Search

Purpose Clinicians and researchers must contextualize a patient’s genetic variants against population-based references with detailed phenotyping. We sought to establish globally scalable technology, policy, and procedures for sharing biosamples and associated genomic and phenotypic data on broadly consented cohorts, across sites of care. Methods Three of the nation’s leading children’s hospitals launched the Genomic Research and Innovation Network (GRIN), with federated information technology infrastructure, harmonized biobanking protocols, and material transfer agreements. Pilot studies in epilepsy and short stature were completed to design and test the collaboration model. Results Harmonized, broadly consented institutional review board (IRB) protocols were approved and used for biobank enrollment, creating ever-expanding, compatible biobanks. An open source federated query infrastructure was established over genotype–phenotype databases at the three hospitals. Investigators securely access the GRIN platform for prep to research queries, receiving aggregate counts of patients with particular phenotypes or genotypes in each biobank. With proper approvals, de-identified data is exported to a shared analytic workspace. Investigators at all sites enthusiastically collaborated on the pilot studies, resulting in multiple publications. Investigators have also begun to successfully utilize the infrastructure for grant applications. Conclusions The GRIN collaboration establishes the technology, policy, and procedures for a scalable genomic research network.

Clinicians and researchers must contextualize a patient's genetic variants against population-based references with detailed phenotyping. We sought to establish globally scalable technology, policy, and procedures for sharing biosamples and associated genomic and phenotypic data on broadly consented cohorts, across sites of care.Three of the nation's leading children's hospitals launched the Genomic Research and Innovation Network (GRIN), with federated information technology infrastructure, harmonized biobanking protocols, and material transfer agreements. Pilot studies in epilepsy and short stature were completed to design and test the collaboration model.Harmonized, broadly consented institutional review board (IRB) protocols were approved and used for biobank enrollment, creating ever-expanding, compatible biobanks. An open source federated query infrastructure was established over genotype-phenotype databases at the three hospitals. Investigators securely access the GRIN platform for prep to research queries, receiving aggregate counts of patients with particular phenotypes or genotypes in each biobank. With proper approvals, de-identified data is exported to a shared analytic workspace. Investigators at all sites enthusiastically collaborated on the pilot studies, resulting in multiple publications. Investigators have also begun to successfully utilize the infrastructure for grant applications.The GRIN collaboration establishes the technology, policy, and procedures for a scalable genomic research network.

American Pediatric Society 2011 Presidential Address: Earthquakes, Tectonic Shifts in Graduate Medical Education, and the Role of the APS in Finding Solid Ground

ACADEMIC EMERGENCY MEDICINE 2011; 18:906–911 © 2011 by the Society for Academic Emergency Medicine Abstract Objectives: The objectives were to identify patient and hospital characteristics associated with the use of computed tomography (CT) imaging of the cervical spine (c-spine) in the evaluation of injured children and, in particular, to examine the influence of hospital setting. Methods: This was a retrospective cohort of children younger than 19 years of age from the Massachusetts Hospital Emergency Department (ED) database who were discharged from the ED with an injury diagnosis from 2005 through 2009. Multivariable logistic regression was used to analyze characteristics associated with CT imaging of the c-spine. Results: Of the 929,626 pediatric patients diagnosed with an injury in Massachusetts EDs and then discharged home, 1.3% underwent CT imaging of the c-spine. Rates of CT imaging nearly doubled over the 5 years. In the multivariable model, patient age (adjusted odds ratio [AOR] = 2.3, 95% confidence interval [CI] = 2.0 to 2.7 for children age 12 to 18 years vs. under 1 year of age) and evaluation outside of a pediatric Level I trauma center (AOR = 2.2, 95% CI = 1.1 to 4.3 for children evaluated at non Level I trauma centers vs. pediatric Level I trauma centers; AOR = 2.1, 95% CI = 0.93 to 4.7 for children evaluated at adult Level I trauma centers vs. pediatric Level I trauma centers) were associated with higher rates of CT imaging of the c-spine. Conclusions: Cervical spine CT imaging for children discharged from the ED with trauma diagnoses increased from 2005 through 2009. Older age and evaluation outside a Level I pediatric trauma center were associated with a higher c-spine CT rate. Educational interventions focused outside pediatric trauma centers may be an effective approach to decreasing CT imaging of the c-spine of pediatric trauma patients.

OBJECTIVES: To determine whether a Web-based educational intervention improves emergency physicians' knowledge about bioterrorism and to survey physicians' knowledge and sources of information on bioterrorism. METHODS: Prospective randomized controlled trial using pre- and postintervention testing in hospitals. Participants were general and pediatric emergency medicine attending physicians, fellows, and fourth-year emergency medicine residents. All participants completed a pretest and attended a lecture on bioterrorism. Participants were then randomized to the Web intervention group that received continuous access to a bioterrorism educational Web site with weekly exposure to case scenarios of diseases due to biological agents, or the control group. Participants were retested after one and six months and surveyed to identify their sources of information and assess their knowledge. RESULTS: Sixty-three physicians completed the pretest. There was no difference in mean +/- standard deviation (SD) pretest scores between Web intervention (45% +/- 10%) and control (44% +/- 10%) groups (mean difference: 1.9%; 95% confidence interval [CI] = -6.7% to 2.9%). There was no significant difference between pre- and posttest scores among groups at one month (Web intervention 48% +/- 10% vs. control 45% +/- 10%; mean difference: 3.3%; 95% CI = -8.5% to 2.0%) and six months (Web intervention 51% +/- 8% vs. control 47% +/- 9%; mean difference: 3.8%; 95% CI = -8.8% to 1.2%). More than 60% of physicians cited media reports as their primary source of information on bioterrorism and believed that their knowledge of bioterrorism was limited after one month. CONCLUSIONS: Providing physicians information on bioterrorism through simulated cases and continuous access to an educational Web site does not increase knowledge of bioterrorism. Physicians are more likely to use media reports for their primary source of information. Language: en

An 11-year-old girl presented with a papulovesicular rash and severe respiratory distress 5 weeks after receiving varicella vaccine. Restriction fragment length-polymorphism analysis of virus isolated from an endotracheal-tube aspirate and from bronchoalveolar lavage revealed that this patient's illness was due to the Oka vaccine strain of varicella. An extensive immunologic analysis failed to identify a known diagnostic entity to explain her susceptibility to this attenuated vaccine strain. Analysis of her lymphocytes on separate occasions, months after recovery from her illness, revealed a profound deficiency of natural killer T (NKT) cells and of NKT-cell activity, suggesting that NKT cells contribute to host defense against varicella virus.

