Skip main navigation

Causes and Predictors of 30‐Day Readmission in Patients With COVID‐19 and ST‐Segment–Elevation Myocardial Infarction in the United States: A Nationwide Readmission Database Analysis

Originally publishedhttps://doi.org/10.1161/JAHA.123.029738Journal of the American Heart Association. 2023;12:e029738

Abstract

Background

Rates, causes, and predictors of readmission in patients with ST‐segment–elevation myocardial infarction (STEMI) during COVID‐19 pandemic are unknown.

Methods and Results

All hospitalizations for STEMI were selected from the US Nationwide Readmissions Database 2020 and were stratified by the presence of COVID‐19. Primary outcome was 30‐day readmission. Multivariable hierarchical generalized logistic regression analysis was performed to compare 30‐day readmission between patients with STEMI with and without COVID‐19 and to identify the predictors of 30‐day readmissions in patients with STEMI and COVID‐19. The rate of 30‐day all‐cause readmission was 11.4% in patients with STEMI who had COVID‐19 and 10.6% in those without COVID‐19, with the adjusted odds ratio (OR) not being significantly different between the two groups (OR, 0.88 [95% CI, 0.73–1.07], P=0.200). Of all 30‐day readmissions in patients with STEMI and COVID‐19, 41% were for cardiac causes. Among the cardiac causes, 56% were secondary to acute coronary syndrome, while among the noncardiac causes, infections were the most prevalent. Among the causes of 30‐day readmissions, infectious causes were significantly higher for patients with STEMI who had COVID‐19 compared with those without COVID‐19 (29.9% versus 11.3%, P=0.001). In a multivariable model, congestive heart failure, chronic kidney disease, low median household income, and length of stay ≥5 days were found to be associated with an increased risk of 30‐day readmission.

Conclusions

Post‐STEMI, 30‐day readmission rates were similar between patients with and without COVID‐19. Cardiac causes were the most common causes for 30‐day readmissions, and infections were the most prevalent noncardiac causes.

Nonstandard Abbreviations and Acronyms

AIC

Akaike Information Criterion

ECI

Elixhauser Comorbidity Index

HCUP

Healthcare Cost and Utilization Project

MACE

major adverse cardiovascular events

NRD

Nationwide Readmission Database

Clinical Perspective

What Is New?

  • This is the first nationwide, all‐payer, real‐world study reporting 30‐day readmission outcomes in patients with ST‐segment–elevation myocardial infarction (STEMI) during COVID‐19 era, where we found no difference in 30‐day all‐cause readmissions after stratification for COVID‐19.

  • Cardiac causes (acute coronary syndrome, arrhythmias, and heart failure) were the most common reasons for 30‐day readmission in patients with STEMI with and without COVID‐19 in 2020.

  • Prior heart failure, chronic kidney disease, median household income in the lowest quartile, and a length of stay ≥5 days were the predictors of 30‐day readmission in patients with STEMI and COVID‐19.

What Are the Clinical Implications?

  • This study helps physicians to understand the impact of COVID‐19 on readmissions after STEMI and to identify the common causes of 30‐day readmissions after STEMI in patients positive for COVID‐19, which can help in addressing these factors during the discharge process.

  • Further research is needed to examine the preventability of costly 30‐day readmissions in patients with STEMI and COVID‐19.

Severe acute respiratory syndrome coronavirus 2, the pathogen responsible for COVID‐19, continues to shape the landscape of health care. The confluence of COVID‐19 and cardiovascular disease represents a milieu with a substantially elevated risk for adverse outcomes, with some estimates demonstrating a 4‐fold elevation in the risk of mortality.1, 2 While the most clinically relevant manifestations of COVID‐19 are often pulmonary, primary cardiac injury in the form of myocarditis and acute coronary syndrome (ACS) are well described.2, 3 The prevalence of myocardial injury in patients with COVID‐19 has been reported with variability, in the range of 15.8% to 51% depending on the cohort of patients being investigated, while the prevalence of COVID‐19 in patients with ST‐segment–elevation myocardial infarction (STEMI) has been estimated at 1.38%.4 Despite this variability, the important influence of myocardial injury on patient outcomes in those with COVID‐19 is clear.

Previous retrospective analyses of patients with STEMI and COVID‐19 have demonstrated an increased risk of in‐hospital mortality when compared with patients with STEMI without COVID‐19.5 Although there are likely numerous culprits responsible for the potential interaction of COVID‐19 and STEMI on clinical outcomes, the prothrombotic state conferred by COVID‐19, the reduced probability of undergoing an invasive management strategy such as percutaneous coronary interventions, and patient complexity are likely contributors.6 Although the impact of COVID‐19 on in‐hospital outcomes of patients with STEMI have been described to some extent, postdischarge readmission outcomes are less well described.7 It is unclear whether the increased risk for adverse cardiovascular outcomes associated with COVID‐19 extends beyond the immediate in‐hospital period.

Given the associations between COVID‐19, myocardial injury, and adverse in‐hospital outcomes in patients with STEMI, we performed a nationwide retrospective analysis to better understand the impact of COVID‐19 on 30‐day readmission outcomes in patients with STEMI, the causes of 30‐day readmissions in patients with STEMI with and without COVID‐19, and predictors of 30‐day readmission during the COVID‐19 pandemic.

