Association of Smoking With Postprocedural Complications Following Open and Endovascular Interventions for Intermittent Claudication

Katherine M Reitz; Andrew D Althouse; Joseph Meyer; Shipra Arya; Philip P Goodney; Paula K Shireman; Daniel E Hall; Edith Tzeng

doi:10.1001/jamacardio.2021.3979

. 2021 Oct 6;7(1):1–10. doi: 10.1001/jamacardio.2021.3979

Association of Smoking With Postprocedural Complications Following Open and Endovascular Interventions for Intermittent Claudication

Katherine M Reitz ^1,^2,³, Andrew D Althouse ⁴, Joseph Meyer ⁵, Shipra Arya ⁶, Philip P Goodney ^7,⁸, Paula K Shireman ^9,¹⁰, Daniel E Hall ^1,^3,^11,¹², Edith Tzeng ^1,^2,^3,^✉

¹Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania

²Division of Vascular Surgery, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania

³Department of Surgery, Veterans Affairs Pittsburgh Healthcare System, Pittsburgh, Pennsylvania

⁴University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania

⁵Department of Cardiology, Johns Hopkins, Baltimore, Maryland

⁶Division of Vascular Surgery, Stanford University School of Medicine, Stanford, California

⁷Section of Vascular Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire

⁸Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, New Hampshire

⁹Department of Surgery, UT Health San Antonio, University of Texas, San Antonio

¹⁰Department of Surgery, South Texas Veterans Health Care System, San Antonio

¹¹Center for Health Equity Research and Promotion, Veterans Affairs Pittsburgh Healthcare System, Pittsburgh, Pennsylvania

¹²Wolff Center at UPMC, Pittsburgh, Pennsylvania

Accepted for Publication: August 4, 2021.

Published Online: October 6, 2021. doi:10.1001/jamacardio.2021.3979

^✉

Corresponding Author: Edith Tzeng, MD, Division of Vascular Surgery, University of Pittsburgh School of Medicine, 200 Lothrop St, South Tower, Room 351.6, Pittsburgh, PA 15213 (tzenge@upmc.edu).

Author Contributions: Drs Reitz and Tzeng had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Hall and Tzeng are co–senior authors.

Study concept and design: Reitz, Meyer, Arya, Goodney, Hall, Tzeng.

Acquisition, analysis, or interpretation of data: Reitz, Althouse, Meyer, Shireman, Hall.

Drafting of the manuscript: Reitz, Meyer.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Reitz, Meyer, Goodney.

Obtained funding: Tzeng.

Administrative, technical, or material support: Reitz, Arya, Tzeng.

Study supervision: Arya, Hall, Tzeng.

Conflict of Interest Disclosures: None reported.

Funding/Support: This research was supported in part by grants I21 HX-002345 and XVA 72-909 from the Veterans Health Administration Office of Research and Development (Dr Hall); grant U01 TR002393 from the National Center for Advancing Translational Sciences (Drs Hall and Shireman); grant T32 HL0098036 from the National Heart, Lung, and Blood Institute (Dr Reitz); and grant L30 AG064730 from the National Institute on Aging (Dr Reitz).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The opinions expressed here are those of the authors and do not necessarily reflect the position of the US Department of Veterans Affairs or the US government.

Meeting Presentation: A subset of these data were presented virtually at the VA Pittsburgh Health Care System Research Week; May 17, 2021.

Additional Contributions: We thank the US Department of Veterans Affairs Office of Rural Health for sharing with us the use of their Rurality Calculator, developed by Daniel Palmeri, BS.

^✉

Corresponding author.

PMCID: PMC8495608 PMID: 34613348

Key Points

Question

Does smoking prior to elective revascularization for intermittent claudication (IC) increase the risk of early postprocedural complications?

Findings

In this cohort study including 14 350 revascularizations (open or endovascular) for IC, more than half of patients were smoking prior to the procedure. Overall, smoking within 1 year of the procedure was associated with a 48% increase in the risk of any early postprocedural complication; compared with active smokers (smoking within 2 weeks before the procedure), the risk of any complication was decreased by 65% for never smokers and by 29% for former smokers (more than 1 year cessation).

Meaning

In this study, smoking was associated with an increased risk of early postprocedural complications; smoking cessation should be emphasized prior to elective revascularization for IC.

This cohort study assesses if preprocedural smoking is associated with an increased risk of early postprocedural complications following elective open and endovascular revascularization.

Abstract

Importance

Smoking is a key modifiable risk factor in the development and progression of peripheral artery disease, which often manifests as intermittent claudication (IC). Smoking cessation is a first-line therapy for IC, yet a minority of patients quit smoking prior to elective revascularization.

Objective

To assess if preprocedural smoking is associated with an increased risk of early postprocedural complications following elective open and endovascular revascularization.

Design, Setting, and Participants

This retrospective cohort study used nearest-neighbor (1:1) propensity score matching of 2011 to 2019 data from the Veterans Affairs Surgical Quality Improvement Program, including all cases with a primary diagnosis of IC and excluding emergent cases, primary procedures that were not lower extremity revascularization, and patients with chronic limb-threatening ischemia within 30 days of the intervention. All data were abstracted June 18, 2020, and analyzed from July 26, 2020, to June 30, 2021.

Exposures

Preprocedural cigarette smoking.

Main Outcomes and Measures

Any and organ system–specific (ie, wound, respiratory, thrombosis, kidney, cardiac, sepsis, and neurological) 30-day complications and mortality, overall and in prespecified subgroups.

Results

Of 14 350 included cases of revascularization, 14 090 patients (98.2%) were male, and the mean (SD) age was 65.7 (7.0) years. A total of 7820 patients (54.5%) were smoking within the preprocedural year. There were a total of 4417 endovascular revascularizations (30.8%), 4319 hybrid revascularizations (30.1%), and 5614 open revascularizations (39.1%). A total of 1594 patients (11.1%) had complications, and 57 (0.4%) died. Among 7710 propensity score–matched cases (including 3855 smokers and 3855 nonsmokers), 484 smokers (12.6%) and 34 nonsmokers (8.9%) experienced complications, an absolute risk difference (ARD) of 3.68% (95% CI, 2.31-5.06; P < .001). Compared with nonsmokers, any complication was higher for smokers following endovascular revascularization (26 [4.3%] vs 52 [2.1%]; ARD, 2.19%; 95% CI, 0.77-3.60; P = .003), hybrid revascularization (204 [17.3%] vs 163 [14.1%]; ARD, 3.18%; 95% CI, 0.23-6.13; P = .04), and open revascularization (228 [15.4%] vs 153 [10.3%]; ARD, 5.18%; 95% CI, 2.78-7.58; P < .001). Compared with nonsmokers, respiratory complications were higher for smokers following endovascular revascularization (20 [1.7%] vs 6 [0.5%]; ARD, 1.17%; 95% CI, 0.35-2.00; P = .009), hybrid revascularization (33 [2.8%] vs 10 [0.9%]; ARD, 1.93%; 95% CI, 0.85-3.02; P = .001), and open revascularization (32 [2.2%] vs 19 [1.3%]; ARD, 0.89%; 95% CI, 0-1.80; P = .06). Wound complications and graft failure were higher for smokers compared with nonsmokers following open interventions (wound complications: 146 [9.9%] vs 87 [5.8%]; ARD, 4.05%; 95% CI, 2.12-5.99; P < .001; graft failure: 33 [2.2%] vs 11 [0.7%]; ARD, 1.50%; 95% CI, 0.63-2.37; P = .001). In a sensitivity analysis, compared with active smokers (n = 5173; smoking within 2 weeks before the procedure), the risk of any complication was decreased by 65% for never smokers (n = 1197; adjusted odds ratio, 0.45; 95% CI, 0.34-0.59) and 29% for former smokers (n = 4755; cessation more than 1 year before the procedure; adjusted odds ratio, 0.71; 95% CI, 0.61-0.83; P = .001 for interaction).

