Rates of Acute Myocardial Infarction During the COVID-19 Pandemic



 

Matthew T Mefford, PhD1; Jaejin An, PhD1,2; Nigel Gupta, MD3; Teresa N Harrison, SM1; Steven J Jacobsen, MD, PhD1,2; Ming-Sum Lee, MD4; Paul Muntner, PhD1,5; Chileshe Nkonde-Price, MD6,7; Lei Qian, PhD1; Kristi Reynolds, PhD, MPH1,2

Perm J 2021;25:21.074

https://doi.org/10.7812/TPP/21.074

ABSTRACT

Background: During the early phases of the COVID-19 pandemic pandemic, stay-at-home orders and fear of acquiring COVID-19 may have led to an avoidance of care for medical emergencies, including acute myocardial infarction (AMI). We evaluated whether a decline in rates of AMI occurred during the COVID-19 stay-at-home order.

Methods: Rates of AMI per 100,000 member-weeks were calculated for Kaiser Permanente Southern California patients from January 1 to March 3, 2020 (prepandemic period) and from March 20 to July 31, 2020 (pandemic period), and during the same periods in 2019. Rate ratios (RRs) were calculated comparing the time periods using Poisson regression. Case fatality rates (CFRs) were also compared.

Results: Rates of AMI were lower during the pandemic period of 2020 compared to the same period of 2019 [3.20 vs 3.76/ 100,000 member-weeks; RR, 0.85; 95% confidence interval (CI) 0.80–0.90]. There was no evidence that rates of AMI differed during the 2020 prepandemic period compared to the same period in 2019 (4.45 vs 4.24/100,000 member-weeks; RR, 0.95; 95% CI, 0.88–1.03). AMI rates were lower during the early pandemic period (March 20–May 7: RR, 0.70; 95% CI, 0.66–0.77), but not during the later pandemic period (May 8–July 31: RR, 0.95; 95% CI, 0.88–1.02) compared to 2019. In-hospital and 30-day case fatality rates were higher during the pandemic period of 2020 compared to 2019 (8.8% vs 6.1% and 6.5% vs 5.0%, respectively).

Conclusion: AMI rates were lower during the COVID-19 pandemic compared to the same period in 2019. During stay-at- home orders, public health campaigns that encourage people to seek care for medical emergencies are warranted.

INTRODUCTION

On March 4, 2020, the governor of California declared a state of emergency after the first confirmed COVID-19 death in the state.1 On March 13, 2020, the US government declared a national emergency, and on March 19, 2020, a stay-at-home order was enacted in California to slow the spread of COVID-19.2,3 After these orders, the number of individuals seeking health care declined dramatically, potentially from fear of contracting the virus.4–6 National surveil- lance data from the US Centers for Disease Control and Prevention and other health-care systems suggest emergency department (ED) visits declined during the COVID-19 pandemic.7–9 During the first 10 weeks of the pandemic, ED visits across the country declined by 42%, despite an increase in individuals presenting with symptoms of COVID-19.9 A study from Kaiser Permanente Northern California found that the incidence of hospitalization for acute myocardial infarction (AMI) declined more than expected from March 4 through April 14, 2020 compared to the same time period in 2019.10

Delays in urgent medical care may have downstream effects on patient outcomes. In prior studies, an increase in deaths from heart disease, stroke, and diabetes and an decrease in ED visits for these conditions were present.11,12 As these findings emerged, health-care systems and professional associations, including the American Heart Association, issued public awareness campaigns about seeking care for urgent medical conditions such as heart attacks.13–15 Estimating the impact of the stay-at-home orders on health-care use, and delays in care for acute emergencies can provide data to guide outreach and policies regarding the ongoing pandemic or future pandemics or natural disasters. The goal of our study was to examine the long-term impact of the pandemic on hospital admissions for AMI in Kaiser Permanente Southern California (KPSC). We also examined in-hospital and 30-day postdischarge mortality among adults hospitalized with AMI before and during the COVID-19 pandemic.

METHODS

Anonymized data that support the findings of our study may be made available from the investigative team with the following conditions: 1) agreement to collaborate with the study team on all publications, 2) provision of external funding for administrative and investigator time necessary for this collaboration, 3) demonstration that the external investigative team is qualified and has documented evidence of training for human subjects protections, and 4) agreement to abide by the terms outlined in data use agreements between institutions.

KPSC is an integrated health-care delivery system with approximately 4.6 million members within a service area comprising 10 Southern California counties and more than 20% of Southern California’s population. KPSC member- ship is diverse and widely representative of the Southern California region.16 For our study, we identified KPSC members 18 years of age or older during the prepandemic period of January 1 to March 13, 2020, the pandemic period of March 20 to July 31, 2020, and the same time periods in 2019. March 14 to 19, 2020 was considered a washout period given the gradual rollout of stay-at-home orders. Data for the washout period are not presented.

AMI was defined as an inpatient hospitalization or ED visit with a principal discharge diagnosis of International Classication of Diseases, Tenth Revision, Clinical Modication (ICD-10-CM) codes I21.x and I22.x from KPSC discharge records and claims.17 For AMI identified in the ED setting, we also required one of the following: 1) measured troponin I . 0.04 ng/mL, 2) death from any cause within 24 hours, or 3) percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) within 24 hours. The first AMI event per member in each period was included. Events occurring within 7 days of each other were combined into a single encounter. Members with less than 12 months of continuous enrollment, allowing up to a 45-day gap in coverage, prior to the date of their AMI (index date) were excluded. Separately, events were classified as ST-segment elevation myocardial infarction (STEMI) using ICD-10-CM codes I21.0, I21.01, I21.02, I21.09, I21.1, I21.11, I21.19, I21.2, I21.21, I21.29, I21.3, I22.0, I22.1, I22.8, and I22.9 or non-ST-segment elevation myocardial infarction (NSTEMI) using codes I21.4, I21.9, I21.A, I21.A1, I21.A9, and I22.2.18,19 The first STEMI and NSTEMI event per member per period was included for analyses specific to these subgroups.

Age on the date of the AMI event, gender, and self- reported race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, Asian/Pacific Islander, other) were identified from electronic membership records. Vital signs (eg, systolic and diastolic blood pressure, heart rate, and body mass index), smoking status, and laboratory measures (eg, total cholesterol, low-density lipoprotein cholesterol, high- density lipoprotein cholesterol, and triglycerides) were identified from outpatient nonurgent care settings most proximal to and within the 12 months prior to the index date. Troponin measures on the date of the ED visit or hospitalization were identified. Comorbidities including coronary heart dis- ease, history of AMI, unstable angina, prior CABG and PCI, ischemic stroke, heart failure, atrial fibrillation, hyper- tension, diabetes, chronic kidney disease, major depression, and history of malignancy were identified using ICD-10- CM codes for the 12 months prior to the index date. Medication fills for statins, nonstatin lipid-lowering agents, b-blockers, angiotensin converting enzyme inhibitors, angiotensin receptor blockers, and antiplatelet agents were identified in the 12 months prior to the index date from outpatient pharmacy records.

Care-related factors including length of stay for the AMI, discharge disposition, revascularization procedures (defined as receiving CABG or PCI up to 30 days including and after admission) were captured from patients’ medical records. Outcomes including in-hospital and, separately, 30-day postdischarge case fatality rates (CFRs) were identified from KPSC administrative and hospital sources.

Sociodemographic, clinical, and care-related characteristics of AMI cases during the prepandemic period of 2020 and the pandemic period of 2020 were compared to the same periods of time in 2019 using t-tests for means, the Mann-Whitney test for medians, and x2 tests for proportions. Rates of AMI per 100,000 member-weeks were calculated and plotted for the prepandemic and pandemic periods of 2020 (January 1–March 3 and March 20–July 31, respectively), and in the same periods of 2019. The weekly incidence of COVID-19 hospitalizations per 100,000 member-weeks was also plotted for the 2020 period. Absolute rate difference (RD), and rate ratios (RRs) and 95% confidence intervals (CIs) were calculated overall and by age group (< 65 years and ³ 65 years), gender, and race/ethnicity, comparing prepandemic and pandemic periods of 2020 to the same periods in 2019 using Poisson regression. In-hospital and 30-day CFRs were calculated using Mantel-Haenszel effect estimation. Analyses were repeated for STEMI and NSTEMI separately. In a sensitivity analysis, in-hospital and 30-day postdischarge CFRs were calculated among AMI cases, excluding patients with 1) laboratory-confirmed COVID-19 or 2) laboratory- confirmed COVID-19 or a COVID-19 diagnosis code in the 14 days prior to AMI admission through the 30-day postdischarge date.

In a secondary analysis, the pandemic period was split into March 20 to May 7, 2020 and May 8 to July 30, 2020, signifying the transition between the early pandemic period and phase 2 reopening in the state of California.20 Rates of AMI were compared to the same periods in 2019, and RD, RRs, and 95% CIs were calculated using Poisson regression. Analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC). All statistical tests were 2-sided, with a p value < 0.05 considered statistically significant. This study was approved by the KPSC institutional review board, and a waiver for informed consent was obtained.

RESULTS

There were 1271 and 2052 KPSC patients with AMI during the prepandemic and pandemic periods of 2020, respectively, compared with 1308 and 2388 patients during the same time in 2019 (Supplemental Figure 1) Patients who had an AMI during the prepandemic period of 2020 were older compared to those having an AMI during same time in 2019 (mean, 70.1 years vs 69.1 years; p = 0.05), but there was no evidence of differences with respect to gender, race/ethnicity, vital signs, laboratory measures, comorbidities, and medication use, or length of stay (all p > 0.05) (Table 1). There was no evidence of a difference with respect to age, gender, race/ethnicity, vital signs, laboratory measures and medication use, or length of stay between patients who had an AMI during the pandemic period of 2020 com- pared to those having an AMI during the same period in 2019. However, compared to patients who had an AMI during March 20 through July 31, 2019, those who had an AMI during the 2020 pandemic period were more likely to have prior AMI (4.2% vs 2.9%, p = 0.01) and less likely to have hypertension (53.7% vs 60.8%, p < 0.001), chronic kidney disease (16.2% vs 18.4%, p = 0.05), and major depression (12.9% vs 16.3%, p = 0.001).

