The hospital and post-discharge course of patients diagnosed with type-2 myocardial infarction (MI), triggered when myocardial oxygen demand outstrips supply, differs in telling ways from those with the more common atherothrombotic type-1 MI, suggests a new registry analysis that aims to lift a cloud of confusion surrounding their management.
The observational study of more than 250,000 patients with either form of MI, said to be the largest of its kind, points to widespread unfamiliarity with distinctions between the two, and the diagnostic and therapeutic implications of misclassification. It suggests, in particular, that type-2 MI may be grossly underdiagnosed and undertreated.
The minority of patients with type-2 MI were more likely female and to have heart failure (HF), renal disease, valve disease, or atrial fibrillation (AF), and less likely to have a lipid disorder compared with those with type-1 MI. They were one fifth as likely to be referred for coronary angiography and 20 times less likely to undergo revascularization.
Indeed, only about 2% of the type-2 cohort ultimately underwent percutaneous coronary intervention (PCI) or coronary bypass surgery (CABG). Yet the analysis suggests that cardiovascular risk climbs regardless of MI type and that in patients with type-2 MI, coronary revascularization might well cut the risk of death in half over the short term.
There were also disparities in clinical outcomes in the analysis, based on data from the final 3 months of 2017 in the Nationwide Readmissions Database, which reportedly documents almost 60% of hospitalizations in the United States.
For example, those with type 1 or type 2 MI — as characterized in the then-current 3rd Universal Definition of Myocardial Infarction (UDMI-3) and today’s UDMI-4 — were comparably at risk for both 30-day all-cause readmission and HF-readmission. But type-2 patients were less likely to die in the hospital or be readmitted within 30 days for recurrent MI.
Importantly, the study’s 3-month observation period immediately followed the debut of a code specifically for type 2 MI in the 10th International Classification of Diseases-Clinical Modification (ICD-10-CM) system.
Type 2 accounted for about 15% of MIs during that period, the percentage climbing sharply from the first to the third month. That suggests clinicians were still getting used to the code during the early weeks, “undercoding” for type-2 MI at first but less so after some experience, observed Cian P. McCarthy, MB, BCh, BAO, Massachusetts General Hospital, Boston, for the theheart.org | Medscape Cardiology.
“I can imagine that as people become more aware of the coding, using it more often, the proportion of type-2 MI relative to the total MI cases will probably be much higher,” said McCarthy, lead author on the study published online February 15 in the Journal of the American College of Cardiology.
What had been understood about type-2 MI came largely from single-center studies, he said. This “first national study of type-2 MI in the United States” sought to determine whether such findings are hospital-specific or “representative of what people are doing nationally.”
The new analysis largely confirms that patients with type-2 MI are typically burdened with multiple comorbidities, McCarthy said, but also suggests that type 2 often was, and likely still is, incorrectly classified as type 1. So, it was “surprising” that they were rarely referred for angiography, he said. “Only 1 in 50 received revascularization.”
Those diagnosed with type-2 MI were far less likely to receive coronary angiography (10.9% vs 57.3%), PCI (1.7% vs 38.5%), or CABG (0.4% vs 7.8%); P < .001 for all three differences, the report notes.
That, McCarthy said, “clearly shows that clinicians are uncertain about whether revascularization is beneficial” in type-2 MI.
Coding Not in Sync with UDMI
If there is confusion in practice about differentiating type 2 from type 1 MI, it likely has multiple sources, and one may be inconsistencies in how the UDMI and relevant ICD codes are applied in practice.
For example, the coding mandate is always to classify ST-segment elevation MI (STEMI) and non-STEMI (NSTEMI) as type 1, yet UDMI-4 itself states that a type 2 MI may be either STEMI or NSTEMI, notes McCarthy, as well as an editorial accompanying the report.
“It also can be difficult at times to distinguish type-2 MI from the diagnosis of myocardial injury,” both of which are partly defined by elevated cardiac troponin (cTn), adds the editorial, from Kristian Thygesen, MD, DSc, Aarhus University Hospital, Aarhus, Denmark, and Allan S. Jaffe, MD, Mayo Clinic, Rochester, Minnesota.
Crucially, but potentially sometimes overlooked, a diagnosis of infarction requires evidence of ischemia along with the biomarker elevation, whereas myocardial injury is defined by raised cTn without evidence of ischemia. Yet there is no ICD-10-CM code for “nonischemic myocardial injury,” Thygesen and Jaffe observe.
“Instead, the new ICD-10-CM coding includes a proxy called ‘non-MI troponin elevation due to an underlying cause,’ ” they write. “Unfortunately, although some have advocated using this code for myocardial injury, it is not specific for an elevated cTn value and could represent any abnormal laboratory measurements.” The code could be “misleading” and thus worsen the potential for miscoding and “misattribution of MI diagnoses.”
In the current study, 84.6% of the cohort were classified with type-1 MI, 14.8% with type 2, and 0.6% with both types. Of those with type-1 MI, 22.1% had STEMI, 76.4% had NSTEMI with the remainder “unspecified.”
“I think the introduction of ICD codes for type-2 MI is helpful in that we can study type-2 MI more broadly, across institutions, and try and get a better sense of its outcomes and how these patients are treated,” McCarthy said. But the coding system’s deficiencies may often lead to misclassification of patients. Especially, patients with type-2 STEMI may be miscoded as having type-1 STEMI, and those with only myocardial injury may be miscoded as having type-2 MI.
Most Type-2 MI Is a Complication
A profile of patients with type-2 MI may be helpful for making distinctions. The analysis showed that, compared with patients with type-1 MI, they were slightly but significantly older and more likely to have clinical depression, alcohol or other substance abuse disorder, and to be female. They also had more heart failure (27.9% vs 10.9%), kidney disease (35.7% vs 25.7%), atrial fibrillation (31% vs 21%), and anemia (26% vs 18.9%), P < .001 for all differences.
Type-2 patients were less likely to have CV risk factors usually associated with plaque instability and atherothrombosis, including a history of smoking, dyslipidemia, MI, PCI, or CABG (P < .001 for all differences), the group noted.
Of the 37,765 patients with type 2-MI, 91% received the diagnosis as secondary to another condition, including sepsis in 24.5%, hypertension in 16.9%, arrhythmias in 6.1%, respiratory failure in 4.3%, and pneumonia in 2.8% of cases.
In multivariate analyses, patients with type-2 MI, compared with type 1, showed lower risks of in-hospital death and readmission for MI within 30 days. Their 30-day risks of readmission from any cause and from MI were similar.
Table. Adjusted Odds Ratio (OR) for Outcomes by MI Type and Revascularization Status
|End points||Type 2 vs Type 1 MI, OR (95% CI)||Revascularization for Type 2 MI, yes vs no, OR (95% CI)|
|In-hospital death||0.57 (0.54 – 0.60)||0.51 (0.30 – 0.85)|
|30-day HF readmission||0.75 (0.52 – 1.07)||1.14 (0.14 – 9.18)|
|30-day MI readmission||0.46 (0.35 – 0.59)||1.65 (0.37 – 7.34)|
In-hospital mortality was lower for patients with type-2 MI who underwent revascularization compared with those who did not, “but they were a very select, small proportion of the patient group. I would say there are probably unmeasured confounders,” McCarthy said.
“There’s a real kind of equipoise, so I think we desperately need a trial to guide us on whether revascularization is beneficial.”
McCarthy has disclosed no relevant financial relationships. Disclosures for the other authors are in the report. Thygesen has disclosed no relevant financial relationships. Jaffe discloses serving as a consultant for Abbott, Roche, Siemens, Beckman-Coulter, Radiometer, ET Healthcare, Sphingotec, Brava, Quidel, Amgen, Novartis, and Medscape for educational activities.