Clinical Profile of Children Requiring Early Unplanned Admission to the PICU

RESEARCH ARTICLE Clinical Profile of Children Requiring Early Unplanned Admission to the PICU abstract OBJECTIVE: The goal of this study was to describe the frequency, characteristics, and outcomes of children who require early unplanned admission to the PICU within 24 hours of hospitalization from the emergency department. METHODS: This study was a retrospective audit of 24 months of prospectively collected medical emergency team records at a tertiary pediatric hospital in Canada. Our review identified 39 hospitalized children who had an activation that resulted in unplanned admission to the PICU within 24 hours of admission from the ED. RESULTS: Forty-six percent of the study subjects were infants aged <1 year, and 64% were male. Respiratory complaints were the most common reason for hospitalization (59%). Preexisting medical conditions (51%), abnormal respiratory rates (46%), abnormal heart rates (33%), abnormal blood gas values (49%), high supplemental oxygen requirement (23%), and treatment with nebulized medications (46%), intravenous fluids (33%), and antibiotics (33%) were common. The median time to medical emergency team activation was 9.4 hours (interquartile range: 4.4 14.5). Nearly one-half (49%) of the patients required a significant intervention after admission to the PICU, with a mean length of stay of 3.4 days and a mortality rate of 5%. CONCLUSIONS: Male subjects, infants aged <1 year, and children with respiratory complaints accounted for a large proportion of children requiring early unplanned admission to the PICU within 24 hours of hospitalization from the ED. Further studies are required to determine which factors are associated with deterioration after hospitalization. INTRODUCTION The identification and management of evolving critical illness are paramount to patient safety. Hospitalized children admitted from the emergency department (ED) may initially appear well enough to be admitted to the general inpatient ward but later experience significant clinical deterioration and subsequently require unplanned admission to the PICU. Children admitted to the PICU from hospital wards have a 1.65 times higher risk of mortality than children admitted directly to the PICU from the ED. 1 AUTHORS Kristina Krmpotic, MD, FRCPC, and Anna-Theresa Lobos, MD, FRCPC Division of Pediatric Critical Care Medicine, Department of Pediatrics, Children s Hospital of Eastern Ontario, Ottawa, Ontario, Canada KEY WORDS critical care, critical illness, early warning, intensive care, medical emergency team, pediatrics, rapid response team ABBREVIATIONS CHEO: Children s Hospital of Eastern Ontario ED: emergency department IQR: interquartile range MET: medical emergency team RRS: rapid response system www.hospitalpediatrics.org doi:10.1542/hpeds.2012-0081 Address correspondence to Kristina Krmpotic, MD, FRCPC, Division of Pediatric Critical Care Medicine, Department of Pediatrics, Children s Hospital of Eastern Ontario, 401 Smyth Rd, Ottawa, Ontario, K1H 8L1, Canada. E-mail: kkrmpotic@cheo.on.ca HOSPITAL PEDIATRICS (ISSN Numbers: Print, 2154-1663; Online, 2154-1671). Copyright 2013 by the American Academy of Pediatrics FINANCIAL DISCLOSURE: The authors have indicated they have no fi nancial relationships relevant to this article to disclose. FUNDING: Funding for this study was provided by the Children s Hospital of Eastern Ontario Research Institute. Implementation of rapid response teams is 1 strategy recommended to improve patient outcomes through the early identification and management of ward patients whose condition is deteriorating. 2 Initially intended to decrease ICU admission rates and severity of illness at ICU admission in resource-limited 212 VOLUME 3 ISSUE 3 www.hospitalpediatrics.org

adult hospitals, introduction of a rapid response system (RRS) by using medical emergency teams (MET) in pediatric hospitals may be associated with improved outcomes for patients whose condition worsens while on the hospital ward. 3 6 A recent multicenter study investigating the use of a pediatric RRS found a 20% reduction in mortality for children requiring unplanned admission to the PICU after implementation of a MET. 3 Many studies have also attempted to provide clinicians with scoring systems to help identify high-risk hospitalized children who are more likely to require unplanned admission to the PICU. 7 11 Although 1 recently published predictive score used nonvital sign patient characteristics to identify children at high risk, 7 the majority of these scores were based on changes in vital signs. 