This prospective, multicenter, observational cohort study was based on data obtained from the 7 tertiary Spanish hospitals with on-site percutaneous coronary intervention (PCI) capability operating a 24-hour, 7-day a week service (RECOVID-SCA Registry). This registry included all consecutive AMI patients with or without ST-segment elevation (STEMI or NSTEMI) who underwent coronary angiography from March 15 to April 15, 2020. This period chronologically corresponds with the COVID-19 surge in Spain. The study was performed in accordance with the 1964 Declaration of Helsinki and its subsequent amendments and was approved by the Ethics Committee from the hospital in charge of coordinating the registry. An independent academic clinical end point committee performed blinded adjudication of events.

Demographic information, cardiovascular history, and risk factors were recorded: sex, body mass index, hypertension, diabetes mellitus, dyslipidemia, chronic kidney disease with creatinine clearance <30mL/min, smoking (current or former), chronic obstructive pulmonary disease, heart failure (both preserved and reduced ejection fraction), prior coronary artery disease, stroke or transient ischemic attack, any pattern of atrial fibrillation,11 cancer, and chronic anticoagulant treatment.

Patients were classified according to their initial diagnosis as STEMI or NSTEMI. STEMI was diagnosed in patients with clinical characteristics of myocardial ischemia and an electrocardiogram with ST-segment elevation in at least 2 contiguous leads.12 Meanwhile, NSTEMI diagnosis required the presence of angina associated with a cardiac biomarker increase above the 99th percentile of the upper reference limit.13 Patients who were finally diagnosed with myocarditis or takotsubo cardiomyopathy were excluded. Likewise, patients were classified according to whether they tested positive for COVID-19. This diagnosis was made in each center according to World Health Organization guidelines.14 In each patient, a real-time fluorescence polymerase chain reaction was performed to detect the positive nucleic acid of SARS-CoV-2 in throat swabs or in lower respiratory tract samples.

Clinical manifestations, electrocardiographic and echocardiographic findings, Killip-Kimball classification, and GRACE score were recorded, as well as the results of the following laboratory tests: cardiac troponin, creatine kinase, N-terminal pro-brain natriuretic peptide, brain natriuretic peptide, C-reactive protein, D-dimer, ferritin, complete blood count, and liver function. Each center measured cardiac troponin I or T using high-sensitivity assays with different upper reference limits. To standardize these results, we used the ratio of the observed troponin value divided by the cutoff points for the 99th percentile of the upper reference limit from every center, as performed previously.15

The following procedure-related findings were assessed: vascular access, number of coronary vessels with severe stenosis (> 70%), culprit vessel, angiographic characteristics (thrombus characterization according to the Thrombolysis in Myocardial Infarction [TIMI] study group grading system and the TIMI flow grade before and after the procedure),16,17 number of stents, and thrombus aspiration during PCI.

The following in-hospital adverse events were considered: bleeding according to the Bleeding Academic Research Consortium (BARC) scale,18 ARDS requiring tracheal intubation, stroke, systemic arterial thromboembolism, myocardial reinfarction, the use of mechanical circulatory support, sustained ventricular arrhythmias, death, and cardiovascular death. Re-infarction was considered when ST-segment elevation ≥ 1mm recurred or new pathognomonic Q waves appeared in at least 2 contiguous leads, particularly when associated with ischemic symptoms and troponin elevation (> 20% increase in the troponin value in the second sample).19 ARDS was defined as an acute condition characterized by bilateral pulmonary infiltrates and severe hypoxemia (ratio of the partial pressure of oxygen in the patient's arterial blood [PaO2] to the fraction of oxygen in inspired air [FiO2] in the absence of evidence of cardiogenic pulmonary edema.20 Cardiovascular deaths included deaths due to AMI, sudden cardiac death, death due to heart failure, death due to stroke, death due to cardiovascular hemorrhage, or death secondary to cardiovascular procedures.21 Embolism refers to systemic embolism and stroke and excludes venous thromboembolism.

