The optimal management and short- and long-term prognoses of spontaneous coronary artery dissection (SCAD) remain not well defined. The aim of this observational multicenter study was to assess long-term clinical outcomes in patients with SCAD. In-hospital and long-term outcomes were assessed in 134 patients with documented SCAD, as well as the clinical impact and predictors of a conservative rather than a revascularization strategy of treatment. The mean age was 52 ± 11, years and 81% of patients were female. SCAD presented as an acute coronary syndromes in 93% of patients. A conservative strategy was performed in 58% of patients and revascularization in 42%. On multivariate analysis, distal versus proximal or mid location of dissection (odds ratio 9.27) and basal Thrombolysis In Myocardial Infarction (TIMI) flow grade 2 or 3 versus 0 or 1 (odds ratio 0.20) were independent predictors of conservative versus revascularization strategy. A conservative strategy was associated with better in-hospital outcomes compared with revascularization (rates of major adverse cardiac events 3.8% and 16.1%, respectively, p = 0.028); however, no significant differences were observed in the long-term outcomes. In conclusion, in this large observational study of patients with SCAD, angiographic features significantly influenced the treatment strategy, providing an excellent short- and long-term prognosis.
Spontaneous coronary artery dissection (SCAD) represents an important cause of myocardial infarction (MI) especially in young or middle-aged women. Although several small reports have demonstrated favorable outcomes with conservative management, as well as with percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG), there is still a lack of consensus with respect to the best treatment approach in this clinical setting. Consequently treatment is usually chosen on an individual case basis rather than according to a standard approach. Of note, few studies have addressed factors influencing the choice of treatment and the outcomes of such patients. The aim of the present observational study was to evaluate management approaches and immediate and long-term prognosis of patients with SCAD. In addition, we sought to identify factors affecting the choice of treatment strategy in this rare clinical setting.
Methods
From 2001 to December 2013, patients with angiographic diagnosis of SCAD admitted to institutions participating in the study were evaluated. Patients were identified using retrospective and prospective methods. Patients were retrospectively identified using a local institutional database that specifically included the term “spontaneous coronary dissection” among the diagnostic variables. From 2010, patients with diagnoses of SCAD were also included in the study. Patients were analyzed as overall cohort and according to treatment strategy. This study was approved by the institutional review committees and conducted in accordance with the Declaration of Helsinki.
Demographics, conventional cardiovascular risk factors, clinical presentation, coronary distribution of dissections, treatment approaches, and in-hospital outcomes of patients were determined through medical records and angiographic review. Long-term outcomes were obtained through clinical visits, telephone follow-up, or medical record review in the case of readmission. The rate of major adverse cardiac events (MACEs), including all-cause death, nonfatal MI, cardiogenic shock, and target vessel revascularization, was evaluated, as well as heart failure and cerebrovascular accidents. Moreover, pharmacologic treatment at discharge and the frequency of functional tests, repeat angiography, or coronary multislice computed tomography at follow-up were recorded.
The diagnosis of SCAD required specific angiographic features. Coronary angiograms were reviewed by 2 interventional cardiologists at each participating center, and SCAD was diagnosed when ≥1 of the following criteria was detected on multiple angiographic views after administering intracoronary nitroglycerin: (1) a longitudinal radiolucent linear image separating true and false lumen, often with contrast staining (intimal flap), according to the National Heart, Lung, and Blood Institute classification or (2) a long obstructive stenosis with abrupt demarcation from normal segments (intramural hematoma). In patients with clinical suspicion of SCAD and lesions not fulfilling these classic angiographic patterns, adjunctive intracoronary imaging (intravascular ultrasound or optical coherence tomography) was considered for SCAD diagnosis when available. To further characterize SCAD lesions, the Thrombolysis In Myocardial Infarction (TIMI) flow grade classification was also used. The coronary segment involved with SCAD was defined according to the Bypass Angioplasty Revascularization Investigation (BARI) classification. Previous coronary interventions, iatrogenic coronary dissection, and the presence of significant atherosclerotic coronary artery disease as underlying conditions causing SCAD were considered exclusion criteria.
