Mycobacterium tuberculosis monoarthritis in a child

  • Derek Rajakumar1 and

    Affiliated with

    • Alan M Rosenberg1Email author

      Affiliated with

      Pediatric Rheumatology20086:15

      DOI: 10.1186/1546-0096-6-15

      Received: 19 February 2008

      Accepted: 18 September 2008

      Published: 18 September 2008

      Abstract

      A child with isolated Mycobacterium tuberculosis monoarthritis, with features initially suggesting oligoarthritis subtype of juvenile idiopathic arthritis, is presented. This patient illustrates the need to consider the possibility of tuberculosis as the cause of oligoarthritis in high-risk pediatric populations even in the absence of a tuberculosis contact history and without evidence of overt pulmonary disease.

      Introduction

      One-third of the world's population is infected with Mycobacterium tuberculosis and the global burden of tuberculosis continues to grow [1, 2]. Approximately one-quarter to one-third of children with tuberculosis develop extrapulmonary involvement [35]. Skeletal tuberculosis, now rare since the advent of antituberculosis therapy, occurs in approximately 5% of children with extrapulmonary tuberculosis [3, 4]. The vertebral body is the most common skeletal site involved followed by lower limb bones [5, 6]. Rarely, intra-articular inflammation can occur in children either as a result of direct invasion of the tuberculous bacillus into the joint or as a consequence of an aseptic reactive arthritis (so-called Poncet's disease) related to an extra-articular tuberculous focus [320]. Intra-articular Mycobacterium tuberculosis infection is especially rare in children in the absence of associated pulmonary disease [7, 8, 13, 14, 17, 19]. Consequently, delays in recognition and treatment of this diagnostically challenging condition occur.

      We report a child with isolated Mycobacterium tuberculosis monoarthritis who presented with features initially suggesting oligoarthritis subtype of juvenile idiopathic arthritis (JIA). The patient we present illustrates the need to maintain heightened awareness that tuberculosis joint infection can occur in high-risk populations even in the absence of overt pulmonary involvement.

      Case report

      At age 2 years 10 months this previously healthy North American Indian girl presented with a 3-week history of left knee swelling and morning stiffness without associated symptoms. There were no infectious contacts reported at first presentation. On initial physical examination, the left knee was moderately swollen and warm with signs of both intra-articular fluid and synovial hypertrophy. Flexion and extension were limited by 10 degrees. The child was afebrile and appeared otherwise healthy. There were no abnormal pulmonary signs and no peripheral lymphadenopathy. The remainder of the examination was normal.

      Hemoglobin and white blood cell counts were normal. Platelet count (529 × 109/L {normal 150–400 × 109/L}) and C-reactive protein (10.6 mg/L {normal 0.0 – 7.0 mg/L}) were both elevated. The antinuclear antibody test was initially reported as negative (titer < 1:80) but on re-testing was weakly positive at a titre of 1:80. An annual tuberculin skin test, done 1 month prior to the onset of joint swelling, was negative.

      Joint swelling persisted despite 3 weeks of non-steroidal anti-inflammatory therapy (naproxen, 20 mg/kg/day). The provisional diagnosis of oligoarthritis subtype of JIA was made and, because of persistent synovitis despite oral non-steroidal anti-inflammatory therapy, the left knee was injected with triamcinolone hexacetonide 1 mg/kg. Synovial fluid cell count revealed an absolute count of 14.4 × 109/L of which 46% were neutrophils, 45% lymphocytes and 8% macrophages. Gram stain and routine bacterial cultures were negative. Synovial fluid glucose and protein levels were not measured.

      There was prompt and complete resolution of the synovitis following intra-articular steroid injection but joint swelling recurred within 2 months and oral naproxen was resumed.

      Fourteen weeks after symptom onset and 8 weeks following intra-articular steroid injection the family became aware of an aunt who had been diagnosed with pulmonary tuberculosis 6 months following our patient's arthritis onset. Seven months after arthritis onset our patient had a normal chest radiograph. A repeat tuberculin skin test at that time was positive with 15 mm of induration; the patient had not received bacillus Calmette-Guerin vaccination. Eight months after arthritis onset a small left knee effusion was aspirated and culture was positive for Mycobacterium tuberculosis. Anti-tuberculous therapy (isoniazid and rifampin) was initiated. Joint swelling resolved within two months after initiation of treatment for tuberculosis and full joint range of motion was restored. Sixteen months after arthritis onset and 8 months following initiation of anti-tuberculous therapy the child developed diffuse circumferential swelling of the left lower leg that was not associated with pain or fever. The swelling was indurated, warm to the touch, and a 1 cm erythematous area was present overlying the distal medial aspect of the leg (Figure 1). No intra-articular fluid of the knee was detected. Incision and drainage of the erythematous patch yielded a milky discharge that stained positive for acid-fast bacilli; cultures of the discharge, however, were negative for Mycobacterium tuberculosis 8 weeks after inoculation. Following approximately 8 weeks of drainage and 1 year of anti-tuberculous therapy the child remains well.
      http://static-content.springer.com/image/art%3A10.1186%2F1546-0096-6-15/MediaObjects/12969_2008_Article_38_Fig1_HTML.jpg
      Figure 1

      Marked swelling of the left lower leg first developed 16 months after arthritis onset and 8 months following initiation of anti-tuberculosis treatment. An erythematous bleb (arrow) was incised and the extruded fluid stained positive for acid fast bacilli.

      Discussion

      The patient we present illustrates the challenge in diagnosing isolated tuberculosis monoarthritis especially in the absence of a contact history and in the context of an initial negative tuberculin skin test and no overt pulmonary involvement. Our case also illustrates that the presentation of tuberculous arthritis can mimic oligoarthritis subtype of JIA. In our patient the diagnosis of tuberculous arthritis was not considered initially as there was no history of tuberculosis contact and a recent tuberculin skin test had been negative. Tuberculous arthritis was first considered when a contact with tuberculosis became evident six months after onset and was confirmed by skin test conversion to positive and eventual isolation of Mycobacterium tuberculosis from synovial fluid.

      Tuberculous arthritis can result from direct invasion of Mycobacterium tuberculosis into a joint space (as in our case) or as a consequence of an aseptic reactive polyarthritis (Poncet's disease) [16, 18]. Tuberculous arthritis associated with direct invasion is often monoarticular and characterized by an insidious onset [7, 8, 14, 19] and not typically associated with signs of acute articular inflammation [13].

      Early diagnosis of tuberculous arthritis is dependent on a high index of clinical suspicion that would be prompted particularly by knowing of an infectious contact or documenting conversion of a tuberculin skin test to positive. As illustrated by our patient, tuberculous arthritis can be difficult to distinguish from oligoarthrtis subtype of JIA in the context of an absent history of a tuberculosis contact and a recently documented negative skin test. In our case, tuberculous synovitis appeared to be a primary site of tuberculosis, rather than reactivation of a latent pulmonary or extra-pulmonary focus. Among 61 reports of tuberculous arthritis in children in which the knee was involved, including the present report, 32 (52.5%) had an apparent primary joint infection with no evidence of a pulmonary or other extra-articular focus [7, 8, 13, 14, 17, 19]. Primary tuberculous arthritis in which one knee is the only joint involved, as in our case, has been reported rarely in children (Table 1) [7, 8, 13, 14, 19].
      Table 1

      Reported cases of tuberculous knee arthritis in children

      Author (reference)

      Number of Cases

      Number with joint/bone infection only

      Appendicular joints involved

      Presentation age

      Diagnosis of tuberculosis

      Chest radiograph

      Skin test

      Southwood (18)

      1

      1

      Knees, ankles

      2 years

      Positive culture from bone

      Hilar adenopathy; no parenchymal involvement

      Positive

      Jacobs (14)

