PEDIATRICS Vol. 105 No. 3 March 2000, pp. 502-509

Reevaluation of Outpatients With Streptococcus pneumoniae Bacteremia

Richard Bachur, MD^* and Marvin B. Harper, MD^*,

From the Divisions of ^* Emergency Medicine and  Infectious Disease, Children's Hospital, Boston, Massachusetts.

	ABSTRACT

Top Abstract Methods Results Discussion References

Background.  The reevaluation process for outpatients recalled for Streptococcus pneumoniae bacteremia has not been standardized. Children who return ill or with new serious focal infections require admission and parenteral antibiotic therapy. Limited data exist to guide the follow-up management of those patients identified as having occult pneumococcal bacteremia.

Objectives.  Characterize the outcomes of outpatients with pneumococcal bacteremia based on their evaluation at follow-up. For patients who are well-appearing without serious focal infection, propose a management scheme for reevaluation.

Methods.  Retrospective review of outpatients with pneumococcal bacteremia. Patients with immunocompromise, those identified with focal bacterial infection at the initial visit, or those admitted at the initial visit were excluded. Data were collected from the initial visit (when blood culture drawn) and follow-up visit with regard to clinical parameters, laboratory data, diagnoses, and any antibiotic treatment. Decision tree analysis was used to generate a model to predict children at high risk for persistent bacteremia (PB).

Results.  A total of 548 episodes of pneumococcal bacteremia were studied. Seventy-three children received no antibiotic, 239 oral antibiotic, and 236 parenteral antibiotic at the initial visit. Median age, temperature, and white blood cell (WBC) count were 13.5 months, 40.0°C, and 20 400/mm³. Forty-one patients had PB or new focal infections (15 with PB alone, 4 had focal infection and PB). Eight patients had meningitis at follow-up. Ninety-two percent returned because of notification of the positive blood culture result. A repeat blood culture was obtained in 92%, 23% had a lumbar puncture, 33% had a chest radiograph, and 12% were admitted. PB was associated with the antibiotic treatment group, elevation of temperature, and WBC count at follow-up. A simple management scheme using 2 sequential decision nodes of antibiotic treatment (none vs any) and then temperature at follow-up (>38.8°C) would have predicted 16/19 patients with PB (sensitivity = .84 and specificity = .86).

Conclusions.  All patients with pneumococcal bacteremia need prompt reevaluation. For well-appearing patients without new focal infection, the utility of diagnostic testing (specifically repeat blood cultures) and the need for admission may be determined by the use of antibiotics at the initial evaluation and the presence of fever at follow-up. The majority of patients can be managed as outpatients entirely. Patients who did not receive antibiotics at the initial evaluation and those treated with oral antibiotics but remain febrile are at the highest risk for persistent bacteremia.  Key words:  antibiotics, bacteremia, fever, outpatient, Streptococcus pneumoniae, evaluation, blood culture.

The evaluation of young, febrile children and the related topic of occult bacteremia has been a featured topic for pediatric journals and meetings for over 2 decades. Despite the exhaustive discussion and numerous publications, controversy persists. In 1993, an expert panel reviewed the literature, performed a meta-analysis of available data, and offered guidelines for the evaluation of young febrile children.¹ Not surprisingly, the guidelines have not resolved the issues about the use of laboratory tests and whether to provide empiric therapy for those at risk for occult bacteremia.² Despite the disagreement about the role of testing and empiric therapy, blood cultures are the gold standard for identifying bacteremia and continue to be obtained by many physicians in febrile patients who are believed to be at risk for bacteremia based on age, temperature, or white blood cell (WBC) count.

