Outcomes of open and endovascular lower extremity revascularization in active smokers with advanced peripheral arterial disease

From the Western Vascular Society

Outcomes of open and endovascular lower extremity revascularization in active smokers with advanced peripheral arterial disease Samuel L. Chen, MD, Matthew D. Whealon, MD, Nii-Kabu Kabutey, MD, Isabella J. Kuo, MD, Michael D. Sgroi, MD, and Roy M. Fujitani, MD, Orange, Calif

ABSTRACT Objective: Concern over perioperative and long-term durability of lower extremity revascularizations among active smokers is a frequent deterrent for vascular surgeons to perform elective lower extremity revascularization. In this study, we examined perioperative outcomes of lower extremity endovascular (LEE) revascularization and open lower extremity bypass (LEB) in active smokers with intermittent claudication (IC) and critical limb ischemia (CLI). Methods: Active smokers undergoing LEE or LEB from 2011 to 2014 were identified in the American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) targeted vascular data set. Patient demographics, comorbidities, anatomic features, and perioperative outcomes were compared between LEE and LEB procedures. Subgroup analysis was performed for patients undergoing revascularization for IC and CLI independently. Results: From 2011 to 2014, 4706 lower extremity revascularizations were performed in active smokers (37% of all revascularizations). In this group, 1497 were LEE (55.6% for CLI, 13.4% for below-knee pathology) and 3209 were LEB (68.9% CLI, 34.7% below-knee). Patients undergoing LEE had higher rates of female gender, hypertension, end-stage renal disease, and diabetes (all P # .02). LEE patients also had a higher frequency of prior percutaneous interventions (22.7% vs 17.2%; P < .01) and preoperative antiplatelet therapy (82.3% vs 78.7%; P ¼ .02). On risk-adjusted multivariate analysis, LEE patients had higher need for reintervention on the treated arterial segment than LEB (5.1% vs 5.2%; odds ratio [OR], 1.52; 95% confidence interval [CI], 1.08-2.13; P ¼ .02) but had lower wound complications (3.1% vs 13.2%; OR, 0.32; 95% CI, 0.23-0.45; P < .01) and no statistically significant difference in 30-day mortality (0.6% vs 0.9%), myocardial infarction or stroke (1.1% vs 2.6%), or major amputation (3.2% vs 2.1%) in the overall cohort of active smokers. In the IC subgroup, myocardial infarction or stroke was significantly higher in the LEB group (1.9% vs 0.6%; OR, 1.83; 95% CI, 1.17-1.97; P ¼ .03), although no difference was found in the CLI subgroup (2.8% vs 1.4%; OR, 0.75; 95% CI, 0.37-1.52; P ¼ .42,). Also in IC group, there was a trend for lower major amputation rates #30 days in the LEE group, whereas in the CLI group, LEE had a trend toward higher risk of early amputation compared with LEB. Conclusions: In active smokers, LEB for IC and CLI requires fewer reinterventions but is associated with a higher rate of postoperative wound complications compared with LEE revascularization. However, the risk for limb amputation is higher in actively smoking patients when treated by LEE compared with LEB for CLI. Importantly, cardiovascular complications are significantly higher in actively smoking patients with IC undergoing LEB compared with LEE. This additional cardiovascular risk should be carefully weighed when proposing LEB for actively smoking patients with nonlimb-threatening IC. (J Vasc Surg 2017;65:1680-9.)

Peripheral arterial disease (PAD) is an affliction affecting nearly 200 million individuals worldwide, ranging from asymptomatic disease to severe limb-threatening From the Division of Vascular and Endovascular Surgery, Department of Surgery, University of California, Irvine Medical Center. Author conflict of interest: none. Presented at the Thirty-first Annual Meeting of the Western Vascular Society, Colorado Springs, Colo, September 24-27, 2016. Correspondence: Roy M. Fujitani, MD, Division of Vascular and Endovascular Surgery, Department of Surgery, University of California, Irvine Medical Center, 333 The City Blvd West, Ste 1600, Orange, CA (e-mail: [email protected]). 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 2017 Published by Elsevier Inc. on behalf of the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvs.2017.01.025

1680

ischemia.1 Smoking has been clearly implicated as one of the most significant modifiable risk factors contributing to both the development of PAD and progression of disease once diagnosed.2-6 Historical observational data have shown that continued smoking is associated with higher rates of amputation, death, and myocardial infarction (MI) in patients with PAD compared with those who quit.7 The Society for Vascular Surgery practice guidelines for atherosclerotic occlusive disease of the lower extremities include smoking cessation as a Grade IA recommendation in the management of asymptomatic disease and intermittent claudication (IC).8 Increasing evidence shows that continued smoking leads to worse outcomes after various types of surgical intervention.9 In vascular surgery, smoking has been associated with early graft failure after infrainguinal

Journal of Vascular Surgery

Chen et al

1681

Volume 65, Number 6

bypass surgery.10-13 In addition, smoking cessation is associated with decreased mortality and improved amputation-free survival among patients with symptomatic PAD.14 Because traditional open surgery and more complex endovascular options are available for treating symptomatic PAD, great variation exists in practice patterns among vascular specialists who perform these treatments for patients with IC and critical limb ischemia (CLI).15 Specifically in smokers, practitioners must weigh the effect of continued smoking on outcomes after peripheral arterial interventions and decide when to offer open or endovascular revascularization to patients with nonlimb-threatening ischemia. In this regard, there are limited contemporary data regarding the comparative outcomes among current smokers undergoing open lower extremity bypass (LEB) and lower extremity endovascular (LEE) revascularization for advanced peripheral arterial disease. In this study, we sought to evaluate the perioperative outcomes of LEE revascularization and LEB in a nationwide contemporary cohort of current smokers with IC and CLI.

