Global vascular guidelines on the management of chronic limb-threatening ischemia

GVG structure

The three major global vascular surgical societies, the European Society for Vascular Surgery (ESVS), the Society for Vascular Surgery (SVS), and the World Federation of Vascular Societies (WFVS), joined efforts to launch the GVG initiative. In this process, the ESVS represents national vascular societies from Europe and the SVS represents national, regional, and local vascular societies in North America. The WFVS represents a large number of non-European, non-North American vascular surgical societies from across the world. These include the Australian and New Zealand Society for Vascular Surgery, the Japanese Society for Vascular Surgery, the Vascular Society of India, the Vascular Society of Southern Africa, the Asian Society for Vascular Surgery, and the Latin American Society of Vascular Surgery and Angiology (this list is not exhaustive). As the primary sponsors, the ESVS, SVS, and WFVS developed the organizational structure, policies on conflict of interest, and committed financial support for the GVG program. All financial support for the GVG was derived directly from the sponsoring societies and without the direct involvement of industry or other external stakeholders. Representatives from the three leading societies were asked to serve as Co-Editors as well as members of the Steering Committee to oversee all aspects of the project and its subsequent communications. Oversight from the societies was limited to budgetary and administrative aspects, including their respective document review policies before public dissemination of the final guideline. The Steering Committee recruited a large and diversified writing group; developed the scope and section briefs for the guideline; identified priority questions for commissioned evidence reviews; and participated in all stages of writing, consensus debate, and editing of the manuscript.

Conflict of interest policy

A primary consideration on inception of the GVG was to create a robust yet practical approach to conflict of interest to enable an unbiased effort at guideline development by experts in the field. A central element to this, in concert with the exclusion of direct commercial funding sources, was full disclosure and specific limits on relevant financial relationships for members of the writing group, Steering Committee, and Co-Editors. A full description of the GVG Conflict of Interest policy is provided at the beginning of this supplement. Financial disclosures for all contributing authors were collected and updated by the Steering Committee. They are detailed in the table of Contributing Authors listed at the beginning of the guideline.

Leadership and writing group

The Co-Editors and Steering Committee were selected by the three major sponsoring societies and were tasked with the recruitment of a multidisciplinary, international writing group of recognized experts. In total, the final writing group comprised 58 individuals from 24 countries across 6 continents. This group represents specialists in vascular surgery, vascular medicine, interventional cardiology and radiology, angiology, epidemiology, podiatry, and orthopedics as well as a methodologist with expertise in guideline development. Authors were assigned to individual sections of the guideline, and all authors reviewed the complete final document before societal review.

Methodology

The Steering Committee drafted a Table of Contents that was divided into distinct sections. Briefs were created to outline the scope and content of each section. Potential authors were then solicited and vetted, and two authors were chosen to co-lead the writing effort for each section. The co-lead authors communicated directly with the Steering Committee on their progress and on iterative cycles of revision as needed. All of the authors of each section reviewed and approved their final versions before compilation of the full document.

The Steering Committee examined the state of recent evidence reviews in the field, including those commissioned by the participating societies, and determined the need for additional evidence reviews and updating. These were commissioned to an external group (Mayo Clinic Evidence-Based Practice Research Program) who performed four systematic reviews that summarized evidence from randomized and nonrandomized studies.^4–7 These systematic reviews underwent peer review and were published in the Journal of Vascular Surgery, one of which is published as an accompaniment to the guideline document in this supplement.⁷

Consensus development during the process occurred through confidential electronic communications, teleconferences, and multiple in-person meetings of the Steering Committee and members of the writing group. The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach was used to determine the quality of evidence and strength of recommendations.⁸ A strong (Grade 1) recommendation implies that the guideline developers are confident as to the balance of benefits and harm and that this recommendation should apply to the majority of patients. A conditional recommendation (Grade 2) implies less certainty and indicates that a different course of action is reasonable. The guideline developers used an imperative verb to denote strong recommendations and used the term “consider” to denote a conditional recommendation. The level of evidence for each recommendation is considered high quality (A), moderate quality (B), or low quality (C). The guideline also includes good practice recommendations. These ungraded good practice recommendations are supported by a wealth of indirect evidence but no direct evidence, and the benefit of pursuing the recommended actions is considered to outweigh any plausible harm. The intention of these good practice recommendations was to draw attention to and remind providers of known and noncontroversial surgical principles or principles about general medical care. For example, there are good practice statements about performing a comprehensive history and physical examination in patients with CLTI.⁹

