Increased warm ischemia time during vessel harvest decreases the primary patency of cryopreserved conduits in patients undergoing lower extremity bypass

From the Southern Association for Vascular Surgery

Increased warm ischemia time during vessel harvest decreases the primary patency of cryopreserved conduits in patients undergoing lower extremity bypass J. Michael Cullen, MD,a J. Hunter Mehaffey, MD, MSc,a Robert B. Hawkins, MD, MSc,a Vikram Gupta,b Rishi A. Roy, MD,c William P. Robinson III, MD,c Margaret C. Tracci, MD, JD,c Kenneth J. Cherry, MD,c John A. Kern, MD,d and Gilbert R. Upchurch Jr, MD,c,e Charlottesville, Va; and Gainesville, Fla

ABSTRACT Objective: Autologous vein is the preferred conduit for lower extremity bypass. However, it is often unavailable because of prior harvest or inadequate for bypass owing to insufficient caliber. Cryopreserved cadaveric vessels can be used as conduits for lower extremity revascularization when autogenous vein is not available and the use of prosthetic grafts is not appropriate. Many studies have shown that donor characteristics influence clinical outcomes in solid organ transplantation, but little is known regarding their impact in vascular surgery. The purpose of this study was to examine the effects donor variables have on patients undergoing lower extremity bypass with cryopreserved vessels. Methods: The tissue processing organization was queried for donor blood type, warm ischemia times (WITs), and serial numbers of cryopreserved vessels implanted at a single center from 2010 to 2016. The serial numbers were then matched with their respective patients using the institutional Clinical Data Repository and patient data were obtained from the Clinical Data Repository and chart review. Primary outcomes were primary patency of the bypass conduits and limb salvage. Time to loss of patency was evaluated using Kaplan-Meier methods and a Cox proportional hazards model determined risk-adjusted predictors of patency and limb salvage. Results: Sixty patients underwent lower extremity bypass with 65 cryopreserved vessels (23 superficial femoral arteries, 41 saphenous veins, 1 femoral vein). Thirty-eight procedures were reoperations. There were 21 inflow, 44 outflow, and 44 infrainguinal procedures. Preexisting comorbidities did not differ significantly between those who lost patency and those who did not. The mean WIT among the entire cohort was 892.3 6 389.1 minutes (range, 158.0-1434.0 minutes). The median follow-up was 394 days. Kaplan-Meier analysis demonstrated an overall 1-year primary patency rate of 51%. Primary patency at 1 year was 67% and 41% for inflow and outflow procedures, respectively, and did not differ significantly between the two groups (P ¼ .15). Donor-to-recipient ABO incompatibility was not associated with loss of primary patency. The 1-year amputation-free survival was 74%. Primary patency significantly decreased with each hourly increase in WIT on risk-adjusted analysis (hazard ratio, 1.1; P ¼ .02). Conclusions: Higher cryopreserved vessel WIT was associated with increased risk-adjusted loss of primary patency in this cohort. At 1 year, the overall primary patency was 51% and amputation-free survival was 74%. Vascular surgeons should be aware that WIT may affect outcomes for lower extremity bypass. (J Vasc Surg 2018;-:1-10.) Keywords: Vascular surgical procedures; Vascular grafting; Tissue transplantation; Lower extremity; Limb salvage; Warm ischemia; Proportional hazards models; Kaplan-Meier estimate

Autologous vein is the preferred conduit for lower extremity bypass in patients with peripheral arterial disease owing to superior patency. However, autologous vein is often unavailable owing to prior harvest or inadequate caliber for successful bypass.1 Cryopreserved veins and arteries are occasionally used as an alternative in these situations, although the literature reports patency

rates inferior to both autologous vein and synthetic materials.2-7 Consequently, some authors have proposed that cryopreserved conduits should be used as a last resort or only for critical limb ischemia.8 Extensive literature demonstrates that numerous cadaveric donor characteristics impact clinical outcomes in solid organ allotransplantation, but little is

From the Department of Surgery,a College of Arts & Sciences,b Division of

Correspondence: Gilbert R. Upchurch Jr, MD, Edward R. Woodward Professor &

Vascular and Endovascular Surgery,c and Division of Thoracic and Cardiovas-

Chair, University of Florida Department of Surgery, PO Box 100286, 1600 SW

cular Surgery,d University of Virginia, Charlottesville; and the Department of Surgery, University of Florida, Gainesville.e Supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under Award Number T32HL007849. Author conflict of interest: none. Presented at the Forty-second Annual Meeting of the Southern Association for Vascular Surgery, Scottsdale, Ariz, January 17-20, 2018.

Archer Rd, Rm 6174, Gainesville, FL (e-mail: [email protected]fl.edu). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2018 by the Society for Vascular Surgery. Published by Elsevier Inc. https://doi.org/10.1016/j.jvs.2018.04.065

1

2

Journal of Vascular Surgery

Cullen et al

---

known regarding the effects donor characteristics have on vascular allotransplantation.9 Variables impacting the patency of cryopreserved conduits remain an understudied area of clinical importance. The purpose of this study was to examine the risk-adjusted impact that patient, operative, and cadaveric donor characteristics have on patients undergoing lower extremity bypass with a cryopreserved vessel conduit. We hypothesized that increased warm ischemia times (WITs) would negatively impact the primary patency of these bypass grafts.

