Fitting a Square Peg in a Round Hole: A Simple Case of Chest Pain

A 39-year-old female presents to the emergency department with chest pain and shortness of breath. Her electrocardiogram suggests ST-elevation myocardial infarction, but she has no atherosclerotic risk factors. She is gravida 4, para 4, and four weeks postpartum from uncomplicated vaginal delivery. She is diaphoretic and anxious, but otherwise her exam is unremarkable. Cardiac enzymes are markedly elevated and point-of-care echocardiogram shows inferolateral hypokinesis and ejection fraction of 50%. In this clinicopathological case, we explore a classically underappreciated cause of acute coronary syndrome in healthy young women.

She receives regular medical care, and has never been diagnosed with any chronic illness or had surgery. She denies family history of coronary artery disease, sudden unexplained death, blood clotting disorders, or other chronic diagnosis. She has never used tobacco, alcohol, or drugs. She lives at home with her loving husband and four children, and denies psychiatric illness.
In fact, the only notable findings on history are regarding reproductive history. She is gravida 4, para 4 (G4P4), and four weeks postpartum. Pregnancy and vaginal delivery were uncomplicated and she has been breastfeeding without issue. She has help from multiple supportive family members, and she does not feel stressed.
Point-of-care echocardiogram reveals inferolateral hypokinesis and left ventricular ejection fraction (EF) of 50%, but no pericardial effusion, valvular abnormalities, or right heart strain. The chest radiograph (CXR) shows nothing acute. The total creatine kinase is 1344 units per liter and troponin I is 2.42 nanograms per milliliter, but remaining labs are normal (Table 1).

