ORIGINAL RESEARCH published: 26 April 2022 doi: 10.3389/fcvm.2022.827212 Dynamics of Emergency Cardiovascular Hospital Admissions and In-Hospital Mortality During the COVID-19 Pandemic: Time Series Analysis and Impact of Socioeconomic Factors Claudia Álvarez-Martín 1,2 , Aida Ribera 1,2,3* , Josep Ramon Marsal 1,2 , Albert Ariza-Solé 4,5,6 , Santiago Pérez-Hoyos 7,8 , Gerard Oristrell 6,9 , Toni Soriano-Colomé 9 , Edited by: Rafael Romaguera 4,5 , Jose Ignacio Pijoan 2,10 , Rosa M. Lidón 6,9 , Josepa Mauri 11,12 and Jose R. Medina Inojosa, Ignacio Ferreira-González 1,2,9 Mayo Clinic, United States 1 Cardiovascular Research and Epidemiology Unit, Cardiology Department, University Hospital Vall d’Hebron and Vall Reviewed by: d’Hebron Research Institute, Barcelona, Spain, 2 Centro de Investigación Biomédica en Red de Epidemiología y Salud Mubashir Ayaz Ahmed, Pública (CIBERESP), Madrid, Spain, 3 Recerca en Envelliment, Fragilitat i Transicions (REFiT) Barcelona Research Group, AMITA Health St Joseph Hospital, Parc Sanitari Pere Virgili and Vall d’Hebron Institute of Research, Barcelona, Spain, 4 Cardiology Department, Hospital United States Universitari de Bellvitge, Barcelona, Spain, 5 Bioheart, Grup de Malalties Cardiovasculars, Institut d’Investigació Biomèdica Maria Irene Barillas Lara, de Bellvitge (IDIBELL), L’Hospitalet de Llobregat, Barcelona, Spain, 6 Centro de Investigación Biomédica en Red en Boston Medical Center, United States Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain, 7 Statistics and Bioinformatics Unit, Vall d’Hebron Research *Correspondence: Institute, Barcelona, Spain, 8 Statistics Department, Faculty of Biology, University of Barcelona, Barcelona, Spain, Aida Ribera 9 Cardiology Department, University Hospital Vall d’Hebron, Universitat Autònoma de Barcelona, Barcelona, Spain, 10 Clinical ariberas@perevirgili.cat Epidemiology Unit, Hospital Universitario Cruces/BioCruces-Bizkaia Health Research Institute, Barakaldo, Spain, 11 Cardiology Department, Hospital German Trias i Pujol, Barcelona, Spain, 12 Director Plan for Cardiovascular Diseases, Pla Specialty section: Director de Malalties Cardiovasculars (PDMCV), Department of Health, Catalan Government, Barcelona, Spain This article was submitted to Cardiovascular Epidemiology and Prevention, Aims: This study aimed to evaluate the decline in urgent cardiovascular hospital a section of the journal admissions and in-hospital mortality during the COVID pandemic in two successive Frontiers in Cardiovascular Medicine waves, and to evaluate differences by sex, age, and deprivation index subgroups. Received: 01 December 2021 Accepted: 24 March 2022 Methods and Results: We obtained acute cardiovascular hospital episodes during the Published: 26 April 2022 years 2019–2020 from region-wide data on public healthcare usage for the population of Citation: Catalonia (North-East Spain). We fitted time models to estimate the incidence rate ratios Álvarez-Martín C, Ribera A, Marsal JR, Ariza-Solé A, (IRRs) of the acute coronary syndrome (ACS) and acute heart failure (HF) admissions Pérez-Hoyos S, Oristrell G, during the first pandemic wave, the between-waves period, and the second wave Soriano-Colomé T, Romaguera R, Pijoan JI, Lidón RM, Mauri J and compared with the corresponding pre-COVID-19 periods and to test for the interaction Ferreira-González I (2022) Dynamics with sex, age, and area-based socioeconomic level. We evaluated the effect of COVID- of Emergency Cardiovascular Hospital 19 period on in-hospital mortality. ACS (n = 8,636) and HF (n = 27,566) episodes were Admissions and In-Hospital Mortality During the COVID-19 Pandemic: Time defined using primary diagnostic ICD-10 codes. ACS and HF admissions decreased Series Analysis and Impact during the first wave (IRR = 0.66, 95%CI: 0.58–0.76 and IRR = 0.61, 95% CI: 0.55– of Socioeconomic Factors. Front. Cardiovasc. Med. 9:827212. 0.68, respectively) and during the second wave (IRR = 0.80, 95%CI: 0.72–0.88 and doi: 10.3389/fcvm.2022.827212 IRR = 0.76, 95%CI: 0.69–0.84, respectively); acute HF admissions also decreased in the Frontiers in Cardiovascular Medicine | www.frontiersin.org 1 April 2022 | Volume 9 | Article 827212 Álvarez-Martín et al. Cardiovascular Admissions During COVID-19 Pandemic period between waves (IRR: 0.81, 95%CI: 0.74–0.89). The impact was similar in all sex and socioeconomic subgroups and was higher in older patients with ACS. In-hospital mortality was higher than expected only during the first wave. Conclusion: During the first wave of the COVID-19 pandemic, there was a marked decline in urgent cardiovascular hospital admissions that were attenuated during the second wave. Both the decline and the attenuation of the effect have been similar in all subgroups regardless of age, sex, or socioeconomic status. In-hospital mortality for ACS and HF episodes increased during the first wave, but not during the second wave. Keywords: COVID-19, acute coronary syndrome, myocardial infarction, heart failure, time-series INTRODUCTION MATERIALS AND METHODS During the first pandemic wave (FW) of coronavirus disease 2019 Population and Data Sources (COVID-19) between March and April 2020, there was a marked We used region-wide data on public healthcare usage for the decline in admissions for acute coronary syndrome (ACS) (1– population of Catalonia, a North-East region in Spain with 11), heart failure (HF) (12–14), and other non-COVID health universal public healthcare coverage, with a reference population conditions (15–17). of more than 7 M and a hospital network of 11 reference tertiary While prioritization of healthcare services explains decreases hospitals performing percutaneous coronary