Differentiating Kingella kingae Septic Arthritis of the Hip from Transient Synovitis in Young Children Pablo Yagupsky, MD1, Gal Dubnov-Raz, MD, MSc2,3, Amadeu Gene, MD4, and Moshe Ephros, MD5, on behalf of the Israeli-Spanish Kingella kingae Research Group* Objective To conduct a retrospective multicenter study to assess the ability of a predictive algorithm to differentiate between children with Kingella kingae infection of the hip and those with transient synovitis. Study design Medical charts of 25 Israeli and 9 Spanish children aged 6-27 months with culture-proven K kingae arthritis of the hip were reviewed, and information on the 4 variables included in the commonly used Kocher prediction algorithm (body temperature, refusal to bear weight, leukocytosis, and erythrocyte sedimentation rate) was gathered. Results Patients with K kingae arthritis usually presented with mildly abnormal clinical picture and normal serum levels of or near-normal acute-phase reactants. Data on all 4 variables were available for 28 (82%) children, of whom 1 child had none, 6 children had 1, 13 children had 2, 5 had 3, and only 3 children had 4 predictors, implying #40% probability of infectious arthritis in 20 (71%) children. Conclusions Because of the overlapping features of K kingae arthritis of the hip and transient synovitis in children younger than 3 years of age, Kocher predictive algorithm is not sensitive enough for differentiating between these 2 conditions. To exclude K kingae arthritis, blood cultures and nucleic acid amplification assay should be performed in young children presenting with irritation of the hip, even in the absence of fever, leukocytosis, or a high Kocher score. (J Pediatr 2014;165:985-9).
Y
oung children who come to medical attention because of limping pose a common and serious diagnostic challenge, and differentiating between septic arthritis of the hip and transient synovitis is of paramount clinical and therapeutic importance. If inadequately diagnosed and treatment delayed, pediatric septic arthritis may result in major morbidity, including joint destruction, necrosis of the femoral head, and permanent disability.1 Transient synovitis, in contrast, is a benign and self-limited condition requiring only rest and analgesic therapy, and does not lead to orthopedic sequelae.2 Erroneous misclassification of transient synovitis as septic arthritis may result in unnecessary hospitalization, surgery, and antibiotic therapy. To assist in differentiating between these 2 conditions, an algorithm combining clinical information and laboratory data was developed by Kocher et al.3 This predictive guideline is based on clinical criteria (body temperature >38.5 C and refusal to bear weight) and blood inflammatory markers (white blood cell [WBC] count >12 000 cells/mm3 and erythrocyte sedimentation rate [ESR] >40 mm/h), complemented by the presence of >50 000 WBCs/mm3 in synovial fluid, when joint aspiration is performed.3 The risk of suppurative arthritis directly correlates with increasing number of positive predictors, and transient synovitis is characterized by low scores. The performance of this diagnostic algorithm and similar guidelines based on Kocher original clinical prediction model with slight modifications has been evaluated and validated in many studies that included pediatric patients with bacterial arthritis caused by traditional pathogens such as Staphylococcus aureus, Streptococcus pyogenes, Streptococcus pneumonia, and Haemophilus influenzae type b.3-8 As the result of increasing use of improved culture methods and sensitive nucleic acid amplification (NAA) assays, K kingae, a gram-negative component of the normal pharyngeal flora, is increasingly recognized as an important human pathogen and the most common etiology of joint infections in children aged 6-48 months in widespread geographic populations.9-11 The clinical presentation of patients with K kingae arthritis is often subtle, requiring a high index of suspicion: the body temperature and levels of acute phase reactants may be From the Clinical Microbiology Laboratory, Soroka normal or only mildly elevated, and leukocytosis is frequently absent.9,12-15 K University Medical Center, Beer-Sheva, Israel; Edmond and Lily Safra Children’s Hospital, Chaim Sheba Medical kingae usually affects large weight-bearing joints and in a large multicenter study Center, Tel-Hashomer, Israel; Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel; Molecular comprising 140 children with culture-proven K kingae arthritis, the hip was Microbiology Department, University Hospital Sant Joan involved in 20 (14.3%) patients and represented the third most commonly infected de Deu, Barcelona, Spain; and Pediatric Infectious Diseases Unit, Carmel Medical Center, and the Faculty of site in this series.15 Although Kocher decision-making algorithm has gained Medicine, Technion-Israel Institute of Technology, Haifa, 1
2
3
4
5
Israel
CRP ESR NAA WBC
C-reactive protein Erythrocyte sedimentation rate Nucleic acid amplification White blood cell
*List of members of the Israeli-Spanish Kingella kingae Research Group is available at www.jpeds.com (Appendix). The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2014.07.060
985
THE JOURNAL OF PEDIATRICS
www.jpeds.com
general acceptance for differentiating between septic hip joints and transient synovitis,16 it was developed when the role of K kingae in pediatric septic arthritis was unrecognized. As none of the patients included in the original and subsequent validation studies had a proven infection caused by this organism, the performance of this tool for diagnosing such infections remains unknown. A retrospective multicenter study, comprising Israeli and Spanish patients, was conducted to assess the ability of Kocher diagnostic algorithm to correctly categorize children with K kingae infection of the hip as having septic arthritis.
