Kosteneffectiviteit rapport

Kosteneffectiviteit rapport (iMTA – Burgers 2015)

REPORT 08/2015

Systematic review of cost-effectiveness analyses of

treatments of peripheral artery disease

LT Burgers, M Luyendijk, MJ Poley, JL Severens, WK Redekop

Erasmus University Rotterdam

Institute of Health Policy & Management

Tel. +31 10 40 88535

Redekop@bmg.eur.nl

Table of Contents

List of abbreviations. 215

English summary. 217

Dutch summary. 222

Introduction. 226

Methods. 228

Results. 232

Cost-effectiveness of interventions for patients with intermittent claudication. 234

Supervised exercise versus unsupervised exercise. 234

Revascularization versus exercise therapy. 235

Revascularization versus revascularization. 237

Medication. 239

Cost-effectiveness of interventions for patients with critical limb ischemia. 241

Amputation versus revascularization. 241

Revascularization versus revascularization. 242

Medication versus revascularization. 245

Cost-effectiveness of percutaneous interventions for patients with PAD.. 246

Discussion. 248

Conclusions. 252

References. 254

APPENDIX 1. 260

APPENDIX 2. 268

APPENDIX 3. 3

List of abbreviations

AAV	autologous alternative vein	N	necrosis
ABI	ankle-brachial index	NA	not applicable
AC	Allocation concealment (selection bias)	NHS	National Health Service
ACD	absolute claudication distance	NS	not stated
AFS	amputation free survival	NT	no treatment
ASA	acetylsalicylic acid	OB	other bias
BMS	bare metal stent	OMC	optimal medical care
BOA	Blinding of outcome assessment (detection bias)	OS	overall survival
BP	bypass surgery	PAD	peripheral artery disease
BPP	Blinding of participants and personnel (performance bias)	PAOD	peripheral arterial occlusive disease
CLI	critical limb ischemia	PES	paclitaxel eluting stent
CLOP	clopidogrel	PFWD	Pain-free walking distance
CMM	conventional medical management	PGE1	prostaglandin E1
DES	drug eluting stent	PRWD	patient reported walking distances
DSA	intra-arterial digital subtraction angiography	PTA	percutaneous transluminal angioplasty
DUS	colour-guided duplex ultrasound	PTFE	polytetrafluoroethylene
ED	extended dominancy	QALY	quality adjusted life years
EES	everolimus eluting stent	QOL	quality of life
ER	endovascular revascularization	RCT	randomized controlled trial
EQ-5D	euroqol 5 dimensions	Rev	revision
EVBT	endovascular brachytherapy	RP	rest pain
GSV	good-quality great saphenous vein	RSG	Random sequence generation (selection bias)
HCP	health care payer/provider	SCS	spinal cord simulation
HCS	health care system	SE	supervised exercise
HRQOL	health related quality of life	SES	sirolimus eluting stent
HUI	health utility index	SF-36	short form 36
IC	intermittent claudication	SG	standard gamble
ICD	intermittent claudication distance	SR	Selective reporting (reporting bias)
ICER	incremental cost-effectiveness ratio	SS	selective stenting
IOD	Incomplete outcome data (attrition bias)	TL	tissue loss
LY	life years	TTO	time trade off
MED	medication	U	ulceration
MRA	magnetic resonance angiography	UE	unsupervised exercise
MWD	maximum walking distance	WA	walk advice

English summary

Patients with lower limb peripheral arterial disease (PAD) often have pain in their calf, buttocks or thigh after walking (intermittent claudication). When pain persists in rest and/or trophic skin changes are present and the systolic blood pressure measured at the ankle by a Doppler ultrasound device is lower than 50mmHg, patients are diagnosed with critical limb ischemia (CLI). Treatment of IC patients has three aims: 1) decrease cardiovascular disease risk, 2) improve walking distance, and 3) improve quality of life. Patients often receive medication (e.g. cilostazol, naftidrofuryl oxalate, pentoxifylline or inositol nicotinate), exercise therapy, revascularization (e.g. (bypass) surgery, percutaneous transluminal angioplasty (PTA) or PTA in combination with stenting) or an amputation.

The interventions named above all have the aim to improve health outcomes of patients with CLI or IC, but may also lead to additional healthcare costs. Therefore, it is necessary to consider the cost-effectiveness of these interventions. Many studies have evaluated the cost-effectiveness of these interventions for IC or CLI or both indications. The aim of this systematic review was to comprehensively summarize the cost-effectiveness of all treatment strategies for patients with CLI or IC focussing on specific decision problems:

What is the cost effectiveness of treatment strategies for the treatment of PAD in patients with intermittent claudication in the Netherlands?
What is the cost effectiveness of treatment strategies for the treatment of PAD in patients with critical limb ischemia in the Netherlands?
What is the cost effectiveness of percutaneous interventions for the treatment of PAD in patients with intermittent claudication or critical limb ischemia in the Netherlands?

A systematic literature search was performed to identify all English language studies (until 13-01-2015) of cost-effectiveness analyses comparing both the costs and consequences of treatment strategies for PAD. Studies should have focused on patients with PAD or specifically on IC or CLI. Interventions of interest were (supervised) exercise, (stented) PTA, bypass surgery, medication (symptom relief) and amputation. Studies were identified using multiple electronic databases, including PubMed publisher, Embase.com, Medline (OvidSP), Web-of-science, Cochrane (NHS EED, DARE), Google scholar, and EconLit. To ensure that all relevant studies were included, reference lists of all included studies were scanned for eligible articles as an additional check to incorporate all relevant studies. The search to identify all potentially relevant publications consisted of two parts: a disease-specific search strategy and a cost-effectiveness search strategy. The searches were based on previously published systematic reviews and the validated CRD search strategy for economic evaluations.

All records retrieved from the databases were merged into one database. After duplicates were removed, the remaining records were screened on title and abstract by two independent reviewers. After this selection the remaining studies were read in their entirety to determine which of them met all eligibility criteria. All remaining studies that fulfilled the inclusion criteria were included in this review. Various characteristics were extracted from the eligible studies by one reviewer, including: study characteristics (year of publication, first author, country, population, discount rates, time horizon, perspective and outcome measure), input parameters (costs of interventions and health utilities), the incremental outcomes, incremental cost-effectiveness ratios and conclusions. Moreover, when studies were based on a randomized controlled trial (RCT) we assessed the risk of bias. In addition, one reviewer assessed the methodological quality of the included studies using the Drummond et al. checklist which was checked by a second reviewer. The transferability of the studies to the Dutch setting was assessed using the Welte et al. checklist.

In total, we included 29 original studies on the cost-effectiveness of PAD interventions. Thirteen studies reported the cost-effectiveness of treatments for patients with IC. Eleven studies focussed on CLI and five studies focussed on the cost-effectiveness of treatments for both CLI and IC.

Based on the literature regarding treatment strategies for the treatment of PAD in patients with intermittent claudication (IC), we can conclude that SE therapy is a cost-effective alternative when compared with UE therapy. Regarding revascularization versus exercise therapy, it seems that revascularization improves health but also increases costs; whether the initial health gain justifies the additional costs is not clear from the studies. Regarding PTA versus BP, PTA is the preferred choice for femoropopliteal interventions according to the literature. PTA with selective stent placement in IC patients with iliac or femoropopliteal revascularizations seems to be cost-effective versus PTA primary stent placement. Of the specific types of endovascular treatments, drug-eluting balloon (paclitaxel) was the dominant strategy (i.e. greater effectiveness combined with cost-savings). Orbital atherectomy in combination with angioplasty was considered cost-effective versus angioplasty alone in femoropopliteal interventions; (Nd/YAG) laser-assisted PTA could also be preferred over PTA alone in femoropopliteal interventions. Concerning medication options, studies have found pentoxifylline to be less effective than cilostazol and naftidrofuryl. However, this comparison is not relevant for Dutch policymaking since cilostazol and naftidofuryl are not available in the Netherlands.

The second aim of this review was to examine the cost effectiveness of treatment strategies for the treatment of PAD in patients with critical limb ischemia (CLI) in the Netherlands. Studies have reported that BP is cost-effective compared with primary amputation. However, it is unclear whether BP or PTA is more cost-effective in patients with CLI. Orbital atherectomy in combination with balloon angioplasty and (Nd/ YAG) laser-assisted angioplasty were both considered cost-effective treatment strategies versus angioplasty alone in femoropopliteal interventions. Cool excimer laser-assisted angioplasty was more effective and more expensive than tibial balloon angioplasty but was considered cost-effective. Of the different types of endovascular treatments, drug-eluting interventions were reported to be more effective and cost-effective than the other types. PTA and BP were considered cost-effective versus PGE1.

The third and final aim of this review was to examine the cost effectiveness of percutaneous interventions for the treatment of PAD in patients with intermittent claudication or critical limb ischemia. Studies of percutaneous interventions in these patients have concluded that PTA with selective stenting is more cost-effective than primary stent placement or PTA alone. Concerning the type of endovascular approach, drug-coated balloons (paclitaxel) are the preferred type of treatment. Studies have found that using a laser may also improve the effectiveness and cost-effectiveness of PTA.

For some decision problems (e.g. supervised exercise therapy versus non-supervised exercise therapy) it was clear which strategy was the most cost-effective strategy since the conclusions of the studies were consistent. However, in most cases it was not possible to indicate which intervention is the most cost-effective solution because of important variation in methodology and conclusions between studies or because the conclusions were based on a single study. Important differences between studies can be found in methodological approaches, health care system characteristics and population characteristics. While estimates of the cost-effectiveness of an intervention can be extracted from a literature review, it is essential to consider the methodological quality and transferability of each study across geography, time, healthcare systems and patient populations. All of these must be considered together; for example, while a high-quality study performed for the Dutch setting may appear to be acceptable in quality and transferability, it may nevertheless be of limited value if it was performed too long ago.

Dutch summary

Patiënten met perifeer vaatlijden hebben vaak pijn in de bil, dijbeen of kuit tijdens het lopen (claudicatio intermittens). Men spreekt van kritieke ischemie wanneer patiënten ook pijn aan de voet of been hebben in rust en/of trofische stoornissen hebben en de systolische enkeldruk, gemeten met een dopplerapparaat, is lager dan 50 mmHg. De behandeling van patiënten met perifeer vaatlijden heeft hoofdzakelijk drie doelen: 1) het risico op hart- en vaatziekte verminderen, 2) het aantal meter lopen verbeteren en 3) de kwaliteit van leven verbeteren. Patiënten worden vaak behandeld met medicatie, looptherapie, revascularisatie of amputatie.

De bovengenoemde interventies hebben allemaal het doel om gezondheidswinst te behalen voor patiënten met kritieke ischemie of claudicatio intermittens, maar deze behandelingen leiden ook tot een toename in de kosten. Daarom is het belangrijk om de kosteneffectiviteit van deze behandelingen in kaart te brengen. Het doel van deze studie was om een uitgebreid overzicht te presenteren van de kosteneffectiviteit van alle behandelingen voor patiënten met kritieke ischemie of claudicatio intermittens. Tevens kunnen de resultaten van de studies helpen bij sommige beslisproblemen (bijv. gesuperviseerde looptraining versus niet gesuperviseerde looptraining). In het bijzonder is er gekeken naar de volgende deelvragen:

Wat is de kosteneffectiviteit van interventies voor PAV in patiënten met claudicatio intermittens?
Wat is de kosteneffectiviteit van interventies voor PAV in patiënten met kritieke ischemie van het onderbeen?
Wat is de kosteneffectiviteit van percutane interventies gefocust op patiënten met etalagebenen of kritieke ischemie?

Een systematische literatuur zoekstrategie is opgezet om alle Engelstalige artikelen te vinden die de kosteneffectiviteit presenteren van interventies voor PAV. De studies moeten origineel zijn en kosten en effecten van de interventies moeten met elkaar vergeleken worden. Interventies die in aanmerking kwamen waren: bewegingstherapie, angioplastiek, bypass chirurgie, medicatie en amputatie. Meerdere databases waren geraadpleegd. De zoek strategieën zijn gebaseerd op bestaande systematische reviews en op een gevalideerd economisch filter voor kosteneffectiviteitsanalyses. Nadat alle databases bekeken waren en de duplicaties eruit waren gehaald hebben twee reviewers onafhankelijk van elkaar de titel en de samenvatting van iedere publicatie bekeken. Hierna zijn de overgebleven artikelen volledig gelezen om te bepalen of deze geschikt waren. Van de studies die geïncludeerd waren zijn verschillende karakteristieken uit de artikelen gehaald: studie karakteristieken, input parameters, resultaten, kwaliteit van de studies, de kwaliteit van de RCT’s waarop de studies gebaseerd waren en de transferabiliteit van de resultaten.

In totaal bleven 29 studies die de kosteneffectiviteit van twee of meerdere PAV interventie evalueerden van de 4318 referenties over. Dertien studies rapporteerden over de kosteneffectiviteit van een interventie voor claudicatio intermittens, elf studies voor kritieke ischemie en vijf studies focusten op beide populaties. Gemiddeld genomen waren deze studies van een goede kwaliteit.

Op basis van de literatuur over behandelingsstrategieën voor de behandeling van PAD patiënten met claudicatio intermittens (IC), kunnen we concluderen dat SE therapie kosteneffectief is vergeleken met de UE therapie. Ten aanzien van revascularisatie versus oefentherapie, lijkt het erop dat revascularisatie tot zowel gezondheidswinst als kostenstijging leidt; of de gezondheidswinst de extra kosten rechtvaardigt is niet duidelijk uit de studies. Bij de vergelijking van PTA versus BP, krijgt PTA de voorkeur voor femoropopliteale interventies volgens de literatuur. PTA met selectieve stentplaatsing bij IC-patiënten met iliacale of femoropopliteale revascularisaties lijkt kosteneffectief versus PTA primaire stentplaatsing. Van de specifieke soorten van endovasculaire behandelingen, was drug-eluting ballon (paclitaxel) de dominante strategie (omdat het tot gezondheidswinst en kostenbesparing leidt). Orbital atherectomie in combinatie met angioplastiek werd beschouwd als kosteneffectief versus angioplastie alleen in femoropopliteale interventies; (ND / YAG) laser-assisted PTA was ook kosteneffectief versus PTA alleen in femoropopliteale interventies. Met betrekking tot de medicinale opties, hebben studies gevonden dat pentoxyfilline minder effectief is dan cilostazol en naftidrofuryl. Echter, deze vergelijking is niet relevant voor de Nederlands situatie omdat cilostazol en naftidofuryl niet beschikbaar zijn in Nederland.

Het tweede doel van dit onderzoek was om de kosteneffectiviteit van de behandeling strategieën voor de behandeling van PAD bij patiënten met kritieke ischemie (CLI) in Nederland te onderzoeken. Studies hebben gemeld dat BP kosteneffectief is in vergelijking met primaire amputatie. Het is echter onduidelijk of BP of PTA kosteneffectiever zijn bij patiënten met CLI. Orbital atherectomie in combinatie met ballon angioplastiek en (ND / YAG) laser-assisted angioplastie werden beide beschouwd als kosteneffectieve behandelstrategieën versus angioplastie alleen in femoropopliteale interventies. Cool excimer laser-assisted angioplastie was effectiever en duurder dan tibiale ballonangioplastiek maar werd toch als kosteneffectief beshouwd. Van de verschillende typen endovasculaire behandelingen, werden drug-eluting interventies gezien als effectiever en kosteneffectiever dan de andere types. PTA en BP werden beschouwd als kosteneffectief versus PGE1.

Het derde en laatste doel van dit onderzoek was om de kosteneffectiviteit van percutane interventies voor de behandeling van PAD bij patiënten met claudicatio intermittens of kritische ischemie van de ledematen te onderzoeken. Studies van percutane interventies bij deze patiënten hebben geconcludeerd dat PTA met selectieve stenting kosteneffectief is vergeleken met primaire stentplaatsing of PTA alleen. Wat betreft de aard van de endovasculaire benadering, zijn drug-coated ballonnen (paclitaxel) de meest voorkomende vorm van behandeling. Studies hebben aangetoond dat gebruik van een laser de effectiviteit en kosteneffectiviteit van PTA kan verbeteren.

Bij sommige beslisproblemen zijn de conclusies van de studies consistent, zoals bij gesuperviseerde looptherapie voor patiënten met claudicatio intermittens, welke in elke studie kosteneffectief is gebleken wanneer deze vergeleken is met niet gesuperviseerde looptherapie. In de meeste gevallen was het niet mogelijk om aan te geven welke interventie kosteneffectief zijn vanwege belangrijke variatie in methodologie en conclusies tussen studies of omdat de conclusies zijn gebaseerd op een enkele studie. Belangrijke verschillen tussen de studies kunnen worden gevonden in de methodologische aanpak, kenmerken van het gebruikte zorgsysteem en kenmerken van de patiëntenpopulatie. Terwijl schattingen van de kosteneffectiviteit van diverse interventies uit een literatuurstudie verkregen kunnen worden, is het essentieel om rekening te houden met de methodologische kwaliteit en de ‘transferability’ van elke studie ten aanzien van het land, de periode (jaar van uitvoering), de gezondheidszorg en de patiëntenpopulatie. Alle factoren moeten worden bekeken; bijvoorbeeld, hoewel een hoogwaardig onderzoek uitgevoerd voor de Nederlandse setting in eerste instantie ongetwijfeld acceptabel lijkt, kan het echter van beperkte waarde zijn als het te lang geleden is uitgevoerd.

Introduction

Patients with lower limb peripheral arterial disease (PAD) often have pain in their leg after walking (intermittent claudication). When the pain persists in rest and/or trophic skin changes are present and the systolic blood pressure measured at the ankle by a Doppler ultrasound device is lower than 50mmHg, patients are diagnosed with critical limb ischemia (CLI) ¹. Patients with CLI are at risk of irreversible ischaemic damage to the leg or foot and have a very high mortality risk in the first year after diagnosis (50%) ². The incidence of PAD increases with age, and people who smoke, people with diabetes, and people with coronary artery disease are at higher risk of PAD. The prevalence of PAD has been estimated at 20% for people aged over 60 years ³.

Treatment of PAD patients has three aims: 1) decreasing cardiovascular disease risk, 2) improving walking distance ² and 3) improving quality of life. For the first aim, patients often receive advice to improve their lifestyle (e.g. quit smoking or improve physical activity). Patients often receive medication (e.g. cilostazol, naftidrofuryl oxalate, pentoxifylline or inositol nicotinate), exercise therapy or a revascularization (e.g. bypass surgery (BP), percutaneous transluminal angioplasty (PTA) or PTA in combination with stenting) to improve the walking distance without pain. When CLI patients develop irreversible damage to the leg or foot, they may require amputation.

The interventions named above all have the aim to improve health outcomes of patients with CLI or IC, but may also lead to additional healthcare costs. Therefore, it is necessary to consider the cost-effectiveness of these interventions. Many studies have evaluated the cost-effectiveness of these interventions for IC or CLI or both indications. Recently, three systematic reviews^2,4,5 were performed summarizing the cost-effectiveness of treatment strategies for patients with PAD. However, these focussed on a specific indication (CLI or IC), on a specific type of intervention (PTA) or did not assess the quality of the studies and the transferability of the studies.

The aim of this systematic review was to comprehensively summarize the cost-effectiveness of all interventions for patients with CLI or IC. In addition, the quality and the transferability of the studies for the Dutch setting were assessed. The results are presented to address three different questions about the cost-effectiveness of existing treatments:

What is the cost effectiveness of treatment strategies for the treatment of PAD in patients with intermittent claudication in the Netherlands?
What is the cost effectiveness of treatment strategies for the treatment of PAD in patients with critical limb ischemia in the Netherlands?
What is the cost effectiveness of percutaneous interventions for the treatment of PAD in patients with intermittent claudication or critical limb ischemia in the Netherlands?

Methods

The process of identifying studies that met inclusion criteria is presented in Figure 1. This systematic review adhered to the PRISMA guidelines ⁶ for reporting of systematic reviews.

Search strategy

A systematic literature search was performed to identify all English language studies (until 13-01-2015) of full economic evaluations comparing both the costs and consequences of treatment strategies for PAD. Studies should have focused on patients with PAD or specifically on IC or CLI. Interventions of interest were (supervised) exercise, (stented) PTA, BP, medication (symptom relief) and amputation. This implies that interventions aimed at preventing PAD, such as clopidogrel, and studies focussing on diagnostic interventions were excluded. The effectiveness outcomes of the studies had to be expressed in quality adjusted life years (QALYs), life years (LYs) or disease specific outcomes [e.g. maximum walking distance (MWD) or amputation free survival (AFS)]. There was no restriction on the perspective that was used in the economic evaluation and no restriction was placed on the date of publication. Reviews, editorials and abstracts were not included in this systematic review. Studies were identified using multiple electronic databases, including PubMed publisher, Embase.com, Medline (OvidSP), Web-of-science, Cochrane (NHS EED, DARE), Google Scholar, and EconLit. To ensure that all relevant studies were included, reference lists of all included articles were scanned for additional eligible articles as an additional check that all relevant studies had been identified. The search for all potentially relevant publications consisted of two parts: a disease specific search strategy and a cost-effectiveness search strategy. The searches were based on previously published systematic reviews ^4,5 and on the validated CRD search strategy for economic evaluations ⁷. The full search strategies for all databases are presented in Appendix 1.