intervention (PCI). in elective admissions, explanations for the decrease in the Data were accessed through the Public Data Analysis for Health attention of emergent conditions remain conjectural (18). Research and Innovation Program (PADRIS). PADRIS allows Possible reasons attributed to the steep decrease in hospital access to information from different sources linked at the patient admissions for acute cardiovascular problems were patients’ level with warranted accomplishment of ethical principles and hesitation to refer to the hospital, missed diagnosis, and the Spanish Law of data protection 3/2018. competing risk with COVID-19 mortality in the elderly and Acute coronary syndrome and HF episodes from 31 December institutionalized patients (19). The analysis of differing trends of 2018 to 27 December 2020 were extracted from the Minimum emergent cardiovascular admissions in successive waves periods Basic Data Set (MBD). Index admissions were identified with the and in the between-waves periods (BWP) might be useful to International Classification of Diseases (ICD)-10 as the primary unravel the underlying mechanisms and to anticipate future diagnostic (see diagnostic codes in Supplementary Table 1). We trends. A few reports of results during the second pandemic selected only emergency admissions. wave (SW, between April and September 2020) indicate a Weekly COVID-19 cases were obtained from the Open Access second decline of unknown magnitude (20) despite information Data Portal of the Catalan Government. Confirmed cases were campaigns targeted at the affected populations. defined as having a positive RT-PCR test or rapid antigen Moreover, the COVID-19 pandemic has been shown to test for SARS-CoV-2. increase geographic, gender, and socioeconomic inequalities in For each episode, we extracted sex, age in 5 years’ intervals, access to healthcare (21). Age, gender, and deprivation level all diagnostic and procedure ICD-10 codes, region of residence, might influence behaviors in such ways that would imply a primary care service areas (PCSA), comorbidity status (AMG), greater impact on more disadvantaged groups. Gender and age and PCSA-based socioeconomic status (PCSA index). differences in admissions for acute myocardial infarction (MI) have been analyzed previously with differing results (22); but to the best of our knowledge, it has not been assessed in longer Primary Care Service Areas Index series (beyond the FW) considering socioeconomic indices. The To strengthen territorial equity in the allocation of primary care potential influence of gender, age, and socioeconomic differences resources, in 2015, the Catalan Health Department developed a on the dynamic of admissions for acute HF decompensations has socioeconomic deprivation indicator representative of the PCSA not been assessed so far. linked to adverse health outcomes. The indicator score ranges Therefore, we aimed to evaluate the variations in ACS and from 0 (less deprived) to 100 (more deprived) (23). acute HF hospital admissions during the COVID-19 pandemic period, including the FW and the SW, in comparison with the hospital admissions of the corresponding pre-COVID-19 Comorbidity pandemic periods. In addition, we aimed to assess whether The comorbidity index [Adjusted Morbidity Groups (AMGs)] age, gender, and socioeconomic status might have influenced score is a morbidity measure that enables assigning a weight these variations. Finally, we analyzed the predictors of in- based on the preexisting comorbidities (24). AMG explanatory hospital death. value has been checked by comparing it with morbidity Frontiers in Cardiovascular Medicine | www.frontiersin.org 2 April 2022 | Volume 9 | Article 827212 Álvarez-Martín et al. Cardiovascular Admissions During COVID-19 Pandemic measures such as the Charlson Index or the number of chronic random slope models and non-structured covariance matrix for diseases (24). random effects to account for repeated episodes of individual patients and for the clustered structure of patients living in Statistical Analysis different PCSA (371 PCSA). All analyses were performed with Based on the visual representation of the weekly number of R version 4.1.0. COVID-19 diagnostics, we defined the following periods of analysis: pre-COVID (from 31 December 2018 to 23 February 2020), FW (from 24 February 2020 to 27 April 2020), BWP RESULTS (from 28 April 2020 to 20 September 2020), and SW (from 21 September 2020 to 27 December 2020). We considered the After excluding elective admissions and patients younger beginning of the FW as the date when the first case in the than 20 years old, we analyzed a series of 36,202 episodes region was reported in the press (24 February 2020), although the (corresponding to 34,575 patients) admitted for ACS (n = 8,424) COVID registry started later. or HF (26,151) between 31 December 2018 and 27 December Patients’ characteristics were compared between each COVID 2020 (Figure 1). period and the corresponding pre-COVID period (same calendar Tables 1, 2 show the characteristics of patients attended period in 2019). Continuous variables were reported as mean for ACS and HF during each predefined COVID-19 period as (SD) or median (P25 to P75) and compared using Student’s t-tests compared with each corresponding pre-COVID-19 period: FW, or Kruskal-Wallis tests. Categorical variables were reported as n BWP, and SW. The decline in admissions for every subgroup is (percentage) and compared using χ2 tests. presented as a% change. The description of the complete cohort Weekly admissions for ACS and HF were depicted from is shown in Supplementary Table 2. 31 December 2018 to 27 December 2020. Negative binomial In general, patients admitted during the pandemic period had regression was used to estimate the incidence rate ratios less comorbidities, as reflected by a consistently lower AMG (IRRs) and their 95% confidence intervals for ACS and acute score compared with the pre-COVID periods. In addition, in HF hospital admissions during the COVID-19 FW, during both conditions, the total hospital length of stay was significantly the BWP, and during the SW compared with the expected lower. Regarding ACS admissions (Table 1), the decline was admissions, based on the pre-COVID-19 period. Using this significantly higher for patients aged 80 years or older admitted approach, each IRR represents the ratio between the weekly during the SW. Diagnostic episodes labeled as other AMI or other number of admissions in each period and the weekly number ACS increased in the whole COVID-19 period. As compared of admissions in the reference period. First, we modeled the with the pre-COVID-19 period, the number of PCI remained pre-COVID-19 time series to estimate the expected number virtually the same during the FW but decreased slightly during of admissions in the COVID-19 periods by fitting sinusoidal the BWP (−23%; p = 0.001) and during the SW (−32%; functions to control seasonality and autocorrelation. Best-fitted p = 0.01). In-hospital mortality increased during the FW [8.86% models were selected on the basis of minimizing deviance vs. 5.51% in the pre-COVID-19 period (p = 0.018)], and there of the model and then fitted to the whole series including were no statistically significant differences neither during the COVID waves as indicators to calculate IRR. To show SW nor in the BWP. graphically the variations in ACS and HF hospital admissions Regarding acute HF admissions (Table 2), the proportion of attributable to the pandemic phases, we depicted the observed older people admitted during the BWP was significantly lower, vs. the expected admissions estimated from the pre-COVID- and in-hospital mortality was higher during the FW (7.41 vs. 19 model. 9.26%; p = 0.022) but lower during the BWP (7.46 vs. 6.38%; Finally, we fitted an additional set of models using fractional p = 0.039) and during the SW (7.63% vs. 5.94%; p = 0.007). polynomial coefficients to parameterize in a flexible way the Figure 2 shows the time series of ACS (Figure 2A) and HF different descending and ascending slopes observed during the (Figure 2B) weekly urgent hospital admissions and the distortion pandemic period. We used the parameters of the fitted models to observed after the arrival of the pandemic wave. Both ACS and estimate the instantaneous rate of change in the different periods HF admissions series showed a significant seasonality, with lower and the absolute decrease in each wave. admissions in summer weeks. After adjusting for seasonality and Different effects of COVID periods [FW, BWP, and SW autocorrelation, there was a marked decrease in ACS episodes among sex, age (<80/≥80), and PCSA index quartiles (first (Supplementary Figure 1A), reaching a maximum of 69.46% quartile/fourth quartile)] were estimated by including two-way cases below the expected [IRR 0.43 (0.37–0.50), p < 0.001] interaction terms in regression models. coinciding with the highest peak of the FW. Then, after a recovery To assess whether changes in hospital mortality during the in the BWP, the number of admissions decreased again in a COVID-19 period were due to the distortion in healthcare smoother way. Admissions for HF (Supplementary Figure 1B) during the pandemic or to changes in the risk profile of also decreased steeply to a 72.83% less than expected [IRR 0.21 patients, we performed logistic mixed-effects regression with in- (0.17–0.26), p < 0.001] coinciding with the peak of the FW and hospital mortality as the response variable and the pandemic then returned to normal levels to immediately decrease again to periods together as a covariable. We adjusted the model by a sustained level about 30% below the expected. patients’ characteristics [e.g., sex, age, comorbidity weight, type The model parameters used to estimate IRRs and changes in of ACS (in ACS model), PCSA index quartiles]. We used the weekly number of admissions are shown in Supplementary Frontiers in Cardiovascular Medicine | www.frontiersin.org 3 April 2022 | Volume 9 | Article 827212 Álvarez-Martín et al. Cardiovascular Admissions During COVID-19 Pandemic FIGURE 1 | Flowchart. Tables 3–6. The IRRs of ACS admissions for the FW, the 95%CI: 1.6–3.27; p < 0.001) and HF episodes (OR: 1.33; 95% BWP, and the SW were 0.66 (0.58–0.76), 0.92 (0.84–1.00), and CI: 1.11–1.59; p = 0.002). In addition, ACS episodes labeled as 0.80 (0.72–0.88), respectively. The IRRs of HF admissions for other MI had the highest mortality risk (8.9; 95% CI 5.44–14.58; the FW, the BWP, and the SW were 0.61 (0.55–0.68), 0.81 p < 0.001) as well as those episodes with a concomitant or recent (0.74–0.89), and 0.76 (0.69–0.84), respectively. Supplementary diagnosis of coronavirus infection (OR: 4.1; 95% CI: 1.78–9.44; Figures 2, 3 show the results of the models with fractional p = 0.001 and OR: 1.87; 95% CI: 1.22–2.86; p = 0.004 for ACS and polynomic coefficients for ACS and HF. HF, respectively). Supplementary Figures 4–15 show the separate time series according to gender, age, and area-level deprivation index. For the IRRs of ACS admissions, there was a significant interaction DISCUSSION with age in the BWP and in the SW (Figure 3), when the relative reduction in the weekly number of admissions was higher for This population-based time-trends analysis shows an important older people. Despite an apparent recovery in the admissions decrease in the number of emergent