Brodie abscess (BA) is a form of subacute hematogenous osteomyelitis characterized by the presence of a well-defined abscess normally in cancellous bone.1–3 The incidence of subacute osteomyelitis is approximately 5 in 100,000 children per year in high-income countries but may be higher in low- and middle-income countries.4–6 Diagnosis of BA is challenging as patients may present with minor symptoms and unrevealing laboratory studies. Biopsy for culture and histologic examination is often needed in addition to radiography to diagnose BA. Knowledge of BA in pediatric patients is mostly limited to individual case reports and case series, with a strong focus on the imaging appearance of the abscess. Prior studies on this entity in adults have reported a wide range and variability in treatment approaches.3 , 7 To determine the current epidemiology and clinical outcomes of children with BA, we conducted a retrospective review of patients with BA seen by the pediatric infectious diseases service at Texas Children’s Hospital (TCH) from 2007 to 2017.
The consultation database of the pediatric infectious diseases service at TCH was queried for patients using the search terms “Brodie abscess,” “subacute osteomyelitis” and “chronic osteomyelitis” between the dates of July 1, 2007, and November 1, 2017. Patients were included in this study if they had a BA defined as subacute or chronic osteomyelitis with radiographic evidence of an intraosseous abscess. Presence of a BA was independently verified by a pediatric musculoskeletal radiologist who remeasured intraosseous abscess size, location and volume. Demographic and clinical information was obtained from the electronic medical record. Age, gender, ethnicity, clinical symptoms, symptom duration, white blood cell count, erythrocyte sedimentation rate, C-reactive protein and microbiologic data were recorded. This study was approved by the Baylor College of Medicine Institutional Review Board.
We identified 21 potential patients for inclusion in the study. Three patients were excluded because of alternative diagnoses: osteoid osteoma, aneurysmal bone cyst and Rosai Dorfman disease; 3 patients were excluded because of the absence of an intraosseous abscess. Fifteen patients met inclusion criteria, and their clinical features are described in Table 1.
The male to female ratio was 1.1:1. The median age was 4 years (range: 4 months to 12 years). One patient had sickle cell trait. The 2 most common presenting symptoms were pain and limp. Subjective fever was reported in 27% (4/15) of the cases. Symptom duration ranged from 4 days to 2 years, although the median duration was 3 weeks. The femur was the most frequently involved bone (10/15, 67%). There were 2 cases of BA involving bones of the ankle. There was one case of bilateral, near symmetric BA involving the proximal femurs of a medically complex child. Three patients had white blood cell counts >12,000 cells/mm3, 5 patients had C-reactive protein >1 mg/dL and 8 patients had an erythrocyte sedimentation rate >20 mm/h. Eight of the 15 patients had blood cultures performed, and 1 was positive. Radiographs and magnetic resonance imaging were performed in each case. The lesions were epiphyseal (5), metaphyseal (5) or metaphyseal crossing into epiphysis (4); no lesions involved the diaphysis.
One patient had no procedure, and the remaining patients underwent either percutaneous bone biopsy (2) or open surgical incision and drainage with debridement (12). All surgical incisions were closed primarily, and 9 patients had a drain placed. Specimens were sent for histology and culture (aerobic, anaerobic, fungal and mycobacterial). The most common etiology of BA was Staphylococcus aureus (9/15, 60%). Seven patients had methicillin-susceptible S. aureus, and 2 patients had methicillin-resistant S. aureus. One of the S. aureus isolates displayed constitutive clindamycin resistance. Bacterial and fungal DNA detection using real-time polymerase chain reaction (University of Washington Department of Laboratory Medicine, Seattle, WA) was performed in 3 cases and established the etiology in 1 case (Mycobacterium tuberculosis). The causative pathogen was not identified in 5 cases (5/15, 33%).
Clindamycin and cephalexin were frequently used for treatment. Eight patients received intravenous antibiotics during admission and were transitioned to oral antibiotics at discharge, 5 patients had a peripherally inserted central catheter placed and were discharged on intravenous therapy and 2 patients received oral antibiotics only. Duration of antibiotic treatment ranged from 4 weeks to 2 years. No patients were readmitted for peripherally inserted central catheter complications. Eight patients had successful treatment, 4 patients required readmission for repeat incision and drainages and had extended antibiotic courses, 2 patients were lost to follow-up and 1 patient remains on treatment. There was one reported case of acquired leg length discrepancy because of femoral overgrowth and no cases of pathologic fracture.
We performed a retrospective review of children with BA. As previously described in the literature, nearly all cases occurred in the lower extremities. Although the tibia has been described as the most commonly involved bone, we had more cases involving the femur.3 , 7 To our knowledge, our 4-month-old patient is the youngest reported patient with a BA in the literature.
The main presenting symptom in this series was pain (12/15, 80%), followed by limp (9/15, 60%). Four patients presented with swelling of the involved extremity. These characteristics differ from that of Kozlowski et al8 who reported that local swelling was present in nearly all cases (10/11), with limping and pain in approximately half the cases (5/11 and 6/11, respectively). The overall benign clinical appearance of most patients with BA presents a challenge for medical providers. Additionally, laboratory findings do not always indicate a diagnosis of infection. Erythrocyte sedimentation rate was the most frequently abnormal value, although it was elevated in only 53% of our patients. Only 5 patients had more than one abnormal laboratory value.