Baseline variables and clinical outcomes during admission were compared between patients with and without COVID-19 using a chi-square or Fisher exact test (if the expected value in each cell was less than 5) for categorical variables and a t test or Mann-Whitney U test (if the assumption of normality was significant [Shapiro-Wilk test P <.05]) for continuous variables, as appropriate. No imputation of missing data was required. In all tests, the P values of all outcomes were 2-sided; a value less than .05 was considered to indicate statistical significance. Confidence intervals were defined as 95% (95%CI). A multivariate logistic regression model was used to identify independent predictors of in-hospital death using “Forward: conditional” selection. Patients with type 2 AMI were excluded from this analysis due to its distinct pathophysiology vs type 1 AMI. Variables with P <.05 in univariate analysis were entered into a multivariate analysis. GRACE score> 140 and left ventricular ejection fraction <30% were analyzed as categorical variables. Results are reported as odds ratios (ORs) with associated 95%CIs. The Hosmer-Lemeshow test and the Nagelkerke R2 from the regression modeling were used as indicators of the goodness-of-fit of each risk model and to assess their calibration ability.

In addition, due to the observational nature of the study, there were some differences in the baseline characteristics between the patients in the 2 groups. Given these differences, a propensity score analysis was performed using generalized boosted models with COVID-19 as the dependent variable and the baseline characteristics outlined in table 1 as covariates. A propensity score-weighted analysis was performed to determine the effect of COVID-19 on mortality.

Analyses were performed with the software package IBM SPSS Statistics for Mac version 25 (IBM Corp, Armonk, New York, United States) and R statistics version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria).

Between March 15, 2020, and April 15, 2020, 193 patients were diagnosed with AMI. Six patients were excluded: 3 who were diagnosed with myocarditis, 1 of which was fulminant myocarditis resulting in death, and 3 who were diagnosed with tako-tsubo cardiomyopathy. A total of 187 patients were included in the study, 76 with NSTEMI (40%) and 111 with STEMI (60%) (figure 1). The mean age at admission was 65±13 years old and 34 patients (18%) were female. Out of all patients, 32 (17%) tested positive for COVID-19: 12 (37%) had NSTEMI and 20 (63%) had STEMI (figure 2). Ten patients with no infectious symptoms did not undergo a real-time fluorescence polymerase chain reaction for SARS-CoV-2 and were considered negative for COVID-19 (8 with STEMI and 2 with NSTEMI). None of the patients had adverse events. There were no statistically significant differences in baseline characteristics between patients who were diagnosed with COVID-19 and those who were not (table 2).

Figure 1.

Acute myocardial infarction during the COVID-19 outbreak. 95%CI, 95%confidence interval; COVID-19, coronavirus disease 2019; ECG, electrocardiogram; GRACE, Global Registry of Acute Coronary Events; LVEF, left ventricular ejection fraction; NSTEMI, non-ST-segment elevation myocardial infarction; OR, odds ratio; STEMI, ST-segment elevation myocardial infarction; TIMI, Thrombolysis in Myocardial Infarction.

(0.6MB).

Figure 2.

Distribution of patients based on their initial diagnoses. AMI, acute myocardial infarction; COVID-19, coronavirus disease 2019; NSTEMI, non-ST-segment elevation myocardial infarction; STEMI, ST-segment elevation myocardial infarction.

(0.11MB).

Non-COVID-19 patients predominantly had chest pain, whereas COVID-19 patients mainly had respiratory symptoms such as fever and cough. Killip-Kimball III/IV and GRACE scores were significantly higher in patients with COVID-19. There were no significant differences in left ventricular ejection fraction between the 2 groups. COVID-19 patients had higher levels of C-reactive protein at admission and of peaks levels during hospitalization, fewer lymphocytes at admission, and higher peak levels of ferritin, D-dimer, and alanine aminotransferase. There were no significant differences in cardiac troponin at admission and in the peak measured during hospitalization between the 2 groups. Regarding antiplatelet treatment, patients with COVID-19 had a higher proportion of treatment with prasugrel. The hospital length of stay was significantly shorter in non-COVID-19 than COVID-19 patients (4.3±2.9 days vs 7.6±6.3 days, P <.01).

