This article is more than 5 years old. Information may no longer be current.
Lab markers may flag knee septic arthritis
Some laboratory criteria may be used to identify those children who may be at risk for knee septic arthritis, according to study data published online.
Study researchers Julia K. Deanehan, MD, and Lise E. Nigrovic, MD, MPH, of the division of emergency medicine, Boston Children’s Hospital, and colleagues published data on a retrospective sample of children with knee monoarthritis who resided in two Lyme disease endemic areas. The researchers derived and validated a clinical prediction rule to distinguish between arthritis types using available peripheral laboratory results.
Julia K. Deanehan
The researchers reported that children with a peripheral blood absolute neutrophil count <10 x 103 cells/mm3 and an erythrocyte sedimentation rate <40 mm/hr were at low risk for septic arthritis.
“Septic arthritis was very rare in our study population,” Deanehan told Infectious Diseases in Children. “In Lyme-endemic areas, children with knee monoarthritis in our low-risk group and may not require diagnostic arthrocentesis.”
In an accompanying blog to the study, Pediatrics Editor Lewis First, MD, wrote: “While it is well known that Lyme disease can cause a monoarthritis, we often worry that a hot knee may mean more than that and can raise suspicion for a bacterial arthritis. Perhaps clues outside of joint fluid could preclude our need to tap the hot knee, but until now, that hypothesis had not been studied.”
Lise E. Nigrovic, MD, MPH, can be reached at the Division of Emergency Medicine, Children’s Hospital, Boston, 300 Longwood Ave., Boston, MA 02115.
Disclosure: Nigrovic and Deanehan report no relevant financial disclosures.
Perspective
Back to Top
The investigators sought to identify factors in children with arthritis of the knee that could be used to distinguish children with Lyme arthritis from those with bacterial arthritis due to organisms other than Borrelia burgdorferi. If such factors could be identified, they could be used to prevent children with Lyme arthritis from undergoing unnecessary invasive procedures, such as arthrocentesis or surgical drainage of the knee.
This study found that, among children in Lyme-endemic areas who present with monoarthritis of the knee, there was substantial overlap in the presenting features. However, all children who had bacterial arthritis due to organisms other than B. burgdorferi had both an absolute neutrophil count >10,000/mm3 and an erythrocyte sedimentation rate >40 mm/hr. The sensitivity of this combination of predictors of bacterial arthritis was 100% (19/19), although its specificity was poor (<60% overall).
The authors of the paper conclude that children with an absolute neutrophil count <10,000/mm3 and an erythrocyte sedimentation rate <40 mm/hr are unlikely to have bacterial arthritis due to organisms other than B. burgdorferi, and that these invasive procedures may not be necessary.
The goal of the authors is sensible; every child with a swollen knee and possible infection does not need arthrocentesis — certainly it is rarely useful in Lyme arthritis. I have seen many instances of unnecessary surgical drainage of the knee under anesthesia in children who turn out to have Lyme disease. The tricky part is to not allow any children with bacterial arthritis not due to Lyme disease to go undiagnosed.
The authors are appropriately cautious and delineate the limitations of their study (in particular the relatively small number of children with bacterial arthritis other than Lyme disease). They call for additional studies before this rule is adopted as routine. I agree and think that this is a decent study (though certainly not earth-shaking). Most of the authors are surgeons and, even though published in Pediatrics, the findings are particularly directed at surgeons and ED docs — many pediatricians are already well aware of this issue and do not routinely tap every swollen knee in Lyme endemic areas.