Comparative practice patterns of emergency medicine physicians and pediatric emergency medicine physicians managing fever in young children.

BACKGROUND/OBJECTIVE: The management of fever in young children is a controversial topic. This study seeks to compare the management approaches between general emergency medicine physicians (GEMPs) and pediatric emergency medicine physicians (PEMPs) and correlate them to existing practice guidelines. DESIGN/METHODS: All charts of children age 3 to 36 months presenting with the complaint of fever at both a children's hospital emergency department (ED) and a general ED from June 1, 1998 to September 1, 1998; December 1, 1998 to April 1, 1999; and June 1, 1999 to September 1, 1999 were retrospectively reviewed. Fever was defined as >/=39 degrees C. Patients with a history of immunodeficiency, chronic illness, ventriculoperitoneal shunt, antibiotic use in the past 48 hours, or focal infection noted on examination were excluded. Data collected included focal exam findings, laboratory tests, diagnosis, treatment, and disposition. Variances from the practice guidelines were tabulated and compared. RESULTS: One thousand three hundred twenty-three eligible children met exclusion criteria and were seen by PEMPs; 755 were eliminated because of exclusion criteria (526 because of focal infection). Twenty-two (4%) of 568 remaining patients were admitted to the hospital. Two hundred twenty-eight eligible children were seen by GEMPs; 147 were excluded (109 because of focal infection). No patients were admitted to the hospital. PEMPs ordered more complete blood counts (324/568 vs 27/81), more blood cultures (321/568 vs 27/81), and more urine cultures (208/568 vs 20/81) than GEMPs. GEMPs ordered more chest radiographs and cerebrospinal fluid analyses than PEMPs; GEMPs ordered less complete blood counts, blood cultures, and urine cultures than PEMPs. GEMPs diagnosed more focal infections (109/228 vs 526/1323), and conflicted more often with the practice guidelines (66/79 vs 225/498) than PEMPs. Patients spent an average of 2.26 +/- 0.16 hours in the pediatric ED versus 3.0 hours +/- 0.18 hours in the general ED. CONCLUSIONS: Significant differences in the management of the young child with fever and no source exist between these two groups of physicians. These variations affect both cost and standard of care. Future studies assessing whether these strategies affect patient outcomes would further elucidate their clinical implication.

Predictors of bacteremia in febrile children 3 to 36 months of age.

PURPOSE: To develop an improved model for the prediction of bacteremia in young febrile children. METHODS: A retrospective review was performed on patients 3 to 36 months of age seen in a children's hospital emergency department between December 1995 and September 1997 who had a complete blood count and blood culture ordered as part of their regular care. Exclusion criteria included current use of antibiotics or any immunodeficient state. Clinical and laboratory parameters reviewed included age, gender, race, weight, temperature, presence of focal bacterial infection, white blood cell count (WBC), polymorphonuclear cell count (PMN), band count, and absolute neutrophil count (ANC). Logistic regression analyses were used to identify factors associated with bacteremia, defined as growth of a pathogen in a blood culture. The model that was developed was then validated on a second dataset consisting of febrile patients 3 to 36 months of age collected from a second children's hospital (validation set). RESULTS: There were 633 patients in the derivation set (46 bacteremic) and 9465 patients in the validation set (149 bacteremic). The mean age of patients in the derivation and validation sets were 15.8 months (95% confidence interval [CI]: 15.2-16.5) and 16.6 months (95% CI: 16.5-16.8), respectively; the mean temperatures were 39.1 degrees C (95% CI: 39. 0-39.2) and 39.8 degrees C (95% CI: 39.7-39.8); 56% were male in the derivation set and 55% male in the validation set. Predictors of bacteremia identified by logistic regression included ANC, WBC, PMN, temperature, and gender. Receiver operator characteristic (ROC) analysis showed similar performance of ANC and WBC as predictors of bacteremia. When placed into a multivariate logistic regression model, band count was not significantly associated with bacteremia. Information regarding focal infection was available for 572 patients in the derivation set. The percentage of patients diagnosed with bacteremia with a focal bacterial infection was not significantly different from the percentage who had bacteremia without a focal bacterial infection (16/200 vs 30/372). Based on this dataset, a logistic regression formula was developed that could be used to develop a unique risk value for each patient based on temperature, gender, and ANC. When the final model was applied to the validation set, the area under the ROC curve (AUC) constructed from these data indicated that the model retained good predictive value (AUC for the derivation vs validation data =.8348 vs 0.8221, respectively). CONCLUSIONS: Use of the formulas derived here allows the clinician to estimate a child's risk for bacteremia based on temperature, ANC, and gender. This approach offers a useful alternative to predictions based on fever and WBC alone.bacteremia, detection, white blood cell.