admissions for ACS and for ACS in the BWP [IRR 0.92 (0.84–1.00)], ACS admissions HF during the FW and during the SW compared with the remained lower than expected among women, older, and in most corresponding pre-COVID-19 periods. This decrease was also deprived areas. Regarding HF, there was a significant decline present during the BWP, and it was similar among socioeconomic of a similar magnitude in all subgroups (Figure 4), and HF subgroups. However, there were less patients aged 80 years or admissions remained significantly lower than expected during older admitted for ACS during the BWP and the SW. Adjusted the whole pandemic period. in-hospital mortality risk compared with the corresponding pre- Table 3 shows the association between the COVID-19 COVID-19 periods was higher during the FW for both ACS epidemic periods and in-hospital mortality adjusted for patient and HF episodes despite a lower comorbidity weight and a demographics, AMG score, type of ACS, and PCSA index. similar rate of PCI. Mortality risk, compared with the corresponding pre-COVID- Although the decrease in cardiovascular admissions was 19 periods, was higher during the FW for both ACS (OR: 2.29; consistent during all the study periods, the impact of the SW Frontiers in Cardiovascular Medicine | www.frontiersin.org 4 April 2022 | Volume 9 | Article 827212 Frontiers in Cardiovascular Medicine | www.frontiersin.org Álvarez-Martín et al. TABLE 1 | Characteristics of patients admitted to hospital for acute coronary syndrome (ACS) in each period compared with the reference pre-COVID-19 period. First wave period (24/02 to 27/04) Between waves period (28/04 to 20/09) Second wave period (21/09 to 27/12) Pre-COVID- COVID-19 % P-value Pre-COVID- COVID-19 % P-value Pre-COVID- COVID-19 % P-value 19 change 19 change 19 change N 817 519 −36 1788 1308 −12 1308 1001 −23 Women 250 (30.60) 171 (32.95) −32 0.368 617 (34.51) 505 (32.21) −18 0.159 449 (34.33) 331 (33.07) −26 0.526 Men 567 (69.40) 348 (67.05) −39 1171 (65.49) 1063 (67.79) −9 859 (65.67) 670 (66.93) −22 Age 0.638 0.352 0.001 < 80 579 (70.87) 374 (72.06) −35 1288 (72.04) 1152 (73.47) −11 883 (67.51) 738 (73.73) −16 ≥ 80 238 (29.13) 145 (27.94) −39 500 (27.96) 416 (26.53) −17 425 (32.49) 263 (26.27) −38 Type of ACS < 0.001 < 0.001 < 0.001 Unstable angina 143 (17.50) 84 (16.18) −41 272 (15.21) 274 (17.47) 1 167 (12.77) 167 (16.68) 0 NSTEMI 560 (68.54) 325 (62.62) −42 1249 (69.58) 997 (63.58) −20 944 (72.17) 671 (67.03) −29 STEMI 97 (11.87) 57 (10.98) −41 228 (12.75) 95 (6.06) −58 173 (13.23) 63 (6.29) −64 Other MI 8 (0.98) 43 (8.29) 438 14 (0.78) 161 (10.27) 1050 10 (0.76) 75 (7.49) 650 5 Other ACS 9 (1.10) 10 (1.93) 11 25 (1.4) 41 (2.61) 64 14 (1.07) 25 (2.50) 79 AMG weight, mean 32.92 (17.46) 20.44 (16.29) −38 <0.001 31.82 (16.47) 18.61 (14.44) −41 <0.001 31.19 (16.27) 17.16 (13.85) −45 <0.001 (SD) PCSA index, mean (SD) 42.11 (15.27) 42.68 (15.00) 1 0.514 42.20 (14.62) 41.24 (15.10) −2 0.064 41.79 (14.56) 42.05 (14.56) 1 0.680 Quantiles of PCSA 0.084 0.799 0.414 index 1st 202 (24.72) 116 (22.35) −43 422 (23.60) 394 (25.13) −7 324 (24.77) 221 (22.08) −32 2nd 188 (23.01) 113 (21.77) −40 428 (23.94) 376 (23.98) −12 306 (23.39) 262 (26.17) −14 Cardiovascular Admissions During COVID-19 Pandemic 3rd 187 (22.89) 139 (26.78) −26 432 (24.16) 370 (23.60) −14 305 (23.32) 236 (23.58) −23 4th 215 (26.32) 145 (27.94) −33 472 (26.40) 389 (24.80) −18 335 (25.61) 249 (24.88) −26 PCI during 353 (43.21) 242 (46.63) −31 0.220 826 (46.20) 633 (40.37) −23 0.001 619 (47.32) 420 (41.96) −32 0.010 hospitalization In−hospital mortality 45 (5.51) 46 (8.86) 2 0.018 69 (3.86) 73 (4.66) 6 0.253 60 (4.59) 48 (4.80) −20 0.814 April 2022 | Volume 9 | Article 827212 Hospital length of stay 9.41 (8.03); 7 7.11 (6.34); 5 −24 <0.001 8.88 (7.71); 7 8.29 (7.21); 6 −7 <0.001 9.47 (9.63); 7 8.03 (6.97); 6 −15 <0.001 (days), mean (SD); (5 to 11) (3 to 9) (4 to 11) (4 to 10) (4 to 11) (4 to 10) median (p25 to p75) Numbers indicate n (%) except if otherwise stated. Frontiers in Cardiovascular Medicine | www.frontiersin.org Álvarez-Martín et al. TABLE 2 | Characteristics of patients admitted to hospital for acute heart failure (HF) in each period compared with the reference pre-COVID-19 period. First wave period (24/02 to 27/04) Between waves period (28/04 to 20/09) Second wave period (21/09 to 27/112) Pre-COVID- COVID-19 % P-value Pre-COVID- COVID-19 % P-value Pre-COVID- COVID-19 % P-value 19 change 19 change 19 change N 3,200 1,749 −45 5,641 4,026 −29 4,090 2795 −32 Women 1,741 (54.41) 946 (54.09) −46 0.830 3130 (55.49) 2,200 (54.16) −30 0.195 2,315 (56.60) 1,525 (54.56) −34 0.094 Men 1,459 (45.59) 803 (45.91) −45 2,511 (44.51) 1,862 (45.84) −26 1,775 (43.40) 1,270 (45.44) −28 Age 0.107 0.001 0.061 <80 1,246 (38.94) 722 (41.28) −42 2,106 (37.33) 1,651 (40.65) −22 1,517 (37.09) 1,099 (39.32) −28 ≥80 1,954 (61.06) 1,027 (58.72) −47 3,535 (62.67) 2,411 (59.35) −32 2,573 (62.91) 1,696 (60.68) −34 AMGweight, mean (SD) 48.33 (15.39) 33.86 (18.22) −30 <0.001 48.74 (15.86) 31.18 (17.69) −36 <0.001 46.40 (15.73) 28.08 (15.59) −39 <0.001 6 PCSA index, mean (SD) 41.03 (15.08) 41.76 (14.66) 2 0.108 40.76 (15.27) 41.73(15.15) 2 0.002 40.73 (15.33) 40.96 (14.99) 1 0.556 Quantiles of PCSA 0.166 0.110 0.751 index 1st 836 (26.13) 426 (24.36) −49 1,534 (27.19) 1,008 (24.82) −34 1,137 (27.80) 754 (26.98) −34 2nd 790 (24.69) 408 (23.33) −48 1349 (23.91) 989 (24.35) −27 995 (24.33) 681 (24.36) −32 3rd 681 (21.28) 419 (23.96) −38 1,220 (21.63) 929 (22.87) −24 861 (21.05) 573 (20.50) −33 4th 790 (24.69) 445 (25.44) −44 1,370 (24.29) 1,006 (24.77) −27 986 (24.11) 712 (25.47) −28 In−hospital mortality 237 (7.41) 162 (9.26) −32 0.022 421 (7.46) 259 (6.38) −38 0.039 