Emergency Medicine I
Effectiveness of a Mentor-Implemented Violence Prevention Intervention for Assault-Injured Youth Presenting to the Emergency Department: Results of a Randomized Trial
Tina L. Cheng, Denise Haynie, Ruth A. Brenner, Shang-en Chung, Joseph L. Wright, Bruce Simons-Morton. Pediatrics, Johns Hopkins University, Baltimore, MD; National Institute of Child Health Human Development, Bethesda, MD; George Washington University, Children National Medical Center, Washington, DC.
BACKGROUND: The emergency department (ED) has been described as an ideal setting to initiate interventions with assault-injured youth to reduce the risk of re-injury and reactive perpetration. Mentoring and community-based strategies are considered best practices in violence prevention. Intervention with assault-injured youth has received little study.
OBJECTIVE: To assess the impact of a mentor-implemented violence prevention intervention on reducing aggression, fighting and re-injury among assault-injured youth.
DESIGN/METHODS: Youth age 10-15 presenting to two large urban hospital EDs with peer assault injury were recruited and randomly assigned to intervention and control groups. Intervention youth received a mentor who implemented a 6 session problem-solving curriculum while parents received 3 home visits with an educator to discuss family needs, parental monitoring and facilitate service use. Controls received a list of community resources with 2 follow-up phone calls to facilitate service use. Youth and parents were interviewed at baseline and 6 months to assess attitudes about violence, aggression, fighting and repeat injury. Negative binomial modeling was conducted.
RESULTS: 227 families were recruited with 23% refusing and 4% partial interviews. 166 families were enrolled; 116 (70%) completed both youth and parent follow-up interviews. Intervention and control groups were not significantly different at baseline on demographics, risk factors or risk behavior except for increased knife carrying and decreased deviant peers in the intervention group. Fighting was common in both groups. After adjustment for baseline differences, there was significant program effect reducing youth-reported aggression scores (rate ratio .6, p<.05) and increasing youth self efficacy (rate ratio 1.06, p<.05). Program impact increased with intervention dose.
CONCLUSIONS: A community-based mentor-implemented program with assault-injured youth reduced youth aggression. The ED may be an important contact location, and injuries an important context, for augmenting self efficacy for violence prevention.
Relationship of Serum S100B and Intracranial Injury in Children with Accidental Closed Head Trauma
Kirsten Bechtel, Sarah Frasure, James Dziura, Christine Simpson. Yale University School of Medicine, New Haven, CT.
BACKGROUND: Symptomatic children with closed head trauma (CHT) often undergo computed tomography (CT) to exclude intracranial injury (ICI). S100B, a time-dependent protein made by CNS glial cells and chondrocytes, has been shown to be elevated in the serum of patients with brain injury and in patients with fractures.
OBJECTIVE: To determine if serum S100B levels are higher in children with ICI and can determine which children will have ICI after CHT; and the effect of fractures and time of venipuncture after injury on S100B levels.
DESIGN/METHODS: Children < 18 years old who initially presented to an urban Pediatric Emergency Department, or were transferred from a referral hospital, within 6 hours after accidental CHT and had cranial CT, were prospectively enrolled. After informed consent, samples were obtained by venipuncture and analyzed for a quantitative serum level of S100B. The primary outcome, mean serum level of S100B between children with ICI and without ICI, and the secondary outcomes, fractures and the time of venipuncture, on the level of S100B, were analyzed by analysis of covariance. Discriminatory value of S100B to detect ICI was evaluated by the area under the ROC curve (AUC).
RESULTS: 152 children were prospectively enrolled-24 with ICI and 128 children without ICI. There were no significant differences between the 2 groups with respect to age, gender or ethnicity. Injury mechanisms were predominantly falls (42%), pedestrian struck by a vehicle (23%), motor vehicle crashes (13%), and sports collisions (9%). 25 children had fractures. Time of venipuncture after injury was significantly later in children with ICI (p=.03). Mean S100B levels were significantly greater in those children with ICI (212.9 ng/L vs. 84.4 ng/L p= .0001), in children with fractures (p=.0008) and in children who were non-white. (p=.03). After controlling for time of venipuncture, fractures and race, mean S100B levels were still greater in children with ICI (409 ng/L vs. 118 ng/L p= .0001). Discriminatory value of S100B to detect ICI (AUC) was .67.
CONCLUSIONS: After controlling for time of venipuncture, fractures and race, S100B levels were still higher in children with ICI than those without. However, discriminatory value of S100B to detect ICI was only .67. Further study of S100B is needed to determine if it is a helpful adjunct to evaluate children for ICI after CHT.
Blunt Head Trauma in the Pediatric Emergency Care Applied Research Network (PECARN)
N. Kuppermann, J.F. Holmes, P.S. Dayan, J. Hoyle, S.M. Atabaki, J.M. Dean, R. Holubkov, PECARN. Univ of Calif, Davis, CA; M.S. Childrens Hosp of NY, NY; DeVos Childrens Hosp, Grand Rapids, MI; Childrens National Med Ctr, Washington, DC; Univ of Utah, Salt Lake City, UT.
BACKGROUND: Blunt head trauma is common in children evaluated in emergency departments (Eds). Computed tomography (CT) is frequently but variably used, and routine CT use has drawbacks.
OBJECTIVE: To describe the spectrum of head trauma in a large pediatric emergency medicine network. For children with minor head trauma, we also sought to derive a decision rule to identify patients at very low and high risk for traumatic brain injuries (TBI).
DESIGN/METHODS: We prospectively enrolled children <18 years old with blunt head trauma at 25 EDs in PECARN from 6/04 to 5/06. Patients with trivial mechanisms were excluded. Head CTs were obtained at the clinicians discretion. A standardized clinical assessment was recorded onto a case report form prior to CT results. Patients discharged home had clinical status determined by phone call, and medical records of hospitalized patients were reviewed. In patients with minor head trauma (i.e. Glasgow Coma Scale [GCS] score 14-15), clinical variables with a priori likelihood of association with TBI and with at least moderate interobserver agreement are being analyzed to develop a decision rule to identify children at very low and high risk of TBI. Site monitoring and an independent data center insured data quality.
RESULTS: Of 35,140 children enrolled, 62% were boys, and mean age was 6.6 years (SD 5.6). Common mechanisms of injury included falls (48.1%), occupant in motor vehicle crashes (9.1%), ran into stationary objects (8.5%), object struck head (8.4%), sports (7.3%), assaults (7.3%), pedestrians (3.3%) or bicyclists (1.2%) struck by automobiles, and bicycle crashes (3.5%). 96.8% had GCS of 14 (3.1%) or 15 (93.7%). CTs were obtained in 36.8% (site-specific range: 9.7% - 71.1%), and were positive in 11.2% (30% of these were isolated skull fractures). Neurosurgery was performed on 0.5%, and 0.1% died from TBI. For patients with GCS 14-15, a decision rule for TBI will be reported at the meeting.
CONCLUSIONS: Blunt head trauma is common in PECARN EDs. Patients are more commonly male, and falls are the dominant mechanism. Most patients have minor head trauma, and CT use is highly variable. A decision rule to identify patients at very low and high risk of TBI may result in more efficient CT use.
