Vagus Nerve Stimulation Paired With Upper-Limb Rehabilitation After Stroke: 2- and 3-Year Follow-up From the Pilot Study.

Objective: To assess whether a long-term home-based intervention using Paired VNS therapy is feasible and whether the benefits of Paired VNS therapy are maintained beyond 1 year.

Design: A long-term follow-up study.

Setting: Three centers in the United States and 1 in the United Kingdom.

Participants: Adults with chronic ischemic stroke (n=15) with moderate to severe arm and hand impairment.

Interventions: Participants were implanted with a VNS device followed by 6 weeks of in-clinic therapy with Paired (Active) or control VNS followed by home-based rehabilitation through day 90 (blinded phase). The control VNS group then crossed over to receive 6 weeks of in-clinic Active VNS. Participants in both groups then continued a long-term home exercise program with self-administered Active VNS.

Main Outcome Measures: Fugl-Meyer Assessment for Upper Extremity (FMA-UE) and Wolf Motor Function Test (WMFT) Functional scores were evaluated at the end of in-clinic therapy and day 90. Since both groups were subsequently receiving home-based rehabilitation with Active VNS during the long term, follow-up outcome assessments were pooled for the analyses at 6, 9, and 12 months, as previously reported. Here, we report pooled analysis of outcomes beyond 1 year.

Results: One year after Paired VNS therapy, FMA-UE improved by an average of 9.2±8.2 points, as previously reported. Overall, the 2- and 3-year FMA-UE gain from baseline was 11.4±8.7 (P<.001) and 14.8±10.2 points (P<.001), respectively. At years 2 and 3, FMA-UE improved by an additional 2.9 (P=.03 for change vs year 1, n=14) and 4.7 (P=.02 for change vs year 1, n=14) points, respectively. At year 1, 73% (11/15) of participants were responders (FMA-UE change ≥6) and by year 3, 85.7% (12/14) were responders. At years 2 and 3, the WMFT score improved by an additional 0.21 points (P=.03 for change vs year 1, n=15) and 0.42 points (P=.01 for change vs year 1, n=13), respectively. Responder rate (WMFT change ≥0.4) was 46.6% (7/15), 73.3% (11/15), and 69.2% (9/13) at years 1, 2, and 3, respectively. Long-term significant improvements were also observed for Motor Activity Log (MAL) and Stroke Impact Scale, Hand section (SIS-Hand). There were no serious long-term adverse events from the stimulation.

Conclusions: Significant effects of Paired VNS therapy at 1 year were maintained at years 2 and 3, and further improvements in both impairment and function were observed in years 2 and 3. These changes were associated with improvements in measures of activity and participation.