Drug‐eluting balloon angioplasty versus uncoated balloon angioplasty for the treatment of in‐stent restenosis of the femoropopliteal arteries

Ahmed Kayssi; Wissam Al‐Jundi; Giuseppe Papia; Daryl S Kucey; Thomas Forbes; Dheeraj K Rajan; Richard Neville; Andrew D Dueck

doi:10.1002/14651858.CD012510.pub2

Angioplastia con balón impregnado de fármacos versus angioplastia con balón no recubierto para el tratamiento de la reestenosis intrastent de las arterias femoropoplíteas

Authors' declarations of interest

Version published: 26 January 2019 Version history

https://doi.org/10.1002/14651858.CD012510.pub2

Collapse all Expand all

Resumen

available in

Antecedentes

Los stents se colocan en las arterias femoropoplíteas por varias razones, como la enfermedad ateroesclerótica, la necesidad de disección y la perforación de las arterias, y que se pueden colocar un stent con el tiempo. Se habla de "reestenosis intrastent" cuando un stent desarrolla una estenosis que limita el flujo. Se cree que la reestenosis intrastent es resultado de un proceso conocido como "hiperplasia intimal" en lugar de por la evolución de la enfermedad ateroesclerótica. El tratamiento de la reestenosis intrastent puede incluir la angioplastia con balón, en que se coloca otro stent dentro del stent estenosado para forzar la apertura y crear una derivación para que llegue sangre alrededor del stent. No está aclarada la función de las tecnologías de recubrimiento con fármacos, como los balones impregnados de fármacos (BIF), en el tratamiento de la reestenosis intrastent. Los balones impregnados de fármacos podrían ser útiles mediante el recubrimiento del interior de los stents estenosados con productos químicos citotóxicos como el paclitaxel, y mediante la inhibición de los procesos hiperplásicos responsables de la reestenosis intrastent. Es importante realizar esta revisión sistemática para evaluar la eficacia del BIF debido a la posibilidad de mayores gastos asociados con los BIF en comparación con la angioplastia con balón no recubierto, también conocida como angioplastia con balón simple antigua (POBA).

Objetivos

Evaluar la seguridad y la eficacia de los BIF en comparación con la angioplastia con balón no recubierto en pacientes con reestenosis intrastent de las arterias femoropoplíteas, evaluadas con criterios como la supervivencia sin amputación, la permeabilidad del vaso, la revascularización de la lesión objetivo, la tasa de reestenosis binaria y la muerte. La "reestenosis intrastent" se define como una estenosis de 50% o mayor de un vaso sanguíneo con un stent previo, diagnosticada por ecografía dúplex o angiografía.

Métodos de búsqueda

El especialista en información del Grupo Cochrane Vascular (Cochrane Vascular Information Specialist) buscó en el registro especializado del Grupo Cochrane Vascular (Cochrane Vascular Specialised Register), en CENTRAL, MEDLINE, Embase, en las bases de datos CINAHL, en la World Health Organization International Clinical Trials Registry Platform y en los registros de ensayos ClinicalTrials.gov hasta el 28 noviembre 2017. Los autores de la revisión también verificaron las referencias para identificar estudios adicionales.

Criterios de selección

Se incluyeron todos los ensayos controlados aleatorios que comparan los BIF versus la angioplastia con balón no recubierto para el tratamiento de la reestenosis intrastent en las arterias femoropoplíteas.

Obtención y análisis de los datos

Dos autores de la revisión (AK, WA) seleccionaron de forma independiente los ensayos apropiados y realizaron la extracción de datos, la evaluación de la calidad de los ensayos y el análisis de los datos. Los desacuerdos fueron resueltos por el autor principal de la revisión (DKR).

Resultados principales

Tres ensayos que asignaron al azar a un total combinado de 263 participantes cumplieron con los criterios de inclusión de la revisión. Los tres ensayos examinaron el tratamiento de la reestenosis sintomática intrastent en las arterias femoropoplíteas. Estos ensayos se realizaron en Alemania y Austria y utilizaron paclitaxel como agente de los balones impregnados de fármacos. Dos de los tres ensayos fueron patrocinados por la industria. Dos empresas fabricaron los balones impregnados de fármacos (Eurocor, Bonn, Alemania; Medtronic, Fridley, Minnesota, EE.UU). Los ensayos examinaron variables de evaluación anatómicas y clínicas. Se observó heterogeneidad en la frecuencia de colocación de stents de rescate entre los estudios así como en la dosis de paclitaxel aplicada por los BIF. Con el uso de los criterios de evaluación GRADE, se determinó que la certeza de la evidencia presentada fue muy baja para los resultados de amputación, revascularización de la lesión objetivo, reestenosis binaria, muerte y mejoría de una o más categorías Rutherford. La mayoría de los participantes se siguieron durante 12 meses, pero un ensayo siguió a los participantes durante 24 meses.

Los resultados de los ensayos no muestran diferencias en la incidencia de amputación entre los BIF y la angioplastia con balón no recubierto. Con los BIF se obtuvieron mejores resultados hasta los 24 meses para la revascularización de la lesión objetivo (odds ratio [OR] 0,05; intervalo de confianza [IC] del 95%: 0,00 a 0,92 a los seis meses; OR 0,24; IC del 95%: 0,08 a 0,70 a 24 meses) y a seis y 12 meses para la reestenosis binaria (OR 0,28; IC del 95%: 0,14 a 0,56 a seis meses; OR 0,34; IC del 95%: 0,15 a 0,76 a los 12 meses). Los pacientes tratados con BIF también mostraron una mejoría de una o más categorías de Rutherford a los seis y 12 meses (OR 1,81; IC del 95%: 1,02 a 3,21 a seis meses; OR 2,08; IC del 95%: 1,13 a 3,83 a los 12 meses). Los datos no muestran diferencias claras de la mortalidad entre los BIF y la angioplastia con balón no recubierto. Los datos fueron insuficientes para realizar análisis de subgrupos o de sensibilidad.

Conclusiones de los autores

Sobre la base de un metanálisis de tres ensayos con 263 participantes, la evidencia indica una ventaja de los BIF en comparación con la angioplastia con balón no recubierto para las variables de evaluación anatómicas, como la revascularización objetivo de la lesión y la reestenosis binaria, y para una variable de evaluación clínica en la categoría de Rutherford posintervención hasta los 24 meses. Sin embargo, la certeza de la evidencia para todos estos resultados es muy baja debido al número reducido de estudios y participantes incluidos y el alto riesgo de sesgo en el diseño del estudio. Se necesitan ensayos controlados aleatorios de poder estadístico apropiado y realizados con atención para investigar adecuadamente la función de las tecnologías de recubrimiento con fármacos en el tratamiento de la reestenosis intrastent.

PICOs

Population

Intervention

Comparison

Outcome

The PICO model is widely used and taught in evidence-based health care as a strategy for formulating questions and search strategies and for characterizing clinical studies or meta-analyses. PICO stands for four different potential components of a clinical question: Patient, Population or Problem; Intervention; Comparison; Outcome.

See more on using PICO in the Cochrane Handbook.

Resumen en términos sencillos

available in

Angioplastia con balón impregnado de fármacos versus angioplastia con balón no recubierto para la estenosis de los stents en las arterias del miembro inferior

Antecedentes
Muchos pacientes presentan enfermedades en las arterias de la pierna, que pueden obstruirse y llevar a la pérdida de la circulación y, en consecuencia, causar dolor, úlceras cutáneas y amputación de la pierna. Con la finalidad de impedir que la falta de sangre dañe la extremidad, se pueden realizar procedimientos que derivan a la arteria obstruida a través de una vena o un injerto artificial, o que introducen un alambre y abren la arteria con un balón y posteriormente colocan un stent para impedir que se repita la obstrucción. Aunque los stents son muy resistentes y pueden durar mucho tiempo, es posible que un stent colocado en el interior de una arteria pueda en ocasiones estrecharse y, con el tiempo, obstruirse. Este proceso se conoce como "reestenosis intrastent". La obstrucción del stent se puede tratar de varias maneras, como la extracción o el tratamiento con balón del coágulo, o mediante la colocación de otro stent dentro del que ha fracasado. Aunque todas las opciones terapéuticas tienen ventajas y desventajas, las tecnologías disponibles para tratar este problema han evolucionado . Uno de estos avances requiere el recubrimiento del balón para tratar la obstrucción del stent con un fármaco citotóxico utilizado en quimioterapia, lo que ralentiza el proceso de obstrucción después del tratamiento del stent. Con estos balones especialmente diseñados, que se conocen como "balones impregnados de fármacos", se han obtenido resultados alentadores en el tratamiento de los pacientes con arteriopatía de la pierna.

Pregunta de la revisión
El objetivo de esta revisión era comparar la angioplastia con balón impregnado con fármaco (BIF) con la angioplastia tradicional con balón no recubierto, también conocida como angioplastia con balón simple antigua (POBA, siglas en inglés), para el tratamiento de la reestenosis intrastent de los stents colocados en arterias de la pierna.

Características de los estudios
La revisión incluyó tres ensayos clínicos que asignaron al azar a 263 participantes (búsqueda más reciente ‐ 28 de noviembre de 2017). Los ensayos, realizados en Europa, analizaron las arterias de la pierna a la altura y por encima de la rodilla; en todos se utilizaron los BEF con el producto químico conocido como "paclitaxel". Dos empresas fabricaron los BEF: Eurocor y Medtronic. La mayoría de los participantes en los estudios fueron seguidos durante seis meses o más (actividad denominada "seguimiento").

Resultados clave
Los resultados mostraron que los BIF no fueron mejores para los pacientes que la angioplastia con balón no recubierto con respecto a la necesidad de amputación. A los 24 meses de seguimiento, los BIF se asociaron con menos revascularización de lesiones objetivo, que se refiere a la necesidad de realizar un procedimiento en un stent que ya había sido tratado con un BIF o una angioplastia con balón no recubierto para la reestenosis intrastent. También se encontró que fueron mejores las tasas de reestenosis binaria con los BIF, lo que se refiere al porcentaje de stents tratados que desarrollan una nueva estenosis después de haber sido tratados con un BIF o una angioplastia con balón no recubierto. Finalmente, más pacientes que fueron tratados con BIF describieron una mejoría de los síntomas de la pierna, según indican los cambios en la categoría de Rutherford. Se halló que los BIF no son más beneficiosos para los pacientes que la angioplastia con balón no recubierto con respecto a la mortalidad.

Certeza de la evidencia
La certeza de la evidencia fue muy baja porque sólo se identificaron tres estudios con un número reducido de participantes, y porque se perdieron muchos pacientes durante el seguimiento. Además, el riesgo de sesgo de rendimiento y de desgaste fue significativo, al igual que el riesgo de otros sesgos, debido a que no se representa el tipo de stent tratado y la necesidad de colocación de stent de rescate.

Authors' conclusions

Implications for practice

All three included trials were powered to detect anatomical rather than clinically relevant endpoints (FAIR; ISAR‐PEBIS; PACUBA). Our analysis showed an advantage for DEB over uncoated balloon angioplasty for treatment of in‐stent restenosis for up to two years in some domains. However, the certainty of the evidence is generally very low because of the small number of included studies and the risk of bias in those studies. As such, evidence is insufficient at present to show the superiority of DEB over uncoated balloon angioplasty for treatment of in‐stent restenosis. This has important implications for practice, given the increased costs associated with the use of DEBs, which are generally more expensive than uncoated balloon angioplasty catheters.

Implications for research

To properly evaluate the role of DEBs in treating patients with in‐stent restenosis, appropriately powered, independently funded clinical trials are needed to generate rigorous and reliable data that can be used in a future iteration of this review and that answer questions that are clinically relevant to patients. Questions that can be answered in future trials involve the characteristics of stents that would benefit most from DEB angioplasty for in‐stent restenosis and the effective dose of paclitaxel that would prevent further restenosis.

