The Impact of Isolated Tibial Disease on Outcomes in the Critical Limb Ischemic Population
Article Outline
Background
Most patients with critical limb ischemia (CLI) have multilevel infrainguinal peripheral arterial disease (M-PAD). One-third of CLI patients will have isolated tibial disease (ITD). The treatments for multilevel disease or ITD differ depending on whether open or endovascular procedures are used, but we questioned whether outcomes from these procedures differ. We evaluated outcomes of CLI patients after open and/or endovascular revascularization for CLI and assessed the impact of disease distribution.
Methods
Four hundred forty-six CLI patients (Rutherford 4-6) who underwent revascularization from 2001 to 2005 were evaluated arteriographically and followed after revascularization with noninvasive testing. Based on arteriographic data, all patients with ITD (occlusions in one or more tibial arteries) were compared with patients with occlusive femoropopliteal disease with or without concomitant tibial occlusions (M-PAD). Patients with disease solely above the inguinal ligament were excluded. Clinical data (survival, amputation-free survival, primary patency, secondary patency, limb salvage, maintenance of ambulation, and maintenance of living status) were acquired from a prospective vascular database, allowing the comparison of revascularization outcomes according to disease distribution.
Results
In this study, 36% of patients had ITD and 64% presented with M-PAD. The severity of ischemia at presentation was rest pain (28.5%), ulceration (42.3%), and gangrene (29.1%). In this study, 92% presented exclusively with infrainguinal disease, and 8% presented with both suprainguinal and infrainguinal disease. Risk factors included diabetes mellitus (61.2%), smoking (61.0%), coronary artery disease (57.9%), hypertension (84.3%), hyperlipidemia (40.4%), obesity (15.5%), and chronic obstructive pulmonary disease (19.3%). In comparing the ITD and M-PAD groups, there was no difference in primary patency at 2 years. All other outcomes were statistically different out to 3 years including survival (50.4% vs. 62.6%; p = 0.0026, hazard ratio [HR] 0.669); amputation-free survival (35.1% vs. 50.2%; p = 0.0062; HR 0.595); limb salvage (65.2% vs. 74.4%; p = 0.0062; HR 0.595); maintenance of ambulation (68.9% vs. 76.9%; p = 0.0352; HR 0.644); maintenance of living status (79.0% vs. 84.8%; p = 0.0403; HR 0.599); and secondary patency (66.8% vs. 74.8%; p = 0.0309; HR 0.665). Multivariate analyses reveal that ITD is not an independent predictor of outcome after controlling for confounding factors, of which tissue loss and end-stage renal disease correlate most consistently with poor clinical outcomes.
Conclusion
After revascularization for CLI, ITD carries a worse prognosis (amputation-free survival, limb salvage, survival, maintenance of ambulation, and independent living status) compared with patients with M-PAD, despite the “greater” disease burden in M-PAD patients. ITD patients are more likely to have confounding factors such as diabetes mellitus, renal disease, and worse ischemia at presentation than those with M-PAD. The recognition of ITD may be helpful in identifying high-risk patients but is not an independent risk factor for poor outcomes.
Introduction
Clinical decision making for patients with critical limb ischemia (CLI) is influenced by physical assessment, complemented by noninvasive testing and formalized based on anatomic assessment. The anatomic assessment consists of the distribution and severity of atherosclerotic lower extremity disease including quality of bypass target/run-off. Most patients have extensive multilevel atherosclerotic peripheral arterial disease, characterized by femoropopliteal and tibial peripheral arterial disease (M-PAD); however, one-third will have isolated tibial disease (ITD), without hemodynamically significant femoropopliteal disease.1 It is unclear why some develop multilevel disease and others have only ITD. It also remains a challenge to predict who will do well after revascularization (and who will not), because many will have at least one of the negative predictive risk factors, such as smoking, diabetes mellitus, chronic kidney disease, and/or advanced age.2, 3, 4, 5
Several infrainguinal atherosclerotic disease classifications have been proposed to assist in risk assessment and treatment preference, but they have limitations in the CLI population. The Trans-Atlantic Inter-Society Consensus (TASC) classified lesion severity from A (simple, focal disease) to D (long, complex occlusions); however, TASC is based on single-level disease and is unable to classify multilevel disease.6 The Society of Vascular Surgery peripheral vascular runoff score used anatomic factors to grade (single-level) tibial disease.7 Unfortunately, this scale has not been consistently predictive of graft patency, limb salvage, or survival.8, 9, 10 Graziani et al.11 proposed a grading scale for multilevel infrainguinal atherosclerotic disease. They described seven morphologic classes and ordered them according to the extent of disease with worsening transcutaneous oxygen levels. It is unknown whether this classification system predicts outcomes after revascularization for CLI. Furthermore, is it the presence of specific risk factors (i.e., diabetes, smoking) or the distribution of disease that predicts outcome?
To answer these questions, we evaluated outcomes of our CLI patients after open and/or endovascular revascularization and assessed the impact of disease distribution and atherosclerotic risk factors.
Methods
The outcomes after revascularization (either endovascular or open surgery) for 446 patients with CLI consecutively treated between January 2001 and December 2005 at the Greenville Hospital System University Medical Center (GHSUMC) were reviewed from a prospective database. Their prerevascularization arteriograms were evaluated and categorized according to the Graziani morphologic class by a single observer (A.A.G.)11 (Table 1). Arteriograms with overlap features were categorized by consensus. Categories 1, 2, and 5 are characterized by ITD, whereas categories 3, 4, 6, and 7 have varying extent of femoropopliteal and/or tibial disease. For the purposes of statistical evaluation, each class was analyzed separately, and those with ITD (ITD) were compared with those with M-PAD.