METHODS Clinical data. Lifenet Health, the tissue processing organization, was queried for the serial numbers, allograft donor blood type, Rhesus immunization status, and WITs for all cryopreserved conduits implanted at a single center from 2010 to 2016. A total of 101 serial numbers for the corresponding cryopreserved conduits were provided during the study period. The institutional Clinical Data Repository was queried to match the serial numbers with the respective patient. Clinical data for all patients who underwent a lower extremity bypass with a cryopreserved superficial femoral artery, femoral vein, or saphenous vein for symptomatic peripheral arterial disease were included for analysis. Cryopreserved vessels used for trauma, upper extremity, carotid, cardiothoracic, or hepatobiliary purposes were excluded. Patient data were obtained from the Clinical Data Repository and chart review. This study was approved by the University of Virginia Institutional Review Board with a waiver of consent owing to its retrospective nature (IRB#17900). Variable definitions and outcomes. This study’s primary outcomes were primary patency of the cryopreserved bypass graft and limb salvage. The outcomes were defined according to the recommended reporting standards as detailed by Rutherford et al.10 In short, a graft is primarily patent if it has uninterrupted patency with no subsequent procedures or has undergone a procedure to treat disease progression in an adjacent native vessel. Loss of primary patency was determined by objective testing including duplex ultrasound examination, angiography, magnetic resonance imaging, or direct observation. The presence or absence of a palpable pulse on physical examination was not used for this determination. A limb salvage procedure is defined as an operation intended to avoid a major amputation. Patients with prior major amputations, defined as an amputation requiring a prosthesis for standing or walking, were not included in the limb salvage analysis in accordance with the reporting standards.10 Preoperative variables of interest were the medical comorbidities, donor blood type, WIT, whether the procedure was an inflow or outflow operation, and

2018

ARTICLE HIGHLIGHTS d

d

d

Type of Research: Single institution retrospective study of patients who underwent bypass with a cryopreserved artery or vein to determine the impact of warm ischemia time (WIT) during harvesting on graft patency Take Home Message: Longer WITs during harvesting were associated with lower cyropreserved vein and artery patency. Recommendation: The authors suggest that the WIT during harvesting of a cryopreserved artery or vein be considered when using it for revascularization.

the use of arterial or venous allograft. Chronic kidney disease was defined as being at least stage III renal failure. Inflow procedures were classified as axillofemoral, iliofemoral, and femorofemoral bypasses. Outflow procedures were classified as femoropopliteal, femoraldistal, and popliteal-distal bypasses. Per the transplant literature, WIT refers to the period of time from arrest of donor cardiac activity to cooling of the explanted solid organ for transport or storage.11 The WITs reported herein were recorded and verified by the tissue processing organization at the time of vessel harvest from the cadaveric donor. Saphenous veins are harvested first from the donor, followed by femoral arteries and veins. These values are reported on the product schematic when distributed to a surgical center for use. For the present study, the WIT represents the elapsed time beginning with either cessation of donor cardiac activity or pronouncement of death and ending with donor vessel explantation and cold storage; it is not inclusive of the duration of time after a cryopreserved vessel is thawed, prepared on the back table, and surgically implanted for bypass. Below-the-knee bypasses were classified as those with any infrageniculate target, including femoral below-the-knee popliteal and distal bypasses. Distal bypasses were classified as those to the tibioperoneal trunk, anterior or posterior tibial artery, or peroneal artery. Statistical analysis. Continuous variables were normally distributed and presented as mean 6 standard deviation and categorical variables as count (%). Vessels were stratified by inflow or outflow procedure status for univariate analysis, using either independent t-test or c2 test as appropriate. Additionally, time to loss of primary patency of the bypass conduits and limb salvage were evaluated using Kaplan-Meier methods and a Cox proportional hazards model was used to determine risk-adjusted predictors for primary patency of the bypass conduits and limb salvage. All statistical analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC) with an a threshold of 0.05 for statistical significance.

Journal of Vascular Surgery Volume

-,

Number

Cullen et al

3

-

Table I. Unadjusted baseline donor, patient, and operative characteristics stratified by inflow or outflow Inflow procedure (n ¼ 21)

Outflow procedure (n ¼ 44)

59.2 6 13.6

62.7 6 10.4

P value

Patient characteristics Age, years

.15

Male sex

12 (57.1)

27 (61.4)

.75

Hyperlipidemia

11 (52.4)

26 (59.1)

.61

Hypertension

11 (52.4)

28 (63.6)

.39

Coronary artery disease

17 (81.0)

28 (63.6)

.16

Chronic kidney disease

8 (38.1)

6 (13.6)

.02

Diabetes mellitus

11 (52.4)

17 (38.6)

.30

10 (47.6)

27 (61.4)

.30

Tobacco abuse Procedure Axillofemoral

5 (23.8)

0 (0)

Iliofemoral

5 (23.8)

0 (0)

Femorofemoral

11 (52.4)

0 (0)

Femoropopliteal

0 (0)

11 (25.0)

Femoral-distal

0 (0)

29 (65.9)

Popliteal-distal

0 (0)

4 (9.1)

Infrageniculate

0 (0)

42 (95.5)

Critical limb ischemia

5 (23.8)

37 (84.1)

Aneurysm

9 (42.9)

1 (2.3)

Claudication

1 (4.8)

5 (11.4)

Infected graft

6 (28.6)

1 (2.3)

Indication

Donor characteristics Cryopreserved artery

16 (76.2)

7 (15.9)

Donor A

4 (22.2)

20 (52.6)

Donor B

5 (27.8)

5 (13.2)

Donor AB

3 (16.7)

1 (2.6)

<.0001

Donor O

6 (33.3)

12 (31.6)

ABO seromatch

9 (42.9)

21 (47.7)

Rhesus positive

15 (83.3)

30 (79.0)

.70

Rhesus seromatch

14 (66.7)

27 (61.4)

>.05

863.0 6 369.7

906.6 6 401.7

.39

6.5 6 1.3

5.1 6 1.2

<.0001

19 (90.5)

36 (81.8)

.37

8 (38.1)

15 (34.1)

.75

Anticoagulation

7 (33.3)

12 (27.3)

.62

30-Day mortality

0 (0)

3 (6.8)

.22

WIT, minutes Conduit diameter, mm

>.05

Postoperative Aspirin Dual antiplatelet

WIT, Warm ischemia time. Values are number (%) or mean 6 standard deviation.

RESULTS Patient characteristics. Data were collected from 60 patients with peripheral arterial disease who underwent lower extremity bypass with 65 cryopreserved conduits during the study period. Among the entire cohort, 60% were male and the average age was 61.6 6 11.5 years. Table I details the unadjusted baseline patient, operative, and donor characteristics stratified by inflow and outflow procedures. There were significantly more

patients with chronic kidney disease in the inflow group compared with the outflow group (38.1% vs 13.6%; P ¼ .02). There were no differences in age, sex, hyperlipidemia, hypertension, coronary artery disease, diabetes mellitus, or tobacco abuse between the two groups (Table I). Additionally, there were no differences in patient demographics or preexisting comorbidities among grafts that lost primary patency and those that did not (Table II).

4

Journal of Vascular Surgery

Cullen et al

---

2018

Table II. Unadjusted univariate analysis of patient, operative, and donor characteristics stratified by patency status Lost primary patency (n ¼ 26)

No loss of primary patency (n ¼ 39)

P value

59.5 6 9.1

62.9 6 12.8

.12 .76

Patient characteristics Age, years Female sex

11 (42.31)

15 (38.46)

Caucasian

20 (76.92)

32 (82.05)

Coronary artery disease

18 (69.23)

27 (69.23)

1.0

Chronic kidney disease

6 (23.08)

8 (20.51)

.81

.54

Diabetes mellitus

11 (42.31)

17 (43.59)

.92

Hyperlipidemia

13 (50.00)

24 (61.54)

.36

Hypertension

16 (61.54)

23 (58.97)

.84

Tobacco abuse

18 (69.23)

19 (48.72)

.10

Procedure Inflow

6 (23.1)

Axillofemoral

2 (7.7)

3 (7.7)

Iliofemoral

3 (11.5)

2 (5.1)

Femorofemoral

1 (3.9)

10 (25.6)

Femoropopliteal Femoral-distal

20 (51.3)

5 (19.2)

6 (15.4)

12 (46.2)

17 (43.6)

Popliteal-distal

3 (11.5)

1 (11.5)

Infrageniculate

18 (69.2)

24 (61.5)

18 (69.2)

24 (61.5)

.52

Indication Critical limb ischemia Aneurysm

3 (11.5)

7 (18.0)

Claudication

2 (7.7)

4 (10.3)

Infected graft

3 (11.5)

4 (10.3)

Cryopreserved artery

7 (26.9)

16 (41.0)

Donor A

9 (40.9)

15 (44.1)

Donor B

4 (18.2)

6 (17.7)

Donor AB

2 (9.1)

2 (5.9)

Donor O

7 (31.8)

11 (32.3)

Donor characteristics .24

ABO seromatch

11 (42.3)

19 (48.7)

.61

Rhesus positive

18 (81.8)

27 (79.4)

.83

Rhesus seromatch

23 (67.7)

.24

964.4 6 378.9

846.1 6 393.4

.22

5.5 6 1.6

5.6 6 1.3

.73

Aspirin

22 (84.62)

33 (84.6)

Dual antiplatelet

14 (53.9)

9 (23.1)

.01

Anticoagulation

4 (15.4)

15 (38.5)

.5

WIT, minutes Conduit diameter, mm

18 (81.8)

Postoperative care 1.0

WIT, Warm ischemia time. Values are number (%) or mean 6 standard deviation.

