key: cord-0979481-7fa2kivt
authors: Ippolito, Mariachiara; Catalisano, Giulia; Marino, Claudia; Fucà, Rosa; Giarratano, Antonino; Baldi, Enrico; Einav, Sharon; Cortegiani, Andrea
title: Mortality after In-Hospital Cardiac Arrest in Patients with COVID-19: A Systematic Review and Meta-Analysis
date: 2021-05-05
journal: Resuscitation
DOI: 10.1016/j.resuscitation.2021.04.025
sha: a64e8b8e8b886ac2a908152b61de5b1a55ce79b6
doc_id: 979481
cord_uid: 7fa2kivt

AIM: To estimate the mortality rate, the rate of return of spontaneous circulation (ROSC) and survival with favorable neurological outcome in patients with COVID-19 after in-hospital cardiac arrest (IHCA) and attempted cardiopulmonary resuscitation (CPR). METHODS: PubMed, EMBASE, Web of Science, bioRxiv and medRxiv were surveyed up to 8th February 2021 for studies reporting data on mortality of patients with COVID-19 after IHCA. The primary outcome sought was mortality (in-hospital or at 30 days) after IHCA with attempted CPR. Additional outcomes were the overall rate of IHCA, the rate of non-shockable presenting rhythms, the rate of ROSC and the rate of survival with favorable neurological status. RESULTS: Ten articles were included in the systematic review and meta-analysis, for a total of 1179 COVID-19 patients after IHCA with attempted CPR. The estimated overall mortality rate (in-hospital or at 30 days) was 89.9% (95% Predicted Interval [P.I.] 83.1%–94.2%; 1060/1179 patients; I(2) = 82%). The estimated rate of non-shockable presenting rhythms was 89% (95% P.I. 82.8%-93.1%; 1022/1205 patients; I(2) = 85%), and the estimated rate of ROSC was 32.9% (95% P.I. 26%-40.6%; 365/1205 patients; I(2) = 82%). The estimated overall rate of survival with favorable neurological status at 30 days was 6.3% (95% P.I. 4%-9.7%; 50/851 patients; I(2) = 48%). Sensitivity analysis showed that COVID-19 patients had higher risk of death after IHCA than non COVID-19 patients (OR 2.34; 95% C.I. 1.37-3.99; number of studies = 3; 1215 patients). CONCLUSIONS: Although one of three COVID-19 patients undergoing IHCA may achieve ROSC, almost 90% may not survive at 30 days or to hospital discharge.

Patients with coronavirus disease 2019 (COVID- 19) can develop severe acute respiratory failure and hypoxemia 1 . Indeed, cardiorespiratory comorbidities have been identified as risk factors for the development of the most severe forms of the disease 1 . Cardiac arrest may thus complicate the in-hospital clinical course of these patients, as a consequence of various triggers, both pneumological (e.g. severe hypoxemia, pneumothorax) and cardiological ones (e.g pulmonary embolism, QT interval prolongation) 2 3 . Moreover, COVID-19 pandemic has determined the hospitalization of an unprecedented number of patients which has been associated with a high number of in-hospital cardiac arrests (IHCA) 4 .

Poor outcomes have been described in patients with COVID-19 undergoing IHCA, despite cardiopulmonary resuscitation (CPR) attempts 5 . At the same time, cardiopulmonary resuscitation of a patient with COVID-19 carries inherent risks for healthcare workers as resuscitation procedures may disperse aerosol 6 7 . Much public and scientific debate has therefore surrounded the ethics of CPR and do-not-attempt resuscitation (DNAR) orders in patients with COVID-19 8 . At this time, the rates of mortality and other relevant clinical outcomes in patients with COVID-19 after IHCA remain unclear and the data surrounding this topic is fragmented 9 .

The aim of this systematic review and meta-analysis was to estimate the mortality rate in patients with COVID-19 after IHCA and attempted CPR from the available literature. We also sought to estimate the rate of return of spontaneous circulation (ROSC) and survival with favorable neurological outcome in the same population of patients. February 2021 without limiting the results by date filters. BioRxiv and medRxiv preprint servers were also searched using 'cardiac arrest', 'cpr' and 'COVID-19' as search terms and keywords. No language restriction was applied. The retrieved records were considered relevant if they provided data on mortality after in-hospital cardiac arrest occurring in adult patients with COVID-19. Nonrandomized studies, both prospective and retrospective, and case series were included. Abstracts and case reports were excluded. The full search strategy is provided as Supplementary Material 1. The retrieved records were independently screened using the titles and abstracts by two authors (MI, CM). The articles selected during this screening were then downloaded in full-text and independently reviewed by the same two authors.

Studies were included if the two screening authors agreed regarding eligibility and, in case of disagreement, a third author adjudicated (AC). The reference lists of relevant articles were also searched for additional potentially relevant papers (i.e. snowballing method). The data were extracted in duplicate by two authors (MI, GC) using a standard data extraction form. Discrepancies in the extracted data were also adjudicated by a third author (AC). To reduce heterogeneity from studies that also provided data on patients developing IHCA who had not received CPR, we extracted only the data on those patients who received CPR. The final version of the database was validated by all the authors involved in data collection (MI, GC, CM, AC). The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) J o u r n a l P r e -p r o o f checklist and the checklist for meta-analysis of observational studies (MOOSE) are provided in Table S1 and Table S2 , Supplementary Material 2.

The primary outcome sought was mortality after IHCA with attempted CPR (inhospital or at 30 days if the former was not reported). Additional outcomes were the overall rate of IHCA, the rate of non-shockable presenting rhythms, the rate of return of spontaneous circulation (ROSC) and the rate of survival with favorable neurological functional status, defined as a Cerebral Performance Category (CPC) score of 1 or 2 (CPC 1 -Good cerebral performance: conscious, alert, able to work, might have mild neurologic or psychologic deficit; CPC 2 -Moderate cerebral disability: conscious, sufficient cerebral function for independent activities of daily life.

Able to work in sheltered environment; CPC 3 -Severe cerebral disability: conscious, dependent on others for daily support because of impaired brain function. Ranges from ambulatory state to severe dementia or paralysis; CPC 4 -Coma or vegetative state: any degree of coma without the presence of all brain death criteria.

Unawareness, even if appears awake (vegetative state) without interaction with environment; may have spontaneous eye opening and sleep/awake cycles. Cerebral unresponsiveness; CPC 5 -Brain death: apnea, areflexia, EEG silence, etc.) 10 .

Two investigators (AC, MI) performed a qualitative analysis of the included studies independently and in duplicate. Disagreements over assessments were resolved by consensus or, if necessary, adjudicated by a third author (SE). As our primary outcome was an incidence, the Methodological Index for Non Randomized studies (MINORS) 11 was used for the qualitative assessment, due to its ability to evaluate the methodological quality of single arm studies. The eight items are scored J o u r n a l P r e -p r o o f 0 (not reported), 1 (reported but inadequate) or 2 (reported and adequate). The global ideal score is 16 for non-comparative studies 11 .

Meta-analysis was performed if two or more of the included studies reported data on the outcomes of interest. To maintain symmetry, the summary estimates were calculated using logit transformation of individual study proportions and presented along with the corresponding 95% prediction interval (P.I.). Random effect models (DerSimonian-Laird) were used for all the analyses. For the outcome of mortality, we performed sub-group analyses based on the level of care (i.e. ICU and non-ICU) and on the number of centres per study (i.e. multicentre, single centre). We also performed a sensitivity analysis on papers comparing the mortality rates of patients with COVID-19 to those of non COVID-19 after IHCA, providing odds ratio as a measure of risk.

The I-squared (I 2 ) statistical model was used to describe the percentage of variation across the included studies due to heterogeneity. All the analyses were performed by AC and MI, with input from SE, using OpenMeta-Analyst 12 .

The search identified a total of 2088 records. After the exclusion of duplicates and not relevant records, ten articles were included, evaluating a total of 1179 COVID-19 patients after IHCA and attempted CPR 5 13-21 . The process of inclusion and exclusion, detailed in a the PRISMA flow diagram, is presented as 

All the included articles reported retrospective studies. Three were multicentre and seven were single centre studies. Nine studies reported data on a mixed cohort of patients in both the intensive care unit (ICU) and other wards. One study included only ICU patients. Eight studies were conducted in the U.S., one was conducted in China and one in Sweden. All the included studies specifically included patients with IHCA and CPR, except one that also provided data on patients with IHCA that did not undergo CPR 13 . All the studies specifically included patients with COVID-19, but three studies also reported data on historical cohorts of non COVID-19 patients 15 19 21 .

The mean age of the patients ranged from 61 to 69 years. Overall, 64.9% Table 1 .

Six studies had 13/16 as overall qualitative assessment score 5 14-16 19 21 , three studies had 12/16 17 18 20 and one had 14/16 13 . The detailed qualitative assessment with individual domain score per study is provided as 

Seven studies provided data on the rate of IHCA with attempted CPR in cohorts of hospitalized patients with COVID-19 5 All the included studies provided information on the presenting rhythms. The estimated overall rate of non-shockable presenting rhythms was 89% (95% P.I. 82.8%-93.1%; 1022/1205 patients; I 2 =85% ; Fig. 2, panel a) .

All the included studies provided information on ROSC. The estimated overall ROSC rate was 32.9% (95% P.I. 26%-40.6%; 365/1205 patients; I 2 =82%; Fig. 2, 

Four studies provided data on the neurological status of survivors at 30 days [13] [14] [15] [16] . The estimated overall rate of a favorable neurological status (defined as a CPC score of 1 or 2) at 30 days was 6.3% (95% P.I. 4%-9.7%; 50/851 patients; I 2 =48% ;   Fig. 2, panel c) .

When the arrest occurred in an ICU, the estimated mortality rate was 85.8% Three studies reported the data on mortality of COVID-19 versus non-COVID-19 patients who developed IHCA 15 19 21 . Patients with COVID-19 had a significantly higher risk of death after IHCA with CPR than non-COVID-19 (OR 2.34; 95% C.I.

1.37-3.99; 1215 patients; Fig. S5, Supplementary Material 2) .

To the best of our knowledge, this is the most updated systematic review and meta-analysis on mortality rate after IHCA with attempted CPR in patients with COVID-19. The main finding of our study was that although one in three patients with COVID-19 may achieve ROSC after IHCA with attempted CPR, only one in ten may survive 30 days or to hospital discharge and 6% may survive with favorable neurological status (defined as CPC score of 1 or 2). The mortality rates seem to be similarly high, regardless of whether the arrest occurred in the ICU or in a ward, although the latter finding is highly heterogeneous.

Recently, an intense debate regarding the ethics of universal do-notresuscitate orders for patients with COVID-19 arose 22 . Pre pandemic data showed the rate of ROSC after IHCA in adult patients has increased from 50% to 60-70% dependent on age group, while the rate of survival to hospital discharge has increased from 15 to 25% over the last decade in all age groups 23 . Less than 30%

were discharged from hospital with more than moderate neurological dysfunction, only one in three of these was severely neurologically impaired 24 hypoxemia). This assumption was reflected by our subgroup analysis on ICU and non-ICU settings, which showed nearly similar outcomes regardless of location. The fact that the overall rate of mortality in non-ICU settings was as high as 95.5% may reflect crisis allocation of resources (e.g. not all the most severe patients could be admitted to an ICU).

Our analysis has several limitations. Any analysis is only as good as the studies included in it. We chose to study incidence without a comparator because early review of the literature suggested that most of the studies we would identify would have no comparator. We also identified only ten studies (1179 patients). These numbers are probably too low to provide a definitive estimation of the rate of mortality and other outcomes. Most of the data came from the U.S. We found no data from European countries other than Sweden (one study). Only one study provided data from China during the first COVID-19 outbreak in Wuhan, including a relatively small sample size (n=136 patients). Our data has large statistical heterogeneity which probably stems from differences in the characteristics of the study cohorts (e.g. severity or setting) and the timing of the studies during the course of the pandemic (e.g. early pandemic, late pandemic). Lastly, all the included studies refer to the first pandemic peak. It cannot be excluded that the outcome of IHCA in COVID-19 patients could be different during the second pandemic peak, considering a different organization of the hospitals, increased availability of ICU beds and increased preparedness of physicians to treat COVID-related complications that can lead to cardiac arrest.

Data from the available evidence suggests an estimated mortality rate of nearly 90% among patients with COVID-19 after IHCA with CPR, regardless of arrest location, even though one patient out of three may achieve ROSC. COVID-19 patients seem to have a significantly higher risk of death after IHCA with CPR than non-COVID-19 patients. The decision to provide CPR, even in critically ill patients, requires a patient centered approach that acknowledges advance directives along J o u r n a l P r e -p r o o f with input from healthcare proxies. Certainly, when CPR is deemed futile or when the risk to those engaged in resuscitation exceed the potential benefit, the decision to proceed with CPR becomes more complicated. More data are needed to clarify for which COVID-19 patients, the benefit derived from CPR may not justify the risk incurred by this procedure to the treating healthcare staff. There is an urgent need for 

Baseline Characteristics and Outcomes of 1591 Patients Infected with SARS-CoV-2 Admitted to ICUs of the Lombardy Region

Pulmonary Embolism in Patients with COVID-19: Awareness of an Increased Prevalence

QTc Interval and Mortality in a Population of SARS-2-CoV Infected Patients

Coronavirus (COVID-19) Hospitalizations -Statistics and Research -Our World in Data

Is Cardiopulmonary Resuscitation Futile in Coronavirus Disease 2019 Patients Experiencing In-Hospital Cardiac Arrest?

Epidemic-prone and pandemic-prone acute respiratory diseases Infection Control Strategies for Specifi c Procedures in Health-Care Facilities A Quick Reference Guide

Cardiopulmonary resuscitation in prone position: A scoping review

Hospitals consider universal do-not-resuscitate orders for coronavirus patients

F2020%2F03%2F25%2Fcoronavirus-patients-do-not-resucitate%2F)

Outcomes of in-hospital cardiac arrest in COVID-19 patients: A proportional prevalence meta-analysis

Open Meta Analyst; Closing the gap between methodologists and end-users: R as a computational back-end

In-hospital cardiac arrest in critically ill patients with covid-19: multicenter cohort study

In-hospital cardiac arrest outcomes among patients with COVID-19 pneumonia in Wuhan

Clinical characteristics and outcomes of in-hospital cardiac arrest among patients with and without COVID-19

In-Hospital Cardiac Arrest in Patients with Coronavirus

Outcomes of in-hospital cardiac arrest in patients with COVID-19 in New York City

COVID-19 and cardiac arrhythmias

Cardiac arrest in COVID-19: characteristics and outcomes of in-and out-of-hospital cardiac arrest. A report from the Swedish Registry for Cardiopulmonary Resuscitation

Clinical outcomes of in-hospital cardiac arrest in COVID-19

Characteristics and Outcomes of In-Hospital Cardiac Arrest Events during the COVID-19 Pandemic: A Single-Center Experience from a New York City Public Hospital

Cardiopulmonary resuscitation after hospital admission with covid-19

Age-dependent trends in survival after adult in-hospital cardiac arrest

Trends in Survival after In-Hospital Cardiac Arrest

Survival after In-Hospital Cardiac