key: cord-1043219-ijlezovw
authors: Ramzy, John; Martin, Catherine A.; Burgess, Sonya; Gooley, Robert; Zaman, Sarah
title: COVID-19 Pandemic Impact on Percutaneous Coronary Intervention for Acute Coronary Syndromes: An Australian Tertiary Centre Experience
date: 2022-02-11
journal: Heart Lung Circ
DOI: 10.1016/j.hlc.2021.10.019
sha: bdfb84500974ca20e2891c17786543399a6100d4
doc_id: 1043219
cord_uid: ijlezovw

BACKGROUND: Countries who suffered large COVID-19 outbreaks reported a decrease in acute coronary syndrome (ACS) presentations and percutaneous coronary intervention (PCI). The impact of the pandemic in countries like Australia, with relatively small outbreaks yet significant social restrictions, is relatively unknown. There is also limited and conflicting data regarding the impact on clinical outcomes, symptom-to-door time (STDT) and door-to-balloon time (DTBT). METHODS: Consecutive ACS patients treated with PCI were prospectively recruited from a tertiary hospital network in Melbourne, Australia. The pre-pandemic period (11 March 2019–10 March 2020) was compared to the pandemic period (11 March 2020–10 May 2020) using an interrupted time series analysis with a primary endpoint of number PCI-treated ACS per day. Secondary endpoints included STDT, DTBT, total mortality and major adverse cardiac events (MACE). RESULTS: A total 984 ACS patients (14.8% during the pandemic period) received PCI. Mean number of PCI-treated ACS per day did not differ between the two periods (2.3 vs 2.4, p=0.61) with no difference in STDT [+51.3 mins, 95% confidence interval (CI) -52.4 to 154.9, p=0.33], 30-day mortality (5% vs 5.3%, p=0.86) or MACE (5.2% vs 6.1%, p=0.68). DTBT was significantly longer during the pandemic versus the pre-pandemic period (+18.1 mins, 95% CI 1.6–34.5, p=0.03) and improved with time (slope estimate: -0.76, 95% CI -1.62 to 0.10). CONCLUSIONS: Despite significant social restrictions imposed in Melbourne, numbers of ACS treated with PCI and 30-day outcomes were similar to pre-pandemic times. DTBT was significantly longer during the COVID-19 pandemic period, likely reflecting infection control measures, which reassuringly improved with time.

On 11 March 2020 the World Health Organisation (WHO) declared coronavirus disease 2019 (COVID-19) a pandemic. On 16 March a state of emergency (SOE) was declared in the state of Victoria, Australia. This was rapidly followed by significant lockdown measures which lasted until the 11 May 2020, at which stage restrictions were eased following the first 'wave' of COVID-19. Countries who suffered large COVID-19 waves of infection have seen decreases in acute coronary syndromes (ACS) and interventional procedures of 26-40% [1] [2] [3] [4] [5] [6] . This has been hypothesised to be multifactorial with causes including missed ACS diagnoses in overwhelmed hospitals, higher rates of thrombolysis, competing risk of death from COVID-19 and changes in health care seeking behaviour of the general population. With regards to the latter, concerns have been raised that during periods of social restrictions, patients with ACS may avoid hospitals due to fear of contracting COVID-19, or dismiss their symptoms to avoid being an additional burden on health care services [7] [8] [9] [10] .

Current guidelines recommend timely reperfusion therapy for ST-elevation myocardial infarction (STEMI) with a doorto-balloon (DTB) time of less than 90 minutes [11, 12] . During the pandemic, additional precautions by emergency and hospital staff may impact the provision of timely care due to the need for COVID-19 precautions. However, much of our data regarding management of STEMI during the COVID-19 era has come from locations where health care services were overwhelmed with COVID-19 patients such as New York and Lombardy.

The effect on ACS care of government-mandated social restrictions, social distancing and heightened health care worker infection precautions, in the absence of high COVID-19 numbers, is not known. The state of Victoria in Australia represents a unique environment to study this, with significant and severe lockdown measures enforced, without a large influx of COVID-19 cases into hospitals. While there are reports that Victorian catheterisation laboratories have seen a reduction in percutaneous coronary intervention (PCI) for ACS during the pandemic, there is no data from high volume centres to confirm this. This study reports ACS and PCI numbers and key performance indicators (such as door-toballoon time), during the first Victorian COVID-19 lockdown.