Selection

All records retrieved from the databases were merged into one database. After duplicates were removed, the remaining records were screened on title and abstract. Studies were excluded based on study design, intervention, comparator or language. Studies were only included if they reported original economic evaluations comparing both costs and effects of two or more strategies. Studies were excluded when only costs and outcomes of the interventions were reported and no comparison with alternative treatments in terms of incremental cost-effectiveness ratio (ICER) was made. This title/abstract selection was performed by two independent reviewers (LB & MP). The results of both reviewers were compared; any discrepancies were discussed and resolved by consensus or by a third reviewer (WR). After this selection the remaining studies were read entirely to find the studies that fully met eligibility criteria. Again, studies were excluded based on study design, intervention or comparator. The full assessment selection was also performed by two reviewers (LB & ML) and the results were discussed and resolved by consensus. Those articles which fulfilled the inclusion criteria were included in this review. After identifying all relevant publications we divided all publications into categories based on relevance for the three specific decision problems. Some publications were eligible for more than one decision problem.

Data extraction

Various characteristics were extracted from the included studies by one reviewer (LB), including study characteristics (year of publication, first author, country, population, discount rates, time horizon, perspective and outcome measure), input parameters (costs of interventions and health utilities), the incremental outcomes, incremental cost-effectiveness ratios and conclusions. Moreover, when studies were based on a randomized controlled trial (RCT) we incorporated the risk of bias ⁸ that was previously assessed in a review of the clinical effectiveness of the treatments for PAD. In addition, one reviewer (LB) assessed the methodological quality of the included studies using the Drummond et al. ⁹ checklist (Appendix 2). This checklist contains thirty-five questions concerning the validity of economic evaluations. The results were checked by a second reviewer (ML).

Transferability

Since the primary focus of this review was on the potential cost-effectiveness of PAD treatments in the Netherlands, we assessed the transferability of the results of the included studies to the Netherlands. To achieve this, we used the Welte et al. ¹⁰ checklist, which includes three types of transferability characteristics: 1) methodological characteristics (i.e. perspective, discounting, medical cost approach and productivity costs approach), 2) healthcare system characteristics (i.e. absolute and relative prices in healthcare, practice variation and technology availability), and 3) population characteristics (i.e. disease incidence/prevalence, case-mix, life expectancy, health status preferences, acceptance/compliance/incentives to patients, productivity and work-loss time and disease spread). Prior to using this tool, we determined the relevance (range: very low - very high) of all these characteristics for the interventions. This led us to eliminate the characteristics of ‘disease incidence/prevalence’ and ‘disease spread’ as these are especially applicable to interventions focussing on prevention, which are outside the scope of our review. We then determined the transferability of the methodological characteristics by estimating the degree of correspondence (range: very low – very high) between the methods used in a particular study and the methods proposed by the Dutch guidelines on pharmacoeconomic research (i.e., the guidelines published in the country of interest). Moreover, the likely effect of the transferability factor on the ICER (e.g. unbiased) was assessed. Studies were considered as very high in correspondence if they were performed using the methods described in the Dutch guidelines on pharmacoeconomic research ¹¹. Lastly, in cooperation with the clinical experts, we determined the degree of correspondence between the health care system and population characteristics considered in a particular study and what should be considered in the Netherlands; the impact of these comparisons on the ICER was then assessed.

Figure 1: Overview methods

Overview methods PAV

Results

Figure 2 describes the process ⁶ of identifying relevant publications. Of the 4318 potentially relevant publications, 4127 were excluded based on title, abstract and keywords. Full-text evaluation was performed for 191 articles leading to 30 relevant studies. Reasons to exclude studies in this phase were: no original cost-effectiveness analysis (n=86), language other than English (n=3), intervention (n=38), population (n=13) and unavailability of full text publication [including studies that were only presented in abstract form (n=21)]. Two additional records were found using the snowball technique, which resulted in a total of 32 eligible publications describing the cost-effectiveness analyses of two or more interventions for PAD. However, in three cases, we found that a full report ^4,5,12 and a paper ^13-15 reported results from the same analyses; in each case, we extracted data from the full report ^4,5,12. Consequently, we included 29 original, unique studies reporting the cost-effectiveness of PAD interventions.

Thirteen studies reported the cost-effectiveness of treatments for patients with IC ^5,12,16-26. Eleven studies focussed on CLI ^27-37 and five studies focussed on the cost-effectiveness of treatments for both CLI and IC ^4,38-41.

Most studies were performed in the UK (38%) or US (38%). Three studies (10%) were conducted in the Netherlands ^17,20,21. The most commonly used perspective was the societal perspective (34%), followed by the National Health Service perspective (31%). In 17 (59%) of the 29 studies a model was used to estimate the costs and effects, often to extrapolate short term outcomes. In 10 studies (34%), a lifetime perspective was chosen. The discount rates varied from 3% - 6%, and the rate used was influenced by the country where the study was conducted. Most studies were of a reasonable quality; on average the studies fulfilled 77% of the Drummond criteria. Transferability, study limitations and the overall scores of the studies are presented in Tables 5-8 in Appendix 3.

Figure 2: PRISMA flowchart ⁶

PRISMA flowchart PAV

Of the 29 included studies, 18 studies reported on the cost-effectiveness of treatment strategies for intermittent claudication (Appendix 3 - Table 1, Table 5, Table 6). The treatments that were evaluated in the studies were (supervised) exercise, revascularization and medication.

Most of the included studies on IC evaluated the cost-effectiveness of (stented) revascularization ^{4,12,17-19,21,22,25,26,38-41}. The cost-effectiveness of (supervised) exercise therapy was evaluated in nine studies^{12,16-20,24-26}. Six studies evaluated (supervised) exercise versus revascularization^{12,17-19,25,26}. Furthermore, medication (e.g. cilostazol) was evaluated in two studies ^5,23. Most studies were conducted in the UK and US. Three studies evaluated treatment options for Dutch patients with IC ^17,20,21. The strategies that were compared in these three studies were: 1) UE therapy versus SE therapy, 2) different PTA strategies, and 3) exercise versus PTA.

Supervised exercise versus unsupervised exercise

Five studies compared SE therapy with UE therapy. One study was performed in the Netherlands using the societal perspective which estimated the one-year cost-effectiveness of SE therapy compared with UE therapy (verbal instructions about walking) based on the EXITPAD trial and concluded that SE therapy is likely to be a cost-effective therapeutic option for patients with claudication (ICER: €28,693/QALY gained)²⁰.

In the NICE clinical guideline 147, appendix L, ¹² the lifetime costs and effects of SE therapy followed by SE therapy (SE-SE) and UE therapy followed by SE therapy (UE-SE) was compared in aorto-iliac artery: SE-SE therapy led to an increase in QALYs (4.447 vs 4.355) and a cost increase of £217 (£4,155 vs £3,938), leading to an ICER of £2,359 per QALY gained compared with UE-SE therapy. In the femoro-popliteal artery group SE-SE therapy led to an increase in both QALYs (0.092) and costs (£217), leading to an ICER of £2,359 per QALY gained. Another analysis using the NHS perspective¹⁶ found that SE therapy led to a cost increase and QALY gain, leading to an ICER ranging from £711- £1,608 per QALY gained. Lee et al. ²⁴ also estimated a slightly higher but nevertheless acceptable ICER (£1,780) for SE therapy. However, the study had a time horizon of only one year and was based on a single non-randomized clinical study. Reynolds et al. ²⁶ estimated an ICER of $24,070 for SE therapy versus UE therapy using decision modelling techniques, which was considerably higher than what was estimated in the NICE clinical guideline^12,16 and the study conducted by Lee et al. ²⁴.

Based on these studies, it seems likely that SE therapy is a cost-effective solution when compared with UE therapy. When we specifically focus on the study that was conducted using the Dutch setting²⁰ this conclusion remains the same. However, one should keep in mind that this study had a relatively short time horizon of 1 year and did not capture long term compliance.

Revascularization versus exercise therapy

Six studies reported on the cost-effectiveness of exercise therapy compared with revascularization. One study ¹⁷ evaluated the cost-effectiveness of supervised exercise therapy and PTA for the Netherlands. Spronk et al. ¹⁷ estimated the one-year cost-effectiveness of PTA versus supervised exercise therapy based on the CETAC trial and concluded that the small incremental QALY gain of PTA with selective stent placement did not offset the initially higher treatment cost (ICER: €231,800 per QALY gained). Another study ¹⁹, performed from the US societal perspective, combined US effectiveness data with Dutch effectiveness data (e.g. Dutch iliac stent trial study) using decision modelling techniques to estimate the lifetime cost-effectiveness of PTA compared with SE therapy alone. They concluded that the cost-effectiveness ratio ($38,000/QALY gained) of PTA was within “the range of those reported for currently accepted technologies”. Furthermore, they also concluded that the small gain in effectiveness achieved with bypass surgery for IC does not justify the additional costs.

Based on a RCT, Mazari et al. ²⁵ estimated the one-year cost-effectiveness of three strategies to treat patients with IC (PTA, SE therapy, PTA + SE therapy) and concluded that SE therapy is the most cost-effective first-line treatment for IC, and, when combined with PTA, is more cost-effective than PTA alone.

When developing the NICE clinical guideline 147, appendix L, ¹², researchers used decision modelling techniques to estimate the lifetime cost-effectiveness of 13 treatment strategies, combining SE therapy, UE therapy, PTA with selective stent and PTA with primary stent placement. In this study, SE therapy followed by PTA with selective stent placement for people with persistent or worsening claudication was found to be the most cost-effective sequence of treatments for people with IC in the aorto-iliac and femoro-popliteal artery. Reynolds et al. ²⁶ estimated the five-year cost-effectiveness of stenting, SE therapy and optimal medical care using decision modelling techniques and concluded that stenting resulted in higher QALYs and was more expensive than SE therapy. However, the ICER of $122,600 per QALY gained was unacceptably high when compared with the threshold of $30,000 per QALY gained. Treesak et al. ¹⁸ estimated a larger improvement in absolute (ACD) and intermittent claudication distance (ICD) for patients treated with SE therapy compared with patients treated with PTA for a one year time horizon. Furthermore, SE therapy was less expensive than PTA ($4,968 versus $9,303) and consequently PTA was dominated by SE therapy.

In conclusion, it seems that a revascularization leads to more effects in terms of QALYs but also to higher costs when compared with exercise therapy. While most studies concluded that the initial higher costs of revascularizations were not offset by the incremental QALY gain it yielded and thus not cost-effective, other studies concluded that PTA was in fact a cost-effective alternative to exercise.

The Dutch study by Spronk et al¹⁷ concluded that the small incremental QALY gain of PTA with selective stent placement did not offset the initially higher treatment cost. However, there are three points to note regardingthis study: it had a relatively short time horizon of 1 year, it was published in 2008 and it was based on a single centre RCT of patients fulfilling strict eligibility criteria. These factors should be considered when assessing its relevance in assessing the potential cost-effectiveness of PTA with selective stent placement in the Netherlands.

Revascularization versus revascularization

Eight studies compared different types of revascularizations. Since the effectiveness of a revascularization may vary depending on the location of the revascularization the results are categorized by location.

Iliac

Two studies estimated the cost-effectiveness of revascularizations performed in the iliac artery in patients with IC. Bosch et al. ²¹ estimated the cost-effectiveness of PTA, PTA with selective stent placement and primary stent placement for Dutch patients using decision modelling techniques. PTA with selective stent placement was considered as a cost-effective solution compared with primary stent replacement or PTA alone from the Dutch societal perspective. Bosch et al. ²² compared the lifetime costs and effects of PTA and PTA with selective stent placement using decision modelling techniques and concluded that PTA with selective stent placement was a more cost-effective treatment for the US (societal perspective). It should be noted that the clinical data used in these two cost-effectiveness studies were based on the same Dutch estimates, which could be the reason for the comparable results.

Femoropopliteal

Hunink et al. ³⁸ used decision modelling techniques to evaluate the lifetime cost-effectiveness of PTA and BP for IC and CLI and concluded that, from the US healthcare system perspective, PTA was the preferred initial treatment in patients with disabling claudication and a stenosis or occlusion. Muradin et al. ³⁹ also compared BP with PTA using decision modelling techniques and also concluded that PTA was more cost-effective than BP from the US healthcare system perspective.

NICE clinical guideline 147, Appendix L, ¹² used decision modelling techniques to estimate the lifetime cost-effectiveness from the UK NHS perspective of 13 treatment strategies, combining UE therapy, SE therapy, PTA with selective stent and PTA with primary stent. PTA with primary stenting was on average less effective and more expensive than PTA with selective stenting. Simpson et al. ⁴ performed an economic evaluation using decision modelling techniques of several endovascular approaches to treat IC: (drug-eluting) stents and balloons, stent-graft, brachytherapy and cryoplasty. The lifetime cost-effectiveness in that study was estimated from the UK NHS perspective. A drug-coated balloon (paclitaxel) was more effective in terms of late failure (loss patency) and return of symptoms than most strategies and was less expensive compared with most of the other strategies. The difference in intervention costs with the least expensive strategy was only £410. Consequently, the drug-coated balloon (paclitaxel) treatment was considered as the dominant strategy since it had the highest lifetime effectiveness (6.120 QALY) and lowest lifetime costs (£12,668).

Weinstock et al. ⁴⁰ evaluated two types of PTA: 1) orbital atherectomy and balloon angioplasty, and 2) balloon angioplasty. The one-year cost-effectiveness conducted from the US third party payer perspective was based on the COMPLIANCE study with a small sample size. The atherectomy intervention was more effective and more expensive but was still considered cost-effective with an ICER of $3,441 per QALY gained. Sculpher et al. ⁴¹ used decision modelling techniques to estimate the 25-year cost-effectiveness of laser-assisted (Nd/YAG) PTA from the UK NHS perspective. Laser assisted PTA resulted in an incremental QALY gain of 0.09 and a cost increase of £260 compared with PTA alone, leading to an acceptable ICER.

In conclusion, when a decision needs to be made between PTA and BP, PTA is the preferred choice for femoropopliteal interventions. PTA with selective stent placement in IC patients with iliac or femoropopliteal revascularizations could be considered as cost-effective compared with primary stent placement. Of the specific types of endovascular treatments, use of a drug-eluting balloon (paclitaxel) was the most effective and dominant strategy (i.e. greater effectiveness combined with cost-savings) amongst femoropopliteal interventions. However, this conclusion was based on one single study. Orbital atherectomy in combination with angioplasty was considered more cost-effective than angioplasty alone in femoropopliteal interventions. Furthermore, (Nd/YAG) laser-assisted PTA could also be preferred over PTA alone in femoropopliteal interventions. Any conclusions regarding the cost-effectiveness of the orbital atherectomy and (Nd/YAG) laser assisted angioplasty can only be based on just one cost-effectiveness study each. Moreover, the clinical studies used in these cost-effectiveness studies had small sample sizes. Please note that the (Nd/YAG) laser assisted PTA is not available anymore in the Netherlands.

The only study ²¹ that could be considered transferable and of sufficient quality concluded that PTA with selective stent placement in an iliac artery was a cost-effective solution when compared with PTA with primary stenting and PTA alone.

Medication

Two studies have evaluated the cost-effectiveness of pentoxifylline, cilostazol, and naftidrofuryl oxalate. Neither of these studies was performed for the Netherlands and both studies concluded that pentoxifylline was the least effective treatment. Guest et al. ²³ concluded that cilostazol was the most effective treatment in terms of maximum walking distance. Cilostazol was more expensive than naftidrofuryl but less expensive when compared with pentoxifylline. However, Squires et al. ⁵ concluded that naftidrofuryl was the most effective treatment and the most cost-effective treatment. Both studies estimated the cost-effectiveness of these interventions from the UK NHS perspective using decision modelling techniques. However, Squires et al. ⁵ used a time horizon of 100 years (and thus a lifetime), which is considerably longer than the 24 weeks time horizon of Guest et al. ²³.

In conclusion, studies have concluded that cilostazol and naftidrofuryl are more effective than pentoxifylline. However, it is not clear if cilostazol or naftidrofuryl is the most cost-effective treatment. Nevertheless, this uncertainty has no practical implications for policymaking in the Netherlands since cilostazol and naftidrofuryl are not available there.

Cost-effectiveness of interventions for patients with critical limb ischemia

Sixteen studies evaluated the cost-effectiveness of treatment strategies for patients with CLI (Appendix 3 - Table 2, Table 7, Table 8). Patients were treated with an amputation, revascularization or medication. In five studies primary amputation was compared with revascularization (e.g. BP or PTA)^{27,28,30,36,37}. All sixteen studies included a revascularization strategy in their economic evaluation. Holler et al. ³² compared the cost-effectiveness of 16 strategies combining medication, revascularization and no treatment. None of the studies evaluating the cost-effectiveness of CLI treatment were performed for the Dutch setting.

Amputation versus revascularization

Five studies evaluated the cost-effectiveness of primary amputation compared with a revascularization^{27,28,30,36,37}. The revascularizations in these studies were focused on the infra-inguinal area and on tibial-peroneal artery occlusive disease.

Infra-inguinal area

Barshes et al. ²⁸ concluded that primary amputation resulted in fewer ambulatory years than BP but was also the least expensive treatment in patients with absence of great saphenous vein conduit. Several BP treatment strategies were compared with amputation using decision modelling techniques; BP was both more effective and more expensive than amputation. BP with autologous alternative vein (AAV), the most effective strategy, was estimated to have an ICER of $5,475 per ambulatory year gained when compared with amputation. Barshes et al. ²⁷ concluded that BP and an endovascular approach were more effective and more expensive than amputation. Moreover, BP was considered cost-effective compared with primary amputation ²⁷. Barshes et al. ³⁷ and Barshes et al. ³⁶ also evaluated the cost-effectiveness of primary amputation compared with revascularization (i.e. bypass or endovascular) and both studies concluded that primary amputation was both less effective (in terms of limb years and ambulatory years) and more expensive than revascularization. However, these two studies focussed on patients with end stage renal disease and patients with marginal functional status. The results of the above-mentioned studies were based on a decision analytical model which estimated the cost-effectiveness of treatments using a 10-year time horizon and a societal perspective.

Tibial-peroneal

Another study, Brothers et al. ³⁰, concluded that BP was more effective and more expensive than primary amputation after a 5 year period. The cost-effectiveness ratio was relatively low (ICER of $3,545 per QALY gained). This study used decision modelling techniques to estimate the cost-effectiveness from an institutional perspective.

In conclusion, revascularization (i.e. bypass or endovascular) was more effective than primary amputation for revascularizations in the infra-inguinal area and for tibial-peroneal artery occlusive disease. The published studies concluded that a revascularization can be considered a cost-effective treatment compared with primary amputation.

Revascularization versus revascularization (comparison of revascularization types)

Fourteen studies evaluated different types of revascularizations for patients with CLI, either in the infra-inguinal area and on femoral, popliteal, tibial revascularizations.

Infra-inguinal area

Bradbury et al. ²⁹ estimated an ICER of BP versus PTA: £134,257 per QALY gained which can be considered as too high. The cost-effectiveness from the UK NHS perspective was based on the BASIL trial using a time horizon of three years. The same BASIL trial data (where UK NHS costs were converted to US dollars) was used by Forbes et al in an analysis with a three year time horizon. ³¹ Like Bradbury et al., ²⁹ they also concluded that the incremental QALY gain (0.03) of BP was small relative to the initial higher treatment costs ($5,521), resultingin an unacceptably high ICER.

Barshes et al. ²⁷ concluded that a BP with endovascular revisions was the most cost-effective approach using a decision analytic model with a time horizon of 10 years. BP strategies were more effective and less expensive than endovascular approaches. However, Barshes et al. ³⁷ and Barshes et al. ³⁶ both concluded that endovascular strategies were more effective and less expensive (i.e. dominant) than BP strategies. However, it should be noted that these two studies focussed on patients with end stage renal disease and patients with marginal functional status.

Femoral-popliteal

Hunink et al. ³⁸ evaluated the lifetime cost-effectiveness of PTA and BP for IC and CLI from the US healthcare system perspective. They found that in patients with CLI and occlusion, BP is the preferred initial treatment. Muradin et al. ³⁹ also evaluated the lifetime cost-effectiveness of BP versus PTA and concluded that BP was the treatment of choice.

Based on a non-randomized clinical study, Sultan et al. ³⁴ estimated the three-year cost-effectiveness of subintimal angioplasty compared with BP and found that subintimal angioplasty was both more effective and expensive; they estimated an ICER of €10,768 per QALY gained.

Using effectiveness data from an RCT, Weinstock et al. ⁴⁰ compared the one-year cost-effectiveness of two types of angioplasty: 1) orbital atherectomy and balloon angioplasty, and 2) balloon angioplasty. While the atherectomy intervention was more effective and also more expensive, it was still considered cost-effective with an ICER of $3,441 per QALY gained. Sculpher et al. ⁴¹ estimated the 25-year cost-effectiveness of (Nd/YAG) laser-assisted PTA versus PTA alone and found that (Nd/YAG) laser-assisted PTA led to an incremental QALY gain of 0.09 and a cost increase of £260 compared with PTA alone, leading to an acceptable ICER.

Simpson et al. ⁴ reported that drug-coated (paclitaxel) balloons were less expensive and more clinically effective over a lifetime time horizon than all of the other endovascular options and therefore dominated all other options. The drug-coated balloon was more effective in terms of late failure (loss patency) and return of symptoms than most strategies and was less expensive compared with most of the other strategies. Katsanos et al. ³³ evaluated the lifetime cost-effectiveness of two treatment decisions: 1) bail-out sirolimus eluting stent (SES) versus bail-out bare metal stent (BMS), 2) primary everolimus eluting stent (EES) versus bail-out BMS. They found that the treatment options with drug-eluting stents (DES) were more effective and cost-effective since the ICERs were acceptable (1: €6,518 per event-free year, 2: €11,581 per event-free year).

Tibial

Sultan et al. ³⁵ compared two types of angioplasty and found that the cool excimer laser-assisted angioplasty was more effective and more expensive than tibial balloon angioplasty after a period of three years. With an ICER of €2,073 per QALY gained the cool excimer laser-assisted angioplasty was considered cost-effective.

No specific area

Holler et al. ³² evaluated the five year cost-effectiveness of sixteen treatment strategies for patients with CLI by combining different interventions, including BP, no treatment, prostaglandin E1 (PGE1), and PTA. Repeated use of BP was the most effective strategy. Compared with repeated use of PTA, repeated use of BP was associated with an ICER of €5,654 per QALY gained.

In conclusion, it is unclear whether BP or PTA is more cost-effective in patients with CLI. Orbital atherectomy in combination with balloon angioplasty and (Nd/YAG) laser-assisted angioplasty compared with angioplasty alone were both cost-effective treatment strategies in femoral-popliteal interventions. However, it should be noted that (Nd/YAG) laser assisted PTA is not available anymore in the Netherlands. Cool excimer laser-assisted angioplasty was also more effective and more expensive than tibial balloon angioplasty and thus considered cost-effective. Subintimal angioplasty in femoral-popliteal lesions is more effective and more expensive but considered cost-effective compared with BP. The conclusions of the orbital atherectomy, laser assisted angioplasty and subintimal angioplasty were based on one single study. Of the specific types of endovascular treatments, drug-eluting interventions were more effective and cost-effective. However, these conclusions were not confirmed in multiple studies.

Medication versus revascularization

Holler et al. ³² is the only study that evaluated medication for patients with CLI. The cost-effectiveness from the German societal perspective of several strategies was estimated using a decision analytical model. PGE1 compared with repeated use of PTA led to an ICER of €23,797 per QALY gained. However, this positive ICER was only observed because of PGE1 was led to reduced costs and effectiveness after a period of five years. The same was true for PGE1 versus repeated use of BP: the positive ICER of €10,450 per QALY gained was observed because PGE1 reduced costs and health benefits.

In conclusion, PGE1 was less effective and less expensive than PTA and BP. PTA and BP were both considered cost-effective compared with PGE1.

Cost-effectiveness of percutaneous interventions for patients with PAD

Seven studies have focussed solely on the cost-effectiveness of specific types of percutaneous interventions for patients with PAD (Appendix 3, Table 3).

Bosch et al. ²¹ used a decision model to estimate the cost-effectiveness of PTA, PTA with selective stent placement and primary stent placement for Dutch patients. PTA with selective stent placement was considered as a cost-effective solution compared with primary stent replacement or PTA alone. Bosch et al. ²² compared the lifetime costs and effects of PTA and PTA with selective stent placement for the US and concluded that PTA with selective stent placement was a more cost-effective treatment. Simpson et al. ⁴ performed an economic evaluation from the UK NHS perspective of several endovascular approaches to treat IC and CLI: (drug-eluting) stents and balloons, brachytherapy, stent-graft, and cryoplasty. Drug-coated balloons (paclitaxel) were considered as the dominant strategy since they yielded the highest lifetime effectiveness and lowest costs. Weinstock et al. ⁴⁰ estimated the cost-effectiveness from an US third party payer perspective of two types of angioplasty: 1) orbital atherectomy and balloon angioplasty, and 2) balloon angioplasty. Atherectomy was more effective and more expensive but was still considered cost-effective with an ICER of $3,441/QALY gained. Sculpher et al. ⁴¹ estimated the 25-year cost-effectiveness of (Nd/YAG) laser-assisted angioplasty using a decision analytical model and found that (Nd/YAG) laser-assisted angioplasty resulted in an incremental QALY gain of 0.09 and a cost increase of £260, leading to an acceptable ICER. Sultan et al. ³⁵ also estimated the cost-effectiveness of two types of angioplasty. The cool-excimer laser-assisted angioplasty was after a period of three years more effective and expensive than tibial balloon angioplasty and led to an ICER of €2,073 per QALY gained. Katsanos et al. ³³ evaluated the lifetime cost-effectiveness of two treatment decisions: 1) bail out SES versus bail out BMS, 2) primary EES versus bail out BMS. The treatment options with DES were more effective and the ICERs were acceptable (1: €6,518 per event-free year, 2: €11,581 per event-free year).

In conclusion, the literature suggests that it is likely that PTA with selective stenting is more cost-effective than primary stent replacement or PTA alone. Concerning the type of endovascular approach, drug-eluting balloons are the preferred type of treatment from a cost-effectiveness standpoint. However, this conclusion was based on one single study adopting the UK NHS perspective. Studies have found that using a laser may also improve the effectiveness of patients undergoing PTA as well as be cost-effective. Please note that the (Nd/YAG) laser assisted PTA is not available anymore in the Netherlands. Only one study focussed on the Netherlands²¹ and this study concluded that PTA with selective stent placement was cost-effective compared with primary stent replacement or PTA alone.

Discussion

This systematic review evaluated the cost-effectiveness of several interventions for the treatment of patients with CLI or IC. In total, we found 29 studies evaluating the cost-effectiveness of interventions for patients with PAD. Revascularization, exercise therapy and medication were interventions that were included in the cost-effectiveness analyses focussing on IC. The cost-effectiveness studies for CLI patients evaluated revascularization, medication and primary amputation.

Several conclusions about the cost-effectiveness of the various interventions can be drawn based on this literature review. The following paragraphs describe the overall conclusions that can be drawn for each of the three aims of this literature review.

The first aim of this review was to examine the cost effectiveness of treatment strategies for the treatment of PAD in patients with intermittent claudication (IC) in the Netherlands. When we examine the literature concerning the IC population, we can conclude that SE therapy is a cost-effective alternative when compared with UE therapy. Regarding studies comparing revascularization with exercise therapy, it seems that revascularization leads to a QALY gain but also to higher costs; whether the initial health gain justifies the additional costs is not clear from the studies. Regarding the choice between PTA and BP, PTA is the preferred choice for femoropopliteal interventions according to the included studies. PTA with selective stent placement in IC patients with iliac or femoropopliteal revascularizations seems to be a cost-effective strategy when compared with PTA primary stent placement. Of the specific types of endovascular treatments, drug-eluting balloon (paclitaxel) was the dominant strategy (i.e. greater effectiveness combined with cost-savings). Orbital atherectomy in combination with angioplasty was considered more cost-effective than angioplasty alone in femoropopliteal interventions. Furthermore, (Nd/YAG) laser-assisted PTA could also be preferred over PTA alone in femoropopliteal interventions. Concerning medication, pentoxifylline was less effective than cilostazol and naftidrofuryl. However, a review of the literature does not help to determine whether cilostazol or naftidrofuryl is more cost-effective treatment for patients with IC. However, this uncertainty is a moot point for Dutch policymaking since both medications are not available in the Netherlands.

The second aim of this review was to examine the cost effectiveness of treatment strategies for the treatment of PAD in patients with critical limb ischemia (CLI) in the Netherlands. Here we can conclude that studies have reported that BP was both more effective and more expensive than primary amputation; perhaps more importantly, these studies have also concluded that BP is cost-effective compared with primary amputation. However, it is unclear whether BP or PTA is more cost-effective in patients with CLI. Orbital atherectomy in combination with balloon angioplasty and (Nd/ YAG) laser-assisted angioplasty were both considered cost-effective treatment strategies versus angioplasty alone in femoropopliteal interventions. Cool excimer laser-assisted angioplasty was more effective and more expensive than tibial balloon angioplasty but was considered cost-effective. Of the specific types of endovascular treatments, drug-eluting interventions were more effective and cost-effective than the other types. PTA and BP were considered cost-effective versus PGE1.

Strengths

This study is the first systematic review of all possible treatment options for patients with PAD, focusing on both CLI and CI patients. As mentioned above, other literature reviews of the cost-effectiveness of treatments of PAD have been done before ^2,4,5. However, those reviews only focussed on specific types of interventions for PAD (e.g., medication) while our review was broader in scope since we included all possible treatments for PAD. Despite these differences in scope, the general findings of our review corresponded well with the findings of the previous reviews. Another strength of our review compared to previous reviews was our assessment of the quality and transferability of cost-effectiveness studies. In addition, this systematic review was thorough since it was based on a combination of validated searches of seven important databases. Furthermore, we increased the specificity of the literature search by expanding our search algorithm to ensure the inclusion of studies identified using previous literature reviews ^4,5. Lastly, the selection process was performed by two independent reviewers and data extraction was performed by one reviewer and checked by a second reviewer.

Limitations

Although the quality of the economic evaluations using the Drummond checklist ⁹ was assessed by one independent reviewer and checked by another reviewer, it was difficult to judge the quality due to the inherent subjectivity of the questions in the checklist; this problem has been recognized in the past ⁴². Besides the subjectivity of the Drummond checklist ⁹, it is possible that studies fulfilled the requirements but did not report this in a transparent way. In addition, for some characteristics, a pragmatic approach was needed to assess the transferability of studies using the Welte et al. ¹⁰ checklist. While methodological characteristics were easy to assess using information reported in the articles, it was much more difficult to judge the transferability of the healthcare and population characteristics to the Dutch setting since these characteristics are often not described sufficiently in articles. With the exception of case-mix, we assessed the transferability of healthcare and population characteristics solely on the differences between the country examined in the study (e.g. US) and the Netherlands. Normally the transferability of a study’s findings is assessed by comparing the study characteristics from the article with the characteristics of the country for which policy decisions are to be made (i.e, the Netherlands in this case).

Implications

This study examined which treatment strategies are preferred for patients with PAD from a cost-effectiveness standpoint. The cost-effectiveness estimates can help to decide between treatments for patients with PAD. However, when discussing the implications of this study, it must be recognized that some of the studies are probably outdated (since the included studies were performed between 1995 and 2014) and this means that their results do not necessarily reflect the current cost-effectiveness of the treatment strategies that were investigated. There are various reasons for this, including (but not limited to) changes in the clinical experience with performing an intervention and the subsequent impact on effectiveness and safety, important shifts in the patient populations (case-mix), and changes in unit costs and resource use. Similarly, the transferability of the results of published studies to a specific decision country is dependent on many factors.

The relevance of some previous studies is clearly limited because of between-country differences and temporal changes in practice. For example, the laser-assisted (Nd/YAG) PTA is not performed anymore in the Netherlands.

Uncertainty about the cost-effectiveness of an intervention also exists because of the lack of published studies. Since the cost-effectiveness of some treatments (e.g. orbital atherectomy or laser assisted PTA) was examined in just one single study, it is particularly difficult to draw any hard conclusions about their cost-effectiveness compared to other treatments.

Lastly, while the treatment of PAD is a dynamic process that involves a sequence of interventions, most studies estimated the cost-effectiveness of several interventions as a first line treatment; very few studies examined the optimal treatment sequence. This represents an important gap in the literature and is an important avenue worth exploring in the future using, for example, decision modelling techniques.

It should be emphasised that some of the conclusions of this report are based on cost-effectiveness studies that were not performed in the Netherlands. Consequently, it is possible that the conclusions may change if the cost-effectiveness for the Dutch situation were to be examined, which could arise due to necessary changes in the analysis, including use of the societal perspective and information regarding Dutch treatment practice and case mix. Moreover, even a well-performed Dutch cost-effectiveness study may not be relevant anymore if it was performed ten years ago. Therefore, in some cases, it would be advisable to perform cost-effectiveness studies focusing on the Dutch setting using decision modelling techniques; these studies would yield a more valid estimate of the cost-effectiveness of one intervention versus another. However, the first step would be to carefully consider which studies are worth performing. The candidates worth considering first might be interventions with a potentially high budget impact that have been shown to be sufficiently safe and effective.

Conclusions

This systematic review provides an overview and analysis of all published studies that have evaluated the cost-effectiveness of treatment strategies for patients with PAD. In the literature, the cost-effectiveness of various interventions has been compared with others for patients with intermittent claudication or critical limb ischemia: bypass, percutaneous transluminal angioplasty, exercise therapy, medication and amputation. The overall conclusions for some decision problems (e.g. supervised exercise versus non-supervised exercise) were clear, meaning that the conclusions have been consistent between studies. However, it is difficult to draw definitive conclusions about the cost-effectiveness of other interventions for patients with CLI or IC because of important variation between studies or because the conclusions were based on a single study. Important differences between studies can be found in methodological approaches, health care system characteristics and population characteristics.

References

(1) Bartelink M, Elsman B, Oostindjer A, Stoffers H, Wiersma T, Geraets J. NHG-Standaard Perifeer arterieel vaatlijden (tweede herziening). Huisarts Wet 2014; 57(2): 81.

(2) National Institute for Health and Clinical Excellence. Lower limb peripheral arterial disease - Diagnosis and management NICE Clinical Guideline 147. 2012; 2015.

(3) Fowkes FG, Housley E, Cawood EH, Macintyre CC, Ruckley CV, Prescott RJ. Edinburgh Artery Study: prevalence of asymptomatic and symptomatic peripheral arterial disease in the general population. Int J Epidemiol 1991; 20(2): 384-392.

(4) Simpson EL, Kearns B, Stevenson MD, Cantrell AJ, Littlewood C, Michaels JA. Enhancements to angioplasty for peripheral arterial occlusive disease: Systematic review, cost-effectiveness assessment and expected value of information analysis. Health Technol Assess 2014; 18: I-XX+1-251.

(5) Squires H, Simpson E, Meng Y, Harnan S, Stevens JW, Wong R, et al. A systematic review and economic evaluation of cilostazol, naftidrofuryl oxalate, pentoxifylline and inositol nicotinate for the treatment of intermittent claudication in people with peripheral arterial disease. Health Technol Assess 2011; 15: i-210.

(6) Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev 2015; 4(1): 1-4053-4-1.

(7) Centre for Reviews and Dissemination. <br />NHS EED Economics Filter: Medline (Ovid) monthly search <br />. 2013.

(8) Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ 2011; 343: d5928.

(9) Drummond MF, Jefferson TO. Guidelines for authors and peer reviewers of economic submissions to the BMJ. Br Med J 1996; 313(7052): 275-283.

(10) Welte R, Feenstra T, Jager H, Leidl R. A decision chart for assessing and improving the transferability of economic evaluation results between countries. Pharmacoeconomics 2004; 22(13): 857-876.

(11) Health Care Insurance Board (CVZ). Guidelines for pharmacoeconomic evaluation, updated version [in Dutch: Richtlijnen voor farmaco-economisch onderzoek, geactualiseerde versie]. 2006;.

(12) National Institute for Health and Clinical Excellence. Lower limb peripheral arterial disease - Diagnosis and management - NICE Clinical Guideline 147 - Appendices - Appendix L. 2012; 2014.

(13) Bermingham SL, Sparrow K, Mullis R, Fox M, Shearman C, Bradbury A, et al. The cost-effectiveness of supervised exercise for the treatment of intermittent claudication. Eur J Vasc Endovasc Surg 2013; 46: 707-714.

(14) Kearns BC, Michaels JA, Stevenson MD, Thomas SM. Cost-effectiveness analysis of enhancements to angioplasty for infrainguinal arterial disease. Br J Surg 2013; 100: 1180-1188.

(15) Meng Y, Squires H, Stevens JW, Simpson E, Harnan S, Thomas S, et al. Cost-effectiveness of cilostazol, naftidrofuryl oxalate, and pentoxifylline for the treatment of intermittent claudication in people with peripheral arterial disease. Angiology 2014; 65: 190-197.

(16) National Institute for Health and Clinical Excellence. Lower limb peripheral arterial disease - Diagnosis and management - NICE Clinical Guideline 147 - Appendices - Appendix K. 2012; 2014.

(17) Spronk S, Bosch JL, den Hoed PT, Veen HF, Pattynama PMT, Hunink MGM. Cost-effectiveness of endovascular revascularization compared to supervised hospital-based exercise training in patients with intermittent claudication: A randomized controlled trial. J Vasc Surg 2008; 48: 1472-1480.

(18) Treesak C, Kasemsup V, Treat-Jacobson D, Nyman JA, Hirsch AT. Cost-effectiveness of exercise training to improve claudication symptoms in patients with peripheral arterial disease. Vasc Med 2004; 9: 279-285.

(19) Vries SO, Visser K, Vries JA, Wong JB. Intermittent Claudication: Cost-effectiveness of Revascularization versus Exercise Therapy 1. Radiology 2002;.

(20) Van Asselt, A. D. I., Nicolai SPA, Joore MA, Prins MH, Teijink JAW. Cost-effectiveness of exercise therapy in patients with intermittent claudication: Supervised exercise therapy versus a 'go home and walk' advice. Eur J Vasc Endovasc Surg 2011; 41: 97-103.

(21) Bosch JL, Tetteroo E, Mali WPTM, Hunink MGM. Iliac arterial occlusive disease: Cost-effectiveness analysis of stent placement versus percutaneous transluminal angioplasty. Radiology 1998; 208: 641-648.

(22) Bosch JL, Haaring C, Meyerovitz MF, K AC, Hunink MGM. Cost-effectiveness of percutaneous treatment of iliac artery occlusive disease in the United States. Am J Roentgenol 2000; 175: 517-521.

(23) Guest JF, Davie AM, Clegg JP. Cost effectiveness of cilostazol compared with naftidrofuryl and pentoxifylline in the treatment of intermittent claudication in the UK. Curr Med Res Opin 2005; 21: 817-826.

(24) Lee HLD, Mehta T, Ray B, Heng MST, McCollum PT, Chetter IC. A Non-randomised Controlled Trial of the Clinical and Cost Effectiveness of a Supervised Exercise Programme for Claudication. Eur J Vasc Endovasc Surg 2007; 33: 202-207.

(25) Mazari FAK, Khan JA, Carradice D, Samuel N, Gohil R, McCollum PT, et al. Economic analysis of a randomized trial of percutaneous angioplasty, supervised exercise or combined treatment for intermittent claudication due to femoropopliteal arterial disease. Br J Surg 2013; 100: 1172-1179.

(26) Cost-effectiveness of supervised exercise, stenting, and optimal medical care for claudication: results from the Claudication: Exercise Versus Endoluminal Revascularization (CLEVER) trial. J Am Heart Assoc 2014; 3: e001233.

(27) Barshes NR, Chambers JD, Cohen J, Belkin M, Optimize MT, Study C. Cost-effectiveness in the contemporary management of critical limb ischemia with tissue loss. J Vasc Surg 2012; 56: 1015-1024.e1.

(28) Barshes NR, Ozaki CK, Kougias P, Belkin M. A cost-effectiveness analysis of infrainguinal bypass in the absence of great saphenous vein conduit. J Vasc Surg 2013; 57: 1466-1470.

(29) Bradbury AW, Adam DJ, Bell J, Forbes JF, Fowkes FGR, Gillespie I, et al. Multicentre randomised controlled trial of the clinical and cost-effectiveness of a bypass-surgery-first versus a balloon-angioplasty-first revascularisation strategy for severe limb ischaemia due to infrainguinal disease. The Bypass versus Angioplasty in Severe Ischaemia of the Leg (BASIL) trial. Health Technol Assess 2010; 14: 1-236.

(30) Brothers TE, Rios GA, Robison JG, Elliott BM. Justification of intervention for limb-threatening ischemia: A surgical decision analysis. Cardiovasc Surg 1999; 7: 62-69.

