[Lymphography and lymph node excision in bladder cancer. Apropos of a series]

The authors reviewed a series of 100 consecutive total cystectomies for carcinoma of the bladder, 78 with bilateral pelvic lymph node dissection and 43 with preoperative lymphography, attempting to determine the reliability of lymphography, to quantify the risk of lymphadenectomy and to assess the usefulness of lymphadenectomy when the nodes are involved. Routine reading of lymphography was associated with a level of accuracy of 68.7% an error rate of 10.8% and a level of uncertainty of 20.5%. Evaluating the results by number of patients rather than by the number of sides opacified by lymphography, the score was even lower: accuracy 60.5%, error 13.2%, doubt 26.3%. Strictly reviewing those cases where all necessary information was present in the record, initial routine reading of lymphographies gave 58% accuracy, 14% error, and 28% doubtful. Retrospective reading by one radiologist only gave much better results, but these did not correspond with the results obtained in everyday routine work. The number of complications related to lymphadenectomy appeared slight: amongst 78 operations, one lymphocoele, two cases of prolonged lymphatic discharge and three cases of oedema of the lower limbs. Of 14 patients with involved nodes, only three are living without recurrence and with follow-up periods which are as yet too short. All patients with involvement extending above the common iliac bifurcation died within a short time, as did all those with prostatic involvement. Comparing the data from their own series and the literature, the authors came to the following conclusions.(ABSTRACT TRUNCATED AT 250 WORDS)