ftDepdatsacc

etlcMivtmlOtiinptatHmtbcprlrtbdiCvc(npccibtpdcliacorrP

Journal of the American College of Cardiology Vol. 60, No. 15, 2012© 2012 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2012.05.043

EDITORIAL COMMENT

Post-Dilation ofSuperficial FemoralArtery Stents With CryoplastyDoes the COBRA Trial Take aBite Out of Diabetic Restenosis?*

Anand Prasad, MD

San Antonio, Texas

Over the past decade there has been an explosion in thenumber of tools available for the endovascular treatment oflower extremity peripheral arterial disease (PAD). Seem-ingly every few months, the interventionalist is confrontedwith a new device that promises to deliver both immediateangiographic improvement and long-term freedom fromrestenosis. Despite the increasing availability of newerangioplasty balloons, nitinol stents, drug-coated devices,and atherectomy cutters, the ideal treatment for an individ-ual patient is unclear and often not based on rigorouslyvetted data. In contrast to the coronary circulation, thisclinical uncertainty is largely the result of a paucity ofrandomized controlled trials to evaluate the utility of thesedevices.

See page 1352

In this issue of the Journal, Banerjee et al. (1) present datarom a prospective multicenter randomized controlled pilotrial (COBRA [Cryoplasty or Conventional Balloon Post-ilation of Nitinol Stents for Revascularization of Periph-

ral Vascular Segments]) to examine the benefit of cryo-lasty in diabetic patients using the PolarCath peripheralilatation system (Boston Scientific, Natick, Massachusetts)s an adjunctive treatment after nitinol stent placement inhe superficial femoral artery (SFA). Cryoplasty using thisystem has been approved since 2004. The widespreaddoption of this technology has been limited by mixedlinical data and a lack of fundamental understanding ofryobiology in relation to PAD. In comparison to drug-

*Editorials published in the Journal of the American College of Cardiology reflect theviews of the authors and do not necessarily represent the views of JACC or theAmerican College of Cardiology.

From the Department of Medicine, Division of Cardiology, University of Texas

wHealth Science Center at San Antonio, San Antonio, Texas. Dr. Prasad has reportedthat he has no relationships relevant to the contents of this paper to disclose.

luting balloons, for example, where the biology of pacli-axel has been studied extensively, the mechanisms under-ying vascular responses to cryoplasty have lagged behind thelinical implementation of this tool.

echanisms of action. Cryoplasty (using the PolarCath)s hypothesized to have 2 pathways of action on peripheralessels. First, there is the standard effect of plaque disrup-ion and vessel dilation using low pressure inflation. Thisicroprocessor-controlled dilation is suggested to induce

ess mechanical barotrauma than conventional angioplasty.f course, while a “low and slow” inflation may prevent

rauma and dissection, it may also limit acute gain and resultn restenosis or reocclusion. Therefore, the second and morenteresting aspect of this device is the delivery of liquiditrous oxide into the balloon system. As the liquid changeshase to a gas, the balloon inflates and the associated heatransfer results in supercooling of the balloon to �10°C. Aslready mentioned, the physiological impact of cooling onhe vascular architecture remains an active area of study.

owever, several mechanisms have consistently emerged inultiple studies—mostly performed using animal-derived

issue and cell preparations (2–5). The pathway of mostiological relevance is apoptosis of vascular smooth muscleells with relative sparing of the endothelial cells. Thisrocess is hypothesized to limit the inflammatory andeactive responses to injury that often follow angioplasty,eading to restenosis. Some data would suggest that both theate and magnitude of cooling directly impact the depth ofreatment within the vessel wall as well as the relativealance between necrosis and apoptosis (5). However, theuration and magnitude of smooth muscle cell suppressionn humans undergoing cryoplasty remain poorly defined.

linical data for use of cryoplasty in humans. Thearious studies that have examined the clinical utility ofryoplasty for PAD in humans are highlighted in Table 16–22). The majority of these studies have consisted ofonrandomized single-center experiences. Accordingly, theatient population, methods, and outcomes have variedonsiderably. In most studies, the acute angiographic suc-ess rates have been excellent; however the long-term ratesn regard to patency or target lesion revascularization haveeen inconsistent—comparable (or worse in some studies)han conventional percutaneous transluminal balloon angio-lasty (PTA). The handful of small randomized trials toate have also not demonstrated an overwhelming benefit ofryoplasty over PTA. In all of these trials, cryoplasty wasooked to as a primary endovascular treatment (with stent-ng reserved for bail-out). Therefore, underlining these trialsre basic mechanical limitations of PTA that are shared byryoplasty, namely, vascular recoil and dissection. Placementf nitinol stents solves the problems associated with acuteecoil and dissection, but continues to invite the bane ofestenosis. Accordingly, the benefits of routine stenting overTA alone have been decidedly mixed in the literature. It is

