key: cord-0942565-w86ll8xv
authors: Mizera, Lars; Rath, Dominik; Schoellmann, Anna; Petersen-Uribe, Alvaro; Avdiu, Alban; Zdanyte, Monika; Jaeger, Philippa; Heinzmann, David; Müller, Karin; Gawaz, Meinrad; Eick, Christian; Duckheim, Martin
title: Deceleration capacity is associated with acute respiratory distress syndrome in COVID-19
date: 2021-08-02
journal: Heart Lung
DOI: 10.1016/j.hrtlng.2021.07.016
sha: 3a1cc02cc9020ff80dc50576cbdde35fc8492a21
doc_id: 942565
cord_uid: w86ll8xv

BACKGROUND: : Acute respiratory distress syndrome (ARDS) is considered the main cause of COVID-19 associated morbidity and mortality. Early and reliable risk stratification is of crucial clinical importance in order to identify persons at risk for developing a severe course of disease. Deceleration capacity (DC) of heart rate as a marker of cardiac autonomic function predicts outcome in persons with myocardial infarction and heart failure. We hypothesized that reduced modulation of heart rate may be helpful in identifying persons with COVID-19 at risk for developing ARDS. METHODS: : We prospectively enrolled 60 consecutive COVID-19 positive persons presenting at the University Hospital of Tuebingen. Arterial blood gas analysis and 24h-Holter ECG recordings were performed and analyzed at admission. The primary end point was defined as development of ARDS with regards to the Berlin classification. RESULTS: : 61.7% (37 of 60 persons) developed an ARDS. In persons with ARDS DC was significantly reduced when compared to persons with milder course of infection (3.2 ms vs. 6.6 ms, p < 0.001). DC achieved a good discrimination performance (AUC = 0.76) for ARDS in COVID-19 persons. In a multivariate analysis, decreased DC was associated with the development of ARDS. CONCLUSION: : Our data suggest a promising role of DC to risk stratification in COVID-19.

The current worldwide pandemic of the 2019 coronavirus disease caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) poses a significant threat to global health and economy. The clinical presentation varies widely among individuals from mild afebrile respiratory symptoms to severe illness causing acute respiratory distress syndrome (ARDS) (1) . Approximately 5% of infected persons require admission to an intensive care unit and approximately 12% of hospitalized COVID-19 persons receive mechanical ventilation (2, 3) . Prediction models for both prognosis and clinical course of COVID-19 are urgently required to identify persons at high risk of developing severe ARDS requiring mechanical ventilation. The quick COVID-19 severity index has been recently postulated as a risk score for short-term respiratory deterioration by assessing respiratory rate, oxygen saturation and oxygen flow rate (4) . Current data suggest, however, that reliable prediction models are missing (5) . Therefore, further tools to support medical decision making are of great interest, especially in the setting of an emergency department or in an outpatient area. Heart rate variability analysis is a non-invasive, objective method assessing the sympathetic and vagal balance of the autonomic nervous system. Reduced heart rate deceleration capacity (DC) indicates impaired cardiac vagal modulations (6) . DC has already been proved to be a powerful predictor of mortality in persons with myocardial infarction, heart failure, cancer, aortic stenosis, syncope, stroke and pneumonia. (6) (7) (8) (9) (10) (11) (12) (13) . Recently, a retrospective, observational case series emphasized direct and indirect affection of the central nervous system in COVID-19 persons with respiratory insufficiency (14) . In the current study, we postulate that impaired DC provides prognostic information in persons with COVID-19.

This prospective study was conducted in the Department of Cardiology, Angiology 

Oxygen saturation and routine blood samples were obtained at hospital admission. If progressive dyspnea, oxygen requirement or reduced oxygen saturation <97% occurred, arterial blood gas analysis was performed. Additional arterial blood gas analysis was completed when persons showed relevant clinical worsening of their symptoms or oxygen insufflation was required. We determined ARDS by the ratio of partial pressure of oxygen in blood to fraction of inhaled oxygen (PaO 2 /FiO 2 ). In this sense, ARDS was classified as mild (PaO 2 /FiO 2 >200 mmHg, but ≤300 mmHg), moderate (PaO 2 /FiO 2 >100 mmHg, but ≤200 mmHg) and severe (PaO2/FiO2 is ≤100 mmHg) according to the Berlin Definition of ARDS (18). Oxygen delivery devices such as nasal cannulas or venturi masks were used in non-ventilated persons to increase the FiO2 up to 60 % based on known equipment standards. With a nasal cannula, the FiO2 increases by approximately 4% for every additional liter of oxygen flow provided. For example, a flow rate at 1L/min is able to increase the FiO2 to 24%, 3L/min to 32% and 6L/min up to 44% (19) .