Associated coronary artery disease was defined as ≥1 coronary lesion (different from the SCAD lesion) with a diameter stenosis >50% on visual assessment. PCI success was defined as TIMI grade 3 flow after balloon dilatation or stent implantation, associated with reduction of the stenosis at visual assessment. Taking into account the frequent finding of residual vessel dissection or narrowing distal or proximal to the treated segment in the subset of SCAD, coronary flow was considered a marker of success rather than complete regression of the stenosis.
Unsuccessful PCI was defined as a lack of improvement or worsened TIMI flow compared with baseline before PCI or extension of dissection. MI (ST-segment elevation MI [STEMI] and non–ST-segment elevation MI) was defined as the detection of increases and/or decreases in cardiac biomarkers (cardiac troponin [cTn] or creatine kinase-MB) with ≥1 value above the 99th percentile upper reference limit, associated with ≥1 of the following criteria: symptoms of ischemia, development of pathologic Q waves, new electrocardiographic changes indicative of ischemia (ST-segment elevation or depression or T-wave inversion or new left bundle block). In case of biomarkers already elevated, recurrent spontaneous MI was defined as an increase of cTn level >20% of baseline values, associated with 1 of the aforementioned criteria. PCI- or CABG-related MI was defined as elevation of cTn level (>5 times the 99th percentile upper reference limit in patients with normal baseline values or an increase of cTn level >20% in case of biomarkers already elevated and stable or decreasing), associated with ≥1 of the following criteria: symptoms suggestive of myocardial ischemia, new ischemic electrocardiographic changes, angiographic findings consistent with procedural complications, and new regional wall motion abnormalities. Repeated revascularization was defined as any repeated PCI or CABG. Cardiogenic shock was defined as systolic blood pressure <90 mm Hg nonresponsive to fluid challenge or requiring the administration of inotropic drugs and associated with tachycardia (heart rate > 100 beats/min) and clinical signs of peripheral hypoperfusion. Stroke was defined as the presence of a focal neurologic deficit of central origin lasting >72 hours or lasting >24 hours with imaging evidence of cerebral infarction or intracerebral hemorrhage. Transient ischemic attack was defined as a transient episode of neurological dysfunction caused by focal brain ischemia.
Continuous variables are presented as mean ± SD or medians and discrete variables as percentages. Between-group differences were analyzed using Student’s t tests or chi-square tests as appropriate. A stepwise logistic regression model with an entry and exit level of significance of 0.10 was used to identify variables independently associated with treatment strategies in the whole population and with procedural success in patients who underwent revascularization. The risk estimates for each variable are reported as odds ratios and 95% confidence intervals. Follow-up analysis was performed using time-to-event data with the event rates estimated by Kaplan-Meier methods and compared using the log-rank test. A p value < 0.05 was considered to indicate statistical significance. Statistical analyses were performed with SAS version 9.2 (SAS Institute Inc., Cary, North Carolina).
Results
We identified 134 patients with SCAD. Forty-six patients (34%) were retrospectively identified, and 88 patients (66%) were prospectively identified. Baseline characteristics and clinical presentation are listed in Table 1 . The average age was 52 ± 11 years (range 28 to 86), and 109 (81%) were women. Almost all patients (124 of 134 [93%]) were admitted with acute coronary syndromes (ACS). Forty-nine percent of patients presented with STEMI and 40% with non–ST-segment elevation MI. Single-vessel and multivessel dissection were present in 116 (87%) and 18 (13%) patients, respectively. The left anterior descending coronary artery (LAD) was the most commonly dissected artery. Coronary artery angiographic findings are listed in Table 2 ; the distribution of dissections location according to vessels is presented in Figure 1 . Intravascular ultrasound and optical coherence tomography were used as ancillary diagnostic tools in 9 and 3 patients, respectively.
