Spontaneous coronary artery dissection (SCAD) is a heterogeneous condition that normally occurs in young, otherwise healthy women. Moreover, SCAD has received more attention recently as this disease process can cause life-threatening ventricular arrhythmias and cardiac arrest. A recent review article by Kaddoura and colleagues provided a comprehensive overview of this disease process. The authors discussed how the pathogenesis of SCAD is most likely secondary to an intimal tear of the epicardial coronary arteries, leading to intramural hematoma formation and subsequent occlusion of the epicardial lumen [1, 2, 3]. Patients with SCAD typically present with symptoms of acute coronary syndrome [4]. Diagnosis is usually made via coronary angiogram, along with intravascular ultrasound and optical coherence tomography being used as adjunct modalities [5]. Based on international guidelines, treatment generally favors early revascularization over medical therapy in unstable patients [6]. However, the characteristics and outcomes of SCAD with cardiac arrest are not yet well characterized. Likewise, the benefit of an implantable cardioverter defibrillator (ICD) in SCAD patients with cardiac arrest is unknown. This study investigated baseline characteristics in SCAD patients with cardiac arrest using the National Inpatient Sample (NIS) database between 2016 and 2020. We also evaluated the outcomes and their predictors in SCAD–cardiac arrest with or without an ICD.

We analyzed 24,620 patients with spontaneous coronary artery dissection, 1125 of which suffered cardiac arrest and 23,495 who did not suffer cardiac arrest. Of the patients with SCAD who also suffered a cardiac arrest, the mean age was 62.4 years, with 57% of the patients being female and most patients being Caucasian. In comparison, of the patients with SCAD who did not have a cardiac arrest, the mean age was 59.8 years, with 59.6% of the patients being female and most patients being Caucasian. The SCAD with cardiac arrest population had significantly higher comorbidities, including ventricular arrhythmias, congestive heart failure, pulmonary circulatory disorders, liver diseases, solid tumors, coagulopathy, obesity, fluid disorders, chronic kidney disease (CKD), psychosis, and neurologic disorders, compared to the SCAD without cardiac arrest population (Table 1).

In addition, we constructed a model to assess which comorbidities were associated with a greater risk of cardiac arrest (Table 2), which according to our model were ventricular arrhythmias (OR 12.11 with 95% CI (8.78–16.70) p-value 0.0001), fluid and electrolyte disorders (OR 1.42 with 95% CI (1.01–2.01) p-value 0.045), and other neurological disorders (OR 2.55 with 95% CI (1.69–3.85) p-value 0.0001). Fig. 1 shows the temporal trend of sudden cardiac arrest in SCAD patients from 2016 to 2020. The rate of sudden cardiac arrest peaked in 2017 at 5.56%, then trended downwards over the next two years, with a low of 3.56% in 2019 before rising to 4.98% in 2020. Fig. 2 shows the temporal trend of in-hospital mortality among patients with SCAD who experienced sudden cardiac arrest from 2016 to 2019. There was high variability, as the in-hospital mortality in 2016 was 33.33% in 2016, 21.57% in 2017, 31.25% in 2018, 21.62% in 2019, and 29.79% in 2020. The trends of sudden cardiac arrest incidence in SCAD patients and in-hospital mortality among sudden cardiac arrest patients were analyzed; however, neither showed statistically significant changes over time (p-trend = 0.317 and 0.838, respectively).

Fig. 1.

Temporal trends of cardiac arrest among patients with spontaneous coronary artery dissection (SCAD).

Fig. 2.

Temporal trend of in-hospital mortality among spontaneous coronary artery dissection (SCAD) patients with cardiac arrest.

Among 24,620 SCAD patients, 0.6% underwent ICD placement, and 99.4% did not. The mean age was 61.8 years among SCAD patients who received an ICD, while the mean age was 59.9 years among SCAD patients who did not receive an ICD. Compared with SCAD patients without an ICD, those with an ICD were associated with congestive heart failure and cardiac arrhythmias (p 0.05). In addition, compared with SCAD patients without an ICD, those with an ICD had a significantly higher prevalence of pulmonary circulatory disorders, neurologic disorders, liver disease, fluid and electrolyte disorders, and alcohol abuse (p 0.05) (Table 3).

Similarly, we conducted a model to predict which comorbidities were associated with a greater risk of ICD placement (Table 4), which according to our model were congestive heart failure (OR 3.99 with 95% CI (1.46–10.87) p-value 0.01), ventricular arrhythmias (OR 33.94 with 95% CI (9.20–125.18) p-value 0.001), pulmonary circulatory disorders (OR 4.40 with 95% CI (1.34–14.44) p-value 0.02), and alcohol abuse (OR 4.41 with 95% CI (1.22–15.94) p-value 0.02).

