Printer Friendly

Rochester criteria modified for better sensitivity.

SAN FRANCISCO -- Anew, large study has concluded that the Rochester criteria--for determining which febrile infants are at low risk of having a serious bacterial infection--are not as sensitive as once thought.

But with a modification that adds information about the infant's age and viral status, the sensitivity and negative predictive value of the Rochester criteria can be maintained, Dr. Carrie L. Byington said at the annual meeting of the Pediatric Academic Societies.

The prospective cohort study involved 1,779 febrile infants 1-90 days of age. Dr. Byington and her colleagues from the University of Utah, Salt Lake City, analyzed these cases according to the original Rochester criteria, and also evaluated several possible modifications of the criteria.

Developed from data collected in the 1970s and 1980s from 233 infants, the 15 Rochester criteria are based on patient history, physical examination, and laboratory values. Infants with no positive criteria are said to have only a 1.4% risk of serious bacterial infection, while infants with one or more positive criteria are considered to be at high risk, having a 21% chance of serious bacterial infection.

But the Rochester criteria can be unwieldy, and Dr. Byington noted that many physicians routinely make decisions based on factors that are not part of the criteria, such as the patient's age and viral status.

In the original studies, the Rochester criteria were said to have a sensitivity of 95.7% and a negative predictive value of 99.3%. But in this series of infants seen in Salt Lake City, the Rochester criteria yielded a sensitivity of 90% and a negative predictive value of 97%. By the original criteria, 33% of the infants were said to have a low risk of serious bacterial infection, and 67% were said to have a high risk. As it turned out, only 10% of the infants, whose average age was 31 days, had a serious bacterial infection.

The investigators determined that four of the criteria were significantly associated with serious bacterial illness. They were:

* A urinalysis with more than 10 white blood cells per high-power field (odds ratio 38.8).

* An absolute band count greater than 1,500 (odds ratio 2.7).

* A white blood cell count of less than 5,000 or greater than 15,000 per [mm.sup.3] (odds ratio 1.9).

* The combination of chronic illness and prematurity (odds ratio 9.7). The investigators defined chronic illness as any major anatomic abnormality or major immune deficiency, and they defined prematurity as a gestational age of 37 weeks or less.

Together they refer to these four criteria as the "Modified Rochester Criteria" (MRC).

In addition, high-risk infants (by MRC) aged 29-90 days had a 42% smaller chance of having a serious bacterial infection than high-risk infants between 1-28 days old. And infants with a confirmed viral illness were 70% less likely to have a serious bacterial infection than those with no confirmed viral illness.

When age and viral illness are added to the MRC, the sensitivity rises to 96%, while the negative predictive value remains at 97%.

In fact, this combination of factors can be used to stratify patients into low-, medium-, and high-risk categories. Patients who are negative for the MRC and positive for virus are at low risk, having only a 0.7% chance of serious bacterial infection if they are 29 days or older, and a 1.7% risk if they are younger.

Patients who have at least one positive MRC finding and who are positive for virus form the medium-risk group. The older infants have a 3.8% chance of having a serious bacterial infection, while the younger ones have a 6.5% chance.

Patients who are MRC positive and negative for virus are at the highest risk. The older infants have a 22% chance of having a serious bacterial infection, and the younger infants have a 33% chance.

"We believe that the diagnosis of viral infections should be a standard component of a rule-out-sepsis evaluation," she said.

In response to a question from the audience, Dr. Byington agreed that even a 1% chance of serious bacterial infection merits consideration. She said that her goal is not to identify a group of infants that can be safely left untreated, but rather to decrease the length of time it takes to do a full sepsis evaluation from 53-72 hours to 24 hours.

"My take-home message is not to stop looking for bacteria, but to speed up the way we can make that decision," she said. "I am working toward developing diagnostic technology that can diagnose both bacterial and viral pathogens--15 pathogens [in] 15 minutes."


San Francisco Bureau
COPYRIGHT 2006 International Medical News Group
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Infectious Diseases; rating risk for febrile infants and bacterial infections
Author:Finn, Robert
Publication:Pediatric News
Geographic Code:1U8UT
Date:Jul 1, 2006
Previous Article:Dexamethasone eases pharyngitis.
Next Article:Routine interval appendectomy unnecessary?

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters