Duplex Ultrasound for Diagnosis of Failing Stents Placed for Lower Extremity Arterial Occlusive Disease

Journal Pre-proof Duplex ultrasound for diagnosis of failing stents placed for lower extremity arterial occlusive disease Hong Zheng, DO, Keith D. Calligaro, MD, Jiah Jang, DO, Samuel Tyagi, MD, Nicholas Madden, DO, Douglas A. Troutman, DO, Matthew J. Dougherty, MD PII:

S0890-5096(19)30765-4

DOI:

https://doi.org/10.1016/j.avsg.2019.08.086

Reference:

AVSG 4628

To appear in:

Annals of Vascular Surgery

Received Date: 22 August 2018 Revised Date:

1 August 2019

Accepted Date: 11 August 2019

Please cite this article as: Zheng H, Calligaro KD, Jang J, Tyagi S, Madden N, Troutman DA, Dougherty MJ, Duplex ultrasound for diagnosis of failing stents placed for lower extremity arterial occlusive disease, Annals of Vascular Surgery (2019), doi: https://doi.org/10.1016/j.avsg.2019.08.086. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Elsevier Inc. All rights reserved.

1

Duplex ultrasound for diagnosis of failing stents

2

placed for lower extremity arterial occlusive disease

3 4 5 6

Hong Zheng, DO, Keith D. Calligaro, MD, Jiah Jang, DO, Samuel Tyagi, MD, Nicholas

7

Madden, DO, Douglas A. Troutman, DO, and Matthew J. Dougherty, MD

8 9 10 11 12

Presented at the Society for Clinical Vascular Surgery.

13

Las Vegas, NV. March 18, 2018.

14

ABSTRACT

15

Objective: Non-invasive diagnostic testing may be beneficial to identify stenotic (failing) stents

16

placed for occlusive lower extremity peripheral arterial disease (LEPAD), especially if

17

subsequent intervention proves useful in maintaining prolonged stent patency. We previously

18

documented the benefit of surveillance duplex ultrasonography (DU) for peripheral covered

19

stents (stent grafts). The purpose of this study was to evaluate whether DU can reliably diagnose

20

failing bare metal stents placed in iliac, femoral and popliteal arteries for LEPAD.

21

Methods: Between January 1, 2013 – December 31, 2016, 172 stents were placed for LEPAD in

22

119 arterial segments (1.4 stents/stenotic artery) in 110 patients who underwent one or more DU

23

surveillance study documenting stent patency. Post-stent DU surveillance was performed in our

24

Inter-societal Accreditation Commission accredited non-invasive vascular lab at one week and

25

then every six months. DU measured peak systolic velocities (PSVs) and ratio of adjacent PSVs

26

(Vr) every 2.0 cm within the stent(s) and adjacent arteries. We retrospectively classified the

27

following factors as “abnormal DU findings”: focal PSVs > 300 cm/s, uniform PSVs < 45 cm/s,

28

and Vr > 3.0.

29

Results: During average follow-up of 22 months (range, 1 week - 48 months), all three of these

30

DU criteria were “normal” in 62 (52%) of the 119 stented segments. Of the other 57 (48%)

31

stented arterial segments that had one or more abnormal DU finding, 40 underwent prophylactic

32

intervention, 12 patients did not undergo intervention and subsequently occluded (5 patient

33

refusal, 4 surgeon-decision, 3 shortened surveillance interval), and 5 remained patent after mean

34

follow-up of 7.2 months. Of the 12 arterial segments that occluded, 6 patients chose not to have

35

further intervention, 4 failed additional endovascular intervention and required an arterial bypass,

36

and 2 required amputation. Therefore, of the 17 stented arterial segments with one or more

37

abnormal DU finding that did not undergo intervention, 12 (70%) went on to occlude vs. 2 of 62

38

(3%) with normal DU findings demonstrating an odds ratio (OR) of 72.0 (95% CI 12.5 – 415.6,

39

P < 0.0001). Of these 12 stented arterial segments with abnormal DU findings that occluded, 7

40

had uniform low PSVs alone, 3 had both abnormal PSV and Vr’s, and 2 had abnormal Vr’s

41

alone.

42

Conclusion: DU surveillance can predict LEPAD stent occlusion. While PSV > 300cm/sec

43

alone is not a statistically significant predictor of stent failure, Vr > 3.0, and most importantly,

44

uniform PSVs < 45 cm/s throughout the stent were statistically reliable markers for predicting

45

stent thrombosis, while the absence of any of these abnormalities strongly predicted stent

46

patency.

47 48

49 50

Introduction As endovascular technologies grow, an increasing number of patients with lower

51

extremity peripheral arterial disease (LEPAD) are being treated with balloon angioplasty,

52

stenting or other endovascular adjuncts. Bare metal stents continue to be widely used in the

53

treatment of ilio-femoral and femoro-popliteal occlusive disease. Duplex ultrasound (DU)

54

surveillance has been suggested to improve patency rates of lower extremity vein and prosthetic

55

bypasses by identifying and prophylactically treating failing grafts. 1-5 Although Baril et al

56

demonstrated a correlation between DU criteria and angiographic evidence of significant in-stent

57

stenosis in superficial femoral arteries6, the value of DU surveillance to identify failing or

58

stenotic peripheral stents remains controversial. The purpose of this study was to determine

59

whether DU criteria can predict failure of bare metal stents placed in iliac and femoro-popliteal

60

arteries.

61 62 63 64

Methods

65

LEPAD by vascular surgeons at Pennsylvania Hospital between January 1, 2013 and December

66

31, 2016 were reviewed. All patients had at least one DU surveillance study documenting stent

67

patency. The post-stent DU surveillance protocol included studies performed in our Inter-

68

societal Accreditation Commission (IAC) accredited non-invasive vascular lab at one week post-

69

intervention and then every three to six months. DU was performed with the patient in the supine

70

position. Imaging was performed from the inflow artery to the outflow artery involving the

71

arterial segment treated with stents. The arteries were measured in the transverse and

72

longitudinal axes. DU measurements were performed every 2.0 cm within the stent(s) and

Patients treated with bare metal stents placed in the iliac and femoro-popliteal arteries for

73

adjacent arteries with B-mode imaging and spectral doppler. DU was performed by registered

74

vascular laboratory technologists using version 6.3.5.7 software (Philips HD-11, Phillips DHI-

75

5000, Philips HDI-3000; Bothell, WA). Patient studies were prospectively maintained in a

76

computer database (Access, Microsoft Corp, Redmond CA). Odds ratio calculations were

77

performed to compare the group who had abnormal duplex ultrasound and did not receive

78

reintervention (the ‘exposed’ group) to the normal duplex ultrasound group (the ‘unexposed’

79

group). Patient consent and Institutional Review Board approval were not obtained due to use of

80

a de-identified database.

81

DU parameters measured included peak systolic velocities (PSVs), the ratio of adjacent

82

PSVs (Vr), and the luminal diameter of the stent and adjacent artery. We classified the following

83

factors as “abnormal” DU findings: focal PSVs > 300 cm/s, uniform PSVs < 45 cm/s throughout

84

the stented segment, or a Vr > 3.0, based on previously published criteria for failing lower

85

extremity bypass grafts and stent grafts (1-6). Studies with none of the DU findings were

86

classified as “normal”.

87 88 89 90

Results

One-hundred seventy-two stents were placed for LEPAD in 119 arterial segments (1.4

91

stents/stenotic artery) in 110 patients and 116 limbs who underwent more than one DU

92

surveillance study documenting stent patency. There were 77 LifeStent (C. R. Bard Inc, Temple,

93

AZ) (45%), 43 Epic (Boston Scientific, Natick, MA) (25%), 26 Luminexx (C. R. Bard Inc,

94

Temple, AZ) (15%), 12 Supera (Abbott, Abbott Park, IL) (7%), 10 Wallstent (Boston Scientific,

95

Natick, MA) (6%), and 4 Protégé Everflex (Medtronic, Minneapolis, MN) (2%) stents inserted in

96

30 iliac and 89 femoro-popliteal arteries. Mean treatment length was 7.5 cm in iliac arteries

97

(range, 4.0 – 15.0 cm) and 12.2 cm in femoro-popliteal arteries (range, 2.0 - 21.0 cm). Mean

98

follow-up was 22 months (range, 1 week - 48 months).

99 100

Patients were classified according to presenting symptoms using the Rutherford

101

Classification for chronic ischemia (Table I). The majority of limbs treated were for Rutherford

102

classification 3-5 (severe claudication to minor tissue loss).

103 104

Of the 119 stented segments, 62 (52%) had “normal” DU criteria, without any of the

105

three abnormal criteria. During follow-up 60 remained patent and two occluded. One occlusion

106

was treated with a stent graft and the other failed endovascular re-intervention and required an

107

arterial bypass for rest pain. The other 57 (48%) stented arterial segments had one or more

108

“abnormal” DU finding: 40 underwent prophylactic intervention and remained patent with

109

normal DU on subsequent follow-up. All 40 of these patients demonstrated a hemodynamically

110

significant lesion at arteriography. This was treated either by scoring balloon or drug-coated

111

balloon angioplasty and selective atherectomy of native lesions, with new stent placement if

112

there was greater than50% residual stenosis following this initial intervention. Twelve patients

113

with abnormal duplex findings did not have reintervention and occluded while 5 did not have

114

reintervention and remained patent at mean follow-up of 7.2 months. Of the 12 patients who did

115

not have reintervention, 5 patients refused, and in 7 patients intervention was deferred in favor of

116

a shortened surveillance interval of three months.

117

Of these 12 arterial segments that occluded with “abnormal” DU findings, 6 patients

118

chose no further intervention and remained with claudication only, 4 failed additional

119

endovascular intervention and required an arterial bypass, and 2 required amputation. Therefore,

120

12 (70%) of the 17 stented arterial segments with one or more “abnormal” DU finding that did

121

not undergo intervention progressed to occlusion vs. 2 of 62 (3%) with “normal” DU findings

122

(Table II). Of these 12 stented arterial segments with abnormal DU findings that occluded, 7 had

123

uniform low PSVs only, 3 had both abnormal elevated PSV > 300cm/sec and Vr, and 2 had

124

abnormal Vr only (Table III). A total of 14 stents occluded. The mean time to occlusion was 5.6

125

months (95% confidence interval 4.3 – 6.9).

126

We performed a standard odds ratio (OR) calculation, comparing the odds of stent

127

thrombosis between the abnormal and normal duplex groups. This calculation was done by

128

dividing the product of the thrombosed stents in the abnormal duplex group and the open stents

129

in the normal duplex group by the product of the open stents in the abnormal duplex group and

130

the thrombosed stents in the normal duplex group. Evaluating for the presence of any abnormal

131

duplex criteria, we found an Odds Ratio of 72.0 (95% CI 12.5 – 415.6, P < 0.0001). We then

132

looked at each individual criterion. We did include patients in the abnormal group which did not

133

have that specific criteria in the control group for these calculations (for example, patients with

134

uniform PSV < 45cm/sec were included in the control group when analyzing patients with PSV

135

> 300cm/sec, etc). For PSV > 300cm/sec the OR is 4.2 (95% CI 0.8 – 21.2, P = 0.08). For Vr >

136

3.0 the OR is 11.5 (95% CI 2.3 – 56.5, P = 0.0027) and for uniform PSV < 45cm/sec the OR is

137

64.0 (95% CI 6.8 – 598.7, P = 0.0003). Combination of PSV > 300cm/sec and Vr > 3.0 had OR

138

of 5.6 (95% CI 1.0 – 31.6, P = 0.0493). This suggests uniform PSVs < 45 cm/sec was the most

139

accurate predictor of stent failure.

140 141 142 143

Discussion The incidence of re-stenosis and occlusion following endovascular interventions for

144

LEPAD has been reported to be as high as 40% after one year.7 With increasing use of stents to

145

treat LEPAD, DU surveillance may prove useful to help maintain patency of these

146

revascularizations should they develop stenotic lesions. However, there are conflicting

147

recommendations in the literature whether this strategy is worthwhile.

148

DU surveillance has been widely accepted for lower extremity vein bypass grafts to

149

detect lesions that may lead to graft failure. It is less accepted for prosthetic bypasses, and there

150

is currently limited evidence for its use in peripheral arterial stents. Intimal hyperplasia is most

151

likely to develop during the first two years after lower extremity vein bypass grafts, hence the

152

SVS Guidelines for DU vein graft surveillance recommending studies every three months for the

153

first year, every six months for the next year, and annually thereafter. There are no established

154

guidelines for peripheral arterial stent DU surveillance. While bypass graft failure is frequently

155

associated with a need for more complex treatment and poor outcomes, this may not be the case

156

for bare metal stent thrombosis.

157

Tielbeek et al described similar sensitivity and specificity for peripheral stent surveillance

158

using DU surveillance compared to following patients with only symptomatic and ankle-brachial

159

index (ABI) changes.9 Bui et al also reported that DU surveillance may not be worthwhile since

160

DU predicted occlusion with a sensitivity and specificity of only 88% and 60%, respectively.10

161

Conversely, Kinney et al have shown that ABI alone is insufficient to distinguish

162

recurrent stent stenosis vs. progression of atherosclerotic disease in untreated areas. 11 Baril and

163

associates from the University of Pittsburgh identified DU criteria to detect in-stent re-stenosis in

164

the superficial femoral artery and support its use.6 This group used PSVs and Vr to determine DU

165

criteria that was both sensitive and specific for in-stent re-stenosis of the femoro-popliteal artery.

166

They concluded that a PSV > 275 cm/sec or a Vr > 3.5 was highly specific and predictive of

167

80% or more in-stent stenosis. A review of the literature by Shames suggested that routine post-

168

operative DU surveillance of lower extremity stents may help to improve primary assisted

169

patency rate to 80% at one year.7 Finally, a recent Reporting Standards document by the Society

170

for Vascular Surgery for endovascular treatment of chronic LEPAD recommended DU imaging

171

as the standard for lower extremity endovascular surveillance. 12 Studies were recommended to

172

be performed in an accredited non-invasive vascular laboratory one week after the endovascular

173

intervention, every three months the first year, and every six months or annually thereafter.12

174

We have previously described the utility of DU for surveillance of lower extremity stent

175

grafts (covered stents) placed for peripheral arterial disease.13 In our current retrospective

176

analysis of peripheral stent surveillance, we applied similar criteria to diagnose a failing (or

177

stenotic) peripheral stent. We defined “abnormal DU findings” as a focal PSV > 300 cm/sec,

178

uniform PSVs < 45 cm/sec, or Vr > 3.0. When taken together the odds ratio (OR) for stent

179

thrombosis with any positive duplex finding is 72.0. When evaluated independently, PSV >

180

300cm/sec alone is not a significant predictor with OR of 4.2 (95% CI 0.8 – 21.2, P = 0.08).

181

While Vr > 3.0 and uniform PSV < 45cm/sec are both predictors of stent failure, uniform PSV <

182

45cm/sec is the strongest predictor for stent thrombosis with an OR of 64.0 and 95% CI from 6.8

183

– 598.7.

184

We included a life-table (Graph I) evaluating the percentage of stents remaining patent in

185

the three different groups of our study which demonstrates the significantly increased risk of a

186

thrombotic complication in the patients with these abnormal findings.

187

If reliable DU criteria are established to predict a failing peripheral arterial stent, the

188

question remains whether subsequent prophylactic treatment of these stenotic lesions prior to

189

onset of symptoms or stent occlusion is worthwhile. There are several advantages of intervening

190

before symptoms develop or stents occlude. First, onset of symptoms is commonly associated

191

with stent occlusion and not just stent stenosis, similar to the reasoning for intervention of failing

192

lower extremity vein bypasses prior to onset of symptoms. Technical success is higher when

193

treating a stenotic stent than an occluded one, especially for a long occlusion. A stent placed for

194

claudication that acutely occludes may result in thrombosis of important collaterals with an

195

associated limb-threatening situation. Thrombolysis may prove necessary to salvage an occluded

196

stent, especially if distal small arteries have also occluded, and has associated morbidity and

197

extra cost. Restoring patency of an occluded stent may be associated with an increased likelihood

198

of distal embolization compared to balloon angioplasty of a stenotic but patent stent that is not

199

filled with thrombus. Lastly, similar to lower extremity bypasses, subsequent patency of a failing

200

but patent stent may be higher than patency of a stent that has already occluded.

201

There are several limitations of this study. These results were based on findings reported

202

by highly skilled technologists in an IAC-accredited non-invasive vascular laboratory that is

203

devoted to vascular duplex surveillance studies. Other vascular labs may not yield the same

204

results. Peripheral stents placed in iliac, femoral and popliteal arteries were analyzed in this

205

study, so recommendations cannot be made regarding tibial stents. Our series had a relatively

206

small sample size. Additionally, our database did not include data regarding demographics,

207

comorbidities or complete TASC classification data and there exists a possibility of a

208

confounding variable explaining our findings. Lastly, DU surveillance of stents may also be

209

limited by the patient’s habitus, dense calcifications, or prior surgeries.

210

Follow-up studies based on this information could include prospective or randomized

211

trials evaluating the natural history and effect of intervention on abnormal duplex findings

212

following peripheral arterial stenting.

213 214 215

Conclusion

216

Our results suggest that DU surveillance can predict LEPAD stent thrombosis. Focal

217

PSVs > 300 cm/s, Vr > 3.0, and most importantly, uniform PSVs < 45 cm/s throughout the stent

218

were statistically reliable markers for predicting stent thrombosis, while the absence of any of

219

these abnormalities strongly predicted stent patency.

220

221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266

REFERENCES 1. Calligaro KD, Syrek JR, Dougherty MD, Rua I, McAffee-Bennett S, Doerr KJ et al. Selective use of duplex ultrasound to replace preoperative arteriography for failing arterial vein grafts. J Vasc Surg 1998;27:89-95. 2. Calligaro KD, Musser DJ, Chen AY, Dougherty MD, McAffee-Bennett S, Doerr KJ et al. Duplex ultrasonography to diagnose failing arterial prosthetic grafts. Surgery 1996;120:455-59. 3. Calligaro KD, Doerr K, McAffee-Bennett S, Krug R, Raviola CA, Dougherty MD. Should duplex ultrasonography be performed for surveillance of femoropopliteal and femorotibial arterial prosthetic bypass? Ann Vasc Surg 2001;15:520-24. 4. Bandyk DF, Schmitt DD, Seabrook GR, Adams MB, Towne JB. Monitoring functional patency of in situ saphenous vein bypasses: The impact of a surveillance protocol and elective revision. J Vasc Surg 1989:9:286-96. 5. Mills JL, Harris EJ, Taylor LM, Beckett WC, Porter JM. The importance of routine surveillance of distal bypass grafts with duplex scanning: A study of 379 reversed vein grafts. J Vasc Surg 1990;12:379-89. 6. Baril DT, Rhee RY, Kim J, Makaroun MS, Chaer RA, Marone LK. Duplex criteria for determination of in-stent stenosis after angioplasty and stenting of the superficial femoral artery. J Vasc Surg 2009;49:133-39. 7. Shames ML. Duplex surveillance of lower extremity endovascular interventions. Perspect Vasc Surg Endovasc Ther 2007;19:370-374 8. Conte MS, Pomposelli FB, Clair DG, Geraghty PJ, McKinsey JF, Mills JL, Moneta GL, Murad MH, Powell RJ, Reed AB, Schanzer A, Sidawy A. Society for Vascular Surgery practice guidelines for atherosclerotic occlusive disease of the lower extremities: management of asymptomatic disease and claudication. J Vasc Surg 2015;61:2Se41S. 9. Tielbeek AV, Rietjens E, Buth J, Vroegindeweij D, Schol FPG. The value of duplex surveillance after endovascular intervention for femoropopliteal obstructive disease. Eur J Vasc Endovasc Surg 1996; 12:145-150 10. Bui TD, Mills JL, Ihnat DM, Gruessner AC, Goshima KR, Hughes JD. The nature history of duplex-detected stenosis after femoropopliteal endovascular therapy suggests questionable clinical utility of routine duplex surveillance. J Vasc Surg 2012;55:346-52 11. Kinney EV, Bandyk DF, Mewissen MW, Lanza D, Bergamini TM, Lipchik EO, Seabrook GR, Towne JB. Monitoring functional patency of percutaneous transluminal angioplasty. Arch Surg. 1991;126(6):743-747 12. Stoner MC, Calligaro KD, Chaer RA, Dietzek AM, Farber A, Guzman RJ, Hamdan AD, Landry GJ, Yamaguchi DJ on behalf of the SVS. Reporting standards of the Society for Vascular surgery for endovascular treatment of chronic lower extremity peripheral artery disease. J Vasc Surg 2016;64:e1-e-21 13. Troutman DA, Madden NJ, Dougherty MJ, Calligaro KD. Duplex ultrasound diagnosis of failing stent grafts placed for occlusive disease. J Vasc Surg 2014;60;1580-1584

267 268 269 270

Table I Clinical Symptoms for patients treated with peripheral artery stents Rutherford Classification 1 2 3 4 5 6

Iliac

Femoro-popliteal

4 18 5 3 3

6 51 20 11 -

271 272

Table II

273

Stent patency based on normal versus abnormal Duplex Ultrasound (DU) criteria Total Cohort

274 275 276 277 278 279 280

281 282

Abnormal DU

Normal DU

Total

Occluded

12

2

14

Patent

5

60

65

Total

17

62

79

Table III Duplex ultrasound (DU) criteria characteristics for occluded stents

DU criteria PSV > 300cm/s alone Vr > 3.0 alone Uniform PSV < 45cm/s (low flow) PSV > 300cm/s + Vr > 3.0

N 0 2 7 3

Total

12

DU, Duplex Ultrasound; PSV, Peak Systolic Velocity; Vr, Velocity Ratio

283 284

Table IV Abnormal DU Criteria in stents which did not occlude DU Criteria PSV > 300cm/s alone Vr > 3.0 alone Uniform PSV < 45cm/s PSV > 300cm/s and Vr > 3.0 Total

285 286 287 288 289

290 291 292 293

DU, Duplex Ultrasound; PSV, Peak Systolic Velocity; Vr, Velocity Ratio

Figure I Life Table Demonstrated as Percent (%) Stent Patency over Time

Table V Patients at risk for each time period

Months

Normal 0 3 6 9 12 15 18

294 295 296

N 1 0 1 3 5

62 60 58 58 57 57 57

Abnormal Duplex + Intervention

Abnormal Duplex + Observation 40 40 39 39 38 38 38

17 15 7 3 3 2 2