Henoch-Schönlein (HS) purpura is a common cause of renal glomerular injury in children. This condition is responsible for 10-15% of glomerulonephritis in children. The outcome is generally favorable, but up to 5% of patients develop kidney failure. The outcome of patients with kidney biopsy is less favorable with 7-50% of them progressing to chronic renal failure.

Prevalence of HS is difficult to determine from literature. Annual incidence is estimated at 6.1 / 100,000 children in the Netherlands and up to 20.4 / 100 000 children in the United Kingdom. The proportion of children with HS who develop renal disease is difficult to determine because the numbers reported in the literature are variable and depend greatly on the type of the reporting center, whether or not specialized in pediatric nephrology. Thus the proportion of renal disease varies from 20% to 100% of children with a HS.

The treatment of HS nephropathy (HSN) usually depends on the severity of histological lesions but histological classification is discussed and there is currently no consensus. Randomized studies are scarce and often do not allow to draw clear conclusions. A meta-analysis suggested a positive effect of corticosteroids on renal prognosis of severe forms but in this study the definition of renal disease was very heterogeneous. The only classification of the HSN recognized is from the International Study Group of Kidney Disease in Childhood (ISKDC) which is the following: grade I: minimal glomerula abnormalities, grade II: pure proliferation, grade III: crescents/ segmental lesions <50%,grade IV: crescents/ segmental lesions 50 to 75%, grade V: crescents/ segmental lesions > 75%, grade VI: pseudomesangiocapillary. However, this classification is questioned because it ignores other significant histological lesions such as interstitial fibrosis, tubular lesions, glomerular and interstitial inflammation, the appearance of crescents (segmental or totally encompassing the glomerulus, fibrous or cellular), segmental sclerosis, fibrosis and arteriolar appearance in immunofluorescence.

There is currently no consensus on the criteria indicating the initiation of corticosteroid therapy whether oral or intra venous bolus. Some patients with severe clinical and / or histological initial presentation can evolve to remission spontaneously while others who have more moderate initial symptoms will evolve later to kidney failure. The management is therefore heterogeneous. In France, some centers perform a kidney biopsy almost always before starting treatment (or in the days following the start of treatment), while in other centers's treatment decision is based on the biology resulting from the glomerular disease, kidney biopsy being performed possibly in a second time in case of failure of the initial treatment.

Principal objective of the study: assessment of the interest for the long term outcome of performing early a kidney biopsy (before the establishment of treatment or within 15 days after the start of treatment) in children with HSN compare to kidney biopsy performed later (depending on the response to initial therapy) or not performed.

Secondary objective: assessment of the impact of early kidney biopsy (before the establishment of treatment or within of 15 days after the start of treatment) on the initial treatment HSN : does it modify or not the treatment started right before it (decided on clinical and biological criteria).

If in the past the HSN could be considered a rather benign disease not requiring specific active treatment, the studies evaluating the long-term outcome have shown the risk of progression to Chronic Kidney Disease (CKD) and have lead to recommend the use of corticosteroids and immunosuppressors even in not rapidly progressive glomerulonephritis forms (24,25). Unfortunately clinical studies are scarce, often with few patients and uncontrolled (17,26,27). However, the efficacy of methylprednisolone pulses followed by oral steroids has been suggested in several studies, as in the study conducted at the Hospital Necker, where prospective patients receiving pulses of methylprednisolone were compared to an historical cohort from the same center (28) in a control arm of a randomized controlled trial (29), and in studies where patients received combinations of several immunosuppressive drugs (30-32).

Thus, the current French way to manage severe HSN is to give methylprednisolone pulses followed by oral steroids. Anti-proteinuric medications are for initially mild forms or sequelae. Immunosuppressor is added to steroid in the forms not responding well to initial corticosteroid therapy.

If these therapies are used by most of pediatric teams, practice in kidney biopsy (KB) varies from one center to another. Some teams routinely perform KB before the start of steroid therapy (and adapt the treatment to the results), while others first establish the treatment and perform the KB only if the evolution is not as expected. This second approach reduces the number of KB since patients with favorable outcome will never be biopsied.

The question is which of these two attitudes is the best. Do the biopsied patients have a better prognosis at 5 years (because the lesions were better evaluated in comparison to clinical evaluation, because the diagnosis was confirmed, despite the risk taken to perform KB, because there is no excess of treatment) than those who were not biopsied at the initial period (with possible errors in assessing the severity of injury ), or is there is no difference (because the treatment is the same whether KB was performed or not, because clinical criteria prevail in the therapeutic decision on histological criteria, because KB may falsely reassure and may lead to stop treatment too early (sample problem)).

The aim of the study is to answer these questions, to improve patients care by identifying the most effective strategy to improve long term prognosis, and to standardize practices to make randomized control trials easier to drive in the future.

Inclusion Criteria:

• Patients aged under 18 at the time of initial management of HSN
• Patient treated for HSN in a French pediatric nephrology unit
• Initial care between January 2006 and December 2010
• Patient who had or not a kidney biopsy

Exclusion Criteria:

• Patients who received treatment with corticosteroids for another complication of HSN or other pathology

Study of historical cohort, from different centers in France, known to have a different policy for kidney biopsy in HSN (agreements: Reims, Hôpital Trousseau Paris, Hôpital Robert Debré Paris, Toulouse, Lille, Nantes, Lyon and Necker).