Summary of findings

Open in table viewer

Summary of findings for the main comparison. Drug‐eluting balloon angioplasty at 6 months compared to uncoated balloon angioplasty at 6 months for in‐stent restenosis of the femoropopliteal arteries

Drug‐eluting balloon angioplasty at 6 months compared to uncoated balloon angioplasty at 6 months for in‐stent restenosis of the femoropopliteal arteries
Patient or population: patients with in‐stent restenosis of the femoropopliteal arteries Setting: hospital Intervention: drug‐eluting balloon angioplasty at 6 months Comparison: uncoated balloon angioplasty at 6 months
Outcomes	*Anticipated absolute effects (95% CI)**		Relative effect (95% CI)	No. of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with uncoated balloon angioplasty	Risk with drug‐eluting balloon angioplasty
Amputation Follow‐up: 6 months	See comments	See comments	See comments	263 (3 RCTs)	⊕⊕⊝⊝ VERY LOW^a	None of the 3 included studies reported any major or minor amputations during follow‐up. Amputation‐free survival was not an outcome in the included studies
Target lesion revascularization (TLR) Follow‐up: 6 months	Study population		OR 0.05 (0.00 to 0.92)	70 (1 RCT)	⊕⊕⊝⊝ VERY LOW^a
	206 per 1000	13 per 1000 (0 to 193)
Binary restenosis Follow‐up: 6 months	Study population		OR 0.28 (0.14 to 0.56)	189 (2 RCTs)	⊕⊕⊝⊝ VERY LOW^a
	407 per 1000	161 per 1000 (88 to 277)
Death Follow‐up: 6 months	Study population		OR 0.93 (0.13 to 6.71)	263 (3 RCTs)	⊕⊕⊝⊝ VERY LOW^a
	8 per 1000	7 per 1000 (1 to 49)
Improvement ≥ 1 Rutherford category Follow‐up: 6 months	Study population		OR 1.81 (1.02 to 3.21)	193 (2 RCTs)	⊕⊕⊝⊝ VERY LOW^a
	427 per 1000	574 per 1000 (432 to 705)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; OR: odds ratio; RCT: randomized controlled trial.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded by three levels because of significant risk of performance and attrition bias, the small numbers of studies and participants, and other biases due to lack of accounting for the type of stent being treated and need for bailout stenting.

Background

Description of the condition

Peripheral arterial disease (PAD) of the lower extremities is closely associated with atherosclerosis, a progressive and chronic disease process that leads to hardened vessels and decreased blood flow (Stary 1994). Although many people with PAD of the lower extremities are asymptomatic, others present with a variety of symptoms, including pain with walking (intermittent claudication), pain at rest, and tissue loss in the affected limbs (McDermott 2001). PAD is a highly prevalent condition that increases with age and occurs in more than 10% of individuals above the age of 60 (Criqui 2015).

Numerous treatment modalities have been used for the management of PAD of the lower extremities. Medical therapies aim to arrest the progression of atherosclerotic disease and improve clinical outcomes by modifying patient risk factors (Ali 2012). Open surgical and endovascular interventions are used in conjunction with medical therapies to improve blood delivery to ischemic tissues by bypassing or otherwise opening arterial blockages with wires, catheters, and other specialized devices (Conte 2015).

Of the growing number of interventions available for treating patients with lower extremity PAD, deployment of stents across a stenotic or occluded arterial segment has become a routine practice in certain circumstances, such as with long lesions and those that do not respond well to balloon angioplasty alone (OHTAC 2010). Stenting provides the mechanical support required to maintain the patency of arterial lesions, but its effectiveness is limited by the occurrence of neointimal proliferation, known as "intimal hyperplasia," and by overgrowth of the inner lining of the artery within the stent. This process, known as "in‐stent restenosis," may limit blood flow through the stent and may sometimes result in recurrence of ischemic symptoms (Alfonso 2014).

