Unfortunately, most in-hospital cardiopulmonary arrests are caused by either asystole (39%) (Fig. 1.2) or pulseless electrical activity (PEA) (37%) (i.e. no pulse, but an ECG trace that would normally be expected to produce a cardiac output, (Fig. 1.3) Both of these non-shockable rhythms are associated with a very poor outcome (12% and 11% respectively – Meaney et al. 2010; Resuscitation Council UK 2010). These arrests are usually not sudden nor are they unpredictable: cardiopulmonary arrest usually presents as a final step in a sequence of progressive deterioration of the presenting illness, involving hypoxia and hypotension (Resuscitation Council UK 2010). In some studies it has been alleged that patients with abnormal vital signs before cardiac arrest have improved survival rates, compared with those who have normal vital signs, thus indicating the preventability of cardiac arrests (Skrifvars et al. 2006; Peberdy et al. 2008). Patients who experience cardiac arrest with non-shockable rhythms have a reduced chance of survival, so a vital approach that is likely to be successful is prevention of the cardiopulmonary arrest if at all possible. For this prevention strategy to be successful, early recognition and effective treatment of patients at risk of cardiopulmonary arrest are paramount. This strategy may prevent some cardiac arrests, deaths and unanticipated ICU admissions (Nolan et al. 2005). The statistics are irrefutable in that antecedents are present in 79% of cardiopulmonary arrests, 55% of deaths and 54% of unanticipated ICU admissions (Kause et al. 2004).
Suboptimal Critical Care
In a seminal study (McQuillan et al. 1998), it was demonstrated that the management of deteriorating inpatients in the UK is frequently suboptimal.
Two external reviewers assessed the quality of care in 100 consecutive unplanned admissions to ICU:
- Twenty patients were deemed to have been well managed and 54 to have received suboptimal management, with disagreement about the remainder.
- Case mix and severity of illness were similar between the groups, but the ICU mortality rate was worse in those whom both reviewers agreed received suboptimal care before ICU admission (48% compared with 25% in the well-managed group).
- Admission to ICU was considered late in 37 patients in the suboptimal group. Overall, a minimum of 4.5% and a maximum of 41% of admissions were considered potentially avoidable.
- Suboptimal care contributed to morbidity or mortality in most instances.
- The main causes of suboptimal care were considered to be failure of organisation, lack of knowledge, failure to appreciate clinical urgency, lack of supervision and failure to seek advice. Junior staff frequently fail to recognise deterioration and appreciate the severity of illness and, when therapeutic interventions are implemented, these have often been delayed or are inappropriate. The management of deteriorating patients is a significant problem, particularly at night and at weekends, when responsibility for these patients usually falls to the on-call team whose main focus is on a rising tide of new admissions (Baudoin and Evans 2002).
Despite gaining criticism for alleged methodological weakness, this groundbreaking study was the catalyst for many other reviews and studies pertaining to the management of the deteriorating patient and the recognition of clinical antecedents.
Even more disturbingly, earlier studies of events leading to ‘unexpected’ in-hospital cardiac arrest indicate that many patients have clearly recorded evidence of marked physiological deterioration before the event, without appropriate action being taken in many cases (Schein, et al. 1990; Franklin and Matthew 1994).
Deficiencies in critical care management frequently involve simple aspects of care, e.g. failure to recognise and effectively treat abnormalities of the patient’s airway, breathing and circulation, incorrect use of oxygen therapy, failure to monitor the patient, failure to ask for help from senior colleagues, ineffective communication, lack of teamwork and failure to use treatment limitation plans (McQuillan et al. 1998; Hodgetts et al. 2002).
The ward nurse is uniquely positioned to recognise that the patient is starting to deteriorate and to call for appropriate help (Adam and Osborne 2005). However, recognition of the deteriorating patient remains inadequate (Beaumont et al. 2008; Odell et al. 2009). Strategies to prevent in-hospital adverse events are based on certain elements: the early identification of the deteriorating patient, escalation of care, timely and appropriate management, and the early transfer to critical care areas as appropriate.
Guidelines have been produced to enable organisations and individual clinicians to design, implement and evaluate a strategy to identify and manage the deteriorating patient to improve outcomes.
Best Practice – Measurement and Documentation of Observation
The accurate measurement of physiological observations is essential in detecting the deteriorating patient and reducing adverse events.
