No abstract

The encapsulation of islets of Langerhans in alginate-poly-l-lysine has been proposed as a method for the immunoprotection of transplanted islets. Although several capsule compositions have been reported, there has been no published study concerning the effect of capsule composition on the severity of the foreign body reaction. Empty capsules were prepared from high mannuronic acid alginate and were coated with: (1) poly-l-lysine alone, (2) poly-l-lysine plus high guluronic acid alginate, or (3) poly-l-lysine plus high mannuronic acid alginate. The capsules were placed in the renal subcapsular space or the peritoneal cavity, and retrieved after three weeks of histological examination. The recipients were WAG/01a, nude (athymic), diabetic BB, and non-diabetes prone BB rats. The severity of reaction to the capsules was determined by measuring the thickness of the pericapsular cell infiltrate or by a scoring system. The severity of the reaction to the capsules was strain-dependent in both the renal and peritoneal sites, with the BB and nude rats displaying the most severe responses. The degree of response was not affected by capsule composition in the renal subcapsular space, but in the peritoneum, the high mannuronic acid alginate capsules provoked the weakest response, and this type of capsule will be used for future transplantation work. The infiltrating cells were characterised by immunohistochemistry and electron microscopy and found to be mostly fibroblasts and macrophages.

No abstract

We have assessed the degree of observer variation of both Breslow depth and Clark's level in a series of 50 thin malignant melanomas. Our findings are similar to those of previous international studies in the Breslow depth is the more reproducible measure. Significant intra- and inter-observer variation exists and in some cases it was up to +/- 0.86 mm. Even small differences will potentially affect patient management at our centre and this was analysed using kappa statistics. Good agreement was found between observers and this could be improved by comparing the mean of two or more measurements. This removes larger errors, but smaller observer errors and differences in subjective interpretation of the deepest malignant cell mean that agreement will never be more than 90 per cent. This is high compared with studies of observer variation in other pathological conditions, e.g., dysplasia of the cervix, but where surgical management is potentially disfiguring it is not high enough. We conclude that Breslow depth and Clark's level should not be the sole basis of wide excision protocols.