Major Adverse Limb Events and Major Adverse Cardiac Events after Contemporary Lower Extremity Bypass and Infrainguinal Endovascular Intervention in Patients with Claudication

Abstract

Objectives:

Major Adverse Limb Events (MALE) and Major Adverse Cardiovascular Events (MACE) at 30 days provide standardized metrics for comparison and have been adopted by the Society for Vascular Surgery’s Objective Performance Goals for critical limb ischemia. However, MALE and MACE have not been widely adopted within the claudication population, and the comparative outcomes after lower extremity bypass (LEB) and infrainguinal endovascular intervention (IEI) remain unclear. The purpose of this study was to compare MALE and MACE after LEB and IEI in a contemporary national cohort and to determine predictors of MALE and MACE after revascularization for claudication.

Methods:

A national dataset of LEB and IEI performed for claudication was obtained using National Surgical Quality Improvement Program vascular-targeted participant use files from 2011-2014. Patients were stratified by LEB versus IEI and compared by appropriate univariate analysis. The primary outcomes were MALE (defined as untreated loss of patency, reintervention on the index arterial segment, or amputation of the index limb) and MACE (defined as stroke, myocardial infarction or death). Multivariable logistic regression was used to identify predictors of MALE and MACE.

Results:

A total of 3,925 infrainguinal revascularization procedures (2155 LEB and 1770 IEI) were performed for claudication. There was no difference in 30-day MALE between LEB and IEI (4.0% vs. 3.2%, P=.17). On multivariable logistic regression, predictors of 30-day MALE included tibial revascularization (OR 2.2, P<.0001) and prior LEB on the same arterial segment (OR 1.8, P=.004). LEB had significantly higher 30-day MACE (2.0% vs. 1.0%, P=.01) but similar mortality (0.5% vs. 0.4%, P=.6). Predictors of MACE included LEB vs IEI (OR 2.1, P=.01), chronic obstructive pulmonary disease (OR=2.2, P=.01), dialysis dependence (OR=4.4, P=.003) and diabetes (OR=1.9, P=.02).

Conclusions:

In this large national cohort, LEB and IEI for claudication are associated with similar 30-day MALE. Tibial revascularization and revascularization after prior failed bypass predict MALE in claudicants and should therefore be undertaken with caution. LEB was associated with higher 30-day MACE but comparable 30-day mortality compared to IEI. Patients with ESRD, COPD, and diabetes are high risk for MACE. The risk of 30-day MACE after LEB should be weighed against the longer-term outcomes of LEB vs IEI and conservative management, particularly in these higher risk patients. This analysis helps define contemporary 30-day outcomes after infrainguinal revascularization performed for claudication and serves as a baseline to which the short term outcomes of future treatments can be compared.

Introduction

Lower extremity peripheral arterial disease (PAD) is a common syndrome worldwide and affects approximately 8 million patients in the United States.¹ Claudication has traditionally been treated conservatively with risk factor modification and supervised exercise programs; revascularization procedures have typically been reserved for those patients with severe disability from their symptoms and a favorable risk/benefit ratio for procedural intervention.^1-3 Revascularization can be performed via either open lower extremity bypass (LEB) or infrainguinal endovascular intervention (IEI), which has gained widespread acceptance over the past several decades.^{2, 4} Because IEI has lower peri-procedural morbidity and mortality when compared to open LEB, increasing numbers of patients with claudication are being treated with IEI as first-line therapy.^{5, 6} However, there remains limited high-quality evidence to guide the optimal revascularization strategy in claudicants. The majority of the current data are limited by heterogeneous study cohorts in which LEB and IEI have been compared in cohorts comprised of both claudicants and patients with critical limb ischemia (CLI). In addition, the use of inconsistent definitions of endpoints and outcome measures has significantly limited the ability to compare treatment strategies across different studies and populations.

The Society for Vascular Surgery (SVS) Objective Performance Goals (OPGs) established standardized metrics for comparison of 30-day outcomes after revascularization procedures in CLI patients.^{7, 8} These OPGs established expected rates of Major Adverse Limb Events (MALE), defined as either major amputation of the revascularized limb and/or reintervention on the revascularized segment. Additionally, they set goals for Major Adverse Cardiovascular Events (MACE), defined as cerebrovascular accident (CVA), myocardial infarction (MI) or death. The OPGs were created to compare outcomes of revascularization almost exclusively in the CLI population. The expected 30-day rates of MALE and MACE after revascularizations for claudication are not well known and OPGs could not be established.

