If you wish to respond to a paper or other item already published in the BJA, please go to the abstract/full text version of that item and click on the link "E-Letters: Submit a response to the article".
Electronic Letters to:
|
|
E-letters: Electronic letters published:
-
![[Read E-letter]](/icons/misc/sectionDOWN.gif)
Early Warning Scores
- Stephen J Fletcher Guy W Glover (20 July 2007)
-
Worthing PSS: would removing the temperature variable help?
- Pierre-Antoine Laloë (11 July 2007)
|
|
|||
|
Stephen J Fletcher Guy W Glover, Intensive Care Unit Bradford Teaching Hospitals
Send letter to journal:
|
Editor - We read the paper from Duckitt and colleagues 1, and the associated editorial from Smith and Cuthbertson 2 with great interest. Properly validated track and trigger systems are long overdue and the Worthing physiological scoring system is therefore welcome. Intensivists who worked through the ‘critical care without walls’ revolution and participated in ‘outreach’, initially found it was striking to discover how many sick patients were present in their hospitals. It was even more striking however, to realise that a large proportion of these patients never presented to the intensive care unit and nor did they come to our attention via the mechanism of the cardiac arrest call. The suspicion was therefore that subsets of the critically ill may survive with or despite the care of non-intensivist medical teams. Genotypic differences that predispose to differing responses to sepsis and therefore differing outcomes may be one explanation for this 3. Another striking finding is that on current evidence, early critical care intervention using systems such as the MET concept 4, has not been proven to improve outcome; Cuthbertson and Smith explore this issue eloquently in their editorial. One possibility not explored is that in some patients critical care intervention may worsen outcome. For example, in a patient with intra-abdominal sepsis and sepsis syndrome, aggressive fluid therapy in the face of modest hypotension not infrequently results in pulmonary oedema (as leakage from pulmonary capillaries occurs at relatively low pulmonary venous pressures due to increased capillary permeability). Thus single organ failure is turned into two-organ failure and mortality is increased in a patient who was otherwise destined to survive. Whilst this may sound like mischievous conjecture, the theory is not groundless. For example, there are data from North America suggesting that patients with low sickness severity scores who are admitted to critical care units fare worse than those managed on ordinary wards. (Mitchell Levy, Oral presentation, European Society of Intensive Care Medicine meeting, Barcelona 2006). Excess mortality may indicate that we are not properly implementing evidence based treatments, or that we are causing harm by applying them to the wrong subset of patients. For example, drotrecogin alfa was found to be ineffective in patients with an APACHE 2 score less than 25 5. The corollary of all this is that the traditional endpoint used in the creation of track and trigger scores, i.e. mortality, may be inappropriate. The Worthing scoring system has been validated as identifying patients likely to die, but by definition, death is still occurring despite those patients presumably being treated in the intensive care unit (or being deemed not for resuscitation). Therefore one could make the argument that the score only predicts those who will fail to benefit from intensive care support. We would contend that a more appropriate task for track and trigger systems is not to identify patients who will die but to identify those who likely to benefit from critical care intervention. To develop such systems may require studies that are impossible to perform because of ethical considerations (i.e. randomisation to critical care management or not). However it may be that in the absence of such data reliance on currently available track and trigger systems, created using mortality as the outcome variable, may never be sophisticated enough to provide significant benefit to the critically ill. S J Fletcher G W Glover Bradford, UK E-mail: sjfletcher@doctors.org.uk 1. Duckitt RW, Buxton-Thomas R, Walker J et al. Worthing physiological scoring system: derivation and validation of a physiological early-warning system for medical admissions. An observational, population- based single-centre study Br J Anaesth 2007; 98:769-74 2. Cuthbertson BH, Smith GB. A warning on early-warning scores! Br J Anaesth 2007; 98: 704-6 3. Villar J, Maca-Meyer N, Pérez-Méndez L, Flores C. Bench-to- bedside review: Understanding genetic predisposition to sepsis. Crit Care 2004; 8: 180-9 4. Lee A, Bishop G, Hillman KM, Daffurn K. The Medical Emergency Team. Anaesth Intens Care 1995; 23: 183-186 5. Abraham E, Laterre P, Garg R et al. Drotrecogin Alfa (Activated) for Adults with Severe Sepsis and a Low Risk of Death. N Engl J Med 2005; 353: 1332-41 Conflict of Interest:None declared |
|||
|
|
|||
|
Pierre-Antoine Laloë
Send letter to journal:
|
Editor – I read about the Worthing physiological scoring system (PSS) with particular interest.1 I work in a district general hospital comparable to that of Duckitt and colleagues where we have recently introduced a similar, although not so robustly devised, physiological early-warning system for surgical inpatients. I have made three observations from the Worthing study. Firstly, in line with previous studies, temperature was only significant in cases of hypothermia < 35.3°C (incidentally temperature was omitted from Methods which I presume to be auricular). Secondly, as shown in table 6, mean temperature values did not differ with age, therefore temperature might not be a confounding factor in the age differences observed in table 4. Thirdly, the overall data collection was perhaps typical of busy admission units with complete data only available for 76% (4286 of 5645) patients. My query is double. Does removing the temperature variable from the Worthing PSS significantly change its area under the receiver operating curve? Does having a five rather than six variable based Worthing PSS render it more user-friendly? I wonder in the 1359 cases of incomplete data how many and which physiological variables were not recorded. I commend the authors' aim to produce a validated physiological early-warning system. However a scoring system where one in every four admissions lacks data clearly has some teething issues. The balance between reliability and usability is vital for the success of any new system.
Pierre-Antoine Laloë
Conflict of Interest:None declared |
|||