Monthly variations in acute coronary syndromes outcomes during the first year of the COVID-19 pandemic

Abstract Background Cardiovascular disease, particularly acute coronary syndromes (ACS), is the leading cause of death in the United States. Minor fluctuations in hospital admissions for different conditions, including ACS, can be seen throughout the year. This study focuses on the impact of admission month on outcomes of acute coronary syndromes during the first year of the COVID-19 pandemic. Methods This was a retrospective observational study of patients hospitalized with ACS from the National Inpatient Sample, during the years 2020 (n = 779,895) and 2019 (n = 935,975). We compared the monthly outcomes for every month to the outcomes for the month of January of that same year. The primary outcomes of interest were in-hospital mortality and time from admission to PCI Results Inpatient mortality for patients admitted with STEMI was significantly higher for admissions in the months of April, October and December of 2020 than January of that same year. For patients admitted with NSTEMI or UA, inpatient mortality was higher for admissions in April and December 2020 when compared to admissions in January 2020. Inpatient mortality for patients with STEMI, NSTEMI and UA was not different based on admission month in the year 2019. Conclusion The month of admission significantly affected outcomes for patients admitted with ACS during the COVID-19 pandemic, with higher inpatient mortality and longer time from admission to PCI for certain months in 2020. Further studies should investigate disparities in monthly ACS outcomes for the year 2021 and onward, now that COVID-19 infections have been steadily declining.


Introduction
Cardiovascular disease (CVD), particularly acute coronary syndrome (ACS), is the leading cause of death in United States.Advancements in reperfusion modalities, particularly percutaneous coronary intervention (PCI), and the commitment of health care facilities to provide timely intervention over the past few decades have resulted in a decrease in the overall mortality related to acute coronary syndrome. 1Providing timely intervention for ACS requires health care facilities to provide resources, not limited to personnel, equipment, and facility space.Over the years, health care facilities have worked on developing strategies in order to minimize the door-to-balloon time for patients presenting with ACS, and especially, ST-elevation myocardial infarction.
While some minor fluctuations in hospital admissions can be seen throughout different times of the year, for example with some months experiencing upticks of admissions due to certain conditions, such as the seasonal flu, in our modern history, there has never been a strain on our healthcare system at the level of a global pandemic that was the COVID-19 virus.On January 21, 2020 the CDC confirmed the first case of COVID-19 infection in the United States. 2 What followed was millions of COVID-19 infections all across the US, with hundreds of thousands of hospital admissions primarily for COVID-19 and a burden on health care facilities unlike any other.
In this study, we analyze if outcomes for ACS were impacted based on the month of admission during the first year of the COVID-19 pandemic.To our knowledge, there has not been a study to date that has analyzed the impact on admission month on the outcomes of acute coronary syndromes.

Study design
This was a retrospective observational study of adult patients hospitalized with acute coronary syndromes, including STEMI, NSTEMI and UA, during the year 2020.We completed a similar analysis for patients hospitalized with ACS in the year 2019.We compared the monthly outcomes for every month to the outcomes for the month of January of that same year.

Data source and sample
Analysis was conducted using the 2020 and 2019 National Inpatient Sample (NIS) database of the Healthcare Utilization Project (HCUP).NIS is one of the largest available databases in United States and consists of discharge data from a 20% stratified sample of US hospitalizations.Our hospitalization sample was selected using the International Classification of Diseases, tenth Revision, Clinical Modification (ICD10-CM) coding system.We identified ICD10-CM codes for primary diagnosis of STEMI (I21, I211, I212), NSTEMI (I214, I222), UA (I25, I200, I222).We excluded patients with ICD10-CM diagnosis of shock (R571, R578, R579, R6521) and patients requiring mechanical support (T884XXD, T88DXXS, T884XXA, T884,5A1522G; supplemental file).

Variables of interest and study outcomes
The primary outcomes of interest for our study were in-hospital mortality and time from admission to PCI.Our secondary outcomes were length of stay and total hospital charges.For each admission month, we compared the outcomes of interest during that month to the month of January of that same year.The independent variable of interest is the admission month.Potential confounders that were adjusted for included age in years, gender, race, hospital bed size and patient comorbidities measured using Charlson Comorbidity Index (CCI).