Conclusions and Relevance

In this cohort study, more than half of patients with IC were smoking prior to elective revascularization, and complication risks were higher across all modalities of revascularization. These findings stress the importance of smoking cessation to optimize revascularization outcomes.

Introduction

Peripheral artery disease (PAD) affects more than 230 million people worldwide and is particularly prevalent among US veterans owing to high rates of tobacco use.^1,2 PAD often manifests as intermittent claudication (IC), defined as ischemic muscle pain with activity.³ Atherosclerosis is a systemic disease that affects coronary, cerebral, and peripheral arteries and confers increased risk of cardiovascular disease (CVD)–related morbidity and mortality.^2,4 Proinflammatory and prothrombotic effects of cigarette smoking^4,5 on PAD contribute to earlier disease onset and progression, limb loss, functional decline, reduced quality of life, and 20% of CVD deaths in the US.^{3,6,7,8,9,10,11,12}

Evidence-based IC treatment guidelines focus primarily on mitigation of associated CVD morbidity and mortality and secondarily on relief of IC.^13,14 The former involves clear prioritization of optimal medical therapy (OMT), which includes smoking cessation.^15,16,17,18 If IC symptoms persist following OMT, revascularization may be considered. Yet despite the unknown effectiveness, lack of durability, elective nature of revascularization for IC, and known benefits of OMT, nearly two-thirds of patients are not medically optimized prior to revascularization.^18,19

Our primary objective was to determine the association between smoking and 30-day postprocedural complications following elective revascularization for IC. Secondarily, we evaluated the use of endovascular and open revascularization for IC over time. We hypothesized that a large proportion of patients undergoing revascularization for IC were smokers and that smoking was associated with an increased risk of early postprocedural complications across all intervention types.

Methods

We used deidentified 2011 to 2019 data from the Veterans Affairs Surgical Quality Improvement Program (VASQIP) in this retrospective cohort study, determined to be exempt by the Veterans Affairs Pittsburgh Healthcare System Institutional Review Board. All data were abstracted June 18, 2020, and analyzed from July 26, 2020, to June 30, 2021. Data are reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.²⁰

Data Source

The VASQIP comprises a systematic and representative sampling of veterans undergoing interventions at all Veterans Affairs surgical centers.^21,22 Each center is mandated to randomly sample up to 6 cases per procedure type and more than 36 cases overall every 8-day cycle, deliberately and systematically oversampling uncommon procedures. Each patient is observed for 30 postprocedural days.

The VASQIP includes patient-level data including more than 200 variables, described in detail by the Veterans Affairs National Surgery Office,²² capturing demographic, medical, surgical center, and procedure data and postprocedure complications in conjunction with healthy user indicators representing health habits (eg, alcohol use) and functionality (eg, activities of daily living, nursing residence).²³ We expanded these data to capture otherwise unmeasured potential confounding, including (1) patient frailty, (2) physiologic stress of the intervention, (3) intervention modality, (4) anatomic level of revascularized PAD, and surgical center specific measures of (5) patient rurality and (6) site complexity. We assessed patient frailty and the physiologic stress of the principal procedure with the validated Risk Analysis Index²⁴ and Operative Stress Score,^25,26 which independently and synergistically predict postprocedural complications.^24,25,26,27 The principal procedure Current Procedural Terminology code defined the intervention modality (ie, endovascular or open) and was recategorized as hybrid if the case had multiple Current Procedural Terminology codes including both modalities. The most proximal intervention defined the anatomic level of revascularized PAD. Surgical site rurality and complexity level were quantified, capturing health care access inequalities and socioeconomic status.^28,29 We quantified rurality by Rural-Urban Commuting Areas and the Veterans Health Administration Office of Rural Health Rurality Calculator.³⁰ Surgical center complexity level was determined by the highest designation of procedure complexity, volume, and infrastructure capabilities.³¹

Cohort Identification

We included all noncardiac cases in the VASQIP (January 3, 2011, to September 30, 2019) with an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10-CM preprocedural diagnosis of IC (eTable 1 in the Supplement) and a principal revascularization procedure (eTable 2 in the Supplement). Our final cohort included only elective cases, excluding (1) patients with preprocedural chronic limb-threatening ischemia and/or (2) American Society of Anesthesiologists Score emergency designation.

Study Exposure and Outcome

Veterans were classified as smokers if the VASQIP abstractor adjudicated that the patient smoked cigarettes within 1 year or as nonsmokers (ie, former and never) if they did not smoke within 1 year prior to admission. The primary outcome was any 30-day postprocedural complication.³² Thirty-day secondary outcomes included major and minor complications (eTable 3 in the Supplement), complications by organ system (eTable 4 in the Supplement), and mortalitly.³²

Statistical Analysis

We expressed categorical variables as counts and percentages and continuous variables as means and SDs or medians and IQRs. Proportions of missing data stratified by smoking status were also quantified.

To minimize the potential bias of smoking status on comorbid conditions and confounding, we generated a propensity score to quantify the likelihood patients had for smoking using covariates determined a priori.³³ We included all data available throughout the study timeframe and excluded variables (1) affecting less than 0.1% of patients eligible for matching, (2) with more than 5% missingness, which could not be assumed to be missing at random, or (3) potentially representing the proinflammatory and prothrombotic effects of smoking (ie, causal pathway).

We generated a matched population of smokers and nonsmokers using a 1:1 nearest-neighbor matching algorithm with calipers of 0.02, without replacement.^34,35,36 We compared preprocedural patient, procedure, and treatment center characteristics before and after matching using a standardized difference (less than 0.15 considered acceptable).^34,37 We also compared characteristics among the patients included and excluded from the match.

In the matched cohort, we quantified the frequency of postprocedural outcomes with and without stratification by intervention modality and anatomic level of disease. We calculated the absolute risk difference (ARD) and compared outcomes using the McNemar test.³⁸ We used logistic regression with the robust variance estimator to calculate matched odds ratios (ORs) for postprocedural complications. The effects of smoking were assessed among patients from self-identified racial minority groups and with prespecified comorbid conditions, including (1) chronic obstructive pulmonary disease (COPD), (2) comorbid CVD (ie, myocardial infarction, congestive heart failure, and stroke), (3) diabetes, (4) frailty (ie, nonfrail [Risk Analysis Index less than 30] and frail [Risk Analysis Index of 30 or more]), (5) intervention modality, and (6) level of revascularization. Significance was set at a P value less than .05, and all P values were 2-tailed.