312 (7.63) 166 (5.94) −47 0.007 Cardiovascular Admissions During COVID-19 Pandemic Hospital length of stay 10.33 8.64 (7.32); 7 −16 <0.001 9.59 (8.66); 7 8.98 (7.75); 7 −6 0.017 9.77 (9.34); 7 8.47 (7.13); 6 −13 <0.001 (days), mean (SD); (10.04); 8 (4 (4 to 11) (4 to 12) (4 to 11) (4 to 12) (3 to 11) median (p25 to p75) to 13) Numbers indicate n (%) except if otherwise stated. April 2022 | Volume 9 | Article 827212 Álvarez-Martín et al. Cardiovascular Admissions During COVID-19 Pandemic FIGURE 2 | Weekly number of hospital admissions for (A) acute coronary syndrome and (B) heart failure, between 31 December 2018 and 27 December 2020. Black line: observed hospital admissions; shaded area: number of COVID-19 cases; red lines: predefined period limits: pre-pandemic, first wave, period between waves, and second wave; small circles: negative binomial model estimated values (pre-COVID-19). was clearly different from the impact of the FW. The impact of Besides patients’ hesitation to go to the hospital, this sustained the SW was quite attenuated, probably due to the rapid reaction decrease might be related in part to the fact that HF symptoms of hospitals to increase their intensive care capacity and to the are closer to COVID-19 symptoms. Since the primary focus of information campaigns launched to raise awareness of the need health professionals has been a rapid COVID-19 diagnosis, it to contact the emergency services when experiencing symptoms might have led to underdiagnoses of HF cases. In addition, acute compatible with MI. HF hospitalizations might have been lower because patients were There were also remarkable differences between ACS episodes preferably attended to in the primary care setting or due to the and acute HF episodes. While the number of ACS admissions increased use of telemedicine, which is less practicable in the gradually returned to the expected levels in the BWP, admissions setting of the acute ischemic heart disease. for acute HF remained low for almost the whole observed It has been speculated that lifestyle changes occurring during pandemic period studied. Whether the number of admissions the lockdown periods might have influenced acute coronary has recovered to usual values or has decreased again to similar disease incidence (25). However, the observations, as in other levels in the successive waves needs to be further addressed. contexts (8), that the decline in admissions preceded the Frontiers in Cardiovascular Medicine | www.frontiersin.org 7 April 2022 | Volume 9 | Article 827212 Álvarez-Martín et al. Cardiovascular Admissions During COVID-19 Pandemic FIGURE 3 | Incidence rate ratios of acute coronary syndrome admissions, in the three predefined pandemic period vs. the corresponding pre-pandemic period (2019) by subgroups of socioeconomic factors, and the p-value for the interaction between period and each socioeconomic factor. FIGURE 4 | Incidence rate ratios of acute heart failure admissions, in the three predefined pandemic period vs. the corresponding pre-pandemic period (2019) by subgroups of socioeconomic factors, and the p-value for the interaction between period and each socioeconomic factor. lockdown and that the return to “normal” levels preceded degree of competitive risk between COVID-19 disease and the reopening suggest that environmental changes, decreased cardiovascular events, which could result in a lower rate of physical activity, or diminished stress are unlikely to be major cardiovascular events in the population with a higher rate contributors to these trends. of severe COVID-19 disease, which was precisely the elderly Some changes in the characteristics of patients attended to and the more fragile population, especially during the FW. during the successive COVID-19 periods compared with the The stronger decrease in admissions of patients aged 80 years corresponding pre-COVID-19 periods might give some hints or older in the BWP and during the SW supports the for the interpretation of trends. Overall, patients admitted hypothesis of competitive risk with the high COVID-19-related during the COVID-19 period had much lower comorbidity mortality during the FW. weight assessed by AMG score. This suggests that those Another indirect finding that suggests the burden of the patients with more comorbidities and higher risk who had a healthcare facilities during certain periods was the hospital length cardiovascular event did not seek care or that the healthcare of stay. It declined steadily during the epidemic waves and, to a system collapsed at some point and could not hospitalize lesser extent, during the BWP. In the future, it will become clear these more fragile patients. Another possibility is a certain whether this indicates a simple trend in a complex context or a Frontiers in Cardiovascular Medicine | www.frontiersin.org 8 April 2022 | Volume 9 | Article 827212 Álvarez-Martín et al. Cardiovascular Admissions During COVID-19 Pandemic TABLE 3 | Association of patient characteristics and COVID-19 periods with highlighted HF as one of the health conditions not returning in-hospital mortality for acute coronary syndrome (ACS) and heart failure (HF). to baseline levels, together with respiratory and nervous system ACS valid (n* = 8,622) HF valid (n** = 27,531) diseases, cancer, acute exacerbations of chronic pulmonary disease, sepsis, and pneumonia. OR (95% CI) P-value OR (95% CI) P-value Although female sex and older age have been linked to increased delay before seeking healthcare (27), the COVID-19 Women 0.93 (0.75 to 1.15) 0.487 0.86 (0.78 to 0.95) 0.002 pandemic does not seem to have affected differently the pattern Age (for 5 years 1.44 (1.36 to 1.52) <0.001 1.30 (1.26 to 1.34) <0.001 increase) of admission for ACS and HF in male and female patients AMG weight 1.02 (1.01 to 1.02) <0.001 1.01 (1.00 to 1.01) <0.001 (Figures 3, 4). This suggests that the potential different behavior Type of ACS of men and women for seeking care shown in other studies is Unstable 1 - attenuated in a context such as COVID-19 pandemic. However, angina (ref) we observed that the decline in hospital admissions was higher Other ACS 2.81 (1.32 to 5.99) 0.007 in older patients, but only after the FW. Huynh et al. (28) NSTEMI 2.22 (1.51 to 3.26) <0.001 - found a higher reduction of acute MI in women ≥70 years old STEMI 6.13 (4.03 to 9.35) <0.001 - during the COVID-19 pandemic in a Swedish healthcare region Other MI 8.67 (5.32 to 14.13) <0.001 - without lockdown. Quartiles of To the best of our knowledge, this is the largest study showing PCSA index the effects of the COVID-19 pandemic on acute cardiovascular 1st 1 1 disease admissions during a sufficiently large time window to 2nd 1.00 (0.75 to 1.33) 0.993 1.05 (0.90 to 1.23) 0.501 include two successive waves. Previous reports have observed 3rd 1.19 (0.90 to 1.57) 0.223 1.10 (0.94 to 1.28) 0.245 lower hospital admission rates for MI (5, 6, 11, 29, 30) and HF 4th 1.09 (0.83 to 1.44) 0.525 1.05 (0.90 to 1.22) 0.553 (13, 14, 31, 32) during the FW that are comparable with our COVID19 study; however, reports of further declines in successive waves are period scarce. In this sense, Wu et al. (20) reported a second decline of a 1st wave 2.14 (1.50 to 3.05) <0.001 1.36 (1.14 to 1.62) 0.001 similar magnitude (41% for HF and 34% for MI) during the SW Between waves 1.20 (0.90 to 1.62) 0.217 0.94 (0.82 to 1.09) 0.432 in the United Kingdom. 2nd wave 1.37 (0.97 to 1.93) 0.072 0.86 (0.72 to 1.03) 0.094 *14 missing values for PCSA; **35 missing values for PCSA. Limitations ACS, acute coronary syndrome; HF, heart failure; OR, odds ratio; AMG, adjusted Despite the use of large region-wide administrative databases morbidity groups; NSTEMI, non-ST elevation myocardial infarction; STEMI, ST elevation myocardial infarction; MI, myocardial infarction; PCSA, primary that offer clear advantages, disadvantages include the following: care service areas. Lack of detail in clinical severity, comorbidities, out-of-hospital mortality and causes of mortality, lower data quality, and higher risk of misclassification. In the case of HF, it would real change in the efficiency of the health system for managing have been informative to merge hospital admissions to primary these cardiovascular events. In any case, it indicates the effort care emergent admissions and telecare attentions. Adoption of of the healthcare professionals to minimize the overload of telecare programs as a response to the pandemic might have the system and the awareness of the risk of infection during influenced the rate of hospital admissions. Finally, we used admission by professionals and by the patients and their families. related ICD-10 diagnostic codes to label cases with concomitant The increase in the number of diagnoses labeled as other COVID-19, as we did not have access to data for laboratory- ACS or other MI, present in only few cases, might be due proven SARS-Cov-2 infection. However, the proportion of to codification problems during the overwhelming working patients with a diagnosis of both cardiovascular disease and conditions of the professionals responsible for codification COVID-19 infection was low, and its impact on overall outcomes during the pandemic peaks. This fact prevents us to draw might not be very relevant. Finally, the depiction of the COVID conclusions about the differential changes by type of ACS. wave does not correspond with the real cases, especially during The analysis of mortality determinants reveals that, unlike the the FW, when fewer RT-PCR tests or rapid antigen tests for results observed in other series (26), there was a significantly SARS-CoV-2 were performed, and specific diagnostic codes for increased risk of in-hospital mortality after ACS and acute HF COVID-19 were not available. episodes in the FW but not in the SW. The increase in mortality was independent of case mix and despite a lower comorbidity weight and a similar rate of PCI compared with the pre-COVID- CONCLUSION AND PRACTICAL 19 period. This probably reflects, on the one hand, the delay in IMPLICATIONS patients seeking care increasing symptoms-to-balloon time (2), together with a more important healthcare overload during the The decline in hospital admissions for ACS and acute HF FW and, on the other hand, a certain learning curve effect that during the FW of the COVID-19 pandemic was marked and benefited the second and subsequent waves. striking. We provide evidence that the decline was attenuated Our results are consistent with those observed in other but still observed in the SW and that the impact has been contexts. The study by Bodilsen et al. (16), in Denmark, similar in all age, sex, and socioeconomic subgroups in the Frontiers in Cardiovascular Medicine | www.frontiersin.org 9 April 2022 | Volume 9 | Article 827212 Álvarez-Martín et al. Cardiovascular Admissions During COVID-19 Pandemic Catalan healthcare system. In addition, the adjusted in-hospital this study in accordance with the national legislation and the mortality risk comparing the corresponding pre-COVID-19 institutional requirements. periods shows a clearly higher risk during the FW for both ACS and HF episodes but not during the SW, which indicates the difficulty of the healthcare system to adapt to new and stressful AUTHOR CONTRIBUTIONS pandemic conditions. The impact of COVID-19 in Spain was among the strongest AR, AA-S, and IF-G conceived