Among young infants presenting with fever, untreated serious bacterial infections can have severe outcomes, so a full sepsis workup is often recommended but may not be necessary. This issue reviews the use of novel diagnostic tools such as procalcitonin, C-reactive protein, and RNA biosignatures as well as new risk stratification tools such as the Step-by-Step approach and the Pediatric Emergency Care Applied Research Network prediction rule to determine which febrile young infants require a full sepsis workup and to guide the management of these patients in the emergency department. The most recent literature assessing the risk of concomitant bacterial meningitis with urinary tract infections and the role for viral testing, specifically herpes simplex virus and enterovirus, are also reviewed.
A 20-day-old boy presents to the ED in August for evaluation of a rectal temperature of 38˚C (100.4˚F). The baby was born by spontaneous vaginal delivery at 39 weeks’ gestational age. The mother’s prenatal labs were normal, including negative screening for group B Streptococcus. The patient felt warm to the parents today but has otherwise been asymptomatic. The baby has been eating 3 ounces every 4 hours and making an appropriate amount of wet diapers. The physical examination is normal, including a flat anterior fontanel and good hydration. When you explain to the mother that the baby will need to undergo the full sepsis workup, including lumbar puncture, she starts asking you questions: Is all of that testing necessary? Since her baby appears well other than the fever, what is the probability that he has a serious infection? Can other infections besides bacterial infections cause a fever, and does the baby need testing to identify those infections? After the testing is completed, will the baby need to be admitted to the hospital?
A 40-day-old girl presents to the ED in January for evaluation of a rectal temperature of 38˚C (100.4˚F). The history and physical examination are similar to the infant you saw in August, except that she has nasal discharge and a cough. Which risk stratification algorithm should you use for this infant? Would your workup change if a respiratory swab was positive for respiratory syncytial virus?
Your next patient is a 50-day-old girl who also presents to the ED with a rectal temperature of 38˚C (100.4˚F). The history and physical examination are similar to the patient you just saw, except this patient does not have nasal discharge or a cough. You send routine blood and urine tests, and the urinalysis results are positive for leukocyte esterase, > 20 white blood cells/high-power field, and many bacteria. Does this baby require a lumbar puncture? What is the likelihood of concomitant bacterial meningitis with a urinary tract infection?
Due to an immature immune system and pathogens often specific to the age group, the young infant (generally aged < 60-90 days, depending on the specific study or review) is at high risk for serious bacterial infections (SBIs); in particular, urinary tract infection (UTI), bacteremia, and bacterial meningitis. Consequently, the febrile young infant with a rectal temperature ≥ 38°C (100.4°F) is commonly encountered in the emergency department (ED).1 The incidence of SBI in febrile infants aged < 90 days is 8% to 12.5%,2 and it is nearly 20% in neonates (aged ≤ 28 days).3 The incidence of potentially life-threatening bacteremia and/or bacterial meningitis (ie, invasive bacterial infection [IBI]) is approximately 2%.4
Due to their lack of social responsiveness (eg, social smile) and verbal cues, even well-appearing febrile young infants may harbor an SBI, in contrast to well-appearing febrile older infants and children who are at lower risk for IBI.5 Multiple studies have demonstrated that both observation scales and clinician suspicion for SBI are poorly predictive of bacterial infection in febrile infants.6,7 Additionally, bacterial meningitis is the most common diagnosis involved in pediatric medical malpractice claims in the emergency department (ED).8
Over 2 decades ago, several risk stratification criteria were created to identify febrile young infants at low risk for SBI, and the criteria have been utilized to potentially avoid hospitalization of certain low-risk patients.9-11 More recently, newer risk stratification algorithms that incorporate biomarkers such as procalcitonin (PCT) and C-reactive protein (CRP) have been developed and validated in febrile infants.12,13
In addition to bacterial disease, the febrile infant aged ≤ 28 days is also at risk for neonatal herpes simplex virus (HSV) infection, a rare but life-threatening disease that is controversial in its workup and management.14 Other current controversies include the utility of the full sepsis workup in febrile young infants with identifiable sources of fever such as respiratory syncytial virus (RSV)15 and bronchiolitis, and the need for cerebrospinal fluid (CSF) testing in infants with presumptive UTI but who are otherwise at low risk for bacteremia and/or bacterial meningitis. Parents understandably question why invasive testing is recommended for their febrile baby, and the emergency clinician needs to clearly communicate the rationale behind the management of patients in this high-risk age group.
