- Email: [email protected]
Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan
G Model
JJCC-1572; No. of Pages 7 Journal of Cardiology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Cardiology journal homepage: www.elsevier.com/locate/jjcc
Original article
Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan Satoshi Kodera (MD)*, Arihiro Kiyosue (MD, PhD, FJCC), Jiro Ando (MD), Issei Komuro (MD, PhD, FJCC) Department of Cardiovascular Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 August 2017 Received in revised form 26 September 2017 Accepted 10 October 2017 Available online xxx
Background: Transcatheter aortic valve implantation (TAVI) is a less invasive treatment for elderly patients with aortic stenosis. However, the cost of TAVI is a major issue. This study analyzed the cost effectiveness of TAVI in Japan. Methods: We developed an economic model to evaluate the quality-adjusted life years (QALYs) and costs of TAVI, surgical aortic valve replacement (SAVR), and medical therapy over a 10-year time horizon from the perspective of Japanese public healthcare payers. The first model compared transapical or transfemoral TAVI with Sapien valve implantation and medical therapy in inoperable patients. The second model compared transfemoral TAVI with Sapien XT valve implantation and SAVR in operable patients with intermediate surgical risk. We assumed a cost-effectiveness threshold of 5,000,000 yen per QALY, and assessed the cost-effectiveness probability with 100,000 simulations. We performed a broad sensitivity analysis to assess the effect of uncertainty on our results. Results: Among inoperable patients, the incremental cost-effectiveness ratio for TAVI compared with medical therapy was 3,918,808 yen per QALY. In operable patients, the incremental cost-effectiveness ratio for TAVI compared with SAVR was 7,523,821 yen per QALY. The cost-effectiveness probability of TAVI was 60% for inoperable patients and 46% for operable patients. Among inoperable patients, the costeffective threshold of TAVI was <7,759,085 yen. Among operable patients, the cost-effective threshold of TAVI was <5,427,439 yen. Conclusions: This study suggests that TAVI has good cost effectiveness for inoperable patients, but not for operable patients. © 2017 Published by Elsevier Ltd on behalf of Japanese College of Cardiology.
Keywords: Transcatheter aortic valve implantation Incremental cost-effectiveness ratio Cost-effectiveness Quality-adjusted life year Aortic stenosis
Introduction Transcatheter aortic valve implantation (TAVI) was developed for the treatment of severe aortic stenosis (AS) [1]. TAVI is a less invasive treatment than surgical aortic valve replacement, and is an effective procedure for elderly and high-risk patients [1– 3]. However, the cost of TAVI is a major issue, with the Sapien valve (Edwards Lifesciences Co., Irvine, CA, USA) costing 4,530,000 yen in Japan. Cost-effectiveness analysis is a research method applied to estimate the incremental benefit and cost of a new treatment compared with the standard treatment [4]. A number of reports in
* Corresponding author at: Department of Cardiovascular Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 1138655, Japan. E-mail address: [email protected] (S. Kodera).
Western countries have examined the cost effectiveness of TAVI [5–11]. However, their conclusions varied, with some studies reporting TAVI to be cost-effective, while others did not. Healthcare systems differ between countries, as does the cost effectiveness of TAVI. However, no cost-effectiveness analysis of TAVI has been conducted in Japan. Considering the low economic growth and burgeoning elderly population in Japan, health economic issues are increasingly important. Thus, in the present study we analyzed the cost effectiveness of TAVI in Japan. Methods We developed an economic model to evaluate the costs and effectiveness of TAVI. The model evaluated the quality-adjusted life years (QALYs) and cost of three treatment options—TAVI, surgical aortic valve replacement (SAVR), and medical therapy—during a 10-year time horizon from the perspective of the Japanese public
https://doi.org/10.1016/j.jjcc.2017.10.008 0914-5087/© 2017 Published by Elsevier Ltd on behalf of Japanese College of Cardiology.
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
2
healthcare payer. We constructed two models. The first model compared transapical (TA) or transfemoral (TF) TAVI with a Sapien valve and medical therapy in inoperable patients. We defined an inoperable patient as a patient with a 30-day mortality and morbidity rate of >50% according to one of the Placement of Aortic Transcatheter Valves (PARTNER) trials [1], which are large randomized controlled trials of AS. The second model compared TF TAVI with a Sapien XT valve and SAVR in operable patients with intermediate surgical risk. We defined an operable patient with intermediate surgical risk as a patient with a 30-day mortality rate of 4%–8% according to another PARTNER trial [3]. We derived various parameters from the PARTNER trials and the Optimized Catheter vAlvular iNtervention (OCEAN) TAVI registry [12,13], which is a Japanese TAVI registry. The analysis was performed according to the Consolidated Health Economic Evaluation Reporting Standards statement and in accordance with a Japanese guideline [4,14]. Model We developed a Markov model with Monte Carlo simulations to evaluate the efficiency of TAVI. The structure of the model is shown in Fig. 1. A monthly cycle was modeled with each cycle, the patient may die, be hospitalized, or become stable. We defined four study phases: study entry, stability, hospitalization, and death. At the entry point, all patients were classified in the study entry phase. If the subsequent 1 month was uneventful, the patients were classified as stable. We combined stroke, myocardial infarction, and vascular complications as hospitalization conditions.
Population The population of patients in each group is shown in Table 1. Our study population reflected that of the PARTNER trials [1,3]. To compare TA or TF TAVI with Sapien valve implantation and medical therapy, we used the PARTNER cohort B, which comprised inoperable patients. The average patient age was 83 years, 46% were men, their average Society of Thoracic Surgeons (STS) score was 11, and 93% had a New York Heart Association class of III or IV [1]. To compare TF TAVI with Sapien XT valve implantation and SAVR, we used the PARTNER 2 cohort A, which comprised patients with an intermediate surgical risk. The average patient age was 82 years, 55% were male, the average STS score was 6, and 77% were New York Heart Association class III or IV [3]. Intervention vs. comparator In the first model, we compared TA or TF TAVI with Sapien valve implantation and medical therapy in inoperable patients. With medical therapy, balloon valvuloplasty was included if necessary. In the second model, we compared TF TAVI with Sapien XT valve implantation and SAVR in operable patients with an intermediate surgical risk. Time horizon We restricted the time horizon to 10 years in both models. The average age of both inoperable and operable patients was approximately 82 years. We considered that a 10-year estimation would be sufficient to evaluate the cost effectiveness of the treatments. In the sensitivity analysis, we changed the time horizon to 3, 5, 10, 15, and 20 years. Mortality and hospitalization The mortality rates associated with each treatment are shown in Table 2. The mortality for all treatments was based on that reported for the PARTNER cohorts and OCEAN TAVI registry [12,13]. For inoperable patients, TAVI was 60% superior to medical therapy in terms of survival. For operable patients, TF TAVI was 10% superior to SAVR in terms of survival. Among patients who underwent TAVI in the OCEAN TAVI registry, the procedural 30-day mortality rate was approximately 2%. The hospitalization rate was based on the PARTNER trial. The reasons for hospitalization included worsening heart failure, stroke, and vascular complications. Among inoperable patients, the hospitalization rate was 50% lower with TAVI than with medical therapy. For operable patients, the hospitalization rate was higher with TAVI than with SAVR. We modified the mortality rate to evaluate the effects of that rate in the sensitivity analysis. Utility The utility value in each situation is shown in Table 2. The PARTNER study determined quality of life (QOL) using the EuroQol Five-Dimensions Questionnaire [5,15]. We used data of the PARTNER study and from other trials to define QOL [16]. In patients with AS before SAVR or TAVI, the QOL was 0.66. In patients with AS after SAVR or TAVI, the QOL was 0.75. We assumed that the QOL of hospitalized patients was 0.48. Following a Japanese guideline, we decreased the utility values by 2% annually [14]. Costs
Fig. 1. Markov model of transcatheter aortic valve implantation. The model applies to each monthly cycle. AS, aortic stenosis; AVR, aortic valve replacement; TAVI, transcatheter aortic valve implantation.
We performed an economic evaluation from the perspective of a public healthcare payer in Japan. The costs included procedural,
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
3
Table 1 Characteristics of the study groups. Inoperable patients [1]
Age (years) Male (%) NYHA function class III–IV (%) STS score (%) Type of valve
Operable patients [3]
TAVI (TA or TF)
Medical therapy
TAVI (TF)
SAVR
83 46 93 11 Sapien valve
83 46 93 11 –
82 55 77 6.0 Sapien XT
82 55 77 6.0 –
NYHA, New York Heart Association; TA, transapical; TAVI, transcatheter aortic valve implantation; TF, transfemoral; SAVR, surgical aortic valve replacement; STS, Society of Thoracic Surgeons.
Table 2 Key parameters used in the models. Inoperable patients TAVI (TA or TF) Perioperative death Perioperative Hospitalization One year mortality One year hospitalization Mortality rate of hospitalization Procedural cost Complication cost Follow-up cost (per month) AS utility Post AS utility Hospitalization utility
Source
Medical therapy
Operable patients Source
TAVI (TF)
Source
SAVR
Source
2.1 0.5% 5.6 1.7%
[12,13] [1]
2.8 1.2% 10.1 2.2%
[1] [1]
2.1 0.5% 5.4 0.8%
[3,12,13] [3,12,13]
4.1 0.7% 6.1 0.8%
[3] [3]
18 10.2% 22.3 10.6% 20 2.8%
[12,13] [1] Expert
49.7 18.0% 44.1 17.4% 20 2.8%
[1] [1] Expert
7.6 3.3% 9.0 3.6% 20 2.8%
[3,12,13] [3,12,13] Expert
8.6 3.4% 8.4 3.4% 20 2.8%
[3] [3] Expert
¥ 6,000,000 3,000,000 ¥ 1,500,000 750,000 ¥ 15,000 7500
[12,13] [16] [16]
¥ 15,000 7500 ¥ 1,500,000 750,000 ¥ 15,000 7500
[16] [16] [16]
¥ 6,000,000 3,000,000 ¥ 1,500,000 750,000 ¥ 15,000 7500
[12,13] [16] [16]
¥ 4,500,000 2,250,000 ¥ 1,500,000 750,000 ¥ 15,000 7500
[17] [16] [16]
0,66 0.2 0.75 0.28 0.48
[5] [5] [4]
0.66 0.2 0.66 0.2 0.48
[5] [5] [4]
0.66 0.2 0.75 0.28 0.48
[15] [15] [15]
0.66 0.2 0.75 0.28 0.48
[15] [15] [15]
AS, aortic stenosis; TA, transapical; TAVI, transcatheter aortic valve implantation; TF, transfemoral; SAVR, surgical aortic valve replacement.
complication, hospitalization, and drug costs. The cost data and their sources are listed in Table 2. We derived the follow-up and complication costs from a previous study [16]. The procedural costs for TAVI included hospital and material costs. The Edwards Sapien valve and the Sapien XT valve cost 4,530,000 yen in Japan. From the OCEAN TAVI registry, we estimated that TAVI costs 6,000,000 yen [12,13]. We estimated the TAVI procedural cost as 5,000,000 yen and the hospital cost as 1,000,000 yen. Following various reports, we determined the SAVR cost as 4,500,000 yen [17]. We decreased the costs by 2% annually [14]. Sensitivity analyses We performed a broad sensitivity analysis to assess the effect of uncertainty on the results. We calculated the impact of uncertainty on all input parameters probabilistically. The distribution of the parameters (beta or gamma distribution) depended on the type of parameter. We estimated the range of parameters based on previous studies [7,10] and performed 100,000 simulations. The outcomes were calculated as the incremental cost-effectiveness ratio (ICER) per QALY gained and per life-year gained. We performed one-way deterministic sensitivity analyses to evaluate the impact of several parameters, and the results were summarized in a tornado diagram. Threshold analyses were used to evaluate the limits of each parameter. We applied a cost-effectiveness cut-off point of 5,000,000 yen based on previous reports [4]. We then generated cost-effectiveness acceptability curves to evaluate the probability that one treatment was more cost effective than another using a maximum willingness-to-pay threshold of 5,000,000 yen per QALY gained. The models were run using commercial software (TreeAge Pro 2016; TreeAge, Williamstown, MA, USA).
Results Main results The main results are shown in Table 3A. For inoperable patients, over the 10-year time horizon the expected costs of TF or TA TAVI and medical therapy were 8,014,886 and 1,639,824 yen, respectively. The expected QALYs of TAVI and medical therapy were 3.02 and 1.27 QALYs, respectively. The ICER of TAVI compared with medical therapy was 3,918,808 yen per QALY gained. In operable patients, over the 10-year time horizon the expected costs of TF TAVI and SAVR were 8,039,694 and 6,316,178 yen, respectively. The expected QALYs of TAVI and SAVR were 4.81 and 4.59 QALYs, respectively. The ICER of TAVI compared with SAVR was 7,523,821 yen per QALY gained. Scenario analyses The results of the scenario analysis are shown in Table 3B. Among inoperable patients, the PARTNER cohort B showed a TAVI patient mortality rate of 30% per year. At that mortality rate, the ICER for TAVI was 6,634,192 yen per QALY gained. Among operable patients, the PARTNER 2 cohort (TA and TF) showed a TAVI patient mortality rate of 8.2% per year. At that mortality rate, the ICER for TAVI was 56,528,188 yen per QALY gained. Deterministic sensitivity analyses Differences in the cost-effectiveness ratio results based on the various one-way sensitivity analyses are shown in Fig. 2. Both
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
4 Table 3A Cost effectiveness of transaortic valve implantation. Cost
Arm
TAVI vs. medical therapy (inoperable patients) ¥ 1,639,824 (463,332–3,962,846) Medical therapy TAVI ¥ 8,014,886 (3,122,490–15,514,629) TAVI vs. SAVR (operable patients) SAVR ¥ 6,316,178 (2,592,184–11,966,389) TAVI ¥ 8,039,694 (3,220,941–15,455,080)
QALY
LY
ICER/QALY
ICER/LY
1.27 (0.51–2.58) 3.02 (1.09–6.23)
1.90 (1.08–3.20) 4.02 (2.02–6.42)
¥ 3,918,808
¥ 3,234,236
4.59 (1.99–8.32) 4.81 (2.08–8.71)
6.09 (4.42–7.52) 6.38 (4.69–7.80)
¥ 7,523,821
¥ 5,715,471
5,000,000 yen, the cost-effectiveness probability of TAVI was 60% in inoperable patients and 46% in operable patients.
