A pragmatic checklist to identify pediatric ICU patients at risk for cardiac arrest or code bell activation


In-hospital cardiac arrest is a rare event associated with significant morbidity and mortality. The ability to identify the ICU patients at risk for cardiac arrest could allow the clinical team to prepare staff and equipment in anticipation.


This pilot study was completed at a large tertiary care pediatric intensive care unit to determine the feasibility of a simple checklist of clinical variables to predict deterioration. The daily checklist assessed patient risk for critical deterioration defined as cardiac arrest or code bell activation within 24 h of the checklist screen. The Phase I checklist was developed by expert consensus and evaluated to determine standard diagnostic test performance. A modified Phase II checklist was developed to prospectively test the feasibility and bedside provider “number needed to train”.


For identifying patients requiring code bell activation, both checklists demonstrated a sensitivity of 100% with specificity of 76.0% during Phase I and 97.7% during Phase II. The positive likelihood ratio improved from 4.2 to 43.7. For identifying patients that had a cardiac arrest within 24 h, the Phase I and II checklists demonstrated a sensitivity of 100% with specificity again improving from 75.7% to 97.6%. There was an improved positive likelihood ratio from 4.1 in Phase I to 41.9 in Phase II, with improvement of “number needed to train” from 149 to 7.4 providers.


A novel high-risk clinical indicators checklist is feasible and provides timely and accurate identification of the ICU patients at risk for cardiac arrest or code bell activation.

Cardiopulmonary resuscitation; Intensive care units; Pediatric; Hospital rapid response team
CA, cardiac arrest; CBA, code bell activation; CPR, cardiopulmonary resuscitation; NNT, number needed to train; PICU, pediatric intensive care unit

A Spanish translated version of the summary of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2015.11.017.

Correspondence to: Department of Critical Care, The Children’s Hospital of Philadelphia, 3401 Civic Center Blvd Philadelphia, PA 19104.

Complete Article – Requires log in

Dana E. Nilesa, Maya Dewana, , , Carleen Zebuhrb, Heather Wolfea, Christopher P. Bonafidea, Robert M. Suttona, Mary Ann DiLibertoa, Lori Boylea, Natalie Napolitanoa, Ryan W. Morgana, Hannah Stinsona, Jessica Leffelmana, Akira Nishisakia, Robert A. Berga, Vinay M. Nadkarnia
a The Children’s Hospital of Philadelphia, Philadelphia, PA, USA
b Children’s Hospital Colorado, Denver, CO, USA
Received 3 August 2015, Revised 8 November 2015, Accepted 26 November 2015, Available online 17 December 2015

Content for tab4

Content for tab6

Proper Development of an Early PICU Readmission Risk Prediction Tool

Purpose: Few tools are available to reduce readmissions to PICU, a proposed quality measure reflecting outcomes important to the patient, family and health system. This retrospective case-control study was designed to a) assess the applicability of an adult risk prediction score (Stability and Workload Index for Transfer, SWIFT) and, b) create a pediatric version (Prevention Of PICU Early Readmissions, PROPER).
Methods: After IRB approval, 76 unplanned early ( Results: Readmitted patients were younger (33% of readmissions were less than 1 year of age, 24% of controls; p=0.05) and weighed less (45% readmissions had weight < 15 centile, 28% in controls; p=0.009). There were no differences with gender, race or admission Pediatric Index of Mortality (PIM) scores. Higher proportion of patients in the readmission group had a Pediatric Cerebral Performance Category (PCPC) in the moderate to severe disability category (42% vs. 25%, p<0.01). More of the readmitted patients were admitted from ED (50% vs. 29% p<0.001) and fewer were postsurgical (32% vs. 52%, p=0.003). No readmission differences were noted for patients admitted in busier PICU seasons or with day of week of discharge. Similarly no difference was observed with median PICU or floor census at the time of PICU discharge. There was a much higher proportion of patients on supplemental oxygen in the readmissions group compared to the non-readmissions group (14% vs 1%, p<0.01). Only 2/5 (Source of admission and GCS score, significant; while last measured P/F ratio and PaCo2, Total PICU LOS, insignificant)) of the categories in SWIFT model were significantly different and the median SWIFT score of 13, was identical in the two groups. A seven category PROPER score (Age 0.21% at discharge, Last GCS<15) was created based on multiple logistic regression model , with weights to the individual components assigned based on the odds ratio of the risk. The AUC for prediction of readmission of this model was 0.757 (0.690 to 0.824).
Conclusion: We have created a preliminary model for predicting patients at risk for early readmissions to the PICU from hospital floor. This score needs to be validated prospectively as well as externally. SWIFT score is not applicable for predicting risk for pediatric population.