This issue of Pediatric Emergency Medicine Practice reviews the most up-to-date evidence for evaluation and management of febrile young infants, including the newer risk stratification algorithms, the need for routine versus selective CSF testing, the role of viral testing, and diagnostic and therapeutic challenges.
A literature search was performed in the PubMed database using multiple combinations of the search terms: febrile young infant, febrile infant, fever, low risk criteria, neonate, serious bacterial infection, invasive bacterial infection, neonatal herpes simplex virus, and infant less than 90 days old. In addition to reviewing articles included in the original version of this review published in 2013, all relevant articles published in or after 2013 were reviewed. Over 140 articles were reviewed, 109 of which were selected for inclusion. Emphasis was placed on reviewing the most important historical evidence, as well as recent studies with evidence that has been incorporated into clinical practice.
The body of research on the evaluation and management of febrile young infants is extensive and growing, but there is a paucity of randomized controlled trials, and there is no universally accepted clinical practice guideline. Additionally, IBIs (particularly bacterial meningitis) are rare in febrile young infants, so there are limited data on the precision of algorithms (eg, Rochester, Boston, and Philadelphia criteria) for the risk stratification of infants with an IBI. There are also limited data on the risk of adverse outcomes among infants who experience a delay in diagnosis of IBI. The Step-by-Step approach is a recently validated risk stratification algorithm, but this approach needs to be evaluated in certain populations of febrile young infants, such as infants with bronchiolitis. Additionally, the statistically derived and newly published Pediatric Emergency Care Applied Research Network (PECARN) prediction rule should, ideally, be further validated.
1. “The neonate had a fever, but he appeared to be well. I couldn’t justify doing the full sepsis workup, since there was little chance he had a serious infection.”
The prevalence of infection is too high for testing to be deferred. The febrile young infant is at high risk for an SBI, especially if he is aged ≤ 28 days, as nearly 1 in 5 febrile neonates will have an SBI.3 Additionally, the well-appearing febrile infant aged ≤ 60 days is also at risk, as 9.6% of these infants have an SBI and 1.8% have an IBI.6
5. “The mother denied any history of HSV, so I thought the 12-day-old neonate who looked ill likely had a bacterial infection and did not have neonatal HSV.”
The highest risk for transmission of neonatal HSV is babies born to mothers who have a primary infection at the time of delivery.31 The infection may be subclinical, so the mother may not know she has HSV when the baby presents to the ED. While the incidence of neonatal HSV is low,14 comprehensive HSV testing should be performed and empiric acyclovir therapy initiated in the ill-appearing, hypothermic, or seizing neonate, or when vesicles or a CSF pleocytosis with lymphocyte predominance are present.105
9. “The 40-day-old febrile baby was very fussy on my exam, but the labs were normal, so he met the low-risk criteria, and I discharged him home.”
All of the low-risk criteria require the infant to be well-appearing on physical examination. (See Table 4.) Even with normal laboratory studies, if the infant is ill-appearing or has a focal infection, the baby should be hospitalized with initiation of empiric antibiotic therapy.
Evidence-based medicine requires a critical appraisal of the literature based upon study methodology and number of patients. Not all references are equally robust. The findings of a large, prospective, randomized, and blinded trial should carry more weight than a case report.
To help the reader judge the strength of each reference, pertinent information about the study is included in bold type following the reference, where available. In addition, the most informative references cited in this paper, as determined by the author, are highlighted.