Study objective: Success of prospective studies, particularly in the emergency department, often depends on immediate identification of eligible patients to ensure timely sample collection and initiation of study interventions. We report use of a real-time automated notification system to identify potential patients for a clinical trial at the time of ED registration on the basis of information routinely collected. We hypothesize that the automated notification system improves the rate of investigator notification. Methods: We performed a prospective comparison of the notification rate by the automated notification system compared with that by ED clinicians. Results: In the 11 months before use of the automated notification system, the investigator was notified by ED staff for 56% of 61 potentially eligible patients. During 10 months of using the automated notification system, the investigator was paged by the automated notification system for 84% of 49 potentially eligible patients. Conclusion: The automated notification system improves study investigator notification. Use requires online linked registration, a database, and paging systems. The automated notification system is a potentially valuable tool in the recruitment of patients for clinical trials. [ Ann Emerg Med. 2003;41:242-246.]

We have purposely expanded on the well-known ATLS paradigm to aid EHCPs in their approach to a potential bioterrorism event. By building on a process that is already familiar, we hope this will aid the EHCP to remember a systematic approach to such an incident. By following this ten-step process, we believe that all EHCPs, and especially those practicing at the first echelons of care in urgent care clinics and EDs, can approach the daunting problem of biological defense with a good deal more confidence. This same model advocated for bioterrorism also may apply to natural infectious disease epidemics, particularly of emerging or re-emerging diseases, that might not be optimally managed by reliance on the conventional public health strategy that requires physician-dependent definitive diagnosis and active reporting mechanisms. The authors hope the acquired knowledge and skills one might gain will rarely be needed, but if the events surrounding the dispersal of anthrax-contaminated mail in the fall of 2001 are any indication of the future, such competencies will be invaluable.

Geographic circumstances: We participated intimately in disaster response after Hurricane Katrina's landfall and other disasters since then. About the lead author (Dr Weiner): I am a pediatric emergency medicine physician at Children's Hospital Boston (CHB)/Harvard Medical School (HMS) and the CHB representative and a charter member of the HMS Section on Disaster Medicine. I have been a National Disaster Medical System member on the Disaster Medical Assistance Team, the International Medical Surgical Response Team, and the Pediatric Specialty Team since 1995. I am a member of the American Academy of Pediatrics Extended Disaster Preparedness Advisory Council Network. I have served as a domestic and international consultant in disaster planning and am developing technology-based resources for disaster training and response. Hurricane Katrina was a disaster unlike others for which the US National Disaster Medical System (NDMS) had been deployed. Never before had the NDMS attempted to provide sustained disaster relief over a geographic area as expansive as the multistate region devastated first by Hurricane Katrina and then by Hurricane Rita. There were challenges we anticipated and those that we did not. Lessons learned from Hurricanes Katrina and Rita, and subsequent disasters, have led and continue to lead to strategic conceptual, organizational, and operational modifications to improve care to disaster victims.1 Increased emphasis on the care of children in disasters is an important component of these initiatives, given that for most disasters more than one-third of victims are children.2 As demonstrated by the hurricanes in the Gulf Coast region, and again by the earthquake in Haiti in January 2010, the NDMS remains a critical asset for disaster relief and one of the few US disaster-relief organizations/agencies with the capacity to provide care for critically injured or ill disaster victims including children. Lessons learned from and actions taken since Hurricanes Katrina … Address correspondence to Debra L. Weiner, MD, PhD, Emergency Medicine, Children's Hospital Boston, 300 Longwood Ave, Boston, MA 02115. E-mail: debra.weiner{at}childrens.harvard.edu

Study objective: Occult pneumococcal bacteremia (OPB) occurs in 2.5% to 3% of highly febrile children 3 to 36 months of age, and 10% to 25% of untreated patients with OPB experience complications, including 3% to 6% in whom meningitis develops. The purpose of this study was to identify predictors of OPB among a large cohort of young, febrile children treated as outpatients using multivariable statistical methods. Methods: We derived and validated a logistic regression model for the prediction of OPB. We evaluated 6,579 outpatients 3 to 36 months of age with temperatures of 39° C or higher who previously had been enrolled in a study of young febrile patients at risk of OPB in the emergency departments of 10 hospitals in the United States between 1987 and 1991; 164 patients (2.5%) had OPB. We randomly selected two thirds of this population for the derivation of the model and one third for validation. In the derivation set, we analyzed the univariate relationships of six variables with OPB: age, temperature, clinical score, WBC count, absolute neutrophil count (ANC), and absolute band count (ABC). All six variables were then entered into a logistic regression equation and those retaining statistical significance were considered to have an independent association with OPB. Results: Patients with OPB were younger, more frequently ill-appearing, and had higher temperatures, WBC, ANC, and ABC than patients without bacteremia. Only three variables, however, retained statistically significant associations with OPB in the multivariate analysis: ANC (adjusted odds ratio [OR] 1.15 for each 1,000 cells/mm 3 increase, 95% confidence interval [CI] 1.06, 1.25), temperature (adjusted OR 1.77 for each 1° C increase, 95% CI 1.21, 2.58), and age younger than 2 years (adjusted OR 2.43 versus patients 2 to 3 years old, 95% CI interval 1.11, 5.34). In the derivation set, 8.1% of patients with ANCs greater than or equal to 10,000 cells/mm 3 had OPB (95% CI 6.3, 10.1%) versus .8% of patients with ANCs less than 10,000 cells/mm 3 (95% CI .5, 1.2%). When tested on the validation set, the model performed similarly. Conclusion: Independent predictors of OPB in children 3 to 36 months of age with temperatures of 39° C or higher treated as outpatients include ANC, temperature, and age younger than 2 years. These predictors may be used to develop clinical strategies to limit laboratory testing and antibiotic administration to those children at greatest risk of OPB. [Kuppermann N, Fleisher GR, Jaffe DM: Predictors of occult pneumococcal bacteremia in young febrile children. Ann Emerg Med June 1998;31:679-687.]

