Non-Invasive Test May Transform Infant Meningitis Screening

TOPLINE:

A high-resolution ultrasound-based device demonstrated excellent accuracy in detecting elevated white blood cell (WBC) counts in cerebrospinal fluids, showing promise as a non-invasive screening tool for meningitis in neonates and infants with a permeable fontanel and potentially offering an alternative to lumbar puncture.

METHODOLOGY:

• The diagnosis of meningitis requires an invasive lumbar puncture to collect cerebrospinal fluid for detecting elevated WBC counts; however, this procedure carries inherent risks and practical limitations.
• Researchers conducted a prospective study across hospitals in Spain, Mozambique, and Morocco to validate the safety and accuracy of Neosonics — a non-invasive high-resolution ultrasound-based device that can detect signals from WBCs in the cerebrospinal fluid via the anterior fontanel — in neonates and infants younger than 24 months with suspected meningitis against the gold-standard techniques.
• High-resolution images of the cerebrospinal fluid were captured and used to train a deep learning model that identified and classified cerebrospinal fluid patterns on the basis of lumbar puncture-confirmed WBC counts, with ≥ 30 cells/mm³ defined as a case.
• The accuracy of the trained algorithm was tested on 3782 high-resolution images of 76 patients.

TAKEAWAY:

• The Neosonics device correctly identified 17 of 18 meningitis cases and 55 of 58 controls participants without meningitis, yielding a sensitivity of 94.4% and a specificity of 94.8%.
• Overall, the device achieved an accuracy of 94.7% in detecting elevated WBC counts in the cerebrospinal fluid.
• Misclassified participants either had low WBC counts or were microbiologically confirmed cases without elevated WBC counts.

IN PRACTICE:

“The non-invasive WBC counter, can lead to a more accurate indication for LP [lumbar puncture] and/or indications of treatment, resulting in better patient outcomes and in a reduction of the risk of long-term sequelae, along with a likely more cost-effective use of resources in all settings,” the authors wrote.

SOURCE:

This study was led by Sara Ajanovic and Quique Bassat, ISGlobal, Barcelona, Spain, and Fernando Cabañas, Quirónsalud Madrid University Hospital, Madrid, Spain. It was published online on July 23, 2025, in Pediatric Research.

LIMITATIONS:

This study faced challenges in obtaining reliable gold-standard comparison results. The device’s precision decreased in participants with thicker fontanels, and it misclassified three participants with very narrow cerebrospinal fluid spaces. Moreover, the image acquisition time was suboptimal.

DISCLOSURES:

This study was supported, in whole or in part, by the Gates Foundation in Mozambique and Morocco; funded by Instituto de Salud Carlos III and co-funded by the European Union in Spain; and funded by the Spanish Agency of Cooperation in Morocco. Five authors were employees of Kriba, the company developing Neosonics. One author and the ISGlobal Biomedical Data Science team disclosed receiving honoraria, training fees, and consulting fees from Kriba.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.