or HSP. Usually preschool but any age! Boys more than girls. Vasculitis, leucocytoclastic with IgA predominance. EULAR criteria 2010.
Features:
- Urticarial rash becoming purpuric but still raised (pathognomic), predominantly on lower limbs, especially extensor surfaces incl buttocks, but sometimes trunk, rarely face (infant). Accompanying soft tissue swelling, as seen in photo in feet.
- Arthralgia (not migratory, cf acute rheumatic fever)
- Abdo pain, diffuse, colicky, often severe – possible intussusception, melaena
- Proteinuria (30+ mmol/mg albumin:creatinine ratio), 2+ red cells on dip or 5+ on microscopy (or casts) indicating glomerulonephritis, usually asymptomatic cf streptococcal, but a few develop diffuse proliferative lesions with irreversible renal damage.
- Scrotal rash/bruising common, rarely torsion
- Rarely encephalitis, seizures, pulmonary haemorrhage
Pathology
Leukocytoclastic vasculitis with glomerular mesangial IgA deposits. Biopsy of patients with IgA nephropathy identical, and indeed both groups have defective IgA1 glycosylation, which may explain why they may aggregate and precipitate IgA in small vessel walls as well as in the glomerular mesangium.
Complications
Complications and relapse associated with age esp over 6yr. 50% relapse, usually within 6 weeks but can be up to a year later; 50% of those will relapse more than once. Treat with NSAIDs (unless renal involvement!) for joint pain, analgesics for abdo pain. Small study from Turkey found that Ranitidine reduced symptoms.
Steroids (prednisolone 1-2 mg/kg/day for 1-2 weeks) are suggested (grade 2 recommendation) within 3 days of onset of severe abdominal pain (defined as pain requiring hospital admission) or acute GI bleeding – after exclusion of intussusception, of course.
Steroids are also suggested for orchitis, after excluding torsion.
Reviewing 101 children with abdo involvement those who did not receive steroids had an average of 5 days of abdominal pain, whereas all those treated recovered within 24 to 48 hours of starting steroids (letter, J Pediatr Gastroenterol Nutr. 31(3):323-4, 2000).
Abdominal pain is a predictor of renal involvement, so maybe that’s the best reason for giving steroids… (Kidney Int 1998; 53:1755-9).
Renal disease
20 to 90% risk of renal disease, but generally mild (mostly just mild proteinuria and/or haematuria)! Normal urinalysis at day 7 has 97% negative predictive value for chronic renal disease [PLoS One 2012;7:e29512].
About 56% of those children with renal disease develop signs and symptoms of renal disease a week or more after presentation, although generally in first month. Can be up to 6 months later. Incidence of renal failure in HSP nephritis is just 2 to 5%. Risk factors are severe abdominal pain (OR=2.1), age >8yrs (OR=2.7), and relapsed HSP (OR=3).[Arch Dis Child. 2010 Nov;95(11):877-82. doi: 10.1136/adc.2009.182394]
A normal urinalysisNB Children who appear to recover may have significant renal disease many years later. (Lancet. 339:280282, 1992).
If more than 50% crescents on biopsy, then poor prognosis. Only 1% get that far.
Archimedes in 2012 found 3 RCTs of steroids, treatment courses 2-4 weeks, seemed to shorten duration of abdominal pain, with most obvious effect when used early.
Dudley trial of early steroids (day 7) failed to find any benefit at 12 months (n=352). Quite a high proportion of drop outs unfortunately, but prob not enough to influence results. Doesn’t answer question of what to do if severe disease at onset eg nephrotic range. [Arch Dis Child. 2013 Oct;98(10):756-63.]
Non-randomized prospective clinical trial of 223 kids showed that steroids were effective in reducing the severity of abdominal and joint pain and in treating renal disease – steroids did not prevent the development of nephritis [Archives of disease in childhood. 2010;95:877–82, doi: 10.1136/adc.2009.182394 ]. Previous systematic review suggested that steroid treatment at diagnosis did not reduce the median time to resolution of abdominal pain but did significantly reduce the mean resolution time, and increased the odds of resolution within 24 hours.
There is no proven benefit of corticosteroids in the treatment of established HSP nephritis (2009 Zaffanello systematic review, evidence is poor), but used anyway.
The major risks of corticosteroid treatment in children with HSP are masking an acute abdomen or intussuception, and GI bleeding.
Follow up
UK Kidney association guideline 2022 – for uncomplicated presentation, do frequent dipstick checks, for example weekly for first 4-6 weeks then monthly. BP check at presentation and then if evidence of nephritis. No need for parents to check at home otherwise. 6 months minimum (audit criteria, rather than recommendation).
If hypertensive or urinalysis pos, then do proper urine protein:creatinine ratio, U&Es, MC&S. Isolated haematuria is benign. As soon as urinalysis becomes normal, child can have routine follow up.
Scottish guideline was that persisting proteinuria of + or more needs more frequent follow up eg 2 weekly, 2 monthly then 3 monthly with the second line investigations.
Refer urgently for confirmed hypertension, or nephrotic range proteinuria (P:CR over 200mg/mmol).
Refer for biopsy if persisting severe proteinuria (UP: UC >250 mg/mmol for up to 4
weeks), persisting moderate proteinuria (UP: UC 100–250 mg/mmol for 3 months),
AKI stage 1 or greater (creatinine >1.5 × previous baseline or >1.5 × upper limit of normal for age). Treatment depends on histology and severity of clinical features.
ACE inhibitor is suggested (grade 2 recommendation) for persisting mild/moderate proteinuria.
A systematic review found no risk of long-term renal impairment in children with Henoch-Schonlein purpura with normal or minimal urinary findings without nephritic or nephrotic syndrome or renal failure (Arch Dis Child 2005;90:916-20). If urine analysis is normal at presentation, then test for 6 months after the last symptoms. If there is renal disease at presentation, then the risk for progression seems to be more associated with rising proteinuria during follow-up rather than presentation features. (Am J Kidney Dis 2006;47:993-1003)