Persons received a 24h-ECG Holter recording within 24 hours after admission to the isolation ward. DC was automatically assessed from a 24h-ECG Holter (Getemed CardioMem CM 3000SM 24h Holter ECG Recorder) according to validated technologies (20). Briefly, sequences of RR intervals were processed by a signal processing algorithm called phase-rectified signal averaging (PRSA). RR intervals longer than their precursors are identified as so-called anchors. Segments with certain length surrounding the anchors are aligned and averaged to obtain the PRSA signal. The DC is acquired by quantifying the central amplitude of the PRSA signal by

Haar-wavelet analysis. The DC can be considered as integral of all decelerationrelated fluctuations during the selected period of observation (6, 21) . We only used the first 10 minutes of the ECG recordings for analysis of at least 200 anchors. In case of low signal quality, the observation period was extended to a maximum of 30 minutes until 200 anchors were identified. ECG recordings with noisy, low-quality signals were excluded using a validated algorithm (22) . According to Bauer et al., a DC ≤ 4.5 ms might be associated with high risk of mortality (6) .

We defined a composite of mild, moderate and/or severe ARDS during hospitalization as the primary endpoint.

All statistical analyses were performed using SPSS version 26.0 2019 (SPSS Inc., Chicago IL). Non-normally distributed data was compared using Mann-Whitney U-Test. Chi-squared cross-tabulations were used to analyze qualitative data. Data was presented as mean ± standard deviation. Univariate and multivariate binary logistic regression analyses were used to evaluate associations between patient characteristics, laboratory results at admission and DC in terms of development of ARDS. The confidence interval was set to 95% and a p-value of ≤ 0.05 was considered significant. Receiver operating characteristic (ROC) curves were used to assess the sensitivity and specificity of DC for prediction of ARDS after admission to ED due to COVID-19. The overall discriminatory capacity of the test is estimated by 

60 persons with COVID-19 infection requiring hospitalization underwent 24h-ECG Holter recording. Baseline characteristics, medical history and outcome of the overall cohort are presented in Table 1 .

The incidence of ARDS was 61.7% (37 of 60 persons), with an intrahospital mortality Hypertension as a comorbidity was more frequent in the ARDS group (p=0.007).

ARDS persons had a mean (± SD) PaO 2 /FiO 2 of 178 ± 58 mmHg. 24 persons (64.9%) developed moderate to severe ARDS and 17 out of 37 ARDS persons (45.9%) required intubation. The mean timespan between DC measurement at isolation ward and admission to the ICU was 3.6 ± 4.6 days. Table 2 compares ARDS and non-ARDS persons in detail. There were no differences with regard to medication with possible impact on DC, such as beta-blockers or prevalence of diabetes between the two groups. Further characteristics comparing ARDS and non-ARDS persons are presented in table 2. One person suffered from congestive heart failure (CHF).

We found DC to be significantly lower in persons with ARDS compared to non-ARDS persons (3.2 ± 2.2 ms vs. 6.6 ± 6.1 ms; p<0.001, OR=0.7 ; 95% CI, 0.5-0.9; p=0.005 (Figure 1 ).

The major findings of the current study are: (1) Compromised DC was associated with an increased risk of developing ARDS in persons with COVID-19 infection. (2) DC may be helpful to characterize persons in a more advanced disease stage of COVID-19 and therefore may be used as a marker of advanced disease.

Severe respiratory failure with development of ARDS is a potential complication of COVID-19 infection. Previous reports also consider severe ARDS to be responsible for the majority of COVID-19 associated morbidity and mortality (1, 23) . Prognostic markers that help to distinguish persons at risk for severe ARDS are urgently needed to establish proper treatment strategies. Coagulation abnormalities and myocardial injury were found to be prevalent in COVID-19 positive persons and to be predictive of adverse outcomes (24) (25) (26) (27) (28) . These studies identified higher levels of the myocardial distress marker troponin I, as well as higher levels of D-dimers, C-reactive protein and LDH on admission being highly predictive for respiratory insufficiency during the course of the COVID-19 disease. In line with these papers, we also found higher levels of troponin I and LDH in the ARDS group. Interestingly, persons in the ARDS group were more likely to suffer from hypertension.

As an early marker for progressive disease, we propose calculation of the DC. DC provides information regarding the cardiac autonomic function. The mechanisms of compromised DC are various and related to altered sympathetic vascular tone, reduced neuro-humoral organ interaction and cardiac injury due to myocardial inflammation (29) . Since DC reflects cardiac vagal modulation, decreased DC may be used to estimate disease severity, independent of the underlying cause. It is certainly not specific to COVID-19. However, as risk predictor DC may aid to identify persons at risk of a severe course of disease including COVID-19. In an emergency setting with patients presenting at different stages of disease, risk stratification can be challenging. Besides respiratory rate, heart rate or blood pressure, DC may allow a quick assessment by evaluation of ECG signals. In a previous study, the prognostic value of DC was demonstrated in emergency patients with pneumonia (11).

The calculation of DC is efficient, user independent and can be performed fullyautomated either from of heart rhythm monitoring or, as carried out in our study, from of 24h-Holter ECGs. According to our findings, DC as well as previously postulated risk factors (troponin I, D-dimers, LDH, hypertension) may be associated with ARDS (25, 30, 31) . Furthermore, DC may be considered a suitable screening method to identify persons at low risk for developing severe ARDS. In times of the COVID-19 pandemic with scenarios of overcrowded emergency departments, this might be of crucial importance.