(31) Forbes JF, Adam DJ, Bell J, Fowkes FGR, Gillespie I, Raab GM, et al. Bypass versus Angioplasty in Severe Ischaemia of the Leg (BASIL) trial: Health-related quality of life outcomes, resource utilization, and cost-effectiveness analysis. J Vasc Surg 2010; 51: 43S-51S.

(32) Holler D, Claes C, Von DS. Cost-utility analysis of treating severe peripheral arterial occlusive disease. Int J Angiol 2006; 15: 25-33.

(33) Katsanos K, Karnabatidis D, Diamantopoulos A, Spiliopoulos S, Siablis D. Cost-effectiveness analysis of infrapopliteal drug-eluting stents. Cardiovasc Intervent Radiol 2013; 36: 90-97.

(34) Sultan S, Hynes N. Five-year irish trial of CLI patients with TASC II type C/D lesions undergoing subintimal angioplasty or bypass surgery based on plaque echolucency. J Endovasc Ther 2009; 16: 270-283.

(35) Sultan S, Tawfick W, Hynes N. Cool excimer laser-assisted angioplasty (CELA) and tibial balloon angioplasty (TBA) in management of infragenicular arterial occlusion in critical lower limb ischemia (CLI). Vasc Endovasc Surg 2013; 47: 179-191.

(36) Barshes NR, Kougias P, Ozaki CK, Goodney PP, Belkin M. Cost-effectiveness of revascularization for limb preservation in patients with end-stage renal disease. J Vasc Surg 2014; 60: 369-374.e1.

(37) Cost-effectiveness of revascularization for limb preservation in patients with marginal functional status. Ann Vasc Surg 2014; 28: 10-17.

(38) Hunink MGM, Wong JB, Donaldson MC, Meyerovitz MF, De Vries J, Harrington DP. Revascularization for femoropopliteal disease: A decision and cost- effectiveness analysis. J Am Med Assoc 1995; 274: 165-171.

(39) Muradin GSR, Hunink MGM. Cost and patency rate targets for the development of endovascular devices to treat femoropopliteal arterial disease. Radiology 2001; 218: 464-469.

(40) Weinstock B, Dattilo R, Diage T. Cost-effectiveness analysis of orbital atherectomy plus balloon angioplasty vs balloon angioplasty alone in subjects with calcified femoropopliteal lesions. Clin Outcomes Res 2014; 6: 133-139.

(41) Sculpher M, Michaels J, McKenna M, Minor J. A cost-utility analysis of laser-assisted angioplasty for peripheral arterial occlusions. Int J Technol Assess Health Care 1996; 12: 104-125.

(42) Handels RL, Wolfs CA, Aalten P, Joore MA, Verhey FR, Severens JL. Diagnosing Alzheimer's disease: A systematic review of economic evaluations. Alzheimers Dement 2013;.

(43) Visser K, de Vries SO, Kitslaar PJEHM, van Engelshoven, J. M. A., Hunink MGM. Cost-effectiveness of diagnostic imaging work-up and treatment for patients with intermittent claudication in the Netherlands. Eur J Vasc Endovasc Surg 2003; 25: 213-223.

APPENDIX 1

Embase.com 2832

('peripheral occlusive artery disease'/de OR claudication/exp OR 'peripheral vascular disease'/de OR 'peripheral ischemia'/exp OR 'leg ischemia'/exp OR 'limb ischemia'/exp OR 'leg revascularization'/exp OR (('leg artery'/exp OR limb/exp OR 'femoral vein'/exp) AND ('artery disease'/de OR 'vascular disease'/de OR occlusion/de OR 'blood vessel occlusion'/exp OR 'artery ligation'/exp OR 'bypass surgery'/exp OR 'vascular surgery'/de OR 'blood vessel graft'/de OR 'vein graft'/de)) OR 'ankle brachial index'/exp OR 'limb salvage'/exp OR (((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit* OR noncoronar* OR non-coronary OR noncardiac* OR non-cardiac) NEAR/3 ('vascular disease' OR 'vascular surgery' OR 'vascular surgical' OR 'vascular operation' OR 'vascular operations' OR bypass* OR 'vascular diseases' OR 'arterial disease' OR 'arterial diseases' OR ischem* OR ischaem* OR occlus* OR obstruct* OR salvage* OR narrow* OR harden* OR steno* OR resteno* OR constric* OR arteriosclero* OR atherosclero* OR atheroma* OR revasculari*)) OR (((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit*) NEAR/3 (vascul* OR vessel* OR arter OR lesion* OR vein*)) AND (arteriosclero* OR atherosclero* OR atheroma*)) OR claudicat* OR 'angina cruris' OR (angiosclerot* NEAR/3 intermitten*) OR paod OR PVD OR (pad NEAR/6 (arter* OR atheroscler*)) OR ((ankle OR toe) NEAR/3 brachial) OR claudicat* OR angiotome*):ab,ti) AND ('economic evaluation'/exp OR economics/exp OR 'economic aspect'/de OR cost/exp OR 'health economics'/de OR 'health care cost'/exp OR pharmacoeconomics/exp OR 'quality adjusted life year'/de OR 'decision support system'/exp OR 'decision tree'/exp OR (economic* OR pharmacoeconomic* OR cost OR costs OR costly OR costing OR price OR prices OR pricing OR budget* OR (expenditure* NOT energy) OR (value NEAR/3 money) OR ((benefit* OR financ*) NEAR/3 (control* OR analy* OR minim* OR illness* OR evaluation* OR effectiv* OR efficac* OR efficien* OR impact* OR model* OR aspect*)) OR (decision* NEAR/3 (support* OR model* OR tree*)) OR (cost* NEAR/3 (benefit* OR utilit* OR medical* OR nation* OR internation* OR hospital* OR patient* OR saving* OR allocat*)) OR (health* NEAR/3 (cost* OR econom* OR expend* OR budget*)) OR (('health technology' OR 'medical technology') NEAR/3 assessment*) OR pharmacoeconomic* OR qaly* OR ('quality adjusted' NEAR/3 ('life year' OR 'life years'))):ab,ti) AND [english]/lim NOT ([Conference Abstract]/lim OR [Letter]/lim OR [Note]/lim OR [Conference Paper]/lim OR [Editorial]/lim)

Medline (OvidSP) 1770

(Peripheral Arterial Disease/ OR Intermittent Claudication/ OR Peripheral Vascular Diseases/ OR ((exp Extremities/) AND (Vascular Diseases/ OR Vascular Grafting/ OR Vascular Surgical Procedures/)) OR Ankle Brachial Index/ OR Limb Salvage/ OR (((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit* OR noncoronar* OR non-coronary OR noncardiac* OR non-cardiac) ADJ3 (vascular disease OR vascular surgery OR vascular surgical OR vascular operation OR vascular operations OR bypass* OR vascular diseases OR arterial disease OR arterial diseases OR ischem* OR ischaem* OR occlus* OR obstruct* OR salvage* OR narrow* OR harden* OR steno* OR resteno* OR constric* OR arteriosclero* OR atherosclero* OR atheroma* OR revasculari*)) OR (((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit*) ADJ3 (vascul* OR vessel* OR arter OR lesion* OR vein*)) AND (arteriosclero* OR atherosclero* OR atheroma*)) OR claudicat* OR angina cruris OR (angiosclerot* ADJ3 intermitten*) OR paod OR PVD OR (pad ADJ6 (arter* OR atheroscler*)) OR ((ankle OR toe) ADJ3 brachial) OR claudicat* OR angiotome*).ab,ti.) AND (exp Economics/ OR Economics.xs. OR Quality-Adjusted Life Years/ OR Markov Chains/ OR Decision Trees/ OR (economic* OR pharmacoeconomic* OR cost OR costs OR costly OR costing OR price OR prices OR pricing OR budget* OR (expenditure* NOT energy) OR (value ADJ3 money) OR ((benefit* OR financ*) ADJ3 (control* OR analy* OR minim* OR illness* OR evaluation* OR effectiv* OR efficac* OR efficien* OR impact* OR model* OR aspect*)) OR (decision* ADJ3 (support* OR model* OR tree*)) OR (cost* ADJ3 (benefit* OR utilit* OR medical* OR nation* OR internation* OR hospital* OR patient* OR saving* OR allocat*)) OR (health* ADJ3 (cost* OR econom* OR expend* OR budget*)) OR ((health technology OR medical technology) ADJ3 assessment*) OR pharmacoeconomic* OR qaly* OR (quality adjusted ADJ3 (life year OR life years))).ab,ti.) AND english.la. NOT (letter OR news OR comment OR editorial OR congresses OR abstracts).pt.

Cochrane central / technology assesments 161 / 6

((((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit* OR noncoronar* OR non-coronary OR noncardiac* OR non-cardiac) NEAR/3 ('vascular disease' OR 'vascular surgery' OR 'vascular surgical' OR 'vascular operation' OR 'vascular operations' OR bypass* OR 'vascular diseases' OR 'arterial disease' OR 'arterial diseases' OR ischem* OR ischaem* OR occlus* OR obstruct* OR salvage* OR narrow* OR harden* OR steno* OR resteno* OR constric* OR arteriosclero* OR atherosclero* OR atheroma* OR revasculari*)) OR (((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit*) NEAR/3 (vascul* OR vessel* OR arter OR lesion* OR vein*)) AND (arteriosclero* OR atherosclero* OR atheroma*)) OR claudicat* OR 'angina cruris' OR (angiosclerot* NEAR/3 intermitten*) OR paod OR PVD OR (pad NEAR/6 (arter* OR atheroscler*)) OR ((ankle OR toe) NEAR/3 brachial) OR claudicat* OR angiotome*):ab,ti) AND ((economic* OR pharmacoeconomic* OR cost OR costs OR costly OR costing OR price OR prices OR pricing OR budget* OR (expenditure* NOT energy) OR (value NEAR/3 money) OR ((benefit* OR financ*) NEAR/3 (control* OR analy* OR minim* OR illness* OR evaluation* OR effectiv* OR efficac* OR efficien* OR impact* OR model* OR aspect*)) OR (decision* NEAR/3 (support* OR model* OR tree*)) OR (cost* NEAR/3 (benefit* OR utilit* OR medical* OR nation* OR internation* OR hospital* OR patient* OR saving* OR allocat*)) OR (health* NEAR/3 (cost* OR econom* OR expend* OR budget*)) OR (('health technology' OR 'medical technology') NEAR/3 assessment*) OR pharmacoeconomic* OR qaly* OR ('quality adjusted' NEAR/3 ('life year' OR 'life years'))):ab,ti)

Cochrane economic evaluations 67

Web-of-science 1573

TS=(((((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit* OR noncoronar* OR non-coronary OR noncardiac* OR non-cardiac) NEAR/3 ("vascular disease" OR "vascular surgery" OR "vascular surgical" OR "vascular operation" OR "vascular operations" OR bypass* OR "vascular diseases" OR "arterial disease" OR "arterial diseases" OR ischem* OR ischaem* OR occlus* OR obstruct* OR salvage* OR narrow* OR harden* OR steno* OR resteno* OR constric* OR arteriosclero* OR atherosclero* OR atheroma* OR revasculari*)) OR (((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit*) NEAR/3 (vascul* OR vessel* OR arter OR lesion* OR vein*)) AND (arteriosclero* OR atherosclero* OR atheroma*)) OR claudicat* OR "angina cruris" OR (angiosclerot* NEAR/3 intermitten*) OR paod OR PVD OR (pad NEAR/6 (arter* OR atheroscler*)) OR ((ankle OR toe) NEAR/3 brachial) OR claudicat* OR angiotome*)) AND ((economic* OR pharmacoeconomic* OR cost OR costs OR costly OR costing OR price OR prices OR pricing OR budget* OR (expenditure* NOT energy) OR (value NEAR/3 money) OR ((benefit* OR financ*) NEAR/3 (control* OR analy* OR minim* OR illness* OR evaluation* OR effectiv* OR efficac* OR efficien* OR impact* OR model* OR aspect*)) OR (decision* NEAR/3 (support* OR model* OR tree*)) OR (cost* NEAR/3 (benefit* OR utilit* OR medical* OR nation* OR internation* OR hospital* OR patient* OR saving* OR allocat*)) OR (health* NEAR/3 (cost* OR econom* OR expend* OR budget*)) OR (("health technology" OR "medical technology") NEAR/3 assessment*) OR pharmacoeconomic* OR qaly* OR ("quality adjusted" NEAR/3 ("life year" OR "life years"))))) AND DT=(Article) AND LA=(english)

PubMed publisher 168

(Peripheral Arterial Disease[mh] OR Intermittent Claudication[mh] OR Peripheral Vascular Diseases[mh] OR ((Extremities[mh]) AND (Vascular Diseases[mh] OR Vascular Grafting[mh] OR Vascular Surgical Procedures[mh])) OR Ankle Brachial Index[mh] OR Limb Salvage[mh] OR (((peripheral*[tiab] OR leg OR extremit*[tiab] OR limb*[tiab] OR femoral*[tiab] OR poplit*[tiab] OR infrapoplit*[tiab] OR femoropoplit*[tiab] OR noncoronar*[tiab] OR non-coronary OR noncardiac*[tiab] OR non-cardiac) AND (vascular disease OR vascular surgery OR vascular surgical OR vascular operation OR vascular operations OR bypass*[tiab] OR vascular diseases OR arterial disease OR arterial diseases OR ischem*[tiab] OR ischaem*[tiab] OR occlus*[tiab] OR obstruct*[tiab] OR salvage*[tiab] OR narrow*[tiab] OR harden*[tiab] OR steno*[tiab] OR resteno*[tiab] OR constric*[tiab] OR arteriosclero*[tiab] OR atherosclero*[tiab] OR atheroma*[tiab] OR revasculari*[tiab])) OR (((peripheral*[tiab] OR leg OR extremit*[tiab] OR limb*[tiab] OR femoral*[tiab] OR poplit*[tiab] OR infrapoplit*[tiab] OR femoropoplit*[tiab]) AND (vascul*[tiab] OR vessel*[tiab] OR arter OR lesion*[tiab] OR vein*[tiab])) AND (arteriosclero*[tiab] OR atherosclero*[tiab] OR atheroma*[tiab])) OR claudicat*[tiab] OR angina cruris OR (angiosclerot*[tiab] AND intermitten*[tiab]) OR paod OR PVD OR (pad AND (arter*[tiab] OR atheroscler*[tiab])) OR ((ankle OR toe) AND brachial) OR claudicat*[tiab] OR angiotome*[tiab])) AND (Economics[mh] OR Economics[sh] OR Quality-Adjusted Life Years[mh] OR Markov Chains[mh] OR Decision Trees[mh] OR (economic*[tiab] OR pharmacoeconomic*[tiab] OR cost OR costs OR costly OR costing OR price OR prices OR pricing OR budget*[tiab] OR (expenditure*[tiab] NOT energy) OR (value AND money) OR ((benefit*[tiab] OR financ*[tiab]) AND (control*[tiab] OR analy*[tiab] OR minim*[tiab] OR illness*[tiab] OR evaluation*[tiab] OR effectiv*[tiab] OR efficac*[tiab] OR efficien*[tiab] OR impact*[tiab] OR model*[tiab] OR aspect*[tiab])) OR (decision*[tiab] AND (support*[tiab] OR model*[tiab] OR tree*[tiab])) OR (cost*[tiab] AND (benefit*[tiab] OR utilit*[tiab] OR medical*[tiab] OR nation*[tiab] OR internation*[tiab] OR hospital*[tiab] OR patient*[tiab] OR saving*[tiab] OR allocat*[tiab])) OR (health*[tiab] AND (cost*[tiab] OR econom*[tiab] OR expend*[tiab] OR budget*[tiab])) OR ((health technology OR medical technology) AND assessment*[tiab]) OR pharmacoeconomic*[tiab] OR qaly*[tiab] OR (quality adjusted AND (life year OR life years)))) AND english[la] AND publisher[sb]

EconLit 47

((((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit* OR noncoronar* OR non-coronary OR noncardiac* OR non-cardiac) AND (vascular disease OR vascular surgery OR vascular surgical OR vascular operation OR vascular operations OR bypass* OR vascular diseases OR arterial disease OR arterial diseases OR ischem* OR ischaem* OR occlus* OR obstruct* OR salvage* OR narrow* OR harden* OR steno* OR resteno* OR constric* OR arteriosclero* OR atherosclero* OR atheroma* OR revasculari*)) OR (((peripheral* OR leg OR extremit* OR limb* OR femoral* OR poplit* OR infrapoplit* OR femoropoplit*) AND (vascul* OR vessel* OR arter OR lesion* OR vein*)) AND (arteriosclero* OR atherosclero* OR atheroma*)) OR claudicat* OR angina cruris OR (angiosclerot* AND intermitten*) OR ((ankle OR toe) AND brachial) OR claudicat* OR angiotome*))

Google Scholar

APPENDIX 2⁹

Appendix2_PAV

APPENDIX 3

Table 1: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for IC patients

Author

Year

Country

Indication

Population

Lesion type

Area

Interventions

Study design

Perspective

Time horizon

Outcomes

Discounting %

Hunink ³⁸

1995

IC & CLI

Patients requiring revascularization because of chronic femoropopliteal disease. Subgroups by age, sex, indication, lesion type and graft type (ie, lifestyle-limiting) claudication, rest pain, or tissue necrosis. 65-year-old men with disabling claudication and a femoropopliteal stenosis or occlusion.

occlusion / stenosis

femoropopliteal

1) NT
2) PTA-NT
3) PTA-PTA
4) PTA-BP
5) BP-NT
6) BP-Rev

Model

HCS

Lifetime

QALY

Sculpher⁴¹

1996

IC/CLI

Patients with moderate to severe claudication, age 65

occlusion

femoral or femoropopliteal

1) PTA
2) laser (Nd/YAG) assisted PTA

Model

NHS

25 years

Life-years

QALY

6.0%

Bosch²¹

1998

60-year old male with life-limiting claudication due to stenoses in the iliac arteries, for whom a percutaneous intervention was indicated

stenosis

Iliac

1) No revascularization
2) PTA followed by no revascularization
3) PTA with SS followed by no revascularization
4) PTA and repeated PTA
5) PTA followed by PTA with SS
6) Initial and repeated PTA with SS
7) Primary stent placement followed by PTA with SS

Model

Societal

QALY

3.0%

Bosch²²

2000

60-year old male with life-limiting claudication due to stenoses in the iliac arteries, for whom a percutaneous intervention was indicated

stenosis

Iliac

Model

Societal

Lifetime

QALY

3.0%

Muradin³⁹

2001

IC/CLI

A 65-year-old man with femoropopliteal arterial disease without comorbidity or other risk factors

stenosis/ occlusion

femoropopliteal

1) PTA
2) BP

Model

HCS

Lifetime

QALY

3.0%

Vries, de¹⁹

2002

Patients with mild and severe IC. 60-year-old man with no history of coronary artery disease

stenosis/ occlusion

iliac & femoropopliteal

1) SE
2) SE±PTA
3) SE±PTA/BP
4) PTA/SE
5) PTA/BP/SE

Model

Societal

Lifetime

QALY

3.0%

Table 1: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for IC patients

Author

Year

Country

Indication

Population

Lesion type

Area

Interventions

Study design

Perspective

Time horizon

Outcomes

Discounting %

Treesak¹⁸

2004

PAD patients with claudication in whom ilio-femoral arterial disease was known, and for whom either exercise rehabilitation or PTA could appropriately be indicated

ilio-femoral arterial disease

1) no treatment
2) SE therapy
3) PTA without primary stent placement

Model

Societal

12 months

ICD

ACD

Guest²³

2005

Patients with intermittent claudication who are 40 years of age or above and have at least six months history of symptomatic intermittent claudication, secondary to lower extremity arterial occlusive disease

Occlusion

no specific location

1) Cilostazol
2) Naftidrofuryl
3) Pentoxifylline

Model

NHS

24 weeks

MWD

Lee²⁴

2007

Patients with IC

no specific location

1) CMT
2) CMT+SE therapy

single study, non-randomized

NHS

12 months

MWD
PRWD
ICD
ABI
QALY

Spronk¹⁷

2008

Patients with symptoms of intermittent claudication due to an iliac or femoro-popliteal arterial lesion. Rutherford category 1,2, or 3 ≥3 months; pain-free MWD of 350m during a treadmill test; ABI 0.9 at rest ABI with a decrease of 0.15 after the treadmill test; (4) vascular stenoses of 50% diameter reduction at the iliac or femoro-popliteal level

Iliac or femoro-popliteal arterial lesion

1) SE therapy
2) ER (PTA, if necessary a stent)

Single study, RCT

Societal

12 months

QALY

3.0%

Asselt, van²⁰

2011

Patients with PAD, Fontaine stage II, ABI< 0.9, ACD <500m

no specific location

1) WA
2) SE therapy (with and without feedback)

Single study, RCT

Societal

12 months

QALY
MWD

Table 1: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for IC patients

Author

Year

Country

Indication

Population

Lesion type

Area

Interventions

Study design

Perspective

Time horizon

Outcomes

Discounting %

Squires⁵

2011

Patients who have stable (at least for the past 3 months) and symptomatic IC, secondary to PAD. Fontaine stage II, 66 years old

no specific location

1) No vascular drug

2) Pentoxifylline (1200mg/d)
3) Cilostazo (200mg/d)
4) Naftidrofuryl oxalate (600mg/d)

5) Inositol nicotinate (4g/d).

