ARTICLE IN PRESS

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/cupe

Current Paediatrics (2006) 16, 259–263

0957-5839/$ - see frodoi:10.1016/j.cupe.

�Corresponding au

E-mail address: s

Henoch–Schonlein purpura

S.E. Tarvin, Susan Ballinger�

Department of Pediatric Rheumatology, Indiana University School of Medicine, Indianapolis, IN 46202-5114, USA

KEYWORDSHenoch–Schonleinpurpura;Vasculitis;Nephritis

nt matter & 20062006.05.003

thor.

balling@iupui.edu

SummaryThis review summarises the evidence from the latest published research on theepidemiology, aetiology, pathophysiology, clinical manifestations, treatment, and prog-nosis of Henoch–Schonlein purpura (HSP). Analysis of the literature indicates theimportance of genetic and infectious aetiologic considerations in the development ofHSP. And, within the last year, multiple inflammatory markers have been studied inassociation with the disease. Although, the common complaints associated with HSP arewell known, the disease can also be associated with sequalae in multiple organ systems, aswell as vasculitis throughout the body. No consensus has yet been agreed upon regardingtreatment methodology of disease complications, but prognostic studies have determinedthat 6 months is appropriate in which to follow-up children to monitor for renalcomplications. Recent literature indicates the continued interest in defining the aetiology,complete clinical manifestations, treatment options and prognostic markers of HSP and thecomplications from the disease.& 2006 Elsevier Ltd. All rights reserved.

Practice points

� Randomised control trials of treatments of HSP withsteroids show no influence on long-term outcome� Most cases of HSP are self-limiting. Therefore the

unusual and unusually severe cases are written upand published� Monitoring for renal complications in patients with

HSP should last at least 6 months

Elsevier Ltd. All rights reserved.

(S. Ballinger).

Introduction

Henoch–Schonlein purpura (HSP) was first recognised byHeberden in 1801 and first described as an associationbetween purpura and arthritis by Schonlein in 1837. Henochadded descriptions of gastrointestinal (GI) involvement in1874 and renal involvement in 1899. HSP is a small vesselvasculitis the major manifestations of which includearthritis, non-thrombocytopenic purpura, abdominal pain,and renal disease. In 1990, the American College ofRheumatology published diagnostic criteria for HSP. Theseincluded: (1) palpable purpura—slightly raised ‘palpable’haemorrhagic skin lesions, not related to thrombocytopenia;(2) age 20 years or younger at onset of first symptoms of thedisease; (3) bowel angina—diffuse abdominal pain, worseafter meals, or the diagnosis of bowel ischaemia, usuallyincluding bloody diarrhea; and (4) wall granulocytes onbiopsy—histologic changes showing granulocytes in the wallsof arterioles or venules. The classification further states:

ARTICLE IN PRESS

S.E. Tarvin, S. Ballinger260

For purposes of classification, a patient shall be said tohave HSP if at least two of these four criteria are present.The presence of any two or more criteria yields asensitivity of 87.1% and a specificity of 87.7%.1

HSP is one of the most common vasculitidies of childhoodand is considered to be self-limiting. One manifestation ofHSP that can continue to cause lifelong problems is renalinvolvement.2

Epidemiology, aetiology and pathophysiology

Dolezalova et al. conducted a prospective epidemiologicincidence survey of childhood vasculitic disease in the CzechRepublic showing an incidence of 10.2 per 100 000,3 whereasacross the globe, a Taiwanese survey of childhood HSPepidemiological characteristics yielded an annual incidenceof 12.9 per 100 000 children under the age of 17 years.4

Although the aetiology of HSP remains unclear, there arecase reports of associations that are worth reviewing. In thelast year, HSP has been reported in adults in association withoesophageal and lung adenocarcinoma,5 pulmonary Myco-bacterium avium-intracellulare complex,6 familial Mediter-ranean fever,7 coumadin exposure,8 and intravesicaladministration of bacillus Calmette Guerrin (BCG) for themanagement of bladder cancer.9 In children, HSP has beenreported in association with protein-losing enteropathy andsevere esophagitis,10 varicella,11 and invasive meningococ-cal disease.12 In all cases, HSP was nonfatal and prolongedrenal involvement was rare. Most cases of disease-asso-ciated HSP presented after the primary diagnosis was made.

A number of studies evaluated the relationship between HSPand inflammatory markers, the studies document involvementof anti-neutrophil antibodies (ANCAs) both P and C, the rareinterleukin (IL)-1b (-511) T allele, transforming growth factor-b, the C4 null allele, increased production of IL-8 but noincrease in expression of ICAM-1, tumour necrosis factor (TNF)-a, and the lectin pathway complement activation in theprogression of renal disease in HSP.15–24 There are conflictingarticles about the role of nitric oxide synthase polymorphismsin susceptibility of developing HSP.20,21

Clinical manifestations

Characteristic clinical manifestations associated with HSPinclude GI complaints, non-thrombocytopenic rash, arthri-tis, and nephritis. Multiple case reports and research articlesover the past year have expounded upon these character-istic symptoms to discuss the extent of symptomology andreport involvement of other organ systems.

Two studies investigated the GI manifestations of HSP.Chang et al.25 used a retrospective analysis of 261 patientswith the goal of assessing the diagnostic value of imagingand stool occult blood tests to identify fatal complications.This study found that 58% of patients with HSP hadabdominal pain, 17.6% suffered overt GI bleed or heme-positive stools. A 3+ or greater stool occult blood had a highincidence of being associated with positive imaging findings,and abdominal ultrasound should be used to excludeintussusception or bowel perforation.25 Chen and Kong26

also used the method of retrospective chart review, but this

study noted the incidence of GI complications and manifes-tations. Of the 162 patients, 98.1% had colicky abdominalpain (the most frequent symptom), followed by vomiting in39.5% of patients. Interestingly, in 25.3% of reported cases,GI symptoms manifested before skin rash.26 Other GImanifestations reported include terminal ileum involvementas the presenting symptom of a 43-year-old male with HSP27;extensive mucosal inflammation with duodenojejunoileitisas pictured by capsule endoscopy in a patient with activeHSP28; and multiple small bowel perforations (0.38%incidence) in a child who died from complications ofsepticaemia secondary to perforation.29 The illustration ofthese reports highlights the importance of observing a widedifferential when discussing the aetiology of abdominalcomplaints.

Nadrous et al. and Gonzalez-Gay et al. both studiedpulmonary pathology in association with HSP but with verydifferent parameters. The former study documented thatpulmonary pathology among a 124-patient adult cohort withHSP is very rare (2.4%), and when present was associatedwith diffuse alveolar haemorrhage and interstitial pneumo-nia.30 The latter study assessed whether children with HSPand a history of upper respiratory tract infection (URI) had adifference in clinical course. Seventy-eight children werestudied, and although children who had a history of URIbefore onset had a higher incidence of renal manifestations,specifically nephritis, there was no difference in long-termrenal complications observed.31

A Swedish study reported evidence pertaining to thegreatest cause of morbidity associated with HSP, renaldisease. The group evaluated renal hemodynamics bymeasuring the glomerular filtration rate (GFR) and proteinexcretion rates in 73 children with HSP nephritis. The studyconcluded that patients with HSP nephritis had lower GFRsthan controls, but that severe morphological changes, alsodemarcated by lower GFRs, and higher blood pressure,occurred in patients with both nephritic range, as well asmild proteinuria. From this data the authors concluded thatproteinuria is a marker of renal damage in HSP nephritis,and that all patients with proteinuria should be monitoredclosely. Kidney biopsy should be readily performed to assessthe extent of morphological damage and initiate treatmentwhen proteinuria increases and/or persists.32

Other clinical manifestations reported over the past yearinclude two reports of children who developed rarehaemorrhagic bullae in association with HSP33; a report ofa 13-year-old male with proximal symmetrical muscleweakness and pain in association with HSP nephritis34; anda report of a 25-week pregnant woman who developed HSPand was successfully treated with steroid therapy, resultingin disease resolution and a healthy, term infant.35 Alsoreported was the case of a 13-year-old girl who developedcerebral vasculitis and intracerebral haemorrhage, asso-ciated with vision loss, whose sight was successfully restoredfollowing plasmapheresis.36

Treatment

HSP, in general, is considered a self-limiting disease.However, in cases of prolonged GI pain, steroids have beenreported to decrease the length of GI complaints, as well as

ARTICLE IN PRESS

Henoch–Schonlein purpura 261

decrease renal complications. A study published in an April2004 edition of BMC Medicine documents a randomised,controlled trial (RCT) of prednisone as an early interventionin those patients diagnosed with HSP. The 41-patient studyrandomised children within 7 days of symptom onset toprednisone therapy or placebo for 2 weeks. Unlike previouscase reports, the study found that there was no difference inthe rate of renal involvement and/or the rate of acute GIcomplications between the placebo and steroid groupsthrough 1-year of follow-up.37

As no single treatment methodology has been proven tobe effective for refractory symptoms associated with HSP,multiple immunosuppressive therapies have been tried in aneffort to decrease complicating symptoms. Sugiyama et al.report a case of an adult patient with HSP nephritis who wassuccessfully treated to clinical remission with tonsillectomyfollowed by intravenous pulse methylprednisolone and oralprednisone therapy.38 In the last year, cyclosporine A hasbeen reported to be an effective treatment of HSP nephritis,with nephritic-range proteinura, in a child refractory tosteroid therapy.39 Dapsone has been reported, in a smallcase series, to provide improvement in skin rash butprovides no known beneficial outcome regarding renaldisease.40 Also in an adult refractory to dapsone, oralsteroids, azathiaprine, IVIG, methotrexate and cyclopho-sphamide, cyclosporin was reported to ameliorate heromnipresent abdominal and joint pains.41

A RCT of cyclophosphamide therapy for 56 children withhistopathologically severe HSP nephritis found no differencesin onset data or outcome between the trial and placebogroups.42 However, Kawasaki et al. reported that cyclopho-sphamide, when combined with methylprednisolone andurokinase pulse therapy, is beneficial for patients with severeHSP nephritis. After 6 months of therapy, the patients treatedwith three-drug therapy had decreased urinary proteinexcretion, lower chronicity index of serial kidney biopsiesand no progression to persistent nephropathy.43

Plasmapheresis has also been reported to be effective forpatients with nephritis. A Taiwanese case report discussesthe case of a 33-year-old man with crescentic glomerulone-phritis refractory to steroid and oral cyclophosphamidetherapy. The patient had successful recovery after ninesessions of simple double-filtration plasmapheresis andremained free of vasculitic events at 18 months follow-up.44 In a case series of six Japanese children with rapidlyprogressive HSP nephritis, the authors report treatmentwith five courses of plasmapheresis followed by multipledrug therapy, including methylprednisolone and urokinasepulse therapy, oral prednisolone, cyclophosphamide, dipyr-idamile and warfarin. After 6 months, each patient hadsignificantly reduced urinary protein excretion, no increasein crescentic and sclerosed glomeruli, and no patientprogressed to renal insufficiency.45

The most original treatment study reported a cohort offive children with biopsy-proven HSP and repeated episodesof haematuria and proteinuria treated with fish oil andangiotensin-converting enzyme inhibitor (ACEI) therapy forhypertension. After an average of 49.2 weeks of follow-up,the protein excretion rate and average blood pressure hadboth significantly decreased with no study participantrequiring ACEI therapy. Also, the GFR and serum creatinineof all study patients remained stable.46

It is evident from these treatment studies that a clearconsensus regarding therapy has yet to be determined,however, many different treatment strategies can be usedto aid patients with complications from HSP.

Prognosis

In general, the prognosis for most children diagnosed withHSP is very good. As HSP is considered a self-limiting diseasewith a low percentage of complications, most childrenrecover without permanent sequelae. The most significantcause of morbidity associated with HSP is renal insufficiency.Two studies have discussed prognostic factors relating to HSPand renal involvement.

Rigante et al. prospectively examined a cohort ofpaediatric patients diagnosed with HSP to look at thepossible relationship between renal involvement or diseaserelapse with regard to multivariant analysis. Those patientswho were treated with steroid therapy or who had renalcomplications at the onset of disease were excluded fromthe analysis. The study revealed that persistent rash,greater than 1 month, was significantly related to renalinvolvement.47 A second prognostic meta-analysis wasundertaken with the goal of determining the duration ofrecommended follow-up for children diagnosed with HSPwithout significant renal complications at the time ofdiagnosis. The study included 12 studies with a total cohortof 1133 children. The results indicated that children shouldbe followed-up using urinalysis for at least 6 months because97% of children who will have abnormal urine findings willpresent in that time period. Encouragingly, children withnormal urinalysis at presentation were found to have nolong-term renal impairment. However the risk of long-termrenal impairment is 12 times higher if the patient presentswith nephritic or nephritic syndrome. Those children thathave abnormal urinalysis at presentation or developabnormal urinalysis within the first 6 months shouldbe followed-up by measurements of serum urea andcreatinine.48

Conclusion

Although HSP has been described for well over a century, weare still struggling to determine the aetiology of this mostcommon vasculitis of childhood, and the best treatment forthe most severe outcomes. Most children with HSP have nosignificant sequelae, but renal involvement can result inlifelong problems. Through review of the most recent year’sliterature, the continued interest in this disease, itsaetiology, and its outcome can be appreciated.

References and recommended reading

Papers of particular interest, published withinthe annual period of review, have been highlightedas: *Of special interest; **Of outstanding interest.

1. Mills JA, Michel BA, Bloch DA, et al. The American Collegeof Rheumatology 1990 criteria for the classification ofHenoch–Schonlein purpura. Arthritis Rheum 1990;33:1114–21.

ARTICLE IN PRESS

S.E. Tarvin, S. Ballinger262

2. Saulsbury FT. Epidemiology of Henoch–Schonlein purpura. CleveClin J Med 2002;69(Suppl. 2):S1187–9.

3. Dolezalova P, Telekesova P, Nemcova D, et al. Incidenceof vasculitis in children in the Czech Republic: 2-yearprospective epidemiology survey. J Rheumatol 2004;31:2295–9.

4. Yang YH, Hung CF, Hsu CR, et al. A nationwide survey ofepidemiological characteristics of childhood Henoch–Schonleinpurpura in Taiwan. Rheumatology 2005;44:618–22.

5. Weiler-Bisig D, Ettin G, Brink T, et al. Henoch–Schonlein purpuraassociated with esophagus carcinoma and adenocarcinoma ofthe lung. Clin Nephrol 2005;63:302–4.

6. Yano, Shuichi. Henoch–Schonlein purpura associated withpulmonary Mycobacterium avium-intracellulare complex. In-tern Med 2004;43:843–5.

7. Lange-Sperandio B, Mohring K, Gutzler F, et al. Variableexpression of vasculitis in siblings with familial Mediterraneanfever. Pediatr Nephrol 2004;19:539–43.

8. Borras-Blasco J, Girona E, Navarro-Ruiz A, et al. Acenocoumar-ol-induced Henoch–Schonlein purpura. Ann Pharmacother2004;38:261–4.

9. Nan DN, Fernandez-Ayala M, Garcia-Ibarbia C, et al.Henoch–Schonlein purpura after intravesical administrationof bacillus Calmette–Guerin. Scand J Infect Dis 2005;37:613–5.

10. Dalgic B, Aktas A, Poyraz A, et al. Severe esophagitis in a childwith Henoch–Schonlein purpura presenting as protein-losingenteropathy. Ind J Gastroenterol 2005;24:80–1.

11. Kalman S, Aydin HI, Atay A. Henoch–Schonlein purpura in a childfollowing varicella. J Trop Pediatr 2005;51:240–1.

12. Tsolia MN, Fretzayas A, Georgouli H, et al. Invasive meningo-coccal disease presenting as Henoch–Schonlein purpura. Eur JClin Microbiol Infect Dis 2004;23:776–9.

*Case report of familial association of HSP presentingin the same environment.

15. Ferraz-Amaro I, Herrero MJ, Jurado A, et al. Dual positivity forcytoplasmic and perinuclear antineutrophil antibodies in apatient with Henoch–Schonlein purpura. Clin Exp Rheumatol2004;22:233–4.

16. Ozaltin F, Bakkaloglu A, Ozen S, et al. The significance of IgAclass of antineutrophil cytoplasmis antibodies (ANCA) in child-hood Henoch–Schonlein purpura. Clin Rheumatol 2004;23:426–9.

17. Amoli MM, Calvino MC, Garcia-Porrua C, et al. Interleukin 1bgene polymorphism association with severe renal manifesta-tions and renal sequelae in Henoch–Schonlein purpura.J Rheumatol 2004;31:295–8.

18. Yang YH, Lai HJ, Kao CK, et al. The association betweentransforming growth factor-b gene promoter C-509T poly-morphism and Chinese children with Henoch–Schonlein pur-pura. Pediatr Nephrol 2004;19:972–5.

19. Stefansson Thors V, Kolka R, Sigurdardottir SL, et al. Increasedfrequency of C4B*Q0 alleles in patients with Henoch–Schonleinpurpura. Scand J Immunol 2005;61:274–8.

20. Amoli MM, Garcia-Porrua C, Calvino MC, et al. Lack ofassociation between endothelial nitric oxide synthase poly-morphisms and Henoch–Schonlein purpura. J Rheumatol 2004;31:299–301.

21. Martin J, Paco L, Ruiz MP, et al. Inducible nitric oxide synthasepolymorphism is associated with susceptibility to He-noch–Schonlein purpura in Northwestern Spain. J Rheumatol2005;32:1081–5.

22. Yang YH, Lai HJ, Huang CM, et al. Sera from children with activeHenoch–Schonlein purpura can enhance the production of

interleukin 8 by human umbilical venous endothelial cells.Ann Rheum Dis 2004;63:1511–3.

*Sera from children with HSP induce expression of IL-8but not ICAM-1.

23. Ha Tae-Sun. The role of tumor necrosis factor-a inHenoch–Schonlein purpura. Pediatr Nephrol 2005;20:149–53.

24. Hisano S, Matsushita M, Fujita T, et al. Activation of the lectincomplement pathway in Henoch–Schonlein purpura nephritis.Am J Kidney Dis 2005;45:295–302.

25. Chang WL, Yang YH, Lin YT, et al. Gastrointestinal manifesta-tions in Henoch–Schonlein purpura: a review of 261 patients.Acta Paediatr 2004;93:1427–31.

26. Chen SY, Kong MS. Gastrointestinal manifestations and compli-cations of Henoch–Schonlein purpura. Chang Gung Med2004;27:175–81.

27. Karagozian R, Turbide C, Szilagyi A. Henoch–Schonlein purpurapresenting with ileal involvement in an adult. Dig Dis Sci2004;49:1722–6.

28. Skogestad E. Capsule endoscopy in Henoch–Schonlein purpura.Endoscopy 2005;37:189.

29. Vigiter M, Bosnali O, Sekmenli T, et al. Multiple and recurrentintestinal perforations: an unusual complication ofHenoch–Schonlein purpura. Eur J Pediatr Surg 2005;15:125–7.

30. Nadrous HF, Yu AC, Specks U, et al. Pulmonary involvementin Henoch–Schonlein purpura. Mayo Clin Proc 2004;79:1151–7.

31. Gonzalex-Gay MA, Calvino MC, Vazquez-Lopez ME, et al.Implications of upper respiratory tract infections anddrugs in the clinical spectrum of Henoch–Schonleinpurpura in children. Clin Exp Rheumatol 2004;22:781–4.

*Retrospective study analyzing the correlationbetween incidence of URI in conjunction with thediagnosis of HSP.

32. Halling SFE, Soderberg MP, Berg UB. Henoch–Schonlein nephri-tis: clinical findings related to renal function and morphology.Pediatr Nephrol 2005;20:46–51.

*Study of 73 children and the clinical findings,including proteinuria and GFR, that relate tosignificant renal pathology. Surprisingly even mild tomoderate proteinuria can be associated with severemorphological changes seen on biopsy.

33. Liu PM, Bong CN, Chen HH, et al. Henoch–Schonlein purpurawith hemorrhagic bullae in children: a report of two cases.J Microbiol Immunol Infect 2004;37:375–8.

34. Watanabe T, Abe Y. Muscle involvement in a patient withHenoch–Schonlein purpura nephritis. Pediatr Nephrol 2004;19:227–8.

35. Koizumi M, Hagino D, Fukuyama C, et al. Schonlein–

Henoch purpura during pregnancy: case report andreview of the literature. J Obstet Gynaecol Res 2004;30:37–41.

36. Wen YK, Yang Y, Chang CC. Cerebral vasculitis and intracerebralhemorrhange in Henoch–Schonlein purpura treated with plas-mapheresis. Pediatr Nephrol 2005;20:223–5.

ARTICLE IN PRESS

Henoch–Schonlein purpura 263

37. Huber AM, King J, McLaine P, et al. A randomized, placebo-controlled trial of prednisone in early Henoch–Schonleinpurpura. BMC Med 2004;2:1–7.

**RCT with data indicating that early interventionwith prednisone therapy has no effect on outcome ofGI or renal complications.

38. Sugiyama H, Watanabe N, Onoda T, et al. Successful treatmentof progressive Henoch–Schonlein purpura nephritis with tonsil-lectomy and steroid pulse therapy. Intern Med 2005;44:611–5.

39. Someya T, Kaneko K, Fujinaga S, et al. Cyclosporine A for heavyproteinuria in a child with Henoch–Schonlein purpura nephritis.Pediatr Int 2004;46:111–3.

40. Iqbal H, Evans A. Dapsone therapy for Henoch–Schonleinpurpura: a case series. Arch Dis Child 2005;90:985–6.

41. Harries MJ, McWinney P, Melsom R. Recurrent Henoch–Schonleinpurpura controlled with ciclosporin. J Roy Soc Med2004;97:184–5.

42. Tarshish P, Bernstein J, Edelmann C. Henoch–Schonlein purpuranephritis: course of disease and efficacy of cyclophosphamide.Pediatr Nephrol 2004;19:51–6.

43. Kawasaki Y, Suzuki J, Suzuki H. Efficacy of methylprednisoloneand urokinase pulse therapy combined with or withoutcyclophosphamide in severe Henoch–Schonlein nephritis: aclinical and histopathological study. Nephrol Dial Transplant2004;19:858–64.

44. Chen TC, Chung FR, Lee CH, et al. Successful treatment ofcrescentic glomerulonephritis associated with adult-onsetHenoch-Schonlein purpura by double-filtration plasmapheresis.Clin Nephrol 2004;61:213–6.

45. Kawasaki Y, Suzuki J, Murai M, et al. Plasmapheresis therapy forrapidly progressive Henoch–Schonlein nephritis. Pediatr Ne-phrol 2004;19:920–3.

46. Dixit MP, Dixit NM, Scott K. Managing Henoch–Schonlein purpurain children with fish oil and ACE inhibitor therapy. Nephrology2004;9:381–6.

*Study of five patients with HSP and renalcomplications treated with fish oil and ACEI therapy.The cohort had significant reductions in renalcomplications.

47. Rigante D, Candelli M, Federico G, et al. Predictive factorsof renal involvement or relapsing disease in childrenwith Henoch–Schonlein purpura. Rheumatol Int 2005;25:45–8.

**Prospective multivariant analyis of factors relatedto prolonged renal involvement in children diagnosedwith HSP. The study found that the only predictivevariable of renal involvement is prolonged classic HSPrash greater than 1 month.

48. Narchi H. Risk of long-term renal impairment and duration offollow-up recommended for Henoch–Schonlein purpura withnormal or minimal urinary findings: a systematic review. ArchDis Child 2005;90:916–20.**A literature review that recom-mends a 6 month duration of follow-up for children diagnosedwith HSP without significant renal involvement at the time ofdiagnosis

Further reading

13. Cakir N, Pamuk ON, Donmez S. Henoch–Schonlein purpura intwo brothers imprisioned in the same jail: presentation twomonths apart. Clin Exp Rheumatol 2004;22:235–7.

14. Patterson CC, Ross P, Pope-Harman AL, et al. Alpha-1 anti-trypsin deficiency and Henoth–Schonlein purpura associatedwith anti-neutrophil cytoplasmic and anti-endothelial cellantibodies of immunoglobulin-A isotype. J Cutan Pathol2005;32:300–6.