Advertisement

Poncet's disease—a novel cause of non-compliance with anti-tuberculous drugs

      Introduction

      We present a patient with isoniazid resistant pulmonary tuberculosis (Tb) associated with large joint polyarthropathy. A diagnosis of Poncet's disease (PD) was made. This case highlights the problems of non-compliance with drug therapy, directly observed therapy (DOT) and difficulty in diagnosing PD. The mechanisms for PD are discussed and the links between mycobacterial infection and rheumatoid arthritis (Rh.A) are reviewed.

      Case report

      A 42-year old married man presented with a 6-month history of cough, sputum, night sweats, dyspnoea and weight loss. He moved to England from Tanzania in 1974 and had subsequently not left the country. He was a heavy consumer of alcohol (20 units/day), moderate smoker, and had intermittently been sleeping rough since selling his house to buy alcohol. A BCG vaccination was given at birth and he denied any Tb contacts, drug abuse or high-risk sexual behaviour. The only previous medical history of note was anaemia secondary to haemorrhoids.
      Clinical examination revealed a temperature of 37.2°C, SaO2 of 92% on air, respiratory rate of 24min−1, heart rate 128min−1 sinus rhythm, BP 140/64mmHg and course inspiratory crackles in the left upper zone anteriorly were heard on auscultation of the chest. The examination was otherwise unrevealing. Initial results were as follows; Na 132mmol/l, K 4.2mmol/l, U 3.4mmol/l, Cr 65μmol/l, Alb24g/l, total protein 50g/l, Alp 257IU/l, Alt 62IU/l, Hb 7.7g/dl, WCC 8.1×109/l, Plt 536×109/l, CRP 196mg/l. Blood clotting was within normal limits. Urinalysis was normal and urine culture was sterile after incubation. Sputum analysis demonstrated acid and alcohol-fast bacilli on Zeihl–Neilsen staining. Chest X-ray showed bilateral mid-zone shadowing with cavitation (Fig. 1).
      Figure thumbnail gr1
      Figure 1Extensive bilateral midzone shaddowing and cavitation as a result of pulmonary Tb.
      The patient was isolated and treated for pulmonary Tb with rifater and ethambutol. One week into his admission, he developed a painful, swollen left knee. Plain X-rays and MRI examination of the joint were normal. Joint aspiration revealed a lymphocytosis only. Synovial fluid cultures were negative. The knee pain settled with simple analgesics. Several days later he developed bilateral ankle discomfort, stiffness, swelling, and lumbar back pain. Again plain X-rays and a lumbar spine MRI scan were normal. Serum urate was 476μmol/l (<450μmol/l), ds DNA, ANCA, HIV, HCV, HBV serology, were negative and rheumatoid factor was elevated at 204IU/ml (0–40IU/ml). The arthropathy settled with analgesia and physiotherapy. At 2 weeks, sputum culture grew isoniazid resistant Mycobacterium tuberculosis. The organism was sensitive to ciprofloxacin, which was added in the place of isoniazid. He was discharged 25 days after admission, on rifampicin, pyrazinamide, ethambutol and ciprofloxacin. At 2 months he claimed to be compliant with his treatment and this was supported by urine discoloration. The ciprofloxacin and pyrazinamide were discontinued. A month later he complained of difficulty in walking due to bilateral ankle pain and swelling. He also reported a painful right knee, right shoulder pain and bilateral wrist stiffness. He had stopped taking his anti-Tb drugs since he attributed his joint symptoms to these. On clinic review, his sputum was found to be positive for isoniazid resistant Tb (fully sensitive to both rifampicin and ethambutol).The patient agreed to DOT with rifampicin and ethambutol, where he is now supervised thrice-weekly on the chest ward.
      In the rheumatology clinic, the affected joints were repeatedly X-rayed and aspirated. The radiographic appearances were normal and aspirate analysis revealed only leucocytosis. A synovial biopsy of the left knee demonstrated a sterile inflammatory infiltrate of lymphocytes with occasional plasma cells and neutrophills. No granulomata or organisms were seen or cultured. Both of the knee joints, right shoulder and right ankle were injected with steroids. Over the course of the next few months, his joint symptoms resolved and his sputum remained negative for Tb. He continues to have DOT.

      Discussion

      The musculoskeletal manifestations of Tb are seen in 2% of cases. Tb arthritis is the commonest form and is usually a large joint monoarthropathy where mycobacteria are isolated from the affected joint. Pott's disease is thoracic spine involvement associated with a perispinous abscess. A very rare form of arthropathy in Tb infection is Poncet's disease. In 1892, Grocco suggested that patients with Tb might present with arthritis unassociated with tubercles or abscesses.
      • Dall L
      • Long L
      • Stanford J
      Poncet's disease Tuberculous rheumatism.
      In 1897, Poncet described a polyarthritis in the acute stage of Tb, which resolved without joint damage.

      Poncet A. De la polyarthrite tuberculeuse deformante ou pseudo-rheumatisme chronique tuberculeux. Congr Fr Chir 1897; 732.