CASE DISCUSSION (Attending Discussion)
I like to start thinking through any case with a little review. This patient was brought in by ambulance with chest pain, shortness of breath, and diaphoresis. Her vitals were Volume IV, no. 1: February 2020 Studies have shown that women are less likely than men to undergo coronary angiography or revascularization and, specifically, they are less likely to receive timely revascularization. 1 Implicit gender bias may have played a role in this case and there are probably some very good reasons why. So let's step back and review the most significant details. This patient was a previously healthy 39-year-old female, whose chest pain presentation was complicated by the fact that she was G4P4 and just four weeks postpartum. Usually when a postpartum woman presents with chest pain or shortness of breath, there are two obvious diagnoses to consider: pulmonary embolism (PE) and postpartum cardiomyopathy.
PE is a leading cause of pregnancy-related death in the developed world. 2 The incidence of venous thromboembolism (VTE) is about 13 in 10,000 pregnancies, with half occurring before delivery and half in the postpartum period. The increased risk of VTE continues for 6-12 weeks postpartum. Whether a patient is recently pregnant or not, this diagnosis is always on my differential for chest pain and/or shortness of breath in a patient with lungs that are clear to auscultation. However, this patient has no clinical signs or symptoms of deep venous thrombosis, and she has normal vital signs. Healthy young patients may have good cardiovascular reserve and may not show any abnormalities until later in the course of their illness, so I would keep PE on the differential for now.
No thought exercise on pregnancy and chest pain would be complete without mentioning amniotic fluid embolism. This is similar to a pulmonary VTE, but it is due to amniotic fluid entering the maternal pulmonary circulation. This presents with a classic triad of hypoxia, hypotension, and coagulopathy. Most of these occur during labor; but about one third happen during the immediate postpartum period. 3 This patient has no hypoxia, hypotension, or coagulopathy and is outside the expected time frame for amniotic fluid embolism, meaning we can likely remove it from the differential.
Next, let's think about peripartum cardiomyopathy. This typically occurs in the postpartum period and is marked by left ventricular dysfunction and heart failure. Research has suggested that peripartum cardiomyopathy is caused by vascular dysfunction, triggered by late-gestational maternal hormones. 4 According to the definition from the European Society of Cardiology, these patients have a reduced EF, usually less than 45%, toward the end of pregnancy or in the months after delivery. Clinically, this patient does not look like she has heart failure. On exam she has clear lungs and no peripheral edema, and on CXR there is no pulmonary edema; so this makes heart failure less likely.
We have another clue in the case that can help us think about both PE and cardiomyopathy: the echocardiogram. There was no right heart strain, and a 50% EF. It seems unlikely to have a PE large enough to cause an elevated troponin without signs of right heart strain, and the 50% EF isn't quite low enough for cardiomyopathy. The combination of her clinical picture and the echocardiogram allows us to remove PE and cardiomyopathy from our differential.
Because neither of these likely diagnoses fit in this patient, we turn again to the ECG, which shows a STEMI. On top of that, the inferolateral hypokinesis on her   is super rare! It complicates only 6.2 of 100,000 pregnancies. 5 Wait, is there a differential diagnosis for STEMI?
The Fourth Universal Definition of Myocardial Infarction Task Force provided an international consensus on the classification of myocardial injury and infarction in 2018. 6 They defined five types of MI.
MI types 4 and 5 are those associated with revascularization procedures, either percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG), respectively. This patient had no recent procedure, so she couldn't have a type 4 or 5 MI.
Type 3 MI occurs when a patient suffers cardiac death, with symptoms or ECG changes that suggest MI, but who die before we can obtain biomarkers. This scenario clearly doesn't fit our alive patient.
Type 2 MI is due to an imbalance between myocardial oxygen supply and demand, unrelated to coronary artery disease. This is what we think of clinically as "demand ischemia." Supply and demand imbalance causes MI in a variety of ways, and some of these are worth exploring (Table 2).
This patient does not have hypertension, respiratory failure or shock; so we can immediately remove those from our differential. Her ECG shows normal sinus rhythm, so we can rule out the tachy-or brady-dysrhythmias. According to her blood work, she is not anemic. Her CXR shows a normal mediastinum, and while that does not entirely rule out aortic dissection, it certainly makes it less likely. That leaves coronary vasospasm (Prinzmetal's angina) and coronary artery dissection (CAD) .
Coronary spasm was first reported by Prinzmetal et al. in the 1950s when they demonstrated reversible myocardial ischemia accompanied by ST-segment elevation on the ECG. Coronary artery spasm is defined as "dynamic, transient reduction in the luminal diameter of the epicardial coronary arteries due to increased vasomotor tone leading to myocardial ischemia." It causes only 1.8% of pregnancy-related MIs, 7 which is pretty rare but not impossible in this patient.
Next, let's turn to CAD. Spontaneous coronary artery dissection (SCAD) is a non-traumatic, non-iatrogenic epicardial CAD. While the cause of pregnancy-associated SCAD is not fully understood, we think the hormonal changes of pregnancy may compromise the arterial wall architecture. Importantly, SCAD isn't limited to the peripartum woman; it's a major cause of MI overall in women ≤50 years of age. 8 The last type of MI (type I) is caused by plaque rupture or erosion leading to thrombus formation in a patient with atherosclerotic coronary artery disease. Our patient has no known atherosclerotic risk factors, making type 1 MI less likely.
We now have three causes of STEMI that remain on this patient's differential: coronary artery vasospasm, CAD, and classic plaque rupture with thrombus formation. The diagnostic test in each case is the same: the patient needs coronary angiography. To diagnose this patient without that study, as asked to do for a clinicopathological case, then becomes about numbers and odds. Pregnancy-associated SCAD is the most common cause of MI among patients who are pregnant or postpartum, accounting for 43% of acute MI in the peripartum population, 9 making this the most likely cause of her MI.

Clinical Diagnosis:
Acute ST-elevation myocardial infarction due to pregnancy-associated spontaneous coronary artery dissection.

