Acute haemorrhagic oedema of infancy: a condition that is not always benign

Background

Acute haemorrhagic oedema of infancy (AHOI), also known as Seidlmayer’s purpura, is a rare leucocytoclastic vasculitis that occurs in paediatric age and, most of the time, it goes to healing in a few weeks without any therapy. Aim of this report is to highlight the rare but potentially severe complications of AHOI in which a paediatrician may come across in the course of clinical practice, in order to recognise them without delay and simultaneously monitor them properly.

Case presentation

A 2-year-old girl came to our centre for a sudden appearance of erythematous lesions spread to the lower limbs with the inability to walk. Noteworthy in her medical history was a Mycoplasma pneumoniae pneumonia from the previous month, followed after 2 weeks by fever and widespread rash interpreted as scarlet fever. When it came to our centre at the initial examination she appeared to be suffering, with erythematous lesions spread to the lower limbs, buttocks, face and diffuse oedema at the forehead, left eyelid, left leg and instep and arthralgias/arthritis of the large joints. No lymphadenopathy was detected. After few days of hospitalisation there was a rapid evolution of the previous lesions into large, round, ecchymotic plaques, with accentuation of the subcutaneous oedema of the hands, feet, pretibial region and face (figures 1–3). Laboratory tests showed normal values of albumin and total proteins and were nonspecific except for a mild anaemia (Hb 104 g/L) and elevate C reactive protein (1.4 mg/dL). Based on the clinic and medical history, we suspected an AHOI. During hospitalisation, since the child experienced an intense abdominal pain followed by an isolated episode of rectorrhagia. An ultrasound was performed and resulted in negative. We started a high-dose steroid therapy with intravenous pulses of methylprednisolone at the dose of 10 mg/kg for 3 days, then at the dose of 2–3 mg/kg/day. Despite this, the child gradually developed a severe anaemia, even though no more episodes of macroscopic bleeding occurred, there were no signs of peritoneal irritation and stools were normocolic. A further abdominal ultrasound was negative and other causes of anaemia were excluded. For the persistence of a severe anaemia (Hb 64 g/L), a blood transfusion was required. In the suspicion of obscure gastrointestinal bleeding from the small bowel, a scintigraphy with marked red blood cells (RBCs) was performed. The test showed a small uptake in the left abdominal area, of intestinal origin, integrated and partly confirmed by two consecutive abdominal ultrasounds. Both of them showed a thickening of some loops and a small ileoileal intussusception which then resolved spontaneously.

Along with intestinal involvement, other organs and systems were affected. During hospitalisation, more urine tests revealed microhematuria, with 30–50 RBCs per high-power microscopic field in urinary sediment and proteinuria, with a urine protein-to-creatinine (UPr/Cr) ratio of 0.7. Other laboratory data, such as total proteins, albumin, immunoglobulins, electrolytes, as well as renal function, were normal. We monitored it through multiple urinalysis and urine tests for UPr/Cr ratio, with a slow disappearance of proteinuria in the weeks following the start of steroid therapy and with persistence of microhematuria even after discharge and at subsequent outpatient checks in the paediatric nephrology clinic, where she is still followed periodically. Moreover, there was the suspicion of a central nervous system involvement, since the patient started reporting intense and persistent drowsiness and sleepiness. No sensory or motor disorders were observed at the clinical examination. For this reason, in order to exclude a vascular involvement of the brain, an MRI was also performed, which was negative, and the symptoms resolved after the administration of steroid therapy.

Differential diagnosis

For its clinical features, a proper diagnosis can be really challenging for the clinician, since AHOI may mimic other medical conditions, such as urticaria, erythema multiforme, idiopathic thrombocytopenia, meningococcemia, Kawasaki disease and Henoch-Schönlein purpura (HSP).1

Indeed, in relation to the clinical scenario reported above, the main differential diagnosis should be made with HSP. Several are the differences with HSP in terms of clinic and prognosis: 80% of the cases occur in children aged 6–24 months (with a median age of 12 months),2 while the incidence of HSP approximately ranges 4–6 years of age.3 Pathogenesis is poorly understood but a history of recent infection, as in our case (M. pneumoniae and Streptococcus pyogenes), drug administration or immunisation is usually reported.4 Classically AHOI is characterised by the triad: fever, purpura and oedema, which appear quite suddenly and without other symptoms in most of the cases. In particular, skin lesions could have heterogeneous features: they could be painful or asymptomatic, appearing like red to purpuric petechiae that rapidly evolve into large, round, ecchymoses or plaques often with a target-like aspect, together with oedema of the distal extremities, ears and face.4 5 Whereas in patients diagnosed with HSP skin lesions are limited to lower limbs and buttocks, in AHOI body surface can be affected without a favoured site, although it is predominantly over the cheeks, ears and extremities, with relative sparing of the trunk.6 From a histopathological point of view, both these two entities show a leucocytoclastic vasculitis, but a substantial difference regards skin biopsy specimens: in fact, in those patients diagnosed with AHOI in whom a skin biopsy is performed, no IgA deposits are detected.7 One more important difference between AHOI and HSP regards systemic complications: differently from HSP, where organ involvement is considered one of the diagnostic criteria, in AHOI extracutaneous manifestations have been described but are rare, occurring in less than 5% of patients.5

Outcome and follow-up

Although with a slow course, the clinical picture progressively improved, with almost complete disappearance of the skin lesions and oedemas (figures 4 and 5), absence of abdominal pain, normalisation of general clinical conditions and laboratory tests, with the exception of the persistence of microhematuria. The total hospitalisation lasted 50 days, after which the patient was discharged in good clinical condition with the indication to continue the steroid weaning with prednisone at the dose of 1 mg/kg/day and then stopped according to tapering schedule. The subsequent follow-up lasted 6 months during which skin lesions completely disappeared and serial urine tests were performed to monitor the trend of microhematuria. For the persistence of this laboratory data, we recommended an outpatient follow-up at the paediatric nephrology clinic of our hospital, where she is still followed through periodic urine tests.