We examined the efficacy of percutaneous cervical cordotomy (PCC) and subarachnoid phenol block using fluoroscopy (SAPB-F) for control of chest and/or back pain from costopleural syndrome. The efficacy of each block was evaluated by changes in pain score (PS), analgesic dose and performance status 1 week after the block, as well as by the complications. Between 1980 and 1986, PCC was performed in 10 patients. SAPB-F was performed in 13 patients between 1987 and 1991. Pain was not well controlled by analgesics in any of these patients. For PCC the follow-up period was 94.7 +/- 71.1 days. PS (VAS, 0-10) reduced from 8.5 +/- 0.9 to 3.0 +/- 2.7. No analgesics were needed in 4 patients. Pain recurred in 1 patient. Hemiparesis occurred in 2 patients. General fatigue occurred in 6 patients. In 4 patients with these complications performance status deteriorated and did not recover during the follow-up period. For SAPB-F the follow-up period was 71.8 +/- 44.0 days. SAPB-F was designed to achieve selective phenol deposit at the targeted nerve root. PS decreased from 7.5 +/- 1.9 to 2.7 +/- 2.6. No analgesics were needed in 5 patients. Pain recurred in 3 patients. There were no complications and no changes in performance status. From this study we concluded that PCC is an effective method of pain control for costopleural syndrome, but a risk of serious complications is involved. SAPB-F is an effective and safe method and should be the first choice of nociceptive pathway block.