Description of the intervention

Numerous endovascular interventions have been described for the treatment of in‐stent restenosis, including repeat uncoated balloon angioplasty, also known as "plain old balloon angioplasty (POBA)" alone, cutting balloon angioplasty, angioplasty with drug‐eluting balloons (DEBs), repeat stenting, and atherectomy (Singh 2014).

For all of these interventions, a wire is used first to cross the stenotic or occlusive lesion within the stent before treatment can be administered. Use of DEBs requires pre‐dilation of the stenotic lesion with a plain balloon. A second balloon is then used to coat the inside of the stent with a cytotoxic chemical in the hope of reducing neointimal proliferation.

How the intervention might work

Uncoated balloon angioplasty has been used widely and successfully for the treatment of lower extremity PAD and in‐stent restenosis, but its use is limited by recurrence of the neointimal hyperplastic process. DEBs address this limitation by applying a chemical to the stent that exerts an antiproliferative effect and arrests the restenotic process (Waksman 2009).

The most commonly used DEB cytotoxic agent is the taxane paclitaxel, a chemotherapeutic medication that exerts its cytotoxic effect by stabilizing microtubules (Marupudi 2007). Paclitaxel is a safe and effective DEB agent for treatment of PAD and has been used successfully to decrease the incidence of binary restenosis in femoropopliteal vessels (Canaud 2014).

Why it is important to do this review

Numerous prospective studies have compared DEB versus uncoated balloon angioplasty for treatment of in‐stent restenosis. DEBs have shown promising results in the treatment of coronary in‐stent restenosis (Indermuehle 2013). However, the certainty of evidence for their use in femoropopliteal vessels remains unclear. Furthermore, DEB technologies are more expensive and are less widely available than uncoated balloon angioplasty. Our review seeks to assess the efficacy of this intervention and to determine whether it is associated with improved clinical outcomes.

Objectives

To assess the safety and efficacy of DEBs compared with uncoated balloon angioplasty in people with in‐stent restenosis of the femoropopliteal arteries as assessed by criteria such as amputation‐free survival, vessel patency, target lesion revascularization, binary restenosis rate, and death. We define "in‐stent restenosis" as 50% or greater narrowing of a previously stented vessel by duplex ultrasound or angiography.

Methods

Criteria for considering studies for this review

Types of studies

We included randomized controlled trials (RCTs) that compared drug‐eluting balloons (DEBs) versus uncoated balloon angioplasty for in‐stent restenosis of the femoropopliteal arteries. We define "in‐stent restenosis" as 50% or greater narrowing of a previously stented vessel by duplex ultrasound or angiography.

Types of participants

We included people undergoing drug‐eluting or uncoated balloon angioplasty for treatment of in‐stent restenosis of the femoropopliteal arteries.

Types of interventions

We sought to compare DEBs versus uncoated balloon angioplasty for treatment of in‐stent restenosis of femoropopliteal arteries. Analysis excluded hybrid studies in which DEBs are utilized simultaneously with open surgery, as well as studies that include atherectomy procedures, cutting balloon angioplasty, vascular brachytherapy, and repeat stenting with plain or drug‐eluting stents for management of in‐stent restenosis.

Types of outcome measures

Primary outcomes

Amputation‐free survival, defined as the probability of being alive without loss of the affected lower extremity
Vessel patency, both primary and secondary, measured by late lumen loss, target lesion revascularization (TLR), and rate of binary restenosis, defined as 50% or greater narrowing of the vessel by duplex ultrasound or angiography
Death

Secondary outcomes

Change in Rutherford category of PAD
Change in the Society for Vascular Surgery Wound, Ischemia, and Foot Infection (WIfI) stage (Mills 2014)
Change in ankle‐brachial index (ABI)
Change in quality of life (QoL) scores, measured by validated instruments such as Short Form 36 (SF‐36)
Change in functional walking ability, measured by change in walking distance (on a treadmill or by other walking test) or use of a walking impairment questionnaire

Search methods for identification of studies

We applied no language, publication year, or publication status restrictions.

Electronic searches

The Cochrane Vascular Information Specialist searched the following databases for relevant trials.

