Author + information
- Received December 11, 2014
- Revision received June 18, 2015
- Accepted July 2, 2015
- Published online September 1, 2015.
- Osamu Iida, MD∗,
- Masato Nakamura, MD, PhD†∗ (, )
- Yasutaka Yamauchi, MD‡,
- Masashi Fukunaga, MD§,
- Yoshiaki Yokoi, MD, PhD‖,
- Hiroyoshi Yokoi, MD¶,
- Yoshimistu Soga, MD¶,
- Kan Zen, MD, PhD#,
- Nobuhiro Suematsu, MD, PhD∗∗,
- Naoto Inoue, MD††,
- Kenji Suzuki, MD††,
- Keisuke Hirano, MD‡‡,
- Yoshiaki Shintani, MD§§,
- Yusuke Miyashita, MD, PhD‖‖,
- Kazushi Urasawa, MD¶¶,
- Ikuro Kitano, MD##,
- Taketsugu Tsuchiya, MD, PhD∗∗∗,
- Kenji Kawamoto, MD†††,
- Terutoshi Yamaoka, MD‡‡‡,
- Michitaka Uesugi, MD§§§,
- Toshiro Shinke, MD, PhD‖‖‖,
- Yasuhiro Oba, MD¶¶¶,
- Norihiko Ohura, MD###,
- Masaaki Uematsu, MD, PhD∗,
- Mitsuyoshi Takahara, MD, PhD∗∗∗∗,
- Toshimitsu Hamasaki, MSc, PhD††††,
- Shinsuke Nanto, MD, PhD‡‡‡‡,
- OLIVE Investigators
- ∗Cardiovascular Center, Kansai Rosai Hospital, Amagasaki, Japan
- †Division of Cardiovascular Medicine, Toho University, Ohashi Medical Center, Tokyo, Japan
- ‡Cardiovascular Center, Kikuna Memorial Hospital, Yokohama, Japan
- §Cardiovascular Division, Department of Internal Medicine, Hyogo College of Medicine, Nishinomiya, Japan
- ‖Department of Cardiology, Kishiwada Tokushukai Hospital, Osaka, Japan
- ¶Cardiovascular Medicine, Fukuoka Sanno Hospital, Fukuoka, Japan
- #Department of Cardiovascular Medicine, Omihachiman Community Medical Center, Omihachiman, Japan
- ∗∗Department of Cardiology, Saiseikai Fukuoka General Hospital, Fukuoka, Japan
- ††Department of Cardiology, Sendai Kousei Hospital, Sendai, Japan
- ‡‡Division of Cardiology, Saiseikai Yokohama-City Eastern Hospital, Yokohama, Japan
- §§Department of Cardiology, Shin-Koga Hospital, Kurume, Japan
- ‖‖Department of Cardiovascular Medicine, Shinshu University Graduate School of Medicine, Matsumoto, Japan
- ¶¶Cardiovascular Center, Tokeidai Memorial Hospital, Sapporo, Japan
- ##Wound Treatment Center, Shin-Suma General Hospital, Kobe, Japan
- ∗∗∗Division of Trans-Catheter Therapeutics, Kanazawa Medical University Hospital, Kahoku, Japan
- †††Department of Cardiology, National Hospital Organization Iwakuni Clinical Center, Iwakuni, Japan
- ‡‡‡Department of Vascular Surgery, Matsuyama Red Cross Hospital, Matsuyama, Japan
- §§§Department of Cardiology, Ogaki Municipal Hospital, Ogaki, Japan
- ‖‖‖Department of Internal Medicine, Kobe University Graduate School of Medicine, Kobe, Japan
- ¶¶¶Department of Surgery, Kasugai Municipal Hospital, Kasugai, Japan
- ###Department of Plastic and Reconstructive Surgery, Kyorin University School of Medicine, Tokyo, Japan
- ∗∗∗∗Department of Metabolic Medicine, Osaka University Graduate School of Medicine, Suita, Japan
- ††††Office of Biostatistics and Data Management, National Cerebral and Cardiovascular Center, Suita, Japan
- ‡‡‡‡Nishinomiya Municipal Central Hospital, Nishinomiya, Japan
- ↵∗Reprint requests and correspondence:
Dr. Masato Nakamura, Division of Cardiovascular Medicine, Toho University Ohashi Medical Center, 2-17-6 Ohashi, Meguro-ku, Tokyo 153-8515, Japan.
Objectives This study sought to investigate the 3-year follow-up results of OLIVE registry patients.