METHODS Data source. We retrospectively analyzed the American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) Targeted Open Lower Extremity Revascularization and Lower Extremity Endovascular Revascularization Participant Use Data Files (PUFs) consisting of all qualifying procedures performed from January 1, 2011, to December 31, 2014, inclusively. The ACS NSQIP is a nationally validated, risk-adjusted, outcomes-based program to measure and improve the quality of surgical care in the United States. The data in the general and procedurally targeted databases are collected and entered by surgical clinical reviewers who are certified by the ACS. Strict variable definitions are used when data are collected to ensure consistency across participating centers, and periodic auditing is used to ensure accuracy.16 The procedurally targeted databases collect procedurespecific demographics, anatomic, and perioperative details, and 30-day postoperative outcomes data, which first became available in 2011. We then merged the targeted databases with the general PUF by the deidentified Case ID, allowing for both procedure-specific and general variables and outcomes collected in NSQIP to be analyzed for all cases. The NSQIP database is exempt from requiring informed consent from individual patients and does not require Institutional Review Board approval for analysis. Cohort and variables. Patients who underwent LEB or LEE revascularization between 2011 and 2014 were identified from the NSQIP targeted database. The patients in this surgical group who were reported to have been

ARTICLE HIGHLIGHTS d

d

d

Type of Research: Retrospective analysis of prospectively collected American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) data Take Home Message: Analysis of 30-day outcome in 4706 active smokers undergoing lower extremity revascularization revealed that open surgery resulted in lower rate of early amputation but more wound infection and a higher rate of cardiovascular complications than endovascular repair. Recommendation: The authors suggest smoking cessation and avoiding open surgical bypass in claudicant patients who smoke.

smokers within the past year, a defined variable in the general NSQIP PUF, were selected for our study group. Preoperative variables analyzed were age, gender, race, and comorbid conditions, including chronic obstructive pulmonary disease, congestive heart failure, hypertension requiring medication, diabetes mellitus, end-stage renal disease, chronic steroid use, obesity, dependent functional status, and high-risk physiology, defined as a recent adverse cardiac event within the past 6 months. Preoperative use of antiplatelet therapy, statins, and b-blockers was evaluated. Data regarding preoperative ankle-brachial indices were also identified, as were whether the procedure performed was deemed an emergency and whether the indication for the procedure was claudication, rest pain, or tissue loss. The latter two comprised the definition of the CLI cohort. Perioperative variables analyzed were type of anesthesia used, below-knee vs above-knee target, bleeding requiring intraoperative or postoperative transfusion, and operative time. A below-knee target was defined as tibial angioplasty or stenting for the LEE group and bypass distal to the popliteal artery for the LEB group. Postoperative outcomes examined were 30-day mortality and postoperative occurrence of MI, stroke, and wound complications, major reintervention on the treated arterial segment, major amputation, venous thromboembolism, frequency of discharge to home vs another facility, and unplanned readmission. Hospital length of stay was also tabulated. Statistical analysis. Univariate and multivariate analyses were performed comparing current smokers who underwent LEE with those who underwent LEB. Demographics, perioperative variables, and postoperative outcomes defined in the previous section were analyzed. Categoric variables were compared by c2 test, and continuous variables were compared by the Student t-test. A priori adjustments for age, gender, race, body mass index, chronic obstructive pulmonary disease,

1682

Chen et al

Journal of Vascular Surgery June 2017

congestive heart failure, diabetes mellitus, hypertension, preoperative end-stage renal disease, preoperative functional status, preoperative use of antiplatelet therapy, recent adverse cardiac events (#6 months), emergency procedures, indication for procedure, and whether the intervention was performed for an above-knee vs below-knee arterial target, were entered into a multivariable logistic regression model to estimate adjusted odds ratios (ORs) with 95% confidence intervals (CIs) for the following primary outcomes: 30-day mortality, major amputation, major reintervention, wound complication, MI or stroke, hospital readmission #30 days, and overall hospital length of stay. A subgroup analysis also compared LEE vs LEB in smokers whose procedure was performed exclusively for IC and those performed for CLI. Univariate and multivariate analyses were performed for these comparisons as well. Statistical significance was defined as P < .05, and the Holm method was used to adjust P values for multiple comparisons. The statistical analysis was performed with SPSS 23.0 software (IBM Corp, Armonk, NY).

RESULTS Between January 1, 2011, and December 31, 2014, there were 12,655 cases of LEE revascularization and LEB identified in the ACS-NSQIP procedurally targeted databases. Of these patients, 4706 (37.2%) were current smokers, which comprised the study cohort. The cohort was a mean age of 62.6 years, with 65.8% male and 69.1% white. Of all procedures done in current smokers, 35.3% (n ¼ 1661) were performed for the indication of IC, and 27.9% (n ¼ 1313) were performed for below-knee target vessels. Overall 30-day mortality was 0.8%. In this study cohort, 1497 procedures (31.8%) were LEE and 3209 (68.2%) were LEB. When LEE revascularization was compared with LEB, age, race, rates of chronic obstructive pulmonary disease, congestive heart failure, obesity, chronic steroid use, preoperative dependent functional status, and frequency of recent adverse cardiac events were similar between the two groups. Patients in the LEE group were less frequently male (62.1% vs 67.5%; P < .01), and had higher rates of hypertension requiring medication, diabetes mellitus, preoperative end-stage renal disease, and preoperative use of antiplatelet therapy (all P < .03). There was no difference in the incidence of the procedure being done emergently vs electively between the two groups (4.7% vs 5.9%; P ¼ .11). IC was the indication for 44.4% of the LEE procedures but for only 31.1% of the LEB procedures. The remaining 68.9% of LEB were performed for patients with CLI (Table I). The procedure was performed under general anesthesia for 30.3% of the LEE patients and for 95.7% of the LEB patients. Mean operative time was 111.9 minutes for the LEE revascularization and was

238.8 minutes for LEB procedures. LEE procedures required fewer intraoperative or postoperative transfusions compared with LEB (7.5% vs 21.0%; P < .01). In addition, LEE revascularization was much less frequently performed for below-knee targets than LEB (13.4% vs 34.7%; P < .01). In the unadjusted outcomes analysis, 30-day mortality remained low in both the LEE and LEB groups (0.6% and 0.9%, respectively). However, the incidence of MI was higher in the LEB group (2.1% vs 0.9%; P ¼ .01). Wound complications were significantly lower in the LEE group (3.1% vs 13.2%; P < .01), but the progression to major amputation #30 days was higher in the LEE group (3.2% vs 2.1%; P < .01). The reintervention rate on the treated arterial segment was similar between the LEE and LEB groups (5.1% vs 5.2%; P ¼ .90). The venous thromboembolism rate was higher in the LEB group (0.9% vs 0.3%; P ¼ .02) as was rate of unplanned readmission (10.3% vs 5.7%; P < .01). Mean hospital length of stay was also longer in the LEB group (7.3 6 7.8 days vs 3.6 6 7.1 days; P < .01; Table II). Multivariate analyses, adjusted for age, sex, race, comorbid conditions, indication, and above-knee vs belowknee intervention, showed LEE revascularization had higher rates of reintervention on the treated arterial segment (OR, 1.52; 95% CI, 1.08-2.13; P ¼ .02) and lower wound complications (OR, 0.32; 95% CI, 0.23-0.45; P < .01), with a significantly shorter mean hospital length of stay (mean difference, 1.34 days; P < .01) compared with LEB. There was no statistically significant difference between LEE and LEB for mortality (OR, 0.44; P ¼ .13), major amputation (OR, 1.54; P ¼ .09), MI (OR, 0.649; P ¼ .21), or risk for unplanned readmission (OR, 0.795; P ¼ .12; Table III). IC subgroup. The 1661 patients undergoing LEE revascularization in the IC subgroup had higher rates of hypertension requiring medication, diabetes mellitus, and preoperative antiplatelet use, and were less frequently male (all P < .01). General anesthesia was used much more frequently in the LEB group (95.3% vs 23.6%; P < .01). While 85.9% of all LEB procedures were performed for above-knee arterial targets, LEE procedures were still more frequently done for above-knee targets (92.8% vs 81.2%; P < .01; Table IV). The LEE and LEB groups both had an extremely low 30-day mortality rate (0.2% and 0.7%, respectively), but MI occurred more frequently in the LEB group (1.0% vs 0.2%; P ¼ .04). Major reintervention on the treated arterial segment and major amputation were both similarly low among the LEE and LEB groups. However, wound complications were again higher in the LEB group than in the LEE revascularization group (9.6% vs 2.0%; P < .01; Table V).

Journal of Vascular Surgery

Chen et al

1683

Volume 65, Number 6

Table I. Demographics and preoperative details of lower extremity revascularization in smokers Lower extremity revascularization All (N ¼ 4706)

Variablea

Endo (n ¼ 1497)

Open bypass (n ¼ 3209)

P value

62.6 6 9.7

62.4 6 9.7

62.7 6 9.7

.328

<60

37.9 (1783)

39.8 (596)

37.0 (1187)

.216

60-69

38.6 (1817)

37.3 (558)

39.2 (1259)

70-79

19.1 (900)

18.0 (270)

19.6 (630)

Age, years

80-89

4.2 (198)

4.7 (70)

4.0 (128)

>90

0.2 (8)

0.2 (3)

0.2 (5) <.001

Gender Male

65.8 (3097)

62.1 (930)

67.5 (2167)

Female

34.2 (1609)

37.9 (567)

32.5 (1042)

White

69.1 (3252)

69.5 (1040)

68.9 (2212)

African American

21.0 (989)

21.9 (328)

20.6 (661)

Race

.191

Hispanic

3.8 (181)

5.9 (88)

2.9 (93)

Asian

0.7 (31)

0.5 (7)

0.7 (24)

Other

0.4 (19)

0.2 (3)

0.5 (16)

Unknown

5.0 (234)

2.1 (31)

6.3 (203)

Comorbidities Smoking COPD Congestive heart failure Hypertension End-stage renal disease Steroid use Diabetes mellitus Dependent functional status

.

.

.