Model

NHS

100 years

QALY

3.5%

NICE (App.K)¹⁶

2012

Patients with mild IC, no CLI, willing and suitable for exercise program. Age 67, male 66%, ABPI: 0.67,

no specific location

1) UE therapy
2) SE therapy

Model

NHS

Lifetime

QALY

3.5%

NICE (App.L)¹²

2012

Patients with intermittent claudication in England and Wales who are suitable for both exercise and angioplasty as first-line treatment options. Age 67, male: 70%, ABI 0.64

stenosis/ occlusion

aorto-iliac & femoropopliteal seperately

1) UE-SE
2) UE-PTA SS
3) UE-BP
4) SE-SE
5) SE-PTA SS
6) SE-BP
7) PTA SS-SE
8) PTA SS-PTA SS
9) PTA SS-BP
10) PTA PS-SE
11) PTA PS-PTA SS
12) PTA PS-BP
13) PTA SS+SE

Model

NHS

Lifetime

QALY

3.5%

Mazari²⁵

2013

A 69-year old man with unilateral IC due to femoropopliteal disease amenable to angioplasty with symptoms unchanged after 3 months of best medical treatment.

femoropopliteal

1) PTA
2) SE therapy
3) PTA+SE therapy

single study, RCT

HCP

12 months

QALY

Reynolds ²⁶

2014

Patients with IC due to aorto iliac PAD

stenosis

aorto-iliac

1) SE therapy + OMC
2) Stenting + OMC
3) OMC

Model

Societal

5 year

QALY

Simpson⁴

2014

IC / CLI

Patients with symptomatic PAD suitable for endovascular treatment for disease distal to the inguinal ligament. IC patients 66 years

No distinction made

Distal to the inguinal ligament. Femoropopliteal and infrapopliteal arteries

1) Paclitaxel-coated balloon
2) PTA with bail-out PES
3) PTA with bail-out BMS
4) PTA, no bail-out stenting
5) BMS
6) PES
7) EVBT
8) Stent-graft
9) Cryoplasty
10) SES

Model

NHS

Lifetime

LY
QALY

3.5%

Weinstock⁴⁰

2014

IC/CLI

Patients with Rutherford class 2–4 classification had lesions of 70% stenosis and fluoroscopically visible calcium

Calcified femoropopliteal lesions

1) orbital atherectomy + balloon angioplasty
2) balloon angioplasty

Single study, RCT

Third party payer

1 year

QALY

Table 1: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for IC patients - continued

Author	Year	Price year (currency)	Intervention costs	Health utilities	Total costs	Incremental costs	Total effectiveness	Incremental effects	ICER	Cost-effective?
Hunink ³⁸	1995	1990 (US $)	PTA: $ 6,608 BP: $ 13.343		1) $ 25,000 3) $ 15,000 - $ 24,000 4) $ 17,000 - $ 27,000 6) $ 25,000 - $ 29,000		1) 4.5 QALY 3) 6.2-7.3 QALY 4) 6.7-7.4 QALY 6) 6.5-6.9 QALY		3) Dominant compared to NT - Dominant compared to all other strategies 4) 2.8 - 64 (compared with 3) 6) Inferior	In patients with femoropopliteal disease, PTA is the preferred initial treatment in those with disabling claudication and a stenosis or occlusion and in those with chronic critical ischemia and a stenosis. BP is the preferred initial treatment in patients with chronic critical ischemia and an occlusion
Sculpher ⁴¹	1996	1993/1994 (£)	1) £1,186 2) £2,133	SF-36	1) £3,669 2) £3,929	£260	1) 6.78 LY; 5.78 QALY 2) 6.79 LY; 5.87 QALY	0.09 QALY	£3,040/QALY gained	The results indicate a cost-effective role for the laser, but important areas of uncertainty exist.
Bosch ²¹	1998	1995 (US $)	PTA $2,343 PTA with SS $ 3,033 Primary stent placement $ 3,990	RAND-36 (SF36), HUI, EQ5D, SG, TTO	1) $3,368 2) $5,868 3) $6,573 4) $7,145 5) $7,504 6) $7,806 7) $8,763		1) 10.30 QALY 2) 11.03 QALY 3) 11.29 QALY 4) 11.36 QALY 5) 11.47 QALY 6) 11.61 QALY 7) 11.61 QALY		1) Reference strategy 2) Inferior by ED 3) $4,073/QALY 4) Inferior by ED 5) Inferior by ED 6) $4,519/QALY 7) Inferior by dominance	PTA with selective stent placement is a cost-effective treatment strategy compared with primary stent replacement or PTA alone in the treatment of IC caused by an iliac arterial stenosis. PTA with selective stent placement for both primary and secondary treatment is more cost effective than both selective stent placement followed by conservative management and primary stent placement followed by selective stent placement.
Bosch²²	2000	1998 (US $)	PTA $5,207 PTA with SS $6,027		1) $4,531 2) $10,048 3) $10,903 4) $12,458 5) $12,830 6) $13,158		1) 7.79 QALY 2) 8.46 QALY 3) 8.63 QALY 4) 8.73 QALY 5) 8.79 QALY 6) 8.89 QALY		1) ref 2) ED 3) $7,624/QALY 4) ED 5) ED 6) $8,519/QALY	PTA followed by selective stent placement for unsatisfactory PTA results is a more cost-effective treatment strategy than PTA alone
Muradin ³⁹	2001	1999 (US$)		HUI	IC, stenosis: PTA $22,758 IC, stenosis: BP $33,229 IC, occlusion: PTA $32,131 IC, occlusion: BP $33,229	PTA vs BP IC, stenosis: -$10,471 IC, occlusion: -$1,098	IC, stenosis: PTA 5.85 QALY IC, stenosis: BP 5.46 QALY IC, occlusion: PTA 5.59 QALY IC, occlusion: BP 5.46 QALY	PTA vs BP IC, stenosis: 0.39 QALY IC, occlusion: 0.13 QALY	PTA dominant	The results illustrate that when currently used procedures are considered, PTA is more cost-effective compared with BP in the treatment of milder forms of femoropopliteal arterial disease, and bypass surgery is the treatment of choice in more severe disease
Vries, de¹⁹	2002	1995 (US $)	Aortic bifurcation bypass $23,490 Iliac PTA with selective stenting $7,550 Femoropopliteal or infrapopliteal bypass surgery $16,490 Femoral or popliteal PTA $4,170 Exercise per year $3,780	EQ5D	1) $ 17,400 2) $ 21,400 3) $ 43,000 4) $ 21,200 5) $ 43,000	4 vs. 1) $3,800 3 vs. 4) $21,800	1) 6.05 QALY 2) 6.14 QALY 3) 6.22 QALY 4) 6.15 QALY 5) 6.21 QALY	4 vs. 1) 0.1 QALY 3 vs. 4) 0.07 QALY	4 vs. 1) $38,000/QALY 3 vs. 4) $311,000/QALY 3 vs. 5) dominant	On average, the small gain in effectiveness achieved with BP for IC does not justify the additional costs. PTA as an alternative to SE therapy, when feasible, was more effective than was SE therapy alone, and the cost-effectiveness ratio was within the generally accepted range

Table 1: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for IC patients - continued

Author	Year	Price year (currency)	Intervention costs	Health utilities	Total costs	Incremental costs	Total effectiveness	Incremental effects	ICER	Cost-effective?
Treesak ¹⁸	2004	2001 ($)	2) $1,734-3,758 ($27.26 p/session) 3) $9,303-11,384		1) $ 290 2) $ 4,968 3) $ 9,303	3 vs 1) $9,303 3 vs 2) $4,335 2 vs 1) $4,678	1) ICD/ACD: 0 meter 2) ICD: 123 meter ACD: 250 meter 3) ICD: 54 meter ACD: 113 meter		2 vs 1) ICD: $38/meter; ACD: $18/meter 3 vs 1) ICD: $177/meter; ACD: $80/meter 2 vs 3) ICD/ACD: 2 dominates	SE therapy provides clinical efficacy, cost-effectiveness, and probable cost-savings for improvement of claudication in individuals with claudication
Guest ²³	2005	2002/2003 (£)	Cilostazol (100 mg): £0.63 Naftidrofuryl (100 mg): £0.09 Pentoxifylline (400 mg): £0.26		1) £ 894.91 2) £ 801.37 3) £ 917.19	1 vs. 2) £94 2 vs. 3) -£116 1 vs. 3) -£22	1) 75% increase MWD 2) 57% increase MWD 3) 45% increase MWD	1 vs. 2) 32% 2 vs. 3) 27% 1 vs. 3) 67%		Cilostazol is expected to be a clinically more effective strategy for improving MWD at 24 weeks than naftidrofuryl and pentoxifylline and potentially the most cost effective strategy in the UK
Lee²⁴	2007	NS (£)	SE therapy: annual costs £2,307 (3h/week) for 12 patients, costs per patient £48 for three months (3h per week)	SF36		£48.06		0.027 QALY	£1,780/QALY gained	SE therapy for claudication is highly cost effective, using a willingness to pay of £30,000 per QALY gained
Spronk ¹⁷	2008	2005 (€)	1) € 485 2) € 2,178	EQ-5D	1) € 2,771 2) € 7,031	€ 2,318	1) 0.17 QALY 2) 0.18 QALY	0.01 QALY	€231,800/QALY	The small gain achieved with endovascular revascularization was non-significant and the incremental cost/QALY gained by revascularization compared to SE therapy was higher than the generally-accepted willingness-to-pay threshold of 50,000 €/QALY.
Asselt, van²⁰	2011	2008 (€)	Exercise session: €24,67	EQ-5D	WA: €2304 SET: €3407	€ 1,104	WA: 400m; 0.67 QALY SE therapy: 600m; 0.71 QALY	0.04 QALY; 200m	€ 4.08/extra metre; €28,693/QALY	For cost per QALY, assumed that €40,000 is an acceptable price to pay for a QALY; SE therapy has a higher probability than WA to be cost-effective.

Table 1: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for IC patients - continued

Author

Year

Price year (currency)

Intervention costs

Health utilities

Total costs

Incremental costs

Total effectiveness

Incremental effects

ICER

Cost-effective?

Squires ⁵

2011

2002/2003 (£)

1)
2) £4.59 weekly
3) )£8.83 weekly
4) £2.26/£4.13 weekly
5) £17.23

SF36

1) £ -
2) £ 493
3) £ 964
4) £ 298

2 vs 1) £492

3 vs 1) £964

4 vs 1) £298

1) 4.975 QALY
2) 4.984 QALY
3) 4.994 QALY
4) 5.024 QALY

2 vs 1) 0.009 QALY
3 vs1) 0.019 QALY
4 vs 1) 0.049 QALY

4 vs 1. ) £6,070/QALY

Naftidrofuryl oxalate has the highest probability of being most cost-effective above willingness-to-pay thresholds of around £6000 per QALY gained

NICE (App.K) ¹⁶

2012

2009/2010 (£)

1) £ -
2) £ 288

SF36 EQ5D

1) £ 2,499
2) £ 2,690 - £ 2,714

£191- £215

1) 5.078 - 5.082 QALY
2) 5.212 - 5.350 QALY

0.134-0.268 QALY

£711- £1,608/QALY

SE therapy is more cost effective than UE therapy for the treatment of people with IC

NICE (App.L) ¹²

2012

2009/2010 (£)

SE: £288
Primary PTA no complications: £3,661
Secondary PTA no complications: £3,695
Bypass with no major complications: £5,988
BMS: £550

SF36 EQ5D

Femoro

popliteal

1) £3,938
2) £5,025
3) £5,588
4) £4,155
5) £5,248
6) £5,811
7) £5,603
8) £6,566
9) £7,064
10) £6,126
11) £6,687
12) £6,946
13) £8,120

aorto iliac

1) £3,938

2) £5,089

3) £5,590

4) £4,155

5) £5,307

6) £5,809

7) £7,548

8) £9,567

9) £10,449

10) £8,428

11) £10,702

12) £11,696

13) £8,329

Femoro-popliteal

4 vs 1) £217
5 vs 1) £1,310
8 vs 1) £2,628

Aorto iliac

4 vs 1) £217
5 vs 1) £1,369
8 vs 1) £5,629

QALYs femoropopliteal

1) 4.303
2) 4.374
3) 4.373
4) 4.395
5) 4.467
6) 4.466
7) 4.427
8) 4.503
9) 4.501
10) 4.345
11) 4.389
12) 4.388
13) 4.351

QALYs

Aorto iliac

1) 4.355

2) 4.427

3) 4.426

4) 4.447

5) 4.519

6) 4.518

7) 4.521

8) 4.661

9) 4.659

10) 4.112

11) 4.269

12) 4.267

13) 4.484

Femoropopliteal

4 vs 1) 0.092
5 vs 1) 0.069
8 vs 1) 0.200

Aorto-iliac

4 vs 1) 0.092
5 vs 1) 0.069
8 vs 1) 0.306

Aorto-iliac

4 vs 1) £2,362
5 vs 1) £8,348
8 vs 1) £18,395

Femoropopliteal

4 vs 1) £2,359

5 vs 1) £7,988

8 vs 1) £13,140

SE therapy followed by PTA with selective stent placement for people with persistent or worsening claudication is the most cost-effective sequence of treatments for people with IC in the aorto-iliac and femoropopliteal artery

Mazari ²⁵

2013

2009/2010 (€)

1) €3,553.83

2) €230.05

3) €3,783.88

SF-36

1) € 7.301,74
2) € 3.866,49
3) € 6.911,68

1 vs. 2) €3,165
3 vs. 2) €3,045
3 vs. 1) -€390,06

1) mean QALY gain: 0.620
2) mean QALY gain: 0.629
3) mean QALY gain:0.649

1 vs. 2) -0.009
3 vs. 2) 0.02
3 vs. 1) 0.029

1 vs. 2) -€381,694
3 vs. 2) €152,259
3 vs. 1) -€13,450

PTA, SE therapy and combined treatment are cost-effective options for the treatment of IC due to femoropopliteal atherosclerotic disease. SE therapy is the cheapest option and should be offered to all claudicants as first-line treatment.

Reynolds ²⁶

2014

2011 (US $)

SE therapy: $4,448
Stenting: $9,211

1) $20,616
2) $25,454
3) $16,103

1 vs 3) $4,513
2 vs 3) $9,351
2 vs 1) $4,838

1) 3.43 QALY
2) 3.47 QALY
3) 3.24 QALY

1 vs 3) 0.19 QALY
2 vs 3) 0.16 QALY
2 vs 1) 0.04 QALY

1 vs 3) $24,070/QALY
2 vs 3) $41,376/QALY
2 vs 1) $122,600/QALY

Because stenting is more costly and provides marginal additional benefit over SE therapy, SE therapy may provide better value, at least in the short term. Longer term results are uncertain.

Simpson ⁴

2014

2009/2010 (£)

1) £4,071
2) £3,949
3) £3,837
4) £3,661
5) £4,316
6) £4,525
7) £6,171
8) £6,561
9) £7,367
10) £4,732

EQ5D

1) £ 12,668
2) £ 13,032
3) £ 14,637
4) £ 14,787
5) £ 15.030
6) £ 15.692
7) £ 15.891
8) £ 16.171
9) £ 17.578

1) 6.120 QALY
2) 6.081 QALY
3) 5.956 QALY
4) 5.931 QALY
5) 5.989 QALY
6) 5.993 QALY
7) 5.984 QALY
8) 5.989 QALY
9) 5.934 QALY

Strategy 1: dominant

For patients with IC and ICL, paclitaxel (drug)-coated balloons, is both less expensive and more clinically effective than all of the other options and, therefore, it dominates them.

Weinstock ⁴⁰

2014

2010 ($)

1) $10,516
2) $6,951

SF36 EQ5D

1) $13,487
2) $12,939

$549

1) 2.33 QALY
2) 2.17 QALY

0.16 QALY

$3,441/QALY

The index procedure costs and cost-effectiveness to 1 year were comparable for orbital atherectomy +PTA vs PTA alone.

Table 2: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for patients with CLI

Author

Year

Country

Indication

Population

Lesion type

Area

Interventions

Study design

Perspective

Time horizon

Outcomes

Discounting %

Hunink ³⁸

1995

IC & CLI

Patients requiring revascularization because of disabling (ie, lifestyle-limiting) claudication, rest pain, or tissue necrosis. 65-year-old men with chronic critical ischemia and a femoropopliteal stenosis.

occlusion / stenosis

femoropopliteal

1) NT
2) PTA-NT
3) PTA-PTA
4) PTA-BP
5) BP-NT
6) BP-Rev

Model

HCS

Lifetime

QALY

5.0%

Sculpher ⁴¹

1996

IC/CLI

Patients with limb threatening ischemia, age 65

occlusion

femoral or femoropopliteal

1) PTA
2) laser (Nd/YAG) assisted PTA

Model

NHS

25 years

Life-years

QALY

6.0%

Brothers ³⁰

1999

CLI

patients with limb-threatening distal lower extremity arterial occlusive disease

occlusion

tibial-peroneal

1) Primary amputation
2) Expectant management
3) BP

Model

institutional perspective

5 years

QALY

Utility: 5%

Costs: 0%

Muradin ³⁹

2001

IC/CLI

A 65-year-old man with femoropopliteal arterial disease without co-morbidity or other risk factors

stenosis/ occlusion

femoropopliteal

1) PTA
2) BP

Model

HCS

Lifetime

QALY

3.0%

Holler ³²

2006

CLI

Hypothetical German cohort of patients with PAOD in stages III/IV

Stenosis / occlusions

no specific location

1 NT + NT
2 NT + PGE1
3 NT + PTA
4 NT + BP
5 PGE1 + NT
6 PGE1 + PGE1
7 PGE1 + PTA
8 PTA + NT
9 PTA + PGE1
10 BP + NT
11 PTA + PTA
12 BP + PGE1
13 BP + PTA
14 PGE1 + BP
15 PTA + BP
16 BP + BP

Model

Societal

5 year

QALY

Costs: 5%

Utilities 0%

Table 2: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for patients with CLI

Author

Year

Country

Indication

Population

Lesion type

Area

Interventions

Study design

Perspective

Time horizon

Outcomes

Discounting %

Sultan ³⁴

2009

CLI

CLI patients with TASC II type C or D lesions who had been selected for treatment based on plaque echolucency. Mean age 73, 45% men

occlusions / stenosis

femoral and popliteal lesions

1) subintimal angioplasty
2) BP

Single study, non-randomized

NS (direct & indirect costs)

5 year

Bradbury ²⁹

2010

CLI

severe limb ischaemia

occlusions / stenosis

aortoiliac excluded only infrainguinal

1) PTA
2) BP

BASIL trial

NHS

3 years (QALY)
7 years (AFS)

AFS
QALY

3.5%

Forbes ³¹

2010

CLI

severe limb ischaemia

stenosis / occlusion

infrainguinal

1) BP
2) PTA

BASIL trial

NHS

3 year

QALY

3.5%

Barshes ²⁷

2012

CLI

Independently living patients presenting with CLI and tissue loss in the context of a salvageable foot (ie, Rutherford category 5 limb ischemia

occlusions

infrainguinal arterial occlusive disease

1) Local wound care
2) Primary amputation
3) BP with surgical revisions
4) BP with endovascular revisions
5) Purely endovascular
6) Endovascular, BP for failure

Model

Societal

10 year

QALY
Salvaged limb-year
Functional limb-year

3.5%

Barshes ²⁸

2013

CLI

CLI associated with non-healing foot wounds in ambulatory, independently living patients

occlusions

infrainguinal, infrapopliteal

1) Conservative management
2) Primary amputation
3) BP with PTFE 6 vein patch
4) BP with AAV
5) Cryopreserved venous allograft
6) Small-caliber GSV
7) Cryopreserved arterial allograft

Model

Societal

10 year

limb-year

3.5%

Barshes ³⁷

2014

CLI

Patients with nonhealing foot wounds and critical limb ischemia (i.e., Rutherford category 5 ischemia) independently living at baseline but had marginal functional status.