the study. AR received funds reported worldwide, and its indirect impact on other non- for the study. AR, JM, RL, and CÁ-M obtained the data. AR, COVID health conditions should be surveilled by health CÁ-M, JRM, and SP-H performed the data management and authorities. Numerous efforts have been made to forecast health statistical analysis. AR and IF-G have drafted the manuscript. resources needs for COVID-19 but not for other health problems All authors participated in the study design, contributed to the that have been strongly impacted by the pandemic. Extending interpretation of the results, revised it critically for important our temporal series to longer time frames, and including other intellectual content, and gave the final approval. contextual factors that might influence the rate of ACS or HF admissions, can serve to anticipate future distortions and prepare the health services to improve the attention of non-COVID FUNDING health problems. This study was funded with a grant from Sociedad Española de Cardiología y Fundación Española del Corazón (SEC/FEC-INV- DATA AVAILABILITY STATEMENT CLI 21/017). The funder had no role in the study development. The data analyzed in this study is subject to the following licenses/restrictions: Restrictions apply to the availability of ACKNOWLEDGMENTS these data, which were used under license for this study. Data are available from the authors upon request with the This study was carried out using anonymized data provided permission of AQuAS. Requests to access these datasets should by the Agency for Health Quality and Assessment of Catalonia be directed to https://aquas.gencat.cat/ca/ambits/analitica-dades/ (AQuAS), within the framework of the Public Data Analysis for padris/. Health Research and Innovation Program (PADRIS). ETHICS STATEMENT SUPPLEMENTARY MATERIAL The studies involving human participants were reviewed and The Supplementary Material for this article can be found approved by Clinical Research Ethics Committee Vall d’Hebron. online at: https://www.frontiersin.org/articles/10.3389/fcvm. Written informed consent for participation was not required for 2022.827212/full#supplementary-material REFERENCES pandemic. J Am Coll Cardiol. (2020) 75:2871–2. doi: 10.1016/j.jacc.2020 .04.011 1. Romaguera R, Ribera A, Güell-Viaplana F, Tomás-Querol C, Muñoz-Camacho 7. Woolf SH, Chapman DA, Sabo RT, Weinberger DM, Hill L. Excess deaths JF, Agudelo V. Decrease in ST-segment elevation myocardial infarction from covid-19 and other causes, March-April 2020. JAMA. (2020) 324:510–3. admissions in Catalonia during the COVID-19 pandemic. Rev Esp Cardiol. doi: 10.1001/jama.2020.11787 (2020) 73:778–80. doi: 10.1016/j.rec.2020.06.001 8. Mafham MM, Spata E, Goldacre R, Gair D, Curnow P, Bray M, et al. 2. Rodríguez-Leor O, Cid-Álvarez B, Pérez de Prado A, Rossello X, Ojeda S, COVID-19 pandemic and admission rates for and management of acute Serrador A, et al. Impact of COVID-19 on ST-segment elevation myocardial coronary syndromes in England. Lancet. (2020) 396:381–9. doi: 10.1016/ infarction care. The Spanish experience. Rev Esp Cardiol. (2020) 73:994–1002. S0140-6736(20)31356-8 3. Mesnier J, Cottin Y, Coste P, Ferrari E, Schiele F, Lemesle G, et al. 9. Gluckman TJ, Wilson MA, Chiu ST, Penny BW, Chepuri VB, Waggoner JW, Hospital admissions for acute myocardial infarction before and after lockdown et al. Case rates, treatment approaches, and outcomes in acute myocardial according to regional prevalence of COVID-19 and patient profile in France: infarction during the coronavirus disease 2019 pandemic. JAMA Cardiol. a registry study. Lancet Public Health. (2020) 5:536–42. doi: 10.1016/S2468- (2020) 5:1419–24. doi: 10.1001/jamacardio.2020.3629 2667(20)30188-2 10. Metzler B, Siostrzonek P, Binder RK, Bauer A, Reinstadler SJ. Decline of acute 4. Tam CCF, Cheung KS, Lam S, Wong A, Yung A, Sze M, et al. Impact coronary syndrome admissions in Austria since the outbreak of COVID-19: of coronavirus disease 2019 (COVID-19) outbreak on ST-segment-elevation the pandemic response causes cardiac collateral damage. Eur Heart J. (2020) myocardial infarction care in Hong Kong, China. Circ Cardiovasc Qual 41:1852–3. doi: 10.1093/eurheartj/ehaa314 Outcomes. (2020) 13:e006631. doi: 10.1161/CIRCOUTCOMES.120.006631 11. Solomon MD, McNulty EJ, Rana JS, Leong TK, Lee C, Sung SH, et al. The 5. De Filippo O, D’Ascenzo F, Angelini F, Bocchino PP, Conrotto F, Saglietto A, covid-19 pandemic and the incidence of acute myocardial infarction. N Engl J et al. Reduced rate of hospital admissions for ACS during Covid-19 outbreak Med. (2020) 383:691–3. in Northern Italy. N Engl J Med. (2020) 383:88–9. doi: 10.1056/NEJMc200 12. Reza N, DeFilippis EM, Jessup M. Secondary impact of the COVID-19 9166 pandemic on patients with heart failure. Circ Heart Fail. (2020) 13:e007219. 6. Garcia S, Albaghdadi M, Meraj P, Schmidt C, Garberich R, Jaffer doi: 10.1161/CIRCHEARTFAILURE.120.007219 FA, et al. Reduction in ST-segment elevation cardiac catheterization 13. König S, Hohenstein S, Meier-Hellmann A, Kuhlen R, Hindricks G, Bollmann laboratory activations in the United States during COVID-19 A, et al. In-hospital care in acute heart failure during the COVID-19 pandemic: Frontiers in Cardiovascular Medicine | www.frontiersin.org 10 April 2022 | Volume 9 | Article 827212 Álvarez-Martín et al. Cardiovascular Admissions During COVID-19 Pandemic insights from the German-wide Helios hospital network. Eur J Heart Fail. insights from Cardarelli Hospital, Naples, Southern Italy. Eur J Prev Cardiol. (2020) 22:2190–201. doi: 10.1002/ejhf.2044 (2021). [Online ahead of print]. doi: 10.1093/EURJPC/ZWAB165 14. Shoaib A, Van Spall HGC, Wu J, Cleland