This study describes a field observation study and use of simulation to quantify the effect of patient arrival rate and physician practices on physician idle time and patient wait time. The observation study measured actual service (diagnosis, therapy, and charting) times for 126 patients. Subsequently, a FORTRAN simulation model examined effects of physician practices and patient arrival rate on physician utilization and patient wait time. Observations were taken in the emergency department of an urban, university-affiliated pediatric teaching hospital. Although times for initial diagnostic evaluation (diagnosis), therapy, and charting averaged 13.3, 13.8, and 11.6 minutes, respectively, maximum patient visits approached six hours. The simulation model showed that, during times of frequent patient arrivals, maximum patient wait times increased greatly. Additionally, the model predicted that physician idle time persists even during periods of frequent patient arrivals and long maximum visit times. Emergency department (ED) senior staff (fellows and attendings) often begin treating new patients when current patients leave for tests external to the ED. This practice increases physician utilization and makes more physician capacity available, but it may lead to small additional patient waits if the physician is treating another patient when the first patient returns from the external test before their original physician becomes free. On the other hand, during periods of high patient arrivals, increased physician utilization and reduced idle time result in reduced average and expected maximum patient visit times. Unfortunately the variability in visit times increased. A very small percentage of patients wait longer, owing to additional waits incurred upon returning from external testing. Overall, most patients benefit from shorter visits. Finally, the study suggested maximum rather than average wait time be considered as a measure of emergency department capacity and quality of service provided. Although average wait times seemed reasonable, maximum wait times were at times quite long and could impact both physician's and patient's perceptions of service quality.

We have developed a model of low-inoculum Streptococcus pneumoniae infection in infant rats. We challenged 4-day-old Sprague-Dawley pups via intraperitoneal or intrapulmonary injection of S. pneumoniae serotypes 1, 3, 4, 5, 6b, 7f, 9v, 14, 19f, and 23f. To achieve bacteremia with low inocula, it was necessary to passage the isolates in rats. Inocula of the 10 S. pneumoniae serotypes producing bacteremia in 50% or more animals ranged from 1 to 400 CFU. Virulence was similar by intraperitoneal and intrapulmonary routes. Lung specimens from animals challenged by the intrapulmonary route grew S. pneumoniae and demonstrated histologic evidence of focal infection. Meningitis was detected in 20 to 50% of bacteremic animals, and mortality invariably followed bacteremia within 24 to 48 h. This model of intrapulmonary infection uses low inocula of S. pneumoniae and results in bacteremia, meningitis, and death in infant rats.

Objective We determined the incidence of serious invasive bacteremia caused by Neisseria meningitidis) and other organisms in febrile infants and children with a petechial rash. Further, we studied the diagnostic value of laboratory and clinical findings in these patients. Study design We conducted this prospective cohort study in the emergency department of an urban pediatric teaching hospital, during an 18-month period, and enrolled consecutive patients with temperature of 38° C or higher and petechiae. Our measures included (1) laboratory tests (leukocyte count, coagulation profile, blood culture, and cerebrospinal fluid bacterial culture); (2) a questionnaire requesting clinical data including general appearance, number and location of petechiae, and presence or absence of purpura; and (3) a follow-up telephone survey documenting health status. Results A total of 411 patients were enrolled, with 57.7% between 3 and 36 months of age. Eight patients (1.9%) had bacteremia or clinical sepsis. Six had serious invasive bacteremia: N. meningitidis (two patients), group A streptococcus (one), or sepsis with negative culture results (three). Two had occult bacteremia caused by Streptococcus pneumoniae and no evidence of sepsis. No patient had a positive cerebrospinal fluid culture result. None of the 357 well-appearing patients (95% confidence interval: 0.0%, 1.0%) had serious invasive bacteremia. Fifty-three patients appeared ill, including all six with serious invasive bacteremia. Ill appearance of the child had a sensitivity of 1.00 (95% confidence interval: 0.60, 1.00), and a leukocyte count of 15,000 or greater, or of less than 5000, had a sensitivity of 1.0 (95% confidence interval: 0.53, 1.00) for detecting serious invasive bacteremia. All children with meningococcemia had purpura. Conclusions Invasive bacteremia occurred less frequently in our study than in previous series and was identified by clinical criteria. Our data support the treatment of selected well-appearmg children “with fever and petechiae as outpatients.

Objective: To compare a recombinant bactericidal/permeability-increasing protein variant and a recombinant endotoxin-neutralizing protein. Design: Randomized, blinded, controlled study, using a rat model of sepsis. Setting: Animal research facility. Subjects: Male Wistar rats. Interventions: An inoculum of 1.5 x 10 7 to 1.8 x 10 8 Escherichia coli O18ac K1, implanted in the peritoneum, produced bacteremia in 95% of animals after 1 hr. One hour after E. coli challenge, animals received recombinant bactericidal/permeability-increasing protein variant, recombinant endotoxin-neutralizing protein, or saline intravenously, followed by ceftriaxone and gentamicin intramuscularly. Measurements and Main Results: Twenty-four (85.7%) of 28 animals receiving recombinant endotoxin-neutralizing protein (p < .001 vs. control) survived 7 days compared with nine (33.3%) of 27 recombinant bactericidal/permeability-increasing protein variant-treated (p

Objective : To quantify the frequency of blood loss necessitating transfusion and identify the clinical factors predictive of severe hemorrhage in children with femoral fractures. Design : Retrospective review of computerized discharge diagnoses and medical records between January 1, 1987, and July 31, 1992. Setting : Tertiary care children's hospital. Patients : Children younger than 18 years of age, discharged between January 1, 1987, and July 31, 1992, with a final diagnosis of femur fracture. Results : The 257 patients ranged in age from birth to 18 years with a mean of 6.5 years ; 183 (71%) were male. Fractures were closed in 250 (98%) and represented isolated injuries in 225 (87%) patients. The mean systolic blood pressure (SBP) was 119 mmHg, the mean heart rate (HR) was 114 beats/min, and the mean hematocrit (Hct) was 35.2% on arrival. Eight (2.9%) patients had a SBP less than 90 mmHg, all of which were normal values for age. There were 19 patients with a HR >150, all were

Objective To assess the benefit of a recombinant endotoxin neutralizing protein from Limulus polyphemus in treating Gram-negative bacterial sepsis in rabbits. Design Prospective, blinded, controlled, laboratory trial. Setting Animal research laboratory. Subjects New Zealand White rabbits. Interventions We established a rabbit model of Escherichia coli peritonitis and bacteremia, with high mortality rate, despite treatment with gentamicin and ceftriaxone. Twenty-five pairs of male New Zealand White rabbits were challenged intraperitoneally with E. coli O18ac K1 in 5% porcine mucin (mean 7 times 109 colony-forming units). All animals were treated with intravenous gentamicin (2.5 mg/kg) and ceftriaxone (100 mg/kg), and with either intravenous endotoxin neutralizing protein (50 mg/kg) or saline 1 hr after E. coli challenge. Measurements and Main Results All animals were bacteremic 1 hr after challenge (mean 3.6 times 105 colony-forming units/mL). Animals in both groups developed tachycardia, hypotension, and acidosis (NS). Geometric mean serum endotoxin and tumor necrosis factor (TNF) concentrations were significantly (p less than .001) higher 1 hr after challenge compared with baseline prechallenge concentrations in both groups. From 1 to 2 hrs after challenge, endotoxin concentrations increased 2.5-fold in control animals (95% confidence interval equals 13.1 to 32.9 endotoxin units/mL, p equals .024), whereas endotoxin concentrations increased only 1.2-fold in endotoxin neutralizing protein-treated animals (95% confidence interval equals 20.4 to 23.8 endotoxin units/mL, NS). TNF concentrations increased significantly (p less than .001) in both groups from 1 to 2 hrs after challenge. Eighteen (72%) of 25 endotoxin neutralizing protein-treated animals vs. 11 (44%) of 25 controls survived 24 hrs (p equals .032). Conclusions Treatment with endotoxin neutralizing protein had the following effects: a) the increase in serum endotoxin was blunted, but not TNF concentrations measured 1 hr after antibiotic treatment; and b) survival in rabbits with E. coli sepsis was improved.