Sensitivity Analysis

We evaluated the robustness of our results in sensitivity analyses. First, we calculated the E-value quantifying the effect size magnitude required for a potential unmeasured confounder to negate the observed association between smoking and any postprocedural complication.^39,40 Second, we evaluated alternative definitions of any and major (ie, Clavien-Dindo grade IV) postprocedural complications.⁴¹ Third, we calculated adjusted ORs of complications in the full cohort eligible for matching to account for the reduction in sample size inherent to the matching methodology. Finally, we evaluated outcomes among all patients eligible for matching with a known smoking timeframe. We calculated adjusted ORs for any postprocedural complication among those who (1) actively smoked (within 2 weeks of the procedure), (2) formerly smoked (cessation more than 1 year before the procedure), and (3) never smoked overall and by intervention modality. All analyses were conducted using Stata version 15.1 (StataCorp) and Prism version 7.0 (GraphPad).

Results

We identified 20 266 cases of revascularization for IC. Of 14 350 included cases of elective revascularizations, 14 090 patients (98.2%) were male, 4490 (31.3%) had concurrent CVD, and the mean (SD) age was 65.7 (7.0) years. A total of 7820 patients (54.5%) were smoking within the preprocedural year (eFigure 1 in the Supplement). Patients with IC and concurrent CVD had a lower rate of smoking (2259 of 4490 [50.3%]) compared with those without CVD (5589 of 9860 [56.7%]; P < .001). There were a total of 4417 endovascular revascularizations (30.8%), 4319 hybrid revascularizations (30.1%), and 5614 open revascularizations (39.1%); and there were 4514 cases (31.6%) of suprainguinal disease and 9780 (68.4%) of infrainguinal disease. Case numbers increased over time (mean [SD] change per year, 98 [234]) with a dramatic rise in the proportion of endovascular revascularizations (mean [SD] change per year, 5.0% [8.8%]) and a progressive decline in open revascularization (mean [SD] change per year, −3.4% [7.9%]; eFigure 2 in the Supplement). Overall, 1594 cases (11.1%) resulted in any complication, and 57 (0.4%) resulted in death by 30 days.

Characteristics according to smoking status are summarized in Table 1 with variable missingness quantified in eTable 5 in the Supplement. In the full cohort eligible for matching, compared with nonsmokers, smokers were younger, more likely to be from a racial and ethnic minority group and have COPD, and less likely to have CVD, diabetes, and frailty. Smokers underwent more physiologically stressful open and hybrid interventions. The rate of smoking for each intervention type was consistent over time (eFigure 2 in the Supplement).

Table 1. Preprocedural Patient and Hospital Characteristics.

Variable^a	All eligible for propensity score matching			Propensity score matched
	No. (%)		Standardized difference^b	No. (%)		Standardized difference^b
	Nonsmoking (n = 6530)	Smoking (n = 7820)	Standardized difference^b	Nonsmoking (n = 3855)	Smoking (n = 3855)	Standardized difference^b
Demographic characteristics
Age, mean (SD), y	68.4 (6.8)	63.5 (6.2)	0.614	66.3 (6.2)	66.2 (5.7)	0
Male	6453 (98.8)	7637 (97.7)	0.088	3795 (98.4)	3818 (99.0)	0.054
Race^c
Black	701 (10.7)	1240 (15.9)	0.151	491 (12.7)	495 (12.8)	0.003
White	5146 (78.8)	5863 (75.0)	0.091	2976 (77.2)	2977 (77.2)	0.001
Other	477 (7.3)	464 (5.9)	0.055	266 (6.9)	257 (6.7)	0.009
Hispanic ethnicity	185 (2.9)	196 (2.6)	0.020	96 (2.6)	116 (3.1)	0.032
BMI, mean (SD)^d	29.0 (4.9)	26.7 (5.0)	0.465	28.0 (4.5)	28.2 (5.1)	0.042
Comorbid conditions
Transient ischemic attack^e	355 (5.4)	298 (3.8)	0.077	179 (4.6)	183 (4.7)	0.005
Stroke^e
Without neurologic deficits	306 (4.7)	462 (5.9)	0.055	225 (5.8)	199 (5.2)	0.030
With neurologic deficits	284 (4.3)	389 (5.0)	0.030	191 (5.0)	172 (4.5)	0.023
Congestive heart failure within 1 mo^e	1378 (21.6)	1158 (15.1)	0.165	762 (20.2)	726 (19.2)	0.024
Myocardial infarction within 6 mo^e	823 (12.6)	695 (8.9)	0.120	481 (12.5)	470 (12.2)	0.009
Hypertension^f	5891 (90.2)	6548 (83.7)	0.193	3410 (88.5)	3455 (89.6)	0.037
COPD	1402 (21.5)	2183 (27.9)	0.150	970 (25.2)	957 (24.8)	0.008
Dialysis within 2 wk	98 (1.5)	39 (0.5)	0.101	54 (1.4)	38 (1.0)	0.038
Diabetes
Oral medications	1390 (21.3)	1279 (16.4)	0.126	789 (20.5)	780 (20.2)	0.006
Insulin medications	1700 (26.0)	1176 (15.0)	0.275	728 (18.9)	804 (20.9)	0.049
Alcohol use	495 (7.6)	948 (12.1)	0.153	379 (9.8)	324 (8.4)	0.050
RAI
Nonfrail (RAI ≤29)	1642 (25.1)	4093 (52.3)	0.581	1276 (33.1)	1413 (36.7)	0.075
Frail (RAI of 30-39)	4381 (67.1)	3473 (44.4)	0.469	2379 (61.7)	2297 (59.6)	0.044
Very frail (RAI ≥40)	507 (7.8)	254 (3.2)	0.199	200 (5.2)	145 (3.8)	0.069
Hospital admission
Hospital complexity
Intermediate	1467 (22.5)	1630 (20.8)	0.039	852 (22.1)	898 (23.3)	0.028
Complex	5047 (77.5)	6189 (79.2)	0.045	3002 (77.9)	2956 (76.7)	0.028
Hospital rurality, %
<25	2494 (39.9)	2804 (36.8)	0.048	1458 (38.9)	1456 (39.0)	0.001
25-50	2534 (40.6)	3429 (45.0)	0.103	1565 (41.8)	1480 (39.7)	0.045
50-75	1057 (16.9)	1209 (15.9)	0.020	618 (16.5)	676 (18.1)	0.040
>75	162 (2.6)	181 (2.4)	0.011	103 (2.8)	119 (3.2)	0.025
Operative stress score
2	4619 (71.0)	4703 (60.4)	0.224	2524 (65.6)	2533 (66.0)	0.005
3	1715 (26.4)	2565 (32.9)	0.144	1181 (30.7)	1186 (30.9)	0.003
≥4	170 (2.6)	522 (6.7)	0.195	145 (3.8)	121 (3.2)	0.034
Anatomic location of disease
Suprainguinal	1820 (28.0)	2694 (34.6)	0.142	1168 (30.3)	1138 (29.6)	0.017
Infrainguinal	4684 (72.0)	5096 (65.4)	0.142	2682 (69.7)	2702 (70.4)	0.011
Intervention modality
Endovascular	2179 (33.4)	2238 (28.6)	0.103	1211 (31.4)	1200 (31.1)	0.006
Hybrid	1891 (29.0)	2428 (31.0)	0.046	1154 (29.9)	1179 (30.6)	0.014
Open	2460 (37.7)	3154 (40.3)	0.055	1490 (38.7)	1476 (38.3)	0.007

Open in a new tab

Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; RAI, Risk Analysis Index.