Implementation of an Ambulatory Medication Management Application in a Pediatric Emergency Department
Usha M. Sankrithi, Diana Wong, Michael DeBusk. Division of Emergency Medicine, Childrens Hospital Regional Medical Center, Seattle, WA; Department of Information Technology, Pharmacy Services, Childrens Hospital of Central California, Madera, CA.
BACKGROUND: In recent years, medication-related mishaps have received significant attention in the literature and from the legislative and regulatory agencies. Childrens Hospital Central California implemented an electronic medication management application in its emergency department to improve patient safety and increase efficiencies in the medication management process.
OBJECTIVE: A commercially-available, electronic medication management application was implemented. Key components of this application include (1) capturing and maintaining patients medication profile, (2) managing patients allergy information, (3) generating computer prescriber order entry (CPOE) prescriptions, (4) providing real-time clinical decision support tools to the providers and (5) generating electronic discharge instructions.
DESIGN/METHODS: The numbers of patients and prescriptions pre-CPOE and post-CPOE were analyzed. The error rate (based on predefined criteria) for written prescriptions pre-CPOE was compared with those prescriptions electronically generated post-CPOE. The quality of the patients medication profile post-CPOE was also audited.
RESULTS: Compared to pre-CPOE, the error rate of the prescriptions post-CPOE was reduced by 84% (5.7% pre-CPOE to 0.9% post-CPOE). After implementation, the average number of patients who received at least one prescription increased by 130 % (30 patient per day pre-CPOE to 69 patients per day post-CPOE). The average number of prescriptions generated increased by 173% (45 per day pre-CPOE to 123 per day post-CPOE). The average number of patients seen in the emergency department increased by 29% (116 patients per day pre-CPOE to 150 patients per day post-CPOE). An audit of the medication profiles revealed that 8% of these contained redundant information.
CONCLUSIONS: Implementation of the electronic medication management system enabled the achievement of a centralized patient medication profile within our integrated health record. It reduced errors in prescription writing and resulted in a more complete, although sometimes redundant, medication record. The medication profile information, including allergies and weights, became instantly available to other clinicians at the healthcare facility.
The Impact of Family Presence at Pediatric Trauma Resuscitations
Nanette C. Dudley, Kristine W. Hansen, Ronald A. Furnival, Kaye Lynn Van Wagenen, Eric R. Scaife. Pediatrics, University of Utah School of Medicine, Salt Lake City, UT; Trauma Surgery, Primary Childrens Medical Center, Salt Lake City, UT; Pediatrics, University of Minnesota, Minneapolis, MN; Surgery, University of Utah School of Medicine, Salt Lake City, UT.
BACKGROUND: Families may want the option to witness resuscitations, but are often not invited. Many health care providers believe that family presence(FP) is inappropriate, and may somehow hinder trauma care. Few studies have objectively evaluated this concern.
OBJECTIVE: To evaluate the impact of FP on pediatric trauma resuscitation care.
DESIGN/METHODS: Prospective data were collected over two years on all children presenting to the Emergency Department for Trauma Team evaluation. Available families were randomized on even days to the option of FP. Families entered the trauma room after the primary survey. Resuscitation time was defined as the time to complete the primary and secondary surveys, labs, and portable radiographs. In a convenience subsample, staff and families were surveyed.
RESULTS: 674 trauma activations were studied, 281 with family presence(FP+) and 393 without family presence(FP-). Age and sex distributions were similar for the two groups. FP- patients were more severely injured, with worse median ISS(9,11 p=0.012) and RTS(7.84,7.55 p<0.001) values. There was no difference in the efficiency of the resuscitation as measured by time intervals (table 1). Analysis of patients with severe(ISS>=15) and less severe (ISS<15) injury also showed no difference between groups (table 1). There was no statistically significant difference in time to IV placement, number of IV attempts or pain or sedation medications given. A large majority of families surveyed(94%) wanted the option of FP and 96% felt it was their right. Staff reported interruption or disruption by family in 19(7%) cases, but agreement about this by more than one staff member occurred in only 3 instances. Staff responses generally suggested FP was beneficial and did not prolong the resuscitation.
CONCLUSIONS: The results of this study suggest FP does not negatively impact pediatric trauma care measures, and is highly valued by families.
Effect of Simulation on Pediatric Resident Performance in Mock Resuscitation
Aaron Donoghue, Dennis Durbin, Frances Nadel, Glenn Stryjewski, Suzanne Kost, Vinay Nadkarni. Emergency Medicine, Childrens Hospital of Philadelphia, Philadelphia, PA; Critical Care Medicine, Childrens Hospital of Philadelphia, Philadelphia, PA; Critical Care Medicine, A.I. duPont Hospital for Children, Wilmington, DE; Emergency Medicine, A.I. duPont Hospital for Children, Wilmington, DE.
BACKGROUND: Mock resuscitations effectively train pediatric housestaff in resuscitation. Simulation is useful for realistic resuscitation training and may teach critical skills to residents superiorly.
OBJECTIVE: To determine whether pediatric residents have improved performance of critical skills in a mock asystole resuscitation following training enhanced by high-fidelity simulation compared to standard manikin simulation.
DESIGN/METHODS: Subjects were randomized to simulator (SIM) or manikin (MAN) groups. Subjects completed 3 study phases: phase 1 (baseline assessment), mock code involving an asystolic infant; phase 2 (intervention), didactic review of the PALS pulseless arrest algorithm; phase 3 (outcome assessment), repeat asystole mock code. SIM subjects performed phases 1 and 3 on an infant simulator; MAN subjects performed phases 1 and 3 on an infant manikin. Phase 2 was identical for both groups, except for simulated physical signs pertinent to the PALS algorithm for the SIM group. Outcomes were improvement in performance of critical tasks (pulse check, CPR, epinephrine, pulse recheck after ROSC), proper sequence of tasks (pulse check CPR epinephrine), and the number of pulse checks with rhythm changes(maximum of 3). Groups were compared using rank-sum analysis for change in tasks completed and number of pulse checks; proportion with tasks in correct sequence was analyzed by Fishers exact test.
RESULTS: 33 subjects completed the trial (SIM=17;MAN=16). Phase 1 performance of tasks was similar between groups as was frequency of correct sequence of tasks. SIM subjects had greater improvement in number of tasks performed than the MAN group (p=0.02) and performed more correct pulse checks with rhythm changes than the MAN group (2 1 vs. 1.2 1, p=0.03) More SIM subjects performed tasks in proper sequence during phase 3 than MAN subjects (12/17 vs. 6/16; p=0.05).
CONCLUSIONS: Pediatric residents had improved performance of critical skills in a mock code following training enhanced by high-fidelity simulation compared to standard manikin simulation.
Utility of Procalcitonin To Identify Young Febrile Infants at Low Risk of Serious Bacterial Infections
Scott Weiss, Andrew Dauber, Vincenzo Maniaci, Eric Nylen, Richard Bachur. Medicine, Childrens Hospital, Boston, MA; Endocrine, VAMC, Washington, DC.
BACKGROUND: Differentiating young febrile infants with serious bacterial infection (SBI) from those with viral infections is difficult without laboratory investigation. There has been little advancement in the diagnostic evaluation and treatment of these infants during the last decade. Procalcitonin (PCT) may be a novel biomarker for SBI.
OBJECTIVE: (1) study the test performance of PCT in young febrile infants; (2) determine an optimal cut-off value of PCT to identify infants at low risk for SBI.
DESIGN/METHODS: A prospective cohort study at a large pediatric ED. Infants 90 days with fever 38C and no bacterial infection by exam were enrolled. Subjects received routine care. Patients were excluded for antibiotic use, immunizations <48 hrs, or underlying conditions. Definite SBIs were defined as a pathogen from blood or CSF, 50,000 cfu/ml pathogen from urine, or 10-50,000 cfu/ml pathogen from urine with a positive urinalysis (UA). Possible SBIs were defined as 10-50,000 cfu/ml from urine with negative UA or focal opacity on chest x-ray. PCT was measured using a quantitative immunometric assay. The diagnostic performance of PCT was determined through ROC analysis; a cut-point was chosen to optimize identification of low-risk patients.
RESULTS: PCT was measured in 136 infants.18 patients (13%) had definite SBIs (13 UTI; 1 UTI w/ bacteremia; 4 bacteremia) and 6 (4%) had possible SBIs. Median ages of all patients and subgroup of SBI patients were 52 days(IQR 31-75) and 43 days(IQR 20-70). Mean PCT (SD) for definite SBI vs non-SBI were 1.79 (2.21) ng/mL vs 0.34 (0.93) [mean diff 1.45, 95%CI 0.85,2.04]. Mean PCT (SD) for combined definite and possible SBI vs non-SBI were 2.75 (4.58) ng/mL vs 0.34 (0.93) [mean diff 2.4, 95%CI 1.48,3.33]. Mean PCT for bacteremic patients was 2.90 ng/mL(range 0.25-7.30). Area under ROC curve for definite SBI and combined definite + possible SBI was 0.80 and 0.81.The optimal cut-point for determining low risk was 0.12 ng/mL yielding sensitivity 96%, specificity 30%, PPV 23%, NPV 97%, and positive and negative likelihood ratios (95%CI) of 1.38 (1.19,1.60) and 0.14 (0.02,0.95).
CONCLUSIONS: Procalcitonin can identify febrile infants at low risk of SBI. Using PCT, 30% of infants can be classified as low risk and might benefit from a more limited evaluation and observation without empiric antibiotics.
Is Neutropenia a Risk Factor for Serious Bacterial Infection among Children without Underlying Disease?
Elliot Melendez, Marvin B. Harper. Emergency Medicine, Childrens Hospital, Boston, Boston, MA; Pediatric Critical Care, Massachusetts General Hospital, Boston, MA.
BACKGROUND: Children with depressed immune function are known to be at increased risk of serious bacterial, viral, and fungal infections. There have been no published reports regarding the risk of infection in neutropenic children without underlying disease.
OBJECTIVE: To determine the risk of serious bacterial infection (SBI) among children without underlying immunodeficiency who present to the emergency department for evaluation and have newly identified and unexpected neutropenia. We hypothesize that the risk of SBI is low and that there is no relationship between the degree of neutropenia and risk of SBI among these children.
DESIGN/METHODS: Retrospective consecutive chart review from October 1995 through September 2003 of an urban emergency department of a tertiary childrens hospital. All patients aged 0-21 years presenting to the Emergency Department documented to have neutropenia defined as an absolute neutrophil count (ANC) of < 1000 cells/mm3 without known underlying risk factor for SBI were included. SBI was defined as growth of a pathogen from culture of blood, urine or cerebrospinal fluid.
RESULTS: There were 3,179 children with an ANC of < 1000/mm3 during the study period. Of these, 1,888 had no underlying immunodeficiency or central venous catheter (CVC). Fifteen of 453 (3%; 95%CI: 1.9, 5.4) infants less than 3 months of age had SBI; 7 bacteremia, 4 meningitis, and 8 urinary tract infection. SBI was rare among children over three months of age (18 of 1,435; 1.3%; 95%CI: 0.7, 2.0); 1 bacteremia, 0 meningitis, and 13 urinary tract infection.
CONCLUSIONS: The risk of SBI is low among children who present to the emergency department for any evaluation with or without fever found to unexpectedly have isolated neutropenia, unless the child was less than 3 months of age. Children older than 3 months with neutropenia but no underlying immunodeficiency or CVC are not at high risk of SBI. Hospital admission or empiric antibiotic treatment is not routinely indicated for the well appearing child greater than 3 months of age on the basis of neutropenia alone.
Emergency Medicine II
Prospective Validation of a Pediatric Dehydration Scale in Children with Acute Gastroenteritis
in the Emergency Department
Ran D. Goldman, Jeremy N. Friedman, Patricia C. Parkin. Divisions of Pediatric Emergency Medicine, Department of Pediatrics, the Hospital for Sick Children, Toronto, ON, Canada; Pediatric Medicine, Department of Pediatrics, the Hospital for Sick Children, Toronto, ON, Canada.
BACKGROUND: We previously described a 4-item clinical dehydration scale (range: 0-8) which includes: general appearance, eyes, mucous membranes, and tears. The scale was found reliable, discriminatory, and responsive to change. Following analysis of the diagnostic test properties, we developed a dehydration severity categories: score = 0 (none); score = 1 to 4 (mild); score = 5 to 8 (moderate to severe).
OBJECTIVE: To prospectively validate the clinical dehydration scale in a new cohort of patients with acute gastroenteritis assessed in a tertiary Emergency Department (ED) in a developed country.
DESIGN/METHODS: Children 1m-5y with symptoms of acute gastroenteritis assessed in the ED were consecutively enrolled during a 4 month period. The ED triage nurse completed the dehydration scale following a history and physical examination. Validation hypotheses included an association between severity category using the dehydration scale and: length of stay (LOS) in the ED; treatment with intravenous fluids; number of episodes of vomiting in the 7 days prior to the ED visit. ANOVA and chi square tests were used for comparison of severity categories, and a p value of < 0.05 was considered significant.
RESULTS: 205 children enrolled. Mean age of 22.4+14.9 months and 103 (50%) were male. The distribution of severity categories was as follows: 117 (57%) score = 0; 83 (41%) score = 1-4; and 5 (2%) score = 5-8. The three dehydration categories were significantly different with respect to the validation hypotheses: LOS (mean, SD in minutes: 245181, 397302, 501389 respectively, p<0.0001); treatment with intravenous fluids (17 (15%), 41 (49%), 4 (80%) respectively, p=0.001); number of vomiting episodes in the seven days prior to the ED visit (8.47.7, 13 10.7, 30.2 14.8 respectively, p<0.000).
CONCLUSIONS: The clinical dehydration scale and the three severity categorizations are valid in a prospectively enrolled cohort of patients assessed in our tertiary ED. The scoring system is valuable in predicting a longer LOS and the need for intravenous fluid rehydration in children with symptoms of acute gastroenteritis. The score may be added to a decision making algorithm to guide management of children with gastroenteritis.
Effect of Antibiotic Pretreatment on Cerebrospinal Fluid Profiles of Children with Culture Proven Bacterial Meningitis
L. Nigrovic, N. Kuppermann, C. Macias, J. Kanegaye, D. Moro-Sutherland, R. Schremmer, S. Schwab, D. Agrawal, K. Mansour, J. Bennett, Y. Katsogridakis, M. Mohseni, B. Bulloch, D. Steele, R. Kaplan, M. Herman, S. Bandyopadhyay, P. Dayan, U. Troung, V. Wang, B. Bonsu, J. Chapman, R. Malley. Pediatric Emergency Medicine Collaborative Research Committee.
BACKGROUND: While antibiotic pretreatment may sterilize bacterial cerebrospinal (CSF) cultures, the effect of antibiotic pretreatment on CSF profiles of patients with bacterial meningitis (BM) is not well defined.
OBJECTIVE: To evaluate the effect of antibiotic administration prior to lumbar puncture (LP) on CSF profiles in children with culture proven BM.
DESIGN/METHODS: We reviewed the medical records of all children 1 month to 18 years with BM who had presented to 20 pediatric emergency departments (2001-2004). BM was defined by a positive CSF culture for a bacterial pathogen or CSF pleocytosis with a positive blood culture and/or CSF latex agglutination. Duration of pretreatment was the time between starting antibiotics and the LP. We performed linear regression to determine the relationship between pretreatment and CSF parameters and identified the time point at which CSF profiles start becoming significantly affected by pretreatment.
RESULTS: We identified 217 patients with culture-proven BM with data available for length of pretreatment. After adjusting for patient age, and duration and severity of illness, longer duration of pretreatment was not associated with a difference in CSF white blood cell count (WBC), or CSF Absolute Neutrophil Count (ANC), but was significantly associated with higher CSF glucose and lower CSF protein. When compared to patients who were either not pretreated or pretreated for < 12 hours, patients with > 12 hours of pretreatment had similar CSF WBC (median 1063 vs. 1120 cells/mm3; p=0.76), CSF ANC (698 vs. 790 cells/mm3; p=0.49) and Gram stain positivity rates (50 vs. 65%; p=0.10) but higher CSF glucose (49 vs. 26 mg/dL; p=0.005), lower CSF protein (122 vs. 183 mg/dL; p= 0.02) and lower rates of positive CSF cultures (59 vs. 86%; p=0.001).
CONCLUSIONS: In patients with culture-proven BM, pretreatment does not modify CSF WBC, CSF ANC or Gram strain positivity but results in higher CSF glucose and lower CSF protein even after adjusting for age and duration and severity of illness. We cannot comment on these effects, however, on culture-negative pretreated patients with BM.
Observation Therapy for Acute Otitis Media in the Emergency Department
Jennifer H. Chao, Sergey Kunkov, Lilia B. Reyes, Stephanie Lichten, Ellen F. Crain. Pediatrics, Jacobi Medical Center, Bronx, NY.
BACKGROUND: Acute otitis media (AOM) is a common diagnosis in the emergency department (ED), and there are few data regarding parental acceptance of delayed antibiotic therapy in this setting.
OBJECTIVE: This study compared parental compliance and satisfaction with management of AOM using two methods of delayed antibiotic therapy.
DESIGN/METHODS: Children 2-12 years diagnosed with AOM who met American Academy of Pediatrics criteria for delayed antibiotic treatment were randomized to one of 2 groups, watchful waiting (WW) or safety net antibiotic protocol (SNAP). Children randomized to WW were discharged home with pain medications but no antibiotic prescription; those randomized to SNAP were discharged with pain medications and an antibiotic prescription. Parents in both groups received instructions on symptom management and indications to return or initiate antibiotic therapy. WW group parents were instructed to return to the ED or their childs health care provider if symptoms persisted 2 or 3 days. SNAP parents were instructed to fill their prescription if symptoms persisted 2 or 3 days. A research assistant blinded to group assignment called parents 7-10 days after the ED visit to assess satisfaction and compliance with delayed antibiotic therapy.
RESULTS: Two hundred twenty three children were enrolled; 113 in WW and 110 in SNAP. The mean age was 5.1 yrs (+ 2.44). Ninety-seven (86%) in the WW group and 98 (89%) in the SNAP group completed telephone follow up and comprised the sample for the remaining analyses. In the WW group 89 parents (92%) reported no antibiotic use within the 2-3 day observation period vs. 65 parents (66%) in the SNAP group (OR:4.88; 95% CI: 2.42-9.84). During the 7-10 day follow-up period, 81% of the WW group did not use antibiotics vs. 50% in the SNAP group (p=<0.001). The WW and SNAP groups did not differ in satisfaction with the visit (95% vs. 97% very or extremely satisfied, respectively, p= 0.44) or willingness to consider observation therapy in the future (65% vs. 62%, p=0.7). No complications were reported.
CONCLUSIONS: Watchful waiting and SNAP were both well accepted by parents of children diagnosed with AOM in an urban ED. Adherence to delayed antibiotic therapy was better for those offered WW than SNAP. These data suggest that in the ED setting, WW reduces antibiotic use without compromising satisfaction with the visit.
Laser Assisted Anesthesia Prior to Intravenous Cannulation; Acceptability and Effectiveness in Children
Deena Berkowitz, Ira T. Cohen. Emergency Medicine, Childrens National Medical Center, Washington, DC; Anesthesiology, Childrens National Medical Center, Washington, DC.
BACKGROUND: Intravenous cannulation is painful and anxiety provoking in children. A low power Erbium: YAG laser can desiccate a small area of the strateum corneum, allowing for rapid penetration of topical anesthetics. This device has been shown to be effective in adults prior to intravenous cannulation.Because children perceive and respond to pain differently than adults, the usefulness and application of this new device needs to be further studied in the pediatric population.
OBJECTIVE: Pretreament of the skin with laser-assisted anesthesia would reduce the pain of intravenous cannulation and phlebotomy in children. The intervention would be well accepted by pediatric patients and their parents.
DESIGN/METHODS: This is a double-blinded, randomized, controlled clinical trial. After IRB approval and parental consent, children ages 3- 18 years old who required intravenous cannulation were enrolled. All subjects had the laser device pressed against their skin, lidocaine 4% cream applied and after five minutes underwent intravenous cannulation. For patients randomized to the control group, the laser was not engaged. Using age-appropriate pain assessment scales, (Wong-Baker or Visual Analog Scale) the children were asked to report on the intensity of pain they experienced with the laser and with the phlebotomy. Patient and parent satisfaction was obtained by survey. Percent difference between groups for pain with needle insertion and comparative pain experience were compared using Chi Square tests and confidence intervals.