Content for tab2


The Cardiac Children’s Hospital Early Warning Score (C-CHEWS)

Inpatient pediatric cardiovascular patients have higher rates of cardiopulmonary arrests than other hospitalized children. Pediatric early warning scoring tools have helped to provide early identification and treatment to hospitalized children experiencing deterioration thus preventing arrests from occurring. However, the tools have rarely been used and have not been validated in the pediatric cardiac population. This paper describes the modification of a pediatric early warning scoring system for cardiovascular patients, the implementation of the tool, and its companion Escalation of Care Algorithm on an inpatient pediatric cardiovascular unit.


Arrest Prevention

Pediatric cardiopulmonary arrests have been reported in 0.7–2% of all pediatric inpatient admissions (Reis, Nadkarni, Perondi, Grisi, & Berg, 2002; Slonim, Patel, Ruttimann, & Pollack, 1997; Suominen, Olkkola, Voipio, Korpela, Palo, & Rasanen, 2000) and 5.5–14% of intensive care unit (ICU) admissions (Reis et al., 2002; Rhodes et al., 1999; Suominen et al., 2000) despite diligent monitoring (Akre, Finkelstein, Erickson, Liu, Vanderbilt, & Billman, 2010; Nadkarni et al., 2006; Reis et al., 2002; Suominen et al., 2000) and advances in medicine and technology. Survival to discharge outcomes are poor (11–37%) for children that experience an in-hospital cardiopulmonary arrest (Brilli et al., 2007; Lopez-Herce et al., 2004; Meaney et al., 2006; Nadkarni et al., 2006; Parra et al., 2000; Reis et al., 2002; Samson, Berg, & Berg, 2006; Samson, Nadkarni, et al., 2006; Slonim et al., 1997; Suominen et al., 2000; Tibballs & Kinney, 2009; Young & Seidel, 1999). Symptoms of deterioration may be present 6–12hours prior to arrest events, had these symptoms been recognized and treated sooner, almost two-thirds of in-hospital pediatric cardiopulmonary arrests may have been prevented (Pearson, Ward- Platt, Harnden, & Kelly, 2010; Akre et al., 2010; Parshuram, Hutchinson, & Middaugh, 2009; Schein, Hazday, Pena, Ruben, & Sprung, 1990; Tibballs & Kinney, 2009; Tume, 2007). “Given the dismal survival rate of in-hospital cardiac arrest, it is critical to develop systems that recognize predictable clinical warning signs and intervene before patients reach the point of arrest” (VanVoorhis & Willis, 2009, p. 919).
To improve outcomes for patients at risk for clinical deterioration and cardiopulmonary arrest, hospitals have been charged by several international committees to implement systems that identify significantly abnormal values and then trigger an immediate treatment response (Berwick, Calkins, McCannon, & Hackbarth, 2005; DeVita et al., 2006; Peberdy et al., 2007). Hospitals initiated rapid response teams (RRTs), also known as patient at risk teams (PART), critical care outreach (CCO), or medical emergency teams (MET), as an adjunct to their code blue teams to provide this immediate treatment for patients that are identified as being at risk for deterioration and possible arrest (Brilli et al., 2007; Hanson et al., 2009; Hillman, Parr, Flabouris, Bishop, & Stewart, 2001; Hunt et al., 2008; Salamonson, Kariyawasam, van Heere, & O’Connor, 2001; Sharek et al., 2007; Tibballs & Kinney, 2009; Tibballs, Kinney, Duke, Oakley, & Hennessy, 2005; ul-Haque, Saleem, Zaidi, & Haider, 2010; VandenBerg, Hutchison, & Parshuram, 2007; VanVoorhis & Willis, 2009; Zenker et al., 2007). The RRTs are defined as an interdisciplinary group that “resemble Code teams in that they are staffed by health care professionals…Unlike a Code team, a RRT is summoned before a code occurs…to initiate changes in care that prevent the arrest, or by facilitating transfer to an intensive care unit” (Berwick et al., 2005, p. 324). Pediatric RRTs have been composed of PICU physicians, ICU RNs, respiratory therapists, ED physicians and/or a supervisor for patient placement (Brilli et al., 2007; Hanson et al., 2009; Sharek et al., 2007; Tibballs & Kinney, 2009; Tibballs et al., 2005; ul-Haque et al., 2010; VanVoorhis & Willis, 2009; VandenBerg et al., 2007; Zenker et al., 2007). Pediatric RRTs typically respond to the bedside within 5–15minutes of activation to assess patients, write orders for any diagnostic studies or interventions, discuss management with the primary team, and determine optimal location for the patient (Brilli et al., 2007; Hanson et al., 2009; Sharek et al., 2007; Tibballs & Kinney, 2009; Tibballs et al., 2005; ul-Haque et al., 2010; Zenker et al., 2007). Studies have reported reduction in pediatric inpatient cardiopulmonary arrests, reduction in mortality rates, and improved survival outcomes post-arrest following the implementation of RRTs (Chan, Jain, Nallmothu, Berg, & Sasson, 2010; Chapman, Grocott, & Franck, 2010; Hunt et al., 2008; Tibballs & Kinney, 2009).
Activation criteria for when to call RRTs have been developed by hospitals based upon retrospective reviews and/or clinician consensus (Brilli et al., 2007). Activation criteria may be a combination of physiological parameters and/or subjective assessments. Early warning scoring tools are tools that may be used as activation triggers for hospitals’ RRTs. There are three types of early warning tools: (1) single and multiple parameter systems which trigger a response when one or more parameters achieve a defined threshold; (2) aggregate systems which weigh observations based upon abnormality and a summary of the scores are achieved; and (3) combination systems which have single or multiple parameter systems with aggregate weighted scoring systems (Gao et al., 2007).

Pediatric Early Warning Scores

Pediatric early warning scores (PEWS) tools have been created based on previously developed adult early warning scoring tools. Pediatrics create a unique challenge in the development of early warning scoring tools in that vital sign norms are aged-based whereas in adults these norms are more finite (Brilli et al., 2007). The PEWS published by Monaghan (2005) (Figure 1) is an aggregate tool based on three assessment domains: behavior, cardiovascular and respiratory with each domains’ score ranging from 0 to 3, with 3 being the highest severity of illness (Monaghan, 2005). Components of the PEWS’ domains are based on bedside physical assessments and do not require familiarity with the patient or patient’s history or clinical values (i.e. recent laboratory values), which contributes to the ease of bedside use compared to other pediatric early warning scoring tools which do require additional patient informa- tion (Duncan, 2006; Duncan, Hutchison, & Parshuram, 2006; Edwards, Powell, Mason, & Oliver, 2009; Haines, Perrott, & Weir, 2006; Tibballs, 2006). Nurses complete the assessment, total the score, and are guided to follow a four- tiered escalation of actions guide based upon the PEWS score (Monaghan, 2005; Tucker, Brewer, Baker, Demeritt, & Vossmeyer, 2008). A separate study of this PEWS tool demonstrated that critical PEWS scores occurred a median of 11.5 hours prior to events with the shortest time preceding the event to be 35 minutes (Akre et al., 2010). This PEWS was validated in a cohort of pediatric patients admitted to a general medicine unit, the area under the receiver operating characteristic curve was 0.89 (95% CI = 0.84–0.94, p b .001) (Tucker et al., 2008).
In 2008, Children’s Hospital Boston, an academic tertiary pediatric institution, participated in the Child Health Corporation of America Collaborative, “Eliminating Codes on the Inpatient Units.” As part of this collaborative, the previously mentioned PEWS was modified into the Children’s Hospital Early Warning Score (CHEWS). The CHEWS incorporated the PEWS’ domains and scoring, plus the addition of two subjective domains of “family concern” and “staff concern” which add one point each to a patient’s score if it is present (Kleinman & Romano, 2010). Higher scores are indicative of higher severity of deterioration symptoms and will ‘trigger’ the nurse to activate resources to the patient’s bedside based on a three-tiered, color-coded Escalation of Care Algorithm (Figure 4). The patient’s nurse is responsible for tallying the score at the time of vital sign assessment, typically every 4hours. A colored indicator (green, yellow or red), representative of the patient’s last CHEWS score, is placed by patients’ names on the unit’s locator board providing easy visibility of all the patients’ CHEWS statuses for the unit. The CHEWS system was first piloted in 2008 on three surgical units with a total of 90 beds. During the pilot phase there was an increase in ICU evaluations (non-urgent assessment by an ICU MD or ICU RN) with an associated decrease in ICU STAT (rapid assessment by an ICU MD, ICU RN, respiratory therapies and an intermediate care unit MD) and code blue calls (Kleinman & Romano, 2010). The tool and algorithm were implemented throughout the rest of the inpatient medical and surgical units following the success of the pilot. Based on the positive evaluation of the CHEWS tool, the hospital’s leadership directed that the tool also be used in the acute care cardiovascular unit.

