PEDIATRICS Vol. 108 No. 4 October 2001, pp. 866-871

Enhanced Urinalysis Improves Identification of Febrile Infants Ages 60 Days and Younger at Low Risk for Serious Bacterial Illness

Sandra M. Herr, MD, Ellen R. Wald, MD, Raymond D. Pitetti, MD, MPH, and Sylvia S. Choi, MD

From the Department of Pediatrics, Children's Hospital of Pittsburgh, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania.

	ABSTRACT

Top Abstract Methods Results Discussion References

Objective.  Investigators have sought to establish "low-risk" criteria to identify febrile young infants who can be observed safely without antibiotics. Previous studies have used criteria for standard urinalysis to identify suspected urinary tract infection; however, cases of urinary tract infection have been missed. Enhanced urinalysis, using hemocytometer cell count and Gram stain performed on uncentrifuged urine, has been shown to have greater sensitivity and negative predictive value than standard urinalysis. The objective of this study was to evaluate the ability of criteria that incorporate enhanced urinalysis to identify febrile young infants who are at low risk for serious bacterial illness (SBI).

Methods.  Institutional guidelines were established in 1999 to evaluate in a retrospective cohort study infants who were 60 days of age with temperature 38.0°C. "Low-risk" criteria included 1) well appearance without focal infection (excluding otitis media); 2) no history of prematurity, illness, or previous antibiotics; 3) peripheral white blood cell count (WBC) between 5 and 15 000/mm³; 4) absolute band count 1500/mm³; 5) cerebrospinal fluid WBC 5/mm³ with a negative Gram stain; 6) enhanced urinalysis with WBC 9/mm³ with a negative Gram stain; 7) stool WBC <5/high power field in infants with diarrhea; and 8) chest radiograph without lobar infiltrate(s) in infants with respiratory signs or symptoms. SBI was defined as a lobar infiltrate on chest radiograph or presence of a bacterial pathogen in blood, urine, cerebrospinal fluid, stool, or culture obtained from the soft tissue. The hospital records of all infants who presented to the emergency department for evaluation of fever after January 1999, including those who did not meet low-risk criteria, were reviewed; data were collected regarding history, physical examination, laboratory test results, treatment, and clinical course.

Results.  During the study period, 434 infants presented to the emergency department for evaluation of fever. Thirty patients were excluded from additional analysis because of incomplete data; 60 patients were identified immediately as "not low risk" on the basis of history or physical examination. Of the 344 remaining infants, 127 were identified as "low risk" on the basis of laboratory criteria; 83 (65.4%) were observed without antibiotics. None of the "low-risk" infants had an SBI. A total of 217 well-appearing infants were classified as "not low risk" on the basis of laboratory criteria; 28 (12.9%) had an SBI. The overall incidence of SBI in infants with complete data was 10.1%, whereas the incidence of SBI in all "not low-risk" infants was 14.8%. The negative predictive value for the "Pittsburgh" criteria was 100% (95% confidence interval: 96.7%-100%); the sensitivity was 100% (95% confidence interval: 89.7%-100%).

Conclusions.  The application of low-risk criteria using enhanced urinalysis improves identification of infants who are at low risk for SBI.  Key words:  bacteremia, low risk criteria, enhanced urinalysis, serious bacterial illness.

Fever is an extremely common symptom among infants and young children who present for evaluation at an emergency department (ED). The incidence of serious bacterial illness (SBI) in those infants in the first 2 months of life has been reported to be as high as 15%.^1-8 Traditionally, a conservative approach has been recommended for febrile infants <2 months of age, in whom the signs and symptoms of infection can be nonspecific. This conservative approach includes obtaining specimens of blood, urine, and cerebrospinal fluid (CSF) for laboratory analysis and culture on all children who subsequently are admitted to the hospital and given intravenous (IV) antibiotics for 48 to 72 hours. Despite these recommendations, there is considerable variability in the extent of evaluation and approach to treatment among general pediatricians, family practitioners, and ED physicians.^9,10