Background Although favorable 12-month clinical outcomes after endovascular therapy (EVT) in OLIVE registry patients with critical limb ischemia (CLI) from infrainguinal disease have been reported, long-term results after EVT remain unknown.
Methods This was a prospective multicenter registry study that consecutively enrolled patients who received infrainguinal EVT for CLI. The primary outcome was 3-year amputation-free survival (AFS), whereas secondary outcome measures were 3-year freedom from major adverse limb events (MALE), wound-free survival, and wound recurrence rate. Prognostic predictors for each outcome were also elucidated by Cox proportional hazard regression analysis or the log-rank test.
Results The completion rate of 3-year follow-up was 95%. Three-year AFS, freedom from MALE, and wound-free survival rates were 55.2%, 84.0%, and 49.6%, respectively. Wound recurrence out to 3 years was 43.9%. After multivariable analysis, age (hazard ratio [HR]: 1.43, p = 0.001), body mass index ≤18.5 (HR: 2.17, p = 0.001), dialysis (HR: 2.91, p < 0.001), and Rutherford 6 (HR: 1.64, p = 0.047) were identified as predictors of 3-year major amputation or death. Statin use (HR: 0.28, p = 0.02), Rutherford 6 (HR: 2.40, p = 0.02), straight-line flow to the foot (HR: 0.27, p = 0.001), and heart failure (HR: 1.96, p = 0.04) were identified as 3-year MALE predictors. Finally, CLI due to isolated below-the-knee lesion was a wound recurrence predictor (HR: 4.28, p ≤ 0.001). Three-year survival, freedom from major amputation, and reintervention rates were 63.0%, 87.9%, and 43.2%.
Conclusions In CLI patients with infrainguinal lesions, 3-year clinical results of EVT were reasonable despite high reintervention and moderate ulcer recurrence rate. (A Prospective, Multi-Center, Three-Year Follow-Up Study on Endovascular Treatment for Infra-Inguinal Vessel in Patients With Critical Limb Ischemia [OLIVE 3-Year Follow-Up Study]; UMIN000014759)
Currently, endovascular treatment (EVT) has become commonplace as a treatment strategy for critical limb ischemia (CLI) (1,2) because it achieves comparable clinical outcomes to bypass surgery (BSX), thanks to development of recent catheter devices and because operative risk is high in patients with severe CLI who also present with concomitant morbidities such as cardiovascular disease and heart failure (3–7). However, prospective multicenter studies especially focused on long-term outcomes are limited in this particular field.
The BASIL (Bypass versus Angioplasty in Severe Ischemia of the Leg) trial is the only prospective multicenter study that evaluated long-term results comparing EVT with BSX for CLI (8,9). However, because the study population did not include dialysis patients, and neither long balloons nor nitinol stents were used, it did not reflect true clinical practice; additionally, EVT device evaluation was not conducted (8,9). We subsequently demonstrated favorable 12-month results of EVT for CLI in the OLIVE (A Prospective, Multi-Center, Three-Year Follow-Up Study on Endovascular Treatment for Infra-Inguinal Vessel in Patients With Critical Limb Ischemia) trial (the first prospective multicenter study by a wound core laboratory) in a real-world setting (6). High intervention and poor survival rates notwithstanding, reasonable 12-month results were obtained for freedom from major amputation and wound healing. Herein, we present the 3-year follow-up data for the OLIVE registry.
This was a prospective multicenter registry study that consecutively enrolled patients who underwent infrainguinal EVT for CLI. The 12-month results were reported previously (6), and medium-term results out to 3 years were evaluated in this study. Informed consent was obtained from each subject before enrollment, and the study protocol was approved by the ethics committee at each participating institution. This study was registered with UMIN-CTR (UMIN000014759; A Prospective, Multi-Center, Three-Year Follow-Up Study on Endovascular Treatment for Infra-inguinal Vessel in Patients With Critical Limb Ischemia).
A total of 314 subjects were enrolled between December 2009 and July 2011 and observed between December 2010 and July 2012. The 3-year follow-up was completed in July 2014. Inclusion and exclusion criteria have been described previously (6). In brief, CLI patients without a history of major amputation whose lesions were limited to the infrainguinal artery were included. CLI patients with dementia, and those whose CLI was due to acute arterial occlusion, nonarterial sclerotic disease, or inflammatory disease were excluded.
Details of the endovascular procedure have also been described previously (6). Briefly, provisional stenting using nitinol stents was conducted on femoral artery lesions, whereas balloon angioplasty alone was employed on infrapopliteal lesions. During this study period, drug-eluting stents and balloons for femoropopliteal lesions and atherectomy devices for femorotibial lesions were not available for peripheral intervention.