Sample and Setting

Pediatric cardiovascular patients have the highest inci- dence of cardiopulmonary arrests as compared to other hospitalized children (Berg, Nadkarni, Zuercher, & Berg, 2008; Hunt et al., 2008; Parra et al., 2000; Rhodes et al., 1999; Samson, Nadkarni, et al., 2006). They are unlike other pediatric populations whose arrest etiology is typically respiratory failure and/or circulatory shock (Berg et al., 2008; Lopez-Herce et al., 2004; Nadkarni et al., 2006; Reis et al., 2002; Samson, Berg, & Berg, 2006; Samson, Nadkarni et al., 2006; Tibballs & Kinney, 2009). Instead, cardiac patients have a different arrest etiology with arrhythmia accounting for 41% of acute decompensation events (Berg et al., 2008; Parra et al., 2000; Rhodes et al., 1999; Samson, Nadkarni, et al., 2006). In addition to arrhythmias, children with cyanotic congenital heart defects may have baseline cyanosis which would be atypical of other pediatric populations. The sample population for the validation study of the original PEWS did not include cardiac patients and was limited to patients admitted to a single medical unit (Tucker et al., 2008). With these differences in mind, it was unclear whether the previous PEWS or CHEWS tools would be effective for pediatric cardiovascular patients. Therefore it was important to study the CHEWS tool for feasibility, relevance, and agreement prior to implementing it in a pediatric cardiac unit.
Our hospital’s cardiovascular unit is a 42-bed cardiac medical and surgical telemetry unit with patients ranging in age from newborn to adult. Within this age range more than half the patient population is less than 1 year of age. Ten of the beds are considered “higher dependency” beds where the nurse to patient ratio is 1:2, the remaining beds are staffed at 1:3. Patients needing the higher dependency beds are those who may be less clinically stable, such as those requiring inotropic support, and require either more nursing care and/ or more frequency assessment.
The purpose of this manuscript is to describe the implementation and subsequent modifications of the CHEWS tool and its companion Escalation of Care Algorithm for pediatric cardiovascular patients and early detection of deterioration and prevention of cardiopulmonary arrests or unplanned transfers to a cardiac ICU (CICU).


Tool Modification

A pilot study consisting of current electronic health record documentation and clinician interview was implemented on the cardiac unit. A single staff nurse, qualified in the use of the CHEWS tool, scored all the patients (n = 27; observations = 157) on the unit during two consecutive 12-hour shifts. Scores were based on docu- mentation in patients’ electronic health records. The pilot study nurse concurrently interviewed the charge nurse and the patients’ nurses, nurse practitioners or fellows and asked each of them to identify which of the their patients were most acute and/or had them concerned. Nurses indicated during the interviews whether patients’ families had concerns or were absent from the bedside and this information was utilized to score the “family concern” domain of the CHEWS. Data from the patients’ clinical events during the pilot, bed assignment (higher dependency bed or not), and the clinicians’ assessments from the interviews were documented.
The clinicians’ assessments, patients’ clinical events during the pilot, bed assignment and the calculated CHEWS scores were compared. There was consistent agreement about patients’ acuity among the clinicians’ assessments, bed assignments, and clinical events and these were used to describe the patients’ clinical presentations. Nearly one-third (29.6%, n = 8) of the patients had lower CHEWS scores than the acuity severity of their clinical presentation should have warranted (Figure 2). Of the patients that scored too low, three patients were urgently transferred to the CICU during the pilot, with one being intubated upon arrival to the CICU. None of the three patients’ CHEWS scores were above a normal range and therefore would not have triggered an escalation of care response using the CHEWS tool.
An expert multidisciplinary panel from the CICU, ICU and cardiac unit reviewed the patients’ clinical presentations and CHEWS scores. The following areas were identified as sources for the score discrepancies:
• Behavior: “sleeping appropriately” was absent from the CHEWS resulting in sleeping patients unnecessar- ily scoring 1 point for this behavior.
• Cardiovascular: presence of arrhythmia was absent from the CHEWS; heart rate range limits in the CHEWS did not account for the wide age range of patients, especially the newborns and infants.
• Respiratory: presence of apnea or cyanosis was absent from the CHEWS; oxygen flow rates on the CHEWS was too high for younger patients; and respiratory rate range limits of the CHEWS did not accommodate the wide age range of patients, especially the newborns and infants.
The tool was then modified to account for these variables and became the Cardiac Children’s Hospital Early Warning Score (C-CHEWS) tool. A second pilot was conducted with the new C-CHEWS tool using the previously described methods (n = 53; observations = 312). Analysis of the data collection revealed 7.5% (n = 4) of the patients’ C-CHEWS scores did not correlate with the acuity of their clinical picture, however this time it was an equal mix of patients either scoring too high or too low (Figure 2). Analysis revealed that the presence of patients’ baseline abnormalities accounted for these discrepancies:
• Behavior: baseline seizures
• Cardiovascular: baseline arrhythmias
• Respiratory: baseline use of supplemental oxygen flow
rate, baseline cyanosis.
The tool was modified such that should a patient have any of these pre-existing abnormalities at baseline they would not score high, whereas if a patient had a new onset of any of those clinical findings, or it was unknown whether this was normal for the patient (i.e. new admission), the findings would still generate a higher C-CHEWS score.
The third and last pilot event (n = 20; observations = 119) with the updated C-CHEWS tool demonstrated 100% of the C-CHEWS scores matched the acuity of patients’ clinical presentations (Figure 2). The final version of the C-CHEWS tool was approved for use on the cardiac unit (Figure 3) and the Escalation of Care Algorithm (Figure 4) conformed with existing critical response structures within the Cardiovas- cular Program.
The Escalation of Care Algorithm is an escalation of resources to a patient’s bedside to assess and treat deterioration based upon the C-CHEWS score. A C- CHEWS score of 0–2 (color code: green) recommends for clinicians to continue routine care, monitoring and assess- ments. A C-CHEWS score of 3–4 (color code: yellow) instructs the patient’s nurse to notify the charge nurse and patient’s resident or nurse practitioner of the elevated score. These clinicians discuss as a team a treatment plan for the
patient, initiate the plan and increase the frequency of patient assessments. The unit’s charge nurse may consider assigning the patient a higher dependency bed status, which would place the patient in a 1:2 patient assignment rather than the typical 1:3, thereby making it more feasible for the patient’s nurse to increase assessments and monitoring of the patient and implement the recommended modifications to the treatment plan. If a patient scores a 5 or greater (color code: red), the same steps are followed as described for color code yellow (score 3–4) with the addition of notifying the patient’s attending physician of their patient’s elevated C- CHEWS score. The patient’s resident or nurse practitioner must also examine the patient when they are notified of the high C-CHEWS score. As a team the clinicians determine if a CICU evaluation should be activated to assess for CICU transfer. It is not uncommon for patient’s C-CHEWS scores to remain elevated while the treatment plan is initiated. Should the score increase during the treatment phase, the algorithm is activated again.
Implementation of the C-CHEWS Tool
All of the cardiovascular staff were required to complete a short, computer-based learning module using three case studies. In addition, this educational material was reinforced during staff meetings and the unit’s monthly newsletter. The C-CHEWS tool and companion Escalation of Care Algorithm were posted throughout the unit for reference before and during implementation. The education initiative occurred over a 2-month period and all clinical staff knew the date designated for the tool to “go-live” and become standard of care.
The C-CHEWS was incorporated into the electronic health record in anticipation of the unit-based implementation. Nurses select from a drop-down list the numbered options that correspond to their patient’s assessment in each of the five C-CHEWS domains and the software calculates the score. Based on the score, the nurse follows the C-CHEWS Escalation of Care Algorithm and documents the actions taken. Documentation of the C-CHEWS score takes the nurse less than 10seconds. The benefit of incorporating the C-CHEWS score into the electronic record is that it allows for trending of scores in relation to patients’ vital signs. The additional assessments help to identify quickly patients that are trending towards deterioration. Patients’ C-CHEWS score and color code are displayed in real-time on the electronic unit census (Figure 5) which helps clinicians to immediately identify the patients who need closer attention.
Audits were done 1-month following the implementation of C-CHEWS documentation as standard of care. The audits were performed two to three times per week on 10 randomized patients’ charts for a total of 16 weeks to assess compliance with C-CHEWS documentation and utilization of the companion algorithm. Staff members were given feedback about documentation and any processes needing improvement. They also were asked if there were any system issues that could be improved. The entire process from the initial pilot to complete implementation totaled 6 months.
Unplanned CICU Transfers After C-CHEWS Implementation
The C-CHEWS tool has provided real-time trigger responses which have activated necessary resources to pediatric cardiovascular patients who are deteriorating on the inpatient cardiac unit. Chart review of patients who have had an unplanned transfer to the CICU or experienced an arrest on the cardiac unit typically had elevated C-CHEWS scores. The exception to this finding was patients who experienced sudden onset of compromising arrhythmia. This review suggests that the C-CHEWS may be effective in identifying the majority of patients in the cardiovascular pediatric population who are at risk for arrest or critical adverse events earlier to allow for intervention and prevent such events from occurring. In addition, the hospital has been tracking unplanned transfer events for several years as part of several quality improvement initiatives. In comparing the rate (transfers per 1000 patient days) of these events 1 year pre- and 1 year post- C-CHEWS implementation, there has been a reduction in unplanned transfers (Figure 6). The frequency has largely remained below the mean number of transfers per 1000 patients. This may suggest that patients are being treated earlier therefore preventing the necessity of a CICU transfer.