Why to Use
When to Use
If the patient is at low risk for SBI according to the Rochester criteria (in the derivation study, SBI occurred in 1% of low-risk infants):
If the patient is not considered to be at low risk for SBI according to the Rochester criteria (in the study, SBI occurred in 12.3% of infants who were identified as not at low risk):
Abbreviations: ED, emergency department; SBI, serious bacterial infection.
Laura Mercurio, MD
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 3 prospective studies that were conducted between 1984 and 1992. Only infants aged ≤ 60 days who had rectal temperatures ≥ 38ºC (100.4ºF) at home or at presentation were included in the validation study. The clinical environments were an emergency department and a pediatric outpatient clinic.
The evaluation of each infant included global assessment, past medical history, physical examination (including for evidence of skin, soft tissue, bone, or joint infection), and laboratory assessment (including blood, urine, and cerebrospinal fluid studies). Chest x-ray and stool studies were only obtained if clinical symptoms were present. Of note, cerebrospinal fluid studies were not part of the Rochester risk stratification criteria. Each infant was then categorized as low risk or not low risk. Among 931 evaluable patients, 437 met all of the low-risk criteria and 511 did not.
The study’s main outcomes were bacteremia and a larger inclusive category of serious bacterial infection (SBI). SBI was defined as bacteremia, meningitis, osteomyelitis, suppurative arthritis, soft tissue infections (cellulitis, abscess, mastitis, omphalitis), urinary tract infection, gastroenteritis, or pneumonia. SBI was identified in 1% of the low-risk infants as compared to 12.3% of non–low-risk infants. The negative predictive value (NPV) of the low-risk criteria was 99.5% for bacteremia and 98.9% for SBI.
In 2012, Hui et al conducted a review of 84 studies to determine the diagnostic accuracy of screening tools for SBI and HSV in infants aged < 3 months. This review also examined the rela-tionship between viral testing and risk of SBI. The various clinical and laboratory criteria (including the Rochester, Philadelphia, Boston, and Milwaukee screening tools) demonstrated similar overall accuracy (84.4%-100% sensitivity; 93.7%-100% NPV) for identifying infants with SBI. The Rochester criteria were more accurate in neonates than in older infants, while the other screening tools were more accurate in older infants than in neonates.
In 2016, Gomez et al conducted a prospec-tive study including infants aged < 90 days who presented to 11 European pediatric emergency departments between September 2012 and August 2014. The study compared the accuracies of the new Step-by-Step approach, the Rochester criteria, and the Lab-score for identifying patients who are at low risk of invasive bacterial infection (IBI). For the study population, the sensitivity and NPV for ruling out IBI were 92.0% and 99.3%, respectively, for the Step-by-Step approach, 81.6% and 98.3% for the Rochester criteria, and 59.8% and 98.1% for the Lab-score. Some infants with IBIs were misclassified by each of the tools in the study: 7 by the Step-by-Step approach,16 by the Rochester criteria, and 35 by the Lab-score.
Ron Dagan, MD
Why to Use
The etiology of fever in infants aged ≤ 90 days can range from self-limiting viral illness (eg, bronchiolitis) to life-threatening IBI (eg, bacteremia or meningitis). The Step-by-Step approach can be used to rule out IBI with a high negative predictive value (99.3%). If IBI can be safely ruled out, these low-risk infants do not require hospital admission and intravenous antibiotics.
When to Use
|Risk Group||IBI Risk||Recommendation|
|Low||0.7%||Full sepsis workup is likely not needed. Consider a period of ED observation, especially if the fever lasts < 2 hours, and ensure outpatient follow-up with a pediatrician.|
|Intermediate||3.4%||Full sepsis workup (including blood, urine, and cerebrospinal fluid cultures), initiation of broad-spectrum intravenous antibiotics, and inpatient hospital admission may be indicated, especially if the patient is aged 21 to 28 days.|
|High||8.1%||Full sepsis workup (including blood, urine, and cerebrospinal fluid cultures), initiation of broad-spectrum intravenous antibiotics, chest x-ray, and inpatient hospital admission are recommended.|
Management of IBI in Infants:
Abbreviations: ED, emergency department; IBI, invasive bacterial infection
Emily Heikamp, MD, PhD
No decision rule should trump clinical gestalt. High suspicion for IBI in a febrile infant should warrant a full sepsis workup.
