Association Between Coronavirus Disease 2019 Vaccination and Mortality After Major Operations.

OBJECTIVE: The objective of this study was to compare postoperative 90-day mortality between (1) fully vaccinated patients with COVID-19-positive and negative diagnosis, and (2) vaccinated and unvaccinated patients with COVID-19 positive diagnosis. BACKGROUND: Societal guidelines recommend postponing elective operations for at least 7 weeks in unvaccinated patients with preoperative coronavirus disease 2019 (COVID-19) infection. The role of vaccination in this infection-operation time risk is unclear. METHODS: We conducted a national US multicenter retrospective, matched cohort study spanning July 2021 to October 2022. Participants were included if they underwent a high-risk general, vascular, orthopedic, neurosurgery, or genitourinary surgery. All-cause mortality occurring within 90 days of the index operation was the primary outcome. Inverse probability treatment weighted propensity scores were used to adjust logistic regression models examining the independent and interactive associations between mortality, exposure status, and infection proximity. RESULTS: Of 3401 fully vaccinated patients in the 8-week preoperative period, 437 (12.9%) were COVID-19-positive. Unadjusted mortality rates were not significantly different between vaccinated patients with COVID-19 (22, 5.0%) and vaccinated patients without COVID-19 (99, 3.3%; P = 0.07). After inverse probability treatment weighted adjustment, mortality risk was not significantly different between vaccinated COVID-19-positive patients compared to vaccinated patients without COVID-19 (adjusted odds ratio = 1.38, 95% CI: 0.70, 2.72). The proximity of COVID-19 diagnosis to the index operation did not confer added mortality risk in either comparison cohort. CONCLUSIONS: Contrary to risks observed among unvaccinated patients, postoperative mortality does not differ between patients with and without COVID-19 when vaccinated against the severe acute respiratory syndrome coronavirus 2 virus and receiving a high-risk operation within 8 weeks of the diagnosis, regardless of operation timing relative to diagnosis.

Association Between Pack-Years and Smoking Cessation on Outcomes After Vascular Interventions.

INTRODUCTION: The influence of pack-year history and smoking cessation timing on postoperative morbidity and mortality in a highly comorbid cohort is uncertain. We examined whether the association between smoking and adverse postoperative events is modified by pack-year history and smoking cessation timing. METHODS: We collected single-institution, retrospective data from consecutive patients undergoing open operations for carotid, aortic, and infrainguinal arterial disease. Active smoking was defined as smoking on the day of the index surgical intervention. Duration of smoking cessation was calculated as the time between smoking cessation and index surgery. The primary outcome was a composite of 30-day mortality and morbidity. Logistic and time-to-event Cox regressions estimated associations with interactions between cessation duration and pack-year history for nonactive smokers. RESULTS: Between 2010 and 2019, 1087 patients underwent 1640 high-risk vascular interventions. Median pack-year history was 40.0 pack-years (interquartile range [IQR]: 20.0-60.0) among nonactive smokers and 46.0 pack-years (IQR: 31.0-61.0) among active smokers (P < 0.001). The median smoking cessation time was 15.5 y (IQR: 4.4-30.9). Smoking status did not independently predict an increased risk of postoperative mortality or morbidity (odds ratio [OR] = 0.99, P = 0.96). Among nonactive smokers, neither smoking cessation duration (OR = 0.99, P = 0.16) nor pack-year history (OR = 1.00, P = 0.88) were significantly associated with adverse events (interaction P = 0.11). CONCLUSIONS: After high-risk vascular interventions in a single institution, active smoking, cessation period, and pack-year history are not associated with an increased risk of postoperative morbidity or mortality-highlighting the benefit of cessation as independent of frequency or intensity.

Timing of a Major Operative Intervention After a Positive COVID-19 Test Affects Postoperative Mortality: Results From a Nationwide, Procedure-matched Analysis.

BACKGROUND: Studies indicate that coronavirus disease 2019 (COVID-19) infection before or soon after operations increases mortality, but they do not comment on the appropriate timing for interventions after diagnosis. OBJECTIVE: We sought to determine what the safest time would be for COVID-19 diagnosed patients to undergo major operative interventions. METHODS: High-risk operations, between January 2020 and May 2021, were identified from the Veterans Affairs COVID-19 Shared Data Resource. Current Procedural Terminology (CPT) codes were used to exact match COVID-19 positive cases (n=938) to negative controls (n=7235). Time effects were calculated as a continuous variable and then grouped into 2-week intervals. The primary outcome was 90-day, all-cause postoperative mortality. RESULTS: Ninety-day mortality in cases and controls was similar when the operation was performed within 9 weeks or longer after a positive test; but significantly higher in cases versus controls when the operation was performed within 7 to 8 weeks (12.3% vs 4.9%), 5 to 6 weeks (10.3% vs 3.3%), 3 to 4 weeks (19.6% vs 6.7%), and 1 to 2 weeks (24.7% vs 7.4%) from diagnosis. Among patients who underwent surgery within 8 weeks from diagnosis, 90-day mortality was 16.6% for cases versus 5.8% for the controls ( P <0.001). In this cohort, we assessed interaction between case status and any symptom ( P =0.93), and case status and either respiratory symptoms or fever ( P =0.29), neither of which were significant statistically. CONCLUSIONS: Patients undergoing major operations within 8 weeks after a positive test have substantially higher postoperative 90-day mortality than CPT-matched controls without a COVID-19 diagnosis, regardless of presenting symptoms.