As an alternative to the conservative approach, several studies have shown that a combination of historical, physical examination, and laboratory criteria can be used to identify a subset of febrile infants who are at low risk for SBI.^1-5,11 Investigators have concluded that these infants can be observed without antibiotic therapy. The previously established criteria include standards for normal white blood cell counts (WBC) in the blood and CSF, negative CSF Gram stain, urinalysis showing <10 WBC/high-power field with a negative Gram stain, and normal chest radiograph (CXR) and stool WBC examinations on selected infants; the negative predictive value of these criteria for SBI has ranged from 95% to 100%. Although many physicians have adopted a less conservative approach to fever in infants as a result of these investigations, concerns remain about the potential for missing cases of SBI.⁶

Urinary tract infection (UTI) is the most common SBI among febrile young infants, as well as the most commonly missed SBI in studies that evaluate low-risk criteria.^1-3,12 In the study of low-risk criteria by Jaskiewicz et al,² 35% of patients with UTI had normal standard urinalyses and 3 patients with UTI met all low-risk criteria; other studies that evaluated low-risk criteria failed to specify what proportion of patients with UTI were identified by standard urinalysis.^2-7 Standard urinalysis and Gram stain have been shown to have a relatively poor sensitivity, ranging from 48% to 65%.^13-15 Enhanced urinalysis, using hemocytometer cell count and Gram stain on uncentrifuged urine specimens, has been shown in several prospective studies to have superior negative predictive value and sensitivity when compared with standard urinalysis and Gram stain.^14,16,17 This modification of the traditional urinalysis results in an improved ability to detect UTI and therefore may reduce the risk of failure to identify cases of SBI in febrile infants.

We sought to evaluate the ability of low-risk criteria that incorporate enhanced urinalysis to identify febrile young infants who are at low risk for SBI. In addition, we hoped to foster a more standardized approach to the evaluation and treatment of febrile young infants at our institution.

	METHODS

Top Abstract Methods Results Discussion References

In January 1999, guidelines were established at Children's Hospital of Pittsburgh, a large, urban, academic children's hospital, for the evaluation and treatment of infants who were 60 days of age and had a documented temperature of 38.0°C. These guidelines included elements of history, physical examination, and laboratory findings to identify "low risk" infants. These criteria are shown in Table 1.

The guidelines recommended that any infant who did not meet low-risk criteria be admitted to the hospital and given IV antibiotics until all culture results were known. The recommendation for infants who met low-risk criteria was to admit the infant to the hospital for observation without antibiotics or with oral antibiotic therapy for infants with otitis media, for 24 hours. If the infant remained well-appearing, cultures remained negative, and close follow-up with a primary care provider could be ensured, then the infant could be discharged to home. Any low-risk infant whose clinical status deteriorated or whose cultures became positive was to be given IV antibiotics until culture results were known. The evaluation, treatment, and disposition of all febrile infants remained at the discretion of the ED physician, in consultation with the family and the primary care provider.

The hospital records of all febrile infants who had been evaluated for fever between January 1, 1999, and August 31, 2000, were reviewed. Demographic and clinical data collected for each patient included age, sex, birth history (gestational age, complications, maternal group B streptococcal colonization status, type of delivery, need for resuscitation or neonatal intensive care unit stay), maximum recorded temperature, pertinent medical history, and documented appearance and examination by the supervising ED physician (fellow or attending). Laboratory data collected included peripheral WBC and absolute band count, CSF cell count, differential and Gram stain, enhanced urinalysis (uncentrifuged urine used for hemocytometer cell count and Gram stain), any radiograph results, stool WBC results on infants with diarrhea, and all culture results. Other information recorded included treatment with antibiotics, disposition, and clinical course. For children who were treated with IV antibiotics despite meeting low-risk criteria, the reason for treatment was documented whenever possible.

Our guidelines, as shown in Table 1, included examination and culture of the CSF. At our institution, the CSF evaluation occasionally is omitted in infants who are older than 6 weeks of age and are to be observed without antibiotics; we included infants in our analysis when the examination of the CSF was omitted intentionally and the infant was observed without antibiotic therapy. Children were excluded from analysis if they received antibiotics and did not have a sample of CSF for cell count, Gram stain, and culture or did not have an enhanced urinalysis, peripheral CBC, or cultures of blood and urine performed.

SBI was defined as a lobar infiltrate on CXR (reviewed by an attending radiologist); growth of a bacterial pathogen from the CSF, blood, stool, or soft tissue; or growth of 50 000 colony forming units (cfu)/mL of a single pathogenic organism from a urine specimen obtained by catheter. Demographic and clinical data are reported as means or proportions. The sensitivity, specificity, and positive and negative predictive values of the low-risk criteria were computed using Yates-corrected ². This study was approved by the Human Rights Committee at the Children's Hospital of Pittsburgh.

	RESULTS

Top Abstract Methods Results Discussion References

Between January 1999 and August 2000, 434 febrile infants <2 months of age presented to the ED for evaluation (Fig 1). Thirty infants, who did not have a complete laboratory database as outlined above, were excluded from additional analysis. Of the remaining 404 infants, 60 were identified as "not low risk" on the basis of history and/or physical examination. Of the 344 previously healthy, well-appearing infants with complete data, 127 were classified as "low risk" according to our established guidelines and 217 were classified as "not low risk." There were a total of 41 cases of SBI (10.1%): 25 infants with UTI; 8 with pneumonia; 3 with bacteremia; 2 with meningitis; and 1 each with bacterial gastroenteritis, pertussis, and chlamydia. The diagnoses and causative agents for infants with SBI are shown in Table 2.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Flow diagram of febrile infants.

Patient Demographics

A total of 222 boys and 212 girls presented to the ED with fever during the study period; data were incomplete for 16 boys and 14 girls. The patients ranged in age from 2 to 60 days, with a mean age of 34 days. The overall age distribution, including breakdown by risk group and proportion with an SBI, is included in Table 3. Among male patients, 139 (67.4%) with complete data were classified as "not low risk," whereas 67 (32.6%) were classified as "low risk"; 15 (7.3%) had an SBI. Among female patients, 138 (69.7%) with complete data were classified as "not low risk," whereas 60 (30.3%) were classified as "low risk"; 27 (13.6%) had an SBI. The overall risk of SBI was not significantly different between boys and girls (P = .05).

Patients With Incomplete Data

Two infants with SBI in the group of 30 infants had incomplete data. One patient was a 45-day-old infant who had a history of short gut and a central IV catheter and was treated with IV antibiotics for presumed line sepsis without examination of the CSF; Enterococcus faecalis was recovered from the blood culture. The second patient was a 25-day-old infant who had clinical findings of omphalitis and did not have a sample of CSF obtained; Escherichia coli grew from the culture of the wound. Both of these patients would have been classified as "not low risk" by historical or examination criteria, respectively; in fact, 17 of 30 of the patients with incomplete data met historical, examination, and/or laboratory criteria for classification as "not low risk."

Infants Classified as "Not Low Risk" on the Basis of History or Examination

Of the 60 infants who were classified as "not low risk" on the sole basis of history and/or physical examination, 13 (21.7%) had and SBI. The reasons for classification as "not low risk," as well as the proportion with SBI, are listed in Table 4. Nine of the 13 infants (69.2%) with SBI also would have been classified as "not low risk" on the basis of laboratory criteria. The 4 infants who had an SBI and met "low-risk" laboratory criteria but were identified as "not low risk" by history or examination included 1) a 38-day-old infant who had respiratory distress and lethargy and received a diagnosis of chlamydial pneumonitis; 2) a 52-day-old infant who had focal findings on examination of the lungs and received a diagnosis of lobar pneumonia by CXR; 3) a 12-day-old infant who had respiratory distress and received a diagnosis of lobar pneumonia by CXR; and 4) a 30-day-old infant who had a history of lethargy and poor feeding and was febrile, mottled, and lethargic on examination and grew Streptococcus agalactiae from a blood culture.

