The effects of local infiltration anesthesia, brachial plexus blockade, isoflurane, or halothane anesthesia on blood flow through the brachial artery and through a newly created forearm arteriovenous fistula (AVF) were compared in 36 patients with endstage renal failure. Brachial artery blood flow was measured at two different times, before anesthesia and during anesthesia but before surgery, using a pulsed Doppler flowmeter. AVF flows were calculated from brachial, radial, and ulnar blood flows at the end of surgery, 2 h after surgery, and 3 and 10 days after the procedure. Mean arterial pressure was lower in patients receiving isoflurane or halothane than in those receiving local anesthesia or brachial plexus blockade (BPB). There was a significant increase in brachial artery blood flow following BPB (43.7 +/- 18.7 to 186.9 +/- 98.2 ml.min-1) during isoflurane anesthesia (46.2 +/- 15.9 to 153.1 +/- 80.5 ml.min-1) and during halothane anesthesia (49.9 +/- 24.1 to 97.6 +/- 62.1 ml.min-1). During anesthesia, the difference in brachial artery blood flow between patients in the BPB and halothane groups was significant. Local anesthesia failed to increase brachial artery blood flow (44.0 +/- 12.7 to 45.6 +/- 11.3 ml.min-1). In the immediate postoperative period, the AVF blood flow was lower in patients in the halothane group than in the other groups, but this difference was only significant when compared with BPB group.(ABSTRACT TRUNCATED AT 250 WORDS)