Table 3B Scenario analysis. TAVI mortality TAVI vs. medical therapy (inoperable OCEAN registry PARTNER cohort B TAVI vs. SAVR (operable patients) PARTNER 2 (TF) PARTNER 2 (TF and TA)
patients) 18%/year 30%/year 7.6% 8.2%
ICER/QALY ¥3,918,808 ¥6,634,192 ¥7,523,821 ¥56,528,188
ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life year; LY, life year; TA, transapical; TAVI, transcatheter aortic valve implantation; TF, transfemoral; SAVR, surgical aortic valve replacement.
models were most sensitive to the long-term mortality rate of TAVI. The TAVI cost was also sensitive to change in both models. Complication rates and follow-up costs were not sensitive to change in either model. Different time horizons resulted in different ICERs (Table 4A). A longer time horizon resulted in a better ICER for TAVI in both models. The results of the threshold analyses are shown in Table 4B. Among inoperable patients, the threshold for TAVI was a mortality rate of <23.4% per year and a TAVI cost of <7,759,085 yen. Among operable patients, the threshold for TAVI was a mortality rate of <6.3% per year and a TAVI cost of <5,427,439 yen. Probabilistic sensitivity analysis The distribution of the simulated cost effectiveness of TAVI is shown in Fig. 3. Among inoperable patients, the cost of TAVI was higher than the cost of medical therapy; however, TAVI achieved more QALYs than medical therapy. Among operable patients, the cost and QALYs were similar in the TAVI and SAVR groups. The costeffectiveness acceptability curves for medical therapy and TAVI are presented in Fig. 4. With a cost-effectiveness threshold of
Discussion This study showed that in inoperable patients, TAVI had good cost effectiveness with an ICER of <5,000,000 yen per QALY gained. In inoperable patients, TAVI achieved around two more QALYs than medical therapy; two more QALYs indicate that a patient will live two more years with full QOL. Compared with chemotherapy for lung cancer, which achieves a life prolongation of only several months, TAVI is a quite an effective treatment for patients with inoperable AS. Although the cost of TAVI was high, the gain in QALYs exceeded the cost. Accordingly, in the treatment of inoperable patients TAVI should be considered not only from a clinical but also a cost-effectiveness standpoint. Conversely, TAVI was not cost-effective for operable patients, with an ICER of >5,000,000 yen. Among operable patients, TAVI achieved almost the same QALYs as SAVR. However, the difference in cost resulted in an unacceptable cost effectiveness for TAVI. If the cost of TAVI treatment diminished, its cost effectiveness would improve. Our threshold analysis showed that if TAVI treatment cost <5,427,439 yen, it would be economically acceptable. If the cost of valve materials for TAVI fell from 4,530,000 to 3,900,000 yen, TAVI would become cost effective for operable patients. Another way to improve the cost effectiveness of TAVI would be to improve long-term mortality. Our one-way sensitivity analysis showed that the mortality difference was the most sensitive factor for the ICER. If mortality associated with TAVI improved beyond that associated with SAVR, the ICER for TAVI would become economically acceptable. Thus, to expand indication of TAVI to operable Japanese patients at intermediate risk from a cost-effectiveness standpoint, either the cost of TAVI must be lower or the outcome of TAVI must improve. A recent study demonstrated that TF TAVI with a Sapien 3 valve achieved a better outcome because of TAVI device
Fig. 2. Tornado plots showing results from deterministic one-way analysis. (A) Transcatheter aortic valve implantation (TAVI) vs. medical therapy. (B) TAVI vs. surgical aortic valve replacement. TAVI, transcatheter aortic valve implantation; Prob, probability; EV, estimated value.
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
5
Table 4A Time horizon (ICER/QALY). TAVI vs. medical (inoperable patients)TAVI vs. SAVR (operable patients) 3 years ¥ 5 years ¥ 10 years¥ 15 years¥ 20 years¥
9,141,474 5,691,221 3,918,808 3,589,914 3,503,678
¥ ¥ ¥ ¥ ¥
25,926,218 15,017,195 7,523,821 5,353,510 4,400,950
Table 4B Threshold analysis.
TAVI cost TAVI operative mortality TAVI operative complication Post TAVI mortality Post TAVI hospitalization
TAVI vs. medical therapy
TAVI vs. SAVR
¥ 7,759,085 16.4%/month 52.0%/month 23.4%/year 37.8%/year
¥ 5,427,439 Less than 0% Less than 0% 6.3%/year 5.7%/year
ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life year; TAVI, transcatheter aortic valve implantation; SAVR, surgical aortic valve replacement.
improvement [18]. In the future, the ICER for TAVI may improve in operable patients. Most studies have found TAVI to be cost effective in inoperable patients [5,9,10]. A UK study determined that the ICER for TAVI was £16,100 per QALY [9], while a Canadian investigation reported that the ICER for TAVI was $51,324 per QALY [10]. A study in Belgium also found TAVI to be acceptable for inoperable patients (ICER of s44,900 per QALY) [7], and a US study determined that the ICER for TAVI was $61,889 per QALY [5]. These findings are consistent with our data. Thus, TAVI appears to be cost effective in inoperable patients. By contrast, there are contrasting findings on the cost effectiveness of TAVI in operable patients. A UK study reported that TF TAVI was definitely cost effective in high-risk patients (lower cost and more QALYs) [19], and a US study found that the ICER for TAVI was $52,773 per QALY [6]. However, an investigation in Canada determined that the ICER for TAVI was $657,237 per QALY [10], while a Belgian study demonstrated that TAVI was not cost effective among operable patients (ICER of s750,000 per QALY gained) [7]. In each of these studies, TF TAVI achieved somewhat more QALYs than SAVR. The differences in these costs of TAVI were because of differences in the respective healthcare systems. The procedural cost of TAVI or SAVR, the valve cost, and the hospital cost differ between countries. If the total TAVI cost is much higher than the total SAVR cost, then the cost effectiveness of TAVI over SAVR is poor. Hospital costs In the Japanese healthcare system are not high. However, the cost of materials, especially that of TAVI valves, is high in Japan. In the present study, the high cost of valves resulted in poor cost effectiveness of TAVI in operable patients. A previous study with a low TAVI cost showed good cost effectiveness of TAVI over SAVR [19]. In the present study, we did not evaluate TA TAVI in operable patients. However, two US studies reported that TA TAVI was superior to SAVR (higher cost and fewer QALYs) [6,15]. Thus, TA TAVI is not cost effective because of its lower QALYs and greater cost than SAVR. Our one-way sensitivity analysis indicated that for both inoperable and operable patients, the ICER for TAVI was largely dependent upon the efficacy and cost of TAVI. We mainly used PARTNER trial data to assess TAVI efficacy, and our results may have differed if we used a different TAVI efficacy. Our threshold analysis demonstrated that as long as the TAVI-associated mortality rate was within 23.4% per year, TAVI would be cost effective for
Fig. 3. Cost-effectiveness plots for transcatheter aortic valve implantation in (A) inoperable patients and (B) operable patients from 100,000 simulations with a decision-analytic model. The ellipses represent the 95% confidence interval. The solid lines represent the willingness to pay with the slope of 5,000,000 yen per quality-adjusted life-year (QALY) gain.
inoperable patients. According to the OCEAN TAVI registry, TAVIassociated mortality is approximately 1–2%, which is much lower than in other countries. These data indicate a better ICER for TAVI in inoperable Japanese patients. Our threshold analysis determined that as long as the cost of TAVI is <7,759,085 yen, then TAVI is cost effective for inoperable patients. We estimated the cost of TAVI from the OCEAN TAVI registry, as there was no national registry in Japan that would have allowed us to confirm this cost. Thus, the actual cost of TAVI may be different from our estimate. Nevertheless, the findings of the present study indicate that TAVI was cost effective in inoperable patients. This study has several strengths. First, it was a non-industrybased investigation. Bell et al. [20] reported that industry-based studies tend to produce good results for certain companies. Another strength of the present investigation with respect to operable patients was the use of data from the PARTNER 2 trial, which addressed intermediate-risk patients. Previous studies have used data from PARTNER cohorts A, B, and C, which covered highrisk patients. At present, TAVI is performed for intermediate-risk patients. The OCEAN TAVI registry showed a mean STS score of 7%. With that score, most Japanese patients undergoing TAVI would be at intermediate risk. The present study is the first to evaluate the cost effectiveness of TAVI over SAVR for intermediate-risk patients. This study also has several limitations. First, our model was mainly based on the PARTNER trial, which is a high-quality randomized controlled trial with a large number of patients. However, there may be differences between the Japanese AS
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
6
Fig. 4. Cost-effective acceptability curve. (A) Transcatheter aortic valve implantation (TAVI) vs. medical therapy. (B) Transfemoral TAVI vs. surgical aortic valve replacement (SAVR).
