Sage Journals: Discover world-class research

Abstract

Critical illness-related cardiac arrest (CIRCA) as a distinct entity is not well described epidemiologically. There is currently a knowledge gap regarding how many occur in the UK or the impact on patient outcome. The CIRCA study is a prospective multi-centre observational cohort study of patients in the United Kingdom experiencing a cardiac arrest while in a Critical Care Unit embedded in the Case Mix Programme and National Cardiac Arrest Audit. The duration of data collection is 12 months, with surviving patients and family members receiving questionnaire follow-up at 90 days, 180 days and 12 months. This paper describes the protocol for the CIRCA study which received favourable ethical opinion from South Central – Berkshire Research Ethics Committee and approval from the Health Research Authority. Study registration is on clinicaltrials.gov (NCT04219384).

Keywords

Cardiac arrest critical illness intensive care observational study

Introduction

Cardiac arrest is often categorised by the location of the event – out-of-hospital (OHCA) or in-hospital (IHCA) – as there are important differences in population characteristics, aetiology and outcome.^1–3

Within a hospital, cardiac arrest is usually not further categorised, yet there may be clinically relevant differences between patients treated by different specialities or in locations with varying levels of patient monitoring available.³ One such location is the critical care unit (here meaning intensive care units providing level 3 care, high dependency units providing level 2 care and combined units that are hereafter collectively referred to as ‘critical care’). critical care units have common features that are known to be associated with improved patient outcomes, for example, electrocardiogram monitoring at the time of cardiac arrest is associated with a 38% decrease in risk of death after IHCA.⁴ Better patient to nurse ratios are also associated with higher probability of survival following IHCA.⁵ In addition, the skill mix of the critical care multidisciplinary team is suited to delivering advanced life support. It is likely therefore that delays in initiation of cardiopulmonary resuscitation (CPR), defibrillation and adrenaline administration, all of which are associated with decreased survival,⁶ will be minimised.

Conversely, patients being cared for in critical care are inherently likely to be more seriously ill, suffering from multiple organ failures. The probability of successful return of spontaneous circulation (ROSC) and favourable long-term outcomes could therefore be less likely compared to a general IHCA patient population. Thus, the risk of cardiac arrest occurring, the probability of ROSC and the ultimate outcome are all likely to be different to IHCAs occurring in other areas.⁷

Accurate epidemiological data on critical illness-related cardiac arrest is currently lacking. A recent systematic review and meta-analysis estimated the potential number of events at 22.7 per 1000 admissions.⁸ Limitations of this review include a high proportion of single centre and retrospective studies, many at risk of bias, and with wide variation in time, location and type of critical care unit. In the United Kingdom, the National Cardiac Arrest Audit (NCAA) collects data on IHCA attended by the resuscitation team following a 2222 call;⁹ however, for the reasons outlined above, a 2222 call is rarely placed when IHCAs occur within critical care and, therefore, a large proportion (estimated 80%) of these are not included.

A prospective study of 45 ICUs in France analysed 677 patients (of 31,399 admitted in 2017) who had at least one attempt at CPR in the ICU and concluded that only one in six was alive with good functional status (cerebral performance category 1 or 2) at 6 months.¹⁰ Although proximate factors considered contributory to cardiac arrest such as hypoxia were identified in a majority of cases, a number of pathophysiological drivers of cardiac arrest (colloquially the ‘4Hs and 4Ts’) are already well recognised.¹¹ Such derangements are to be expected in critical care. There is no information on which patient populations are most at risk beyond those who were ‘sicker’, that is, in receipt of more organ support – nor data on inter-hospital variation in the incidence of cardiac arrest that could be used to drive quality improvement. While the contemporary French healthcare system has similarities to that in the United Kingdom, there are also differences in critical care staffing and patients that could affect study outcomes.^12,13 Finally, follow-up was limited to 6 months using cerebral performance category rather than a global measure of health-related quality of life, such as the Euroqol EQ-5D-5L.^14,15

In this paper, we present the protocol for the CIRCA study, a prospective multi-centre, observational cohort study of patients in England and Wales experiencing a cardiac arrest while in the critical care unit.

Aim and objectives

The aim of the CIRCA study aim is to determine the incidence and outcomes of critical illness-related cardiac arrest (defined as cardiac arrest while in the critical care unit) in England and Wales and explore associated risk factors with critical care and hospital survival and quality of survival following hospital discharge.

The primary objective is to determine the incidence of critical illness-related cardiac arrest in adults in England and Wales.

Secondary objectives are:

a. What are the outcomes for critical illness related cardiac arrest?

b. What are the risk factors for critical illness related cardiac arrest?

c. What risk factors are associated with outcomes (critical care and hospital survival and quality of survival) in patients experiencing critical illness related cardiac arrest?

d. What is the quality of life for patients after critical illness related cardiac arrest?

e. What is the quality of life for family members of patients surviving critical illness related cardiac arrest?

Definition of outcomes

Primary outcome

The primary outcome is the incidence proportion of cardiac arrests, defined as the total number of patients who arrest divided by the total number of patients admitted to the critical care unit during an active recruitment period to a CIRCA participating site. The incidence rate will also be calculated, defined as the number of arrests during a critical care stay divided by the total patient days at risk (observed duration in critical care during an active recruitment period).

Secondary outcomes

Mortality in critical care is defined as death from any cause before discharge from critical care. Mortality in acute hospital stay is defined as death from any cause prior to discharge from acute hospital.

Duration of stay in critical care will be calculated in days between first cardiac arrest within critical care and date of discharge from critical care, or death in critical care, plus the duration of any subsequent admissions to critical care within the same acute hospital stay. Duration of stay in acute hospital will be calculated in days from first cardiac arrest within critical care to the date of acute hospital discharge or date of death within acute hospital.

Number of subsequent arrests in the critical care unit is calculated as the number of arrests excluding their first arrest within the critical care admission in which the patient was recruited to CIRCA. The number of subsequent arrests in acute hospital will be calculated as the number of cardiac arrests resulting in a 2222 call and a team visit recorded in NCAA between discharge from critical care and discharge from acute hospital.

Quality of life scores from EQ-5D-5L, Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE, short version) and Family Reported Outcome measures 16 (FROM-16) will be calculated at each follow-up time point.

Methods and analysis

Study design

The CIRCA study is a prospective observational cohort study in adult general critical care units (with variation by geographical area, university teaching status and size of unit) participating in the Case Mix Programme (CMP) and based on hospitals that participate in the National Cardiac Arrest Audit (NCAA).

Participants

The inclusion criteria are:

1. Age 18 years old or more; and either

2. Cardiac arrest (defined as receipt of chest compressions or defibrillation) occurring while in-hospital and within critical care (defined as either intensive care unit, high dependency unit or combined ICU/HDU); or

3. Family member of a patient surviving to discharge from critical care after a cardiac arrest within critical care. A family member is defined as a person with a close familial, social or emotional relationship to the patient and is not restricted solely to next-of-kin.

Critical care areas providing surge capacity were eligible to be included in CIRCA providing the site inclusion criteria were met (see Study Design above). Cardiac arrests occurring in critical care patients during transfer or invasive procedures were not eligible for inclusion.

Recruitment and consent

Enrolment

At the time of eligibility and enrolment onto the CIRCA study, participants will be unable to give informed consent and given the primary aim of this study is to describe accurately the incidence of critical illness related cardiac arrest it is not appropriate to target a smaller subset of patients for prospective consent. As such, the CIRCA study has been granted Section 251 approval (CAG reference: 19/CAG/0173) from the Confidentiality Advisory Group to access and collect minimal patient information without consent for the purposes of the study.

Patients surviving to critical care discharge will be provided information about the study and invited to participate in questionnaire follow-up. Patients and their family members are able to ‘opt out’ of the study and their data for the purposes of the study.

Data collection

Data collection for the CIRCA study will be embedded within the CMP and NCAA national clinical audits. Data collection comprises of cardiac arrest event and data linkage and questionnaire follow-up.

Cardiac arrest event data and additional treatment data will be collected on paper event and case report forms (CRFs), respectively, prior to entry onto a secure electronic data entry system. Minimal patient identifiable information (CMP and NHS number) will be collected on the CRF to allow for data linkage with routine sources (CMP, NCAA and NHS Digital).

Identification of in-hospital cardiac arrest event

Data is collected on all patients experiencing cardiac arrest while in a participating critical care unit. At the time of the event, the clinical team completes an event report form (paper) based on a modified NCAA dataset and captures basic event data, such as:

• CMP/NHS numbers;

• Critical care unit/Hospital;

• time and date of arrest;

• whether chest compressions/defibrillation were received;

• other treatments received (e.g. adrenaline and dose);

• presenting rhythm and

• outcome (including time to return of spontaneous circulation).

A small amount of additional treatment data surrounding the event is captured on a secure electronic CRF (eCRF) to determine risk factors associated with cardiac arrest in critical care. This includes:

• documented comorbidities;

• pre-event organ support (recorded in the 24 hours prior to first IHCA event in critical care) and

• data recorded in the 24 hours following the first IHCA event in critical care (physiological and neurological data, organ support).

Data linkage

Critical care and hospital outcomes

The patient CMP and NHS numbers (individual patient identifiers) are collected to allow for data linkage with the CMP and NCAA national clinical audits. Baseline demographic and clinical data on admission to critical care, and critical care resource usage and survival data will be obtained from the CMP. For patients surviving to discharge to the ward, subsequent IHCA events will be identified via data linkage with NCAA. The outcomes from critical illness-related cardiac arrest will be compared with those from IHCA reported in NCAA.

Questionnaire follow-up

At each time-point (90 days, 180 days and 12 months), survivors receive a post questionnaire containing the EQ-5D-5L and IQCODE questionnaires on Cognitive Decline in the Elderly (IQCODE, short version) questionnaires. At the same timepoints, an identified family member of a survivor receives a postal questionnaire containing the FROM-16 questionnaire and some additional demographic data (age, ethnicity and relationship to patient). The questionnaire packs include an information sheet and a refusal form to be completed and returned to the site if patients or family members do not wish to participate in follow-up.

In the event of no response to the patient questionnaire, sites will attempt to make telephone contact, and where appropriate a second questionnaire is sent. If no response is received for the 90-day follow-up, questionnaires will be sent again at all subsequent time points provided no refusal form has been received.

Withdrawal

If a patient or family member returns a refusal form or questionnaire to a participating site indicating that they do not wish to take part, no further contact or follow-up is made. Follow-up data collected up to that point will be retained unless the patient or family member requests otherwise. However, data collected without prior consent on the in-hospital cardiac arrest event will not be discarded, unless requested by the patient.

Long-term outcomes

Data on longer-term outcomes will be obtained from data linkage to routine sources (e.g. NHS Digital) and will include survival status at 12 months, and if applicable, date of death following discharge from acute hospital.

Sampling

The best current estimate (from the literature based on single critical care unit) of UK incidence is that between 5446 and 7329, cardiac arrests occur in critical care in 4066 to 5472 patients annually. Based on this estimate, we intend to recruit from 100 units for 12 months, giving us an anticipated number of events of 1320 to 2200 among 986 to 1643 patients from a total of around 75000 admissions. This sample size would enable us to estimate the rate of CIRCA with a standard error of approximately 0.5 to 0.6 per 1000 admissions. Survival to hospital discharge (pooled estimate from literature) is 17% and assuming 10% refusal of consent, which represents 151 to 251 patients and families for follow-up. Assuming loss to follow-up or death of a further 10%, the number followed up at 12 months is estimated to be between 136 and 226 patients and families.

Statistical analysis

The overall incidence proportion of cardiac arrests in critical care will be reported in terms of critical care admissions, and overall incidence rate in terms of observed days in critical care. Incidence proportions and rates for each site will be summarised by median and interquartile range. Factors associated with key secondary outcomes will be determined using multilevel logistic, linear, Poisson and Cox regression models as appropriate, using site as a random effect.

Risk factors prior to cardiac arrest will be obtained from the CMP and include patient characteristics on admission to critical care, and physiology during the first 24 hours in critical care.

Risk factors during arrest will be obtained from CIRCA data collection forms and include presenting/first documented rhythm, previous medical history and receipt of organ support in the 24 hours prior to cardiac arrest.

Risk factors post arrest will be obtained from CIRCA data collection forms and include duration of (first) arrest, level of sedation and/or paralysis, physiology 24 hours post-arrest and receipt of post-event organ support.

A complete list of potential risk factors for each outcome is defined in full in the statistical analysis plan which will be published on the study website (https://www.icnarc.org/Our-Research/Studies/Circa/About). Where appropriate, risk factors will be removed as necessary following checks on final sample size and assessing relevant model assumptions. Final models for each secondary outcome analysed will be reported as adjusted and unadjusted estimates (odds ratio, relative risk, mean difference and hazard ratio, as appropriate), 95% confidence intervals and p-values.

Handling of missing data

Missing data will be assessed for all data items collected and outcome measures. Multiple imputation methods will be considered for any high levels of missing data, assuming the data are missing at random (MAR).

Statistical software

All analyses will be conducted in Stata/MP version 16.1 (StataCorp, Texas).

