Abstract
Purpose:
The underlying difference between intermittent claudication (IC) and critical limb-threatening ischemia (CLTI) still remains unclear. This prospective multicenter observational study aimed to clarify differences in clinical features and prognostic outcomes between IC and CLTI, and prognostic factors in patients undergoing endovascular therapy (EVT).
Materials and Methods:
A total of 692 patients with 808 limbs were enrolled from 20 institutions in Japan. The primary measurements were the 3-year rates of major adverse cardiovascular event (MACE) and reintervention.
Results:
Among patients, 79.0% had IC and 21.0% had CLTI. Patients with CLTI were more frequently women and more likely to have impaired functional status, undernutrition, comorbidities, hypercoagulation, hyperinflammation, distal artery disease, short single antiplatelet and long anticoagulation therapies, and late cilostazol than patients with IC. Aortoiliac and femoropopliteal diseases were dominant in patients with IC and infrapopliteal disease was dominant in patients with CLTI. Patients with CLTI underwent less frequently aortoiliac intervention and more frequently infrapopliteal intervention than patients with IC. Longitudinal change of ankle-brachial index (ABI) exhibited different patterns between IC and CLTI (pinteraction=0.002), but ABI improved after EVT both in IC and in CLTI (p<0.001), which was sustained over time. Dorsal and plantar skin perfusion pressure in CLTI showed a similar improvement pattern (pinteraction=0.181). Distribution of Rutherford category improved both in IC and in CLTI (each p<0.001). Three-year MACE rates were 20.4% and 42.3% and 3-year reintervention rates were 22.1% and 46.8% for patients with IC and CLTI, respectively (log-rank p<0.001). Elevated D-dimer (p=0.001), age (p=0.043), impaired functional status (p=0.018), and end-stage renal disease (p=0.019) were independently associated with MACE. After considering competing risks of death and major amputation for reintervention, elevated erythrocyte sedimentation rate (p=0.003) and infrainguinal intervention (p=0.002) were independently associated with reintervention. Patients with CLTI merely showed borderline significance for MACE (adjusted hazard ratio 1.700, 95% confidence interval 0.950–3.042, p=0.074) and reintervention (adjusted hazard ratio 1.976, 95% confidence interval 0.999–3.909, p=0.05).
Conclusions:
The CLTI is characterized not only by more systemic comorbidities and distal disease but also by more inflammatory coagulation disorder compared with IC. Also, CLTI has approximately twice MACE and reintervention rates than IC, and the underlying inflammatory coagulation disorder per se is associated with these outcomes.
Clinical Impact
The underlying difference between intermittent claudication (IC) and critical limb-threatening ischemia (CLTI) still remains unclear. This prospective multicenter observational study, JPASSION study found that CLTI was characterized not only by more systemic comorbidities and distal disease but also by more inflammatory coagulation disorder compared to IC. Also, CLTI had approximately twice major adverse cardiovascular event (MACE) and reintervention rates than IC. Intriguingly, the underlying inflammatory coagulation disorder per se was independently associated with MACE and reintervention. Further studies to clarify the role of anticoagulation and anti-inflammatory therapies will contribute to the development of post-interventional therapeutics in the context of peripheral artery disease.
Keywords: peripheral artery disease, endovascular therapy, coagulation, inflammation, outcome
Introduction
The presence of peripheral artery disease (PAD) is associated with cardiovascular events and mortality. 1 However, during the last 3 decades, the general and morphological characteristics of PAD have transformed with continuing increase of diabetes mellitus and chronic kidney disease.2–4 To date, the increasing global burden of PAD is an emerging public health issue. 5 Meanwhile, medical management has developed with the dissemination of guidelines on antithrombotic therapy and risk factor modification, and endovascular therapy (EVT) has been sophisticated with tremendous improvement in armamentarium and techniques. 6 However, there is still substantial room for improvement in clinical outcomes in PAD patients undergoing EVT. The essential difference between intermittent claudication (IC) and critical limb-threatening ischemia (CLTI) still remains unclear in the setting of PAD. Insights into the underlying difference between IC and CLTI may contribute to further development of therapeutics. The aim of this prospective multicenter observational study was to clarify differences in clinical features and prognostic outcomes between IC and CLTI, and prognostic factors after EVT.
Materials and Methods
Study Design
The JPASSION (Japan Peripheral Artery disease: endovaScular revaScularizatION prospective multicenter observational study) is a prospective multicenter cohort study with a 3-year follow-up period. A total of 20 institutions participated in this study between 2013 and 2014. The study was developed in line with the principles of the Declaration of Helsinki and the study protocol was approved by the ethics review board of each institution. All patients who gave informed consent to participate in the study were enrolled. The protocol was registered in the University Hospital Medical Information Network Clinical Trial Registry (000010503).
Study Cohort
Patients who met both of the following criteria were eligible for enrollment: (1) 20 years of age or older and (2) scheduled EVT after being diagnosed with symptomatic PAD. Symptomatic PAD was diagnosed based on symptoms, signs, and hemodynamic assessment of the lower limb. Symptoms were divided into IC and CLTI. In cases of oligosymptomatology because of less opportunity to walk, emphasis was placed on physical signs, including muscle atrophy or hair loss. 7 Ankle-brachial index (ABI) was measured as macrocirculation, and dorsal or plantar skin perfusion pressure was additionally measured as microcirculation in CLTI because of the potential of falsely high ankle pressure and concomitant pedal artery disease. Hemodynamic criteria of PAD were considered to be ABI less than 0.9 in cases of IC, and skin perfusion pressure less than 50 mmHg in cases of CLTI.1,3 At baseline, demographics, risk factors, comorbidities, medication, and laboratory data were collected. The extent of lesions was assessed by catheter-based angiography, computed tomography angiography, magnetic resonance angiography, or duplex ultrasonography. Significant lesions were defined as greater than angiographical 50% stenosis on visual estimation or physicians’ clinical judgment if needed. Preinterventional and postinterventional medical management and endovascular procedures were performed based on physicians’ discretion. In cases of staged intervention, the date of the first procedure was recorded as time of the index procedure on the limb. In cases of bilateral intervention, the date of initial procedure was recorded as time of the index procedure on the patient. Wounds were assessed based on the University of Texas classification grade. Patients were followed post-EVT and at 1, 3, 6, 12, 24, and 36 months. Reintervention was implemented with EVT or surgery on a clinically-driven basis for symptomatic restenosis or new lesions (angiographically greater than 50% restenosis or de novo stenosis) in the index limb. Baseline and follow-up data were collected online and managed by the study office.
Endpoint and Definitions
The primary measurements were the 3-year major adverse cardiovascular event (MACE) rate and reintervention rate. The MACE was defined as a composite of myocardial infarction, stroke, and all-cause death. In addition, Rutherford category, major amputation, cardiovascular event (myocardial infarction or stroke), death, and amputation-free survival were analyzed. Myocardial infarction was defined as positive troponin and creatine kinase-MB twice the upper reference limit with clinical ischemic findings. 8 Major amputation was defined as above-the-ankle amputation of the index limb.
Statistical Analysis
Data are expressed as mean ± standard deviation, median (interquartile range), numbers, or percentages. Categorical data were compared using the χ 2 test or Fisher exact test as appropriate. Differences between the 2 independent groups were evaluated with Student t test or Welch t test for parametric continuous variables or the Mann-Whitney U test for nonparametric continuous variables. The extent of lesions, frequency of antithrombotic treatment, ABI, and skin perfusion pressure were analyzed using a mixed model with a variance component. The IC or CLTI, time point, dorsal or plantar side, or reintervention was appropriately modeled as fixed effects. In the models, random intercepts were modeled by patient ID. Multiple comparisons of estimated marginal means were performed by post hoc analyses of the linear mixed models with Bonferroni corrections. Kaplan-Meier analysis with the log-rank test was used for survival analysis. The univariate Cox proportional hazards model was used to assess the influence of factors on MACE and first reintervention. The following covariates were assessed: PAD stage, age, sex, impaired functional status, geriatric nutritional risk index, hypertension, diabetes mellitus, dyslipidemia, end-stage renal disease, current smoker, prior coronary revascularization, atrial fibrillation, heart failure, stroke, prior lower limb revascularization, D-dimer, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), hemoglobin, low-density lipoprotein cholesterol, statin use post-EVT, angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker use post-EVT, β-blocker use post-EVT, and post-EVT antithrombotic therapy such as any antiplatelet therapy, single antiplatelet therapy (SAPT), dual antiplatelet therapy (DAPT), aspirin, P2Y12 inhibitor, anticoagulation, or cilostazol. Post-EVT antithrombotic therapy was treated as a time-varying covariate. Multilevel intervention and infrainguinal intervention were additionally assessed on reintervention. For the adjustment, factors with a value of p<0.05 were entered into the multivariate Cox proportional hazards model with a backward stepwise elimination. Given competing risks for the first reintervention, multivariate competing-risk Cox regression was analyzed based on Fine and Grey model. 9 SPSS version 22 (IBM Corp, Armonk, New York) or R version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria) was used for statistical analysis. All p values <0.05 were considered statistically significant.
Results
Initially, a total of 861 limbs were registered, of which 47 limbs were withdrawn. Of 814 limbs, 6 limbs were excluded because their disease was nonatherosclerotic in origin. A total of 808 limbs in 692 patients (73.2±8.6 years old, range 34–98) were enrolled in this study. A total of 547 patients with 642 limbs were IC and 145 patients with 166 limbs were CLTI.
Clinical and Laboratory Characteristics
Patients with CLTI were more frequently women and more likely to have impaired functional status and low geriatric nutritional risk index (Table 1). As for risk factors and comorbidities, patients with CLTI had a higher prevalence of diabetes mellitus, heart failure, stroke, atrial fibrillation, chronic kidney disease, end-stage renal disease, and prior any amputation. Patients with IC had a higher prevalence of hypertension, dyslipidemia, smoking history, prior coronary revascularization, and carotid disease. Levels of inflammatory coagulation biomarkers such as CRP, ESR, and D-dimer were higher in patients with CLTI than those in patients with IC. Although levels of blood pressure and HbA1c were similar between patients with IC and CLTI, level of low-density lipoprotein cholesterol was lower, and statin and antihypertensive agents were less frequently used in patients with CLTI than in patients with IC. Medical management after EVT showed a similar trend to before EVT. Approximately three-quarters of CLTI cases were complicated by wounds. Deep wounds (University of Texas classification grade 2 or 3) were observed in 33.3% of the wounds, bacterially-infected wounds in 31.7%, and osteomyelitis in 18.3%.
Table 1.
Demographics.
| Variables | Intermittent claudication | Critical limb-threatening ischemia | p value |
|---|---|---|---|
| Patients, n | 547 | 145 | |
| Age, y | 73.0±8.1 | 74.1±10.4 | 0.221 |
| Female, % | 23.6 | 36.6 | 0.002 |
| Impaired functional status, % | 10.1 | 44 | <0.001 |
| GNRI | 102±10 | 95±11 | <0.001 |
| GNRI<92/ 92≤GNRI<98 / 98≤GNRI, % | 14.7/14.5/70.8 | 33.1/25.7/41.2 | <0.001 |
| Body mass index, kg/m2 | 22.6±3.3 | 21.3±3.1 | <0.001 |
| Hypertension, % | 87.7 | 78.6 | 0.005 |
| Antihypertensive agent for hypertension, % | 93.5 | 85.1 | 0.003 |
| Systolic blood pressure (mmHg) | 140±23 | 139±25 | 0.513 |
| Diastolic blood pressure (mmHg) | 73±13 | 75±16 | 0.318 |
| Guideline hypertension goal attainment: Blood pressure<140/90 (mmHg), % | 53.4 | 51.4 | 0.679 |
| Antihypertensive agent for hypertension post-EVT, % | 90.1 | 75.9 | <0.001 |
| ACEI or ARB use post-EVT, % | 57.1 | 45.5 | 0.027 |
| β-blocker use post-EVT, % | 28 | 16.1 | 0.004 |
| Diabetes mellitus, % | 49.7 | 62.1 | 0.008 |
| Antidiabetic agent or insulin for diabetes mellitus, % | 80.5 | 81.1 | 0.901 |
| HbA1c, % | 6.4±1.0 | 6.5±1.2 | 0.391 |
| Guideline diabetes mellitus goal attainment: HbA1c<7.0 (%), % | 76.4 | 69.8 | 0.127 |
| Antidiabetic agent or insulin for diabetes mellitus post-EVT, % | 79.2 | 81.4 | 0.665 |
| Dyslipidemia, % | 66.8 | 43.8 | <0.001 |
| Statin for dyslipidemia, % | 78.4 | 56.5 | <0.001 |
| LDL-C (mg/dL) | 100±31 | 92±34 | 0.027 |
| Guideline LDL-C goal attainment: LDL-C<70 (mg/dL), % | 15.9 | 22.0 | 0.152 |
| Statin for dyslipidemia post-EVT, % | 79.7 | 62.9 | 0.004 |
| Triglyceride (mg/dL) | 145±91 | 109±58 | <0.001 |
| Current smoking, % | 23.5 | 12.8 | 0.006 |
| Smoking history (cessation>1 month), % | 74.1 | 55.3 | <0.001 |
| Family history, % | 15.2 | 15.4 | 0.951 |
| Prior coronary revascularization, % | 44.8 | 27.6 | <0.001 |
| Old myocardial infarction, % | 10.1 | 10.3 | 0.918 |
| Atrial fibrillation, % | 6.6 | 14.5 | 0.002 |
| Heart failure, % | 8.8 | 15.9 | 0.012 |
| Stroke, % | 17.0 | 33.8 | <0.001 |
| Carotid artery disease, % | 11.9 | 4.7 | 0.017 |
| Prior lower limb revascularization, % | 30 | 26.9 | 0.468 |
| Prior major amputation, % | 0.7 | 7.1 | <0.001 |
| Prior minor amputation, % | 0.7 | 8.6 | <0.001 |
| Chronic obstructive pulmonary disease, % | 7.0 | 2.9 | 0.076 |
| Hemoglobin (g/dL) | 12.6±1.8 | 11.6±1.8 | <0.001 |
| C-reactive protein (mg/dL) | 0.16 (0.06–0.41) | 0.83 (0.20–2.42) | <0.001 |
| Erythrocyte sedimentation rate (mm/h) | 22 (11–37) | 46 (30–78) | <0.001 |
| Albumin (g/dL) | 4.0±0.5 | 3.7±0.6 | <0.001 |
| eGFR (mL/min/1.73 m2) | 49.9±28.3 | 39.4±34.2 | 0.001 |
| eGFR<60 (mL/min/1.73 m2), % | 60.3 | 78.3 | <0.001 |
| End-stage renal disease, % | 20 | 39.3 | <0.001 |
| Corrected calcium (mg/dL) | 9.2 (8.9–9.5) | 9.4 (9.0–9.8) | 0.001 |
| Phosphate (mg/dL) | 3.7±1.0 | 3.7±1.0 | 0.703 |
| D-dimer (μg/mL) | 0.90 (0.67–1.70) | 1.94 (1.00–3.18) | <0.001 |
| Brain natriuretic peptide (pg/mL) | 65 (24–139) | 182 (79–454) | <0.001 |
| Left ventricular ejection fraction on echocardiography (%) | 59±11 | 54 ±14 | 0.002 |
| Limbs, n | 642 | 166 | |
| Right, % | 53.6 | 53.6 | 0.994 |
| Rutherford category 0/1/2/3/4/5/6, % | 4.6/14.8/40.5/40/0/0/0 | 0/0/0/0/22.2/73.5/4.3 | <0.001 |
| Ankle-brachial index | 0.65±0.21 | 0.55±0.38 | 0.003 |
| Dorsal skin perfusion pressure (mmHg) | 32.4±17.9 | ||
| Plantar skin perfusion pressure (mmHg) | 31.0±16.9 | ||
| Wounds, n | 126 | ||
| University of Texas classification grade 0/1/2/3, % | 9.2/57.5/12.5/20.8 | ||
| Infectious wound, % | 31.7 | ||
| Osteomyelitis, % | 18.3 |
Data are presented as n, %, mean ± standard deviation, or median (interquartile range).
Abbreviations: ACEI, angiotensin-converting enzyme inhibitors; ARB, angiotensin II receptor blockers; eGFR, estimated glomerular filtration rate; EVT, endovascular therapy; GNRI, geriatric nutrition risk index; LDL-C, low-density lipoprotein cholesterol.
Anatomical Characteristics and Endovascular Intervention
The spectrum of lower limb artery lesions, including chronic total occlusions, differed between patients with IC and CLTI (pinteraction<0.001) (Figure 1). When broadly divided into 3 segments, patients with IC more frequently had aortoiliac disease (p<0.001) and femoropopliteal disease (p=0.001) and patients with CLTI more frequently had infrapopliteal disease (p<0.001). Based on detailed segment-based analysis, the distal superficial femoral artery showed no significant difference in lesion frequency between patients with IC and CLTI (34.4% vs 27.9%, p=0.105). Having the distal superficial femoral artery as a boundary, upstream lesions were more frequent in patients with IC and downstream lesions were more frequent in patients with CLTI, except for abdominal aorta, common femoral artery, and deep femoral artery (Figure 1).
Figure 1.
Lesion spectrum of peripheral artery disease. Slash in box and the number in parentheses indicate chronic total occlusion. IC, intermittent claudication; CLTI, critical limb-threatening ischemia; CIA, common iliac artery; EIA, external iliac artery; Femopop, femoropopliteal artery; CFA, common femoral artery; DFA, deep femoral artery; SFA, superficial femoral artery; POP, popliteal artery; Infrapop, infrapopliteal artery; ATA, anterior tibial artery; TP, tibioperoneal artery; Pero, peroneal artery; PTA, posterior tibial artery; DPA, dorsalis pedis artery; LP, lateral plantar artery; MP, medial plantar artery; pro, proximal segment; mid, mid segment; dis, distal segment.
The patients with CLTI underwent less frequently aortoiliac intervention (48.9% vs 17.5%, p<0.001) and more frequently infrapopliteal intervention (15.3% vs 78.3%, p<0.001) than patients with IC. Femoropopliteal intervention was performed at a similar frequency between patients with IC and CLTI (62.1% vs 55.4%, p=0.114). Also, patients with CLTI more frequently underwent multilevel intervention (24.3% vs 43.4%, p<0.001) and infrainguinal intervention (66.2% vs 95.8%, p<0.001) than patients with IC. Consequently, the usage of stent was less frequent in CLTI than in IC (78.7% vs. 42.8%, p<0.001), but the frequency of drug-eluting stent when using stent was similar between IC and CLTI (10.7% vs 15.5%, p=0.231). No drug-coated balloon was used because of no availability during the study period. Hybrid endovascular and open surgical procedure was uncommonly performed at a similar frequency (1.3% vs 1.9%, p=0.476). Total number of EVT on the index limb was higher in CLTI (1 [1–3]) compared with IC (1 [1–1]) (p<0.001) patients.
Antithrombotic Treatment
The pattern of any antiplatelet therapy was identical over time between IC and CLTI (pinteraction=0.522) (Figure 2A). In the context of any antiplatelet therapy, selection of SAPT or DAPT exhibited a different pattern over time between IC and CLTI (pinteraction=0.002); DAPT-dominated periods were from post-EVT to 6 months in IC and from post-EVT to 3 months in CLTI; in particular, CLTI was more frequently treated with SAPT from post-EVT to 1 month (post-EVT: 21.9% vs 38%, p<0.001; 1 month: 32.3% vs 44.9%, p=0.01); in other words, IC was more frequently treated with DAPT from post-EVT to 1 month (post-EVT: 78.1% vs 62%, p<0.001; 1 month: 67.7% vs 55.1%, p=0.01) (Figure 2B). In the context of SAPT, selection of aspirin or P2Y12 inhibitor exhibited a similar pattern over time between IC and CLTI (pinteraction=0.584) (Figure 2C). The anticoagulation pattern was identical over time between IC and CLTI (pinteraction=0.259), but frequency of anticoagulation therapy was consistently higher in CLTI (approximately 30%) than in IC (less than 15%) (all time points p<0.001) (Figure 2D). Cilostazol pattern was different over time between IC and CLTI (pinteraction=0.016); cilostazol was more frequent in IC than in CLTI before EVT (16.1% vs 9%, p=0.031), but was more frequent in CLTI than in IC at 3 years (14.8% vs 26.9%, p=0.027) (Figure 2E).
Figure 2.
Antithrombotic treatment pattern. (A) With or without APT. (B) SAPT or DAPT in APT. (C) Aspirin or P2Y12 inhibitor in SAPT. (D) With or without anticoagulation. (E) With or without cilostazol. IC, intermittent claudication; CLTI, critical limb-threatening ischemia; APT, antiplatelet therapy; SAPT, single antiplatelet therapy; DAPT, dual antiplatelet therapy.
Lower Limb Circulation and Symptoms
Longitudinal change of ABI exhibited different patterns between IC and CLTI (pinteraction=0.002). However, both in IC and in CLTI, ABI improved after EVT, which was sustained throughout 3 years (pre-EVT vs all time points, p<0.001). Regarding microcirculation in CLTI, dorsal and plantar skin perfusion pressure showed a similar improvement pattern (pinteraction=0.181) (Supplemental Figure A). Distribution of Rutherford category improved both in IC and in CLTI (each p<0.001) (Supplemental Figure B). In IC, relief of claudication (Rutherford category 0) was observed in 75.4% at 1 month, which was sustained over time. Any claudication was observed in approximately in 25% over time after EVT, and progression from IC to CLTI was observed in 3.2% at 3 years. On the contrary, in CLTI, Rutherford category 4, or resting pain, decreased from 22.8% before EVT to 1.8% at 1 month, which was sustained over time. Minor tissue loss (Rutherford category 5) gradually decreased from 73.5% to 43.9% at 1 month, 12.6% at 1 year, 7.9% at 2 years, and 7.5% at 3 years. Major tissue loss (Rutherford category 6) decreased from 4.3 to 1.8% at 1 month, and thereafter averaged at around 2%.
Clinical Course
During the median follow-up of 3.3 years, coronary revascularization was performed in 9.1% (percutaneous intervention 7.9%, bypass grafting 1.2%), but there was no difference between IC and CLTI (9.1% vs 9.0%, p=0.948). According to Kaplan-Meier analysis, amputation-free survival, death, cardiovascular event, and major amputation were all different between IC and CLTI. Death, cardiovascular event, and major amputation competitively increased, especially in patients with CLTI until 6 months after EVT (Figure 3A). The MACE was observed in 23.3% (161 patients). Kaplan-Meier analysis revealed a significant difference in MACE between IC and CLTI (log-rank p<0.001); 3-year MACE rates were 20.4% in IC and 42.3% in CLTI (Figure 3B). The MACE rate per 100 patient-years was 7.2 in IC and 17.2 in CLTI. After adjustments of factors associated with MACE in univariate Cox regression analysis, multivariate Cox regression analysis found borderline significance of CLTI for MACE (p=0.074), and that elevated D-dimer (p=0.001), age (p=0.043), impaired functional status (p=0.018), and end-stage renal disease (p=0.019) were independently associated with MACE (Table 2).
Figure 3.
Clinical events. (A) Death, cardiovascular event, major amputation, and amputation-free survival. (B) Major adverse cardiovascular event. (C) Reintervention. IC, intermittent claudication; CLTI, critical limb-threatening ischemia; AFS, amputation-free survival; CV, cardiovascular event; MA, major amputation; MACE, major adverse cardiovascular event.
Table 2.
Predictors of Clinical Events.
| Univariate | p value | Multivariate | p value | |
|---|---|---|---|---|
| Hazard ratio [95% CI] | Hazard ratio [95% CI] | |||
| MACE | ||||
| Critical limb-threatening ischemia | 2.357 [1.683, 3.301] | <0.001 | 1.700 [0.950, 3.042] | 0.074 |
| Age, per 1 year | 1.031 [1.010, 1.052] | 0.003 | 1.034 [1.001, 1.067] | 0.043 |
| Impaired functional status | 3.036 [2.114, 4.360] | <0.001 | 2.241 [1.151, 4.362] | 0.018 |
| Geriatric nutritional risk index | 0.965 [0.951, 0.979] | <0.001 | ||
| End-stage renal disease | 2.957 [2.158, 4.053] | <0.001 | 1.917 [1.113, 3.304] | 0.019 |
| Prior coronary revascularization | 1.559 [1.144, 2.126] | 0.005 | ||
| Prior lower limb revascularization | 1.405 [1.019, 1.937] | 0.038 | ||
| Low-density lipoprotein cholesterol, per mg/dL | 0.993 [0.987, 0.999] | 0.026 | ||
| Hemoglobin, per 1 g/dL | 0.757 [0.696, 0.824] | <0.001 | ||
| D-dimer, per 1 μg/mL | 1.146 [1.078, 1.219] | <0.001 | 1.125 [1.047, 1.208] | 0.001 |
| C-reactive protein, per 1 mg/dL | 1.106 [1.036, 1.182] | 0.003 | ||
| Reintervention | ||||
| Critical limb-threatening ischemia | 2.858 [2.092, 3.904] | <0.001 | 1.976 [0.999, 3.909] | 0.05 |
| Female | 1.426 [1.047, 1.941] | 0.024 | ||
| Impaired functional status | 1.657 [1.123, 2.445] | 0.011 | ||
| Geriatric nutritional risk index | 0.98 [0.967, 0.995] | 0.007 | 1.026 [0.995, 1.058] | 0.095 |
| Diabetes mellitus | 1.512 [1.127, 2.029] | 0.006 | ||
| Dyslipidemia | 0.686 [0.513, 0.917] | 0.011 | ||
| End-stage renal disease | 2.574 [1.907, 3.473] | <0.001 | ||
| Prior lower limb revascularization | 1.704 [1.275, 2.277] | <0.001 | 1.737 [0.954, 3.160] | 0.071 |
| Hemoglobin, per 1 g/dL | 0.830 [0.769, 0.897] | <0.001 | ||
| C-reactive protein, per 1 mg/dL | 1.121 [1.059, 1.187] | <0.001 | ||
| Erythrocyte sedimentation rate, per 1 mm/h | 1.018 [1.009, 1.026] | <0.001 | 1.014 [1.004, 1.025] | 0.005 |
| Statin after the index procedure | 0.684 [0.513, 0.911] | 0.009 | 0.58 [0.318, 1.057] | 0.075 |
| Infrainguinal intervention | 5.353 [3.206, 8.940] | <0.001 | 3.485 [1.326, 9.161] | 0.011 |
| Multilevel intervention | 1.849 [1.376, 2.485] | <0.001 | ||
Abbreviations: CI, confidence interval; MACE, major adverse cardiovascular event.
Reintervention was observed in 23.1% (187 limbs; EVT 181 limbs, surgery 12 limbs). Kaplan-Meier analysis revealed a significant difference in reintervention between IC and CLTI (log-rank p<0.001); 3-year reintervention rates were 22.1% in IC and 46.8% in CLTI (Figure 3C). Reintervention rate per 100 limb-years was 8.0 in IC and 25.2 in CLTI. There was no significant association with reintervention and post-EVT antithrombotic therapy pattern. After adjustment of factors associated with reintervention in univariate Cox regression analysis, multivariate Cox regression analysis found borderline significance of CLTI for reintervention (p=0.05) and that elevated ESR (p=0.005) and infrainguinal intervention (p=0.011) were independently associated with reintervention (Table 2). After considering competing risks of a composite of death and major amputation for reintervention, multivariate Cox competing risk regression analysis found that elevated ESR (subhazard ratio 1.014, 95% confidence interval 1.005–1.023, p=0.003) and infrainguinal intervention (subhazard ratio 4.356, 95% confidence interval 1.681–11.290, p=0.002) remained predictors of reintervention.
Discussion
The main findings of this study are as follows: (1) CLTI was characterized not only by more systemic comorbidities and distal disease but also by more inflammatory coagulation disorder compared with IC; (2) CLTI had a preference for short SAPT, long anticoagulation therapy, and late cilostazol, whereas IC had a preference for early cilostazol and short DAPT; (3) 3-year MACE rates were 20% in IC and 42% in CLTI; (4) 3-year reintervention rates were 22% in IC and 47% in CLTI; and (5) elevated D-dimer and ESR were independently associated with MACE and reintervention, respectively, whereas CLTI merely showed borderline significance for both MACE and reintervention (Figure 4).
Figure 4.
Clinical phenotypes with intermittent claudication and critical limb-threatening ischemia. Red overlay on the cartoon illustrates the association of peripheral artery disease stage with arterial lesion spectrum. AF, atrial fibrillation; AI, aortoiliac; CKD, chronic kidney disease; CLTI, critical limb-threatening ischemia; DAPT, dual antiplatelet therapy; ESR, erythrocyte sedimentation rate; ESRD, end-stage renal disease; EVT, endovascular therapy, FP, femoropopliteal; HF, heart failure; MACE, major adverse cardiovascular event; PAD, peripheral artery disease; SAPT, single antiplatelet therapy; SFA, superficial femoral artery.
Comorbidities and Inflammatory Coagulation Disorders
To date, comparable clinical data between IC and CLTI are sparse. The present study found that, compared with patients with IC, patients with CLTI were more frequently women and more likely to have impaired functional status and low geriatric nutritional risk index. Also, patients with IC were more likely to have atherosclerotic risk factors of hypertension, dyslipidemia, and smoking history, and have major atherosclerotic diseases such as coronary and carotid artery diseases. In contrast, patients with CLTI were more likely to have diabetes mellitus and multiple comorbidities in the heart, brain, and kidney. Three-quarters of CLTI cases were complicated by wounds: More than 30% of the wounds were deep and bacterially infected, and approximately 20% of the wounds were complicated by osteomyelitis. These findings suggest that patients with CLTI are exposed to more disability, undernutrition, and frailty, and have more systemic comorbidities encompassing many disciplines compared with patients with IC. Also, previous studies reported elevated CRP and D-dimer in patients with PAD compared with control subjects.10,11 The present study found more elevated CRP, ESR, and D-dimer in CLTI than in IC, indicating that more severe inflammatory coagulation disorder is the hallmark of CLTI compared with IC.
Proximal and Distal Diseases Separated by the Distal Superficial Femoral Artery
Atherosclerotic risk factor profiles are associated with the extent of atherosclerotic lesions in the lower limbs. 12 The present study found heterogeneous arterial lesion spectrum between IC and CLTI. In IC, femoropopliteal disease was predominant, followed by aortoiliac disease and infrapopliteal disease. In CLTI, infrapopliteal disease was predominant, followed by femoropopliteal disease and aortoiliac disease (Figure 1). It is noteworthy that, having the “distal superficial femoral artery” as a boundary, IC more frequently had upstream lesions, whereas CLTI more frequently had downstream lesions. Therefore, IC and CLTI can be rephrased as “proximal disease” and “distal disease,” respectively. Also, there was a discrepancy between lesion frequency and intervention intensity, implying the implementation of personalized endovascular intervention presumably based on risk-benefit trade off.
Antithrombotic Preference
Little is known about patterns of antithrombotic treatment in patients with PAD. According to a German retrospective study with health insurance data between 2008 and 2018, 13 antiplatelet use before revascularization was only 50% to 60% (IC: 52%, CLTI: 59%). This prospective study found a similar intensity of antiplatelet therapy before EVT (IC: 63%, CLTI: 55%), and then temporary escalation of any antiplatelet therapy and DAPT after EVT presumably due to the prevention of subacute thrombosis (Figure 2A). Of note, in the context of antiplatelet therapy, SAPT was more frequent in CLTI than in IC (conversely, DAPT was more frequent in IC than in CLTI) from post-EVT to 1 month, suggesting “short SAPT” preference in CLTI and “short DAPT” preference in IC (Figure 2B). Also, the present study found “long anticoagulation therapy” preference in CLTI (less than 15% in IC and approximately 30% in CLTI) presumably due to concomitant atrial fibrillation or lesion complexity (Figure 2D). Secondary use of cilostazol was more frequent in IC before EVT (16% in IC and 9% in CLTI), but more frequent in CLTI at 3 years (15% in IC and 27% in CLTI), indicating “early cilostazol” preference in IC and “late cilostazol” preference in CLTI (Figure 2E). Plausible explanation for these differences in antithrombotic treatment patterns is a lack of alignment of antithrombotic recommendation across the United States and Europe, the differences in lesion characteristics and endovascular procedure between IC and CLTI, and physician’s discretion. 5
Predictors of MACE
In a retrospective study from the Swedish national health care registry, 3-year MACE rates after any lower limb revascularization were 14% in IC and 34% in CLTI. 14 In this prospective study, 3-year MACE rates after EVT were 20% in IC and 42% in CLTI (Figure 3B), suggesting that CLTI has approximately twice MACE rate compared with IC. However, CLTI was merely a borderline signal for MACE. Strikingly, elevated D-dimer, or hypercoagulation, was the strongest predictor of MACE irrespective of PAD stages. In meta-analysis with 2 studies, elevated D-dimer was associated with increased risk of cardiovascular mortality in patients with PAD.15–17 Given that CLTI had more elevated D-dimer than IC, an underlying hypercoagulable state ensures to explain higher MACE risk in CLTI than in IC. Most recently, subanalysis of the COMPASS (Cardiovascular Outcomes for People Using Anticoagulation Strategies) reported that a non-vitamin K antagonist oral anticoagulant, rivaroxaban plus aspirin reduced MACE compared with aspirin alone in PAD patients. 18 A close relationship between hypercoagulation and MACE in the present study supports the benefits of add-on rivaroxaban on the reduction of MACE in PAD patients.
Predictors of Reintervention
Reintervention is a less invasive and repeatable mode as opposed to major amputation. In the present study, 3-year reintervention rates after EVT were 22% in IC and 47% in CLTI (Figure 3C), suggesting that CLTI has approximately twice reintervention rate compared with IC. However, CLTI was merely a borderline signal for reintervention. Intriguingly, elevated ESR, or increased chronic inflammation, was an independent predictor of reintervention irrespective of PAD stages. Previous retrospective studies reported that elevated CRP and inflammatory cytokines were associated with restenosis after femoropopliteal intervention.19,20 Given that CLTI had more elevated ESR than IC, an underlying chronic inflammation ensures to explain higher reintervention risk in CLTI than in IC. In a recent randomized study, an anti-interleukin-1β neutralizing antibody, canakinumab improved maximum and pain-free walking distance in patients with PAD, albeit in the short term. 21 A close relationship between hyperinflammation and reintervention in the present study warrants further investigation for the role of anti-inflammatory therapy on the postinterventional process.
Study Limitations
The present study has several potential limitations. First, our data are prospective and observational in nature, and may be affected by confounding factors. Therefore, results rather than the primary measurements need to be interpreted as hypothesis generating. Second, endovascular strategy, device usage, or management might have varied during the study period. Third, clinical practice behavior due to regional or institutional circumstances might result in incomplete data collection. Indeed, the primary endpoints of MACE and reintervention were censored in 16% and 14% of subjects. Fourth, frequency of antithrombotic agents was regularly assessed, but no data are available regarding switch, interruption, and resumption between each time point, and the effects of “on” and “off” medication fluctuation are unclear. Fifth, no central adjudication of clinical events was done and only those events counted by physicians in each institution were analyzed.
Conclusions
In the context of contemporary PAD, IC and CLTI are considerably heterogeneous phenotypes (Figure 4). The CLTI is characterized not only by more systemic comorbidities and distal disease but also by more inflammatory coagulation disorder as compared with IC. Also, CLTI has approximately twice MACE and reintervention rates than IC, and the underlying inflammatory coagulation disorder per se is independently associated with these outcomes.
Supplemental Material
Supplemental material, sj-tif-1-jet-10.1177_15266028221134886 for Characteristics, Antithrombotic Patterns, and Prognostic Outcomes in Claudication and Critical Limb-Threatening Ischemia Undergoing Endovascular Therapy by Osami Kawarada, Kan Zen, Koji Hozawa, Hideaki Obara, Kentaro Matsubara, Yoshito Yamamoto, Tatsuki Doijiri, Nozomu Tamai, Shigenori Ito, Akihiro Higashimori, Daizo Kawasaki, Hideki Doi, Kensuke Matsushita, Kengo Tsukahara, Katsuo Noda, Masahisa Shimpo, Yuki Tsuda, Shinjo Sonoda, Takuya Taniguchi, Katsuhisa Waseda, Masato Munehisa, Eiji Taguchi, Tatsuya Kinjo, Yohei Sasaki, Kenichiro Yuba, Shinichiro Yamaguchi, Takuo Nakagami, Shinobu Ayabe, Shingo Sakamoto, Takeshi Yagyu, Soshiro Ogata, Kunihiro Nishimura, Hisashi Motomura, Teruo Noguchi, Masaharu Ishihara, Hisao Ogawa and Satoshi Yasuda in Journal of Endovascular Therapy
Acknowledgments
This article was supported in part by Asian Chapter of The International Society of Endovascular Specialists (ISEVS), Endovascular Asia (https://endovascularasia.com), a nonprofit physician education and research meeting. Also, the present study was presented at Endovascular Asia 2022 (https://endovascularasia.com) which was held in Kyoto, on January 22, 2022.
Footnotes
Correction (June 2023): Article updated online to correct “A total of 47 patients with 642 limbs were IC and 145 patients with 166 limbs were CLTI.” to “A total of 547 patients with 642 limbs were IC and 145 patients with 166 limbs were CLTI.” in the Results section.
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: O.K. has received a research grant and consulting fees from Terumo and has received lecture fees from Medicon, Otsuka Pharmaceutical, and Kowa Pharmaceutical. H.O. has received lecture fees from Bayer Pharmaceutical, Bristol-Meyers Squibb, Eisai, Novartis Pharma, Pfizer, Towa Pharmaceutical, and Toa Eiyo. S.Y. has received remuneration for lecture from Bristol-Meyers, Bayer, and Daiichi-Sankyo; has received trust research/joint research funds from Takeda, NEC, Daiichi-Sankyo, Bayer, and Abbott; and is affiliated with endowed department sponsored by Abbott, Terumo, Nihon-Kohden, Medtronic, Japan Lifeline, Otsuka, Ono, Boehringer Ingelheim, Takeda, Kowa, Zeon, Shionogi, Nippon-Shinyaku, Mochida, and Tesco. The remaining authors have nothing to disclose.
Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Scientific Research Grant and Health Labor Sciences Research Grant from Japanese government.
ORCID iDs: Osami Kawarada 
https://orcid.org/0000-0001-5869-3821
Daizo Kawasaki
https://orcid.org/0000-0002-4413-6773
Supplemental Material: Supplemental material for this article is available online.
References
- 1. Weitz JI, Byrne J, Clagett GP, et al. Diagnosis and treatment of chronic arterial insufficiency of the lower extremities: a critical review. Circulation. 1996;94:3026–3049. [DOI] [PubMed] [Google Scholar]
- 2. Aday AW, Matsushita K. Epidemiology of peripheral artery disease and polyvascular disease. Circ Res. 2021;128:1818–1832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kawarada O, Zen K, Hozawa K, et al. Contemporary critical limb ischemia: Asian multidisciplinary consensus statement on the collaboration between endovascular therapy and wound care. Cardiovasc Interv Ther. 2018;33:297–312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Shirasu T, Hoshina K, Yamamoto S, et al. Poor prognosis in critical limb ischemia without pre-onset intermittent claudication. Circ J. 2015;79:1618–1623. [DOI] [PubMed] [Google Scholar]
- 5. Song P, Rudan D, Zhu Y, et al. Global, regional, and national prevalence and risk factors for peripheral artery disease in 2015: an updated systematic review and analysis. Lancet Glob Health. 2019;7:e1020–e1030. [DOI] [PubMed] [Google Scholar]
- 6. Kithcart AP, Beckman JA. ACC/AHA versus ESC guidelines for diagnosis and management of peripheral artery disease: JACC guideline comparison. J Am Coll Cardiol. 2018;72:2789–2801. [DOI] [PubMed] [Google Scholar]
- 7. Creager MA, Loscalzo J. Chapter 275: Arterial diseases of the extremities. In: Jameson J, Fauci AS, Kasper DL, Hauser SL, Longo DL, Loscalzo J. eds. Harrison’s Principles of Internal Medicine. 20th ed. New York, NY: McGraw-Hill. [Google Scholar]
- 8. Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, et al. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project. Registration procedures, event rates and case fatality in 38 populations from 21 countries in 4 continents. Circulation. 1994;90:583–612. [DOI] [PubMed] [Google Scholar]
- 9. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94:496–509. [Google Scholar]
- 10. Brevetti G, Giugliano G, Brevetti L, et al. Inflammation in peripheral artery disease. Circulation. 2010;122:1862–1875. [DOI] [PubMed] [Google Scholar]
- 11. Cooke JP, Wilson AM. Biomarkers of peripheral arterial disease. J Am Coll Cardiol. 2010;55:2017–2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Diehm N, Shang A, Silvestro A, et al. Association of cardiovascular risk factors with pattern of lower limb atherosclerosis in 2659 patients undergoing angioplasty. Eur J Vasc Endovasc Surg. 2006;31:59–63. [DOI] [PubMed] [Google Scholar]
- 13. Peters F, Kuchenbecker J, Kreutzburg T, et al. Long-term effectiveness and safety of initiating statin therapy after index revascularization in patients with peripheral arterial occlusive disease. J Am Heart Assoc. 2020;9:e018338. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Sigvant B, Kragsterman B, Falkenberg M, et al. Contemporary cardiovascular risk and secondary preventive drug treatment patterns in peripheral artery disease patients undergoing revascularization. J Vasc Surg. 2016;64:1009–1017. [DOI] [PubMed] [Google Scholar]
- 15. Kremers B, Wübbeke L, Mees B, et al. Plasma biomarkers to predict cardiovascular outcome in patients with peripheral artery disease: a systematic review and meta-analysis. Arterioscler Thromb Vasc Biol. 2020;40:2018–2032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Vidula H, Tian L, Liu K, et al. Biomarkers of inflammation and thrombosis as predictors of near-term mortality in patients with peripheral arterial disease: a cohort study. Ann Intern Med. 2008;148:85–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Bosevski M, Kostoska S, Tosev S, et al. Usefulness of D-Dimers and fibrinogen plasma determination in patients with polyvascular disease. Angiol Sosud Khir. 2006;12:9–15. [PubMed] [Google Scholar]
- 18. Kaplovitch E, Eikelboom JW, Dyal L, et al. Rivaroxaban and aspirin in patients with symptomatic lower extremity peripheral artery disease: a subanalysis of the COMPASS randomized clinical trial. JAMA Cardiol. 2021;6:21–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Schillinger M, Haumer M, Schlerka G, et al. Restenosis after percutaneous transluminal angioplasty in the femoropopliteal segment: the role of inflammation. J Endovasc Ther. 2001;8:477–483. [DOI] [PubMed] [Google Scholar]
- 20. Guo S, Zhang Z, Wang L, et al. Six-month results of stenting of the femoropopliteal artery and predictive value of interleukin-6: comparison with high-sensitivity C-reactive protein. Vascular. 2020;28:715–721. [DOI] [PubMed] [Google Scholar]
- 21. Russell KS, Yates DP, Kramer CM, et al. A randomized, placebo-controlled trial of canakinumab in patients with peripheral artery disease. Vasc Med. 2019;24:414–421. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material, sj-tif-1-jet-10.1177_15266028221134886 for Characteristics, Antithrombotic Patterns, and Prognostic Outcomes in Claudication and Critical Limb-Threatening Ischemia Undergoing Endovascular Therapy by Osami Kawarada, Kan Zen, Koji Hozawa, Hideaki Obara, Kentaro Matsubara, Yoshito Yamamoto, Tatsuki Doijiri, Nozomu Tamai, Shigenori Ito, Akihiro Higashimori, Daizo Kawasaki, Hideki Doi, Kensuke Matsushita, Kengo Tsukahara, Katsuo Noda, Masahisa Shimpo, Yuki Tsuda, Shinjo Sonoda, Takuya Taniguchi, Katsuhisa Waseda, Masato Munehisa, Eiji Taguchi, Tatsuya Kinjo, Yohei Sasaki, Kenichiro Yuba, Shinichiro Yamaguchi, Takuo Nakagami, Shinobu Ayabe, Shingo Sakamoto, Takeshi Yagyu, Soshiro Ogata, Kunihiro Nishimura, Hisashi Motomura, Teruo Noguchi, Masaharu Ishihara, Hisao Ogawa and Satoshi Yasuda in Journal of Endovascular Therapy