Nurses often verbalize that they “feel or sense something is not right” with their patient however the subtle differences in the patient’s presentation causing their unease may not be evident to the physicians as quantifiable changes are minimal and there is nothing obvious to treat (Andrews & Waterman, 2005). An early warning score can be an effective tool for nurses to use when communicating concern about subtle changes in the patient as the score provides a common language between nurses and physician colleagues (Andrews & Waterman, 2005). Early warning scoring tools provide an agreed upon framework (algorithm) for escalation of assessing and treating patients, which can empower nurses and interns to contact attending physicians more readily (Andrews & Waterman, 2005).
An objective scoring tool adjusts for familiarity with the patient and can heighten awareness of slow deterioration. The C-CHEWS score is calculated using the current vital signs and clinical assessment of the patient thus providing a real-time score to the clinician as to whether the patient may be deteriorating. The score does not rely on further information (i.e. calculated urine output) obtained away from the bedside to provide essential or necessary data. This fact helps to decrease any reliance on knowledge of the patient, patient’s history or level of clinician experience for interpreting additional data.
Pediatric cardiac patients are at higher risk for cardiopul- monary arrests than other hospitalized children. Arrest prevention will improve patient outcomes and survival for hospitalized pediatric cardiovascular patients. The C- CHEWS is an early warning scoring tool specific for this high-risk, vulnerable population and may assist clinicians in recognizing and treating these patients early and prevent arrests from occurring.
This is a single center experience of an acute pediatric cardiovascular care unit and may not be generalized to all pediatric cardiovascular units. Formal validation of the C- CHEWS tool, including sensitivity and specificity are needed. At the time of this writing, formal validity testing is proceeding within the institution for the C-CHEWS tool in both the inpatient cardiac and non-cardiac units with favorable preliminary data. The institution is also tracking whether there is a sustainable decrease in unplanned CICU transfers since the implementation of the C-CHEWS tool and companion Escalation of Care Algorithm on the cardiac unit.


The C-CHEWS is a tool that was specifically created for identifying pediatric cardiovascular patients at risk for deterioration, the tool previously used at our hospital was not effective for this patient population. The C-CHEWS tool and companion Escalation of Care Algorithm provides a standardized assessment and approach to deteriorating patients, ensuring that there is the appropriate dispersal of resources allocated to the acuity of the patients. Early activation of resources to at-risk patients’ bedsides provides early treatment of deterioration and may prevent cardiopul- monary arrests or unplanned CICU transfers.


We would like to thank Roger E. Breitbart MD, Jane C. Romano MS, RN, Monica Kleinman, MD and Suzanne Reidy MS, RN, NE-BC for participating in the expert multi- disciplinary panel.

Mary C. McLellan BSN, RN, CPNa,⁎, Jean A. Connor DNSc, RN, CPNPb aCardiovascular Program Inpatient Unit, Boston Children’s Hospital, Boston, MA
bCardiovascular and Critical Care Services, Boston Children’s Hospital, Boston, MA

