Introduction
The association between respiratory infections and cardiovascular sequalae are well established and cardiac arrhythmias are frequently diagnosed in patients with lower respiratory tract infection.1
Atrial fibrillation (AF) is a common cardiac arrythmia, responsible for at least a fifth of all ischaemic strokes in the United Kingdom,2 yet an individual's stroke risk can be reduced by about 65% with appropriate anticoagulation treatment.3 A first diagnosis of AF (nAF), discovered during admission to hospital with an infection is common.4 Furthermore, 8–11% of patients hospitalised with COVID-19 are demonstrated to have AF,5,6 with studies consistently demonstrating poor outcomes in this group of patients.7 However, the incidence of nAF in the setting of COVID-19 is reported only in a few studies.
We wanted to quantify the incidence of nAF during hospitalisation with COVID-19 compared with a lower respiratory tract infection (LRTI), as well as compare anticoagulation rates at discharge, reasons for non-prescription of anticoagulation and determine factors associated with developing nAF with COVID-19.
Methods
We retrospectively analysed collected coding data on patients hospitalised due to COVID-19+/-AF or LRTI +/-AF, including those with nAF, between 1 March 31 December 2020 at a university hospital in the UK.
Results
2,243 patients were hospitalised with LRTI and 488 with COVID-19. nAF was diagnosed in significantly more COVID-19 patients (COVID-19 +nAF) compared to LRTI (LRTI +nAF) (7.0% vs 3.6%, P=0.003). Excluding patients who died during their inpatient stay, and those who were already on anticoagulation at admission, significantly fewer COVID-19 +nAF patients were discharged on anticoagulation (19.2% vs 55.9%, P = 0.003); despite being younger, having similar CHA2DS2-VASc scores, and lower ORBIT scores compared to LRTI +nAF (Table 1).
Table 1.
Incidence, demographics and anticoagulation at discharge details of first-diagnosed AF in COVID-19 and LRTI patients
| LRTI (+ nAF) | COVID-19 (+ nAF) | p | |
|---|---|---|---|
| N (%) | 73 (3.6) | 31 (7.0) | 0.003* |
| Age (Q1–Q3) | 83 (74–90) | 75 (64–84) | 0.02 |
| Female (%) | 30 (41.1) | 10 (32.3) | 0.51 |
| Length of stay (Q1–Q3) | 10 (4–21) | 15 (8–28) | 0.16 |
| Discharged on anticoagulation, n (%) | 33 (55.9) | 5 (19.2) | 0.003 |
| Admission values | |||
| Hemoglobin (g\L) ±SD | 126.1 ± 22.8 | 124.7±33.5 | 0.81 |
| White cell count (X) (Q1–Q3) | 9.8 (8–14) | 6.5 (5–10) | 0.007 |
| Neutrophil count (X) (Q1–Q3) | 8.2 (6–12) | 5.4 (4–10) | 0.03 |
| C-Reactive protein (Q1–Q3) | 78.0 (19–135) | 61.2 (20–125) | 0.81 |
| NEWS score | 4 (2–7) | 3 (1–7) | 0.37 |
| Risk factors for thrombotic disease, n (%) | |||
| Hypertension | 46 (63.0) | 18 (58.1) | 0.80 |
| Congestive cardiac failure | 18 (24.7) | 7 (22.6) | 0.82 |
| Diabetes | 14 (19.2) | 7 (22.6) | 0.90 |
| IHD/PVD | 17 (23.3) | 11 (35.5) | 0.30 |
| Previous stroke | 7 (1.0) | 2 (6.5) | 0.90 |
| Median CHA2DS2-VASc score (Q1–Q3) | 3 (2.5–4) | 3 (1–5) | 0.32 |
| Median ORBIT score (Q1–Q3) | 2 (1–4) | 1 (0–2) | 0.02 |
| Events since discharge, n (%) | |||
| Deaths during admission | 17 (23.3) | 8 (25.8) | 0.98 |
| Outpatient cardiac monitoring | 1 (1.3) | 1 (3.2) | – |
| Myocardial infarction at 6 months | 0 | 1 (3.2) | – |
| Ischaemic stroke at 6 months | 0 | 0 | – |
| Haemorrhagic stroke at 6 months | 1 (1.4) | 0 | – |
| Deaths at 1 year | 34 (46.6) | 13 (41.9) | 0.83 |
Individual analysis of medical notes in patients not prescribed anticoagulation at discharge revealed that 14/26 (53.8%) LRTI +nAF patients had a clear contraindication not to be anticoagulated (9 limited life expectancy, 3 recent or current intracerebral haemorrhage, 1 CHA2DS2-VASc = 0, 1 end-stage renal failure), whereas this was only the case in 1/12 (8.3%) patients with COVID-19 +nAF (1 end-stage renal failure). In 10/26 (38.4%) LRTI +nAF and 8/12 (66.7%) COVID-19 +nAF, a reason for non-prescription of anticoagulation was not documented in the notes.