^{^a}

A subset of the 44 variables included in the propensity score match; the full table is available in eTable 6 in the Supplement.

^{^b}

Standardized differences were defined as difference in means, proportions, or ranks divided by the mutual standard deviation. A standardized difference less than 0.15 was considered acceptable.

^{^c}

Race and ethnicity data are abstracted from the Veterans Health Information System and Technology Architecture Patient Information Management System, which catalogs patient registration, admission, discharge, transfer, and appointment scheduling data in the US Department of Veterans Affairs. Other includes American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, and Unknown by Patient.

^{^d}

Calculated as weight in kilograms divided by height in meters squared.

^{^e}

Constitutes a history of cardiovascular disease.

^{^f}

Hypertension defined as the presence of antihypertensive medication(s) use.

After propensity score matching (3855 smokers and 3855 nonsmokers), characteristics were equally distributed between groups (Table 1; eTable 6 and eFigures 3 and 4 in the Supplement). Compared with matched nonsmokers, unmatched nonsmokers were older, more likely to be White and have CVD, and less likely to have COPD. Unmatched smokers demonstrated an opposite trend (eTable 7 in the Supplement).

Outcomes in the Propensity Score–Matched Cohort

Smokers had a higher risk of any 30-day postprocedural complication compared with nonsmokers (484 [12.6%] vs 34 [8.9%]), with an ARD of 3.68% (95% CI, 2.31-5.06; P < .001). The increased risk for smokers was observed across subgroups; compared with nonsmokers, any complication was higher for smokers following endovascular revascularization (26 [4.3%] vs 52 [2.1%]; ARD, 2.19%; 95% CI, 0.77-3.60; P = .003), hybrid revascularization (204 [17.3%] vs 163 [14.1%]; ARD, 3.18%; 95% CI, 0.23-6.13; P = .04), and open revascularization (228 [15.4%] vs 153 [10.3%]; ARD, 5.18%; 95% CI, 2.78-7.58; P < .001) (Figure 1). However, the risk of any complication among smokers was significantly increased for non-White Hispanic patients and those with COPD and nonfrail status compared with non-Hispanic White patients and those without COPD and frailty.

Figure 1. — Heterogeneity of smoking association visualized by forest plot of matched odds ratios and 95% CIs for any 30-day complication by prespecified subgroup and overall. The dotted vertical line represents an odds ratio of 1.00, indicating no difference between smoking and nonsmoking groups. The dashed vertical line represents the overall matched odds ratio for the association of smoking with any 30-day complication. COPD indicates chronic pulmonary obstructive disease; CVD, cardiovascular disease.

^aP value of the interaction term.

^bAmong patients with infrainguinal disease, there were 5194 femoral/popliteal interventions (96.6%) and 190 tibial interventions (3.4%).

In evaluating secondary outcomes, smokers had a higher risk of major, minor, wound, respiratory, and thrombotic complications compared with nonsmokers (Figure 2; Table 2). Compared with nonsmokers, respiratory complications were higher for smokers following endovascular revascularization (20 [1.7%] vs 6 [0.5%]; ARD, 1.17%; 95% CI, 0.35-2.00; P = .009), hybrid revascularization (33 [2.8%] vs 10 [0.9%]; ARD, 1.93%; 95% CI, 0.85-3.02; P = .001), and open revascularization (32 [2.2%] vs 19 [1.3%]; ARD, 0.89%; 95% CI, 0-1.80; P = .06). However, wound and thrombotic complications were only significantly higher for smokers compared with nonsmokers undergoing open procedures (wound: 146 [9.9%] vs 87 [5.8%]; ARD, 4.05%; 95% CI, 2.12-5.99; P < .001; thrombotic: 41 [2.8%] vs 16 [1.1%]; ARD, 1.71%; 95% CI, 0.72-2.69; P = .001). Notably, the risk of vascular graft failure was also significantly higher for smokers compared with nonsmokers undergoing open revascularization (33 [2.2%] vs 11 [0.7%]; ARD, 1.50%; 95% CI, 0.63-2.37; P = .001). Smoking was associated with increased 30-day mortality (23 [0.6%] vs 2 [0.1%]; ARD, 0.54%; 95% CI, 0.29-0.80).

Figure 2. — Complications by severity (A) and wound, respiratory, and thrombosis complications (B) were significantly higher among smokers compared with nonsmokers. Error bars indicate 95% CIs.

^aComplication categorization defined in eTable 3 in the Supplement.

^bComplication system defined in eTable 4 in the Supplement.

^cGraft failure is a subset of the thrombosis-related complications category.

Table 2. 30-Day Outcomes in Propensity Score–Matched Cohort and Full Cohort Eligible for Matching.

30-d Outcome	Propensity score–matched cohort (n = 7710)						All eligible for match (N = 14 350)
	Events, No. (%)		Absolute risk difference, % (95% CI)	P value^a	Matched OR (95% CI)	P value	Adjusted OR (95% CI)^b	P value
	Nonsmoking	Smoking	Absolute risk difference, % (95% CI)	P value^a	Matched OR (95% CI)	P value	Adjusted OR (95% CI)^b	P value
Primary outcome
Any complication	342 (8.9)	484 (12.6)	3.68 (2.31 to 5.06)	<.001	1.48 (1.27 to 1.71)	<.001	1.36 (1.19 to 1.55)	<.001
Secondary outcomes
Complication by severity^c
Major	195 (5.0)	255 (6.6)	1.56 (0.51 to 2.60)	.004	1.33 (1.10 to 1.61)	<.001	1.21 (1.03 to 1.13)	.02
Minor	163 (4.2)	265 (6.9)	2.65 (1.63 to 3.67)	<.001	1.67 (1.37 to 2.04)	<.001	1.51 (1.27 to 1.81)	<.001
Complication by organ system^d
Wound	209 (5.4)	280 (7.3)	1.84 (0.75 to 2.93)	.001	1.37 (1.14 to 1.64)	.001	1.35 (1.14 to 1.60)	<.001
Respiratory	35 (0.9)	85 (2.2)	1.30 (0.75 to 1.85)	<.001	2.46 (1.66 to 3.66)	<.001	1.85 (1.28 to 2.69)	.001
Thrombosis	51 (1.3)	77 (2.0)	0.67 (0.10 to 1.24)	.02	1.52 (1.06 to 2.17)	.02	1.24 (0.91 to 1.68)	.17
Graft failure^e	45 (1.2)	63 (1.6)	0.47 (0.01 to 0.10)	.08	1.41 (0.96 to 2.07)	.08	1.24 (0.84 to 1.87)	.27
Kidney	36 (0.9)	43 (1.1)	0.18 (−0.27 to 0.63)	.43	1.20 (0.77 to 1.87)	.43	1.08 (0.70 to 1.66)	.72
Cardiac	34 (0.9)	30 (0.8)	0.01 (−0.51 to 0.30)	.62	0.88 (0.54 to 1.44)	.62	0.88 (0.59 to 1.31)	.52
Sepsis	23 (0.6)	36 (0.9)	0.34 (−0.05 to 0.73)	.09	1.57 (0.93 to 2.66)	.09	1.22 (0.76 to 1.95)	.40
Neurological	5 (0.1)	11 (0.3)	0.16 (−0.05 to 0.36)	.14	2.20 (0.76 to 6.35)	.14	1.42 (0.81 to 2.49)	.23

Open in a new tab

Abbreviation: OR, odds ratio.

^{^a}

Differences between matched groups evaluated with McNemar test.

^{^b}

Adjusted results obtained from multivariable logistic regression including the full cohort eligible for propensity score matching (eTable 8 in the Supplement).

^{^c}

Complication categorization defined in eTable 3 in the Supplement; patients may have both major and minor 30-day complications.

^{^d}

Complication system defined in eTable 4 in the Supplement; patients may have 1 or more system-based 30-day complications.

^{^e}

Graft failure is a subset of thrombosis complications.

Sensitivity Analysis

Our results in the propensity score–matched cohort were robust to both potential confounders and alterative definitions of the outcomes. The E-value indicates a single unmeasured confounder with an ARD greater than 2.13% or an OR greater than 2.21 is required to negate the association between smoking and any complication. In the matched cohort, the association between smoking and complications persisted when altering the definition of any complication (ARD, 2.20%; 95% CI, 0.79-3.62; P = .002) and major complications (ARD, 1.63%; 95% CI, 0.31-2.95; P = .02).

Among the full cohort eligible for matching, the association between smoking and complications was consistently observed (eTable 8 in the Supplement). The risk of postprocedural complications also increased with the magnitude of the smoking exposure timeframe (eTable 9 in the Supplement). Compared with active smokers (n = 5173; smoking within 2 weeks before the procedure), the risk of any complication was decreased by 65% for never smokers (n = 1197; adjusted odds ratio, 0.45; 95% CI, 0.34-0.59) and 29% for former smokers (n = 4755; adjusted odds ratio, 0.71; 95% CI, 0.61-0.83; P = .001 for interaction), which differed across intervention types (Figure 3).

Figure 3. — Heterogeneity of smoking timeframe visualized by forest plot of adjusted odds ratios and 95% CIs for any 30-day complication comparing active smokers (smoking within 2 weeks before the procedure) with never smokers and with former smokers (more than 1 year cessation) overall and by treatment modality. There were a total of 4417 endovascular revascularizations (30.8%), 4319 hybrid revascularizations (30.1%), and 5614 open revascularizations (39.1%). The dotted vertical line represents an odds ratio of 1.00, indicating no difference between timeframe exposure groups. The dashed vertical lines indicate the overall active vs never smoking (top) and active vs former smoking (bottom) adjusted odds ratios. Error bars indicate 95% CIs.

^aCigarette smoking exposure timeframe is available only in the 2012 to 2019 Veterans Affairs Quality Improvement Program data sets (n = 13 184). Analysis included those with cigarette smoking timeframe available (n = 12 794 [97.0%]) and excluding those who stopped smoking between 2 and 52 weeks prior to the procedure (n = 1669 [12.7%]). Adjusted odds ratios generated after controlling for baseline demographic characteristics, including age, sex, race, ethnicity, and body mass index; comorbid conditions, including stroke, congestive heart failure, hypertension, diabetes, alcohol use, and frailty; history of lower extremity revascularization or amputation for ischemia and percutaneous coronary intervention; preprocedural hematocrit level; hospital complexity level; procedure stress and intervention type; and procedure year.

^bThe interaction P value overall and by intervention modalities by exposure subgroups.

Discussion

Our large, contemporary study of 14 350 cases evaluated the association of preprocedural smoking with early postprocedural outcomes following revascularization for IC. We confirmed a high rate of smoking in patients undergoing elective revascularization. In this well-balanced propensity score–matched cohort, smoking was associated with a 48% relative increase in the risk of any 30-day postprocedural complication compared with nonsmokers (ARD, 3.68%; 95% CI, 2.31-5.06). All smokers undergoing revascularization, regardless of intervention modality and anatomic level of disease, had a higher risk of early postprocedural complications. This risk fell by 29% among former smokers (adjusted odds ratio, 0.71; 95% CI, 0.61-0.83) with more than 1 year of smoking cessation and by 65% among never smokers (adjusted odds ratio, 0.45; 95% CI, 0.34-0.59). Our findings were robust to adjustments for potential confounders, including patient comorbidity, frailty, healthy user indicators, health care access inequalities with investigation by multiple methods, across subgroups, and in sensitivity analysis.

A large body of evidence has established that cigarette smoking is a key modifiable risk factor in the earlier onset of arterial changes contributing to the pathogenesis of atherosclerotic CVD, including PAD.^{1,2,3,7,8,9,10,12,13,14} Over the past several decades, the national smoking rate has decreased.⁴² In conjunction with new, effective therapies and increasing rates of other OMT, CVD-related mortality has declined. Despite clear and consistent evidence-based guidelines promoting smoking cessation,¹⁷ patients with PAD repeatedly demonstrate higher rates of smoking compared with those with other CVD.^15,19 In the Reduction of Atherothrombosis for Continued Health (REACH) registry,⁴³ patients with isolated PAD smoked at twice the rate of patients with other CVD. In the Patient‐Centered Outcomes Related to Treatment Practices in Peripheral Arterial Disease: Investigating Trajectories (PORTRAIT) registry,¹⁷ more than 30% of patients evaluated for IC were smokers but only 25% were referred to smoking cessation counseling or received assistive pharmacotherapy. Observational data suggest less than 25% of patients with IC achieve OMT prior to elective open revascularization.^18,19 Our data reinforce these findings. More than half the patients smoked within 1 year of an elective intervention, and the smoking rate was lower among those with other concurrent CVD. In contrast to declining rates of smoking nationally, the proportion of smoking veterans remained strikingly unchanged throughout our 9-year study period. The alarmingly high and stable rate of smokers undergoing revascularization for IC may be because of both the causal relationship between smoking and PAD and an underappreciation of the benefits of smoking cessation for patients with PAD. Further, the barriers to revascularization are lower with endovascular approaches such that surgeons and interventionalists may proceed with revascularization without fully achieving OMT.

Early postprocedural complications decrease patient satisfaction and increase hospital length of stay, readmission, and health care costs.⁴⁴ Our data support a large body of cross-discipline evidence demonstrating increased early postoperative complications, especially respiratory and wound-related events, for smokers following a variety of procedures.^45,46,47 We observed the risk of any complication for smokers was higher among those with COPD, as expected, and among patients from racial and ethnic minority groups, mirroring ubiquitously observed health care disparities.^48,49 Nonfrail patients also demonstrated an increased relative risk of any complication associated with smoking compared with frail patients. As observed across diverse and PAD-specific procedures,^25,50,51,52 surgeons and interventionalists may apply more rigorous patient selection and preprocedural optimization (eg, prehabilitation, nutritional support) for frail patients undergoing physiologically stressful procedures.^24,25,27,41 We therefore hypothesize that these mitigation efforts may exceed the relative risk reduction achieved with smoking cessation. In contrast, robust patients may be subject to less selection scrutiny and preprocedural optimization. This is supported by the relative increase in any postprocedural complication associated with smoking in nonfrail patients. Alternatively, the depleted physiologic reserve of frailty increases the risk of decompensation in response to a physiologic stressor (ie, procedure) so strongly that it may dominate the effect of smoking (eg, among frail patients, the reserve is so greatly reduced such that the effects of smoking are blunted).

