Clinical

Accurately diagnosing kids' meningitis clinically remains challenging: report

Last Updated: 2010-10-26 10:40:14 -0400 (Reuters Health)

NEW YORK (Reuters Health) - In young children, no single isolated clinical feature is diagnostic of bacterial meningitis, and the best combination of clinical features to raise or lower suspicion of meningitis remains "unclear," clinicians from the US and Canada conclude in an article in the November issue of Pediatrics, available online now.

It's no secret that pediatric bacterial meningitis can be hard to diagnose clinically, particularly in young infants who often don't have the classic features of the disease, first author Dr. Sarah Curtis of University of Alberta, Edmonton, Canada, and colleagues note in the article.

"Ideally, primary clinical assessment should provide an estimate of the probability of disease and help to determine if further diagnostic testing is required," they write. However, to their knowledge, no one has systematically reviewed prospective data pertaining to clinical features suggestive of meningitis.

Dr. Curtis and colleagues set out to do this. They identified 14,145 studies in a structured literature search and found 10 that met their inclusion criteria. The articles contained historical and physical features of children with lumbar puncture-confirmed bacterial meningitis.

The authors calculated sensitivity, specificity and likelihood ratios (LRs) for signs and symptoms of bacterial meningitis in the nearly 5,000 children included in the studies.

On history, they found that the likelihood of meningitis increased about 8-fold when the child had a bulging fontanel or a stiff neck (LR 8.00 and 7.70, respectively). A seizure outside of the typical age range for febrile seizure increased the likelihood of meningitis more than 4-fold (LR 4.40), while a history of reduced feeding "only somewhat" increased the likelihood of meningitis (LR 2.00).

On clinical examination, factors that independently raised the likelihood of meningitis (in descending order) were jaundice (LR 5.90), being toxic or moribund (LR, 5.80), having meningeal signs (LR 4.50), neck stiffness (LR 4.00), bulging fontanel (LR 3.50), Kernig sign (LR 3.50), tone up (LR 3.20), fever > 40 degrees Celsius (LR 2.90), and Brudzinski sign (LR 2.50).

The absence of meningeal signs and an abnormal cry independently reduced the likelihood of meningitis (LR 0.41 and 0.30, respectively). The absence of fever didn't rule out meningitis.

The authors also note that while petechiae was a "strong" sign of meningitis (LR 37.00), it was "surprisingly" only examined in one small prospective study and only four patients displayed this feature. "Thus, the relevance of this well-known feature is currently uncertain."

Based on their review, the authors say the presence of a combination of meningeal signs, bulging fontanel, and a high fever raises an infant's probability of meningitis to 84%.

"Although the presence or absence of these findings, in combination or separately, hardly confirms or refutes a diagnosis of meningitis, they raise the probability high enough that a lumbar puncture must be performed," they say.

Nonetheless, the authors say, "the ideal clinical model for pediatric meningitis is still unclear, and prospective evaluation and validation of known and new prediction models in varying populations are imperative."

They also advise clinicians to "remain prudent" when making decisions regarding possible meningitis in young infants because several investigators noted infants with meningitis who displayed few or no classic features of the disease.

SOURCE: http://link.reuters.com/quq32q

Pediatrics 2010;126:952-960.