Late Longitudinal Comparison of Endovascular and Open Popliteal Aneurysm Repairs

Accepted Manuscript Late Longitudinal Comparison of Endovascular and Open Popliteal Aneurysm Repairs Mathew Wooster, MD, Murray Shames, MD, Martin Back, MD PII:

S0890-5096(15)00670-6

DOI:

10.1016/j.avsg.2015.07.012

Reference:

AVSG 2533

To appear in:

Annals of Vascular Surgery

Received Date: 12 April 2015 Revised Date:

12 July 2015

Accepted Date: 15 July 2015

Please cite this article as: Wooster M, Shames M, Back M, Late Longitudinal Comparison of Endovascular and Open Popliteal Aneurysm Repairs, Annals of Vascular Surgery (2015), doi: 10.1016/ j.avsg.2015.07.012. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

Late Longitudinal Comparison of Endovascular and Open Popliteal Aneurysm

2

Repairs

3

Mathew Wooster MD, Murray Shames MD, Martin Back MD

SC

RI PT

USF Health Morsani College of Medicine, Division of Vascular and Endovascular Surgery. USF Health South Building, 7th floor 2a Columbia drive Tampa, FL 33606 [email protected] [email protected] [email protected]

EP

TE D

Mathew Wooster 843-384-1460 [email protected] [email protected] USF Health South Building, 7th floor 2a Columbia Dr Tampa, FL 33606

M AN U

Corresponding Author:

AC C

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

1

ACCEPTED MANUSCRIPT

Abstract

24

Objective: We sought to define suitable anatomy predicting durable exclusion of popliteal artery

25

aneurysms (PAA) and define optimal patient selection criteria for endovascular repair.

26

Methods: Seventy-five PAA were repaired in 66 patients (64 male, 2 female) over the past 13 years.

27

Fifty-two aneurysms (69%) were treated with open surgical exclusion/bypass using autologous vein

28

(69%) or PTFE (31%) conduit. Extended bypass targets required inflow from the common femoral

29

artery in 15% of limbs and outflow via a tibial artery in 31%. Since May 2001, endovascular repair

30

was considered in patients with high medical risk, limited vessel tortuosity, absence of significant

31

occlusive disease (ABI > 0.9), and PAA not involving below knee segments. Interventions were

32

performed via antegrade femoral access in 23 limbs (31%) using commercially available endografts.

33

Device diameters ranged between 7-13mm, with a median of 2 devices per PAA, and mean treatment

34

length was 22 cm (range 5-36 cm). All patients were followed with duplex ultrasound surveillance

35

and were prescribed clopidogrel and/or aspirin.

36

Results: Patients treated endovascularly were older (82 v. 70 yo, P=.01), but had shorter LOS (2 v. 12

37

days, P=.01) and lower complication rates (8% v. 17%, P=.02). Mean surveillance interval was 39

38

months with similar 4-year survival (67.9% open, 73.7% endo). Primary and secondary patencies

39

were 67.2%, 67.2% after endovascular repair and 65.5%, 78.4% for open at 4 yrs respectively. Four

40

of 6 endovascular failures were thrombosis within 4 mo of intervention and had conversions to open

41

repair. Secondary interventions were required after 48.1% of endovascular and 54.1% of open

42

repairs. Three limbs were lost in the series (2 open, 1 endo).

43

Conclusion: Similar outcomes can be expected after endovascular and open PAA repair with

44

adherence to specific anatomic and technical selection requisites.

46 47

SC

M AN U

TE D

EP

AC C

45

RI PT

23

Introduction

With an incidence of 7.39/100,000 people, popliteal artery aneurysms

48

account for 80% of all peripheral aneurysms1. Commonly accepted indications for

49

repair include size >/= 2 cm or presence of symptoms (most commonly acute

2

ACCEPTED MANUSCRIPT

thrombosis or embolism with limb ischemia, or pain/swelling associated with

51

extrinsic popliteal vein compression). Traditional repair has been open surgical

52

excision of the aneurysm with interposition bypass or proximal and distal ligation of

53

the aneurysmal segment with bypass of the excluded segment. More recently,

54

however, endovascular management has taken a larger role with multiple series

55

now published citing high technical success rates with relatively low complications

56

and primary/secondary patency greater than 70% as far out as 5 years2-6.

SC

RI PT

50

57

Recent review of the past quarter century of literature suggested very

59

appreciable overall results for endovascular repair with primary patency of 74% at

60

one year and secondary patency of 85% at three years compared to 87% and 81%

61

for open repair respectively7. Unfortunately, the literature is incomplete with a

62

wide range of endpoints used for comparison, leaving considerable debate as to the

63

preferred management of PAAs. To this end, we seek to present our experience

64

with open and endovascular repair of PAAs, the largest comparative cohort

65

presented in the literature to date.

EP

TE D

M AN U

58

67 68

AC C

66 Methods

All patients treated for PAA repair from 1999 to 2013 by a single surgeon in

69

two hospitals (one Veteran Affairs hospital and one quaternary care academic

70

hospital) were retrospectively reviewed and all were included for whom operative

71

records were available. Aneurysms were treated electively when >2cm in size and

72

non-electively in the setting of acute limb ischemia. Patients were offered

3

ACCEPTED MANUSCRIPT

endovascular repair beginning in 2001 if they were deemed high risk for surgery,

74

had minimal occlusive (ABI >/= 0.9, >1 vessel tibial runoff), and aneurysm

75

confinement to the above knee popliteal segment. Patients were non-randomized to

76

open (OR) and endovascular repair (ER) at surgeon discretion, using the

77

aforementioned endovascular consideration criteria.

78

RI PT

73

Endovascular repair was performed using ipsilateral open femoral exposure with antegrade access via 9-12French sheaths. Wallgrafts (Boston Scientific) were

80

utilized from 2001-2003 until the Viabahn stentgraft (Gore) became available at our

81

institution, at which time it became the preferred device. Post operatively patients

82

were treated with dual anti-platelet therapy (aspirin/Plavix) for 30 days, followed

83

by lifetime aspirin therapy (unless already on alternative antiplatelet or

84

anticoagulation for comorbidities).

M AN U

Surgical repair was performed using either a posterior approach for

TE D

85

SC

79

aneurysmectomy with interposition graft or a medial approach for bypass with

87

aneurysm exclusion at the surgeon’s discretion. In situ greater saphenous vein was

88

used when available with reversed GSV or PTFE being used when in situ NRGSV was

89

not technically appropriate. Arm vein was not used in this cohort. Post operatively

90

patients were treated with a minimum of aspirin, and Plavix or Coumadin were

91

added based on comorbidities and duplex criteria8.

93 94 95

AC C

92

EP

86

Results Seventy-seven PAAs were repaired in our series (25 endovascular, 52 open repair) with a mean follow up greater than three years. The endovascular cohort

4

ACCEPTED MANUSCRIPT

96

was significantly older than the open repair group and experienced a significantly

97

shorter length of stay, but the cohorts were otherwise similar with respect to sex,

98

follow up, limb loss, and complication and reintervention rates (Table 1). The majority of endovascular repairs were performed with Viabahn

RI PT

99

endoprosthesis (80%), which were exclusively used since available at our

101

institution in 2003. A median of 2 devices (range 1-3) was used to cover a mean

102

length of 22.3cm (Table 2).

103

SC

100

Open repair was primarily indicated for patients with occlusive disease (60%) requiring tibial bypass targets for 19% of patients (Table 3). Greater

105

saphenous vein was the preferred conduit, however greater than 30% of patients

106

were treated with PTFE due to lack of adequate saphenous vein. Similar proportion

107

of bypasses were to tibial targets for PTFE (4/16) and vein (6/36). Patient survival at four years neared 70% for both groups, with the greatest

TE D

108

M AN U

104

mortality during the third year of follow up (Figure 1a). Endovascular repair had

110

worse primary patency of nearly 70% at 1 year, but this difference was largely lost

111

by four years of follow up with each reaching less than 70% primary patency

112

(Figure 1b). Primary assisted and secondary patency were both better for the open

113

repair group as only one endoprosthesis achieved secondary patency with lysis

114

whereas multiple interventions for maintenance and restoration of patency were

115

performed for the open repair group (Figure 1c-d).

116

AC C

EP

109

Only half of each cohort remained free from reintervention by four years

117

(Figure 2a). Endovascular cohort reinterventions were performed for extension of

118

aneurysmal disease, endoleak, and occlusion (one successfully lysed, five converted

5

ACCEPTED MANUSCRIPT

to open bypass). The open repair group saw an early reintervention requirement

120

for anastamotic stenoses and pseudoaneurysms as well as for infection. Late

121

interventions were required for bypass stenosis, thrombosis, and extension of

122

disease. Interventions included jump bypasses, angioplasty, stenting, open

123

thrombectomy, and catheter-directed lysis (Figure 2b).