The final grading of all guideline recommendations was determined by the guideline developers and the methodologist. After approval by the full writing group, the sections were compiled into one document and reviewed concurrently by the document oversight bodies of each of the three sponsoring societies. An open comment period was subsequently enabled on a secure website (http://vsweb.org/GlobalVascularGuidelines) to provide an opportunity for external stakeholders to review the document. The Co-Editors collated all reviews and made final revisions to the document, which was then approved by the sponsoring societies before publication and dissemination.

Target population

The target population of patients includes adults with CLTI, defined as a patient with objectively documented PAD and any of the following clinical symptoms or signs:

• Ischemic rest pain with confirmatory hemodynamic studies
• Diabetic foot ulcer (DFU) or any lower limb ulceration present for at least 2 weeks
• Gangrene involving any portion of the lower limb or foot

Specifically excluded are patients with pure venous ulcers, pure traumatic wounds, acute limb ischemia (symptoms present for 2 weeks or less), embolic disease, and nonatherosclerotic chronic vascular conditions of the lower extremity (eg, vasculitis, Buerger disease, radiation arteritis).

Target audience

The primary target audience for this guideline includes all clinicians who are directly involved in the management of patients with CLTI, to include surgeons (vascular, general, plastic, and orthopedic), interventionalists (radiologists, cardiologists), podiatrists, wound care providers, rehabilitation medicine specialists, orthotists and physical therapists, and trainees in these disciplines.

Secondary audiences include referring providers, such as primary care physicians, medical specialists, nurses, and other allied health providers, who may care for the at-risk population and who are critical for awareness and timely specialist referral of patients with suspected CLTI. Other key targets for this guideline are third parties with influence over the current and future treatment of CLTI, including government agencies, payers (funders), industry stakeholders, investigators, and research organizations.

CLTI: A new paradigm for treatment and research

This clinical practice guideline (CPG) intentionally seeks to create a new conceptual framework for the treatment of CLTI. It encompasses nomenclature, disease staging, and a platform for evidence-based revascularization (EBR) that will allow future evolution and quality improvement in the field. A brief introduction to the key elements introduced in this document is provided here.

Dissemination, translation to practice, and future revisions of the guideline

Translation of expert guidelines into clinical practice is known to be a major obstacle to evidence-based medicine. Reasons are multifactorial and include limited provider and patient engagement, lack of consensus, economic conflicts, and resource constraints. The international scope of the GVG mandated an attempt to survey differences in practice patterns, resources, and potential hurdles to implementation around the globe (Section 13). Dissemination of the guideline by the sponsoring societies is planned to include an array of print media, web and social media, mobile apps, and communications at multiple national and regional meetings to facilitate discussion. The incorporation of suggested staging systems and end points into national and multinational registries will greatly facilitate use and future refinement of this effort. It is anticipated that the GVG will be translated into the other major world languages.

To remain current and evidence based, practice guidelines must be periodically reviewed and updated. Ongoing RCTs and prospective cohort studies will provide critical new evidence in the management of CLTI during the next several years. The sponsoring societies of the GVG recognize the importance of stewardship of this practice guideline, both as new key evidence arises and as a planned interval exercise.

Previous leg ischemia definition and classification systems

Lower extremity threatened limb classification system

The definitions summarized in Table 1.1 were developed primarily to describe patients suffering from pure ischemia due to atherosclerosis. This was when the predominant risk factor was tobacco smoking and before the global epidemic of diabetes mellitus (DM). As such, these definitions were ischemia-dominant models of limb threat. However, because patients with DM now make up the majority of patients with CLTI, absolute perfusion now needs to be considered in the context of neuropathy, wound characteristics, and infection. To address this unmet need, the SVS Lower Extremity Guidelines Committee created the SVS Lower Extremity Threatened Limb Classification System. This system stratifies amputation risk according to wound extent, degree of ischemia, and presence and severity of foot infection (Wound, Ischemia, and foot Infection [WIfI]).¹⁰ Although it may require some adjustments, WIfI appears to correlate strongly with important clinical outcomes. This includes those set forth in the SVS objective performance goals (OPGs) that focus on limb amputation, 1-year amputation-free survival (AFS), and wound healing time (Table 1.2).^{10,68–72,162–167}

The WIfI classification system is currently being evaluated in multicenter trials including the U.S. National Institutes of Health-funded BEST-CLI trial¹³ and the UK National Institute for Health Research Health Technology Assessment-funded BASIL-2 and BASIL-3 trials.^14,15 WIfI is also being incorporated into the U.S. SVS Vascular Quality Initiative registry of lower extremity interventions.

Hemodynamic criteria

Although previous guidelines have suggested a range of AP and toe pressure (TP) thresholds for defining limb-threatening ischemia, such thresholds must be used with great caution and considered in the clinical context because of multiple confounding factors and the lack of a clear and reliable relationship to outcomes. Patients with limb-threatening ischemia should be defined primarily in terms of their clinical presentation, supplemented by physiologic studies that demonstrate a degree of ischemia sufficient to cause pain, to impair wound healing, and to increase amputation risk.

In addition to patients who meet the proposed new definition of CLTI, there are a significant number of patients whose PAD is so severe that they are likely to be at increased risk for development of CLTI in the foreseeable future.¹⁶⁸ Although data are lacking, it is logical to suggest that such individuals should be monitored closely for clinical disease progression.

Risk factors for PAD.

Modifiable risk factors for PAD have been comprehensively studied in HICs and include smoking, DM, hypertension, hypercholesterolemia, and air pollution. A global study suggested that although these risk factors may be equally applicable to LMICs, for most, the strength of the association was greater in HICs. This may be because HIC studies often include a larger number of older patients and because the exposure time tends to be shorter in LMICs.¹

Smoking is unarguably a significant risk factor in the development and progression of PAD. Nevertheless, whereas smoking rates are falling in most HICs, this is not the case in LMICs (Fig 2.2). DM is also strongly associated with the development of PAD, and risk increases with the duration of DM in affected individuals. Patients with DM are widely recognized to be at markedly higher risk of amputation.^191,192 The rapidly increasing worldwide prevalence of type 2 DM is concerning and likely to have a significant impact on the future incidence and prevalence of PAD and CLTI as well as their morbid end points.

Open in a separate window

Fig 2.2.

Odds ratios (ORs) for peripheral artery disease (PAD) in high-income countries (HICs) and low- and middle-income countries (LMICs). BMI, Body mass index; CRP, C-reactive protein; CVD, cardiovascular disease; HDL, high-density lipoprotein. (Reprinted from Criqui MH, Aboyans V. Epidemiology of peripheral artery disease. Circ Res 2015;116:1509–26.)

The link between obesity and PAD is inconsistent. Many studies have suggested the existence of an “obesity paradox,” with lower rates of PAD being observed in patients with a higher body mass index (BMI).¹⁸⁶ By contrast, other studies that have adjusted for smoking, which is associated with a generally lower BMI,¹⁹³ reported a positive correlation between BMI and PAD. Hypertension is associated with the development of PAD and is another common risk factor in the adult population.

The association between dyslipidemia and the development and progression of atherosclerosis has been extensively studied. Whereas elevated levels of total cholesterol and low-density lipoprotein cholesterol (LDL-C) are widely accepted as risk factors for PAD, reduced high-density lipoprotein cholesterol levels also appear to be associated with increased mortality in PAD patients.¹⁹⁴ A ratio of the two may also be a useful predictor of PAD.¹⁹⁵ Whereas hypertriglyceridemia appears to be atherogenic,¹⁹⁶ its role in the development and progression of PAD remains incompletely defined.

Chronic kidney disease (CKD), particularly ESRD, is a strong risk factor for PAD and limb loss, especially in association with DM. Affected patients frequently have heavily calcified arteries and a distal pattern of arterial disease.¹⁸⁶

The association between alcohol consumption and PAD is inconsistent, making it difficult to draw any firm conclusions.¹⁹⁷ However, heavy alcohol consumption is often associated with other risk factors for PAD, such as smoking, and as with DM, the presence of alcoholic neuropathy increases the risk of tissue loss for any given perfusion deficit.