We further evaluated the predictors of mortality among SCAD patients who did not receive an ICD. Based on our model, increasing age (OR 1.06 with 95% CI (1.04–1.07) p-value 0.001), fluid and electrolyte disorders (OR 2.34 with 95% CI (1.73–3.16) p-value 0.001), diabetes (OR 1.64 with 95% CI (1.21–2.23) p-value 0.001), neurological disorders (OR 2.42 with 95% CI (1.63–3.59) p-value 0.001), peripheral vascular disease (OR 1.48 with 95% CI (1.04–2.10) p-value 0.03), pulmonary circulatory disorders (OR 2.01 with 95% CI (1.20–3.39) p-value 0.01), ventricular arrhythmias (OR 3.25 with 95% CI (2.36–4.48) p-value 0.001), and congestive heart failure (OR 1.57 with 95% CI (1.18–2.10) p-value 0.001) were associated with increased risk of mortality (Table 5). Notably, given the small sample size (n = 140), we did not evaluate the predictors of mortality among SCAD patients who received an ICD.

In our study, we found that SCAD patients who suffered a cardiac arrest had higher comorbidities, such as congestive heart failure, pulmonary diseases, liver diseases, cancers, coagulopathy, and CKD, compared to SCAD patients who did not suffer a cardiac arrest. SCAD patients who underwent cardiac arrest were associated with AMI and ventricular arrhythmias (ventricular tachycardia (VT), ventricular fibrillation (VF)). We also found that the trend of cardiac arrest in SCAD patients has continually been trending downward, whereas the in-hospital mortality of these patients has remained quite variable. The decline in trends of cardiac arrest in SCAD patients is perhaps due to increased recognition of SCAD and improvement in medical therapy for SCAD.

One prior study also tried to answer the question of which characteristics are inherently present in SCAD patients with cardiac arrest. Phan and colleagues performed a retrospective cohort analysis of 208 SCAD patients from 2006 to 2016. Of those who suffered cardiac arrest, the investigators concluded that this subset was more likely to have coronary lesions involving the left main or left anterior descending artery (LAD) territory. This could explain the results of our findings, as patients who have ischemic disease are more likely to develop cardiac arrhythmia and heart failure [9, 10, 11]. In other words, in SCAD patients who presented with cardiac arrest, it seems they carried an increased risk of developing cardiac arrest secondary to underlying coronary artery disease. Interestingly, they also found that secondary prevention with an ICD did not significantly benefit this population. This brings into question whether ICD therapy is beneficial in SCAD patients with cardiac arrest. Sharma and colleagues also evaluated 102 patients presenting to Massachusetts General Hospital with SCAD. Comparing those presenting with cardiac arrest to those without cardiac arrest, they found that the cardiac arrest subset was more likely to smoke and present with ST elevation myocardial infarction (STEMI). Although these data are from a very small sample size [12], the findings echo our current findings. Both of these studies seem to stress that SCAD patients presenting with cardiac arrest occurred in the presence of underlying heart disease and comorbidities such as smoking.

Although SCAD patients are likely to be healthy and young, SCAD patients with cardiac arrest tend to have risk factors that would predispose them to cardiac arrest. These factors include coronary artery disease, congestive heart failure, and atrial or ventricular arrhythmia. In our prior study from a single healthcare system, we found that ventricular arrhythmia and atrial fibrillation were independently associated with in-hospital mortality in patients with SCAD [13]. The trend in SCAD patients experiencing cardiac arrest decreased from 2017 to 2019, which could be explained by better detection of SCAD in general. The in-hospital mortality variability for SCAD with sudden cardiac arrest (SCA) is probably a combination of factors, including the expertise of each hospital and physicians in treating SCAD patients.

Our study has certain limitations. Notable limitations include the small number of SCAD patients who received ICD. This potentially explained the relatively wide confidence intervals of certain ICD placement predictors, suggesting that these predictors may be imprecise; thus, the results are inconclusive. Other limitations include the lack of essential laboratory and medication data.

SCAD patients with certain comorbidities (e.g., pulmonary diseases, liver diseases, cancers, coagulopathy, and CKD) who presented with AMI or congestive heart failure should be monitored closely for ventricular arrhythmias as they have a higher chance of progressing to cardiac arrest. More research needs to be conducted to effectively determine how SCAD patients with cardiac arrest should be treated and managed going forward. ICD therapy can be considered for these patients, but data on the success of this treatment option are limited, and more research needs to be performed to determine whether the benefits of this outweigh the risks.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

CK designed the research study. CK, YKQ, ZW, SPA, MA, SS, HJ performed the research. SPA analyzed the data. ZW, MA performed validation of analyses. CK, YKQ, ZW, SPA, MA, SS, HJ wrote, reviewed and edited the manuscript. CK, SS, MA, HJ supervision. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Given that the data contained within NIS are deidentified, our study did not require approval from the Institutional Review Board. The participants provided written informed consents before enrollment.

Not applicable.

This research received no external funding.

The authors declare no conflict of interest. Hani Jneid is serving as one of the Editorial Board members of this journal. We declare that Hani Jneid had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to Giuseppe Boriani.