124

Endovascular repair of PAAs failed to match open repair patency rates in our cohort. Interestingly, however, the mortality of the two groups was nearly

128

equivalent out to four years despite the significantly greater age and perceived

129

preoperative risk of the endovascular group. Further, the need for open

130

reintervention following loss of endovascular patency did not seem to increase

131

mortality, suggesting that despite its higher rate of failure endovascular repair

132

remains a worthwhile undertaking in high risk patients, a proposal supported by the

133

larger scale Basil trials9. Alternatively, this also suggests that the endovascular

134

group was perhaps not as high risk as subjectively felt, and thus would tolerate

135

initial open intervention.

136

EP

TE D

M AN U

127

AC C

126

Discussion

SC

125

RI PT

119

It becomes challenging to compare the similarly high reintervention rates

137

between the two groups given that the endovascular group largely was not offered

138

attempts at secondary patency, instead deferring to open conversion in most

139

instances. Early reintervention for the endovascular group was primarily for

140

occlusion with one stent graft undergoing successful lysis and the remaining failures

141

being converted to open repair. Early reintervention for the open group was

6

ACCEPTED MANUSCRIPT

primarily for infection or anastomotic complications (pseudoaneurysm or stenosis).

143

Midterm re-interventions for endovascular repair were performed for stenosis (1)

144

and extension of disease (3) as opposed to the open group, which required

145

reintervention primarily for thrombosis in the PTFE conduits and stenosis in GSV

146

conduits.

147

RI PT

142

The high early failure rate for endovascular repair forces a closer inspection of the root cause. Reviewing our failures, we have identified several underlying

149

problems. Early device design led to failure when the stent graft failed to fully

150

expand despite angioplasty, leaving a longitudinal fold in the graft predisposing to

151

turbulent flow and early occlusion. Next, anatomy requiring long segment, below

152

knee, or tortuous segment coverage failed early due to similar hemodynamic

153

changes all causing low in-stent velocities with early stenosis developing. Lastly,

154

small diameter popliteal artery/stent (<6mm) induced early failure.

M AN U

TE D

155

SC

148

While the four-year patency rates for open and endovascular repair were similar in our cohort, endovascular repair led to distinctly lower patency. This

157

finding must give pause and question its utility. Recognizing that new

158

technology/techniques involve a learning curve and the distinct advances in stent-

159

graft technology since the inauguration of our experience with endovascular PAA

160

repair, one realizes that over the course of the review period early failure decreased.

161

Further review over time and increased cohort size is required before being able to

162

unquestionably establish endovascular repair as non-inferior to traditional open

163

repair, however given the significant reduction in length of stay and ability to

164

provide a less invasive/strenuous option for a higher medical risk group while

AC C

EP

156

7

ACCEPTED MANUSCRIPT

165

maintaining equivalent mortality and limb loss rates, it is crucial to continue to

166

advance endovascular viability and continually evaluate for efficacy.

167

Conclusion Endovascular repair of popliteal artery aneurysms demonstrates lower

RI PT

168

patency compared to open surgical bypass. However, given the general benefits of

170

endovascular options, it remains a viable alternative particularly for high surgical

171

risk patients without adequate autologous conduit.

AC C

EP

TE D

M AN U

SC

169

8

ACCEPTED MANUSCRIPT

References

173 174 175

1.

Lawrence PF, Lorenzo-Rivero S, Lyon JL. The incidence of iliac, femoral, and popliteal artery aneurysms in hospitalized patients. J Vasc Surg. 1995 Oct;22(4):409–15–discussion415–6.

176 177 178

2.

Tielliu IFJ, Verhoeven ELG, Zeebregts CJ, Prins TR, Span MM, van den Dungen JJAM. Endovascular treatment of popliteal artery aneurysms: Results of a prospective cohort study. J Vasc Surg. 2005 Apr;41(4):561–6.

179 180 181 182

3.

Antonello M, Frigatti P, Battocchio P, Lepidi S, Dall'Antonia A, Deriu GP, et al. Endovascular treatment of asymptomatic popliteal aneurysms: 8-year concurrent comparison with open repair. J Cardiovasc Surg (Torino). 2007 Jun;48(3):267–74.

183 184 185

4.

Curi MA, Geraghty PJ, Merino OA, Veeraswamy RK, Rubin BG, Sanchez LA, et al. Mid-term outcomes of endovascular popliteal artery aneurysm repair. J Vasc Surg. 2007 Mar;45(3):505–10.

186 187 188

5.

Idelchik GM, Dougherty KG, Hernandez E, Mortazavi A, Strickman NE, Krajcer Z. Endovascular exclusion of popliteal artery aneurysms with stent-grafts: a prospective single-center experience. J Endovasc Ther. 2009 Apr;16(2):215–23.