RESULTS: 94 patients ages 3-18 were enrolled. Mean age of control group was 11.2, mean age of treatment group was 8.9. Median pain of laser application was 0 in control group and 0 in treatment group. The mean pain of IV cannulation was significantly less for patients pretreated with laser (2.94 cm, 95% CI= 2 to 3.9) than for the control group (5.36cm 95%CI = 4.4 6.3). For parents of patients enrolled in the treatment group, 72% reported less pain with IV cannulation after laser compared to 38% reporting less pain in control group (p = 0.003).
CONCLUSIONS: In this study, laser-assisted penetration of local anesthesia in children undergoing phlebotomy and intravenous cannulation was found to be effective and well accepted.
PAMPPER; a Novel Pediatric Pain Management Educational Program for EMS Providers
Halim Hennes, Deborah Simpson, Ronald Pirrallo, Judith Rehm, Kenneth Sternig, Del Szewczuga, Michael Kim. Pediatrics, Medical College of Wisconsin, Milwaukee, WI; Medical Education, Medical College of Wisconsin, Milwaukee, WI; Emergency Medicine, Medical College of Wisconsin, Milwaukee, WI; Emergency Medical Services, Milwaukee County, Milwaukee, WI.
BACKGROUND: Emergency Medical Service (EMS) providers frequently transport children with painful conditions to emergency departments (ED). Despite the availability of analgesic agents most children do not receive prehospital analgesia. Studies identified inability to assess pain and poor knowledge of pediatric pain management as potential barriers.
OBJECTIVE: Evaluate the effect of 2-part educational intervention on EMS providers knowledge of pediatric pain management and determine its impact on pain management of children with extremity fracture or burn.
DESIGN/METHODS: The program was implemented in an urban EMS system with 234 paramedics. Part one is a 1-hour PowerPoint presentation with audio narration on pediatric pain physiology, assessment, and management. Part two a 1-hour facilitator-led small group discussion centered on six short video triggers scripted to portray prehospital pain assessment and management scenarios. EMS providers completed a 26-item validated multiple choice test before (pre) and after (post) participating in each part. Prehospital patient care records of eligible children transported during the study period were reviewed and data abstracted. Data analysis included Chi Square and Student t-test.
RESULTS: 145 providers completed both parts of PAMPPER and the corresponding pre and post-tests. Mean pre-test score for part one was 68.6% (95%CI; 66.6 - 70.4) and post-test score 84.5% (95%CI; 82.9-86.1), difference 16% (p=0.001). For part two mean pre-test score was 80.1%(95%CI; 78.1-82.0) and post-test score 87.9% (95%CI; 86.4-89.3), difference 7.8% (p=0.001). Proportion of patients with documented pain score before part 1 was 14.5% and following part 2 53.3% (p=0.0001). 7.2% of the patients received prehospital analgesia prior to part 1 compared to 33.3% follwing part 2 (p=0.0001).
CONCLUSIONS: Preliminary results indicate that our intervention improved EMS providers knowledge of pediatric pain assessment and management. The proportion of patients with documented pain scores and those who received prehospital analgesia increased significantly.
Propofol vs. Pentobarbital for Sedation of Children Undergoing MR Imaging:
Results from the Pediatric Sedation Research Consortium
Michael D. Mallory, Amy Baxter, Susanne Kost. Pediatric Emergency Medicine Associates, Childrens Healthcare of Atlanta, Atlanta, GA; Pediatrics, Thomas Jefferson University, Philadelphia, PA.
BACKGROUND: Pentobarbital (PB) and propofol (PF) are commonly used to sedate children for MRI.The Pediatric Sedation Research Consortium (PSRC) was created in 2003 to improve pediatric sedation process and outcomes.
OBJECTIVE: To use prospectively collected data from the PSRC to compare the effectiveness, efficiency and adverse events of PF versus PB sedation in large numbers of children undergoing MRI.
DESIGN/METHODS: We identified PSRC records of children who underwent sedation for MRI. We included American Society of Anesthesiologists (ASA) classification I-III patients age 6 months to 6 years, who recieved either intravenous PF or PB. We excluded patients who received other anesthetics or sedatives with the exception of midazolam. Outcomes included conditions produced during sedation, complications, and recovery time.
RESULTS: Of 15,175 MRI sedations, 7,079 met inclusion criteria (PF=5,072; PB=2,007). Age, weight and ASA classifications were similar between the two groups. Ideal sedation conditions were produced in 96.45% of those who received PB and in 96.8% of those who got PF, but PB use was about 10 times more likely to result in inadequate sedation than was PF (OR:9.71; CI:3.19, 25.59). PF resulted in physiologic changes more frequently than did PB (OR:7.77; CI:1.39, 43.30). PB was more likely to result in vomiting (OR:20.11; CI:2.04, 198.5). The incidence of apnea and overall airway complications was not significantly different between the two.
CONCLUSIONS: Among institutions contributing data to the PSRC, PF provides more efficient and effective sedation than PB for children receiving sedation for MR imaging, and is not associated with more airway complications.
Efficacy and Impact of Intravenous Morphine before Surgical Consult in Children with Right Lower Quadrant Pain Suggestive of Appendicitis: A Randomized Controlled Trial
Benoit Bailey, Sylvie Bergeron, Jocelyn Gravel, Jean-Francois Bussieres, Arie Bensoussan. Division of Emergency Medicine, Department of Pediatrics, CHU Ste-Justine, Montral, QC, Canada; Department of Pharmacy, CHU Ste-Justine, Montral, QC, Canada; Department of Surgery, CHU Ste-Justine, Montral, QC, Canada.
BACKGROUND: The evidence supporting the use of analgesia in children with abdominal pain suggestive of appendicitis is limited.
OBJECTIVE: To evaluate the risks and benefits of intravenous morphine before surgical consult in children with right lower quadrant pain suggestive of appendicitis in a pediatric emergency department.
DESIGN/METHODS: All children between the age of 8 and 18 years presenting to a pediatric emergency department with right lower quadrant abdominal pain and a presumptive diagnosis of appendicitis were eligible to be enrolled in a randomized double-blind placebo controlled trial if their initial pain was at least 5 out of 10 on a verbal numeric scale. Patients received either 0.1 mg/kg of morphine (maximum 5 mg) or placebo. The primary outcomes were: 1) the difference in pain using a visual analog scale at baseline and 30 minutes after the completion of the intervention, and 2) the time delay between arrival in the emergency department and the final surgical decision (surgery or discharge from surgery care).
RESULTS: Ninety patients with a suspected diagnosis of appendicitis were randomized to received morphine or placebo. Of these, 58/87 (67%) had a final diagnosis of appendicitis. Both groups were similar in terms of demographic, history, physical findings, probability of appendicitis, and initial pain score. There was no difference in the decrease of pain between the morphine and placebo groups: 24 23 mm and 20 18 mm, respectively (difference of 4 mm [95% CI -5, 12]). There was also no difference in the time delay between arrival in the emergency department and the surgical decision: median of 269 minutes (95% CI 240, 355) for morphine, and of 307 minutes (95% CI 239, 415) for placebo (difference of -34 minutes [95% CI -105, 40]).
CONCLUSIONS: The use of morphine in children with right lower quadrant abdominal pain with a presumptive diagnosis of appendicitis did not delay the surgical decision. In our group of patients, however, morphine was not more effective than placebo in diminishing their pain.
Use of Intranasal Fentanyl for the Relief of Orthopedic Trauma Pain
Mary Saunders, Kathleen Adelgais, Doug Nelson. Divison of Pediatric Emergency Medicine, University of Utah, Salt Lake City, UT.
BACKGROUND: Pediatric patients with painful orthopedic injuries often do not receive analgesia, due to NPO status and the invasive nature of IV placement. Intranasal fentanyl administration provides adequate and rapid analgesia, but has not been studied objectively for painful orthopedic injuries.
OBJECTIVE: To prospectively evaluate the use of intranasal fentanyl as analgesia for painful pediatric orthopedic injuries.
DESIGN/METHODS: We conducted a prospective interventional trial on a convenience sample of patients 3-18 years with painful orthopedic trauma. Eligible patients with pain scores greater than 3 faces on the Wong-Baker Faces (WBS) for 3-8 years or greater than 40mm on a Visual Analog Scale (VAS) for 9-18 years were enrolled. Fentanyl (2 mcg/kg, 100 mcg maximum) was administered intranasally using a syringe-tip atomization device. Pain scores and vital signs were obtained at baseline and at 10, 20, and 30 minutes after fentanyl administration. Satisfaction surveys were then completed using a 100mm VAS. The change in pain score was the primary outcome.
RESULTS: 81 patients were enrolled (28 in the VAS group and 53 in the WBS group). The mean patient age was 7.9 years and 60% were male. The fracture types included forearm (47%; 32% required reduction), supracondylar (21%), clavicle (8%), tibia/fibula (6%), and other (18%). Pain scores after intranasal fentanyl administration are shown in Tables 1 and 2. Mean satisfaction scores were 78.6 (95% CI: 73.9-83.3) for providers, 74.3 (95% CI: 69.3-79.3) for parents, and 61.6 (95% CI: 53.4-69.8) for patients. No adverse events were recorded.
CONCLUSIONS: Intranasal fentanyl at a dose of 2 mcg/kg is an effective method for providing analgesia to pediatric patients with moderate to severe pain due to orthopedic trauma.
Emergency Medicine III
Platelet Granzyme B Expression Is Increased in Sepsis:
A Potential Mechanism for Sepsis-Associated Endothelial Apoptosis
Robert J. Freishtat, JoAnne Natale, Joanna Cohen, Reecha Sachdeva, Lindsay M. De Biase, Eunis Ngor, Andrew Degnan, Angela S. Benton, Matthew Chow, Ervand Kristosturyan, Eric P. Hoffman. Research Center for Genetic Medicine, Childrens National Medical Center, Washington, DC; Division of Emergency Medicine, Childrens National Medical Center, Washington, DC; School of Medicine and Health Sciences, George Washington University, Washington, DC; Division of Pediatric Critical Care Medicine, University of California, Davis, Sacramento, CA; University of California, Davis, Davis, CA; George Washington University, Washington, DC; Johns Hopkins University, Baltimore, MD.
BACKGROUND: The role of platelets in sepsis-associated endothelial cell apoptosis is poorly understood. However, platelets have been shown to induce endothelial cell apoptosis via TGF1.
OBJECTIVE: We hypothesized that platelets are capable of inducing endothelial cell apoptosis in sepsis via up-regulation of other platelet cell death mediators.
DESIGN/METHODS: BALB/c mice (7-12 weeks old) were bled 0 and 24h after cecal-ligation and puncture (CLP). Platelet mRNA was used to determine cell death (GO:0008219) gene expression with Affymetrix Mouse Genome 430 2.0 GeneChips. Genes were filtered by ANOVA with an active cross-gene error model and p<0.1 followed by quantitative RT-PCR validation of select genes. Flow cytometry data was generated for proteins of interest from fresh whole blood samples.
RESULTS: All mice developed signs of sepsis. PLT cDNA microarrays showed up-regulation of 13 cell death mediators, particularly a 10-fold up-regulation of pro-apoptotic, Granzyme (Gzm) B (p=0.06), validated by RT-PCR [24h:0h relative expression (meanSE) = 10.120.07, p=0.04]. This correlated well with megakaryocytic mRNA sampled at the same time points (r=0.996). Flow cytometry of whole blood, gated on platelets by forward/side-scatter properties showed intracellular GzmB expression of 4.41.3% at 0h. At 24h, 19.66.3% (p=0.04) of platelets expressed intracellular GzmB.
CONCLUSIONS: These data show platelets up-regulate GzmB expression in a mouse model of severe sepsis. This supports the hypothesis that platelets are capable of inducing endothelial apoptosis in sepsis. If GzmB-mediated apoptosis is supported by our ongoing functional studies, this would provide a novel target for sepsis therapy.
Lorazepam Pharmacokinetics in Children with Status Epilepticus
Edmund V. Capparelli, John van den Anker, Kathleen Brown, Jill Baren, Prashant Mahajan, Cheryl Vance, Kathleen Lillis, Rachel Stanley, Colleen Davis, Tasmeen Singh, Jeffrey Blumer, James Chamberlain, the Pediatric Pharmacology Research Unit Network, the Pediatric Emergency Care Applied Research Network. Pedatrics, UC San Diego PPRU, San Diego, CA; Pediatrics, Emerg Med, and Pharmacology, Childrens Nat Med Ctr, Washington, DC; Pharmacology and Critical Care, Rainbow Babies and Childrens Hospital, Cleveland, OH; Emerg Med, Childrens Hosp Philadelphia, Philadelphia, PA; Emerg Med, Childrens Hosp of Detroit, Detroit, MI; Emerg Med, UC Davis, Sacramento, CA; Emerg Med, Childrens Hosp Buffalo, Buffalo, NY; Emerg Med, Childrens Hosp Michigan, Ann Arbor, MI; Emerg Med, Strong Memorial Hosp, Rochester, NY.
BACKGROUND: Despite widespread use for the treatment of pediatric status epilepticus (SE), the pharmacokinetic (PK) disposition of lorazepam has not been well defined in children.
OBJECTIVE: To determine the PK of lorazepam in pediatric SE patients.
DESIGN/METHODS: Pediatric patients age 0.4-18 yrs with SE (sparse sampling) or a history of SE (elective subjects [ELCT], intensive sampling) were studied for 48 hours. SE subjects received 0.05-0.1 mg/kg (4 mg max) per routine clinical care. ELCT received a dose of 0.05 mg/kg (2 mg max). PK data were analyzed with a 2-compartment population model (NONMEM) and non-compartmental methods (ELCT only).
RESULTS: There were 50 SE and 15 ELCT subjects with a median age of 7 years. A total of 331 lorazepam concentrations were included. The overall lorazepam post-hoc parameters from the population analysis were CL 0.070+0.051 L/hr/kg (mean + s.d.), Vdss 1.52+0.52 L/kg and half-life 17.3+7.3 hrs. The free-fraction was 10%. Weight adjusted CL and Vdss were modestly higher in younger infants and children compared to older children. Neither concomitant therapy with valproate or enzyme-inducing anticonvulsants nor seizure activity affected lorazepam CL. A dose of 0.1 mg/kg maintained lorazepam concentrations in the range associated with anticonvulsant effects for 12 hours. A maximum dose of 4 mg achieved lower serum levels in adolescents.