Consecutive patients who underwent percutaneous coronary intervention (PCI) for ACS were prospectively recruited and followed for 30 days as part of the Victorian Cardiac Outcomes Registry (VCOR) (2013-current). Patients who received PCI at MonashHeart, a tertiary referral hospital, treating ACS patients from four hospitals (Monash Medical Centre, Dandenong Hospital, Casey Hospital and Jessie McPherson Private Hospital) in Melbourne, Victoria, Australia, were included. The patient population served by this hospital group is between 913,000 and 1,294,000 people.

Results were analysed according to time periods. The 'prepandemic' period was determined as the time from 11March 2019 to 10 March 2020 while the 'pandemic' period was defined as 11 March 2020 to 10 May 2020 (the day the WHO declared COVID-19 a pandemic to the day the stage 3 COVID-19 restrictions in Victoria ended). Daily ACS and PCI numbers, patient clinical characteristics, medication use, in-hospital and 30-day adverse outcomes were compared between the pre-pandemic and pandemic group. For the door-to-balloon time (DTBT) analysis only STEMI patients who presented directly to a PCI capable facility within 12 hours of symptom onset were included, while those with symptom onset while in hospital or who received PCI after thrombolytics were excluded. Identical data collection methods were used for the pre-pandemic and COVID-19 periods. The study was approved by the Monash Human Research Ethics Committee with an optout consent. De-identified VCOR data is available upon request to the VCOR Data Access, Research and Publications Committee by emailing vcor@monash.edu.

The primary endpoint was the number of PCI performed each day for ACS. Secondary endpoints included symptomto door time (STDT) and DTBT time in STEMI patients, inhospital and 30-day major adverse cardiac events, major adverse cardiac and cerebrovascular events, and major bleeding for all ACS patients. The relevant times were defined as follows: STDT as time from symptom onset to arrival at a PCI-capable centre, DTBT as time from arrival at a PCI-capable hospital to time of establishing flow in the culprit vessel (independent of which device was used for revascularisation). Major adverse cardiac events (MACE) were defined as death, new or recurrent MI, stent thrombosis, and target vessel revascularisation. Major adverse cardiac and cerebrovascular events (MACCE) were defined as for MACE with the addition of stroke. The international Bleeding Academic Research Consortium standardised bleeding definitions were used to report on major bleeding events (identified by Bleeding Academic Research Consortium categories 3 and 5), including bleeding that required blood transfusion, cardiac tamponade, intracranial haemorrhage, and/or any fatal bleeding.

Interrupted time series analysis was chosen as the most appropriate method of statistical analysis to assess the impact of the COVID-19 pandemic on the volume of PCI for ACS as well as STDT and DTBT in STEMI. The study periods were initially collapsed by month, producing a time series of number of PCIs per month to determine if seasonality was present. For analysis of STDT and DTBT the data was collapsed by day. In exploratory analyses stationarity of the time series was assessed using the Dickey-Fuller unit root test. Determination of any autocorrelation and an appropriate lag to model the autocorrelation was assessed by visualising the autocorrelation function (ACF), the partial autocorrelation function (PCF) and by Stata's actest. Time series was modelled using generalised linear modelling (glm) [13] . For count or binary outcomes glm with a negative binomial probability distribution and a log link function was used. A heteroskedastic and autocorrelation consistent variance estimator was used, with Newey West standard errors. Seasonality was tested by including indicator variables for winter in the model. The adequacy of the models was investigated by examining the plot of observed modelestimated monthly event count against the observed event count and using the estimated error variance. The ratio of estimated regression slopes was tested with a slope ratio of 1 showing no difference in slopes (no difference in the rate of change over time). The ratio of the midpoints of estimated regression slopes was tested, with a midpoint ratio of 1 showing no difference in midpoint estimates. Statistical significance was defined as a two-sided p-value 0.05. Stata Statistical Software: Release 15 (StataCorp LP, College Station, TX, USA) was used for all analyses.