population and the PARTNER trial population. The event rate also differs according to the stage of AS. However, we conducted a scenario analysis and sensitivity analysis to adjust for such differences. In a Japanese retrospective study, the mortality of inoperable patients with heart failure was as high as PARTNER cohort B [21]. A second limitation is that we made a number of assumptions in our data set, including procedural complications, follow-up costs, and QOL. Nevertheless, our one-way sensitivity analysis revealed that procedural complications, follow-up costs, and QOL had little effect on the ICER for TAVI. Thus, we assume that our results are robust despite these various assumptions. A third limitation is the relatively short follow-up time. Although the PARTNER trial obtained 5-year follow-up data [22], longer-term data for TAVI are largely absent. Thus, we should expand our data from the present study to establish a 10-year Markov model, as our estimation may be different from real-world data. A fourth limitation relates to the type of valve. At present, the Sapien 3 is the most commonly used valve which can be used for bicuspid valve, while the CoreValve and CoreValve Evolut R (Medtronic, Inc., Minneapolis, MN, USA) are other commonly used self-expandable TAVI valves [23,24]. However, our analysis only used Sapien valve or Sapien XT valve data. Different results would likely have been obtained with the Sapien 3 or CoreValve Evolut R. In addition, because a cost-effectiveness analysis requires published data from a large clinical trial it cannot be up to date from a clinical point of view. The final limitation relates to patient selection. In our analysis, we compared TAVI and SAVR in intermediate-risk patients with AS, but did not compare TAVI and SAVR in high-risk patients, who may show different results. Conclusions Our findings suggest that TAVI has good cost effectiveness for inoperable patients. However, TAVI is not cost effective for operable patients. Funding This research received no grant from any funding agency in the public, commercial, or not-for-profit sectors. Conflict of interest The authors declare that there is no conflict of interest.
Acknowledgments We thank Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript.
References [1] Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 2010;363:1597–607. [2] Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med 2011;364:2187–98. [3] Leon MB, Smith CR, Mack MJ, Makkar RR, Svensson LG, Kodali SK, et al. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med 2016;374:1609–20. [4] Anderson JL, Heidenreich PA, Barnett PG, Creager MA, Fonarow GC, Gibbons RJ, et al. ACC/AHA statement on cost/value methodology in clinical practice guidelines and performance measures: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures and Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63:2304–22. [5] Reynolds MR, Magnuson EA, Lei Y, Leon MB, Smith CR, Svensson LG, et al. Health-related quality of life after transcatheter aortic valve replacement in inoperable patients with severe aortic stenosis. Circulation 2011;124:1964–72. [6] Gada H, Kapadia SR, Tuzcu EM, Svensson LG, Marwick TH. Markov model for selection of aortic valve replacement versus transcatheter aortic valve implantation (without replacement) in high-risk patients. Am J Cardiol 2012;109: 1326–33. [7] Neyt M, Van Brabandt H, Devriese S, Van De Sande S. A cost-utility analysis of transcatheter aortic valve implantation in Belgium: focusing on a well-defined and identifiable population. BMJ Open 2012;2(3). pii:e001032. [8] Reynolds MR, Magnuson EA, Lei Y, Wang K, Vilain K, Li H, et al. Cost-effectiveness of transcatheter aortic valve replacement compared with surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results of the PARTNER (Placement of Aortic Transcatheter Valves) trial (Cohort A). J Am Coll Cardiol 2012;60:2683–92. [9] Watt M, Mealing S, Eaton J, Piazza N, Moat N, Brasseur P, et al. Cost-effectiveness of transcatheter aortic valve replacement in patients ineligible for conventional aortic valve replacement. Heart 2012;98:370–6. [10] Doble B, Blackhouse G, Goeree R, Xie F. Cost-effectiveness of the Edwards SAPIEN transcatheter heart valve compared with standard management and surgical aortic valve replacement in patients with severe symptomatic aortic stenosis: a Canadian perspective. J Thorac Cardiovasc Surg 2013;146. 52–60.e3. [11] Cao C, Liou KP, Pathan FK, Virk S, McMonnies R, Wolfenden H, et al. Transcatheter aortic valve implantation versus surgical aortic valve replacement: meta-analysis of clinical outcomes and cost-effectiveness. Curr Pharm Des 2016;22:1965–77. [12] Watanabe Y, Kozuma K, Hioki H, Kawashima H, Nara Y, Kataoka A, et al. Comparison of results of transcatheter aortic valve implantation in patients with versus without active cancer. Am J Cardiol 2016;118:572–7. [13] Yamamoto M, Shimura T, Kano S, Kagase A, Kodama A, Koyama Y, et al. Impact of preparatory coronary protection in patients at high anatomical risk of acute coronary obstruction during transcatheter aortic valve implantation. Int J Cardiol 2016;217:58–63. [14] Fukuda T, Shiroiwa T, Ikeda S, Igarashi A, Akazawa M, Ishida H, et al. Guideline for economic evaluation of healthcare technologies in Japan. J Natl Inst Public Health 2013;62:625–40.
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx [15] Reynolds MR, Magnuson EA, Wang K, Thourani VH, Williams M, Zajarias A, et al. Health-related quality of life after transcatheter or surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results from the PARTNER (Placement of AoRTic TraNscathetER Valve) Trial (Cohort A). J Am Coll Cardiol 2012;60:548–58. [16] Tsutani K, Igarashi A, Fujikawa K, Evers T, Kubin M, Lamotte M, et al. A health economic evaluation of aspirin in the primary prevention of cardiovascular disease in Japan. Intern Med (Tokyo Japan) 2007;46:157–62. [17] Matsuda S. Current situation of aortic stenosis treatment from DPC data point of view. J JCS Cardiol 2011;19:80–3. [18] Wendler O, Schymik G, Treede H, Baumgartner H, Dumonteil N, Neumann FJ, et al. SOURCE 3: 1-year outcomes post-transcatheter aortic valve implantation using the latest generation of the balloon-expandable transcatheter heart valve. Eur Heart J 2017;38(36):2717–26. [19] Fairbairn TA, Meads DM, Hulme C, Mather AN, Plein S, Blackman DJ, et al. The cost-effectiveness of transcatheter aortic valve implantation versus surgical aortic valve replacement in patients with severe aortic stenosis at high operative risk. Heart 2013;99:914–20.
7
[20] Bell CM, Urbach DR, Ray JG, Bayoumi A, Rosen AB, Greenberg D, et al. Bias in published cost effectiveness studies: systematic review. BMJ 2006;332:699–703. [21] Izumo M, Takeuchi M, Seo Y, Yamashita E, Suzuki K, Ishizu T, et al. Prognostic implications in patients with symptomatic aortic stenosis and preserved ejection fraction: Japanese multicenter aortic stenosis, retrospective (JUSTR) registry. J Cardiol 2017;69:110–8. [22] Mack MJ, Leon MB, Smith CR, Miller DC, Moses JW, Tuzcu EM, et al. 5-Year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet 2015;385:2477–84. [23] Arai T, Lefevre T, Hovasse T, Morice MC, Romano M, Benamer H, et al. The feasibility of transcatheter aortic valve implantation using the Edwards SAPIEN 3 for patients with severe bicuspid aortic stenosis. J Cardiol 2017;70:220–4. [24] Gonska B, Seeger J, Baarts J, Rodewald C, Scharnbeck D, Rottbauer W, et al. The balloon-expandable Edwards Sapien 3 valve is superior to the self-expanding Medtronic CoreValve in patients with severe aortic stenosis undergoing transfemoral aortic valve implantation. J Cardiol 2017;69:877–82.
