By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Due to essential maintenance the subscribe/renew pages will be unavailable on Wednesday 26 October between 02:00- 08:00 BST/ 09:00 – 15:00 SGT/ 21:00- 03:00 EDT. Apologies for the inconvenience.
Henoch–Schönlein purpura is the most common systemic vasculitis of childhood. In the majority of children, the outcome of Henoch–Schönlein purpura is excellent with spontaneous resolution of symptoms and signs. However, a small subset of patients will develop long-term sequelae in the form of chronic kidney disease. While the clinical presentation and diagnosis of Henoch–Schönlein purpura is straightforward, treatment of Henoch–Schönlein purpura nephritis and long-term renal outcomes of more severely affected children are less certain. This review article gives a general overview of Henoch–Schönlein purpura with emphasis on recently published information, including the new classification of childhood vasculitis, insights into pathogenesis of Henoch–Schönlein purpura and a summary of various treatments of established Henoch–Schönlein purpura nephritis.
Henoch–Schönlein purpura is the most common systemic vasculitis of childhood presenting with a tetrad of purpura, arthritis or arthralgia, abdominal pain and renal disease. While the presence of purpura is a compulsory criterion for the diagnosis of Henoch–Schönlein purpura, other signs and symptoms are more variably present.
Abnormal glycosylation of immunoglobulin A1 molecules predisposes patients with Henoch–Schönlein purpura to formation of large immune complexes. Clearance of these large molecules is impaired, they deposit in small vessel walls of the affected organs and trigger immune response leading to inflammatory reaction presenting as clinical signs and symptoms.
The long-term morbidity of Henoch–Schönlein purpura is related to nephritis. Based on the current evidence, early immunosuppressive treatment of children with Henoch–Schönlein purpura should be reserved for those presenting with severe kidney involvement (rapidly progressive glomerulonephritis, nephrotic syndrome). There might be a role for immunosuppression in patients with ongoing nephritis (persistent/increasing proteinuria), but this approach will have to be tested in large prospective studies before it can be widely accepted in clinical practice.
The first clinical description of Henoch–Schönlein purpura (HSP) comes from English physician William Heberden, who described two boys with clinical findings suggestive of HSP including purpuric rash, arthralgia and abdominal pain. HSP carries the names of two 19th century German physicians, Johann Schönlein and his student Eduard Henoch. Schönlein described the association of non-thrombocytopaenic purpura and joint pain in 1837, which he called purpura rheumatica. Henoch added gastrointestinal and renal involvement in 1874. Purpura is a latin word for purple, which has its origin in Greek word porphyra, a name of the Tyrian purple dye secreted by sea snails Murex trunculus and Murex brandaris, used for centuries as an ‘imperial dye’. Anaphylactoid purpura, another commonly used name for HSP, is incorrect and should not be used because anaphylaxis does not play a significant role in the pathogenesis of this vasculitis.
Definition and Classification
HSP is a systemic vasculitis characterised by the deposition of immunoglobulin A (IgA)-containing immune complexes in the walls of small vessels (arterioles, capillaries and venules). According to the recently endorsed European League Against Rheumatism, Paediatric Rheumatology European Society and Paediatric Rheumatology International Trials Organisation classification of childhood vasculitis, HSP belongs to the group of non-granulomatous, predominantly small vessel vasculitides (Table 1).
Table 1. New EULAR/PRINTO/PRES endorsed classification of childhood vasculitis (with permission from reference 4)
I Predominantly large vessel vasculitis
II Predominantly medium sized vessel vasculitis
Childhood polyarteritis nodosa
III Predominantly small vessel vasculitis
Isolated cutaneous leucocytoclastic vaculitis
Hypocomplementic urticarial vasculitis
IV Other vasculitides
Vasculitis secondary to infection (including hepatitis B associated polyrateritis nodosa), malignancies, and drugs, including hypersensitivity vasculitis
Vasculitis associated with connective tissue diseases
Isolated vasculitis of the central nervous system
HSP is the most common childhood vasculitis with a reported annual incidence that varies between 10 and 30 cases per 100 000 children < 17 years based on hospital and overall population estimates.[5, 6] These reports are likely to underestimate the true prevalence of HSP given the voluntary nature of reporting to these surveys. The mean age of presentation is 6 years with most cases in children < 10 years of age, and recent studies suggesting an equal incidence in males and females. HSP occurs predominantly in cold months of the year and has been reported worldwide.
Aetiology and Pathogenesis
The majority of HSP cases are preceded by an upper respiratory tract infection suggesting a potential infectious trigger. Streptococcus, staphylococcus and parainfluenza are most commonly implicated but there are multiple case reports describing the association between virtually all respiratory pathogens and HSP. Anecdotal reports also describe HSP cases after vaccination, with multiple vaccines implicated, including the recently developed pandemic influenza A (H1N1) vaccine. An association between drugs and HSP has also been reported, although the role of these medications in pathogenesis of HSP is uncertain given that most were being used at the time of onset of the disease for treatment of concurrent infection.
The characteristic pathological feature of HSP vasculitis is a deposition of IgA-containing immune complexes in vessel walls of affected organs and in the kidney mesangium. Histologically, the appearance of HSP nephritis is identical to IgA nephropathy and recent studies in patients with HSP and IgA nephropathy have detailed a potential role of IgA1 in the pathogenesis of both of these conditions.[11, 12]
IgA is found in serum and mucosal secretions and is a major class of immunoglobulin that plays an important role in mucosal immunity. In man, more IgA is produced than all other immunoglobulin classes combined because of the high mucosal synthesis and short half-life of IgA of 5–6 days. Of two IgA subclasses, IgA1 is phylogenetically younger and differs from IgA2 by insertion of a 13–17 amino acid sequence in the hinge region of the IgA1 molecule. The hinge region of IgA1 is heavily glycosylated in normal individuals. N-acetylgalactosamine (GalNAc) is O-linked to serine residues (O-glycosylation) and elongation of the glycan chains is achieved by further addition of galactose (Gal) and N-acetylneuraminic acid (NeuNAc) (galactosylation and sialylation) to GalNAc (Fig. 1). Glycosylation of IgA1 seems to play important role in facilitating clearance of IgA1 molecules. Normally glycosylated IgA1 molecules interact with the asialoglycoprotein receptor (ASGP-R) expressed on the hepatocytes, followed by internalisation and degradation of these molecules. Patients with HSP and IgA nephropathy express inherited Gal deficient glycosylation of IgA1 molecules. While the absence of Gal exposes GalNAc as a terminal glycan, the stimulus for the formation of antibodies against GalNAc is unknown. However, many microorganisms express GalNAc-containing sugars on their surface. During infection by these microorganisms, antibodies to GalNAc on bacteria or viruses could potentially cross-react with GalNAc on IgA1 molecule with subsequent formation of large IgA1-IgG immune complexes that cannot reach ASGP-R in the space of Disse in the liver but are able to cross endothelial fenestrae in the glomerulus and deposit in the mesangium (Fig. 2). Deposited immune complexes activate the alternative complement pathway (with deposition of C3) and recruit inflammatory cells causing glomerulonephritis.[12, 15] Deposition of IgA1-containing immune complexes in other sites (skin, gut, joints) leads to organ-specific clinical manifestations of HSP.
HSP is a systemic vasculitis with multiorgan involvement. The classic tetrad of signs and symptoms includes: 1/ palpable purpura, 2/ arthritis or arthralgia, 3/ abdominal pain, and 4/ renal disease.
Skin involvement is present in all children with HSP. Petechiae and palpable purpura are the most common, but erythematous, macular, urticarial or even bullous rashes have also been observed. Purpura is characteristically distributed symmetrically over the extensor surfaces of the lower limbs, buttocks and forearms (Fig. 3) with involvement of trunk and face described occasionally in younger children. Recurrence of purpura, which might be associated with more severe renal involvement, is observed in 25% of children with HSP.
Arthritis/arthralgia is present in three quarters of children with HSP. Joint involvement is usually oligoarticular with large joints of the lower extremities (knee, ankle, hip) most commonly affected. There is usually prominent periarticular swelling, tenderness and pain; erythema and joint effusion are rare. Arthritis is non-deforming and heals without chronic damage within a few weeks.
Approximately two thirds of children with HSP develop abdominal pain, usually diffuse, increasing after meals, and sometimes associated with nausea and vomiting. These symptoms are caused by submucosal haemorrhage and oedema of the bowel wall, predominantly affecting the proximal small bowel. The most severe gastrointestinal complication is intussusception, affecting 3–4% patients with HSP. In 60% of these cases, it is limited to small bowel. Clinical presentation of intussusception is characterised by severe abdominal pain, often colicky in nature and vomiting. Other significant, though less common gastrointestinal complications are gangrene of the bowel, bowel perforation and massive haemorrhage.
Renal involvement is reported in 20–55% of children with HSP.[18, 19] The most common finding is isolated microscopic haematuria, usually developing within 4 weeks of the onset of the disease. Proteinuria of variable degree might be present, and if severe can present as nephrotic syndrome. Hypertension might develop at the onset or during recovery. Renal function is usually normal but the occasional patient might present with a progressive glomerulonephritis with significant renal impairment.
Other less common clinical manifestations of HSP include cerebral vasculitis, scrotal or testicular haemorrhage, and interstitial pulmonary haemorrhage.[20-22] Distal ureteric vasculitis resulting in ureteric stenosis, presenting as renal colic has also been described. Potential complications of HSP are summarised in Table 2.
Table 2. Possible complications of Henoch–Schönlein purpura
Gangrene of the bowel
Central nervous system
Diagnosis of HSP is based on the presence of purpura (palpable) or petechiae (without thrombocytopaenia) with lower limb predominance (mandatory criterion) plus at least one of the flowing four features: (1) abdominal pain; (2) arthritis or arthralgia; (3) leukocytoclastic vasculitis or proliferativeglomerulonephritis with predominant deposition of IgA on histology; (4) renal involvement (haematuria, red blood cell casts or proteinuria). Laboratory tests are complementary in assessing renal involvement (urinalysis, urine microscopy, serum creatinine), and imaging studies are helpful in the evaluation of abdominal involvement and its potential complications (intussusception). In children with incomplete or unusual presentation, biopsy of the affected organ (skin, kidney) confirms the diagnosis.
Every child with HSP should have urinalysis performed at diagnosis and during follow-up. Dipstick assessment of urine for blood and protein is a good screening test for nephritis. Urine microscopy may reveal dysmorphic red cells and red-cell casts. Positive dipstick reading for protein requires quantification of protein excretion either by measuring protein/creatinine ratio on a first morning urine sample or protein excretion on a timed urine sample (24-hour collection).
There are no blood tests specific for HSP and measurement of serum levels of total IgA is not helpful in confirming the diagnosis or providing prognostic information. Galactose-deficient IgA1 serum levels seem to distinguish patients with HSP nephritis from patients without nephritis, and might become an important commercially available biomarker in the future.[14, 25]
Not all patients with HSP require diagnostic imaging, which is generally reserved for children with abdominal pain in whom intussusception is suspected. Abdominal ultrasound is the technique of choice with the accuracy in diagnosing intussusception approaching 100% in experienced hands. Concentric rings of tissue representing components of bowel and mesenteric fat create a classic ‘target sign’ (Fig. 4). The classic ‘meniscus sign’ of intussusception on contrast enema, where the apex of the intussusception projects into the contrast material, is not present in cases of intussusception limited to small bowel.