ith these limitations in mind that the COBRA trial makes

emg

tdrdccRdi(tctcrncwvcdgiidtipt

dp

pPCcifmcLrssabppmoptcmcbtafpfimrsared

hcaPsl

1361JACC Vol. 60, No. 15, 2012 PrasadOctober 9, 2012:1360–3 The COBRA Trial and Diabetic Restenosis

sense: use a stent to counter the acute mechanical limita-tions of PTA, and use cryoplasty to address the smoothmuscle mediated restenosis process.What did we learn from the COBRA trial? Banerjee et al.(1) used the aforementioned framework to perform a pilotstudy of SFA nitinol stenting followed by post-dilation withcryoplasty versus conventional balloon PTA. Forty-fivepatients were randomly allocated to each group, andnotably, despite the small sample size, the 2 groups werefairly well matched in terms of comorbid conditions. Theinvestigators used binary restenosis as their primaryendpoint, defined by a �2.5-fold increase in peak systolicvelocity within the stented segment. Patients in thecryoplasty arm had PolarCath treatment once the nitinolstent was in place as the final post-dilation treatment orafter conventional post-dilation. The entire length of thestent was treated with cryotherapy. The key finding of thestudy was a significant difference in the 12-month binaryrestenosis rates (55.8% in the PTA arm and 29.3% in thecryoplasty arm, p � 0.01). Furthermore, as a secondaryndpoint, there was a sustained improvement at 12onths in the ankle-brachial index in the cryoplasty

roup that was not seen in the conventional PTA arm.Given the unique post-dilation cryotherapy strategy of

he COBRA trial, it is difficult to directly compare theseata with previous studies in the literature. However, 2andomized studies have examined cryoplasty specifically iniabetics. Spiliopoulos et al. (6) performed a small single-enter randomized trial of cryoplasty (n � 24) versusonventional PTA (n � 26) involving diabetic patients withutherford stage 3 to 6 symptoms for femoropoplitealisease. In this particular study—as in all the prior random-

zed cryoplasty trials—stenting was used only for bail-out26% usage in cryoplasty arm). A particular strength of thisrial was the use of invasive follow-up and an angiographi-ally determined binary restenosis definition. Immediateechnical success rates were relatively low: 58.0% in theryoplasty arm and 64.0% in the PTA arm (p � 0.29). Theesearchers found that in-lesion angiographic binary reste-osis was significantly higher with cryoplasty at follow-up asompared with PTA. On multivariable analyses, cryoplastyas associated with increased TLR events and decreasedessel patency. Fossaceca et al. (21) also compared PTA toryoplasty in 48 diabetic patients with femoropoplitealisease in a prospective randomized trial. Echoing theeneral findings of the study by Spiliopoulos et al. (6), thenvestigators found that cryoplasty resulted in both anntermediate (6 months) and a longer term (12 months)ecrease in vessel patency as well as a lower rate of initialechnical success. Furthermore, they noted more pain dur-ng inflation with cryoplasty as compared with PTA—ahenomenon that is not universally commented on in otherrials.

Given the small sample sizes, it is difficult to drawefinitive conclusions, but these data involving diabetic

atients would suggest that, at a minimum, cryoplasty as a

rimary strategy is not a better treatment than traditionalTA. The results, however, do add to the rationale for theOBRA trial. Stenting often provides a better acute pro-

edural result than PTA alone, and can improve themmediate technical success in complex calcified and dif-usely diseased vessels as seen in diabetic patients. Withechanical scaffolding in place, cryoplasty can then be

onsidered as an adjuvant therapy to deter restenosis.imitations of the COBRA trial and future strategies for

estenosis prevention. The findings of the COBRA trialhould be taken in the light of a preliminary study. As amall trial, there were several notable limitations. First, theuthors chose to study only diabetic patients. Although theurden, severity, and poor outcomes of PAD in diabeticatients are well documented, inclusion of a more variedatient population would have allowed for a larger study andore generalizability to the general PAD population. Sec-

nd, as an inherent limitation of a small pilot study, it is notossible to know if the strategy of post-dilation employed inhe study will lead to any true difference in clinical out-omes. Interestingly, both groups had an improvement inean walking impairment scores even though the signifi-