Cochrane Vascular Specialised Register (February 2, 2017).
Cochrane Central Register of Controlled Trials (CENTRAL; 2017, Issue 1), in the Cochrane Library, via the Cochrane Register of Studies Online.

See Appendix 1 for details of the search strategy that were used to search CENTRAL.

The Information Specialist also searched the following trials registries for details of ongoing and unpublished studies (February 2, 2017).

ClinicalTrials.gov (clinicaltrials.gov).
World Health Organization International Clinical Trials Registry Platform (who.int/trialsearch).
International Standard Randomized Controlled Trials Number (ISRCTN) Register (isrctn.com/).

The Information Specialist subsequently conducted top‐up searches of the following databases.

Cochrane Vascular Specialised Register via the Cochrane Register of Studies (CRS‐Web searched from January 1, 2017, to November 28, 2017).
Cochrane Central Register of Controlled Trials (CENTRAL), in the Cochrane Library, via the Cochrane Register of Studies Online (CRSO; 2017, Issue 10).
MEDLINE (Ovid MEDLINE® Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, Ovid MEDLINE® Daily, and Ovid MEDLINE®) (searched from January 1, 2017, to November 28, 2017).
Embase Ovid (searched from January 1, 2017, to November 28, 2017).
Cumulative Index to Nursing and Allied Health Literature (CINAHL) Ebsco (searched from January 1, 2017, to November 28, 2017).
Allied and Complementary Medicine Database (AMED) Ovid (searched from January 1, 2017, to November 28, 2017).

The Information Specialist modeled search strategies for the listed databases on the search strategy designed for CENTRAL. When appropriate, the search strategies were combined with adaptations of the highly sensitive search strategy designed by Cochrane for identifying RCTs and controlled clinical trials (as described in Chapter 6, of the Cochrane Handbook for Systematic Reviews of Interventions; Higgins 2011). We have provided search strategies for the major databases in Appendix 3.

The Information Specialist also performed top‐up searches of the following trials registries on November 28, 2017.

World Health Organization International Clinical Trials Registry Platform (who.int/trialsearch).
ClinicalTrials.gov (clinicaltrials.gov).

Searching other resources

We further examined the bibliographies of studies identified in our search to identify other relevant articles.

Data collection and analysis

Selection of studies

Two review authors (AK, WA) independently selected trials for inclusion in this review from studies identified by the search. A third review author (AD) assessed these trials, determined their suitability, and adjudicated disagreements between the first two review authors. We have outlined in the Criteria for considering studies for this review section the inclusion criteria used to determine suitability.

Data extraction and management

Two review authors (AK, WA) extracted relevant data from the included studies. We collected participant demographics (age, gender, PAD comorbidities, Rutherford PAD category, and baseline ABI), interventions (types and duration of DEBs and uncoated balloon angioplasty), and outcomes (as specified in the Criteria for considering studies for this review section). A third review author (AD) cross‐checked data.

Statistical analysis complied with the standard methods of Cochrane Vascular. We used the computer software package Review Manager 5.3 (RevMan) to perform all statistical analyses and to generate figures (Review Manager 2014).

Assessment of risk of bias in included studies

Two review authors (AK, WA) carried out a thorough risk of bias assessment of all included studies using the Cochrane tool for assessing risk of bias (Higgins 2011). This tool assesses bias in six different domains, with each domain receiving a score of high, low, or unclear, depending on each review author’s judgement. A third review author (AD) adjudicated disagreements. We contacted study authors when we needed clarification to better assess risk of bias.

Measures of treatment effect

We used the mean difference (MD) to calculate and report continuous outcome measures, such as change in QoL scores. If included studies used different scales, we calculated the standardized mean difference (SMD) instead. We also calculated the associated 95% confidence interval (CI) between the two treatment groups. We calculated and reported dichotomous (binary) outcome measures, such as target lesion revascularization, using the hazard ratio (HR) or the risk ratio (RR) with associated 95% CI, depending on reported data. We carried out analyses at different time points, as reported by the trials. We based our calculations on an intention‐to‐treat approach.