By using temporal artery and rectal thermometers, we followed temperatures in 45 febrile (38.5 degrees C) infants given an antipyretic drug. Sixty and 90 minutes after drug administration, temporal artery temperatures had decreased significantly more than rectal temperatures. When body arterial temperature changes rapidly, changes in rectal temperature may lag.

Objective. To quantitate the increase in invasive group A β-hemolytic streptococcal (GABHS) infections and to define a possible association between GABHS bacteremia and primary varicella zoster virus (VZV) infections. Methods. This was a retrospective chart review conducted at Children's Hospital. Participants were patients with documented GABHS bacteremia occurring from January 1977 through December 1993. Measurements/Main Results. We identified 63 episodes of GABHS bacteremia in 62 patients. From 1977 to 1992, a mean of 3.2 ± 2 cases occurred per year (range, 0 to 6), increasing by a factor of 3 (10 cases) in 1993. The median age was 4 years (range, 1 day to 20 years; mean, 8 years ± 3 months); 36 were male; five children were immunocompromised. One child was dead on arrival and one had a cardiac arrest during evaluation in the emergency department. Primary sites of infection (oropharynx, skin, or middle ear) were identified in 40 (75%) of the cases; in addition, 10 cases occurred in patients with primary VZV. From 1977 to 1992, we identified five VZV-associated cases; an average of 7 ± 11.5% of the patients with GABHS had concurrent VZV infection annually, with no more than one case per year. In 1993, 50% of the 10 new GABHS cases were in children with VZV infection (P = .003, Fisher's exact test). The diagnosis of invasive GABHS infection in patients with VZV was not readily recognized, requiring a median of two (range, one to four) physician visits before admission and the administration of antibiotics. All 10 children were diagnosed on the fourth or fifth day of the exanthem and were febrile (39.6 ± 1.1°C, range, 38.3 to 40.8°C), with a mean white blood cell count (WBC) of 11 500 ± 8 400/mm3 (8 of 10 cases had a WBC less than 15 000/mm3). None of the five VZV-associated cases in 1993 had signs of cutaneous bacterial superinfection; among these were two cases of streptococcal toxic shock syndrome (one death), one case of osteomyelitis, and two cases of occult bacteremia. Of the five VZV-associated cases before 1993, one patient was diagnosed with supraglottitis, one with septic arthritis, one with orbital cellulitis, and two solely with impetiginized or cellulitic lesions. Conclusions. We found that the incidence of invasive GABHS infections has risen dramatically, increasing by a factor of 3 over the past year. In 1993, 50% of new cases of invasive GABHS disease were associated with VZV infection. Invasive GABHS should be considered in children with VZV who manifest fever on or beyond the fourth day of the exanthem. The absence of an elevated WBC and impetiginized or cellulitic lesions should not eliminate this diagnosis from consideration.

Objective : To determine the efficacy of a murine anti-tumor necrosis factor (TNF) monoclonal antibody in the treatment of Escherichia coli peritonitis and sepsis in the rabbit. Design : Prospective, paired, randomized, blinded, controlled animal trial. Setting : Animal research laboratory. Subjects : Male New Zealand white rabbits. Interventions : Anesthetized rabbits were cannulated with indwelling femoral arterial and venous catheters. Peritonitis and sepsis were induced by intraperitoneal challenge using live E. coli O18ac bacteria. All animals were treated with gentamicin and ceftriaxone 1 hr after challenge. One group (prophylaxis experiment) consisting often rabbit pairs (the prophylaxis group), was treated with either murine anti-TNF monoclonal antibody or an equivalent volume of 5% albumin 3 hrs before E. coli challenge. A second group (therapeutic experiment) of 17 rabbit pairs, the treatment group, was also treated with murine anti-TNF monoclonal antibody or albumin control 1 hr after E. coli challenge. Measurements and Main Results : All animals were bacteremic 1 hr after challenge. Physiologic measures of sepsis (heart rate, mean arterial pressure, serum bicarbonate, and arterial pH) did not differ between control, prophylaxis, and treatment groups. Peak serum TNF concentration was significantly (p

Study objective: To describe a population of patients who arrived in a pediatric emergency department with pulse and respirations but then sustained cardiopulmonary or respiratory arrest while in the ED. Design: Retrospective case series of patients from July 1987 to June 1993. Setting: Urban, tertiary care pediatric ED. Participants: All patients who sustained cardiopulmonary or respiratory arrest while in the ED. Results: Thirty-two cases of cardiopulmonary (n=18) or respiratory arrest (n=14) were identified, for an incidence of 1.2 arrests per 10,000 patient visits. Causes of arrest varied widely. Excluding those patients with do-not-resuscitate orders (n=2), an initial response to resuscitative efforts was obtained in 22 of 30 (73%) patients. Overall rates of survival to discharge from the ED and from the hospital were 21 of 30 (70%) and 15 of 30 (50%), respectively. Of those patients who survived to hospital discharge,12 of 15 (80%) were discharged at a baseline level of overall function. Conclusion: Cardiopulmonary or respiratory arrest in the pediatric emergency department is rare. The rate of survival of such an arrest is superior to that in outpatient arrests but inferior to that in inpatient arrests. [Teach SJ, Moore PE, Fleisher GR: Death and resuscitation in the pediatric emergency department. Ann Emerg Med June 1995;25:799-803.]

