2. Tugwell BD, Lee LE, Gillette H, Lorber EM, Hedberg K, Cieslak PR.

Chickenpox outbreak in a highly vaccinated school population. Pediatrics 2004;113:455-9. 3. Galil K, Lee B, Strine T, Carraher C, Baughman AL, Eaton M, et al. Outbreak of varicella at a day-care center despite vaccination. N Engl J Med 2002;347:1909-15.

bacterial illness, but nothing proved more effective than the clinical judgment of the PROS practitioners. A concern that was an initial stumbling block to this study being funded was that testing would not be done on every infantso how would we know that all serious bacterial illnesses were identied? The authors responded that outcome is a more important endpoint than the number of abnormal tests. Therefore, some infants in this study may well have had unrecognized bacteremia or urinary tract infectionssome self-resolving, some treated successfullybut it seems quite unlikely that the most serious condition, bacterial meningitis, was missed, even though only a third of infants underwent lumbar puncture. How should this study be applied? Were the practitioners really good or just lucky? Are practitioners vindicated for not applying published protocols? Should residents and emergency department (ED) physicians stop using protocols? The answers may rest on two observations: (1) In only 4% of cases did the practitioners have only 1 contact with the families during the course of the febrile illness. That is quite different from the situation in EDs; (2) Outcomes such as bacterial meningitis not suspected clinically are rare in practice but are more likely discovered in EDs, simply because of the frequency with which young febrile infants are seen. It is clear that the rate of serious bacterial infections is much higher in the rst month of life than thereafter, therefore infants aged 0 to 3 months should not be considered a homogeneous group. Infants older than 1 month who do not appear clinically ill and have only a low-grade fever are very unlikely to have bacteremia or bacterial meningitis. Moreover, the fear of missing bacteremia and bacterial meningitis should not direct attention away from the possibility of urinary tract infection, a much more likely condition, particularly in girls and uncircumcised boys. Despite the best research efforts, no set of guidelines or clinical prediction tools identify all infants who need treatment, short of hospitalizing and treating all febrile infants, an approach that is not without its own risks as well as costs. Studies like this one demonstrate the benet of collaborative research in ofce settings and the limits of extrapolating ndings from studies in academic medical centers and EDs to ofce practices.

Management and outcomes of care of fever in early infancy

PANTELL RH, NEWMAN TB, BERNZWEIG J, BERGMAN DA, TAKAYAMA JI, SEGAL M, ET AL. JAMA 2004;291:1203-12
Context Fever in infants challenges clinicians to distinguish between serious conditions, such as bacteremia or bacterial meningitis, and minor illnesses. To date, the practice patterns of ofce-based pediatricians in treating febrile infants and the clinical outcomes resulting from their care have not been systematically studied. Objectives To characterize the management and clinical outcomes of fever in infants, develop a clinical prediction model for the identication of bacteremia/bacterial meningitis, and compare the accuracy of various strategies. Design Prospective cohort study. Setting Ofces of 573 practitioners from the Pediatric Research in Ofce Settings (PROS) network of the American Academy of Pediatrics in 44 states, the District of Columbia, and Puerto Rico. Participants Consecutive sample of 3066 infants aged 3 months or younger with temperatures of at least 388C seen by PROS practitioners from February 28, 1995, through April 25, 1998. Main outcome measures Management strategies, illness frequency, and rates and accuracy of treating bacteremia/ bacterial meningitis. Results The PROS clinicians hospitalized 36% of the infants, performed laboratory testing in 75% (74% with complete blood cell count or blood culture; 54% with urine testing; 33% with lumbar punctures), and initially treated 57% with antibiotics. The majority (64%) were treated exclusively outside of the hospital. Bacteremia was detected in 1.8% of infants (2.4% of those tested) and bacterial meningitis in 0.5%. Well-appearing infants aged 25 days or older with fever of <38.68C had a rate of 0.4% for bacteremia/bacterial meningitis. Frequency of other illnesses included urinary tract infection, 5.4%; otitis media, 12.2%; upper respiratory tract infection, 25.6%; bronchiolitis, 7.8%; and gastroenteritis, 7.2%. Practitioners followed current guidelines in 42% of episodes. However, in the initial visit, they treated 61 of the 63 cases of bacteremia/bacterial meningitis with antibiotics. Neither current guidelines nor the model developed in this study performed with greater accuracy than observed practitioner management. Conclusions Pediatric clinicians in the United States use individualized clinical judgment in treating febrile infants. In this study, relying on current clinical guidelines would not have improved care but would have resulted in more hospitalizations and laboratory testing. Comment These authors tried hard to develop a clinical prediction tool that would identify infants with serious

Kenneth B. Roberts, MD
Moses Cone Health System Greensboro, NC 27401

Diagnosis and testing in bronchiolitis: A systematic review

BORDLEY WC, VISWANATHAN M, KING VJ, SUTTON SF, JACKMAN AM, STERLING L, ET AL. ARCH PEDIATR ADOLESC MED 2004;158:119-26
Context The diagnosis of bronchiolitis is based on typical history and results of a physical examination. The indications for and utility of diagnostic and supportive laboratory testing (eg, chest radiograph lms, complete blood cell counts, and respiratory syncytial virus [RSV] testing) are unclear.

Clinical Research Abstracts for Pediatricians

417

Objectives To review the data on diagnostic and supportive testing in the management of bronchiolitis and to assess the utility of such testing. Design Systematic review of studies addressing diagnostic and supportive testing for bronchiolitis. Main outcome measures Sensitivity and specicity of various diagnostic tests, and impact of these tests on clinical outcome. Study identication In conjunction with an expert panel, the authors derived relevant terms to search the literature published from 1980 to November 2002 in MEDLINE and the Cochrane Collaboration Database of Controlled Clinical Trials. Studies reviewed Of the 797 abstracts identied, 82 trials met the authors inclusion criteria (17 are primary articles on diagnosis of bronchiolitis and 65 are reports of treatment or prevention trials). Results Numerous studies demonstrate that rapid RSV tests have acceptable sensitivity and specicity, but no data show that RSV testing affects clinical outcomes in typical cases of the disease. Seventeen studies presented chest radiographic lm data. Abnormalities on chest radiographs ranged from 20% to 96%. Insufcient data exist to show that chest radiographs reliably distinguish between viral and bacterial disease or predict severity of disease. Ten studies included complete blood cell counts, but most did not present specic results. In one study, white blood cell counts correlated with radiologically dened disease categories of bronchiolitis. Conclusions A large number of studies include diagnostic and supportive testing data. However, these studies do not dene clear indications for such testing or the impact of testing on relevant patient outcomes. Given the high prevalence of bronchiolitis, prospective studies of the utility of such testing are needed and feasible. Comment Bronchiolitis is one of the most common childhood diseases and has been variably diagnosed over decades with differing sobriquets such as infectious asthma, wheezy bronchitis, and infectious bronchitis. Nevertheless, no gold standard exists for its diagnosis. With our current technologic advances, the diagnosis of this clinical entity has become increasingly laboratory-based, but not evidence-based, according to Bordley et al. Over 12 years only 82 trials met the inclusion criteria for their review, and presumably there would have been fewer if they had limited the age to that usually applied to the diagnosis of bronchiolitisthe rst two years of life. Of these studies, diagnosis was the primary focus of only 17, and none evaluated the usefulness of supportive testing in diagnosing bronchiolitis. The other 65 studies were primarily on treatment and prevention but contained some data on diagnosis and testing. The included populations, however, had differing or poorly dened inclusion criteria. Most required an etiologic diagnosis, but none asked or answered whether knowing RSV to be the cause affected the clinical outcome. Most focused on disease severity assessed by clinical scales, the reliability of which may be more dependent on the assessor than what is assessed.

Bordley et al, recognizing these differences and decits, concluded that the studies did not dene clear indications for such testing or for the impact of testing on relevant patient outcomes. This may be interpreted, though incorrectly in my opinion, as evidence against the use of supportive diagnostic testing. One may conclude that specic or supportive testing in managing bronchiolitis seems unlikely to be benecial for most children, but may be warranted in individual circumstances, which currently are not able to be dened. Second, the best diagnostic method currently available is the test of time, performed by an experienced clinician. Third, this analysis substantiates the authors suggestion that prospective trials of the value of such testing are needed. However, although they state that such studies are feasible, they do not dene how to do them. The design, implementation, and funding of such a trial will I hope be the subject of a future publication from this group.

Caroline Breese Hall, MD

Departments of Pediatrics and Medicine, Infectious Diseases University of Rochester, School of Medicine and Dentistry Rochester, NY 14642

Otitis media and speech and language: A meta-analysis of prospective studies

ROBERTS JE, ROSENFELD RM, ZEISEL SA. PEDIATRICS 2004;113:e238-48
Context Considerable controversy surrounds whether a history of otitis media with effusion (OME) in early childhood causes later speech and language problems. Objectives To determine (1) whether a history of OME in early childhood is related to receptive language, expressive language, vocabulary, syntax, or speech development in children 1 to 5 years of age; and (2) whether hearing loss caused by otitis media in early childhood is related to childrens receptive language or expressive language through 2 years of age. Design Systematic review of prospective studies examining how OME in early childhood relates to later speech and language skills. Main outcome measures Speech and language outcomes in children tested between 1 and 5 years of age. Study identication The authors searched online databases and bibliographies of OME studies and reviews for prospective or randomized clinical trials published between January 1966 and October 2002 that examined the relationship of OME or OME-associated hearing loss in early childhood to childrens later speech and language development. Studies reviewed Of the 38 studies identied, 14 had data suitable for calculating a pooled correlation coefcient (correlational studies) or standard difference between parallel groups (group studies). Results There were no signicant ndings for the analyses of OME during early childhood versus receptive or expressive language during the preschool years in the correlation studies. Similarly, there were no signicant ndings for OME versus vocabulary, syntax, or speech during the preschool years. Conversely, there was a signicant negative association between OME and preschoolers receptive and expressive language (lower language, 0.24 and 0.25 standard difference, respectively) in the group studies. In addition, hearing was also

418

Clinical Research Abstracts for Pediatricians

The Journal of Pediatrics  September 2004