Recent data suggest that air pollution from sources such as motor vehicles, power plants, wood burning, and some industrial processes may be associated with increased cardiovascular morbidity and mortality.¹⁹⁸ Likewise, chronic inflammation, characterized by elevated levels of C-reactive protein and other biomarkers, has been shown to be associated with PAD.¹⁸⁶ Homocysteine levels are higher in several case-control PAD cohort studies, although the benefits of folate supplementation appear to be negligible.^186,199

The significance of family history and genetic makeup is uncertain.^200,201 Studies have yielded varying results, with some identifying a small number of candidate genes or even single-nucleotide polymorphisms and others failing to identify any association at all.

Finally, people of lower socioeconomic status and educational attainment tend to have a higher prevalence of IC and probably also of CLTI, although the association is not always strong and can often be explained in part by their increased exposure to other risk factors, such as smoking.^180,183,202 However, there is increasing evidence that chronic mental and psychosocial stress may have direct effects on cardiovascular health.²⁰³

Incidence and prevalence of CLTI.

As noted before, high-quality data on the epidemiology of CLTI are lacking, especially from LMICs, with many estimates being extrapolated from the incidence and prevalence of IC, amputation, and DM. Unfortunately, such estimates can be highly misleading for a number of reasons. First, IC does not progress to CLTI in a predictable manner. Second, CLTI probably represents <10% of all PAD patients, and those undergoing amputation for CLTI are at very high risk of premature death (and so more likely to be absent from population-based studies). Third, the clinical and hemodynamic data required to reliably diagnose CLTI are difficult to obtain in large populations. This is particularly true in patients with DM, who often have incompressible vessels. Thus, although it is estimated that approximately half of all patients with a DFU in western Europe and North America also have significant PAD, the disease may often appear relatively mild (not fulfilling the criteria for CLTI) on hemodynamic assessment.²⁰⁴

For many years, the annual incidence of what has typically been termed CLI was estimated at 500 to 1000 new cases per million individuals in Western countries.²⁰⁵ Unfortunately, there are no reliable contemporary epidemiologic data that take into account recent changes in lifestyle (such as reduced smoking rates), identification and medical management of cardiovascular risk factors, prevalence of obesity and diabetes, and overall increasing life expectancy around the world.

In 2013, a meta-analysis involving 6 studies and close to 83,000 patients showed the overall prevalence of severe chronic limb ischemia (defined by Fontaine stage, AP <70 mm Hg, and ABI <0.60) to be 0.74% (95% confidence interval [CI], 0.26–1.46), with marked heterogeneity between studies (prevalence, 0.11%−1.59%).²⁰⁶

In an analysis of the U.S. MarketScan database (Truven Health Analytics, Ann Arbor, Mich), composed of approximately 12 million Americans aged 40 years and older receiving care from Medicare and Medicaid between 2003 and 2008, the prevalence and annual incidence of CLTI were estimated at 1.33% and 0.35%, respectively. This equates to around 3500 new cases per million individuals per year.¹⁸⁷ The study defined primary CLTI as patients with no prior PAD or subsequent PAD diagnostic code >30 days after a CLTI diagnostic code. Secondary CLTI included patients with prior PAD (or subsequent PAD diagnostic codes within 30 days of a CLTI diagnostic code). The annual incidence rate of primary and secondary CLTI was 0.19% and 0.16%. CLTI patients represented 11.08% (95% CI, 11.03%−11.13%) of total PAD patients annually. As noted before, although one might expect similar rates of CLTI in other developed nations and regions, data from LMICs are lacking. Even within HICs, the epidemiology of CLTI is likely to be complex and evolving.

Amputation and CLTI.

A number of studies have used major lower limb amputation as a surrogate for CLTI on the basis that most (>80%) are due to CLTI. However, it can be difficult to distinguish reliably between minor (below the ankle) and major (above the ankle) amputations in some administrative data. Furthermore, the number of amputations that are performed for trauma, tumor, or infection, including patients with DM and neuropathy (but without PAD), is likely to vary considerably from country to country, particularly in comparing HICs and LMICs.

In the United States in 2015, an estimated 504,000 individuals (of a total estimated population of 295.5 million) were living with a major amputation due to PAD, a number that was projected to more than double by 2050.²⁰⁷ In Minnesota, a state with low overall rates of cardiovascular disease (CVD), one study showed that between 2005 and 2008, the age-adjusted annual incidence of ischemic lower limb amputation (amputations not due to trauma or cancer) remained unchanged at 20 per 100,000.²⁰⁸

A systematic review found that the rate of major amputation varied considerably (3.6 to 68.4 per 100,000 per year) across the world, probably because of differences in ethnicity, social deprivation, and, in particular, the prevalence of DM.²⁰⁹ In some countries, including England, the incidence of amputations unrelated to DM appears to be decreasing.²¹⁰ However, in most parts of the world, the incidence of DM-related limb amputations is increasing.²¹¹

Natural history of untreated CLTI.

A meta-analysis (13 studies and 1527 patients) of the natural history of untreated CLTI found that during a median follow-up of 12 months, both the mortality rate and the per-patient amputation rate were 22%, although there was marked heterogeneity between studies.⁵ With regard to disease progression, one study estimated that only 5% to 10% of patients with either asymptomatic PAD or IC went on the develop CLTI during a 5-year period.²¹² However, another meta-analysis suggested that this progression rate may be significantly higher at 21% (range, 12%−29%) during 5 years.²¹³ Approximately 50% of patients presenting with CLTI have no prior history of PAD.^214,215

Patients with CLTI present with a wide spectrum of clinical, hemodynamic, and anatomic disease. Outcomes depend on the availability and quality of primary and secondary care and may be further influenced by factors such as social stigmatization and cultural and religious beliefs. Those living in regions with poor access to health care often present late with advanced disease and unsalvageable limbs. Indeed, it has been estimated that approximately half of all patients with CLTI do not undergo revascularization.²¹⁶ Even in HICs with advanced health care systems, such as Germany and the United States, many patients with suspected CLTI do not receive angiography or any attempt at revascularization.²¹⁷ This may be because patients are too sick or frail, are thought to have no revascularization option, or present too late. Unfortunately, whereas reasonable data are available on amputation rates, data on processes of care that can help explain the shortfall and differences in revascularization and amputation are lacking.

The recently published VASCUNET report showed large (almost sixfold) differences but an overall decline in major amputation rates in 12 European and Australasian countries between 2010 and 2014.²¹⁸ DM prevalence, age distribution, and mortality rates were also found to vary between countries. Despite limitations inherent to the use of registry data, these findings are important and may indicate disparities in access to vascular surgical intervention across the countries studied. Further research is clearly required to improve limb salvage in different demographic and geographic settings.²¹⁸

In patients with known PAD, the risk for development of CLTI appears to be greater in men, in patients who have had a stroke or are in heart failure, and in patients with DM.¹⁸⁷ Patients who present de novo with CLTI (no prior diagnosis of PAD) seem more likely to be older and male and to have pre-existing CVD (including hypertension, myocardial infarction, heart failure, or stroke) and renal failure.¹⁸⁷ Not surprisingly, because of the associated high prevalence of neuropathy, DM had the strongest association with a new presentation of CLTI (odds ratio [OR], 7.45; 95% CI, 7.19–7.72). The medical management of patients who have or are at risk of having CLTI is covered elsewhere in the guideline (Section 4). Still, there is growing evidence that aggressive medical management of risk factors can significantly improve the overall prognosis for patients with PAD. This may in part explain the decline in mortality observed in patients with IC and CLTI in The Netherlands between 1998 and 2010.²¹⁹

The risk of amputation is high in CLTI patients, even in those undergoing a successful revascularization.²²⁰ Unsurprisingly, patients who present late and with the greatest degree of tissue loss are at highest risk. In one analysis, the rates of amputation at 4 years were 12.1%, 35.3%, and 67.3% for Rutherford class 4, class 5, and class 6, respectively.²¹⁷

Anatomic patterns of disease.

CLTI is usually the result of multilevel arterial occlusive disease. Involvement of parallel vascular beds, such as the superficial femoral artery (SFA) and profunda femoris artery (PFA), is also common. Below-knee arteries typically become increasingly involved as the overall severity of disease worsens. However, FP and IP disease does not always progress in parallel. The general requirement is that there needs to be two levels of arterial occlusive disease to cause CLTI. However, an increasingly observed exception is diffuse disease involving the IP and pedal arteries in patients with DM or CKD. In patients with CLTI and IP disease, the PT artery tends to be the most diseased, often with relative sparing of the peroneal artery. In patients with DM, there may also be sparing of the DP artery. A number of specific factors appear to drive the distribution of lower limb PAD (Fig 2.3). Thus, women may be more prone to development of FP disease, whereas elderly male patients and those with diabetes are more likely to develop IP disease.²²¹ There is also some evidence that black people and Asians are more likely to develop distal disease.^222,223

Open in a separate window

Fig 2.3.

