Septic arthritis is undeniably the most common form of osteoarticular disease of K. kingae. In fact, some reports demonstrated that septic arthritis represents between 53 and 82.8% of all OAI due to this pathogen [26,27,28]. K. kingae septic arthritis usually involves large weight-bearing articulations, such as the hip, knee, ankle, shoulder, or elbow [20,25,27,28,29,30,31,32]. In terms of frequency, the infection tends to involve the lower extremity the most, and the knee is the joint that is the most frequently incriminated. However, atypical joints, such as the sternoclavicular, acromioclavicular, tarsal, or metacarpophalangeal/metatarsophalangeal joints, are overrepresented in K. kingae arthritis compared with septic arthritis caused by other pathogens [20,25,27,28,29,30,31,32]. The clinical and biological aspects of septic arthritis can, unfortunately, be truncated when K. kingae is responsible for infection [33]. The synovial fluid examination in children with culture-proven arthritis due to K. kingae demonstrated low leukocyte counts in a substantial number of cases. On that point, the recognized cut-off value of 50,000 WBC/mL in synovial fluid aspirates, used as a diagnostic factor defining bacterial arthritis, may erroneously exclude the diagnosis of K. kingae arthritis and should, therefore, be used with great caution [33]. In fact, a few studies demonstrated that the WBC count of the synovial fluid showed less than 50,000 WBC/mL in a quarter of the cases, and that and the examination of the Gram-stain is, most of the time, negative [21,22,34]. Thus, the afebrile presentation, the mild clinical symptoms, and the absence or the little disturbance of acute phase reactants in K. kingae arthritis do not meet the diagnostic criteria of a septic joint [35]. Even worse, when using the application of Kocher’s predictive algorithm [36], it seems that three-quarters of children with culture-proven K. kingae septic arthritis of the hip would have been considered to have transient synovitis [35]. The clinical experience has taught us that this algorithm should not be used for children less than 4 years old. Here again, the pediatric orthopedist must be extremely vigilant in the face of clinical situations not very suggestive of arthritis.

Finally, the clinical course of septic arthritis due to K. kingae is very different from those of arthritis caused by pyogenic microorganisms, which stimulate a huge influx of neutrophils to the site. In septic arthritis due to pyogenic pathogens, the clinical course is characterized by a potent activation of the immune response, in association with high levels of cytokines and reactive oxygen species, and an increase in the release of host matrix metalloproteinases and other collagen-degrading enzymes, which, in conjunction with bacterial toxins, lead to joint cartilage destruction [37,38,39]. In addition, the antigen-induced inflammatory response may persist and continue to damage the joint architecture even after the infection has been cleared [38,39]. Fortunately, the clinical experience acquired during the management of OAI with K. kingae seems to demonstrate that arthritis due to this particular pathogen does not follow this pattern of cartilage destruction. Therefore, many cases of K. kingae’s septic arthritis may even be treated by a simple syringe injection–aspiration procedure, and thus, especially when the leukocyte count appears low in the synovial fluid examination, the attitude will not be defensible in pyogenic septic arthritis.

The anatomical sites affected by K. kingae osteomyelitis include long bones, such as the femur, tibia, humerus, radius, and ulna [3,4,5,19,21,22]. Nevertheless, any bone rarely infected by other pathogens, such as the sternum, the clavicle, the talus, or the calcaneum, may also be affected by K. kingae osteomyelitis [3,4,5,19,30]. K. kingae osteomyelitis represents between 15 to 31% of all OAI due to this pathogen, and more than a quarter of such osteomyelitis is concomitant with septic arthritis [3,5].

It is also interesting to highlight that epiphysis or apophysis, which are almost never affected by other organisms, may be frequently involved in K. kingae osteomyelitis [40,41]. Some reports have suggested that patients with isolated osteomyelitis had a more prolonged duration of symptoms before admission and presented with a lower body temperature compared with children with septic arthritis alone [30]. Without an appropriate treatment, acute osteomyelitis can sneakily evolve into subacute osteomyelitis. Subacute osteomyelitis is an atypical osteomyelitis that is most likely attributed to an atypical host–pathogen relationship that may be explained by any combination of increased host resistance, decreased virulence of the causative pathogen, and/or prior antibiotic exposure [42,43,44,45,46]. In most cases, culture-based methods fail to identify the causative organism, especially when fine-needle aspiration is performed. Surgical drainage may yield positive cultures in 40% to 75% of patients [41]. Subacute osteomyelitis can be divided into two main clinical forms according to the children’s age and to its bacteriological etiology [41]. The first form, called the infantile form, affects children aged between 6 months and 4 years. Approximately 90% of all PSAHO affect patients in this age group, with K. kingae as the main observed microorganism [41]. In these young children, the clinical course is most likely attributable to the natural low virulence of K. kingae. Many children in this age group are usually recognized late as having an osteoarticular infection, and the accurate diagnosis is generally delayed after a bony lytic lesion has occurred [40,41]. The second form, entitled the juvenile form, affects children older than 4 years and S. aureus appears as the major responsible pathogen. Subacute osteomyelitis due to K. kingae follows a benign course, and the recommended treatment for sub-acute osteomyelitis, when radiographic investigations demonstrate lucent lesions or nidus, is curettage, biopsy, and culture, followed by antibiotics [40,41,47]. However, a few authors have suggested that antibiotics alone may be adequate and that surgery should be reserved only for “aggressive lesions”, as well as those that do not respond to antibiotics [40,41,47]. However, it is generally agreed that treatment should not be initiated until proper drainage and bacteriological samples have been obtained [40,41,47].

Excepting arthritis and osteomyelitis, spondylodiscitis is probably the main frequent form of invasive K. kingae infection in children less than 4 years old. Childhood spondylodiscitis is a term frequently used to describe a continuum of spinal infections, from discitis to spondylodiscitis as well as vertebral osteomyelitis with occasional associated soft tissue abscesses. Three main clinical forms of spondylodiscitis have been described in the pediatric population according to the age of the patient [48,49,50]. The neonate form affects infants under 6 months and is the most serious manifestation of the disease, often presenting with Staphylococcus aureus septicemia and multiple infectious foci. The infantile form affects children from 6 months (end of maternally derived immunity) to 48 months of age, and this age group represents 80% of childhood spondylodiscitis. In this group, some studies have suggested that K. kingae could be the most frequent microorganism responsible for the spinal infection [48,49,50]. Finally, in the third form, affecting children older than 4 years, patients are more prone to be febrile and ill-appearing and to sustain vertebral osteomyelitis due to S. aureus. This triphasic age distribution is currently widely used among pediatric orthopedists since it explains both the different microbiologic epidemiology and the different clinical forms among the age groups.

Invasive K. kingae infections can also give rise to atypical osteoarticular infections, such as cellulitis, pyomyositis, bursitis, and tendon sheath infections [31,51,52]. Thus, we suggest using and incorporating MRI into modern diagnostic algorithms for OAI to better identify unusual locations of OAI.

Several controversies concerning the treatment of OAI are still contributing to the debate, and this debate seems even more relevant when osteoarticular infections due to K. kingae must be treated. In fact, there is, nowadays, no consensus about the treatment of OAI considering which infections may be treated medically and which will need a surgical approach.

The clinical presentation and outcome may be vastly different when considering the microbiological causes of OAI, and, thus, the required treatment can be diametrically opposed from one case to another. For example, OAI due to S. aureus present a more severe course of disease, a slower clinical response, and a potentially worse outcome, requiring, most of the time, invasive diagnostic and therapeutic procedures and rapid antibiotic treatment [54,57], whereas many OAI due to K. kingae could be theorically treated only with antibiotherapy [3].

Thus, the main question to be answered, especially when treating a child with an OAI due to K. kingae, is “does the child need surgery?” [3,4,5]. Most of the time, there are three basic reasons for proposing a surgical intervention in many bone and joint infections: i.e., microbiologic diagnosis, infectious source control, and preservation of maximal function [3,4,5]. Starting with microbiologic diagnosis, identification of the pathogen responsible for infection would be extremely helpful for tailoring definitive antibiotic treatment regarding the choice, the route, and the duration for antibiotic therapy.

Theoretically, children with skeletal system infections due to K. kingae would not require invasive surgical procedures, except maybe for excluding pyogenic germs’ implication. On that point, a diagnostic procedure has been developed to avoid unnecessary surgical procedures with a bacteriological diagnostic aim. It has been suggested that the performance of a sensitive K. kingae-specific NAAA on an oropharyngeal specimen may provide strong evidence that this microorganism is responsible for OAI [58]. However, it should be kept in mind that a positive test is not an irrefutable proof of the etiology of the disease since around 10% of young children carry the organism. Contrariwise, a negative PCR result rules out K. kingae as the causative pathogen of OAI [58].

Most of the patients with K. kingae’s OAI respond promptly to conservative treatment with appropriate antibiotics to such an extent that the antibiotic treatment could be reasonably shortened. In many cases, any patients with K. kingae’s OAI may follow an abortive course, even when no antimicrobials therapy is administered. Nowadays, there is still a lack of controlled studies that permit the formulation of evidence-based recommendations on the most efficient antibiotic, on the necessity or not to investigate production of beta-lactamase, and on the optimal length of therapy for K. kingae OAI [32]. Drug therapy for osteoarticular infections usually consists of the intravenous administration of oxacillin/nafcillin or a second- or third-generation cephalosporin while pending culture results [18,19,59]. In areas in which community associated methicillin-resistant Staphylococcus aureus is prevalent and when the clinical presentation is acute, a combination of a b-lactam antibiotic and vancomycin may be suggested [60]. The clinical response is then used to guide switching to oral antibiotics (usually within less than 3 days) The current trend is to shorten as much as possible the length of antibiotic treatment since it is noted that K. kingae’s OAI respond well even to short antibiotics treatments. Antibiotic treatment generally varies from 2 to 3 weeks for K. kingae arthritis, from 3 to 6 weeks for osteomyelitis, and from 3 to 12 weeks for spondylodiscitis [30]. However, the current trend is to shorten as much as possible the length of antibiotic treatment since it is noted that K. kingae’s OAI respond well even to short antibiotics treatments. In fact, many authors consider that OAI due to K. kingae, characterized by a subacute course and low bacterial concentrations, could be treated solely by shorter oral antibiotic therapy [3,4,5].

During the last few years, the susceptibility of K. kingae to antibiotics that are generally given to children with suspected or confirmed invasive infection has been greatly studied, and this microorganism’s characteristics are currently better recognized. K. kingae is considered to be extremely susceptible to ampicillin, penicillin, second- and third-generation cephalosporins, cotrimoxazole, ciprofloxacin, macrolides, tetracycline, and chloramphenicol [30,61]. The organism exhibits reduced susceptibility to oxacillin and clindamycin [62,63] and appears totally resistant to vancomycin and trimetoprim [62,63,64,65,66,67]. Some reports have described occasional in vitro resistance to cotrimoxazole, erythromycin, and ciprofloxacin [62,63,64,65,66,67]. Yagupsky et al. have demonstrated that 38.5% of the isolates from their healthy respiratory carriers and patients with invasive infection enclosed K. kingae strains that were resistant to clindamycin [62,63]. More alarming is the discovery of strains producing b-lactamase recovered from an HIV-positive patient in the USA [68] and in three isolates from children in Iceland [69]. The culture of K. kingae remains suboptimal if not, frankly, ineffective, even when the samples from infected joint or bone are inoculated directly into blood culture vials.

Because of the large-scale use of nucleic acid amplification assays, the susceptibility of the organism to antimicrobial drugs administered to children with invasive infection due to K. kingae is being less frequently or not at all investigated. Therefore, clinicians will not have any information about the antibiotic susceptibility of the organism in many clinical situations, and there is the potential risk to treat patients affected with an antimicrobial-resistant organism infection ineffectively. On this subject, a few studies demonstrated that the strain of K. kingae isolated in the throat was responsible for OAI; this observation highlighted the possibility of isolating with significant effectiveness the K. kingae strain responsible for OAI using throat swabs [58,70]. Thus, there is a real interest to perform throat swabs both to isolate the bacteria and study its antimicrobial sensitivity profile.

OAI due to K. kingae have a good character: they are considered as benign with a mild-to-moderate clinical presentation, they have a favorable prognosis after adequate antibiotic treatment, and they seldom lead to long-term sequelae [18,21,22,32,57]. There is a world of difference between infections due to pyogenic germs and those caused by K. kingae. On that point, the clinical course is usually better for children with OAI caused by K. kingae, as evidenced by shorter hospitalization and fewer adverse events [54,57]. This can be explained both by the low virulence of K. kingae [22] and by its high susceptibility to β-lactam antibiotics [22,30,32].

However, the initial benign clinical presentation of K. kingae may result in delayed diagnosis [30,32,54] and may thus lead to subsequent severe infectious bone lesions [71]. In fact, the mildness of the clinical signs during K. kingae osteomyelitis may lead to a diagnostic delay with a more destructive nature of the lesions, presence of intraosseous abscesses, and potential damage to the growth cartilage. Such subacute osteomyelitis is rare and often misdiagnosed because of insidious symptoms [43,44,45,72]. Fortunately, such complicated cases are rare (less than 5% of K. kingae osteoarticular infections), and the functional results are quite satisfactory even in the case of physeal damage [71]. Even if the emergency character is not the same as for pyogenic infections, care must be taken, and antibiotic treatment should be introduced as soon as the diagnosis of OAI due to K. kingae is suspected.