Osteomyelitis/Septic Arthritis Caused by Kingella kingae Among Day Care Attendees --- Minnesota, 2003
Kingella kingae is a fastidious gram-negative coccobacillus that colonizes the respiratory and oropharyngeal tract in children. K. kingae occasionally causes invasive disease, primarily osteomyelitis/septic arthritis in young children, bacteremia in infants, and endocarditis in school-aged children and adults (1--8). Although diagnosis of this organism frequently is missed, invasive disease is uncommon. Only sporadic, non-epidemiologically linked cases have been reported previously. In October 2003, the Minnesota Department of Health (MDH) investigated a cluster of two confirmed cases and one probable case of osteomyelitis/septic arthritis caused by K. kingae among children aged 17--21 months attending the same toddler classroom in a day care center. All reported within the same week with onset of fever, preceding or concurrent upper respiratory illness (URI), and refusal to bear weight on the affected limb. This report summarizes these cases and describes the epidemiologic investigation of the day care center. The findings underscore the need for clinicians and laboratorians to consider K. kingae infection in young children with Gram stain--negative or culture-negative skeletal infections.
Case 1. In mid-October 2003, a boy aged 21 months was taken to his pediatrician after 6 days of worsening limp. He had a history of prematurity (32 weeks' gestation), reflux disease, reactive-airway disease, and eczema, but was otherwise healthy with no history of recent antibiotic use. Nine days before, he had an isolated temperature reaching 103º F (39.4º C), which resolved with acetaminophen. His white blood cell (WBC) count and erythrocyte sedimentation rate (ESR) were normal. A right hip radiograph showed an abnormality in the proximal femur, which was diagnosed as a possible fracture. The patient's limp deteriorated to a crawl, and he underwent surgery 7 days after the initial examination. Femoral neck osteomyelitis and hip septic arthritis were diagnosed. At the time of surgery, his WBC count and ESR were elevated (17,200 cells/mm3 and 51 mm/hr, respectively [normal ranges: 6,000--17,000 cells/mm3 and 0--10 mm/hr, respectively]) and C-reactive protein (CRP) was normal (<0.5 mg/dL). Gram stain of synovial fluid showed several WBCs but no organisms; synovial fluid and bone cultures were positive for K. kingae 5 days later.
Case 2. In mid-October, a previously healthy girl aged 20 months who had completed a 14-day course of amoxicillin/clavulanic acid for otitis media in early October was reported with irritability, refusal to bear weight on her right foot, and a warm right ankle. During the next few days, she had a temperature reaching 101.6º F (38.7º C), and her right ankle became swollen and red. An MRI revealed fluid in her ankle joint. She had a normal WBC count and slightly elevated ESR (38 mm/hr). Three days later, she underwent surgery for ankle and subtalar septic arthritis. Gram stain of synovial fluid was negative, but K. kingae was identified from culture 4--5 days later.
Case 3. In mid-October, a previously healthy boy aged 17 months with recent but transient neutropenia related to viral illness, was taken to the emergency department with irritability, a limp of 2 days' duration, and a temperature of 102º F (38.9º C). His right ankle was moderately warm and swollen, and radiographs were normal. His WBC was high-normal (12,500 cells/mm3), and his CRP was slightly elevated (1.27 mg/dL). He had synovitis and otitis media diagnosed and was administered a 7-day course of oral amoxicillin. Although his clinical status improved, he continued to limp after foot manipulation; MRI of his right ankle 16 days later revealed evidence of distal tibial osteomyelitis. No specimens were obtained.
The day care staff members were interviewed, and a site visit was performed; absentee, illness, and biting records were examined. No obvious outbreak source was identified, and no unusual practices or behaviors were noted that could explain this cluster of illnesses. Oropharyngeal cultures were obtained from center staff and from children aged 0--5 years to assess K. kingae colonization. Of 122 children, 115 (94%) were cultured; 16 (14%) children were colonized with K. kingae. The highest prevalence occurred in the toddler class (nine [45%] of 20 tested positive; patients 1--3 tested negative, but all had received antibiotics). The remaining seven colonized children were distributed among four classes of older children; two were siblings of colonized children in the toddler class. No staff or children aged <16 months were colonized.
Pulsed-field gel electrophoresis (PFGE) of the K. kingae isolates from the two confirmed patients and from 15 of the 16 colonized children (nine toddlers and six from older classes) demonstrated indistinguishable PFGE patterns. Antimicrobial susceptibility testing revealed a minimum inhibitory concentration (MIC) of 0.047--0.125 µg/mL to rifampin and a MIC of 0.004--0.047 µg/mL to penicillin.
