The Heart Attack She Did Not Know She Had

Some heart attacks announce themselves with crushing chest pain, an ambulance, and a name that lasts a lifetime. Others happen during what felt like a bad week at work and are found years later on a routine ECG ordered before an unrelated procedure. Silent or unrecognized myocardial infarction is not rare, and it is not equally distributed: women and people with diabetes bear a disproportionate share of these events, which places them at the center of the argument this pillar makes about under-recognized coronary disease in women.

The Mechanism

A myocardial infarction is classified as silent or unrecognized when ischemia causes cardiac muscle damage without producing symptoms that the patient or a clinician identifies as a cardiac event at the time it occurs. That absence of recognition can happen through at least two distinct mechanisms, and both likely operate in women who experience these events.

The first mechanism is genuinely asymptomatic ischemia: coronary blood flow is sufficiently interrupted to cause myocardial cell death, but the event generates no pain signal at all. The second, and arguably more consequential from a health-system standpoint, is atypical symptom presentation: the event does produce symptoms, but those symptoms are fatigue, jaw or neck discomfort, shortness of breath, nausea, or a vague sense of being unwell rather than the substernal crushing pressure that defines textbook presentations. When symptoms do not fit the pattern clinicians are trained to recognize, the cardiac cause goes uninvestigated.

In the cardiac pain pathway, sensory afferents from the myocardium travel through the sympathetic nervous system and enter the spinal cord at levels T1 through T4, eventually producing the referred chest and arm pain characteristic of ischemia. This system is not invariably intact. Diabetic autonomic neuropathy impairs sensory transmission throughout the peripheral and autonomic nervous system, including these cardiac afferents. The result is that ischemia sufficient to cause cell death may not generate a pain signal detectable as cardiac in origin. Women, independent of diabetes, also show higher rates of atypical symptom presentation during acute MI compared with men, though the neurological basis for this sex difference is less well defined than the neuropathic mechanism.

Standard cardiac testing compounds the detection failure. Resting electrocardiograms are performed periodically, not continuously, so they capture only the electrical footprint left behind after the event rather than the event itself. That footprint, typically Q-wave changes or loss of R-wave progression in the anterior leads, is present only in transmural infarcts affecting enough myocardium to alter surface-electrode signals. Smaller or non-transmural infarcts, which are disproportionately common in women, may leave no Q-wave signature. Stress testing identifies inducible ischemia but does not detect prior events. Coronary calcium scoring quantifies calcified plaque burden but does not show myocardial scar. The only test that directly identifies prior myocardial injury without requiring the event to be witnessed in real time is cardiac magnetic resonance imaging with late gadolinium enhancement, which marks scar tissue in the myocardium regardless of when it formed or what symptoms accompanied it.

What the Evidence Shows

The Atherosclerosis Risk in Communities (ARIC) study, which followed more than 9,000 participants prospectively over multiple years, found that approximately 45 percent of incident myocardial infarctions were unrecognized. That proportion was higher in women than in men across the cohort. This study established the basic epidemiology of unrecognized MI in a community population and confirmed that it is not a rare edge case but a substantial fraction of all cardiac events.

More recent CMR-based data have substantially revised estimates upward. The ICELAND-MI study used cardiac MRI with late gadolinium enhancement in community-dwelling adults without a known history of cardiac events and found silent MI prevalence significantly higher than ECG-based estimates from the same population. Because CMR detects myocardial scar directly rather than relying on surface electrical changes, it captures non-transmural infarcts and events too small to alter the ECG, categories that ECG surveillance misses entirely. The gap between CMR-detected and ECG-detected silent MI prevalence is a measure of how large the detection failure actually is.

The prognostic implications of unrecognized MI are not trivial. Data from ARIC and subsequent analyses consistently show that individuals with silent MI have elevated all-cause and cardiovascular mortality compared with those without prior infarction, and that this excess risk is not substantially smaller than the risk associated with recognized events. A 2019 analysis in Circulation by Yeh and colleagues followed ARIC participants through a 20-year period and found that silent MI was associated with significantly worse long-term survival. The silence is in the presentation only; the myocardial scar, the impaired left ventricular function in the territory of the event, and the elevated vulnerability to subsequent coronary events are all present.