8 11 Given that a significant number of pediatric MET activations occur within 24 hours of hospital admission, 12 it may be possible to identify factors present at the time of hospital admission from the ED that are associated with clinical deterioration in pediatric inpatients. The purpose of the current study was to determine the frequency at which unplanned admission to the PICU occurs within the first 24 hours of hospitalization from the ED, and to describe the clinical characteristics and outcomes for hospitalized children who required early unplanned admission to the PICU. METHODS Study Setting This study was conducted at the Children s Hospital of Eastern Ontario (CHEO) in Ottawa, Canada. CHEO, a freestanding, tertiary pediatric hospital affiliated with the University of Ottawa, has 166 beds, including 10 ICU beds. The hospital admits >6000 patients each year. Medical Emergency Team The MET at CHEO is composed of a PICU physician, a critical care nurse, and a respiratory therapist, available 24 hours per day, 7 days per week. The MET can be activated by any health care provider on the hospital ward (eg, resident, nurse, respiratory therapist) who has concerns regarding the clinical status of a patient, by using the previously published, age-specific, physiologic criteria for MET activation identified on posters throughout the hospital (Table 1). All MET members are simultaneously notified of the activation (through dedicated pagers) to coordinate a response at the bedside within 5 to 10 minutes of activation. The patient is assessed by the MET, and if necessary, the patient is transferred from the inpatient ward to the PICU, referred to as an unplanned admission. Since the implementation of the MET at CHEO in January 2007, the only way that pediatric inpatients can be admitted to the PICU directly from the inpatient ward is via MET activation. The MET has entirely replaced PICU consultations for inpatients. The MET at CHEO has >200 activations per year, with 30% resulting in an unplanned admission to the PICU. 3 Methods of Data Collection After approval by the CHEO Research Ethics Board, an audit of prospectively collected MET records was conducted TABLE 1 Physiologic Criteria for MET Activation Any 1 or more of: 1 Nurse or physician worried about clinical state 2 Airway threat 3 Hypoxemia: Spo 2 <90% in any amount of oxygen Spo 2 <60% in any amount of oxygen (cyanotic heart disease) 4 Severe respiratory distress, apnea, or cyanosis 5 Tachypnea Age Respiratory Rate (breaths/min) Term 3 mo >60 4 12 mo >50 1 4 y >40 5 12 y >30 12 y >30 6 Bradycardia or tachycardia Age Bradycardia (beats/min) Tachycardia (beats/min) Term 3 mo <100 >180 4 12 mo <100 >180 1 4 y <90 >160 5 12 y <80 >140 12 y <60 >130 7 Hypotension Age SBP (mm Hg) Term 3 mo <50 4 12 mo <60 1 4 y <70 5 12 y <80 12 y <90 8 Acute change in neurologic status or convulsion 9 Cardiac or respiratory arrest SBP, systolic blood pressure; Spo 2, pulse oxygen saturation. Adapted from Tibballs and Kinney. 6 213

to identify a case series of pediatric inpatients who had a MET activation that resulted in transfer to the PICU within 24 hours of hospitalization during a 24-month period (February 2009 January 2011). A time duration of 24 hours from admission to hospital was selected based on the literature, which suggests that changes in these physiologic parameters may be present up to 24 hours before clinical deterioration of acutely ill pediatric patients. 8 11 Children who were not admitted via the ED were excluded, as were children who received a general anesthetic between the time of their assessment in the ED and the MET activation, given that an anesthetic could significantly affect their hemodynamic parameters. The hospital charts of eligible children were retrospectively reviewed by 2 trained research assistants to collect information, including patient demographic characteristics, medical history, chief complaint on presentation, vital signs, and results of initial laboratory investigations in the ED, interventions in the ED, admission and discharge diagnoses, time to MET activation, interventions in the PICU, and PICU length of stay. Data Analysis Preexisting medical conditions, presenting complaint, and admission and discharge diagnoses were coded into the following categories: respiratory, cardiac, gastrointestinal, genitourinary, neurologic/neuromuscular, endocrine, hematologic/oncologic, and infectious. Vital signs documented at triage and before transfer to the ward, as well as extremes of vital signs (eg, highest heart rate, highest respiratory rate) documented at any time in the ED, were coded as normal, high, or low, compared