BACKGROUND

The NECTAR-HF study evaluated safety and feasibility of vagal nerve stimulation (VNS) for the treatment of heart failure patients. The first six-month randomized phase of the study did not show improvement in left ventricular remodelling in response to VNS. This study reports the 18-month results and provides novel findings aiming to understand the lack of efficacy of VNS, including a new technique assessing the effects of VNS.

METHODS

Ninety-six patients were randomized 2:1 to active or inactive VNS for 6months, thereafter VNS was activated for all patients. The primary safety endpoint was 18-month all-cause mortality.

RESULTS

Ninety-one patients continued in the long-term evaluation with active VNS. The on-therapy survival estimate at 18months was 95% with a 95% one-sided lower confidence limit of 91%, (better than the predefined criterion). Left ventricular systolic volume decreased in the crossover group (VNS OFF→ON; 144±37 to 139±40, p<0.05) after VNS activation; LVESD (5.02±0.77 to 4.96±0.82, p>0.05) and LVEF (33.2±4.9 to 33.3±6.5, p>0.05) did not change. A new technique to detect subtle heart rate changes during Holter recordings, i.e. "heat maps", revealed that VNS evoked heart rate response in only 13/106 studies (12%) at 6 and 12months with active VNS.

CONCLUSIONS

Although a favourable long-term safety profile was found, improvements in the efficacy endpoints were not seen with VNS. A new technique for detecting acute heart rate responses to VNS suggests that the recruitment of nerve fibres responsible for heart rate changes were substantially lower in NECTAR-HF than in pre-clinical models.