Methods The medical records of pediatric patients with culture-proven K kingae hip arthritis, defined as presence of a painful hip joint and from whom the bacterium was recovered by culture of synovial fluid and/or blood, diagnosed in 7 Israeli hospitals and 1 large Spanish medical center located in the city of Barcelona, dating from 1989-2013, were examined. Patients were identified through the clinical microbiology laboratory databases of participating hospitals. Blood and synovial fluid specimens obtained from Israeli patients were processed using the Bactec blood culture system (Becton Dickinson), whereas the BacT/ Alert (bioMerieux) was employed in the Spanish medical center. Medical records were reviewed and relevant demographic and clinical information was extracted using a uniform data collection form. Some patients were included in previous reports in the medical literature. Data retrieval was approved by the local Institutional Review Boards of each medical center.15 Information, thus, collected included the clinical variables (body temperature on admission and refusal to bear weight) and laboratory data (blood WBC count and ESR) of Kocher predictive algorithm. The number of predictors (score) recorded for each patient was calculated, and the likelihood of a child having septic arthritis was interpreted according to Kocher original study model: from 0.2% when none was present, 3% for 1 predictor, 40% for 2 predictors, 93.1% for 3 predictors, and 99.6% if all 4 criteria were met.3 Additional information included demographic data (age, sex), clinical data (duration of symptoms), laboratory results (leukocyte count in joint aspirate [> or #50 000 WBCs/mm3 in synovial fluid]), and, if performed, joint drainage modality (aspiration/ arthrotomy). Statistical Analyses The study examined the distribution of the individual clinical and laboratory components of Kocher diagnostic algorithm among children with bacteriologically-proven K kingae arthritis of the hip, and the sensitivity of Kocher algorithm to correctly assign children with culture-proven K kingae infection of the hip to the septic arthritis category. The statistical significance of the differences between continuous variables was assessed by the Mann-Whitney test and that of categorical variables by the c2 test. A P value of .05 for all variables). All patients were previously healthy and had no significant underlying medical problem. K kingae was isolated from a synovial fluid specimen in 22 (65%) children (from a seeded blood culture vial in 21 children and from direct plating on routine solid media in one). In the remaining 12 children, the organism was recovered from blood culture. None of the 34 cases had the organism isolated from both blood and synovial fluid. Fifteen patients (44%) underwent open arthrotomy and another 2 (6%) closed needle aspiration of the affected joint. In 17, no drainage of the affected joint was performed. Two children (patients 3 and 13 in Table I) who presented with an irritable hip joint and from whom K kingae was isolated from blood culture recovered without antibiotic therapy, indicating a self-limited clinical course. All other patients were administered a variety of b-lactam antibiotics. Although body temperature on admission was recorded