This study examined the effect of the aldose reductase inhibitor (ARI), ponalrestat, on decreased motor nerve conduction velocity (MNCV) and increased resistance to hypoxic conduction block (RHCB) in diabetic rats. The effects of 5 mmol/L, and 25 mmol/L glucose on RHCB were also determined. Twenty streptozotocin-diabetic rats formed two groups; untreated and ponalrestat-treated (300 mg/kg diet/day); 10 non-diabetic rats acted as controls. MNCV was measured in vivo after 4 weeks of diabetes +/- treatment in the sciatic/tibialis system and rats were killed 48-72 h later. The median nerves were removed and assayed for polyol pathway metabolites by gas chromatography. The sciatic nerves were dissected to form endoneurial preparations for the recording of compound action potentials (CAPs) in vitro and maintained in media with either 5 (standard) or 25 (high) mmol/L glucose and initially gassed with 95% O2/5% CO2. Oxygen content was then reduced to 8% for 40 min to study the effect of this period of hypoxia on CAP amplitude. MNCV (m/s +/- SD) in diabetic rats (43.86 +/- 4.86) was decreased compared to controls (52.24 +/- 6.59) and this decrease was absent in the ARI-treated group (52.24 +/- 6.90). The decline in CAP amplitude during a 40-min hypoxic period was greater in controls than in diabetics. Ponalrestat treatment caused a decline which was mid-way between these two in standard medium and closer to that seen in control preparations in high glucose medium. These findings give further support to the involvement of the sorbitol pathway in the development of the acute MNCV deficit in diabetic rats and indicate that it may have a partial role in the development of increased resistance to hypoxic conduction block in peripheral nerves.