[Contribution of cytological punctures to the diagnosis of kidney tumors].

Analysis concerned 47 kidney tumours. Puncture was under ultrasonographic control using 22 G needles. The sample was obtained by suction. The authors then describe the technique of fixing and reading of smears. Four types of indications were adopted; diagnostic uncertainty between benign and malignant nature (22 cases), high risk surgical patients in whom a pre-operative histological diagnosis is very desirable (10 cases), search for a primary neoplasm in the presence of metastatic spread (5 cases), routine puncture of tumours seen by ultrasonography (10 cases). Puncture material was uninterpretable in 6 cases. In 26 cases of puncture-cytology could be compared with those of histological examination (nephrectomy or surgical biopsy). 3 puncture-cytology results were un-interpretable, whilst in 21 cases the correlation was perfect (17 malignant tumours and 4 benign tumours). In 1 case puncture-cytology led to suspicion of the malignant nature of the lesion. Puncture cytology was erroneous in one case (a malignant tumour confused with an angiomyolipoma). In 21 cases no sample was obtained and it was not possible to establish the correlation between cytological and histological results. Amongst these 21 cases 3 were uninterpretable whilst in 18 there was no evidence of error on the basis of the subsequent clinical course. No complications were seen and in particular no abdominal wall dissemination. Sensitivity of the method was of the order 80 to 87%. Interpretation of puncture-cytology is difficult in urothelial carcinomas, renal adenoma and oncocytoma. By contrast puncture-cytology is particularly reliable in angiomyolipoma and metastatic lesions of the kidney.