Herdf�rmige tumorartige Bindegewebshyperplasien des Magens

Fibroid polyps of intestinal tract are inflammatory-reactive proliferations of CD34-positive perivascular cells

AIMS

Our aim was to determine the histogenesis of fibroid polyps (FP). These polyps are rare inflammatory-reactive, tumour-like lesions of unknown aetiology, arising in the submucosa or mucosa of the gastrointestinal tract. They are mainly due to a proliferation of characteristic spindle cells.

METHODS AND RESULTS

Nine FP were investigated by light microscopy and immunohistochemistry with endothelial markers (Factor VIII, CD34, CD31), a neuronal marker (S100), muscular markers (desmin, alpha-smooth muscle actin) and histiocytic markers (PGMI, KP1, MAC387) using the highly sensitive avidin-biotin-peroxidase technique. We demonstrate, for the first time, a consistent positivity of the characteristic spindle cells of FP for CD34. The proposed endothelial, histiocytic or neuronal origin of FP could be completely ruled out.

CONCLUSIONS

Because of the consistent positivity of the spindle cells of FP for CD34 we suggest an origin of these lesions from primitive perivascular or vascular cells. This origin and a probable relationship to gastrointestinal stromal tumours (GIST) is discussed.

Granuloblastomas of the stomach (so-called eosinophilic granulomas)-- a variant of fibrous histiocytomas?

25 cases of focal connective tissue proliferations in the submucosa of the stomach are presented. These lesions are termed "granuloblastomas" and have many features in common with so-called eosinophilic granulomas of the stomach. We found that granuloblastomas may be subdivided into two groups: (1) 8 cases are considered to be the result of the proliferation of a peculiar granulation tissue with abundant eosionophilic granulocytes, and (2) 17 cases show a more or less marked storiform pattern and the cellularity is constituted by fibroblasts and histiocytes as well as differing amounts of eosinophilic granulocytes. After discussing the concept of "fibrous histiocytomas" it is concluded that at least the second group of granuloblastomas may be interpreted as a pseudotumorous variant of fibrous histiocytomas. It remains to be clarified in future if submucosal neurofibromas can show the histological features of lesions which we designate granuloblastomas.