METHODS

Data Source

We conducted a retrospective cohort study using the Nationwide Readmission Database (NRD) to derive patient‐relevant information from January 1, 2020, to November 30, 2020. The NRD was developed by the Agency for Healthcare Research and Quality as part of the Healthcare Cost and Utilization Project (HCUP). It is the largest all‐payer, publicly available database in the United States. It utilizes a deidentified unique number for tracking each patient's admission, discharge, readmissions, and death across hospitals within a calendar year. The NRD includes the data derived from US hospitals in 31 geographically dispersed states, which encompasses ≈16 million unweighted and 32 million weighted discharges in 2020 in the United States. It represents ≈62.2% of the total US resident population and 60.8% of all US hospitalizations. Additional details on the NRD are available online and in previous studies.3, 8, 9, 10, 11 Institutional review board approval and informed consent were not required for the current study because all data collection was derived from a deidentified administrative database. The NRD is publicly available and can be procured from the HCUP website. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.12

Study Population

We queried the NRD with the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD‐10‐CM) codes to identify all adults (n=189 325) who had a STEMI hospitalization (primary or secondary diagnosis, dx‐1 to dx‐40 in the NRD). Subsequently, we identified patients positive for COVID‐19 (n=586 393) using appropriate and previously validated ICD‐10‐CM codes (secondary diagnosis, dx‐2 to dx‐40 in NRD). The codes used to identify the cohorts in this study are summarized in Table S1 and have been used in previous studies.4, 7 All patients were divided into 2 groups according to the presence or absence of COVID‐19. Only the initial admission for calendar year per patient was included for analysis as an index admission. The flow diagram for patient selection is summarized in Figure 1. Patients discharged from January through November were included in the study to allow for completeness of data on 30 days of follow‐up after discharge. Patients with missing values of death and length of stay (LOS) (n=65, 0.03%) were excluded from the analysis and therefore were not imputed. Patient‐ and hospital‐level variables were included as baseline characteristics. NRD variables were used to identify age; sex; admission day; median household income quartiles; primary payer; hospital teaching status, location, and bed size; and discharge disposition. The comorbidities and overall severity of comorbidities were defined by using the Elixhauser Comorbidity Index (ECI) score.13 In‐hospital complications were identified by using ICD‐10‐CM procedure codes (in either primary or secondary procedure field) and diagnosis codes of acute conditions (in secondary diagnosis field), which distinguished them from comorbidities (Table S1). LOS was stratified to ≤2 days, 3 to 4 days, and ≥5 days.

Figure 1. Patient selection flow diagram.

ICD‐10‐CM indicates International Classification of Diseases, Tenth Revision, Clinical Modification; NRD, Nationwide Readmission Database; and STEMI, ST‐segment–elevation myocardial infarction.

Outcomes

The primary outcome of interest was 30‐day all‐cause readmission rate among patients with STEMI stratified by the presence of COVID‐19. According to the methodology described by the HCUP, time to readmission was computed as the number of days from discharge date of index admission to readmission date.3 Only the first readmission within 30 days after discharge was included, and transfer to another hospital was not counted as a readmission. Secondary outcomes were 30‐day major adverse cardiovascular events (MACE; which are a composite of myocardial infarction, atrial fibrillation, stroke, heart failure, and death) and 30‐day revascularization rate (composite of thrombolytic therapy, percutaneous coronary intervention, and coronary artery bypass grafting). The primary cause of 30‐day readmission was identified based on the first diagnosis field of each readmission record and was categorized into cardiac and noncardiac causes and their further subcategories.

Statistical Analysis

Discharge weight provided by the NRD was used for all analyses to obtain national estimates. For descriptive analyses, we compared baseline patient and hospital characteristics of patients with STEMI stratified by the occurrence of COVID‐19. Continuous variables are presented as means and SDs or medians and interquartile ranges (IQRs), and categorical variables are presented as percentages. For comparison, the χ2 test was used for categorical variables and either the Wilcoxon rank sum test or survey‐specific linear regression was used for continuous variables. We constructed multivariable hierarchical logistic regression models with multiple patient and hospital covariates. Factors adjusted in the model were selected based on both forward and backward selection models and based on clinical significance. The following variables were adjusted in the multivariable model for 30‐day outcomes: age, sex, heart failure, chronic kidney disease (CKD), chronic pulmonary disease, ECI, primary payer, and median household income. Hospital identification was included as a random effect and survey statistics were used to account for within‐hospital correlation and between‐hospital variation. Patients who died at the end of index admission were excluded from the regression models. The appropriateness of the multivariable hierarchical generalized logistic models for 30‐day readmission predictors and for 30‐day outcomes was assessed by C‐statistic, which was >0.75 for all models. Akaike Information Criterion (AIC) was used to choose the best‐fit regression model. The AICs for 30‐day readmission outcomes models were <1000, and the AIC for the 30‐day all‐cause readmission predictors model was 831.

Additional subgroup analysis was performed according to patient's age (<65 versus ≥65 years), sex (men versus women), location at the time of the STEMI (in‐hospital versus out of hospital), need for mechanical ventilation (no mechanical ventilation versus mechanical ventilation), heart failure (absence versus presence), CKD (absence versus presence), and chronic pulmonary disease (absence versus presence) to compare the risk of 30‐day all‐cause readmission in patients with STEMI with versus without COVID‐19.

To evaluate the robustness of our findings, we conducted a sensitivity analysis to determine the validity of the study. Sensitivity analysis was performed using 1:1 propensity score matching of patients with STEMI with and without COVID‐19. Propensity scores were generated using the following covariates: age, sex, patients' comorbidities (as mentioned in Table 1), ECI, primary payer, median household income, hospital bed size, hospital teaching status, and hospital location. Patients with similar propensity scores in 2 groups were matched using a 1‐to‐1 scheme without replacement using a greedy method. Patients without matched observations were excluded.

Table 1. Baseline Characteristics of Patients With STEMI Without COVID‐19 Versus Patients With STEMI With COVID‐19