Early experience with venous arterialization for limb salvage in no-option patients with chronic limb-threatening ischemia.

OBJECTIVE: Chronic limb-threatening ischemia (CLTI) is associated with adverse limb outcomes and increased mortality. However, a small subset of the CLTI population will have no feasible conventional methods of revascularization. In such cases, venous arterialization (VA) could provide an alternative for limb salvage. The objective of the present study was to review the outcomes of VA at our institution. METHODS: We performed a single-institution review of 41 patients who had been followed up prospectively and had undergone either superficial or deep VA. The data collected included patient demographics, comorbidities, VA technique (endovascular vs hybrid), and WIfI (wound, ischemia, and foot infection) limb staging. Data were collected at 1-month, 6-month, and 1-year intervals and included the following outcomes: patency, wound healing, major adverse limb events, major amputation, and death. Descriptive statistics were used for analysis. RESULTS: The study group included 41 patients who had undergone successful open hybrid superficial or deep endovascular VA; 21 (51.2%) had undergone a purely endovascular procedure and 20 (48.8%), hybrid VA. The WIfI clinical stage was as follows: stage 4, 33 (80.5%); stage 3, 6 (14.6%); and stage 2, 1 (2.4%). Of the 41 patients, 24 (58.5%) had completed follow-up at 6 months and 16 (39%) at 1 year. At 1 year, the VA primary patency was 28.6% (95% confidence interval [CI], 0.15%-0.43%), primary assisted patency was 44.3% (95% CI, 0.27%-0.60%), and secondary patency was 67% (95% CI, 0.49%-0.80%). The complete wound healing rate was 2.7% (n = 1) at 1 month, 62.5% (n = 15) at 6 months, and 18.8% (n = 3) at 1 year. Overall wound healing at 1 year was 46.3% (n = 19). The number of major adverse limb events at 1 year was 15 (36.5%) and included 8 reinterventions (19.5%) and 7 major amputations (17%). The number of deaths was zero (0%) at 1 month and four (19%) at 6 months. Two deaths (9.5%) were attributed to COVID-19 (coronavirus disease 2019). No further deaths had occurred within 1 year. The limb salvage survival probability at 1 year was 81%. CONCLUSIONS: These findings suggest that for a select subset of CLTI patients presenting with a high WIfI clinical limb stage and no viable options for conventional open or endovascular arterial revascularization, superficial and deep VA are feasible options to achieve limb salvage.

Treatment of Early Deep Wound Infections after Arterial Reconstructions: A Meta-Analysis and Survey.

BACKGROUND: Peripheral vascular graft infections, a serious concern after open lower extremity interventions, are treated using multiple strategies. Yet, there is no consensus on the optimal treatment. This study summarizes the literature and compares aggregate effect sizes between graft preservation with antibiotic beads and total graft excision. METHODS: Manuscripts published between 1972 and 2019 were systematically queried using Ovid Medline and PubMed. Studies were included if they described early (≤4 months of the index procedure) infection-related outcomes after extracavitary and infrainguinal arterial graft infections that were managed with antibiotic-loaded beads or total excisions. Outcomes assessed included the prevalence of graft preservation failure, reinfection, and major amputation. To examine current preferences on this subject, a voluntary, anonymous survey was administered to practicing members of the Society for Clinical Vascular Surgery. RESULTS: Six graft preservation studies (n = 147 patients) were included in the meta-analysis, based on PRISMA guidelines. The meta-analytic pooled proportion of patients with: (1) graft preservation failure was 0.09 (95% CI: 0.00, 0.46, I2 = 88.8%), (2) reinfection was 0.04 (95% CI: 0.00, 0.18, I2 = 79.7%), and (3) major amputation was 0.00 (95% CI: 0.00, 0.04, I2 = 0%). Five studies addressing total excisions were identified via the systematic review however, their combined sample size (n = 28 patients) impeded use of a meta-analysis. Ninety (19%) licensed surgeons participated in the survey. In a hemodynamically stable, nonseptic patient, 67% (60) of respondents routinely excise the graft, while 31% (28) prefer preservation. CONCLUSIONS: Study design and patient characteristic-related heterogeneity limited our ability to generate robust, clinical evidence-level outcome estimates. A prospective study is necessary to definitively establish the efficacy of antibiotic beads in the treatment and preservation of vascular graft infections.