For women with diabetes, the evidence layers compound. A 2020 analysis in the Journal of the American College of Cardiology examining sex-stratified silent MI prevalence found that women with diabetes had among the highest rates of unrecognized infarction of any subgroup studied, reflecting the convergence of atypical symptom patterns in women with the neuropathy-driven pain blunting of diabetes. These women are at the intersection of the two risk factors most strongly associated with the silent subtype.

What the retrospective literature also shows is that these events were rarely entirely asymptomatic in the way that word implies. In interviews with patients found to have unrecognized MI on subsequent testing, many describe a period of unusual fatigue, exertional intolerance, or atypical discomfort in the timeframe that CMR or ECG evidence localizes the event. The attribution failure is as important as any neurological mechanism: atypical symptoms in women are documented in medical records as gastrointestinal, musculoskeletal, or anxiety-related before cardiac causes are considered, and if the woman does not return with escalating symptoms, that initial misattribution becomes permanent.

Sex-Specific Symptom Patterns and the Attribution Gap

The 2009 GRACE (Global Registry of Acute Coronary Events) analysis, which examined presentations across more than 25,000 patients across 14 countries, found that women presenting with MI were significantly more likely than men to present without chest pain as the dominant symptom, particularly in older age groups. In women over 65, the proportion presenting without chest pain exceeded 40 percent. These were not silent events in the purely asymptomatic sense; they were events producing atypical symptom profiles that did not trigger appropriate clinical response.

The mechanism behind the sex difference in symptom presentation is incompletely understood, but several contributing factors are supported by evidence. Women with coronary artery disease are more likely to have microvascular disease alongside or instead of large-vessel obstructive disease. Microvascular ischemia activates cardiac sensory afferents differently than transmural large-vessel occlusion, potentially generating sensations perceived as fatigue, breathlessness, or diffuse discomfort rather than the focal crushing pressure of proximal LAD or RCA occlusion. Hormonal differences in pain processing, particularly estrogen’s modulation of pain sensitivity through central and peripheral pathways, may also contribute, though this remains an area of active investigation.

What is less uncertain is the consequence. When a woman presents with fatigue and shortness of breath and no chest pain, the cardiac workup is less likely to be initiated promptly. The same symptom cluster in a 55-year-old woman is statistically more likely to receive a workup for anemia, thyroid disease, or depression before cardiac evaluation is ordered than the same cluster would in a 55-year-old man. This is not a theoretical bias detected only in controlled studies; it has been documented in real-world emergency department triage data from multiple countries.

The Framingham Heart Study investigators examined sex differences in MI presentation in a community cohort over several decades and found that women were more likely than men to have unrecognized MI as a proportion of all their MIs, and that this excess held across age groups. The sex ratio for unrecognized-to-recognized MI was consistently above 1.0 for women relative to men. Epidemiologically, this translates into a larger pool of women walking around with prior myocardial injury that was not identified when it occurred and therefore was not treated with the secondary prevention strategies, aspirin, statin, beta-blocker after MI, that recognized events trigger.

The secondary prevention gap is the other dimension of the prognosis story. A recognized MI triggers a protocol: revascularization if indicated, antiplatelet therapy, statin, beta-blocker, angiotensin-converting enzyme inhibitor, cardiac rehabilitation referral, and follow-up imaging. An unrecognized MI triggers none of that, because it was never recognized. The myocardial scar is there, left ventricular function may be subtly impaired, and the residual coronary disease that caused the event is still present and untreated. This is why unrecognized MI carries similar long-term mortality risk to recognized events despite its apparently less dramatic presentation: it is not treated in any of the ways that reduce post-MI mortality.