Baseline characteristics and comorbiditiesSTEMI without COVID‐19STEMI with COVID‐19P value
n=185 700n=3560
Age, mean (SD), y64.27 (13.18)65.83 (13.80)<0.001
Age group, y<0.001
<5013.3%12.0%
50–6438.1%33.2%
65–7934.7%37.3%
≥8013.9%17.5%
Women31.4%32.4%0.354
Comorbidities
Hypertension75.1%73.7%0.199
Diabetes34.2%44.7%<0.001
Hyperlipidemia65.9%56.2%<0.001
Heart failure42.6%41.4%0.299
Pulmonary circulation disorder4.6%5.8%0.015
Atrial fibrillation7.8%8.1%0.561
Known CAD*38.2%29.9%<0.001
Ischemic cardiomyopathy20.8%14.4%<0.001
Prior myocardial infarction13.9%10.5%<0.001
Prior PCI12.2%10.3%0.023
Prior CABG4.7%4.1%0.272
Family history of CAD14.4%7.4%<0.001
Fluid/electrolyte disorders30.3%55.5%<0.001
History of pacemaker/ICD2.0%1.5%0.177
Valvular heart disease/prosthetic valve10.2%7.1%<0.001
Prior stroke/TIA7.6%9.2%0.019
Peripheral vascular disease8.6%6.7%0.006
Chronic pulmonary disease17.0%15.5%0.118
CKD16.5%23.0%<0.001
Chronic liver disease6.6%12.0%<0.001
Nonadherence to medications5.6%4.4%0.023
Anemia2.8%3.7%0.017
Coagulopathy7.3%17.1%<0.001
Obesity20.2%21.5%0.175
Smoking26.9%20.8%<0.001
ECI, median (IQR)3 (2–5)4 (3–6)<0.001
Treatment during index STEMI admission
Revascularization
Thrombolytic therapy1.00%3.4%<0.001
All PCI75.5%51.5%<0.001
CABG5.2%2.1%<0.001
Admission day0.962
Weekday72.1%72.1%
Weekend27.9%27.9%
Primary payer<0.001
Medicare48.6%52.3%
Medicaid10.6%12.0%
Private insurance30.6%26.5%
Self‐pay/no charge/other10.2%9.2%
Median household income quartile<0.001
0–25th28.5%33.3%
26–50th30.1%29.3%
51–75th23.5%22.4%
76–100th18.0%15.0%
Hospital bed size§0.941
Small15.0%15.1%
Medium27.5%27.0%
Large57.5%57.8%
Hospital teaching statusǁ0.955
Nonteaching24.1%24.0%
Teaching75.9%76.0%
Hospital location0.664
Nonmetropolitan area6.3%6.6%
Metropolitan area93.7%93.4%
LOS category, d<0.001
≤249.7%31.2%
3–423.0%17.0%
≥527.3%51.9%
Discharge disposition<0.001
Discharged home with self‐care77.0%65.9%
Short‐term hospital1.7%3.4%
Discharge to extended‐care facility7.7%15.3%
Home health care12.3%13.7%
Against medical advice1.3%1.6%
Discharge alive, destination unknown0.04%0.18%

CKD indicates chronic kidney disease; ECI, Elixhauser Comorbidity Index; ICD, implantable cardioverter‐defibrillator; IQR, interquartile range; LOS, length of stay; STEMI, ST‐segment–elevation myocardial infarction; and TIA, transient ischemic attack.

*Known coronary artery disease (CAD) is a composite of ischemic cardiomyopathy, prior myocardial infarction, prior percutaneous coronary intervention (PCI), and prior coronary artery bypass grafting (CABG).

Medicare includes both fee‐for‐service and managed care Medicare patients, Medicaid includes both fee‐for‐service and managed care Medicaid patients, and private insurance includes Blue Cross, commercial carriers, and private health maintenance organizations and preferred provider organizations (https://www.hcup‐us.ahrq.gov/db/vars/pay1/nrdnote.jsp).

Represents a quartile classification of the estimated median household income of residents within the patient's zip code (https://www.hcup‐us.ahrq.gov/db/vars/zipinc_qrtl/nrdnote.jsp).

§The bed size cutoff points were divided into small, medium, and large such that approximately one‐third of the hospitals in a given region, location, and teaching status combination would fall within each bed size category (https://www.hcup‐us.ahrq.gov/db/vars/hosp_bedsize/nrdnote.jsp).

ǁA hospital is considered to be a teaching hospital if it has an American Medical Association–approved residency program (https://www.hcup‐us.ahrq.gov/db/vars/hosp_ur_teach/nrdnote.jsp).

Extended‐care facility includes skilled nursing facilities, intermediate care facilities, and another type of facility.

Odds ratios (ORs) with 95% CIs were used as a measure of effect size. All P values were 2‐sided, with P<0.05 considered statistically significant. All statistical analyses were conducted using appropriate weighting, stratifying, and clustering samples using the svy package of STATA version 17 (StataCorp LLC).

RESULTS

Study Population and Baseline Characteristics

We identified 189 325 index STEMI hospitalizations from January 1, 2020, to November 30, 2020. Of these, 3560 (1.9%) patients were diagnosed with COVID‐19, and 185 700 (98.1%) patients were not. Among all patients with STEMI who were discharged alive, the 30‐day readmission rate was 10.6% (n=17 640). Baseline characteristics of the study population are presented in Table 1. The mean age was 64.30 years (SD, 13.2 years), and most patients were men (68.6%). Patients with STEMI with COVID‐19 were older (mean age, 65.8 versus 64.3 years) and had a statistically significant higher burden of comorbidities such as diabetes (44.7% versus 34.2%), prior stroke/transient ischemic attack (9.2% versus 7.6%), CKD (23.0% versus 16.5%), chronic liver disease (12.0% versus 6.6%), coagulopathy (17.1% versus 7.3%), and median ECI (4 versus 3) compared with those without COVID‐19. Patients with STEMI who had COVID‐19, however, had statistically significant lower rates of smoking (20.8% versus 26.9%), known coronary artery disease (CAD) (29.9% versus 38.2%), valvular heart disease (7.1% versus 10.2%), and family history of CAD (7.4% versus 14.4%) compared with those without COVID‐19. Patients with STEMI and COVID‐19 had significantly lower rates of revascularization with percutaneous coronary intervention (51.5% versus 75.5%) or coronary artery bypass grafting (2.1% versus 5.2%), longer LOS ≥5 days (51.9% versus 27.8%), and higher rates of discharge to an extended‐care facility (15.3% versus 7.7%) compared with those without COVID‐19.