in 33 of 34 children and blood WBC counts on admission were available for all children, ESR determination was performed in only 28 (82%). Because of the patients’ young age, refusal to bear weight could not be assessed in 2 infants aged 5 and 6 months. Therefore, results for the 4 variables of Kocher algorithm were available for 28 (82%) children, 3 variables were evaluable in 4 (12%) children, and only 2 variables in the remaining 2 (6%) children. The 2 patients who did not receive antibiotics had Kocher scores of 0 and 1, respectively. Spanish children presented with a significantly lower body temperature (median: 38.5 C), compared with Israeli children (median: 38.9 C [P = .023]), and lower erythrocyte sedimentation rate values (median values: 19 mm/h and 35 mm/h, respectively [P = .029]), but the 2 populations did not have significantly different WBC counts in peripheral blood. Results of the leukocyte count in joint fluid were available for 7 (21%) children, and the Gram-stain examination of the fluid was negative for bacteria in all 7 specimens. Table I shows the patients’ demographic, clinical, and laboratory results recorded on admission. Table II summarizes the results of the individual variables and their sensitivity for correctly identifying children with cultureproven K kingae septic arthritis. The Figure depicts the patients’ Kocher algorithm score of the total number of evaluable variables. The mean SD score of the 28 children with 4 evaluable variables was 2.1 1.0, of whom 20 (71%) had scores #2, indicating that nearly threequarters of children with proven K kingae infection would Yagupsky et al
ORIGINAL ARTICLES
November 2014
Table I. Demographic, clinical, and laboratory features of 34 children with culture-proven K kingae arthritis of the hip joint Patient
Country
Year
Age (mo)
Duration of symptoms (d)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Spain Israel Spain Spain Spain Israel Spain Spain Israel Spain Spain Israel Spain
1989 1993 1994 1998 1999 1999 1999 1999 2001 2002 2002 2002 2003 2003 2003 2003 2003 2004 2004 2004 2005 2006 2007 2007 2007 2007 2008 2008 2008 2009 2011 2011 2013 2013
10 12 23 14 19 18 16 18 6 13 10 5 27 16 8 11 20 15 12 16 10 17 27 20 15 21 6 24 20 18 11 7 17 8
4 2 1 2 1 3 5 3 1 2 2 1 2 1 4 14 5 1 1 1 1 1 4 3 1 1 1 1 1 2 3 5 2 7
T, C
Blood WBC count (WBCs/mm3)
ESR (mm/h)
39.0 39.0 38.8 39.4 37.6 39.3 38.5 38.6 39.5 37.5 39.5 39.0 37.8 38.5 39.6 40.0 37.7 38.9 38.2 39.1 38.3 38.1 37.5 38.0 37.8 38.0 39.0 37.9 38.0 38.2 38.0 38.7 39.0 NR
11.0 17.9 9.7 11.0 19.0 25.6 11.7 20.0 15.2 13.5 11.9 20.5 11.0 14.5 15.0 6.6 15.6 7.8 10.1 9.3 11.5 20.9 12.5 13.7 13.9 24.7 18.5 10.7 15.6 20.5 13.0 22.7 16.3 15.0
55 50 25 36 35 50 25 50 ND 30 5 25 21 80 ND 58 5 20 85 30 42 27 35 17 24 13 15 11 ND 55 21 ND ND ND
Refusal to bear weight + + + + + + NA + NA + + + + + + + + + + + + + + + + + +
Synovial fluid WBC count (WBCs/mm3) 20.0 194.0 ND ND 125.5 ND ND ND ND ND ND ND ND ND 130.0 ND ND 139.9 144.0 ND 5.2 ND ND ND ND ND ND ND ND ND ND ND ND ND
Isolation Joint Joint Blood Blood Joint Blood Joint Blood Joint Blood Blood Joint Blood Blood Joint Blood Joint Joint Joint Blood Joint Joint Blood Joint Joint Joint Blood Joint Joint Joint Joint Joint Joint Joint
NA, not applicable; ND, not done; NR, not recorded.
have been considered at low (50.0 >12.0 >38.5 >40
32 7 34 33 28
NA 108.4 69.3 (44.3-172.5) 14.9 4.8 (13.2-16.6) 38.6 0.7 (38.4-38.8) 33.8 20.4 (16.4-21.2)
NA 130.0 14.2 38.5 28.5
NA 5.2-144.0 6.6-25.7 37.5-40.0 5-85
28 5 22 16 9
88% 71% 65% 48% 32%
some young patients in which this diagnosis is suspected have, in fact, undetected K kingae infection. The relatively benign clinical picture, and the observation that a few patients with culture-proven K kingae arthritis had a seemingly abortive course (ie, improving without antibiotic therapy),28 raise questions about the clinical implications of misdiagnosing a suppurative hip joint infection caused by this organism. However, severe osteoarticular K kingae infections, characterized by metaphyseo-ephyseal and epiphyseal abscess and femoral growth plate damage, have been reported,29,30 and seeding of the organism to the endocardium also can occur,9,15 indicating that appropriate antibiotics should be administered to all patients from whom the organism is recovered from a normally sterile body site. The distribution of the clinical and laboratory findings among the 34 children enrolled in the current study shows that refusal to bear weight had the best sensitivity for detecting K kingae septic arthritis, and body temperature on admission, blood WBC count, and ESR values exceeded cutoff values of the Kocher algorithm only in approximately onehalf of cases. This observation is consistent with the concept that K kingae infection in early childhood, especially when affecting the skeletal system, frequently is characterized by paucity of symptoms, lack of leukocytosis, and relatively low concentrations of acute-phase serum reactants.12-15 Moreover, in 2 of the 7 children who underwent joint aspiration, the synovial leukocyte count was
Y
oung children who come to medical attention because of limping pose a common and serious diagnostic challenge, and differentiating between septic arthritis of the hip and transient synovitis is of paramount clinical and therapeutic importance. If inadequately diagnosed and treatment delayed, pediatric septic arthritis may result in major morbidity, including joint destruction, necrosis of the femoral head, and permanent disability.1 Transient synovitis, in contrast, is a benign and self-limited condition requiring only rest and analgesic therapy, and does not lead to orthopedic sequelae.2 Erroneous misclassification of transient synovitis as septic arthritis may result in unnecessary hospitalization, surgery, and antibiotic therapy. To assist in differentiating between these 2 conditions, an algorithm combining clinical information and laboratory data was developed by Kocher et al.3 This predictive guideline is based on clinical criteria (body temperature >38.5 C and refusal to bear weight) and blood inflammatory markers (white blood cell [WBC] count >12 000 cells/mm3 and erythrocyte sedimentation rate [ESR] >40 mm/h), complemented by the presence of >50 000 WBCs/mm3 in synovial fluid, when joint aspiration is performed.3 The risk of suppurative arthritis directly correlates with increasing number of positive predictors, and transient synovitis is characterized by low scores. The performance of this diagnostic algorithm and similar guidelines based on Kocher original clinical prediction model with slight modifications has been evaluated and validated in many studies that included pediatric patients with bacterial arthritis caused by traditional pathogens such as Staphylococcus aureus, Streptococcus pyogenes, Streptococcus pneumonia, and Haemophilus influenzae type b.3-8 As the result of increasing use of improved culture methods and sensitive nucleic acid amplification (NAA) assays, K kingae, a gram-negative component of the normal pharyngeal flora, is increasingly recognized as an important human pathogen and the most common etiology of joint infections in children aged 6-48 months in widespread geographic populations.9-11 The clinical presentation of patients with K kingae arthritis is often subtle, requiring a high index of suspicion: the body temperature and levels of acute phase reactants may be From the Clinical Microbiology Laboratory, Soroka normal or only mildly elevated, and leukocytosis is frequently absent.9,12-15 K University Medical Center, Beer-Sheva, Israel; Edmond and Lily Safra Children’s Hospital, Chaim Sheba Medical kingae usually affects large weight-bearing joints and in a large multicenter study Center, Tel-Hashomer, Israel; Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel; Molecular comprising 140 children with culture-proven K kingae arthritis, the hip was Microbiology Department, University Hospital Sant Joan involved in 20 (14.3%) patients and represented the third most commonly infected de Deu, Barcelona, Spain; and Pediatric Infectious Diseases Unit, Carmel Medical Center, and the Faculty of site in this series.15 Although Kocher decision-making algorithm has gained Medicine, Technion-Israel Institute of Technology, Haifa, 1
2
3
4
5
Israel
CRP ESR NAA WBC
C-reactive protein Erythrocyte sedimentation rate Nucleic acid amplification White blood cell
*List of members of the Israeli-Spanish Kingella kingae Research Group is available at www.jpeds.com (Appendix). The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2014.07.060
985
THE JOURNAL OF PEDIATRICS
www.jpeds.com
general acceptance for differentiating between septic hip joints and transient synovitis,16 it was developed when the role of K kingae in pediatric septic arthritis was unrecognized. As none of the patients included in the original and subsequent validation studies had a proven infection caused by this organism, the performance of this tool for diagnosing such infections remains unknown. A retrospective multicenter study, comprising Israeli and Spanish patients, was conducted to assess the ability of Kocher diagnostic algorithm to correctly categorize children with K kingae infection of the hip as having septic arthritis.
Methods The medical records of pediatric patients with culture-proven K kingae hip arthritis, defined as presence of a painful hip joint and from whom the bacterium was recovered by culture of synovial fluid and/or blood, diagnosed in 7 Israeli hospitals and 1 large Spanish medical center located in the city of Barcelona, dating from 1989-2013, were examined. Patients were identified through the clinical microbiology laboratory databases of participating hospitals. Blood and synovial fluid specimens obtained from Israeli patients were processed using the Bactec blood culture system (Becton Dickinson), whereas the BacT/ Alert (bioMerieux) was employed in the Spanish medical center. Medical records were reviewed and relevant demographic and clinical information was extracted using a uniform data collection form. Some patients were included in previous reports in the medical literature. Data retrieval was approved by the local Institutional Review Boards of each medical center.15 Information, thus, collected included the clinical variables (body temperature on admission and refusal to bear weight) and laboratory data (blood WBC count and ESR) of Kocher predictive algorithm. The number of predictors (score) recorded for each patient was calculated, and the likelihood of a child having septic arthritis was interpreted according to Kocher original study model: from 0.2% when none was present, 3% for 1 predictor, 40% for 2 predictors, 93.1% for 3 predictors, and 99.6% if all 4 criteria were met.3 Additional information included demographic data (age, sex), clinical data (duration of symptoms), laboratory results (leukocyte count in joint aspirate [> or #50 000 WBCs/mm3 in synovial fluid]), and, if performed, joint drainage modality (aspiration/ arthrotomy). Statistical Analyses The study examined the distribution of the individual clinical and laboratory components of Kocher diagnostic algorithm among children with bacteriologically-proven K kingae arthritis of the hip, and the sensitivity of Kocher algorithm to correctly assign children with culture-proven K kingae infection of the hip to the septic arthritis category. The statistical significance of the differences between continuous variables was assessed by the Mann-Whitney test and that of categorical variables by the c2 test. A P value of .05 for all variables). All patients were previously healthy and had no significant underlying medical problem. K kingae was isolated from a synovial fluid specimen in 22 (65%) children (from a seeded blood culture vial in 21 children and from direct plating on routine solid media in one). In the remaining 12 children, the organism was recovered from blood culture. None of the 34 cases had the organism isolated from both blood and synovial fluid. Fifteen patients (44%) underwent open arthrotomy and another 2 (6%) closed needle aspiration of the affected joint. In 17, no drainage of the affected joint was performed. Two children (patients 3 and 13 in Table I) who presented with an irritable hip joint and from whom K kingae was isolated from blood culture recovered without antibiotic therapy, indicating a self-limited clinical course. All other patients were administered a variety of b-lactam antibiotics. Although body temperature on admission was recorded in 33 of 34 children and blood WBC counts on admission were available for all children, ESR determination was performed in only 28 (82%). Because of the patients’ young age, refusal to bear weight could not be assessed in 2 infants aged 5 and 6 months. Therefore, results for the 4 variables of Kocher algorithm were available for 28 (82%) children, 3 variables were evaluable in 4 (12%) children, and only 2 variables in the remaining 2 (6%) children. The 2 patients who did not receive antibiotics had Kocher scores of 0 and 1, respectively. Spanish children presented with a significantly lower body temperature (median: 38.5 C), compared with Israeli children (median: 38.9 C [P = .023]), and lower erythrocyte sedimentation rate values (median values: 19 mm/h and 35 mm/h, respectively [P = .029]), but the 2 populations did not have significantly different WBC counts in peripheral blood. Results of the leukocyte count in joint fluid were available for 7 (21%) children, and the Gram-stain examination of the fluid was negative for bacteria in all 7 specimens. Table I shows the patients’ demographic, clinical, and laboratory results recorded on admission. Table II summarizes the results of the individual variables and their sensitivity for correctly identifying children with cultureproven K kingae septic arthritis. The Figure depicts the patients’ Kocher algorithm score of the total number of evaluable variables. The mean SD score of the 28 children with 4 evaluable variables was 2.1 1.0, of whom 20 (71%) had scores #2, indicating that nearly threequarters of children with proven K kingae infection would Yagupsky et al
ORIGINAL ARTICLES
November 2014
Table I. Demographic, clinical, and laboratory features of 34 children with culture-proven K kingae arthritis of the hip joint Patient
Country
Year
Age (mo)
Duration of symptoms (d)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Israel Spain Israel Spain Spain Spain Israel Spain Spain Israel Spain Spain Israel Spain
1989 1993 1994 1998 1999 1999 1999 1999 2001 2002 2002 2002 2003 2003 2003 2003 2003 2004 2004 2004 2005 2006 2007 2007 2007 2007 2008 2008 2008 2009 2011 2011 2013 2013
10 12 23 14 19 18 16 18 6 13 10 5 27 16 8 11 20 15 12 16 10 17 27 20 15 21 6 24 20 18 11 7 17 8
4 2 1 2 1 3 5 3 1 2 2 1 2 1 4 14 5 1 1 1 1 1 4 3 1 1 1 1 1 2 3 5 2 7
T, C
Blood WBC count (WBCs/mm3)
ESR (mm/h)
39.0 39.0 38.8 39.4 37.6 39.3 38.5 38.6 39.5 37.5 39.5 39.0 37.8 38.5 39.6 40.0 37.7 38.9 38.2 39.1 38.3 38.1 37.5 38.0 37.8 38.0 39.0 37.9 38.0 38.2 38.0 38.7 39.0 NR
11.0 17.9 9.7 11.0 19.0 25.6 11.7 20.0 15.2 13.5 11.9 20.5 11.0 14.5 15.0 6.6 15.6 7.8 10.1 9.3 11.5 20.9 12.5 13.7 13.9 24.7 18.5 10.7 15.6 20.5 13.0 22.7 16.3 15.0
55 50 25 36 35 50 25 50 ND 30 5 25 21 80 ND 58 5 20 85 30 42 27 35 17 24 13 15 11 ND 55 21 ND ND ND
Refusal to bear weight + + + + + + NA + NA + + + + + + + + + + + + + + + + + +
Synovial fluid WBC count (WBCs/mm3) 20.0 194.0 ND ND 125.5 ND ND ND ND ND ND ND ND ND 130.0 ND ND 139.9 144.0 ND 5.2 ND ND ND ND ND ND ND ND ND ND ND ND ND
Isolation Joint Joint Blood Blood Joint Blood Joint Blood Joint Blood Blood Joint Blood Blood Joint Blood Joint Joint Joint Blood Joint Joint Blood Joint Joint Joint Blood Joint Joint Joint Joint Joint Joint Joint
NA, not applicable; ND, not done; NR, not recorded.
have been considered at low (50.0 >12.0 >38.5 >40
32 7 34 33 28
NA 108.4 69.3 (44.3-172.5) 14.9 4.8 (13.2-16.6) 38.6 0.7 (38.4-38.8) 33.8 20.4 (16.4-21.2)
NA 130.0 14.2 38.5 28.5
NA 5.2-144.0 6.6-25.7 37.5-40.0 5-85
28 5 22 16 9
88% 71% 65% 48% 32%
some young patients in which this diagnosis is suspected have, in fact, undetected K kingae infection. The relatively benign clinical picture, and the observation that a few patients with culture-proven K kingae arthritis had a seemingly abortive course (ie, improving without antibiotic therapy),28 raise questions about the clinical implications of misdiagnosing a suppurative hip joint infection caused by this organism. However, severe osteoarticular K kingae infections, characterized by metaphyseo-ephyseal and epiphyseal abscess and femoral growth plate damage, have been reported,29,30 and seeding of the organism to the endocardium also can occur,9,15 indicating that appropriate antibiotics should be administered to all patients from whom the organism is recovered from a normally sterile body site. The distribution of the clinical and laboratory findings among the 34 children enrolled in the current study shows that refusal to bear weight had the best sensitivity for detecting K kingae septic arthritis, and body temperature on admission, blood WBC count, and ESR values exceeded cutoff values of the Kocher algorithm only in approximately onehalf of cases. This observation is consistent with the concept that K kingae infection in early childhood, especially when affecting the skeletal system, frequently is characterized by paucity of symptoms, lack of leukocytosis, and relatively low concentrations of acute-phase serum reactants.12-15 Moreover, in 2 of the 7 children who underwent joint aspiration, the synovial leukocyte count was
