Introduction

Coronavirus disease 2019 (COVID-19) is a disease caused by the severe acute respiratory syndrome (SARS-CoV-2) virus that causes a spectrum of mild-to-severe illness, manifesting most commonly within the hospitalized critically unwell population as a severe pneumonitis progressing to respiratory failure and multiorgan dysfunction. On March 11, 2020, the World Health Organization declared a global COVID-19 pandemic which was followed by national lockdowns, mass hospitalizations, and a significant expansion in critical care capacity to facilitate the use of mechanical ventilation.¹ The global research effort has since focused on establishing effective treatment options, as well as biochemical and bedside prognostic markers of disease severity.

Fragmentation of the QRS complex (fQRS), characterized by various RSR’ or notching patterns, is a depolarization disorder that can be detected from routine 12-lead electrocardiography (ECG) (Figure 1). In a general population, the prevalence of fQRS was found to be 19.7%, with a greater prevalence observed in those with underlying cardiac disease.² There is established evidence that the presence of fQRS on ECG is strongly indicative of myocardial scarring in patients with coronary artery disease and a marker of further cardiac events.^3,4 It is pathologically also present in those with cardiomyopathy, valvular, and congenital heart disease.⁵ Fragmented QRS has previously been shown to have a high positive predictive value and specificity for the presence of dysfunctional segmental myocardial contractility.⁶ Until recently its significance in the COVID-19 population had not been assessed, but there is emerging evidence in these patients that fQRS is linked to an increased risk of all-cause mortality and need for invasive mechanical ventilation.^7,8

In contrast to previous studies that have focused on the cardiac population, our focus pertained to the critically ill COVID-19 population. We aimed to explore the relationship between fQRS and all-cause mortality, and to investigate the significance of associated cardiac structural and functional abnormalities on transthoracic echocardiography (TTE) in this population.

Methods

Population and Demographics

This was a retrospective case-cohort study of consecutive patients treated for COVID-19 in a critical care setting (high dependency or intensive care unit) at our institution in the United Kingdom between February 2020 and May 2021. Data were collected from the local electronic patient health record. Those who had pre-existing ceilings of treatment were excluded. Patients with active COVID-19 were defined as those with positive real-time reverse transcriptase-polymerase chain reaction assay for SARS-CoV-2 on nasopharyngeal swab. Patient demographics and comorbidities classified using the Charlson Comorbidity Index were extracted from patient records.⁹

Primary and secondary outcomes were in-hospital mortality and the need for mechanical invasive ventilation respectively. Known prognostic laboratory markers of cytokine storm were also recorded. These included peak values for platelets, leukocytes, neutrophils, C-reactive protein (CRP), lactate dehydrogenase (LDH), ferritin, troponin T, D-dimer, creatinine kinase (CK), alanine aminotransferase (ALT), and trough values for lymphocyte count.¹⁰ Where values were above or below the range quantifiable by the laboratory, the highest or lowest quantifiable value was used, respectively.

Electrocardiography

Standard 12-lead surface ECG (0.5-150 Hz, 25 mm/s, 10 mm/mV) recordings performed throughout the admission were collated. All patients had ECGs recorded on initial presentation to the hospital, on admission to the critical care unit, and at subsequent points thereafter at the treating team’s discretion. For each patient, the longest PR interval (in ms), longest QRS duration (in ms), largest ST segment change (in mm), and longest Bazett’s corrected QT interval (in ms) measured throughout the admission were recorded. Any new arrhythmic events were also recorded.

In ECGs with a QRS duration ≤120 ms, fQRS was defined as notching in the R or S wave, RSR’ pattern or multiple R’ (Figure 1).¹¹ In patients with a QRS >120 ms, fQRS was defined as the presence of >2 notches on the R wave or the S wave.¹² Fragmented QRS was considered present if these changes appeared on at least 1 ECG recording and in at least 2 consecutive leads corresponding to the major coronary artery territories. The territories were categorized as anterior (V1-V5), lateral (I, aVL, and V6), and inferior (II, III, and aVF).