Comparison of 30‐Day Outcomes in Patients With STEMI Without COVID‐19 Versus Patients With STEMI With COVID‐19

The 30‐day all‐cause readmission rate was 10.6% in patients without COVID‐19 and 11.4% in patients with COVID‐19 (P=0.362), 30‐day MACE rates were 4.6% and 4.2% (P=0.463), and 30‐day revascularization rates were 1.7% and 2.0%, respectively (P=0.424). The adjusted risk for 30‐day all‐cause readmission (adjusted OR, 0.88 [95% CI, 0.73–1.07], P=0.200), for 30‐day MACE (adjusted OR, 0.81 [95% CI, 0.60–1.07], P=0.138), and for 30‐day revascularization (adjusted OR, 1.19 [95% CI, 0.80–1.76], P=0.387) were not significantly different in patients with STEMI without COVID‐19 compared with patients with STEMI with COVID‐19 (Table 2).

Table 2. Comparison of 30‐Day Outcomes in Patients With STEMI Without Versus With COVID‐19

Discharged aliveSTEMI without COVID‐19 (reference)STEMI with COVID‐19P valueUnivariable regression modelMultivariable regression model
n=164 700n=2175OR95% CIP valueAdjusted OR95% CIP value
30‐d all‐cause readmissions10.6%11.4%0.3621.090.911.310.3620.880.731.070.200
30‐d MACE*4.6%4.2%0.4630.900.681.190.4630.810.601.070.138
Myocardial infarction1.8%2.1%0.429
Atrial fibrillation0.25%0.56%0.084
Stroke0.29%0.16%0.368
Heart failure1.90%0.53%0.001
Death0.60%0.85%0.258
30‐d revascularization1.7%2.0%0.4241.170.791.740.4251.190.801.760.387

OR indicates odds ratio; and STEMI, ST‐segment–elevation myocardial infarction.

*30‐day major adverse cardiovascular events (MACE) is a composite of myocardial infarction, atrial fibrillation, stroke, heart failure, and death.

Revascularization is a composite of thrombolytic therapy, all percutaneous coronary interventions, and coronary artery bypass grafting.

Causes of 30‐Day Readmission After STEMI in Patients With and Without COVID‐19

Figure 2A and 2B demonstrates the causes of 30‐day readmissions for patients with STEMI and a co‐diagnosis of COVID‐19. A total of 41% of 30‐day readmissions were attributable to cardiac causes, and 59% of 30‐day readmissions were attributable to noncardiac causes. Among the cardiac causes of 30‐day readmissions, 56% were secondary to ACS, 20% were secondary to arrhythmias, and 13% were secondary to heart failure. Among noncardiac causes of 30‐day readmissions, infectious causes, followed by respiratory and gastrointestinal/hepatic/pancreatic causes were most prevalent (Table S2).

Figure 2. Causes of 30‐day readmissions after index admission for concurrent ST‐segment–elevation myocardial infarction (STEMI) and COVID‐19.

A, All causes for 30‐day readmission. B, Cardiovascular causes for 30‐day readmission. Acute coronary syndrome followed by arrhythmias and heart failure were most prevalent among cardiac causes. Bleeding in gastrointestinal, COVID‐19 in infectious, pulmonary embolism in respiratory, stroke in neurological, and diabetes in endocrine causes were most prevalent. Please refer to Table S2 for a detailed breakdown.

Figure 3 demonstrates the causes of 30‐day readmission for patients with STEMI after stratification for COVID‐19. Among patients without COVID‐19, 53% of 30‐day readmissions were attributed to cardiac causes, and 47% of 30‐day readmissions were attributed to noncardiac causes (Table S3). Among the cardiac causes of 30‐day readmissions, 42% were secondary to ACS, 11% were secondary to arrhythmias, and 37% were secondary to heart failure. Among noncardiac causes of 30‐day readmissions, infectious causes, followed by gastrointestinal/hepatic/pancreatic and respiratory causes were most prevalent.

Figure 3. Comparison of causes of 30‐day readmissions after ST‐segment–elevation myocardial infarction (STEMI) between patients with and without COVID‐19.

Infectious causes of 30‐day readmissions were significantly higher in patients with STEMI with COVID‐19 as compared with those without COVID‐19 (P=0.001). For patients with STEMI without COVID‐19, acute coronary syndrome followed by heart failure, and arrhythmias were most prevalent among cardiac causes, and bleeding in gastrointestinal, sepsis from unspecified organism in infectious, pneumonia in respiratory, stroke in neurological, acute kidney injury in kidney‐related, anemia in hematological, and diabetes in endocrine causes were most prevalent. Please refer to Table S3 for detailed categorization.

While comparing the causes of 30‐day readmissions, infectious causes were significantly higher for patients with STEMI and COVID‐19 as compared with those without COVID‐19 (29.9% versus 11.3%, P=0.001).

Timing of 30‐Day Readmission After STEMI

Figure 4A and 4B demonstrates the timing of readmission after STEMI for subgroups stratified by COVID‐19. Among patients with COVID‐19, 53.6% and 75.4% were readmitted within 7 and 14 days, respectively. Among patients without COVID‐19, 47.1% and 70.6% were readmitted within 7 and 14 days, respectively. Patients with STEMI with COVID‐19 had significantly shorter median time to readmission as compared with those without COVID‐19 (median, 7 [IQR, 3–14 days]) versus 8 days (IQR, 4–16 days), P=0.044.

Figure 4. Timing of 30‐day readmission by postdischarge day after index admission for ST‐segment–elevation myocardial infarction (STEMI).

A, Patients with STEMI with COVID‐19. B, Patients with STEMI without COVID‐19. Among patients with COVID‐19, 53.6% and 75.4% were readmitted within 7 and 14 days, respectively. Among patients without COVID‐19, 47.1% and 70.6% were readmitted within 7 and 14 days, respectively. Patients with STEMI with COVID‐19 had significantly shorter median time to readmission compared with those without COVID‐19 (median, 7 days [interquartile range, 3–14 days] vs 8 [interquartile range, 4–16 days], P=0.044).