Streptococcus pneumoniae currently causes 90% of occult bacteremia after the introduction of the Haemophilus influenzae conjugate vaccine.³ The outcomes of outpatients with bacteremia have been previously reported.^4-11 The risk of serious focal infections or sepsis has been estimated to be 6% for S pneumoniae.^4,10,12 Despite a relatively low complication rate, S pneumoniae causes most of the complications of occult bacteremia. The reevaluation process of outpatients with proven S pneumoniae bacteremia has been the subject of a single previous study¹³; understandably, given the limited data, the reevaluation process is not standardized. All patients who are identified to have a positive blood culture result need prompt reevaluation. Everyone would agree that patients who return ill or with serious focal infections would require admission for intravenous antibiotics. However, we postulate that the majority of children will be well-appearing without focal infection and can be discharged from the hospital after careful assessment with close follow-up. We undertook this study to review and better define the reevaluation process for outpatients with S pneumoniae bacteremia.

	METHODS

Top Abstract Methods Results Discussion References

We retrospectively reviewed the charts of patients with S pneumoniae bacteremia at Children's Hospital, Boston, MA, from January 1, 1987 through December 31, 1996. Records from the bacteriology laboratory were reviewed to identify these positive blood culture results. Subjects were included in our analysis if they were not hospitalized at the visit from which the blood culture was drawn (defined as the initial visit) and were not immunocompromised. Patients also were excluded if they had a focal bacterial infection (presumed bacterial, requiring antibiotics) other than otitis media, sinusitis, or pharyngitis. Clinical data obtained from the initial and follow-up visits included the chief complaint (as recorded on the nursing triage note), reported duration of fever, antibiotic therapy given in the preceding 48 hours, the presence of any immunodeficiency or other chronic illness, initial vital signs, laboratory tests, radiographs, initial and final discharge diagnoses, and antibiotic therapy. The date and time of the initial and follow-up visits were also recorded. The parental reports of condition and fever at follow-up were categorized as improved, same, worse, and unknown according to the description in the medical record. Any subsequent hospital admissions were noted. Outcomes of patients not returning to our institution were obtained by telephone or review of the medical record at the offices of their physicians. Antibiotic administration was considered parenteral if given intravenously or intramuscularly. Sixty-six patients also were part of a separate prospective occult bacteremia study previously reported by Fleisher et al.¹¹ Outcomes of some other patients (seen before 1994) were previously published by Harper et al.¹⁰

Rectal temperatures were performed on the vast majority of the children and were used for our analysis. Other measured temperatures were adjusted to a rectal temperature by adding 1.0°C to oral and 1.5°C to axillary measurements. Afebrile was defined as <38°C, rectally.

Persistent bacteremia (PB) was defined as a positive blood culture result drawn at the follow-up visit. A new diagnosis was defined as a diagnosis of a new focal infection not made at the initial visit but established in follow-up. Diagnoses of meningitis, pneumonia, septic arthritis, osteomyelitis, and cellulitis were included as new focal infections but not otitis media, pharyngitis, and sinusitis.

The Bactec blood-culturing system (Beckton Dickinson Co, Sparks, MD) was used during the study period. During this time, the system underwent several upgrades in both detection sensitivity as well as in the actual handling of the cultures: originally, blood cultures were manually removed from incubators to a testing platform (twice per day during the first 2 days, then once per day for the next 5 days), and a continuous, automated alarm system was added in October 1995.

Statistical analyses were conducted using Statistical Program for the Social Sciences, Version 6.1.1 (SPSS Inc, Chicago, IL). Medians and interquartile ranges (25th-75th percentile) were given for non-normal data. Mean values of interval data were compared between groups by using a 2-tailed Student's t test. ² and Fisher's exact test were used to test nominal data. Confidence intervals for proportions were calculated using Stata Version 6 (Stata Inc, College Station, TX). Variables found to be associated with PB, lumbar puncture, or hospitalization by univariate analysis were also tested in a multivariate logistic regression model to determine which variables have an independent association.

