Sacral nerve stimulation versus the magnetic sphincter augmentation device for adult faecal incontinence: the SaFaRI RCT

Health Technol Assess. 2021 Mar;25(18):1-96.doi: 10.3310/hta25180.

David G Jayne 1, Annabelle E Williams 2, Neil Corrigan 3, Julie Croft 3, Alison Pullan 3, Vicky Napp 3, Rachel Kelly 3, David Meads 4, Armando Vargas-Palacios 4, Adam Martin 4, Claire Hulme 5, Steven R Brown 6, Karen Nugent 7, Jen Lodge 8, David Protheroe 9, Sushil Maslekar 10, Andrew Clarke 11, Pasha Nisar 12, Julia M Brown 3


Author information

  • 1Academic Surgery, Leeds Institute of Medical Research at St James's, University of Leeds, Leeds, UK.
  • 2Colorectal and General Surgery, Milton Keynes University Hospital, Milton Keynes, UK.
  • 3Clinical Trials Research Unit, Leeds Institute of Clinical Trials Research, University of Leeds, Leeds, UK.
  • 4Academic Unit of Health Economics, Leeds Institute of Health Sciences, University of Leeds, Leeds, UK.
  • 5Health Economics Group, Institute of Health Research, University of Exeter Medical School, Exeter, UK.
  • 6Department of Colorectal Surgery, Northern General Hospital, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK.
  • 7University of Southampton, Southampton, UK.
  • 8Bowel Health and Pelvic Floor Dysfunction Community Urology and Colorectal Service (CUCS), Leeds Community Healthcare NHS Trust, Leeds, UK.
  • 9Department of Liaison Psychiatry, Leeds General Infirmary, Leeds and York Partnership NHS Foundation Trust, Leeds, UK.
  • 10St James's Hospital, Leeds Teaching Hospitals NHS Trust, Leeds, UK.
  • 11Poole Hospital NHS Foundation Trust, Poole, UK.
  • 12St Peter's Hospital, Ashford and St Peter's Hospitals NHS Foundation Trust, Chertsey, UK.


Background: Preliminary studies using the FENIX™ (Torax Medical, Minneapolis, MN, USA) magnetic sphincter augmentation device suggest that it is safe to use for the treatment of adult faecal incontinence, but efficacy data are limited.

Objective: To compare FENIX with sacral nerve stimulation for the treatment of adult faecal incontinence in terms of safety, efficacy, quality of life and cost-effectiveness.

Design, setting and participants: Multicentre, parallel-group, unblinded, randomised trial comparing FENIX with sacral nerve stimulation in participants suffering moderate to severe faecal incontinence.

Interventions: Participants were randomised on an equal basis to either sacral nerve stimulation or FENIX. Follow-up occurred 2 weeks postoperatively and at 6, 12 and 18 months post randomisation.

Main outcome and measure: The primary outcome was success, defined as device in use and ≥ 50% improvement in Cleveland Clinic Incontinence Score at 18 months post randomisation. Secondary outcomes included complication rates, quality of life and cost-effectiveness. Between 30 October 2014 and 23 March 2017, 99 participants were randomised across 18 NHS sites (50 participants to FENIX vs. 49 participants to sacral nerve stimulation). The median time from randomisation to FENIX implantation was 57.0 days (range 4.0-416.0 days), and the median time from randomisation to permanent sacral nerve stimulation was 371.0 days (range 86.0-918.0 days). A total of 45 out of 50 participants underwent FENIX implantation and 29 out of 49 participants continued to permanent sacral nerve stimulation. The following results are reported, excluding participants for whom the corresponding outcome was not evaluable. Overall, there was success for 10 out of 80 (12.5%) participants, with no statistically significant difference between the two groups [FENIX 6/41 (14.6%) participants vs. sacral nerve stimulation 4/39 (10.3%) participants]. At least one postoperative complication was experienced by 33 out of 45 (73.3%) participants in the FENIX group and 9 out of 40 (22.5%) participants in the sacral nerve stimulation group. A total of 15 out of 50 (30%) participants in the FENIX group ultimately had to have their device explanted. Slightly higher costs and quality-adjusted life-years (incremental = £305.50 and 0.005, respectively) were observed in the FENIX group than in the sacral nerve stimulation group. This was reversed over the lifetime horizon (incremental = -£1306 and -0.23 for costs and quality-adjusted life-years, respectively), when sacral nerve stimulation was the optimal option (net monetary benefit = -£3283), with only a 45% chance of FENIX being cost-effective.

Limitations: The SaFaRI study was terminated in 2017, having recruited 99 participants of the target sample size of 350 participants. The study is, therefore, substantially underpowered to detect differences between the treatment groups, with significant uncertainty in the cost-effectiveness analysis.

Conclusions: The SaFaRI study revealed inefficiencies in the treatment pathways for faecal incontinence, particularly for sacral nerve stimulation. The success of both FENIX and sacral nerve stimulation was much lower than previously reported, with high postoperative morbidity in the FENIX group.

Future work: Further research is needed to clarify the treatment pathways for sacral nerve stimulation and to determine its true clinical and cost-effectiveness.

Trial registration: Current Controlled Trials ISRCTN16077538.

Funding: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 25, No. 18. See the NIHR Journals Library website for further project information.

© Copyright 2013-2022 GI Health Foundation. All rights reserved.
This site is maintained as an educational resource for US healthcare providers only. Use of this website is governed by the GIHF terms of use and privacy statement.