Predictors of 30‐Day Readmission After STEMI in Patients With COVID‐19

Table 3 lists univariable and multivariable predictors of 30‐day readmission after initial hospitalization with STEMI in patients with COVID‐19. After adjusting for clinical and hospital characteristics in a multivariable model, heart failure (adjusted OR, 1.67 [95% CI, 1.03–2.72], P=0.0.36) and CKD (adjusted OR, 1.94 [95% CI, 1.11–3.39], P=0.019) were found to be associated with an increased risk of 30‐day readmission. In addition, self‐pay status was associated with a lower risk of 30‐day readmission compared with Medicare (adjusted OR, 0.32 [95% CI, 0.12–0.85], P=0.022) and low median household income was associated with a higher risk of 30‐day readmission compared with higher median household income (adjusted OR, 2.35 [95% CI, 1.17–4.74], P=0.017). LOS ≥5 days during the index hospitalization was highly predictive of 30‐day readmission compared with LOS ≤2 days (adjusted OR, 2.16 [95% CI, 1.15–4.04], P=0.017). Discharge to an extended‐care facility compared with discharge to home with self‐care disposition was associated with a lower risk of 30‐day readmission (adjusted OR, 0.45 [95% CI, 0.24–0.86], P=0.016).

Table 3. Predictors of 30‐Day All Cause Readmission in Patients With STEMI With COVID‐19

Predictors of 30‐day all cause readmissionUnivariable regression modelMultivariable regression model
OR95% CIP valueAdjusted OR95% CIP value
Age, y1.010.991.020.2031.000.981.020.728
Age group, y
<50Reference
50–640.710.401.270.254
65–791.070.611.890.813
≥801.420.742.750.295
Female1.020.651.590.9430.910.541.540.726
Comorbidities
Hypertension1.540.962.470.0731.080.611.920.790
Diabetes1.440.982.120.0641.010.651.570.953
Hyperlipidemia1.130.771.670.532
Heart failure2.341.583.470.0001.671.032.720.036
Pulmonary circulation disorder1.910.904.030.0911.140.472.790.768
Atrial fibrillation1.640.942.880.0841.180.592.330.642
Known CAD1.591.092.330.0171.140.751.730.545
Family history of CAD0.900.461.740.745
Fluid/electrolyte disorders1.370.942.000.1000.820.511.340.434
History of pacemaker/ICD1.910.645.740.247
Valvular heart disease/prosthetic valve1.881.013.470.0451.270.632.550.498
Prior stroke/TIA1.040.581.890.885
Peripheral vascular disease1.690.933.070.0831.210.562.590.628
Chronic pulmonary disease1.480.882.490.1391.290.722.320.396
CKD2.591.703.950.0001.941.113.390.019
Chronic liver disease2.221.094.520.0271.990.874.540.101
Nonadherence to medications1.180.562.470.6641.100.532.320.804
Anemia1.490.693.240.312
Coagulopathy1.350.772.370.298
Obesity0.980.621.560.941
Smoking1.090.681.750.726
ECI1.231.141.330.0001.000.851.190.959
In‐hospital events
Cardiac complications1.450.942.230.0900.770.471.270.314
Cardiogenic shock2.041.233.380.006
Cardiac arrest1.420.742.740.294
Vasopressor use0.990.422.310.982
Short‐term mechanical circulatory support (Impella, IABP, ECMO)2.011.083.710.026
Respiratory complications1.901.262.870.0021.210.712.040.481
Need for mechanical ventilation1.360.832.220.228
24 h0.690.143.240.634
24–96 h2.431.344.390.003
>96 h0.730.351.530.410
Acute hypoxic respiratory failure1.711.142.570.010
Acute respiratory distress syndrome1.420.633.220.399
Need for nonmechanical ventilation0.700.261.900.486
Septic shock1.280.702.330.421
Acute DVT/PE2.101.024.290.0431.700.733.980.217
Acute kidney injury1.921.272.900.0021.200.791.840.392
Major bleeding requiring blood transfusion1.100.343.600.873
Vascular complications1.240.265.910.791
Treatment during index STEMI admission
Revascularization0.860.571.280.4501.370.822.280.229
Admission day
WeekdayReference
Weekend0.920.611.410.715
Primary payer
MedicareReference
Medicaid0.940.551.590.8150.800.421.530.501
Private insurance0.560.340.900.0170.660.361.210.178
Self‐pay/no charge/other0.430.190.970.0410.320.120.850.022
Median household income quartile
76–100thReference
0–25th2.311.194.490.0132.351.174.740.017
26–50th1.390.672.880.3731.260.602.660.540
51–75th1.490.743.000.2631.470.713.050.296
Hospital bed size
SmallReference
Medium1.850.993.460.055
Large1.330.732.440.351
Hospital teaching status
NonteachingReference
Teaching1.040.671.630.851
Hospital location
Nonmetropolitan areaReference
Metropolitan area1.810.664.980.248
LOS category, d
≤2Reference
3–41.290.682.450.4331.100.572.140.776
≥52.971.834.830.0002.161.154.040.017
Discharge disposition
Discharged home with self‐careReference
Short‐term hospital1.870.754.670.1802.010.775.230.151
Discharge to extended‐care facility1.190.711.980.5140.450.240.860.016
Home health care1.410.862.310.1760.600.331.080.092
Against medical advice1.330.404.440.6411.260.334.860.735

Known coronary artery disease (CAD) is a composite of ischemic cardiomyopathy, prior myocardial infarction, prior percutaneous coronary intervention (PCI), and prior coronary artery bypass grafting (CABG); revascularization is a composite of thrombolytic therapy, all PCIs, and CABG; respiratory complications are a composite of acute hypoxic respiratory failure, acute respiratory distress syndrome, and need for mechanical ventilation; and cardiac complications are a composite of cardiac arrest, cardiogenic shock, vasopressor use, and short‐term mechanical circulatory support. CKD indicates chronic kidney disease; DVT, deep venous thrombosis; ECI, Elixhauser Comorbidity Index; ECMO, extracorporeal membrane oxygenation; IABP, intra‐aortic balloon pump; ICD, implantable cardioverter‐defibrillator; LOS, length of stay; OR, odds ratio; PE, pulmonary embolism; STEMI, ST‐segment–elevation myocardial infarction; and TIA, transient ischemic attack.