Three authors independently analyzed all ECGs performed on all patients while being blinded to the patient’s outcome to assess for the presence of fQRS. Where there was disagreement between the initial assessors, the ECG was evaluated by a fourth senior cardiologist author to serve as a tiebreak. Agreement between the initial assessors was high, with 9% of ECGs sent to the fourth author.

Echocardiography

Transthoracic echocardiography was performed at the critical care team’s discretion by experienced British Society of Echocardiography (BSE) accredited technicians according to the BSE COVID-19 consensus pathway.¹³ The time interval between ECG and TTE was based on clinical need during the critical care admission. Technically inadequate studies were excluded. Left ventricular ejection fraction was calculated by the Simpson’s biplane method. Diastolic dysfunction was defined as present if severity was grade II or above according to European Association of Cardiovascular Imaging (EACVI) guidelines.¹⁴ A dilated left ventricle (LV) internal end-systolic diameter (LVIDs) was defined as >41 mm in males and >37 mm in females, while a dilated LVID (LVIDd) was defined as >56 mm in males and >51 mm in females.¹⁵ Left atrial dilatation was defined as a volume of >54 mL. The presence of any regional wall abnormalities was also recorded.

The right atrium (RA) and right ventricle (RV) were assessed in a dedicated focused view. Right atrial dilatation was defined as an area of >22 cm² for males and >19 cm² for females.¹⁵ Right ventricle dilatation was defined as a basal diameter >47 mm in males and >43 mm in females. Right ventricle systolic dysfunction was defined as a fractional area change <35% in females or <30% in males, a tricuspid annular plane systolic excursion <17 mm or an RV s′ <9 cm/s. Pulmonary hypertension was marked as present if the echocardiographic probability of pulmonary hypertension was high: defined as peak tricuspid regurgitation velocity (TRV) > 3.4 m/s or peak TRV 2.9-3.4 m/s with ≥2 supporting signs of pulmonary hypertension.¹⁶

Statistical Analysis

Continuous variables were assessed for normality using Kolmogorov–Smirnov and Shapiro–Wilk tests. When normally distributed, continuous variables were expressed as mean ± SD and compared using the independent samples t-test. Non-normal data were presented as median with interquartile range and compared using the Mann–Whitney U-test. Categorical variables were summarized as frequencies and percentages and compared using either the Fisher’s exact test or chi-squared test. A multiple logistic regression analysis was carried out using covariates implicated in COVID-19 patient outcomes. Mortality was used as a binary outcome, against which the recorded parameters were assessed. All parameters were included in the analysis; however, peripheral vascular disease and malignancy were excluded due to the low incidence among the study population. Significance was determined by a confidence interval entirely greater than 1 and a P-value less than .05. All statistical analyses were performed using GraphPad Prism 9 for MacOS using a P-value of less than .05 to determine statistical significance.

Ethics

This study was approved by our Research, Quality Improvement, and Audit department with the reference number FH205. All data were collected locally, anonymized, and handled in accordance with local data protection guidelines.

Results

A total of 212 patients were included for this study (Figure 2), with 6% of those treated for COVID-19 excluded from the final analysis due to incomplete data. Baseline characteristics and clinical data of all patients in our cohort are summarized in Table 1. In this study population, 120 patients (57%) exhibited fQRS during their admission. There was a high incidence of mortality and a requirement for invasive mechanical ventilation (58% and 71% of the total population respectively). Older age was a significant predictor of mortality but not invasive mechanical ventilation. Gender and ethnicity were not significant predictors of mortality or invasive mechanical ventilation. Overall, the most common comorbidities were hypertension and diabetes mellitus (47% and 42% of the total population, respectively). There was a low prevalence of pre-existing heart failure in the patient cohort overall (3%). Diabetes mellitus, ischemic heart disease and the Charlson Comorbidity Score were significantly associated with mortality; however, neither specific comorbidities nor overall co-morbidity burden were associated with invasive mechanical ventilation. Certain laboratory markers were associated with mortality, including peak leukocyte, peak neutrophil, trough lymphocyte, peak CRP, peak troponin T, and peak LDH concentrations. Peak CRP concentrations were associated with the need for invasive mechanical ventilation.