A common cause of altered mental status occurring at an uncommon age.

Wernicke's encephalopathy is a neurologic disorder due to a nutritional deficiency of thiamine, characterized by ocular palsies, ataxia, and altered mental activity. While Wernicke's encephalopathy is commonly attributed to alcoholism in the adult population, it has been described in children receiving prolonged parenteral nutrition and those with malignancies and AIDS. The disease, however, is rarely diagnosed in the pediatric population during life. We report a case of Wernicke's encephalopathy in a child with prolonged starvation and aim to improve awareness of a potentially fatal but treatable disease.

Accuracy of a polymerase chain reaction-based assay for detection of pneumococcal bacteremia in children.

OBJECTIVE: To evaluate the utility of a polymerase chain reaction (PCR)-based assay for identifying pneumococcal DNA in the blood of pediatric patients with suspected bacteremia. METHODS: Children evaluated at the Children's Hospital of Pittsburgh who were having blood drawn for culture had an additional 2 to 3 mL of blood (from the same sampling) obtained and placed in a sodium citrate tube for PCR processing (study group). The control group for this study consisted of children having blood drawn for biochemical analysis who were afebrile, well-appearing, and had no recent illnesses. Specimens were frozen at -70 degrees C and then batch-processed for PCR-based analyses with the JM201/202-204 primer/probe set. Amplified products were detected after liquid hybridization format wherein a 32P end-labeled probe was annealed to the amplified DNA and visualized by autoradiographic analysis after gel retardation. RESULTS: Four hundred eighty study group patients and 103 controls had specimens tested by both PCR and blood culture. Twenty-six (5%) patients had a positive blood culture for a pathogenic organism (21 of which were Streptococcus pneumoniae). Twelve (57%) of the 21 patients with blood cultures positive for S pneumoniae also were positive by PCR. In addition, 206 study group patients and 16 controls with negative blood cultures had positive PCR results. A greater proportion of study group patients were PCR-positive/culture-negative than were controls (206/459 vs 16/103). CONCLUSION: Although this assay currently lacks adequate sensitivity and specificity for clinical use, the high frequency of PCR-positive cases in patients with suspected bacteremia may indicate a greater role for S pneumoniae than had previously been appreciated. Further refinement of this assay as well as the development of a rapid PCR-based assay appears warranted.

A simple intervention for improving telephone contact of patients discharged from the emergency department.

OBJECTIVE: To determine if emergency department (ED) follow-up contact rates can be improved by confirming a best contact telephone number with the patient prior to discharge. DESIGN/SETTING: Prospective comparison of intervention and control groups taken from convenience samples of ED patients from Children's Hospital of Pittsburgh (CHP) and Children's Hospital of Wisconsin (CHW). PARTICIPANTS: One hundred eighty-eight (188) patients (138 from CHP and 50 from CHW) who had x-rays and laboratory studies done in the ED were interviewed by the investigators prior to discharge (intervention group) and 305 control patients (256 from CHP, 49 from CHW) identified from ED log books. INTERVENTION: Prior to discharging the patient from the ED, the investigators verified and/or corrected the best contact number for a follow-up phone call with each intervention patient. Within 24 hours of each visit, a follow-up call was made to each intervention and control patient during one of three time intervals spaced between 8 AM and 10 PM. RESULTS: A total of 29 patients, or 15.4%, of the intervention group, gave a telephone number that differed from the one listed in the patient's medical record. Of the CHP group, 93.5% (129/138) of intervention patients and 78.5% (201/256) of the control patients were successfully contacted (P < 0.001). Of the CHW group, 96% (48/50) of intervention patients, and 94% (46/49) of control patients were successfully contacted (P = NS). Successful contact of control patients was greater in CHW than CHP (46/49 vs 201/256, P = 0.02). CONCLUSIONS: A significant proportion of telephone numbers listed in the ED medical records are incorrect, but the frequency of inaccuracy may be institution-dependent. Confirming the patient's "best contact" number can significantly increase the successful contact of ED patients.

Child restraint device use in patients leaving a children's hospital.