Subgroup Analysis

Figure 5 shows a forest plot comparing patients with STEMI with versus without COVID‐19 according to various subgroups for 30‐day all‐cause readmission outcome. The adjusted risk for 30‐day all‐cause readmission in patients with STEMI with COVID‐19 was not significantly different than those without COVID‐19 when stratified by patient's age (<65 years versus ≥65 years), sex (men versus women), location at the time of the STEMI (in‐hospital versus out of hospital), need for mechanical ventilation (no mechanical ventilation versus mechanical ventilation), heart failure (absence versus presence), CKD (absence versus presence), and chronic pulmonary disease (absence versus presence).

Figure 5. Subgroup analysis of 30‐day readmissions in patients with ST‐segment–elevation myocardial infarction (STEMI) with and without COVID‐19.

CKD indicates chronic kidney disease.

Sensitivity Analysis

Sensitivity analysis using 1:1 propensity score matching of patients with STEMI with and without COVID‐19 demonstrated that the adjusted risk for 30‐day all‐cause readmission and 30‐day MACE were not significantly different in patients with STEMI without COVID‐19 compared with those with COVID‐19 (Table S4).

DISCUSSION

In this nationwide real‐world retrospective observational study, we investigated 30‐day readmissions in patients presenting with STEMI stratified by a co‐diagnosis of COVID‐19. The main findings of the study were: (1) when evaluating 30‐day outcomes in patients with STEMI, there was no difference in the rate of all‐cause readmissions, MACE, or revascularization after stratification for COVID‐19; (2) the leading causes of 30‐day readmission, regardless of COVID‐19 status, were cardiac (ACS, arrhythmias, and heart failure), followed by infectious; (3) while comparing the causes of 30‐day readmissions, infectious causes were significantly higher for patients with STEMI with COVID‐19 as compared with those without COVID‐19; and (4) in a multivariate model, independent predictors for 30‐day readmission in patients with STEMI and COVID‐19 included prior heart failure, CKD, Medicare as a primary payer as compared with self‐pay, median household income in the lowest quartile, and LOS ≥5 days during index hospitalization (Figure 6).

Figure 6. Thirty‐day readmissions in patients with ST‐segment–elevation myocardial infarction (STEMI) with and without COVID‐19.

CAD indicates coronary artery disease; and MACE, major adverse cardiac events

The 30‐day readmission rates for patients with STEMI with COVID‐19 (11.4%) and without COVID‐19 (10.6%) were lower than reported in studies prior to the COVID‐19 era (ranging from 12.3% to 55.6%).14, 15, 16, 17 Historically, there has been a decline in the rate of 30‐day readmissions following acute myocardial infarction (AMI).18 Although some of the decline in the rate of 30‐day readmissions we observed may be a continuation of this trend, COVID‐19 likely independently contributed to this decline. This may in large part be related to the significant reluctance of patients to return to the hospital due to fear of COVID‐19. This is in line with evidence that admissions for AMI declined significantly during the COVID‐19 pandemic as compared with years prior.19, 20

Regardless of COVID‐19 status, we observed cardiac causes as the most common causes for readmission within 30‐days of STEMI. On further assessment of the specific cardiac causes, we identified ACS (56%), arrhythmias (20%), and heart failure (13%) as the leading causes among the patients with COVID‐19. Studies prior to the rise of COVID‐19 identified similar trends, with cardiac causes as the leading cause for readmission among patients with AMI. More specifically, ischemic heart disease, heart failure, and arrhythmias are cited as the most common cardiac causes in previous studies.14, 21

COVID‐19–related hypercoagulability is associated with arterial thrombi and serves as an important factor that may theoretically influence readmissions in patients with STEMI and COVID‐19.15 Importantly, the incidence of arterial thrombi in patients with COVID‐19 has been shown to decline rapidly after the index diagnosis of COVID‐19 and returned to near baseline levels after 1 to 2 weeks in one study, suggesting that COVID‐19–related arterial hypercoagulability may not have significantly impacted the causes of readmission beyond the first 1 to 2 weeks in our cohort.15 Furthermore, there was no statistically significant difference in the need for 30‐day revascularization after stratification for COVID‐19 in our study, which is concordant with the near resolution of COVID‐19–related arterial hypercoagulability within 1 to 2 weeks after diagnosis that likely impacts outcomes during hospitalization but subsequently improves.

Among noncardiac causes, infections were the most common cause of readmission in both groups of STEMI. Similarly, in the pre–COVID‐19 era, septicemia and pneumonia were important leading causes of readmission after STEMI.14, 16, 21 However, the percentage of 30‐day readmissions attributable to infectious causes was significantly higher in patients with STEMI who had COVID‐19 compared with those who did not have COVID‐19 (29.9% versus 11.3%, P<0.05). Thus, infectious causes for readmission have shifted to those related to COVID‐19 in our cohort, whereas they were largely related to pneumonia and sepsis in the past.14, 16, 21

Data from the Centers for Medicare & Medicaid Services and the National Cardiovascular Data Registry have identified percutaneous coronary intervention as 1 of the 7 conditions responsible for almost one‐third of readmission costs in the United States.17 In recognition of this, the Hospital Readmission Reduction Program, which originated from the Affordable Care Act of 2010, targeted both AMI and heart failure readmissions to curtail the high costs, under the premise that there may be a simultaneous increase in the quality of care received by patients.17 How COVID‐19 impacts the determinants of 30‐day readmissions in patients with STEMI is unknown. To continue to address the high burden of readmissions following AMI, it is important to identify predictors of readmission. In our study, many of the important predictors of 30‐day readmissions in patients with STEMI and a co‐diagnosis of COVID‐19 were similar to those that have been described in patients from the pre‐COVID era: prior heart failure, CKD, Medicare as a primary payer, median household income in the lowest quartile, and an LOS ≥5 days.14, 21 Patients with such comorbidities may need to have an early outpatient follow‐up to avoid such readmissions. Further research is needed to examine the preventability of 30‐day readmissions and to explore whether short‐term readmissions after STEMI should serve as a reliable quality metric of hospital performance and clinical outcomes in patients with STEMI and COVID‐19.