CONCLUSIONS: Pediatric lorazepam PK indicate that a dose of 0.1 mg/kg maintains lorazepam concentrations in the range associated with anticonvulsant effects for 12 hours.
Use of a Novel Triage Protocol Decreased Time in Department (TID) for Neonates with Hyperbilirubinemia
Presenting to a Pediatric Emergency Department (ED)
Hannah F. Smitherman, Chantal A. Caviness, Ann R. Stark. Pediatrics, Section of Emergency Medicine, Baylor College of Medicine, Houston, TX; Pediatrics, Section of Neonatology, Baylor College of Medicine, Houston, TX.
BACKGROUND: Neonates with hyperbilirubinemia often present to an ED and require urgent attention. Data are lacking on time to begin definitive therapy and disposition. Triage protocols may expedite patient care.
OBJECTIVE: To determine the duration of time between presentation and admission for definitive treatment of jaundiced neonates presenting to one ED after implementation of a triage algorithm that emphasized education, rapid recognition of hyperbilirubinemia and fast tracking to phototherapy.
DESIGN/METHODS: Using an electronic database (EMSTAT, A4 Health Systems), we identified neonates with jaundice and/or hyperbilirubinemia < 14 days of age presenting to the Texas Childrens Hospital ED. Prevalence of jaundice/hyperbilirubinemia was recorded along with TID and disposition for one year (06/01/04-05/31/05). Using a multi-disciplinary approach, a novel triage protocol based on the 2004 American Academy of Pediatrics Guidelines for hyperbilirubinemia management was instituted on 06/15/2006. TID and disposition for 3 months post-protocol (6/15-9/14/2006) was compared to the same periods in the 2 previous years. Statistical analysis used Mann-Whitney-U tests.
RESULTS: In the pre-protocol year, 966 of 96,726 ED visits were neonates < 14 days of age, including 259 (30%) related to jaundice. After protocol implementation, TID decreased significantly from 4.7 to 2.9 hrs (p = 0.025). More than 40% of neonates evaluated were admitted for phototherapy.
CONCLUSIONS: Jaundice represented nearly one-third of annual neonatal ED visits. Implementation of a triage protocol reduced TID by approximately 60%. Educational programs should target jaundice as a priority and utilize system analysis to improve patient care.
Low Risk Criteria for Pelvic Radiography in Pediatric Blunt Trauma Patients
Andrew T. Wong, KeriAnne B. Brady, David H. Rubin, David A. Listman. Dept of Emergency Med, St Barnabas Hospital, Bronx, NY; Dept of Pediatrics, St Barnabas Hospital, Bronx, NY; Dept of Pediatrics, Weill Cornell School of Medicine, New York, NY.
BACKGROUND: The American College of Surgeons recommends any patient with blunt trauma undergo radiographic evaluation including a pelvis x-ray (PXR). Studies have questioned the utility of routine PXR in pediatric blunt trauma victims. To date there are no accepted criteria to determine which patients do not require PXR. Selective elimination of PXR would save time, money and unshielded radiation exposure to the gonads.
OBJECTIVE: To determine which patients do not warrant a PXR based on mechanism of injury and physical exam.
DESIGN/METHODS: Medical records of blunt trauma patients birth-25 years from 1/1/2002 to 6/30/06 from an urban level 1 trauma center with PXRs were examined. Variables included sex, mechanism of injury (MOI), GCS, Pediatric Trauma Score, fall height, lower extremity injury, blood on rectal exam, blood at meatus and clinical need for CT. Outcomes were abnormal PXR, CT result and need for surgery.
RESULTS: 579 patients underwent 580 trauma evaluations (76% male). 22 (4 %) patients had a fracture identified on PXR. The age was 185.7 yrs (meanSD). Mechanisms of injury included fall (12.1%), MVC occupant (27.4%), MVC pedestrian (35.7%), assault (21%) and other (3.8%). Mean GCS was 14.4; with 4% of patients range 3-8, 4.3% range 9-13 and 91.7% range 14-15. There was no significant association between age, MOI, GCS, blood on rectal exam and presence of a pelvic fracture. There was a strong association between pelvic fracture and clinical lower extremity injury (p=.007), abnormal pelvis exam (p<.001) and clinical need for CT scan of abdomen and pelvis (p<.001). Patients requiring surgery for pelvis fracture had similarly significant associations. The negative predictive value for pelvis fracture was 98% for lower extremity injury, 99% for abnormal exam of pelvis, and >99% for clinical need for CT scan.
CONCLUSIONS: Clinical findings in pediatric blunt trauma patients have strong predictive value for pelvis fractures. The combination of lack of lower extremity injury, normal exam of pelvis and lack of clinical need for abdomino-pelvic CT scan reliably rule out the presence of pelvic fracture and the need for PXR. Retrospectively applying these criteria to our cohort would have eliminated the need for PXR in 263 (45%) patients.
Electronic Order Sets Promote Evidence-Based Corticosteroid Prescribing
in Emergency Dept. Patients with Acute Asthma
Scott D. Reeves, Ed Mendez, Beth Scheid, Stephanie Kennebeck. Pediatrics, Univ of Cincinnati College of Medicine, Cincinnati, OH; Division of Clinical Effectiveness, Cincinnati Childrens Hosp Medical Center, Cincinnati, OH; Division of Emergency Medicine, Cincinnati Childrens Hosp Medical Center, Cincinnati, OH; Pediatrics, Cincinnati Childrens Hosp Medical Center, Cincinnati, OH.
BACKGROUND: Pre-printed and electronic order sets have been used previously to simplify ordering tests and treatments on patients with specific diagnoses, and to direct care to coincide with best practice.
OBJECTIVE: We compared rates of appropriate steroid dosing (1 mg/kg of prednisone or equivalent) in patients whose ED orders were initiated using an electronic asthma order set versus those ordered using traditional computerized physician order entry (CPOE).
DESIGN/METHODS: This was a simple interventional study. An electronic asthma order set was created based on current asthma guidelines as published in the Evidence-based Practice Guideline at our institution. The order set was built into the electronic documentation system (EMSTAT, A4 Corp,Cary, NC.) It pre-filled doses of beta-agonist and steroid based on previously entered patient weights. The order set was implemented September 22, 2005. Use of the order set was not mandatory. Data collection took place from November 1, 2005 to September 1, 2006. Data was collected on patients who had a diagnosis of asthma and received at least one albuterol treatment during their ED stay. Outcomes included correct steroid dosing, time to first albuterol and appropriate documentation of respiratory scores. Patients who received care with the Asthma Order Set and were compared to those whose care was ordered using the standard CPOE method.
RESULTS: 1556 patients with acute asthma were seen in our ED during the study period. Patients who were cared for using the Order Set received the recommended steroid dose 97% of the time versus 72% of the time for those for which the order set was not used (p<0.0001). The average time to first albuterol treatment was twelve minutes shorter using the electronic order set and documentation of respiratory scores was 25% higher.
