Diagnostic approach to a patient with atypical Henoch-Schönlein purpura

We read with great interest the recent contribution by Uza et al. [1]. They reported an 81-year-old man who presented with hematochezia and hemorrhagic ascites, but with no abdominal pain or skin rash. Initial duodenoscopy and colonoscopy demonstrated multiple ulcers and nonspecific inflammation. Palpable purpura developed 8 days later, and hematochezia and renal function rapidly worsened and did not respond to methylprednisolone pulse therapy (MPT), leading to death. IgA depositions on the small vessels of the duodenum and ileum were demonstrated on the autopsy specimens. The authors regarded this patient as having atypical Henoch-Schönlein purpura (HSP) due to old age at onset, delayed appearance of characteristic purpura, an absence of abdominal pain, and the presence of hemorrhagic ascites; they wondered why the jejunum could be spared despite systemic vasculitis.

To increase the diagnostic accuracy, we should focus on the definition of HSP. Recently, the new classification criteria for HSP was modified by the EULAR/PreS (European League Against Rheumatism/Paediatric Rheumatology European Society) [2], which deleted the age at onset (because HSP also occurs in adults), included predominant IgA deposition (because this is a main pathogenesis of HSP), and added arthritis and renal involvement (because these are common presentations of HSP) compared with the ACR (American College of Rheumatology) criteria [3]. Nevertheless, it is very difficult to suspect HSP at the early course of the disease when a patient presents with isolated gastrointestinal manifestations, especially without abdominal pain or bowel angina as in this case, and which might lead to delayed diagnosis and unexpected fatal outcome. Therefore, some authors have suggested that the measurement of factor XIII activity might be useful when the characteristic rash of HSP is absent [4] [5], although it is not a diagnostic criterion of HSP. If the immunofluorescence of IgA had been examined at the initial endoscopic biopsies in this patient, the diagnosis of HSP would have been confirmed without skin rash according to the new EULAR/PreS criteria [2], even though the initial endoscopic biopsy showed nonspecific inflammation.

The locations of gastrointestinal involvement in adults with HSP are variable. Mahamedi et al. [6] reported six adult patients with HSP, in whom the lesions were located in the duodenum (n = 2), the jejunum (n = 2), and the ileum (n = 5) on abdominopelvic computed tomography. Two segments were involved in one patient and three segments in another patient. Therefore, the jejunum is not always involved in HSP.

We should also consider the possibility of polyangitis overlap syndrome in this case, because Watanabe et al. [7] reported a fatal case (56-year-old man) of systemic vasculitis combined with HSP and polyarteritis nodosa, the clinical course of which was very similar to Uza’s case. He also died of progressive renal failure 1 month after the onset of the disease. He complained of arthralgia, purpura of both lower extremities, nasal bleeding, and tarry stool, and acute renal failure was noted at the time of admission to hospital. A skin biopsy from the purpura lesion exhibited leucocytoclastic vasculitis with IgA deposition, and HSP was considered. However, renal failure progressed rapidly and subsequently was complicated by acute myocardial infarction. Postmortem examination revealed polyarteritis nodosa-type necrotizing vasculitis in the kidneys, heart, and mesentery resulting in acute multiple infarctions of these organs. Therefore, they described that it would be important to suspect the polyangitis overlap syndrome positively when progressive acute renal failure is seen in a patient with HSP, and to begin appropriate therapy immediately [7]. We suggest that intensive treatment such as cyclophosphamide or plasmapheresis in addition to MPT should be initiated as early as possible in a patient with atypical course of vasculitis, considering the fatal course of polyarteritis nodosa in contrast to the benign nature of HSP [8].

In conclusion, when a patient shows atypical presentation of HSP without skin rash, the measurement of factor XIII activity should be performed at an early stage, and polyangitis overlap syndrome should also be suspected to avoid a fatal outcome.

References

• 1 Uza N, Yazumi S, Tanabe K. et al . Henoch-Schonlein purpura in an elderly patient with unusual manifestation.  Endoscopy . 2007 Feb 7;  [Epub ahead of print]
• 2 Ozen S, Ruperto N, Dillon M J. et al . EULAR/PReS endorsed consensus criteria for the classification of childhood vasculitides.  Ann Rheum Dis. 2006;  65 936-941
• 3 Mills J A, Michel B A, Bloch D A. et al . The American College of Rheumatology 1990 criteria for the classification of Henoch-Schonlein purpura.  Arthritis Rheum. 1990;  33 1114-1121
• 4 Kaneko K, Fujii S, Shono T. et al . Diagnostic value of plasma factor XIII in Henoch-Schonlein purpura.  Pediatr Nephrol. 2004;  19 702-703
• 5 Gunasekaran T S, Berman J, Gonzalez M. Duodenojejunitis: is it idiopathic or is it Henoch-Schonlein purpura without the purpura?.  J Pediatr Gastroenterol Nutr. 2000;  30 22-28
• 6 Mahamedi H, Andre M, Privat C. et al . Value of abdominal-pelvic computed tomography in adult rheumatoid purpura.  Rev Med Interne. 2001;  22 132-140
• 7 Watanabe K, Abe H, Mishima T. et al . Polyangitis overlap syndrome: a fatal case combined with adult Henoch-Schonlein purpura and polyarteritis nodosa.  Pathol Int. 2003;  53 569-573
• 8 Colmegna I, Maldonado-Cocco J A. Polyarteritis nodosa revisited.  Curr Rheumatol Rep. 2005;  7 288-296