Nephrosis: a clinical and histological study of 38 children.

Many attempts have been made to classify renal diseases in general and glomerulonephritis in particular since Richard Bright (1836) reported his classical researches into the clinical and pathological changes in his patients suffering from kidney disease. The word nephrosis was coined by Muller (1905) to define a clinical condition characterized by oedema, albuminuria, and hypoproteinaemia. Volhard and Fahr (1914) classified Bright's disease into three groups, (A) nephrosis, (B) nephritis, and (C) arteriosclerosis. The best known classification in this country is that of Ellis (1942) who divided nephritis into types 1 and 2: its great merit was that it attempted to correlate clinical features with histological changes. In our experience, however, it is often impossible to divide renal lesions into these two types, for there are often transitions from one type to another, and even the correlation between histological features and clinical findings is poor. The attempts by Dodge, Daeschner, Rosenberg, Brennan, Travis, and Hopps (1962) to introduce modifications of the Ellis system only lead to further confusion, especially when comparing their results with those of other workers. We have, therefore, come to the conclusion that the only helpful classification of nephritis is one based on descriptive morphology; in our experience this classification is of considerable help in prognosis, though it might not be of as much assistance in unravelling the aetiology of this group of diseases. It is likely that the kidney responds in much the same way to a variety of insults, but some of the changes may be permanent, some may develop slowly over a period of years, some may be progressive, and yet others may resolve completely. We believe that renal biopsy may be of considerable help in prognosis by providing evidence of the types of changes occurring in the kidney. The patients we have studied have all shown clinical evidence of nephrosis. We apply this term to patients presenting a characteristic group of symptoms, signs, and biochemical features, i.e. oedema, albuminuria, hypoproteinaemia, and hypercholesterolaemia; all are often of insidious onset and vary in degree over a period of years. This clinical view of nephrosis is well described by Kramer, Goldman, and Cason (1952), who add that nephrosis occupies a position shared by few diseases in medicine: its aetiology is unknown, its course unpredictable, its prognosis uncertain, and its treatment unsatisfactory. Over the past decade knowledge of the natural history of nephrosis has increased, due in no small measure to the improved biochemical and histological techniques which have been employed, and treatment has been more satisfactory since corticosteroids became available. It is our purpose in presenting the clinical, biochemical, and histological features of 38 children suffering from nephrosis to suggest that a more accurate diagnosis is possible, a more valid' prognosis can be made, and the response to corticosteroids can be predicted.