Skip to main content
Skip main navigation

Bleeding Complications in Lower-Extremity Peripheral Vascular Interventions: Insights From the NCDR PVI RegistryFree Access

Peripheral

J Am Coll Cardiol Intv, 12 (12) 1140–1149
Sections

Central Illustration

Abstract

Objectives:

This study sought to assess periprocedural bleeding complications in lower-extremity peripheral vascular interventions (PVIs).

Background:

Few studies have examined the incidence, predictors, or outcomes of periprocedural bleeding after lower-extremity PVI.

Methods:

The study examined patients undergoing PVI at 76 hospitals in the National Cardiovascular Data Registry PVI registry from 2014 to 2016. Post-PVI major bleeding was defined as any overt bleeding with a hemoglobin (Hb) drop of ≥3 g/dl, any Hb decline of ≥4 g/dl, or blood transfusion in patients with pre-procedure Hb >8 g/dl within 72 h of their procedure. Hierarchical multivariable logistic regression was used to identify factors independently associated with post-PVI bleeding. The study also examined adjusted in-hospital mortality among patients with or without major bleeding complications.

Results:

Among 18,289 PVI procedures, major bleeding occurred in 744 (4.10%). Patient characteristics independently associated with bleeding included age, female sex, heart failure, pre-procedural hemoglobin <12 g/dl, nonelective PVI, and critical limb ischemia on presentation. Procedural characteristics associated with bleeding included nonfemoral vascular access, use of thrombolytic therapy, PVI of the aortoiliac segment, and multilesion interventions, whereas use of closure devices was associated with less bleeding. All-cause in-hospital mortality was higher in patients who experienced bleeding than in those who did not (6.60% vs. 0.30%; p < 0.001; adjusted hazard ratio: 10.9; 95% confidence interval: 6.9 to 17.0).

Conclusions:

Major bleeding occurred in 4.10% of lower-extremity PVI procedures and was associated with several patient and procedural characteristics, as well as in-hospital mortality. These insights can be incorporated into strategies to reduce periprocedural bleeding after PVI.

Introduction

Periprocedural bleeding at the time of percutaneous coronary intervention (PCI) is associated with increased morbidity and mortality and is often preventable (1–10). Few studies have examined the frequency of bleeding and its association with clinical outcomes in the setting of endovascular peripheral vascular intervention (PVI) (11–15). There is marked heterogeneity among existing studies regarding the definition of bleeding and its incidence, with reported rates ranging from 2.20% to 4.60% (11–15). To better define the incidence, predictors, and risks associated with major bleeding after PVI, we sought to describe periprocedural bleeding in the National Cardiovascular Data Registry (NCDR) PVI Registry, a large, contemporary, nationally representative registry of consecutive patients undergoing PVI at 76 hospitals across the United States, and identify patient and procedural characteristics associated with bleeding, as well as the association of bleeding with in-hospital mortality. Collectively, these data can inform future strategies to reduce the risks of bleeding and improve the safety of PVI.

Methods

Study population and data collection

The NCDR PVI Registry collects data on the demographic and clinical characteristics, comorbidities, treatments, and in-hospital outcomes of all patients undergoing percutaneous treatments for peripheral artery disease (PAD) at participating institutions. The data elements collected in the NCDR PVI Registry, along with their definitions, are available online (16).

This study analyzed a total of 19,537 PVI procedures at 76 U.S. sites participating in the NCDR PVI Registry from 2014 to 2016 (Online Figure 1). All patients >18 years of age undergoing lower-extremity PVI, including endovascular aortoiliac, femoropopliteal, and infrapopliteal interventions were included in the study. Patients with acute limb ischemia were excluded (n = 1,248), yielding a final analytic cohort of 18,289 patients.

Data definitions

The aim of the study was to assess post-PVI bleeding within 72 h of the procedure. Major bleeding was defined as overt bleeding with a hemoglobin (Hb) drop of ≥3 g/dl, any Hb decline ≥4 g/dl, or blood transfusion in patients with pre-procedure Hb >8 g/dl. This definition was not clinically adjudicated but was applied consistently to all patients in the registry after sites had been trained in the use of the NCDR data collection form and its definitions. The rationale for restricting the definition of bleeding associated with blood transfusions to only those with Hb >8 g/dl was to avoid including patients whose transfusions were performed to treat severe anemia, irrespective of the PVI procedure. This definition is derived from the components of bleeding definitions used in NCDR Cath-PCI and ACTION (Acute Coronary Treatment and Intervention Outcomes Network) registries (8,17). Post-procedural blood transfusion in patients with Hb >8 g/dl was derived from the Cath-PCI registry (8) and an absolute Hb decrease ≥4 g/dl was taken from the ACTION registry (17). Nonfemoral access used for PVI in the registry included popliteal, dorsalis pedis, posterior tibial, radial, and brachial access sites.