with the same age-based norms cited in the MET activation criteria (Table 1). Laboratory values were coded as normal, high, or low, compared with age-based norms used by the CHEO hematology and biochemistry laboratories. All data were entered into an Ac cess database (Microsoft Corporation, Red mond, WA) and analyzed by a statistician to provide descriptive statistics, and were summarized as percentages, with median and interquartile ranges (IQRs) provided for continuous variables. RESULTS MET Activity and Study Eligibility During the 2-year study period, 8193 children were admitted from the ED, and there were 413 MET activations. Of these, 41 children met inclusion criteria for the study, having had a MET activation that resulted in transfer to the PICU within 24 hours of hospitalization. Two patients were excluded because they received general anesthesia between the time they were assessed in the ED and the time of the initial MET activation. Therefore, 39 patients were included in the analysis. During the study period, 4 children not admitted via the ED had a MET activation that resulted in transfer to the PICU within 24 hours of hospitalization and were not included in the analysis. Patient Demographic Characteristics Of the 39 children included in the analysis, 64% were male (n = 25) (Table 2). The median age at admission for the study group was 14 months (IQR: 3 71), with 46% of calls being for infants aged <1 year. Respiratory diagnoses (ie, bronchiolitis, asthma, pneumonia, croup) were among the most common reasons for hospital admission in these study patients (59%), TABLE 2 Patient Characteristics Characteristic Study Sample (N = 39) Gender Male 25 (64) Female 14 (36) Age, mo 0 12 18 (46) 13 24 5 (13) 25 36 0 (0) 37 48 2 (5) 49 60 1 (3) 60 13 (33) Reason for hospital admission Respiratory 23 (59) Infectious 6 (15) Neurologic 2 (5) Hematologic 2 (5) Gastrointestinal 2 (5) Cardiac 1 (3) Other 3 (8) Data are presented as n (%). followed by infection (eg, rule-out sepsis, fever) in 15% of patients. Clinical Characteristics Many of the children in the study group (51%) had 1 preexisting medical condition, most commonly respiratory disease (26%). Other preexisting medical conditions included neurologic/ neuromuscular (15%), gastrointestinal (13%), genitourinary (8%), cardiac (5%), hematologic (3%), or endocrine (3%) conditions. While in the ED, 56% of the children fulfilled at least 1 of the MET activation criteria (46% for tachypnea and 33% for tachycardia). No children fulfilled the activation criteria for hypotension. However, >30% of patients had no documented blood pressure measurement while in the ED. Laboratory derangements (Fig 1) included abnormal blood gas values (49%), glucose (33%), electrolytes (33%), lactate (20%), white blood cell count (23%), and hemoglobin (13%). ED Interventions Interventions initiated in the ED are shown in Fig 2. Administration of 214 VOLUME 3 ISSUE 3 www.hospitalpediatrics.org

and 2 required intensive monitoring during initiation of antiarrhythmic medications. The remaining 28% of the study population was admitted for monitoring but did not require any of the aforementioned interventions. The mean length of stay in the PICU was 3.4 days (range: 1 9 days), with a mortality rate of 5%. Both of the study patients who died had significant preexisting medical conditions. FIGURE 1 Frequency of abnormal laboratory values in the ED. supplemental oxygen >6 L/min or frac tion of inspired oxygen of 40% to main tain oxygen saturation measures >92% was required by 23% of patients. Nebulized medications, including hypertonic saline, salbutamol, and epinephrine, were administered to 46% of patients. A fluid bolus (ie, isotonic saline in an amount of 5 20 ml/kg) was administered to 33% of patients. One or more intravenous antibiotics were administered to 33% of patients. The median time from triage to antibiotic administration was 3 hours 20 minutes (IQR: 2 6). PICU was consulted for only 13% of the study patients. Outcomes As demonstrated in Fig 3, median time from arrival on the inpatient ward to MET activation was 9.4 hours (IQR: 4.4 14.5). Time lapse between MET activation and transfer to the PICU ranged from 10 minutes to 9 hours 40 minutes, with a median time of 66 minutes (IQR: 45 158). Data regarding frequency of interventions that required admission to the PICU at our institution are shown in Fig 4. These interventions included administration of continuous nebulized medications for 18% of patients; noninvasive positive pressure ventilation, including continuous positive airway pressure and bilevel positive airway pressure, for 36% of patients; endotracheal intubation for 10% of patients; and mechanical ventilation for 1 additional patient with tracheostomy. One patient required inotropic support, DISCUSSION During the 2-year study period at CHEO, 3% of all hospitalized children required assistance from the MET (32 activations/1000 admissions). In the pediatric literature, the use of METs vary and range from 2.8 to 44 activations per 1000 hospital admissions. 