      1

      1

      Knee

      4 years

      Histopathology (giant cells); cultures and stains for TB negative

      Not stated

      Initially negative then positive

      Jacobs (14)

      1

      0

      Knee

      8 months

      Positive synovial fluid culture

      Positive

      Initially negative then positive

      Haygood (12)

      45

      Unknown

      Hip – 18

      Knee – 17

      Ankle – 5

      Shoulder – 2

      Midfoot, elbow, wrist, hand – 1 each

      6 months to 19 years

      Positive culture, histopathology and/or guinea pig inoculation

      Not stated

      Not stated

      Al-Matar (7)

      1

      1

      Knee

      2 years

      Positive culture and histopathology

      Normal

      Positive

      Al-Matar (7)

      1

      1

      Knee

      6 years

      Positive culture and histopathology

      Normal

      Positive

      Hoffman (13)

      52

      24

      Knees

      8 months to 13 years

      Predominantly histopathologic but details not stated

      Normal in 24 (47%)

      91% Positive

      Sawhney (17)

      1

      1

      Knees, elbow, shoulder

      13 years

      Positive culture, polymerase chain reaction and histopathology

      Normal

      Positive

      Aloui (8)

      1

      1

      Knee

      1 year

      Histopathology of bone lesion

      Normal

      Negative

      Aloui (8)

      1

      1

      Knee

      2 years

      Histopathology and positive culture of bone lesion

      Normal

      Not stated

      Uzel (19)

      1

      1

      Knee

      9 years

      Synovial histopathology; cultures negative

      Normal

      Positive

      Rajakumar (present case)

      1

      1

      Knee

      2 years

      Synovial fluid culture

      Normal

      Initially negative then positive

      Most patients presenting with tuberculous arthritis of a knee will present with indolent joint swelling due to synovial hypertrophy and fluid that can persist for years before diagnosis [13]. Approximately 15% of patients with tuberculous arthritis of a knee will have warmth and redness accompanying the swelling suggesting septic arthritis [13]. Non-specific indicators of inflammation, including fever and elevated acute phase reactants, are inconsistently present and cannot be relied on as discriminating diagnostic indicators. A positive Mantoux skin test is expected in almost all patients with tuberculous arthritis [13] but, as in our case, a positive skin test might not necessarily be present at initial arthritis presentation. Definitive diagnosis requires microbiological confirmation by isolating Mycobacterium tuberculosis from synovial fluid or synovium. Synovial membrane histopathology showing caseating granuloma, without documented microbiological confirmation, is suggestive of tuberculous arthritis. Polymerase chain reaction (PCR) has been reported to have low sensitivity (but high specificity) in detecting Mycobacterium tuberculosis from the knee in children [13]; however, the reliability of PCR in detecting Mycobacterium tuberculosis has been documented in other reports [21]. Synovial fluid analysis is not associated with sufficiently characteristic features in tuberculous synovitis to be consistently helpful diagnostically [22].

      Tuberculous arthritis can be associated with erosion through the joint capsule to create draining sinuses, a circumstance which likely occurred in our patient and accounted for lower leg swelling and chronic drainage. Approximately 10% of patients with Mycobacterium tuberculosis knee arthritis will have a sinus at the time of initial presentation [13]. In our patient discharge from the sinus stained positive for acid-fast bacilli but cultures were negative suggesting the presence of non-viable, post-treatment remnants of the offending organism.

      In high-risk populations with unexplained monoarthritis mimicking JIA, documenting a negative tuberculin skin test would seem prudent before proceeding to intra-articular corticosteroid injection. However, as demonstrated in our patient in whom a negative skin test was recorded 1 month prior to the onset of joint swelling, a delay in conversion to a positive skin test can occur after exposure to Mycobacterium tuberculosis. In our patient, treatment with intra-articular corticosteroids resulted in prompt but unsustained resolution of synovitis. The recurrence of joint swelling within 2 months of injection might have been an indication that the synovitis was different than oligoarthritis subtype of JIA in which a more sustained response to intra-articular triamcinolone hexacetonide therapy is typical.

      In our patient intra-articular corticosteroids, which traditionally have been considered to be contraindicated in septic arthritis, were administered inadvertently but without any apparent adverse consequences. There are no substantive data to support the use of intra-articular steroids as adjunctive therapy in septic arthritis. However, concomitant use of systemic antibiotics and intra-articular steroids are not deleterious in experimental animals [23] and it has been suggested that the use of intra-articular steroids might be chondroprotective [24]. Thus, it is conceivable, though unproven, that intra-articular corticosteroids could be effective adjunctive therapy in septic arthritis by blunting inflammatory mediators that contribute to cartilage and bone degradation. The theoretical role for intra-articular steroids as adjunctive therapy for septic arthritis might prove to be analogous to the putative therapeutic efficacy of steroids in sepsis occurring at other sites including in the treatment of tuberculosis of the pericardium, pleura and meninges [25].

      Conclusion

      Our experience illustrates the need to maintain a high index of suspicion for the possibility of tuberculosis as the cause of arthritis in high-risk populations even when a tuberculosis contact is not immediately evident, when there is no overt pulmonary involvement, and when a recent tuberculin skin test is negative.

      Declarations

      Consent

      Written consent for publication of this article has been obtained from a parent of the child.

      Authors’ Affiliations

      (1)
      Department of Pediatrics, University of Saskatchewan

      References

      1. Dye C, Scheele S, Dolin P, Pathania V, Raviglione MC: Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. Journal of the American medical association. 1999, 282: 677-686. 10.1001/jama.282.7.677.View ArticlePubMed
      2. Global tuberculosis control 2002: Surveillance, Planning, Financing. WHO Report.
      3. Maltezou HC, Spyridis P, Kafetzis DA: Extra-pulmonary tuberculosis in children. Archives of disease in childhood. 2000, 83: 342-346. 10.1136/adc.83.4.342.PubMed CentralView ArticlePubMed
      4. Nelson LJ, Schneider E, Wells CD, Moore M: Epidemiology of childhood tuberculosis in the United States, 1993–2001: the need for continued vigilance. Pediatrics. 2004, 114: 333-341. 10.1542/peds.114.2.333.View ArticlePubMed
      5. Rieder HL, Snider DE, Cauthen GM: Extrapulmonary tuberculosis in the United States. The American review of respiratory disease. 1990, 141: 347-351.View ArticlePubMed
      6. Vallejo JG, Ong LT, Starke JR: Tuberculous osteomyelitis of the long bones in children. The Pediatric infectious disease journal. 1995, 14: 542-546. 10.1097/00006454-199506000-00013.View ArticlePubMed
      7. Al-Matar MJ, Cabral DA, Petty RE: Isolated tuberculous monoarthritis mimicking oligoarticular juvenile rheumatoid arthritis. The Journal of rheumatology. 2001, 28: 204-206.PubMed
      8. Aloui N, Nessib N, Jalel C, Ellouze S, Essayed M, Bellagha I, Ben Chehida F, Ghachem M, Hammou A: [Tuberculosis of the knee: MRI findings in two pediatric cases]. Journal de radiologie. 2003, 84: 861-864.PubMed
      9. Berney S, Goldstein M, Bishko F: Clinical and diagnostic features of tuberculous arthritis. The American journal of medicine. 1972, 53: 36-42. 10.1016/0002-9343(72)90113-1.View ArticlePubMed
      10. Ellis ME, el-Ramahi KM, al-Dalaan AN: Tuberculosis of peripheral joints: a dilemma in diagnosis. Tubercle and lung disease. 1993, 74: 399-404. 10.1016/0962-8479(93)90084-B.View ArticlePubMed
      11. Garrido G, Gomez-Reino JJ, Fernandez-Dapica P, Palenque E, Prieto S: A review of peripheral tuberculous arthritis. Seminars in arthritis and rheumatism. 1988, 18: 142-149. 10.1016/0049-0172(88)90007-8.View ArticlePubMed
      12. Haygood TM, Williamson SL: Radiographic findings of extremity tuberculosis in childhood: back to the future?. Radiographics. 1994, 14: 561-570.View ArticlePubMed
      13. Hoffman EB, Allin J, Campbell JA, Leisegang FM: Tuberculosis of the knee. Clinical orthopaedics and related research. 2002, 100-106. 10.1097/00003086-200205000-00014.
      14. Jacobs JC, Li SC, Ruzal-Shapiro C, Kiernan H, Parisien M, Shapiro A: Tuberculous arthritis in children. Diagnosis by needle biopsy of the synovium. Clinical pediatrics. 1994, 33: 344-348. 10.1177/000992289403300606.View ArticlePubMed
      15. Papavasiliou VA, Petropoulos AV: Bone and joint tuberculosis in childhood. Acta orthopaedica Scandinavica. 1981, 52: 1-4.View ArticlePubMed
      16. Poncet A: De la polyarthrite tuberculose deformate ou pseudorheumatism chronique tuberculeux. Congres Francais Chirugie. 1887, 1: 732-739.
      17. Sawhney S, Murray KJ: Isolated tuberculosis monoarthritis mimicking juvenile rheumatoid arthritis. The Journal of rheumatology. 2002, 29: 857-859. author reply 860PubMed
      18. Southwood TR, Hancock EJ, Petty RE, Malleson PN, Thiessen PN: Tuberculous rheumatism (Poncet's disease) in a child. Arthritis Rheum. 1988, 31: 1311-1313. 10.1002/art.1780311014.View ArticlePubMed
      19. Uzel M, Garipardic M, Cetinus E, Bilgic E, Karaoguz A, Boran C, Oztoprak I: Tuberculosis of the knee in a child. Journal of tropical pediatrics. 2004, 50: 182-184. 10.1093/tropej/50.3.182.View ArticlePubMed
      20. Zahraa J, Johnson D, Lim-Dunham JE, Herold BC: Unusual features of osteoarticular tuberculosis in children. The Journal of pediatrics. 1996, 129: 597-602. 10.1016/S0022-3476(96)70126-9.View ArticlePubMed
      21. Negi SS, Gupta S, Khare S, Lal S: Comparison of various microbiological tests including polymerase chain reaction for the diagnosis of osteoarticular tuberculosis. Indian journal of medical microbiology. 2005, 23: 245-248.PubMed
      22. Wallace R, Cohen AS: Tuberculous arthritis: A report of two cases with review of biopsy and synovial fluid findings. The American journal of medicine. 1976, 61: 277-282. 10.1016/0002-9343(76)90177-7.View ArticlePubMed
      23. Wysenbeek AJ, Leitman M, Amit M, Lushkov G, Robinson D, Boldur I, Nevo Z: Experimental septic arthritis in rabbits treated by a combination of antibiotic and steroid drugs. Clin Exp Rheumatol. 1996, 14 (5): 507-512.PubMed
      24. Stricker SJ, Lozman PR, Makowski AL, Gunja-Smith Z: Chondroprotective effect of betamethasone in lapine pyogenic arthritis. Journal of pediatric orthopedics. 1996, 16: 231-236.View ArticlePubMed
      25. Evans DJ: The use of adjunctive corticosteroids in the treatment of pericardial, pleural and meningeal tuberculosis: Do they improve outcome?. Respiratory medicine. 2008, 102: 793-800. 10.1016/j.rmed.2008.01.018.View ArticlePubMed

      Copyright

      © Rajakumar and Rosenberg; licensee BioMed Central Ltd. 2008

      This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://​creativecommons.​org/​licenses/​by/​2.​0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

      Advertisement