All children (n = 20) and staff (n = six) in the toddler classroom received a prophylactic 2-day course of rifampin, and oropharyngeal cultures were collected again 10--14 days later. Of the nine toddlers colonized originally, three (33%) remained positive on reculture. An additional toddler, who was initially culture-negative, was positive for K. kingae on reculture.
Reported by: R Faville, MD, S Koop, MD, Gillette Children's Hospital, St. Paul; F Ogunmodede, MBBS, R Lynfield, MD, R Danila, PhD, B Juni, MS, D Boxrud, MS, A Glennen, E Shade, K Penterman, Minnesota Dept of Health. K Kiang, MD, EIS Officer, CDC.
This report describes the first reported cluster of epidemiologically linked cases of invasive K. kingae disease. The high incidence in the toddler class and the matching PFGE pattern are consistent with child-to-child transmission. This report also describes the first reported attempt to use rifampin to eliminate K. kingae carriage; this attempt proved to be moderately effective.
K. kingae constitutes part of the normal respiratory flora in children but can cause isolated cases of invasive disease, primarily osteomyelitis/septic arthritis (65%--75% of cases) in young children and bacteremia (20%--30% of cases) in infants (1,2,7,8). The majority of children who have invasive disease are previously healthy without immunosuppressive conditions; >90% are aged <2 years (1,2,5--8).
Invasive disease is associated frequently with concomitant or precedent URI or stomatitis (3,4); disrupted respiratory or buccal mucosa might facilitate bacterial invasion and hematogenous dissemination. Biting might be an alternative means of introducing oropharyngeal pathogens into the bloodstream.
The presence of K. kingae is difficult to detect without immediate clinical suspicion. Gram stain of synovial fluid shows WBCs but frequently is negative for organisms. Recovery of the organism in culture is difficult because of its fastidious nature, and might require laboratories to hold culture plates for up to 7 days. For cases described in this report, cultures were held longer than routine laboratory protocol recommends (usually 3 days) because an atypical organism was suspected. Studies of cases in Israel indicate that 40%--50% of culture-negative septic arthritis cases in children aged <2 years might be attributable to K. kingae (5,8). Inoculating synovial fluid or bony exudates directly into blood-culture bottles with a continuous monitoring system increases the rate of K. kingae recovery substantially, compared with direct plating of specimens on solid media (5,8). The increased awareness and enhanced capability of laboratories to isolate this organism might lead to an observed increase in incidence of K. kingae invasive disease.
Although limited data are available about the epidemiology and transmission of K. kingae, the organism most likely is transmitted through respiratory secretions and saliva. In one study of an Israeli day care center, the monthly prevalence of K. kingae colonization ranged from 6% to 35%, and approximately 70% of children were colonized at some point during the 11-month study period. No invasive disease was observed (9). Subtyping by PFGE, immunoblotting, and ribotyping of the isolates demonstrated that children were colonized continuously, or intermittently with different subtypes over weeks to months. Two distinct subtypes with temporal clustering represented approximately 75% of the isolates (10). In comparison, a cohort of epidemiologically unrelated cases showed substantially more subtype variability (10). These findings suggest person-to-person transmission within the facility.
The pattern of colonization and invasive disease described in this report is consistent with previous studies. The indistinguishable PFGE pattern of the isolates further indicates the person-to-person mode of K. kingae transmission among children who attend day care centers. The incidence of invasive disease was exceptionally high among these children. Further examination into potential risk factors and DNA sequencing of the day care K. kingae isolates are being conducted by MDH.
The findings in this report underscore the need for clinicians to suspect infection with K. kingae and other atypical organisms in young children with Gram stain--negative or culture-negative skeletal infections and for laboratorians to perform appropriate laboratory diagnostic testing. The use of blood-culture bottles for inoculation and cultivation of synovial fluid/bone tissue and the incubation of culture plates for >1 week might increase the diagnosis of pediatric skeletal infections attributed to K. kingae.
This report is based on contributions by P Yagupsky, Clinical Microbiology Laboratory and Pediatric Infectious Disease Unit, Soroka Univ Medical Center, Beer-Sheva, Israel. K Marbin, Gillette Children's Hospital, St. Paul, Minnesota. J Jorgenson, Health Science Center, Univ of Texas-San Antonio, Texas. F Tenover, D Jernigan, Div of Healthcare Quality Promotion; M Fischer, Div of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, CDC.
**Questions or messages regarding errors in formatting should be addressed to email@example.com.
Page converted: 3/25/2004
This page last reviewed 3/25/2004