Biopsy of the affected skin reveals leukocytoclastic vasculitis with deposition of IgA-containing immune complexes, predominantly in small vessels in the papillary dermis (primarily venules). Neutrophils undergo destruction (leukocytoclasis) with destructive fragmentation of the nuclei of dying cells (karyorrhexis) during apoptosis or necrosis (Fig. 5). Deposits of IgA and C3 in the dermal capillaries of purpuric lesions and uninvolved skin by immune-fluorescent staining are considered valid diagnostic criterion, with 100% specificity in combination with leukocytoclastic vasculitis.
Kidney biopsy is usually performed in patients with uncertain diagnosis and in those with more severe kidney involvement (rapidly progressive nephritis, nephrotic syndrome). In general, there is a correlation between the severity of renal manifestations and findings on kidney biopsy. Light microscopy findings can range from mild mesangial proliferation to severe crescentic glomerulonephritis. Diffuse mesangial IgA deposits seen on immunofluorescence are the hallmark of HSP nephritis (Fig. 6) and co-deposition of C3 complement (75%) might also be present. The absence of the classical complement pathway components (C1q and C4) distinguishes HSP nephritis from other forms of immune-mediated glomerulonephritis, such as lupus nephritis. Electron microscopy shows electron dense deposits in the mesangial areas. The current classification of HSP nephritis is based on the extent of proliferation and the presence of crescents on light microscopy, but other histological findings, such as mesangial/subendothelial deposits, the extent of tubulointerstitial damage or glomerular sclerosis might be better predictors of the outcome.[28, 29]
Management of HSP
Management of HSP includes supportive care, symptomatic therapy and, in some cases, immunosuppressive treatment.
The basic principles of supportive care consist of maintenance of good hydration, symptomatic pain relief and monitoring for the development of complications. If adequate hydration cannot be maintained orally, intravenous fluids should be considered. Parenteral nutrition is usually unnecessary, except in cases with prolonged severe abdominal involvement precluding enteral feeding. Patients with severe abdominal pain need prompt evaluation and investigations to exclude intussusception. In cases with sudden change of mental status, intracranial haemorrhage should be excluded with appropriate imaging. Arthritis/arthralgia usually responds well to non-steroidal anti-inflammatory drugs (NSAIDs), but occasionally requires opioids for adequate symptomatic relief. Treatment is usually well tolerated and is not associated with an increased risk of gastrointestinal bleeding. Patients with compromised renal function taking NSAIDs need close monitoring of their fluid status, blood pressure and renal function.
The use of glucocorticosteroids (GCS) in HSP has been source of controversy for many years. While the suggested benefits of early GCS treatment have included shortened duration of abdominal pain, decreased risk of intussusception and decreased risk of surgical intervention,[31-33] the quality of evidence is generally poor, having come from mostly from small studies or case reports. In clinical practice, short courses of GCS are being used in patients with severe abdominal pain, usually with rapid symptomatic improvement. This treatment cannot be recommended in all patients with HSP since the majority will improve spontaneously. While some reports have suggested that early treatment with GCS might prevent development of nephritis and chronic kidney disease, a recent Cochrane review concluded that there is no evidence from randomised controlled trials that the use of GCS prevents kidney disease in children with HSP.
Immunosuppressive treatment of HSP nephritis is used in patients with severe kidney involvement (nephrotic range proteinuria and/or progressive renal impairment). In these cases, renal biopsy should be considered before treatment. Mild renal involvement (microscopic haematuria or mild proteinuria) does not require biopsy or immunosuppressive treatment, but these children need close follow-up.
In patients with rapidly progressive glomerulonephritis or nephrotic syndrome (usually accompanied by crescents on kidney biopsy), pulse intravenous methylprednisolone followed by 3 to 6-month course of oral steroids is most commonly used. A current KDIGO guideline suggests adding cyclophosphamide to steroid treatment for crescentic glomerulonephritis even though the quality of evidence is low with a lack of demonstrated improvement in renal outcome.[37, 38] Plasmapheresis has also been used in children with rapidly progressive glomerulonephritis, but it is difficult to assess its efficacy due to selection bias (used in the most severe cases) and concurrent administration of other immunosuppressive treatments. Recent studies in children with HSP nephritis and nephrotic syndrome suggest a potential benefit of cyclosporine A (CsA) in achieving remission of proteinuria and histological improvement of nephritis on follow-up kidney biopsies.[40, 41] Other treatments used with some success in small studies include intravenous immunoglobulin, combined therapy of immunosuppression and anti-clotting therapy (warfarin, dipyridamol and acetylsalicylic acid), tonsillectomy, and B-cell depletion with rituximab and mycophenolate mofetil.[42-46] The efficacy of these treatments is yet to be tested in prospective clinical trials.
Use of angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs) has become an accepted treatment of HSP nephritis with persistent proteinuria, with beneficial effects not only on reduction of proteinuria but also on inhibition of renal fibrosis. Although there are no available studies on the efficacy of ACEIs or ARBs in HSP nephritis, long-term data showing their beneficial effect on renal survival and improvement of proteinuria in patients with IgA nephropathy, a disease with the same pathophysiology, are encouraging.
Prognosis of HSP
In the majority of children, the outcome of HSP is excellent with spontaneous resolution of symptoms and signs. HSP recurs in approximately one third of patients, typically within 4 months of the initial presentation. Recurrent purpura can be occasionally associated with joint complaints and episodes of gross haematuria although each subsequent episode is generally milder and shorter. The long-term morbidity of HSP is related to the degree of HSP nephritis.
In unselected cohorts of children, HSP nephritis is a mild disease, characterised by microscopic haematuria and minimal proteinuria, with <1% risk of progression to end-stage kidney disease (ESKD). Reports from tertiary centres indicating that 10–30% of children will develop ESKD are likely to overestimate the true risk of ESKD due to the selection of cases with more severe renal impairment seen in these centres.[48, 49] Children at risk are those with nephrotic or nephritic/nephrotic syndrome or renal failure at presentation, and those with impaired kidney function and persistent proteinuria after several years of follow-up. Children with uncomplicated HSP are usually managed in the primary care setting either by a GP or a paediatrician. The aim of the initial follow-up is to identify patients with worsening kidney involvement and is based on serial urinalyses, blood pressure measurement, blood tests to assess kidney function and exclusion of other causes of glomerulonephritis. A practical pathway for detection and referral of children with HSP nephritis to a paediatric nephrologist during the first 6–12 months after diagnosis has been developed (Fig. 7). The involvement of a paediatric rheumatologist in cases of severe arthritis/arthralgia might also be warranted.
Histological recurrence of HSP nephritis (IgA deposition) in transplanted kidneys can be as high as 60% but is rarely associated with clinical recurrence. Long-term outcomes of transplanted kidneys in patients with HSP nephritis are comparable to other primary diseases with ∼90% survival at 10 years.
HSP is a common childhood vasculitis with a good outcome in the majority of affected children. However, there is a small subgroup of children who will develop significant renal impairment and some of them will eventually progress to ESKD and require kidney transplantation. To predict which patients are at risk of long-term renal sequelae, we need better biomarkers reflecting the tissue damage, such as urinary biomarkers of tubulointersitial (profibrotic cytokines) or glomerular (urinary podocytes) injury. Furthermore, we need well-designed multicentre randomised controlled prospective studies on treatment of HSP nephritis to answer questions such as who should be treated, when and with what medication. It is a hope that with the early detection of HSP patients at risk of developing ESKD and their appropriate and timely treatment, the outcomes of these children will improve.
I thank Dr Leo Francis, Pathology Queensland, Royal Brisbane and Women's Hospital, Brisbane for providing the histological samples and Dr Steve McTaggart, Queensland Child and Adolescent Renal Service, Brisbane for reviewing this manuscript.
Multiple Choice Questions
Q1. Which one of the following mechanisms plays the most important role in pathogenesis of Henoch–Schönlein purpura:
Anaphylactoid reaction to various medications (non-steroidal anti-inflammatory drugs, antibiotics)
Immune reaction to infective agents (bacteria, viruses) involving the classical complement pathway
Inherently deficient glycosylation of IgA1 molecules predisposing to formation of immune complexes and poor clearance of IgA1 from circulation
Abnormally expressed asialoglycoprotein receptor on hepatocytes leading to poor clearance of IgA1 molecules from circulation with subsequent formation of immune complexes
Increased production of normally glycosylated IgA1 immunoglobulin in affected individuals leading to the deposition of this immunoglobulin in the vessels of affected organs
A1. Correct answer is c.
People with HSP (and IgA nephropathy) have inherited predisposition to abnormal glycosylation of IgA1 molecules. Following intercurrent infection, abnormally glycosylated molecules form large immune complexes that are poorly cleared by the liver and deposit in the vessels of the affected organs. Anaphylactoid reactions do not play role in pathogenesis of HSP. Activation of the alternative complement pathway plays a role in organ damage; the classical complement pathway is not involved in pathogenesis of HSP. Individuals with HSP have normal asialoglycoprotein receptors in the liver. Abnormal glycosylation of IgA1 molecules rather than increased production of normally glycosylated IgA1 is the main predisposing factor in pathogenesis of HSP.
Q2. Which one of the following statements regarding diagnosis of Henoch–Schönlein purpura is correct:
There is no specific test for HSP and the diagnosis is made based on the clinical findings
Total IgA in serum is elevated in the majority of patients with HSP
Measurement of galactose-deficient IgA1 in serum is a useful and widely available test for HSP
Contrast enema is an imaging test of choice for diagnosis of intussusception in children with HSP
Kidney biopsy is usually performed in children with HSP who have haematuria or proteinuria on presentation
A2. Correct answer is a.
Diagnosis of HSP is clinical and is based on presence of purpura or petechiae plus one of the following: abdominal pain, arthritis or arthralgia, histological presence of leukocytoclastic vasculitis or proliferative glomerulonephritis, or renal involvement (haematuria, red blood cell casts or proteinuria). There are no tests specific for HSP. Serum level of total IgA is not clinically useful test since it is elevated in ∼50% of patients with HSP. Serum levels of galactose-deficient IgA1 can distinguish patients with HSP from healthy controls, but this test is not widely available for clinical purposes. Contrast enema would miss intussusception limited to the small bowel; abdominal ultrasound is the imaging test of choice. Kidney biopsy is usually done in patients with uncertain diagnosis and in those with more severe kidney involvement (rapidly progressive nephritis, nephrotic syndrome).
Q3. Which one of the following is the correct answer with regard to management of Henoch–Schönlein purpura:
All children with HSP should be admitted to hospital for close monitoring and intravenous hydration
Treatment with non-steroidal anti-inflammatory drugs is contraindicated in children with HSP because of the potential adverse effects on the kidneys
Early treatment with glucocorticosteroids will prevent development of HSP nephritis and chronic kidney disease
Children with HSP who have persistent microscopic haematuria require kidney biopsy and immunosuppressive treatment
Treatment with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers is an accepted treatment of HSP nephritis in children with persistent proteinuria
A3. Correct answer is e.
Majority of children with HSP can be managed out of hospital with close monitoring in an outpatient setting. NSAIDs are useful treatment for arthralgia/arthritis in children with HSP, but potential side effects on the kidney must be kept in mind and close monitoring of kidney function is important. There is no evidence from randomised controlled trials that early use of glucocorticoids prevents kidney disease in children with HSP. Persistent microscopic haematuria is a common finding in children with mild HSP nephritis; most of these children continue to have normal kidney function and will do well. In children with HSP nephritis and persistent proteinuria (especially those who are also hypertensive), treatment with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers seems to slow down the progression of kidney disease. At what level of proteinuria one should start this treatment is, however, unclear.