Results of laparoscopic adrenalectomy for large and potentially malignant tumors

Endoscopic adrenalectomy is the procedure of choice for patients with small functioning adrenal tumors. For most surgeons invasive adrenal carcinoma is an absolute contraindication for laparoscopic adrenalectomy (LA). Whether LA should be proposed for large (> 6 cm), potentially malignant tumors is questionable. The aim of this study was to evaluate the risks and outcome of LA performed in our department in patients with tumors > 6 cm and potentially malignant. We performed a retrospective study of 216 patients who underwent 233 LAs in our department from 1994 to 2000. We selected 19 patients with a tumor > 6 cm and potentially malignant: 8 nonfunctional tumors, 4 cortisol-secreting tumors, 1 virilizing tumor, and 6 pheochromocytomas. In none of these patients did preoperative investigations demonstrated invasive carcinoma. The median tumor size was 70 mm. LA was performed by a transperitoneal flank approach. Conversion to open adrenalectomy was performed in two patients owing to intraoperative evidence of invasive carcinoma. The median operating time was 150 minutes (range 95-240 minutes). Capsular disruption occurred during the dissection of two pheochromocytomas. There was no postoperative morbidity. Six patients had an adrenocortical carcinoma on pathologic diagnosis: three of the eight nonfunctional tumors, one of the four cortisol-secreting tumors, and one virilizing tumor. One patient presented with liver metastases 6 months after surgery and died. The five other patients are disease-free with a follow-up ranging from 8 to 83 months. The 13 patients with benign lesions (6 cortical adenomas, 1 ganglioneuroma, 6 pheochromocytomas) are disease-free with a median follow-up of 47 months (range 10-81 months). In experienced hands LA can be proposed for large, potentially malignant tumors. Conversion to open adrenalectomy should be performed if local invasion is observed during surgery. At present the risk of intraabdominal recurrence is unknown.