Rochester Criteria for Febrile Infants
Use in febrile infants ≤60 days of age (rectal temp ≥38°C or 100.4°F).
- Well-appearing infants ≤60 days old presenting to the emergency department for a chief complaint of fever ≥38ºC (100.4ºF), or found to have fever on presentation for other complaint.
- Ill-appearing infants should be redirected to sepsis guidelines.
- While the validation study looked at infants any age ≤60 days, in clinical practice infants <28 days are often considered NOT low risk due to age.
- Premature infants should be assessed based on corrected age (e.g. for infant born at 30 weeks gestational age, subtract 7 weeks from chronologic age).
- Identifies infants at low risk for serious bacterial infection (SBI), defined as bacteremia, meningitis, osteomyelitis, suppurative arthritis, soft tissue infections (cellulitis, abscess, mastitis, omphalitis), urinary tract infection, gastroenteritis, or pneumonia.
- Infants ≤60 days may present with minimal signs and symptoms, or appear similarly to those with viral infection. The criteria can help identify SBI; prevalence is 10-12% in this group, with the majority (>90%) representing UTI (Biondi et al 2013, Greenhow et al 2014)
- May reduce over-testing and treatment of well-appearing febrile infants.
Please fill out required fields.
- Herpes simplex virus (HSV) risk factors should be carefully assessed, including: maternal history of HSV infection or primary lesions at delivery, household contacts with lesions, vesicular rash, presentation with seizures, or pleocytosis on cerebrospinal fluid (CSF) testing.
- Positive viral testing (e.g. RSV, influenza) reduces serious bacterial infection (SBI) likelihood by ~50%, but the risk of concurrent SBI is NOT zero (Greenhow 2014, Krief 2009).
- The gold standard for urine culture is a sample obtained via straight catheterization. “Bag” urine introduces risk of specimen contamination with skin flora.
- Obtain blood, urine, and CSF samples BEFORE starting antibiotics, if possible.
- Differential diagnosis of ill-appearing infants <60 days of age should also include the following: congenital heart disease, metabolic disease (e.g. galactosemia), congenital adrenal hyperplasia with adrenal crisis, and non-accidental trauma.
If LOW risk (in the derivation study, SBI occurred in 1% of low risk infants):
- Limited testing, including complete blood count, blood culture, urinalysis, and urine culture, is recommended.
- These infants generally do not require antibiotics.
- Generally safe to be discharged home, given no social concerns or question of ability to follow up with their primary care provider (PCP).
If NOT low risk (in the study, SBI occurred in 12.3% of infants not at low risk):
- Further testing is required, including complete blood count, blood culture, urinalysis, urine culture, and cerebrospinal fluid (CSF) testing.
- Empiric broad spectrum antibiotic coverage is indicated.
- Admission is recommended, pending negative cultures at 24-36 hours.
Selection of the appropriate criteria:
- Infant appears generally well.
- No evidence of focal infection (skin, soft tissue, bone, joint, or ear).
- No prior illness:
- Born at term (≥37 weeks gestation)
- No perinatal antibiotics
- No unexplained hyperbilirubinemia
- No previous hospitalizations
- No chronic or underlying illness
- Not hospitalized longer than mother after delivery
- Lab values
- WBC 5,000-15,000/mm³ (5 to 15 × 10⁹/L)
- Band neutrophils ≤1,500/mm³ (≤1.5 × 10⁹/L)
- If diarrhea present, fecal leukocytes <5 WBC/hpf
- Urine WBC <10/hpf
If all of the above are true, the infant is low risk for SBI.
Facts & Figures
Risk for SBI
Low risk (<1%)
Not low risk
The Rochester criteria were first proposed by Dagan et al in 1985 at the University of Rochester Medical Center in New York. In 1994, Jaskiewicz et al validated the criteria by aggregating data from three prospective studies conducted from 1984-1992. Only infants ≤60 days with rectal temperature ≥38ºC at home or at presentation were included. Clinical environment included an emergency department and a pediatric outpatient clinic.
Evaluation of each infant included global assessment, past medical history, physical examination (no evidence of skin, soft tissue, bone, or joint infection), and laboratory assessment including blood, urine, and cerebrospinal fluid (CSF) studies. Chest x-ray and stool studies were only obtained if clinical symptoms were present. Of note, CSF studies were NOT a part of the Rochester risk-stratification criteria.
Each infant was then stratified as low-risk or not low risk. Of 931 evaluable, 437 met all of the low-risk criteria, and 511 did not. The study’s main outcomes were bacteremia and a larger, inclusive category: serious bacterial infection (SBI). SBI included: bacteremia, meningitis, osteomyelitis, suppurative arthritis, soft tissue infections (cellulitis, abscess, mastitis, omphalitis), urinary tract infection, gastroenteritis, and pneumonia.
SBI was identified in 1% of low risk infants as compared to 12.3% of non-low risk infants. The negative predictive value of the low-risk criteria was 99.5% for bacteremia and 98.9% for SBI.
Hui et al conducted a review of 84 studies to determine the diagnostic accuracy of screening tools for SBI and herpes simplex virus (HSV) in infants less than 3 months. This review also examined the relationship between viral testing and risk of SBI.
The combined clinical and laboratory criteria (Rochester, Philadelphia, Boston, and Milwaukee) demonstrated similar overall accuracy (sensitivity: 84.4-100%; negative predictive value: 93.7-100%) for identifying infants with SBI. The Rochester criteria were more accurate in neonates than in older infants; the other screening tools were more accurate in older infants as opposed to neonates.
In 2016, Gomez et al conducted a prospective study including infants <90 days presenting to 11 European pediatric emergency departments between September 2012 and August 2014. They compared the accuracies of the new Step-by-Step approach, Rochester Criteria, and Lab-score in identifying those patients at low risk of invasive bacterial infection (IBI).
For their study population, the sensitivity (“sens”) and negative predictive value (NPV) for ruling out IBI were as follows:
- Step-by-Step: sens 92.0%, NPV 99.3%.
- Rochester: sens 81.6%, NPV 98.3%.
- Lab-score: sens 59.8, NPV 98.1%.
Seven infants with an IBI were misclassified by the Step by Step, 16 by Rochester criteria, and 35 by the Lab-score.
Original/Primary ReferenceDagan 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(6):855-60.
ValidationJaskiewicz 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. Febrile Infant Collaborative Study Group. Pediatrics. 1994;94(3):390-6.
Other ReferencesBiondi E, Evans R, Mischler M, et al. Epidemiology of bacteremia in febrile infants in the United States. Pediatrics. 2013;132(6):990-6.Greenhow TL, Hung YY, Herz AM, Losada E, Pantell RH. The changing epidemiology of serious bacterial infections in young infants. Pediatr Infect Dis J. 2014;33(6):595-9.Krief WI, Levine DA, Platt SL, et al. Influenza virus infection and the risk of serious bacterial infections in young febrile infants. Pediatrics. 2009;124(1):30-9.Hui C, Neto G, Tsertsvadze A, et al. Diagnosis and management of febrile infants (0-3 months). Evid Rep Technol Assess (Full Rep). 2012;(205):1-297.Biondi EA, Byington CL. Evaluation and Management of Febrile, Well-appearing Young Infants. Infect Dis Clin North Am. 2015;29(3):575-85.Gomez B, Mintegi S, Bressan S, et al. Validation of the "Step-by-Step" Approach in the Management of Young Febrile Infants. Pediatrics. 2016;138(2)
From the Creator
Why did you develop the Rochester Criteria? Was there a particular clinical experience or patient encounter that inspired you to create this tool for clinicians?
Prior to the development of the Rochester Criteria, the consensus was that there was no way to predict which child under 3 months of age will have serious bacterial infection (SBI). Too many children that obviously did not have much risk to have SBI were hospitalized, submitted to sepsis workup, and treated with antibiotics for 2-3 days at least. My thought was that instead of looking for who has SBI, we needed to look at those who are very likely NOT to have SBI.
I have used my clinical experience from my residency in Israel to set a list of criteria that would suggest no SBI and tested them first in Rochester (during my fellowship) and then submitted them for validation later in Israel. These criteria were proved to be able to rule out more than 2/3 of the "suspected" children under 3 months of age, and thus only ~1/3 needed to be subjected to sepsis workup.
What pearls, pitfalls and/or tips do you have for users of the Rochester Criteria? Do you know of cases when it has been applied, interpreted, or used inappropriately?
- Remember that part of these criteria is the absence of previous history that suggests SBI (previously healthy!).
- Remember that any real sick looking baby (i.e., toxic) should be treated anyhow.
- Remember that very low risk is still not 100% proof! The group of neonates under 2-3 weeks was the minority in these studies, so be even more careful with this group before you decide they are low risk.
- Remember to follow all those who are not treated very carefully!
What recommendations do you have for doctors once they have applied the Rochester Criteria? Are there any adjustments or updates you would make to the score based on new data or practice changes?
Just remember that infants <2-3 months may show a dynamic disease, so follow up very carefully.
How do you use the Rochester Criteria in your own clinical practice? Can you give an example of a scenario in which you use it?
In our pediatric emergency room at the Soroka University Medical Center in Beer-Sheva, Israel, the Rochester criteria are the gold standard guidelines for suspected sepsis or febrile children <2-3 months. However, we always warn that in doubt, treat. Low risk infants can be sent home if there is no other reason for hospitalization (e.g. dehydration, hypoxemia).
About the Creator
Ron Dagan, MD, is a distinguished professor of pediatrics and infectious diseases at the Ben-Gurion University of the Negev in Beer-Sheva, Israel. He also created the pediatric infectious disease unit at the Soroka University Medical Center, where he was director until 2014. Dr. Dagan is an active researcher, primarily focusing on new conjugate vaccines, epidemiology of vaccine-preventable diseases, and other infectious disease topics.
To view Dr. Ron Dagan's publications, visit PubMed
About the Creator
Julie Jaskiewicz, MD, is a pediatrician affiliated with HealthSource of Ohio, the Cincinnati Children’s Hospital Medical Center, and Mercy Health Anderson Hospital. She is also a fellow of the American Academy of Pediatrics. Dr. Jaskiewicz's research interests include evaluation and management of febrile infants as well as impact of night shift work on pediatric residents.
To view Dr. Julie Jaskiewicz's publications, visit PubMed
- Laura Mercurio, MD