24.1 Orthopaedics and rheumatology
Introduction
The child who has pathology of the bones and joints may present in a variety of ways. Acute paediatric musculoskeletal conditions range from the many infective, inflammatory and other causes of limp and limb pains, to the isolated acute limb fracture as a result of moderate trauma, which is among the top ten paediatric emergency department (ED) presentations in Australia.1 An awareness of the range of possible conditions and their clinical appearance, and skilled physical examination and clinical reasoning, are essential for optimal outcomes.
The child with acute musculoskeletal pain or dysfunction
General approach
Infant and child development have extensive influence on the musculoskeletal pathology seen and its manifestations, as well as the techniques used in assessment. These differences are outlined in Table 24.1.1.
MBA, motorbike accident; MVA, motor vehicle accident.
The limping child
The wide spectrum of causes of limp in children is illustrated in Table 24.1.2. In terms of frequency, common causes in different age groups are outlined in Table 24.1.3.
Trauma |
Infection (point focus) |
Post-infective |
Inflammatory |
Primary bone disorders |
• Osteochondroses, e.g. Osgood–Schlatter disease (patellar tendon insertion), Sever’s disease (Achilles tendon insertion) |
Neoplastic |
Haematological |
Physical |
Psychological and idiopathic pain syndromes |
Abdominal |
Spine |
Assessment
Pain localisation
This is the foundation of acute orthopaedic diagnosis and efficient investigation use. A suggested sequence of examination for young children with musculoskeletal pain is shown in Table 24.1.4.
Observation and examination
When assessing hip range of motion, the ‘flexion adduction’ test2 may be helpful: with buttocks flat on the bed, flex the hip (first the unaffected and then the affected) to 90 degrees while supporting the lower leg. Then gently attempt to fold the knee over the contralateral leg without child lifting the buttock off the couch. In the normal hip this adduction should allow the ipsilateral knee to be positioned over the opposite leg whereas an inability to adduct past the midline is common in the irritable hip or other causes of hip joint pathology.
The child whose examination findings suggest an isolated irritable hip is likely to have one of the pathologies outlined in Table 24.1.5. A suggested pathway for investigation is shown in the algorithm in Fig. 24.1.1. This algorithm incorporates the findings of a number of studies of the prognostic significance of various features of the limping child with respect to septic arthritis, which are further discussed below.
Inflammatory markers
FBC is useful primarily to flag bone marrow dysfunction. Each cell line (haemoglobin, white cells and platelets) should be individually assessed. The sensitivity of a raised white cell count (WCC) (>12 × 109 L–1) for the identification of septic arthritis is variable (20–75%).3,4 The absence of an elevated white cell response should never be used to rule out septic arthritis. ESR has shown higher sensitivity (90–95%) in identifying the subgroup with septic arthritis; however, this may miss early infection.3 CRP has shown superior sensitivity (up to 80%)4 and specificity in the early identification of invasive bacterial infection in general and of septic arthritis in particular. CRP rises within 24 hours of acute illness, and also falls rapidly and can be used to monitor effectiveness of treatment.5
Radiology
X-ray
X-rays are usually non-contributory in children under 10 years of age with acute onset limp (<1 week) without a specific trauma history. Radiological findings in osteomyelitis are usually not present for up to 10 days from the onset of illness, at which time there may be periosteal elevation outlined by new bone formation and/or lucent areas.6 Due to the higher burden of pelvic or gonadal irradiation, non-selective or routine pelvic/lower limb radiology should not be encouraged. An effusion is better diagnosed by ultrasound, and deep soft tissue infection such as osteomyelitis, by magnetic resonance imaging (MRI). X-rays are required in trauma with focal tenderness or non-weight-bearing, to rule out SUFE (see below) in the adolescent child with hip pain/dysfunction with or without a trauma history, and in cases of persistent unexplained pain or limb dysfunction.
Ultrasound
Ultrasound is a sensitive, non-invasive assessment tool for evaluation of the irritable hip. It will detect even small hip effusions, may show a subperiosteal pus collection in some cases of osteomyelitis, and may be diagnostic for Perthes’ disease and SUFE,7 although it is not usually the first-line investigation for these pathologies. Disadvantages of ultrasound include variations in quality from different operators, and difficulty in after-hours access. An ultrasound cannot distinguish between reactive and infective effusions; however, the presence of an effusion confirms organic pathology and its absence in a clinically abnormal presentation should prompt a search for an extrasynovial focus. There is controversy over the place of ultrasound-guided hip aspiration in the evaluation of the irritable hip with an effusion.8 The majority of Australian paediatric orthapaedic units favour orthopaedic evaluation of irritable hips with selective aspiration and arthrotomy/washout under general anaesthesia. However, ultrasound-guided diagnostic hip aspiration has a role in some units.9–11
Bone scan
Isotope bone scans with three-phase technetium-99m MDP are sensitive early, and well tolerated by children, but lack specificity. Confusion may arise particularly in relation to physeal sites, where uptake is already above base-line. ‘Hot’ scans indicate increased osteoblastic uptake and generally relate to a response to infection or injury. A ‘cold’ scan suggests infarction, which in severe cases may be a consequence of osteomyelitis. Bone scan is of particular value in children in whom multifocal disease is suspected e.g. neonatal osteomyelitis or chronic recurrent multifocal osteomyelitis (CRMO).6
MRI
MRI is excellent for detailed delineation of soft-tissue and bony pathology, particularly when surgery is planned.12 It is the investigative modality of choice to delineate the site of possible bacterial infection in a child with localised musculoskeletal pain and a septic picture in whom septic arthritis has been ruled out by clinical assessment, ultrasound or arthrocentesis.13 Limitations include cost, access limitations, and the need to remain still for a longer time period, necessitating general anaesthetic in most small children.