COVID-19 +nAF patients were older (P <0.001), had pre-existing congestive cardiac failure (P =0.004), ischaemic heart disease (IHD) or peripheral vascular disease (PVD) (P <0.001), and a higher CHA2DS2-VASc score (P =0.02) (Table 2). Older age (Odds ratio (OR) 1.03, P =0.007) and IHD/PVD (OR 2.87, P =0.01) increased the odds of developing nAF with COVID-19.
Table 2.
Factors associated with developing first-diagnosed AF during hospitalisation with COVID-19
| COVID-19 (No AF) | COVID-19 (+nAF) | p | |
|---|---|---|---|
| n | 411 | 31 (7.0) | – |
| Age (Q1–Q3) | 62 (46–79) | 75 (64–84) | <0.001 |
| Female (%) | 205 (49.9) | 10 (32.3) | 0.06 |
| Admission Values (Q1-Q3) | |||
| NEWS score | 3 (1–7) | 3 (1–7) | 0.29 |
| Length of stay (days) | 5 (1–12) | 15 (8–28) | <0.001 |
| Hemoglobin (g\L) ±SD | 129.5 ± 20.4 | 129.1±24.8 | 0.82 |
| C-Reactive protein (Q1-Q3) | 42 (13–104) | 61 (20–125) | 0.16 |
| Risk factors for thrombotic disease, n (%) | |||
| Hypertension | 173 (42.2) | 18 (58.1) | 0.08 |
| Congestive cardiac failure | 31 (7.5) | 7 (22.6) | 0.004 |
| Diabetes | 82 (20.0) | 7 (22.6) | 0.72 |
| IHD/PVD | 44 (10.7) | 11 (35.5) | <0.001 |
| Previous stroke* | 10 (2.4) | 2 (6.5) | 0.18 |
| CHA2DS2-VASc | 2 (1–3) | 3 (1–5) | 0.02 |
Conclusion
Higher incidence of nAF and lower anticoagulation rates in COVID-19 patients were observed, compared with LRTI. The majority of COVID-19 +nAF patients had no reason for non-prescription of anticoagulation documented in the notes, a much higher proportion than that of LRTI +nAF, despite CHA2DS2-VASc scores suggestive of a high future ischaemic stroke risk. Older patients with pre-existing cardiovascular risk factors are more likely to develop nAF during hospitalisation with COVID-19. While we await further research and clear guidelines, clinicians must holistically consider anticoagulation in all patients with COVID-19 and AF/nAF with a high ischaemic stroke risk.
References
- 1.Corrales-Medina VF, Musher DM, Wells GA, et al. Cardiac complications in patients with community-acquired pneumonia incidence, timing, risk factors, and association with short-term mortality. Circulation 2012;125:773–81. [DOI] [PubMed] [Google Scholar]
- 2.Royal College of Physicians Sentinel Stroke National Audit Programme (SSNAP) . National clinical audit annual results portfolio April 2020–March 2021. www.strokeaudit.org/results/Clinical-audit/National-Results.aspx (accessed 6 October 2022).
- 3.Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007;146:857–67. [DOI] [PubMed] [Google Scholar]
- 4.Walkey AJ, Wiener RS, Ghobrial JM, et al. Incident stroke and mortality associated with new-onset atrial fibrillation in patients hospitalized with severe sepsis. JAMA 2011;306: 2248–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Pellicori P, Doolub G, Wong CM, et al. COVID-19 and its cardiovascular effects: a systematic review of prevalence studies. Cochrane Database of Systematic Reviews 2021. Epub ahead of print 11 March 2021. DOI:10.1002/14651858.CD013879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Li Z, Shao W, Zhang J, et al. Prevalence of atrial fibrillation and associated mortality among hospitalized patients with COVID-19: a systematic review and meta-analysis. Front Cardiovasc Med 2021;8:1314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yang H, Liang X, Xu J, et al. Meta-analysis of atrial fibrillation in patients with COVID-19. Am J Card 2021;144:152–6. [DOI] [PMC free article] [PubMed] [Google Scholar]