OBJECTIVE: To determine the frequency of and factors associated with the use of child restraint devices (CRDs) in patients leaving an urban children's hospital. DESIGN: Verbally administered survey, followed by observation of CRD use. SETTING: Main parking area at an urban children's hospital with mandated use of parking attendants. PARTICIPANTS: A convenience sample of 295 parents or guardians who were leaving the hospital and were accompanied by at least 1 child younger than 48 months. RESULTS: Eighty percent of the respondents were female; 77% of the respondents were white and 22% were African American. Two hundred thirty-five (80%) used CRDs. Subjects who were at risk of CRD noncompliance included nonwhite individuals (adjusted odds ratio [OR], 6.6; 95% confidence interval [CI], 3.0-14.5), those whose primary source of medical care was from a public clinic (OR, 2.4; 95% CI, 1.1-5.3) or from a specialty clinic (OR, 2.4; 95% CI, 0.8-6.6), and those who were older than 24 months (OR, 8.7; 95% CI, 3.5-21.9). Parental education and income level were not important predictors of CRD compliance controlling for race, primary medical care source, and age. Study participants were observed for incorrect CRD use; 30% (30/ 99 of infants younger than 12 months were incorrectly placed in the forward-facing position, and 23% (54/ 235 of all CRD users did not use its harness. CONCLUSIONS: Patients who use public clinics or pediatric subspecialists as their primary source of medical care, especially those who are African Americans, are at risk of CRD noncompliance. Children's hospitals should take an active role in improving CRD use in these patients.

Effect of a research nurse on patient enrollment in a clinical study.

OBJECTIVES: To assess the effect of adding a shared research nurse on patient enrollment in an ongoing clinical study. DESIGN: Before/after comparison of recruitment rates and consent rates involving patients eligible for a study of blood culture collection strategies for methods on the detection of bacteremia and determination of inter-rater agreement regarding study eligibility between the research nurse and primary investigator. SETTING: An urban pediatric emergency department (ED). PARTICIPANTS: Patients presenting to the ED between February 1, 1991, and February 1, 1993, with suspected bacteremia. INTERVENTION: Addition of a research nurse to the study group to identify, educate, and enroll eligible study patients. MAIN OUTCOME MEASURE: Patient enrollment rates. RESULTS: Successful enrollment of eligible study patients was 14% (40/281) during the year prior to the addition of the research nurse and 50% (126/253) after the addition (P < 0.001). Refusal rates were 21% (22/106) when recruitment was carried out by the nurse and 15% (13/86) when done by physicians (P = 0.41). Review of records to identify eligible study patients showed agreement between the study nurse and physician on 78/80 (97.5%) of the charts reviewed (kappa = 0.94). CONCLUSIONS: Addition of a shared research nurse increased recruitment of patients and demonstrated equivalent judgment of the research nurse and physician investigators regarding patient eligibility for the study. This arrangement demonstrates a practical and effective means of increasing the efficiency of ED-based clinical research.

Effect of number of blood cultures and volume of blood on detection of bacteremia in children.

OBJECTIVE: To determine whether bacteremia can be detected more rapidly and completely by (1) obtaining two blood cultures instead of one and/or (2) collecting a larger volume of blood. STUDY DESIGN: Prospective comparison of different strategies in 300 patients undergoing blood culture for suspected bacteremia. Each patient had two samples of blood, A (2 ml) and B (9.5 ml), obtained sequentially from separate sites. The B sample was divided into three aliquots: B1 (2 ml), B2 (6 ml), and ISO (1.5 ml, quantitative culture). RESULTS: A pathogen was isolated from one or more blood cultures in 30 patients (10% of cases). When measured at 24 hours, the pathogen recovery rate for the B2 sample (72%) was higher than that for the individual small-volume samples (A = 37%, B1 = 33%; p < 0.01 for each comparison) and for the combination of the two small-volume samples (A + B1 = 47%; p = 0.04). At final (7-day) reading the pathogen recovery rate for the B2 sample (83%) was higher than that for B1 (60%; p = 0.02) and similar to the recovery rate observed with the combination of the two small-volume cultures (A + B1 = 73%; p = 0.55). CONCLUSIONS: Increasing the volume of blood inoculated into blood culture bottles improves the timely detection of bacteremia in pediatric patients and spares the patients the cost and pain of an additional venipuncture.

Identification of a patient with Streptococcus pneumoniae bacteremia and meningitis by the polymerase chain reaction (PCR).

A polymerase chain reaction (PCR) assay based on the penicillin-binding protein gene PBP2B identified the presence of DNA specific for Streptococcus pneumoniae in the serum and CSF of a patient with culture-proven bacteremia and meningitis. Positive signals were seen to dilutions of 1:125 and 1:390,625 for the blood and CSF specimens, respectively. Potential advantages of PCR over conventional culture include exquisite sensitivity, faster results and the ability to identify the organisms by the presence of species-specific DNA even in patients pretreated with antibiotics.

Detection of Streptococcus pneumoniae in whole blood by PCR.

