Spontaneous coronary artery dissection (SCAD) is defined as a non-traumatic, non-iatrogenic separation of the coronary artery walls, creating a false lumen.1 Traditionally thought of as rare, for many years only anecdotal cases were reported. However, in recent years, probably since the increased use of coronary angiography and intracoronary imaging techniques in acute coronary syndromes (ACS), it has been recognized as an important cause of myocardial infarction, especially in women.2–4 The true prevalence of SCAD in the general population is unfortunately unknown because of under-diagnosis. However, based on modern series, its prevalence in ACS is 1.7-4%.5,6 This study aimed to investigate the characteristics, clinical presentation, therapeutic approach and outcome in patients with ACS presenting with SCAD over a period of seven years.

This was a descriptive and retrospective analysis of patients admitted to a Cardiology Department over seven consecutive years, from January 2010 to December 2016, with a diagnosis of ACS due to SCAD. SCAD was diagnosed in the presence of angiographic characteristics on coronary angiography and was categorized according to the Saw classification7 as follows: Type 1, contrast dye staining of the artery wall with multiple radiolucent lumens (the pathognomonic angiographic appearance of SCAD); Type 2, diffuse stenosis of varying severity and length (typically longer than 20 mm), with frequent subtle abrupt changes in the artery caliber from normal diameter to diffuse smooth narrowing. This diffuse narrowing may be bordered by normal artery segments, proximally and distally (type 2A variant) or it may extend to the apical tip of the artery (type 2B variant). Lastly, Type 3, focal or tubular (typically less than 20 mm) stenosis that mimics atherosclerosis and requires intracoronary imaging to confirm diagnosis.

Clinical, angiographic, therapeutic, and follow-up data up to the end of December 2016 were analyzed. Data were obtained from medical records of hospital stay and subsequent consultations.

Spontaneous coronary artery dissection is uncommon and often misdiagnosed. Diagnosis can be challenging, with a high degree of clinical suspicion necessary.7

The underlying etiology of SCAD appears to be multifactorial and it is believed to be triggered by a precipitating stressor in an individual with underlying predisposing arterial disease.2

A variety of potential predisposing factors have been described, including fibromuscular dysplasia, pregnancy, connective tissue disorders, SID, HT and coronary artery spasm. A predisposing factor was identified in two fifths of our population. In relation to arterial integrity, the Eleid et al. study showed that tortuosity is present in a higher proportion of patients with SCAD and repeat dissections were common within tortuous segments.8 In fact, we found this anatomical variation in a fifth of our cases, but whether arterial tortuosity predisposes to arterial fragility or is only a marker of SCAD remains undetermined. In our population, the only precipitating factors identified were intense physical exercise and Valsalva-type activities. However, other factors, like intense emotional stress, labor and delivery, use of recreational drugs and intense HT have been described.

Hence, in view of the potential association between SCAD and systemic disorders, clinicians should thoroughly investigate to find underlying causes.

The clinical presentation of SCAD varies from asymptomatic to sudden death, but a large proportion of patients present with ACS.9–11 In our population, ACS was the main form of presentation. It can affect one or more coronary arteries, but, as in our study, is most common in the anterior descending artery.12

Coronary angiography is widely available and is the gold standard for imaging in patients with SCAD. Nonetheless, this examination is notoriously suboptimal to visualize intimal tears, as it only provides a two-dimensional image.

There are three main patterns of SCAD2: type 1 angiographic SCAD appearance which, although not the most common pattern, can be demonstrated well on angiography with multiple radiolucent and contrast dye stains in the artery wall; type 2 angiographic SCAD appearance, which is the most common, with long diffuse stenosis of varying severity, commonly in mid to distal segments (this pattern was observed in about two-thirds of our series); and type 3 angiographic pattern, which is more challenging to diagnosis, because it mimics atherosclerosis (not seen in our population).

Complementary techniques such as IVUS and OCT contribute to better identification and classification of SCAD. IVUS has a lower spatial resolution (150-200 μm), but has deeper penetration, allowing full vessel wall visualization. It can determinate the true and false lumen and detect intramural hematoma (IMH) in its full extension.13 Although the poorer penetration of OCT hampers visualization of the full extent of IMHs, it is more sensitive, has much higher resolution (10-20 μm) and can visualize intimal tears, and true and false lumens. It is also superior in visualizing strut apposition, allowing stent optimization if angioplasty is performed.4,14-16 We only used these tools to confirm diagnosis or to guide SCAD treatment in two procedures, as we only recently acquired the equipment.