Clinical decision making in a child with a limp
Relative weightings of various symptoms and signs in the various acute paediatric hip pathologies are expressed in Table 24.1.5 and Fig. 24.1.1. However, spinal, abdominal, and pelvic pathology can present as a limp, and must be considered prior to this narrowed focus. The infant or toddler, because of their increased risk for invasive bacterial illness, also represents a special circumstance and the ill child under 2 with an abnormal acute musculoskeletal assessment is best admitted for combined paediatric and orthopaedic assessment and investigation.
The child over 2 with acute atraumatic localised knee or hip pain and an abnormal hip examination is most likely to have an irritable hip (transient synovitis), and the main issue is to exclude septic arthritis. An ultrasound assessment should confirm the suspicion of a hip effusion as suggested by examination, but does not differentiate between transudate and exudate. Table 24.1.6 lists conditions at higher risk of septic arthritis which must be flagged; however, the majority of infections occur in children without underlying pathology. Various attempts have been made to create a valid decision rule with a high sensitivity and specificity for bacterial infection,4,14–16 however none have demonstrated sustained power prospectively. The four most important variables to give a predicted probability of septic arthritis in a given child with an acutely irritable hip are shown in Table 24.1.7. Degree of pain and range of motion of the joint are auxiliary clinically important variables in the differentiation.
Relative immune deficit |
Injury mechanism |
Joint disease |
Non-weight-bearing Febrile >38.5°C WCC > 12×109 L–1 ESR > 40 mm hr–1 CRP > 20 mg L−1 |
Specific syndromes
Transient synovitis
Transient synovitis is a self-limited inflammatory disorder of uncertain aetiology, occurring particularly in boys (70%) in the age group 3–10 years. The onset of lower limb pain is generally gradual and initially may be localised to hip, knee, groin or thigh. The child may be mildly unwell or have had a recent non-specific upper respiratory tract illness, although this is not uniformly present. Range of movement at the hip joint is usually mildly diminished, with discomfort at the end-points of the range. A hip effusion is present. Transient synovitis is a diagnosis of exclusion, the major differential being septic arthritis. Children with localised hip pain should be managed according to the algorithm in Fig. 24.1.1, and non-weight-bearing children with a hip effusion who do not have needle aspiration to rule out septic arthritis should be admitted under the close observation of the orthopaedic team. Weight-bearing children without unusual or high risk features can be managed with parental instruction, ibuprofen, and ED follow up in 2 days.17,18
Prognosis
Taylor et al showed a 15% recurrence rate in transient synovitis,19 and there is some concern about occult Perthes’ being an underlying diagnosis, especially in those with delayed bone age at the first imaging.20 All children discharged with this presumptive diagnosis should have orthopaedic follow up if pain or limp persists or recurs.
Septic arthritis and osteomyelitis
Introduction
Septic arthritis is infection of the synovial lining and fluid of a joint. Bacteria are the usual pathogens through haematogenous spread from other, sometimes occult, sites. Direct spread from adjacent bone infection may also occur, particularly where the metaphysis is intracapsular, as in the proximal femur. Phagocytic and neutrophil responses to the bacteria result in proteolytic enzyme release and cytokine production, with synovial abscess formation and cartilage necrosis.21 Pus under pressure may also reduce epiphyseal blood flow.
Most infections in children are community acquired and occur in normal joints. Infants and children under 3 are at particular risk of septic arthritis, comprising one-third and one-half, respectively, of a large paediatric series.22 As well as focal clinical inflammation, children may present with occult infection, pseudo-paralysis or generalised sepsis. Comorbidity or deficient host defences, such as those shown in Table 24.1.6, predispose to infection that may be more rapidly progressive or occur in the older child.
The disorder of CRMO has been recognised in infants and children. This is an inflammatory process of unknown origin, which demonstrates culture-negative bony inflammation with histological evidence of necrosis and chronic inflammatory cell infiltrates. Biopsy and culture are mandatory to diagnose the disorder, but antibiotics may be discontinued if pathology is consistent with CRMO. In mild cases, treatment of symptoms may be possible with non-steroidal anti-inflammatory drugs (NSAIDs); however, in more severe cases systemic steroids or treatment with intravenous bisphosphonates, and long-term follow up will be required.23–27