16.3 (767)

15.6 (233)

16.6 (534)

.352

2.2 (105)

2.1 (31)

2.3 (74)

.611

78.5 (3695)

80.8 (1209)

77.5 (2486)

.010

4.2 (196)

5.1 (77)

3.7 (119)

.022

3.1 (147)

3.6 (54)

2.9 (93)

.193

39.4 (1856)

45.1 (675)

36.8 (1181)

<.001

5.0 (234)

5.7 (86)

4.6 (148)

.096

BMI, kg/m

27.2 6 6.0

27.3 6 6.0

27.1 6 6.0

.274

Obese (BMI >30 kg/m2)

27.4 (1291)

29.0 (434)

27.0 (865)

.730

HRP (recent cardiac event)

12.4 (583)

12.8 (192)

12.2 (391)

.559

Prior percutaneous intervention

19.0 (893)

22.7 (340)

17.2 (553)

<.001

Prior bypass

21.6 (1015)

19.8 (297)

22.4 (718)

.049

Antiplatelet

79.8 (3757)

82.3 (1232)

78.7 (2525)

.016

b-Blocker

53.5 (2520)

54.8 (821)

52.9 (1699)

.130

Statin

66.1 (3113)

66.8 (1000)

65.8 (2113)

.786

2

Preoperative status

Medication use

Emergency procedure

5.5 (260)

4.7 (71)

5.9 (189)

.109 <.001

Indication IC

35.3 (1661)

44.4 (664)

31.1 (997)

CLI rest pain

32.7 (1540)

25.7 (385)

36.0 (1155)

CLI tissue loss

32.0 (1505)

29.9 (448)

32.9 (1057)

BMI, Body mass index, CLI, critical limb ischemia; COPD, chronic obstructive pulmonary disease; HRP, high-risk physiology; IC, intermittent claudication. a Categoric data are shown as percentage (number) and continuous data as mean 6 standard deviation.

On multivariate analysis, major reintervention on the treated arterial segment remained higher in the LEE group (OR, 2.03; 95% CI, 1.01-4.09; P ¼ .04). But the LEE group had a lower risk of wound complications (OR, 0.23; 95% CI, 0.11-0.46; P < .01). In addition, multivariate

analysis showed MI or stroke was significantly lower in the LEE revascularization group than in the LEB group (OR, 0.17; 95% CI, 0.03-0.83; P ¼ .03). There was no difference in risk of mortality, progression to early amputation, or hospital readmission #30 days (Table III).

1684

Journal of Vascular Surgery

Chen et al

June 2017

Table II. Operative details and outcomes of lower extremity revascularization in smokers Lower extremity revascularization Variablea

All (n ¼ 4706)

Endo (n ¼ 1497)

Open bypass (n ¼ 3209)

74.9 (3523)

30.3 (453)

95.7 (3070)

P value <.001

Anesthesia type General Spinal/epidural

2.6 (122)

Regional/local/MAC/sedation

22.1 (1042)

Other/unknown

0.4 (19)

0.5 (7)

3.6 (115)

68.2 (1021)

0.6 (21)

1.1 (16)

0.1 (3)

Operative details 198.4 6 120.6

Operative time, minutes Bleeding requiring transfusion

16.8 (791)

Below-knee target

27.9 (1313)

Above-knee target

72.1 (3393)

111.9 6 79.2 7.7 (116) 13.4 (200) 86.6 (1297)

238.8 6 115.3

<.001

21.0 (675)

<.001

34.7 (1113)

<.001

65.3 (2096)

<.001

Postoperative outcomes 30-day mortality MI Stroke

0.8 (38)

0.6 (9)

1.7 (80)

0.9 (14)

2.1 (66)

.006

0.3 (5)

0.5 (16)

.430

0.4 (21)

0.9 (29)

MI or stroke

2.1 (98)

1.1 (16)

2.6 (82)

Wound complications

10 (470)

3.1 (46)

13.2 (424)

Major reintervention

5.2 (245)

5.1 (77)

5.2 (168)

Major amputation

2.5 (116)

3.2 (48)

2.1 (68)

VTE (DVT or PE)

0.7 (32)

Discharge to home

0.3 (4)

82.0 (3860)

0.9 (28)

.280

.001 <.001 .895 <.001 .019

89.9 (1346)

78.3 (2514)

<.001

10.3 (332)

<.001

7.3 6 7.8

<.001

Unplanned readmission

8.9 (417)

5.7 (85)

Hospital length of stay, days

6.1 6 7.8

3.6 6 7.1

DVT, Deep venous thrombosis; MAC, monitored anesthesia care; MI, myocardial infarction; PE, pulmonary embolism; VTE, venous thromboembolism. a Categoric data are shown as percentage (number) and continuous data as mean 6 standard deviation.

Table III. Risk-adjusted outcomes of lower extremity endovascular (LEE) revascularization vs open bypass (LEB) in active smokers Outcome 30-day mortality

OR

95% CI

P value .133

0.444

0.154-1.282

IC

.

.

.

CLI

0.574

0.183-1.803

.342

1.542

0.940-2.528

.086

IC

0.657

0.157-2.750

.565

CLI

1.583

0.931-2.691

.090

Major amputation

Major reintervention IC CLI Wound infection or complication

1.518

1.080-2.134

.016

2.034

1.012-4.087

.046

1.375

0.931-2.031

.110

0.318

0.225-0.450

<.001

IC

0.226

0.111-0.459

<.001

CLI

0.364

0.243-0.543

<.001

0.569

0.302-1.073

.081

IC

0.166

0.033-0.829

.029

CLI

0.748

0.369-1.519

.422

MI or stroke

CI, Confidence interval; CLI, critical limb ischemia; IC, intermittent claudication; MI, myocardial infarction; OR, odds ratio.

