ORIGINAL STUDIES Contribution of a Broad Range Polymerase Chain Reaction to the Diagnosis of Osteoarticular Infections Caused by Kingella kingae Description of Twenty-four Recent Pediatric Diagnoses Isabelle Verdier, DPharm,* Ange `le Gayet-Ageron, MD,‡ Christine Ploton, MD,* Patricia Taylor, MD,† Yvonne Benito, Ingenior,§ Anne-Marie Freydiere, DPharm,* Franck Chotel, MD,† Je ´ro ˆme Be ´rard, MD, PhD,† Philippe Vanhems, MD, PhD,‡ and Franc ¸ois Vandenesch, MD, PhD§ Background: Microbiologic diagnosis of septic arthritis and osteo- myelitis in children is hindered by the less than optimal yield of blood and osteoarticular fluid cultures. Patients and Methods: All patients admitted to a pediatric unit for osteoarticular infections (OAI) between January 2001 and February 2004 were enrolled in this prospective study. Osteoarticular fluid and biopsy samples that were negative by conventional culture were tested by polymerase chain reaction (PCR) with universal 16S ribosomal DNA primers. Results: We enrolled 171 children. Culture was positive in 64 cases (37.4%), yielding Kingella kingae in 9 cases. The 107 culture- negative specimens were tested by 16S ribosomal DNA PCR. Fifteen samples (14%) were positive, all for Kingella DNA se- quences. K. kingae was the second cause of OAI in this population (30.4%), after Staphylococcus aureus (38%). Patients with Kingella infection diagnosed by culture (9 cases) did not differ from those diagnosed by PCR (15 cases) in terms of their clinical characteristics (including prior antibiotic therapy). The characteristics of the 24 children with arthritis (n = 17) or osteomyelitis (n = 7) were similar to those reported elsewhere. Fever (38°C) and symptom onset shortly before hospitalization (median, 4.5 days) were significantly associated with arthritis. Conclusion: Use of molecular diagnostic methods increases the identification of K. kingae in osteoarticular infections. Key Words: osteoarticular infections, Kingella kingae, polymerase chain reaction, molecular diagnosis (Pediatr Infect Dis J 2005;24: 692– 696) P rimary osteoarticular infections (OAI), which include ar- thritis and osteomyelitis, are far from rare in children. The main etiologic agents of OAI, which are cultured in fewer than 40% of cases, are Gram-positive cocci such as Staphy- lococcus aureus and -hemolytic streptococci, and Gram- negative rods such as Haemophilus influenzae and Kingella kingae. 1,2 K. kingae is involved in 14% of cases of OAI in which a microorganism is isolated. 3–5 The organism was first de- scribed in the early 1960s and typically colonizes the upper respiratory tract. 6 It can cause septicemia, endocarditis and bone and joint disease, mostly in infants and children. 7–18 This fastidious bacterium is difficult to isolate on solid medium. Rapid inoculation of clinical specimens into en- riched blood culture systems enhances the recovery rate. 19 –21 Inoculation on agar medium during surgery has been pro- posed as an alternative or complementary method. 22 Synovial fluid and bone biopsy specimens nonetheless remain culture- negative in a substantial proportion of patients with OAI. The place of Kingella in OAIs might therefore be underesti- mated. 5,23 Some authors reported the detection of K. kingae in culture-negative specimens by mean of polymerase chain reaction amplification (PCR). 24,25 The main objective of this study was to evaluate the incidence of K. kingae OAI in a series of consecutive hospi- talized children. All microorganisms were identified by cul- ture or by using a universal eubacterial 16S ribosomal DNA (rDNA) PCR method validated on other clinical specimens in our laboratory. 26 –28 A secondary objective was to describe the clinical, biologic and radiologic features of patients with OAI caused by K. kingae. PATIENTS AND METHODS Patients and Definitions. The study was based on a prospec- tive cohort of all patients admitted for OAI to Debrousse Pediatric Hospital (Lyon, France) during the 3-year period from January 1, 2001 to January 31, 2004. OAI was suspected in children with fever (temperature 38°C) or chills, bone pain and/or limping and/or limited limb movement. The diagnosis of arthritis or osteomyelitis Accepted for publication March 10, 2005. From *Laboratoire de Bacte ´riologie and †Service de Chirurgie Pe ´diatrique, Ho ˆpital Debrousse, ‡Laboratoire d’E ´ pide ´miologie et Sante ´ Publique, Universite ´ Claude Bernard, and §Laboratoire de Bacte ´riologie, Ho ˆpital Cardiologique Louis Pradel, Lyon, France Address for reprints: F. Vandenesch, MD, PhD, Laboratoire de Bacte ´riolo- gie, Ho ˆpital Cardiologique Louis Pradel, 28 Avenue Doyen Lepine, 69394 Lyon Cedex 03. Fax 33 4 72 35 73 35; E-mail denesch@ @univ-lyon1.fr. Copyright © 2005 by Lippincott Williams & Wilkins ISSN: 0891-3668/05/2408-0692 DOI: 10.1097/01.inf.0000172153.10569.dc The Pediatric Infectious Disease Journal • Volume 24, Number 8, August 2005 692