Table 1. Baseline characteristics of acute myocardial infarction cases before and during the COVID-19 pandemic and during the same period in 2019
Baseline characteristics January 1–March 3, 2019 (n = 1308) January 1–March 3, 2020 (n = 1271) p Value March 20–July 31, 2019 (n = 2388) March 20–July 31, 2020 (n = 2052) p Value
Demographics
Age, y; mean (SD) 69.1 (14.0) 70.1 (13.3) 0.05 69.0 (13.8) 69.2 (13.7) 0.66
Age group, n (%) 0.02 0.53
< 65 years 479 (36.6) 409 (32.2) 878 (36.8) 736 (35.9)
³ 65 years 829 (63.4) 862 (67.8) 1510 (63.2) 1316 (64.1)
Male, n (%) 830 (63.5) 788 (62.0) 0.44 1504 (63.0) 1289 (62.8) 0.91
Race/ethnicity, n (%) 0.66 0.21
White 625 (47.8) 587 (46.2) 1120 (46.9) 1004 (48.9)
Hispanic 373 (28.5) 363 (28.6) 661 (27.7) 565 (27.5)
Non-Hispanic Black 149 (11.4) 141 (11.1) 287 (12.0) 228 (11.1)
Asian/Pacific Islander 143 (10.9) 156 (12.3) 271 (11.3) 201 (9.8)
Other 18 (1.4) 24 (1.9) 49 (2.1) 54 (2.6)
Vital signs, mean (SD)
Systolic BP, mmHg 131.5 (18.4) 131.4 (17.1) 0.93 131.3 (17.9) 131.8 (17.7) 0.34
Diastolic BP, mmHg 69.5 (12.9) 68.8 (12.9) 0.21 70.1 (12.9) 70.0 (13.0) 0.86
Heart rate, bpm 75.5 (14.0) 76.0 (13.8) 0.35 75.5 (14.3) 75.4 (13.6) 0.77
Body mass index, kg/m2 29.0 (6.2) 29.2 (6.6) 0.45 29.1 (6.5) 28.9 (6.2) 0.31
Smoking status, n (%) 0.82 0.35
Current 116 (8.9) 107 (8.4) 214 (9.0) 155 (7.6)
Former 521 (39.8) 511 (40.2) 923 (38.7) 807 (39.3)
Never 648 (49.5) 636 (50.0) 1211 (50.7) 1050 (51.2)
Unknown 23 (1.8) 17 (1.3) 40 (1.7) 40 (1.9)
Medication use prior to admission, n (%)
Statins 849 (64.9) 794 (62.5) 0.20 1465 (61.3) 1230 (59.9) 0.34
Nonstatin lipid-lowering therapy 67 (5.1) 62 (4.9) 0.78 83 (3.5) 79 (3.8) 0.51
b-Blockers 603 (46.1) 572 (45.0) 0.58 1075 (45.0) 884 (43.1) 0.20
ACE-I 423 (32.3) 377 (29.7) 0.14 800 (33.5) 674 (32.8) 0.64
ARB 298 (22.8) 287 (22.6) 0.90 496 (20.8) 403 (19.6) 0.35
Antiplatelets 273 (20.9) 265 (20.8) 0.99 473 (19.8) 414 (20.2) 0.76
Cardiovascular history, n (%)
Coronary heart disease 502 (38.4) 471 (37.1) 0.49 800 (33.5) 667 (32.5) 0.49
Prior AMI 99 (7.6) 76 (6.0) 0.11 69 (2.9) 87 (4.2) 0.01
Unstable angina 75 (5.7) 51 (4.0) 0.04 95 (4.0) 74 (3.6) 0.52
Coronary artery bypass graft 8 (0.6) 4 (0.3) 0.27 13 (0.5) 12 (0.6) 0.86
Percutaneous intervention 74 (5.7) 47 (3.7) 0.02 74 (3.1) 74 (3.6) 0.35
Ischemic stroke 58 (4.4) 42 (3.3) 0.14 83 (3.5) 89 (4.3) 0.14
Heart failure 203 (15.5) 191 (15.0) 0.73 282 (11.8) 234 (11.4) 0.67
Atrial fibrillation 143 (10.9) 144 (11.3) 0.75 233 (9.8) 180 (8.8) 0.26
Comorbidities, n (%)
Hypertension 804 (61.5) 791 (62.2) 0.69 1452 (60.8) 1102 (53.7) < 0.001
Diabetes 552 (42.2) 562 (44.2) 0.30 972 (40.7) 783 (38.2) 0.08
Chronic kidney disease 273 (20.9) 265 (20.8) 0.99 439 (18.4) 332 (16.2) 0.05
Major depression 192 (14.7) 174 (13.7) 0.47 389 (16.3) 264 (12.9) 0.001
Malignancy 88 (6.7) 73 (5.7) 0.30 139 (5.8) 131 (6.4) 0.43
 Laboratory measures
Total cholesterol, mg/dL; mean (SD) 165.3 (52.6) 170.5 (53.6) 0.07 173.0 (51.1) 174.5 (53.4) 0.50
HDL-C, mg/dL; mean (SD) 47.2 (14.7) 46.1 (13.6) 0.17 46.8 (13.7) 46.6 (14.1) 0.71
LDL-C, mg/dL; mean (SD) 92.9 (42.7) 96.4 (44.6) 0.15 98.5 (41.9) 99.7 (45.9) 0.54
Triglycerides, mg/dL; mean (SD) 152.7 (116.2) 164.1 (121.7) 0.09 160.5 (128.0) 164.9 (121.5) 0.43
Troponin, ng/mL; median (IQR) 1.9 (0.6–7.1) 1.9 (0.5–6.8) 0.41 1.8 (0.5–6.4) 1.9 (0.5–7.0) 0.21
Length of stay, d; mean (SD) 5.9 (7.3) 5.6 (5.9) 0.30 5.7 (6.3) 5.8 (6.4) 0.37

ACE-I = angiotensin converting enzyme inhibitor; AMI = acute myocardial infarction; ARB = angiotensin receptor blocker; BP = blood pressure; HDL-C = high-density lipoprotein cholesterol; IQR = interquartile range; LDL-C = low-density lipoprotein cholesterol; SD = standard deviation.

Rates of AMI in the prepandemic period of 2020 and the same period in 2019 were 4.24 and 4.45 per 100,000 member-weeks, respectively (Figure 1 and Table 2). During the pandemic period of 2020 and the same period in 2019, rates were 3.20 and 3.76 per 100,000 member-weeks, respectively. Although there was no evidence of a difference in AMI rates during the prepandemic period of 2020 com- pared to the same period in 2019 (RD, –0.21, 95% CI, –0.55 to 0.13; and RR, 0.95; 95% CI, 0.88–1.03), the rates of AMI were lower during the pandemic period of 2020 compared to the same period of 2019 (RD, –0.56; 95% CI, –0.77 to –0.35; and RR, 0.85; 95% CI, 0.80–0.90). These findings were consistent across age, gender, and race/ethnic subgroups. Rates of NSTEMI and STEMI were each lower during the pandemic period of 2020 compared to the same period in 2019 (NSTEMI: RD, –0.39; 95% CI, –0.58 to –0.21; and RR, 0.87; 95% CI, 0.81–0.93; STEMI: RD, –0.26; 95% CI, –0.38 to –0.14; and RR, 0.80; 95% CI, 0.72–0.88, respectively) (Supplemental Table 1).

Table 2. Rates per 100,000 member-weeks, rate ratios, and rate differences comparing acute myocardial infarction during the prepandemic and pandemic periods of 2020 with the same periods in 2019
   January 1–March 3, 2019 (n = 1308) January 1–March 3, 2020 (n = 1271) Rate ratio (95% CI) Rate difference (95% CI) March 20–July 31, 2019 (n = 2388) March 20–July 31, 2020 (n = 2052) Rate ratio (95% CI) Rate difference (95% CI)
Overall 4.45 4.24 0.95 (0.88–1.03) –0.21 (–0.55 to 0.13) 3.76 3.20 0.85 (0.80–0.90) –0.56 (–0.77 to –0.35)
Age group
< 65 years 1.99 1.69 0.85 (0.74–0.97) –0.30 (–0.55 to –0.05) 1.70 1.42 0.83 (0.75–0.92) –0.29 (–0.44 to –0.13)
³ 65 years 14.51 14.29 0.98 (0.89–1.08) –0.22 (–1.61 to 1.16) 12.06 10.14 0.84 (0.78–0.91) –1.92 (–2.75 to –1.10)
Gender
Male 6.02 5.64 0.94 (0.85–1.03) –0.38 (–0.95 to 0.20) 5.06 4.29 0.85 (0.79–0.91) –0.77 (–1.12 to –0.42)
Female 3.07 3.02 0.98 –0.05 (–0.44 2.63 2.27 0.86 –0.37 (–0.61
      (0.86–1.12) to 0.33)     (0.78–0.95) to –0.13)
Race/ethnicity
White 5.97 5.60 0.94 (0.84–1.05) –0.37 (–1.02 to 0.29) 4.98 4.53 0.91 (0.84–0.99) –0.45 (–0.85 to –0.04)
Black 5.95 5.56 0.93 (0.74–1.18) –0.39 (–1.71 to 0.94) 5.33 4.24 0.80 (0.67–0.95) –1.09 (–1.91 to –0.26)
Hispanic 3.07 2.87 0.94 –0.20 (–0.63 2.51 2.12 0.84 –0.39 (–0.65
      (0.81–1.08) to 0.23)     (0.75–0.94) to –0.13)
API 3.71 4.22 1.14 (0.90–1.44) 0.51 (–0.43 to 1.45) 3.46 2.46 0.71 (0.59–0.86) –1.00 (–1.55 to –0.44)
Other 2.09 2.04 0.98 (0.24–3.90) –0.05 (–2.91 to 2.81) 2.18 1.68 0.77 (0.29–2.07) –0.50 (–2.39 to 1.39)

AMI, acute myocardial infarction; API = Asian/Pacific Islander; CI = confidence interval.

In the time period between the issue of the stay-at-home order and phase 2 reopening (March 20–May 7, 2020), overall rates of AMI were lower compared to the same period in 2019 (RR, 0.70; 95% CI, 0.63– 0.77), and among patients with NSTEMI (RR, 0.72; 95% CI, 0.64–0.81) and STEMI (RR, 0.71; 95% CI, 0.60–0.85) (Supplemental Table 2). During the phase 2 reopening through the end of the study period (May 8–July 31, 2020), only rates of STEMI (RR, 0.85; 95% CI, 0.75–0.97) were lower compared with the same period in 2019.

tpj21074f1

Figure 1. Acute myocardial infarction (AMI) per 100,000 member-weeks before and during the COVID-19 pandemic and during the same period in 2019. Rates of confirmed AMI per 100,000 member-weeks were plotted using a 4-week moving average between January 1, 2020 and July 31, 2020, and the same time period in 2019. Overall, rates of AMI were not different in the prepandemic period of 2020 compared with the same time period in 2019. In contrast, rates of AMI were less in the pandemic period of 2020 compared with the same time in 2019. NSTEMI = non-ST-segment elevation myocardial infarction; STEMI = ST-segment elevation myocardial infarction.

There was no evidence of a difference in revascularization procedures among patients with AMI, NSTEMI, and STEMI, separately, comparing the prepandemic period of 2020 to the same time in 2019 (Figure 2). During the pandemic period of 2020 compared to the same time in 2019, there was no evidence of a difference in the proportion of patients with AMI receiving revascularization procedures; however, the proportion of patients having a CABG procedure was less (Figure 2). Revascularization procedures among STEMI patients were not different when comparing the pandemic period of 2020 with the same period in 2019.

tpj21074f2

Figure 2. Revascularization procedures among members with acute myocardial infarction (AMI) before and during the COVID-19 pandemic and during the same period in 2019. (A) The proportions of AMI, non-ST-segment elevation myocardial infarction (NSTEMI), and ST-segment elevation myocardial infarction (STEMI) patients receiving revascularization overall, and separately by  coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) during the prepandemic period of 2020 compared to 2019 were not different. (B) During the pandemic period of 2020, the proportion of CABG among AMI and NSTEMI patients was less, whereas procedures were not different among STEMI patients. *p < 0.05.

In-hospital CFRs during the pandemic period of 2020 and the same period in 2019 were 8.8% and 6.1%, respectively (p , 0.001) (Table 3). The 30-day postdischarge CFRs in the pandemic period of 2020 and the same period in 2019 were 6.5% and 5.0%, respectively (p 5 0.03). The in-hospital and 30-day postdischarge CFRs among patients with NSTEMI in the pandemic period of 2020 were higher compared with the same period in 2019 (7.8% vs 4.6%, p , 0.001, and 7.1% vs 5.2%, p 5 0.02, respectively). There was no evidence of a difference in CFRs among patients with STEMI or when comparing the prepandemic period of 2020 with the same period in 2019. Gender and race/ethnicity-specific in-hospital and 30-day CFRs are listed in Supplemental Table 3. Statistically significant increases in in-hospital CFRs during the pandemic period of 2020 com- pared to 2019 were observed among women and white and Hispanic patients; 30-day CFRs during the pandemic period of 2020 compared to 2019 increased among men, Hispanic, and black patients. Excluding patients with either laboratory-confirmed COVID-19 or a COVID-19 diagnosis did not change the overall CFR among AMI cases reported during the pandemic period, and rates remained higher compared to the same period in 2019 (Supplemental Table 4).

Table 3. Case fatality rates for acute myocardial infarction during the prepandemic and pandemic periods of 2020 and the same periods in 2019
  January 1–March 3, 2019 January 1–March 3, 2020 p Value March 20–June 30, 2019 March 20–June 30, 2020 p Value
AMI n = 1308 n = 1271 n = 2388 n = 2052
In-hospital CFR, n (%) 90 (7.0) 86 (6.7) 0.79 145 (6.1) 182 (8.8) < 0.001
30-Day post-discharge CFR, n (%) 71 (5.5) 67 (5.2) 0.68 119 (5.0) 135 (6.5) 0.03
NSTEMI n = 1035 n = 999 n = 1867 n = 1626
In-hospital CFR, n (%) 61 (6.0) 60 (6.0) 0.96 85 (4.6) 128 (7.8) < 0.001
30-Day postdischarge CFR, n (%) 61 (6.0) 53 (5.2) 0.41 94 (5.2) 117 (7.1) 0.02
STEMI n = 435 n = 430 n = 818 n = 665
In-hospital CFR, n (%) 41 (9.5) 42 (9.7) 0.91 80 (9.7) 76 (11.5) 0.28
30-Day postdischarge CFR, n (%) 16 (3.7) 19 (4.4) 0.64 38 (4.6) 35 (5.3) 0.52

AMI = acute myocardial infarction; CFR = case fatality rate; NSTEMI = non-ST-segment elevation myocardial infarction; STEMI = ST-segment elevation myocardial infarction.