The rise of COVID‐19 shifted the landscape of health care and cardiology with a reduction in the use of invasive strategies for the management of ACS, an increase in the rate of late presentations for AMI, and poor in‐hospital outcomes in patients with STEMI with a co‐diagnosis of COVID‐19 compared with those with STEMI without COVID‐19.2, 4, 7 Our data suggest that 30‐day readmission MACE and mortality are similar after stratification for COVID‐19 but we may have confounders that were not accountable in the analysis such as patient's personal preference regarding treatment during index admission. Thus, efforts to reduce 30‐day readmissions by improving patient education, triage mechanisms, and care coordination should be continued.17

Limitations

There are notable limitations in our study, many of which are tied to the use of an administrative database. As a result of this, errors related to coding discrepancies may influence our results as the fidelity of our data is dependent on the rigor of coding practices. However, HCUP quality‐control procedures are routinely performed to confirm that NRD data values are valid, consistent, and reliable.22 In addition, this database lacks information regarding important clinical data, such as laboratory results and medications. Although we have reported mortality during the 30‐day period following the index admission despite this not being the main focus of our analysis, the NRD does not have data regarding out‐of‐hospital mortality in patients discharged after STEMI. Therefore, our post‐STEMI mortality estimates may be lower than the actual 30‐day post‐STEMI mortality value. Although a diagnosis of COVID‐19 was linked to patients at the time of admission for STEMI, it is unclear whether these patients had acute infections or prior infections with residual positive test results, which could impact the results of this study. Finally, we did not have data on the severity of the COVID‐19, patient's preference/family discussion regarding goals of care, and rates of delayed diagnosis of STEMI, which might affect the rate of revascularization for STEMI as seen in this study. Although the NRD is subject to these important limitations, it is the largest publicly available all‐payer readmission database in the United States that encompassed over 16 million unweighted and 32 million weighted discharges in the United States in 2020, representing >60% of the total US resident population. Moreover, results from sensitivity analysis with 1:1 propensity score matching are consistent with the primary conclusion. Thus, this analysis provides valuable data on 30‐day outcomes for patients with STEMI and COVID‐19 from the most current nationally available database, and the multi‐institutional sample makes our results generalizable.

CONCLUSIONS

We demonstrate a lower rate of 30‐day readmissions in patients with STEMI in comparison to the pre–COVID‐19 era. This was independent of a co‐diagnosis of COVID‐19. There was no difference in 30‐day all‐cause readmissions in patients with STEMI after stratification for COVID‐19. Cardiac causes (ACS, arrhythmias, and heart failure) were the most common reasons for 30‐day readmission in patients with STEMI with and without COVID‐19. Prior heart failure, CKD, median household income in the lowest quartile, and an LOS ≥5 days were the predictors of 30‐day readmission in patients with STEMI and COVID‐19. Further research is needed to examine the preventability of 30‐day readmissions in patients with STEMI and COVID‐19.

Sources of Funding

None.

Disclosures

None.

Acknowledgments

We thank Ms Vibhuti Bhatt for her help in creating Figure 6. Part of Figure 6 was created with BioRender.com. All authors participated in the research and preparation of the article per the International Committee of Medical Journal Editors (ICMJE) recommendations.

Footnotes

*Correspondence to: Kamal Gupta, MD, University of Kansas Medical Center, 3901 Rainbow Boulevard, Mailstop 4023, Kansas City, KS 66160. Email:

*K. N. Patel and M. Majmundar contributed equally to this article and are co‐first authors.

This manuscript was sent to Tiffany M. Powell‐Wiley, MD, MPH, Associate Editor, for review by expert referees, editorial decision, and final disposition.

Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.123.029738

For Sources of Funding and Disclosures, see page 15.

