The effects of the aldose-reductase inhibitor, tolrestat, on chronic symptomatic diabetic sensorimotor neuropathy were studied during a placebo-controlled, randomised, 52-week multicentre trial. Of the four tolrestat doses investigated, only the highest dose group, 200 mg once daily, showed subjective and objective benefit over baseline and placebo, and further analyses are confined to this group (n = 112) and placebo (n = 107). Painful and paraesthetic symptoms were analysed separately: improvement in paraesthetic symptoms were seen at one year (p = 0.04), though painful symptoms improved on both placebo and active therapies. Significant improvement in both tibial and peroneal motor nerve conduction velocities were seen at 52 weeks. Tolrestat 200 mg once daily was significantly better than placebo in producing concordant improvements in both motor nerve conduction velocities and paraesthetic symptom scores at 24 weeks (p = 0.01), 42 weeks (p = 0.01) and 52 weeks (p = 0.02). Long-term benefit [concordant improvement at 24 weeks maintained until 52 weeks] was seen in 28% of treated patients compared to 5% on placebo (p = 0.001). It is concluded that some sustained improvement in symptomatic diabetic neuropathy may be obtained following aldose-reductase inhibition with tolrestat 200 mg once daily.