infrainguinal

1) Wound care only; major amputation as needed
2) Primary amputation
3) Initial surgical bypass; surgical revision(s) as needed
4) Initial surgical bypass; endovascular revision(s) as needed
5) Initial endovascular intervention; repeat intervention(s) as needed
6) Initial endovascular intervention; surgical bypass +/ revision(s) as needed for failure

Model

unclear

10 year

years of ambulation/ limb preservation years

3.5%

Table 2: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for patients with CLI

Author

Year

Country

Indication

Population

Lesion type

Area

Interventions

Study design

Perspective

Time horizon

Outcomes

Discounting %

Barshes ³⁶

2014

CLI

Patients with non-healing foot wounds (Rutherford
category 5 chronic limb ischemia12) and end-stage renal disease

infrainguinal

1) Wound care, with selective major amputation as indicated
2) Primary amputation
3) Infrainguinal surgical bypass using an autologous vein conduit and open surgical intervention as needed
4) Infrainguinal surgical bypass using an autologous vein conduit and subsequent endovascular intervention as needed
5) Endovascular intervention, subsequent endovascular reinterventions as needed
6) Endovascular intervention, surgical bypass as needed

Model

unclear

10 year

years of ambulation

3.5%

Katsanos ³³

2013

CLI

The primary EES cohort: at least 1 infrapopliteal lesion (4.5 cm)
The Bail-out SES cohort: A CLI population was treated in both cases with the primary aim of limb salvage

infrapopliteal

1) Bail-out SES cohort
1.1 Bail-out use of SES after suboptimal balloon angioplasty
1.2 BMS after suboptimal balloon angioplasty
2) Primary EES cohort
2.1 Full-lesion primary EES
2.2 Plain balloon angioplasty and bail-out BMS as necessary

two single centre prospectively controlled studies

direct healthcare expenditures

Lifetime

event-free survival

Sultan ³⁵

2013

CLI

Patients with CLI, Rutherford categories 4, 5, or 6 with tibial TASC D lesion

Occlusion

tibial artery occlusive disease

1) cool excimer laser-assisted angioplasty
2) tibial balloon angioplasty

Single study

3 year

QALY (Q-twist)

Simpson ⁴

2014

IC / CLI

Patients with symptomatic PAD suitable for endovascular treatment for disease distal to the inguinal ligament. CLI patients 74 years

No distinction made

distal to the inguinal ligament. femoropopliteal and infrapopliteal arteries

1) Paclitaxel-coated balloon
2) PTA with bail-out PES
3) PTA with bail-out BMS
4) PTA, no bail-out stenting
5) BMS
6) PES
7) EVBT
8) Stent-graft
9) Cryoplasty
10) SES

Model

NHS

Lifetime

LY
QALY

3.5%

Weinstock ⁴⁰

2014

IC/CLI

Patients with Rutherford class 2–4 classification had lesions of 70% stenosis and fluoroscopically visible calcium

stenosis

calcified femoropopliteal lesions

1) orbital atherectomy + balloon angioplasty
2) balloon angioplasty

Single study, RCT

Third party payer

1 year

QALY

Table 2: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for patients with CLI - continued

Author	Year	Price year (currency)	Intervention costs	Health utilities	Total costs	Incremental costs	Total effectiveness	Incremental effects	ICER	Cost-effective?
Hunink ³⁸	1995	1990 (US $)	PTA: $ 11,809 BP: $ 16,820		RP 1) $ 43,000 RP 3) $ 33,000 - $ 51,000 RP 4) $ 33,000 - $ 52,000 RP 6) $ 35,000 - $ 47,000 N 1) $ 41,000 N 3) $ 28,000 - $ 45,000 N 4) $ 27,000 - $ 44,000 N 6) $ 29,000 - $ 39,000		RP 1) 2.5 QALY RP 3) 3.6-5.7 QALY RP 4) 4.4-6.3 QALY RP 6) 4.6-5.9 QALY N 1) 0.9 QALY N 3) 1.9-3.3 QALY N 4) 2.9-3.7 QALY N 6) 2.7-3.4 QALY		RP 3) Dominant compared to NT - Dominant vs all strategies RP 4) inferior - $32,000 RP 6) inferior - Dominant vs all strategies 3 N) inferior – Dominant vs all strategies 4 N) inferior - Dominant vs all strategies 6 N) inferior - Dominant vs all strategies	In patients with femoropopliteal disease, PTA is the preferred initial treatment in those with disabling claudication and a stenosis or occlusion and in those with chronic critical ischemia and a stenosis. BP is the preferred initial treatment in patients with chronic critical ischemia and an occlusion
Sculpher ⁴¹	1996	1993/1994 (£)	1) £1,186 2) £2,133	SF-36	1) £8,716 2) £8,823	£107	1) 5.44 LY; 4.40 QALY 2) 5.46 LY; 4.46 QALY	0.06 QALY	£1,810/QALY	The results indicate a cost-effective role for the laser, but important areas of uncertainty exist.
Brothers ³⁰	1999	NS (US$)	Primary revascularization: $11,000 Primary or revision amputation: $10,000 Clinical follow-up patients receiving no intervention: $1,000		1) $22,700 2) $20,500 3) $26,600	1 vs 2) $2,200 3 vs 2) $6,100 3 vs 1) $3,900	1) 2.2 QALY 2) 2.14 QALY 3) 3.3 QALY	1 vs 2) 0.06 QALY 3 vs 2) 1.16 QALY 3 vs 1) 1.1 QALY	1 vs 2) $39,000/QALY 3 vs 2) $5,280/QALY 3 vs 1) $3,545/QALY	Our analysis has demonstrated that attempts at limb salvage for limb-threatening lower extremity ischemia appear to be justified by offering substantial cost-savings compared with other forms of therapy.
Muradin ³⁹	2001	1999 (US$)	1) $18,171 2) $25,881	HUI	RP, stenosis PTA: $42,372 RP, stenosis BP: $44,694 RP, occlusion PTA: $55,074 RP, occlusion BP: $44,694 TL, stenosis PTA: $48,589 TL, stenosis BP: $53,346 TL, occlusion PTA: $65,578 TL, occlusion BP: $53,346		RP, stenosis PTA: 5.26 QALY RP, stenosis BP: 5.00 QALY RP, occlusion PTA: 4.83 QALY RP, occlusion BP: 5.00 QALY TL, stenosis PTA: 5.20 QALY TL, stenosis BP: 4.92 QALY TL, occlusion PTA: 4.74 QALY TL, occlusion BP: 4.92 QALY			The results illustrate that when currently used procedures are considered, PTA is more cost-effective compared with BP in the treatment of milder forms of femoropopliteal arterial disease, and bypass surgery is the treatment of choice in more severe disease
Holler ³²	2006	2001 (€)	NT: €1,044 - €4,964 PGE1: €2,108 - €4,238 PTA: €3,372 - €6,637 BP: €6,353 - €10,499	EQ5D	1) €5,941.80 2) €6,036.70 3) €6,169.00 4) €6,442.70 5) €7,473.80 6) €7,890.90 7) €8,569.70 8) €8,721.40 9) €9,029.40 10) €9,335.20 11) €9,413.60 12) €9,517.70 13) €9,525.60 14) €9,917.40 15) €10,257.90 16) €10,420.10	11 vs 6) €,1523 16 vs 6) €2,529 16 vs 11) €1,006	1) 0.802 QALY 2) 0.830 QALY 3) 0.827 QALY 4) 0.837 QALY 5) 1.457 QALY 6) 1.596 QALY 7) 1.575 QALY 8) 1.611 QALY 9) 1.683 QALY 10) 1.782 QALY 11) 1.660 QALY 12) 1.837 QALY 13) 1.798 QALY 14) 1.613 QALY 15) 1.679 QALY 16) 1.838 QALY	11 vs 6) 0.064 QALY 16 vs 6) 0.242 QALY 16 vs 11) 0.178 QALY	2 vs. 1) €3,365.98/QALY 3 vs. 2) Dominated 4 vs. 3) €54.314,69/QALY 5 vs. 4) €1,664.64/QALY 6 vs. 5) €2,996.59/QALY 7 vs. 6) Dominated 8 vs. 7) €56,871.85/QALY 9 vs. 8) €4,264.25/QALY 10 vs. 9) €3,097.67/QALY 11 vs. 10) Dominated 12 vs. 11) €3,303.73/QALY 13 vs. 12) Dominated 14 vs. 13) Dominated 15 vs. 14) Dominated 16 vs. 15) €771,521.54/QALY	The remaining life expectancy of a patient in the stage III/IV (aged 65–75 years) is maximized by the repeated use of BP on average of 4.7 years. Considering quality of life, this therapy strategy can provide the best results with 2.92 QALY. Repeated infusion of PGE1 is cost-effective with €4,944.19/QALY.

Table 2: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for patients with CLI - continued

Author	Year	Price year (currency)	Intervention costs	Health utilities	Total costs	Incremental costs	Total effectiveness	Incremental effects	ICER	Cost-effective?
Sultan ³⁴	2009	NS (€)	1) €10,770.89 2) €17,678.65	Q-twist	1) €11,655.90 2) €18,726.05		1) 24.7 months Q-twist 2) 24.5 months Q-twist		€10,768/QALY	SIA lead to a substantial cost reduction and better Q-TWiST
Bradbury ²⁹	2010	2006/2007 (£)	PTA: £2,115 BP: £3,634	EQ-5D	1) 1-3 years: £25,472; 1-7 year: £31,228 2) 1- 3 years: £29,006; 1-7 year: £33,539	1- 3 years: £3,533 1-7 years: £2,310	1) 984 days AFS; 1.13 QALY (414 days), 1134 OS 2) 1017 days AFS; 1.16 QALY (424 days), 1155 OS	32.4 days AFS 0.026 QALY (9.4 days), 20.4 OS	£26,032/AFS year £ 41,402/ OS £134,257/QALY	BP may be cost-ineffective at broadly accepted willingness-to-pay thresholds both strategies are more or less equally cost-effective.
Forbes ³¹	2010	2006 (US$)	1) $5,677 2) $3,305	EQ-5D	1) $45,322 2) $39,801	$5,521	1) 1.158 QALY (423 days) 2) 1.128 QALY (412 days)	0.03 QALY	$184,492/QALY	The additional cost of BP compared with the small increase in survival adjusted for HRQOL, leads to relatively high cost per QALY
Barshes ²⁷	2012	2009 (US $)	Infrainguinal BP: $44,634 Endovascular intervention$26,509 Major amputation: $28,701 Surgical revision of a vein graft: $28,039 Endovascular intervention of a vein graft: $13,137 Minor amputation and foot debridement: $17,751 Local wound care (p/year): $21,029		1) $68,736 2) $78,958 3) $84,961 4) $81,920 5) $89,040 6) $88,306	2 vs. 1) $10,222 3 vs. 1) $16,225 4 vs. 1) $13,184 5 vs. 1) $20,304 6 vs. 1) $19,570 3 vs 2) $6,003 4 vs 2) $2,962	1) 2.280 QALY 2) 2.178 QALY 3) 2.556 QALY 4) 2.556 QALY 5) 2.448 QALY 6) 2.472 QALY	2 vs. 1) - 0.102 QALY 3 vs. 1) 0.276 QALY 4 vs. 1) 0.276 QALY 5 vs. 1) 0.168 QALY 6 vs. 1) 0.192 QALY 3 vs 2) 0.378 QALY 4 vs 2) 0.378 QALY	2 vs. 1) Dominated 3 vs. 1) $58,749/QALY 4 vs. 1) $47,738/QALY 5 vs. 1) $121,010/QALY 6 vs. 1) $101,702/QALY 3 vs 2) $15,881/QALY 4 vs 2) $7,836/QALY	Endovascular-first and surgery-first strategies can both achieve comparable long-term limb salvage rates, surgical BP (especially with endovascular revisions as needed) is the most cost-effective alternative to local wound care alone.
Barshes ²⁸	2013	2011 (US $)	Endovascular intervention: $26,509 BP: $44,635 Major amputation: $34,251 Revision of BP: $28,039 Endovascular reintervention: $13,138	NA	1) $73,948 2) $84,906 3) $87,463 4) $93,814 5) $95,557 6) $95,741 7) $100,575		1) 1.912 limb year 2) 2.931 limb year 3) 4.256 limb year 4) 4.558 limb year 5) 4.242 limb year 6) 4.537 limb year 7) 4.352 limb year		2 vs. 3) Weakly dominated 3 vs. 1) $5,325/limb year 4 vs. 3) $21,228/limb year 5 vs. 4) Dominated 6 vs. 4) Dominated 7 vs. 4) Dominated 4 vs 2) $5,475/limb year	Infrainguinal BP with PTFE or AAV may be a reasonably cost-effective alternative to conservative management for patients with non-healing foot wounds and CLI. BP with PTFE appears to have a lower total cost and good outcomes. BP with AAV appears to have better outcomes but at a higher total cost, although both options have ICERs that appear reasonable when compared with conservative management
Barshes ³⁷	2014	2011 (US $)	Wound care $23,065 Major amputation $40,541 Surgical bypass $49,228 Endovascular intervention$29,297 Operative reintervention $30,910 Endovascular reintervention $14,449	NA	1) $129,651 2) $185,955 3) $113,944 4) $110,910 5) $104,118 6) $108,794		1) 0.834 ambulatory years; 1.473 limb salvage years 2) 1.585 ambulatory years; 0 limb salvage years 3) 2.410 ambulatory years; 2.941 limb salvage years 4) 2.410 ambulatory years; 2.941 limb salvage years 5) 2.468 ambulatory years; 3.031 limb salvage years 6) 2.459 ambulatory years; 3.015 limb salvage years		5 dominant strategy	Revascularization and limb preservation attempts appear less costly and provide more health benefits than wound care alone or primary amputation, even among patients with marginal functional status at baseline.

Table 2: Study characteristics and results of studies evaluating the cost-effectiveness of treatments for patients with CLI - continued

Author	Year	Price year (currency)	Intervention costs	Health utilities	Total costs	Incremental costs	Total effectiveness	Incremental effects	ICER	Cost-effective?
Barshes ³⁶	2014	2011 (US $)		NA	1) $118,086 2) $152,426 3) $128,517 4) $126,487 5) $121,478 6) $124,696		1) 1.71 ambulatory years 2) 1.19 ambulatory years 3) 1.82 ambulatory years 4) 1.82 ambulatory years 5) 1.93 ambulatory years 6) 1.87 ambulatory years		5 vs. 1) $15,403 per additional year of ambulation 6 vs. 1 $40,594 per additional year of ambulation 3 en 4 vs 1) > $70,000 2 vs 1) dominated	Endovascular intervention appears to be a cost-effective alternative to local wound care alone for patients with ESRD, PAD, and non-healing foot wounds. Primary amputation does not appear to be a cost-effective option in this patient population
Katsanos ³³	2013	NS (€)	DES: €1,500	NA		1) €358,500 2) €498,000		1) 55 years (0.89 years p/patient) 2) 43 years (0.91 years p/patient)	1) €6,518/year 2) €11,581/year	In conclusion, both strategies of bail-out and primary drug-eluting stent placement in the infrapopliteal arteries for CLI exhibit single-digit NNT and relatively low corresponding ICER.
Sultan ³⁵	2013	NS (€)	1) €8,304.97 2) €7,449.59	Q-twist	1) €8910.25 2) €9293.79	€ 383,54	1) 10.48 months Q-twist 2) 7.17 months Q-twist	-3.31 months Q-twist	€ 2073/QALY	The cool excimer laser-assisted angioplasty provides an improved Q-TwiST in a cost-effective manner.
Simpson ⁴	2014	2009/2010 (£)	1) £4,071 2) £3,949 3) £3,837 4) £3,661 5) £4,316 6) £4,525 7) £6,171 8) £6,561 9) £7,367 10) £4,732		1) £49,890 2) £52,335 3) £55,199 4) £56,539 5) £54,775 6) £55,012 7) £55,928 8) £55,852 9) £58,097		1) 3.402 QALY 2) 3.297 QALY 3) 3.047 QALY 4) 2.988 QALY 5) 3.144 QALY 6) 3.157 QALY 7) 3.134 QALY 8) 3.144 QALY 9) 3.003 QALY		Strategy 1: dominant	For patients with IC and CLI, paclitaxel (drug)-coated balloons, is both less expensive and more clinically effective than all of the other options and, therefore, it dominates them.
Weinstock ⁴⁰	2014	2010 ($)	1) $10,516 2) $6,951		1) $13,487 2) $12,939	$549	1) 2.33 QALY 2) 2.17 QALY	0.16 QALY	$3,441/QALY	The index procedure costs and cost-effectiveness to 1 year were comparable for OAS+BA vs BA alone.

Table 3: Study characteristics and results of studies evaluating the cost-effectiveness of percutaneous interventions

Author

Year

Country

Indication

Population

Lesion type

Area

Interventions

Study design

Perspective

Time horizon

Outcomes

Discounting %

Sculpher ⁴¹

1996

IC/CLI

Patients with moderate to severe claudication or limb threatening ischemia, age 65,

occlusion

femoral or femoropopliteal

1) PTA
2) laser (Nd/YAG) assisted PTA

Model

NHS

25 years

Life-years QALY

6.0%

Bosch²¹

1998

60 year old man with claudication due to an iliac arterial stenosis

stenosis

iliac

Model

Societal

QALY

3.0%

Bosch²²

2000

60-year old male with life-limiting claudication due to stenoses in the iliac arteries, for whom a percutaneous intervention was indicated

stenosis

iliac

Model

Societal

Lifetime

QALY

Katsanos ³³

2013

CLI

The primary EES cohort: at least 1 infrapopliteal lesion (4.5 cm)
The Bail-out SES cohort: a CLI population was treated in both cases with the primary aim of limb salvage

infrapopliteal

two single centre prospectively controlled studies

direct healthcare expenditures

Lifetime

event-free survival

Table 3: Study characteristics and results of studies evaluating the cost-effectiveness of percutaneous interventions

Author

Year

Country

Indication

Population

Lesion type

Area

Interventions

Study design

Perspective

Time horizon

Outcomes

Discounting %

Sultan³⁵

2013

CLI

Patients with CLI, Rutherford categories 4, 5, or 6 with tibial TASC D lesion

Occlusion

tibial artery occlusive disease

1) Cool excimer laser-assisted angioplasty
2) Tibial balloon angioplasty

Single study

3 year

QALY (Q-twist)

Simpson⁴

2014

IC / CLI

Patients with symptomatic PAD suitable for endovascular treatment for disease distal to the inguinal ligament. IC patients 66 years; CLI patients 74 years

Distal to the inguinal ligament. femoropopliteal and infrapopliteal arteries

1) Paclitaxel-coated balloon
2) PTA with bail-out PES
3) PTA with bail-out BMS
4) PTA, no bail-out stenting
5) BMS
6) PES
7) EVBT
8) Stent-graft
9) Cryoplasty
10) SES

Model

NHS

Lifetime

LY
QALY

3.5%

Weinstock⁴⁰

2014

IC/CLI

Patients with Rutherford class 2–4 classification had lesions of 70% stenosis and fluoroscopically visible calcium

calcified femoropopliteal lesions

1) Orbital atherectomy + balloon angioplasty
2) Balloon angioplasty

Single study

Third party payer

1 year

QALY

Table 3: Study characteristics and results of studies evaluating the cost-effectiveness of percutaneous interventions - continued

Author	Year	Price year (currency)	Intervention costs	Health utilities	Total costs	Incremental costs	Total effectiveness	Incremental effects	ICER	Cost-effective?
Sculpher ⁴¹	1996	1993/1994 (£)	1) £1,186 2) £2,133	SF-36	IC 1) £3,669 IC 2) £3,929 RP/U 1) £8,716 RP/U 2) £8,823	IC: £260 RP/U: £107	IC 1) 6.78 LY; 5.78 QALY IC 2) 6.79 LY; 5.87 QALY RP/U 1) 5.44 LY; 4.40 QALY RP/U 2) 5.46 LY; 4.46 QALY	IC: 0.09 QALY RP/U: 0.06	IC: £3,040/QALY RP/U: £1,810/QALY	The results indicate a cost-effective role for the laser, but important areas of uncertainty exist.
Bosch ²¹	1998	1995 (US $)	PTA $2,343 PTA with SS $ 3,033 Primary stent placement $ 3,990	RAND-36 (SF36) /EQ5D	1) $3,368 2) $5,868 3) $6,573 4) $7,145 5) $7,504 6) $7,806 7) $8,763		1) 10.30 QALY 2) 11.03 QALY 3) 11.29 QALY 4) 11.36 QALY 5) 11.47 QALY 6) 11.61 QALY 7) 11.61 QALY		1) Reference strategy 2) Inferior by ED 3) $4,073/QALY 4) Inferior by ED 5) Inferior by ED 6) $4,519/QALY 7) Inferior by dominance	PTA with selective stent placement is a cost-effective treatment strategy compared with primary stent replacement or PTA alone in the treatment of intermittent claudication caused by an iliac arterial stenosis.
Bosch ²²	2000	1998 (US $)	PTA $5,207 PTA with SS $6,027	SF36	1) $4,531 2) $10,048 3) $10,903 4) $12,458 5) $12,830 6) $13,158		1) 7.79 QALY 2) 8.46 QALY 3) 8.63 QALY 4) 8.73 QALY 5) 8.79 QALY 6) 8.89 QALY		1) ref 2) ED 3) $7,624/QALY 4) ED 5) ED 6) $8,519/QALY	PTA followed by selective stent placement for unsatisfactory PTA results is a more cost-effective treatment strategy than PTA alone.
Katsanos ³³	2013	NS (€)	DES: €1,500	NA		1) €358,500 2) €498,000		1) 55 years (0.89 years p/patient) 2) 43 years (0.91 years p/patient)	1) €6,518/year 2) €11,581/year	In conclusion, both strategies of bail-out and primary drug-eluting stent placement in the infrapopliteal arteries for CLI treatment exhibit single-digit NNT and relatively low corresponding ICER.

Table 3: Study characteristics and results of studies evaluating the cost-effectiveness of percutaneous interventions - continued

Author	Year	Price year (currency)	Intervention costs	Health utilities	Total costs	Incremental costs	Total effectiveness	Incremental effects	ICER	Cost-effective?
Sultan ³⁵	2013	NS (€)	1) €8,304.97 2) €7,449.59	Q-twist	1)€8,910.25 2) €9,293.79	€ 383,54	1) 10.48 months Q-twist 2) 7.17 months Q-twist	-3.31 months Q-twist	€ 2073.19/QALY	The cool excimer laser-assisted angioplasty provides an improved Q-TwiST in a cost-effective manner.
Simpson ⁴	2014	2009/2010 (£)	1) £4,071 2) £3,949 3) £3,837 4) £3,661 5) £4,316 6) £4,525 7) £6,171 8) £6,561 9) £7,367 10) £4,732		1) IC: £12,668 CLI: £49,890 2) IC: £13,032 CLI: £52,335 3) IC: £14,637 CLI: £55,199 4) IC: £14,787 CLI: £56,539 5) IC: £15,030 CLI: £54,775 6) IC: £15,692 CLI: £55,012 7) IC: £15,891 CLI: £55,928 8) IC: £16,171 CLI: £55,852 9) IC: £17,578 CLI: £58,097		1) IC: 6.120 QALY; CLI: 3.402 QALY 2) IC: 6.081 QALY; CLI: 3.297 QALY 3) IC: 5.956 QALY; CLI: 3.047 QALY 4) IC: 5.931 QALY; CLI: 2.988 QALY 5) IC: 5.989 QALY; CLI: 3.144 QALY 6) IC: 5.993 QALY; CLI: 3.157 QALY 7) IC: 5.984 QALY; CLI: 3.134 QALY 8) IC: 5.989 QALY; CLI: 3.144 QALY 9) IC: 5.934 QALY; CLI: 3.003 QALY		1) Dominant	For patients with IC and CLI, paclitaxel (drug)-coated balloons are both less expensive and more clinically effective than all of the other options and, therefore dominates them.
Weinstock ⁴⁰	2014	2010 ($)	1) $10,516 2) $6,951	EQ5D SF36	1) $13,487 2) $12,939	$549	1) 2.33 QALY 2) 2.17 QALY	0.16 QALY	$3,441/QALY	The index procedure costs and cost-effectiveness to 1 year were comparable for OAS+BA vs BA alone.

Table 4: Study characteristics and results of other studies evaluating the cost-effectiveness of treatments for PAD

Author	Year	Country	Indication	Population	Lesion type	Interventions	Study design	Perspective	Time horizon	Outcomes	Discounting %
Visser⁴³	2003	NL	IC	60-year-old men with severe unilateral intermittent claudication of at least one year duration		1) No test‡ + SE therapy 2) DUS‡ + PTA/SE therapy 3) MRA‡ + PTA/SE therapy 4) DSA‡ + PTA/SE therapy 5) DUS + ‡PTA/BP/SE therapy 6) MRA‡ + PTA/BP/SE therapy 7) DSA + ‡PTA/BP/SE therapy	Model	Societal	Lifetime	QALY	3.0%

Table 4: Study characteristics and results of other studies evaluating the cost-effectiveness of treatments for PAD - continued

Author	Year	Price year (currency)	Intervention costs	Health utilities	Total costs	Incremental costs	Total effectiveness	Incremental effects	ICER	Cost-effective?
Visser⁴³	2003	1999 (€)	MRA €494 Colour-guided DUS €184 Intra-arterial DSA €1,062 Planned PTA and stopped after angiography €357 PTA for suprainguinal lesions €1,934 PTA for infrainguinal lesions €1,655 BP for suprainguinal lesions €10,179 BP for infrainguinal lesions €5,452 SE therapy, time costs per year €1,267		1) €6,793 2) €8,546 3) €8,566 4) €8,997 5) €18,720 6) €18,440 7) €18,583		1) 6.0606 QALY 2) 6.1465 QALY 3) 6.1487 QALY 4) 6.1498 QALY 5) 6.2002 QALY 6) 6.2136 QALY 7) 6.2254 QALY		1) ref 2) ED 3) €20,138/QALY 4) ED 5) Dominated 6) ED 7) €130,557/QALY	PTA preceded by MRA is the most cost-effective initial intervention, with SE therapy for those who are not suitable for PTA

Table 5.1: Quality of studies, quality of effectiveness, study limitations: Intermittent claudication – supervised exercise versus unsupervised exercise
Study	Population	Indication	Comparison	Interventions	Clinical data		Study limitations/ other comments	Drummond Score
					Data sources effectiveness	Quality of data source
					Data sources effectiveness	Quality of data source
Lee²⁴	IC	no specific location	SE vs CMT	1) CMT 2) CMT+SE therapy	Non-randomised Controlled Trial		1) Non-randomized trial 2) The study did not sufficiently capture the long-term compliance to each type of treatment nor was designed to evaluate the benefit to cardiovascular health that is associated with exercise 3) short time horizon (1 year) 4) no uncertainty analyses 5) non-preference based method of QALY calculation	Y: 19 N: 5 NA: 11
Asselt, van²⁰	IC	no specific location	SE vs UE	1) WA 2) SE therapy (with and without feedback)	EXITPAD	RSG: Low risk AC: Low risk BPP: High risk BOA: low risk IOD: low risk SR: low risk OB: unclear	1) The study did not sufficiently capture the long-term compliance to each type of treatment nor was designed to evaluate the benefit to cardiovascular health that is associated with exercise. 2) Short time horizon (1 year)	Y: 26 N: 4 NA: 5
NICE (App.K) ¹⁶	IC	no specific location	SE vs UE	1) UE therapy 2) SE therapy	Synthesis of literature		1) not possible to determine whether certain types of supervised programs are more cost effective than others 2) no published RCT data exist to inform the relative risk of cardiovascular events and mortality in people who exercise compared to those who do not in people with IC. 3) data on short and long term compliance to these regimens is not available 4) lifetime time horizon	Y: 30 N: 1 NA: 4
NICE (App.L) ¹²	IC	Aorto iliac & femoropopliteal separately	SE vs UE SE vs PTA PTA vs PTA	1) UE-SE 2) UE-PTA SS 3) UE-BP 4) SE-SE 5) SE-PTA SS 6) SE-BP 7) PTA SS-SE 8) PTA SS-PTA SS 9) PTA SS-BP 10) PTA PS-SE 11) PTA PS-PTA SS 12) PTA PS-BP 13) PTA SS+SE	Synthesis of literature		1) secondary interventions are associated with the same relative risk of mortality and morbidity as those observed in primary procedures 2) literature based model 3) not possible to determine whether certain types of supervised programs are more cost effective than others. 4) no published RCT data exist to inform the relative risk of cardiovascular events and mortality in people who exercise compared to those who do not in people with IC. 5) lifetime time horizon	Y: 30 N: 1 NA: 4
Reynolds ²⁶	IC	Aorto- iliac	SE vs UE SE vs PTA	1) SE therapy + OMC 2) Stenting + OMC 3) OMC	CLEVER		1) 5 year time horizon 2) assumption QOL	Y: 27 N: 6 NA: 2

Table 5.2: Quality of studies, quality of effectiveness, study limitations: Intermittent claudication – supervised exercise versus PTA
Study	Population	Indication	Comparison	Interventions	Clinical data		Study limitations/ other comments	Drummond Score
					Data sources effectiveness	Quality of data source
					Data sources effectiveness	Quality of data source
Vries, de¹⁹	IC	iliac & femoropopliteal	SE vs PTA	1) SE 2) SE±PTA 3) SE±PTA/BP 4) PTA/SE 5) PTA/BP/SE	Synthesis of literature		1) Assumed angioplasty preceded by catheter angiography 2) literature based model - combining evidence 3) lifetime time horizon	Y: 29 N: 3 NA: 3
Treesak ¹⁸	IC	ilio-femoral arterial disease	SE vs PTA	1) no treatment 2) SE therapy 3) PTA without primary stent placement	Synthesis of literature		1) short time horizon (1 year) 2) ICD/ACD outcome 3) literature based model - small studies 4) cost diagnostic workup not included	Y: 26 N: 5 NA: 4
Spronk ¹⁷	IC	Iliac or femoropopliteal arterial lesion	SE vs PTA	1) SE therapy 2) ER (PTA, if necessary a stent)	CETAC (Dutch RCT)		1) Short time horizon (1 year) 2) CETAC was a single centre study with adherence to strict in- and exclusion criteria	Y: 27 N: 2 NA: 6
NICE (App.L) ¹²	IC	Aorto-iliac & femoropopliteal separately	SE vs UE SE vs PTA PTA vs PTA	1) UE-SE 2) UE-PTA SS 3) UE-BP 4) SE-SE 5) SE-PTA SS 6) SE-BP 7) PTA SS-SE 8) PTA SS-PTA SS 9) PTA SS-BP 10) PTA PS-SE 11) PTA PS-PTA SS 12) PTA PS-BP 13) PTA SS+SE	Synthesis of literature		1) secondary interventions are associated with the same relative risk of mortality and morbidity as those observed in primary procedures 2) literature based model 3) not possible to determine whether certain types of supervised programs are more cost effective than others. 4) no published RCT data exist to inform the relative risk of cardiovascular events and mortality in people who exercise compared to those who do not in people with IC. 5) lifetime time horizon	Y: 30 N: 1 NA: 4
Mazari ²⁵	IC	femoropopliteal	SE vs PTA	1) PTA 2) SE therapy 3) PTA+SE therapy	NCT00798850		1) short time horizon (1 year) 2) This study did not include any adjunctive procedures 3) diagnostic angiography was used in this study before angioplasty for identification of suitable patients. MRA/DUS nowadays often used 4) costs of drugs could vary	Y: 24 N: 4 NA: 7
Reynolds ²⁶	IC	Aorto-iliac	SE vs UE SE vs PTA	1) SE therapy + OMC 2) Stenting + OMC 3) OMC	CLEVER		1) 5 year time horizon 2) assumption QOL	Y: 27 N: 6 NA: 2

Table 5.3: Quality of studies, quality of effectiveness, study limitations: Intermittent claudication – Revascularizations
Study	Population	Indication	Comparison	Interventions	Clinical data		Study limitations/ other comments	Drummond Score
					Data sources effectiveness	Quality of data source
					Data sources effectiveness	Quality of data source
Hunink ³⁸	IC/CLI	femoropopliteal	BP vs PTA	1) NT 2) PTA-NT 3) PTA-PTA 4) PTA-BP 5) BP-NT 6) BP-Rev	Synthesis of literature		1) Quality of life estimated using Torrence Multi Attribute Scale by healthcare workers; 2) Patency failure assumed to be equivalent to symptom progression & re-intervention (according to the GDG, not all patients who experience failure or symptom progression following angioplasty will undergo reintervention. Only those who return to their healthcare provider will be considered, and of those, the probability of treatment will depend on the location and extent of the lesion) 3) Progression of symptoms not modelled due to lack of data 4) lifetime time horizon	Y: 24 N: 8 NA: 3
Sculpher ⁴¹	IC/CLI	femoral or femoropopliteal	PTA vs PTA	1) PTA 2) laser (Nd/YAG) assisted PTA	Lammer et al.		1) the cost-effectiveness of the laser when used as a secondary intervention (on immediate failure) is based on only seven patients 2) literature based model 3) laser technology not available anymore in NL 4) time horizon 25 years 5) disease progression was not modelled	Y: 28 N: 4 NA: 3
Bosch ²¹	IC	iliac	PTA vs PTA	1) No revascularization 2) PTA followed by no revascularization 3) PTA with selective stent placement followed by no revascularization 4) PTA and repeated PTA 5) PTA followed by PTA with selective stent placement 6) Initial and repeated PTA with selective stent placement 7) Primary stent placement followed by PTA with selective stent placement	Synthesis of literature		1) outdated 2) follow-up DIST 9 months 3) assumed SP and PTA same mortality/complications 4) subgroup with occlusions or critical ischemia was underrepresented. Meta-analysis, only 122 (15%) of 816 patients had critical ischemia, and 254 (28%) of 906 lesions were occlusions. In the RCT, only 18 (6%) of 279 patients were treated for critical ischaemia, and 29 (8%) of 356 lesions were occlusions. 5) literature based model - DIST trial + meta-analysis 6) time horizon unclear	Y: 29 N: 4 NA: 2
Bosch²²	IC	iliac	PTA vs PTA	1) No revascularization 2) PTA followed by no revascularization 3) PTA with selective stent placement followed by no revascularization 4) PTA and repeated PTA 5) PTA followed by PTA with selective stent placement 6) Initial and repeated PTA with selective stent placement	Synthesis of literature		1) cost accounting was based on a small sample size 2) cost accounting for the additional hospital costs not performed 3) generalizability 4) literature based model	Y: 25 N: 9 NA: 1
Muradin ³⁹	IC/CLI	femoropopliteal	BP vs PTA	1) PTA 2) BP	Synthesis of literature		1) teaching hospital 2) assumptions on complications 3) literature based model (RCT, meta-analysis) 4) lifetime time horizon	Y: 27 N: 5 NA: 3

Table 5.3: Quality of studies, quality of effectiveness, study limitations: Intermittent claudication – Revascularizations continued
Study	Population	Indication	Comparison	Interventions	Clinical data		Study limitations/ other comments	Drummond Score
					Data sources effectiveness	Quality of data source
					Data sources effectiveness	Quality of data source
NICE (App.L) ¹²	IC	Aorto- iliac & femoropopliteal separately	SE vs UE SE vs PTA PTA vs PTA	1) UE-SE 2) UE-PTA SS 3) UE-BP 4) SE-SE 5) SE-PTA SS 6) SE-BP 7) PTA SS-SE 8) PTA SS-PTA SS 9) PTA SS-BP 10) PTA PS-SE 11) PTA PS-PTA SS 12) PTA PS-BP 13) PTA SS+SE	Synthesis of literature		1) secondary interventions are associated with the same relative risk of mortality and morbidity as those observed in primary procedures 2) literature based model 3) not possible to determine whether certain types of supervised programs are more cost effective than others 4) no published RCT data exist to inform the relative risk of cardiovascular events and mortality in people who exercise compared to those who do not in people with IC 5) lifetime time horizon	Y: 30 N: 1 NA: 4
Simpson ⁴	IC/CLI	femoropopliteal and infrapopliteal arteries	PTA vs PTA	1) Paclitaxel-coated balloon 2) PTA with bail-out paclitaxel-eluting stents 3) PTA with bail-out BMSs 4) PTA, no bail-out stenting 5) BMS 6) PES 7) EVBT 8) Stent-graft 9) Cryoplasty 10) Sirolimus-eluting stent	Synthesis of literature		1) small trials included in the SR for literature based model, matching populations? 2) clinical outcomes, such as claudication distance, quality of life and reintervention could not be linked with patency due to lack of evidence 3) The assumption prolonged patency led to cost savings as a result of fewer reinterventions was based on relatively little direct evidence 4) lifetime time horizon	Y: 29 N: 3 NA: 3
Weinstock ⁴⁰	IC/CLI	calcified femoropopliteal lesions	PTA vs PTA	1) orbital atherectomy + balloon angioplasty 2) balloon angioplasty	COMPLIANCE		1) short time horizon (1 year) 2) compliance study small sample size and many important parameters not estimated	Y: 19 N: 5 NA: 11

Table 5.4: Quality of studies, quality of effectiveness, study limitations: Intermittent claudication – Medication
Study	Population	Indication	Comparison	Interventions	Clinical data		Study limitations/ other comments	Drummond Score
					Data sources effectiveness	Quality of data source

Guest ²³	IC	no specific location	MED vs MED	1) Cilostazol 2) Naftidrofuryl 3) Pentoxifylline	Synthesis of literature		1) short time horizon (24 weeks) 2) MWD outcome 3) literature based model - RCT’s 4) older clinical trials with naftidrofuryl compared to cilostazol, and less stringent criteria were used to assess patients’ walking distances. Consequently, the outcomes pertaining to naftidrofuryl may have been over-estimated. 5) resource use expert opinion 6) no vasoactive comparator included	Y: 25 N: 3 NA: 7
Squires ⁵	IC	no specific location	MED vs MED	1) No vascular drug 2) Pentoxifylline (1200mg/d) 3) Cilostazo (200mg/d) 4) Naftidrofuryl oxalate (600mg/d) 5) Inositol nicotinate	Synthesis of literature		1) uncertainty regarding the utility estimates and discontinuation rate beyond 24 weeks because most RCT’s do not have follow-up beyond this time point. The analysis takes the conservative assumption that there is no benefit of the vasoactive drugs following discontinuation. 2) The model assumes that the drugs are prescribed for symptom relief and have no impact on the progression of disease or serious cardiovascular events. 3) literature based model 4) time horizon 100 years	Y: 27 N: 3 NA: 5

Table 6.1: transferability: Intermittent claudication - supervised exercise versus unsupervised exercise

Study

Transferability (specific knock-out criteria)

Methodological characteristics

Healthcare system characteristics

Population characteristics

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Lee²⁴

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- not relevant (study horizon 1 year)
- very high
- not relevant (not included)

- low (UK, NHS)
-
- high (Finance Department of the Hull Royal Infirmary & literature)
-

- biased

- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (UK)

- high (UK)
- very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium
- low (old patients)

- very high (UK)
- high (UK)
- medium (SF-36)
- high (UK)
- high (UK)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Asselt, van²⁰

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- not relevant (study horizon 1 year)
- very high
- high

- very high (NL, societal)
-
- very high (retrospective cost collection EXITPAD, Dutch cost manual, Dutch pharmaceutical compass, financial department Atrium)
- very high

- unbiased
-
- unbiased
- unbiased

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very high (NL)
- very high (NL)
- very high (NL)

- unbiased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium
- low (old patients)

- very high (NL)
- very high (NL)
- very high (EQ-5D)
- very high (NL)
- very high (NL)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

NICE (App.K) ¹⁶

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- medium (NHS PSSRU & literature)
-

- unbiased
- biased
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (UK)
- medium (UK)
- very high (UK)

- biased
- biased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium
- low (old patients)

- very high (UK)
- high (UK)
- high (EQ-5D, however mapped Sf-36)
- high (UK)
- high (UK)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

NICE (App.L) ¹²

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- medium (NHS PSSRU & literature)
-

- biased
- unbiased
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (UK)
- high (UK)
- very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium
- low (old patients)

- very high (UK)
- high (UK)
- high (EQ-5D, however mapped SF-36)
- high (UK)
- high (UK)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Reynolds ²⁶

- Perspective
- Discounting
- medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- high

- high (US, societal)
- medium (3%)
- very high: resource use and costs measured alongside CLEVER trial
- medium

- unbiased
- unbiased

- unbiased
- unbiased: no friction costs but replacement is not necessary and small percentage working

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (US)
- high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium
- low (old patients)

- very high (US)
- high (US)
- very high (EQ-5D)
- high (US, PTA very high)
- high (US)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Table 6.2: transferability: Intermittent claudication - supervised exercise versus PTA

Study

Transferability (specific knock-out criteria)

Methodological characteristics

Healthcare system characteristics

Population characteristics

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Vries, de¹⁹

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- high (US, societal)
- medium (3%)
- medium (costs based on hospital database with charges (charge: cost ratio used))
-

- unbiased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very low (US, outdated)
- very high (clinical data NL)
- very high (clinical data NL)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium (SE), low (PTA)
- low (old patients)

- medium (NL data, young patients)
- very high (NL data)
- very high (EQ-5D)
- very high (NL, observational data)
- high (US)

- biased
- unbiased
- unbiased
- unbiased
- unbiased

Treesak ¹⁸

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- not relevant (study horizon 1 year)
- very high
- high

- high (US, societal)
-
- medium (costs of a sample of procedures)
- forgone earnings of the average working person residing in the USA. Average monetary values of wage rates

- unbiased
-
- unbiased
- unbiased: no friction costs but replacement is not necessary and small percentage working

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very low (US, outdated)
- high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- not relevant (no QALYS)
- medium (SE), PTA (low)
- low (old patients)

- unclear (no details population)
- high (UK)
-
- high (UK)
- high (UK)

-
- unbiased
-
- unbiased
- unbiased

Spronk ¹⁷

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- very high (NL societal)
- Medium (3%)
- high (CETAC RCT)
-

- unbiased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very high (NL)
- very high (NL)
- very high (NL)

- unbiased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium (PTA low)
- low (old patients)

- high (NL Dutch RCT, PFWD short, small % smokers)
- very high (NL Dutch RCT)
- very high (EQ-5D)
- very high (NL compliance SE RCT vs compliance real world)
- very high (NL)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

NICE (App.L) ¹²

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- medium (NHS PSSRU & literature)
-

- biased
- unbiased

- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium: UK, 3months SE: £288p/p
- high (UK)
- Very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium (PTA: low)
- low (old patients)

- very high (UK)
- high (UK)
- high (EQ-5D, however mapped SF-36)
- high (UK)
- high (UK)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Mazari ²⁵

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- not relevant (study horizon 1 year)
- very high
- not relevant (not included)

- Low (UK, NHS)
-
- Medium (Finance Department of the Hull Royal Infirmary & literature)
-

- biased
-
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (UK)
- high (UK)
- very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium (PTA low)
- low (old patients)

- medium (UK, young patients)
- high (UK)
- medium (SF36)
- high (UK)
- high (UK)

- biased
- unbiased
- biased
- unbiased
- unbiased

Reynolds ²⁶

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- high

- high (US, societal)
- medium (3%)
- very high: resource use and costs measured alongside CLEVER trial
- medium: participants time * nominal wage (not friction costs)

- unbiased
- unbiased
- unbiased
- unbiased: no friction costs but replacement is not necessary and small percentage working

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- low
- low

- low (US, $4,088 6 months)
- high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium (PTA: low)
- low (old patients)

- very high (US)
- high (US)
- very high (EQ-5D)
- high (US, PTA very high)
- high (US)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Table 6.3: transferability: Intermittent claudication - supervised exercise versus PTA - revascularization

Study

Transferability (specific knock-out criteria)

Methodological characteristics

Healthcare system characteristics

Population characteristics

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Hunink ³⁸

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (US, health care system)
- medium (5%)
- very high (micro-costing)
-

- biased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very low (1990 prices, US, charges)
- very high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (US)
- high (US)
- low (Torrance Multi-Attribute Scale, physicians)
- high (US)
- high (US)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Sculpher ⁴¹

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (6%)
- medium (NHS PSSRU & literature)
-

- biased
- unbiased

- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (UK, outdated)
- very high
- very low

- biased
- unbiased
- biased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (UK)
- high (UK)
- medium (SF-36 /EQ5D/physicians, patients)
- high (UK)
- high (UK)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Bosch ²¹

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- very high (NL, societal)
- medium (3%)
- very high: resource use and costs measured alongside DIST trial
-

- unbiased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (NL, outdated)
- very high (NL)
- very high (NL)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (NL)
- very high
- medium (HUI, SF-36, EQ5D)
- very high (NL)
- very high (NL)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Bosch²²

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- high (US, societal)
- medium (3%)
- very high (micro-costing)
- medium (multiplying the mean number of hospitalization days by the average gross earnings of a full-time 60-year-old male employee, including fringe benefits per day)

- unbiased
- unbiased
- unbiased
- unbiased: no friction costs but replacement is not necessary and small percentage working

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very low (US, outdated)
- very high (NL)
- very high (NL)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (NL)
- very high
- medium (HUI, SF-36, EQ5D)
- very high (NL)
- high (US)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Table 6.3: transferability: Intermittent claudication - supervised exercise versus PTA – revascularization continued

Study

Transferability (specific knock-out criteria)

Methodological characteristics

Healthcare system characteristics

Population characteristics

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Muradin ³⁹

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (US, health care system)
- medium (3%)
- very high (based on published and unpublished data (micro-costing))
-

- biased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very low (US, outdated, charges)
- very high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (US)
- high (US)
- low (physicians, abbreviate form of HUI)
- high (US)
- high (US)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

NICE (App.L) ¹²

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- medium (NHS PSSRU & literature)
-

- biased
- unbiased
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium: UK, 3months SE: £288p/p
- high (UK)
- Very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium (PTA: low)
- low (old patients)

- very high (UK)
- high (UK)
- high (EQ-5D, however mapped SF-36)
- high (UK)
- high (UK)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Simpson ⁴

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- medium (NHS PSSRU & literature)
-

- biased
- unbiased - biased

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (UK)
- very high (UK)
- very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (UK)
- high (UK)
- very high (EQ5D)
- high (UK)
- high (UK)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Weinstock ⁴⁰

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- not relevant (study horizon 1 year)
- very high
- not relevant (not included)

- low (US, third party payer)
-
- medium (DRG)
-

- biased
-
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (US)
- medium
- low (orbital atherectomy)

- biased
- biased
- biased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (US)
- high (US)
- medium (SF36/EQ5D)
- high (US)
- high (US)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Table 6.4: transferability: Intermittent claudication - medication

Study

Transferability (specific knock-out criteria)

Methodological characteristics

Healthcare system characteristics

Population characteristics

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Guest ²³

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- low (time horizon 24 weeks)
- not relevant (time horizon 24 weeks)
- very high
- not relevant (not included)

- low (UK, NHS)
-
- medium (NHS PSSRU & drug tariff)
-

- unbiased
-
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- very high

- medium (UK)
- very high
- very low (cilostazol, naftidrofuryl NA in NL)

- biased
- unbiased
- biased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- not relevant (QALY not included)
- medium
- low (old patients)

- high (UK)
- high (UK)
-
- high (UK)
- high (UK)

- unbiased
- unbiased
-
- unbiased
- unbiased

Squires ⁵

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- very high (drug tariff)
-

- biased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- very high

- medium (UK)
- very high
- very low (cilostazol, naftidrofuryl NA in NL)

- biased
- unbiased
- biased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- medium
- low (old patients)

- high (UK)
- high (UK)
- medium (SF36)
- high (UK)
- high (UK)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Table 7.1: Quality of studies, quality of effectiveness, study limitations: Critical limb ischaemia – amputation versus bypass

Study

Population

Indication

Comparison

Interventions

Clinical data

Study limitations/ other comments

Drummond Score

Data sources effectiveness

Quality of data source

Brothers ³⁰

CLI

tibial-peroneal

AMP vs BP

1) Primary amputation
2) Expectant management
3) BP

observational study

1) Long-term patient survival, limb salvage rate, and primary and cumulative secondary patency rates were obtained from the results of retrospective analyses previously conducted by the authors with no evidence of a systematic search;
2) utility values were obtained from people with CLI rather than patients who had experienced each health state QALY gain was considered only over a 5-year horizon, therefore, this study will underestimate the long-term effect of reduced operative mortality expected from both the expectant management and primary amputation strategies;
3) unclear method of QALY elicitation and valuation.
4) decision tree

5) costs not discounted

Y: 24
N: 9
NA: 2

Barshes ²⁷

CLI

infrainguinal arterial occlusive disease

AMP vs BP/PTA

1) Local wound care
2) Primary amputation
3) BP with surgical revisions
4) BP with endovascular revisions
5) Purely endovascular
6) Endovascular, BP for failure

Synthesis of literature

1) 10 year time horizon
2) literature based model

Y: 23
N: 9
NA: 3

Barshes ²⁸

CLI

infrainguinal, infrapopliteal

AMP vs BP

1) Conservative management
2) Primary amputation
3) BP with PTFE 6 vein patch
4) BP with AAV
5) BP with cryopreserved venous allograft
6) BP with small-caliber GSV
7) BP with cryopreserved arterial allograft

Synthesis of literature

1) 10 year time horizon
2) combination studies (cost single centre)
3) literature based model

4) no QALYs

Y: 19
N: 13
NA: 3

Barshes ³⁷

CLI

infrainguinal

AMP vs BP/PTA

Synthesis of literature

1) 10 year time horizon
2) combination studies
3) literature based model
4) no QALYs

Y: 21
N: 9
NA: 5

Barshes ³⁶

CLI

infrainguinal

AMP vs BP/PTA

1) local wound care, with selective major amputation as indicated
2) primary amputation
3) infrainguinal surgical bypass using an autologous vein conduit and open surgical intervention as needed
4) infrainguinal surgical bypass using an autologous vein conduit and subsequent endovascular intervention as needed
5) endovascular intervention, subsequent endovascular reinterventions as needed
6) endovascular intervention, surgical bypass as needed

Synthesis of literature

1) 10 year time horizon
2) combination studies
3) literature based model
4) perspective unclear
5) no QALYs

Y: 18
N: 12
NA: 5

Table 7.2: Quality of studies, quality of effectiveness, study limitations: Critical limb ischaemia – BP versus PTA

Study

Population

Indication

Comparison

Interventions

Clinical data

Study limitations/ other comments

Drummond Score

Data sources effectiveness

Quality of data source

Hunink ³⁸

IC/CLI

femoropopliteal

BP vs PTA

1) NT
2) PTA-NT
3) PTA-PTA
4) PTA-BP
5) BP-NT
6) BP-Rev

Synthesis of literature

1) Quality of life estimated using Torrence Multi Attribute Scale by healthcare workers;
2) Patency failure assumed to be equivalent to symptom progression & re-intervention (according to the GDG, not all patients who experience failure or symptom progression following angioplasty will undergo reintervention. Only those who return to their healthcare provider will be considered, and of those, the probability of treatment will depend on the location and extent of the lesion)
3) Progression of symptoms not modelled due to lack of data
4) lifetime time horizon

Y: 24
N: 8
NA: 3

Muradin ³⁹

IC/CLI

femoropopliteal

BP vs PTA

1) PTA
2) BP

Synthesis of literature

1) teaching hospital
2) assumptions on complications
3) literature based model (RCT, meta-analysis)
4) lifetime time horizon

Y: 27
N: 5
NA: 3

Holler ³²

CLI

no specific location

BP vs PTA
PTA vs MED
BP vs MED

1 NT + NT
2 NT + PGE1
3 NT + PTA
4 NT + BP
5 PGE1 + NT
6 PGE1 + PGE1
7 PGE1 + PTA
8 PTA + NT

9 PTA + PGE1
10 BP + NT
11 PTA + PTA
12 BP + PGE1
13 BP + PTA
14 PGE1 + BP
15 PTA + BP
16 BP + BP

Synthesis of literature

1) 5 year time horizon
2) literature based - systematic review
3) lack of data, assumptions effectiveness
4) interaction interventions not included

Y: 27
N: 7
NA: 1

Sultan ³⁴

CLI

femoral and popliteal lesions

BP vs PTA

1) subintimal angioplasty
2) BP

observational study

1) non-preference based method of QALY calculation (Qtwist)
2) 5 year time horizon
3) based on observational study

Y: 15
N: 12
NA: 8

Bradbury ²⁹

CLI

aortoiliac excluded only infrainguinal

BP vs PTA

1) PTA
2) BP

BASIL

AC: low risk
RSG: low risk
“By the end of the follow-up period in the BASIL study, 99% of patients had been followed up at one year, 74% at two years, 48% at three years, 22% at four years, and only 8% at five years.”

1) short time horizon (3 year QALYs)
2) resource use and unit costs not reported
3) analysis of uncertainty based on undiscounted costs and discounted QALYs
4) cost of amputation not accounted for

Y: 25
N: 3
NA: 7

Forbes ³¹

CLI

infrainguinal

BP vs PTA

1) BP
2) PTA

BASIL

1) short time horizon (3 years)
2) missing data HRQOL

Y: 24
N: 5
NA: 6

Table 7.3: Quality of studies, quality of effectiveness, study limitations: Critical limb ischaemia – PTA versus PTA
Study	Population	Indication	Comparison	Interventions	Clinical data		Study limitations/ other comments	Drummond Score
					Data sources effectiveness	Quality of data source
					Data sources effectiveness	Quality of data source
Sculpher ⁴¹	IC/CLI	femoral or femoropopliteal	PTA vs PTA	1) PTA 2) laser (Nd/YAG) assisted PTA	Lammer et al.		1) the cost-effectiveness of the laser when used as a secondary intervention (on immediate failure) is based on only seven patients 2) literature based model) 3) time horizon 25 years	Y: 28 N: 4 NA: 3
Katsanos ³³	CLI	infrapopliteal	PTA vs PTA	1) Bail-out SES cohort 1.1 Bail-out use of SES after suboptimal balloon angioplasty 1.2 BMS after suboptimal balloon angioplasty 2) Primary EES cohort 2.1 Full-lesion primary EES 2.2 Plain balloon angioplasty and bail-out bare metal stenting as necessary	observational studies		1) event-free survivals 2) lifetime time horizon	Y: 18 N: 8 NA: 9
Sultan ³⁵	CLI	tibial artery occlusive disease	PTA vs PTA	1) cool excimer laser-assisted angioplasty 2) tibial balloon angioplasty	observational study		1) non-preference based method of QALY calculation (Qtwist) 2) 3 year time horizon 3) observational study, no matching	Y: 14 N: 14 NA: 7
Simpson ⁴	IC/CLI	femoropopliteal and infrapopliteal arteries	PTA vs PTA	1) Paclitaxel-coated balloon 2) PTA with bail-out paclitaxel-eluting stents 3) PTA with bail-out BMSs 4) PTA, no bail-out stenting 5) BMS 6) PES 7) EVBT 8) Stent-graft 9) Cryoplasty 10) Sirolimus-eluting stent	Synthesis of literature		1) small trials included in the SR for literature based model, matching populations? 2) clinical outcomes, such as claudication distance, quality of life and reintervention could not be linked with patency due to lack of evidence 3) The assumption prolonged patency led to cost savings as a result of fewer reinterventions was based on relatively little direct evidence 4) lifetime time horizon	Y: 29 N: 3 NA: 3
Weinstock ⁴⁰	IC/CLI	calcified femoropopliteal lesions	PTA vs PTA	1) orbital atherectomy + balloon angioplasty 2) balloon angioplasty	COMPLIANCE		1) 1 year time horizon 2) compliance study small and many important parameters not estimated	Y: 19 N: 5 NA: 11

Table 8.1: transferability: Critical limb ischaemia – amputation vs revascularization

Study

Transferability (specific knock-out criteria)

Methodological characteristics

Healthcare system characteristics

Population characteristics

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Brothers ³⁰

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (US, institutional)
- low (utility 5%, costs 0%)
- medium (patient charges)
-

- biased
- biased
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (US, outdated)
- very high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- medium (US)
- high (US)
- unclear from article
- high (US)
- high (US)

- biased
- unbiased
- biased/unclear
- unbiased
- unbiased

Barshes ²⁷

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (US, societal)
- medium (3.5%)
- medium (registries at the Brigham and Women’s Hospital combined with institution’s transition cost accounting system)
-

- unbiased
- unbiased
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (US)
- very high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (US)
- high (US)
- unclear (combination of measures)
- high (US)
- high (US)

- unbiased
- unbiased
- biased/unclear
- unbiased
- unbiased

Barshes ²⁸

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (US, societal)
- medium (3.5%)
- medium
-

- unbiased
- unbiased
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (US)
- very high (US)
- medium (US)

- biased
- unbiased
- biased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- medium (US)
- high (US)
-
- high (US)
- high (US)

- biased
- unbiased
-
- unbiased
- unbiased

Barshes ³⁷

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (US, societal)
- medium (3.5%)
-
-

- unbiased
- unbiased
-
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (US)
- very high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- medium (US)
- high (US)
-
- high (US)
- high (US)

- biased
- unbiased
-
- unbiased
- unbiased

Barshes ³⁶

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (US, societal)
- medium (3.5%)
-
-

- unbiased
- unbiased
-
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (US)
- very high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- medium (US)
- high (US)
-
- high (US)
- high (US)

- biased
- unbiased
-
- unbiased
- unbiased

Table 8.2: transferability: Critical limb ischaemia – BP versus PTA

Study

Transferability (specific knock-out criteria)

Methodological characteristics

Healthcare system characteristics

Population characteristics

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Hunink ³⁸

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (US, health care system)
- medium (5%)
- very high (micro-costing:)
-

- biased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very low (1990 prices, US, charges)
- very high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- low (US)
- high (US)
- low (Torrance Multi-Attribute Scale, physicians)
- high (US)
- high (US)

- biased
- unbiased
- biased
- unbiased
- unbiased

Muradin ³⁹

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (US, health care system)
- medium (3%)
- very high (micro-costing)
-

- biased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- very low (US, outdated, charges)
- very high (US)
- very high (US)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- low (US)
- high (US)
- low (physicians, abbreviate form of HUI)
- high (US)
- high (US)

- biased
- unbiased
- biased
- unbiased
- unbiased

Holler ³²

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- medium (DE, societal)
- low (only costs 5%)
- unclear
-

- unbiased
- biased
- unclear
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (DE)
- very high (DE)
- very high (DE)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (DE)
- high (DE)
- very high (EQ5D)
- high (DE)
- high (DE)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Sultan ³⁴

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- High (indirect costs included)

- low (IR, NS)
- unclear (NS)
- unclear (NS)
- unclear (NS)

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (IR)
- very high (IR)
- very high (IR)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (IR)
- high (IR)
- low (Qtwist)
- high (IR)
- high (IR)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Bradbury ²⁹

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- very high (micro costing)
-

- biased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (UK)
- very high (UK)
- very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (UK)
- high (UK)
- very high (EQ5D)
- high (UK)
- high (UK)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Forbes ³¹

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- very high (micro costing)
-

- biased
- unbiased
- unbiased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (UK)
- very high (UK)
- very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (UK)
- high (UK)
- very high (EQ5D)
- high (UK)
- high (UK)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Table 8.3: transferability: Critical limb ischaemia – PTA versus PTA

Study

Transferability (specific knock-out criteria)

Methodological characteristics

Healthcare system characteristics

Population characteristics

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Estimated relevance

Estimated correspondence (study vs decision country)

Estimation of CER

Sculpher ⁴¹

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (6%)
- medium (NHS PSSRU & literature)
-

- biased
- unbiased
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (UK, outdated)
- very high
- very low

- biased
- unbiased
- biased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (UK)
- high (UK)
- medium (SF36/EQ5D/ physicians, patients)
- high (UK)
- high (UK)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Katsanos ³³

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar))
- very high
- not relevant (not included)

- low (GR, direct HC expenditures)
- unclear (NS)
- unclear (publicly available information on net DES costs/no ref)
-

- biased
- unclear
- unclear
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (GR)
- very high (GR)
- very high (GR)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- not included
- low (PTA)
- low (old patients)

- medium
- medium
-
- medium
- medium

- biased
- unbiased
-
- unbiased
- unbiased

Sultan ³⁵

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- unclear (IR, NS)
- unclear (NS)
- unclear
-

- unclear
- unclear
- unclear
-

- absolute and relative prices in healthcare,
- practice variation,
- technology availability
-

- very high
- medium
- low

- medium (IR)
- very high
- very high (cool eximer laser)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (IR)
- high (IR)
- low (Qtwist)
- high (IR)
- high (IR)

- unbiased
- unbiased
- biased
- unbiased
- unbiased

Simpson ⁴

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- low (survival/costs long term similar)
- very high
- not relevant (not included)

- low (UK, NHS)
- medium (3.5%)
- medium (NHS PSSRU & literature)
-

- biased
- unbiased
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- medium (UK)
- very high (UK)
- very high (UK)

- biased
- unbiased
- unbiased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (UK)
- high (UK)
- very high (EQ5D)
- high (UK)
- high (UK)

- unbiased
- unbiased
- unbiased
- unbiased
- unbiased

Weinstock ⁴⁰

- Perspective
- Discounting
- Medical cost approach
- Productivity costs approach

- very high
- not relevant (study horizon 1 year)
- very high
- not relevant (not included)

- low (US, third party payer)
-
- medium (UB04-DRG)
-

- biased

-
- biased
-

- Absolute and relative prices
- Practice variation
- Technology availability

- very high
- medium
- low

- low (US)
- medium
- low (orbital atherectomy)

- biased
- biased
- biased

- case-mix
- life-expectancy
- health-status preferences
- acceptance / compliance / incentives to patients
- productivity and work-loss time

- medium
- medium
- medium
- low
- low (old patients)

- high (US)
- high (US)
- medium (SF36/EQ5D)
- high (US)
- high (US)

- unbiased
- unbiased
- biased
- unbiased
- unbiased