Operative details. Among the entire cohort, 38 operations were redo procedures. Two patients received a cryopreserved vessel bypass in both lower extremities and three patients had cryopreserved vessel bypasses that failed, were excised, and a redo bypass was performed using another cryopreserved conduit. Autologous vein options were either inadequate or absent in 27 cases. Nineteen cryopreserved conduits were

implanted after failed catheter-directed attempts at revascularization. For 19 procedures, cryopreserved conduits were selected because of infection in a synthetic graft, aneurysm, or the adjacent tissue. A total of 21 inflow procedures and 44 outflow procedures were performed using the 65 cryopreserved conduits (Table I). Inflow procedures included axillofemoral (23.8%), iliofemoral (23.8%), and femorofemoral

Journal of Vascular Surgery Volume

-,

Number

Cullen et al

5

-

Fig 1. Kaplan-Meier curves demonstrating percent primary patency of overall, inflow, and outflow procedures (A), and cryopreserved arterial versus vein bypass conduits (B).

(25%) bypasses. Synthetic grafts were avoided in these procedures mostly owing to infectious concerns because four axillofemoral (80%), three iliofemoral (60%), and nine femerofemoral (81.8%) operations were performed in the face of infection. Outflow procedures included femoropopliteal (25%), popliteal-distal (9.1%), and femoral-distal bypasses (65.9%). Infrageniculate bypasses accounted for 95.5% of the outflow procedures. The most common surgical indication overall was critical limb ischemia (64.6%; P < .01), and significantly more outflow operations were indicated for critical limb ischemia compared with inflow procedures (84.1% vs 23.8%; P < .0001). Donor characteristics. Twenty-three cryopreserved superficial femoral arteries and 42 cryopreserved veins (41 saphenous veins and 1 femoral vein) were implanted. Cryopreserved arteries were used as conduits in significantly more inflow procedures compared with outflow procedures (76.2% vs 15.9%; P < .0001). Overall, cryopreserved arteries were used in 1 axillofemoral bypass (20%), 5 iliofemoral bypasses (100%), 10 femorofemoral bypasses (90.9%), 4 femoropopliteal bypasses (36.4%), 3 femoral-distal bypasses (10.3%), and in no popliteal-distal bypasses. Surgical indication did not differ between inflow and outflow groups, or grafts that lost patency. Donor blood types did not differ significantly between groups, and donorto-recipient ABO/Rhesus compatibility did not differ between groups (Tables I and II). The mean WIT among the entire cohort was 892.3 6 389.1 minutes (range, 158.0-1434.0 minutes). On univariate analysis there was no difference in the average WIT between the inflow and outflow groups (Table I). The average minimum conduit diameter was significantly larger in inflow procedures compared with outflow procedures (6.5 6 1.3 vs 5.1 6 1.2; P < .0001). Patency and limb salvage. Fig 1 displays the KaplanMeier analysis for time to loss of primary patency, stratified by inflow versus outflow (Fig 1, A) and arterial versus venous conduit (Fig 1, B). Overall primary patency

at 1 year was 51.1%. Primary patency at 1 year for inflow procedures was 67.2% compared with 41.4% for outflow procedures (P ¼ .15). Primary patency at 1 year was 62.9% for arterial conduits and 42.9% for venous conduits (P ¼ .09); importantly, the average WIT was significantly shorter among cryopreserved arteries versus veins (732.3 6 310.2 vs 982.0 6 403.2; P ¼ .02). Fig 2 displays boxplots of the WIT among patent and nonpatent grafts at 100 days (Fig 2, A) and 300 days (Fig 3, A). Overall, 26 bypasses lost primary patency. Analysis of primary assisted and secondary patency did not reveal a correlation with WIT. Table III displays the patient, operative, and donor characteristics of the 36 operations meeting criteria for limb salvage. Eight of these patients ultimately underwent a major amputation. Limb salvage operations were significantly more likely to involve an outflow procedure (86.1% vs 13.9%; P < .05). Kaplan-Meier analysis demonstrates the amputation-free survival at 1 year to be 74% (Fig 3). Table IV displays the hazard ratios (HRs) for loss of primary patency and limb loss. After adjusting for chronic kidney disease, diabetes, tobacco abuse, and inflow and outflow vessels, the risk-adjusted primary patency independently decreased with each hourly increase in WIT (HR, 1.1; P ¼ .02). Chronic kidney disease was an independent predictor for loss of primary patency (HR, 5.13; P ¼ .04). Distal and popliteal outflow procedures were significant predictors for loss of primary patency (HR, 49.6 and 87.4; P ¼ .03 and .01, respectively). Cryopreserved artery compared with cryopreserved vein did not independently predict patency in this population (HR, 0.65; P ¼ .60). WIT did not independently predict limb loss (HR, 1.0; P ¼ .60). ABO compatibility was not significantly associated with an increased likelihood of limb salvage (HR, 1.87; P ¼ .51).

DISCUSSION The present study reports the impact of cadaveric donor characteristics on the primary patency of cryopreserved vascular conduits used for lower extremity bypass and represents the first to evaluate WIT and find that WIT

6

Journal of Vascular Surgery

Cullen et al

---

p = 0.27

1500 1000 500 0

Patent vs. Non-Patent at 300 days p = 0.06

2000 1500 1000 500

(n te nt

te nt Pa

Pa

N

N

ot

ot

Pa

=2 2)

=2 6) (n

(n te nt

te nt

(n

=1 3)

0

Pa

2000

B Warm Ischemia Time (mins)

Patent vs. Non-Patent at 100 days

=4 2)

Warm Ischemia Time (mins)

A

2018

% Amputation-Free Survival

Fig 2. Boxplot graphs of warm ischemia times (WITs) among patent and not patent grafts at 100 days (A) and 300 days (B).

100

Number At Risk

50

0

0

100

36

28

200 Days 24

300 20

17

Fig 3. Kaplan-Meier curve of percent amputation-free survival among limb salvage operations.

affects patency. These data showed that primary patency decreased for every hour increase in WIT (HR, 1.1; P ¼ .02). ABO compatibility did not affect limb salvage. Overall primary patency at 1 year was 51.1%. Amputation-free survival at 1 year was 74%. These results provide additional insight into using cryopreserved conduits for these procedures. Overall primary patency at 1 year in the present study was 51.1%, and is one of the higher rates compared with similar series.2-7 In a series of 35 patients, Walker et al4 reported 28% patency at the same interval. The largest series to date, including 177 patients and 240 conduits, by Farber et al7 had a similar rate of 30% at 12 months. The finding of a limb salvage rate of 74% at 1 year is analogous to similar series. Martin et al5 reported outstanding salvage rates of up to 81% at 24 months in 82 patients, and Farber et al8 implanted 240 conduits with 199 qualifying as limb salvage procedures and a reported success rate of 81% at 12 months. Overall, in the present study, failure of these grafts was due to a variety of causes, including focal stenosis, neointimal hyperplasia, and aneurysmal degeneration.

The findings of no significant increase in loss of primary patency with differing donor blood types are consistent with that of Walker et al,4 but other studies have found outcomes were affected by ABO incompatibility.3,8 O’Banion et al8 found that ABO incompatibility was a significant risk factor for 30-day graft failure. Zehr et al3 reported that only 6 of 20 noncompatible grafts provided limb salvage. In other studies, centers seromatched donor veins to recipients, but did not find a difference between different donor blood type or Rhesus immunization and patency.2,5-7 Postoperative factors may have influenced the clinical outcomes in this study. A possible confounding variable regarding our limb salvage data is that, during the study period, our group hired a full-time wound care and management nurse responsible for rounding on these patients and seeing them in clinic. This dedicated wound care professional may have had an effect on the relatively high rate of limb salvage. Additionally, limb salvage is likely reflective of ultimate patency, and the present analysis did not find a consistent correlation between WIT and primary assisted and secondary patency. Limb salvage is more reflective of ultimate patency and explains why WIT did not ultimately correlate with limb salvage. Tables I and II report postoperative antiplatelet and anticoagulation management, which did not follow a formal, standardized pathway. Dual antiplatelet therapy was used in significantly more grafts that lost patency (Table II). This may reflect that patients in whom these procedures were performed were deemed to be high risk and were more aggressively treated. Cryopreserved femoral arteries are much stronger tissue than saphenous vein, and the improved patency of arterial conduits compared with venous conduits in cardiac surgery is well-known.11 Few studies have reported outcomes with cryopreserved arterial conduits and these found a high rate of degeneration and loss of

Journal of Vascular Surgery Volume

-,

Number

Cullen et al

7

-

Table III. Unadjusted baseline patient and operative characteristics among limb salvage operations Limb loss (n ¼ 8)

Limb salvage (n ¼ 28)

P value

56.9 6 10.4

65.3 6 8.5

.04

Patient characteristics Age, years Male sex

6 (75.0)

19 (67.9)

.70

Hyperlipidemia

3 (37.5)

18 (64.3)

.17

Hypertension

7 (87.5)

17 (60.7)

.16

Coronary artery disease

4 (50.0)

19 (67.9)

.35 .72

Chronic kidney disease

1 (12.5)

5 (17.9)

Diabetes mellitus

3 (37.5)

12 (42.9)

.79

Tobacco abuse

7 (87.5)

17 (60.7)

.16

Procedure Iliofemoral

0 (0)

3 (10.7)

Femorofemoral

0 (0)

2 (7.1)

Femoropopliteal

1 (12.5)

3 (10.7)

Femoral-distal

6 (75.0)

17 (60.7)

Popliteal-distal

1 (12.5)

3 (10.7)

Infrageniculate

7 (87.5)

22 (78.6)

1059.6 6 361.4

897.6 6 376.5

.16

4.9 6 0.4

5.6 6 1.6

.24

Cryopreserved artery

1 (12.5)

9 (32.1)

.27

Donor A

5 (71.4)

6 (25.0)

Donor B

0 (0)

5 (20.8)

Donor AB

0 (0)

2 (8.3)

Donor O

2 (28.6)

11 (45.8)

ABO seromatch

5 (62.5)

14 (50.0)

