Lymph node metastases have been proven to be the main prognostic factor in medullary thyroid carcinoma (MTC). This retrospective study was undertaken to evaluate the efficiency of two surgical techniques of regional lymph node dissection with regard to the normalization of pentagastrin-stimulated serum calcitonin level and patient survival: selective lymphadenectomy, i.e., the excision of macroscopically or microscopically involved lymph nodes, versus a systematic lymphadenectomy performed by the new technique of a compartment-oriented microdissection. From 1970 to 1990, 82 patients with sporadic (n=57) and hereditary (n=25) MTC underwent a total of 142 operations including 63 selective lymphadenectomies and, since 1986, 35 systematic lymphadenectomies. The study revealed that in node-positive MTC the rate of interventions with a postoperative normalization of pentagastrin-stimulated serum calcitonin was higher after systematic lymphadenectomy (29.2%) than after selective lymphadenectomy (8.5%) (P<0.01). The rate of patients undergoing repeat surgery due to a recurrence of MTC was 48% after selective lymphadenectomy and 10% after systematic lymphadenectomy. Survival was significantly better for patients after systematic versus selective lymphadenectomy (P<0.005). This study thus emphasizes that systematic lymphadenectomy, using the technique of a compartment-oriented microdissection of cervicomediastinal lymph nodes, represents the preferred surgical treatment as well as the optimum technique in primary as well as secondary node-positive MTC.