ant improvement in binary restenosis and the ankle-rachial index was confined to the cryoplasty arm. Third,he lack of angiographic follow-up in all patients limits thenatomical analyses in regard to stent fracture, collateralormation, and progression of plaque. Last, we cannotredict where the strategy outlined in the COBRA trial willt in with the current and upcoming endovascular treat-ents for PAD. With the concerns of stent fracture and

estenosis, many interventionalists have migrated away fromtenting. In this regard, the role of primary debulking beforedditional treatments will have to be evaluated, as will theole of biodegradable technologies. In addition, the relativefficacy of cryoplasty versus drug-eluting balloons for post-ilation of nitinol stents will need to be studied.Widespread adoption of this technology will also be

elped by a better understanding of the vascular biology ofryotherapy. What is clear is the need for further fundingnd interest in performing randomized controlled trials forAD revascularization. In this regard, Banerjee et al. (1)hould be commended for their novel approach to a chal-enging problem.

Reprint requests and correspondence: Dr. Anand Prasad, De-partment of Medicine, Division of Cardiology, University of TexasHealth Science Center at San Antonio, 7703 Floyd Curl Drive,San Antonio, Texas 78229. E-mail: anandprasadmd@gmail.com.

REFERENCES

1. Banerjee S, Das TS, Abu-Fadel MS, et al. Pilot trial of cryoplasty orconventional balloon post-dilation of nitinol stents for revasculariza-tion of peripheral arterial segments (COBRA trial). J Am Coll Cardiol2012;60:1352–9.

2. Basco MT, Yiu WK, Cheng SW, Sumpio BE. The effects of freezingversus supercooling on vascular cells: implications for balloon cryo-

plasty. J Vasc Interv Radiol 2010;21:910–5.

1362 Prasad JACC Vol. 60, No. 15, 2012The COBRA Trial and Diabetic Restenosis October 9, 2012:1360–3

Selected Cryoplasty Data for Lower Extremity Peripheral Arterial DiseaseTable 1 Selected Cryoplasty Data for Lower Extremity Peripheral Arterial Disease

First Author(Ref. #) Year

Treatment,Patients (n) Type of Study Key Findings/Comments

Spiliopoulos (6) 2010 Cryoplasty, 24 PTA, 26 Single-center randomized controlled trial(PTA vs. cryoplasty) in diabetic patients

Immediate technical success similar in bothgroups (58% to 64%); bail-out stenting ratewas 25.8% in cryoplasty group; cryoplastyassociated with increased TLR events,requiring repeat interventions; cryoplastyassociated with lower primary patency

Laird (7) 2005 Cryoplasty, 102 Prospective multicenter registry of cryoplasty in SFAand popliteal arteries

Technical success 85.3%; dissection rate6.9%; bail-out stent rate 8.8%; 9-monthprimary patency 70.1%; 9-month clinicalpatency 82.2%

Laird (8) 2006 Cryoplasty, 70 Extended follow-up of preceding (Laird [7]) registrypatients

At 3.4 years of follow-up, clinical patencyrate 75%; authors concluded cryoplasty isdurable, with low incidence of morbidity

Samson (9) 2007 Cryoplasty, 32 Single-center experience of cryoplasty for SFA andpopliteal disease

Technical success rate 96%; stent rate 8.5%;freedom from restenosis at 12 months82.2%; authors commented that low rateof spiral dissections may be a particularbenefit of cryoplasty

Bakken (10) 2007 Cryoplasty, 39PTA, 85

Retrospective analysis of cryoplasty and PTA forcomplex SFA disease in patients with critical limbischemia

Technical success rates were similar in bothgroups; 1-year primary and secondarypatency were the same; fewer stents wererequired in cryoplasty group

Das (11,12) 2007 Cryoplasty, 108 Prospective multicenter nonrandomized experience ofbelow-the-knee cryoplasty for critical limb ischemia

Overall success rate (inline flow to foot with�50% stenosis) was 97.3%; clinicallysignificant dissection was low (0.9%);6-month freedom from amputation93.4%; 6-month amputation free survival89.3%; 12-month freedom fromamputation 85.2%

Samson (13) 2008 Cryoplasty, 64 Single-center experience of cryoplasty for SFA andpopliteal disease

