Abstract

A series of 93 rectal biopsies performed for diagnosis of suspected progressive neurometabolic disease between 1967 and 1973 is reviewed, and the results of this and of two previously published series totalling 307 biopsies are analysed.

In G_M1 and G_M2 gangliosidosis, Batten's disease, and certain other neuronal storage diseases rectal biopsy is a reliable diagnostic alternative to brain biopsy. However, the need for biopsy has diminished with improvement in other diagnostic methods, particularly enzyme assay, the availability of which should determine the extent to which biopsy is used. It is suggested that rectal biopsy is necessary in the various forms of Batten's disease and in the neurovisceral storage disease with supranuclear ophthalmoplegia described by Neville et al. (1973). In certain diseases its use is unjustified, either because the result would be negative or because other less traumatic, reliable investigations are available. On rare occasions it is justifiable to use rectal biopsy either as an `excluding investigation' to exclude Batten's disease for certain in a healthy sib of a known case with this disorder or to detect the disease before onset of clinical symptoms.

The need for a full thickness biopsy and for a full range of staining methods is emphasized. Without these the investigation cannot be expected to give diagnostic information and may be misleading, giving `false negative' results.