Content for tab3

Akre, M., Finkelstein, M., Erickson, M., Liu, M., Vanderbilt, L., & Billman, G. (2010). Sensitivity of the Pediatric Early Warning Score to indentify patient detioration. Pediatrics, 125, 763–769.
Andrews, T., & Waterman, H. (2005). Packaging: A grounded theory of how to report physiological deterioration effectively. Journal of Advanced Nursing, 52, 473–481.
Berg, M. D., Nadkarni, V. M., Zuercher, M., & Berg, R. A. (2008). In- hospital pediatric cardiac arrest. Pediatric Clinics of North America, 55, 589–604.
Berwick, D. M., Calkins, D. R., McCannon, C. J., & Hackbarth, A. D. (2005). The 100,000 Lives Campaign. Setting a goal and a deadline for improving health care quality. Journal of the American Medical Association, 295, 324–327.
Brilli, R. J., Gibson, R., Luria, J. W., Wheeler, T. A., Shaw, J., Linam, M., & McBride, M. (2007). Implementation of a medical emergency team in a large pediatric teaching hospital prevents respiratory and cardiopulmo- nary arrests outside the intensive care unit. Pediatric Critical Care Medicine, 8, 236–246.
Chapman, S. M., Grocott, M. P., & Franck, L. S. (2010). Systematic review of paediatric alert criteria for identifying hospitalized children at risk for clinical deterioration. Intensive Care Medicine, 36, 600–611.
Chan, P. S., Jain, R., Nallmothu, B. K., Berg, R. A., & Sasson, C. (2010). Rapid response teams: A systematic review and meta-analysis. Archives of Internal Medicine, 170, 18–26.
DeVita, M. A., Bellomo, R., Hillman, K., Kellum, J., Rotondi, A., Teres, D., & Galhotra, S. (2006). Findings of the First Consensus Conference on Medical Emergency Teams. Critical Care Medicine, 34, 2463–2478.
Duncan, H. P. (2006). Survey of early identification systems to identify inpatient children at risk of physiological deterioration. Archives of Diseases of Childhood, 828.
Duncan, H., Hutchison, J., & Parshuram, C. S. (2006). The pediatric early warning system score: A severity of illness score to predict urgent medical need in hospitalized children. Journal of Critical Care, 21, 271–279.
Edwards, E. D., Powell, C. V. E., Mason, B. W., & Oliver, A. (2009). Prospective cohort study to test the predictability of the Cardiff and Vale paediatric early warning system. Archives of Disease in Childhood, 94, 602–606.
Gao, H., McDonnell, A., Harrison, D. A., Moore, T., Adam, S., Daly, K., & Harvey, S. (2007). Systematic review and evaluation of physiological track and trigger warning systems for identifying at-risk patients on the ward. Intens Care Medicine, 33, 667–679.
Haines, C., Perrott, M., & Weir, P. (2006). Promoting care for acutely ill children—Development of a Paediatric Early Warning Tool. Critical Care Nursing, 22, 73–81.
Hanson, C. C., Randolph, G. D., Erickson, J. A., Mayer, C. M., Bruckel, J. T., Harris, B. D., & Willis, T. S. (2009). A reduction in cardiac arrests and duration of clinical instability after implementation of a paediatric rapid response system. Quality & Safety in Health Care, 18, 500–504.
Hillman, K., Parr, M., Flabouris, A., Bishop, G., & Stewart, A. (2001). Redefining in-hospital resuscitation: The concept of the medical emergency team. Resuscitation, 48, 105–110.
Hunt, E. A., Zimmer, K. P., Rinke, M. L., Shilkofski, N. A., Matlin, C., Garger, C., & Miller, M. R. (2008). Transition from a traditional code team to a medical emergency team and categorization of cardiopulmo- nary arrests in a children’s center. Archives of Pediatric and Adolescent Medicine, 162, 117–122.
Kleinman, M & Romano, J. (2010). Children’s Hospital Boston Early Warning Score: early detection + early treatment = better outcomes. FIRST Do No Harm. Retrieved from http://www.mass.gov/eohhs/ docs/borim/newsletters/qps-august-2010.pdf.
Lopez-Herce, J., Garcia, C., Dominguez, P., Carrillo, A., Rodriguez-Nunez, A., Calvo, C., & Delgado, M. A.for the Spanish Group of Cardiopul- monary Arrest in Children. (2004). Characteristics and outcome of cardiopulmonary arrest in children. Resuscitation, 63, 311–320.
Meaney, P. A., Nadkarni, V. M., Cook, F., Testa, M., Helfaer, M., Kaye, W., & Berg, R. A.for the American Heart Association National Registry of Cardiopulmonary Resuscitation Investigators. (2006). Higher survival rates among younger patients after pediatric intensive care unit cardiac arrests. Pediatrics, 2424–2433.
Monaghan, A. (2005). Detecting and managing deterioration in children. Paediatric Nursing, 17, 32–35.
M.C. McLellan, J.A. Connor
Resuscitation Council); American Heart Association Emergency Cardiovascular Care Committee; the Council on Cardiopulmonary, Perioperative and Critical Care; and the Interdisciplinary Working Group on Quality of Care and Outcomes Research. Circulation, 116, 2481–2500.
Reis, A. G., Nadkarni, V. M., Perondi, M. B., Grisi, S., & Berg, R. A. (2002). A prospective investigation into the epidemiology of in-hospital pediatric cardiopulmonary resuscitation using the International Utstein reporting style. Pediatrics, 109, 200–209.
Rhodes, J. F., Blaufoux, A. D., Sieden, H. S., Asnes, J. D., Gross, R. P., Rhodes, J. P., & Rossi, A. F. (1999). Cardiac arrest in infants after congenital heart surgery. Circulation, 100(supl II), II194–II199.
Salamonson, Y., Kariyawasam, A., van Heere, B., & O’Connor, C. (2001). The evolutionary process of Medical Emergency Team (MET) implementation: Reduction in unanticipated ICU transfers. Resuscita- tion, 49, 135–141.
Samson, R. A., Berg, M. D., & Berg, R. A. (2006). Cardiopulmonary resuscitation algorithms, defibrillation and optimized ventilation during resuscitation. Current Opinion in Anaesthesiology, 19, 146–156.
Samson, R. A., Nadkarni, V. M., Meaney, P. A., Carey, S. M., Berg, M. D., & Berg, R. A.for the American Heart Association National Registry of CPR Investigators. (2006). Outcomes of in-hospital ventricular fibrillation in children. The New England Journal of Medicine, 22, 2328–2339.
Sharek, P. J., Parast, L. M., Leong, K., Coombs, J., Earnest, K., Sullivan, J., & Roth, S. J. (2007). Effect of a rapid response team on hospital-wide mortality and code rates outside the ICU in a children’s hospital. Journal of the American Medical Association, 298, 2267–2274.
Schein, R. M., Hazday, N., Pena, M., Ruben, B. H., & Sprung, C. L. (1990). Clinical antecedents to in-hospital cardiopulmonary arrest. Chest, 98, 1388–1392.
Slonim, A. D., Patel, K. M., Ruttimann, U. E., & Pollack, M. M. (1997). Cardiopulmonary resuscitation in pediatric intensive care units. Critical Care Medicine, 25, 1951–1955.
Suominen, P., Olkkola, K. T., Voipio, V., Korpela, R., Palo, R., & Rasanen, J. (2000). Utstein style reporting of in-hospital pediatric cardiopulmo- nary resuscitation. Resuscitation, 45, 17–25.
Tibballs, J. (2006). Evaluation of a paediatric early warning tool—Claims unsubstantiated. Intensive & Critical Care Nursing, 22, 315–316.
Tibballs, J., & Kinney, S. (2009). Reduction of hospital mortality and of preventable cardiac arrest and death on introduction of a pediatric medical emergency team. Pediatric Critical Care Med, 10, 306–312.
Tibballs, J., Kinney, S., Duke, T., Oakley, E., & Hennessy, M. (2005). Reduction of paediatric in-patient cardiac arrest and death with a medical emergency team: Preliminary results. Archives of Disease in Childhood, 90, 1148–1152.
Tucker, K. M., Brewer, T. L., Baker, R. B., Demeritt, B., & Vossmeyer, M. T. (2008). Prospective evaluation of a pediatric inpatient early warning scoring system. Journal for Specialists in Pediatric Nursing, 14, 79–85.
Tume, L. (2007). The deterioration of children in ward areas in a specialist children’s hospital. Nursing in Critical Care, 12, 12–19.
ul-Haque, A., Saleem, A. F., Zaidi, S., & Haider, S. R. (2010). Experience of pediatric rapid response team in a tertiary care hospital in Pakistan. Indian Journal of Pediatrics, 77, 273–276.
VandenBerg, S. D., Hutchison, J. S., & Parshuram, C. S. (2007). A cross- sectional survey of levels of care and response mechanisms for evolving critical illness in hospitalized children. Pediatrics, 119, 940–946.
VanVoorhis, K. T., & Willis, T. S. (2009). Implementing a pediatric rapid response system to improve quality and patient safety. Pediatric Clinics of North America, 56, 919–933.
Young, K. D., & Seidel, J. S. (1999). Pediatric cardiopulmonary resuscitation: A collective review. Annals of Emergency Medicine, 33, 195–205.
Zenker, P., Schlesinger, A., Hauck, M., Spencer, S., Hellmich, T., Finkelstein, M., & Billman, B. (2007). Implementation and impact of a rapid response team in a children’s hospital. Joint Commission Journal on Quality and Patient Safety, 33, 418–425.
Nadkarni, V. M., Larkin, G. L., Peberdy, M. A., Carey, S. M., Kaye, Mancini, M. E., & Berg, R. A. (2006). First documented rhythm clinical outcomes from in-hospital cardiac arrest among children adults. Journal of the American Medical Association, 295, 50–57.
W., and and
Parra, D. A., Totapally, B. R., Zahn, E., Jacobs, J., Aldousany, A., Burke, R. P., & Chang, A. C. (2000). Outcome of cardiopulmonary resuscitation in a pediatric cardiac intensive care unit. Critical Care Medicine, 28, 3296–3300.
Parshuram, C. S., Hutchinson, J., & Middaugh, K. (2009). Development and initial validation of the Bedside Paediatric Early Warning System score. Critical Care, 13, R135, http://dx.doi.org/10.1186/cc7998.
Pearson, G. A., Ward-Platt, M., Harnden, A., & Kelly, D. (2010). Why children die: Avoidable factors associated with child deaths. Archives of Disease in Childhood, http://dx.doi.org/10.1136/adc.2009.177071.
Peberdy, M. A., Cretikos, M., Abella, B. S., DeVita, M., Goldhill, D., Kloeck, W., & Young, L. (2007). Recommended guidelines for monitoring, reporting and conducting research on medical emergency team, outreach and rapid response systems: An Utstein-Style scientific statement: A scientific statement from the International Liaison Committee on Resuscitation (American Heart Association, Australian Resuscitation Council, European Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of South Africa, and the New Zealand

Content for tab5

Frequency, Risk Factors, and Outcomes of Early Unplanned Readmissions to PICUs

Abstract Objectives—To determine the rate of unplanned PICU readmissions, examine the characteristics of index admissions associated with readmission, and compare outcomes of readmissions versus index admissions. Design—Retrospective cohort analysis. Setting—Ninety North American PICUs that participated in the Virtual Pediatric Intensive Continue reading Frequency, Risk Factors, and Outcomes of Early Unplanned Readmissions to PICUs