Unit of analysis issues

The unit of analysis was the treated limb for anatomical outcomes such as late lumen loss or binary restenosis. Conversely, the unit of analysis was the patient for outcomes such as quality of life or death.

Dealing with missing data

We contacted study authors to inquire about missing or incomplete data. In the event that a significant quantity of data was missing from an included study and attempts to contact study authors were unsuccessful, we assessed the risk of bias associated with including the study in the meta‐analysis and performed a sensitivity analysis.

Assessment of heterogeneity

We assessed interstudy heterogeneity by using a forest plot (Schünemann 2011). We calculated Chi² and I² values to measure the amount of heterogeneity (Higgins 2003). I² values less than 50% indicated low heterogeneity, I² values between 50% and 75% indicated moderate heterogeneity, and I² values greater than 75% indicated significant heterogeneity (Deeks 2011).

Assessment of reporting biases

We planned to construct a funnel plot to assess publication bias when 10 or more studies were available for a particular outcome (Sterne 2011).

Data synthesis

To calculate the pooled treatment effect data, we used random‐effects or fixed‐effect models, depending on the degree of interstudy heterogeneity. If the calculated degree of interstudy heterogeneity was significant (defined as I² greater than 75%), we used the random‐effects model. Otherwise, we used the fixed‐effect model. We calculated 95% CIs for continuous and dichotomous outcome variables, as detailed above. We created a forest plot for each treatment effect as per Cochrane Vascular guidelines.

Subgroup analysis and investigation of heterogeneity

We planned to perform subgroup analyses by type of DEB, type and dose of pharmacological agent used in the DEB, and clinical severity of PAD, as defined by the change in Rutherford category or the WIfI stage. We also planned to perform subgroup analyses for people with diabetes and for the types of antiplatelet agents prescribed for participants in the included trials.

Sensitivity analysis

We planned to exclude studies at high risk of bias from the pooled analysis and to repeat the analysis to assess their impact. We planned to performed a sensitivity analysis by sequentially excluding studies at high risk of bias.

"Summary of findings"

We prepared a "Summary of findings" table using the GRADEpro Guideline Development Tool to present the main findings of the review for the time point at which the most relevant data were available from the included studies (www.gradepro.org;Atkins 2004). The population consisted of people with PAD, and we compared DEBs versus uncoated balloon angioplasty for treatment of in‐stent restenosis of femoropopliteal arteries. We included in our "Summary of findings" table the main outcomes listed under Types of outcome measures that we considered essential for decision‐making. We evaluated the certainty of the evidence using the GRADE approach (Schünemann 2011b). We assigned one of four levels of quality: high, moderate, low, or very low, based on overall risk of bias, directness of evidence, inconsistency of results, precision of estimates, and risk of publication bias, as previously described (Higgins 2011).

Results

Description of studies

Results of the search

Please see Figure 1 for search results.

Figure 1

Study flow diagram.

Included studies

This review includes three randomized controlled trials that compared DEB versus uncoated balloon angioplasty for treatment of symptomatic in‐stent restenosis of the femoropopliteal arteries (FAIR; ISAR‐PEBIS; PACUBA). The Characteristics of included studies tables present the details of those studies.

The three included studies were conducted in Germany ‐ FAIR; ISAR‐PEBIS ‐ and in Austria ‐ PACUBA. All studies included patients with in‐stent restenosis in the femoropopliteal arteries and used paclitaxel‐coated balloons, although at different dosages (Medtronic IN.PACT Admiral 3.5 mcg/mm² in FAIR and ISAR‐PEBIS vs Eurocor Freeway 3 mcg/mm² in PACUBA). We also identified two ongoing trials (Copa Cabana; TRC‐14004848). Details of the ongoing trials are presented in the Characteristics of ongoing studies tables.

We contacted the authors of all included and ongoing studies to request study data.

