Cardiac procedural deferral during the coronavirus (COVID‐19) pandemic – Yong – – Catheterization and Cardiovascular Interventions

1 INTRODUCTION

With the rapid spread of COVID‐19 (SARS‐CoV‐2) across the United States in early 2020,1 large‐scale procedural deferral was implemented in cardiac catheterization laboratories (CCL) nationwide in order to minimize use of limited resources in anticipation of COVID‐19 surge needs and reduce risk of exposure for both patients and providers. Many CCLs adopted these practices per professional society recommendations2 and federal mandates.3, 4 Early data now suggest that patient preference also contributed to a decline in demand for cardiac procedures.5 The degree to which the full spectrum of cardiovascular procedures has been deferred due to this pandemic is unprecedented, and to our knowledge, has not been studied.

We sought to evaluate the impact of COVID‐19 from the perspective of interventional cardiologists. The goal was to improve our understanding of the effects of cardiac procedural deferral when applied at the national level, variability in implementation based on provider and hospital characteristics including COVID‐19 disease burden, and its immediate impact on clinical practice and patient outcomes.

2 METHODS

A 20‐item web‐based survey instrument (Supplement), with embedded subquestions, was administered in May 2020 under the direction of the Society for Cardiovascular Angiography and Interventions (SCAI), in collaboration with the American College of Cardiology (ACC) Interventional Council. From 333 email invitations that were sent to the SCAI CCL director list and opened by the recipient, 288 responses were received over a 2‐week time period, representing an 86% response rate. The ACC listserv had 5,036 members (including some not eligible to participate), from which we received an additional 126 surveys over a 1‐week time period, representing a 3% response rate (underestimate). Weighted proportionally, this resulted in an overall 60% response rate.

Respondents self‐identified as a CCL director or not. For hospitals without a responding CCL director, “CCL director surrogates” were identified from among the interventional cardiologists who responded from those hospitals. Unique respondents from a given CCL were identified using state and zip code, with earliest respondents used as tie‐breakers. The CCL director and surrogate responses were used to analyze CCL‐specific questions, while the entire cohort was used to analyze questions reflecting individual perspectives.

Linear regression models were used to model association between continuous variables. p‐Values were calculated using chi‐square test or binomial test of proportions for categorical variables and Welch’s two‐sample t test for continuous variables. Pairwise comparison across more than two groups was performed using analysis of variance. Analyses of responses were performed using R version 4.0. IRB exemption was granted by the Stanford University Institutional Review Board.

3 RESULTS

A total of 414 responses were received from the United States (59 international responses were excluded, Table 1). One hundred fourteen respondents self‐identified as CCL directors, and we additionally identified 246 CCL director surrogates, representing data from a total of 360 unique CCLs from across the United States. Respondents came from 48 states, with the most from California (11%), New Jersey (7%), and New York (6%). Private (68%), Public (27%), and VA (5%) hospitals were represented, with 63% reporting an academic affiliation. Urban (43%), suburban (41%), and rural (16%) settings were represented. Most interventionalists were in the 40–60 year‐old age bracket, with mean number of females per CCL 0.5 ± 0.8 (64% of CCLs reported no female interventionalists in their group). A shelter‐in‐place mandate was present in 91% of CCL, with 88% starting in March 2020.

3.2 Procedural deferral and volumes

There was significant variability in degree of deferral across the spectrum of CCL procedures (Tables S2 and S3). Likelihood of deferral decreased as the urgency of the procedure increased, with most deferring left atrial appendage closure and transcatheter mitral valve replacement (TMVR), and almost none deferring angiogram/percutaneous coronary intervention (PCI) for unstable angina (UA), non‐ST‐elevation myocardial infarction (NSTEMI) or ST‐elevation myocardial infarction (STEMI, Figure 1). Deferral of angiogram/PCI for UA (p < .001), NSTEMI (p = .04) and STEMI (p < .001) increased as inpatient COVID‐19 disease burden increased (Table S4).

Significant reductions in procedural volumes were experienced across the country. Median monthly PCI volumes from March 15 to April 15, 2020 ranged from 0 to 500, a 55% reduction from 2019 (annual volumes in 2019: 23–6,500 per hospital, mean 795, median 600). Median monthly TAVR volumes during the same time frame ranged from 0 to 40, a 64% reduction from 2019 (annual volumes in 2019: 1–600 per hospital, mean 90, median 45). There was a significant association between a hospital’s COVID‐19 disease burden with reduction in PCI volume (Pearson coefficient 0.3, p < .0001, Figure 2a) and TAVR volume (Pearson coefficient 0.3, p = .004, Figure 2b).

3.4 Outcomes

Almost 40% of CCL directors reported an increase in late presenting STEMIs and almost 20% reported increased mortality in patients waiting for TAVR (Figure S1). Geographic mapping of late presenting STEMIs overlaid on the map of COVID‐19 mortality burden by state6 revealed no consistent overlap of areas with increased late presenting STEMIs and high COVID‐19 burden regions (Figure 3).

3.6 Interventionalist salary

Almost 1/2 of CCL directors reported a salary reduction due to COVID‐19. Among them, 95% reported a drop in base salary (mean 23%, maximum 70%) and 99% reported a drop in wRVUs (mean 46%, maximum 90%). There was no association between salary reduction and degree of angiogram/PCI deferral for either stable ischemic heart disease (SIHD) or UA (Figure 4). Most CCL directors reported they are in the process of reinstituting normal CCL operations, but this was not driven by magnitude of salary reductions (Figure S2).