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
JJCC-1572; No. of Pages 7 Journal of Cardiology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Cardiology journal homepage: www.elsevier.com/locate/jjcc
Original article
Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan Satoshi Kodera (MD)*, Arihiro Kiyosue (MD, PhD, FJCC), Jiro Ando (MD), Issei Komuro (MD, PhD, FJCC) Department of Cardiovascular Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 August 2017 Received in revised form 26 September 2017 Accepted 10 October 2017 Available online xxx
Background: Transcatheter aortic valve implantation (TAVI) is a less invasive treatment for elderly patients with aortic stenosis. However, the cost of TAVI is a major issue. This study analyzed the cost effectiveness of TAVI in Japan. Methods: We developed an economic model to evaluate the quality-adjusted life years (QALYs) and costs of TAVI, surgical aortic valve replacement (SAVR), and medical therapy over a 10-year time horizon from the perspective of Japanese public healthcare payers. The first model compared transapical or transfemoral TAVI with Sapien valve implantation and medical therapy in inoperable patients. The second model compared transfemoral TAVI with Sapien XT valve implantation and SAVR in operable patients with intermediate surgical risk. We assumed a cost-effectiveness threshold of 5,000,000 yen per QALY, and assessed the cost-effectiveness probability with 100,000 simulations. We performed a broad sensitivity analysis to assess the effect of uncertainty on our results. Results: Among inoperable patients, the incremental cost-effectiveness ratio for TAVI compared with medical therapy was 3,918,808 yen per QALY. In operable patients, the incremental cost-effectiveness ratio for TAVI compared with SAVR was 7,523,821 yen per QALY. The cost-effectiveness probability of TAVI was 60% for inoperable patients and 46% for operable patients. Among inoperable patients, the costeffective threshold of TAVI was <7,759,085 yen. Among operable patients, the cost-effective threshold of TAVI was <5,427,439 yen. Conclusions: This study suggests that TAVI has good cost effectiveness for inoperable patients, but not for operable patients. © 2017 Published by Elsevier Ltd on behalf of Japanese College of Cardiology.
Keywords: Transcatheter aortic valve implantation Incremental cost-effectiveness ratio Cost-effectiveness Quality-adjusted life year Aortic stenosis
Introduction Transcatheter aortic valve implantation (TAVI) was developed for the treatment of severe aortic stenosis (AS) [1]. TAVI is a less invasive treatment than surgical aortic valve replacement, and is an effective procedure for elderly and high-risk patients [1– 3]. However, the cost of TAVI is a major issue, with the Sapien valve (Edwards Lifesciences Co., Irvine, CA, USA) costing 4,530,000 yen in Japan. Cost-effectiveness analysis is a research method applied to estimate the incremental benefit and cost of a new treatment compared with the standard treatment [4]. A number of reports in
* Corresponding author at: Department of Cardiovascular Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 1138655, Japan. E-mail address: [email protected] (S. Kodera).
Western countries have examined the cost effectiveness of TAVI [5–11]. However, their conclusions varied, with some studies reporting TAVI to be cost-effective, while others did not. Healthcare systems differ between countries, as does the cost effectiveness of TAVI. However, no cost-effectiveness analysis of TAVI has been conducted in Japan. Considering the low economic growth and burgeoning elderly population in Japan, health economic issues are increasingly important. Thus, in the present study we analyzed the cost effectiveness of TAVI in Japan. Methods We developed an economic model to evaluate the costs and effectiveness of TAVI. The model evaluated the quality-adjusted life years (QALYs) and cost of three treatment options—TAVI, surgical aortic valve replacement (SAVR), and medical therapy—during a 10-year time horizon from the perspective of the Japanese public
https://doi.org/10.1016/j.jjcc.2017.10.008 0914-5087/© 2017 Published by Elsevier Ltd on behalf of Japanese College of Cardiology.
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
2
healthcare payer. We constructed two models. The first model compared transapical (TA) or transfemoral (TF) TAVI with a Sapien valve and medical therapy in inoperable patients. We defined an inoperable patient as a patient with a 30-day mortality and morbidity rate of >50% according to one of the Placement of Aortic Transcatheter Valves (PARTNER) trials [1], which are large randomized controlled trials of AS. The second model compared TF TAVI with a Sapien XT valve and SAVR in operable patients with intermediate surgical risk. We defined an operable patient with intermediate surgical risk as a patient with a 30-day mortality rate of 4%–8% according to another PARTNER trial [3]. We derived various parameters from the PARTNER trials and the Optimized Catheter vAlvular iNtervention (OCEAN) TAVI registry [12,13], which is a Japanese TAVI registry. The analysis was performed according to the Consolidated Health Economic Evaluation Reporting Standards statement and in accordance with a Japanese guideline [4,14]. Model We developed a Markov model with Monte Carlo simulations to evaluate the efficiency of TAVI. The structure of the model is shown in Fig. 1. A monthly cycle was modeled with each cycle, the patient may die, be hospitalized, or become stable. We defined four study phases: study entry, stability, hospitalization, and death. At the entry point, all patients were classified in the study entry phase. If the subsequent 1 month was uneventful, the patients were classified as stable. We combined stroke, myocardial infarction, and vascular complications as hospitalization conditions.
Population The population of patients in each group is shown in Table 1. Our study population reflected that of the PARTNER trials [1,3]. To compare TA or TF TAVI with Sapien valve implantation and medical therapy, we used the PARTNER cohort B, which comprised inoperable patients. The average patient age was 83 years, 46% were men, their average Society of Thoracic Surgeons (STS) score was 11, and 93% had a New York Heart Association class of III or IV [1]. To compare TF TAVI with Sapien XT valve implantation and SAVR, we used the PARTNER 2 cohort A, which comprised patients with an intermediate surgical risk. The average patient age was 82 years, 55% were male, the average STS score was 6, and 77% were New York Heart Association class III or IV [3]. Intervention vs. comparator In the first model, we compared TA or TF TAVI with Sapien valve implantation and medical therapy in inoperable patients. With medical therapy, balloon valvuloplasty was included if necessary. In the second model, we compared TF TAVI with Sapien XT valve implantation and SAVR in operable patients with an intermediate surgical risk. Time horizon We restricted the time horizon to 10 years in both models. The average age of both inoperable and operable patients was approximately 82 years. We considered that a 10-year estimation would be sufficient to evaluate the cost effectiveness of the treatments. In the sensitivity analysis, we changed the time horizon to 3, 5, 10, 15, and 20 years. Mortality and hospitalization The mortality rates associated with each treatment are shown in Table 2. The mortality for all treatments was based on that reported for the PARTNER cohorts and OCEAN TAVI registry [12,13]. For inoperable patients, TAVI was 60% superior to medical therapy in terms of survival. For operable patients, TF TAVI was 10% superior to SAVR in terms of survival. Among patients who underwent TAVI in the OCEAN TAVI registry, the procedural 30-day mortality rate was approximately 2%. The hospitalization rate was based on the PARTNER trial. The reasons for hospitalization included worsening heart failure, stroke, and vascular complications. Among inoperable patients, the hospitalization rate was 50% lower with TAVI than with medical therapy. For operable patients, the hospitalization rate was higher with TAVI than with SAVR. We modified the mortality rate to evaluate the effects of that rate in the sensitivity analysis. Utility The utility value in each situation is shown in Table 2. The PARTNER study determined quality of life (QOL) using the EuroQol Five-Dimensions Questionnaire [5,15]. We used data of the PARTNER study and from other trials to define QOL [16]. In patients with AS before SAVR or TAVI, the QOL was 0.66. In patients with AS after SAVR or TAVI, the QOL was 0.75. We assumed that the QOL of hospitalized patients was 0.48. Following a Japanese guideline, we decreased the utility values by 2% annually [14]. Costs
Fig. 1. Markov model of transcatheter aortic valve implantation. The model applies to each monthly cycle. AS, aortic stenosis; AVR, aortic valve replacement; TAVI, transcatheter aortic valve implantation.