Table I. Morphologic classification of infrainguinal disease according to Graziani et al.11 with each category representing progressive severity of disease
| Class | Angiographic findings | Patients, n |
|---|---|---|
| 1 | F-P patent with 1 occluded tibial | 31 (6.9%) |
| 2a | F-P patent with 2 occluded tibials (w/ occluded peroneal) | 27 (6.1%) |
| 2b | F-P patent with 2 occluded tibials (with patent peroneal) | 41 (9.2%) |
| 3 | F-P occlusion with patent tibials | 58 (13.0%) |
| 4 | F-P occlusion with 1 occluded tibial | 58 (13.0%) |
| 5 | F-P patent with 3 occluded tibials | 62 (13.9%) |
| 6 | F-P occlusion with 2 occluded tibials | 101 (22.6%) |
| 7 | F-P occlusion with 3 occluded tibials | 68 (15.2%) |
Each patient underwent either endovascular or open surgical repair. All treatments were performed by members of the Vascular Surgery Department. Endovascular therapy consisted of balloon angioplasty with provisional stenting for infrainguinal disease. The 8% of patients with concomitant aortoiliac disease underwent primary stenting prior to infrainguinal revascularization. The open procedures consisted of femoropopliteal, femorotibial, and popliteal-tibial bypass using autologous vein as the primary conduit. If unavailable, PTFE was used with distal anastomotic vein cuff.
The Lower Extremity Prospective Registry
The Vascular Surgery Service at GHSUMC has maintained a prospective vascular registry of all cases performed (∼2500 lower extremity revascularizations) since 1992. A subset of patients with lower extremity PAD has been actively followed with Institutional Review Committee approval. Patients were defined as having CLI if the ankle pressure was <50 mm Hg or toe pressure <30 mm Hg in the Rutherford 4/Fontaine III category or ankle pressure <70 mm Hg or toe pressure <50 mm Hg in the Rutherford 5 or 6 or Fontaine IV category. Each procedure was entered on an Excel spreadsheet (Microsoft Corp., Redmond, WA). Preoperative demographics were obtained at presentation and entered into the database. Functional information (ambulatory status and living situation) was recorded for each patient. Follow-up information was recorded at each follow-up visit. For patients receiving open infrainguinal bypass, follow-up with noninvasive duplex scan–derived graft flow velocities was obtained at 1 month, then every 3 months for the first 18 months, and then every 6 months thereafter. Patients receiving all other types of revascularization (endovascular intervention) were followed with a patient visit and a noninvasive vascular study at 1 month, then every 6 months for 2 years, then annually thereafter. The database information is updated with each visit. As well, the database was scrutinized each summer by independent research workers identifying and entering missing data points or missing patients. For the purpose of the life-table analysis, patients with incomplete follow-up were censored at the last confirmed follow-up date.
Sources used to attain follow-up include the hospital computerized Lifetime Clinical Record, the computerized radiology Picture Archiving Communications System (PACS), the Medicare national database, and the online obituary services of all statewide newspapers.
Description of Time-Dependent Outcomes with Definitions
Patients were analyzed for the following outcome endpoints: reconstruction patency, limb salvage, maintenance of ambulation, maintenance of independent living status, and survival. Influence of the following patient characteristics on outcomes with primary focus on amputation-free survival were examined: the presence of coronary artery disease (high risk by the Eagle criteria),12 chronic obstructive pulmonary disease (COPD), presence of diabetes mellitus, end-stage renal disease, presence of hypertension, hyperlipidemia, obesity (body mass index ≥30 kg/m2), race, gender, history of smoking, type of intervention (endovascular versus open surgery), and level of disease (ITD versus M-PAD). Patient outcomes dependent on the presence or absence of diabetes were also evaluated for the ITD and M-PAD groups.
Patency of the vascular intervention was assessed by physical examination, ankle-brachial index determination, duplex ultrasound evaluation, and, in some cases, arteriography. An ankle-brachial index of greater than 0.15 after intervention was thought to represent a successful, patent intervention. Conversely, a drop in ankle-brachial index of greater than 0.15 was considered significant, and an absolute ankle-brachial index within 0.15 of baseline was considered a failed intervention. For the purpose of this study, all reinterventions, whether for a thrombosed original intervention or a failing intervention, were considered to result in secondary patency.
Preoperative ambulatory status prospectively entered into the database was characterized as ambulatory (independent ambulation out of house), ambulatory/homebound (ambulatory in the home only), nonambulatory/transfer (e.g., uses legs to transfer from bed to chair or from the chair to the commode), or bedridden. Abnormal ambulatory status was usually a function of other medical comorbidities such as arthritis, cerebrovascular and cardiovascular disease, or advanced age and nonvascular morbidity. Preoperative living status was characterized as independent, defined as living in an independent dwelling without external assistance; or nonindependent, defined as living in an assisted living environment or a private residence with external assistance for activities of daily living. As with ambulatory status, living status was determined as the living status immediately prior to intervention.
A change in ambulatory function was defined as permanent postoperative change, despite full recovery from the revascularization, in ambulatory classification (e.g., ambulatory to nonambulatory/transfer, ambulatory to nonambulatory/bedridden, or nonambulatory/transfer to nonambulatory/bedridden). A decline in independent living status was defined as permanent change after intervention to an assisted living residence or incorporation of permanent help into the postoperative domicile to enhance functions of daily living. Short-term assisted living during recovery was not considered loss of independence.