4,5,13 16 The rates of unplanned admission to the PICU via MET activations range from 30% to 57%. 3 5,12 14 Whereas most of the RRS literature does not comment on timing of MET activation in relation to hospital admission, Bonafide et al 12 recently reported that nearly 40% of the 780 MET activations in their study occurred within 24 hours of hospital admission. Furthermore, 41% of those patients seen by the MET within 24 hours of hospital admission required an unplanned PICU admission (C.P. Bonafide, MD, personal communication, 2012). The results of our study revealed that nearly 10% of all MET activations were both within 24 hours of hospitalization from the ED and resulted in an unplanned admission to the PICU, with >70% of these children requiring a significant intervention. Given the variability in duration between admission to the inpatient ward and the MET activation, it is likely that our study population consisted of 2 groups: patients who were mistakenly triaged 215

FIGURE 2 Frequency of interventions initiated in the ED. and arrived on the ward critically ill, and patients who rapidly deteriorated after admission to the ward. The small patient population included in this study prohibits further exploration of characteristics that may distinguish these groups. Prospective study by using electronic documentation of vital signs would allow for the collection of multiple data points for more detailed analysis. Male patients and infants aged <1 year, as well as children with preexisting FIGURE 3 Kaplan-Meier curve demonstrating time to MET activation. medical conditions (particularly respiratory disease), were prevalent in our study population. These findings are similar to the data published by Kinney et al, 13 who reviewed the charts of 172 pediatric inpatients requiring MET activation and found that infants aged <1 year were overrepresented (43%) compared with their baseline rate of hospital admission (14%). Kinney et al also showed that 49% of children who required a MET activation had 2 preexisting chronic diseases. Similarly, Bonafide et al 7 also used age <1 year, diagnosis of epilepsy or congenital/ genetic conditions, and history of transplant in their score to estimate probability of clinical deterioration after hospital admission. Kinney et al 13 also found that the most frequent reason for MET activation was respiratory compromise. Similarly, a large majority of the children in our study presented to the ED and were admitted with signs and symptoms of respiratory disease. Future studies comparing our study population with a control group of children who did not require early unplanned admission to the PICU are required to determine if factors such as gender, age, and respiratory conditions are associated with clinical deterioration after transfer from the ED to the hospital ward. In addition to these clinical characteristics, preexisting chronic medical conditions, ageadjusted abnormalities in heart rate, respiratory rate, and laboratory values (particularly blood gas values), as well as need for interventions including nebulized medications, intravenous fluid, antibiotics, and high amounts of supplemental oxygen, may be potential indicators of increased risk of clinical deterioration after hospitalization. 216 VOLUME 3 ISSUE 3 www.hospitalpediatrics.org

ACKNOWLEDGMENTS We thank Dr Amy Plint and Dr Franco Momoli for their collaboration on this project, Christa Ramsay for maintenance of CHEO MET records, Geoff MacDonald and Kristine De Jesus for completing the chart reviews, and Elham Sabri at Ottawa Health Research Institute for statistical support. REFERENCES FIGURE 4 Frequency of interventions initiated in the PICU. There are several limitations to the current study. This was a retrospective chart review limited by the ability to extract patient information from medical records. It is possible that some information, particularly regarding the presence of a preexisting chronic medical condition, might not be available from these records. In addition, the clinical significance of these conditions may be wideranging, depending on the severity and duration of disease. Another limitation is our small sample size. This study received limited