.53

Rhesus positive

6 (85.7)

18 (75.0)

.55

Rhesus seromatch

6 (85.7)

17 (70.8)

.43

Aspirin

7 (87.5)

23 (82.1)

.72

Dual antiplatelet

4 (50.0)

11 (39.3)

.59

Anticoagulation

1 (12.5)

6 (21.43)

.57

Donor characteristics WIT, minutes Diameter, mm

Postoperative care

WIT, Warm ischemia time. Values are number (%) or mean 6 standard deviation.

patency.12,13 In the current study, cryopreserved arterial conduits were not predictive of patency. Arterial conduits had a higher rate of primary patency at 1 year in the present study, although not statistically significant (Fig 1, B). However, arterial conduits were also used in more inflow procedures and, also, had shorter WITs. WIT is a known variable affecting outcomes in organ transplantation and allograft cardiac valves.9,14-16 St. Louis et al15 published a report in which they assessed injury to allograft cardiac valves and found that cellular adenosine triphosphate was depleted after just 2 hours of hypoxia.15 Another study used flow cytometry to assess cell viability in allograft valves and found a significant negative correlation between WIT and cell viability.16 Currently, many techniques are used in clinical practice to condition organs that have been subject to extended ischemia times to improve

outcomes.17 However, femoral arteries and saphenous veins are harvested after the organs, and thus exposed to longer WIT. Loss of the mechanical force of blood flow through the vasculature in transplanted organs has been shown to result in a propagation of atypical cellular responses leading to the production of damaging free radicals.18,19 Furthermore, biomechanical studies have shown that ischemia time causes derangements in the cellular ultrastructure of allograft valves, and reperfusion injury may also lead to apoptosis of endothelial cells.20,21 However, this finding has not been previously reported in the vascular surgery literature and is unique to the current analysis. Future studies should continue to evaluate the impact of ischemia time on patency to determine if this variable will continue to have a role in patency, as well as the underlying mechanisms.

8

Journal of Vascular Surgery

Cullen et al

---

2018

Table IV. Likelihood estimates for loss of primary patency and amputation HR

95% CI

P value

Loss of primary patency WIT, hours

1.11

1.02-1.21

.02

Age, years

0.95

0.91-0.99

.01

Chronic kidney disease

5.13

1.10-23.94

.04

Cryopreserved artery

0.65

0.13-3.22

.60

Diabetes mellitus

0.98

0.37-2.60

.97

Diameter, mm Inflow e Femoral vs aorta

1.13 0.05

0.75-1.68

.57

0.001-1.78

.10

Inflow e Iliac vs aorta

8.92

0.29-272.76

.21

Inflow e Popliteal vs aorta

0.45

0.01-20.68

.68

Outflow e Distal vs femoral

49.57

1.56e1577.21

.03

Outflow e Popliteal vs femoral

87.40

Tobacco abuse

2.66-2870.12

.01

1.60

0.52-5.00

.42

1.00

0.998-1.003

.60 .51

Amputation WIT, hours ABO compatibility

1.87

0.29-12.32

Age, years

0.94

0.86-1.02

.15

Chronic kidney disease

2.82

0.19-42.81

.45

Diabetes mellitus

0.75

0.12-4.67

.76

Tobacco abuse

3.35

0.24-45.84

.37

Diameter, mm

0.66

0.22-1.95

.45

CI, Confidence interval; HR, hazard ratio; WIT, warm ischemia time.

It is recognized by the authors that there are many other known variables that far outweigh the effect WIT possibly plays in the patency of cryopreserved bypass conduits. We do not believe that WIT is a primary determinant of patency. Variables such as conduit length, bypass location, inflow/outflow procedure, and inherent patient disease certainly play a greater role. However, we believe that the novel finding of WIT and its effect on primary patency is important because it represents a modifiable variable with the potential to affect outcomes and further investigation is certainly warranted. In addition, if this factor does indeed affect outcomes, it should be tracked and, at minimum, be a variable collected on the Vascular Quality Initiative. Vascular surgeons should be aware of all of this information, and in the event thatdall things being equaldwhen presented an opportunity to use a graft with a shorter WIT, we now do at our institution. Furthermore, these grafts should typically be used as a last resort or in circumstances in which less expensive prosthetic grafts are inappropriate owing to infection. Limitations of the present study include its retrospective nature with the inherent risk of selection bias, as well as the limited number of grafts (n ¼ 65). Owing to the limited number of subjects in this retrospective study, a cut-point analysis was not possible, and we are not able to give a recommendation as to what duration

of WIT of the donor graft is clinically acceptable. The tissue processing organization uses a cut-point of 24, and we believe that this is suitable, but further investigation is warranted to determine if a shorter time may be more clinically acceptable. Further, the reported observations are merely associations between donor characteristics and primary patency; this does not imply a direct cause and effect relationship. At this institution, the surgeon’s back table processing of the conduits is almost uniformly the same, but the decision as to when to use a cryopreserved conduit could be different among the surgeons. Rhesus and ABO status were missing for nine donors grafts. In addition, the conduit diameters reported herein are from the product insert schematic; it is possible that intraoperative trimming could have affected the accuracy of these measurements.

CONCLUSIONS The current results indicate that increased WIT is associated with increased risk-adjusted loss of primary patency for lower extremity bypasses using cryopreserved conduits and reflect the outcomes of a single institution cohort of vascular surgery patients. Further investigation is certainly warranted, but vascular surgeons should be aware that donor WITs may affect outcomes for lower extremity bypasses with cryopreserved donor vessels. Surgeons can

Journal of Vascular Surgery Volume

-,

Number

Cullen et al

9

-

find information regarding WIT on the product schematic, along with vessel length and diameter, when requesting a cryopreserved vessel for an operation. Larger future analyses related to the impact of donor characteristics on cryopreserved conduits are warranted owing to the potential to significantly impact patient outcomes.

8.

9.

AUTHOR CONTRIBUTIONS Conception and design: MC, HM, MT, JK, GU Analysis and interpretation: MC, HM, RH, MT, WR, KC, JK, GU, RR Data collection: MC, VG, RR Writing the article: MC, HM, RH, VG, JK, GU Critical revision of the article: MC, HM, RH, MT, WR, KC, JK, GU, RR Final approval of the article: MC, HM, RH, VG, MT, WR, KC, JK, GU, RR Statistical analysis: MC, HM, RH Obtained funding: Not applicable Overall responsibility: GU

10.

11. 12.

13.

14.

REFERENCES 1. Harris L, O’brien-Irr M, Ricotta JJ. Long-term assessment of cryopreserved vein bypass grafting success. J Vasc Surg 2001;33:528-32. 2. Shah RM, Faggioli GL, Mangione S, Harris LM, Kane J, Taheri SA, et al. Early results with cryopreserved saphenous vein allografts for infrainguinal bypass. J Vasc Surg 1993;18: 965-9. 3. Zehr BP, Niblick CJ, Downey H, Ladowski JS. Limb salvage with CryoVein cadaver saphenous vein allografts used for peripheral arterial bypass: role of blood compatibility. Ann Vasc Surg 2011;25:177-81. 4. Walker PJ, Mitchell RS, McFadden PM, James DR, Mehigan JT. Early experience with cryopreserved saphenous vein allografts as a conduit for complex limb-salvage procedures. J Vasc Surg 1993;18:561-8. 5. Martin RS 3rd, Edwards WH, Mulherin JL Jr, Edwards WH Jr, Jenkins JM, Hoff SJ. Cryopreserved saphenous vein allografts for blow-knee lower extremity revascularization. Ann Surg 1994;219:664-70. 6. Buckley CJ, Abernathy S, Lee SD, Arko FR, Patterson DE, Manning LG. Suggested treatment protocol for improving patency of femoral-infrapopliteal cryopreserved saphenous vein allografts. J Vasc Surg 2000;32:731-8. 7. Farber A, Major K, Wagner WH, Cohen JL, Cossman DV, Lauterbach SR, et al. Cryopreserved saphenous vein

15.

16.

17.

18.

19.

20.

21.

allografts in infrainguinal revascularization: analysis of 240 grafts. J Vasc Surg 2003;38:15-21. O’Banion LA, Wu B, Eichler CM, Reilly LM, Conte MS, Hiramoto JS. Cryopreserved saphenous vein as a last-ditch conduit for limb salvage. J Vasc Surg 2017;66:844-9. Aubert O, Kamar N, Vernerey D, Viglietti D, Martinez F, Duong-Van-Huyen JP, et al. Long term outcomes of transplantation using kidneys from expanded criteria donors: prospective, population based cohort study. BMJ 2015;351: h3557. Rutherford RB, Baker JD, Ernst C, Johnston KW, Porter JM, Ahn S, et al. Recommended standards for reports dealing with lower extremity ischemia: Revised edition. J Vasc Surg 1997;26:517-38. Taggart DP. Current status of arterial grafts for coronary artery bypass grafting. Ann Cardiothorac Surg 2013;2:427-30. Touma J, Kashi M, Oubaya N, Cochennec F, Allaire E, Marzelle J, et al. Results of cryopreserved arterial allografts to treat infections of peripheral bypasses. Ann Vasc Surg 2017;38:e2. Schiava N, Mathevet JL, Boudjelit T, Ariscot M, Feugier P, Lermusiaux P, et al. Cryopreserved arterial allografts and ABO and Rhesus compatibility. Ann Vasc Surg 2016;33: 173-80. Halazun KJ, Al-Mukhtar A, Aldouri A, Willis S, Ahmad N. Warm ischemia in transplantation: search for a consensus definition. Transplant Proc 2007;39:1329-31. St Louis J, Corcoran P, Rajan S, Conte J, Wolfinbarger L, Hu J, et al. Effects of warm ischemia following harvesting of allograft cardiac valves. Eur J Cardiothorac Surg 1991;5:458-64. Niwaya K, Sakaguchi H, Kawachi K, Kitamura S. Effect of warm ischemia and cryopreservation on cell viability of human allograft valves. Ann Thorac Surg 1995;60:S114-7. Hsin MK, Iskender I, Nakajima D, Chen M, Kim H, dos Santos PR, et al. Extension of donor lung preservation with hypothermic storage after normothermic ex vivo lung perfusion. J Heart Lung Transplant 2016;35:130-6. de Perrot M, Liu M, Waddell TK, Keshavjee S. Ischemiareperfusion-induced lung injury. Am J Respir Crit Care Med 2003;167:490-511. Inoue R, Jensen LJ, Shi J, Morita H, Nishida M, Honda A, et al. Transient receptor potential channels in cardiovascular function and disease. Circ Res 2006;99:119-31. Messier RH Jr, Bass BL, Domkowski PW, Hopkins RA. Interstitial cellular and matrix restoration of cardiac valves after cryopreservation. J Thorac Cardiovasc Surg 1999;118:36-49. Hilbert SL, Luna RE, Zhang J, Wang Y, Hopkins RA, Yu ZX, et al. Allograft heart valves: the role of apoptosis-mediated cell loss. J Thorac Cardiovasc Surg 1999;117:454-62.