We performed an economic evaluation from the perspective of a public healthcare payer in Japan. The costs included procedural,
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
3
Table 1 Characteristics of the study groups. Inoperable patients [1]
Age (years) Male (%) NYHA function class III–IV (%) STS score (%) Type of valve
Operable patients [3]
TAVI (TA or TF)
Medical therapy
TAVI (TF)
SAVR
83 46 93 11 Sapien valve
83 46 93 11 –
82 55 77 6.0 Sapien XT
82 55 77 6.0 –
NYHA, New York Heart Association; TA, transapical; TAVI, transcatheter aortic valve implantation; TF, transfemoral; SAVR, surgical aortic valve replacement; STS, Society of Thoracic Surgeons.
Table 2 Key parameters used in the models. Inoperable patients TAVI (TA or TF) Perioperative death Perioperative Hospitalization One year mortality One year hospitalization Mortality rate of hospitalization Procedural cost Complication cost Follow-up cost (per month) AS utility Post AS utility Hospitalization utility
Source
Medical therapy
Operable patients Source
TAVI (TF)
Source
SAVR
Source
2.1 0.5% 5.6 1.7%
[12,13] [1]
2.8 1.2% 10.1 2.2%
[1] [1]
2.1 0.5% 5.4 0.8%
[3,12,13] [3,12,13]
4.1 0.7% 6.1 0.8%
[3] [3]
18 10.2% 22.3 10.6% 20 2.8%
[12,13] [1] Expert
49.7 18.0% 44.1 17.4% 20 2.8%
[1] [1] Expert
7.6 3.3% 9.0 3.6% 20 2.8%
[3,12,13] [3,12,13] Expert
8.6 3.4% 8.4 3.4% 20 2.8%
[3] [3] Expert
¥ 6,000,000 3,000,000 ¥ 1,500,000 750,000 ¥ 15,000 7500
[12,13] [16] [16]
¥ 15,000 7500 ¥ 1,500,000 750,000 ¥ 15,000 7500
[16] [16] [16]
¥ 6,000,000 3,000,000 ¥ 1,500,000 750,000 ¥ 15,000 7500
[12,13] [16] [16]
¥ 4,500,000 2,250,000 ¥ 1,500,000 750,000 ¥ 15,000 7500
[17] [16] [16]
0,66 0.2 0.75 0.28 0.48
[5] [5] [4]
0.66 0.2 0.66 0.2 0.48
[5] [5] [4]
0.66 0.2 0.75 0.28 0.48
[15] [15] [15]
0.66 0.2 0.75 0.28 0.48
[15] [15] [15]
AS, aortic stenosis; TA, transapical; TAVI, transcatheter aortic valve implantation; TF, transfemoral; SAVR, surgical aortic valve replacement.
complication, hospitalization, and drug costs. The cost data and their sources are listed in Table 2. We derived the follow-up and complication costs from a previous study [16]. The procedural costs for TAVI included hospital and material costs. The Edwards Sapien valve and the Sapien XT valve cost 4,530,000 yen in Japan. From the OCEAN TAVI registry, we estimated that TAVI costs 6,000,000 yen [12,13]. We estimated the TAVI procedural cost as 5,000,000 yen and the hospital cost as 1,000,000 yen. Following various reports, we determined the SAVR cost as 4,500,000 yen [17]. We decreased the costs by 2% annually [14]. Sensitivity analyses We performed a broad sensitivity analysis to assess the effect of uncertainty on the results. We calculated the impact of uncertainty on all input parameters probabilistically. The distribution of the parameters (beta or gamma distribution) depended on the type of parameter. We estimated the range of parameters based on previous studies [7,10] and performed 100,000 simulations. The outcomes were calculated as the incremental cost-effectiveness ratio (ICER) per QALY gained and per life-year gained. We performed one-way deterministic sensitivity analyses to evaluate the impact of several parameters, and the results were summarized in a tornado diagram. Threshold analyses were used to evaluate the limits of each parameter. We applied a cost-effectiveness cut-off point of 5,000,000 yen based on previous reports [4]. We then generated cost-effectiveness acceptability curves to evaluate the probability that one treatment was more cost effective than another using a maximum willingness-to-pay threshold of 5,000,000 yen per QALY gained. The models were run using commercial software (TreeAge Pro 2016; TreeAge, Williamstown, MA, USA).
Results Main results The main results are shown in Table 3A. For inoperable patients, over the 10-year time horizon the expected costs of TF or TA TAVI and medical therapy were 8,014,886 and 1,639,824 yen, respectively. The expected QALYs of TAVI and medical therapy were 3.02 and 1.27 QALYs, respectively. The ICER of TAVI compared with medical therapy was 3,918,808 yen per QALY gained. In operable patients, over the 10-year time horizon the expected costs of TF TAVI and SAVR were 8,039,694 and 6,316,178 yen, respectively. The expected QALYs of TAVI and SAVR were 4.81 and 4.59 QALYs, respectively. The ICER of TAVI compared with SAVR was 7,523,821 yen per QALY gained. Scenario analyses The results of the scenario analysis are shown in Table 3B. Among inoperable patients, the PARTNER cohort B showed a TAVI patient mortality rate of 30% per year. At that mortality rate, the ICER for TAVI was 6,634,192 yen per QALY gained. Among operable patients, the PARTNER 2 cohort (TA and TF) showed a TAVI patient mortality rate of 8.2% per year. At that mortality rate, the ICER for TAVI was 56,528,188 yen per QALY gained. Deterministic sensitivity analyses Differences in the cost-effectiveness ratio results based on the various one-way sensitivity analyses are shown in Fig. 2. Both
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
4 Table 3A Cost effectiveness of transaortic valve implantation. Cost
Arm
TAVI vs. medical therapy (inoperable patients) ¥ 1,639,824 (463,332–3,962,846) Medical therapy TAVI ¥ 8,014,886 (3,122,490–15,514,629) TAVI vs. SAVR (operable patients) SAVR ¥ 6,316,178 (2,592,184–11,966,389) TAVI ¥ 8,039,694 (3,220,941–15,455,080)
QALY
LY
ICER/QALY
ICER/LY
1.27 (0.51–2.58) 3.02 (1.09–6.23)
1.90 (1.08–3.20) 4.02 (2.02–6.42)
¥ 3,918,808
¥ 3,234,236
4.59 (1.99–8.32) 4.81 (2.08–8.71)
6.09 (4.42–7.52) 6.38 (4.69–7.80)
¥ 7,523,821
¥ 5,715,471
5,000,000 yen, the cost-effectiveness probability of TAVI was 60% in inoperable patients and 46% in operable patients.
Table 3B Scenario analysis. TAVI mortality TAVI vs. medical therapy (inoperable OCEAN registry PARTNER cohort B TAVI vs. SAVR (operable patients) PARTNER 2 (TF) PARTNER 2 (TF and TA)
patients) 18%/year 30%/year 7.6% 8.2%
ICER/QALY ¥3,918,808 ¥6,634,192 ¥7,523,821 ¥56,528,188
ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life year; LY, life year; TA, transapical; TAVI, transcatheter aortic valve implantation; TF, transfemoral; SAVR, surgical aortic valve replacement.
