• 
• 

Three Part Question

In [children with Henoch-Schonlein purpura] does [early use of steroid when compared to non–steroid group] prevent [subsequent nephropathy]?

Clinical Scenario

A 5 year old boy presented to the A & E with mild abdominal pain & florid rash particularly involving the lower limbs. The rash was popular, non-blanching and primarily over the buttocks & lower limbs. Apart from the rash rest of clinical examination including blood pressure were normal. His urine dip stick did not show any significant blood or protein. In presence of a classical rash a diagnosis of Henoch-Schonlein purpura (HSP) was made. The child was started on regular analgesia & was sent home with plans of subsequent follow-ups. In the morning hand over the question was raised whether the child might have benefited from early steroid with the aim of preventing subsequent nephropathy.

Search Strategy

Medline via PubMed, Embase & Cinahl. Only clinical trials with primary objectives of studying effect of steroids on subsequent HSP nephropathies were included . The search was done separately by both the contributors with identical results.
PUBMED (1966 - till date)
Search words : "Purpura , Schonlein-Henoch " [MeSH] AND "Steroids" [MeSH] AND ("Kidney Diseases" [MeSH] OR "Kidney Failure", " Chronic" [MeSH] OR "Kidney Failure" [MeSH] OR "Proteinuria" [MeSH] OR "Hypertension , Renal" [MeSH] OR "Albuminuria"[MeSH]) Field: All Fields , Limits: All Child: 0-18 years , English, Humans
EMBASE - 1974 to date; Search term: "Anaphylactoid-Purpura#.MJ." AND "Steroid#.W..MJ" AND NEPHRITIS#. W..MJ. OR KIDNEY- DISEASE#.MJ. OR GLOMERULONEPHRITIS#. W..MJ. OR HEMATURIA# .W..MJ. OR STEROID-THERAPY #.MJ. OR IMMUNOGLOBULIN-A- NEPHROPATHY#.MJ. OR KIDNEY#.W..MJ." AND CHILD=YES, ENGLISH=YES
CINAHL (R) - 1982 to date; Search term: "Purpura-Schoenlein-Henoch#.MJ." "Steroids#.W..MJ." Results 1, but one relevant COCHRANE database – A Cochrane systemic review on this topic is on-going and awaiting final publication. We did get an important additional randomised control study in process of being indexed for pubmed from the references cited by the Cochrane Reviewer.

Search Outcome

Pubmed - : 34 articles but only 2 relevant
Embase - 15 articles, 2 relevant of which duplicated in pubmed
Cinahl - 1 relevant

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Huber et al 2004	40 children (2-15 yrs) with HSP as initial presentation, seen in a single tertiary centre, were randomised to either receive early oral prednisone (n=21) or an identical appearing placebo (n=19).	Randomised prospective double blind placebo controlled study.(Level = 1b)	Co-primary outcome renal involvement at 1 year & acute gastrointestinal complications	No difference in rate of involvement between prednisone ( 3/21) vs. placebo group (2/19),p=1.0	Early prednisone therapy (within 7 days of onset) in HSP did not appear to reduce the risk of renal involvement at 1 year. Relatively small study (N=40) but had a power of over 90% to show a 50% reduction in renal involvement at 1 year. Used surrogate marker of renal involvement at 1 year for the primary outcome which was chronic renal insufficiency.
Saulsbury 1993	Medical records over a period of 12 yrs in a tertiary centre of all children (1-15 yrs) with acute HSP (n=69) were reviewed. 19 of them had features of nephritis at initial presentation & were excluded. Of the rest, 20 were treated with steroid, primarily for GI manifestations (steroid group n=20). 30 were found not to have steroid (non steroid n=30)	Retrospective case control study (Level=3b)	Utility of steroid for prevention of subsequent nephritis in HSP.	No significant difference was found in the incidence of subsequent nephritis between the steroid & non steroid group. Nephritis developed in 4 children in 'steroid group' (4/20 - 20%).In 'non steroid group' 6 developed nephritis ( 6 /30-20%)	Steroid therapy did not seem to prevent nephritis in HSP. Children with features of nephritis at initial presentation were excluded. Mean interval between onset of symptoms & initiation of steroid therapy was 5±4 days. Small sample size (N=50),but power analysis indicated 90% chance of detecting a difference of 20% or more between the two groups at 0.05 level of significance.
Mollica et al 1992	Over a period of 10 yrs a total of 221 unselected children with HSP presented in a University hospital. 53 were excluded from study , 34 as they had haematuria &/or proteinuria on initial presentation & 19 because of insufficient follow –up. Remaining 168 children were assigned alternately into either steroid group(n=84) or non steroid(n=84). Following recovery from acute phase both groups were frequently followed up over 24-36 months .	Prospective randomised controlled trial (Level =2b)	Incidence of subsequent HSP nephropathy.	None in the 'steroid group' developed nephropathy(0/84) whereas in the 'non steroid group' 10 showed subsequent nephropathy. (10/84- 11.9%).The difference in prevalence of nephropathy between the two group was highly significant, p<0.001.	Good sample Criteria used for diagnosis of HSP was not detailed. Details of randomisation were not mentioned. Interval between onset of HSP & starting of steroid was also not mentioned.
Buchanec et al 1988	Over a ten yrs period case notes of 33 children with HSP & normal urinary findings were reviewed.23 were given prednisone (n=23) & 10 were in the non steroid group(n=10).	Retrospective case control study (Level=3b)	Incidence of renal complication in HSP in dependence of early steroid treatment.	It did show a positive effect in favour of early administration of prednisone. In the 'steroid group' urinary changes occurred in only 1 child (1/23-4.3%) .In contrast 5 children had urinary changes i.e. proteinuria &/or haematuria in the' non steroid group'(5/10-50%) The difference between two groups was statistically significant, p<0.01.	Children with HSP with abnormal renal parameters on admission were excluded. Mean interval between start of the disease & steroid treatment was 21 days.

Comment(s)

Henoch Schonlein Purpura (HSP) is the most common systemic vasculitis in childhood (Tizard). Though the overall prognosis is good, significant morbidity can be associated with disease of the GI tract in the short term & in the long term with nephritis. It can have a wide spectrum of renal manifestations ranging from microscopic haematuria & mild proteinuria to nephritic / nephrotic syndrome or even rapidly progressive crescentic glomerulonephritis & end stage renal failure (Koskimes, Stewart). Persistent evidence of renal involvement is reported to be present at 1 year follow-up in 35-55% of children, although the figure is variable depending upon the referral base (Counhan). Chronic renal involvement may be associated with some degree of renal insufficiency (5% ) & rarely progress to end stage renal failure in up to 1.5% (Koskimes). Despite the low incidence of ESRD, HSP does account for 10% of children entering into end stage renal programme (Koskimes, Kobayashi). HSP being a vasculitis it has been hypothesised for a long time that steroids might be of benefit in preventing subsequent nephropathy. Unfortunately there are very few studies investigating as primary outcome the role of steroids in preventing subsequent nephropathy (Buchanec, Mollica, Saulsbury, and Huber). Of these only 2 were identified as prospective randomised controlled trials (Mollica and Huber). The retrospective studies by Buchanec et al and Saulsbury had conflicting conclusions. Moreover interpretation of retrospective studies are hampered with the likely bias of steroid being used for severe disease & hence preventing any useful interpretation. Though the first randomised controlled trial by Mollica et al had a good sample size (n=168), the lack of proper randomisation & blinding as well as absence of well described methodology leaves concerns regarding possibility of bias in this study. Hence despite the striking significant result in favour of steroids it can not be taken as definitive evidence. Last year Huber et al presented the only other prospective randomised control trial dealing with this particular issue. This was a double blind randomised placebo controlled trial over 4 years. Although relatively small (sample size – 40) it was a well designed, methodologically sound & statistically robust study. It had a power of 90% to show a 50% reduction in renal involvement in 1 year. The other criticism might be that this study used renal involvement at 1 year as a surrogate marker for their primary objective, i.e. chronic renal insufficiency. This might be justified as in 80% of HSP with renal involvement it becomes apparent within first 4 weeks of illness whereas in the rest it usually occurs within next couple of months (Meadow). The study did conclude showing no added benefit of early steroids for preventing subsequent nephropathy. As the article by Huber et al remains the only properly conducted randomised controlled trial, the use of steroids for preventing nephropathy can not be fully justified. More long-term multi-centre double blind placebo controlled randomised trials are required to arrive at a definite conclusion.

Clinical Bottom Line

Urgent need for large long term properly randomised controlled trials dealing with utility of steroids in preventing subsequent HSP nephropathy. Till the availability of further evidence, routine use of early steroids for preventing subsequent nephropathy can not be justified.

References

1. Huber AM, King J, McLaine P, Klassen T, Pothos M. A randomised, placebo-controlled trial of prednisone in early Henoch Schonlein purpura(ISRCTN85109383). BMC Medicine 2004, 2:7.
2. Saulsbury FT. Corticosteroid therapy does not prevent nephritis in Henoch-Schonlein purpura. Pediatr Nephrol (1993) 7:69-71.
3. Mollica F, Volti SL, Garazzo R, Russo G. Effectiveness of early prednisone treatment in preventing the development of nephropathy in anaphylactoid purpura. Eur J Pediatr (1992) 151 140-144.
4. Buchanec J, Galanda V, Belakova S, Minarik M, Zibolen M. Incidence of renal complications in Schonlein-Henoch Purpura syndrome in dependence of an early administration of steroids. International Urology and Nephrology 20(4),pp. 409-412 (1988).
5. Tizard EJ. Henoch- Schonlein purpura. Arch Dis Child 1999; 80:380-383 (April).
6. Koskimes O, Sevgi M, Rapola J, Vilska J. Henoch Schonlein nephritis long-term prognosis of unselected patients. Arch Dis Child 1981 56:482-484.
7. Stewart M , Savage JM , Bell B , McCord B. Long term renal prognosis of Henoch-Schonlein purpura in an unselected childhood population. Eur J Pediatr 1988 147:113-115.
8. Counhan R, Winterborn MH, White RHR, Heaton JM, Meadow SR, Bluett NH, Swetschin H, Cameron JS, Chantler C. Prognosis of Henoch Schonlein nephritis in children. BMJ 1977, 2:11-14.
9. Kobayashi O, Wada H, Okawa K, Takeyama I. Schonlein- Henoch's syndrome in children. Contrib Nephrol 1975;4:48-71.
10. Meadow SR, Glasgow EF, White RHR, Moncireff MW, Cameron JS, Ogg CS. Schonlein –Henoch nephritis. Q J Med 1972; 41 : 241-258.