Statistics from Altmetric.com
‘First, do no harm’ are oft misquoted words from Hippocrates’ 2500-year-old work Epidemics. A more complete and accurate translation of his exhortation reads:
The physician must be able to tell the antecedents, know the present, and foretell the future — must mediate these things, and have two special objects in view with regard to disease, namely, to do good or to do no harm.1
This is laudable sentiment but is it realistic for a diagnosis surrounding which much uncertainty remains?
Looking back only 20 years, a case report in this publication described spontaneous coronary artery dissection (SCAD) as ‘an extremely rare cause of unstable angina and acute myocardial infarction’.2 That is no longer the case. There has been a surge in the recognition of the condition. One recent study found 33 (2.4%) cases of SCAD among 1375 patients who underwent invasive coronary angiography for acute coronary syndrome in 2019.3 This represents a 10-fold increase in relative diagnostic frequency since an earlier registry (1999–2007) identified only 22 (0.2%) cases from 11 175 patients.4 Presuming that the true incidence is unchanged, this growth reflects a remarkable ‘paradigm shift in clinical appreciation’.5
The challenge we face
In contrast to many cardiovascular disorders, SCAD is predominantly a disease affecting middle-aged women with comparatively few traditional atherosclerotic risk factors. The epidemiology, pathophysiology and approach to diagnosis of SCAD have been informed by retrospective observational cohort studies but optimal management remains unclear. Invasive coronary angiography represents the primary imaging modality for identifying the condition, meaning definitive treatment decisions are commonly made in the catheter laboratory and herein lies the challenge. In contrast with atherosclerotic disease, percutaneous coronary intervention (PCI) is associated with less predictable outcomes in the setting of SCAD. Complications—including haematoma propagation requiring additional stents, acute vessel closure and need for emergent coronary artery bypass graft surgery—are more common, while the majority of these lesions may resolve entirely if allowed to heal without stenting. Conversely, a small proportion of cases represent genuine ‘high-risk’ presentations and for these, early intervention may be lifesaving. The difficulty lies in balancing those factors for and against intervention in the time-constrained and clinically pressured setting of the procedure room.
It is in this context that the paper by Kotecha et al 6 presents useful insight and the authors are to be congratulated on an international collaboration with the largest such analysis to date. Two hundred and fifteen patients with angiographically confirmed SCAD who underwent PCI were identified from three national registries. To allow a crude comparison of clinical characteristics, a similar number of conservatively managed patients with SCAD were also included in the description. Overall, the population is as we would expect, predominantly female and young middle-aged. However, although of similar demographics and comorbidity status, it is clear that these subgroups represent two very distinct patterns of presentation. Compared with the conservatively managed patients, the PCI group presented more commonly with ST-segment elevation myocardial infarction, had a twofold incidence of cardiac arrest, and had much greater involvement of the left main stem or proximal left anterior descending artery. Greater than one in four of those proceeding to PCI had type 4 lesions (figure 1), namely those associated with complete vessel occlusion—more than four times that seen among those conservatively managed. Fifteen patients had failed a trial of conservative management. Although haemodynamic data are notably lacking, this sounds like a true high-risk cohort. So how did the coronary intervention proceed? This is where the present manuscript provides its most helpful details.
First, the procedural caveats
Any intervention was not without risk and complications related to PCI were identified in 39%. Even diagnostic imaging alone resulted in a small but important risk of iatrogenic dissection. Other important hazards included one-third of patients who required either unplanned left main stem stenting, ≥4 stents or ≥50 mm of stent length. Further, there were three instances of loss of flow in a proximal coronary segment, and two individuals required bypass graft surgery. Fortunately, the majority of procedural complications were more minor and related to haematoma extension without loss of coronary flow. The authors have attempted to identify useful predictors of complications, which included type 2 and 3 lesions (those without contrast staining of the wall—implying lack of communication between the vessel lumen and the dissection plane), particularly when extending over multiple coronary segments or involving large proximal vessels. Some other postulated predictors, namely number and length of stents, are confounded by indication as these seem likely a consequence rather than a cause of complications.
Second, the commendations
As might be expected, the majority of interventions resulted in coronary stenting and necessitated a greater than typical number (mean 2.3) and length (median 46 mm) of stents. Nevertheless, a substantial minority (21%) received balloon-only angioplasty, which tended to be associated with fewer subsequent complications, and the procedural operators demonstrated admirable restraint, leaving two-thirds of cases with unstented regions of dissection. This is a notable contrast to typical coronary intervention, reflecting operator desire to allow a more natural healing process to occur once coronary patency had been restored. It is in keeping with the growing recognition that angiographic perfection is not required in the SCAD setting given the likelihood of vascular healing over time.7 In the current study, 84% of those with reduced blood flow at baseline achieved an improvement, while flow worsened in only 7%. Further, outcomes were broadly similar for those undergoing PCI than the lower risk cohort who were managed conservatively. This is encouraging indeed.
Finally, the limitations
This dataset arose from three national registries that by merit of requiring patient consent were unable to include early non-survivors following PCI for SCAD. This introduces potential survivorship bias with underestimation of risks associated with coronary intervention. Given reported in-hospital mortality rates of between 0.1% and 5.0%, it is difficult to quantify the importance of this bias. It should also be highlighted that limited comparison between the PCI and conservative arms can be made given the marked discrepancy in baseline risk of these groups for whom no propensity matching was undertaken. Further, despite the international collaboration, there are actually surprisingly few procedures captured and the cohort described are likely to reflect a small fraction of the total number of patients with SCAD undergoing coronary intervention. Given the limitations of angiography to reliably identify this condition, it seems highly plausible that many more uncomplicated interventions are performed with operators unaware that the underlying diagnosis is non-atherosclerotic. The development of complications likely prompts greater recognition of SCAD, and these patients may be more readily captured in registries. Finally, the small sample size results in wide confidence intervals regarding the relative risk of long-term adverse outcomes for the PCI and conservative cohorts. The point estimates all lean heavily against those who underwent PCI suggesting that outcomes are almost certainly (and understandably) worse when coronary intervention is required.
How then should cardiologists incorporate this knowledge in day-to-day practice? We might do well to consider this new information with a view to the advice of Hippocrates. We ‘can tell (some of) the antecedents’—middle-aged women with few traditional risk factors. We can ‘know the present’—current clinical status and coronary anatomy. Can we ‘foretell the future’? Excellent outcomes for most conservatively managed patients suggest our default should be to ‘do no harm’. But, for an important minority where intervention is deemed a true necessity—complete vessel occlusion, haemodynamic instability—we ought to take courage that there is scope ‘to do good’.
Contributors PDA wrote the editorial and commissioned the figure.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.