References

  • 1 Goel A, Malik AH, Bandyopadhyay D, Isath A, Gupta R, Hajra A, Shrivastav R, Virani SS, Fonarow GC, Lavie CJ, et al. Impact of COVID‐19 on outcomes of patients hospitalized with STEMI: a nationwide propensity‐matched analysis. Curr Probl Cardiol. 2023; 48:101547. doi: 10.1016/j.cpcardiol.2022.101547CrossrefMedlineGoogle Scholar
  • 2 Lasica R, Djukanovic L, Mrdovic I, Savic L, Ristic A, Zdravkovic M, Simic D, Krljanac G, Popovic D, Simeunovic D, et al. Acute coronary syndrome in the COVID‐19 era‐differences and dilemmas compared to the pre‐COVID‐19 era. J Clin Med. 2022; 11:11. doi: 10.3390/jcm11113024CrossrefGoogle Scholar
  • 3 Healthcare Cost and Utilization Project (HCUP) . Overview of the Nationwide Readmissions Database (NRD). Agency for Healthcare Research and Quality. Accessed January 3, 2023. https://www.hcup‐us.ahrq.gov/nrdoverview.jspGoogle Scholar
  • 4 Patel KN, Majmundar M, Vasudeva R, Doshi R, Kaur A, Mehta H, Gupta K. Impact of gender, race, and insurance status on inhospital management and outcomes in patients with COVID‐19 and ST‐elevation myocardial infarction (a nationwide analysis). Am J Cardiol. 2023; 198:14–25. doi: 10.1016/j.amjcard.2023.04.030CrossrefMedlineGoogle Scholar
  • 5 Santoro F, Nunez‐Gil IJ, Vitale E, Viana‐Llamas MC, Reche‐Martinez B, Romero‐Pareja R, Feltez Guzman G, Fernandez Rozas I, Uribarri A, Becerra‐Munoz VM, et al. Antiplatelet therapy and outcome in COVID‐19: the Health Outcome Predictive Evaluation Registry. Heart. 2022; 108:130–136. doi: 10.1136/heartjnl-2021-319552CrossrefMedlineGoogle Scholar
  • 6 Azzalini L, Seth M, Sukul D, Valle JA, Daher E, Wanamaker B, Tucciarone MT, Zaitoun A, Madder RD, Gurm HS. Trends and outcomes of percutaneous coronary intervention during the COVID‐19 pandemic in Michigan. PLoS One. 2022; 17:e0273638. doi: 10.1371/journal.pone.0273638CrossrefMedlineGoogle Scholar
  • 7 Saad M, Kennedy KF, Imran H, Louis DW, Shippey E, Poppas A, Wood KE, Abbott JD, Aronow HD. Association between COVID‐19 diagnosis and In‐hospital mortality in patients hospitalized with ST‐segment elevation myocardial infarction. JAMA. 2021; 326:1940–1952. doi: 10.1001/jama.2021.18890CrossrefMedlineGoogle Scholar
  • 8 Majmundar M, Doshi R, Patel KN, Zala H, Kumar A, Kalra A. Prevalence, trends and outcomes of cardiovascular diseases in pregnant patients in the United States: 2010 to 2019. Eur Heart J. 2022; 44:726–737. doi: 10.1093/eurheartj/ehac669CrossrefGoogle Scholar
  • 9 Majmundar M, Patel KN, Doshi R, Anantha‐Narayanan M, Kumar A, Reed GW, Puri R, Kapadia SR, Jaradat ZA, Bhatt DL, et al. Comparison of 6‐month outcomes of endovascular vs surgical revascularization for patients with critical limb ischemia. JAMA Netw Open. 2022; 5:e2227746. doi: 10.1001/jamanetworkopen.2022.27746CrossrefMedlineGoogle Scholar
  • 10 Suri K, Majmundar M, Kumar A, Patel KN, Zhou XY, Zala H, Kalra A. Outcomes and readmission in patients with retinal artery occlusion (from the Nationwide Readmission Database). Am J Cardiol. 2022; 183:105–108. doi: 10.1016/j.amjcard.2022.07.040CrossrefMedlineGoogle Scholar
  • 11 Majmundar M, Doshi R, Vasudeva R, Patel KN, Kumar A, Bangalore S, Kalra A. Hospital procedural volume predicts short‐term outcomes for patients undergoing procedures requiring large‐bore access. Eur Heart J Acute Cardiovasc Care. 2023; 12:237–242. doi: 10.1093/ehjacc/zuad011CrossrefMedlineGoogle Scholar
  • 12 von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, Initiative S. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007; 370:1453–1457. doi: 10.1016/S0140-6736(07)61602-XCrossrefMedlineGoogle Scholar
  • 13 Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998; 36:8–27. doi: 10.1097/00005650-199801000-00004CrossrefMedlineGoogle Scholar
  • 14 Kim LK, Yeo I, Cheung JW, Swaminathan RV, Wong SC, Charitakis K, Adejumo O, Chae J, Minutello RM, Bergman G, et al. Thirty‐day readmission rates, timing, causes, and costs after ST‐segment‐elevation myocardial infarction in the United States: a National Readmission Database analysis 2010–2014. J Am Heart Assoc. 2018; 7:e009863. doi: 10.1161/JAHA.118.009863LinkGoogle Scholar
  • 15 Knight R, Walker V, Ip S, Cooper JA, Bolton T, Keene S, Denholm R, Akbari A, Abbasizanjani H, Torabi F, et al. Association of COVID‐19 with major arterial and venous thrombotic diseases: a population‐wide cohort study of 48 million adults in England and Wales. Circulation. 2022; 146:892–906. doi: 10.1161/CIRCULATIONAHA.122.060785LinkGoogle Scholar
  • 16 Krittanawong C, Yue B, Mahtta D, Narasimhan B, Kumar A, Wang Z, Sharma SK, Tamis‐Holland JE, Brar SS, Mehran R, et al. Readmission in patients with ST‐elevation myocardial infarction in 4 age groups (<45, >45 to <60, 60 to <75, and >75). Am J Cardiol. 2022; 173:25–32. doi: 10.1016/j.amjcard.2022.02.045CrossrefMedlineGoogle Scholar
  • 17 Wasfy JH, Yeh RW. Future of the PCI readmission metric. Circ Cardiovasc Qual Outcomes. 2016; 9:186–189. doi: 10.1161/CIRCOUTCOMES.115.002472LinkGoogle Scholar
  • 18 Zuckerman RB, Sheingold SH, Orav EJ, Ruhter J, Epstein AM. Readmissions, observation, and the hospital readmissions reduction program. N Engl J Med. 2016; 374:1543–1551. doi: 10.1056/NEJMsa1513024CrossrefMedlineGoogle Scholar
  • 19 Fox DK, Waken RJ, Johnson DY, Hammond G, Yu J, Fanous E, Maddox TM, Joynt Maddox KE. Impact of the COVID‐19 pandemic on patients without COVID‐19 with acute myocardial infarction and heart failure. J Am Heart Assoc. 2022; 11:e022625. doi: 10.1161/JAHA.121.022625LinkGoogle Scholar
  • 20 Gluckman TJ, Wilson MA, Chiu ST, Penny BW, Chepuri VB, Waggoner JW, Spinelli KJ. Case rates, treatment approaches, and outcomes in acute myocardial infarction during the coronavirus disease 2019 pandemic. JAMA Cardiol. 2020; 5:1419–1424. doi: 10.1001/jamacardio.2020.3629CrossrefMedlineGoogle Scholar
  • 21 Wang H, Zhao T, Wei X, Lu H, Lin X. The prevalence of 30‐day readmission after acute myocardial infarction: a systematic review and meta‐analysis. Clin Cardiol. 2019; 42:889–898. doi: 10.1002/clc.23238CrossrefMedlineGoogle Scholar
  • 22 Barrett M, McCarty J, Kenney T, Liang L. Methods applying AHRQ quality indicators to Healthcare Cost and Utilization Project (HCUP) data for the 2022 National Healthcare Quality and Disparities Report (NHQDR). Agency for Healthcare Research and Quality; 2022. Accessed January 3, 2023. www.hcup‐us.ahrq.gov/reports/methods/methods.jspGoogle Scholar

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.