Since the DC can be determined both easily and non-invasively with the help of software algorithms, its use is also conceivable in an outpatient setting. Due to advancement of technologies, the requirements of a proper ECG recording with wristworn heart rate monitors, such as the Apple Watch (Apple, Cupertino, California) are already in place. The accuracy of variables was proven in various studies (32, 33) .

This novel technology provides adequate signal quality of ECG recordings and could be used for automated assessment of DC within 10 minutes (34) . Thus, the DC might be a useful tool for risk prediction in COVID-19 persons and therefore could support decision-making in the outpatient area, help to establish treatment strategies and avoid unnecessary hospitalization for low-risk persons. In this sense, both material and personal resources might be preserved and exposition of health care personal can be reduced.

Various limitations of our study need to be mentioned. First, due to the small cohort of COVID-19 infected persons in this study, the present data should be considered hypothesis generating and multivariate analysis needs cautious interpretation.

Further verification in prospective control studies is required. Second, the proportion of persons who developed ARDS in this study population is high compared to previous investigations. This might be due to the fact, that only hospitalized persons were included. A continuous rhythm monitoring providing data for real-time analysis of DC would be desirable. This could improve the ability for risk prediction of this patient collective. Third, calculation of DC can only be performed in the presence of sinus rhythm. Persons with either atrial fibrillation or permanent stimulation by a pacemaker need to be excluded. It should further be noted that ARDS has cofounders such as CHF which is a risk factor for a severe course of COVID-19 itself (35) . DC may serve as an indicator of poor prognosis in persons with both COVID-19 as well as CHF.

To conclude, this is the first study of biosignal analysis in hospitalized persons with ARDS secondary to COVID-19. DC seems to be a convenient and promising prognostic tool that may identify persons with advanced COVID-19 that could develop ARDS. Tables: Table 1 : Baseline characteristics and outcomes of the study population (n=60). 

COVID-19 pathophysiology: A review

COVID-19 pandemic -A focused review for clinicians

Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the

Development and validation of the quick COVID-19 severity index (qCSI): a prognostic tool for early clinical decompensation

Prediction models for diagnosis and prognosis of covid-19 infection: systematic review and critical appraisal

Deceleration capacity of heart rate as a predictor of mortality after myocardial infarction: cohort study

Heart rate variability for rapid risk stratification of emergency patients with malignant disease

Heart rate turbulence predicts all-cause mortality and sudden death in congestive heart failure patients

Deceleration capacity of heart rate predicts 1-year mortality of patients undergoing transcatheter aortic valve implantation

Deceleration capacity as a risk predictor in patients presenting to the emergency department with syncope: A prospective exploratory pilot study. Medicine (Baltimore)

Autonomic Nervous System Activity for Risk Stratification of Emergency Patients With Pneumonia

Autonomic nervous system activity as risk predictor in the medical emergency department: a prospective cohort study

Deceleration capacity for rapid risk stratification in patients suffering from acute ischemic stroke: A prospective exploratory pilot study

Neurologic Manifestations of Hospitalized Patients With Coronavirus Disease 2019 in Wuhan, China. JAMA Neurol. 2020. 15. World Medical Association Declaration of Helsinki. Recommendations guiding physicians in biomedical research involving human subjects

Guideline: Guideline for Good Clinical Practice

on the approximation of the laws, regulations and administrative provisions of the member states relating to the implementation of good clinical practice in the conduct of clinical trials on medicinal products for human use

Fraction of Inspired Oxygen (FiO2). StatPearls

Automated assessment of cardiac autonomic function by means of deceleration capacity from noisy, nonstationary ECG signals: validation study

Risk prediction in post-infarction patients with moderately reduced left ventricular ejection fraction by combined assessment of the sympathetic and vagal cardiac autonomic nervous system

A simple method to detect atrial fibrillation using RR intervals

Clinical Features of 85 Fatal Cases of COVID-19 from Wuhan. A Retrospective Observational Study

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

COVID-19 and the cardiovascular system: implications for risk assessment, diagnosis, and treatment options

The role of biomarkers in diagnosis of COVID-19 -A systematic review

Impaired Cardiac Function is Associated with Mortality in Patients with Acute COVID-19 Infection

Coronavirus-induced myocarditis: A meta-summary of cases

Autonomic dysfunction in the ICU patient

Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease

The association of D-dimers with mortality, intensive care unit admission or acute respiratory distress syndrome in patients hospitalized with coronavirus disease 2019 (COVID-19): A systematic review and meta-analysis

Smartwatch Algorithm for Automated Detection of Atrial Fibrillation

Variable Accuracy of Wearable Heart Rate Monitors during Aerobic Exercise

Recording of Bipolar Multichannel ECGs by a Smartwatch: Modern ECG Diagnostic 100 Years after Einthoven

Impaired cardiac function is associated with mortality in patients with acute COVID-19 infection