DISCUSSION

In the current study, rates of AMI during the pandemic period of 2020 compared to the same time in 2019 were lower overall, within age, gender, and race/ethnicity groups. This was also observed among NSTEMI and STEMI, separately. The proportion of patients with AMI receiving CABG during the pandemic period of 2020 was lower com- pared with the same period in 2019, driven primarily by NSTEMI but not STEMI cases. CFRs for in-hospital and 30-day postdischarge mortality were higher during the pandemic period of 2020 compared to the same time in 2019.

   The CFR remained higher during the pandemic period of 2020 vs the same period in 2019 after excluding AMI cases with laboratory-confirmed or diagnosed COVID-19.

Prior studies have examined the impact of the COVID- 19 pandemic on rates of hospitalization for acute coronary syndromes including AMI. However, many were conducted in shorter time periods during the beginning of the COVID-19 pandemic and do not reflect changes in public health and clinical messaging that may have affected care- seeking behavior. A nationwide survey of PCI centers in Austria found admission rates for acute coronary syndromes decreased by 39% comparing the last week vs the first week of March 2020.21 In a study by Solomon et al10 at Kaiser Permanente Northern California, weekly rates of AMI decreased by 48% during the COVID-19 period of March 4 through April 14, 2020 compared to the same time in 2019, with comparable decreases observed among patients with STEMI and NSTEMI. In our study, rates of AMI at KPSC were 15% lower during the pandemic period of 2020 compared to the same period in 2019. This was largely explained by the early phase of the pandemic (March 20–May 7, 2020). A study of AMI hospitalizations in the UK found an early decline during Spring 2020 compared to the same periods in 2018 and 2019, and although hospitalizations increased during the summer months, a second decline was observed from October through mid-November 2020.22 Given this decline predated the second national lockdown in the UK, the authors suggest it likely reflects ongoing hesitation to seek care for medical emergencies during the pandemic.

In a retrospective cross-sectional study of AMI hospitalization from the Providence St Joseph Health system, Gluckman et al23 found that during the early period of COVID-19, from February 23 to March 28, AMI cases decreased by 19 cases per week for 5 consecutive weeks. Rates then increased by 10.5 cases each week from March 29 to May 16. Similar rebounds in AMI cases were present in a Canadian study.24 Our study provides further evidence that rates of AMI, STEMI, and NSTEMI were lower during the early phase of the pandemic period (March 20–May 7). Rates of AMI and NSTEMI were not different during the later portion of our study period (May 8–July 31) compared to the same period in 2019. Rates of STEMI, however, remained lower during this time. This means that as California entered phase 2 reopening, individuals with acute emergency conditions were presenting more regularly for medical care.

 Patients hospitalized during the pandemic period of 2020 were more likely to have prior AMI compared to those during the same period of 2019, suggesting that awareness— either through experience or education—could have influenced willingness to seek care for medical emergencies during the pandemic. This suggests that earlier or more routine patient education outside of pandemics may be important for minimizing gaps in care among high-risk adults.

Revascularization procedures overall were not lower during the pandemic period of 2020 compared with 2019. CABG procedures, however, which require extensive post- operative care and resources, including ventilators, intensive care unit care, personal protective equipment, blood products, anesthesia, and nursing support, were lower. Because of small sample sizes, these results should be interpreted with caution. In a prior study, there was a 38% reduction in cardiac catheterization lab activations for STEMI during the COVID-19 pandemic.25 Reductions in revascularization procedures were also noted outside the US.26,27 A recent study by Wadhera et al12 using National Center for Health Statistics data found an 11% and 17% increase in deaths attributable to ischemic heart disease and hypertensive dis- ease, respectively, between January and June 2020 compared to the same period in 2019. In our study, in-hospital and 30-day CFRs were higher during the pandemic period of 2020 compared to the same period in 2019, even after the exclusion of AMI cases with COVID-19.

Strengths of our study include the use of a large, diverse population of KPSC patients with comprehensive electronic health records to examine AMI hospitalization or ED visits, and revascularization and CFRs. In addition to calculating  in-hospital and 30-day postdischarge CFRs for AMI, STEMI, and NSTEMI, we also examined these outcomes among AMI cases with lab-confirmed or a diagnosis of COVID-19—an area in which current research is limited. We also acknowledge some potential limitations. First, death records from the state of California and Social Security Administration death files during the pandemic period were unavailable because of a time lag to obtain these data; therefore, we were limited in our ability to understand the  impact of the pandemic on more severe or fatal AMIs. Revascularization procedures were low, which is cause for  additional investigation, but may be attributable in part to incomplete coding or procedures occurring outside of Kaiser  Permanente hospitals that were not captured in the electronic health record. We did not examine AMI hospitalizations beyond July 2020, because this was beyond the scope of our study. However, more recent updates on AMI hospitalization during the last half of 2020 in the US and  abroad provide conflicting evidence regarding whether AMI cases have rebounded to prepandemic levels or continue to remain lower.22,28 Last, given the health-care structure of KPSC, in which the health plan, hospitals, and medical groups are integrated to create a system for  coordinated and comprehensive patient care, our patients may be healthier than the general population. Therefore,  these results may not be generalizable to individuals in less-integrated settings, which often have less robust quality improvement activities regarding cardiovascular  prevention, or among uninsured individuals.