Objective: To assess the efficacy of the Yale Observation Scale (YOS) in detecting occult bacteremia in febrile, ambulatory pediatric patients with no apparent signs or symptoms of severe infection and with no focal infection. Design: YOS scores were assigned as part of a prospective, multicenter, randomized, interventional trial of oral and intramuscular antibiotics in preventing the complications of occult bacteremia in febrile children. Setting: Pediatric emergency departments at eight urban medical centers. Participants: Children, 3 to 36 months of age with a temperature at least 39.0° C, a nonfocal, non-toxic-appearing illness (or uncomplicated otitis media), treated as outpatients. Interventions: None. Results: There were 6611 assessable patients, who had both a blood culture result and a YOS score assigned. The median YOS score for both patients with bacteremia (n = 192) and patients without bacteremia (n = 6419) was 6, but the mean rank among patients with bacteremia was significantly higher ( p Conclusions: The YOS scores are higher among patients with bacteremia than among patients without bacteremia, but the difference is not clinically useful in detecting occult bacteremia in febrile children, with nonfocal, apparently nontoxic infection, treated as outpatients in this age group. (J PEDIATR 1995; 126:877-81)

In order to evaluate the impact of prehospital intravenous fluid therapy on the outcome of pediatric trauma patients and to evaluate the effect of such therapy on the on-scene interval, we performed a retrospective chart review of 50 pediatric trauma patients less than 18 years old transported directly from the field by Emergency Medical Services personnel with an intravenous catheter in place and admitted to the Trauma Service of a level I urban pediatric trauma center. As judged by an expert panel using a new grading system, prehospital intravenous fluid therapy was inconsequential to outcome in 47 of 50 patients, possibly beneficial in two of 50 patients, and possibly detrimental in one of 50 patients. Patients who received their catheters at the scene had significantly longer on-scene intervals than those who received them in the ambulance (15.4 vs 11.4 minutes, P < 0.05). The mean volume of fluid administered was 4.4 ml/kg body weight. Placement of the catheter (at the scene vs in the ambulance) and prehospital fluid volume administered were independent of the Injury Severity Score. The role of prehospital fluid therapy in pediatric trauma patients in an urban setting requires reevaluation.

A recombinant endotoxin neutralizing protein was evaluated for its ability to ameliorate the effects of Escherichia coli sepsis in rats.Prospective, controlled animal trial.Hospital animal research laboratory.Wistar rats, treated with gentamicin 1 hr after challenge with intraperitoneal E. coli O18ac.The animals received a recombinant endotoxin neutralizing protein, in doses of 5, 25, or 50 mg/kg, either 30 or 60 mins after challenge; controls received saline.Geometric mean serum endotoxin concentrations in endotoxin neutralizing protein-treated animals did not differ from control animals. Tumor necrosis factor concentrations in animals treated with endotoxin neutralizing protein 30 mins after challenge were significantly lower than controls. Animals treated with 25 or 50 mg/kg of endotoxin neutralizing protein 30 mins after E. coli challenge had significant improvements in survival compared with controls. Animals treated with 50 mg/kg of endotoxin neutralizing protein 60 mins after E. coli challenge had significant improvements in survival compared with controls.Endotoxin neutralizing protein significantly reduces mortality from Gram-negative sepsis in an antibiotic-treatment model of E. coli peritonitis and bacteremia in rats, mediated by a neutralization of the biological effects of endotoxin.

Study objective: To evaluate the safety and efficacy of twodoses of oral transmucosal fentanyl citrate (OTFC) for premedication of children undergoing laceration repair. Design: Prospective, randomized, nonblinded study. Setting: Urban pediatric emergency department. Participants: Thirty children aged 2 to 8 years requiring laceration repair. Interventions: Premedication with either 10 to 15 μg/kg or 15to 20 μg/kg of OTFC. Results: Activity score, vital signs, oxygen saturation, and painscores were recorded before and after administration of OTFC. Activity scores decreased significantly 15 to 60 minutes after OTFC. The physician suturing the wound rated the child's sedation/pain control as excellent or good in 83% of patients. Vital signs changes were not clinically remarkable. Oxygen saturations remained at 95% or more except in one child who experienced a transient decrease to 91%. Adverse effects were not serious but included vomiting in 20% of the lower-dose group and 47% of the higher-dose group. There were no significant differences between dose groups for activity or pain score changes, physician assessment, discharge times, or adverse events. Conclusion: Both doses of OTFC reduced activity with comparableefficacy, with no serious vital signs changes. However, the higher-dose group had a greater number ( P =NS) of adverse effects.

Study objective: To compare intranasal sufentanil and midazolam (IN-SM) with intramuscular meperidine, promethazine, and chlorpromazine (IM-MPC) for sedation in children. Design: Single-blind, randomized, controlled study. Setting: Urban children's emergency department. Participants: A convenience sample of children aged 1 to 4 years requiring suturing. Interventions: IN-SM or IM-MPC. Results: Vital signs, O 2 saturation, and anxiety and pain scores were recorded. A 6-point scale was used to assess response to medication, and a 12-point recovery score was used to determine readiness for discharge. Both groups were similar in age and sex distribution. There were no significant adverse effects in either group. Patients tolerated the IN regimen better than the IM regimen. Behavioral scores were lower during repair than at baseline within each group; however, they were not different between groups. Time to discharge was longer and recovery scores were lower (worse) among the IM-MPC group. Conclusion: IN-SM is as effective as IM-MPC for sedation in children. [Bates BA, Schutzman SA, Fleisher GR: A comparison of intranasal sufentanil and midazolam to intramuscular meperidine, promethazine, and chlorpromazine for conscious sedation in children. Ann Emerg Med October 1994;24:646-651.]

A recombinant endotoxin-neutralizing protein (ENP) from Limulus polyphemus and a monoclonal IgM anti-lipid A antibody (HA-1A) were compared in a rat model of Escherichia coli sepsis. One hour after intraperitoneal challenge with 10(6) cfu of E. coli O18ac K1, animals were sensitized to endotoxin with lead acetate and treated with ENP, HA-1A, or saline, followed by ceftriaxone and gentamicin. Before treatment, 95% of rats had high-grade bacteremia and high serum endotoxin concentrations, which were similar in all treatment groups (P > .60). One hour after treatment, there was no bacterial growth in any blood sample, and endotoxin concentrations were significantly lower in the ENP group than in the HA-1A and saline groups (P < .01). At 24 h after challenge, survival in the ENP group was significantly higher than in the HA-1A saline group (P < .001). ENP improved survival in a rat model of E. coli sepsis with high mortality despite effective antibiotic therapy.