Revascularization was considered to be clinically successful if each of the following occurred: reconstruction patency until wound healing, limb salvage for 1 year, maintenance of ambulation for 1 year, and survival for 6 months.
Statistical Analysis
Two-group comparisons (ITD vs. M-PAD) were analyzed using Student's t-test for continuously distributed data and the χ2 for categorical data (i.e., bivariate analyses). The Kaplan-Meier product limit method was used to estimate median survival times at 1, 2, and 3 years. The logrank test was used to assess the difference in survival curves. Cox proportional hazards multivariate modeling was used to assess each time-dependent outcome (i.e., dependent variables) and possible independent predictors and/or confounders. p-Values <0.05 were considered indicative of statistical significance. All analyses were conducted using SAS Statistical software (Version 9.1; SAS Institute Inc., Cary, NC).
Results
There were 446 patients with CLI and classified as having rest pain (28.5%), ischemic ulceration (42.3%), and tissue loss (29.1%). The mean age of the entire cohort was 70.0 ± 11.7 years (range, 36-93 years). Mean follow-up time was 30.7 months (SD = 21.3, minimum = 0, maximum = 85 months). The cohort was predominantly male (58.3%) with a high prevalence of diabetic patients (60.5%) and tobacco users (60.9%). The initial arteriograms were classified according to Graziani (Table I). Classes 1, 2a, 2b, and 5 were identified as having ITD, which included 36% of patients. Classes 3, 4, 6, and 7 were identified as having M-PAD, which included 64% of patients. Figure 1 shows the median amputation-free survival of each Graziani class as well as the composite of the ITD and M-PAD groups. There was no clear relationship between Graziani class and amputation-free survival or any other objective endpoint. However, when analyzing for distribution of disease, there was a clear statistically significant difference between ITD versus M-PAD groups (Table II).
Fig. 1
Amputation-free survival according to Graziani class and grouped according to distribution of isolated tibial disease (ITD) versus multilevel peripheral arterial disease (M-PAD).
Table II. Kaplan-Meier analysis and logrank p-value to assess the relationship between patient/disease factors and clinical outcomes
| Factor | Survival | Amputation-free survival | Limb salvage | Maintenance of ambulation | Maintenance of living status | Primary patency | Secondary patency |
|---|---|---|---|---|---|---|---|
| ITD vs. MPAD | 0.002 | <0.001 | 0.005 | 0.033 | 0.040 | 0.504 | 0.023 |
| Age: <65 vs. ≥65 | 0.011 | 0.258 | 0.197 | 0.307 | 0.071 | 0.275 | 0.208 |
| Race | 0.922 | 0.335 | 0.007 | 0.143 | 0.651 | <0.001 | 0.002 |
| Gender | 0.795 | 0.976 | 0.174 | 0.002 | 0.071 | 0.397 | 0.292 |
| Ischemic presentation | <0.001 | <0.001 | <0.001 | 0.092 | 0.446 | 0.141 | <0.001 |
| Smoker | 0.601 | 0.866 | 0.582 | 0.267 | 0.381 | 0.143 | 0.822 |
| COPD | <0.001 | 0.008 | 0.061 | 0.848 | 0.217 | 0.221 | 0.358 |
| Diabetic | 0.050 | 0.001 | <0.001 | 0.003 | 0.388 | 0.074 | <0.001 |
| Diabetic and smoker | 0.383 | 0.047 | 0.031 | 0.166 | 0.585 | 0.027 | 0.031 |
| ESRD | <0.001 | <0.001 | <0.001 | <0.001 | 0.010 | <0.001 | <0.001 |
| CAD | 0.003 | 0.007 | 0.011 | <0.001 | 0.039 | 0.351 | 0.013 |
| Hypertension | 0.773 | 0.668 | 0.366 | 0.385 | 0.381 | 0.760 | 0.968 |
| Hyperlipidemia | 0.883 | 0.576 | 0.595 | 0.923 | 0.466 | 0.176 | 0.779 |
| Obesity | 0.763 | 0.684 | 0.553 | 0.453 | 0.041 | 0.989 | 0.672 |
| Hypertension, smoking, obesity | 0.608 | 0.463 | 0.444 | 0.702 | 0.078 | 0.139 | 0.640 |
| Dementia | <0.001 | <0.001 | 0.789 | 0.013 | <0.001 | 0.327 | 0.945 |
| Preambulation | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | 0.115 | 0.002 |
| Pre–living status | 0.085 | 0.045 | 0.353 | 0.504 | 0.321 | 0.358 | 0.347 |
| Treatment type | <0.001 | 0.001 | 0.011 | 0.011 | 0.062 | 0.981 | 0.014 |
Bivariate analyses of the ITD and M-PAD groups are presented in Table III. These groups differ on the following factors: age, ischemic presentation, presence of diabetes, end-stage renal disease (ESRD), smoking, COPD, and treatment received. Patients with ITD compared with M-PAD were slightly older (71.5 vs. 69.2 years, p = 0.055), were more likely to be diabetic (73.9% vs. 54.0%; p <0.001) and to have chronic kidney disease or ESRD (43.5% vs. 29.1%; p = 0.004), were less likely to smoke or have COPD (60.3% vs. 91.6%; p < 0.001), and had worse ischemia at the time of presentation (ulceration or gangrene: 87.6% vs. 62.5%; p < 0.001). Endovascular treatment (balloon angioplasty or balloon angioplasty plus stent) was more common (74.4%) than open surgery (bypass surgery with autologous or synthetic bypass) (25.6%) in the ITD patients. Open surgery (56.1%) was more commonly used for M-PAD patients compared with endovascular treatment (43.9%). These differences were statistically significant to p < 0.001.
Table III. Patient characteristics by type of critical limb ischemia
| Patient characteristic | ITD group (n = 161) | M-PAD group (n = 285) | p-Value |
|---|---|---|---|
| Age (mean ± SD), y | 71.5 ± 11.9 | 69.2 ± 11.6 | 0.055 |
| Follow-up (mean ± SD), mo | 27.8 ± 19.2 | 32.4 ± 22.3 | 0.021 |
| Race, No. (%) | |||
 White | 109 (67.7) | 208 (73.0) | 0.24 |
| African American/other | 52 (32.3) | 77 (27.0) | |
| Gender, No. (%) | |||
| Male | 95 (59.0) | 165 (57.9) | 0.82 |
| Female | 66 (41.0) | 120 (42.1) | |
| Ischemic presentation, No. (%) | |||
| Rest pain | 20 (12.4) | 107 (37.5) | <0.001 |
| Ulceration | 77 (47.8) | 112 (39.3) | |
| Tissue loss | 64 (39.8) | 66 (23.2) | |
| Smoker, No. (%) | 75 (46.6) | 197 (69.1) | <0.001 |
| COPD, No. (%) | 22 (13.7) | 64 (22.5) | 0.024 |
| Diabetic, No. (%) | 119 (73.9) | 154 (54.0) | <0.001 |
| Diabetic and smoker, No. (%) | 53 (32.9) | 100 (35.1) | 0.64 |
| ESRD, No. (%) | |||
| None | 91 (56.5) | 202 (70.9) | 0.004 |
| Renal insufficiency | 28 (17.4) | 41 (14.4) | |
| Dialysis dependent | 42 (26.1) | 42 (14.7) | |
| CAD, No. (%) | 95 (59.0) | 163 (57.2) | 0.71 |
| Hypertension, No. (%) | 135 (83.9) | 241 (84.6) | 0.84 |
| Hyperlipidemia, No. (%) | 56 (34.8) | 124 (43.5) | 0.07 |
| Obese (body mass index >30 kg/m2), No. (%) | 26 (16.1) | 43 (15.1) | 0.77 |
| Hypertension, smoking, obesity (%) | 13 (8.1) | 21 (7.4) | 0.79 |
| Dementia, No. (%) | 14 ( 8.7) | 27 ( 9.5) | 0.78 |
| Preambulation, No. (%) | |||
| Ambulatory out of home | 122 (75.8) | 230 (80.7) | 0.47 |
| Ambulatory in home only | 31 (19.2) | 43 (15.1) | |
| Transfer only | 8 ( 5.0) | 12 ( 4.2) | |
| Pre–living status, No. (%) | |||
| Independent | 150 (93.2) | 274 (96.1) | 0.33 |
| Nonindependent | 11 ( 6.9) | 11 ( 3.9) | |
| Treatment type, No. (%) | |||
| Endovascular | 119 (74.4) | 122 (43.9) | <0.001 |
| Open surgery | 41 (25.6) | 156 (56.1) |
The ITD and M-PAD patients had similar primary patency rates at 1-year (54.2% vs. 50.2%; p = 0.52, hazard ratio [HR] 1.09) or 2-year follow-up whether treated with endovascular or open techniques (Table II). Yet the patients with ITD fared worse regarding survival (50.4% vs. 62.6%; p = 0.0026, HR 0.669), amputation-free survival (35.1 vs. 50.2%; p = 0.0062, HR 0.595), limb salvage (65.2% vs. 74.4%; p = 0.0062, HR 0.595), maintenance of ambulatory (68.9% vs. 76.9%; p = 0.0352, HR 0.644), maintenance of independent living status (79.0% vs. 84.8%; p = 0.0403, HR 0.599), and secondary patency (66.7% vs. 74.8%; p = 0.0309, HR 0.665) at 3-year follow-up (Table III).
Table II shows the Kaplan-Meier analysis and logrank p-value for each clinical outcome variable. This assesses the relationship between patient/disease factors and outcome variables. We conducted a multivariate analysis to adjust for the differences in Table II as well as in Figure 2. These multivariate analyses are shown in Table IV, Table V, Table VI, Table VII, Table VIII. Independent predictors of outcome were as follows—survival: ischemic tissue loss, COPD, dialysis dependency, and dementia; amputation-free survival: ischemic tissue loss, dialysis dependency, dementia, and perambulation status; limb salvage: COPD, diabetes, dialysis dependency, coronary artery disease, and preambulation status; primary patency: race and dialysis dependency; and scondary patency dialysis dependency and preambulation status. After running all of the models, CLI type (ITD vs. M-PAD) was found to NOT be an independent predictor of outcome, after controlling for other important risk factors.
Fig. 2
Kaplan-Meier analysis of CLI patients with isolated tibial disease (ITD) versus multilevel peripheral arterial disease (M-PAD).
Table IV. Full Cox proportional hazards model for dependent variable—Survival
| Factor | Hazard ratio (95% confidence interval) | p-Value |
|---|---|---|
| Age (y) | 1.02 (1.01-1.03) | 0.005a |
| Presentation | ||
| Rest pain | Reference group | … |
| Ulceration | 1.09 (0.75-1.58) | 0.654 |
| Tissue loss | 1.67 (1.13-2.49) | 0.011a |
| Smoker | ||
| No | Reference group | … |
| Yes | 1.16 (0.87-1.56) | 0.313 |
| COPD | ||
| No | Reference group | … |
| Yes | 1.91 (1.34-2.72) | <0.001a |
| Diabetes | ||
| No | Reference group | … |
| Yes | 1.00 (0.73-1.37) | 0.999 |
| ESRD | ||
| None | Reference group | … |
| Renal insufficiency | 1.12 (0.76-1.64) | 0.580 |
| Dialysis dependent | 2.51 (1.79-3.52) | <0.001a |
| CAD | ||
| No | Reference group | … |
| Yes | 1.25 (0.93-1.67) | 0.135 |
| Dementia | ||
| No | Reference group | … |
| Yes | 1.60 (1.00-2.55) | 0.050a |
| Preambulation | ||
| None; transfer only | Reference group | … |
| Yes, some ambulation | 1.22 (0.93-1.59) | 0.146 |
| Treatment type | ||
| Endovascular | Reference group | … |
| Open | 0.89 (0.66-1.21) | 0.460 |
| Disease type | ||
| ITD | Reference group | … |
| M-PAD | 0.78 (0.58-1.06) | 0.107 |
aStatistically significant. |
Table V. Full Cox proportional hazards model for dependent variable—Amputation-free survival
| Factor | Hazard ratio (95% confidence interval) | p-Value |
|---|---|---|
| Age (y) | 1.01 (0.99-1.02) | 0.346 |
| Presentation | ||
| Rest pain | Reference group | … |
| Ulceration | 1.07 (0.76-1.50) | 0.711 |
| Tissue loss | 1.71 (1.20-2.44) | 0.003a |
| Smoker | ||
| No | Reference group | … |
| Yes | 1.16 (0.88-1.53) | 0.289 |
| COPD | ||
| No | Reference group | … |
| Yes | 1.37 (0.99-1.89) | 0.057 |
| Diabetes | ||
| No | Reference group | … |
| Yes | 1.19 (0.89-1.59) | 0.232 |
| ESRD | ||
| None | Reference group | … |
| Renal insufficiency | 0.96 (0.67-1.38) | 0.834 |
| Dialysis dependent | 2.05 (1.48-2.82) | <0.001a |
| CAD | ||
| No | Reference group | … |
| Yes | 1.19 (0.91-1.55) | 0.197 |
| Dementia | ||
| No | Reference group | … |
| Yes | 1.64 (1.07-2.49) | 0.022a |
| Preambulation | ||
| None; transfer only | Reference group | … |
| Yes, some ambulation | 1.51 (1.18-1.93) | 0.001a |
| Preliving status | ||
| Independent | Reference group | … |
| Nonindependent | 0.75 (0.54-1.03) | 0.073 |
| Treatment type | ||
| Endovascular | Reference group | … |
| Open | 0.94 (0.72-1.23) | 0.642 |
| Disease type | ||
| ITD | Reference group | … |
| M-PAD | 0.78 (0.60-1.03) | 0.076 |
aStatistically significant. |
Table VI. Full Cox proportional hazards model for dependent variable—Limb salvage
| Factor | Hazard ratio (95% confidence interval) | p-Value |
|---|---|---|
| Age (y) | 0.98 (0.97-1.00) | 0.077 |
| Race | ||
| Caucasian | Reference group | … |
| African American/other | 1.39 (0.92-2.10) | 0.115 |
| Presentation | ||
| Rest pain | Reference group | … |
| Ulceration | 0.95 (0.54-1.68) | 0.867 |
| Tissue loss | 1.45 (0.82-2.57) | 0.204 |
| Smoker | ||
| No | Reference group | … |
| Yes | 1.13 (0.74-1.73) | 0.580 |
| COPD | ||
| No | Reference group | … |
| Yes | 0.47 (0.24-0.92) | 0.026a |
| Diabetes | ||
| No | Reference group | … |
| Yes | 1.71 (1.03-2.83) | 0.037a |
| ESRD | ||
| None | Reference group | … |
| Renal insufficiency | 0.74 (0.39-1.39) | 0.352 |
| Dialysis dependent | 1.65 (1.02-2.66) | 0.042a |
| CAD | ||
| No | Reference group | … |
| Yes | 1.54 (1.01-2.36) | 0.046a |
| Preambulation | ||
| None; transfer only | Reference group | … |
| Yes, some ambulation | 1.72 (1.24-2.37) | 0.001a |
| Treatment type | ||
| Endovascular | Reference group | … |
| Open | 0.83 (0.54-1.28) | 0.407 |
| Disease type | ||
| ITD | Reference group | … |
| M-PAD | 0.86 (0.56-1.30) | 0.460 |
aStatistically significant. |
Table VII. Full Cox proportional hazards model for dependent variable—Primary patency
| Factor | Hazard ratio (95% confidence interval) | p-Value |
|---|---|---|
| Age (y) | 1.00 (0.99-1.01) | 0.907 |
| Race | ||
| White | Reference group | … |
| African American/Other | 1.39 (1.04-1.87) | 0.028a |
| Presentation | ||
| Rest pain | Reference group | … |
| Ulceration | 0.91 (0.64-1.29) | 0.587 |
| Tissue loss | 1.11 (0.76-1.62) | 0.592 |
| Smoker | ||
| No | Reference group | … |
| Yes | 1.33 (0.97-1.82) | 0.078 |
| COPD | ||
| No | Reference group | … |
| Yes | 0.71 (0.48-1.06) | 0.094 |
| Diabetes | ||
| No | Reference group | … |
| Yes | 1.25 (0.91-1.73) | 0.164 |
| ESRD | ||
| None | Reference group | … |
| Renal insufficiency | 1.04 (0.69-1.60) | 0.824 |
| Dialysis dependent | 1.80 (1.25-2.60) | 0.002a |
| Treatment type | ||
| Endovascular | Reference group | … |
| Open | 1.03 (0.77-1.37) | 0.862 |
| Disease type | ||
| ITD | Reference group | … |
| M-PAD | 1.28 (0.93-1.75) | 0.127 |
aStatistically significant. |
Table VIII. Full Cox proportional hazards model for dependent variable—secondary patency
| Factor | Hazard ratio (95% confidence interval) | p-Value |
|---|---|---|
| Age (y) | 0.99 (0.97-1.00) | 0.128 |
| Race | ||
| White | Reference group | … |
| African American/other | 1.47 (0.98-2.20) | 0.061 |
| Presentation | ||
| Rest pain | Reference group | … |
| Ulceration | 0.89 (0.52-1.52) | 0.666 |
| Tissue loss | 1.39 (0.80–2.41) | 0.241 |
| Smoker | ||
| No | Reference group | … |
| Yes | 1.11 (0.73-1.71) | 0.622 |
| COPD | ||
| No | Reference group | … |
| Yes | 0.65 (0.36-1.17) | 0.149 |
| Diabetes | ||
| No | Reference group | … |
| Yes | 1.55 (0.95-2.53) | 0.080 |
| ESRD | ||
| None | Reference group | … |
| Renal insufficiency | 0.93 (0.52-1.68) | 0.817 |
| Dialysis dependent | 1.77 (1.10-2.84) | 0.019a |
| CAD | ||
| No | Reference group | … |
| Yes | 1.43 (0.94-2.17) | 0.093 |
| Preambulation | ||
| None ; transfer only | Reference group | … |
| Yes, some ambulation | 1.51 (1.08-2.11) | 0.016a |
| Treatment type | ||
| Endovascular | Reference group | … |
| Open | 0.78 (0.51-1.19) | 0.252 |
| Disease type | ||
| ITD | Reference group | … |
| M-PAD | 0.97 (0.64-1.47) | 0.879 |
aStatistically significant. |
Further evaluation of the ITD patients revealed that 119 of the 161 (74%) were diabetic. The diabetics compared with the nondiabetics in this group (n = 42) were more likely to have dialysis-dependent renal disease (32.8% vs. 7.1%; p = 0.003) and either ulceration or gangrene on presentation (92.4% vs. 73.8%; p = 0.012). The nondiabetics were older (75.9 ± 13.3 vs. 69.9 ± 11.0 years; p = 0.0047). There were no other statistically significant differences in the risk factors or characteristics of these ITD patients, such as race, sex, smoking status, or duration of follow-up. Logrank testing showed no difference in time dependent outcomes of survival (p = 0.37), amputation-free survival (p = 0.93), limb salvage (p = 0.25), maintenance of ambulation (p = 0.36), maintenance of independent living status (p = 0.32), primary (p = 0.72) or secondary patency (p = 0.45) based on diabetic status.
Of the 285 patients in the M-PAD group, 154 (54%) were diabetic. The diabetics compared with the nondiabetics in this group (n = 131) had more severe ischemia (ulceration or gangrene: 76.6% vs. 45.8%; p < 0.001) and dialysis-dependent renal disease (22.1% vs. 6.1%; p < 0.001). The diabetics were more likely to have lipid abnormalities (52.6% vs. 32.8%; p < 0.001) and obesity (22.1% vs. 6.9%; p < 0.001). There were no other statistically significant differences in the risk factors or characteristics of these M-PAD patients, such as age, race, sex, smoking status, or duration of follow-up. Logrank testing showed statistically significant worse time-dependent outcomes of amputation-free survival (p = 0.002), limb salvage (p < 0.001), and maintenance of ambulation (p = 0.008) in diabetic patients with M-PAD. There are no significant differences in time-dependent outcomes of survival (p = 0.29), maintenance of independent living status (p = 0.85), or primary (p = 0.09) or secondary (p = 0.41) patency based on diabetic status. Cox models reveal that the independent predictors of amputation-free survival were the presence of severe ischemia (ulceration or gangrene) or dialysis-dependent renal disease.
Clinical success (ulcer healing, survival, limb salvage, maintenance of ambulation) after revascularization occurred similarly in the ITD and M-PAD groups (40.4% vs. 45.6%, p = 0.28). In the diabetic ITD patients, clinical success occurred in 37.8% compared with 47.6% in the nondiabetic group (p = 0.28). However, in the diabetic M-PAD patients, clinical success occurred in 36.4% compared with 56.5% (p < 0.001) in the nondiabetics.
Discussion
This study evaluated the outcomes after revascularization of patients with CLI. Despite more extensive disease in patients with M-PAD compared with ITD, their functional outcomes and clinical success rates were better. The ITD patients had a higher incidence of diabetes, dialysis-dependent renal disease, and worse ischemia at the time of presentation, all well-recognized negative predictive factors.
This is the first study describing the applicability of the Graziani classification system after revascularization for CLI. The original classification used transcutaneous oxygen levels as objective evidence for the system.11 Our evaluation failed to identify a direct relationship between worsening Graziani class and outcomes such as amputation-free survival or any of the functional measures of quality of life. Clearly there was a difference in outcomes based on the distribution of disease that combined classes 1, 2, and 5 as ITD versus classes 3, 4, 6, and 7, which reflect M-PAD. Even when controlling for number of diseased tibial arteries (Table I, Fig. 1), patients with M-PAD did better than those with ITD.
Most studies fail to differentiate ITD from M-PAD.13, 14, 15, 16, 17, 18 Outcomes from tibial intervention are reported as “tibial or below knee interventions” despite the inclusion of a substantial number of concomitant femoropopliteal interventions. Giles et al.14 published their series of 176 CLI limbs treated with infrapopliteal angioplasty. Concomitant femoropopliteal angioplasty and stenting was performed on 102 of the 163 (63%) patients. They analyzed their data according to the tibial artery TASC classification but did not separately account for the type, extent, and impact of the proximal disease on outcome. In the Bypass versus Angioplasty in Severe Ischemia of the Leg (BASIL) study, 80% of patients underwent superficial femoral artery angioplasty with 62% also requiring more distal intervention.15 Dorros et al.16 reported on 235 patients who underwent tibial artery angioplasty for CLI. Proximal artery intervention was performed in 59% of these patients without a separate analysis of each group. Kudo et al.17 reported a proximal intervention rate of 81% in those undergoing below knee angioplasty for CLI. Faglia et al.18 analyzed procedures on 221 patients with diabetic ulcers (42% had infrapopliteal angioplasty, 52% had both femoropopliteal and tibial intervention) without separate analysis. Our data suggest that patients with ITD are more likely to have adverse outcomes than those with M-PAD. Multivariate analyses clarifies the impact of disease distribution as a nonindependent risk factor, but rather these patients are more likely to have confounding variables such as ESRD or severe ischemia at the time of presentation.
Wound healing has been infrequently studied as an endpoint after revascularization.19, 20 Soderstrom et al.20 reported on 148 patients treated with infrainguinal bypass for CLI. Their patient demographics were not dissimilar to ours, with 58% male, 50% diabetic, median age of 76 years, and 46% presenting with gangrene. They combined amputation-free survival with complete wound healing and noted only a 50% success rate at 1 year. They did not include maintenance of ambulation in their definition of clinical success, which may explain our lower overall rates. Our diabetic patients had less success (36% ITD, 37% M-PAD) compared with the nondiabetic patients (47% ITD, 56% M-PAD). Our patients with dialysis-dependent renal failure were more likely to be diabetic and to have worse ischemia at the time of presentation. The combination of these potent risk factors in association with predominantly ITD portends a worse prognosis particularly in regard to amputation-free survival (p = 0.07).
We cannot exclude the method of revascularization as a contributing factor to overall success. We have previously shown that treatment choice, either endovascular or open surgery, can be standardized by using a Lower Extremity Grading Score (LEGS).21, 22 When using the LEGS algorithm, similar outcomes after revascularization were noted regardless of treatment type. These observations were reproduced in this study because the primary patency rates of both open and endovascular revascularization did not differ, and the type of revascularization, when analyzed as an independent variable (HR 1.10 [95% confidence interval (CI) 0.85-1.43], p = 0.47), did not predict success. The small number of patients in the ITD group who underwent open revascularization limits further comparison of revascularization method. Endovascular and open treatment was included in all multivariate models and was found to be unrelated to clinical outcome. These findings are consistent with other publications from our institution.23, 24 Other recent reports suggest that multilevel endovascular treatment outperforms endovascular treatment for ITD.25
Thirty-nine percent of our ITD patients presented with gangrene compared with only 23% of the M-PAD patients (HR 1.79 [95% CI 1.25-2.55]; p = .001). Anecdotally, patients with CLI who present with a palpable popliteal pulse are at the highest risk for unsuccessful revascularization and limb loss. These data seem to substantiate this thought. Perhaps we hesitate to intervene and delay revascularization, leading to more advanced ischemia (gangrene vs. rest pain) at the time of presentation. We may also underestimate the degree of ischemia in a patient with a palpable popliteal pulse but no pedal pulse.
What about the patients with ITD who were nondiabetic? These patients were older (75.9 vs. 69.9 years, p = 0.047), had less dialysis-dependent renal disease (7.1% vs. 32.8%, p = 0.003), and were otherwise just like the diabetics. Ohare et al.26 established that patients with dialysis-dependent renal disease are extremely high risk for amputation within the first year after revascularization. Other studies have shown a secondary patency rate of 54.7% in the dialysis-dependent population after open revascularization, with a limb salvage rate of 53% at 36 months. Only 40% achieved clinical success.27 So, it remains unclear as to why these nondiabetic patients had outcomes consistent with diabetics or patients with dialysis-dependent renal disease. Because glucose tolerance tests or glycosylated hemoglobin levels were not obtained routinely, many “nondiabetic” ITD patients may indeed have a preclinical diabetic state or glucose intolerance. This raises the question as to whether hematologic screening of these patients for diabetes would be of benefit. These demographics further suggest that Buerger's disease or a systemic vasculitis, which can present with ITD, is an unlikely explanation as to etiology.
Multivariate analysis also identified COPD as a significant risk factor for M-PAD and as less prevalent in the ITD patients. More smokers were identified in the M-PAD group (32% more than ITD), and COPD was 38% more common. Clearly, smoking leads to COPD and lower blood oxygen levels would predispose to lower tissue oxygen levels. These patients are also thought to have a shorter life expectancy, lower functional levels, and higher operative morbidity and mortality than those without COPD.28
Study Limitations
This is a retrospective review of our experience over a 5-year period. It carries all the limitations of a nonrandomized trial. Clearly, interventions evolve that may affect outcomes in a nonquantifiable manner. The impact of diabetes on tibial disease is a well-recognized risk factor, and biochemical analysis should have been performed on all CLI patients. Metabolic syndrome, prediabetes, or glucose intolerance could be involved in the disease processes of those “nondiabetic” patients. There may be some overlap of morphologic classes that could influence our results, but arteriograms were assigned to a Graziani class by one observer to maximize consistency.
Conclusion
After revascularization for CLI, patients with ITD carry a worse prognosis (amputation-free survival, limb salvage, survival, maintenance of ambulation, and independent living status) compared with patients with M-PAD (Fig. 3). There did not appear to be a linear relationship between the number of diseased tibial arteries and outcomes. Despite the “greater” disease burden in M-PAD patients, revascularization is clinically more successful. These differences are most pronounced in diabetic patients regarding functional outcomes and clinical success. Patients with dialysis-dependent renal disease and those with COPD also experience worse outcomes. Risk assessment must go beyond severity of ischemia, placing ITD patients in the highest risk category.
Fig. 3
Summary of outcomes for patients with CLI.
References
- Angiographic evaluation of peripheral arterial occlusive disease and its role as a prognostic determinant for major amputation in diabetic subjects with foot ulcers. Diabetes Care. 1998;21:625–630
- . Meta-analysis of infrapopliteal angioplasty for chronic critical limb ischemia. J Vasc Surg. 2008;47:975–981
- . Infrapopliteal angioplasty for limb salvage in the setting of renal failure: do the results justify its use?. Ann Vasc Surg. 2005;19:762–768
- . Angiographic distribution of lower extremity atherosclerosis in patients with and without diabetes. Diab Med. 2002;19:366–370
- 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
- . Inter-Society Consensus for the Management of Peripheral Arterial Disease. Transatlantic Intersociety Consensus (TASC). J Vasc Surg. 2007;45:S8–S9
- . Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg. 1997;26:517–538
- . Angiographic runoff score as a predictor of outcome following femorocrural bypass surgery. Eur J Vasc. Endovasc Surg. 1999;17:480–485
- . Predictive factors for adverse outcome of pedal bypasses. Eur J Vasc Endovasc Surg. 1999;18:138–143
- . Evaluation of a proposed standard reporting system for preoperative angiograms in infrainguinal bypass procedures: angiographic correlates of measured runoff resistance. J Vasc Surg. 1988;7:379–385
- Vascular involvement in diabetic subjects with ischemic foot ulcer: a new morphologic categorization of disease severity. Eur J Vasc Endovasc Surg. 2007;33:453–460
- Guidelines for perioperative cardiovascular evaluation for non-cardiac surgery. Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). J Am Coll Cardiol. 1996;27:910–948
- Patterns of artery disease in 450 patients undergoing revascularization for critical limb ischemia: Implications for clinical trial design. J Vasc Surg. 2008;47:995–1000
- . Infrapopliteal angioplasty for critical limb ischemia: relation of TransAtlantic InterSociety Consensus class to outcome in 176 limbs. J Vasc Surg. 2008;48:128–136
- . Bypass versus Angioplasty in Severe Ischaemia of the Leg (BASIL): multicenter randomized controlled trial. Lancet. 2005;366:1925–1934
- . Tibioperoneal (outflow lesion) angioplasty can be used as primary treatment in 235 patients with critical limb ischemia: 5-year follow-up. Circulation. 2001;104:2057–2062
- . The effectiveness of percutaneous transluminal angioplasty for the treatment of critical limb ischemia: a 10-year experience. J Vasc Surg. 2005;41:423–435
- Extensive use of peripheral angioplasty, particularly infrapopliteal, in the treatment of ischaemic diabetic foot ulcers: clinical results of multicntric study of 221 consecutive diabetic subjects. J Intern Med. 2002;252:225–232
- . Complete ulcer healing as primary endpoint in studies on critical limb ischemia? A critical reappraisal. Eur J Vasc Endovasc Surg. 2007;25:229–234
- . Healing of ischaemic tissue lesions after infrainguinal bypass surgery for critical leg ischaemia. Eur J Vasc Endovasc Surg. 2008;36:90–95
- The invasive treatment of chronic limb ischemia according to the Lower Extremity Grading System (LEGS) score: a 6-month report. J Vasc Surg. 2004;39:1268–1276
- Does a standardization tool to direct invasive therapy for symptomatic lower extremity peripheral arterial disease improve outcomes?. J Vasc Surg. 2004;40:907–915
- Determinants of functional outcome after revascularization for critical limb ischemia: an analysis of 1000 consecutive vascular interventions. J Vasc Surg. 2006;44:747–755
- . One-year prospective quality-of-life outcomes in patients treated with angioplasty for symptomatic peripheral arterial disease. J Vasc Surg. 2006;44:296–302
- . Improved outcomes are associated with multilevel endovascular intervention involving the tibial vessels compared with isolated tibial intervention. J Vasc Surg. 2009;49:638–643
- Influence of renal insufficiency on limb loss and mortality after initial lower extremity surgical revascularization. J Vasc Surg. 2004;39:709–716
- . Surgical revascularization in patients with end-stage renal disease: results using a new paradigm in outcomes assessment. Am Surg. 2007;73:598–605
- . A simple subclassification of American Society of Anesthesiology III patients undergoing peripheral revascularization based on functional capacity. J Vasc Surg. 2008;47:766–772
PII: S0890-5096(09)00315-X
doi:10.1016/j.avsg.2009.07.034
© 2010 Published by Elsevier Inc.