Infants Classified as "Not Low Risk" on the Basis of Laboratory Results

Of the 217 well-appearing, previously healthy infants who were classified as "not low-risk" according to laboratory criteria, 28 (12.9%) had SBI. The laboratory abnormalities, excluding infants from the low-risk group and the proportion with an SBI, are shown in Table 5. All "not low-risk" infants with SBI received IV antibiotics after their initial evaluation, and all did well. Twenty-two infants who were classified as "not low risk" by laboratory examination were observed without antibiotic therapy; none had an SBI, and all did well. Fifty-six patients in our study were classified as "not low risk" on the sole basis of the CSF findings; the WBC in the CSF ranged from 6 to 673/mm³, and 1 patient had a positive CSF culture. Six of the 8 patients with lobar infiltrates on CXR also had abnormal WBCs and/or absolute band counts; 2 of these patients required admission to the intensive care unit for severe respiratory distress. Five infants who were classified as "not low risk" by laboratory examination had otitis media documented on examination; all received IV antibiotics for at least 48 hours, and none had an SBI.

Low-Risk Infants

None of the 127 infants who were classified as "low-risk" by our criteria had SBI. Eighty-three (65.3%) were observed without IV antibiotic therapy, whereas 44 (34.7%) were treated with IV antibiotics despite meeting low-risk criteria. The reason for treatment was not documented in the majority of cases, although concurrent jaundice, high percentage of bands (although absolute band count was within acceptable range), and primary care physician's request were cited in several cases. Twenty-one infants who were observed without antibiotic therapy did not have a lumbar puncture performed; none of these patients had an SBI, and all did well. Four infants who were not treated initially with antibiotics were started on antibiotics within 48 hours. Two patients developed increased work of breathing and were found later to have respiratory syncytial virus infection. One patient had increased work of breathing with a normal CXR and was treated for 24 hours until culture results were known to be negative. One patient had persistent fever and was treated for 48 hours, at which time all cultures were negative. There were 14 "low-risk" infants who were diagnosed with otitis media; 5 were treated with oral antibiotics, whereas 9 were given IV antibiotics for 48 hours. None of the infants with otitis media had bacteremia or other SBI. The negative predictive value for our low-risk criteria was 100% (95% confidence interval [CI]: 96.7%-100%).

Infants Who Were 28 Days

A total of 166 infants (38.2%) who were 28 days of age were evaluated for fever during our study period. Of these, 153 had complete data for evaluation. Among eligible infants, 17 (11.1%) were classified as "not low risk" by history or examination, 93 (60.8%) were classified as "not low risk" by laboratory criteria, and 43 (28.1%) were classified as "low risk." There were 12 patients (7.8%) with SBI in this age group, all of whom were classified as "not low risk": UTI (6), meningitis (2), pneumonia (2), pertussis (1), and bacteremia in a central line (1). Among the 12 infants who were 28 days with an SBI, 11 were identified as "not low risk" by laboratory criteria; 2 of these infants also were deemed ill-appearing on physical examination, and 1 had a history of surgery and hospitalization. Only 1 of the 12 infants had low-risk laboratory values: a 12-day-old infant with a lobar infiltrate on CXR and respiratory distress that required admission to the intensive care unit. Eighteen (41.9%) of the 43 low-risk infants who were 28 days of age were observed without antibiotics, and 25 (58.1%) were treated; all infants did well. There was a statistically significant difference in the percentage of "low-risk" patients observed without antibiotics between the group of patients who were >28 days and those who were 28 days (P < .005).

Infants With Positive Urine Cultures

Twenty-five infants (6.2% of eligible infants) had growth of >50 000 cfu/mL of a single organism from a urine culture. This group included 19 girls (76%) and 6 boys (24%), with an age range from 13 to 58 days and a mean age of 38 days. The difference in proportion with UTI between male and female patients was statistically significant (P < .01). Twenty-one (84%) infants had growth of >100 000 cfu/mL, whereas 4 (16%) had growth of >50 000 cfu/mL. All of the patients with positive urine cultures were identified as "not low risk" by our criteria. Although our criteria for abnormal enhanced urinalysis included either a positive Gram stain or 10 WBC/mm³, the majority of patients (21 [84%] of 25 ) with UTI had bacteriuria and pyuria; 17 (68%) of 25 also had abnormal peripheral WBC and/or absolute band counts. Only 1 patient with growth of >50 000 cfu/mL had a normal enhanced urinalysis; this patient was described as ill-appearing and had abnormal WBC and absolute band counts. The sensitivity and negative predictive value of the enhanced urinalysis in our study were 96% (95% CI: 78.9%-99.8%) and 99.7% (95% CI: 98.5%-100%), respectively. Four of the infants who received a diagnosis of UTI also had bacteremia; the same organism was isolated from blood and urine cultures (16%). All of the patients with bacteremia had growth of >100 000 cfu/mL from the urine culture. Only 3 infants in our study group had growth of between 10 000 and 50 000 cfu/mL of a single organism from a urine culture. All of these infants were identified as high risk on the basis of the results of the enhanced urinalysis but were not considered to have a positive culture by our criterion.

	DISCUSSION

Top Abstract Methods Results Discussion References

The evaluation and treatment of infants who are <2 months of age and present to the ED with fever remains controversial. Concerns about health care costs, emerging antibiotic resistance, and potential complications from routine hospitalization and antibiotic therapy have prompted many clinicians to question the traditional conservative approach to the treatment of febrile young infants.^1-5,1118-20 The establishment of treatment guidelines at our institution was an attempt to identify systematically infants who are at low risk of SBI and who could be observed without antibiotic therapy.

Our guidelines differ from previously published criteria in several important ways: 1) we incorporated an enhanced urinalysis and Gram stain in an attempt to improve the sensitivity and negative predictive value of the criteria, 2) we did not consider acute otitis media to be an SBI and allowed infants with otitis media to be classified as "low risk" when they met the remaining criteria, and 3) we included infants who were 28 days of age.

In many of the previous studies, pneumonia was considered an SBI only when there was confirmation of bacterial infection by positive blood culture, tracheal aspirate culture, or bacterial antigen testing. We chose to include infants with lobar infiltrates on CXR in the SBI group for several important reasons. First, it is widely known that microbiologic confirmation of bacterial pneumonia is extremely difficult and not frequently achieved in childhood pneumonia; bacteremia occurs in between 1% and 8% of cases, bacterial antigen tests are not readily available, and tracheal aspirate or lung biopsy is not practical for the vast majority of patients.^21,22 Second, all of the patients who received a diagnosis of pneumonia in our study had respiratory signs and symptoms, fever, and lobar infiltrates on CXR; in addition, 6 of 8 had abnormal WBC or absolute band counts, and 3 of 8 had severe respiratory distress that required admission to the intensive care unit.

Some previous studies did not include CSF findings in their low-risk criteria, although they included meningitis among the SBIs identified in their study patients.^1,2,4 Although infants and children with meningitis may be ill-appearing or have laboratory abnormalities other than CSF pleocytosis, it is possible for a patient with bacterial meningitis, especially a young infant, to have only CSF findings to suggest the diagnosis. In fact, our patient who received a diagnosis of Enterobacter cloacae meningitis was identified as "not low risk" on the sole basis of CSF pleocytosis. We believe that it is important to include CSF examination in any low-risk criteria for febrile young infants, and we recommend lumbar puncture for all of these patients. Although we did classify as "low risk" a small percentage of well-appearing patients who did not have a CSF examination performed, these patients were observed carefully without antibiotic therapy, which would allow for detection of any deterioration in clinical status that might suggest bacterial meningitis or other SBI.

In our study population, our guidelines allowed for the identification and safe observation of infants who were at low risk for SBI without antibiotic therapy. In agreement with previous studies, otitis media among the febrile young infants in our cohort was not associated with a significant risk of SBI.^1,2 Although the number of patients with otitis media in this study was small, it seems that these infants can be treated safely with oral antibiotics and close observation when the remainder of the evaluation is unremarkable. All components of the evaluation, including a thorough history, complete examination, and full laboratory testing, are necessary for classifying an infant as low risk for SBI.

There are several limitations to this study. Not all infants who were identified as "low risk" for having SBI according to these guidelines were observed without antibiotics. Although no cases of SBI were diagnosed, we cannot know with certainty that the outcome for these infants was not affected by the antibiotic therapy. The assessment of the infants and the ultimate treatment decisions were made by numerous physicians with variable approaches to the febrile infant. It is impossible to determine whether infants who were deemed "ill-appearing" by one physician would have been classified as "well-appearing" and low risk by another; infants who had an SBI and were treated solely on the basis of ill appearance may have been observed without administration of antibiotics by a different physician. Finally, important historical information, such as siblings with group B streptococcal disease, gestational age, previous antibiotics, and previous hospitalizations, was not always recorded systematically.

As observed in previous studies, UTI was the most commonly diagnosed SBI in our group of febrile infants. UTI occurs in 5% or more of such patients, and a substantial proportion of these infants have concomitant bacteremia.^15,23 Our criterion for significant bacteriuria (50 000 cfu/mL) was based on studies by Hoberman et al,^14,16,17 who found that the majority of patients with UTI have growth of >50 000 cfu/mL. They also showed that the likelihood of contamination and asymptomatic bacteriuria is much higher in those with growth of <50 000 cfu/mL. Other investigators have found that UTI is the SBI most likely to be missed with the application of low-risk criteria, because of the relatively poor sensitivity of standard urinalysis. The use of hemocytometer cell counts on uncentrifuged specimens eliminates the variability of cell counts associated with centrifugation time and volume of resuspension commonly encountered with standard urinalysis. Enhanced urinalysis uses the same technique as cell counts on other body fluids, such as blood and CSF; the training, cost, and time required to perform the test do not differ substantially from standard urinalysis. The enhanced urinalysis has been proved to be superior to standard urinalysis, and its use results in improved identification of UTI in febrile infants and children. The incorporation of the enhanced urinalysis in our evaluation of febrile infants led to the identification of 100% of patients with positive urine cultures and aided in the identification of 100% of patients with SBI.

	FOOTNOTES

Received for publication Oct 11, 2000; accepted Feb 22, 2001.

Reprint requests to (S.H.) University of Pittsburgh School of Medicine, Children's Hospital of Pittsburgh, Department of Pediatrics, Division of Pediatric Emergency Medicine, 3705 Fifth Ave, Pittsburgh, PA 15213-2583. E-mail: herrs{at}chplink.chp.edu

	ABBREVIATIONS

ED, emergency department; SBI, serious bacterial illness; CSF, cerebrospinal fluid; IV, intravenous; WBC, white blood cell count; CXR, chest radiograph; UTI, urinary tract infection; cfu, colony forming units; CI, confidence interval.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Dagan R, Powell KR, Hall CB, Menegus MA Identification of infants unlikely to have serious bacterial infection although hospitalized for suspected sepsis. J Pediatr 1985; 107:855-860 [CrossRef][Medline]
2. Jaskiewicz JA, McCarthy CA, Richardson AC, Febrile infants at low risk of serious bacterial infectionan appraisal of the Rochester criteria and implications for management. Pediatrics 1994; 94:390-396 [Abstract/Free Full Text]
3. Baskin M, O'Rourke EJ, Fleisher GR Outpatient treatment of febrile infants 28 to 89 days of age with intramuscular administration of ceftriaxone. J Pediatr 1992; 120:22-27 [CrossRef][Medline]
4. Dagan R, Sofer S, Phillip M, Shachak E Ambulatory care of febrile infants younger than 2 months of age classified as being at low risk for having serious bacterial infections. J Pediatr 1988; 112:355-360 [CrossRef][Medline]
5. Baker MD, Bell LM, Avner JR The efficacy of routine outpatient management without antibiotics of fever in selected infants. Pediatrics 1999; 103:627-631 [Abstract/Free Full Text]
6. Baker MD, Bell LM Unpredictability of serious bacterial illness in febrile infants from birth to 1 month of age. Arch Pediatr Adolesc Med 1999; 153:508-511 [Abstract/Free Full Text]
7. Baker MD, Bell LM, Avner JR Outpatient management without antibiotics of fever in selected infants. N Engl J Med 1993; 329:1437-1441 [Abstract/Free Full Text]
8. Baker MD, Avner JR, Bell LM Failure of infant observation scales in detecting serious illness in febrile, 4- to 8-week-old infants. Pediatrics 1990; 85:1040-1043 [Abstract/Free Full Text]
9. Baraff LJ Management of the febrile child: a survey of pediatric and emergency medicine residency directors. Pediatr Infect Dis J 1991; 10:795-800 [Medline]
10. Jones RG, Bass JW Febrile children with no focus of infection: a survey of their management by primary care physicians. Pediatr Infect Dis J 1993; 12:179-183 [Medline]
11. McCarthy CA, Powell KR, Jaskiewicz JA, Outpatient management of selected infants younger than two months of age evaluated for possible sepsis. Pediatr Infect Dis J 1990; 9:385-389 [Medline]
12. Hoberman A, Chao HP, Keller DM, Hickey R, Davis HW, Ellis D Prevalence of urinary tract infection in febrile infants. J Pediatr 1993; 123:17-23 [CrossRef][Medline]
13. Lin DS, Huang IY, Chiu NC, Comparison of hemocytometer leukocyte counts and standard urinalyses for predicting urinary tract infections in febrile infants. Pediatr Infect Dis J 2000; 19:223-227 [CrossRef][Medline]
14. Hoberman A, Wald E Urinary tract infections in young febrile children. Pediatr Infect Dis J 1997; 16:11-17 [CrossRef][Medline]
15. Crain EF, Gershel JC Urinary tract infections in febrile infants younger than 8 weeks of age. Pediatrics 1990; 86:363-367 [Abstract/Free Full Text]
16. Hoberman A, Wald E, Penchansky L, Reynolds EA, Young S Enhanced urinalysis as a screening test for urinary tract infection. Pediatrics 1993; 91:1196-1199 [Abstract/Free Full Text]
17. Hoberman A, Wald E, Reynolds EA, Penchansky L, Charron M Pyuria and bacteriuria in urine specimens obtained by catheter from young children with fever. J Pediatr 1994; 124:513-519 [CrossRef][Medline]
18. Baraff LJ, Bass JW, Fleisher GR, Practice guideline for the management of infants and children 0-36 months of age with fever without source. Pediatrics 1993; 92:1-12 [Abstract/Free Full Text]
19. Baraff LJ, Oslund SA, Schriger DL, Stephen ML Probability of bacterial infections in febrile infants less than three months of age: a meta-analysis. Pediatr Infect Dis J 1992; 11:257-265 [Medline]
20. DeAngelis C, Joffe A, Wilson M, Willis E Iatrogenic risks and financial costs of hospitalizing febrile infants. Am J Dis Child 1983; 137:1146-1149 [Abstract/Free Full Text]
21. Ruuskanen O, Mertsola J Childhood community-acquired pneumonia. Semin Respir Infect 1999; 14:163-172 [Medline]
22. Chumpa A, Bachur RG, Harper MB Bacteremia-associated pneumococcal pneumonia and the benefit of initial parenteral antimicrobial therapy. Pediatr Infect Dis J 1999; 18:1081-1085 [CrossRef][Medline]
23. Ginsburg CM, McCracken GH Urinary tract infections in young infants. Pediatrics 1982; 69:409-412 [Abstract/Free Full Text]