Prior work demonstrated that early postprocedural morbidity and mortality are associated with both the physiologic stress of the intervention^25,53 and smoking status.^6,8,9,10 As previously observed among smokers,⁵³ complications were more common after hybrid revascularization (204 [17.3%]) and open revascularization (228 [15.4%]) than endovascular revascularization (26 [4.3%]). We extend prior findings in 2 important ways. First, smoking was associated with a 48% increase in any complication overall (Table 2) that was not limited to high-stress open interventions but was observed across all modalities. Among patients undergoing endovascular revascularization, smoking was associated with a 2% absolute risk (ARD, 2.19%; 95% CI, 0.77-3.60) and 100% relative risk (Table 1) increase in any complication. Second, smoking was associated with an increased risk of early vascular graft failure following a broad range of open revascularization procedures performed exclusively to treat IC. Therefore, regardless of the associated surgical stress or the perceived risk of the intervention, smoking conferred an elevated risk of postprocedural complications.

Evidence-based societal guidelines reflect the uncertain effectiveness, lack of durability, and inability to impact CVD risks attributed to revascularization for IC. In 2015, the Society of Vascular Surgery endorsed a Grade 1B recommendation for revascularization following OMT only for ongoing “significant functional or lifestyle-limiting disabilities.”¹³ Similarly, in the 2018 American College of Cardiology/American Heart Association Appropriate Use Criteria, endovascular or open revascularization was deemed to be appropriate or may be appropriate for aortoiliac, superficial femoral, and popliteal artery disease only following OMT.¹⁴ Smoking cessation, a significant focus of OMT, is an important modifiable risk factor. Following smoking cessation, a slow decline in CVD risk is observed over decades.⁵⁴ Our data support these findings, demonstrating no risk reduction in the early postprocedural rate of neurologic or cardiovascular complications. However, smoking within 1 year of a procedure was associated with an increase in early postprocedural complications. Although our data did not permit direct testing of PAD-specific smoking cessation, we did confirm the benefits of more than 1 year of smoking cessation with a 29% reduction in early postprocedural complications (adjusted OR, 0.71; 95% CI, 0.61-0.83). Surprisingly, the largest relative risk reduction occurred in those undergoing endovascular interventions. Our data highlight the persistently and unacceptably high rate of smoking in patients with IC and the significant risk associated with smoking on revascularization outcomes across all modalities and strongly suggest patients should receive comprehensive smoking cessation counseling prior to revascularization (including endovascular modalities). Further investigation is required to understand the duration of smoking cessation required to minimize the risks of early postprocedural complications, long-term CVD morbidity and mortality, and maximize functional outcomes.

Limitations

Our investigation has limitations. First, VASQIP data are abstracted from billing information and medical records reviewed in full by trained nurse reviewers and are at risk of measurement error. A specific set of data are abstracted, and the prescription status for lipid lowering, antiplatelet, and antithrombotic therapies, plaque morphology, and postprocedural outcomes beyond 30 days are not captured. However, in sensitivity analyses, the E-value suggests unmeasured confounding is unlikely to explain the entirety of the observed association. Second, smoking status is mainly self-reported, which may result in misclassification. Third, because our data could not discern if smoking cessation was a part of PAD-related OMT, smoking status may be a surrogate indicator of other healthy behaviors (eg, cessation may represent patient compliance). Fourth, smoking increases the onset and may alter the underlying pathophysiology of comorbid conditions. These factors are captured in the differences among those who did and did not match, possibly limiting the generalization of our matched analysis. However, multivariable modeling of the full cohort eligible for matching demonstrated similar point estimates, suggesting accuracy. Fifth, the VASQIP provides a representative yet incomplete sample of all surgical procedures, with an unquantified portion of interventions not captured. Therefore, the yearly incidence of procedures for IC cannot be fully quantified.⁵⁵ Sixth, the associations between smoking, sex, and postprocedural complications cannot be adequately assessed in the dominantly male VASQIP sample.

Conclusions

In this cohort study, following elective open and endovascular revascularization for IC, smoking was associated with an increased risk of any and respiratory-specific early postprocedural complications. Further, smoking within 1 year of a procedure was associated with an increase in complications, especially for endovascular interventions. Our findings support the strong societal guidelines emphasizing OMT and call for using comprehensive OMT programs that include smoking cessation therapies prior to revascularization for IC to reduce both early postprocedural complications and long-term CVD morbidity and mortality.

Supplement.

eTable 1. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10-CM procedure diagnosis codes identifying intermittent claudication.

eTable 2. Current Procedural Terminology codes, descriptions, and Operative Stress Scores identifying revascularization procedures.

eTable 3. Postprocedural complications quantified within the Veterans Affairs Surgical Quality Improvement Program (VASQIP) and their associated categorization—by major or minor complication.

eTable 4. Postprocedural complications quantified within the Veterans Affairs Surgical Quality Improvement Program (VASQIP) and their associated categorization—by organ system.

eTable 5. Preprocedural variable missingness for all patients eligible for propensity score matching.

eTable 6. Demographic, patient, procedure, and surgical site characteristics included in propensity score matching.

eTable 7. Baseline characteristics of smoking (n = 7820) and nonsmoking (n = 6530) patients who were included (n = 8420) and excluded (n = 7049) from propensity sore matching.

eTable 8. Multivariable logistic regression for any postprocedural complication among all patients eligible for propensity score matching (N = 14 350).

eTable 9. Patient characteristics by smoking timeframe (N = 12 794).

eFigure 1. Cohort identification.

eFigure 2. Revascularization for claudication by intervention (A) and by smoking status (B) per year in the full cohort eligible for propensity score matching.

eFigure 3. Distribution of propensity score before (N = 14 350) and after (n = 7710) propensity score matching.

eFigure 4. Standardized percentage bias across covariates among matched (n = 7710) and unmatched (n = 7301) cohorts.

Click here for additional data file.^{(1.4MB, pdf)}

References

1.Brown DW. Smoking prevalence among US veterans. J Gen Intern Med. 2010;25(2):147-149. doi: 10.1007/s11606-009-1160-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Fowkes FG, Murray GD, Butcher I, et al. ; Ankle Brachial Index Collaboration . Ankle brachial index combined with Framingham Risk Score to predict cardiovascular events and mortality: a meta-analysis. JAMA. 2008;300(2):197-208. doi: 10.1001/jama.300.2.197 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Sigvant B, Lundin F, Wahlberg E. The risk of disease progression in peripheral arterial disease is higher than expected: a meta-analysis of mortality and disease progression in peripheral arterial disease. Eur J Vasc Endovasc Surg. 2016;51(3):395-403. doi: 10.1016/j.ejvs.2015.10.022 [DOI] [PubMed] [Google Scholar]
4.Brevetti G, Giugliano G, Brevetti L, Hiatt WR. Inflammation in peripheral artery disease. Circulation. 2010;122(18):1862-1875. doi: 10.1161/CIRCULATIONAHA.109.918417 [DOI] [PubMed] [Google Scholar]
5.Anand SS. Smoking: a dual pathogen for arterial and venous thrombosis. Circulation. 2017;135(1):17-20. doi: 10.1161/CIRCULATIONAHA.116.025024 [DOI] [PubMed] [Google Scholar]
6.Selvarajah S, Black JH III, 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(5):1308-1314. doi: 10.1016/j.jvs.2013.12.011 [DOI] [PubMed] [Google Scholar]
7.Resnick HE, Lindsay RS, McDermott MMG, et al. Relationship of high and low ankle brachial index to all-cause and cardiovascular disease mortality: the Strong Heart Study. Circulation. 2004;109(6):733-739. doi: 10.1161/01.CIR.0000112642.63927.54 [DOI] [PubMed] [Google Scholar]
8.Jones DW, Goodney PP, Eldrup-Jorgensen J, et al. ; Vascular Study Group of New England . Active smoking in claudicants undergoing lower extremity bypass predicts decreased graft patency and worse overall survival. J Vasc Surg. 2018;68(3):796-806.e1. doi: 10.1016/j.jvs.2017.12.044 [DOI] [PubMed] [Google Scholar]
9.Young JC, Paul NJ, Karatas TB, et al. Cigarette smoking intensity informs outcomes after open revascularization for peripheral artery disease. J Vasc Surg. 2019;70(6):1973-1983.e5. doi: 10.1016/j.jvs.2019.02.066 [DOI] [PubMed] [Google Scholar]
10.Kalbaugh CA, Gonzalez NJ, Luckett DJ, et al. The impact of current smoking on outcomes after infrainguinal bypass for claudication. J Vasc Surg. 2018;68(2):495-502.e1. doi: 10.1016/j.jvs.2017.10.091 [DOI] [PubMed] [Google Scholar]
11.Duncan MS, Freiberg MS, Greevy RA Jr, Kundu S, Vasan RS, Tindle HA. Association of smoking cessation with subsequent risk of cardiovascular disease. JAMA. 2019;322(7):642-650. doi: 10.1001/jama.2019.10298 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Redgrave JNE, Lovett JK, Rothwell PM. Histological features of symptomatic carotid plaques in relation to age and smoking: the Oxford Plaque study. Stroke. 2010;41(10):2288-2294. doi: 10.1161/STROKEAHA.110.587006 [DOI] [PubMed] [Google Scholar]
13.Conte MS, Pomposelli FB, Clair DG, et al. ; Society for Vascular Surgery Lower Extremity Guidelines Writing Group; Society for Vascular Surgery . 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(3)(suppl):2S-41S. doi: 10.1016/j.jvs.2014.12.009 [DOI] [PubMed] [Google Scholar]
14.Bailey SR, Beckman JA, Dao TD, et al. ACC/AHA/SCAI/SIR/SVM 2018 appropriate use criteria for peripheral artery intervention: a report of the American College of Cardiology Appropriate Use Criteria Task Force, American Heart Association, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, and Society for Vascular Medicine. J Am Coll Cardiol. 2019;73(2):214-237. doi: 10.1016/j.jacc.2018.10.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Arya S, Khakharia A, Binney ZO, et al. Association of statin dose with amputation and survival in patients with peripheral artery disease. Circulation. 2018;137(14):1435-1446. doi: 10.1161/CIRCULATIONAHA.117.032361 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Rothenberg KA, Stern JR, George EL, et al. Association of frailty and postoperative complications with unplanned readmissions after elective outpatient surgery. JAMA Netw Open. 2019;2(5):e194330. doi: 10.1001/jamanetworkopen.2019.4330 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Patel KK, Jones PG, Ellerbeck EF, et al. Underutilization of evidence-based smoking cessation support strategies despite high smoking addiction burden in peripheral artery disease specialty care: insights from the international PORTRAIT registry. J Am Heart Assoc. 2018;7(20):e010076. doi: 10.1161/JAHA.118.010076 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Levin SR, Farber A, Cheng TW, et al. Patients undergoing interventions for claudication experience low perioperative morbidity but are at risk for worsening functional status and limb loss. J Vasc Surg. 2020;72(1):241-249. doi: 10.1016/j.jvs.2019.06.062 [DOI] [PubMed] [Google Scholar]
19.Williams CR, Jellison A, Martin L, et al. Optimal medical management before lower extremity bypass for claudication in the veteran population. J Vasc Surg. 2019;69(2):545-554. doi: 10.1016/j.jvs.2018.05.222 [DOI] [PubMed] [Google Scholar]
20.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495-1499. doi: 10.1016/j.ijsu.2014.07.013 [DOI] [PubMed] [Google Scholar]
21.Massarweh NN, Kaji AH, Itani KMF. Practical guide to surgical data sets: Veterans Affairs Surgical Quality Improvement Program (VASQIP). JAMA Surg. 2018;153(8):768-769. doi: 10.1001/jamasurg.2018.0504 [DOI] [PubMed] [Google Scholar]
22.Khuri SF, Daley J, Henderson W, et al. ; National VA Surgical Quality Improvement Program . The Department of Veterans Affairs’ NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care. Ann Surg. 1998;228(4):491-507. doi: 10.1097/00000658-199810000-00006 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Dormuth CR, Patrick AR, Shrank WH, et al. Statin adherence and risk of accidents: a cautionary tale. Circulation. 2009;119(15):2051-2057. doi: 10.1161/CIRCULATIONAHA.108.824151 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Arya S, Varley P, Youk A, et al. Recalibration and external validation of the Risk Analysis Index: a surgical frailty assessment tool. Ann Surg. 2020;272(6):996-1005. doi: 10.1097/SLA.0000000000003276 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Shinall MC Jr, Arya S, Youk A, et al. Association of preoperative patient frailty and operative stress with postoperative mortality. JAMA Surg. 2020;155(1):e194620. doi: 10.1001/jamasurg.2019.4620 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Yan Q, Kim J, Hall DE, et al. Association of frailty and the Expanded Operative Stress Score with preoperative acute serious conditions, complications and mortality in males compared to females: a retrospective observational study. Ann Surg. Published online June 25, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Rothenberg KA, George EL, Trickey AW, et al. Assessment of the risk analysis index for prediction of mortality, major complications, and length of stay in patients who underwent vascular surgery. Ann Vasc Surg. 2020;66:442-453. doi: 10.1016/j.avsg.2020.01.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Vemulapalli S, Carroll JD, Mack MJ, et al. Procedural volume and outcomes for transcatheter aortic-valve replacement. N Engl J Med. 2019;380(26):2541-2550. doi: 10.1056/NEJMsa1901109 [DOI] [PubMed] [Google Scholar]