Streptococcus pneumoniae is a major cause of bacteremia in both children and adults. Currently, the diagnosis of pneumococcal bacteremia relies on the isolation and identification of the bacteria from blood cultures. We have developed a sensitive assay for the detection of S. pneumoniae in whole blood by the PCR. A specific primer-probe set (JM201 and JM202 primers with JM204 probe) designed from the penicillin-binding protein 2B gene was demonstrated to reproducibly detect between 10 and 100 fg of input purified S. pneumoniae DNA. This assay system was shown to be inclusive for all strains of S. pneumoniae evaluated, including 15 different serotypes and a battery of penicillin-resistant and -sensitive strains. The specificity of this PCR-based assay was demonstrated by its inability to support amplification from a series of human, bacterial, and yeast genomic DNAs. A general specimen preparation method which should be suitable for the purification of DNA from any pathogens in whole blood was developed. With this protocol it was possible to detect S. pneumoniae-specific DNA from whole blood specimens inoculated with as little as 4 CFU/ml. Copurified human blood DNA, ranging from 0 to 4.5 micrograms per PCR, did not affect the sensitivity of S. pneumoniae detection by PCR. A blinded clinical trial was used to compare the PCR-based assay with standard microbiological blood culture for the detection of S. pneumoniae bacteremia in 36 specimens obtained from pediatric patients seen in the emergency room of Children's Hospital of Pittsburgh. With culture as the "gold standard," the PCR-based assay had a sensitivity of 80% (4 of 5 culture-positive specimens were PCR positive) and a specificity of 84% (26 of 31 culture-negative specimens were PCR negative). However, three patients whose specimens were PCR positive and culture negative had histories suggestive of bacteremia, including recent positive blood cultures, treatment with antibiotics, cellulitis, and multiple emergency room visits for fever within a 24-h period. These data suggest that PCR-based assays for S. pneumoniae may prove useful to augment current methods of detection for S. pneumoniae bacteremia.

Utility of routine laboratory testing for detecting intra-abdominal injury in the pediatric trauma patient.

OBJECTIVE: To assess the prevalence of laboratory abnormalities (complete blood cell count, electrolytes, blood urea nitrogen, creatinine, glucose, aspartate aminotransferase, alanine aminotransferase, amylase, lipase, urinalysis [U/A]) and the sensitivity and specificity of the physical examination (PE) and screening laboratory tests for identifying intra-abdominal injury (IAI) in moderately injured pediatric patients. DESIGN, PARTICIPANTS, AND SETTING: Phase I: Retrospective chart review of 285 consecutive level II (moderately injured) trauma patients seen at a children's hospital emergency department/pediatric trauma center. All patients were received directly from the scene and had the following data recorded: mechanism of injury, Glasgow coma score, trauma score, pediatric trauma score, systematically recorded PE findings, laboratory results, and injuries detected during hospitalization. Phase II: To confirm the sensitivity of the PE and U/A found in phase I, the model was applied to 91 additional trauma patients identified by International Classification of Diseases, 9th revision (ICD-9) codes as having IAI. INTERVENTION: None. RESULTS: Phase I: A total of 3939 tests were ordered for the 285 patients entered in phase I. Aspartate aminotransferase and alanine aminotransferase values were obtained in 59% of patients; glucose level was obtained in 78% of patients; complete blood cell count, U/A, and levels of electrolytes, blood urea nitrogen, creatinine, amylase, and lipase were obtained in more than 85% of patients. The overall prevalence of laboratory abnormalities was 5.7%. Fourteen patients (4.8%) were identified who had a total of 23 significant IAIs (9 pancreatic, 6 splenic, 5 renal, 3 hepatic). The PE combined with U/A showing more than five red blood cells per high-power field had a sensitivity of 100%, specificity of 64%, positive predictive value of 13%, and negative predictive value of 100% for the detection of IAI. The presence of laboratory abnormalities suggesting injury did not increase the sensitivity of the model and significantly decreased both specificity and positive predictive value. Phase II: The PE combined with U/A identified an abnormality in 89 (97.8%) of 91 cases (95% confidence interval = 94.8% to 100%). CONCLUSIONS: In the moderately injured pediatric trauma patient, (1) there is a low prevalence of laboratory abnormalities; (2) the PE combined with U/A is a highly sensitive screen for IAI; and (3) in patients with a normal PE of the abdomen and a normal U/A, laboratory testing seldom identifies unsuspected IAI.

Pediatric emergency medicine: state of the art.