Presence of fQRS was significantly associated with mortality (65% presence in mortality group vs. 44% in survivor group, P = .003). The inferior leads (II, III, and aVF) were most commonly involved (88% of patients) (Table 3) and inferior fQRS alone was a significant predictor of mortality (P = .03). On multivariate logistic regression, fQRS remained a significant independent predictor of mortality (odds ratio (OR) = 2.96, 95% CI: 1.42-6.40, P = .005) along with age (OR = 1.07, 95% CI: 1.02-1.13, P = .01) and peak leukocyte count (OR = 1.35, 95% CI: 1.05-1.83, P = .03) (Table 2). Connective tissue disease was negatively associated with mortality.

Upon assessing other ECG parameters in those with fQRS compared to those without, there was no significant difference in PR interval, QRS duration, ST segment change, and QTc interval (Table 3). There was also no significant difference in the incidence of arrhythmic events. In total, 112 patients had TTE performed. These patients were not associated with any significant increase in mortality compared to those who did not undergo TTE (58% mortality in TTE group vs. 59%, P = .89). Demographic data were also similar between those who underwent TTE compared to those who did not. However, there was a greater incidence of diabetes mellitus in the TTE group as well as higher peak troponin T and CRP concentrations and lower trough lymphocyte counts. There was a greater incidence of RV dilatation (16% vs. 2%, P = .02) and pulmonary hypertension (33% vs. 14%, P = .03) based on echocardiographic probability criteria in the fQRS group (Table 3). Notably, there was no difference in RV or LV function between the 2 groups nor any difference in other echocardiographic parameters.

Discussion

The presence of fQRS reflects myocardial conduction abnormality. It arises when myocardial tissue is replaced by scar or fibrosis, resulting in an irregular course of myocardial activation that retards or blocks conduction.¹⁷ Slow conduction within these zones facilitates re-entry, predisposing to the development of ventricular tachyarrhythmias.¹⁸ It has been proposed that the manifestation of fQRS may be a surrogate marker of cardiac events among patients with coronary artery disease¹² and its presence might be utilized as a prognosticator of response to treatment in patients with cardiomyopathy.¹⁹

Our study clearly demonstrates that fQRS is a strong predictor of mortality in COVID-19, corroborating recent literature.^7,8,20 This is the first study that we are aware of demonstrating a significant relationship between fQRS, RV dilatation and pulmonary hypertension on TTE. Right ventricle dilatation corresponded to fQRS within the inferior leads in 88% of patients. We propose that fQRS in patients with severe COVID-19 arises from structural changes within the RV in response to pressure and volume overload leading to aberrant conduction and an abnormal depolarization segment. We suggest that RV dilatation and pulmonary hypertension translates to the higher mortality seen within this sub-group. To date, a few retrospective analyses with small sample sizes have corroborated our findings of increased all-cause mortality and need for invasive mechanical ventilation in COVID-19 patients exhibiting fQRS on ECG.^7,8,20 Özdemir et al ⁷ and Bektas et al²⁰ hypothesized that the higher mortality observed in those with fQRS was related to cytokine storm and subsequent myocardial injury. The aforementioned studies were conducted relatively early in the pandemic, whereas our timescale looked at data from patients as late as May 2021, when the RECOVERY trial had led to the protocolization of treatment in COVID-19 with dexamethasone (June 2020) and interleukin 6 (IL-6) inhibitors tocilizumab and sarilumab (February 2021).^21,22 While this study demonstrated an association between certain markers and mortality (namely peak leukocyte, peak neutrophil, trough lymphocyte, peak CRP, peak troponin T, and peak LDH concentrations), novel treatments may have contributed to cytokine storm playing less of a role in our study group.