Excluded studies

We excluded seven studies from our review (Bosiers 2015; ConSeQuent; DEBATE‐ISR; EXCITE ISR; NCT00481780; NCT02832024; RELINE), and we listed reasons for exclusion in the Characteristics of excluded studies tables. We excluded studies because they compared stenting in Bosiers 2015 and RELINE, laser atherectomy in EXCITE ISR, or cutting balloon angioplasty in NCT00481780 versus uncoated balloon angioplasty for in‐stent restenosis. The ConSeQuent trial compared DEB versus uncoated balloon angioplasty for native vessel restenosis rather than in‐stent restenosis. The DEBATE‐ISR study compared a prospective cohort of patients receiving DEB therapy for in‐stent restenosis against a historical cohort of diabetic patients. Finally, the NCT02832024 study compared stent deployment versus atherectomy versus uncoated balloon angioplasty alone for in‐stent restenosis.

Risk of bias in included studies

Please refer to the "Risk of bias" tables in the Characteristics of included studies tables and summary results in Figure 2 and Figure 3. All three included studies were at a high risk of performance and attrition bias because of significant differences in bailout stenting rates and lack of clarity on the types of stents treated for in‐stent restenosis.

Figure 2

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

The included studies were at low risk of selection bias in random sequence generation or allocation concealment. The FAIR trial used a block randomization study design with a block size of 10 and an allocation ratio of 1:1, and concealed the allocation sequence from investigators by using sequentially numbered and opaque sealed envelopes. The ISAR‐PEBIS study used computer sequence generation to randomize participants after making the decision to proceed with an intervention. These researchers similarly accomplished the allocation of treatment by using sealed and opaque envelopes. The PACUBA trial investigators used a 1:1 randomization protocol with computer‐generated sequences and concealed allocation via use of sealed envelopes.

Blinding

All three included studies were at high risk of performance bias because the operators were not blinded (FAIR; ISAR‐PEBIS; PACUBA). Conversely, however, the outcomes assessors in two trials were blinded to the treatment arm (ISAR‐PEBIS; PACUBA), placing them at low risk for detection bias. It is unclear whether the outcomes assessors in the FAIR trial were blinded to the treatment arm.

Incomplete outcome data

All three included studies were at high risk of attrition bias due to loss to follow‐up (FAIR; ISAR‐PEBIS; PACUBA). In the FAIR trial, approximately one‐third of participants were lost to follow‐up at 12 months. Similarly, in the ISAR‐PEBIS trial, approximately 25% of DEB participants and 21% of uncoated balloon angioplasty participants did not undergo angiography as per the trial's primary endpoint requirement. In the PACUBA trial, eight participants in the DEB group (23%) and six in the uncoated balloon angioplasty group (15%) were lost to follow‐up.

Selective reporting

All three included studies were at low risk for reporting bias (FAIR; ISAR‐PEBIS; PACUBA). Trial authors reported all outcomes that were specified in the study protocols.

Other potential sources of bias

Two major sources of bias in all included trials were specifying the types of stents treated for in‐stent restenosis and addressing the differences in bailout stenting requirements between treatment arms (FAIR; ISAR‐PEBIS; PACUBA).

Study authors did not specify whether stents that were treated were bare‐metal, covered, or drug‐eluting, nor did they stratify the results by length of the stent or duration of time since stent implantation.

The other major source of bias is the disparity in the incidence of bailout stenting between treatment arms in all three included studies. In the FAIR study, four uncoated balloon angioplasty participants received bailout stenting compared with one participant in the DEB arm. Conversely, in the ISAR‐PEBIS study, 26% of DEB participants received bailout stenting compared with 6% of uncoated balloon angioplasty participants (P = 0.02). In the PACUBA study, four DEB participants received bailout stenting compared with one uncoated balloon angioplasty participant. Authors of the included studies did not stratify reported results by bailout stent implantation status.

Furthermore, study populations in the PACUBA trial were not similar, as the uncoated balloon angioplasty group reported more advanced lesions (Tosaka II and TASC B and C) than were noted in the DEB group (67% Tosaka class II lesions in the uncoated balloon angioplasty group vs 46% in the DEB group, 36% TASC B and 26% TASC C lesions in the uncoated balloon angioplasty group vs 23% and 14% in the uncoated balloon angioplasty group, respectively).