JGF, McDonagh TA, Rashid M, 27. Moser DK, McKinley S, Dracup K, Chung ML. Gender differences in et al. Substantial decline in hospital admissions for heart failure accompanied reasons patients delay in seeking treatment for acute myocardial infarction by increased community mortality during COVID-19 pandemic. Eur symptoms. Patient Educ Couns. (2005) 56:45–54. doi: 10.1016/j.pec.2003. Hear J Qual Care Clin Outcomes. (2021) 7:378–87. doi: 10.1093/ehjqcco/ 11.011 qcab040 28. Huynh J, Barmano N, Karlsson J-E, Stomby A. Sex and age differences in 15. Rosenbaum L. The untold toll - the pandemic’s effects on patients without the incidence of acute myocardial infarction during the COVID-19 pandemic Covid-19. N Engl J Med. (2020) 382:2368–71. doi: 10.1056/NEJMms2009984 in a Swedish health-care region without lockdown: a retrospective cohort 16. Bodilsen J, Nielsen PB, Søgaard M, Dalager-Pedersen M, Speiser LOZ, study. Lancet Healthy Longev. (2021) 2:e283–9. doi: 10.1016/s2666-7568(21)00 Yndigegn T, et al. Hospital admission and mortality rates for non-covid 085-4 diseases in Denmark during covid-19 pandemic: nationwide population based 29. De Rosa S, Spaccarotella C, Basso C, Calabrò MP, Curcio A, Filardi PP, cohort study. BMJ. (2021) 373:n1135. doi: 10.1136/BMJ.N1135 et al. Reduction of hospitalizations for myocardial infarction in Italy in the 17. Ullrich A, Schranz M, Rexroth U, Hamouda O, Schaade L, Diercke M, COVID-19 era. Eur Heart J. (2020) 41:2083–8. et al. Impact of the COVID-19 pandemic and associated non-pharmaceutical 30. Campo G, Fortuna D, Berti E, De Palma R, Di Pasquale G, Galvani M, et al. interventions on other notifiable infectious diseases in Germany: an analysis In- and out-of-hospital mortality for myocardial infarction during the first of national surveillance data during week 1–2016 – week 32–2020. Lancet Reg wave of the COVID-19 pandemic in Emilia-Romagna, Italy: a population- Health Eur. (2021) 6:100103. doi: 10.1016/j.lanepe.2021.100103 based observational study. Lancet Reg Health Eur. (2021) 3:100055. doi: 10. 18. Mercier G, Arquizan C, Roubille F. Understanding the effects of COVID-19 1016/j.lanepe.2021.100055 on health care and systems. Lancet Public Health. (2020) 5:e524. doi: 10.1016/ 31. Blecker S, Jones SA, Petrilli CM, Admon AJ, Weerahandi H, Francois S2468-2667(20)30213-9 F, et al. Hospitalizations for chronic disease and acute conditions in the 19. Cameli M, Pastore MC, Mandoli GE, D’Ascenzi F, Focardi M, Biagioni G, time of COVID-19. JAMA Intern Med. (2021) 181:269–71. doi: 10.1001/ et al. COVID-19 and acute coronary syndromes: current data and future jamainternmed.2020.3978 implications. Front Cardiovasc Med. (2021) 7:593496. doi: 10.3389/fcvm.2020. 32. Sokolski M, Gajewski P, Zymliñski R, Biegus J, Berg JMT, Bor W, 593496 et al. Impact of coronavirus disease 2019 (COVID-19) outbreak on 20. Wu J, Mamas MA, de Belder MA, Deanfield JE, Gale CP. Second decline in acute admissions at the emergency and cardiology departments across admissions with heart failure and myocardial infarction during the COVID- europe. Am J Med. (2021) 134:482–9. doi: 10.1016/j.amjmed.2020. 19 pandemic. J Am Coll Cardiol. (2021) 77:1141–3. doi: 10.1016/j.jacc.2020.12. 08.043 039 21. Abrams E, Szefler S. COVID-19 and the impact of social determinants of Conflict of Interest: The authors declare that the research was conducted in the health. Lancet Respir Med. (2020) 8:659–61. doi: 10.1016/S2213-2600(20) absence of any commercial or financial relationships that could be construed as a 30234-4 potential conflict of interest. 22. Moreno R, Alonso JJ, Caballero R, Del Corral E, Elízaga J, Asenjo RM, et al. Age and gender influence on time of arrival for STEMI patients during Covid-19 Publisher’s Note: All claims expressed in this article are solely those of the authors pandemic. Am J Emerg Med. (2021) 42:244–5. doi: 10.1016/j.ajem.2020.06.013 and do not necessarily represent those of their affiliated organizations, or those of 23. Colls C, Mias M, García-Altés A. Un índice de privación para reformar el the publisher, the editors and the reviewers. Any product that may be evaluated in modelo de financiación de la atención primaria en Cataluña. Gac Sanit. (2020) this article, or claim that may be made by its manufacturer, is not guaranteed or 4:44–50. doi: 10.1016/j.gaceta.2018.07.015 endorsed by the publisher. 24. Monterde D, Vela E, Clèries M. Adjusted morbidity groups: a new multiple morbidity measurement of use in Primary Care. Aten Primaria. (2016) 48:674– Copyright © 2022 Álvarez-Martín, Ribera, Marsal, Ariza-Solé, Pérez-Hoyos, 82. doi: 10.1016/j.aprim.2016.06.003 Oristrell, Soriano-Colomé, Romaguera, Pijoan, Lidón, Mauri and Ferreira-González. 25. Tsigkas G, Koufou EE, Katsanos K, Patrinos P, Moulias A, Miliordos I, et al. This is an open-access article distributed under the terms of the Creative Commons Potential relationship between lifestyle changes and incidence of hospital Attribution License (CC BY). The use, distribution or reproduction in other forums admissions for acute coronary syndrome during the covid-19 lockdown. Front is permitted, provided the original author(s) and the copyright owner(s) are credited Cardiovasc Med. (2021) 8:604374. doi: 10.3389/fcvm.2021.604374 and that the original publication in this journal is cited, in accordance with accepted 26. Scalise M, Bossone E, Mannelli MP, Ranieri B, Franzese M, Castaldo R, academic practice. No use, distribution or reproduction is permitted which does not et al. STEMI-PCI incidence rates and outcomes during COVID-19 pandemic: comply with these terms. Frontiers in Cardiovascular Medicine | www.frontiersin.org 11 April 2022 | Volume 9 | Article 827212