Routine Lumbar Puncture in Infants Under Review

Preventing missed diagnoses of serious bacterial infections in febrile infants while minimizing unnecessary invasive procedures, such as lumbar puncture, and avoiding unnecessary antibiotic use remains a major goal in pediatric care.Advances in diagnostic tools and a growing understanding of biomarkers are helping clinicians to achieve a better balance between patient safety and overtreatment.During the Congress of Pediatric Societies in Paris, France, and in an interview with Medscape Medical News, Christèle Gras-Le Guen, MD, PhD, head of the Women, Children and Adolescents Division at Nantes University Hospital in Nantes, France, said, “The challenge is to identify, among the many children who present with fever as their only symptom, those who are at higher risk.”To address this challenge, the multicenter DIAFEVER trial published in 2026 validated a risk-stratification algorithm for managing infants with isolated fever.“Our randomized controlled trial was designed to determine whether clinical management could be simplified without compromising patient safety,” said Gras-le-Guen, who led the DIAFEVER study. “Nearly 5000 children presenting to the emergency department with recent isolated fever between 2018 and 2020 were enrolled across approximately 30 centers.”The study found no increase in adverse events or missed diagnoses, while showing a substantial reduction in lumbar punctures, venipunctures, and antibiotic prescriptions.Biomarker EvaluationBiomarkers that may support the clinical evaluation of isolated fever in young children include biochemical markers, particularly procalcitonin (PCT) and C-reactive protein (CRP), as well as imaging studies and microbiologic tests used to detect viral and bacterial pathogens.“Interpreting protein biomarkers requires an understanding of their kinetics, which vary according to the duration of fever,” Gras-le-Guen said. “For example, PCT levels begin to increase about 6 hours after fever onset, peaks between 12 and 24 hours, and then declines rapidly.”CRP follows a similar pattern but rises more slowly, increasing over approximately 8 hours and reaching peak concentrations around 72 hours.PCT was the principal biomarker in the DIAFEVER trial. The risk-stratification algorithm was displayed in participating emergency departments using a color-coded system that identified high-risk (red), intermediate-risk (orange), and low-risk (green) patients, together with corresponding management recommendations.Gras-le-Guen cited the example of children younger than 1 month of age. The first step involved an initial clinical evaluation to determine whether the child’s condition was a cause for concern. If the clinical presentation was concerning, a lumbar puncture was performed. In the absence of signs of severity, urinary leukocytes were tested, followed by a PCT test, if negative. When the off-site PCT result was negative, a lumbar puncture was not indicated, and the child was kept under observation for 48 hours. In cases of early presentation, particularly within 6 hours of fever onset, which applies to most children, a second PCT measurement was performed after 6 hours to assess the biomarker’s kinetics. “This algorithm made it possible to identify a significant proportion of children who could be managed without a lumbar puncture when their clinical condition was reassuring and the PCT result was negative,” she said.Reduced InterventionsAmong children younger than 3 months of age with fever who did not present to the emergency department, the rate of antibiotic treatment was 38% vs 26% in the “usual care” group vs the “DIAFEVER algorithm” group. This represented a relative reduction of approximately 30%, with no difference in adverse events, particularly severe morbidity and death, and no evidence that the use of the algorithm increased the risk.The algorithm was also associated with fewer invasive procedures, including venipuncture, urinary catheterization, and lumbar puncture. The reduction was particularly marked among infants, with a 35% decrease in lumbar punctures.Emerging BiomarkersIn addition to CRP and PCT, new biomarkers are emerging. One example is the MeMed Key platform, which measures three inflammatory proteins — TNF-related apoptosis-inducing ligand, interferon-gamma-induced protein 10, also known as CXCL10, and CRP — to estimate the likelihood of bacterial infection.Researchers in Switzerland have also developed transcriptomic approaches on the basis of host messenger RNA expression profiles that distinguish viral from bacterial infections.Rapid multiplex polymerase chain reaction (PCR) platforms represent another promising development. Systems such as SPOTFIRE can generate results from a noninvasive anterior nasal swab in approximately 16 minutes.The assay detects 13 respiratory pathogens, including 10 viral targets (adenovirus, influenza A (H1), influenza A(H3), influenza B, parainfluenza viruses, rhinovirus/enterovirus, respiratory syncytial virus, human metapneumovirus, seasonal coronaviruses, and SARS-CoV-2) and three bacterial targets (Mycoplasma pneumoniae, Chlamydia pneumoniae, and Bordetella pertussis/parapertussis).This technology was evaluated in the OPTIPAC trial, a randomized trial conducted in 11 pediatric emergency departments in France that enrolled 499 children with community-acquired pneumonia. The children received either standard care with syndromic multiplex PCR or standard care alone. The primary endpoint, appropriate initial antimicrobial therapy, was achieved in 68.6% of children in the PCR group compared with 48.2% in the standard care group, representing an absolute improvement of 20.4 percentage points (relative risk, 1.42; 95% CI, 1.22-1.66; P < .0001).According to Gras-le-Guen, “Combined protein biomarkers, transcriptomic signatures, and multiplex PCR offer a promising approach for accelerating clinical decision-making and tailoring patient care. However, their use requires careful oversight because of their cost (approximately 100 euros for certain multiplex PCR assays) and the risk of inappropriate use.”Routine Reconsidered“The end of the practice of routinely performing lumbar puncture in cases of fever in febrile newborns is part of a broader shift in clinical practices,” Gras-le-Guen said. “The epidemiology of serious infections has changed since the 1990s, and both our understanding and our diagnostic tools have advanced.”For example, cerebrospinal fluid analysis, which was previously performed routinely in infants younger than 1 month of age, is increasingly being reserved for high-risk infants or those younger than 22 days of age. This trend suggests that the routine use of lumbar puncture should be reconsidered, provided that indications remain strictly defined. The priority is to avoid missed diagnoses, particularly bacterial meningitis.Other countries have adopted similar risk-stratification strategies. Canada, for example, published recommendations in 2024 on the basis of the same principle of reducing unnecessary testing and antibiotic use while maintaining patient safety.Gras-le-Guen said, “We invite all expert groups involved in the management of fever in children, as well as other organizations with expertise in this field, to participate in the development of new recommendations under the auspices of the French Society of Pediatrics. These recommendations are expected to be completed within 12-18 months.”She added that the findings of the DIAFEVER study, together with future research on this topic, will be taken into account during the development of the recommendations.Gras-le-Guen reported having no relevant conflicts of interest.This story was translated from Medscape’s French edition.