By Philip R. Fischer, MD, DTM&H
Professor of Pediatrics, Department of Pediatric and Adolescent Medicine, Mayo Clinic, Rochester, MN
Dr. Fischer reports no financial relationships relevant to this field of study.
SYNOPSIS: Practices vary significantly as clinicians evaluate and manage febrile infants younger than 2 months of age. A retrospective review suggests that meningitis is extremely unlikely in well-appearing babies with initial laboratory results suggestive of urinary tract infection, and that cerebrospinal fluid analysis may not be necessary.
SOURCE: Wang ME, Biondi EA, McCulloh RJ, et al. Testing for meningitis in febrile well-appearing young infants with a positive urinalysis. Pediatrics 2019;144:e20183979.
Fever is a common presenting symptom in young infants, and urinary tract infection is the most likely serious bacterial infection in this group. Various criteria have been established to guide appropriate evaluation and management of febrile babies, but clinical practices still vary widely. Specifically, there has been controversy about the value of cerebrospinal fluid testing in febrile but otherwise well-appearing babies during the second month of life, especially when initial testing suggests that a urinary tract infection is the (or a) cause of the fever. Choosing not to look for meningitis in these babies carries the risk of under-treating a central nervous system infection and leaving the baby with lasting adverse outcomes. However, there are risks to conducting “excessive” diagnostic testing, and the actual risk of meningitis co-occurring with urinary tract infection during the first months of life is unknown.
Thus, Wang and colleagues performed a secondary analysis of data within the Reducing Excessive Variability in Infant Sepsis Evaluation quality improvement project that is led by a network associated with the American Academy of Pediatrics. Overall, the study group’s goal is to increase the rate of appropriate evaluation and care for well-appearing infants ages 7 to 60 days who present with fever. This is an observational study, and the collection of spinal fluid samples was left to treating clinicians and was not mandated by the study group. A total of 124 hospitals participated in the study, including both academic and community institutions.
Patients who appeared “toxic,” “ill,” “sick,” or “lethargic” were not included in the study. Patients with underlying chronic conditions that might alter the risk of bacterial infection and children with obvious bronchiolitis were excluded. Thus, included patients were previously well and looked well (other than having fever) and did not have an obvious site of infection prior to laboratory analysis.
Included patients were classified by age (7-30 days vs. 31-60 days), evidence of probable urinary tract infection (white blood cells, nitrite, or leukocyte esterase at higher than normal levels in the urine), and the presence or absence of inflammatory markers (abnormal white blood cell count, C-reactive protein level, or procalcitonin level). Collected data also included whether the physician had treated a urinary tract infection or meningitis based on microbial growth on sampled fluids.
During the two years of the study, 20,570 well-appearing febrile infants 7-60 days of age were included in the study. Of these, 89% underwent a urinalysis prior to decisions about hospital admission or dismissal from the emergency department; 19% of the urinalyses were consistent with the likelihood of a urinary tract infection. Of those patients with presumed urinary tract infection, 70% underwent spinal fluid analysis (vs. 58% of those without an abnormal urine test). Of those with abnormal urinalyses, the younger patients (first month of life), males, and those with abnormal inflammatory markers were more likely to have cerebrospinal fluid testing.
The rate of spinal fluid testing varied significantly by study site. For children in the first month of life with abnormal urinalyses, rates of spinal fluid analysis varied by institution from 64% to 100%. For children in the second month of life with abnormal urinalyses, rates of spinal fluid analysis varied from 10% to 100%. Hospitals with the highest number of febrile infants were most likely to conduct spinal fluid testing.
There were 1,061 infants with a positive urinalysis who did not have spinal fluid evaluation. Of these, 734 received empiric antibiotics, and 505 received a full course of antibiotics designed to treat urinary tract infection. (The selection and dosing of antibiotics vary in babies with urinary tract infection vs. meningitis.) Of all the patients who did not have spinal fluid testing, none subsequently developed meningitis.
The authors summarized by noting that, for well-appearing febrile babies who had abnormal urinalyses, there was a wide variation in practices for performing tests to rule out concurrent meningitis. Spinal fluid testing was most common for babies younger than 1 month of age, for those with abnormal inflammatory markers, and for those seen at high-volume pediatric centers. However, 30% of febrile young infants still did not receive spinal fluid testing, and none of the non-tested babies had adverse outcomes. It seems that cerebrospinal fluid analysis might not be as necessary as some practitioners think.
The United States spends relatively more money on medical care than many other countries. Some think that this is because of an excessive reliance on expensive testing (instead of using clinical judgment) and because of a fear of missing rare possibilities. Indeed, there is a tension between settling for a near-certain diagnosis and ruling out less likely dangerous diagnoses. The authors of this new paper suggest that perhaps some of the current testing of otherwise well-appearing febrile young infants is not necessary. With significant practice variations, whether spinal fluid analysis was done for well-appearing febrile babies with evidence of a urinary tract infection was not associated with missed diagnoses of meningitis.
Of course, meningitis can co-occur with urinary tract infections. Children who appear ill and children who have underlying medical conditions would be less likely to forgo spinal fluid testing when febrile as young infants. However, otherwise healthy young infants who, other than being febrile, appear well and have abnormal urinalyses should be able to be treated safely for the urinary tract infection without conducting additional testing to rule out meningitis.