To our knowledge, our study is the first to incorporate TTE and electrocardiographic data to explore the relationship between cardiac structure, function, and fQRS in COVID-19. While the link between myocardial scarring and fQRS is increasingly accepted in cardiac patients, we suggest the mechanism of fQRS in the COVID-19 critical care population is more complex. Here, several pathophysiological processes likely contribute to RV dilatation and pulmonary hypertension, either as a direct consequence of COVID-19 pneumonitis, or in combination with pre-existing respiratory pathology and mechanical ventilation. Hypoxemia from pneumonitis increases RV afterload via pulmonary vasoconstriction which is compounded in latter stages by poor CO₂ clearance as the lung loses compliance. Mechanical ventilation necessitates the generation of higher mean airway pressures and alveolar overdistension, which contribute to pulmonary artery hypertension, RV pressure overload followed by volume overload, and subsequent failure. In situ pulmonary thrombosis and coronary artery hypoperfusion in hypovolemic septic shock also directly attenuate RV function.²³ Finally, acutely elevated RV pressure can decrease the preload of the LV and lead to RV ischemia or infarction.

We therefore propose that in critically ill COVID-19 patients, especially those ventilated, fQRS is not only a predictor of mortality but of structural changes in the RV with accompanying pulmonary arterial hypertension. We postulate that fQRS in this context is a phenomenon of abnormal myocardial activation arising from structural and electrical changes within the RV in response to pressure and volume overload, and from the cumulative effect of the above physiological processes. It is not unreasonable to assume that areas of redundant myocardium, irrespective of underlying pathomechanisms, remain substrate for reentry and ventricular tachycardias. While not studied formally, in some of our study patients fQRS was transient or episodic suggesting that the depolarization segment abnormality might be dynamic and corresponding to episodes of acute right heart strain and pulmonary arterial hypertension. This implies that fQRS could be acute and reversible, related to myocardial stretch and stunning, however further study is required to assess this phenomenon.

Other relevant findings from this study included older age and increased comorbidity burden (specifically ischemic heart disease and diabetes mellitus) to be significantly associated with mortality, in agreement with previous studies.^24,25 Patients who underwent TTE were characterized by worsened blood markers of cytokine storm and cardiac injury (CRP, troponin T, and lymphocyte counts). This appropriately reflects how this group of patients would have fulfilled the clinical criteria to warrant TTE.

Our hypotheses and findings require validation in further, larger studies. Longer-term data, utilizing more sophisticated functional imaging modalities such as perfusion magnetic resonance imaging, might elucidate the natural history and clinical significance of electrocardiographic and TTE abnormalities following hospitalization. The other question remains whether myocardial scarring and fibrosis are long-term sequelae from protracted high-pressure ventilation and a high burden of parenchymal +/- vaso-occlusive disease. Nevertheless, emphasis is laid upon recognition of fQRS by clinicians, which should prompt TTE assessment and consideration of targeted lung-protective ventilation and pharmacotherapy to best offload the RV.

Study Limitations

Despite achieving significant results in a number of outcomes, we acknowledge the risk of a type 2 error occurring with our experimental sample size. Therefore, we were unable to discount an association between fQRS, RV dilatation, and pulmonary hypertension and those variables that did not achieve statistical significance with observed effect sizes. The high mortality rate observed may have also generated a bias.

During the 15-month period of our study, treatment protocols for COVID-19 were rapidly changing, as were the variants of SARS-COV-2 identified within the population. Changes in protocol overlapping with our study, namely, dexamethasone (June 2020) and IL-6 inhibitors tocilizumab and sarilumab (February 2021), may have contributed to cytokine storm being less prominent in our study group. These factors make it difficult to generalize to populations where acuity and standard treatment were different.

We also acknowledge that extracting information from medical notes requires second-hand interpretation and may not be representative of the full clinical picture. There was also clear limitations given we relied on observational, retrospective data with no clear protocol in terms of clinician decisions to monitor ECG and TTE parameters. Electrocardiograms were performed on admission to hospital, critical care, and then at clinician discretion rather than at defined time points. TTE was also performed based on clinical discretion, meaning the time interval between ECG and TTE varied between each patient. Therefore, a definitive relationship between fQRS and RV volume and pressure overload cannot be confirmed for certain. Finally, we have no data to ascertain abnormalities in patient pre-morbid ECGs, TTE, or ongoing data post-critical care admission. This remains an area for future study.

Conclusion

Our study is the first to demonstrate a possible relationship between fQRS, RV dilatation, pulmonary hypertension, and mortality in critically ill COVID-19 patients. The mechanism surrounding fQRS in these patients may reflect anomalous myocyte activation arising from mechanical and electrical delay within RV myocardial tissue in response to pressure and volume overload. We suggest RV dilatation and pulmonary hypertension are important prognosticators within the context of current treatment protocols for severe COVID-19. Furthermore, our findings may be of particular interest when considering therapeutic options targeted at lung-protective ventilation and pulmonary arterial vasodilation to improve outcomes. Acknowledgment of the importance of fQRS could also lead to improvement in ECG analysis algorithms to facilitate broader clinical application for diagnosis and prognostication.

Footnotes

Ethics Committee Approval: As an analysis using clinically collected, nonidentifiable data, this work does not fall under the remit of National Health Service Research Ethics Committees. This study was approved by our institution’s Research, Quality Improvement, and Audit Department with the reference number FH205. It was not appropriate or possible to involve patients or the public in the design, conduct, reporting, or dissemination plans of our study.

Informed Consent: As a non-interventional, observational study, specific informed consent was not required by the Research, Quality Improvement and Audit department.

Peer-review: Externally peer-reviewed.

Author Contributions: Concept – R.S.H., A.K.J.M., P.S., C.G.M.; Design – R.S.H., A.K.J.M., P.S., C.G.M.; Supervision – A.K.J.M., P.S., C.G.M.; Data Collection and/or Processing – V.B., S.M., H.M., K.K.M.; Analysis and/or Interpretation – R.S.H., A.K.J.M., V.B.; Literature Search – R.S.H., F.L.S.; Writing – R.S.H., F.L.S., A.K.J.M.; Critical Review – All authors.

Declaration of Interests: The authors have no conflicts of interest to declare.

