With the optimization of conventional culture methods1 and current molecular techniques, Kingella kingae now seems to have supplanted Staphylococcus aureus as the primary pathogen of arthritis in children, especially in those less than 3 years of age in some countries.2–4
Septic arthritis caused by S. aureus can be severe,5,6 and the emergence of community-acquired methicillin-resistant S. aureus isolates has rendered presumptive antibiotic treatment more difficult.5 In contrast, K. kingae infections seem to be mild with few complications7 and almost all isolates are susceptible to beta-lactam antibiotics.8
To date, no published studies have compared the initial clinical and biologic characteristics of patients with S. aureus and K. kingae septic arthritis in a large series of children. We conducted a retrospective study to describe and compare patients with K. kingae and S. aureus arthritis at our institution.
PATIENTS AND METHODS
Patients and Initial Diagnosis.
After approval by the institutional review board, the microbiology laboratory register of the Robert Debré Hospital, a tertiary referral center for pediatric orthopedics, was used to identify all patients with positive culture or molecular diagnosis of S. aureus or K kingae from synovial fluid between 2000 and 2009. Microbiologic methods for the diagnosis were as previously described.2 We included children with the following criteria: age less than 18 years; hospitalized for S. aureus or K. kingae arthritis with pain and/or limited limb movement and/or joint effusion with onset of less than 14 days and treated on the first day following diagnosis by surgical drainage and irrigation of the joint (arthrotomy or arthroscopy or joint aspiration); and antibiotic therapy for which the strain was susceptible. We excluded cases with associated osteomyelitis or subperiosteal abscess, or following traumatic wound or surgical joint injury.
Patient Characteristics, Biologic Data, and Follow-up.
On a standardized form, clinical characteristics and laboratory findings were recorded at diagnosis and at 3, 5, and 7 days after start of treatment. To evaluate the severity of the disease, we also recorded the duration of hospital stay and antibiotic therapy, as well as certain adverse events during treatment (eg, need for a second surgical procedure for drainage and/or hospitalization in intensive care unit for sepsis).
Medians were compared using Student t-test or Mann-Whitney U test depending on normality and group size. Frequencies were compared by Fisher exact test or χ2 test depending on group size. The values of P < 0.05 were considered to denote significant difference.
We identified 109 patients with K. kingae (n = 68) and S. aureus (n = 41) arthritis. Four patients with K. kingae arthritis were excluded (3 missing clinical files, 1 osteomyelitis associated). Fifteen patients with S. aureus infection were excluded (4 missing clinical files, 8 osteomyelitis or subperiosteal abscess associated, 2 patients had intervals between surgery and antibiotic therapy longer than 24 hours, 1 patient was symptomatic for 1 year). We analyzed data of 90 patients, including 64 with K. kingae arthritis among whom 31 were included in a previous study,2 and 26 with S. aureus arthritis. Among the 64 cases of K. kingae arthritis, 9 were positive for joint fluid culture; the 55 others cases, for which joint fluid culture was sterile, were diagnosed by real-time polymerase chain reaction.
Demographic Characteristics, Initial Clinical, and Biologic Features.
The major demographic and initial clinical and laboratory features of patients are described in Table 1. Children with K. kingae arthritis were significantly younger. At admission, none of the criteria examined was significantly different between K. kingae and S. aureus arthritis, except platelet count (Table 1). In patients with K. kingae arthritis, the maximal initial C-reactive protein (CRP) value was 151 mg/L, whereas 7/26 patients (26.9%) with S. aureus arthritis had an initial CRP higher than this value, to a maximum of 650 mg/L. However, the rate of normal CRP values (≤10 mg/L) for K. kingae arthritis (5/63 = 7.9%) was lower than that for S. aureus arthritis (5/21 = 23.8%), but was not significantly different (Fig., Supplemental Digital Content 1, http://links.lww.com/INF/A870). The 2 most common joints infected were similar (knees in 57.8% vs. 38.5% and hips in 23.4% vs. 38.5% in K. kingae and S. aureus arthritis, respectively).
Surgical procedures of the 2 groups were comparable including arthrotomy (49/64 vs. 16/26), arthroscopy (8/64 vs. 7/26), and joint aspiration with lavage (7/64 vs. 3/26). For all episodes, isolates were susceptible to the initial treatment. Notably, 2 methicillin-resistant S. aureus infections were initially treated with cefotaxime and fosfomycin, and vancomycin and rifampin, respectively.
Median CRP values were significantly lower in the K. kingae group at day 3, 5, and 7 compared with those of S. aureus (Fig., Supplemental Digital Content 1, http://links.lww.com/INF/A870). Median fibrinogen values were also lower for K. kingae arthritis compared with S. aureus arthritis at day 3 (5.63 vs. 7.26 g/L, respectively; P < 0.001) and at day 7 (4.55 vs. 6.01 g/L, respectively; P < 0.001). Other biologic features were not significantly different between the 2 groups at days 3 and 7. The mean duration of fever and of hospital stay (days) were markedly shorter in the K. kingae group compared with the S. aureus group (0.2 [0–3] vs. 3.5 [0–27] (P = 0.0001) and 6.6 [3–11] vs. 13.8 [7–39] (P = 0.0001), respectively). A total of 9 (34.6%) of the 26 patients with S. aureus arthritis versus 1 of 64 (1.6%) with K. kingae arthritis had fever longer than 2 days. Complications were more frequent in the S. aureus group (6/26 vs. 1/64; P = 0.002). Because of clinical relapse of the arthritis, 1 (1.6%) patient with K. kingae arthritis needed a second surgical drainage 10 days after treatment initiation. Adverse events were experienced by 6 (23%) patients with S. aureus arthritis. Of them, 4 needed a second surgical drainage, 1 required transfer to the intensive care unit, and 1 needed both.
