Impact of respiratory infectious epidemics on STEMI incidence and care

The effect of respiratory infectious diseases on STEMI incidence, but also STEMI care is not well understood. The Influenza 2017/2018 epidemic and the COVID-19 pandemic were chosen as observational periods to investigate the effect of respiratory virus diseases on these outcomes in a metropolitan area with an established STEMI network. We analyzed data on incidence and care during the COVID-19 pandemic, Influenza 2017/2018 epidemic and corresponding seasonal control periods. Three comparisons were performed: (1) COVID-19 pandemic group versus pandemic control group, (2) COVID-19 pandemic group versus Influenza 2017/2018 epidemic group and (3) Influenza 2017/2018 epidemic group versus epidemic control group. We used Student’s t-test, Fisher’s exact test and Chi square test for statistical analysis. 1455 patients were eligible. The daily STEMI incidence was 1.49 during the COVID-19 pandemic, 1.40 for the pandemic season control period, 1.22 during the Influenza 2017/2018 epidemic and 1.28 during the epidemic season control group. Median symptom-to-contact time was 180 min during the COVID-19 pandemic. In the pandemic season control group it was 90 min (p = 0.183), and in the Influenza 2017/2018 cohort it was 90 min, too (p = 0.216). Interval in the epidemic control group was 79 min (p = 0.733). The COVID-19 group had a door-to-balloon time of 49 min, corresponding intervals were 39 min for the pandemic season group (p = 0.038), 37 min for the Influenza 2017/2018 group (p = 0.421), and 38 min for the epidemic season control group (p = 0.429). In-hospital mortality was 6.1% for the COVID-19 group, 5.9% for the Influenza 2017/2018 group (p = 1.0), 11% and 11.2% for the season control groups. The respiratory virus diseases neither resulted in an overall treatment delay, nor did they cause an increase in STEMI mortality or incidence. The registry analysis demonstrated a prolonged door-to-balloon time during the COVID-19 pandemic.

Multiple triggers of ST-elevation myocardial infarction (STEMI) were described previously, but the effect of respiratory infectious diseases on STEMI incidence is not well understood. The effect of these infectious diseases on treatment of patients with STEMI is not well investigated. The Influenza 2017/2018 season and the coronavirus disease 2019 (COVID-19) pandemic were chosen as observational periods to investigate the effect of respiratory virus diseases on STEMI incidence and treatment in a metropolitan area with a preexisting STEMI network. COVID-19 caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was declared a pandemic by the World Health Organization on March 11, 2020 1 . Previous studies and expert opinions around the world raised concerns about a trend towards decreased ST-elevation myocardial infarction (STEMI) admissions, prehospital treatment delay, as well as an overwhelmed medical system caused by the pandemic [2][3][4][5][6][7] . STEMI patients suffer from high mortality and morbidity rates, thus requiring prompt diagnosis and treatment 8 . After the initial outbreak in Wuhan, China in December 2019, the first case of COVID-19 was documented in Germany on January 27, 2020. On February 25, 2020, the first person in the German state of North Rhine Westphalia was diagnosed with COVID-19. The government of North Rhine Westphalia issued containment recommendations to hospitals on March 3, 2020, and passed a coronavirus executive order on March 22, 2020 9 . This lockdown resulted in contact restrictions, remote work and an extreme reduction of public life and mobility.
On the contrary, the Influenza 2017/2018 season-which was declared as epidemic in Europe and was a prototype of a seasonal respiratory infectious disease with high disease burden-did not lead to the implementation of comparable containment recommendations or executive orders 10 . The influenza season in general was described as predictor of increased incidence of myocardial infarction and might raise the cardiovascular risk of patients with coronary artery disease [11][12][13] . During the Influenza 2017/2018 epidemic in Germany 334,000 patients were diagnosed with influenza, of these, 60,000 required hospitalization 10 . The Influenza 2017/2018 epidemic resulted in a doubling of the registered influenza-caused deaths compared to the prior season 2016/2017 10 . The effect of this disease burden on the treatment of STEMI patients was not investigated before.
The current study aims to clarify the role of the recurrent influenza season on STEMI care exemplified by the 2017/2018 season. This study was also conducted to evaluate the hypothesized concerns about a decrease in STEMI admissions and treatment delay during the COVID-19 pandemic. Specifically, it addressed patient-and system-related delay in a myocardial infarction network in the city of Cologne, Germany 14,15 .

Methods
Study design and population. This prospective, observational, multicenter cohort study included all STEMI patients diagnosed and treated within the period of December 1, 2013 to April 30, 2020 in the city of Cologne, Germany. The concept of the Cologne Infarction Model network ("Kölner Infarkt Modell", KIM) has been described previously 14,15 . Briefly, KIM is a co-operation between all sixteen hospitals and the emergency medical services in the 400 km 2 area of Cologne, Germany, populated by about 1 million people. Seven out of sixteen hospitals feature percutaneous coronary intervention (PCI) capabilities 24 h a day, 7 days a week. STEMI patients who first presented to emergency medical service were directly transferred to the catheterization laboratories of those PCI centers. Patients who presented to non-PCI hospitals were immediately transferred to hospitals with catheterization laboratory capabilities. Guideline conform STEMI diagnosis required typical symptoms in the presence of either ST-segment elevation in at least two contiguous leads or assumed new onset left bundle branch block (LBBB) on a 12-lead electrocardiogram 8,16 . All patients with STEMI and a complete report on treatment periods reflecting pre-clinical and in-hospital delay were eligible for the current analysis. The study complies with the Declaration of Helsinki. Ethical approval was issued by the local ethics committee of the University of Cologne (No. 06-064) and written informed consent was obtained from all patients. The KIM registry is funded by the KIM registered association.
Observational period. The first wave of the COVID-19 pandemic began on January 27, 2020 and ended around April 30, 2020 in Germany 17 . We chose this timeframe for the COVID-19 pandemic observational group. This group was compared to a seasonal control group (pandemic season control group), including all patients treated between January 27 and April, 30 between the years 2014 to 2019. All other patients were excluded because of a potential seasonal effect on health care provision and patients' behavior.
The www.nature.com/scientificreports/ results and peri-interventional complications were extracted from the registry. Peri-interventional data included the access route, culprit lesion, antiplatelet therapy, the maximum amount of creatin kinase, local bleeding rates and peri-interventional ventricular arrhythmia.
Statistical analysis. Data  In the comparison of both respiratory infectious diseases, the COVID-19 pandemic had longer S2C, C2B and D2B intervals, but none of these differences was significantly different (see Table 2b).

Procedural data. The radial artery was the predominant access route in the COVID-19 pandemic and the
Influenza 2017/2018 epidemic group (see Table 3). These proportions were statistically significant different compared to those from the seasonal control groups, as the femoral artery puncture were performed in 64.2 and 70.2 of all procedures in the historical cohorts (p < 0.001). Angiography revealed that LAD occlusion was the most frequent cause of STEMI in all patients, but the proportion of LAD occlusion was significantly higher in patients of the Influenza 2017/2018 epidemic group (55.6%) in comparison to the epidemic season control group (36.7%, p = 0.011). Myocardial infarction of the inferior wall either caused by occlusion of RCA or subsequent branches was the second most common cause for STEMI. After primary angiography 39 patients were transferred to urgent CABG operation. In the remaining cohort, interventional reperfusion (TIMI I-III) was achieved in all but 142 patients.
Peri-interventional complications occurred in 8.5% of patients in the COVID-19 pandemic and 5.9% of patients in the Influenza 2017/2018 epidemic group (see Table 4). The corresponding rates were 14.6% and 14.8% for the seasonal control groups. Specifically, in-hospital mortality was 6.1% for COVID-19 pandemic group and 11.0% for the pandemic season control group (p = 0.129). Corresponding rates were 5.9% for the Influenza 2017/2018 epidemic and 11.2% for the related epidemic season control group (p = 0.333).

Discussion
Major findings. The incidence of STEMI in the metropolitan area of Cologne was steady during the first wave of COVID-19 pandemic or the Influenza 2017/2018 epidemic. We observed adequate intervals between first medical contact and revascularization. In detailed analysis, both diseases demonstrated a trend towards an increased patient-related delay compared to each corresponding seasonal control group without significant differences. Additionally, the current analysis showed a prolonged door-to-balloon time during the COVID-19 pandemic. The direct comparison of the COVID-19 pandemic and Influenza 2017/2018 epidemic did not demonstrate a significant difference regarding system-or patient-related delay.  www.nature.com/scientificreports/ STEMI incidence. The incidence of STEMI was not significantly affected by the respiratory virus diseases in this registry analysis. To our knowledge, this is the first analysis on the effect of the Influenza 2017/2018 season on STEMI care. Over the past years, only few studies investigating the effect of influenza season on myocardial infarction and especially on STEMI incidence or care were published [11][12][13] . An analysis of the SWEDEHEART registry demonstrated an increased risk of myocardial infarction during the yearly influenza season and a correlation with influenza burden 11 . In a subgroup analysis STEMI patients were not at higher risk for cardiovascular mortality during the influenza season 11 . This result is in line with the current study 11 . Registry data in the United States demonstrated that influenza infection was an independent predictor of in-hospital mortality in patients with myocardial infarction 13 . Concomitant influenza infection and myocardial infarction reduced the rate of revascularization rate significantly 12,13 . On the contrary, a large amount of studies regarding the COVID-19 pandemic and the effect on patients with STEMI have been published over the last year. Studies done in Europe and China documented a decrease in hospital admissions for STEMI 2,18-20 . De Rosa et al. reported a 26.5% reduction of STEMI admission during a 1 week period and Scholz et al. detected a 12.6% decrease during a 1 month period 18,20 . All of these studies were observational studies or registries and might have underreported the true incidence of STEMI during the respiratory virus epidemic and pandemic. Treatment delay. Regarding the STEMI treatment during the COVID-19 pandemic the S2C, C2B and D2B were longer compared to the pandemic season control group, but only the D2B was significantly prolonged. The median S2C was 180 min and this was a doubling in comparison to the pandemic season control group. As this was not a statistical significant difference, careful interpretation is required. Nevertheless this difference might have clinical impact and is in line with observations in Europe and Asia. De Rosa et al. and Gramegna et al. observed a relevant patient-related delay with a prolonged symptom-to-hospital and contact-to-balloon time during the COVID-19 pandemic 18,21 . Interestingly, these studies took place in outbreak areas during the climax of the early COVID-19 pandemic in Italy. A German analysis of 15,800 patients documented comparable S2C, C2B and D2B times for the COVID-19 pandemic and control group without any significant differences 20 . Multiple studies documented an increase or a trend towards a patient-related delay during the COVID-19 pandemic, but the extent differs between the trials [3][4][5]19,20  The reasons for patient-related delay are still under consideration. Altruistic behavior has been intensively discussed. One might further argue that patients avoided hospitals as they represent the focus of this viral disease, and because hospitals were confronted by an excessive demand of intensive care unit capacities by COVID-19 patients. These circumstances might also be an explanation for the observed prolonged D2B time in the current analysis. Results by a meta-analysis from Rattka et al. evolving around 50,000 participants also demonstrated a similar increase in D2B time for the COVID-19 group 22 . Mortality. The registry data indicate a deceased mortality during the Influenza 2017/2018 epidemic and COVID-19 pandemic period. In the setting of the COVID-19 pandemic, we speculate that the availability of resources due to the reduction in elective procedures could have had an impact on these outcomes. On the other hand, a potentially increased local sudden cardiac death rate might have improved the in-hospital outcome (survivorship bias). The current local death rate of the city of Cologne, Germany is not available, hence the true explanation of this phenomenon is still pending and might be addressed in future studies. Until April 30, 2020, less than 0.24% (N = 2346) of the total population had a SARS-CoV-2 infection 17 . In contrast, reports by De Rosa et al. and Gramegna et al. included hospitals from outbreak areas with much higher incidence of COVID-19 infections. This might not only have influenced the system, but also the patient-related delay 18,21 . Inherent limitations of registry studies also affected the current analysis. A short follow-up period, underreporting of outcome data, limited availability of patient characteristics including comorbidities and treatment data leading to missing data restrict the generalizability of our results. Assessment of subjective variables (e.g. symptom onset) and data management by diverse centers might also influence the study results. One additional limitation is the lack of information on the incidence of influenza infection or antiviral treatment in the cohorts. Unfortunately these variables are not part of the KIM registry, but might have biased the current analysis.
Another important aspect is the local structure of Cologne with a dense structure of hospitals with short transfer periods and the small sample size in the KIM registry. The current analysis is also restricted by the low number of events especially in the observational groups.

Conclusion
A well-structured co-operation of emergency medical service and local hospitals in a metropolitan area was able to overcome the challenges in the treatment of patients with STEMI during the COVID-19 pandemic and the Influenza 2017/2018 epidemic. In the setting of this network, these respiratory infectious diseases did not result in a significant overall treatment delay. These virus infections did neither cause an increase in "STEMI incidence" section, nor did they raise the STEMI mortality. Nonetheless, both diseases showed a trend towards an increased patient-related delay. Educational programs should address the observed patient-related delay and emphasize the need for immediate contact of the medical system. Additionally, the results demonstrated a significantly prolonged door-to-balloon time during the COVID-19 pandemic. Adequate strategies to minimize this delay and improve health care service balancing prompt treatment and infection protection are needed.