A kidney biopsy is clearly mandatory to confirm the indication of plasma exchanges in adult haemolytic uraemic syndrome.

In case of haemolytic uraemic syndrome, it is not always possible to identify on a pure clinical basis the different kidney lesions responsible for the syndrome. We report a series of six cases without thrombotic microangiopathy, which emphasizes the need to perform a kidney biopsy as early as possible, so as to confirm the actual usefulness of plasma exchanges (PE) commonly carried out in emergency in every case of adult haemolytic uraemic syndrome. PATIENTS AND METHODS--Files of patients who were treated for haemolytic uraemic syndrome over the past 14 years were reviewed. Patients in whom thrombotic microangiopathy had been excluded by renal histology data were studied. Every patient was promptly treated with hypotensive drugs, so as to obtain blood pressure levels not exceeding 160-90 mmHg. Dialysis was performed in two patients. Daily PE with fresh frozen plasma were carried out in three patients as early as the first 24 hours after admission, and discontinued immediately after thrombotic microangiopathy could be excluded. RESULTS--All the patients met the usual criteria for diagnosis of haemolytic uraemic syndrome. Elevated liver enzymes were also found in the four cases of preeclampsia, consisting with diagnosis of severe HELLP syndrome. One case was associated with oestrogen therapy. Glomerular lesions were seen in four patients: slight endotheliosis in three cases of preeclampsia; marked lesions of IgA mesangial deposits in the patient who had been treated by contraceptive pill. Three patients had acute tubular necrosis and three had intense lesions of nephrosclerosis. Complete remission was obtained in every case of preeclampsia. Renal failure persisted in two cases (IgA glomerulopathy and one case of nephrosclerosis). DISCUSSION--The histological heterogeneity of haemolytic uraemic syndrome has been already well demonstrated. Typical lesions of thrombotic microangiopathy are usually classified into predominant glomerular lesions, pure arteriolar and mixed lesions. In other cases, thrombotic microangiopathy is not found: kidney lesions may be glomerular (endotheliosis, various subtypes of glomerulonephritis), tubular (acute tubular necrosis) or vascular (nephroangiosclerosis). In every aetiological circumstance, several different lesions may be found together. The usefulness of PE has been proved in thrombotic thrombocytopenic purpura, has been suggested in haemolytic uraemic syndrome and to a lesser extent in persistently severe HELLP syndrome. Unfortunately, none of these reports gave any information about kidney lesions responsible of acute renal failure. CONCLUSION--The haemolytic uraemic syndrome is a syndrome: thrombotic microangiopathy has to be proven when treatment by PE is planned, except in some severe clinical circumstances.