CONCLUSIONS: In our ED setting the use of an electronic order set improved correct steroid dosing in acute asthma as compared to no order set use. Secondary outcomes (time to first albuterol and respiratory score documentation) improved as well.
A Randomized Trial of a Three-Part Intervention To Improve Primary Care Follow-Up
after an Emergency Department Visit for Asthma
Joseph J. Zorc, Amber Chew, Julian L. Allen, Kathy N. Shaw. Pediatrics, The Childrens Hospital of Philadelphia Univ of Pennsylvania, Philadelphia, PA.
BACKGROUND: Prior studies in urban emergency departments (EDs) have found poor quality of chronic asthma care and low rates of follow-up with a primary care provider (PCP) after an ED visit.
OBJECTIVE: To assess the efficacy of a three-part ED-based intervention including symptom-screening, viewing an educational video, and a mailed follow-up reminder on the rates of follow-up with a PCP and asthma-related outcomes after an ED visit.
DESIGN/METHODS: This randomized trial enrolled children age 1-18 years discharged after asthma treatment in an urban pediatric ED. Control subjects received standard instructions to follow up with a PCP within 3-5 days. In addition, intervention subjects: 1. Received a letter to take to their PCP if they screened positive for persistent asthma symptoms; 2. Viewed a video featuring families and providers discussing the importance of asthma control; 3. Were mailed a reminder to follow up with a PCP. All subjects were contacted by phone at 1, 3, and 6 months after the ED visit and follow-up was confirmed by PCP record review. Asthma-related quality of life (AQOL), symptoms, and beliefs about asthma care were assessed using validated measures.
RESULTS: 433 subjects were enrolled and baseline measures were similar between study groups. After the intervention and prior to ED discharge, subjects rated satisfaction with the visit highly and were more likely to strongly endorse beliefs about the benefits of regular care than controls. However, the percentage following up with a PCP during the 4 weeks after the ED visit (45.1%) was similar to controls (43.9%). During the 6 months after enrollment, the groups were also similar in subsequent ED visits, use of inhaled corticosteroids, and AQOL .
CONCLUSIONS: An ED-based intervention to improve long-term asthma control influenced short-term beliefs and was acceptable to families but did not improve PCP follow-up or asthma-related outcomes. Additional interventions are needed to improve asthma care and outcomes after ED discharge.
Home Oxygen Therapy from the Emergency Department for Patients with Uncomplicated Bronchiolitis and Hypoxia
Stephanie J. Todd, Genie E. Roosevelt, Joan P. Bothner, Lalit Bajaj. Pediatrics, Section of Emergency Medicine, The Childrens Hospital- UCDHSC, Denver, CO.
BACKGROUND: Bronchiolitis is the most common reason for hospital admission in children <1 year of age. A previous study demonstrated the feasibility of home oxygen from the Emergency Department (ED). This study is a continuation of evaluating home oxygen therapy for patients with hypoxia and uncomplicated bronchiolitis.
OBJECTIVE: To prospectively evaluate an ED protocol of home oxygen therapy and to characterize risk factors associated with unscheduled follow up and subsequent admission.
DESIGN/METHODS: We performed a prospective observational study of all patients discharged from the ED with home oxygen who presented to a tertiary care childrens hospital at 5280 ft (1600 m) between Dec 2005 and April 2006. The protocol included an 8 hour observation period prior to discharge on home oxygen. A subgroup of these patients with uncomplicated bronchiolitis (defined as a: URI with associated wheezing and/or crackles, ages 3-18 months and with no prior wheezing history) were analyzed. We recorded demographic and clinical variables to characterize which patients may be at increased risk for unscheduled return or subsequent admission after discharge on home oxygen.
RESULTS: 117 patients were discharged home from the ED on oxygen. 84 (72%) met our definition for uncomplicated bronchiolitis: mean age 8.4 months, mean observation time 10 hours, mean discharge oxygen 0.4 L/min. 37 patients (44%) had saturations < 85%. 68 patients (81%) did not have unscheduled follow up (f/u) and completed their home oxygen treatment without complication. 16 (19%) patients returned for unscheduled f/u, of which 8 (50%) were admitted. Patients with saturations < 85% were 3.5 (RR 95% CI: 1.2, 9.9) times more likely to have an unscheduled f/u than those who had oxygen saturations 85%. However, 25 (68%) of patients with initial saturation < 85% completed their home oxygen therapy without complications. There were no differences between these groups with respect to age or vital signs at presentation. There were no ICU admissions and no adverse events at home.
CONCLUSIONS: The majority of patients with uncomplicated bronchiolitis and hypoxia were safely managed as outpatients with supplemental oxygen. More cautious management may be indicated for patients with presenting oxygen saturations < 85% at 5280 ft (1600m).
Predicting Need for Intensive Care in Children Hospitalized with RSV Bronchiolitis
Frank D. Petruzella, Elizabeth Powell. Pediatrics, Childrens Memorial Hospital, Chicago, IL; Emergency Medicine, Childrens Memorial Hospital, Chicago, IL.
BACKGROUND: Respiratory Syncytial Virus (RSV) lower respiratory tract infection (LRI) is a common cause of hospitalization of healthy infants.
OBJECTIVE: To determine if clinical and demographic variables assessed at time of hospital admission can be used to identify those who are at higher risk of a severe disease course.
DESIGN/METHODS: Retrospective case-control study of children 0-24 months admitted to the inpatient wards at Childrens Memorial Hospital from the Emergency Department (ED) with RSV infection during the RSV seasons of 2000-2005. Cases were defined as patients admitted to the wards that were transferred to PICU, and controls were defined as patients admitted to and managed on wards until discharge. Cases and controls were matched on age and season of admission. Patients with congenital heart disease, chronic lung disease, gestational age <36 wks, or immunodeficiency were excluded. Clinical and demographic variables were collected at various times during hospital admission.
RESULTS: There were 6483 visits to the ED for RSV LRI; 2204 (34%) were admitted to the hospital. 20 children were admitted to the ward and were later transferred to the PICU. No significant differences between case and control groups were found in gender, race or presence of smoker at home. At presentation to the ED, cases (requiring transfer to PICU) had fewer days of symptoms (2 1 days vs. 3.8 1.9 days; p<0.01), lower O2 saturation (92.6% 6.2% vs. 96.8% 2.9%; p<0.04), and higher clinical asthma score (3.5 1.6 vs. 2.3 1.7). On arrival to the inpatient floor, more case patients required O2 therapy than controls (65% vs. 30%; p=0.01). Among those transferred to the PICU, 15% were transferred within 12 hours of admission, 20% were transferred 12-24 hours after admission, and 65% were transferred >24 hours after admission.
CONCLUSIONS: Among healthy children with RSV LRI admitted to the hospital ward, only a small percentage will require transfer to the PICU. Patients requiring transfer presented earlier in their disease course, and most were transferred more than 24 hours after admission. Although O2 saturation and asthma scores at presentation were statistically different between groups, there was significant overlap which limited the usefulness of these measures to predict disease severity.