models were most sensitive to the long-term mortality rate of TAVI. The TAVI cost was also sensitive to change in both models. Complication rates and follow-up costs were not sensitive to change in either model. Different time horizons resulted in different ICERs (Table 4A). A longer time horizon resulted in a better ICER for TAVI in both models. The results of the threshold analyses are shown in Table 4B. Among inoperable patients, the threshold for TAVI was a mortality rate of <23.4% per year and a TAVI cost of <7,759,085 yen. Among operable patients, the threshold for TAVI was a mortality rate of <6.3% per year and a TAVI cost of <5,427,439 yen. Probabilistic sensitivity analysis The distribution of the simulated cost effectiveness of TAVI is shown in Fig. 3. Among inoperable patients, the cost of TAVI was higher than the cost of medical therapy; however, TAVI achieved more QALYs than medical therapy. Among operable patients, the cost and QALYs were similar in the TAVI and SAVR groups. The costeffectiveness acceptability curves for medical therapy and TAVI are presented in Fig. 4. With a cost-effectiveness threshold of
Discussion This study showed that in inoperable patients, TAVI had good cost effectiveness with an ICER of <5,000,000 yen per QALY gained. In inoperable patients, TAVI achieved around two more QALYs than medical therapy; two more QALYs indicate that a patient will live two more years with full QOL. Compared with chemotherapy for lung cancer, which achieves a life prolongation of only several months, TAVI is a quite an effective treatment for patients with inoperable AS. Although the cost of TAVI was high, the gain in QALYs exceeded the cost. Accordingly, in the treatment of inoperable patients TAVI should be considered not only from a clinical but also a cost-effectiveness standpoint. Conversely, TAVI was not cost-effective for operable patients, with an ICER of >5,000,000 yen. Among operable patients, TAVI achieved almost the same QALYs as SAVR. However, the difference in cost resulted in an unacceptable cost effectiveness for TAVI. If the cost of TAVI treatment diminished, its cost effectiveness would improve. Our threshold analysis showed that if TAVI treatment cost <5,427,439 yen, it would be economically acceptable. If the cost of valve materials for TAVI fell from 4,530,000 to 3,900,000 yen, TAVI would become cost effective for operable patients. Another way to improve the cost effectiveness of TAVI would be to improve long-term mortality. Our one-way sensitivity analysis showed that the mortality difference was the most sensitive factor for the ICER. If mortality associated with TAVI improved beyond that associated with SAVR, the ICER for TAVI would become economically acceptable. Thus, to expand indication of TAVI to operable Japanese patients at intermediate risk from a cost-effectiveness standpoint, either the cost of TAVI must be lower or the outcome of TAVI must improve. A recent study demonstrated that TF TAVI with a Sapien 3 valve achieved a better outcome because of TAVI device
Fig. 2. Tornado plots showing results from deterministic one-way analysis. (A) Transcatheter aortic valve implantation (TAVI) vs. medical therapy. (B) TAVI vs. surgical aortic valve replacement. TAVI, transcatheter aortic valve implantation; Prob, probability; EV, estimated value.
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
5
Table 4A Time horizon (ICER/QALY). TAVI vs. medical (inoperable patients)TAVI vs. SAVR (operable patients) 3 years ¥ 5 years ¥ 10 years¥ 15 years¥ 20 years¥
9,141,474 5,691,221 3,918,808 3,589,914 3,503,678
¥ ¥ ¥ ¥ ¥
25,926,218 15,017,195 7,523,821 5,353,510 4,400,950
Table 4B Threshold analysis.
TAVI cost TAVI operative mortality TAVI operative complication Post TAVI mortality Post TAVI hospitalization
TAVI vs. medical therapy
TAVI vs. SAVR
¥ 7,759,085 16.4%/month 52.0%/month 23.4%/year 37.8%/year
¥ 5,427,439 Less than 0% Less than 0% 6.3%/year 5.7%/year
ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life year; TAVI, transcatheter aortic valve implantation; SAVR, surgical aortic valve replacement.
improvement [18]. In the future, the ICER for TAVI may improve in operable patients. Most studies have found TAVI to be cost effective in inoperable patients [5,9,10]. A UK study determined that the ICER for TAVI was £16,100 per QALY [9], while a Canadian investigation reported that the ICER for TAVI was $51,324 per QALY [10]. A study in Belgium also found TAVI to be acceptable for inoperable patients (ICER of s44,900 per QALY) [7], and a US study determined that the ICER for TAVI was $61,889 per QALY [5]. These findings are consistent with our data. Thus, TAVI appears to be cost effective in inoperable patients. By contrast, there are contrasting findings on the cost effectiveness of TAVI in operable patients. A UK study reported that TF TAVI was definitely cost effective in high-risk patients (lower cost and more QALYs) [19], and a US study found that the ICER for TAVI was $52,773 per QALY [6]. However, an investigation in Canada determined that the ICER for TAVI was $657,237 per QALY [10], while a Belgian study demonstrated that TAVI was not cost effective among operable patients (ICER of s750,000 per QALY gained) [7]. In each of these studies, TF TAVI achieved somewhat more QALYs than SAVR. The differences in these costs of TAVI were because of differences in the respective healthcare systems. The procedural cost of TAVI or SAVR, the valve cost, and the hospital cost differ between countries. If the total TAVI cost is much higher than the total SAVR cost, then the cost effectiveness of TAVI over SAVR is poor. Hospital costs In the Japanese healthcare system are not high. However, the cost of materials, especially that of TAVI valves, is high in Japan. In the present study, the high cost of valves resulted in poor cost effectiveness of TAVI in operable patients. A previous study with a low TAVI cost showed good cost effectiveness of TAVI over SAVR [19]. In the present study, we did not evaluate TA TAVI in operable patients. However, two US studies reported that TA TAVI was superior to SAVR (higher cost and fewer QALYs) [6,15]. Thus, TA TAVI is not cost effective because of its lower QALYs and greater cost than SAVR. Our one-way sensitivity analysis indicated that for both inoperable and operable patients, the ICER for TAVI was largely dependent upon the efficacy and cost of TAVI. We mainly used PARTNER trial data to assess TAVI efficacy, and our results may have differed if we used a different TAVI efficacy. Our threshold analysis demonstrated that as long as the TAVI-associated mortality rate was within 23.4% per year, TAVI would be cost effective for
Fig. 3. Cost-effectiveness plots for transcatheter aortic valve implantation in (A) inoperable patients and (B) operable patients from 100,000 simulations with a decision-analytic model. The ellipses represent the 95% confidence interval. The solid lines represent the willingness to pay with the slope of 5,000,000 yen per quality-adjusted life-year (QALY) gain.
inoperable patients. According to the OCEAN TAVI registry, TAVIassociated mortality is approximately 1–2%, which is much lower than in other countries. These data indicate a better ICER for TAVI in inoperable Japanese patients. Our threshold analysis determined that as long as the cost of TAVI is <7,759,085 yen, then TAVI is cost effective for inoperable patients. We estimated the cost of TAVI from the OCEAN TAVI registry, as there was no national registry in Japan that would have allowed us to confirm this cost. Thus, the actual cost of TAVI may be different from our estimate. Nevertheless, the findings of the present study indicate that TAVI was cost effective in inoperable patients. This study has several strengths. First, it was a non-industrybased investigation. Bell et al. [20] reported that industry-based studies tend to produce good results for certain companies. Another strength of the present investigation with respect to operable patients was the use of data from the PARTNER 2 trial, which addressed intermediate-risk patients. Previous studies have used data from PARTNER cohorts A, B, and C, which covered highrisk patients. At present, TAVI is performed for intermediate-risk patients. The OCEAN TAVI registry showed a mean STS score of 7%. With that score, most Japanese patients undergoing TAVI would be at intermediate risk. The present study is the first to evaluate the cost effectiveness of TAVI over SAVR for intermediate-risk patients. This study also has several limitations. First, our model was mainly based on the PARTNER trial, which is a high-quality randomized controlled trial with a large number of patients. However, there may be differences between the Japanese AS
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx
6
Fig. 4. Cost-effective acceptability curve. (A) Transcatheter aortic valve implantation (TAVI) vs. medical therapy. (B) Transfemoral TAVI vs. surgical aortic valve replacement (SAVR).
population and the PARTNER trial population. The event rate also differs according to the stage of AS. However, we conducted a scenario analysis and sensitivity analysis to adjust for such differences. In a Japanese retrospective study, the mortality of inoperable patients with heart failure was as high as PARTNER cohort B [21]. A second limitation is that we made a number of assumptions in our data set, including procedural complications, follow-up costs, and QOL. Nevertheless, our one-way sensitivity analysis revealed that procedural complications, follow-up costs, and QOL had little effect on the ICER for TAVI. Thus, we assume that our results are robust despite these various assumptions. A third limitation is the relatively short follow-up time. Although the PARTNER trial obtained 5-year follow-up data [22], longer-term data for TAVI are largely absent. Thus, we should expand our data from the present study to establish a 10-year Markov model, as our estimation may be different from real-world data. A fourth limitation relates to the type of valve. At present, the Sapien 3 is the most commonly used valve which can be used for bicuspid valve, while the CoreValve and CoreValve Evolut R (Medtronic, Inc., Minneapolis, MN, USA) are other commonly used self-expandable TAVI valves [23,24]. However, our analysis only used Sapien valve or Sapien XT valve data. Different results would likely have been obtained with the Sapien 3 or CoreValve Evolut R. In addition, because a cost-effectiveness analysis requires published data from a large clinical trial it cannot be up to date from a clinical point of view. The final limitation relates to patient selection. In our analysis, we compared TAVI and SAVR in intermediate-risk patients with AS, but did not compare TAVI and SAVR in high-risk patients, who may show different results. Conclusions Our findings suggest that TAVI has good cost effectiveness for inoperable patients. However, TAVI is not cost effective for operable patients. Funding This research received no grant from any funding agency in the public, commercial, or not-for-profit sectors. Conflict of interest The authors declare that there is no conflict of interest.
Acknowledgments We thank Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript.
References [1] Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 2010;363:1597–607. [2] Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med 2011;364:2187–98. [3] Leon MB, Smith CR, Mack MJ, Makkar RR, Svensson LG, Kodali SK, et al. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med 2016;374:1609–20. [4] Anderson JL, Heidenreich PA, Barnett PG, Creager MA, Fonarow GC, Gibbons RJ, et al. ACC/AHA statement on cost/value methodology in clinical practice guidelines and performance measures: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures and Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63:2304–22. [5] Reynolds MR, Magnuson EA, Lei Y, Leon MB, Smith CR, Svensson LG, et al. Health-related quality of life after transcatheter aortic valve replacement in inoperable patients with severe aortic stenosis. Circulation 2011;124:1964–72. [6] Gada H, Kapadia SR, Tuzcu EM, Svensson LG, Marwick TH. Markov model for selection of aortic valve replacement versus transcatheter aortic valve implantation (without replacement) in high-risk patients. Am J Cardiol 2012;109: 1326–33. [7] Neyt M, Van Brabandt H, Devriese S, Van De Sande S. A cost-utility analysis of transcatheter aortic valve implantation in Belgium: focusing on a well-defined and identifiable population. BMJ Open 2012;2(3). pii:e001032. [8] Reynolds MR, Magnuson EA, Lei Y, Wang K, Vilain K, Li H, et al. Cost-effectiveness of transcatheter aortic valve replacement compared with surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results of the PARTNER (Placement of Aortic Transcatheter Valves) trial (Cohort A). J Am Coll Cardiol 2012;60:2683–92. [9] Watt M, Mealing S, Eaton J, Piazza N, Moat N, Brasseur P, et al. Cost-effectiveness of transcatheter aortic valve replacement in patients ineligible for conventional aortic valve replacement. Heart 2012;98:370–6. [10] Doble B, Blackhouse G, Goeree R, Xie F. Cost-effectiveness of the Edwards SAPIEN transcatheter heart valve compared with standard management and surgical aortic valve replacement in patients with severe symptomatic aortic stenosis: a Canadian perspective. J Thorac Cardiovasc Surg 2013;146. 52–60.e3. [11] Cao C, Liou KP, Pathan FK, Virk S, McMonnies R, Wolfenden H, et al. Transcatheter aortic valve implantation versus surgical aortic valve replacement: meta-analysis of clinical outcomes and cost-effectiveness. Curr Pharm Des 2016;22:1965–77. [12] Watanabe Y, Kozuma K, Hioki H, Kawashima H, Nara Y, Kataoka A, et al. Comparison of results of transcatheter aortic valve implantation in patients with versus without active cancer. Am J Cardiol 2016;118:572–7. [13] Yamamoto M, Shimura T, Kano S, Kagase A, Kodama A, Koyama Y, et al. Impact of preparatory coronary protection in patients at high anatomical risk of acute coronary obstruction during transcatheter aortic valve implantation. Int J Cardiol 2016;217:58–63. [14] Fukuda T, Shiroiwa T, Ikeda S, Igarashi A, Akazawa M, Ishida H, et al. Guideline for economic evaluation of healthcare technologies in Japan. J Natl Inst Public Health 2013;62:625–40.
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008
G Model
JJCC-1572; No. of Pages 7 S. Kodera et al. / Journal of Cardiology xxx (2017) xxx–xxx [15] Reynolds MR, Magnuson EA, Wang K, Thourani VH, Williams M, Zajarias A, et al. Health-related quality of life after transcatheter or surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results from the PARTNER (Placement of AoRTic TraNscathetER Valve) Trial (Cohort A). J Am Coll Cardiol 2012;60:548–58. [16] Tsutani K, Igarashi A, Fujikawa K, Evers T, Kubin M, Lamotte M, et al. A health economic evaluation of aspirin in the primary prevention of cardiovascular disease in Japan. Intern Med (Tokyo Japan) 2007;46:157–62. [17] Matsuda S. Current situation of aortic stenosis treatment from DPC data point of view. J JCS Cardiol 2011;19:80–3. [18] Wendler O, Schymik G, Treede H, Baumgartner H, Dumonteil N, Neumann FJ, et al. SOURCE 3: 1-year outcomes post-transcatheter aortic valve implantation using the latest generation of the balloon-expandable transcatheter heart valve. Eur Heart J 2017;38(36):2717–26. [19] Fairbairn TA, Meads DM, Hulme C, Mather AN, Plein S, Blackman DJ, et al. The cost-effectiveness of transcatheter aortic valve implantation versus surgical aortic valve replacement in patients with severe aortic stenosis at high operative risk. Heart 2013;99:914–20.
7
[20] Bell CM, Urbach DR, Ray JG, Bayoumi A, Rosen AB, Greenberg D, et al. Bias in published cost effectiveness studies: systematic review. BMJ 2006;332:699–703. [21] Izumo M, Takeuchi M, Seo Y, Yamashita E, Suzuki K, Ishizu T, et al. Prognostic implications in patients with symptomatic aortic stenosis and preserved ejection fraction: Japanese multicenter aortic stenosis, retrospective (JUSTR) registry. J Cardiol 2017;69:110–8. [22] Mack MJ, Leon MB, Smith CR, Miller DC, Moses JW, Tuzcu EM, et al. 5-Year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet 2015;385:2477–84. [23] Arai T, Lefevre T, Hovasse T, Morice MC, Romano M, Benamer H, et al. The feasibility of transcatheter aortic valve implantation using the Edwards SAPIEN 3 for patients with severe bicuspid aortic stenosis. J Cardiol 2017;70:220–4. [24] Gonska B, Seeger J, Baarts J, Rodewald C, Scharnbeck D, Rottbauer W, et al. The balloon-expandable Edwards Sapien 3 valve is superior to the self-expanding Medtronic CoreValve in patients with severe aortic stenosis undergoing transfemoral aortic valve implantation. J Cardiol 2017;69:877–82.
Please cite this article in press as: Kodera S, et al. Cost effectiveness of transcatheter aortic valve implantation in patients with aortic stenosis in Japan. J Cardiol (2017), https://doi.org/10.1016/j.jjcc.2017.10.008