CASE OUTCOME (Resident Presentation)
Given this patient's STEMI in the setting of multiparity and recent postpartum status, a provisional diagnosis of SCAD was made by the ED provider. The patient was further managed in the ED with a heparin bolus and drip, and metoprolol. Cardiology was called for immediate consultation. Over the following four hours, the ST elevations improved on repeat ECG and the troponin climbed slightly but thereafter plateaued. The patient was admitted to the cardiac care unit (CCU) and was prepared for cardiac catheterization.
Coronary angiography (Image 2) revealed a long but subtle lesion in a large branch of the right coronary artery. To better characterize this lesion, the adjunctive intravascular ultrasound imaging modality optical coherence tomography (OCT) was used during cardiac catheterization. The OCT of The patient stayed in the CCU for several days on the heparin drip and metoprolol. Her symptoms resolved, and her lab values and ST elevations normalized. Her ECG on discharge from the hospital showed a persistent T-wave inversion in aVL. She continued daily metoprolol and baby aspirin, and declined the option of clopidogrel because she was breastfeeding. She had a negative outpatient workup for fibromuscular dysplasia, and has had no recurrence of symptoms.

RESIDENT DISCUSSION Pathophysiology
SCAD is defined as "separation of the coronary arterial wall by intramural hemorrhage creating a false lumen, with or without an intimal tear." 10 It develops in much the same way as aortic dissection; an intimal tear often results in high-pressure accumulation of hematoma between the vessel layers. Alternatively, the vasa vasorum can bleed between weakened vessel layers to create intramural hematoma. If this hematoma compresses the true lumen, it can cause complete or near-complete coronary vessel occlusion, resulting in STEMI or non-STEMI, respectively. There is an atherosclerotic variant of SCAD that tends to be self-limited by medial scarring and atrophy, but nonatherosclerotic SCAD is the real killer; it results in more extensive and severe dissection and has an acute mortality rate of 28-82%. 11,12 In 2018, Circulation published an American Heart Association (AHA) scientific statement noting that SCAD is more common than previously believed. Typical SCAD patients are healthy young women without conventional atherosclerotic risk factors. The AHA also expressed that SCAD must be evaluated and treated differently from atherosclerotic acute coronary syndrome (ACS). 8 These statements drive home three important points: • Epidemiology: SCAD is more common than previously thought • Demographics: SCAD and type 1 MI patients differ markedly in age and risk factors. • Management: SCAD and type 1 MI clinical management differs markedly.

Epidemiology
SCAD is more common than previously thought. According to recent literature, it accounts for up to 4% of ACS overall, and for up to 35% of ACS occurring in women aged 50 and younger. 8 A retrospective review of an 32,869-patient angiography database found that women constituted 77% of SCAD cases, and all of these women had undergone at least one prior pregnancy. 13 Historical underdiagnosis of SCAD is likely due to the subtle coronary angiography findings of long and diffusely narrowed coronary vessel segments. This triggered the need for OCT, which came into use in the mid-1990s. The true incidence and prevalence of SCAD remain unknown. 14

Demographics
Although patients with SCAD present with the same symptoms as those with atherosclerotic ACS, the demographics differ markedly. In the ED, this means we have a different population of chest pain patients to be worried about: healthy young women.
SCAD is associated with fibromuscular dysplasia and other predisposing arteriopathies. Aside from genetic and structural abnormalities, the mechanism for vessel wall weakening is poorly understood. A widely accepted theory is that chronically elevated circulating hormone levels increase the risk of SCAD in multiparous (≥4 births), pregnant, and/or postpartum women. Interestingly, postpartum status was the sole risk factor in 18% of women, with a mean postpartum period of 38 days at the time of SCAD diagnosis. 15

Management
Although SCAD is managed differently from atherosclerotic ACS, the initial ED workup is the same. The ECG for these patients often shows STEMI (25-50% of cases) or non-STEMI. The left anterior descending coronary artery is most commonly involved, and multivessel disease is not rare. 13 Cardiac enzymes may be elevated, echocardiogram can reveal left ventricular dysfunction, and up to 14% of cases are complicated by ventricular dysrhythmias. 16 Medical management is strongly preferred for SCAD. While in the ED, SCAD patients should still get aspirin, nitroglycerin, and heparin whenever applicable. They should not routinely get thrombolytics; there are prior descriptions of thrombolysisprecipitated extension of the dissection or even rupture leading to cardiac tamponade. 17 SCAD patients should undergo coronary angiography with OCT for definitive diagnosis. Stenting is