In conclusion, AMI rates were lower during the first  months of the COVID-19 pandemic compared to the preceding year, and the mortality rate was higher among those  with AMI. The lower rates of AMI during the pandemic suggest that patients may have delayed care because of concerns of contracting COVID-19. Effective public health campaigns are needed to advise people to seek care for medical emergencies during stay-at-home orders.

Supplemental Material

Supplemental Material is available at: www.thepermanentejournal.org/files/2021/21.074supp.pdf.

Disclosure Statement

The authors have no conflicts of interest to disclose.

Author Affiliations

1 Department of Research & Evaluation, Kaiser Permanente Southern California, Pasadena, CA

2 Department of Health Systems Science, Kaiser Permanente Bernard J. Tyson School of Medicine, Pasadena, CA

3 Department of Cardiac Electrophysiology, Kaiser Permanente Los Angeles Medical Center, Los Angeles, CA

4 Department of Cardiology, Kaiser Permanente Los Angeles Medical Center, Los Angeles, CA 

5 Department of Epidemiology, University of Alabama at Birmingham, Birmingham, AL 

6 Department of Cardiology, Kaiser Permanente West Los Angeles Medical Center, Los Angeles, CA 

7 Department of Clinical Science, Kaiser Permanente Bernard J. Tyson School of Medicine, Pasadena, CA

Corresponding Author

Matthew T Mefford, PhD (matthew.t.mefford@kp.org)

Financial Support

This work was supported by the Southern California Permanente Medical Group.

Author Contributions

All authors approved the manuscript and contributed to its design, interpretation, and critical revisions.

Acknowledgments

We thank Ran Liu, MS, of the Kaiser Permanente Southern California Department of Research & Evaluation for her programming and statistical contributions to this study. We also thank the patients of Kaiser Permanente for helping to improve care via the use of information collected through our electronic health record systems.

References

1. Office of Governor Gavin Newsom. Governor Newsom declares state of emergency to help state prepare for broader spread of COVID-19. 2020. Accessed November 9, 2021. https://www.gov.ca.gov/2020/03/04/governor-newsom-declares-state-of-emergency-to-help-state-prepare-for-broader-spread-of-covid-19/

2. U.S. Department of Homeland Security. COVID-19 emergency declaration. 2020. Accessed October 22, 2020. https://www.fema.gov/news-release/20200726/covid-19- emergency-declaration

3. Office of Governor Gavin Newsom. Governor Gavin Newsom issues stay at home order. 2020. Accessed October 22, 2020. https://www.gov.ca.gov/2020/03/19/governor-gavin-newsom-issues-stay-at-home-order/

4. Mehrotra A, Chernew M, Linetsky D, Hatch H, Cutler D. The impact of the COVID-19 pandemic on outpatient visits: A rebound emerges. 2020. Accessed October 22, 2020. https://www.commonwealthfund.org/publications/2020/apr/impact-covid-19-outpatient-visits

5. Wong LE, Hawkins JE, Langness S, Murrell KL, Iris P, Sammann A. Where are all the patients? Addressing COVID-19 fear to encourage sick patients to seek emergency care. 2020. Accessed October 22, 2020. https://catalyst.nejm.org/doi/full/10.1056/CAT.20. 0193

6. Czeisler ME, Marynak K, Clarke KEN, et al. Delay or avoidance of medical care because of COVID-19-related concerns: United States, June 2020. MMWR Morb Mortal Wkly Rep 2020 Sep;69(36):1250–7. DOI: https://doi.org/10.15585/mmwr. mm6936a4

7. Lange SJ, Ritchey MD, Goodman AB, et al. Potential indirect effects of the COVID-19 pandemic on use of emergency departments for acute life-threatening conditions: United States, January–May 2020. MMWR Morb Mortal Wkly Rep 2020 Jun;69(25):795–800. DOI: https://doi.org/10.15585/mmwr.mm6925e2

8. Oseran AS, Nash D, Kim C, et al. Changes in hospital admissions for urgent conditions during COVID-19 pandemic. Am J Manag Care 2020 Aug;26(8):327–8. DOI: https://doi. org/10.37765/ajmc.2020.43837

9. Hartnett KP, Kite-Powell A, DeVies J, et al. Impact of the COVID-19 pandemic on emergency department visits: United States, January 1, 2019–May 30, 2020. MMWR Morb Mortal Wkly Rep 2020 Jun;69(23):699–704. DOI: https://doi.org/10.15585/mmwr. mm6923e1

10. Solomon MD, McNulty EJ, Rana JS, et al. The COVID-19 pandemic and the incidence of acute myocardial infarction. N Engl J Med 2020 Aug;383(7):691–3. DOI: https://doi. org/10.1056/NEJMc2015630

11. Woolf SH, Chapman DA, Sabo RT, Weinberger DM, Hill L, Taylor DDH. Excess deaths from COVID-19 and other causes, March–July 2020. JAMA 2020 Oct;324(15):1562–4. DOI: https://doi.org/10.1001/jama.2020.19545

12. Wadhera RK, Shen C, Gondi S, Chen S, Kazi DS, Yeh RW. Cardiovascular deaths during the COVID-19 pandemic in the United States. J Am Coll Cardiol 2021 Jan;77(2): 159–69. DOI: https://doi.org/10.1016/j.jacc.2020.10.055