A total of 984 ACS patients were treated with PCI over the study period, 839 during the defined pre-pandemic period and 145 during the pandemic period. A time series analysis of the number of PCI procedures performed for ACS each month from 1 January 2016 to 10 May 2019 showed that there was no seasonal component (Supplemental Figure 1 ). The addition of a variable for winter demonstrated to not be significant in a model of monthly PCI count. We therefore performed an interrupted time series analysis comparing the 8-week pandemic period (11 March 2020-10 May 2020) to a 12month pre-pandemic period (11 March 2019-10 March 2020).

The mean number of PCI for ACS per day did not differ between the pre-pandemic and pandemic periods (2.3 vs 2.4, p=0.61). The proportion of PCI performed for each sub-type of ACS was also similar between both time periods (Table 1) 

There were no significant differences in the baseline clinical characteristics of the ACS patients in the pre-pandemic and proportion of patients presenting with an ACS following an out-of-hospital cardiac arrest, and no rise in pre-procedural intubation during the pandemic period. Similarly, the baseline characteristics of the patients presenting with STEMI did not differ between the two periods ( Table 2) .

Following exclusion of patients with in-hospital STEMI there were a total of 531 STEMI PCI procedures available for STD time analysis. Baseline and procedural characteristics did not differ between the 446 STEMI patients in the pre-pandemic period and 85 patients in the pandemic period ( Table 2) .

Interrupted time series analysis showed that there was a trend towards longer STDT during the pandemic period although this was not statistically significant (difference in midpoint estimates: 151.3 mins, 95% CI -52.4 to 154.9, p=0.33). There was also no significant change in STDT throughout the course of each period as demonstrated by the slope estimates ( Figure 2 ). There were 246 STEMI patients in the pre-pandemic period and 44 in the pandemic period who presented within 12 hours to a PCI-capable hospital and were included in the DTBT analyses. Baseline characteristics did not differ between these two groups (Appendix Table 1 ). There was a significant increase in the DTBT during the pandemic period with a difference in the mid-point estimate of 18.1 minutes (95% CI 1.6-34.5, p=0.03) ( Table 3) . Interrupted time series analysis also demonstrated that the prolongation in DTBT during the pandemic period was greatest at the beginning of this period and improved throughout the course of the pandemic (Figure 3) . Amongst the STEMI patients who met the criteria for DTBT analysis, there were no significant differences in in-hospital outcomes between the two periods, although event rates were low (Appendix Table 2 ).

Procedural success was high in both the pandemic and prepandemic ACS groups (94.6% vs 95.4%, p=0.71) and there were no significant differences in in-hospital outcomes such as length of stay (LOS), bleeding and stroke (Table 4) . MACE and MACCE were also similar between the two groups (5.2% vs 6.1%, p=0.68 and 5.4% vs 6.1%, p=0.73 respectively). Similarly, 30-day mortality did not differ between the pre-pandemic and pandemic periods (5% vs 5.3%, p=0.86).

This is the first study from a high-volume Australian tertiaryreferral centre to evaluate the impact of the COVID-19 pandemic on volume, service delivery and outcome of PCI for ACS. It had three major findings. Firstly, using interrupted time series analyses, no change was observed for total ACS numbers, ACS type, or number of PCI, between the prepandemic and pandemic period. Secondly, a significant increase in door-to-balloon time for STEMIs was observed during the pandemic period that was most marked at the start of the COVID-19 pandemic and improved over time.

Lastly, despite delays to STEMI treatment, no difference in short-term ACS outcomes were apparent during the pandemic compared to the pre-pandemic period. The Australian experience provides valuable insight into the effects of COVID-19 social restrictions and heightened health care worker infection precautions in the absence of high COVID-19 numbers. When the state of emergency was declared in Victoria on the 16 March there had been a total of 78 cases of COVID-19 in the state since the start of the pandemic and no deaths [14] . Over the course of the pandemic period studied there were 1,473 cases in the state and 18 deaths [14] . In contrast, there were 208,950 cases and 33,012 deaths in the UK over the same period [15] . In contrast to our findings, reports from around the world have described a marked decrease in ACS presentations and interventional procedures during the pandemic [1] [2] [3] [4] [5] [6] . Data from Italy showed a 32% decrease in PCI for ACS [5] , and a 26% relative decline in ACS admissions at the start of the pandemic [1] . Similarly, a survey of 73 centres in Spain found a 40-56% reduction in PCI procedures during the pandemic [4] . An Austrian study showed there were 275 fewer ACS presentations than expected at 17 centres in March 2020 [3] . An observed decrease in STEMI presentations has been of particular concern. A 38% reduction in cathlab activations for STEMI was demonstrated among nine high volume centres in the United States (US) [2] . While the United Kingdom (UK) found a 43% reduction in STEMI [6] . However, our findings are consistent with those of the only other Australian study to examine PCI volumes for ACS prior to and post the pandemic and found no difference (20 vs 18 cases/month, p=0.2) [16] . The unchanged PCI volumes at our institution may be due to the much smaller scale of the initial COVID-19 outbreak in Australia. A prevailing theory for the reduction in ACS presentations worldwide, is concern regarding nosocomial COVID-19 infection deterring patients from presenting. This was illustrated in a survey of British cardiologists, 71% of whom believed this was the cause and 46% of whom believed patients were hesitant to place an additional burden on overwhelmed health services [17] . These apprehensions may be amplified by stay-at-home orders and alerting news media coverage [18] . Our results suggest that, in an environment where COVID-19 case numbers, hospitalisations and deaths are low, these concerns are either less prevalent or do not lead to a significant change in the behaviour of patients. Similarly, it is possible that stayat-home orders and alarming news media coverage, both of which were present in Victoria, may not exert the same influence on patients when case numbers are low.

Our study represents the second largest examination of the impact of the pandemic on STDT and DTBT in STEMI performed to date. While there was no significant change in STDT, there was a statistically significant increase in DTBT. Our STDT findings differ to those from a retrospective but larger cohort study by Kwok [6] . Two (2) smaller studies have also examined STDT, including one from another tertiary centre in Victoria and one from Hong Kong [9, 16] . Although both found an increase in STDT during the pandemic, they included only three and seven STEMI patients in the pandemic period respectively. Analyses of DTBT during the pandemic have produced mixed results. Kwok et al. reported Table 4 In hospital outcomes and 30-d mortality for STEMI and NSTEMI, only exclusion is inpatient event. a statistically significant 11-minute increase in median door to balloon time after the UK lockdown [6] . Smaller studies from the US and Austria found no significant difference in DTBT for STEMI during the pandemic [19, 20] . Unlike in the UK where large influxes of COVID-19 patients may have impacted service delivery and DTBT, the DTBT delay observed in our study is likely a direct reflection of the introduction of stringent personal protective equipment (PPE) protocols, screening questions and mandatory COVID-19 testing in the emergency department prior to proceeding to the cathlab. A unique feature of our analysis is the use of an interrupted time series analysis which revealed that delays in DTBT improved with time during the pandemic period. This finding likely reflects the health care service adapting over time with streamlining of processes.

Our study found no significant change in the proportion of PCI performed for each subtype of ACS during the pandemic. This is in contrast to several reports from Europe that found larger decreases in NSTEACS than STEMI [18] . Importantly this occurred in settings where overall ACS presentations declined, which was not the case in our study. An explanation for this may be offered by the findings of an analysis of ED presentations in the neighbouring state of New South Wales between 29 March and 30 May 2020 [21] . While there was a significant 12% decrease in presentations with 'chest pain not resulting in another diagnosis' compared to the same period in 2019, there was no change in the number of ACS presentations [21] . Along with our findings, this suggests that, while there was hesitancy amongst patients to present to ED during the pandemic in Australia, this primarily affected patients with non-specific chest pain not those who were found to have an ACS.

Our study found no increase in short-term adverse outcomes among both our overall ACS population and our STEMI cohort during the pandemic, despite longer DTBT. This is in keeping with studies from the UK and Austria that showed no change in short-term outcomes among STEMI patients despite longer ischaemic times [6, 20] . Conversely a prospective single centre study from France found a twofold higher in-hospital mortality among 83 patients with STEMI during the pandemic, however this was in the setting of a very large (3.6-hour) increase in the mean symptom-to-balloon time during the outbreak [8] . The present study is the largest to examine outcomes in an overall ACS cohort during the pandemic. The only other study to examine this included 64 ACS patients presenting during the pandemic who found a significantly higher rate of the composite endpoint of in-hospital death, cardiogenic shock, sustained ventricular tachycardia/ventricular fibrillation (VT/VF) and use of mechanical circulatory support compared to 85 patients presenting prior (14.1% vs 29.7%, p=0.02). While our findings are reassuring that ACS and STEMI patients have not been negatively impacted in the short term by the pandemic and consequent longer DTBT, studies examining the longer term effects of the pandemic are needed.

This study has several limitations. We studied ACS patients from across four hospitals within a tertiary health service and our findings may not be representative of the impact of the pandemic more widely in Victoria or Australia. Additionally, the study was limited only to ACS patients who were treated with PCI as these patients were prospectively captured in the VCOR Registry. We therefore cannot comment on numbers of ACS that were not deemed suitable for PCI. Finally, while we corrected for seasonality with use of the interrupted time series analysis, we cannot exclude other factors that may have influenced ACS numbers such as pollution and patientlevel lifestyle changes due to stay-at-home orders. Study size may also have limited power to demonstrate difference in ACS presentations over this period and larger studies are also warranted, these studies may also benefit from data collection beyond that of the current study to evaluate any late impact on presentations due to Victoria's much larger "second wave" of COVID-19 cases from June to September.

Unlike the decline in ACS presentations and PCI reported broadly overseas, our study found that in the largest cardiac interventional health care service in Victoria, Australia there was no change in the volume of PCI for ACS during the COVID-19 pandemic. Using an interrupted time series analysis, we found that ACS and PCI numbers as well as time from symptom onset to hospital presentation were similar between the pre-pandemic and pandemic period. While a significant increase in DTBT was seen during the COVID-19 pandemic reassuringly, this delay improved with time, and we did not appear to impact short-term outcomes after PCI for ACS.

Reduced rate of hospital admissions for ACS during Covid-19 outbreak in Northern Italy

Reduction in ST-segment elevation cardiac catheterization laboratory activations in the United States during COVID-19 pandemic

Decline of acute coronary syndrome admissions in Austria since the outbreak of COVID-19: the pandemic response causes cardiac collateral damage

Impact of the COVID-19 pandemic on interventional cardiology activity in Spain. REC Interv Cardiol

Population trends in rates of percutaneous coronary revascularization for acute coronary syndromes associated with the COVID-19 outbreak

Impact of COVID-19 on percutaneous coronary intervention for STelevation myocardial infarction

Be prepared. Am Heart Assoc

Changes in characteristics and management among patients with STelevation myocardial infarction due to COVID-19 infection

Impact of coronavirus disease 2019 (COVID-19) outbreak on ST-segment-elevation myocardial infarction care in Hong Kong

Acute coronary syndromes undergoing percutaneous coronary intervention in the COVID-19 Era: comparable case volumes but delayed symptom onset to hospital presentation

ACCF/AHA guideline for the management of STelevation myocardial infarction: executive summary: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines

Prognostic implications of door-to-balloon time and onset-to-door time on mortality in patients with ST-segment-elevation myocardial infarction treated with primary percutaneous coronary intervention

Segmented regression analysis of interrupted time series studies in medication use research

Victorian COVID-19 data

The official UK government website for data and insights on coronavirus (COVID-19)

Cardiovascular mortality following liver transplantation: predictors and temporal trends over 30 years

Lives at risk due to 50% drop in heart attack A&E attendances

The impact of the COVID-19 pandemic on the care and management of patients with acute cardiovascular disease: a systematic review

Impact of COVID-19 pandemic on ST-elevation myocardial infarction in a non-COVID-19 epicenter

Effect of the COVID-19 pandemic on treatment delays in patients with ST-segment elevation myocardial infarction

Fewer presentations to metropolitan emergency departments during the COVID-19 pandemic

COVID-19 Pandemic Impact on Percutaneous Coronary Intervention for Acute Coronary Syndromes: An Australian Tertiary Centre Experience

SZ is supported by a National Heart Foundation Fellowship. She has also obtained research funding from Abbott Vascular and Biotronik Australia and speaking honoraria or consulting fees from AstraZeneca, Amgen and Medtronic.RG has received consulting fees from Abbott Vascular, Medtronic and Boston Scientific.

This project did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.