Outcomes after Endovascular Stent Placement for Long-Segment Superficial Femoral Artery Lesions.

BACKGROUND: Endovascular intervention is commonly pursued as first-line management of symptomatic, long-segment superficial femoral artery (SFA) disease. The relative effectiveness and comparative long-term outcomes among bare metal stents (BMS), covered stents (CS), and drug-eluting stents (DES) for long-segment SFA lesions remain uncertain. METHODS: A retrospective cohort study identified patients with symptomatic SFA lesions measuring at least 15 cm in length who successfully received an endovascular stent (BMS, CS, or DES). The outcomes were patency, patient presentation upon stent occlusion, amputation-free survival (AFS), and all-cause mortality. Proportional hazards regressions and a multinomial logistic regression model were used to control for significant confounders. RESULTS: A total of 226 procedures were analyzed (BMS: 95 [42%]; CS: 74 [33%]; DES: 57 [25%]). There were no significant differences among the 3 stent types with respect to age, prevalence of either diabetes or end-stage renal disease, or smoking history. The median length of the SFA lesion varied across the cohorts (BMS: 28 cm [interquartile range, IQR 20-30]; CS: 26 cm [IQR 20-30]; DES: 20 cm [IQR 16-25]; P = 0.002). The unadjusted primary patency of BMS at 12, 24, and 48 month following index stent placement was 57%, 47%, and 44%, respectively. This is compared to 62%, 49%, and 42% for CS, and 81%, 66%, and 53% for DES, respectively (log-rank P = 0.044). In adjusted models, however, there were no significant differences in primary patency among the stent types. Compared to CS however, DES was associated with improved primary-assisted patency (hazard ratio [HR] for patency loss: 0.35, P = 0.008) and secondary patency (HR: 0.32, P = 0.011). Across the entire follow-up period, stent occlusions occurred in 38 (40%) BMS cases, 42 (57%) CS, and 11 (19%) DES (P < 0.001). Of these, acute limb ischemia (ALI) occurred in 2 (5%) BMS cases, 14 (33%) CS, and 1 (9%) DES (P = 0.010). After adjustment, the relative risk of presenting with ALI as opposed to claudication was 27 times greater among patients re-presenting with occluded CS compared to BMS (P = 0.020). There were no significant differences in AFS or all-cause mortality across the 3 cohorts. CONCLUSIONS: For long-segment SFA lesions, DES is associated with improved primary-assisted and secondary patency over long-term follow-up. In the event of stent occlusion, CS is associated with an increased risk of ALI.

Long-term outcomes of lower extremity graft preservation using antibiotic beads in patients with early deep wound infections after major arterial reconstructions.

OBJECTIVE: Bypass graft preservation with wound sterilization using serial antibiotic bead exchange has been described in patients presenting with deep wound infections after extremity bypass. The long-term benefits of this approach remain poorly understood. We examined whether graft preservation and wound sterilization with antibiotic beads affect amputation rates and patient survival. METHODS: Patients who underwent operations for aortoiliac or infrainguinal aneurysmal or occlusive arterial disease were retrospectively analyzed. The Infection group included those with patent vascular grafts who developed Szilagyi class II or III deep wound infections within 90 days of index reconstruction and had no evidence of anastomotic or arterial bleeding. All patients in the infection group were managed with graft preservation using serial antibiotic bead exchange every 3 to 5 days until wound cultures became negative. This group was compared with a contemporary group of controls who underwent similar interventions but did not develop wound infections postoperatively. The primary outcome was amputation-free survival, defined as survival without major amputation. Secondary outcomes included major amputations and the occurrence of anastomotic pseudoaneurysms necessitating repair. Inverse propensity score weighting was used for risk adjustment between the groups. RESULTS: Over an 8-year period, we treated 701 patients (infection, 68; controls, 633). Compared with controls, patients in the infection group had a higher body mass index (mean, 28.5 vs 26.3, P = .002) and more prosthetic conduits placed during the index reconstruction. Amputation-free survival for the infection vs the control group was 78 vs 76% at 2 years, 61 vs 66% at 4 years, and 51 vs 57% at 6 years postoperatively (log-rank test, P = .516). Freedom from major amputation for the infection vs the control group was 82 vs 86% at 2 years, 80 vs 82% at 4 years, and 80 vs 76% at 6 years postoperatively (log-rank test, P = .568). In the risk-adjusted model, the presence of treated infection did not affect amputation-free survival (hazard ratio, 0.82; P = .440) or major amputation (hazard ratio, 1.02; P = .949). Anastomotic pseudoaneurysms occurred only in the Infection group (4.4%; P = .001), and were treated with interposition grafts without complications. CONCLUSIONS: Bypass graft preservation with wound sterilization using serial antibiotic bead exchange is associated with excellent limb salvage and survival rates, similar to those of noninfected wounds. With the use of this preservation strategy, close follow-up for timely detection of anastomotic pseudoaneurysms is recommended.

Leg Amputations Among Texans Remote From Experienced Surgical Care.

BACKGROUND: Surgical outcomes may differ between low-volume and experienced hospitals. We sought to identify characteristics of remote patients-those living more than 50 miles from an experienced center-who underwent leg amputations for peripheral artery disease (PAD) and foot complications at low-volume and experienced hospitals and identify regions of Texas where such patients live. MATERIALS AND METHODS: Publicly available Texas hospitalization data from 2004 through 2009 were used to identify patients with PAD who underwent leg amputation for foot complications, including foot ulcers, foot infections, and gangrene. Geocoding was used to further identify a subset of remote patients and to estimate distances from zip code of residence to hospital in which care was received. RESULTS: Among all leg amputations, 850 (18.6%) were performed on patients classified as remote, and 3723 (81.4%) were performed on patients classified as nonremote. Compared with nonremote patients, remote patients were more often categorized as white and more frequently received Medicare and/or Medicaid. Of the subset of remote patients, those at low-volume hospitals were older, were less often categorized as Hispanic, more often had Medicaid coverage, were also more frequently admitted through the emergency department, and often had a foot infection compared with those at experienced centers. Geospatial analysis identified five concentrated geographic areas of remote patients who live more than 50 miles from an experienced center. CONCLUSIONS: These findings suggest travel distance may at least influence, if not constrain, the choice of hospital for patients with PAD and foot complications. Efforts to decrease leg amputations among remote patients should be focused on five specific geographic areas of Texas.

Wound Complications and Reoperations after Transtibial Amputation of the Leg.

BACKGROUND: Transtibial amputations (TTAs) of the leg have been associated with high rates of wound complications. We assessed outcomes of TTAs to determine if bundled interventions implemented at our hospital had an impact on lowering wound complications, including surgical site infections. METHODS: We assessed the impact of a surgical site infection prevention bundle (negative-pressure wound therapy, minimizing the use of staples, and a decontamination protocol for methicillin-resistant Staphylococcus aureus) on 90-day wound complications. The year of implementation of the prevention bundle was excluded, and the pre-eras and posteras were defined as the four-year period before and after implementation. The study sample consisted of a single-center cohort, with TTA cases identified using operating room scheduling software. RESULTS: A total of 182 TTAs were performed: 110 in the pre-era and 72 in the postera. The wound complication rate decreased from 22 to 17% despite fewer two-stage operations, less imaging to identify peripheral artery disease, and an increased proportion of patients with end-stage renal disease. Wound complications and revision to a higher level of amputation were more associated with indication (especially no-option peripheral artery disease with ischemic rest pains) than with any particular aspect of surgical technique. The use of drains was associated with reoperations but not higher level revision. CONCLUSIONS: Higher rates of wound complications and revision to a higher level of amputations should be expected among patients with no-option peripheral artery disease with ischemic rest pains undergoing TTAs. Drains should be avoided.

Effect of Postoperative Permissive Anemia and Cardiovascular Risk Status on Outcomes After Major General and Vascular Surgery Operative Interventions.

OBJECTIVES: To determine the effect of postoperative permissive anemia and high cardiovascular risk on postoperative outcomes. METHODS: The Veterans Affairs Surgical Quality Improvement Program and Corporate Data Warehouse databases were queried for patients who underwent major vascular or general surgery operations. The status of cardiovascular risk was assessed by calculating the Revised Cardiac Risk Index. Primary endpoint was a composite of mortality, myocardial infarction, acute renal failure, coronary revascularization, or stroke within 90 days postoperatively. RESULTS: We analyzed 142,510 procedures performed from 2000 to 2015. Postoperative anemia was the strongest independent predictor of the primary endpoint whose odds increased by 43% for every g/dL drop in postoperative nadir Hb [95% confidence interval (95% CI): 41-45]. Cardiac risk status as described by the RCRI also independently predicted the primary endpoint, with an additive effect particularly evident at postoperative nadir Hb values below 10 gm/dL. Postoperative anemia, after age, was the second strongest independent predictor of long-term (12 years) mortality (hazard ratio: 1.18, 95% CI: 1.17-1.19). CONCLUSION: Postoperative anemia is strongly associated with postoperative ischemic events, 90-day mortality, and long-term mortality. Restrictive transfusion should be used cautiously after major general and vascular operations, particularly in patients at a high cardiovascular risk.

A Statistical Model-driven Surgical Case Scheduling System Improves Multiple Measures of Operative Suite Efficiency: Findings From a Single-center, Randomized Controlled Trial.

OBJECTIVE: We sought to determine whether a data-driven scheduling approach improves Operative Suite (OS) efficiency. BACKGROUND: Although efficient use of the OS is a critical determinant of access to health care services, OS scheduling methodologies are simplistic and do not account for all the available characteristics of individual surgical cases. METHODS: We randomly scheduled cases in a single OS by predicting their length using either the historical mean (HM) duration of the most recent 4 years; or a regression modeling (RM) system that accounted for operative and patient characteristics. The primary endpoint was the imprecision in prediction of the end of the operative day. Secondary endpoints included measures of OS efficiency; personnel burnout captured by the Maslach Burnout Inventory; and a composite endpoint of 30-day mortality, myocardial infarction, wound infection, bleeding, amputation, or reoperation. RESULTS: Two hundred and seven operative days were allocated to scheduling with either the RM or the HM methodology. Mean imprecision in predicting the end of the operative day was higher with the HM approach (30.8 vs 7.2 minutes, P = 0.024). RM was associated with higher throughput (379 vs 356 cases scheduled over the course of the study, P = 0.04). The composite rate of adverse 30-day events was similar (2.2% vs 3.2%, P = 0.44). The mean depersonalization score was higher (3.2 vs 2.0, P = 0.044), and mean personal accomplishment score was lower during HM weeks (37.5 vs 40.5, P = 0.028). CONCLUSIONS: Compared to the HM scheduling approach, the proposed data-driven RM scheduling methodology improves multiple measures of OS efficiency and OS personnel satisfaction without adversely affecting clinical outcomes.

Ischemia-induced lower extremity neurologic impairment after fenestrated endovascular aneurysm repair.

OBJECTIVE: Placement of large sheaths in the iliac system during fenestrated endovascular aneurysm repair (FEVAR) leads to lower extremity (LE) ischemia that can be associated with serious neurologic complications. We sought to determine the effect of LE ischemic time on neurologic impairment after FEVAR. METHODS: Consecutive patients who underwent FEVAR at a single institution were analyzed. LE ischemic time was calculated from the time of large sheath (≥18F) insertion to the time of sheath removal from the iliac arteries that led to continuous LE ischemia. The primary outcome was neurologic impairment defined as any new sensory or motor deficit in either LE. Outcomes were analyzed using descriptive statistics and modeled with logistic regression with interaction terms. Each individual LE was used as a unit of analysis. RESULTS: We examined 101 patients (202 lower extremities) who underwent FEVAR over a 5-year period. The median LE ischemic time was 2.75 hours (range, 0.8-5.2 hours). Neurologic impairment developed in 18 extremities (9%). Of those, 12 (67%) developed mild sensory loss, 6 (33%) complete sensory loss, 4 (22%) loss of proprioception, and 2 (11%) motor dysfunction. Sensory deficit was permanent in four limbs (2%) and motor dysfunction in one limb (0.5%). In all other cases, the neurologic examination returned to baseline by postoperative day 15. Duration of LE ischemic time (odds ratio, 6.3; 95% confidence interval, 3.1-12.4; P < .001) and common iliac artery (CIA) stenosis to a lumen of 8 mm or less (odds ratio, 2.7; 95% confidence interval, 1.5-7.3; P = .002) were independent predictors for the development of neurologic impairment. An interaction term between LE ischemic time and CIA stenosis was statistically significant (P = .042), indicating that the presence of CIA stenosis modifies the effect of LE ischemic time. In those with CIA stenosis to a lumen of 8 mm or less, the risk of neurologic impairment increased rapidly after 2.5 hours of LE ischemia, and became nearly certain after 4 hours of ischemic time. By contrast, patients without CIA stenosis tolerated longer ischemic times and demonstrated a less steep increase in the risk for LE neurologic impairment. CONCLUSIONS: LE neurologic impairment after FEVAR is strongly associated with LE ischemic time and CIA occlusive disease to a lumen of 8 mm or less. Our data indicate that, when the LE ischemic time is expected to exceed 2.5 hours (in patients with CIA stenosis) or 3 hours (in patients without CIA stenosis), measures to ensure LE perfusion should be given consideration.

Health Care Resource Distribution of Texas Counties With High Rates of Leg Amputations.

BACKGROUND: Lower extremity amputation rates associated with peripheral arterial disease in Texas are high and vary disproportionately among different populations. We sought to assess the impact of socioeconomic status and health care resource distribution on the geographic prevalence of lower extremity amputation in Texas counties. MATERIALS AND METHODS: We collated 2005-2009 data on all 254 Texas counties. The primary outcome was the number of nontraumatic lower extremity amputations. Population-adjusted regressions identified factors that could explain increasing amputation rates. RESULTS: We identified 33 counties with population-weighted amputation rates in the highest 25%. These counties had higher rates of diabetes, larger populations of people categorized as black, fewer health care resources, and lower health care utilization. In the presence of more emergency room visits, dual Medicare/Medicaid eligibility decreased total amputations. In counties with more than 70% rural communities, additional primary care providers also significantly decreased amputations per 100,000 residents (mean difference = -0.12, 95% confidence interval: -0.23, -0.0008). CONCLUSIONS: Policy-driven strategic health care resource allocation at the local level may benefit patients in high-need, low-resource areas and promote a reduction in amputations.

Delayed gratification and adherence to exercise among patients with claudication.

Few studies have explicitly identified factors that explain an individual's willingness to engage in community-based exercise for claudication. Identifying the unique characteristics of those inclined toward physical activity would inform interventions that encourage walking. We examined the utility of behavioral economics-related concepts in understanding walking among Veterans with claudication. Patients who received care at the Michael E. DeBakey Veterans Affairs Medical Center in Houston, Texas, were surveyed on symptom severity, behavioral economics, stress, and depression. The primary outcome was a binary variable measuring current walking for exercise and defined as walking for at least 30 minutes every day. Multivariable logistic regression models were used to identify variables, both clinically and statistically significant, at a p-value < 0.05. Between April 2017 and March 2018, we received 148 (30%) responses. A total of 35% (n = 51) of respondents indicated that they walked recreationally for exercise compared to 65% (n = 94) who did not. Characteristics that were significantly associated with walking included regularly saving money (adjusted odds ratio (aOR) = 10.7, p = 0.001), seeking complex problem-solving (aOR = 0.12, p = 0.002), and severe symptoms (aOR = 0.24, p = 0.017). Individuals describing a preference for the future rather than immediate benefit also reported currently walking for exercise. Defining the characteristics of those who exercise may help inform strategies designed to increase walking among those who do not adhere to recommendations.

Successful Smoking Cessation Associated with Walking Behavior in Patients with Claudication.

BACKGROUND: Behavioral economics theories suggest that a preference for delayed benefits promotes positive behavioral change, a concept relevant to both smoking cessation and community-based exercise regimens for claudication. Given the high rate of smoking among older veterans, we were interested in examining the association between smoking cessation, exercise regimen adherence, and preferences for delayed versus immediate benefits. METHODS: Between April 2017 and March 2018, patients with claudication at the Michael E. DeBakey Veterans Affairs Medical Center in Houston, Texas, received questionnaires collecting information on social, behavioral, and psychological characteristics. A dual validation system, via the electronic medical record and survey data, measured the primary outcome-smoking cessation versus current smoking. Self-reported physical activity was measured through the validated Ainsworth's compendium of Physical Activities and binary survey questions. The Walking Impairment and Barratt's Impulsivity Questionnaires measured subjective symptom severity and behavioral economics factors, respectively. Multivariable, logistic regression models identified significant associations. RESULTS: The survey was mailed to 500 patients who met the eligibility criteria. We received responses from 148 individuals (30%), and 67 of 141 (48%) indicated that they had successfully quit smoking. In unadjusted comparisons, the median cognitive complexity score in the smoking cessation group was higher than that in the current smoking group. A greater proportion of patients who reported walking for exercise (n = 46) also reported successful smoking cessation (28/46, 61%). Among those who were not walking for exercise (n = 88), more individuals reported current smoking (49/88, 56%). In the multivariable model, individuals who had successfully stopped smoking were older (odds ratio [OR]: 7.59, P < 0.001), more likely to walk for exercise (OR: 3.94, P = 0.009), more interested in the future than in the present (OR: 1.73, P = 0.030), and more likely to regularly save money (OR: 3.49, P = 0.046). CONCLUSIONS: We found that participants who reported successful smoking cessation were more likely to report walking for exercise. Our findings suggest that adherence to walking may be less challenging for patients who have already successfully implemented and continue to implement another beneficial health behavior (smoking cessation). Patients with claudication who are current smokers may be less likely to adopt exercise recommendations.

The impact of foot infection on infrainguinal bypass outcomes in patients with chronic limb-threatening ischemia.

BACKGROUND: Despite advances in endovascular therapy, infrainguinal bypass continues to play a major role in achieving limb salvage. In this study, we sought to compare outcomes of infrainguinal bypass in patients with limb-threatening ischemia who presented with or without foot infection. METHODS: We conducted a retrospective cohort study of patients who underwent infrainguinal bypass for chronic limb-threatening ischemia at a single institution. End points of interest included long-term mortality, 45-day readmission, postoperative length of stay (LOS), major amputation, and time to wound healing. Multivariable Cox, logistic, and robust regressions were used to model time to event outcomes, readmission rates, and LOS. RESULTS: There were 454 infrainguinal bypass procedures analyzed. Demographics and baseline characteristics were similar, except congestive heart failure and diabetes were more common in the infection group. Presence of foot infection had no impact on mortality (hazard ratio [HR], 0.78; P = .243). Significant predictors of long-term mortality included increasing age, hypoalbuminemia, and congestive heart failure; preoperative use of clopidogrel was protective. Presence of foot infection was an independent predictor of major amputation. In the multiple regression model, the presence of foot infection was independently associated with amputation rate (HR, 2.14; 95% confidence interval, 1.42-3.22; P < .001); use of venous conduit and increasing age and body mass index were protective. Foot infection was an independent predictor of prolonged LOS (mean LOS was 1.54 days longer in patients with vs those without infection; P = .001). Other independent predictors of prolonged LOS included intraoperative blood loss and reoperation; history of continuous preoperative aspirin use and normal baseline renal function and albumin levels were associated with decreased LOS. Readmission was influenced by reoperation (odds ratio [OR], 2.51; P < .001) but not by presence of foot infection (OR, 1.21; P = .349). There was a strong trend for prolonged wound healing time in patients with diabetes (HR, 1.58; P = .05) but not in those with foot infection (OR, 0.74; P = .36). CONCLUSIONS: Among patients requiring infrainguinal bypass for limb-threatening ischemia, infection was more common in patients with diabetes and was a significant predictor of major amputation and prolonged LOS. Infection was not predictive of mortality, wound healing time, or readmission. These findings lend support to the inclusion of infection in risk stratification schemes for patients with chronic limb-threatening ischemia, as recommended in the Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification system, because of its adverse impacts on limb salvage.

Missing Data in Surgical Data Sets: A Review of Pertinent Issues and Solutions.

Incomplete data is a common problem in research studies. Methods to address missing observations in a data set have been extensively researched and described. Disseminating these methods to the greater research community is an ongoing effort. In this article, we describe some of the basic principles of missing data and identify practical, commonly used methods of adjustment relevant to surgical data sets. Through an example data set, we compare models generated through complete case analysis, single imputation (SI), and multiple imputation (MI). We also provide information on the steps to conduct MI using Stata IC. In our comparisons, we found that differences in odds ratios were greatest between the results from complete case analysis compared to the SI and MI models indicating that in this case the reduction in statistical power has a non-negligible effect on the parameter estimates. Odds ratio estimates from the SI and MI methods were largely similar. In some instances, when compared to the MI method, the SI method tended to overestimate effect sizes. While in this example the differences in odds ratios do not vary greatly between the SI and MI methods, there are clear indications supporting the use of MI over SI. By describing the issues surrounding missing data and the available options for adjustment, we hope to encourage the use of robust imputation methods for missing observations.

Effect of postoperative anemia and baseline cardiac risk on serious adverse outcomes after major vascular interventions.

OBJECTIVE: Published data suggest that permissive anemia strategies that allow nadir hemoglobin (nHb) values of 7 g/dL or lower are safe in a variety of clinical settings. The appropriateness of these strategies in patients at high risk for adverse postoperative cardiac events remains unclear. We sought to determine the combined effect of postoperative nHb and cardiac risk status on major complications after vascular surgical interventions. METHODS: This was a single-institution retrospective analysis of consecutive patients who underwent elective open procedures for occlusive vascular disease and aneurysm repair, either open or endovascular. The Revised Cardiac Risk Index (RCRI) was used to assess baseline cardiac risk. Primary outcome was a composite end point of mortality or major ischemic events (myocardial infarction, stroke, acute kidney injury, or coronary revascularization) within 90 days from the index operation. Secondary outcomes included intensive care unit (ICU) length of stay (LOS) and 90-day respiratory complications (pneumonia, ventilator dependence for >48 hours postoperatively, or reintubation). Hierarchical multivariable regression was used to model each outcome with adjustment for age, type of operation, baseline comorbidities, and intraoperative covariates. RESULTS: We analyzed 2508 operations performed during 8 years in 2106 patients with a mean age of 67 years (range, 45-90 years). In the fully adjusted multivariable model, lower values of nHb increased the risk of the primary composite end point (odds ratio [OR], 1.24; P < .001, representing a 24% increase in the odds of the composite end point for each 1-g/dL increase in nHb). In the same model, RCRI class II (OR, 1.8; P < .001), class III (OR, 2.06; P < .0001), and class IV (OR, 2.35; P < .0001) were associated with progressively increasing odds of the composite end point compared with RCRI class I. An interaction term between transfusion and nHb was not significant statistically, indicating that the harmful effect of anemia was independent of blood transfusion. Lower values of nHb also increased the risk of respiratory complications (OR, 1.41; P = .002) and ICU LOS (average 2.6-day increase per 1-g/dL increase of nHb; P < .0001). CONCLUSIONS: Postoperative anemia increases the rate of early postoperative mortality and major ischemic events, particularly in patients at higher baseline cardiac risk. It also adversely affects respiratory complications and ICU LOS. Until a randomized trial definitively settles the issue, restrictive transfusion strategies should be practiced with caution in patients undergoing major vascular interventions.

The influence of pain-related beliefs on physical activity and health attitudes in patients with claudication: A pilot study.

We examined how pain beliefs are related to symptom severity, expectations of risk/benefits, and baseline physical activity among claudicants. Eligible patients at the Michael E DeBakey Veterans Affairs Medical Center were administered questionnaires that measured: fear-avoidance beliefs (Fear-Avoidance Beliefs Questionnaire [FABQ]), walking impairment, baseline physical activity, claudication type, and risk/benefit attitudes. Among 20 participants, the median age was 69 years (IQR: 66-75). In our efforts to understand how fear-avoidance beliefs influenced physical activity among people with claudication, we found that 12 out of 19 participants (63%) thought that the primary etiology of their pain was walking, while 18 (out of 20) (90%) people thought that walking would exacerbate their leg symptoms - suggesting that there was some confusion regarding the effects of walking on claudication. Those who expected that walking would benefit their symptoms more than surgery reported fewer fear-avoidance beliefs ( p=0.01), but those who believed that walking would make their leg pain worse expected greater benefit from surgery ( p=0.02). As symptom severity increased, fear-avoidance beliefs also increased ( p=0.001). The association between symptom severity and fear-avoidance beliefs indicates that as pain or impairment increases, the likelihood of avoiding behaviors that are thought to cause pain might also increase. Accounting for pain-related beliefs when recommending physical activity for claudication should be considered.

The impact of clinical, psychological, behavioral, social, and environmental factors on self-perceived symptom severity in a male cohort with intermittent claudication.

OBJECTIVE: To understand the relationship between self-perceived severity of intermittent claudication and various associated nonclinical factors, we examined how correlates in domains of physical activity (ie, clinical, psychological, behavioral, social, and environmental factors) relate to exertional limb symptoms. METHODS: A survey was administered to individuals with intermittent claudication during their initial outpatient assessment. The subjects' self-reported exertional limb symptom severity and classic-versus-atypical claudication classification was based on the Walking Impairment Questionnaire (WIQ) and San Diego Claudication Questionnaire (SDCQ), respectively. We evaluated psychosocial and environmental factors, osteoarthritis symptoms, health, behaviors, and beliefs. Logistic and linear regressions identified factors with a strong independent association with total WIQ scores and the SDCQs. RESULTS: A cohort of 102 subjects (99.0% male) was enrolled in the study. The median age was 65 years with a median ankle-brachial index of 0.69. Forty-three subjects (43%) had "typical" claudication per SDCQs. Individuals with atypical claudication were more likely to report higher Aberdeen Clinical Back Pain Questionnaire scores (odds ratio, 1.04; P = .04) and no depressive symptoms (odds ratio, 8.30; P = .03). Exertional limb symptom severity among the entire cohort was significantly associated with increasing osteoarthritis symptoms (P <.001), age (P = .02), a reserved personality (P = .008), and the belief that an exercise regimen would not improve symptoms (P = .005), self-perceived levels of boredom (P = .002), and the belief that exercise (P = .002) was the best way to improve symptoms were associated with decreased symptom severity. When restricted to those with atypical pain, significant factors associated with increasing exertional symptom severity included age greater than 60 years (P = .005), osteoarthritis (P = .02), alcohol use (P = .01), belief that exercise would not improve walking (P = .03), and difficulty walking around the neighborhood (P = .02). When restricted to those with classic claudication, significant factors associated with increasing exertional limb symptom severity included frequent pain or aching in the calves while walking or sitting (P = .03 [walking]; P = .01 [sitting]) and occasional morning joint stiffness (P = .007). Exertional limb symptom severity was also associated with high limitations at home (P = .003) and a belief that exercise would not improve walking (P = .005) among those with classic claudication. CONCLUSIONS: Symptom severity and type of pain are associated with a number of nonclinical factors. A multidomain approach, as indicated by the models above, would benefit the continuum of care for intermittent claudication, where management is integrated and coordinated among multiple lines of care.