funding, which restricted our sample size to eligible cases over a 2-year period at a single institution. Thus, no comparison group was examined to determine if children who do not require unplanned admission to the PICU within 24 hours of hospitalization also have similar clinical characteristics or if these are significant early indicators of clinical deterioration shortly after admission. The results of this study also may not be generalizable to other institutions. Finally, because of the low mortality rate in pediatric medicine, it is a relatively poor outcome measure. Although the current study also included need for life-sustaining therapies (eg, noninvasive positive pressure ventilation, endotracheal intubation, vasopressor infusions), and length of stay, these factors may not be the best indicators of morbidity in this patient population. CONCLUSIONS The findings of the current study suggest that it may be possible to identify children who are at increased risk of early unplanned admission to the PICU based on clinical characteristics noted in the ED at the time of their hospitalization. A well-funded, multicenter study would allow for a larger sample of cases and the inclusion of a comparison cohort to identify factors associated with early unplanned admission to PICU. 1. Odetola FO, Rosenberg AL, Davis MM, Clark SJ, Dechert RE, Shanley TP. Do outcomes vary according to the source of admission to the pediatric intensive care unit? Pediatr Crit Care Med. 2008;9(1):20 25. 2. Berwick DM, Calkins DR, McCannon CJ, Hackbarth AD. The 100,000 lives campaign: setting a goal and a deadline for improving health care quality. JAMA. 2006;295(3):324 327. 3. Kotsakis A, Lobos AT, Parshuram C, et al; Ontario Pediatric Critical Care Response Team Collaborative. Implementation of a multicenter rapid response system in pediatric academic hospitals is effective. Pediatrics. 2011;128(1):72 78. 4. Sharek PJ, Parast LM, Leong K, et al. Effect of a rapid response team on hospital-wide mortality and code rates outside the ICU in a Children s Hospital. JAMA. 2007; 298(19):2267 2274. 5. Brilli RJ, Gibson R, Luria JW, et al. Implementation of a medical emergency team in a large pediatric teaching hospital prevents respiratory and cardiopulmonary arrests outside the intensive care unit. Pediatr Crit Care Med. 2007;8(3):236 246, quiz 247. 6. Tibballs J, Kinney S. Reduction of hospital mortality and of preventable cardiac arrest and death on introduction of a pediatric medical emergency team. Pediatr Crit Care Med. 2009;10(3):306 312. 7. Bonafide CP, Holmes JH, Nadkarni VM, Lin R, Landis JR, Keren R. Development of a score to predict clinical deterioration in hospitalized children. J Hosp Med. 2012;7(4):345 349. 8. Akre M, Finkelstein M, Erickson M, Liu M, Vanderbilt L, Billman G. Sensitivity of the pediatric early warning score to identify patient deterioration. Pediatrics. 2010;125(4). Available at: www.pediatrics.org/cgi/content/ full/126/4/e763. 217

9. Edwards ED, Powell CV, Mason BW, Oliver A. Prospective cohort study to test the predictability of the Cardiff and Vale paediatric early warning system. Arch Dis Child. 2009;94(8):602 606. 10. Parshuram CS, Hutchison J, Middaugh K. Development and initial validation of the Bedside Paediatric Early Warning System score. Crit Care. 2009;13(4):R135. 11. Duncan H, Hutchison J, Parshuram CS. The Pediatric Early Warning System score: a severity of illness score to predict urgent medical need in hospitalized children. J Crit Care. 2006;21(3):271 278. 12. Bonafide CP, Roberts KE, Priestley MA, et al. Development of a pragmatic measure for evaluating and optimizing rapid response systems. Pediatrics. 2012;129(4). Available at: www.pediatrics.org/cgi/content/full/129/ 4/e874. 13. Kinney S, Tibballs J, Johnston L, Duke T. Clinical profile of hospitalized children provided with urgent assistance from a medical emergency team. Pediatrics. 2008; 121(6). Available at: www.pediatrics.org/ cgi/content/full/121/6/e1577. 14. Tibballs J, Kinney S, Duke T, Oakley E, Hennessy M. Reduction of paediatric in-patient cardiac arrest and death with a medi cal emergency team: preliminary results. Arch Dis Child. 2005;90(11):1148 1152. 15. Zenker P, Schlesinger A, Hauck M, et al. Implementation and impact of a rapid response team in a children s hospital. Jt Comm J Qual Patient Saf. 2007;33(7): 418 425. 16. Hunt EA, Zimmer KP, Rinke ML, et al. Transition from a traditional code team to a medical emergency team and categorization of cardiopulmonary arrests in a children s center. Arch Pediatr Adolesc Med. 2008;162(2):117 122. 218 VOLUME 3 ISSUE 3 www.hospitalpediatrics.org