Immediate success rate with cryoplasty 88%;bail-out stenting after cryoplasty 9.8%; authorsnoted poor response to cryoplasty in heavilycalcified lesions, vein grafts, and ISR; lowfreedom from restenosis rates at 12 months(47%) and 24 months (38%); authors suggestpoor clinical findings coupled with cryoplastycost are discouraging, despite findings of theirearlier study

Korteweg (14) 2009 Cryoplasty, 32 Single-center experience of cryoplasty forfemoropopliteal disease

No bail-out stenting with immediate technicalsuccess in all patients; 12-month patencyrates 64% for TASC A lesions, 31% forTASC B lesions, 33% for TASC C lesions;authors concluded cryoplasty is safe andprovides patency results comparable to PTA

Banerjee (15) 2009 Cryoplasty, 27 Two-center experience of cryoplasty for SFA disease Immediate technical success in all patients;72% underwent cryoplasty; 28%underwent cryoplasty plus stenting; at1-year follow-up, primary patency was75% in cryoplasty-only group and 46% inadjunctive stenting group

Schmieder (16) 2010 Cryoplasty, 71 Retrospective analysis from single center of cryoplastythat included iliac, femoropopliteal, andinfrapopliteal lesions

Initial cryoplasty technical success rate 68%for native vessels and 71% for in-stentprocedures; 1-year primary patency rateswere low at 17% for native lesions and28% for ISR lesions; freedom fromclaudication at 1 year was also low, at21% for ISR and 32% for native lesions;authors concluded that cryoplasty offerslittle clinical benefit

Bosiers (22) 2010 Cryoplasty, 100 Multicenter registry of cryoplasty forinfrapopliteal lesions

Immediate technical success rate 95%; bail-out stent rate 17%; 12- month primarypatency rate 56%; 12-month limb salvagerate 94%; overall data were encouragingfor below-the-knee treatment, particularlyfor longer lesions

Contiued on next page

al arter

1363JACC Vol. 60, No. 15, 2012 PrasadOctober 9, 2012:1360–3 The COBRA Trial and Diabetic Restenosis

3. Grassl ED, Bischof JC. In vitro model systems for evaluation ofsmooth muscle cell response to cryoplasty. Cryobiology 2005;50:162–73.

4. Spiliopoulos S, Diamantopoulos A, Katsanos K, Ravazoula P, Karn-abatidis D, Siablis D. PolarCath cryoplasty enhances smooth musclecell apoptosis in a rabbit iliac artery model: an experimental in vivocontrolled study. Cryobiology 2011;63:267–72.

5. Balasubramanian SK, Venkatasubramanian RT, Menon A, BischofJC. Thermal injury prediction during cryoplasty through in vitrocharacterization of smooth muscle cell biophysics and viability. AnnBiomed Eng 2008;36:86–101.

6. Spiliopoulos S, Katsanos K, Karnabatidis D, et al. Cryoplasty versusconventional balloon angioplasty of the femoropopliteal artery indiabetic patients: long-term results from a prospective randomizedsingle-center controlled trial. Cardiovasc Intervent Radiol 2010;33:929–38.

7. Laird J, Jaff MR, Biamino G, et al. Cryoplasty for the treatment offemoropopliteal arterial disease: results of a prospective, multicenterregistry. J Vasc Interv Radiol 2005;16:1067–73.

8. Laird JR, Biamino G, McNamara T, et al. Cryoplasty for thetreatment of femoropopliteal arterial disease: extended follow-upresults. J Endovasc Ther 2006;13 Suppl 2:52–9.

9. Samson RH, Showalter DP, Lepore MR Jr., Ames S. Cryoplastytherapy of the superficial femoral and popliteal arteries: a single centerexperience. Vasc Endovasc Surg 2006;40:446–50.

10. Bakken AM, Saad WE, Davies MG. Cryoballoon angioplasty broad-ens the role of primary angioplasty and reduces adjuvant stenting incomplex superficial femoral artery lesions. J Am Coll Surg 2008;206:524–32.

11. Das TS, McNamara T, Gray B, et al. Primary cryoplasty therapyprovides durable support for limb salvage in critical limb ischemiapatients with infrapopliteal lesions: 12-month follow-up results fromthe BTK Chill trial. J Endovasc Ther 2009;16:II19–30.

12. Das T, McNamara T, Gray B, et al. Cryoplasty therapy for limbsalvage in patients with critical limb ischemia. J Endovasc Ther2007;14:753–62.

13. Samson RH, Showalter DP, Lepore M Jr., Nair DG, Merigliano K.Cryoplasty therapy of the superficial femoral and popliteal arteries: a

ContinuedTable 1 Continued

First Author(Ref. #) Year Treatment, Patients (n)

Jahnke (17) 2010 Cryoplasty, 40PTA, 46

Single-center rand(PTA vs. cryopl

Gonzalo (18) 2010 Cryoplasty, 11 Single-center expfemoropoplitea

Silva (19) 2011 Cryoplasty, 99 Retrospective sinthat included iinfrapopliteal l

Diaz (20) 2011 Cryoplasty, 86PTA, 69

Single-center randcryoplasty)

Fossaceca (21) 2011 Cryoplasty, 24PTA, 24

Single-center randdiabetic patien

ISR � in-stent restenosis; PTA � percutaneous transluminal angioplasty; SFA � superficial femor

reappraisal after 44 months’ experience. J Vasc Surg 2008;48:634–7.

14. Korteweg MA, van Gils M, Hoedt MT, et al. Cryoplasty for occlusivedisease of the femoropopliteal arteries: 1-year follow-up. CardiovascIntervent Radiol 2009;32:221–5.

15. Banerjee S, Brilakis ES, Das TS, Lichtenwalter CS. Treatment ofcomplex superficial femoral artery lesions with PolarCath cryoplasty.Am J Cardiol 2009;104:447–9.

16. Schmieder GC, Carroll M, Panneton JM. Poor outcomes withcryoplasty for lower extremity arterial occlusive disease. J Vasc Surg2010;52:362–8.

17. Jahnke T, Mueller-Huelsbeck S, Charalambous N, et al. Prospective,randomized single-center trial to compare cryoplasty versus conven-tional angioplasty in the popliteal artery: midterm results of theCOLD study. J Vasc Interv Radiol 2010;21:186–94.

18. Gonzalo B, Solanich T, Bellmunt S, et al. Cryoplasty as endovasculartreatment in the femoropopliteal region: hemodynamic results andfollow-up at one year. Ann Vasc Surg 2010;24:680–5.

19. Silva GV, Fernandes MR, Cardoso CO, et al. Cryoplasty for periph-eral artery disease in an unselected patient population in a tertiarycenter. Tex Heart Inst J 2011;38:122–6.

20. Diaz ML, Urtasun F, Barberena J, Aranzadi C, Guillen-Grima F,Bilbao JI. Cryoplasty versus conventional angioplasty in femoropopli-teal arterial recanalization: 3-year analysis of reintervention-free sur-vival by treatment received. Cardiovasc Intervent Radiol 2011;34:911–7.

21. Fossaceca R, Guzzardi G, Di Terlizzi M, et al. Comparison ofcryoplasty and conventional angioplasty for treating stenotic-occlusivelesions of the femoropopliteal arteries in diabetic patients: immediate,mid-term and long-term results. Radiol Med 2012 Feb 10 [E-pubahead of print].

22. Bosiers M, Deloose K, Vermassen F, et al. The use of the cryoplastytechnique in the treatment of infrapopliteal lesions for critical limbischemia patients in a routine hospital setting: one-year outcome of theCryoplasty CLIMB Registry. J Cardiovasc Surg (Torino) 2010;51:193–202.

Key Words: cryoplasty y diabetes mellitus y peripheral arterial disease

e of Study Key Findings/Comments

d controlled trialor popliteal stenoses

Initial technical success was 35% forcryoplasty versus 54% for PTA (p � 0.02);bail-out stent placement was 30% forcryoplasty versus 39% for PTA (p � NS);no significant difference in target lesionpatency at 9 months between the2 groups

of cryoplasty forns

Immediate technical success in all patients;primary patency was poor (55%) at12 months

ter experience of cryoplastymoropopliteal, and

Immediate success rate of 99%; dissectionrate of 12.2% (no bail-out stenting used);frequent need for reintervention, 42% ofpatients at 6 months and 53% at 2 years

d controlled trial (PTA vs. Overall complications similar in both arms;rate of dissection 13.5% in cryoplasty armwith 22% rate of stent placement;cryoplasty group had patency rates similarto (but numerically lower) PTA at 6, 12,24, and 36 months

d trial (PTA vs. cryoplasty) infemoropopliteal disease

Conventional PTA superior to cryoplasty interms of technical success rates andvessel restenosis at 6 months and12 months

y; TASC � � TransAtlantic InterSociety Consensus; TLR � target lesion revascularization.

Typ

omizeasty) f

eriencel lesio

gle-cenliac, feesions

omize

omizets with

y superficial femoral artery.