outcomes after T2MI and T1MI in this population (1). We now: 1. Report detailed long-term causes of death and 2. Provide outcomes in the group classified as myocardial Injury. All patients provided informed consent. The study was approved by the institutional review board. Patients with prior MI were excluded. Cases were retrospectively classified by two cardiologists following review of clinical records and investigations including angiography. MI was defined as a rise/fall in cTnT associated with at least one of: ischemic symptoms, new ECG changes or new ischemia on imaging (2). T1MI was defined as MI due to atherothrombosis and T2MI due to supply/demand mismatch. Cases not meeting criteria for MI were classified as myocardial injury. Disparity was resolved through consensus. Cause of death was adjudicated at time of death and divided into cardiovascular and non-cardiovascular causes (3) . Cardiovascular death was further classified into acute MI, sudden cardiac death (SCD), heart failure (HF), stroke and vascular. Adjustment for competing risk was performed using the Fine and Gray model.

Of the 4665 patients with a cTnT≥0.01ng/mL, 1022 were classified as T2MI, 1345 as T1MI, 2282 as myocardial injury and 16 as other MI (Table 1). Diagnostic criteria for T2MI were ischemic symptoms (82%), ECG changes (22%) and imaging findings (1%). While cardiovascular death at 5 years was ~20% in all groups, rates of fatal acute MI were lower after T2MI and myocardial injury (4%,3%) compared to T1MI (10%) and even lower after adjustment for age/sex (Table 1). Cardiovascular death after T2MI and myocardial injury was most commonly due to HF and stroke/vascular disease.

Table 1:

Baseline demographics and outcomes.

	Type I MI (N=1345)	Type II MI (N=1022)	Myocardial Injury (N=2282)	p-value
Maximum troponin, median (Q1, Q3), ng/mL	0.33 (0.09, 1.50)	0.11 (0.05, 0.27)	0.05 (0.03, 0.10)	<0.001
Delta troponin, ng/mL	0.2 (0.04, 1.05)	0.06 (0.03, 0.17)	0.02 (0.01, 0.05)	<0.001
Age, years	68.5±14.8	73.7±15.8	73.9±16.2	<0.001
Sex (% male)	847 (63%)	470 (46%)	1118 (49%)	<0.001
Hypertension	955 (71%)	695 (68%)	1506 (66%)	0.003
Diabetes	309 (23%)	143 (14%)	274 (12%)	<0.001
Hypercholesterolemia	780 (58%)	347 (34%)	686 (30%)	<0.001
Prior CABG	125 (9%)	18 (2%)	52 (2%)	<0.001
Prior stroke	40 (3%)	21 (2%)	22 (1%)	<0.001
Prior heart failure	26 (2%)	86 (8%)	284 (12%)	<0.001
COPD	159 (12%)	163 (16%)	409 (18%)	<0.001
Sepsis during admission	29 (2%)	133 (13%)	190 (8%)	<0.001
Angiogram	1200 (89%)	401 (39%)	-	<0.001
5 year unadjusted mortality *
All cause	411 (31%)	605 (59%)	1486 (65%)	<0.001
Cardiovascular	239 (18%)	216 (21%)	525 (23%)	<0.001
MI/SCD	132 (10%)	50 (5%)	71 (3%)	<0.001
MI	119 (9%)	42 (4%)	62 (3%)	<0.001
SCD	13 (1%)	8 (0.8%)	9 (0.4%)	0.10
Heart failure	78 (6%)	102 (10%)	254 (11%)	<0.001
Stroke/Vascular	27 (2%)	54 (5%)	176 (8%)	<0.001
Hazard ratios for 5 year unadjusted mortality (T1MI as reference population) †
All cause		2.5 (2.2-2.8)	2.9 (2.6-3.2)
Cardiovascular		1.2 (1.0-1.4)	1.3 (1.1-1.5)
AMI/SCD		0.48 (0.35-0.67)	0.30 (0.23-0.41)
HF		1.8 (1.3-2.4)	2.0 (1.6-2.6)
Stroke/Vascular		2.7 (1.7-4.3)	4.0 (2.7-6.0)
5 year mortality adjusted for age/sex *
All cause	33%	55%	60%	<0.001
Cardiovascular	18%	16%	18%	<0.001
MI/SCD	10.4%	4.4%	2.6%	<0.001
HF	5.6%	7.4%	7.7%	<0.001
Stroke/Vascular	1.9%	4.3%	7.3%	<0.001

^*p-values from Cox-models using competing risk framework.

^†HR for adjusted mortality after T2MI and myocardial injury, respectively were: fatal MI/SCD: 0.4, 0.24, HF: 1.3, 1.5, stroke/vascular: 2.2, 3.2.

This community cohort study demonstrates that while rates of cardiovascular mortality after T2MI and myocardial injury were similar to T1MI, subtypes of cardiovascular death were different. In contrast to T1MI, cardiovascular deaths after both T2MI and myocardial injury were seldom from fatal MI and predominantly from HF and stroke/vascular disease. These findings contrast to a prior report of similar rates of fatal MI (4) in a tertiary setting. Our study was in a community-based geographic population rather than selected hospital setting, a strength, which likely explains the difference.

T2MI and myocardial injury are heterogenous populations and optimum management may differ according to underlying cause. There remains a need to establish targets for cardiovascular mortality risk reduction after both T2MI and myocardial injury. The lower risk of fatal MI in the current study together with the low risk of non-fatal MI after T2MI that we previously demonstrated (1) suggest that routine secondary preventative therapy may be less impactful than in T1MI. The cause-specific mortality observations suggest that interventions directed toward prevention of HF, stroke and vascular deaths after T2MI may be higher yield, with reported event rates useful for trial design. Further work are needed, including studies using a high-sensitivity troponin assay.

Abbreviations

CABG

coronary artery bypass graft

COPD

chronic obstructive lung disease

cTnT

cardiac troponin T

MI

myocardial infarction

SCD

sudden cardiac death

ECG

electrocardiogram

HF

heart failure

HR

hazard ratio

T1MI

type 1 myocardial infarction

T2MI

type 2 myocardial infarction