      In PD, no evidence of bacteriological involvement of the joint is found and other causes of polyarthropathy are excluded. It has rarely been reported in this country, where its’ existence has seldom been accepted. It is not included in many rheumatology textbooks.
      • Issacs A.J
      • Sturrock R.D
      Poncet's disease—fact or fiction?.
      One case series examined 50 patients with acute or treated Tb. No evidence of arthropathy was found and reported that the links with arthritis were fortuitous.
      • Summers G.D
      • Jaylon M.I.V
      Does Poncet's disease exist?.
      Some authors argue that whilst organisms cannot be isolated from the affected joints, mycobacterial degradation products and overt mycobacterial DNA sequences are present in the synovium.
      • Toivanen P
      From reactive arthritis to rheumatoid arthritis.
      With this in mind, it is reportedly essential to include mycobacterial PCR analysis of synovial fluid and synovial biopsies before entertaining the diagnosis of PD.
      • Hameed K
      • Karim M
      • Islam N
      • Gibson T
      The diagnosis of Poncet's disease.
      Contrary to this statement, the finding of mycobacterial sequences in both inflamed and normal joints can be attributed to the trafficking of antigens from BCG vaccination.
      • Kempsell K.E
      • Cox C.J
      • McColm A.A
      • et al.
      Detection of mycobacterium tuberculosis group organisms in human and mouse joint tissue by reverse transcriptase PCR prevalence in diseased synovial tissue suggests lack of specific association with rheumatoid arthritis.
      The proposed mechanism to explain the pathogenesis of PD is that of molecular mimicry. Heat shock proteins from mycobacteria share amino acid sequences with cartilage proteoglycans, leading to a T-cell mediated cross reactivity and host cell damage. There may also be a genetic predisposition. Those who are HLA DR3 and/or HLA DR4, show T-cell hyper responsiveness to mycobacterial antigens.
      • Southwood T.R
      • Gaston J.S.H
      The molecular basis for Poncet's disease.
      This same mechanism is believed to be the cause of the skin manifestation erythema nodosum. These two conditions may represent expressions of a common immunogenic response to tuberculin.
      The treatment of PD is the same as for pulmonary Tb. The polyarthropathy responds rapidly to treatment with only one reported case of recurrence.
      • Gupta K.B
      • Prakash P
      Recurrent Poncet's disease- a rare presentation.
      In the case reported, the initial accumulation of joint symptoms is explained by non-compliance, and improved with DOT. Our case demonstrates the difficulty in diagnosing PD. The rheumatoid factor was elevated, raising the possibility of Rh.A. However, a raised rheumatoid factor is almost universal in Tb. On the basis of his symptoms; early morning stiffness, bilateral joint involvement, arthritis of >3 joints and rheumatoid factor positivity, he fulfilled the American rheumatism society criteria for Rh.A. This is unlikely here since the arthropathy is improving with regular anti-Tb drugs, and the synovial biopsy and radiology were not typical for Rh.A. It has previously been observed that Rh.A rarely occurs in patients with Tb.
      • Issacs A.J
      • Sturrock R.D
      Poncet's disease—fact or fiction?.
      Infection with mycobacteria and other bacteria have long been implicated in the aetiopathogenesis of Rh.A. Suggested mechanisms involve; direct joint invasion, molecular mimicry and reactive phenomena or immune dysfunction leading to autoimmunity. Advances in molecular techniques continue to cloud the association, since PCR analysis identifies mycobacterial antigens in rheumatoid and normal synovium and can identify multiple matched sequences between bacterium and host cells. Mycobacteria certainly lead to altered immune responses as demonstrated by BCG vaccination enhancing TH1-cell mediated responses and suppression of TH2-cell activity.
      • Choi I.S
      • Koh Y.I
      Therapeutic effects of BCG vaccination in adult asthmatic patients a randomized, controlled trial.
      Rigorous proof implicating mycobacteria in Rh.A pathogenesis is lacking and moreover, no effective treatment for the disease has come from work in this area.
      • Albert L.J
      Infection and rheumatoid arthritis guilty by association.

      Conclusion

      As the incidence of Tb in developed countries rises, more reports of PD are likely. Heightened clinician awareness may lead to more rapid identification of this complication in the interest of patient care.

      References

        • Dall L
        • Long L
        • Stanford J
        Poncet's disease.
        Rev Infect Dis. 1989; 11: 105-107
      1. Poncet A. De la polyarthrite tuberculeuse deformante ou pseudo-rheumatisme chronique tuberculeux. Congr Fr Chir 1897; 732.

        • Issacs A.J
        • Sturrock R.D
        Poncet's disease—fact or fiction?.
        Tubercle. 1974; 55: 135-142
        • Summers G.D
        • Jaylon M.I.V
        Does Poncet's disease exist?.
        Rheumatol Rehabil. 1980; 19: 149-150
        • Toivanen P
        From reactive arthritis to rheumatoid arthritis.
        J Autoimmun. 2001; 16: 369-371
        • Hameed K
        • Karim M
        • Islam N
        • Gibson T
        The diagnosis of Poncet's disease.
        Br J Rheumatol. 1993; 32: 824-826
        • Kempsell K.E
        • Cox C.J
        • McColm A.A
        • et al.
        Detection of mycobacterium tuberculosis group organisms in human and mouse joint tissue by reverse transcriptase PCR.
        Infect Immun. 2001; 69: 1821-1831
        • Southwood T.R
        • Gaston J.S.H
        The molecular basis for Poncet's disease.
        Br J Rheumatol. 1990; 29: 491
        • Gupta K.B
        • Prakash P
        Recurrent Poncet's disease- a rare presentation.
        Indian J Tuber. 2001; 48: 31-33
        • Choi I.S
        • Koh Y.I
        Therapeutic effects of BCG vaccination in adult asthmatic patients.
        Ann Allergy Asthma Immunol. 2002; 88: 584-591
        • Albert L.J
        Infection and rheumatoid arthritis.
        J Rheumatol. 2000; 27: 564-566