Because studies of the treatment of children with occult bacteremia have yielded conflicting results, we compared ceftriaxone with amoxicillin for therapy. Inclusion criteria were age 3 to 36 months, temperature ≥39° C, an acute febrile illness with no focal findings or with otitis media (6/10 centers), and culture of blood. Subjects were randomly assigned to receive either ceftriaxone, 50 mg/kg intramuscularly, or amoxicillin, 20 mg/kg/dose orally for six doses. Of 6733 patients enrolled, 195 had bacteremia and 192 were evaluable: 164 Streptococcus pneumoniae , 9 Haemophilus influenzae type b, 7 Salmonella , 2 Neisseria meningitidis , and 10 other. After treatment, three patients receiving amoxicillin had the same organism isolated from their blood (two H. influenzae type b, one Salmonella ) and two from the spinal fluid (two H. influenzae type b), compared with none given ceftriaxone. Probable or definite infections occurred in three children treated with ceftriaxone and six given amoxicillin (adjusted odds ratio 0.43, 95% confidence interval 0.08,to 1.82, p =0.31). The five children with definite bacterial infections (three meningitls, one pneumonia, one sepsis) recelved amoxicillin (adjusted odds ratio 0.00, 95% confidence interval 0.00 to 0.52, p =0.02). Fever persisted less often with ceftriaxone (adjusted odds ratio 0.52, 95% confidence interval 0.28 to 0.94, p =0.04). Although the difference in total infections was not significant, ceftriaxone eradicated bacteremia, prevented significantly more definite focal bacterial complications, and was associated with less persistent fever.

Forty-three nonhemophiliac, confirmed HIV-positive children followed by the Children's Hospital AIDS Program made 184 visits to the children's Emergency Department (ED) during 1988 and 1989. The mean age was 30 +/- 28 months with a median of 25 months, a mode of 10 months, and a range from two days to 19 years. CD4 counts from within six months of the visit were available in 87% and were low enough to require Pneumocystis carinii pneumonia prophylaxis under current guidelines in 52%. Chief complaints included fever in 50%, respiratory symptoms in 21%, and gastrointestinal symptoms in 8%. The ED discharge diagnosis included fever/possible sepsis in 25%, pneumonia in 17%, otitis media in 9%, and upper respiratory tract infection or viral syndrome in 9%. Overall, an acute infection was identified at 62% of visits; of these, 33% were judged to be serious in nature. A total of 92 blood cultures were drawn, of which eight were positive with the following organisms: Streptococcus pneumoniae (3), Streptococcus faecalis (2), Escherichia coli (1), Torulopsis glabrata (1), and Staphylococcus non-aureus (1, a probable contaminant). Overall, 53% of patient encounters resulted in hospitalization. Patients with a white blood cell count more than 15,000/mm3 were more likely to be hospitalized (87 vs 62%, P < 0.01), though the white blood cell count was not helpful in identifying patients with bacteremia or serious infections. The mean temperature of patients admitted was higher than in those discharged (38.7 vs 37.9 degrees C, P < 0.01). In 1989, an estimated six per 1000 visits to our facility were by HIV-infected children.

Computed tomography (CT) imaging plays an important role in the acute evaluation and management of children with head trauma. When routine quality improvement (QI) meetings with representatives from the Children’s Hospital radiology and emergency departments revealed disagreement regarding the utilization and appropriateness of CT in children presenting with head trauma, an interdepartmental QI team was formed to address this issue. Because formal criteria for obtaining CTs for head trauma were unavailable, internal institutional criteria were developed by consensus after literature review. Contrary to perceptions of some staff members, the majority (95%) of children who received CT met at least one of the established criteria over a one-year study period. There was little relationship between the presence of criteria and abnormal CT results, but decisions whether to admit patients to the hospital or to send them home were influenced by CT results. Follow-up studies suggested that patients who were discharged home with a normal CT or no CT had uniformly good outcomes.

To assess the consultation patterns of pediatric emergency physicians in the management of injured children and to describe the spectrum of pediatric trauma, we retrospectively reviewed 601 patients treated in the emergency department for injuries during four one-week periods at a designated level I regional pediatric trauma center (50,000 patients/year) with a pediatric emergency medicine fellowship. The majority (94%) of pediatric trauma was minor; only 2% of children had injuries severe enough to require direct transfer to the operating room. The highest volume of patients, the greatest number of consultations, and the majority of admissions to the operating room occurred between 4 PM and midnight. No patients went to the operating room on the night shift. Musculoskeletal injuries constituted the predominant category of pediatric trauma, and lacerations were the most common specific injury. One half of all procedures involved laceration repair, and one third involved splinting or casting. Four hundred ten patients (68%) were managed by physicians in the emergency department without consultation. The orthopedic service performed one half of all consultations and admitted the largest number of patients; the majority of patients taken directly to the operating room had musculoskeletal injuries.(ABSTRACT TRUNCATED AT 250 WORDS)

Knowledge of the range of pediatric illness presenting to a general emergency department (ED) is needed to optimize the quality of care delivered there. It was hypothesized that the pediatric population treated at a general ED exhibited a broad range of medical complaints, while differing significantly from children seen in a pediatric ED. General ED records from 1 week each season were reviewed, and patient age, chief complaint, diagnosis, time of arrival, season, and disposition were recorded. Data on 874 patients were analyzed and compared with pediatric ED data. General ED patient age affected chief complaint, diagnosis, and admission rate (9.5% ≤1 year admitted vs 2.6% >1 year, P < .001). General ED patients were older (7.9 vs 6.0 years, P < .001) and admitted less frequently (3.8% vs 11%, P 001). Admission rates varied by arrival time only at the general ED, where minor trauma was more common (41% vs 22%, P < .001). It is concluded that a wide range of pediatric illness is treated in a general ED, supporting the decision to have pediatric emergency physicians on staff, and that significant differences exist in the spectrum and frequency of pediatric illness seen in a general ED and pediatric ED.

Young infants with fever are at risk for serious bacterial infection, but no consensus exists on the optimal approach to diagnosis and treatment. Although the traditional recommendation is always to perform all sepsis tests, including lumbar puncture, and administer intravenous (IV) antibiotics until culture results are negative, recent studies suggest administering intramuscular (IM) ceftriaxone with outpatient follow-up or using laboratory and clinical data to exclude low-risk patients from hospitalization, further testing, and antibiotic treatment. A decision analysis model was used to evaluate six strategies for the diagnosis and treatment of infants aged 28 to 90 days with temperature ≥38.0°C. Data from the literature, data from a 1991 study of 503 febrile infants, and direct, short-term costs from the Children's Hospital of Philadelphia were used as model inputs. The model was run for a hypothetical cohort of 100 000 febrile infants who did not require admission for focal infection or for other reasons that clearly necessitated admission. The model included six strategies: (1) no intervention; (2) all sepsis tests (lumbar puncture, blood culture, urine culture, white blood cell count, and urinalysis) followed by hospitalization and IV antibiotics for all infants; (3) all sepsis tests followed by IM ceftriaxone and outpatient management for most infants; (4) blood and urine cultures with white blood cell count and urinalysis followed by either lumbar puncture and IV antibiotics for high-risk infants or outpatient management without antibiotics for low-risk infants; (5) white blood cell count and urinalysis followed by either lumbar puncture, blood and urine cultures, and IV antibiotics for high-risk infants or outpatient management without antibiotics for low risk infants; and (6) clinical judgment followed by either all sepsis tests and IV antibiotics for high-risk infants or outpatient management without antibiotics for low-risk infants. The two "all sepsis tests" strategies prevented the most cases of death or neurologic impairment, 78% (when IV antibiotics were used) and 76% (when IM ceftriaxone was used) of all potential cases. The most cost-effective strategy was to use all sepsis tests followed by IM ceftriaxone for all patients without meningitis, at an incremental cost of only $3900 per sequela prevented relative to no intervention. Strategies under which only those patients selected as high-risk by laboratory criteria received antibiotic treatment were less effective but incurred lower rates of antibiotic complications. Clinical judgment alone was the least clinically effective and the second least cost-effective strategy. The model's results depended most on assumptions about the effectiveness of IM ceftriaxone, the sensitivity of the white blood cell count, and the sensitivity of clinical judgment in identifying young infants with serious bacterial infection. Combining all sepsis tests with IM ceftriaxone has superior clinical and cost-effectiveness compared with other strategies for managing febrile infants in this model. Strategies that use selective antibiotic treatment based on laboratory tests or clinical judgment are acceptable when the sensitivity of the criterion used is high, but they do not surpass strategies that combine all sepsis tests and antibiotic treatment until the criterion's sensitivity is greater than 96%.

Limulus antilipopolysaccharide factor (LALF), an 11.8-kDa peptide isolated from amebocytes of Limulus polyphemus, neutralizes meningococcal lipooligosaccharide (LOS)-induced gelation of limulus amebocyte lysate. Rabbits challenged with an LD90 of LOS (10 micrograms/kg) premixed with LALF in vitro (n = 10) had significantly higher mean arterial pressure, arterial pH, serum bicarbonate concentrations, and survival (90% vs. 8%, P = .005) than did rabbits challenged with LOS alone. Relative to untreated controls, rabbits pretreated with LALF intravenously (iv) at 1.2 mg/kg (n = 21) also had significant improvements in physiologic measurements and higher survival (52% vs. 8%, P = .003). Even when LALF (1.2 mg/kg iv) was given 1/2 h after LOS challenge, animals showed significant improvements in physiologic measurements and survival (33% vs. 8% in untreated controls P = .028). LALF-treated animals also had significantly lower circulating endotoxin activity and tumor necrosis factor concentrations. Thus, LALF attenuates the toxic effects of meningococcal LOS in rabbits even when administered after LOS challenge and deserves further evaluation as a potential therapeutic agent for treating gram-negative septic shock.

Study objective : To determine the outcome of outpatient treatment of febrile infants 28 to 89 days of age with intramuscular administration of ceftriaxone. Design Prospective consecutive cohort study. Setting Urban emergency department. Patients Five hundred three infants 28 to 89 days of age with temperatures ≥38° C who did not appear ill, had no source of fever detected on physical examination, had a peripheral leukocyte count 9 cells/L, had a cerebrospinal fluid leukocyte count 6 /L, did not have measurable urinary leukocyte esterase, and had a caretaker available by telephone. Follow-up was obtained for all but one patient (99.8%). Intervention After blood, urine, and cerebrospinal fluid cultures had been obtained, the infants received 50 mg/kg intramuscularly administered ceftriaxone and were discharged home. The infants returned for evaluation and further intramuscular administration of ceftriaxone 24 hours laters; telephone follow-up was conducted 2 and 7 days later. Results Twenty-seven patients (5.4%) had a serious bacterial infection identified during follow-up; 476 (94.6%) did not. Of the 27 infants with serious bacterial infections, 9 (1.8%) had bacteremia (8 of these had occult bacteremia and 1 had bacteremia with a urinary tract infection), 8 (1.6%) had urinary tract infections without bacteremia, and 10 (2.0%) had bacterial gastroenteritis without bacteremia. Clinical screening criteria did not enable discrimination between infants with and those without serious bacterial infections. All infants with serious bacterial infections received an appropriate course of antimicrobial therapy and were well at follow-up. One infant had osteomyelitis diagnosed 1 week after entry into the study, received an appropriate course of intravenous antimicrobial therapy, and recovered fully. Conclusions After a full evaluation for sepsis, outpatient treatment of febrile infants with intramuscular administration of ceftriaxone pending culture results and adherence to a strict follow-up protocol is a successful alternative to hospital admission.

Study objectives: To determine whether potentially life-threatening intra-abdominal injuries occur in the absence of multisystem trauma in children, and to determine the usefulness of physical examination and a pediatric triage score in the assessment of liver and spleen injuries in children. Design: Retrospective study. Setting: Admissions to a children's hospital from October 1982 through September 1989 who were found to have liver or spleen injuries. Seventy-seven patients were identified; 55 were male, 22 were female. Mean age was 9 years and 3 months (± 56 months) with a range of 22 months to 20 years, 4 months. Clinical signs were recorded and a Pediatric Trauma Score (PTS) and Injury Severity Score (ISS) were calculated for each patient. Measurements and main results: Fifty-four patients (70%) had a spleen injury, 18 (23%) had a liver injury, and five (7%) had both liver and spleen injuries. Patients were managed without surgery in 63 of 77 cases (82%); two died. Fifty-one of 77 patients (66%) received an ISS of 18 or less; 26 patients (34%) received a score of more than 18. Forty-four of the 51 patients (86%) with an ISS of 18 or less had a normal pulse (120 or less); 48 of the 51 (94%) had a normal systolic blood pressure (90 mm Hg or more). A strong negative correlation ( r = −0.80; P = .001) was found between the two scores for children with multiple severe injuries (ISS of more than 18); there was poor correlation ( r = −0.04; P > .05) between the two scores for isolated liver or spleen injury (ISS of 18 or less). Conclusions: We conclude that liver or spleen damage may be present in children without other injuries and must be considered with a high index of suspicion, and that neither the initial clinical findings nor the PTS reliably predicts liver or spleen injuries in children with focal abdominal injuries.

This study was designed to quantify more precisely the accuracy of magnitude of rectal temperature and total white blood cell (WBC) count as indicators of bacteremia in children with an obvious focal bacterial infection. A total of 955 children, aged 3 to 36 months, who had rectal temperature ≥39.0°C and were seeking care at either of two urban pediatric emergency departments had blood drawn for culture; 885 had blood drawn for WBC count. Twenty-seven had bacteremia. Various combinations of temperature and WBC count were selected to construct receiver-operating-characteristic curves by plotting sensitivity vs false-positive rate (1 - specificity). The receiver-operating-characteristic curve of WBC count provided significantly better diagnostic information than the curve for temperature increments above 39.0°C. Each increment of 0.5°C led to large decrements in sensitivity and false-positive rates. At a WBC count cutoff of 10 000/mm3, the sensitivity was 92% while the false-positive rate was 57%. Using this cutoff point, the clinician could have avoided performing 368 of 955 blood cultures and missed only 2 of 26 children with bacteremia. Receiver-operating-characteristic curves combining WBC count and temperature increments above 39.0°C provided no better diagnostic information than that of WBC count at a temperature cutoff of 39.0°C. It is concluded that increments in temperature above 39.0°C provided additional diagnostic specificity for bacteremia only at the expense of unacceptable decreases in sensitivity. Total WBC count provided better information. A WBC count cutoff of 10 000/mm3 increased specificity with minimal decrease in sensitivity. Receiver-operating-characteristic curve analysis allows selection of cutoff criteria by individual practitioners based on the prevalence of bacteremia in their communities and on the perceived risks of bacteremia.

Knowledge of the spectrum and relative frequencies of pediatric emergencies is an important factor in developing appropriate training curricula for physicians treating children in emergency departments. To provide these data, we reviewed the records for four one-week periods (January, April, July, and October) of a large pediatric emergency department to describe the population in terms of age, chief complaints, diagnoses, time of arrival, seasonal variation, and disposition. There were 3796 log entries. Complete information on all variables was obtained on 3784 patients. Age ranged from one day to 39 years, and the mean age was 6.0 +/- 6.15 years. One half of all emergency department visits were by children three years old or younger. On the other hand, 12% of visits were by adolescents (ages 13 to 18), and one in 25 visits was made by an adult (greater than 18 years old). The majority of chief complaints and final diagnoses were related to infection and trauma. More than half of the patients arrived on the evening shift, between 4 pm and 12 am. Eleven percent of the children seen on day and evening shifts and 13% from the night shift were admitted. From the analysis of our data we recommend expanded skills in the management of minor trauma for pediatric residents, an emphasis on management of infections for nonpediatric emergency specialists, and extensive training in both pediatric and adult trauma for physicians in pediatric emergency medicine fellowships.

Occult bacteremia, which precedes many serious infections in children, is most often due Streptococcus pneumoniae, Haemophilus influenzae, Neisseria meningitidis, or Salmonella species. Diagnosis on the basis of clinical judgment is unreliable, although the presence of certain risk factors may suggest the diagnosis. These risk factors include an age of 3 months to 3 years, a temperature of greater than or equal to 39.0 degrees C, and a white blood cell count of greater than or equal to 15,000/mm3. Although results are delayed, a culture of blood is the only definitive test. Studies suggest that treatment with various antibiotics may be helpful, but that some drugs, particularly orally administered amoxicillin, should not be relied on to eliminate occult bacteremia or prevent its most serious sequela, meningitis.

To investigate the occurrence and outcome of bacteremia associated with otitis media, charts were reviewed from patients who were 3 to 36 months of age, had temperatures greater than or equal to 39 degrees C, and were diagnosed with isolated clinical otitis media. A total of 2982 patients were identified. Blood cultures were obtained from 1666 (56%). Of the 1666 patients, who had blood drawn for cultures, 50 (3.0%) had bacteremia. These included 39 with Streptococcus pneumoniae, 4 with Haemophilus influenzae, 2 with Neisseria meningitidis, 3 with Salmonella species, and 2 with Staphylococcus aureus. The incidence of bacteremia increased at higher temperatures, being 1.9% at temperatures less than or equal to 40 degrees C and 5.0% at temperatures greater than 40 degrees C. Younger children were more likely to have bacteremia; 3.7% less than or equal to 12 months of age, 2.4% 13 to 24 months of age, and 1.9% 25 to 36 months of age had blood culture results that were positive (not significant). Reevaluation of the 50 bacteremic patients showed that 9 patients had continued fever, 3 patients had persistent bacteremia, pneumonia developed in 1 patient, and meningitis developed in 1 patient. It was concluded that (1) 3% of young febrile children with otitis media have bacteremia at the time of evaluation, a rate comparable to that previously reported in children with no focus of infection; (2) the incidence of bacteremia increases at higher temperatures; and (3) most febrile children with otitis media do well. The clinician must therefore weigh the potential benefit of drawing a blood culture to identify children at risk for complications against the inherent cost, inconvenience, and discomfort.

Decision analysis was used to evaluate the probable health benefits, complications, and costs of six management strategies for febrile children at risk for occult bacteremia. The strategy that combined blood culture with empiric oral antibiotic treatment for all patients was predicted to prevent the highest number of major infections and to have the lowest cost per major infection prevented. The stragegy that combined a leukocyte count and blood culture for all patients, followed by empiric antibiotic treatment for those with leukocyte count ≥10,000/mm 3 , had almost equal cost and clinical effectiveness and avoided many antibiotic complications. Culture of blood specimens from all patients and no empiric treatment constituted the third most clinically effective intervention but was the least cost-effective in this model. Giving a 2-day oral course of amoxicillin without testing had the lowest average cost per febrile patient but was the least clinically effective intervention. However, the low degree of effectiveness of empiric treatment alone was based on the assumption that oral amoxicillin therapy was only 20% effective in preventing major infections after bacteremia. At higher estimates of effectiveness, treatment alone became a more viable strategy. We conclude that approaches which combine blood culture with empiric antibiotic treatment are the most clinically effective and the most cost-effective strategies for children at risk for occult bacteremia.

Arterial catheters are used frequently to monitor critically ill children in ICUs, but their role in the emergency department has not been described. The medical records of all patients admitted to the ICU from the ED of a children's hospital during a 30-month period were reviewed retrospectively. Forty-one patients were identified as having arterial catheters placed in the ED. Catheterization was performed for serial blood gas determinations (59%) or continuous blood pressure monitoring (41%). The radial artery was catheterized in 73% of the 41 patients, and a 22-gauge catheter was used in 78% of the patients. The original catheter was discontinued before the end of medical indication in only 24% of cases — 17% were discontinued for failure to draw and 7% for vascular changes at the site of the catheter. We conclude that arterial access can be used for children in the ED to monitor blood pressure and that blood gas tensions and complications are few and generally minor.