Gomez et al (2016) conducted a prospective validation study of previously derived criteria, which they applied to 2185 infants aged ≤ 90 days who presented to pediatric emergency departments at 11 European hospitals. Among this group, 3.9% were diagnosed with an IBI and 19.1% were diagnosed with a noninvasive bacterial infection such as urinary tract infection.
In a post-hoc analysis, the Step-by-Step approach demonstrated superior sensitivity and negative predictive value as compared to other risk assessment tools such as the Rochester criteria and the Lab-score (Shaughnessy 2016). Sensitivity and negative predictive value for ruling out IBI were 92.0% and 99.3% for the Step-by-Step approach, 81.6% and 98.3% for the Rochester criteria, and 59.8% and 98.1% for the Lab-score, respectively.
Santiago Mintegi, MD, PhD.
Why to Use
A physical examination alone is unreliable in ruling out SBI in febrile infants. The PECARN prediction rule may help to decrease unnecessary admissions and/or lumbar punctures. It can be used to help determine the disposition of some well-appearing infants who have reliable access to follow-up with a primary care pediatrician or in the same ED in 24 hours, or whose caregivers can be relied upon to return the patient to the ED if a pending culture has a positive result.
When to Use
Use the PECARN prediction rule in well-appearing infants aged ≤ 60 days, to stratify the risk of SBI (defined as urinary tract infection, bacteremia, or bacterial meningitis).
Some well-appearing infants considered to be at low risk for SBI may be suitable for discharge from the ED with follow-up with their primary care pediatrician or in the same ED in 24 hours for reassessment, as opposed to the traditional practice of admitting all febrile infants aged 0 to 60 days.
Abbreviations: ED, emergency department; PECARN, Pediatric Emergency Care Applied Research Network; SBI, serious bacterial infection.
Derek Tam, MD, MPH
Hector Vazquez, MD
Christopher Tainter, MD, RDMS
Consider a critical congenital heart defect (and empiric prostaglandin treatment) in a neonate who presents in shock.
The derivation study by Kuppermann et al (2019) included 1896 previously healthy febrile infants aged ≤ 60 days who had a serum procalcitonin level test at the time of their sepsis evaluation; participants whose procalcitonin samples were lost or mislabeled were excluded. The overall cohort was 1821 patients (908 in the derivation sample and 913 in the validation sample). The primary outcome was the presence or absence of an SBI, defined as urinary tract infection, bacteremia, or bacterial meningitis.
The prediction rule had a sensitivity of 98.8% (95% confidence interval [CI], 92.5%-99.9%) in the derivation study and 97.7% sensitivity (95% CI, 91.3%- 99.6%) in the validation study. The negative predictive value for SBI was 99.8% (95% CI, 98.8%-100.0%) and 99.6% (95% CI, 98.4%- 99.9%) in the derivation and validation studies, respectively. Because the validation study was not conducted independently, there is a risk of diminished external validity.
The benefits of using this rule are: (1) unnecessary admissions may be decreased and (2) unnecessary lumbar punctures may be avoided. A key difference in this prediction rule as compared to other similar rules is that the sensitivity remained high despite the fact that lumbar puncture results were not used as criteria in the rule. However, there is a low prevalence of bacterial meningitis in the general population due to the use of Haemophilus influenzae type B and pneumococcal vaccinations, so there were few cases of bacterial meningitis included in this study’s data set.
Finally, 3 infants in the study were misclassified by the prediction rule as being at low risk but had SBIs (2 had a urinary tract infection and 1 had Enterobacter cloacae bacteremia). All 3 were treated appropriately based on culture results and had uneventful clinical courses.
Nathan Kuppermann, MD, MPH
Lauren Palladino, MD; Christopher Woll, MD; Paul L. Aronson, MD, MHS
Jeffrey R. Avner, MD, FAAP; Jessica S. Williams, MD
July 2, 2019
August 1, 2022
4 AMA PRA Category 1 Credits™, 4 ACEP Category I Credits, 4 AAP Prescribed Credits, 4 AOA Category 2-A or 2-B Credits. Specialty CME Credits: Included as part of the 4 credits, this CME activity is eligible for 4 Infectious Disease CME credits
Date of Original Release: July 1, 2019. Date of most recent review: June 15, 2019. Termination date: July 1, 2022.
Accreditation: EB Medicine is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians. This activity has been planned and implemented in accordance with the accreditation requirements and policies of the ACCME.
Credit Designation: EB Medicine designates this enduring material for a maximum of 4 AMA PRA Category 1 CreditsTM. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
Specialty CME: Included as part of the 4 credits, this CME activity is eligible for 4 Infectious Disease CME credits, subject to your state and institutional approval.
ACEP Accreditation: Pediatric Emergency Medicine Practice is also approved by the American College of Emergency Physicians for 48 hours of ACEP Category I credit per annual subscription.
AAP Accreditation: This continuing medical education activity has been reviewed by the American Academy of Pediatrics and is acceptable for a maximum of 48 AAP credits per year. These credits can be applied toward the AAP CME/CPD Award available to Fellows and Candidate Fellows of the American Academy of Pediatrics.
AOA Accreditation: Pediatric Emergency Medicine Practice is eligible for 48 Category 2-A or 2-B credit hours per year by the American Osteopathic Association.
Needs Assessment: The need for this educational activity was determined by a survey of medical staff, including the editorial board of this publication; review of morbidity and mortality data from the CDC, AHA, NCHS, and ACEP; and evaluation of prior activities for emergency physicians.
Target Audience: This enduring material is designed for emergency medicine physicians, physician assistants, nurse practitioners, and residents.
Goals: Upon completion of this activity, you should be able to: (1) demonstrate medical decision-making based on the strongest clinical evidence; (2) cost-effectively diagnose and treat the most critical ED presentations; and (3) describe the most common medicolegal pitfalls for each topic covered.
Discussion of Investigational Information: As part of the journal, faculty may be presenting investigational information about pharmaceutical products that is outside Food and Drug Administration approved labeling. Information presented as part of this activity is intended solely as continuing medical education and is not intended to promote off-label use of any pharmaceutical product.
Faculty Disclosures: It is the policy of EB Medicine to ensure objectivity, balance, independence, transparency, and scientific rigor in all CME-sponsored educational activities. All faculty participating in the planning or implementation of a sponsored activity are expected to disclose to the audience any relevant financial relationships and to assist in resolving any conflict of interest that may arise from the relationship. Presenters must also make a meaningful disclosure to the audience of their discussions of unlabeled or unapproved drugs or devices. In compliance with all ACCME Essentials, Standards, and Guidelines, all faculty for this CME activity were asked to complete a full disclosure statement. The information received is as follows: Dr. Palladino, Dr. Woll, Dr. Aronson, Dr. Avner, Dr. Williams, Dr. Mishler, Dr. Skrainka, Dr. Claudius, Dr. Horeczko, and their related parties report no relevant financial interest or other relationship with the manufacturer(s) of any commercial product(s) discussed in this educational presentation.
Commercial Support: This issue of Pediatric Emergency Medicine Practice did not receive any commercial support.
Earning Credit: Two Convenient Methods: (1) Go online to www.ebmedicine.net/CME and click on the title of this article. (2) Mail or fax the CME Answer And Evaluation Form with your June and December issues to Pediatric Emergency Medicine Practice.
Hardware/Software Requirements: You will need a Macintosh or PC with internet capabilities to access the website.
Additional Policies: For additional policies, including our statement of conflict of interest, source of funding, statement of informed consent, and statement of human and animal rights, visit https://www.ebmedicine.net/policies.