Submitted Nov 30, 2017; accepted Apr 27, 2018.

DISCUSSION Dr John F. Eidt (Dallas, Tex). I would like to commend you on a nice presentation and for sending the manuscript several weeks in advance of the meeting. The authors performed a retrospective single institution review of the performance of cryopreserved vessels, specifically femoral artery, femoral vein, and saphenous vein, in the management of lower extremity vascular

disease. They were able to identify 60 patients in whom 65 cryopreserved vessels were used. Grafts were used in a variety of configurations including axial-femoral, ilio-femoral, femoral-femoral, femoral-popliteal, femoraldistal, and popliteal-distal. The indication for surgery was predominantly critical limb-threatening ischemia, but also included a few claudicants, aneurysms, and graft

10

Cullen et al

Journal of Vascular Surgery ---

infections. Primary 12-month patency was 26%, which is typical of many other publications and is certainly consistent with my personal experience. Limb salvage was 75%, which is not surprising in that limb salvage is invariably superior to graft patency. There were no patient-specific factors that were associated with the loss of primary patency. In addition, ABO and Rh seromatch was not predictive of graft failure. There was a trend favoring arterial conduits in preference to venous conduits, but small numbers precluded a statistically significant conclusion. But of most interest, each increase in 60 minutes of warm ischemia time, was associated with a decrease in patency. The limitations of this study are apparent and acknowledged in the article: heterogeneity of indications, anatomic configurations, and relatively small numbers eliminates the possibility of achieving sweeping conclusions. This is a hypothesis-generating study and raises more questions than it answers. But given those caveats it certainly highlights important potential issues related to the use of allogeneic vascular grafts. I have long been concerned that we know really very little about how the tissue banks, harvest, prepare, store, and deliver allogeneic blood vessels. Frankly, I was not aware that the warm ischemia time was a factor that was available to me as a consumer. Several years ago, we looked at the preservation of endothelial function as measured by the presence of thrombomodulin on the surface in saphenous vein segments harvested for coronary bypass from the same patient. We were somewhat surprised to find that even after relatively short periods of time there was a significant decrease in the presence of measurable endothelial function, regardless of the storage method. So my first question relates to the biological function of the cryopreserved arterial and venous conduits. Are you aware of any data assessing the functional activity of these conduits? Second, please make a comment regarding the lack of association between blood type seromatching and performance. It just seems counterintuitive.

2018

Finally, should we be refusing to use conduits with excessive warm ischemia time? And if so, what is the cutoff point that we should be requesting? Again, we appreciate you and your colleagues for choosing to present these data at the Southern Vascular. Dr J. Michael Cullen. Thank you, Dr. Eidt. There have been some studies conducted in cryopreserved cardiac valves, which have shown prolonged ischemia times, and even the preservation process itself, affects the functional activity and quality of the valve. I am unaware of similar published literature for cryopreserved cadaveric vessel conduits. To address the way that these vessels are curated by the tissue processing agents and a cutoff time, there is a proprietary process. The individuals who do process these vessels undergo a substantial amount of training, and they do everything by internal standards that have been determined by the tissue processing organizations, in our case LifeNet Health. Those operating procedure manualsdI do not believe that they are available for us to seeddictate how these vessels are curated and how they are modified when they are explanted. The tissue processing agencies have cutoff ischemia times, and there are some stipulations to this, but generally it is 24 hours. After 24 hours, they will not explant the vessels and then preserve them for bypasses. Now, I briefly mentioned this in the presentation, for an individual who is having their organs removed and will be transplanted into a recipient, the warm ischemia time for those individuals is typically around or less than 8 hours. For individuals that have died in the field after an accident and are transported directly to a funeral home, agents from the tissue processing organizations will go to the funeral home to process the tissue, and these vessels are typically associated with longer ischemic times. We cannot determine a cutoff time point from our study. There may potentially be one, but larger analyses will be required. Likewise, I think blood type seromatching could play a role in performance of these grafts, and has been shown in some studies, but larger numbers will likely be necessary to tease out its true role.