Do Subspecialty Patients Need Special Evaluation to Screen for Deterioration?*

    loading  Checking for direct PDF access through Ovid


To paraphrase George Orwell’s famous quote from “Animal Farm,” all patients are equal, but some patients are more equal than others. Despite the successful implementation of rapid response teams (RRTs) in pediatrics, the promise of objective criteria for high-risk patients remains unfulfilled. Since the introduction of Early Warning Scores, there have been criticisms and concerns that the representation of subspecialty patients is insufficient in the scoring system, and thus, it would likely either miss deteriorating subspecialty patients or, worse, classify everyone as deteriorating. This further distinguishes pediatrics from adult medicine where there has been improvement in early warning systems to detect at-risk patients and evaluate them (1–3). This success in the adult world is likely due to a larger population of high-risk patients, a higher rate of events, and a smaller range of “normal” vital signs and laboratory data leading to robust, sensitive, and specific warning systems (4). More modest improvements in early warning systems and RRTs (5–8) in the pediatric world have led to a significant decrease in code events outside of the ICU to only 5% of all pediatric in-hospital cardiac arrests (9).
This lack of standardized objective criteria for activating a RRT on the pediatric ward is elucidated in the large Get With the Guidelines Registry (10). Following respiratory complaints, the second most common trigger in this registry was due to staff member concerns about a patient. Interestingly, those rapid response evaluations triggered by staff concerns alone, resulted in a much smaller transfer rate to the PICU, only 19.5% and no significant patient events. The median length of time at the bedside for all rapid response evaluations was 29 minutes, raising concerns about the safety of an ICU-based team outside of this ICU for prolonged periods. This data tell us that there is room for improvement in identification of these patients to decrease both the unnecessary evaluations as well as to prevent episodes of harm. Furthermore, the rarity of code events, lack of standards for PICU admission criteria, and the inability to generalize easily demarcated events such as RRT activation or ICU transfer hinder the ability to objectively compare across patient populations the validity of scoring systems in evaluating deteriorating patients. This is true across institutions as well as within a single institution but across specialties. The need for select ICU interventions, sometimes referred to as “critical deterioration events,” have been suggested as a potential surrogate. The concept that the need for early intervention after transfer implies that the lack of recognition or action outside the ICU may have resulted in more consequential events requiring greater intervention and possible arrest before transfer.
In this issue of Pediatric Critical Care Medicine, Dean et al (11) demonstrate the use of an objective measure to predict clinical deterioration events in general and subspecialty pediatric patients. Applying the modified Brighton Pediatric Early Warning Score (PEWS) to different subspecialty patients to determine the rates of clinical deterioration events, they demonstrate an area under the curve ranging from 0.88 for general medical patients to 0.94 for cardiac patients. This translates to a number needed to evaluate ranging from 3.5 for a high PEWS threshold with low sensitivity to 43 for a lower PEWS threshold with higher sensitivity. These numbers seem high, but it is important to remember that this is not predictive of PICU transfer but rather of clinical deterioration events or those events which we seek to prevent outside the PICU. Based upon the wide range of number needed to evaluate, the authors recommend a tiered approach to the response to elevated PEWS scores.

Related Topics

    loading  Loading Related Articles