High-Sensitivity Troponin: Detecting Myocardial Injury That Standard Assays Miss

The transition from conventional troponin assays to high-sensitivity cardiac troponin testing represents one of the most significant diagnostic shifts in cardiology over the past decade, and its implications for unrecognized myocardial injury in women are still being integrated into clinical practice. High-sensitivity assays can detect cardiac troponin concentrations at levels approximately tenfold lower than conventional assays, with measurable values in more than 50% of a healthy reference population compared to fewer than 10% with older testing. That increased sensitivity was expected. What emerged as these assays entered widespread clinical use was more clinically important: the 99th percentile reference thresholds — the values above which results are considered elevated — differ substantially between men and women.

Shah and colleagues, reporting in JAMA in 2015, demonstrated that applying sex-specific diagnostic cutoffs for high-sensitivity cardiac troponin I significantly increased the sensitivity for detecting acute MI in women presenting with possible acute coronary syndrome, without reducing specificity. The higher cutoff derived predominantly from male reference populations was underperforming in female patients — classifying women with genuine myocardial injury as within normal range because their troponin, though elevated relative to their own biology, did not cross a threshold calibrated to male physiology.

This sex-calibration gap has direct relevance to the silent MI question in women. If women with subclinical or atypical acute events presented to emergency departments during the acute phase and had troponins measured with conventional assays using male-derived thresholds, some proportion of genuine myocardial injury would have been classified as non-diagnostic and the patient discharged without a cardiac diagnosis. The retrospective ECG and CMR findings of unrecognized MI documented in ARIC, Framingham, and ICELAND-MI data likely undercount the true event frequency, because some proportion passed through acute care not because the event was electrically invisible, but because the diagnostic threshold did not capture it.

The clinical message is not that every woman needs prospective high-sensitivity troponin monitoring. It is that a historical record of a “normal troponin” during a past presentation does not carry the exculpatory weight it appears to. Particularly for presentations that occurred before high-sensitivity assays were in widespread use, or in facilities where sex-specific thresholds are not applied, that prior negative result leaves meaningful diagnostic uncertainty. When a woman presents today with an unexplained CMR finding or a new Q wave and a prior emergency department record showing “troponin normal,” the era of that testing and the threshold applied belong in the clinical interpretation.

What to Do This Week

1. If you have diabetes, ask your clinician to include cardiovascular risk management as a standing agenda item at every visit, not only when specific cardiac symptoms arise. Silent MI risk in diabetes is not an emergency consideration; it is a chronic surveillance issue that belongs in longitudinal care.

2. If you experienced a period of unexplained fatigue, unexpected exertional intolerance, jaw or neck discomfort, or persistent nausea in the past few years without a clear explanation, mention it explicitly when discussing cardiac history. Clinicians do not always know to ask, and the history is clinically meaningful even if no one documented it as cardiac at the time.

3. If a routine ECG shows Q-wave changes and no prior cardiac event is on record, ask directly whether further evaluation is appropriate before the finding is dismissed as nonspecific. A wall motion abnormality found incidentally on echocardiography in a woman with no documented MI deserves the same question.

4. If your cardiac imaging history is limited to ECG and stress testing, ask your cardiologist whether cardiac MRI is appropriate given your risk profile. CMR with late gadolinium enhancement detects myocardial scar that ECG misses and is the only test that can identify a prior non-transmural infarct retrospectively.

5. If you are a woman over 50 with two or more cardiovascular risk factors and have never had formal cardiac risk assessment beyond a basic lipid panel, initiate that conversation. The ARIC data establish that the base rate of unrecognized MI in this population is not negligible, and the downstream risk implications are real.

The premise embedded in the phrase “silent heart attack” deserves examination. These events are not silent in the sense of leaving no trace: they scar the myocardium, impair regional function, and elevate the risk of future events in ways that are identical to recognized infarcts. What is silent is the clinical recognition at the time the event occurs, and that silence is partly biological and partly a product of how the medical system responds to atypical symptoms in women. Both are addressable, but only if the possibility is on the table to begin with.