Statistical analysis

The demographic characteristics, comorbidities, clinical presentation, procedural variables, and lesion characteristics were compared among those with and without a bleeding event using chi-square and Fisher exact tests, as appropriate, for categorical variables and using Student’s t-tests for continuous variables. Hierarchical multivariable logistic regression was performed to account for clustering at the site level when identifying patient characteristics independently associated with bleeding. Candidate variables were selected based on clinical expertise and prior risk models for coronary interventions. They included demographic characteristics (age, sex), clinical comorbidities (baseline creatinine, baseline hemoglobin, dialysis, history of myocardial infarction or heart failure, left ventricular dysfunction), disease severity (critical limb ischemia [CLI], lesion in graft, multiple lesions), procedural urgency (elective vs. nonelective procedure), pre- and intraprocedural medications (antiplatelet agents, heparin analogues, warfarin, novel oral anticoagulant agents, vasodilators, thrombolytics), access site (femoral vs. nonfemoral), use of closure devices, and the vascular segment involved (aortoiliac, femoropopliteal, and infrapopliteal).

After unadjusted comparison of in-hospital mortality between those with and without major bleeding was performed, adjusted in-hospital mortality rates were compared between patients with and without major bleeding using multivariable logistic regression. This model adjusted for, age, sex, hypertension, diabetes mellitus, current dialysis, baseline creatinine, baseline hemoglobin, prior heart failure, coronary artery disease, prior myocardial infarction, severe lung disease, CLI, and nonelective procedures. Given that there are different components to the bleeding definition, the association of each type of bleeding with mortality was performed. All statistical tests were 2 sided and a p value of <0.05 was considered statistically significant. Missing data on all candidate variables were minimal (<1%), except for pre-procedure creatinine, which was missing in 60% of the cohort. Missing data were imputed with sequential regression using the IVEWare software version 0.3 (IVEWare, Ann Arbor, Michigan). All analyses were performed with SAS version 9.4 (SAS Institute, Cary, North Carolina).

Results

Demographic and procedural characteristics associated with bleeding

A total of 18,289 PVI procedures were included in the analysis, of whom 744 (4.10%) had a major bleeding complication. Among those with bleeding, 358 (48.10%) experienced overt bleeding with a 3-g/dl Hb drop, 161 (21.60%) developed a ≥4-g/dl Hb drop without overt bleeding, and 225 (30.20%) with a baseline hemoglobin ≥8 g/dl received a blood transfusion. The baseline demographic and clinical characteristics of patients with and without periprocedural bleeding at the time of PVI are presented in Table 1. Among the patients with overt major bleeding, the majority experienced access site bleeding (58.40%), followed by retroperitoneal bleeding (22.60%), gastrointestinal bleeding (3.10%), and genitourinary bleeding (1.40%) (Central Illustration). Among the 744 patients with overt bleeding, 81 developed retroperitoneal bleeding. The overall incidence of retroperitoneal bleeds in all patients undergoing PVI was 0.40% (81 of 18,289). Patients with bleeding were older, more likely to be women, had severe or very severe lung disease, had a prior history of myocardial infarction or heart failure, and had a pre-procedural hemoglobin <12 g/dl. Urgent (nonelective) procedures were also associated with more bleeding complications as compared with elective procedures. Patients with CLI were at increased risk for bleeding, as were patients with more complex PAD (suggested by longer lesion length, in-stent thrombosis, multilesion disease, graft lesions, aneurysmal disease, and thrombotic and bifurcation lesions). The use of thrombolytic agents was more common in patients who experienced a bleeding event, whereas the use of heparin and bivalirudin were not significantly different between the 2 groups. Patients with bleeding complications had more target vessel perforations and more frequent requirement of new dialysis initiation. Patients with major bleeding had more unplanned vascular interventions, surgeries, and major amputations. The details of the bleeding events among patients with unplanned procedures are described in Online Table 1. The bleeding complications by access site are reported in Table 2. Higher rates of bleeding were observed from brachial (4.15% vs. 3.98%; p < 0.001) and popliteal (4.69% vs. 3.98%; p < 0.001) access, as compared with femoral access. We also noted a wide variation in the rates of bleeding complications across sites. The median hospital rate of the bleeding complications was 3.65% (interquartile range: 1.20% to 5.90% (Online Figure 2).

