European Review for Medical and Pharmacological Sciences 2019; 23(2 Suppl.): 145-158 Acute osteomyelitis and septic arthritis in children: a systematic review of systematic reviews A. GIGANTE1, V. COPPA2, M. MARINELLI3, N. GIAMPAOLINI3, D. FALCIONI3, N. SPECCHIA3 1Orthopaedics and Traumatology, Polytechnic University of Marche, Ancona, Italy 2Clinical Orthopaedics, Department of Clinical and Molecular Sciences, School of Medicine, Università Politecnica delle Marche, Ancona, Italy 3Clinic of Adult and Paediatric Orthopaedics, Azienda Ospedaliero-Universitaria, Ospedali Riuniti di Ancona, Ancona, Italy Abstract. – OBJECTIVE: Septic arthritis and Osteomyelitis (OM) is an inflammation of osteomyelitis are rare in children, but they are bone, usually due to infection with bacteria or difficult to treat and are associated with a high other micro-organisms (e.g., fungi), that is asso- rate of sequelae. This paper addresses the main ciated with bone destruction. Septic arthritis (SA) clinical issues related to septic arthritis and os- teomyelitis by means of a systematic review of is an infection of the synovial space that involves systematic reviews. the synovial membrane, the joint space, and joint 2 MATERIALS AND METHODS: The major elec- structures . tronic databases were searched for systematic SA and OM are commonly divided into three reviews/meta-analyses septic arthritis and os- types based on aetiology: haematogenous, sec- teomyelitis. The papers that fulfilled the inclu- ondary to contiguous infection, and secondary sion/exclusion criteria were selected. to direct inoculation. The distinctive anatomical RESULTS: There were four systematic re- views on septic arthritis and four on osteomy- characteristics of younger children, especial- elitis. Independent assessment of their meth- ly the presence of vessels between metaphysis odological quality by two reviewers using and epiphysis and of intracapsular metaphyses, AMSTAR 2 indicated that its criteria were not involve that a bone infection may lead to SA sec- consistently followed. ondary to OM and vice versa2. Unless an effec- CONCLUSIONS: Collectively, these works tive treatment is promptly initiated, the intense provide strong evidence regarding a large num- ber of issues including classification, epidemiol- inflammatory reaction that is often associat- ogy and risk factors, causative organisms, clin- ed with SA or OM has the potential to destroy 3 ical presentation, laboratory markers, imaging, structures such as the articular cartilage and the diagnostic needle aspiration, antibiotic therapy, epiphyseal growth plate, resulting in long-term surgical therapy, and prognosis. A clinical sum- disability due to functional impairment of the mary based on the best evidence is supplied. joint or to limb asymmetry2. Since timely diag- nosis is vital for a satisfactory outcome, SA and Key Words: OM must be ruled out in any child presenting Osteomyelitis, Septic arthritis, Children. with a painful limb or joint or with a fever of unknown origin4. The main issues related to SA and OM are classification, epidemiology and risk factors, Introduction causative organisms, clinical presentation, lab- oratory markers, imaging, diagnostic needle Osteoarticular infections are complex con- aspiration, antibiotic therapy, surgical therapy, ditions, especially in paediatric patients, which and prognosis1,3,5,6. unless treated promptly and correctly can result The aim of this systematic review of systematic in limb impairment or in life-threatening con- reviews is to identify the best available evidence re- ditions1. garding the management of paediatric SA and OM. Corresponding Author: Antonio Gigante, MD; e-mail: [email protected] 145 A. Gigante, V. Coppa, M. Marinelli, N. Giampaolini, D. Falcioni, N. Specchia Materials and Methods Septic Arthritis (Table III) Literature Search Strategy Classification A search of the Medline (via Pubmed), Emba- None of the SRs included in the study exami- se, Amed, and CISCOM electronic databases was ned SA classification. performed for works published from 1 January 2000 to 31 October 2017. The search terms used were: septic arthritis, osteomyelitis, children, cli- Epidemiology and Risk Factors nical trial, meta-analysis, and systematic review. Two of the SRs addressed SA epidemiology The reference lists of all the papers thus identified and risk factors. were examined for further relevant references. No Kang et al5 reported an incidence of 1 in language restrictions were applied. 100,000 in industrialized countries and a higher incidence in developing countries (1 in 20,000 Study Selection in Africa and 1 in 5000 in Malawi). These data Only systematic reviews and meta-analyses of were confirmed by Rutz et al3. The two SRs studies involving paediatric patients were included. agreed on the fact that SA affects more frequent- Non-systematic reviews, overviews, clinical trials, ly young male children (infants and toddlers and reviews of non-clinical investigations were being involved most frequently) and children excluded. The most common reasons for exclusion with respiratory distress syndrome, an umbilical were the facts that a review was not systematic or artery catheter, and conditions associated with that it did not include paediatric populations. Study heightened susceptibility to infection (e.g., pre- selection was performed separately by each author. maturity, low birthweight, sickle-cell haemoglo- The following clinical information on SA and OM binopathy, and small size for age). was extracted from the papers: classification, epi- demiology and risk factors, causative organisms, Causative Organisms clinical presentation, laboratory markers, imaging, According to both SRs3,5 Staphylococcus au- diagnostic needle aspiration, antibiotic therapy, reus is the most frequent causative pathogen of surgical therapy, and prognosis. SA, followed by group A Streptococci (GAS) The methodological quality of the systematic and Enterobacter spp. An emerging problem ac- reviews that were included in the study was as- cording to both SRs is the increasing incidence sessed independently by two reviewers using the of infections caused by methicillin-resistant S. AMSTAR 2 tool7. aureus (MRSA), also in patients who are not at significant risk of MRSA. These community-as- Evidence Synthesis sociated (CA) MRSA strains exhibit lower an- A meta-analysis of systematic reviews is com- tibiotic resistance than hospital-acquired (HA) plex, especially due to overlaps, since some of the strains. primary studies may be included in more than one The incidence of arthritis due to Haemophilus review. To address the problem, a qualitative evi- influenzae, a highly common paediatric patho- dence synthesis was provided for each of the key gen in the past, is declining due to vaccination topics listed above. programmes, whereas Kingella kingae is increas- ingly being isolated, probably due to the greater accuracy of current diagnostic tools. Results According to Taj-Aldeen et al8, the most com- mon fungal osteoarticular infections are due to The search identified a total number of 97 pa- Aspergillus and Candida spp, whereas hyalohy- pers on SA and 224 on OM. After evaluation of phomycetes, dematiaceous moulds, and Mucor- the abstracts, the full text of 37 publications on ales are the non-Aspergillus filamentous fungi SA and 53 on OM were examined; of these, 9 sy- isolated most frequently. The main aetiology of stematic reviews (SRs), 5 on SA and 4 on OM, fungal arthritis is direct inoculation, followed were included. Only one SR from each set discus- by haematogenous dissemination and contigu- sed all the topics addressed in this work, whereas ous spread. In immunocompetent patients, the the others focused on specific issues. most common mechanism of infection is road The methodological quality of most SRs was accidents involving knee puncture or penetrat- low to moderate (Tables I and II). ing wounds. 146 Table I. AMSTAR synopsis (modified from Cargnelli et al 2017). Status Conflict A prior Two Literature of List of Study Scientific Combining Publications of Study design reviewers search publication studies characteristics quality Conclusions findings bias interest Septic Kang et al. 2009 NO CANNOT ANSWER YES YES NO NO YES YES YES NOT YES APPLICABLE Arthritis Rutz et al. 2013 NO CANNOT ANSWER NO YES NO NO NO CANNOT CANNOT NO YES ANSWER ANSWER Farrow et al. 2015 NO NO YES YES YES YES YES YES YES YES YES Taj-Aldeen et al. 2015 NO CANNOT ANSWER YES YES NO NO NO YES YES YES YES Table I. YES YES YES YES YES YES YES YES YES NO YES Zhao et al. 2017 YES YES YES NO YES YES YES YES YES YES YES Osteomyelitis Hotchen et al. 2017 YES CANNOT ANSWER YES YES YES YES NO YES YES NO YES Dartnell et al. 2012 NO YES YES YES NO NO NO YES YES NO YES Howard-Jones et al. 2013 NO CANNOT ANSWER YES YES YES YES YES YES YES NO YES Taj-Aldeen et al. 2015 NO CANNOT ANSWER YES YES NO NO NO YES YES YES YES Gill et al. 2017 NO CANNOT ANSWER NO YES NO NO NO YES YES NO YES Mooney et al. 2016 NO CANNOT ANSWER YES YES YES YES NO YES YES NO yes; no; cannot answer; not applicable 147 Table I. AMSTAR synopsis (modified from Cargnelli et al 2017). A prior Selection Literature Selection in Extraction Excluded Study PICO design criteria search duplicate in duplicate studies Septic arthritis Kang et al. 2009 Yes No No No Partial yes No No No Rutz et al. 2013 No No No No No No No Farrow 2015 Yes No Yes Partial yes No No No Taj-Aldeen et al. 2015 No No No Partial yes No No No De Sa et al. 2015 No Partial yes No Partial yes Yes Yes No Zhao et al. 2017 Yes Partial yes Yes P Yes Yes No Osteomyelitis Hotchen et al. 2017 No Yes No Partial yes Yes No No No Dartnell et al. 2012 No No No Partial yes Yes Yes No No Howard-Jones et al. 2013 No No No Partial yes Yes No No No Taj-Aldeen et al 2015 No No No Partial yes Yes No No No Gill et al 2017 No No No Partial yes Yes No No No Mooney et al.