Ten years follow up of a chronic tubercular osteomyelitis of the proximal tibia in infancy

INTRODUCTION

Primary tuberculous osteomyelitis accounts for less than 10% of extra-pulmonary tuberculosis (TB) in childhood.¹ In about 50% of the patients, the vertebrae are the sites of skeletal involvement.² Solitary bone TB is rare in infancy and has non-specific findings and can be misdiagnosed easily.³ Skeletal lesions tend to be isolated to one anatomical site and multifocal tuberculosis with more than one tuberculous osteo-articular lesion is rare, especially in immunocompetent individuals.

In infants the epiphysis can be destroyed by bacterial infection spreading from a metaphyseal focus. At this young age transphyseal vessels provide a pathway for infection to spread.^5,6 Skeletal Tuberculosis is often difficult to diagnose initially; several weeks or months may be required to diagnose it correctly. In the case reported by Chen SC et al, it took 80 days to confirm the diagnosis. The variable radiological appearance of isolated bone lesions in children can resemble various bone lesions including subacute and chronic osteomyelitis, simple and aneurysmal bone cysts, cartilaginous tumors, osteoid osteoma, granulomatous lesions, hematological disease and certain malignant tumors. Surgical debridement and curettage is indicated for cases with diffuse bone destruction.⁴ To the best of our knowledge, tuberculosis of proximal metaphysis of tibia in infancy is very rare and not reported yet.

The objective of the study was to closely follow up the patient, identify any limb length inequality and angular deformities and treat those at the earliest.

CASE REPORT

A 12 month old infant, presented with fever, swelling, pain, difficulty in using the left lower limb and inability to bear weight for 15 days. At the first visit, clinically, the infant was mildly toxic with features suggestive of osteomyelitis of proximal tibia on the left side. No other joints were involved. Heamotological parameters revealed raised total leucocyte count (TLC) 24,000 cells/cm³ with neutrophilia 82%, and erythrocyte sedimentation rate (ESR) 60 mm/hr, whereas C-reactive protein was within normal limits. Blood cultures were negative. Radiographically, lytic pathology was seen in and around the central part of the growth plate on the metaphyseal side. It was associated with attenuation of growth plate, with peri articular osteopenia, subchondral erosion, periosteal reaction and increased joint space of the left knee joint (Figure 1). Aspiration of the lesion, and gonocentesis was done by two separate portals by a thick needle and no organisms were isolated. Hence, she was managed conservatively initially with injectable antibiotics and slab support. The TLC came down to 17,900 cells/cm³ with neutrophils being 32% and ESR normalized to 6 mm/hr.

At the second visit, after 15 days, in view of recurrence of fever (2 episodes) and swelling, a proper biopsy was performed under anesthesia. Intra Operatively a pocket of purulent fluid was found. The biopsy showed plenty of polymorphs, with negative staining for acid fast Bacilli and fungal elements. No organism was isolated on culture. Histopathology revealed possible suppurative inflammation. At this juncture, chronic atypical Brodie’s abscess and chronic recurrent multifocal osteomyelitis was thought of and continued with broad spectrum antibiotics with a wishful thinking of remission.

At the third visit, even after a fortnight with antibiotics, she was worsening (TLC 25,100 cells/cm³, neutrophils 40%, lymphocytes 50%, ESR 56 mm/hr) so we stopped the antibiotics to get a positive culture, and planned for a repeat biopsy. She was operated by means of thorough debridement and gentle curettage of the left proximal tibial metaphysis to mitigate the iatrogenic proximal tibial physeal injury. Intraoperatively cheesy caseous material was found which was meticulously collected for biopsy and culture. The lesion was extending to the entire proximal tibia with involvement of the proximal physis. Once again culture and gram stain was negative and histopathology revealed chronic caseating granulomatous inflammation suggestive of Mycobaterial tuberculosis pathology (Figure 2). Simultaneously the material also subjected to TB culture, which showed growth of mycobacteria species in BACTEC-460 TB system at the end of four weeks, which was sensitive to all primary line of drugs.

Titrated anti tubercular treatment for her body weight was started immediately with four drugs after her biopsy report. She responded very well both clinically and radiologically. The TLC improved to 10,800 cells/cm³, neutrophils 49%, lymphocytes 42% and ESR was 32 mm/hr. The inflammatory signs disappeared and the surgical wound healed uneventfully. At the end of one month the slab was removed and started with knee range of movement. Weight bearing was delayed till 10 weeks as radiologically we felt the involucrum was inadequate till then. Later on she was switched to maintenance therapy with two drugs rifampicin and isoniazid for next 10 months.

The patient made a dramatic recovery in relation to pain and tenderness. Radiologically the lesion showed features of healing and the laboratory parameters started to normalize (Figure 3). The TLC improved to 10,800 cells/cm³, neutrophils 49%, lymphocytes 42% and ESR was 32 mm/hr.