Of the 187 patients who underwent coronary angiography, 7 (3.7%) were finally diagnosed with type 2 AMI with nonobstructive coronary arteries; only 1 patient was COVID-19 positive (3.2%), whereas 6 were COVID-19 negative (3.8%). There were no differences in terms of mortality or major adverse cardiovascular events between patients with type 1 AMI or type 2 AMI.

Primary PCI was performed in 106 of the 111 patients with STEMI. The delay from symptom onset to guidewire crossing in STEMI patients was similar in COVID-19 and non-COVID-19 patients (240 [interquartile range, 157-315] minutes vs 241 [99-420] minutes, P=.80). There was no evidence of a higher amount of thrombus measured by TIMI thrombus score or a higher rate of thrombectomy in any of the groups.

The presence of multivessel coronary artery disease was similar in the 2 groups. There were no significant differences with respect to the treatment of nonculprit vessels during admission, with rates of 21% in COVID-19 patients and of 23% in non-COVID-19 patients (P=.20). We also found no differences in the anatomical characteristics of the culprit vessel or number of implanted stents. Procedure-related complications were also similar in the 2 groups.

There were 14 in-hospital deaths (7.5%, 95%CI, 4-11). Total mortality (8 [25%] vs 6 [3.8%], P <.001) and cardiovascular death (5 [15.2%] vs 3 [1.8%], P=.001) were significantly higher in the COVID-19 group than in the non-COVID-19 group (figure 3). No significant differences were found in the remaining adverse events. COVID-19 patients with AMI and bilateral pneumonia who needed mechanical ventilation because of ARDS had higher mortality than patients without ARDS (83% vs 12%, P=.002).

Figure 3.

Association of COVID-19 with adverse events in patients admitted for acute myocardial infarction. AMI, acute myocardial infarction; ARDS, acute respiratory distress syndrome; BARC, Bleeding Academic Research Consortium scale; COVID-19, coronavirus disease 2019; CV death, cardiovascular death; Embolism, refers to stroke or systemic arterial embolism; MechSupp, mechanical circulatory support; Re-AMI, myocardial re-infarction; V.Arrhyt, ventricular arrhythmias.

(0.27MB).

The causes of death in COVID-19 patients were as follows: ARDS in 5 patients, cardiogenic shock after distal thrombus embolization from a left main coronary lesion in 2, and cardiac rupture as a mechanical complication of a STEMI in 1. In contrast, the causes of death in non-COVID-19 patients comprised 2 cardiogenic shocks, 1 possible stent thrombosis (Academic Research Consortium definition of stent thrombosis), 1 cerebellar stroke, 1 hypoxic encephalopathy after prolonged out-of-hospital cardiac arrest, and 1 fulminant hepatitis.

Univariate logistic regression analysis of the potential risk factors for total mortality is shown in table 1 of the supplementary data. Risk factors for in-hospital mortality were age, Killip-Kimball> 2, GRACE score> 140, TIMI flow post-PCI <3, severe coronary calcification, left ventricular ejection fraction <30%, C-reactive protein, total creatine kinase, D-dimer, aspartate aminotransferase, ferritin, BARC> 2, ARDS, and COVID-19. When all of these variables were entered into a multivariate conditional logistic regression model, only COVID-19 (OR, 6.61; 95%CI, 1.82-24.43; P=.02) and GRACE score> 140 (OR, 23.45; 95%CI; 2.52-62.51; P=.005) remained independent predictors of in-hospital death (Hosmer and Lemeshow, P=.65; Nagelkerke, R2=0.54; figure 1). After propensity score matching, adequate comparability was achieved by a decrease in the standardized differences to less than 20% for all covariates (table 2). In the propensity score-adjusted analysis, mortality was significantly higher in the group of patients with COVID-19 (25.8% vs 18.0%, P=.045).

There are several limitations to be considered in the interpretation of our study. First, this was an observational and nonrandomized study and, as such, the outcomes may have been influenced by both identified and unidentified confounders. Second, the number of patients studied with AMI and COVID-19 was small, which could limit the number of independent predictors identified and the consistency of the results. Finally, all variables were included via consultation of possibly incomplete medical records. Despite these limitations, the statistical significance was sufficient to allow us to draw preliminary conclusions.