Cost effectiveness studies of tremor treatment should not focus on ultrasound while neglecting radiofrequency lesioning

To the Editor, We read with great interest the paper by Jameel and colleagues entitled “The cost-effectiveness of unilateral magnetic resonance-guided focused ultrasound in comparison to unilateral deep brain stimulation for the treatment of medically refractory essential tremor in England.” published recently in BJR.1 The authors employed an elegant statistical model and suggested an estimated cost per case of magnetic resonance-guided focused ultrasound (MRgFUS) lower than deep brain stimulation. This was based on a longer usage time period of half a decade and expert opinion alone of high volume usage shown in supplementary tables. No explicit mention is made in the main paper of the capital procurement cost of an MRgFUS machine which to our knowledge is approximately £2,000,000. Two elephants also remain in the room. Firstly, no mention is made of the cheap alternative procedure of radiofrequency ablation (RFA) or of gamma knife radiosurgery. Secondly, unilateral DBS is used as the comparator, whereas the vast majority of patients with tremor have bilateral symptoms best treated by bilateral DBS. The cost of an RFA machine is £20,000 (1% of an MRgFUS machine) and most hospitals already own one for spinal pain procedures. Operative time of RFA is shorter at 1 h vs 4–6 for MRgFUS, the patient also remains awake, there is no need for hair shave and no costly MRI scanner time is required during the procedure. Risks and efficacy are similar. That MRgFUS incisionless is heavily marketed, but its good efficacy and low risks are similar to RFA and unilateral DBS.2 Importantly, bilateral DBS can be modulated to optimise bilateral tremor control and minimise side-effects upon speech and balance, whereas lesioning is irreversible and least risky if unilateral. Furthermore, much is made by the authors of patients unsuitable for neurosurgery. There are hardly any absolute contraindications to DBS for essential tremor. Age, anticoagulation, implant infection risk, suitability for general anaesthesia and mildness of tremor are all relative contraindications with most of this tiny subgroup remaining as eligible for RFA as MRgFUS. In conclusion, despite the growing popularity of MRgFUS in treating people with tremor syndromes, we advocate DBS as most efficacious for bilateral essential tremor and RFA as most cost effective in those with unilateral tremor who might prefer not to have an implant and those who have a genuine contraindication to neuroprosthesis implantation. The movement disorders neurology and functional neurosurgery communities should raise awareness of and compare all lesioning modalities, in particular the established safe, efficacious and cheap treatment of RFA. 1,2,3Ayesha Jameel, 4Anne Meiwald, 3Peter Bain, 5Neekhil Patel, 5Dipankar Nandi, 2Brynmor Jones, 4Georgie Weston, 4Elisabeth J Adams and 2Wladyslaw Gedroyc 1Department of Surgery and Cancer, Imperial College, London, United Kingdom 2Department of Radiology, Imperial College Healthcare NHS Trust, London, United Kingdom 3Division of Brain Sciences, Imperial College London, London, United Kingdom 4UK Aquarius Population Health Limited, London, United Kingdom 5Department of Neurosciences, Imperial College Healthcare NHS Trust, London, United Kingdom Response to a Letter to the Editor Dear Editor, We thank the authors for their letter on our paper “The cost-effectiveness of unilateral magnetic resonance-guided focused ultrasound (MRgFUS) in comparison with unilateral deep brain stimulation (DBS) for the treatment of medically refractory essential tremor in England.” We explored a specific comparison (as is typical of economic models of this type) and we agree that there are additional considerations that merit further research and analysis in clinical decision making. We thank the authors for their complimentary comments on our economic modelling. The Markov model and statistical analyses utilised are standard for such health economic studies. The model was based on the 2019 Canadian cost-effectiveness study,1 and adapted for the perspective of NHS England. The robustness of data inputs (including probabilities, utilities, disutilities, costs etc.) and set model parameters (age, time horizon, patient care, discounting etc.) were explored through the uncertainty analyses (deterministic sensitivity analyses, probabilistic sensitivity analyses, scenario analyses) as described in the paper and in further detail in the appendices for interested parties. We would like to address specific points raised: The 5-year time horizon in the base case is favourable to DBS, as no costing for battery replacement is included. A longer timeframe would significantly increase the cost of DBS, as demonstrated in the scenario analyses (Table S10; Appendix 3). For both DBS and MRgFUS costs were based on NHS reimbursement which consider the capital costs of equipment, including those for MRgFUS (ExAblate 4000 system) and for DBS (neuronavigational and theatre equipment). Cost inputs are summarised in Table 1 and detailed costing for both DBS and MRgFUS in Appendix 3 (Table S3-7). Of note, this cost-effectiveness study was performed from the perspective of NHS England. Population modelling and treatment eligibility is complex. The model approach is described in Table 1 and detail in Appendix 3 (Table S1). Population parameters were based on the Office for National statistics. Scenario analyses considered the population inputs and eligibility for DBS inputs used (Tables 4 and 5). MRgFUS is economically dominant in all scenarios where more than 30% of population is eligible for DBS. This includes the scenario where there is 100% eligibility for DBS (Table 5). In clinical practice, consideration must be given to the safety of the specific procedure and the patients desire both for and against the various treatment options available. We discussed this in detail with NHS England. There are further established treatment options for tremor, beyond DBS and MRgFUS, which include radiofrequency ablation (RFA) and gamma knife (GK) thalamotomy. Several previous cost-effectiveness studies have included radiofrequency and/or radiosurgery in their analyses,1- 3 with similar results. In a clinical context (taking into account individual patient choice), it is important to be mindful that these further methods of thalamic lesioning have certain comparative drawbacks. MRgFUS is a non-invasive, non-ionising treatment that provides real-time clinical feedback allowing intra-procedural tailoring of treatment to patient response. RFA is an invasive procedure with associated risks including, although small, cognitive decline and intracerebral haemorrhage.2 GK although non-invasive, employs indirect anatomical targeting of ionising radiation and does not permit real-time clinical feedback.2 We encourage the authors and other centers that regularly perform RF and/or GK for the treatment of essential tremor to conduct cost-effectiveness analyses, using their own contemporary data. We agree there is a role for collaborative multi-centre studies, utilising individual centres expertise to explore different comparisons between MRgFUS (unilateral and bilateral), DBS (unilateral and bilateral), RFA and GK. International experience in bilateral MRgFUS treatments is growing3–5 and with time there will be a critical mass of outcome data suitable for health economic analysis. We look forward to seeing the results of such a comparative analysis. Yours Sincerely, Dr Ayesha Jameel, Anne Meiwald, Dr Peter Bain, Mr Neekhil Patel, Prof Dipankar Nandi, Dr Brynmor Jones, Georgie Weston, Dr Elisabeth J Adams and Prof Wladyslaw Gedroyc

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.