29.Weeks WB, Kazis LE, Shen Y, et al. Differences in health-related quality of life in rural and urban veterans. Am J Public Health. 2004;94(10):1762-1767. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Veterans Health Administration . Office of Rural Health Rurality Calculator. Accessed January 10, 2020. http://vaww.vashare.vha.va.gov/sites/ruralhealth/Pages/Rurality.aspx
31.US Department of Veterans Affairs . Veterans Health Administration Operative Complexity. Accessed December 1, 2020. https://www.va.gov/health/surgery/
32.Ghaferi AA, Birkmeyer JD, Dimick JB. Variation in hospital mortality associated with inpatient surgery. N Engl J Med. 2009;361(14):1368-1375. doi: 10.1056/NEJMsa0903048 [DOI] [PubMed] [Google Scholar]
33.Brookhart MA, Wyss R, Layton JB, Stürmer T. Propensity score methods for confounding control in nonexperimental research. Circ Cardiovasc Qual Outcomes. 2013;6(5):604-611. doi: 10.1161/CIRCOUTCOMES.113.000359 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Austin PC. Some methods of propensity-score matching had superior performance to others: results of an empirical investigation and Monte Carlo simulations. Biom J. 2009;51(1):171-184. doi: 10.1002/bimj.200810488 [DOI] [PubMed] [Google Scholar]
35.Leuven E, Sianesi B. PSMATCH2: Stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. Accessed October 1, 2020. https://ideas.repec.org/c/boc/bocode/s432001.html
36.Reitz KM, Marroquin OC, Zenati MS, et al. Association between preoperative metformin exposure and postoperative outcomes in adults with type 2 diabetes. JAMA Surg. 2020;155(6):e200416. doi: 10.1001/jamasurg.2020.0416 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Haukoos JS, Lewis RJ. The propensity score. JAMA. 2015;314(15):1637-1638. doi: 10.1001/jama.2015.13480 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Bernaudin F, Verlhac S, Peffault de Latour R, et al. ; DREPAGREFFE Trial Investigators . Association of matched sibling donor hematopoietic stem cell transplantation with transcranial Doppler velocities in children with sickle cell anemia. JAMA. 2019;321(3):266-276. doi: 10.1001/jama.2018.20059 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-value. Ann Intern Med. 2017;167(4):268-274. doi: 10.7326/M16-2607 [DOI] [PubMed] [Google Scholar]
40.Mathur MB, Ding P, Riddell CA, VanderWeele TJ. Web site and R package for computing E-values. Epidemiology. 2018;29(5):e45-e47. doi: 10.1097/EDE.0000000000000864 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Hall DE, Arya S, Schmid KK, et al. Development and initial validation of the Risk Analysis Index for measuring frailty in surgical populations. JAMA Surg. 2017;152(2):175-182. doi: 10.1001/jamasurg.2016.4202 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Reitsma MB, Fullman N, Ng M, et al. ; GBD 2015 Tobacco Collaborators . Smoking prevalence and attributable disease burden in 195 countries and territories, 1990-2015: a systematic analysis from the Global Burden of Disease Study 2015. Lancet. 2017;389(10082):1885-1906. doi: 10.1016/S0140-6736(17)30819-X [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Cacoub PP, Abola MTB, Baumgartner I, et al. ; REACH Registry Investigators . Cardiovascular risk factor control and outcomes in peripheral artery disease patients in the Reduction of Atherothrombosis for Continued Health (REACH) registry. Atherosclerosis. 2009;204(2):e86-e92. doi: 10.1016/j.atherosclerosis.2008.10.023 [DOI] [PubMed] [Google Scholar]
44.Glance LG, Kellermann AL, Osler TM, et al. Hospital readmission after noncardiac surgery: the role of major complications. JAMA Surg. 2014;149(5):439-445. doi: 10.1001/jamasurg.2014.4 [DOI] [PubMed] [Google Scholar]
45.Mills E, Eyawo O, Lockhart I, Kelly S, Wu P, Ebbert JO. Smoking cessation reduces postoperative complications: a systematic review and meta-analysis. Am J Med. 2011;124(2):144-154.e8. doi: 10.1016/j.amjmed.2010.09.013 [DOI] [PubMed] [Google Scholar]
46.Grønkjær M, Eliasen M, Skov-Ettrup LS, et al. Preoperative smoking status and postoperative complications: a systematic review and meta-analysis. Ann Surg. 2014;259(1):52-71. doi: 10.1097/SLA.0b013e3182911913 [DOI] [PubMed] [Google Scholar]
47.Hawn MT, Houston TK, Campagna EJ, et al. The attributable risk of smoking on surgical complications. Ann Surg. 2011;254(6):914-920. doi: 10.1097/SLA.0b013e31822d7f81 [DOI] [PubMed] [Google Scholar]
48.Winkleby MA, Kraemer HC, Ahn DK, Varady AN. Ethnic and socioeconomic differences in cardiovascular disease risk factors: findings for women from the Third National Health and Nutrition Examination Survey, 1988-1994. JAMA. 1998;280(4):356-362. doi: 10.1001/jama.280.4.356 [DOI] [PubMed] [Google Scholar]
49.Elfassy T, Swift SL, Glymour MM, et al. Associations of income volatility with incident cardiovascular disease and all-cause mortality in a US cohort. Circulation. 2019;139(7):850-859. doi: 10.1161/CIRCULATIONAHA.118.035521 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Arya S, Kim SI, Duwayri Y, et al. Frailty increases the risk of 30-day mortality, morbidity, and failure to rescue after elective abdominal aortic aneurysm repair independent of age and comorbidities. J Vasc Surg. 2015;61(2):324-331. doi: 10.1016/j.jvs.2014.08.115 [DOI] [PubMed] [Google Scholar]
51.Rothenberg KA, George EL, Barreto N, et al. Frailty as measured by the Risk Analysis Index is associated with long-term death after carotid endarterectomy. J Vasc Surg. 2020;72(5):1735-1742.e3. doi: 10.1016/j.jvs.2020.01.043 [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Varley PR, Borrebach JD, Arya S, et al. Clinical utility of the Risk Analysis Index as a prospective frailty screening tool within a multi-practice, multi-hospital integrated healthcare system. Ann Surg. Published online February 28, 2020. doi: 10.1097/SLA.0000000000003808 [DOI] [PubMed] [Google Scholar]
53.Chen SL, Whealon MD, Kabutey NK, Kuo IJ, Sgroi MD, Fujitani RM. Outcomes of open and endovascular lower extremity revascularization in active smokers with advanced peripheral arterial disease. J Vasc Surg. 2017;65(6):1680-1689. doi: 10.1016/j.jvs.2017.01.025 [DOI] [PubMed] [Google Scholar]
54.Cahill K, Lindson-Hawley N, Thomas KH, Fanshawe TR, Lancaster T. Nicotine receptor partial agonists for smoking cessation. Cochrane Database Syst Rev. 2016;(5):CD006103. doi: 10.1002/14651858.CD006103.pub7 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Shunk KA, Zimmet J, Cason B, Speiser B, Tseng EE. Development of a Veterans Affairs hybrid operating room for transcatheter aortic valve replacement in the cardiac catheterization laboratory. JAMA Surg. 2015;150(3):216-222. doi: 10.1001/jamasurg.2014.1404 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials