Incidence of in-hospital cardiac arrest at general wards before and after implementation of an early warning score

In-hospital cardiac arrest (IHCA) is a substantial burden for the patients, for their family and for the healthcare system in general [1]. However, throughout time, research in IHCA has received little attention compared to other high-risk in-hospital conditions even though IHCA is regarded a reversible condition [2] if caught in time. The critical role of early identification of patients with deviating vital signs is underlined in the European Resuscitation Council’s (ERC) guidelines. Here, the chain of survival states that an effective track and trigger system and staff education are needed to decrease the incidence of IHCA. Especially, patients admitted to unmonitored general wards are at risk of deteriorate unnoticed by the ward staff [3]. A deterioration of vital signs prior to an IHCA seems to occur in as many as 80% of patients [4]. Medical emergency team (MET) has been implemented to assist general wards as the efferent part of such a track and trigger system [5]. These teams help to intervene before patients become so ill that an IHCA happens or to initiate a discussion regarding end-of-life care planning and to deescalate treatment rather than to intensify it [6].

In 2012, the National Early Warning Score (NEWS) was developed by the Royal College of Physicians [7]. Hereby, standardising the more than 100 track and trigger systems used at the time [5]. The NEWS system is a systematic approach for general wards used for observation of patients. It is the afferent part of the track and trigger system [5]. The effectiveness of the system is dependent on systematic monitoring and critical, clinical assessment of patients. Furthermore, the whole response system only works if actions are taken upon deviating vital signs by the ward staff.

A Danish study by Andersen et al has estimated that 2500 IHCA happen at Danish hospitals annually. Furthermore, the 30-day survival after IHCA is estimated to be 27% [8]. The NEWS system was implemented in the Capital Region of Denmark on June 1st 2012 [9] followed by an escalation protocol (Table 1) in attempt to decrease deterioration of acutely ill patients and thereby decrease the risk of IHCA among patients admitted to general wards.

This study aims to assess the incidence of IHCA at general wards before and after the implementation of the NEWS system in 2012.

This was conducted as a retrospective quality assurance study based on data from our institution. Data about the IHCA events and return of spontaneous circulation (ROSC) was obtained from The Danish In-Hospital Cardiac Arrest Registry (DANARREST) from 2017 and forward, while earlier data was obtained from a local database. From DANARREST, data on personal identification number (CPR number), sex, ROSC and 30-day survival were provided [11]. The IHCA were divided into two groups relative to the implementation of the NEWS system, which was implemented on 1st of June 2012 (Pre-EWS (2006–2011) versus Post-EWS (2013–2018)) [9]. All patients suffering an IHCA in 2012 were excluded to avoid bias.

During the two time periods, the cardiac arrest team was activated 3073 times overall. In total 1734 confirmed IHCA with the need for resuscitation were identified. Of these, we excluded 723 IHCA due to location, and 73 IHCA due to age. The remaining 938 IHCA in general wards were analysed. Of these, 444 IHCA occurred before the 1st of June 2012 (pre-EWS group (2006–2011)) while 494 IHCA happened after that date (post-EWS group (2013–2018)) (Fig. 1).

OR of ROSC was 1.14 (95% CI [0.88;1.47], p = 0.32). For 30-day survival, the OR was 1.30 (95% CI [0.96;1.75], p = 0.09). The p-values of adjusted ORs for age, sex, and primary rhythm indicate that ROSC was not affected, while 30-day survival significantly improved.

Roberts et al showed that a medium or high EWS was correlated to an increased risk IHCA [1]. However, another study has shown that the implementation did not affected the incidence or even an increase in the incidence rate occurred [12], as we found in our study. The continuous monitoring of patients with the NEWS system is a staff-intensive assignment and it was expected that a beneficial effect could be detected after implementation. Even though, it has not been possible to show a positive effect on IHCA incidence at our institution, the NEWS system has given nursing staff and physicians clinical tools for handling the acutely deteriorating patients with the escalation protocol [1]. The incidence of IHCA at our institution is somewhat lower than reported in other studies in western countries where incidence of IHCA ranges between 1.5 to 10.2 per 1000 admissions [13, 14]. One possible explanation for the low incidence in our study could be due to the exclusion of cardiac arrest in intensive care units, trauma units, and cardiac catherization labs where a fair number of IHCA happen (Fig. 1).

Even though, NEWS can be used to stratify patients at risk of IHCA, it does not consider the increasing age in the population (Table 1). Physiological parameters are known to change in the elderly population [15]. Age has been identified as an independent risk factor for development of cardiac arrest [16]. Studies have shown that the accuracy of EWS systems decrease with increasing age, especially when age is above 65 years. This might have contributed to the non-significant result of our study.

Several other problems have been identified as reasons for the constant incidence rate. Petersen et al found that under-staffing and time constrains were reasons for not adhering to the protocol [17]. Provided that the NEWS protocol was followed, and vital signs were monitored continuously, the system only works if actions are taken upon deviating measurements. Nurses should contact the on-call physician according to the escalation protocol if the NEWS score was 3 in a single variable (Table 1). That did not always happen and that seemed to be related to lack of experience with acutely ill patients or bad experiences with the MET previously [18]. Hereby, treatment of deteriorating patients was delayed, and the risk of cardiac arrest was increased. Problems with assistance from the MET have also been identified [17] even though, the latest guidelines of ERC specifically underline the importance of teamwork between staff at general wards and the MET [3]. These factors could, perhaps, explain why a decrease in the incidence of IHCA was not seen after the implementation of the NEWS. In addition, the number of admissions has increased during the post-EWS (2013–2018) time period and this may also affect the primary result of the study, drawing the conclusion towards the null.

An increase in OR of ROSC and 30-days survival was seen. With the NEWS system ward staff are giving a systematic approach to the acutely deteriorating patient. This could mean that if the monitoring frequency was high enough some patients could be caught right after the cardiac arrest and thereby improving their chance of survival. However, this is speculative, and the system is not intended to improve outcome after IHCA. ROSC and 30-day survival are more likely to be influenced by changing guidelines, response time and training level of the cardiac arrest team, the ward staff, and post resuscitation care. Our institution has followed these guidelines since 2000, and all members of the cardiac arrest team are regularly trained and recertified regarding resuscitation according to the newest guidelines.

This study suggests that the implementation of the EWS system at our hospital did not decrease the incidence rate of in-hospital cardiac arrest at general wards.