The patient was followed up regularly to identify and treat the sequelae of the infection as we expected growth disturbances, angular deformities and knee stiffness as the lesion was large, involving the physis, the patient being an infant and proximal tibia being the main growing end of the bone. Lesion healed well with a significant decrease in the size of the lesion, without any formation of a physeal bar and no residual deformity (Figures 4 and 5).

Over 10 years good remodeling was noted and the patient had excellent range of motion (0-130^°), no limb length discrepancy and no angular deformity (Figures 6-8).

DISCUSSION

Tuberculous affection of bone in this part of the World is fairly common and tuberculosis is the greatest masquerader. One has to have high index of suspicion to pick up the infection early. Because of relative rarity of of skeletal tuberculosis and non specific early clinical findings, the diagnosis is usually considerably delayed. The average duration of symptoms before treatment was 2.3 months (range 2 weeks to 6 months).⁸ Chen et alreported an average 6.6 month delay in the diagnosis of tuberculous osteomyelitis.⁹ Tekali et al reported an average delay in diagnosis of 10 months (range 10 days to 6 years), indicating the difficulty in making a prompt diagnosis. In our case it took 55 days, from the date of presentation to confirm the diagnosis.¹⁰ Despite tuberculosis being common, proximal tibial involvement in infancy is rare and sparingly reported. The common involvement at this region in infancy is heamotagenous osteomyelitis.

The nearest differential was Brodie’s abscess, but it was not typical as no sclerosis of the cavity was seen and the child continued to have progressive changes clinico-radiologically. Brodie’s abscess presents in early childhood and it’s relatively indolent and radiologically it does not progress rapidly. Its slow clinical course is similar to the course of patients with tuberculosis, thus, they are often mixed. Metaphyseal involvement is usually seen among children while diaphyseal involvement is seen among adults.¹¹ Vohra et al reported about a case series of 28 patients with tuberculosis osteomyelitis and mentioned of six cases who had pyogenic osteomyelitis and exhibited clinical and radiological findings similar to Brodie’s abscess.¹²

In 75% of tuberculous osteomyelitis cases, causative organism in the lungs is spread via hematogenous ways.¹² The commonly affected sites in order of frequency are, Spine, femur, tibia and fibula.¹³ The clinical features did not support the diagnosis of acute on chronic osteomyelitis as the infant was not toxemic or septicemic. The most common symptoms include pain, swelling and limp.⁴

Four types of lesions are observed in tuberculous osteomyelitis i.e. cystic, infiltrative, regional erosion and spina ventosa.¹³ They usually localize in the metaphyses of long bones or flat bones as single or, less commonly, multiple foci. The lesions are radiolucent, round to oval with slight marginal sclerosis. These ‘cysts’ can mimic lesions such as simple bone cysts and aneurysmal bone cysts, tumors and granulomatous bone lesions. Infiltrative lesions are diffuse areas of permeation with little or no periosteal reaction, resembling chronic osteomyelitis, benign and malignant tumors such as leukaemias, round cell tumors and Ewing’s sarcoma. Pathological fractures usually occur with this radiological type.¹⁴

While treating these entities one should be aware of early stages of chronic recurrent multifocal osteomyelitis (CRMO) which may not be multifocal to start with. It was first described by Gideon in 1972 as an unusual form of multifocal bone lesions with subacute and chronic symmetrical osteomyelitis. The age of onset is childhood and it’s again culture negative osteitis. These inflammatory conditions are characterized by episodes of systemic inflammation including serological signs of inflammation (CRP, ESR, interleukin-6, tumor necrosis factor-a) occurring in the absence of autoantibodies and antigen specific T cells.¹⁵ The treatment of CRMO is mainly anti-inflammatories and not antibiotics. The entire management of this entity revolves around good biopsy and culture. In this case the biopsy was repeated as the first biopsy couldn’t yield anything and the appropriate treatment was delayed. A meticulous biopsy after studying the scans will entail us to procure the samples from the affected area. Some patients with CRMO may have a markedly raised CRP but these are likely to require a biopsy in order to execute an infectious agent. In cases of deep seated areas, MRI or PET CT scans do find a place in identifying the area of maximum involvement. Biopsies preferably should remain extra-articular and joint aspirations should not be clubbed with it. A properly executed biopsy will expedite the treatment and improves the outcome. Often, the tumor can be mixed with cases of infection. In order to achieve rapid diagnosis and treatment, “culturing for every tumor and biopsy for every infection” should always be kept in mind.¹² Culture results are positive in only 29%-61% of subacute osteomyelitis cases. This rate decreases to 10%-30% in tuberculosis cases.¹⁴

The outcome of infections in growing bones is a matter of concern. The ill effects of physis affection due to infection will lead to growth disturbances, in form of angular deformities and limb length discrepancy. In this case, fortunately the child didn’t have these problems despite the involvement of physis. We believe this is due to the chronic hyperemia caused by the infection and slow destruction by the mycobacterial infections and relatively early detection and treatment and patient’s ability of child to remodel. Partial physeal destruction with chronic hyperemia may be the cause of no limb length discrepancy.

The destruction was in the central area which might have neutralizing effect in relation to angular deformities which is common sequelae after pyogenic infections. The Knee range was not captured owing to its metaphyseal location. The lesion crossing the epiphyseal plate, but not involving the joint must have preserved the knee range. As the treatment was initiated early, natural progression of the disease process could be curtailed with good remodelling potential and sequential follow up. The ill effects of bone infection were not found in this case.

CONCLUSION

Skeletal Tuberculosis in infancy is extremely rare. In treating infections of bone, one has to have this entity in the thought process. Tuberculosis has very few clinical and radiological parameters to diagnose the problem. With a high index of suspicion the biopsy should be well planned and meticulous and always samples to be sent for mycobacteria. Tuberculosis should be an essential differential diagnosis while dealing with cases with similar presentation in an endemic region like India. For every case of osteomyelitis in infancy, tuberculosis should be ruled out. Long follow up is essential for monitoring the sequelae of the bone growth. As clinical and radiological findings may be indistinguishable from malignant disease, tuberculous osteitis should be excluded in infants presenting with undiagnosed bone lesions despite been vaccinated with Bacille Calmette Guerin at birth. Even if radiographs reveal a substantial destruction of the bony epiphysis and metaphysis, the appearance may be deceptive as the remaining chondro-epiphysis can restore the epiphysis and lead to normal joint development. In immature skeleton the lesion has to be followed till skeletal maturity to detect and treat the sequelae at the earliest.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: Not required

REFERENCES

1. 1. Teo HE, Peh WC. Skeletal tuberculosis in children. Pediatr Radiol. 2004;34(11):853-60.
2. 2. Abdelwahab IF, Present DA, Gould E, Klein MJ, Nelson J. Tuberculosis of the distal metaphysis of femur. Skeletal Radiol. 1988;17:199-202.
3. 3. Gümüştaş SA, Çağırmaz T, Orak MM, Pehlivanoğlu G, Öktem S. Cystic tuberculosis osteomyelitis of the distal tibia in infancy. Turk Pediatri Ars. 2017;52(1):53-6.
4. 4. O’Connor BT, Steel WM, Sanders R. Disseminated Tuberculosis. J Bone Joint Surg. 1970;52:537-42.
5. 5. Halberstein BM. Bone regeneration in infantile osteomyelitis: report of a case with fourteen year follow up. J Bone Joint Surg (Am). 1967;49-A:149-52.
6. 6. Ogden JA. Pediatric osteomyelitis and septic arthritis: the pathology of neonatal disease. Yale J Diol Med. 1979;52:423-48.
7. 7. Dhammi IK, Jain AK, Singh S, Aggarwal A, Kumar S. Multifocal skeletal tuberculosis in children: a retrospective study of 18 cases. Scand J Infect Dis. 2003;35:797-9.
8. 8. Hsuan-Kai K, Yang WE, Shih HN, Chang CH. Physeal change after tuberculous osteomyelitis of the long bone in children. Chang Sung Med J. 2010; 33(4):453-60.
9. 9. Chen SC, Huang SC, Wu CT. Non Spinal tuberculous osteomyelitis in children. J Formos Med Association. 1998;97:26-31.
10. 10. Teklali Y, Alami ZF, Madhi T, Gourinda H, Miri A. Peripheral osteoarticular tuberculosis in children: 106 Case reports. Joint, bone, spine:revue du rhumatisme. 2003;70:282-6.
11. 11. Olasinde AA, Oluwadiya KS, Adegbehingbe OO. Treatment of Brodie’s abscess: excellent results from curet-tage, bone grafting and antibiotics. Singapore Med J. 2011;52(6):436-9.
12. 12. Vohra R, Kang HS, Dogra S, Saggar RR, Sharma R. Tuberculous osteomyelitis. J Bone Joint Surg Br. 1997;79:562-6.
13. 13. Vallejo JG, Ong LT, Starke JR. Tuberculous osteomyelitis of long bones in children. Pediatr Infect Dis J. 1995;14:524-6.
14. 14. Rasool MN, Govender S, Naidoo KS. Cystituberculosis of bone in children. J Bone Joint Surg Br. 1994;76:113-7.
15. 15. Hedrich CM, Hofmann SR, Pablik J, Morbach H, Girschick HJ. Autoinflammatory bone disorders with special focus on chronic recurrent multifocal osteomyelitis (CRMO). Pediatr Rheumatol Online J. 2013;11:47.