There are no prospective randomized studies that specifically address the management of SCAD, so its ideal management remains unclear. In terms of medical therapy, there is a general consensus that beta blockers play an important role. They reduce artery wall stress and should therefore be used acutely and long-term following SCAD. The role of antiplatelet therapy for patients with SCAD not treated with stenting is less clear. Based on the benefits of aspirin in acute coronary syndrome and its relatively safe profile, it would appear to be a reasonable option for acute and long-term treatment. As a proportion of SCAD involves intimal tears that can be pro-thrombotic, in theory, double antiplatelet therapy could be beneficial, reducing the false lumen thrombus burden. Although evidence is lacking, clopidogrel is often administered for one to 12 months following SCAD and then discontinued. The role of novel P2Y12 has not been defined. Glycoprotein IIb/IIIa inhibitors are not recommended due to the potential risk of extending the dissection. Anticoagulation for SCAD is subject to debate and it should be discontinued as soon as the diagnosis of SCAD is made due to the potential risk of extending the dissection. Thrombolysis is harmful for SCAD and should be avoided. ACE inhibitors and angiotensin receptor blockers have not been studied in SCAD and tend to be administered in patients with left ventricular dysfunction. Statins should be prescribed if the patient has concomitant dyslipidemia.

The decision to bypass a dissected artery depends on the patient's hemodynamic status and coronary anatomy; site of dissection, the number of vessels involved, distal flow and the possibility of intervention.17,18 In stable patients with normal coronary flow, the conservative approach is preferable.17 This option relies on expert opinion based on observation that dissected arteries heal spontaneously in most cases.

Nonetheless, bypass should be considered in patients with high-risk features, such as ongoing/recurrent ischemia or chest pain, cardiogenic shock, ventricular arrhythmias (ventricular tachycardia or fibrillation) and LMCA dissection.2 Angioplasty is indicated in cases with suitable coronary anatomy (one-vessel disease), due to the high risk of propagation of the dissection related to the procedure17,19; there is still no ideal stent for the treatment of such lesions. We usually follow this rationale to decide which patients should undergo angioplasty. Due to the extent of the dissections, long stents or two stents were frequently used to cover the lesion. A long stent should be selected to provide adequate coverage of both lesion edges (at least 5-10 mm longer proximally and distally), in order to prevent extension of the IMH when compressed by the stent. PCI of dissected coronary arteries can be notoriously challenging and often ends with suboptimal results. First of all, it can be difficult to advance the guidewire through the distal true lumen; second, the IMH can propagate anterogradely or retrogradely with angioplasty; third, the dissections often involve distal coronary arteries, which are too small to implant stents; fourth, even if the dissected arteries are large, the dissections are often extensive, requiring long stents and thus increasing the risks of restenosis; last of all, IMHs resorb and heal over time, and this can result in strut malapposition, increasing the risk of very late stent thrombosis.

CABG may be appropriate in multi-vessel disease or when the LMCA is affected.17 The risk of the procedure is related to not identifying the true lumen. In the only case we had of LMCA dissection, due to the patient's hemodynamic instability and the impossibility to transfer the patient to an operating room within a few minutes, we opted to perform a PCI, the result of which was optimal.

After the acute phase, estimated survival is good, at about 70% to 90%.20 However, there is a risk of recurrence in 50% of patients.21,22

In view of the high rates of major adverse cardiac events in the long term, close follow-up is recommended for patients who suffer a SCAD. In patients treated by PCI, OCT should also be considered to assess for malapposed or uncovered struts before discontinuing dual antiplatelet therapy.

In our population, the prevalence of SCAD was 0.5% among patients who underwent catheterization for suspected ACS. It was more prevalent in middle-aged women. Although SCAD should be considered in the presence of ACS, sudden cardiac arrest could be one form of presentation. It can be associated with SID and patients should be suitably screened. Despite the high prevalence of re-infarction in-hospital or during follow-up, prognosis was good.

The authors have no conflicts of interest to declare.