CLI subgroup. In the subgroup of current smokers, 3045 patients underwent procedures for the indications of rest pain or tissue loss. The demographic makeup of patients undergoing the LEE revascularization and LEB was similar except for male gender (60.4% and 66.7%, respectively), diabetes mellitus (52.5% vs 40.5%; P < .01), and rate of preoperative end-stage renal disease (8.6% vs 5.1%; P < .01). The preoperative pharmacotherapy profiles were similar, as were the rates of preoperative dependent functional status. There was no difference in the emergent nature by which the procedures were performed. Below-knee targets were much more common in those undergoing LEB in this subgroup (41.9% vs 18.2%; P < .01). The use of general anesthesia remained higher in the LEB group compared with the LEE revascularization group (95.8% vs 35.5%; P < .01; Table VI). With regard to unadjusted outcomes, 30-day mortality, need for reintervention on the treated arterial segment, and rate of venous thromboembolism were similar between the two groups. The rate of MI or stroke, however, was higher in the LEB group (2.8% vs 1.4%; P ¼ .03), as was the rate of wound complications (14.8% vs 4.0%; P < .01). The need for unplanned readmissions was also higher in

Journal of Vascular Surgery

Chen et al

1685

Volume 65, Number 6

Table IV. Demographics and preoperative details of lower extremity revascularization in smokers with intermittent claudication (IC) Lower extremity revascularization a

Variable

Age, years

All (N ¼ 1661)

Endo (n ¼ 664)

Open bypass (n ¼ 997)

P value

62.4 6 9.6

62.4 6 9.4

62.4 6 8.9

.115 .005

<60

36.5 (607)

38.0 (607)

35.6 (355)

60-69

42.1 (699)

40.8 (271)

42.9 (428)

70-79

18. (308)

16.7 (111)

19.8 (197)

80-89

2.8 (46)

4.4 (29)

>90

0.1 (1)

0.2 (1)

1.7 (17) 0.0 (0)

Gender

.033

Male

67.3 (1118)

64.3 (427)

69.3 (691)

Female

32.7 (543)

35.7 (237)

30.7 (306)

Race

.262

White

74.0 (1229)

72.4 (481)

75.0 (748)

African American

17.0 (283)

19.4 (129)

15.4 (154)

Hispanic

2.8 (47)

4.1 (27)

2.0 (20)

Asian

0.7 (12)

0.5 (3)

0.9 (9)

Other

0.3 (5)

0.2 (1)

0.4 (4)

Unknown

5.1 (85)

3.5 (23)

6.2 (63)

16.9 (280)

17.6 (117)

16.3 (163)

Comorbidities COPD Congestive heart failure Hypertension

.498

1.4 (24)

1.8 (12)

1.2 (12)

.313

77.9 (1294)

81.8 (543)

75.3 (751)

.002 .710

End-stage renal disease

0.7 (11)

0.8 (5)

0.6 (6)

Steroid use

2.3 (38)

2.7 (18)

2.0 (20)

.347

31.5 (523)

35.8 (238)

28.6 (285)

.002 .293

Diabetes mellitus Dependent functional status

1.6 (26)

2.0 (13)

1.3 (13)

BMI, kg/m2

27.6 6 8.0

27.9 6 5.6

27.5 6 8.9

.429

Obese (BMI >30 kg/m2)

29.9 (496)

32.1 (313)

28.4 (283)

.511

6.3 (63)

.212

HRP (recent cardiac event)

6.8 (113)

7. (50)

Preoperative status Prior percutaneous intervention

18.2 (303)

22.4 (149)

15.4 (154)

<.001

Prior bypass

17.9 (297)

18.7 (124)

17.4 (173)

.491

Antiplatelet

82.5 (1370)

85.5 (568)

80.4 (802)

.022

b-Blocker

50.0 (830)

52.7 (350)

48.1 (480)

.120

Statin

69.5 (1155)

72.1 (479)

67.8 (676)

.103

2.2 (36)

2.3 (15)

2.1 (21)

.834

Medication use

Emergency procedure

BMI, Body mass index, CLI, critical limb ischemia; COPD, chronic obstructive pulmonary disease; HRP, high-risk physiology. a Categoric data are shown as percentage (number) and continuous data as mean 6 standard deviation.

the LEB group (12.3% vs 7.2%; P < .01). Progression to major amputation #30 days was higher in the LEE treatment group (5.3% vs 2.8%; P < .01; Table VII). On multivariate analysis, there was a trend toward increased major amputation in the LEE group, although this was not statistically significant (OR, 1.58; 95% CI, 0.93-2.69; P ¼ .09). The need for major reintervention on the treated arterial segment was similar between the LEE revascularization and LEB groups (OR, 1.38; 95% CI,

0.93-2.03; P ¼ .11). The incidence of MI or stroke (OR, 0.75; 95% CI, 0.37-1.52; P ¼ .42) between the two groups did not approach statistical significance. The LEE revascularization group again had lower risk of wound complications (OR, 0.36; 95% CI, 0.24-0.54; P < .01; Table III). Risk for readmission was similar after adjustment (OR, 0.93; 95% CI, 0.59-1.16; P ¼ .28), and hospital length of stay remained shorter for the LEE revascularization group (mean difference, 1.17 days; P ¼ .02).