Study oversight and ethics

The study has received a favourable opinion in ethical review (REC reference: 19/SC/0465) and Section 251 approval (CAG reference: 19/CAG/0173).

Discussion

The CIRCA study opened on 27/01/2020 and a total of 93 adult, general intensive care units from 89 hospital sites across England and Wales have participated overall. With COVID-19 related pauses taken into account, enrolment will close on 14^th June 2021 and the final follow-up completed 12 months later.

The CIRCA study will determine the incidence and outcomes of critical illness-related cardiac arrest in critical care units in England and Wales. In addition, it will link with national clinical audits (Case Mix Programme and NCAA) to provide detailed information on patient groups at risk, potentially allowing identification of those at greatest risk at admission to critical care. In addition, and unanticipated at the outset of the study, CIRCA will include data from COVID-19 admissions to critical care in England and Wales.

Results will be presented in separate papers and conference abstracts and/or presentations addressing the primary research question and follow-up in line with the protocol and statistical analysis plan and STROBE reporting guidelines.¹⁶

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Study is funded by the Resuscitation Council (UK) (Reference: 6135455598).

ORCID iDs

Robert Darnell

David Harrison

Doug Gould

Matt Thomas

References

Buanes

Heltne

. Comparison of in-hospital and out-of-hospital cardiac arrest outcomes in a Scandinavian community. Acta Anaesthesiol Scand 2014; 58: 316–322.

Fredriksson

Aune

Bång

, et al. Cardiac arrest outside and inside hospital in a community: mechanisms behind the differences in outcome and outcome in relation to time of arrest. Am Heart J 2010; 159: 749–756.

Andersen

Holmberg

Berg

, et al. In-hospital cardiac arrest: a review. JAMA 2019; 321: 1200–1210.

Thoren

Rawshani

Herlitz

, et al. ECG-monitoring of in-hospital cardiac arrest and factors associated with survival. Resuscitation 2020; 150: 130–138.

McHugh

Rochman

Sloane

, et al. Better nurse staffing and nurse work environments associated with increased survival of in-hospital cardiac arrest patients. Med Care 2016; 54: 74–80.

Bircher

Chan

, et al. Delays in cardiopulmonary resuscitation, defibrillation, and epinephrine administration all decrease survival in in-hospital cardiac arrest. Anesthesiology 2019; 130: 414–422

Myrianthefs

Kalafati

Lemonidou

, et al. Efficacy of CPR in a general, adult ICU. Resuscitation 2003; 57: 43–48.

Armstrong

Kane

Oglesby

, et al. The incidence of cardiac arrest in the intensive care unit: a systematic review. J Intensive Care Soc 2019; 20: 144–154.

ICNARC. National Cardiac Arrest Audit . https://www.icnarc.org/Our-Audit/Audits/Ncaa/About (accessed 11th May 2021).

10.

Leloup

Briatte

Langlois

, et al. Unexpected cardiac arrests occurring inside the ICU: outcomes of a French prospective multicentre study. Intensive Care Med 2020; 46: 1005–1015.

11.

Deakin

Soar

Davies

, et al. Special circumstances guidelines. In Resuscitation Council UK 2021 Resuscitation Guidelines. Available at: https://www.resus.org.uk/library/2021-resuscitation-guidelines/special-circumstances-guidelines#specific-causes (accessed 11th May 2021).

12.

Rhodes

Ferdinande

Flaatten

, et al. The variability of critical care bed numbers in Europe. Intensive Care Med 2012; 38: 1647–1653.

13.

Organisation for Economic Cooperation and Development (OECD). Health at a Glance: Europe 2020: State of Health in the EU Cycle . Available at: https://www.oecd-ilibrary.org/sites/82129230-en/index.html?itemId=/content/publication/82129230-en (accessed 11th May 2021).

14.

Hsu

Madsen

Callaham

. Quality-of-life and formal functional testing of survivors or out-of-hospital cardiac arrest correlates poorly with traditional neurologic outcome scales. Ann Emerg Med 1996; 28: 597–605.

15.

EuroQol Group . EQ-5D-5L. Available at: EQ-5D-5L – EQ-5D (euroqol.org) (accessed 11th May 2021).

16.

von Elm

Altman

Egger

, et al. The strengthening the reporting of observational studies in epidemiology (STROBE) Statement: guidelines for reporting observational studies. Available at: https://www.equator-network.org/reporting-guidelines/strobe/ (accessed 10th June 2021).