Table 1. Baseline Patient Characteristics

Overall (N = 18,289)Bleeding (n = 744)No Bleeding (n = 17,545)p Value
Age, yrs69.0 ± 11.270.2 ± 11.268.9 ± 11.20.002
Female7,427 (40.40)406 (53.90)6,979 (39.80)<0.001
White15,302 (83.20)612 (81.30)14,690 (83.20)0.202
BMI, kg/m228.4 ± 12.527.8 ± 9.928.4 ± 12.50.216
History
Hypertension16,534 (90.00)682 (90.90)15,852 (89.90)0.438
Dyslipidemia15,413 (82.40)609 (81.20)14,534 (82.50)0.379
Diabetes mellitus9,574 (52.10)405 (54.40)9,131 (52.00)0.199
ESRD on dialysis1,289 (7.10)87 (11.70)1,211 (6.90)<0.001
Coronary artery disease9,542 (52.20)394 (53.00)9,148 (52.10)0.727
Severe/very severe lung disease2,944 (16.10)169 (22.70)2,775 (15.80)<0.001
Prior myocardial infarction3,923 (21.50)195 (26.10)3,729 (21.30)0.002
Cardiomyopathy or LV systolic dysfunction1,831 (10.00)95 (12.80)1736 (9.90)0.010
Prior heart failure3,075 (16.80)190 (25.50)2,885 (16.40)<0.001
Procedural characteristics
Procedure status<0.001
Elective15,422 (84.30)445 (59.80)14,997 (85.40)
Urgent (nonelective)2,867 (15.70)299 (40.20)2,568 (14.50)
Vascular segment group<0.001
Aortoiliac only3,300 (18.60)167 (23.30)31,333 (18.40)
Femoropopliteal only7,898 (44.60)248 (34.50)7,650 (45.00)
Below the knee only2,178 (12.30)89 (12.40)2,089 (12.30)
Aortoiliac with others1,506 (8.50)83 (11.60)1,423 (8.40)
Femoropopliteal with below the knee2,831 (16.00)130 (18.10)2701 (15.90)
Procedure indication<0.001
Critical limb ischemia7,710 (41.90)441 (58.70)7,269 (41.20)
Worst PAD Presentation<0.001
Asymptomatic498 (2.70)41 (5.50)57 (2.60)
Atypical Claudication189 (1.00)6 (0.80)183 (1.00)
Claudication
Mild (Rutherford 1)480 (2.60)12 (1.60)468 (2.70)
Moderate (Rutherford 2)1,716 (9.40)30 (4.00)1,686 (9.60)
Severe (Rutherford 3)7,292 (39.90)202 (27.20)7,090 (40.40)
Critical limb ischemia
Ischemic pain at rest (Rutherford 4)2,425 (13.30)139 (18.70)2,286 (13.00)
Minor tissue loss (Rutherford 5)4,007 (21.90)179 (24.10)3,828 (21.80)
Major tissue loss (Rutherford 6)1,648 (9.00)133 (17.90)1,515 (8.60)
Access site<0.001
Femoral17,290 (94.60)688 (92.60)16,602 (94.70)
Nonfemoral1,006 (5.50)55 (7.30)951 (5.40)
Popliteal128 (0.750)6 (0.80)122 (0.70)
Dorsalis pedis160 (0.90)6 (0.80)154 (0.90)
Posterior tibial141 (0.80)5 (0.70)136 (0.80)
Radial84 (0.50)3 (0.40)81 (0.50)
Brachial313 (1.70)13 (1.75)300 (1.70)
Other164 (0.90)22 (3.00)143 (0.80)
Access directionality0.059
Antegrade3,382 (19.30)155 (22)3,227 (19.2)
Closure used11,356 (62.10)395 (53.1)10,961 (62.5)<0.001
Lesion characteristics
>1 lesion8,214 (44.70)391 (52.10)7,823 (44.30)<0.001
Chronic total occlusion5,858 (32.00)263 (35.30)5,595 (31.90)0.047
Any lesion in graft644 (3.50)50 (6.70)594 (3.40)<0.001
Lesion length, mm137.5 ± 128.9163.3 ± 158.7136.7 ± 127.3<0.001
Any thrombus present1,494 (8.10)148 (19.80)1,346 (7.60)<0.001
Any bifurcation lesion1,401 (7.60)90 (12.00)1,311 (7.40)<0.001
In-stent restenosis2,105 (11.50)88 (11.80)2,017 (11.50)0.781
In-stent thrombosis324 (1.8)43 (5.8)281 (1.60)<0.001
Medications
Antiplatelet agents15,105 (82.80)574 (77.6)14,531 (83.00)<0.001
Aspirin12,950 (71.00)485 (65.5)12,461 (71.20)<0.001
Clopidogrel11,078 (60.70)403 (54.6)10,675 (51.50)<0.001
Ticagrelor326 (1.80)16 (2.2)310 (1.80)0.429
Oral anticoagulant agents841 (4.60)31 (4.2)810 (4.60)0.580
Warfarin490 (2.70)21 (2.8)469 (2.70)0.792
Apixaban121 (0.70)3 (0.4)118 (0.70)0.378
Rivaroxaban160 (0.90)5 (0.7)155 (0.90)0.549
Unfractionated heparin14,236 (77.60)589 (79.0)13,647 (77.60)0.377
Bivalirudin884 (4.80)41 (5.5)843 (4.80)0.389
Glycoprotein IIb/IIIa antagonist146 (0.80)13 (1.7)133 (0.80)0.002
Thrombolytics737 (4.00)114 (15.3)623 (3.50)<0.001
Laboratory values
Pre-procedure creatinine value1.4 ± 1.61.7 ± 1.91.4 ± 1.5<0.001
Pre-procedure hemoglobin value12.7 ± 2.211.8 ± 2.812.8 ± 2.1<0.001
Intra- and post-procedure events
New requirement of dialysis26 (0.10)13 (1.80)13 (0.10)<0.001
Perforations95 (0.50)33 (4.50)62 (0.40)<0.001
Other vascular complications requiring treatment96 (0.50)46 (6.20)50 (0.30)<0.001
Unplanned vascular intervention/surgery149 (0.80)73 (9.90)76 (0.40)<0.001
Major amputations (unplanned)57 (0.30)20 (2.70)37 (0.20)<0.001

Values are mean ± SD or n (%).

BMI = body mass index; ESRD = end-stage renal disease; LV = left ventricle; PAD = peripheral arterial disease.

Central Illustration.
Central Illustration.

Distribution of Bleeding Events and the Relationship of Bleeding With In-Hospital Mortality

(A) Distribution of bleeding events by location or source (left) and (B) the unadjusted in-hospital mortality according to bleeding category (right). Hb = hemoglobin.

Table 2. Bleeding Complications Based on the Access Site

Bleeding (n = 743)No Bleeding (n = 17,537)Total (N = 18,280)p Value
Femoral688 (92.60)16,602 (94.60)17,290<0.001
Popliteal6 (0.80)122 (0.69)128
Dorsalis pedis6 (0.80)154 (0.88)160
Posterior tibial5 (0.67)136 (0.77)141
Radial3 (0.40)81 (0.46)84
Brachial13 (1.75)300 (1.71)313
Other22 (2.96)142 (0.81)164

Values are n (%).

After multivariable adjustment, the patient and treatment characteristics independently associated with bleeding are presented in Figure 1. Patient characteristics independently associated with bleeding included age (odds ratio [OR]: 1.15/decade; 95% confidence interval [CI]: 1.07 to 1.24), female sex (OR: 1.76; 95% CI: 1.50 to 2.08), prior heart failure (OR: 1.24; 95% CI: 1.00 to 1.53), nonelective PVI (OR: 2.50; 95% CI: 2.03 to 3.07), multilesion disease (OR: 1.22; 95% CI: 1.01 to 1.47), aortoiliac disease (OR: 1.89; 95% CI: 1.52 to 2.36), and presentation with CLI (OR: 1.29; 95% CI: 1.06 to 1.57). Procedural factors associated with bleeding included nonfemoral vascular access (OR: 1.37; 95% CI: 1.01 to 1.85) and use of thrombolytic therapy (OR: 3.34; 95% CI: 2.61 to 4.53). Use of closure devices (OR: 0.63; 95% CI: 0.52 to 0.75) and pre-procedural hemoglobin >12 g/dl (OR: 0.89; 95% CI: 0.84 to 0.95) were independently associated with fewer bleeding complications.

Figure 1.
Figure 1.

Independent Predictors of Bleeding

Forest plot of independent predictors of major bleeding complications showing odds ratios and 95% confidence intervals from our multivariable model. Characteristics associated with a higher risk of bleeding complications are to the right side of the axis whereas the characteristics associated with a low risk of bleeding are towards the left. AI = aortoiliac; ASA = aspirin; BTK = below the knee; CLI = critical limb ischemia; FP = femoropopliteal; G2B3A= glycoprotein IIb/IIIa inhibitors; HGB = hemoglobin; HF = heart failure; LMWH = low-molecular-weight heparin; LVSD = left ventricular systolic dysfunction; MI = myocardial infarction; NOAC = novel oral anticoagulant agent.

Association of bleeding and in-hospital mortality

Among the 4.10% of patients who had a peri-PVI bleeding event, the all cause in-hospital mortality rate was significantly higher than in those who did not (6.60% vs. 0.30%; p < 0.001). The Central Illustration presents the unadjusted mortality rates for each component of the bleeding definition. Compared with the 0.30% in-hospital mortality in those without bleeding events, those with a ≥4-g/dl drop in hemoglobin in the absence of overt bleeding had an in-hospital mortality of 3.10% (p < 0.001). Those who received a blood transfusion had 6.20% in-hospital mortality (p < 0.001), whereas those with an overt bleed had 8.80% in-hospital mortality (p < 0.001) (Central Illustration). After multivariable adjustment, post-PVI major bleeding remained independently associated with in-hospital mortality (OR: 10.87; 95% CI: 6.95 to 17.02), as did age (OR/10 years 1.43; 95% CI: 1.17 to 1.74) and being on dialysis before the procedure (OR: 5.27; 95% CI: 2.46 to 11.31) (Figure 2).

Figure 2.
Figure 2.

Independent Predictors of In-Hospital Mortality

Forest plot of adjusted mortality of patients undergoing peripheral vascular intervention showing odds ratio and 95% confidence intervals from our multivariable model. Variables associated with higher in-hospital mortality are towards the right side of the axis whereas variables associated with lesser in-hospital mortality are on the left side. CAD = coronary artery disease; CLI = critical limb ischemia; DM = diabetes mellitus; HF = heart failure; HGB = hemoglobin; HTN = hypertension; MI = myocardial infarction.

Discussion

Over the past 20 years, there has been a rapid growth in the use of PVI for the treatment of PAD symptoms (18,19). However, there are limited data regarding safety outcomes of PVI in routine clinical practice. Although reducing periprocedural bleeding after percutaneous coronary intervention has been a major focus of efforts to improve procedural safety, less attention has been given to bleeding reduction after PVI (17,20). In the current largest multicenter study of bleeding complications after inpatient PVI to date, this study found that bleeding occurred in 4.10% of patients. The analysis identified a number of patient factors, including increasing age, female sex, comorbidities, multisegment disease, aortoiliac interventions, and procedural treatment strategies, including nonfemoral access and thrombolytic therapy, to be independently associated with bleeding. The use of closure devices and higher pre-procedural hemoglobin levels were independently associated with less bleeding. The high rate of bleeding (49%) among patients requiring unplanned vascular surgery highlights the consequences of unsuccessful PVI requiring either rescue surgeries or subsequent PVI procedures for immediate complications. The analysis also found that major bleeding was associated with increased in-hospital mortality in adjusted analysis. These findings underscore the importance of bleeding and patient factors associated with this complication after PVI.

Our study extends prior efforts to understand bleeding at the time of PVI (11–15). Previous studies, using different definitions of bleeding, have demonstrated an association between both baseline anemia and transfusion with mortality after PVI (12,21–23). However, these studies did not explicitly describe their bleeding definition. Ortiz et al. (24) studied prevalence and predictors of access site complications, defined as a hematoma at the site of arterial puncture with or without pseudoaneurysm, in 22,226 patients from 2007 to 2013 using the Vascular Quality Initiative database. The incidence of access site complications in this study was 3.50% and was associated with >75 years of age, female sex, no prior PVI, white race, nonfemoral access site, sheath size >6-F, procedural time >30 min, nonuse of the femoral closure device, and preoperative ambulatory status before the procedure. Although this study did not describe mortality, length of stay was longer in those with access site complications (24). Our study, which examined a related outcome, found similar patient factors to be associated with bleeding, highlighting the strong association among age, sex, and nonfemoral access with bleeding complications. Our study extends these insights by confirming the strong association between these variables and bleeding, and underscores additional clinical correlates of major bleeding, including treated vessel segments and baseline anemia.

Among the patient-level characteristics and procedural strategies associated with bleeding after PVI in this study, many, including older age, female sex, renal dysfunction, and disease severity are known bleeding predictors for both PCI and PVI and are not modifiable, although they can be useful in informing safer strategies to mitigate the risk of bleeding in these higher-risk patients (25–32). Combining patients’ risk factors into a bleeding risk score could potentially identify higher-risk patients for whom bleeding avoidance strategies might be indicated. In the setting of PCI, risk models from the NCDR Cath-PCI registry were developed and, when prospectively implemented at the time of treatment, were associated with a 44% reduction in periprocedural bleeding (8,10). Moreover, this study found potentially modifiable procedural strategies that might represent modifiable targets to reduce the risk of bleeding for high-risk patients undergoing PVI. For example, nonfemoral access and the use of thrombolytic agents were associated with increased bleeding and consideration of alternative approaches in higher-risk patients might improve procedural safety. Although nonfemoral access is sometimes required for anatomic reasons during PVI, these patients may benefit from application of other bleeding avoidance strategies, such as careful attention to periprocedural anticoagulation, to minimize their risk of bleeding when alternative access is required.

Study limitations

First, this study adopted the strategies used in the NCDR Cath PCI and ACTION registries for combining several components to comprise our bleeding definition (8,17). Although each component of this bleeding definition might not be equally important, all components of the bleeding definition were strongly associated with increased in-hospital mortality as compared with not experiencing a bleeding complication. Additionally, the study only includes in-hospital bleeding and mortality events; therefore, our results may underestimate the rates of bleeding. An additional limitation is that hospital participation in the registry is voluntary and these results may not be generalizable to all hospitals and operators in the United States. Future studies may identify other opportunities to explain and mitigate bleeding. For example, this study did not have data regarding the sheath size or crossing strategies (e.g., only 1 access site could be recorded per procedure in the NCDR PVI Registry) and data regarding the access technique (ultrasound or micropuncture technique) were not collected. Accordingly, it was unclear how many vascular access sites were used in patients undergoing nonfemoral access and details of operators’ approaches could provide additional insight into mediators and potential modifiers of bleeding risk. Similarly, data regarding involvement of trainee during a PVI procedure was not collected. The follow-up data were available in only ∼50% of the patients. Moreover, as an observational study, there may be residual or unmeasured confounding factors that were not directly measured. For example, the absence of an index of patients’ functional status, as shown to be important in the Vascular Quality Initiative, highlights the need for additional study in this area. Finally, these analyses were designed to describe patient and procedural characteristics associated with bleeding and did not create a formal risk prediction model based on patient characteristics alone. This will require future investigations and can form the foundation for future interventions to test strategies to minimize bleeding.

Conclusions

In a large national database of PVI procedures, our study found that bleeding occurred in 4.10% of patients and was independently associated with a range of demographic, clinical, disease severity, and treatment characteristics. Bleeding was also strongly associated with in-hospital mortality. Further work is needed to develop strategies that would maximize the safety of PVI.

Perspectives

WHAT IS KNOWN? Bleeding is a common periprocedural complication of endovascular procedures. Its role for poor outcomes after coronary interventions is well studied, but few studies have examined the risk factors and mortality associated with bleeding complications after PVI.

WHAT IS NEW? Bleeding is associated with important patient and procedural characteristics including age, sex, comorbidities, disease severity, and intra/periprocedural medications. It is strongly associated with higher in-hospital mortality.

WHAT IS NEXT? Further studies are warranted to build a robust pre-procedural risk model for bleeding complications and to define strategies to reduce the risks of bleeding after PVI.

  • 1. Chhatriwalla A.K., Amin A.P., Kennedy K.F.et al. : "Association between bleeding events and in-hospital mortality after percutaneous coronary intervention". JAMA 2013; 309: 1022.

    CrossrefMedlineGoogle Scholar
  • 2. Doyle B.J., Rihal C.S., Gastineau D.A. and Holmes D.R. : "Bleeding, blood transfusion, and increased mortality after percutaneous coronary intervention: implications for contemporary practice". J Am Coll Cardiol 2009; 53: 2019.

    View ArticleGoogle Scholar
  • 3. Lindsey J.B., Marso S.P., Pencina M.et al. : "Prognostic impact of periprocedural bleeding and myocardial infarction after percutaneous coronary intervention in unselected patients: results from the EVENT (evaluation of drug-eluting stents and ischemic events) registry". J Am Coll Cardiol Intv 2009; 2: 1074.

    View ArticleGoogle Scholar
  • 4. Manoukian S.V., Feit F., Mehran R.et al. : "Impact of major bleeding on 30-day mortality and clinical outcomes in patients with acute coronary syndromes: an analysis from the ACUITY Trial". J Am Coll Cardiol 2007; 49: 1362.

    View ArticleGoogle Scholar
  • 5. Marso S.P., Amin A.P., House J.A.et al. : "Association between use of bleeding avoidance strategies and risk of periprocedural bleeding among patients undergoing percutaneous coronary intervention". JAMA 2010; 303: 2156.

    CrossrefMedlineGoogle Scholar
  • 6. Mehta S.K., Frutkin A.D., Lindsey J.B.et al. : "Bleeding in patients undergoing percutaneous coronary intervention: the development of a clinical risk algorithm from the National Cardiovascular Data Registry". Circ Cardiovasc Interv 2009; 2: 222.

    CrossrefMedlineGoogle Scholar
  • 7. Ndrepepa G., Berger P.B., Mehilli J.et al. : "Periprocedural bleeding and 1-year outcome after percutaneous coronary interventions: appropriateness of including bleeding as a component of a quadruple end point". J Am Coll Cardiol 2008; 51: 690.

    View ArticleGoogle Scholar
  • 8. Rao S.V., McCoy L.A., Spertus J.A.et al. : "An updated bleeding model to predict the risk of post-procedure bleeding among patients undergoing percutaneous coronary intervention: a report using an expanded bleeding definition from the National Cardiovascular Data Registry CathPCI: Registry". J Am Coll Cardiol Intv 2013; 6: 897.

    View ArticleGoogle Scholar
  • 9. Rao S.V., O'Grady K., Pieper K.S.et al. : "Impact of bleeding severity on clinical outcomes among patients with acute coronary syndromes". Am J Cardiol 2005; 96: 1200.

    CrossrefMedlineGoogle Scholar
  • 10. Spertus J.A., Decker C., Gialde E.et al. : "Precision medicine to improve use of bleeding avoidance strategies and reduce bleeding in patients undergoing percutaneous coronary intervention: prospective cohort study before and after implementation of personalized bleeding risks". BMJ 2015; 350: h1302.

    CrossrefMedlineGoogle Scholar
  • 11. Danetz J.S., McLafferty R.B., Schmittling Z.C.et al. : "Predictors of complications after a prospective evaluation of diagnostic and therapeutic endovascular procedures". J Vasc Surg 2004; 40: 1142.

    CrossrefMedlineGoogle Scholar
  • 12. Jaffery Z., Shamai L. and White C.J. : "Preprocedural hemoglobin predicts mortality following peripheral vascular interventions". Catheter Cardiovasc Interv 2011; 78: 599.

    CrossrefMedlineGoogle Scholar
  • 13. Shammas N.W., Lemke J.H., Dippel E.J.et al. : "Bivalirudin in peripheral vascular interventions: a single center experience". J Invasive Cardiol 2003; 15: 401.

    MedlineGoogle Scholar
  • 14. Shammas N.W., Lemke J.H., Dippel E.J.et al. : "In-hospital complications of peripheral vascular interventions using unfractionated heparin as the primary anticoagulant". J Invasive Cardiol 2003; 15: 242.

    MedlineGoogle Scholar
  • 15. Sheikh I.R., Ahmed S.H., Mori N.et al. : "Comparison of safety and efficacy of bivalirudin versus unfractionated heparin in percutaneous peripheral intervention: a single-center experience". J Am Coll Cardiol Intv 2009; 2: 871.

    View ArticleGoogle Scholar
  • 16. "NCDR PVI Registry v1.0 Coder’s Data Dictionary". Available at: https://cvquality.acc.org/docs/default-source/ncdr/Data-Collection/pvi-registry_datadictionary.pdf?sfvrsn=c0748dbf_0. Accessed January 5, 2019.

    Google Scholar
  • 17. Mathews R., Peterson E.D., Chen A.Y.et al. : "In-hospital major bleeding during ST-elevation and non-ST-elevation myocardial infarction care: derivation and validation of a model from the ACTION Registry(R)-GWTG". Am J Cardiol 2011; 107: 1136.

    CrossrefMedlineGoogle Scholar
  • 18. Goodney P.P., Beck A.W., Nagle J., Welch H.G. and Zwolak R.M. : "National trends in lower extremity bypass surgery, endovascular interventions, and major amputations". J Vasc Surg 2009; 50: 54.

    CrossrefMedlineGoogle Scholar
  • 19. Rowe V.L., Lee W., Weaver F.A. and Etzioni D. : "Patterns of treatment for peripheral arterial disease in the United States: 1996-2005". J Vasc Surg 2009; 49: 910.

    CrossrefMedlineGoogle Scholar
  • 20. Mehran R., Rao S.V., Bhatt D.L.et al. : "Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium". Circulation 2011; 123: 2736.

    CrossrefMedlineGoogle Scholar
  • 21. Ambulgekar N.V., Grey S.F., Rosman H.S.et al. : "Association of anemia with outcomes in patients undergoing percutaneous peripheral vascular intervention: insights from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2 VIC)". J Invasive Cardiol 2018; 30: 35.

    MedlineGoogle Scholar
  • 22. Henke P.K., Park Y.J., Hans S.et al. : "The association of peri-procedural blood transfusion with morbidity and mortality in patients undergoing percutaneous lower extremity vascular interventions: insights from BMC2 VIC". PLoS One 2016; 11: e0165796.

    CrossrefMedlineGoogle Scholar
  • 23. Toor I.S., Jaumdally R.J., Moss M.S. and Babu S.B. : "Preprocedural hemoglobin predicts outcome in peripheral vascular disease patients undergoing percutaneous transluminal angioplasty". J Vasc Surg 2009; 50: 317.

    CrossrefMedlineGoogle Scholar
  • 24. Ortiz D., Jahangir A., Singh M., Allaqaband S., Bajwa T.K. and Mewissen M.W. : "Access site complications after peripheral vascular interventions: incidence, predictors, and outcomes". Circ Cardiovasc Interv 2014; 7: 821.

    CrossrefMedlineGoogle Scholar
  • 25. Gerhard-Herman M.D., Gornik H.L., Barrett C.et al. : "2016 AHA/ACC Guideline on the Management of Patients With Lower Extremity Peripheral Artery Disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines". J Am Coll Cardiol 2017; 69: e71.

    View ArticleGoogle Scholar
  • 26. Best P.J., Lennon R., Ting H.H.et al. : "The impact of renal insufficiency on clinical outcomes in patients undergoing percutaneous coronary interventions". J Am Coll Cardiol 2002; 39: 1113.

    View ArticleGoogle Scholar
  • 27. Jackson E.A., Munir K., Schreiber T.et al. : "Impact of sex on morbidity and mortality rates after lower extremity interventions for peripheral arterial disease: observations from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium". J Am Coll Cardiol 2014; 63: 2525.

    View ArticleGoogle Scholar
  • 28. Kawamura A., Piemonte T.C., Nesto R.W., Bilazarian S.D., Riskalla N.S. and Chauhan M.S. : "Impact of gender on in-hospital outcomes following contemporary percutaneous intervention for peripheral arterial disease". J Invasive Cardiol 2005; 17: 433.

    MedlineGoogle Scholar
  • 29. Ohno Y., Maekawa Y., Miyata H.et al. : "Impact of periprocedural bleeding on incidence of contrast-induced acute kidney injury in patients treated with percutaneous coronary intervention". J Am Coll Cardiol 2013; 62: 1260.

    View ArticleGoogle Scholar
  • 30. Plaisance B.R., Munir K., Share D.A.et al. : "Safety of contemporary percutaneous peripheral arterial interventions in the elderly insights from the BMC2 PVI (Blue Cross Blue Shield of Michigan Cardiovascular Consortium Peripheral Vascular Intervention) registry". J Am Coll Cardiol Intv 2011; 4: 694.

    View ArticleGoogle Scholar
  • 31. Valle J.A., McCoy L.A., Maddox T.M.et al. : "Longitudinal risk of adverse events in patients with acute kidney injury after percutaneous coronary intervention: insights from the National Cardiovascular Data Registry". Circ Cardiovasc Interv 2017; 10: e004439.

    CrossrefMedlineGoogle Scholar
  • 32. Vouyouka A.G., Egorova N.N., Salloum A.et al. : "Lessons learned from the analysis of gender effect on risk factors and procedural outcomes of lower extremity arterial disease". J Vasc Surg 2010; 52: 1196.

    CrossrefMedlineGoogle Scholar

Abbreviations and Acronyms

CLI

critical limb ischemia

CI

confidence interval

Hb

hemoglobin

NCDR

National Cardiovascular Data Registry

OR

odds ratio

PAD

peripheral artery disease

PCI

percutaneous coronary intervention

PVI

peripheral vascular intervention

Footnotes

Dr. Pokharel has received funding support from the National Heart, Lung, and Blood Institute of the National Institutes of Health under Award Number T32HL110837. Dr. Spertus has a research contract from the American College of Cardiology to analyze the National Cardiovascular Data Registry; has served as a consultant for Janssen, Bayer, Novartis, and AstraZeneca; owns equity in Health Outcomes Sciences and copyright to the Peripheral Artery Questionnaire; and has served on the board for Blue Cross Blue Shield of Kansas City. Dr. Jones has received research grant support from the Agency for Healthcare Research and Quality, AstraZeneca, American Heart Association, Bristol-Myers Squibb, Doris Duke Charitable Foundation, Merck, and Patient-Centered Outcomes Research Institute; and received honoraria/other from the American College of Physicians, Bayer, Bristol-Myers Squibb, Daiichi Sankyo, and Janssen Pharmaceuticals. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.