Wound care and evaluation
Ischemic wounds were assessed and managed by a plastic surgeon or dermatologist. Wound management was carried out regularly. Adjunctive therapies, including maggot and hyperbaric therapy, were not performed, to ensure standardized wound care. Wounds were photographed and evaluated independently out until 12 months as previously described (6), after which they were followed and assessed according to the protocol of each institution’s plastic surgery department.
Follow-up was mandatory at 1, 3, 6, and 12 months. Thereafter, the follow-up protocol was determined by each institution but was generally conducted every 3 months. Reintervention was only indicated for recurrence of rest pain or wounds or for delayed wound healing. Patient survival, major amputation, reintervention, and ulcer recurrence were confirmed by telephone call only if patients were unable to visit the hospital.
Primary outcome was 3-year amputation-free survival (AFS). The following were evaluated: 1) difference in predictors for AFS in 1 and 3 years; 2) timing of influence of each factor; and 3) changes in the hazard rate of major amputation or death. Freedom from major adverse limb events (MALE) and ulcer recurrence rates were evaluated as secondary outcome measures, and changes in the hazard rate for MALE were analyzed. Predicting factors for 3-year MALE were evaluated. Regarding wound assessment, wound-free survival at 1, 2, and 3 years, and post-recovery recurrence rates were evaluated. Additionally, recurrence-predicting factors were assessed. Survival, freedom from major amputation, and freedom from reintervention were also analyzed.
Definition of outcome parameters
AFS was defined as the absence of death or major amputation. Major amputation was defined as above-ankle amputation. Either major amputation or major reintervention was classified as a MALE. Reinterventions consisted of both major and minor reinterventions. Major reinterventions included bypass grafting, jump grafting, and interposition graft revision, as well as thrombectomy and thrombolysis for stents with loss of patency. Minor reintervention was defined as percutaneous transluminal angioplasty and/or stenting (10). Ulcer recurrence after more than 1 year included recurrences at the same site or at different sites.
Continuous variables were expressed as the mean ± SD, whereas categorical data were expressed as numbers with percentages. Either the proportional hazard (PH) regression analysis or log-rank test was performed to examine variables for associations with major amputation or death and time to MALE. Logistic regression analysis or the chi-square test was employed to examine variables for associations with ulcer recurrence. The PH assumption was investigated graphically using its covariate-specific plot and also with a test based on Schoenfeld’s residuals. Each variable showing significant association (p < 0.25) and additional nonsignificant variables with potential clinical importance were then introduced into the stepwise multivariable PH regression model to identify predictors of each time-to-event outcome. We set p = 0.25 for entry into the model and p = 0.05 to remain in the model. We also investigated time-dependent factors for major amputation or death and MALE that could change during the observation period, and incorporated them into the PH model. The hazard ratio (HR) and its 95% confidence interval (CI) were calculated for each possible risk factor. Furthermore, Kaplan–Meier curves were drawn for AFS and time to freedom from MALE. The rate of wound recurrence out to 3 years after wound healing was analyzed by univariable and multivariable stepwise logistic regression analysis to identify risk factors on recurrence outcome. In the analysis of logistic or Cox regression for AFS, MALE, and wound healing, the between-center variance (in the final model) could be ignored enough to be small (p values were 0.22, 0.52, and 0.98 in AFS, MALE, and wound healing, respectively). The odds ratio and its 95% CI were calculated for each possible risk factor. All statistical analyses were performed with SAS version 9.3 for Windows (SAS Institute, Cary, North Carolina).
A total of 314 patients were enrolled, with 312 evaluable patients remaining. Ninety-five percent of patients were observed for the entire 3-year period.
Patient/lower limb characteristics
Mean age was 73 ± 10 years, and 65% of the patients were male (Table 1). Diabetes mellitus and hemodialysis were observed in 71% and 52% of patients, respectively. With respect to limb condition, tissue loss, and wound infection were 88% (Rutherford 5: 73%, Rutherford 6: 15%), and 15%, respectively.
Underlying disease and initial success rates
Isolated below-the-knee (BTK) lesions were culprit lesions in 42% of the patients, and a combination of femoropopliteal artery and BTK lesions was observed in 41% of the patients (Tables 2 and 3). Initial success, defined as straight-line flow to the foot, was achieved in 93%, and the procedure complication rate was 4.0% (12 of 304).
AFS and factors predicting 1-year and 3-year major amputation or death
Figure 1A shows 3-year AFS. Three-year AFS rate was 55.2%. Factors associated with 3-year major amputation or death were age (HR: 1.43, 95% CI: 1.15 to 1.78, p = 0.001), body mass index (BMI) ≤18.5 (HR: 2.17, 95% CI: 1.36 to 3.45, p = 0.001), chronic dialysis (HR: 2.91, 95% CI: 1.88 to 4.51, p < 0.001), and Rutherford classification 6 (HR: 1.64, 95% CI: 1.01 to 2.65, p = 0.047), as shown in Table 4. Previously reported factors associated with 1-year major amputation or death were BMI ≤18.5 (HR: 2.22, 95% CI: 1.23 to 4.01, p = 0.0078), history of heart failure (HR: 1.73, 95% CI: 1.02 to 2.91, p = 0.041), and wound infection (HR: 1.89, 95% CI: 1.07 to 3.32, p = 0.0283). BMI was the only factor common to both. Heart failure and wound infection would affect only early outcomes, and analysis was conducted by incorporating them as covariates that affect AFS within 1 year, but not thereafter. Wound infection was found not to be significant. When this was examined at 6 months instead of 1 year, all factors became significant. In other words, heart failure and wound infection affected AFS only within the first 6 months. Figure 1B shows the change in hazard rate of major amputation or death followed out to 3 years, demonstrating that the highest hazard rate was observed at 0 to 6 months.
Freedom from MALE and predicting factors for 3-year MALE
Three-year freedom from MALE was 84.0% (Figure 2A). Table 5 shows factors associated with 3-year MALE: statin use (HR: 0.28, 95% CI: 0.10 to 0.78, p = 0.02), heart failure (HR: 1.96, 95% CI: 1.02 to 3.77, p = 0.04), Rutherford classification 6 (HR: 2.40, 95% CI: 1.12 to 5.13, p = 0.02), and straight-line flow to the foot (HR: 0.27, 95% CI: 0.11 to 0.66, p = 0.001). Figure 2B shows change in hazard rate of MALE followed out to 3 years, demonstrating that the highest hazard rate was observed at 0 to 6 months.
Wound outcome and recurrence predicting factors
Rate of wound recurrence out to 3 years was 43.9%. Multivariable analysis showed CLI due to isolated BTK lesions as a predicting factor for wound recurrence (Table 6). Wound-free survival rate is shown in Figure 4⇓. At 1, 2, and 3 years, wound-free survival rates were, respectively, 58.9%, 54.8%, and 49.6% in the overall population; 55.9%, 50.6%, and 40.7% in patients with tissue loss; and 80.3%, 83.3%, and 69.4% in patients with rest pain.
Three-year survival, freedom from major amputation and reintervention
Survival, major amputation-free rates, and reintervention-free rates are shown in Figure 3. At 3 years, overall survival rate was 63.0%, the major amputation-free rate was 87.9%, and the freedom from reintervention rate was 43.2%.
In our OLIVE 12-month follow-up paper (6), we reported favorable EVT outcomes for CLI in real world patients, 50% of whom were on hemodialysis. The 3-year outcome study further provided information regarding long-term durability of EVT for CLI. In addition, we found that the prognostic factors for AFS seemed to change over time.
Changes over time in factors related to AFS
Recent prospective trials have not systematically studied the long-term durability of EVT. This study demonstrated that AFS continued to decrease over time after year 1 and reached 55.2% after year 3. This is comparable with 3-year outcomes of previous trials, although the present study included sicker patients such as those on hemodialysis (Table 7). It is noteworthy that factors predicting major amputation or death changed over time. BMI was a predictor at both 1 year and 3 years, whereas heart failure and wound infection only influenced AFS within 6 months after EVT. That is, the short-term and long-term predictors of AFS were different. Factors associated with ulcer healing were short-term predictors for AFS, and presumably, were linked to survival. After ulcer healing, these factors were no longer significant, and other factors related to long-term prognosis tended to emerge. Six months also coincides with the period when leg ulcers were healed in the OLIVE study, suggesting that this might be an appropriate duration when evaluating the usefulness of new methods of limb salvage. The previous notwithstanding, long-term observation is necessary to assess patient survival.
Freedom from MALE
In this study, the MALE-free rate was a favorable 84.0% after 3 years, whereas AFS declined from 73.6% at 1 year to 55.2% after 3 years. Thus, the difference between AFS and the MALE-free rates was 14.7% after 1 year and nearly doubled to 28.8% by 3 years. This result is similar to previous reports that patient survival is worse than limb prognosis. A patent direct vascular connection and Rutherford 6 classification were significant factors affecting 3-year freedom from MALE in this study. Accordingly, short-term limb salvage appears to positively influence long-term prognosis, which may support the difference between vascular patency and limb prognosis. In other words, post-EVT wound healing leads to fewer amputations, and these results are maintained at 3 years. Statin therapy was a predictor of good long-term limb prognosis. This finding agreed with the results of recent studies demonstrating clinical benefits of statins for CLI patients (17,18).
CLI due to isolated BTK lesions was identified as the only predicting factor for ulcer recurrence. Restenosis after EVT for BTK lesions was high, and the risk of wound recurrence is commonly present, suggesting the significance of foot care and meticulous outpatient follow-up.
There is a paucity of prospective multicenter studies especially focused on long-term outcomes of EVT for CLI. The follow-up data for the OLIVE registry demonstrated that the 3-year clinical results of EVT were reasonable in CLI patients with infrainguinal lesions despite high reintervention and moderate ulcer recurrence rates; meanwhile, prognostic factors for clinical endpoints seemed to change over time. Our finding that prognostic factors differed over time could be clinically useful in identifying the treatment goals that change over time for CLI patients with complex morbidity. Moreover, CLI due to isolated BTK lesions was the only identified predicting factor for ulcer recurrence, suggesting the significance of foot care and meticulous outpatient follow-up.
1. Study design: This was a single-arm study of EVT, rather than a randomized controlled trial of EVT versus BSX. However, this study evaluated 3-year outcomes in patients who fit the common clinical profile (approximately 50% on maintenance dialysis) in a prospective and multicenter manner.
2. Study population: This study included Japanese patients only; therefore, results may not necessarily be extrapolated to apply to other populations.
3. Treatment strategy: The results of EVT were not evaluated at a central lab. However, new devices, including drug-eluting balloons, stents, and atherectomy devices, were not approved in Japan during the study period, so there was little heterogeneity in treatment strategy.
4. Clinical assessment: Although ulcer recurrence from year 1 onward was not evaluated by a central lab, the standard wound treatment protocol was used for all patients enrolled in this study, and the outcomes of ulcer, patient, and vascular lesions were confirmed in 95% of the subjects.
The 3-year clinical results of EVT were reasonable in CLI patients with infrainguinal lesions despite a high reintervention rate and moderate ulcer recurrence rate. Prognostic factors for clinical endpoints changed over time.
WHAT IS KNOWN? Although EVT has become commonplace as a treatment strategy for patients with CLI, prospective studies, especially those focused on long-term outcomes and dialysis patients, are limited.
WHAT IS NEW? This OLIVE registry demonstrated that the 3-year clinical results of EVT were reasonable, despite high reintervention and moderate ulcer recurrence rates. Prognostic factors for clinical endpoints seemed to change over time.
WHAT IS NEXT? It is essential that treatment goals are identified and revised as necessary, in accordance with the changing risk profile of CLI patients. EVT efficacy in these patients may be further improved by the advent of new techniques and technologies. Finally, an additional challenge may be establishing a treatment strategy focused on improving CLI patient life prognosis.
The authors would like to thank the cardiac catheterization laboratory medical staff and clinical research coordinators in the participating centers. They also wish to express their special gratitude to Mr. Toru Nakaso.
Funded by Associations for Establishment of Evidence in Interventions. Dr. Y. Yokoi is a consultant for Boston Scientific and Terumo. Dr. Inoue is a consultant for Kaneka Medix Corporation, Medicon Inc., and Japan Lifeline Company. Dr. Nanto has received grant support from Boston Scientific Japan K.K. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- amputation-free survival
- body mass index
- bypass surgery
- confidence interval
- critical limb ischemia
- endovascular treatment
- hazard ratio
- major adverse limb event(s)
- proportional hazard
- Received December 11, 2014.
- Revision received June 18, 2015.
- Accepted July 2, 2015.
- American College of Cardiology Foundation
- Dorros G.,
- Jaff M.R.,
- Dorros A.M.,
- et al.
- Iida O.,
- Nakamura M.,
- Yamauchi Y.,
- et al.,
- OLIVE Investigators
- Liistro F.,
- Porto I.,
- Angioli P.,
- et al.
- Zeller T.,
- Baumgartner I.,
- Scheinert D.,
- et al.
- Westin G.G.,
- Armstrong E.J.,
- Bang H.,
- et al.