13. Lindberg E. Get the care you need during COVID-19, health experts urge. 2020. Accessed October 22, 2020. https://news.usc.edu/171287/get-medical-care-covid-19- pandemic-usc-health-experts/

14. Harrington RA, Poppas A, Albert M, et al. The new pandemic threat: People may die because they’re not calling 911. 2020. Accessed October 22, 2020. https://newsroom. heart.org/news/the-new-pandemic-threat-people-may-die-because-theyre-not-calling-911

15. BetterTogether.Health. Get care when you need it. 2020. Accessed November 9, 2021. https://la.bettertogether.health/p/1

16. Koebnick C, Langer-Gould AM, Gould MK, et al. Sociodemographic characteristics of members of a large, integrated health care system: Comparison with US Census Bureau data. Perm J 2012 16(3):37–41. DOI: https://doi.org/10.7812/TPP/12-031

17. Reynolds K, Go AS, Leong TK, et al. Trends in incidence of hospitalized acute myocardial infarction in the Cardiovascular Research Network (CVRN). Am J Med 2017 Mar;130(3):317–27. DOI: https://doi.org/10.1016/j.amjmed.2016.09.014

18. Yeh RW, Sidney S, Chandra M, Sorel M, Selby JV, Go AS. Population trends in the incidence and outcomes of acute myocardial infarction. N Engl J Med 2010 Jun; 362(23):2155–65. DOI: https://doi.org/10.1056/NEJMoa0908610

19. Carr C, Romanello L. Coding an acute myocardial infarction: Unraveling the mystery. 2018. Accessed January 11, 2021. https://acdis.org/resources/coding-acute-myocardial- infarction-unraveling-mystery

20. Office of Governor Gavin Newsom. Governor Newsom provides update on California’s progress toward stage 2 reopening. 2020. Accessed November 11, 2020. https://www.gov.ca.gov/2020/05/04/governor-newsom-provides-update-on-californias-progress-toward- stage-2-reopening/

21. Metzler B, Siostrzonek P, Binder RK, Bauer A, Reinstadler SJ. Decline of acute coronary syndrome admissions in Austria since the outbreak of COVID-19: The pandemic response causes cardiac collateral damage. Eur Heart J 2020 May;41(19):  1852–3. DOI: https://doi.org/10.1093/eurheartj/ehaa314

22. Wu J, Mamas MA, de Belder MA, Deanfield JE, Gale CP. Second decline in admissions with heart failure and myocardial infarction during the COVID-19 pandemic. J Am Coll  Cardiol 2021 Mar;77(8):1141–3. DOI: https://doi.org/10.1016/j.jacc.2020.12.039

 23. Gluckman TJ, Wilson MA, Chiu ST, et al. Case rates, Treatment approaches, and outcomes in acute myocardial infarction during the coronavirus disease 2019 pandemic.  JAMA Cardiol 2020 Dec;5(12):1419–24. DOI: https://doi.org/10.1001/jamacardio.2020.3629

24. Fardman A, Oren D, Berkovitch A, et al. Post COVID-19 acute myocardial infarction  rebound. Can J Cardiol 2020;36(11):1832.e15–1832.e16. DOI: https://doi.org/10.1016/j.  cjca.2020.08.016

25. Garcia S, Albaghdadi MS, Meraj PM, et al. Reduction in ST-segment elevation cardiac catheterization laboratory activations in the United States during COVID-19 pandemic. J Am Coll Cardiol 2020 Jun;75(22):2871–2. DOI: https://doi.org/10.1016/j. jacc.2020.04.011

26. Piccolo R, Bruzzese D, Mauro C, et al. Population trends in rates of percutaneous coronary revascularization for acute coronary syndromes associated with the COVID-19 outbreak. Circulation 2020 Jun;141(24):2035–7. DOI: https://doi.org/10.1161/ CIRCULATIONAHA.120.047457

27. Kwok CS, Gale CP, Curzen N, et al. Impact of the COVID-19 pandemic on percutaneous coronary intervention in England: Insights from the British Cardiovascular Intervention Society PCI database cohort. Circ Cardiovasc Interv 2020 Nov;13(11): 210–221. DOI: https://doi.org/10.1161/CIRCINTERVENTIONS.120.009654

28. Solomon MD, Nguyen-Huynh M, Leong TK, et al. Changes in patterns of hospital visits for acute myocardial infarction or ischemic stroke during COVID-19 surges. JAMA 2021 Jul;326(1):82–4. DOI: https://doi.org/10.1001/jama.2021.8414

Keywords: acute myocardial infarction, COVID-19, epidemiology, rates

ETOC

Click here to join the eTOC list or text ETOC to 22828. You will receive an email notice with the Table of Contents of The Permanente Journal.

CIRCULATION

2 million page views of TPJ articles in PubMed from a broad international readership.

Indexing

Indexed in MEDLINE, PubMed Central, EMBASE, EBSCO Academic Search Complete, and CrossRef.


                                             

 

 

ISSN 1552-5775 Copyright © 2021 thepermanentejournal.org

All Rights Reserved