PEDIATRICS Vol. 110 No. 4 October 2002, pp. 846-847
Okay, Enhanced Urinalysis, But Questions Cerebrospinal Fluid Findings in Low-Risk Group
To the Editor.—
Dr Herr and colleagues1 studied an enhanced urinalysis (
9 white blood cells [WBC]/mm3 and negative Gram stain of unspun urine) instead of <10 WBC/high-power field in a spun urine sediment as one of the criteria to identify infants at low risk for serious bacterial infection. Using this criterion, no children with urinary tract infections were missed.1 I applaud the authors for this innovation and would hope that the enhanced urinalysis becomes the method of choice to evaluate for urinary tract infection in infants and young children.
In contrast, I find the inclusion of cerebrospinal fluid (CSF) findings as one of the criteria for an infant to be included in the low-risk group troublesome. The vast majority of infants <60 days of age with meningitis (or pneumonia) have viral infections that will not benefit from antimicrobial therapy and might or might not benefit from hospitalization. In the study by Herr et al, 56 infants had 
6 WBC/mm3 in the CSF and 1 had a positive culture (was this the infant with Streptococcus pneumoniae meningitis?). One infant, excluded from the low-risk group only on the basis of CSF pleocytosis, "received a diagnosis of Enterobacter cloace meningitis."1 I find this an interesting statement. Did the infant have E cloace meningitis or not?
The authors are a bit disingenuous in their recommendation to include CSF findings as a low-risk criterion. Thirty infants were excluded from the study because of "incomplete" data. Nonetheless, "21 infants who were observed without antibiotic therapy did not have a lumbar puncture performed." These infants were included in and represent 17% of the 127 infants in their low-risk group.1 How can the authors "recommend lumbar puncture for all of these patients" and yet fail to do one in 17% of their patients? The practice is clearly inconsistent with the recommendation.
In prospective studies of over 1000 infants 0 to 60 days of age in which a lumbar puncture was performed in all study patients, none of the 511 patients meeting the Rochester criteria for inclusion in the low-risk group had bacterial meningitis.2 Although a lumbar puncture was required for inclusion in the study, CSF findings were not included in the low-risk criteria.2 A meta-analysis of low-risk criteria by Baraff et al3 found that the Rochester criteria did not miss a single case of bacterial meningitis. This is not to imply that the Rochester criteria will never miss a case of bacterial meningitis but that such a miss will be rare.
I raise these points because the purpose of low-risk criteria is to identify infants unlikely to have serious bacterial infections. Keeping the criteria simple, noninvasive, and inexpensive will make them easier to use in a wider variety of nonstudy, non-emergency department settings. Herr et al justify including the 17 patients who did not have lumbar punctures in the low-risk group because "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 serious bacterial infection."1 Jaskiewicz et al2 recommended that a lumbar puncture not be done on infants meeting the Rochester criteria and concluded: "Final management decisions for all febrile infants in this age group [0–60 days] will depend on the ability of caregivers to carefully observe their infants, the ease of access to medical care if the infants should clinically worsen, and the assurance of careful follow-up." Herr et al have failed to convince me that infants meeting the Rochester low-risk criteria require a lumbar puncture.
Keith R. Powell, MD
Children’s Medical Center of Akron
Department of Pediatrics
Akron, OH 44308-1062
REFERENCES
- Herr SM, Wald ER, Pitetti RD, Choi SS. Enhanced urinalysis improves identification of febrile infants ages 60 days and younger at low risk for serious bacterial illness.
Pediatrics.2001; 108
:866
–871
[Abstract/Free Full Text] - Jaskiewicz JA, McCarthy CA, Richardson AC, et al. Febril infants at low risk for serious bacterial infection—an appraisal of the Rochester Criteria and Implications for Management.
Pediatrics.1994; 94
:390
–396
[Abstract/Free Full Text] - 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[Web of Science][Medline]
In Reply.—
We appreciate Dr Powell’s comments. Our study involved the establishment of guidelines for evaluation and treatment of febrile infants <60 days, seeking to incorporate an enhanced urinalysis and refine previously established low-risk criteria. Our guidelines included criteria for history, examination, and laboratory findings. Laboratory criteria included cell count, Gram stain, and culture of blood, urine (enhanced urinalysis), and cerebrospinal fluid (CSF). We evaluated 434 infants; 127 were classified as low risk. None of the "low-risk" infants had a serious bacterial illness (SBI), whereas 14.8% of those infants classified as "not low risk" had an SBI. Dr Powell disagrees with the inclusion of CSF findings in the low-risk criteria, asserting that previously established criteria, which did not include CSF findings, did not miss any cases of bacterial meningitis. There were 56 infants in our study group who were classified as "not low risk" based solely upon the CSF findings (normal urinalysis and cell blood count), and 1 of these infants (1.8%) had a positive CSF culture for Enterobacter cloacae. There were several infants who were <28 days of age and had CSF pleocytosis, including 3 with CSF white blood cell counts of 145 to 275 cells, without any other finding that would classify them as "not low risk." Included in this group is the patient with E cloacae meningitis. One additional infant <28 days had Streptococcus pneumoniae meningitis, but also had an abnormal peripheral white blood cell count.
As noted by Dr Powell, there were 21 patients in our low-risk group who did not have a lumbar puncture performed. After the guidelines were established, the care and evaluation of each infant was at the discretion of the treating physician, in conjunction with the primary care provider. Twenty of these 21 patients were older than 28 days of age. None of these infants received antibiotics, and all were carefully observed with no change in clinical status. Although we did not endorse this approach, these infants were included in the study group because the outcome could still be assessed.
We agree that the low-risk criteria should be as simple to implement as possible. However, it is equally imperative that such criteria be as informative and reassuring as possible. Despite the development of various "low-risk criteria," many practitioners in our community remain uncomfortable with and resistant to the management of febrile young infants without routine antibiotic therapy. Having a complete historical, examination, and laboratory database provides critical information regarding a particular infant’s risk of SBI and makes missed diagnoses less likely.
In previous studies published by Jaskiewicz et al,1 lumbar puncture was required for study inclusion. The authors stated that although a lumbar puncture was required for inclusion, the results were not considered in the low-risk criteria. If findings were suggestive of bacterial meningitis, however, antibiotics would be instituted. The authors did not specify what qualified as "suggestive of bacterial meningitis." What degree of pleocytosis is acceptable in an infant <60 days of age? Is a positive Gram stain required for the presumption of bacterial meningitis? While Dr Powell notes that the Rochester criteria "did not miss a single case of bacterial meningitis," it should also be noted that there were no cases of bacterial meningitis in their 1057 patients.
We agree with Dr Powell and others that the use of low-risk criteria allows us to manage a selected group of febrile young infants without antibiotics. We do not agree, however, that examination of the CSF can be omitted from the criteria used to evaluate the risk of SBI including meningitis.
Sandra M. Herr, MD
Children’s Hospital of Pittsburgh
Division of Pediatric Medicine
Pittsburgh, PA 15213-2583
REFERENCE
- Jaskiewicz JA, McCarthy CA, Richardson AC, et al. Febrile infants at low risk for serious bacterial infection—an appraisal of the Rochester criteria and implications for management. Pediatrics.1994; 94 :390 –396
PEDIATRICS (ISSN 1098-4275). ©2002 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