4 DISCUSSION

In the first national survey of CCL directors and interventional cardiologists about cardiac procedural deferral in the United States during the COVID‐19 pandemic, we identified unprecedented large‐scale procedural deferrals, plummeting procedural volumes, and negative impacts on patient outcomes. We also identify physician COVID‐19 positivity and salary reductions, but with reassuringly imperceptible overt influence on clinical decision‐making. We highlight perspectives regarding the greatest challenges moving forward, with a focus on the most pressing unresolved needs as we make plans to reinstitute elective clinical care.

Our findings about COVID‐19 disease burden among patients reveal dramatic variability by center and geography. We found that centers with higher COVID‐19 burden performed more universal testing, but it is unclear to what degree these interact. Many are concerned about insufficient rapid testing, which weighs disproportionately on lower income patients and those who travel long distances for procedures. Preprocedural reliability also decreases, given possible interim exposures.7

Clear prioritization of procedural deferral reflects a shared understanding of their evidence‐base. For example, deferral of PCI for SIHD was high—which comes in the wake of trials supporting the safe deferral of PCI as the initial approach in SIHD.8, 9 Given that left atrial appendage closure and TMVR were the most likely to be deferred, this suggests that both chronicity of disease and risk of transesophageal echocardiogram were other influential factors.

Interventional cardiologists perceived that noninvasive imaging studies were 14–24% more likely to be used to risk stratify patients in lieu of angiography. Just as the COVID‐19 pandemic accelerated use of telehealth, it remains unknown if the increases in noninvasive imaging may have lasting effects on the practice of cardiology, potentially even leading to overuse. Similarly, while primary PCI has long replaced thrombolytic therapy as standard of care in the United States,10 the new challenges of COVID‐19 have at least temporarily made it an option at primary PCI centers.11 However, our practice patterns have not mirrored more thrombolytic heavy strategies adopted in China and other countries,12, 13 Many practicing interventionalists in the United States trained in the primary PCI era, with minimal direct experience with thrombolytics or late presentation myocardial infarctions, which could impact management and patient outcomes during this period.

Despite prioritization of procedures with immediate mortality benefits, cardiologists perceived that patients are presenting later with STEMI. These reports are consistent with early findings of increased out‐of‐hospital arrests and other late presentations during the pandemic.5, 14–20 Our data on late presenting STEMIs overlaid on the map of nationwide COVID‐19 mortality burden suggests that patient fears may be driving behaviors regardless of actual risk in a region—in contrast to physician decision‐making, which correlated with local COVID‐19 burden. Patient education efforts about the importance of seeking timely healthcare even in the face of the pandemic are critical to improve patient outcomes.

The finding that almost a third of CCL directors reported at least one COVID‐19 positive operator in their laboratory likely reflects community exposures, but also raises the question of the degree to which hospital exposure could contribute. With emerging data on the overlap of COVID‐19 and cardiac presentations,21 it is plausible that insufficient testing may result in poor identification of COVID‐19 infection among cardiac patients. It is unclear if the associations between COVID‐19 infection with age22 and sex23, 24 reflect variable testing due to heightened concerns/symptoms among older individuals, or simply confounding data (e.g., a larger CCL may be more likely to have females and positive testing, even if they are not directly linked).

5 STUDY LIMITATIONS

There are many limitations to this study. First, despite a 60% response rate (compared to the 11% average in web‐based surveys25) representing almost all states and a broad spectrum of COVID‐19 burden, we cannot be sure that it captures the perspectives of all 2,000+ CCLs nationwide. We recognize that a high proportion of our respondents reported academic affiliation with a high bed count. We likely underestimated the response rate from the ACC listserv since we did not have the breakdown of recipients not eligible for the study (e.g., noninterventional cardiologists). Given that the survey was open for a 3‐week time period in May 2020, it is also possible that perspectives shifted during even this short period, potentially confounding direct comparisons. While we assured anonymity, we did request identification of hospital and zip code and recognize this may have impacted responses.

6 CONCLUSIONS

As we live and work through the worst pandemic in our lifetimes, a nationwide survey of CCL directors and interventional cardiologists demonstrates that we have adapted quickly to protect our patients amidst the spread of the SARS‐CoV‐2. We rapidly implemented unprecedented nationwide procedural deferrals across the spectrum of cardiovascular care, while prioritizing continued interventions for the most life‐threatening indications. Regardless, patients are suffering the cost in the form of negative cardiovascular outcomes, some driven by heightened patient fears, which do not appear to correlate with risk of exposure. As we transition into a new era of cardiovascular care in the time of COVID‐19, we can use the priorities identified here to help build our strategies moving forward.

CONFLICT OF INTEREST

F.G.P.W. reports consulting relationship with Medtronic Inc. E.M. reports clinical trial research support (Corindus, Abbott Vascular, CSI), consulting (Medtronic, Abiomed). W.F.F. receives institutional research support from Abbott Vascular, Medtronic Inc., and Edwards LifeSciences, has a consulting relationship with CathWorks and minor stock options with HeartFlow. C.M.Y., L.A., S.G., T.D.H., D.S.P., D.C., P.W., and S.A. report no relevant disclosures.