The National Surgical Quality Improvement Program (NSQIP) is a validated database that provides information regarding a national sampling of cases, including 30-day follow-up data.^9-11 In 2011, vascular targeted modules were added to the existing participant use file (PUF), expanding the database and accruing data on vascular-related patient variables as well as limb and cardiovascular outcomes. The goal of this study was to compare MALE and MACE after LEB and IEI to help in starting to define OPGs for these patients, as well as to determine the factors predictive of MALE and MACE after revascularizations performed for claudication.

Methods

The NSQIP Vascular Targeted PUF (2011-2014) for both LEB and IEI procedures were merged to obtain a representative national dataset. All emergent procedures and those performed for CLI or asymptomatic peripheral vascular disease were excluded.

The primary outcomes were MALE and MACE within 30 days. MALE was defined as either untreated loss of patency of the revascularization, re-intervention on the revascularized segment, or major amputation (above or below knee) of the revascularized limb. MACE was defined as CVA, MI or death. Secondary outcomes included the individual component outcomes of MALE (untreated loss of patency of the revascularization, re-intervention on the revascularization, major amputation) and MACE (stroke/MI [a combined outcome variable], and death). Additional secondary outcomes included the operative complications of deep incisional surgical site infection (SSI), bleeding, acute renal failure, discharge to home, readmission and return to OR. Appropriate parametric and nonparametric statistical tests were used, including Chi Square test, independent t-test and Mann-Whitney U test to compare LEB and IEI cases. Multivariable logistic regression was then performed to identify predictors of MALE and MACE. The decision was made a priori to include the most significant predictors from univariate analysis up to a total of one predictor for every ten events in the model. Statistical significance was set to P <.05. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC). Informed consent was not obtained from subjects as The University of Virginia Institutional Review Board exempted all studies using the deidentified NSQIP dataset.

Results

Patients

A total of 13,294 LEB and IEI procedures were identified in the 2011-2014 NSQIP vascular targeted modules. Of these, 3,925 were performed for claudication, of which 2,155 (54.9%) were LEB and 1,770 (45.1%) were IEI. At baseline, these groups had significant differences in many preoperative characteristics (Table I). LEB patients were more commonly younger, male, Caucasian, high-risk ASA, and had history of tobacco use and COPD. They were also more likely than IEI patients to be undergoing revascularization of an infrageniculate arterial segment (21.8% vs. 9.5%, P<.0001). IEI patients were more likely to have a history of bleeding disorders, diabetes and dialysis dependence and were more frequently taking antiplatelet, antihypertensive and statin medications. IEI patients also had a higher history of prior percutaneous vascular interventions than those who were undergoing LEB (24.3% vs. 15.7%, P<.0001).

Table I.

Baseline and procedural characteristics

Baseline characteristics	Endovascular Intervention	Lower Extremity Bypass	p-value
Patients	1770 (45.1%)	2155 (54.9%)
Age	68 ± 11	66 ± 10	<.0001
Sex (male)	1082 (61.1%)	1504 (69.8%)	<.0001
Race (white)	1484 (83.8%)	1861 (86.4%)	.03
High Risk ASA class (>2)	1076 (60.8%)	1975 (91.7%)	<.0001
Preoperative antiplatelet therapy	1538 (86.9%)	1777 (82.5%)	.0001
Preoperative statin therapy	1305 (73.7%)	1507 (69.9%)	.009
Preoperative beta blocker	992 (56.1%)	1157 (53.7%)	.140
Steroid Use	66 (3.7%)	82 (3.8%)	.901
Bleeding Disorder	582 (32.9%)	398 (18.5%)	<.0001
Ascites	1 (0.1%)	1 (0.1%)	.889
Congestive Heart Failure	26 (1.5%)	24 (1.1%)	.323
Chronic Obstructive Pulmonary Disease	197 (11.1%)	303 (14.1%)	.006
Hypertension Medication	1486 (84.0%)	1717 (79.7%)	.001
Tobacco Use	647 (36.6%)	969 (45.0%)	<.0001
Dialysis	46 (2.6%)	27 (1.3%)	.002
Diabetes	735 (41.5%)	716 (33.2%)	<.0001
Functional Status			.563
Independent	1735 (98.0%)	2121 (98.4%)
Partially Dependent	33 (1.9%)	31 (1.4%)
Totally Dependent	2 (0.1%)	3 (0.1%)
Procedural Characteristics
Infrageniculate Target	168 (9.5%)	469 (21.8%)	<.0001
Prior Bypass of Current Segment	329 (18.6%)	389 (18.1%)	.67
Prior Percutaneous Intervention of Current Segment	430 (24.3%)	338 (15.7%)	<.0001
Transfer	23 (1.3%)	33 (1.5%)	.541
Inpatient procedure	677 (38.3%)	2131 (98.9%)	<.0001

ASA = American Society of Anesthsiologists

Procedures

Of the procedures performed for claudication (Table II), 3,288 (83.7%) were femoropopliteal revascularizations while the remaining 637 (16.3%) were completed on infrageniculate arterial targets. Of the total surgeries, 42.9% were LEB and 40.8% were IEI performed on femoropopliteal targets; 12% were LEB and 4.3% were IEI performed on infrageniculate targets.

Table II.

Procedures Performed for Claudication

Procedure	n (%)
Femoropopliteal Revascularization
LEB:
Femoropopliteal bypass w/ single segment saphenous vein	896 (22.8%)
Femoropopliteal bypass w/prosthetic/spliced vein/composite	790 (20.1%)
IEI:
Femoropopliteal angioplasty/stenting/atherectomy	1602 (40.8%)
Infrageniculate revascularization
LEB:
Femoral distal bypass w/ single segment saphenous vein	256 (6.5%)
Femoral distal bypass w/ prosthetic/spliced vein/composite	109 (2.8%)
Popliteal distal w/ single segment saphenous vein	69 (1.8%)
Popliteal distal bypass w/ prosthetic/spliced vein/composite	35 (0.9%)
IEI:
Tibial angioplasty/atherectomy/stenting	168 (4.3%)

LEB = lower extremity bypass; IEI = infrainguinal endovascular intervention

Primary Endpoints

At 30 days, MALE was similar in LEB and IEI groups (4.0% LEB vs. 3.2% IEI; P=.17) with no difference in any individual component outcome of MALE (Table III). On multivariable logistic regression, independent predictors of MALE (Table IV) included infrageniculate revascularization (OR 2.2, P<.0001) and a history of prior open bypass in the same arterial segment (OR 1.8, P=.004). Increasing age was protective against MALE (OR 0.98, P=.04).

Table III.

30-Day Major Adverse Limb and Cardiac Event Outcomes By Procedure Type

Parameter	Endovascular Intervention	Lower Extremity Bypass	p-value
Major Adverse Limb Events	56 (3.2%)	86 (4.0%)	.168
Untreated Loss of Patency	11 (0.6%)	22 (1.0%)	.173
Re-intervention	42 (2.4%)	61 (2.8%)	.372
Amputation	7 (0.4%)	14 (0.7%)	.277
Major Adverse Cardiac Events	17 (1.0%)	42 (2.0%)	.011
CVA or MI	12 (0.7%)	38 (1.8%)	.003
Mortality	7 (0.4%)	11 (0.5%)	.596

Table IV.

Major Adverse Limb Events Logistic Regression

Covariates	Odds Ratio	Confidence Interval		p-value
Infrageniculate Revascularization	2.163	1.473	3.176	<.0001
Prior Open Bypass on Current Segment	1.763	1.201	2.588	.004
Age	0.984	0.969	1	.045
Antiplatelet Therapy	0.664	0.438	1.008	.055
White Race	1.5	0.983	2.287	.06
Hight Risk ASA	1.392	0.867	2.236	.171
Chronic Obstructive Pulmonary Disease	0.671	0.367	1.229	.196
Bleeding Disorder	0.814	0.53	1.248	.345
Wound Class (Clean)	0.619	0.215	1.782	.374
Approach Open LEB Prosthetic/Spliced/Composite vs IEI	0.815	0.503	1.319	.405
Diabetes Mellitus	1.142	0.803	1.623	.46
Approach Open LEB SSV vs IEI	1.111	0.729	1.693	.624

c-statistic = 0.657

MACE was significantly higher in the LEB group (2.0% LEB vs. 1.0% IEI, P=.01) at 30 days (Table III). This was driven primarily by a significantly higher rate of CVA/MI (1.8% LEB vs. 0.7% IEI, P=.003). There was no difference in mortality between groups (0.5% LEB vs. 0.4% IEI, P=.6). On multivariable logistic regression, revascularization approach with LEB was an independent predictor of MACE (OR 2.1, P=.01; Table V). Other independent predictors of MACE included a history of COPD (OR 2.2, P=.01), dialysis dependence (OR 4.4, P=.003) and diabetes (OR 1.9, P=0.02).

Table V.

Major Adverse Cardiac Events Logistic Regression

Covariates	Odds Ratio	Confidence Interval		p-value
Dialysis	4.404	1.647	11.775	.003
LEB vs IEI	2.096	1.17	3.756	.013
Chronic Obstructive Pulmonary Disease	2.16	1.17	3.991	.014
Diabetes	1.859	1.094	3.16	.022
Infrageniculate Revascularization	1.595	0.879	2.896	.125
Beta-Blocker Therapy	1.388	0.797	2.415	.246

c-statistic = 0.694

Secondary Endpoints

Operative complications were different between LEB and IEI groups (Table VI), with LEB patients having a higher incidence of deep incisional SSI (1.6% LEB vs. 0.2% IEI, P<.0001), bleeding (11.7% LEB vs. 2.8% IEI, P<.0001) and return to OR (7.9% LEB vs. 4.9% IEI, P=.0001). LEB patients also had lower rates of discharge to home (90.4% LEB vs 97.7% IEI, P=<.0001).

Table VI:

Secondary Outcomes by Procedure Type

Parameter	Endovascular Intervention	Lower Extremity Bypass	p-value
Deep incisional SSI	4 (0.2%)	34 (1.5%)	<.0001
Bleeding	49 (2.8%)	253 (11.7%)	<.0001
Acute renal failure	3 (0.2%)	6 (0.3%)	.478
Discharge to home	1730 (97.7%)	1948 (90.4%)	<.0001
Readmission	10 (0.6%)	20 (0.9%)	.194
Return to operating room	86 (4.9%)	170 (7.9%)	.0001

SSI = surgical site infection

Discussion

Analysis of this large, national cohort demonstrated that patients who receive IEI have different risk profiles than those undergoing LEB when considering comorbidities and history of prior vascular interventions. After adjusting for these differences, LEB and IEI are associated with similar MALE in claudicants. MALE was independently associated with prior open surgery and infrageniculate revascularizations. Rates of 30-day MACE were higher in the LEB cohort, and open bypass was an independent predictor of MACE. This risk is primarily driven by a higher rate of CVA/MI and not death. Other factors independently associated with MACE are COPD, dialysis dependence and diabetes.

Our results are consistent with prior studies that demonstrate that patients with more comorbid disease and thus higher surgical risk are often primarily considered for IEI.⁵ As a result, it is not surprising that the LEB and IEI groups in this study had many baseline differences. Patients who underwent IEI were older and more likely to have significant cardiac risk factors including diabetes, hypertension and dialysis dependence. The LEB group did, however, have more smoking history and pulmonary disease.

This study identified comparable rates of 30-day MALE in the LEB and IEI group at 4% and 3.2%, respectively. Although these rates are surprisingly high for patients presenting only with claudication, the data shows that this is driven primarily by the need for reintervention and not, fortunately for the patients, by amputation. Several studies of PAD have shown open bypass patients have reduced risk of MALE when compared to those who undergo endovascular revascularizations, though these studies often include only CLI patients.^12-14 Currently there is no high quality data regarding 30-day outcomes after revascularizations for claudicants, particularly utilizing the OPG definitions of MALE. However, a study by Siracuse et al compared longer-term outcomes after LEB and IEI in those with superficial femoral artery (SFA) disease. Their results suggested that LEB leads to greater 3-year freedom from restenosis when compared to IEI (73% LEB vs. 42% IEI) in addition to greater freedom from recurrent claudication symptoms (70% vs. 36%).¹⁵ While they did not report data regarding progression to amputation, IEI patients had similar reintervention as LEB patients.

Our analysis identified two important, limb-specific factors that were independent predictors of MALE in the claudicant population: history of ipsilateral prior open bypass on the same arterial segment and revascularization of an infrageniculate artery. Previous studies confirm poor prognosis for redo revascularizations undertaken after prior failed open bypass.^16-18 Likewise, below-knee arterial segments indicate advanced disease and are typically small vessels associated with a high rate of failure in both open and endovascular procedures.^19-21 Tibial and redo revascularizations for claudication should therefore be undertaken with caution. Patients with these non-modifiable risk factors must be counseled regarding the increased possibility of adverse limb outcomes when discussing the risks and benefits of conservative vs operative management of their claudication symptoms. Although NSQIP does not have detailed anatomic data regarding vascular lesions, previous studies have suggested that, in claudicants, restenosis and symptom recurrence are both predicted by lesions with higher Trans-Atlantic Inter-Society Consensus II (TASC II) classification.¹⁵ The American Heart Association (AHA) and TASC II document recommend that IEI should be considered for claudicants with focal, short atherosclerotic segments while LEB should be considered for long-segment, diffuse disease. The quality of evidence regarding these recommendations has been called into question, however, and the SVS recommends the decision of operative approach should be an individualized one, based on the patient’s anatomy, available conduit, risk factors and preference.³

LEB was associated with a two-fold increase in 30-day MACE as compared to IEI patients. This difference was driven primarily by an increased rate of CVA and MI. The analysis by Siracuse et al comparing IEI and LEB in claudicants with SFA disease did not find a difference in 30-day occurrence of MI or death between groups, though the complication rate was very low. Patients with PAD have disease that affects multiple vascular beds, not only those vessels located in the lower extremities; they are therefore known to be at a high risk of MACE, even prior to undergoing an open or endovascular intervention.^22-24 The general anesthesia required for most LEB procedures coupled with potential blood loss drives the increased theoretical risk compared to IEI. For this reason, preference for IEI over LEB increases as surgical risk increases, as demonstrated in the baseline demographics of this study. While fully controlling for all comorbid states is impossible since the outcome of interest (MACE) was so rare, multivariable regression is able to control for some of the major difference between the IEI and LEB cohorts. Accounting for some of this bias, our multivariable regression shows that LEB is an independent predictor of MACE when compared to IEI. Other independent predictors of MACE in claudicants include known risk factors for CVA and MI, including COPD, dialysis dependence and diabetes.

Limitations of this study include the short-term follow up of the NSQIP dataset. Although no conclusions can be made regarding longer-term outcomes, it is important to define the expected rates of limb and cardiovascular outcomes in the immediate perioperative period. Further, while MACE typically occurs in the early perioperative period, MALE often does not occur until later in a patient’s disease course. Although surgeons may focus on long-term patency, early loss of patency requiring reintervention is a fortunately rare event that becomes all the more important considering the low risk for limb loss with conservative management. The study is retrospective in nature, thereby limiting determination of causality of the investigated outcomes. NSQIP does not provide details of vascular anatomy or imaging data which would allow better comparison of patient groups. Further, NSQIP does not capture all IEI performed at every participating institution as endovascular procedures performed by non-surgeons are not reported. This underestimates the number of IEIs being performed nationally and potentially introduces bias into the patient demographics which in turn could affect the results of the analysis. Finally, the variables available for a multivariable model may not fully account for the multifactorial process of procedure selection leaving some measure of selection bias.

Conclusion

This study helps establish the expected 30-day outcomes for LEB and IEI in the claudicant population using a large, national cohort of vascular patients. Although LEB and IEI have comparable 30-day rates of MALE, the increased risk of claudication patients undergoing infrageniculate and redo revascularizations should trigger caution when proceeding with revascularization. This study also demonstrated that LEB patients experience 30-day MACE twice as frequently as IEI patients. Serious consideration of IEI should therefore be given to claudicants at high risk of cardiovascular morbidity when their anatomy is anatomically favorable. This analysis helps define contemporary outcomes after infrainguinal revascularizations performed for claudication and can serve as a baseline to which future studies can be compared.