Tree-structured analysis (via recursive partitioning) by CART (Salford Systems, San Diego, CA) was performed to develop a management scheme for outpatients reevaluated for pneumococcal bacteremia. In using the software, a single outcome variable and all potential predictor variables are assigned. Recursive partitioning analysis develops prediction rules in a stepwise fashion, using bivariate analysis at each step to determine which of the clinical variables partitions the sample with the lowest probability of false negative and false positive assignments. The partitioning is repeated until either the subgoups contain a homogeneous group (defined by the outcome variable) or the subpopulations are too small for additional subdivision. CART then proceeds to prune back some branches by recombining subgroups if classification errors are not significantly increased in the process of simplifying the tree. A parameter reflecting the relative seriousness of misclassifications (ie, false-positive and false-negative decisions) can be modified to influence the development of the tree. Previous medical publications have used CART in developing rules to diagnose or predict lupus erythematosus, myocardial infarction, and compliance with vaccination.^14-18

This study was approved by the institutional review board at Children's Hospital, Boston.

	RESULTS

Top Abstract Methods Results Discussion References

Study Group

A total of 867 patients were identified with S pneumoniae bacteremia during the 9 study years. Six hundred eighty-six were not hospitalized at the initial visit. Fourteen patients with immunodeficiency and 124 patients with focal infections diagnosed at the initial visit were excluded, leaving 548 patients with outpatient pneumococcal bacteremia.

Initial Visit

Seventy percent of patients had fever for 24 hours; only 1% had fever 1 week. Three percent had been treated with antibiotics in the previous 48 hours. The common primary diagnoses at discharge from the initial visit included otitis media (43%), fever without source (30%), and viral syndrome (22%).

Treatment Subgroups

Patients were divided into 3 groups based on therapy at the initial evaluation: those who were not given or prescribed antibiotics (NoAbx), those who had oral antibiotics administered in the emergency department and/or prescribed (OAbx), and those who received a parenteral dose of an antibiotic with or without prescribed oral antibiotics (PAbx). Table 1 provides the median age, initial temperature, and WBC counts for the treatment groups.

Follow-up Visit

The reason for return visit was recorded in 94% (n = 514) of the patients. Ninety-two percent (475) of patients returned because they were called about the positive blood culture result. Two percent (9) of the patients returned because they were perceived by the parents to be more ill including 1 patient who had a seizure, and 6% (30) returned as a scheduled follow-up. The parental report of fever and the child's condition are noted in Table 2. The median time between the initial and follow-up visits was 33 hours (interquartile range: 24-44). The return temperatures are noted in Table 3.

New Focal Infections and PB

Complications at follow-up, including new foci of infection and PB, are noted in Table 4 (stratified by temperature at follow-up). For the 8 patients with meningitis, the median age, WBC count, and temperature at follow-up were 11.5 months old (range: 3.9-35.5 months), 16 600/µL (range: 11 400- 29 600/µL), 39.2°C (range: 37.1-40.5°C), respectively. Details of those diagnosed with meningitis are presented in Table 5.

Evaluation at Follow-up

The frequency of diagnostic studies and admission rates by treatment group and by temperature at follow-up are presented in Table 6. A total of 126 patients (23%) had a lumbar puncture: 59 at the first visit, 63 at follow-up, and 4 patients had lumbar punctures at both visits. Eight of the 126 patients (6%) had significant cerebrospinal fluid findings (positive culture or pleocytosis). One hundred eighty-two patients (33%) had chest radiographs: 137 at the first visit, 35 at the follow-up visit, and 5 at both visits. Eight of 40 (20%) chest radiographs revealed pneumonia. PB was noted in 19/442 (4%) of the follow-up blood culture results.

Predictors of PB

Mean ages for those with and without PB were not statistically different (12.5 vs 56.2 months; P = .17); mean temperature and WBC count at the repeat visit were different (38.8 vs 37.7°C; P < .001 and 18 900 vs 14 900/mm³; P = .02) but only 33% of patients had a repeat WBC count at follow-up. Patients in the NoAbx group were more likely than either OAbx or PAbx patients to have PB: 13/65 (NoAbx) versus 6/227 (OAbx) or 0/206 (PAbx; P < .001 and P < .001, respectively). Additionally, OAbx patients were more likely than PAbx patients to have PB (P = .04). The time interval between the initial and follow-up visits as well as temperature and WBC count at the initial visit was similar for those with and without PB. Age, initial and follow-up temperature, initial and follow-up WBC count, and treatment group were analyzed for association with PB using univariate logistic regression; temperature at follow-up (P < .01), treatment group (P < .01), and follow-up WBC count (P = .03) were associated with PB. Decision tree analysis (CART) using age, temperature at follow-up, and treatment group yielded 2 models for predicting PB (see Fig 1).

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Decision tree models for predicting PB.

Predictors of Hospitalization and Lumbar Puncture at Follow-up

Although univariate analysis identified patient age, temperature at follow-up, and treatment group as factors associated with lumbar puncture and hospitalization, multivariate logistic regression identified the temperature at follow-up and the patient's age as independent predictors of lumbar puncture and only fever at follow-up to be independently associated with hospitalization.

Duration of Antibiotic Therapy

The length of therapy was recorded in 445 (81%) of patients. Of these, 319 patients (71%) received 10 days of antibiotic treatment. Five percent of patients received antibiotic therapy for <7 days, and only 3% of patients were treated for >14 days. No patient was known to have a second episode of bacteremia within 7 days of completing an antibiotic course.

	DISCUSSION

Top Abstract Methods Results Discussion References

Blood cultures remain the gold standard for diagnosing bacteremia and are performed in selected outpatients. Once the blood culture is known to be positive, clinicians are confronted with decisions about diagnostic testing and necessary treatment. We limited our study to S pneumoniae because this pathogen now represents the major pathogen for bacteremia among young febrile children as well as most of the focal complications associated with bacteremia. Further, there is less consensus on follow-up management strategies for S pneumoniae, compared with the other, less common, but more virulent, pathogens. Previous studies have described the complications of outpatient pneumococcal bacteremia.^{4,58-10,1319-27} It is generally agreed that patients who are ill or have new serious focal infections require inpatient parenteral antibiotic therapy; however, most children with pneumococcal bacteremia do not develop complications before follow-up. The management of this latter group is the focus of our study.

A study published by Korones and Shapiro¹³ addressed the management at the follow-up visit for outpatients who are reevaluated for pneumococcal bacteremia. Patients who were well-appearing and afebrile at follow-up were discussed in detail; 76% had received an antibiotic (65%: oral; 11%: parenteral) at the initial visit when the blood culture was drawn. Four (3.6%) of these well-appearing, afebrile children had persistently culture-positive invasive pneumococcal infections (3 with bacteremia and 1 with meningitis); 3 of these patients had been treated with an oral antibiotic before the reevaluation, and 1 had not received any antibiotic therapy. The authors noted that 19 additional patients with pneumococcal bacteremia who were afebrile, but ill appearing, did not have any new focus of infection or PB. The outcomes and diagnostic testing for children who returned febrile were not discussed.

Bratton et al,⁴ in a review of 97 episodes of pneumococcal bacteremia, found PB in 2 of the 46 children treated with an antibiotic at the initial evaluation and in 13 of the 51 untreated patients (3 with a new diagnosis of pneumonia and 1 with meningitis; P = .007). One of the 15 patients with PB was afebrile (an untreated patient). Of the 46 treated patients, 31 (67%) were followed as outpatients and all did well. Of the 51 untreated patients, 20 (40%) were managed as outpatients and 1 developed meningitis (this patient was 1 of 5 patients not prescribed any antibiotics at the second visit). In a prospective multicenter trial of antibiotic treatment of occult pneumococcal bacteremia (amoxicillin/clavulanate versus ceftriaxone), Bass et al⁵ noted that none of the 33 ceftriaxone treated patients had fever at the 24-hour follow-up visit compared with 4 of 18 oral antibiotic treated patients. None of the patients in either group had PB, and most patients were continued on an oral antibiotic for 7 to 10 days.

In the current study, 87% of patients received either oral (44%) or parenteral antibiotics (43%) at the initial evaluation. The observation that only 13% of patients did not receive an antibiotic at the initial evaluation is a direct result of our protocol for screening and empirically treating those at risk for occult bacteremia. This protocol targets children 3 to 36 months old with fever 39°C and no identifiable source for fever; in such patients, a blood culture and complete blood count are obtained, urine is collected in those at risk for urinary tract infection, and patients with WBC count over 15 000/µL are given a single dose of ceftriaxone. Lee and Harper³ have demonstrated that this WBC count threshold to have a 86% sensitivity for detecting occult pneumococcal bacteremia.^1,28,29

Among the study patients, 92% of patients were reevaluated because of notification about the positive blood culture result, including 76% of the patients with PB or a new serious focal infection. Only 1 of the 8 patients with meningitis, the child who had a seizure, returned because of being more ill. The proportion of patients with complications that were called to return because of the positive culture result is much higher than in previous reports,^9,12 where 50% of the patients with complications returned before the positive culture result was known. These differences may stem from the inclusion of Haemophilus influenzae type B and Neisseria meningitidis as well as differences in blood culture and notification systems. Despite the differences, each study indicates that a substantial proportion of patients with complications, including meningitis, can be reevaluated and treated earlier when a blood culture is obtained at the initial visit.

Patients who did not receive an antibiotic at the initial evaluation were at a significantly higher risk for focal complications and PB compared with children treated with an antibiotic regardless of temperature at follow-up. Patients who were initially treated with an oral antibiotic but returned febrile were also at higher risk for PB, compared with those who were treated but afebrile. Patients treated with an antibiotic were more likely to have improvement in fever and overall condition as judged by the parents and were much more likely to be afebrile at follow-up. It is likely that these factors account for the differences in diagnostic testing and admission rate seen between the treatment groups at the follow-up visit. This finding, in addition to the development of new complications, must be considered when weighing the costs and benefits of empiric antibiotic therapy.

We believe that all patients with a positive blood culture result must be reexamined. Patients who are ill, who have any immunodeficiency, or who return with serious focal infections should be hospitalized and treated with intravenous antibiotics. For those who are well-appearing without serious focal infections at reevaluation, consideration of age, temperature at follow-up, and initial antibiotic therapy should influence diagnostic testing (such as repeat blood culture, chest radiograph, and lumbar puncture) and treatment. Our data set is too small to both derive and then validate a complete management strategy; nonetheless, it does seem that decision tree models for predicting PB can identify patients at high risk. Recursive partition analysis yielded 2 predictive models that can be used clinically; in practice, the exact age and temperature cutoff must be chosen conservatively. In addition, any algorithm for the reevaluation of children with S pneumoniae bacteremia must take into account not only the risk of PB but also the risk of meningitis. Known risk factors such as age and continued fever need also be considered when deciding to perform a lumbar puncture in patients who would not otherwise be suspected for meningitis based on their physical examination. In our study, physicians were more likely to examine the cerebrospinal fluid in young children and those with persistent fever, and no patient subsequently identified to have meningitis was missed at reevaluation. Additionally, use of an antibiotic at the initial visit did not confuse the management of any patient with meningitis. Finally, it should be noted that 88% of all patients recalled for S pneumoniae bacteremia were managed completely as outpatients.

The general applicability of our data depends on similar blood-culturing techniques and prompt notification of positive cultures. The ability for follow-up and consideration of the likelihood of drug-resistance must be factored into any management scheme. Although outpatient management of invasive drug-resistant pneumococcus has been reported,^30,31 more aggressive treatment may be necessary in areas with a high prevalence of resistant pneumococcus.

In review of the literature, our current practice, and the data from this study, we can propose an algorithm for the reevaluation process as diagrammed in Fig 2. In the well-appearing child, physical examination looking for new focal infections, consideration of a repeat blood culture and planning a course of outpatient oral antibiotics, and close follow-up are sufficient. For those with fever or those who had not received parenteral therapy at the first visit, a parenteral dose of antibiotics should be given. The most difficult decision in the well-appearing child is whether to perform a lumbar puncture. Unfortunately, insufficient data are available for this decision other than the prevalence of meningitis by age and the knowledge that signs of meningitis may be absent in young infants. We believe it is reasonable to perform a lumbar puncture on those patients with persistent fever especially in infants <1 year of age. Outpatients with continued fevers should be admitted for parenteral antibiotic therapy and evaluation of secondary sites of infection.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Guidelines for the reevaluation of outpatients with proven pneumococcal bacteremia.

A consensus may never be reached on the best approach to young, febrile children especially with evolving issues of antibiotic resistance and new vaccines. Nonetheless, blood cultures will continue to be a component of the diagnostic evaluation in some febrile children. Hopefully our data can be used to create thoughtful strategies that enable physicians to manage young febrile children with identified S pneumoniae bacteremia in a conservative manner that safely identifies children at the highest risk for complications while minimizing unnecessary testing and hospitalization.

	FOOTNOTES

Received for publication Jun 14, 1999; accepted Aug 30, 1999.

Reprint requests to (R.B.) Division of Emergency Medicine, Children's Hospital, 300 Longwood Ave, Boston, MA 02115. E-mail: bachur{at}a1.tch.harvard.edu

	ABBREVIATIONS

WBC, white blood cell; PB, persistent bacteremia; NoAbx, not given or prescribed antibiotics; OAbx, oral antibiotics administered in the emergency department and/or prescribed; PAbx, received a parenteral dose of an antibiotic with or without prescribed oral antibiotics.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Baraff LJ, Lee SI. Fever without source: management of children 3 to 36 months of age. Pediatr Infect Dis J. 1992;11:146-151. Review
2. Green SM, Rothrock SG. Evaluation styles for well-appearing febrile children: are you a "risk- minimizer" or a "test-minimizer"? Ann Emerg Med. 1999;33:211-214. Editorial comment
3. Lee GM, Harper MB Risk of bacteremia for febrile young children in the post-Haemophilus influenzae type b era. Arch Pediatr Adolesc Med 1998; 152:624-628 [Abstract/Free Full Text]
4. Bratton L, Teele DW, Klein JO Outcome of unsuspected pneumococcemia in children not initially admitted to the hospital. J Pediatr 1977; 90:703-706 [CrossRef][Medline]
5. Bass JW, Steele RW, Wittler RR, Antimicrobial treatment of occult bacteremia: a multicenter cooperative study. Pediatr Infect Dis J 1993; 12:466-473 [Medline]
6. Marshall R, Teele DW, Klein JO Unsuspected bacteremia due to Haemophilus influenzae: outcome in children not initially admitted to hospital. J Pediatr 1979; 95:690-695 [CrossRef][Medline]
7. Dashefsky B, Teele DW, Klein JO Unsuspected meningococcemia. J Pediatr 1983; 102:69-72 [CrossRef][Medline]
8. Woods ER, Merola JL, Bithoney WG, Spivak H, Wise PH Bacteremia in an ambulatory setting: improved outcome in children treated with antibiotics. Am J Dis Child 1990; 144:1195-1199 [Abstract]
9. Joffe M, Avner JR Follow-up of patients with occult bacteremia in pediatric emergency departments. Pediatr Emerg Care 1992; 8:258-261 [Medline]
10. Harper MB, Bachur R, Fleisher GR Effect of antibiotic therapy on the outcome of outpatients with unsuspected bacteremia. Pediatr Infect Dis J 1995; 14:760-767 [Medline]
11. Fleisher GR, Rosenberg N, Vinci R, Intramuscular versus oral antibiotic therapy for the prevention of meningitis and other bacterial sequelae in young, febrile children at risk for occult bacteremia. J Pediatr 1994; 124:504-512 [CrossRef][Medline]
12. Alario AJ, Nelson EW, Shapiro ED Blood cultures in the management of febrile outpatients later found to have bacteremia. J Pediatr 1989; 115:195-199 [CrossRef][Medline]
13. Korones DN, Shapiro ED Occult pneumococcal bacteremia: what happens to the child who appears well at reevaluation? Pediatr Infect Dis J 1994; 13:382-386 [Medline]
14. Goldman L, Cook EF, Johson P, Brand D, Rouan G, Lee T Prediction for the need for intensive care in patients who come to the emergency department with acute chest pain. N Engl J Med 1996; 334:1498-1504 [Abstract/Free Full Text]
15. Edworthy SM, Zatarain E, McShane DJ, Bloch DA Analysis of the 1982 ARA lupus criteria data set by recursive partitioningmethodology: new insights into relative merit of individual criteria. J Rheumatol 1988; 15:1493-1498 [Medline]
16. Humiston SG, Rodewald LE, Szilagyi PG, Decision rules for predicting vaccinationstatus of preschool-age emergency department patients. J Pediatr 1993; 123:88-892
17. McConnochie KM, Roghmann KJ, Pasternack J Developing prediction rules and evaluating observation patterns using categorical clinical markers: two complementary procedures. Med Decis Making 1993; 13:30-42
18. Carter WB, Beach LR, Inui TS, Kirscht JP, Prodzinski JC Developing and testing a decision model for predicting influenza vaccination compliance. Health Serv Res 1986; 20:897-932 [Medline]
19. Myers MG, Wright PF, Smith AL, Smith DH Complications of occult pneumococcal bacteremia in children. J Pediatr 1974; 84:656-660 [CrossRef][Medline]
20. McCarthy PL, Grundy GW, Spiesel SZ, Dolan T Jr Bacteremia in children: an outpatient clinical review. Pediatrics 1976; 57:861-868 [Abstract/Free Full Text]
21. Feder H Jr Occult pneumococcal bacteremia and the febrile infant and young child. Clin Pediatr 1980; 19:457-462
22. Teele DW, Pelton SI, Grant MJ, Bacteremia in febrile children under 2 years of age: results of cultures of blood of 600 consecutive febrile children seen in a "walk-in" clinic. J Pediatr 1975; 87:227-230 [CrossRef][Medline]
23. McGowan J Jr, ., Bratton L, Klein JO, Finland M Bacteremia in febrile children seen in a "walk-in" pediatric clinic. N Engl J Med 1973; 288:1309-1312
24. Carroll WL, Farrell MK, Singer JI, Jackson MA, Lobel JS, Lewis ED Treatment of occult bacteremia: a prospective randomized clinical trial. Pediatrics 1983; 72:608-612 [Abstract/Free Full Text]
25. Jaffe DM, Tanz RR, Davis AT, Henretig F, Fleisher G Antibiotic administration to treat possible occult bacteremia in febrile children. N Engl J Med 1987; 317:1175-1180 [Abstract]
26. Bell LM, Alpert G, Campos JM, Plotkin SA Routine quantitative blood cultures in children with Haemophilus influenzae or Streptococcus pneumoniae bacteremia. Pediatrics 1985; 76:901-904 [Abstract/Free Full Text]
27. Sullivan TD, LaScolea L Jr, Neter E Relationship between the magnitude of bacteremia in children and the clinical disease. Pediatrics 1982; 69:699-702 [Abstract/Free Full Text]
28. Jaffe DM, Fleisher GR Temperature and total white blood cell count as indicators of bacteremia. Pediatrics 1991; 87:670-674 [Abstract/Free Full Text]
29. Kuppermann N, Fleisher GR, Jaffe DM Predictors of occult pneumococcal bacteremia in young febrile children. Ann Emerg Med 1998; 31:679-687 [CrossRef][Medline]
30. Leggiadro RJ, Barrett FF, Chesney PJ, Davis Y, Tenover FC Invasive pneumococci with high level penicillin and cephalosporin resistance at a mid-south children's hospital. Pediatr Infect Dis J 1994; 13:320-322 [Medline]
31. Kaplan SL, Mason EO Jr, ., Barson WJ, et al Three-year multicenter surveillance of systemic pneumococcal infections in children. Pediatrics 1998; 102:538-545 [Abstract/Free Full Text]