189 190 191

6.

Garg K, Rockman CB, Kim BJ, Jacobowitz GR, Maldonado TS, Adelman MA, et al. Outcome of endovascular repair of popliteal artery aneurysm using the Viabahn endoprosthesis. J Vasc Surg. 2012 Jun;55(6):1647–53.

192 193 194

7.

Tsilimparis N, Dayama A, Ricotta JJ2. Open and endovascular repair of popliteal artery aneurysms: tabular review of the literature. Annals of Vascular Surgery. 2013 Feb;27(2):259–65.

195 196 197

8.

Brumberg RS, Back MR, Armstrong PA, Cuthbertson D, Shames ML, Johnson BL, et al. The relative importance of graft surveillance and warfarin therapy in infrainguinal prosthetic bypass failure. J Vasc Surg. 2007 Dec;46(6):1160–6.

198 199 200 201 202

9.

AC C

EP

TE D

M AN U

SC

RI PT

172

Bradbury AW, Adam DJ, Bell J, Forbes JF, Fowkes FGR, Gillespie I, et al. Bypass versus Angioplasty in Severe Ischaemia of the Leg (BASIL) trial: An intentionto-treat analysis of amputation-free and overall survival in patients randomized to a bypass surgery-first or a balloon angioplasty-first revascularization strategy. J Vasc Surg. 2010 May;51(5 Suppl):5S–17S.

203 204

9

82 +/- 9.5 (66-98) 69.9 +/- 6.8 (51-74)

Sex

25/25 Male

50/52 Male

LOS: Days1 (Range)

2 +/- 2.3 (1-10)

12 +/- 27 (1-135) 38.5 +/- 31 (1-117)

Limb Loss

1

2

Complications

2

92

Re-interventions 9

27

Concurrent AAA 19 (76%)

36 (70%)

SC

Follow Up: 37.9 +/- 36.3 Months (Range) (1-90)

205 206 207 208 209 210 211 212 213 214 215 216 217

Table 1 The endovascular cohort was significantly older and experienced significantly shorter length of stay (1P <0.05). The two groups were otherwise similar with expected male predominance and high rate of concomitant AAA presence. 2Of the 9 complications in the open group, 6 were superficial wound infections treated with antibiotics Devices

20%

Viabahn

80%

Number of devices

TE D

Wallgraft

M AN U

218 219 220 221 222 223 224 225

Age (range)1

RI PT

Endo ACCEPTEDOpen MANUSCRIPT

2 (range 1-3)

Treatment length (cm) 22.3 (+/- 6.8)

Device Diameter 8-9mm

226 227 228 229 230 231 232 233 234 235 236 237

8

16

AC C

>10mm

10

EP

<7mm

Table 2 The Viabahn endoprosthesis was used exclusively from the time of availability at our institution in 2003. Device diameters were well nearly evenly divided, though stentgrafts >10mm were most common (non-statistically significant).

10

ACCEPTED MANUSCRIPT

Procedure

BK Pop

Tibial

10

SFA

Bypass

50 AK Pop

Interposition

2 Outflow

RI PT

Acute presentation

31 PTFE (%) GSV (%) 17 Inflow 4 CFA

SC

Occlusive disease Long SFA/ BK pop involved

Table 3 The majority of open repairs were via bypass due to surgeon preference. A significant portion of the open surgical patients required tibial bypasses due to extensive athero-occlusive disease and nearly one-third required prosthetic conduit due to insufficient saphenous vein. These factors predisposed to a lower anticipated patency compared to a cohort of simple above-below knee popliteal bypasses with vein.

M AN U

249 250 251 252 253 254 255 256 257 258 259

238 239 16 (30.8) 240 241 36 (69.2) 242 243 5 244 245 35 246 12 247 248 42

Conduit

Reason for Open

Figure Legends

261 262 263 264 265 266 267 268

Figure 1 Patient survival was similar between the two cohorts and consistent with expected 5-year survival for patients with known peripheral arterial disease. Primary patency was similar over the study period, however, with a much lower one-year primary patency for the endovascular group. Primary assisted and secondary patency improved for the open surgical group, a benefit not recognized in the endovascular cohort.

269 270 271 272 273

Figure 2 Both endovascular and open surgical management required a high rate of reintervention over the study period. This was primary for occlusion of the stentgrafts, occlusion for PTFE bypasses, and stenosis of venous bypasses.

AC C

EP

TE D

260

274 275

11

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT