• mitral valve insufficiency
• delivery of health care
• health care economics and organisations
• heart failure
• systolic

Transcatheter edge-to-edge repair (TEER) may be considered in inoperable patients with severe mitral regurgitation secondary to left ventricular systolic dysfunction, who remain symptomatic despite optimal heart failure guideline-directed medical therapy including cardiac resynchronisation therapy when appropriate. All current guidelines agree in this regard. The European (ESC/EACTS) and American (ACC/AHA) professional societies guidelines define the recommendation for TEER as class IIa with level of evidence B. The UK National Institute for Heath and Care Excellence (NICE) guidelines recommend ‘consider TEER’, which is the NICE equivalent of a class II indication, by comparison with a class I indication that would have been worded as ‘offer TEER’. The weak recommendation in all guidelines reflects the uncertainty of the clinical evidence, with one randomised control trial (COAPT)1 finding symptomatic and prognostic benefit in heart failure patients with reduced left ventricular systolic function and severe secondary mitral regurgitation and one randomised control trial (MITRA-FR)2 finding no benefit. Several attempts of explaining this discrepancy were made. The generally accepted explanation is that TEER can be of benefit if the mitral regurgitation rather than the left ventricular systolic dysfunction drives the heart failure symptoms and the guidelines recommendation refers strictly to this clinical scenario.

In the UK, the introduction and use in clinical practice of novel technologies or new indications for existent technologies are regulated by NICE Technology Appraisals, NHS England Commissioning through Evaluation (CtE) supported by NICE, NHS England Clinical Commissioning Policy and Clinical Guidelines developed by NICE. Currently, TEER is commissioned by NHS England only for primary mitral regurgitation, based on the CtE study performed3 before the development of the first NICE clinical guidelines on heart valve disease.4 However, the recently published NICE clinical guidelines recommend TEER for secondary mitral regurgitation as well. Furthermore, as part of the development process of the NICE guidelines, a cost-effectiveness analysis complementary to the NHS England CtE was performed. This analysis found TEER for severe secondary mitral regurgitation in heart failure with reduced ejection fraction to have an incremental cost per QALY gained of £30 175 (probabilistic base case) and of £28 488 (deterministic lower cost case). The NICE guidelines recommendations are based on both clinical and cost-effectiveness. However, the slightly above the £30 000 threshold probabilistic base case incremental cost per QALY gained did not prevent a recommendation similar with the ESC/EACTS and ACC/AHA recommendations from being made, based on the clinical effectiveness evidence provided by the COAPT trial. Notably, the ESC/EACTS and ACC/AHA guidelines are based solely on clinical effectiveness, so the current recommendation strength could not be upgraded by demonstrating better cost-effectiveness. However, in this Heart paper, COAPT trial investigators and coauthors report a further assessment of cost-effectiveness for the use of TEER in secondary mitral regurgitation in the UK NHS.5

The COAPT trial investigators have previously published a similar cost-effectiveness analysis from a US healthcare system perspective, demonstrating cost-effectiveness in the US.6 Despite the essential differences of the US healthcare system with the UK NHS, the health outcomes from this analysis can be compared and are comparable with the health outcomes from the NICE guidelines cost-effectiveness model. The US analysis6 estimated an increase in life expectancy of 1.13 years and in QALYs of 0.82, similar with the results of the NICE guideline analysis which found TEER to increase life expectancy by 1.18 years and QALYs by 0.69. Additionally, a previously published cost-effectiveness analysis based on COAPT and conducted from the UK NHS perspective7 reported an incremental cost-effectiveness ratio of £30 057 per QALY gained, which is very close to the NICE guidelines analysis result (£30 175).

The main strength of the current COAPT trial-based cost-effectiveness analysis reported by Cohen et al is the realistic and accurate analysis of quality of life performed.5 The values reported in the text for health outcomes appear higher than the above-mentioned corresponding health values in the US analysis and in the NICE guidelines analysis, with an estimated increase in life expectancy of 1.57 years and in QALYs of 1.12. Nevertheless, these reported values seem to be undiscounted and to correspond to discounted values similar with the ones reported in the US analysis and, indeed, in the NICE guidelines analysis. The use of undisclosed undiscounted values in the text may be misleading at first glance for the cardiologist reader of Heart, however not for the health economist. For the actual analysis, the authors discounted the QALYs at 3.5% obtaining the same estimated increase in QALYs of 0.82 as in the US analysis.

Cohen et al aimed to assess cost-effectiveness of TEER in a COAPT-like population from a UK NHS perspective, yet the principles followed resulted in deviation from this aim. Whereas UK NHS health economic evaluations follow the NICE standards and principles, the results of this study were reported following the consolidated health economic evaluation reporting standards of the International Society for Pharmacoeconomics and Outcomes Research.8 Furthermore, while health economics are sensitive and specific to a healthcare system, the economic analysis was approved by the institutional review board of an American hospital rather than of a UK NHS hospital, despite having UK NHS hospital-based cardiologists as co-authors.

The analysis performed by Cohen et al reports a very low incremental cost-effectiveness ratio, considerably lower than the incremental cost-effectiveness ratio reported by Shore et al 7 and by the NICE guidelines. 4 This low incremental cost-effectiveness ratio is the result of using costs that seemed to have been chosen from the lowest available estimates, without providing any explanation or rationale, despite referring to a COAPT-like population with heart failure, advanced age and multiple comorbidities (table 1). Does this approach provide confirmation of cost-effectiveness or does it blur the waters?

The TEER device belongs to the high-cost tariff device category, paid for separately, not included in the cost of the intervention to avoid skewing the average intervention costs of the Healthcare Resource Group (HRG). The cost of high-cost tariff devices used in health economic evaluations is the supply chain cost, which for the TEER device is £16 500. This supply chain cost was used in the Shore et al cost-effectiveness analysis,7 however, the Cohen et al analysis used a lower device cost (£16 218) of undisclosed provenience.

The cost of the intervention used is curtailed even more substantially. The authors do not appear to have consulted the two UK micro-cost analyses available to verify the appropriateness of the costs they collected from national sources. These UK micro-cost analyses are the NHS England CtE3 and the Shore et al cost-effectiveness analysis.7 Cohen et al used the English tariffs and the Payment by Result (PbR) tariffs, not appropriate for a cost-effectiveness analysis performed from a UK NHS perspective. The PbR tariffs refer to the costs incurred by a certain healthcare provider for a procedure, rather than to the cost incurred by the UK NHS to make the respective procedure available to patients, cost which can be higher than the respective PbR tariff. Furthermore, from the values reported, Cohen et al seem to have selected the lowest available cost, corresponding to the lowest comorbidity and complication (CC) level. It should be noted that the HRG code (EY22) that was used to estimate the cost of the procedure represents several transcatheter procedures of variable complexity; the difference in cost in between these procedures within the HRG is only determined by the CC score of the patient category addressed. For example, balloon valvotomy for rheumatic mitral stenosis is included under the same HRG code; this well-established procedure can be described by a low CC score corresponding to low cost, as it has low level of complications and it usually addresses young individuals with normal left ventricular systolic function and no or low number of comorbidities. The baseline characteristics of the COAPT population and the reported TEER outcomes in these patients cannot be described by the same low CC score.

The preferred source for NHS-related costs is the NHS Reference Costs,9 the source used by NHS England for the cost of the TEER procedure in the commissioning policy. Furthermore, patients with heart failure with reduced ejection fraction and severe secondary mitral regurgitation are better represented by the highest CC score that increases the cost of the procedure and explains the higher cost estimated in Shore et al and in the NHS England CtE. Using NHS Reference Costs and the highest CC score, the cost of the procedure excluding the device would have been estimated as £12 319. This cost would have been similar with the CtE micro-cost analysis derived central estimation of £12 760, which was also used in the health economic model that informed the NICE guidelines. Furthermore, this cost would have compared relatively well also with the estimated cost of the procedure of £11 208 from the micro-cost analysis published by Shore et al.7

Therefore, at current device cost, in the UK NHS, the incremental cost per QALY gained for TEER in secondary mitral regurgitation remains at levels reported by NICE. Confirmation of cost-effectiveness is not needed, as only improvement in clinical effectiveness evidence could change the current recommendations strength.