Statistical analysis
Statistical analysis was performed using STATA version 17.0 software 3 .Data was expressed as percentages for categorical variables and mean ± SD for continuous variables.Continuous variables were compared using Student's t-test and categorical variables were compared using chi-square test.Univariate regression analysis was used to calculate unadjusted odds ratios for the study outcomes.Multivariate linear (for continuous outcomes) and logistic (for binary outcomes) regression analysis was used to calculate the adjusted odds ratios (aOR).The models were built by including the significant variables that were associated with the outcome of interest.Those variables are summarized in Table 1.All p values were 2-sided, with 0.05 as a threshold for statistical significance.

Population demographics
The total number of patients that fit our inclusion criteria with acute coronary syndrome in the year 2020 was 779,895, with 65% being male and the average age being 66.3 years old.The majority of these patients were Caucasian at 74%, and African Americans made up 10.5%.Most patients had a comorbidity with the highest comorbidity being hypertension, 42.2% complicated and 43.3% uncomplicated hypertension.Diabetes mellitus was another common comorbidity in the patient population with 27.7% of patients having diabetes mellitus with chronic complications and 15.3% without chronic complications.For the year 2019, similar data was extracted from NIS, with a total of 935,975 patients with a primary diagnosis of ACS, 65% males and average age of 66.7 years old.More detail pertaining to patient demographics, Charlson comorbidity index and comorbidities can be found in Table 1.

Inpatient mortality
In the data analyzed for the year 2020, we found that inpatient mortality for patients with STEMI was significantly higher in April (aOR 1.39, CI 1.39-1.69,p 0.001), October (aOR 1.25, CI 1.04-1.51,p 0.02), and December (aOR 1.33, CI 1.11-1.59,p 0.002) than in January for that same year.Figure 1 showed Forest plot predictors of inpatient mortality in STEMI patients.
For 2019 monthly inpatient mortality rates for patients admitted with STEMI, we find that mortality was not higher in any of the months when compared to January 2019.Comparing April, October and December 2019 to January 2019, we find that mortality was not increased in these months the same way it was in 2020 (2019: April aOR 0.74 CI 0.61-0.89,p 0.002; October aOR 0.70, CI 0.58-0.84,p 0.000; December aOR 0.79, CI 0.66-0.94,p 0.01).Mortality rates for the remainder of the months with corresponding p-values can be found in Table 2. Figure 2 showed Predictive margins of admission month on inpatient mortality.
Inpatient mortality for patients admitted with NSTEMI or unstable angina in 2020 was significantly higher in April (aOR 1.37, CI 1.08-1.73,p 0.01) and December (aOR 1.33, CI 1.08-1.64,p 0.01) when compared to January of the same year.Supplemental Figure 1 showed Forest plot predictors of inpatient mortality in UA/NSTEMI.
In contrast, for the year 2019, the inpatient mortality rate for patients admitted with NSTEMI and UA was not higher for any month from February through December when compared to January of 2019.Inpatient mortality odds ratio for each month as compared to January can be found in Table 2.

Time from admission to PCI
For patients admitted with STEMI, when average time from admission to PCI was calculated for each month, we found that on average PCI was not delayed in any of the months as compared to January 2020.For most months, excluding November and December, the time from admission to PCI was actually significantly shorter.For the month of February, the month before the pandemic, results showed no difference in time to PCI when compared to January 2020.Supplemental Figure 2 shows Forest plot predictors of time from admission to PCI in STEMI patients.
Similarly, there was no difference in time from admission to PCI for all months in the year 2019 when compared to January 2019 for patients admitted with STEMI.
For patients admitted with NSTEMI or unstable angina, there was no delay in time from admission to PCI for each month of 2020 when compared to January 2020.The year 2019 showed no difference when comparing each month's average time to PCI to January 2019.Data for time to PCI for patients admitted with NSTEMI and UA can be found in Table 3. Supplemental Figure 3 shows Forest plot predictors of time from admission to PCI in UA/NSTEMI patients.

Length of stay
The average monthly length of stay for patients admitted with STEMI in 2020 as compared to January 2020 showed that there was not an increase in hospital length of stay for all months in 2020 when compared to January 2020.For the year 2019, results show that monthly length of stay was significantly shorter for nearly all months as compared to January 2019 for patients admitted with STEMI.
For NSTEMI and unstable angina, there was not an increase in hospital length of stay for all months of 2020 when compared to January 2020.Similarly, in 2019, results showed significantly shorter length of stays for nearly all months as compared to January 2019 for patients admitted with unstable angina or NSTEMI.Individual averages for all months of 2019 and 2020 can be found in: Supplemental Table 1.

Hospital cost
The average hospital charges were higher for most months in 2020 when compared to January 2020 for patients admitted with STEMI.For 2019, there was no difference in cost for most months when compared to January 2019 for patients admitted with STEMI.For the months that did show a difference, there was actually a decrease in cost as compared to January 2019.
For NSTEMI and unstable angina admissions in 2020, the monthly average hospital charges were either higher or not significantly different for nearly every month as compared to January of 2020.For NSTEMI and unstable angina admissions in 2019, there was no difference in average monthly hospital charges as compared to January 2019.The average monthly cost for each month as compared to January of 2020 or 2019 can be found in: Supplemental Table 2.

Discussion
In this nationwide observational study, we analyze the impact of month of admission on inpatient mortality, time from admission to PCI, length of hospital stay and hospital charges for patients admitted for acute coronary syndromes during the COVID-19 pandemic in the year 2020.This is the largest analysis to date that evaluates month of admission as an independent variable on outcomes of ACS.It is also the first to analyze and attempt to correlate between the onset of the COVID-19 pandemic and outcomes of acute coronary syndromes by admission month.Data analysis from our study showed first that inpatient mortality for patients admitted with STEMI was significantly higher for admissions in the months of April, October and December of 2020 than January of that same year.For the year 2019, inpatient mortality was not higher for any of the months when compared to admissions in January 2019 for patients admitted with STEMI.For patients admitted with NSTEMI or UA, inpatient mortality was higher for admissions in April and December 2020 when compared to admissions in January 2020.Inpatient mortality for patients with NSTEMI and UA was not different based on admission month in the year 2019.Second, time from admission to PCI was shorter for most of the months of 2020 when compared to January 2020 for patients admitted with STEMI.For the year 2019, time to admission to PCI was not different for any month when compared to January 2019 for patients admitted with STEMI.For patients admitted with NSTEMI and UA in 2020 and in 2019, there was no delay in time from admission to PCI based on admission month.Third, for acute coronary syndrome admissions in 2020 and in 2019, there was no increase in length of stay based on month of admission.Finally, when analyzing hospital charges for STEMI, NSTEMI and UA, there was significantly higher charges for October, November, and December of 2020 when compared to January 2020, but the same increase in charge was not present in 2019.
With rapid and enormous pressure demands, and without anticipation or warning, on an arguably already understaffed, burdened health care system, it is not surprising that studies are observing significant changes in the morbidity and mortality outcomes for a wide variety of different conditions during the pandemic.For instance, a study by Janus et el that explored over 3 million deaths due to cardiovascular disease in the US from 2018-2021 found a 6.7% increase in mortality due to cardiovascular disease, and 2.5% mortality increase from myocardial infarction in particular, during the COVID-19 pandemic. 4ther studies have attempted to explore the impact of COVID-19 on existing comorbidities and on acute coronary syndromes.One study in Geneva, Switzerland revealed a longer delay from symptom onset to first medical contact in patients with STEMI after the outbreak of COVID-19.The authors of the study attributed this delay partially to the widespread fear of infection in patients who had symptoms of STEMI but preferred to avoid contracting the viral infection. 5nother study in Japan analyzed the number of ACS cases per week during and after the pandemic and found a significant decline in hospitalizations for acute coronary syndromes during the pandemic. 6A few studies have explored door-toballoon time during the pandemic.Door-to-balloon time was significantly longer post-COVID-19 pandemic for patients with STEMI in one study that included 121 patients in the post-COVID group and 209 patients in the pre-COVID group at one medical center in Japan 7 .Confounders are likely at play with the geographical limitation of this study and with its limited power.This is unlike our study which included over a million patients across the US and showed no delay in time to PCI.
There are studies that have evaluated cardiovascular hospital admissions during the waves of the COVID-19 pandemic.A study in North-East Spain found an increase in hospital admissions for cardiovascular emergencies during the first wave of the pandemic but decreased admissions during the second wave 8 .Our study is the first to analyze developments in COVID-19 pandemic and correlate that with outcomes of acute coronary syndromes by admission month.
A few observations can be made when reviewing the timeline of COVID-19 developments in the year 2020 in: Supplemental Table 3 COVID-19 was declared a pandemic in March 2020.Of note, prior to March, there was no significant increases in mortality based on month of admission for patients admitted with ACS.In April, there was an increase in inpatient mortality for patients admitted with STEMI and also for patients admitted with NSTEMI or UA.There was not an increase in mortality for patients admitted with ACS in April 2019, prior to the pandemic.Towards the end of the year 2020, the CDC reported rising weekly cases of COVID-19 In the United States, with close to half a million new cases per week in October and surpassing 1 million new cases of COVID-19 per week in November and December 2020. 9Notably, there were increases in inpatient mortality for patients admitted with STEMI towards the end of the year, in October and December 2020.In 2019, there were no increases in mortality based on month of admission for ACS patients, even when analyzing the latter months of the year.Average monthly hospital charges were higher in October through December for patients admitted with STEMI, NSTEMI or UA in 2020, and similar increases were not found during those months in 2019.When contrasting monthly admissions of ACS in 2019 and those of 2020, there are notable correlations between worsening developments in the COVID-19 pandemic and outcomes for patients admitted with acute coronary syndromes.

Limitations
One of the limitations of this study is the possibility that our data does not include the entire population that otherwise fits the inclusion criteria.Coding and documentation errors could have excluded patients that should have been included in our data analysis from being selected for in the database we used.We anticipate that the actual number of patients who were wrongfully included or wrongfully excluded to be small in number and unlikely to change the significance of our results.Another limitation of our study is that not all patient-specific characteristics are accounted for, such as certain past medical history, individual lab results, oncological diagnosis, imaging results, medications, social history, that could potentially affect the outcomes of our study.We did attempt to minimize potential confounders as described in the methods section.Also, while inpatient mortality was analyzed, we did not include mortality after discharge from the hospital, or mortality related to ACS cases that never made it to the hospital.Lastly, a limitation of this study, and any observation study, is selection bias, and unmeasured bias.We completed a multivariable analysis to reduce allocation bias.

Conclusion
The month of admission significantly affected outcomes for patients admitted with acute coronary syndrome in the year 2020 during the COVID-19 pandemic, with higher inpatient mortality for certain months, for patients admitted with STEMI, NSTEMI, and UA for the year 2020 but not before the pandemic, in the year 2019.These results suggest that healthcare facilities are capable of further reducing time from admission to PCI and in turn, decreasing the morbidity and mortality associated with acute coronary syndromes even under the stress of a pandemic.These findings can be correlated with COVID-19 pandemic developments with the CDC reporting rising weekly cases of COVID-19 in the United States during the same time period, close to half a million new cases per week in October and surpassing 1 million new cases of COVID-19 per week in November and December 2020.Further studies should investigate disparities in monthly ACS outcomes for the year 2021 and onward, now that COVID-19 infections have been steadily declining.Additional studies should focus on predicting how future stressful events on our healthcare system may affect outcomes for patients admitted with acute coronary syndromes and, more importantly, on how to prepare for and mitigate the stress inciting from such events to properly and effectively manage and treat patients presenting with acute coronary syndromes.

Table 1 .
Average monthly inpatient length of stays for STEMI and NSTEMI/UA admissions.
Figure 1.Forest plot predictors of inpatient mortality in STEMI patients.

Table 2 .
Average monthly hospital charges for STEMI and NSTEMI/UA hospital admission.