The mild to moderate clinical course of K. kingae arthritis has been previously emphasized.7 In our study, patients with K. kingae arthritis exhibited a mild clinical and biologic picture with medians of fever, WBC count, and CRP very close to those reported in the largest previously published study.7
However, we observed no significant difference between K. kingae and S. aureus groups at initial presentation, except age. The higher platelet counts observed in patients with K. kingae arthritis was likely related to their younger age.9 Neither clinical signs nor the biologic markers we observed appear consistently helpful to discriminate the initial presentation of K. kingae from that of S. aureus arthritis in children. Nonetheless, although medians of CRP were similar in both groups, we observed a wider range of CRP values in S. aureus arthritis. An initial CRP value higher than 150 mg/L, a value similar to the highest CRP (170 mg/L) found by Dubnov-Raz et al,7 might help to eliminate K. kingae as the causative agent of septic arthritis. Nevertheless, our data also clearly indicate that at the initial stage, a low or a normal initial CRP cannot be considered as indicative of K. kingae arthritis and cannot eliminate S. aureus arthritis. We observed a trend toward higher WBC counts in synovial fluid infected by K. kingae than by S. aureus. These results indicate that the WBC count in joint fluid may not contribute to etiologic diagnosis and that the benign outcome of K. kingae arthritis might not be related to lower local inflammatory response.
In contrast to similar initial presentations, several factors such as fever, CRP, and fibrinogen return to normal more rapidly in K. kingae than in S. aureus arthritis. Among patients infected with S. aureus, 35.5% in the study of Carrillo-Marquez et al5 and 34.6% in our study were still febrile for more than 2 days after beginning treatment, versus 1.6% of our patients infected with K. kingae. Of note, 61.9% (13/21) of the S. aureus group had an increased or stable CRP 3 days after start of treatment, in contrast with a decrease in CRP observed in 78.2% (43/55) of the K. kingae group. The difference in the early course of CRP value could not be attributed to a difference in surgical treatments, which were comparable in both groups. The delay of CRP decrease that we observed in S. aureus arthritis was consistent with the study of Peltola et al.10
Finally the lesser virulence of K. kingae compared with S. aureus in arthritis was clearly evidenced by shorter hospital stays and fewer adverse events.
This largest series in Europe confirms the leading rank of K. kingae in septic arthritis in children and the outstanding benefit of molecular methods for its diagnosis. It provides the first evidence that although K. kingae arthritis is less virulent than S. aureus arthritis in terms of clinical and biologic course, no criteria may be consistently used to distinguish these pathogens at initial presentation. The sole exception may be a CRP higher than 150 mg/L, which was only seen in some of the children with S. aureus arthritis. Presumptive antibiotic treatment of septic arthritis in young children should take into account both these pathogens.
1. Yagupsky P. Kingella kingae
: from medical rarity to an emerging paediatric pathogen. Lancet Infect Dis
2. Ilharreborde B, Bidet P, Lorrot M, et al. New real-time PCR-based method for Kingella kingae
DNA detection: application to samples collected from 89 children with acute arthritis. J Clin Microbiol
3. Chometon S, Benito Y, Chaker M, et al. Specific real-time polymerase chain reaction places Kingella kingae
as the most common cause of osteoarticular infections in young children. Pediatr Infect Dis J
4. Ceroni D, Cherkaoui A, Ferey S, et al. Kingella kingae
osteoarticular infections in young children: clinical features and contribution of a new specific real-time PCR assay to the diagnosis. J Pediatr Orthop
5. Carrillo-Marquez MA, Hulten KG, Hammerman W, et al. USA300 is the predominant genotype causing Staphylococcus aureus septic arthritis
in children. Pediatr Infect Dis J
6. Dohin B, Gillet Y, Kohler R, et al. Pediatric bone and joint infections caused by Panton-Valentine leukocidin-positive Staphylococcus aureus
. Pediatr Infect Dis J
7. Dubnov-Raz G, Ephros M, Garty BZ, et al. Invasive pediatric Kingella kingae
infections: a nationwide collaborative study. Pediatr Infect Dis J
8. Yagupsky P, Katz O, Peled N. Antibiotic susceptibility of Kingella kingae
isolates from respiratory carriers and patients with invasive infections. J Antimicrob Chemother
9. Lainey E, Boirie M, Fenneteau O. Hémogramme en pédiatrie: variations physiologiques. Rev Francoph Lab
10. Peltola H, Paakkonen M, Kallio P, et al. Prospective, randomized trial of 10 days versus 30 days of antimicrobial treatment, including a short-term course of parenteral therapy, for childhood septic arthritis
. Clin Infect Dis