Propylthiouracil Induced ANCA Positive Vasculitis: a Case Report

This Article


Article Information:

Group: 2004
Subgroup: Volume 2, Issue 1, Winter
Date: March 2004
Type: Case Repport
Start Page: 47
End Page: 50


  • Khanolkar MP
  • Department of Medicine, University of Wales College of Medicine, Wales, UK
  • Owen PJD
  • Department of Medicine, University of Wales College of Medicine, Wales, UK
  • Lazarus JH
  • Department of Medicine, University of Wales College of Medicine, Wales, UK


      Affiliation: Department of Medicine, University of Wales College of Medicine
      City, Province: Wales,
      Country: UK


Anti-thyroid drugs, methimazole, carbimazole and propylthiouracil are commonly prescribed for the treatment of hyperthyroidism due to Graves' disease_ We present here the report of a Caucasian patient with Graves' disease who developed ANCA positive propylthiouracil induced vasculitis_ The episode was characterised by a vasculitic skin rash associated with polyarthralgia but no renal involvement_ While carbimazole/ methimazole has a lower incidence of reported ANCA positive side effects than PTU, the use of the latter in pregnancy may still be considered as reasonable because the immunosuppression of pregnancy may prevent the appearance of immune problems.

Keywords: Propylthiouracil;Carbimazole;Vasculitis;ANCA;Hyperthyroidism;Graves' disease

Manuscript Body:


Antithyroid drug treatment with thionamides such as carbimazole and propylthiourac il (PTU) is commonly used in the treatment of Graves' disease. I Anti neutrophil cytoplasmic antibody (ANCA) was first detected around 20 years ago in patients with segmental necrotizing glomerulonephritis? Shortly afterwards, ANCA was found to be closely associated with vasculitic conditions and since then has been increasingly used as
a diagnostic marker of vasculitis.3 PTU has been known to promote formation of ANCA,4 although only a very few of these patients with ANCA positivity go on to develop clinical vasculitis.5 To our knowledge, 59 cases of ANCA positive vasculitis related to antithyroid medication have been reported in English literature with the first case being reported in 1992;6 almost half of these are from Japan. We report a case of PTU induced vasculitis in a Caucasian lady with Graves' disease which has occurred followin g a prior similar clinical event while on a course of carbimazole.

Case Report

A 26-year-old Caucas ian lady was admitted to the hospital with a two-week history of migrating polyarthralgia and a rash around both her ankles. The patient was diagnosed with Graves ' hyperthyroidism 6 years before, when she was commenced on carbimazole 10 mg thrice daily. Carbimazole was stopped five weeks after its commencement as she developed polyarthralgia and rash on this medication. No immunological studies were performed at that time. Polyarthralgia and rash disappeared within a week on discontinuing carbimazole. PTU was started instead at a dose of 50 mg twice daily which she took for a year without any adverse effects. Subsequently her Graves' disease was in remission for 3 years. PTU was recommenced 2 years ago at 50 mg twice daily when her Graves' disease relapsed and she had been on this dose until her present admission with polyarthralgia and rash. There was no other significant past medical history or exposure to any other drugs.
On examination, she appeared to be clinically euthyroid and had a diffuse goiter. There was a non-tender, non-blanching, nonpruritic, maculopapular rash around both her ankles. There was no clinical evidence of arthritis although she complained of a migratory polyarthralgia affecting her ankle, knee and shoulder joints.
Investigations revealed Hb 10.8 g/dL (normal range: 12-16 g/dl), WBC 5.5x109/L (normal range: 4-11x109/L), neutrophils 3.8 xl09/L (normal range: 1.2-7 .6x109/L), eosi nophils 0.lx109/L (normal rangd: 0.1- 1.25x109/L), platelets 273x109/L (normal range: l40-350x 109/L), urea 3.4 mmollL, creatinine 61 p.mollL, normal liver function tests, chestl Ankle Xrays and urinalysis. Thyroid function tests showed a normal free T4 level of 22.3 pmol/L (normal range: 9.8- 23.1), nom1al free T3 level of 5.4 pmollL (normal range: 3.5-6.5) and a TSH of 0.33 mUlL (nom1al range: 0.35 -5.50). Inflammatory markers revealed a C-reactive protein of 56 mg/L (normal<11) and an ESR of 61 mmlhr. TSH receptor antibodies were positive consistent with a diagnosis of Graves' disease. Antinuclear antibodies were weakly positive at a titer of 1: 100; however antidsDNA antibodies, ENA screen and rheumatoid factor were all negative. ANCA titres were strongly positive (> 1 :320; a titer of 1:80 is considered positive) and of a perinuclear type.  ELISA assays were positive for MPO (myeloperoxidase) antibodies.
Based on the above findings, the patient was diagnosed to have PTU induced ANCA positive vasculitis. PTU was discontinued immediately which was followed by a prompt relief from symptoms of polyarthralgia and fading of the rash within a week.
The patient attended for follow-up after two months when she was well and her symptoms of polyarthralgia and rash had disappeared completely. Clinically she continued to remain euthyroid off antithyroid medications. This was confirmed biochemically (free T4=20.9 pmol/L, free T3=5.9 pmollL and TSH=0.30 mUlL). Her inflammatory markers had normalized, however the ANCA titers continued to remain positive (>1:320). TSH receptor antibodies were now undetectable. Further treatment options with radioactive iodine therapy were discussed in the event of her becoming hyperthyroid again in the future.


This case report is an attempt to highlight the importance of testing for ANCA in patients with suspicious symptoms whilst on antithyroid medication. The case is unusual in that the patient probably developed the disease in response to carbimazole as well as PTU and it is said that there is normally a low chance of side effect cross reactivity between the two drugs, although this is not unknown. he patient had symptoms of polyarthralgia and rash developing much earlier (within 5 weeks) when commenced on carbimazole, which prompted to switch her to PTU. Having been on PTU (50 mg twice daily) for around 2 years, the patient had a recurrence of her symptoms of polyarthralgia and rash. On both occasions, her symptoms settled promptly on discontinuation of the medication. Diagnosis of ANCA positive vasculitis was strongly suspected on clinical grounds supported by ANCA titers and by clinical resolution on stopping the medication. Previous studies have shown ANCA positivity to be significantly associated with the d uration of therapy with PTU, but there has been no such association with duration of carbimazole therapy.7
Recent studies have shown high frequency of ANCA positivity in patients with Graves'disease treated with antithyroid medications, especially with PTU where the prevalence is estimated to be as high as 25% as against carbimazole/methimazole, which have a much lower prevalence of 3.4%.8 Most cases of ANCA positivity have occurred in patients on long term (>18 months) therapy with antithyroid medications or in those with a recent commencement. 7 However only a small percentage of these patients go on to develop  features of vasculitis. 7-9 A Chinese study by Guo et al. 11 has reported ANCA positivity of 5.9% in untreated Graves'disease as against 22.6% in those treated with PTU and 6.5% in those treated with methimazole. However none of the untreated patients had clinical signs and symptoms of vasculitis. ANCA vasculitis associated with antithyroid medication occurs more frequently in women, which may simply be accounted for by the female predilection for Graves' disease. The underlying thyroid disease is usually Graves' disease, although cases have also been reported in association with toxic multinodular goiters. IO The pathogenesis behind PTU inducing ANCA is not clearly understood, however PTU has been shown to accumulate within neutrophils, binding to myeloperoxidase to alter its config.uration, which subsequently promotes antibody formation by polyclonal activation of B lymphocytes in susceptible individuals.4,11 The commonest skin lesion is leukocytoclastic vasculitis, which typically causes purpuric lesions in lower limbs. 12 In this case there was no skin biopsy taken and no overt renal involvement. The spectrum of vasculitis may range from a milder form with arthralgia and rash to a more severe form with renal or pulmonary involvement. We recommend biopsy confirmation of vasculitis (skin biopsy wherever possible or renal biopsy if indicated).
Mild forms of vasculitis respond well to discontinuation of the responsible medication but severe forms, especially with renal or pulmonary involvement may need more aggressive treatment with steroids, cyclophosphamide and in rare cases, plasmapheresis. 12Once the vasculitic process is controlled, it is also important to treat the underlying hyperthyroidism which may be achieved by considering destructive treatment with radioactive iodine or using other antithyroid medications like lithium or ipodate. Some patients may also be offered treatment with thyroidectomy if appropriate. Fortunately, this patient  remained euthyroid off antithyroid medication at follow up.
It appears reasonable to select carbimazole rather than PTU as initial antithyroid therapy as the prevalence of ANCA positivity is much lower. The issue of monitoring ANCA in patients on antithyroid medication needs to be addressed although it has been suggested that ANCA should be tested in patients on long term treatment (>18 months) or in those developing suspicious vasculitic features on these medications. 7 The significance of ANCA positivity in asymptomatic patients on these medications remains unclear, but this should lead to earlier consideration of definitive treatment with radioiodine or surgery. In our case, ANCA titers continued to remain positive at follow-up after stopping PTU despite normalization of inflammatory markers. lthough ANCA titers are expected to fall with time, some patients do continue to remain ANCA positive for many years as noted by Gunton et al in his review of 32 cases of thionamide associated ANCA positive vasculitis. 7 The significance of this persistent ANCA positivity rema ins unclear. It should be remembered that vasculitis may also be triggered by exposure to other drugs (eg. Hydralazine and sulfonamides. 13 and some viral infections (eg. hepatitis CI4 ).
Patients developing ANCA vasculitis on PTU may be offered treahnent with carbimazole for treating underlying hyperthyroidism; however prescribing PTU to patients with carbimazole induced ANCA vasculitis seems inappropriate given the fact that PTU is associated with a very high prevalence of ANCA positivity (25%). PTU is the preferred antithyroid medication during pregnancy but no significant vasculitic cases have been re- ported in this group of patients which may simply be explained by the shorter duration antiof PTU course and the immunosuppression associated with pregnancy.