| Variable | Overall (n=134) | Conservative (n=78) | Revascularization (n=56) | p-value |
|---|---|---|---|---|
| Age, (years) | 52 ±11 | 53 ±10 | 52 ±12 | 0.58 |
| Female | 109 (81%) | 67 (86%) | 42 (75%) | 0.12 |
| Body mass index (kg/m 2 ) | 24.3±3.5 | 24.5±3.7 | 23.7±3.2 | 0.20 |
| Familial history of CAD | 33 (24.6%) | 19 (24.3%) | 14 (25%) | 1.0 |
| Hypertension | 69 (51.4%) | 40 (51.3%) | 29 (51.8%) | 1.0 |
| Hyperlipidemia | 44 (32.8%) | 27 (34.6%) | 17 (30.3%) | 0.71 |
| Smoke habit | 46 (34.3%) | 19 (24.4%) | 27 (48.2%) | 0.006 |
| Diabetes mellitus | 3 (2.2%) | 2 (2.56%) | 1 (1.79%) | 1.0 |
| Ejection Fraction (%) | 52±10 | 53.1±9.4 | 51.7±10.9 | 0.42 |
| Chronic Renal failure | 2 (1.4%) | 1 (1.28%) | 1 (1.79%) | 1.0 |
| Previous myocardial infarction | 7 (5%) | 5 (6%) | 2 (3.5%) | 0.69 |
| Clinical presentation | 0.09 | |||
| Non acute coronary syndrome | 10 (7.5%) | 8 (10.3%) | 2 (3.6%) | |
| Unstable angina pectoris | 4 (3%) | 2 (2.6/) | 2 (3.6%) | |
| NSTEMI | 54 (40.3%) | 37 (47.4%) | 17 (30.4%) | |
| STEMI | 66 (49.2%) | 31 (39.7%) | 35 (62.5%) | |
| Anterior | 34 (51.5%) | 15 (48.4%) | 19 (54.3%) | |
| Inferior | 25 (37.9%) | 11 (35.5%) | 14 (40%) | |
| Lateral | 7 (10.6%) | 5 (16.1%) | 2 (5.7%) | 0.39 |
| Out-of hospital cardiac arrest | 4 (2.8%) | 1 (1.3%) | 3 (5.4%) | 0.31 |
| Cardiogenic shock | 3 (2.2%) | 1 (1.3%) | 2 (3.6%) | 0.57 |
| Associated CAD | 16 (11.9%) | 7 (9%) | 9 (16.1%) | 0.28 |
| Variable | Overall | Conservative | Revascularization | p-value |
|---|---|---|---|---|
| (n=134) | (n=78) | (n=56) | ||
| Lesions | 144 | 82 | 62 | |
| Multivessel dissection | 18 (13%) | 6 (7.7%) | 12 (21.4%) | 0.03 |
| Lesion length (mm) | 36±20 | 37.8±23 | 33.6±16 | 0.23 |
| Coronary artery | 0.29 | |||
| Left main | 4 (2.8%) | 2 (2.4%) | 2 (3.2%) | |
| Left-anterior descending | 52 (36.1%) | 26 (31.7%) | 26 (41.9%) | |
| Left circumflex | 21 (14.6%) | 12 (14.7%) | 9 (14.5%) | |
| Right | 39 (27.1%) | 21 (25.6%) | 18 (29.1%) | |
| Diag/OM/PDA/PL | 28 (19.4%) | 21 (25.6%) | 7 (11.3%) | |
| Location of dissection | <0.0001 | |||
| Proximal | 54 (37.5%) | 22 (26.9%) | 32 (51.7%) | |
| Mid | 34 (23.6%) | 15 (18.0%) | 19 (30.6%) | |
| Distal | 56 (38.9%) | 45 (55.1%) | 11 (17.7%) | |
| Basal TIMI flow | 0.007 | |||
| 0-1 | 58 (40%) | 25 (30.5%) | 33 (53.2%) | |
| 2-3 | 86 (60%) | 57 (69.5%) | 29 (46.8%) |
During index hospitalization, 78 patients (58%) were initially treated conservatively, whereas 56 (42%) underwent coronary revascularization as first-choice therapy. Two patients who were initially treated conservatively underwent subsequent revascularization because of clinical destabilization and angiographic progression of the dissection. Among patients who underwent revascularization, 51 initially received PCI and 5 underwent CABG for multivessel dissection or left main coronary artery involvement ( Figure 2 ). Treatment details of revascularized patients are listed in Table 3 . On multivariate analysis, distal location of dissection and the presence of basal TIMI grade 2 or 3 flow were found to be independent predictors of conservative versus revascularization strategy ( Table 4 ).
| Percutaneous revascularization | 51 (91.1%) |
| Plain old balloon angioplasty | 7/51 (13.7%) |
| Stenting | 44/51 (86.3%) |
| Stent /lesion | 1.9 ±0.9 |
| Stent diameter (mm) | 3.09 ± 0.43 |
| Stent length (mm) | 36.9 ± 18.6 |
| Bare metal stent | 16 (36.3%) |
| Drug-eluting stent | 23 (52.3%) |
| Drug-eluting stent + bare metal stent | 4 (9.1%) |
| Bioreabsorbable vascular scaffold | 1 (2.3%) |
| PCI procedural success | 37 (72.5%) |
| Coronary artery bypass graft | 5 (8.9%) |
| Arterial graft | 5 |
| Saphenous vein graft | 4 |
| Graft/patient | 1.8 |
| Univariate Analysis | Multivariable Analysis | |
|---|---|---|
| OR 95% CI | OR 95% CI | |
| Smoke habit | 0.68 (0.51-0.91) | |
| STEMI versus other clinical presentation | 0.62 (0.42-0.92) | |
| Multi vessel vs single vessel dissection | 0.85 (0.73-0.99) | |
| Distal versus prox-mid location of dissection | 1.83 (1.39-2.41) | 9.27 (3.36-25.5) |
| TIMI flow 0-1 vs 2-3 | 0.64 (0.46-0.89) | 0.20 (0.08-0.53) |
Procedural success of PCI was achieved in 72.5% of patients. Three patients (5.8%) treated with PCI were switched to urgent CABG for procedural failure ( Figure 2 ). In PCI-treated patients, although no clinical or angiographic variables were found to be predictive of procedural failure, body mass index < 23 kg/m 2 showed a trend toward a significant association with procedural outcome (22.2 ± 2.8 kg/m 2 in patients with procedural failure vs 24.2 ± 3.3 kg/m 2 in those with procedural success, p = 0.06).
Three patients (2.2%) died during index hospitalization; of these, 1 died from retrograde aortic dissection after PCI on the proximal LAD, 1 after emergency CABG for STEMI and multivessel SCAD complicated by cardiogenic shock, and 1 as a consequence of cardiogenic shock after out-of-hospital cardiac arrest with coronary anatomy not suitable for revascularization. There was no significant difference in mortality rate between patients treated conservatively and those revascularized (1.3% vs 3.6%, p = 0.99). During index hospitalization, 5 revascularized patients experienced MI, due to stent thrombosis in 40%. Acute or subacute stent thrombosis occurred in 2 of 44 patients (4.5%). Two revascularized patients but no patients treated conservatively developed transient ischemic attacks; 1 patient with anterior STEMI complicated by out-of-hospital cardiac arrest and treated with PCI developed postanoxic cerebral damage with persistent comatose state. The hierarchical in-hospital MACE rate in the overall population was 8.9%, with a significantly higher incidence in patients who underwent revascularization compared with patients medically managed (16.1% and 3.8%, respectively, p = 0.028). In-hospital events are listed in Table 5 .