1686

Journal of Vascular Surgery

Chen et al

June 2017

Table V. Operative details and outcomes of lower extremity revascularization in smokers with intermittent claudication (IC) Lower extremity revascularization Variablea

All (N ¼ 1661)

Endo (n ¼ 664)

Open bypass (n ¼ 997)

66.6 (1107)

23.6 (157)

95.3 (950)

<.001

Anesthesia type General Spinal/epidural Regional/local/MAC/IV sedation Other/unknown

P value

2.8 (47) 30.1 (500) 0.4 (7)

0.8 (5) 74.7 (496) 0.9 (6)

4.2 (42) 0.4 (4) 0.1 (1)

Operative details Operative time, minutes Bleeding requiring transfusion

163.1 6 94.2

101.8 6 78.1

204.5 6 99.7

<.001

7.2 (120)

3.3 (22)

9.8 (98)

<.001

Below-knee target

14.1 (235)

7.2 (48)

18.8 (187)

<.001

Above-knee target

85.9 (1426)

92.8 (616)

81.2 (810)

<.001

Postoperative outcomes 30-day mortality

0.5 (8)

0.2 (1)

0.7 (7)

.112

MI

0.7 (11)

0.2 (1)

1.0 (10)

.036

Stroke

0.4 (7)

0.3 (2)

0.5 (5)

.537

MI or stroke

1.4 (23)

0.6 (4)

1.9 (19)

.026

Wound complications

6.6 (109)

2.0 (13)

9.6 (96)

<.001 .452

Major reintervention

2.6 (44)

3.0 (20)

2.4 (24)

Major amputation

0.7 (11)

0.6 (4)

0.7 (7)

.798

VTE (DVT or PE)

0.4 (7)

0.0 (0)

0.7 (7)

.030

Discharge to home

94.3 (1565)

97.8 (649)

91.9 (916)

<.001

Unplanned readmission

5.2 (86)

3.8 (25)

6.1 (61)

.034

Hospital length of stay, days

3.2 6 5.4

1.6 6 4.5

4.3 6 5.2

.001

DVT, Deep vein thrombosis; IV, intravenous; MAC, monitored anesthesia care; MI, myocardial infarction; PE, pulmonary embolism; VTE, venous thromboembolism. a Categoric data are shown as percentage (number) and continuous data as mean 6 standard deviation.

DISCUSSION Among active smokers undergoing LEE and LEB captured in the ACS NSQIP procedurally targeted databases, mortality within 30 days is low (0.8%). This did not differ between the two treatment groups, even when categorized by surgical indication. However, patients undergoing LEE were more frequently female and had higher rates of comorbidities, including hypertension, diabetes, end-stage renal disease on dialysis, and preoperative antiplatelet therapy. The LEB group appeared to have more advanced disease, as demonstrated by the higher rate of below-knee pathology and higher frequency of CLI. After adjustment, no differences were found in perioperative mortality, progression to amputation, MI or stroke, or need for readmission between the LEE and LEB populations. There was, however, higher need for reintervention on the treated arterial segment #30 days in the LEE group, whereas wound complications were higher in the LEB group. That the larger incisions needed for open arterial dissection and vein harvesting resulted in statistically significantly higher rates of wound complications in LEB among active smokers is not surprising. Conversely, because endovascular procedures are commonly

performed percutaneously, this likely results in decreased wound complications. The incidence of perioperative cardiovascular complications was similar between the LEE and LEB groups despite general anesthesia being more frequently used for the open procedures. This is consistent with a real-world study that examined smoking status and infrainguinal bypass outcomes and showed <2% risk of perioperative MI among active smokers.12 When performing subgroup analysis of patients by indication, we again found that active smokers undergoing procedures for IC similarly had a higher need for reintervention #30 days in the LEE group, although progression to amputation, mortality, and risk for readmission were similar between the two groups. In the IC patients, however, risk for MI or stroke appeared significantly higher among those undergoing LEB compared with those undergoing LEE, a finding not seen in the analysis of the full cohort. This highlights the independent contribution of the endovascular approach to lower systemic cardiovascular complications among smokers with IC. In patients with CLI, the primary outcomes of mortality, progression to amputation, need for reintervention, MI or stroke, and readmission were not statistically different

Journal of Vascular Surgery

Chen et al

1687

Volume 65, Number 6

Table VI. Demographics and preoperative details of lower extremity revascularization in smokers with critical limb ischemia (CLI) Lower extremity revascularization a

Variable

Age, years

All (N ¼ 3045)

Endo (n ¼ 833)

Open bypass (n ¼ 2212)

P value

62.8 6 10.5

62.5 6 10.0

62.9 6 10.0

.951 .432

<60

38.6 (1176)

41.3 (344)

37.6 (832)

60-69

36.7 (1118)

34.5 (287)

37.6 (831)

70-79

19.4 (592)

19.1 (159)

19.6 (433)

80-89

5.0 (152)

4.9 (41)

5.0 (111)

>90

0.2 (7)

0.2 (2)

0.2 (5)

Gender

.001

Male

35.0 (1066)

39.6 (330)

33.3 (736)

Female

65.0 (1979)

60.4 (503)

66.7 (1476)

Race

.515

White

66.4 (2023)

67.1 (559)

66.2 (1464)

African American

23.2 (706)

23.9 (199)

22.9 (507)

Hispanic

4.4 (134)

7.3 (61)

3.3 (73)

Asian

0.6 (19)

0.5 (4)

0.7 (15)

Other

0.5 (14)

0.2 (2)

0.5 (12)

Unknown

4.9 (149)

1.0 (8)

6.4 (141)

16.0 (487)

13.9 (116)

16.8 (371)

.056 .425

Comorbidities COPD Congestive heart failure Hypertension End-stage renal disease Steroid use Diabetes mellitus Dependent functional status

2.7 (81)

2.3 (19)

2.8 (62)

78.9 (2401)

80.0 (666)

78.4 (1735)

.361

6.1 (185)

8.6 (72)

5.1 (113)

<.001

3.6 (109)

4.3 (36)

3.3 (73)

.176

43.8 (1333)

52.5 (437)

40.5 (896)

<.001 <.001

6.8 (208)

8.8 (73)

6.1 (135)

BMI, kg/m2

26.9 6 6.5

26.9 6 6.2

26.9 6 6.2

.645

Obese (BMI >30 kg/m2)

25.9 (689)

26.5 (221)

25.7 (578)

.840

HRP (recent cardiac event)

15.4 (470)

17.0 (142)

14.8 (328)

.318

Preoperative status Prior percutaneous intervention

19.4 (590)

22.9 (191)

18.0 (399)

.002

Prior bypass

23.6 (718)

20.8 (173)

24.6 (545)

.025 .402

Medication use Antiplatelet

78.4 (2387)

79.7 (664)

77.9 (1723)

b-Blocker

555 (1690)

56.5 (471)

55.1 (1219)

.341

Statin

64.3 (1958)

62.5 (21)

65.0 (1437)

.426

7.6 (168)

.411

Emergency procedure

7.4 (224)

6.7 (56)

BMI, Body mass index, COPD, chronic obstructive pulmonary disease; HRP, high-risk physiology. a Categoric data are shown as percentage (number) and continuous data as mean 6 standard deviation.

between the two groups on adjusted analysis. Interestingly, there was a trend toward higher need for major amputation among smokers undergoing endovascular revascularization in this CLI subgroup. Conversely, there was a trend toward higher risk for major amputation in the LEB group for patients with IC. This may reflect that certain patients selected for endovascular revascularization with limb-threatening ischemia were at an inherently higher risk of requiring major amputation and were thought to be less ideal candidates for open bypass

if there was any expectation of imminent failure. Overall rates of all major complications were higher in both LEE and LEB patients with CLI, likely due to the more acute and ill-nature of these patients. The constant finding among both IC and CLI subgroups and in the analysis of the entire cohort was that wound complications were higher and hospital lengths of stay were consistently longer among those undergoing bypass, although the difference in the hospital length of stay was 1.4 days, the clinical significance of which is

1688

Journal of Vascular Surgery

Chen et al

June 2017

Table VII. Operative details and outcomes of lower extremity revascularization in smokers with critical limb ischemia (CLI) Lower extremity revascularization Variablea

All (N ¼ 3045)

Endo (n ¼ 833)

Open bypass (n ¼ 2212)

P value

79.3 (2416)

35.5 (296)

95.8 (2120)

<.001

Anesthesia type General Spinal/epidural Regional/local/MAC/IV sedation Other/unknown

2.4 (75) 17.8 (542) 0.3 (11)

0.2 (2)

3.3 (73)

63.1 (525)

0.8 (17)

1.2 (10)

0.1 (1)

Operative details Operative time, minutes Bleeding requiring transfusion

217.6 6 135.3 22.0 (671)

120.0 6 79.1

254.3 6 118.5

<.001

11.3 (94)

26.1 (577)

<.001

Below-knee target

35.4 (1078)

18.2 (152)

41.9 (926)

<.001

Above-knee target

64.6 (1967)

81.8 (681)

58.1 (1286)

<.001

1.0 (30)

1.0 (8)

1.0 (22)

.932 .108

Postoperative outcomes 30-day mortality MI

2.3 (69)

1.6 (13)

2.5 (56)

Stroke

0.5 (14)

0.4 (3)

0.5 (11)

.618

MI or stroke

2.5 (75)

1.4 (12)

2.8 (63)

.025

Wound complications

11.9 (361)

4.0 (33)

14.8 (328)

Major reintervention

6.6 (201)

6.8 (57)

6.5 (144)

Major amputation

3.4 (105)

5.3 (44)

2.8 (61)

VTE (DVT or PE)

0.8 (25)

Discharge to home

75.4 (2295)

Unplanned readmission

10.9 (331)

Hospital length of stay, days

7.8 6 8.5

0.5 (4)

0.9 (21)

<.001 .742 <.001 .201

83.7 (697)

72.2 (1598)

<.001

7.2 (60)

12.3 (271)

<.001

5.3 6 8.3

8.7 6 8.4

.345

DVT, Deep vein thrombosis; IV, intravenous; MAC, monitored anesthesia care; MI, myocardial infarction; PE, pulmonary embolism; VTE, venous thromboembolism. a Categoric data are shown as percentage (number) and continuous data as mean 6 standard deviation.

debatable. Interestingly, despite the higher rate of wound complications, readmission and length of stay did not differ significantly after adjustment, although cost-analysis to determine any differences could not be performed. This likely reflects the stringent reporting of many minor wound complications that did not significantly alter the patient’s postoperative hospital course. In addition, a surprisingly high 3.1% wound complication rate was noted for the LEE group, but may have represented the learning curve associated with percutaneous procedures which dated back to 2011, rather than infections, which are more likely to predominate the LEB wound complications. This study has several limitations, starting with its retrospective nature and possible selection bias, because these data are accrued from medical centers and practitioners who electively choose to report data into the NSQIP targeted database. Although this reflects realworld data from academic and community practitioners across the United States with a growing diversity of participation, may still be a source of bias. Because the ACS NSQIP database is inherently a procedural database, we were unable to assess smokers who were not offered revascularization. In our opinion, all modifiable risk

factors, including efforts at smoking cessation, should be addressed before any intervention for nonlimbthreatening lower extremity ischemia is offered. Although patients who underwent reinterventions #30 days were not considered as new patients in the database, we were also unable to specifically identify whether the same patient had multiple procedures within the data set. Another limitation is the lack of lesion-specific characteristics that were treated by LEB or LEE revascularization, such as lesion length, percentage of stenosis, and exact location. The extent of tissue loss or gangrene was similarly unavailable. We stratified by below-knee vs above-knee interventions, which was a variable reported in the database.

CONCLUSIONS In active smokers undergoing lower extremity revascularization in the NSQIP targeted vascular database, perioperative mortality is low. LEE revascularization was more frequently performed in patients who were female, had more comorbidities, and for symptoms of IC. LEB was performed more frequently in patients with CLI and below-knee target vessels. Even when stratified by clinical risk factors and above-knee vs below-knee

Journal of Vascular Surgery

Chen et al

1689

Volume 65, Number 6

pathology, endovascular approaches demonstrated benefits of lower wound complications and shorter hospital lengths of stay. However, patients undergoing LEE required more early reinterventions than those undergoing LEB. Although LEB had similar cardiovascular risk profile to endovascular revascularization among those with CLI, the risk of MI or stroke was higher in the LEB when treating patients with IC. Therefore, therapeutic strategy may be guided more by indication and cardiovascular risk profile in patients who may be candidates for both endovascular revascularization and open bypass.

AUTHOR CONTRIBUTIONS Conception and design: SC, MS, RF Analysis and interpretation: SC, MW, N-KK, IK, MS, RF Data collection: SC, MW Writing the article: SC, RF Critical revision of the article: SC, MW, N-KK, IK, MS, RF Final approval of the article: SC, MW, N-KK, IK, MS, RF Statistical analysis: SC, MW Obtained funding: Not applicable Overall responsibility: RF

REFERENCES 1. Fowkes FG, Rudan D, Rudan I, Aboyans V, Denenberg JO, McDermott MM, et al. Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010:a systematic review and analysis. Lancet 2013;382:1329-40. 2. Cole CW, Hill GB, Farzad E, Bouchard A, Moher D, Rody K, et al. Cigarette smoking and peripheral arterial occlusive disease. Surgery 1993;114:753-7. 3. Criqui MH, Aboyans V. Epidemiology of peripheral artery disease. Circ Res 2015;116:1509-26. 4. Willigendael EM, Teijink JA, Bartelink ML, Kuiken BW, Boiten J, Moll FL, et al. Influence of smoking on incidence and prevalence of peripheral arterial disease. J Vasc Surg 2004;40:1158-65.

5. Agarwal S. The association of active and passive smoking with peripheral arterial disease: results from NHANES 1999-2004. Angiology 2009;60:335-45. 6. Ouriel K. Peripheral arterial disease. Lancet 2001;358:1257-64. 7. Faulker KW, House AK, Castleden WM. The effect of cessation of smoking on the accumulative survival rates of patients with symptomatic peripheral vascular disease. Med J Aust 1983;1:217-9. 8. Society for Vascular Surgery Lower Extremity Guidelines Writing Group, Conte MS, Pomposelli FB, Clair DG, Geragty PJ, McKinsey JF, et al. Society for Vascular Surgery practice guidelines for atherosclerotic occlusive disease of the lower extremities: management of asymptomatic disease and claudication. J Vasc Surg 2015;61:2S-41S. 9. Khullar D. The impact of smoking on surgical outcomes. J Am Coll Surg 2012;215:418-26. 10. Willigendael EM, Teijink JA, Bartelink ML, Peters RJ, Buller HR, Prins MH. Smoking and the patency of lower extremity bypass grafts: a meta-analysis. J Vasc Surg 2005;42: 67-74. 11. Lassila RR, Lepantalo MM. Cigarette smoking and the outcome after lower limb arterial surgery. Acta Chir Scand 1988;154:635-40. 12. Selvarajah S, Black JH, Malas MB, Lum YW, Propper BW, Abularrage CJ. Preoperative smoking is associated with early graft failure after infrainguinal bypass surgery. J Vasc Surg 2014;59:1308-14. 13. Singh N, Sidawy AN, DeZee KJ, Neville RJ, Akbari C, Henderson W. Factors associated with early failure of infrainguinal lower extremity arterial bypass. J Vasc Surg 2008;47:556-61. 14. Armstrong EJ, Wu J, Singh GD, Dawson DL, Pevec WC, Amsterdam EA, et al. Smoking cessation is associated with decreased mortality and improved amputation-free survival among patients with symptomatic peripheral artery disease. J Vasc Surg 2014;60:1565-71. 15. Farber A, Eberhardt RT. The current state of critical limb ischemia: a systematic review. JAMA Surg 2016;151:1070-7. 16. Shiloach M, Frencher SK, Steeger JE, Rowell KS, Bartzokis K, Tomeh MG, et al. Toward robust information: data quality and inter-rater reliability in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg 2010;210:6-16.

Submitted Oct 30, 2016; accepted Jan 2, 2017.