References: (14)

  1. Cooper DS. Antithyroid drugs. N Engl J Med. 1984 Nov 22;3 II (21): 1353-62.
  2. Davies DJ, Moran JE, Niall JF, Ryan GB. Segmental necrotising glomerulonephritis with antineutrophil antibody: possible arb'ovirus aetiology? Br Med 1 (Clin Res Ed). 1982 Aug 28;285(6342):606.
  3. Hall 18, Wadham BM, Wood Cl, Ashton V, Adam WR. Vasculitis and glomerulonephritis: a subgroup with an antineutrophil cytoplasmic antibody. Aust N Z J Med. 1984 Jun; 14(3):277-8.
  4. Lam DC, Lindsay RH. Accumulation of 2- [14C]propylthiouracil in human polymorphonuclear leukocytes. Biochem Pharmacol. 1979 Aug I ;28( 15):2289-96.
  5. Noh JY, Asari T, Hamada N, Makino F, Ishikawa N, Abe Y, Ito K, Ito K. Frequency of appearance of myeloperoxidase-antineutrophil cytoplasmic antibody (MPO-ANCA) in Graves' disease patients treated with propylthiouracil and the relationship between MPO-ANCA and clinical manifestations. Clin Endocrinol (Oxf). 200 I May;54(5):65 1-4.
  6. Stankus SJ, Johnson NT. Propylthiouracil-induced hypersen sitivity vasculitis presenting as respiratory failure. Chest. 1992 Nov;102(5):1595-6.
  7. Gunton JE, Stiel l, Clifton-Bligh P, Wilmshurst E, McElduff A. Prevalence of positive anti-neutrophil cytoplasmic antibody (ANCA) in patients receiving anti-thyroid medication. Eur J Endocrinol. 2000 lun; 142(6):587.
  8. Wada N, Mukai M, Kohno M, Notoya A, Ito T, Yoshioka N. Prevalence of serum anti-myeloperoxidase antineutrophil cytoplasmic antibodies (MPO-ANCA) in patients with Graves' disease treated with propylthiouracil and thiamazole. Endocr J. 2002 Jun;49(3):329-34.
  9. Dolman KM, Gans RO, Vervaat TJ, Zevenbergen G, Maingay D, Nikkels RE, Donker AJ, von dem Borne AE, Goldschmeding R. Vasculitis and antineutrophil cytoplasmic autoantibodies associated with propylthiouracil therapy. Lancet. 1993 Sep II ;342(8872):651-2.
  10. Gunton lE, Stiel J, Caterson Rl, McElduff A. Clinical case seminar: Anti-thyroid drugs and antineutrophil cytoplasmic antibody positive vasculitis. A case report and review of the literature. J Clin Endocrinol Metab. 1999 Jan;84(1):13-6.
  11. Guo XH, Zhao MH, Gao Y, Wang SF, Gao Y. [Antineutrophil cytoplasmic antibody associated vasculitis induced by antithyroid agents] Zhonghua Yi Xue Za Zhi. 2003 lun 10;83(11):932-5. Chinese.
  12. Jennette lC, Falk RJ. Small-vessel vasculitis. N Engl J Med. 1997 Nov 20;337(2 I): I 512-23.
  13. Choi HK, Merkel PA, Walker AM, Niles lL. Drugassociated antineutrophil cytoplasmic antibodypositive vasculitis: prevalence among patients with high titers of antimyeloperoxidase antibodies. Arthritis Rheum. 2000 Feb;43(2):405-13.
  14. Kallenberg CG, Heeringa P. Pathogenesis of vasculitis. Lupus. 1998;7(4):280-4