Whither critical care outreach?
Article Outline
The development of Critical Care Outreach Services (CCOS) has been exponential albeit in various guises such as the Medical Emergency Team (MET) and Intensive Care Liaison nurse in Australia, the Rapid Response Team in the USA and Patient at Risk teams in the UK. Their effectiveness or otherwise on patient survival and morbidity has yet to be demonstrated in prospective multi-centre studies, the evidence base being predominantly founded on retrospective level 3 studies. Systematic reviews of both CCOS and the use of physiologically based early warning systems (frequently referred to as track and trigger systems) are currently being undertaken as part of a larger study funded by the UK National Health Service Research and Development Service Delivery and Organisation Programme but these have yet to be published. This has not, however, prevented influential organisations such as the National Confidential Enquiry into Patient Death and Outcome (NCEPOD) recommending CCOS be established in all hospitals on a 24/7 basis alongside the use of track and trigger systems ‘which should be linked to a response team that is appropriately skilled to assess and manage the clinical problems.’ (NCEPOD, 2005, p. 3).
There are clearly great methodological difficulties associated with research which seeks to investigate interventions which are systems based, not least of which are the vagaries of the participants and the resultant Hawthorne effect. However does the way we work, and the track and trigger systems with which we work, also lead to less than effective management of the deteriorating patient? This editorial seeks to address the following questions:
It is vital that the resuscitation of a patient with suspected severe sepsis or sepsis induced hypoperfusion should commence as soon as possible. It should not be delayed by a skills deficit in identifying and managing the deteriorating patient or dependent on admission to the intensive care unit. However the measurement of blood pressure, one of the key variables associated with the diagnosis of severe sepsis or septic shock, is neither sensitive nor specific. We are all aware of the problems associated with automated versus manual recordings, the competency of the practitioner recording the blood pressure and whether the patient has a relative hypotension even if the systolic pressure is above 90
mmHg. There are also patient groups in whom even accurate blood pressure recordings lack sensitivity and specificity. In particular young fit people who can compensate almost to the point of cardiac arrest and women with raised cardiac output in the third trimester of pregnancy. Given these difficulties it would be more effective to know the degree of tissue hypoperfusion a patient is actually experiencing, particularly if infection is suspected and general inflammatory variables are triggering, i.e. core temperature<36°C>38.3°C, heart rate >90/min, respiratory rate >25, altered mental status. To this end the measurement of serum lactate is important because it can identify tissue hypoperfusion in patients who are not known to be hypotensive.
One general drawback to this is that serum lactate only reflects a global reduction in perfusion rather than one which is organ specific. Nevertheless it may be definitive in those discussions revolving around whether a patient requires more intravascular ‘filling’ or not. A serum lactate of >2mmol aids the diagnosis of tissue hypoperfusion and >4mmol signals the need for goal directed fluid resuscitation, achieving a central venous pressure of >8mmHg together with a superior vena cava oxygen saturation (ScvO2) of >70% or a mixed venous oxygen saturation (SvO2) of >65% (Townsend et al., 2005).
The measurement of serum lactate requires arterial blood gas aspiration. This is to avoid ‘false’ high lactate readings from peripheral venous samples either due to poor peripheral blood flow or the damage caused by the use of tourniquets. Goal directed fluid resuscitation and the restoration of circulating volume necessitate the insertion of a central venous catheter. These are advanced skills both for nurses involved in the CCOS and for junior doctors. The MET system in Australia ensures the attendance of an intensive care doctor skilled in these techniques; however in the UK where CCOS are predominantly led by nurses, this is not always the case.
There are several solutions to this dilemma. The introduction of advanced nurse practitioners with acute care skills, as in the USA, is one possibility. If this were to be considered, the role should be one which extends beyond cannulation to incorporate the ordering of diagnostic tests, in particular chest X-ray, chemical pathology, full blood count and clotting to identify inflammatory variables together with prescribing rights for antibiotics and independence in relation to intravascular filling. It is unlikely that in severe sepsis a single 250ml colloid challenge will be sufficient to restore tissue perfusion. This introduction of the advanced nurse practitioner would not necessarily develop the acute care skills of the patient's medical and nursing team however. Problems could also arise in the interpretation of specific tests if the advanced practice nurse were unaware of these reflecting the patient's underlying pathology rather than the effect of inflammatory mediators. An example of this could be an elevated lactate primarily due to decreased clearance by the liver. Another solution is the introduction of the acute physician (NCEPOD, 2005) who could be accessed by the CCOS for support in the management of deteriorating patients. This would have the added benefit of being a ward-based role which potentially could, through role modelling, improve the management of deteriorating patients. It would also increase the critical mass of staff able to manage acutely deteriorating patients rather then rely solely on the intensive care registrar.
Another possible role for the CCOS in tandem with acute physicians would be to engage in preventative management. Currently, care bundles are used within the domain of intensive care but their introduction to ward areas could potentially reduce the incidence of acute deterioration and critical illness. Based on the evidence that exists already it would be possible to introduce an ‘acute care’ care bundle to ward areas for every patient admitted to acute hospitals. This would include DVT and stress ulcer prophylaxis, sitting patients up at 45° or more, keeping the blood sugar below 10mmol/l (as this was correlated with mortality in van den Bergh's (2001) study) and the institution of the sepsis resuscitation bundle should the patient demonstrate early signs of sepsis (Townsend et al., 2005).
Integration of CCOS and cardiac resuscitation teams is a further possibility, where medical staff are already aligned to resuscitation. The outcome of cardiac arrest is very poor particularly in those who present with pulseless electrical activity (Cohn et al., 2004). The introduction of CCOS was, in part, a response to this indicating the need to manage patients more effectively and reduce the incidence of cardiac arrest, although research has yet to demonstrate this effect (Hillman et al., 2005). One of the main limitations of this study was the Hawthorne effect, alluded to earlier in this editorial, where resuscitation teams had prior knowledge of the MET system and began to function in a similar manner. Given that outcomes are so poor and that prevention is better than cure a combined CCOS/Resus force, aligned to acute physicians, focused on prevention, swift identification and effective management may together decrease the morbidity and improve the survival associated with acute deterioration and critical illness which has yet to be demonstrated across multiple sites.
In conclusion, the future of CCOS continues to be one of change rather than atrophy. An increased emphasis now needs to be placed on the prevention of acute deterioration through the avoidance of adverse events such as pulmonary embolus, acute gastro-intestinal bleeding and respiratory tract infection. Earlier identification of tissue hypoperfusion should be evident through measurement of serum lactate rather than an over reliance on blood pressure readings. Finally, systems should be in place to achieve this through the potential introduction of advanced nurse practitioners or acute physicians and a combined team of resuscitation and CCOS.
References
- Analysis of clinical outcomes following in hospital cardiac arrest. Internal Med J. 2004;34:398–402
- Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial. Lancet. 2005;365:2091–2097
- NCEPOD. An Acute Problem June; 2005.
- Townsend S, Dellinger RP, Levy MM, Ramsey G. Implementing the Surviving Sepsis Campaign. Society of Critical Care Medicine, European Society of Intensive Care Medicine, International Sepsis Forum; 2005.
- Intensive insulin therapy in the critically ill patients. N Engl J Med. 2001;345:1359–1367
PII: S0964-3397(06)00049-8
doi:10.1016/j.iccn.2006.04.001
© 2006 Elsevier Ltd. All rights reserved.