To assess the current practice of pediatric emergency care in the United States, a questionnaire was mailed to the directors of all 240 emergency departments (EDs) affiliated with pediatric residency training programs in the United States. One hundred seventy-two programs (72%) returned completed surveys, which comprised 32 questions highlighting staffing patterns, ancillary services, clinical issues, and resident education. The mean annual ED census was 39,290; the mean number of visits for children 0 to 18 years of age was 17,473. Seven percent of pediatric visits were categorized as critical, 23% as urgent, and 70% as nonurgent. Eleven percent of patients were admitted to the hospital. During peak periods, patients whose visits were triaged as nonurgent waited an average of 1.5 hours to be seen by a physician. Twenty-eight percent of programs provided 24-hour on-site coverage by a pediatric attending physician or fellow. Of the remaining programs, the average daily on-site pediatric coverage was 8.6 +/- 6.2 hours. Eighteen percent of programs used physician assistants or nurse practitioners in the ED. During their first, second, and third years of training, pediatric residents spent an average of 5.2, 5.8, and 3.5 weeks in the ED, respectively. The majority of EDs handled all levels of pediatric trauma (84%), had dedicated trauma teams (73%), employed social workers specifically assigned to the ED (62%), and had child abuse teams (72%). Ninety-one percent of EDs had radio communications with prehospital care vehicles and 67% provided medical command for incoming pediatric patients.(ABSTRACT TRUNCATED AT 250 WORDS)

Reliability of observation variables in distinguishing infectious outcome of febrile young infants.

We prospectively evaluated 7 observation variables (level of activity, level of alertness, respiratory status/effort, peripheral perfusion, muscle tone, affect, feeding pattern) which qualify patient clinical appearance in order to determine reliability in distinguishing the infectious outcome of 233 febrile infants ages 0 to 8 weeks. Each variable was graded either 1, 3, or 5, with a higher score indicative of a greater degree of compromise. All infants received physical examination and sepsis evaluation (lumbar puncture, complete blood count/blood culture, urinalysis/urine culture). The 3 outcome groups compared were 29 cases of serious bacterial infections, (+SBI; 10 with bacterial meningitis, 12 with bacteremia, 7 with urinary tract infection), 45 cases of aseptic meningitis (AM) and 159 cases culture-negative with normal cerebrospinal fluid (CN-NCSF). The mean score for each of the 7 variables was significantly greater in the +SBI group compared with both the AM and CN-NCSF groups (P < 0.05), whereas there was no significant difference in mean score for each of the 7 variables between the AM and CN-NCSF groups. Stepwise discriminant analysis identified 3 variables that best distinguished outcome: affect; respiratory status/effort; and peripheral perfusion, which constituted the Young Infant Observation Scale. The mean total Young Infant Observation Scale score generated from assessing these 3 variables was significantly greater (P = 0.0001) in the +SBI, group (9) compared with both the AM (5) and CN-NCSF (5) groups. A total Young Infant Observation Scale score > or = 7 had a sensitivity of 76%, specificity of 75% and negative-predictive value of 96% for outcome of +SBI.

Standardized instructions: do they improve communication of discharge information from the emergency department?

To determine whether standardized instructions enhance communication of discharge information, we provided 197 parents of children in whom otitis media was diagnosed with one of three types of instruction at the time of discharge from a pediatric emergency department: (1) instruction by individual housestaff and medical students after consultation with an attending physician (control group); (2) standardized verbal instructions given by housestaff and students trained in their use (verbal group); or (3) the same instructions given to the verbal group, together with a type-written copy of the information to take home (verbal + written group). Prior to leaving the emergency department and, again, by phone, 1 and 3 days later, parents were questioned concerning the prescribed medication's name, dose, frequency, and duration of administration (medication data), three signs of improvement, and eight signs indicating the need for medical advice (worrisome signs). The mean percentage of correct responses per parent in each group was computed for each information category. Both at exit interview and at follow-up, parents receiving either form of standardized instructions showed significantly greater knowledge of information related to their child's illness than did controls. Information regarding medication data was more likely to be communicated to parents in all groups than were signs of improvement or worrisome signs. The addition of written instructions to standardized verbal instructions did not improve parental recall of discharge information.

The wheezer that wasn't.

The expression of cardiac dysfunction in pediatric patients with myocarditis may not be conspicuous. While older children with myocarditis may abruptly present with pleuritic or angina-like pain, infants and toddlers with fulminant disease are unable to verbalize such complaints. Cardiac compromise in preverbal children may only be inferred from variable examination findings that include gallop rhythm, tachycardia, malignant dysrhythmias, murmur, rub, and signs of congestive heart failure. The emergency physician is likely to overlook a cardiac origin for wheezing in a child with a past medical history of asthma. Therapeutic modalities chosen for reactive airway disease may adversely influence the outcome of a patient with myocarditis.