Radiographically, BA appears as a lytic lesion with a sclerotic rim. A “target sign” or “penumbra sign” has been described on magnetic resonance imaging in which 4 concentric layers are observed consisting of a central necrotic core, granulation tissue, sclerotic or fibrotic tissue and edema.2 The penumbra sign, although highly suggestive of BA, is not pathognomonic. The differential for osseous lesions is broad, including chondroblastoma, osteoid osteoma, simple bone cysts, localized Langerhans cell histiocytosis and rarely malignant bone tumors.2 In our review, 3 patients were excluded because of alternative diagnoses. There are several examples in the literature that underscore the difficulty of diagnosing mimickers of BA, including osteoid osteoma and Langerhans cell histiocytosis.9 , 10 These reports highlight the importance of pursuing a tissue diagnosis to exclude other causes of osseous lesions.
Most case reports and series in the literature describe S. aureus as the primary pathogen causing BA, which we also observed in our study. Spyropoulou et al11 found Kingella kingae to be a common cause of subacute hematogenous osteomyelitis in patients younger than 4 years of age. Polymerase chain reaction for K. kingae is not routinely performed in our center and may account for a portion of the culture-negative cases. Negative intraoperative cultures should increase suspicion of BA secondary to fastidious organisms. Increased availability of DNA-based testing may improve diagnostic yield in cases of negative bone cultures.
Management of BA typically requires surgical incision and drainage in combination with antibiotic therapy. Bone graft may be required for correction of larger bony defects, >3 cm.3 One patient in our study received a bone graft. Duration and route of antimicrobials may vary. The timing of transition from intravenous to oral therapy for osteomyelitis in children is not well defined, although several studies suggest good outcomes with short-term parenteral treatment of uncomplicated osteomyelitis.5 , 12 We found that treatment courses for BA at our institution often extended beyond 6 weeks. Successful treatment was frequently achieved, although some patients required multiple surgical procedures. Stephens and MacAuley3 reported recurrence in 8 of 21 patients with BA while Olasinde et al7 reported no recurrence among 20 patients treated for BA. In the absence of an unusual presentation or pathogen, incision and drainage plus antibiotic treatment until radiographic improvement were sufficient to achieve a good outcome.
There are some limitations to our retrospective study. In addition to the limitations inherent in retrospective design, it is likely that use of the pediatric infectious diseases service database led to an underestimate of the number of cases of BA. We would not have captured information on patients for whom infectious diseases consultation was not sought; therefore, the number of cases reported should not be taken as a measure of overall incidence of BA. Furthermore, we would have missed patients presenting to outside institutions for recurrence or treatment failure. Results of our single institution study may not be representative of other geographic areas.
The authors thank Frances Morales for her assistance in querying the TCH pediatric infectious diseases consultation database.
1. Brodie BC. An account of some cases of chronic abscess of the Tibia. Med Chir Trans. 1832;17:239–249.
2. Gould CF, Ly JQ, Lattin GE Jr, et al. Bone tumor mimics: avoiding misdiagnosis. Curr Probl Diagn Radiol. 2007;36:124–141.
3. Stephens MM, MacAuley P. Brodie’s abscess. A long-term review. Clin Orthop Relat Res. 1988:211–216.
4. Riise ØR, Kirkhus E, Handeland KS, et al. Childhood osteomyelitis
-incidence and differentiation from other acute onset musculoskeletal features in a population-based study. BMC Pediatr. 2008;8:45.
5. Dartnell J, Ramachandran M, Katchburian M. Haematogenous acute and subacute paediatric osteomyelitis
: a systematic review of the literature. J Bone Joint Surg Br. 2012;94:584–595.
6. Peltola H, Pääkkönen M. Acute osteomyelitis
. N Engl J Med. 2014;370:352–360.
7. Olasinde AA, Oluwadiya KS, Adegbehingbe OO. Treatment of Brodie’s abscess: excellent results from curettage, bone grafting and antibiotics. Singapore Med J. 2011;52:436–439.
8. Kozlowski K. Brodie’s abscess in the first decade of life. Report of eleven cases. Pediatr Radiol. 1980;10:33–37.
9. Chang WF, Hsu YC, Wu YD, et al. Localized Langerhans cell histiocytosis masquerading as Brodie’s abscess in a 2-year-old child: a case report. EXCLI J. 2016;15:33–37.
10. Agrawal P, Sobti A. A Brodie’s abscess of femoral neck mimicking osteoid osteoma: diagnostic approach and management strategy. Ethiop J Health Sci. 2016;26:81–84.
11. Spyropoulou V, Dhouib Chargui A, Merlini L, et al. Primary subacute hematogenous osteomyelitis
: a clearer bacteriological etiology. J Child Orthop. 2016;10:241–246.
12. Peltola H, Pääkkönen M, Kallio P, et al; Osteomyelitis
-Septic Arthritis Study Group. Short- versus long-term antimicrobial treatment for acute hematogenous osteomyelitis
of childhood: prospective, randomized trial on 131 culture-positive cases. Pediatr Infect Dis J. 2010;29:1123–1128.