- All patients in acute care settings should have observations performed
- All patients should have observations performed on admission (NICE 2007)
- Observations should comprise as a minimum:
- Respiratory rate
- Oxygen saturation
- Heart rate
- Blood pressure
- Temperature
- Level of consciousness (NICE 2007; ACSQHC 2010)
- Respiratory rate
- Frequency of observations should be consistent with the condition of the patient, but at least once every 8 hours and documented in the monitoring plan
- Observation charts should display observations in graphic format
Best Practice
Escalation process should:
- be a formal documented escalation process that should be available and applicable to all patients at all times of the day
- support and authorise the clinician to escalate care until a satisfactory response is achieved
- be tailored to the organisation’s characteristics and available resources
- allow for a graded response according to the degree of physiological abnormality, e.g. increase in frequency of observations, interventions by ward staff, review by medical team, summoning urgent assistance, transfer to a higher level of care
- specify the levels of abnormality for which care should be escalated, for whom the care is escalated, who else to contact when care is escalated, timeframes for requested response and back-up options for obtaining a response if this fails
- consider the needs of patients with advance directives
(ACSQHC 2010)
CLINICAL SIGNS OF DETERIORATION
The clinical signs of deterioration and critical illness are usually similar regardless of the underlying cause, because they reflect compromise of the respiratory, cardiovascular and neurological systems (Nolan et al. 2005). These clinical signs are commonly:
- tachypnoea/breathlessness
- hypotension
- altered conscious level (e.g. lethargy, confusion, restlessness or falling level of consciousness). (Kause et al. 2004)
Tachypnoea, a particularly important indicator of an at-risk patient (Goldhill et al. 1999), is the most common abnormality found in critical illness (Goldhill et al. 2004). In an early study it was identified that a raised respiratory rate (>27/min) occurred in 54% of patients in the 72 hours preceding cardiac arrest, most of which occurred 72 hours before the event (Fieselmann et al. 1993).
TRACK-AND-TRIGGER SYSTEMS
Most hospitals in the UK now use some form of track-and-trigger system, e.g. early warning scoring system (EWS), to identify deteriorating patients, and are an important component of risk management strategies (Donahue and Endacott 2010). There is a lack of rigorous evidence to support any single track-and-trigger system above the others, but it is important that organisations introduce a tool based on best evidence surrounding reliability, validity, specificity and sensitivity, and on local requirements (Gao et al. 2007). It has been recommended that some type of physiological track and trigger system should be used to monitor all adult patients in acute hospital settings (NICE 2007).
EARLY WARNING SCORING SYSTEMS
The recommendations from Comprehensive Critical Care (Department of Health 2000) were the widespread implementation of EWS systems and outreach services. The EWS systems have been developed as a tool to enable ward staff to combine their regular observations to produce a physiological score (Sharpley and Holden 2004). They are based on the premise that there is a common physiological pathway of deterioration in the critically ill patient, which can be detected by simple ward-based observations (Goldhill 2001).
The most commonly used scoring system is an aggregated system wherein a weighted score is attached to a combination of blood pressure, pulse, respiratory rate, oxygen saturations, temperature, urine output and simplified level of consciousness (AVPU) (Fig. 1.4). A trigger response is activated when the patient reaches a certain score and the designated escalation process is triggered. Nursing and other healthcare staff must then alert the designated expert help following local protocols. Escalation policies are put in place whereby a failure to improve (or to receive prompt help) results in the immediate contact of more senior members (including consultant staff) (Baudouin and Evans 2002). Clear guidelines should be drawn up to guide the nurse in when escalation is necessary and whom to contact for help (Fig. 1.5).
(Reproduced by kind permission of Fiona Legg and Beverley Ewens.)
Each hospital should instigate a track-and-trigger system that allows rapid detection of the signs of early clinical deterioration and an early and appropriate response (NCEPOD 2005). These track-and-trigger systems should be robust, cover all in-patients and be linked to a response team that is appropriately skilled to assess and manage the clinical problems (NCEPOD 2005).
The main advantages of EWS systems are as follows (Gwinnutt 2008):
- Simplicity: only the basic monitoring equipment is required (usually readily available on acute wards)
- Reproducibility between different observers
- Applicability to multiprofessional team
- Minimal staff training required.
The early recognition and treatment of the deteriorating patient can minimise the occurrence of adverse events and reduce the level of intervention that a patient may have otherwise needed (ACSQHC 2010). However, these clinical signs of deterioration are often subtle and can go unnoticed. It is therefore essential that tools and systems, based on best evidence and designed for individual organisations, be developed, put in place and evaluated, and available to help the practitioner identify signs of deterioration. Ultimately this may prevent adverse events and improve patient outcomes. However, over-reliance on a scoring system to identify the deteriorating patient should be used as an adjunct to clinical judgement not as a replacement for it.
The efficacy of a variety of EWS tools designed to enable practitioners to identify the deteriorating patient has undergone extensive evaluation and discussion in the literature. However, the reliability and validity of these varying systems are lacking and further work to validate them is recommended (Gao et al. 2007).
Best Practice – Early Warning Scores (EWS)
EWS systems should be based on the best available evidence.
EWS should reflect validity, reliability, specificity and sensitivity:
- EWS should be designed to reflect subtle changes in condition
- EWS chart should be straightforward to use and unambiguous in its design
- Implementation should be planned and coordinated
- Extensive education strategy is needed before implementation
- Specific escalation policy should be attached to the EWS, e.g. whom to call and when
- EWS calling criteria can be adjusted for specific patients, e.g. chronic disease
- Ongoing audit of EWS charts should be undertaken to assess completeness and accuracy
- Ongoing review of specific incidents where calling criteria were not adhered to
- There should be on-going education of staff and auditing of tool
- EWS systems should be designed for the organisation’s requirements and resources
- EWS systems should be used as an adjunct to clinical decision making not as a replacement for.