References

1. Aziz S, Arabi YM, Alhazzani W. Managing ICU surge during the COVID-19 crisis: rapid guidelines. Intensive Care Med. 2020;46(7):1303-1325. https://doi.org/10.1007/s00134-020-06092-5
2. Haukilahti MAE, Holmström L, Vähätalo J. Gender differences in prevalence and prognostic value of fragmented QRS complex. J Electrocardiol. 2020;61():1-9. https://doi.org/10.1016/j.jelectrocard.2020.05.010
3. Pietrasik G, Zaręba W. QRS fragmentation: diagnostic and prognostic significance. Cardiol J. 2012;19(2):114-121. https://doi.org/10.5603/cj.2012.0022
4. Sadeghi R, Dabbagh VR, Tayyebi M, Zakavi SR, Ayati N. Diagnostic value of fragmented QRS complex in myocardial scar detection: systematic review and meta-analysis of the literature. Kardiol Pol. 2016;74(4):331-337. https://doi.org/10.5603/KP.a2015.0193
5. Supreeth RN, Francis J. Fragmented QRS – its significance. Indian Pacing Electrophysiol J. 2020;20(1):27-32. https://doi.org/10.1016/j.ipej.2019.12.005
6. Torales JM, Centurión OA, Aquino NJ. The relation between QRS complex fragmentation and segmental abnormalities of the myocardial contractility in patients with coronary artery disease. Indian Heart J. 2021;73(3):325-330. https://doi.org/10.1016/j.ihj.2021.03.010
7. Özdemir İH, Özlek B, Özen MB. Fragmented QRS is a marker of mortality in patients with severe COVID-19: A retrospective observational study. Anatol J Cardiol. 2021;25(11):811-820. https://doi.org/10.5152/AnatolJCardiol.2021.62
8. Yildirim A, Karaca IO, Yilmaz FK, Gunes HM, Cakal B. Fragmented QRS on surface electrocardiography as a predictor of cardiac mortality in patients with SARS-CoV-2 infection. J Electrocardiol. 2021;66():108-112. https://doi.org/10.1016/j.jelectrocard.2021.03.001
9. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-383. https://doi.org/10.1016/0021-9681(87)90171-8
10. Lippi G, Plebani M. Laboratory abnormalities in patients with COVID-2019 infection. Clin Chem Lab Med. 2020;58(7):1131-1134. https://doi.org/10.1515/cclm-2020-0198
11. Das MK, Saha C, El Masry H. Fragmented QRS on a 12-lead ECG: A predictor of mortality and cardiac events in patients with coronary artery disease. Heart Rhythm. 2007;4(11):1385-1392. https://doi.org/10.1016/j.hrthm.2007.06.024
12. Das MK, Suradi H, Maskoun W. Fragmented wide QRS on a 12-lead ECG: A sign of myocardial scar and poor prognosis. Circ Arrhythm Electrophysiol. 2008;1(4):258-268. https://doi.org/10.1161/CIRCEP.107.763284
13. Augustine D, Willis J, Robinson S. Covid-19 BSE TTE Consensus pathway. British Society of Echocardiography. 2020;():-. https://bsecho.org/Common/Uploaded%20files/Education/COVID-19/COVID-19%20BSE%20TTE%20consensus%20pathway.pdf
14. Nagueh SF, Smiseth OA, Appleton CP. Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2016;29(4):277-314. https://doi.org/10.1016/j.echo.2016.01.011
15. Harkness A, Ring L, Augustine DX, Oxborough D, Robinson S, Sharma V. Normal reference intervals for cardiac dimensions and function for use in echocardiographic practice: A guideline from the British Society of Echocardiography. Echo Res Pract. 2020;7(1):G1-G18. https://doi.org/10.1530/ERP-19-0050
16. Augustine DX, Coates-Bradshaw LD, Willis J. Echocardiographic assessment of pulmonary hypertension: a guideline protocol from the British Society of Echocardiography. Echo Res Pract. 2018;5(3):G11-G24. https://doi.org/10.1530/ERP-17-0071
17. Bazoukis G, Garcia-Zamora S, Çinier G. Association of electrocardiographic markers with myocardial fibrosis as assessed by cardiac magnetic resonance in different clinical settings. World J Cardiol. 2022;14(9):483-495. https://doi.org/10.4330/wjc.v14.i9.483
18. de Bakker JM, van Capelle FJ, Janse MJ. Slow conduction in the infarcted human heart. ‘Zigzag’ course of activation. Circulation. 1993;88(3):915-926. https://doi.org/10.1161/01.cir.88.3.915
19. Das MK, Maskoun W, Shen C. Fragmented QRS on twelve-lead electrocardiogram predicts arrhythmic events in patients with ischemic and nonischemic cardiomyopathy. Heart Rhythm. 2010;7(1):74-80. https://doi.org/10.1016/j.hrthm.2009.09.065
20. Bektaş O, Çerik İB, Çerik HÖ. The relationship between severe acute respiratory syndrome coronavirus 2 (SARS - COV - 2) pandemic and fragmented QRS. J Electrocardiol. 2020;62():10-13. https://doi.org/10.1016/j.jelectrocard.2020.07.009
21. Horby P, Lim WS. Dexamethasone in hospitalized patients with Covid-19. N Engl J Med. 2021;384(8):693-704. https://doi.org/10.1056/NEJMoa2021436
22. Abani O, Abbas A, Abbas F. Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial. Lancet. 2021;397(10285):1637-1645. https://doi.org/10.1016/S0140-6736(21)00676-0
23. Mandal AKJ, Kho J, Ioannou A, Van den Abbeele K, Missouris CG. Covid-19 and in situ pulmonary artery thrombosis. Respir Med. 2021;176():106176-. https://doi.org/10.1016/j.rmed.2020.106176
24. Sahni S, Gupta G, Sarda R, Pandey S, Pandey RM, Sinha S. Impact of metabolic and cardiovascular disease on Covid-19 mortality: A systematic review and meta-analysis. Diabetes Metab Syndr. 2021;15(6):102308-. https://doi.org/10.1016/j.dsx.2021.102308
25. Güney BÇ, Taştan YÖ, Doğantekin B. Predictive value of car for in-hospital mortality in patients with COVID-19 pneumonia: A retrospective cohort study. Arch Med Res. 2021;52(5):554-560. https://doi.org/10.1016/j.arcmed.2021.02.006