Association of risk factors with the level of atherosclerotic target lesions. The red overlay on the anatomic cartoon illustrates the association of risk factor with patterns of atherosclerotic disease.²¹⁷ (Reprinted from Diehm N, Shang A, Silvestro A, Do DD, Dick F, Schmidli J, 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.)

CVD and mortality risk.

Despite some evidence of recent improvements in HICs, patients who develop PAD and CLTI remain at high risk of premature death. Thus, in a German study, 4-year mortality was 18.9% in Rutherford class 1 to class 3, 37.7% in class 4, 52.2% in class 5, and 63.5% in class 6.²¹⁷ However, interestingly, up to 40% of the deaths were not cardiovascular, perhaps because better medical therapy and management of risk factors have improved overall survival from CVD.^224,225

In 2014, the Global Burden of Disease (2010) database was used to estimate PAD deaths, disability-adjusted life-years, and years of life lost in 21 regions worldwide between 1990 and 2010. In 1990, the age-specific PAD death rate per 100,000 population ranged from 0.05 among those aged 40 to 44 years to 16.63 among those aged 80 years or older. In 2010, the corresponding estimates were 0.07 and 28.71. Death rates increased consistently with age in 1990 and 2010, and the rates in 2010 were higher than they were in 1990 in all age categories.

The overall relative change in median disability-adjusted life-years was greater for men and women in developing than in developed nations. The overall relative change in the median years of life lost rate in developed countries was larger in women than in men. Researchers concluded that disability and mortality associated with PAD increased during the 20 years of the study and that this increase in burden was greater among women than men. In addition, the burden of PAD is no longer confined to the elderly population and now includes young adults. Finally, the relative increase in PAD burden in developing regions of the world is striking and exceeds the increases in developed nations.²²⁶

Management strategies in CLTI.

A study based in South Carolina identified patients who underwent revascularization for CLTI in 1996 and 2005 and examined the requirement for subsequent amputations and further revascularizations. Although revascularization procedures increased by 33%, the 1-year and 3-year amputation rates did not change significantly between 1996 (34% and 43%) and 2005 (34% and 40%). However, the percentage of patients who required further revascularization in the same calendar year increased from 8% to 19%. Investigators concluded that the shift to endovascular interventions increased the number of secondary procedures required to maintain limb salvage rates. Although the absolute number of amputations appeared to decrease despite the increasing population at risk, they concluded that it could be misleading to suggest a direct relationship to the increase in revascularization rates. Thus, whereas the number of amputations fell by approximately 500, the number of revascularization procedures rose by only 187.²²⁷ As noted before, improved risk factor management and use of best medical therapy are likely to have been important factors. The increased number of revascularization procedures may also be due to the increasing availability of endovascular technology and techniques. Indeed, there is some suggestion that practitioners have become more liberal with the use of all revascularization techniques, including bypass and angioplasty.²²⁸ Data from the United Kingdom suggest that an increasing number of patients are undergoing attempts at revascularization.²²⁸

Undoubtedly, there is an increase in the number and proportion of revascularization procedures performed using an endovascular approach. In the South Carolina study, the endovascular approach was used in 26% of CLTI revascularization procedures performed in 1996 compared with 51% in 2005.²²⁷ It is difficult to establish whether this change in management strategy has resulted in the salvage of more limbs and prevention of premature deaths. Such questions can only be answered by RCTs. There are, however, consistent data to suggest that more modern vascular strategies (including a more widespread adoption of endovascular techniques as first- or second-line therapies) are associated with an increased number of patients requiring repeated revascularization (increasing from 8% to 19% in the South Carolina study).²²⁷ Alternative explanations may be that vascular surgeons are becoming more aggressive at retreating patients or that patients are living longer.

History

Ischemic rest pain usually affects the forefoot, is frequently worse at night, and often requires opiate analgesia for management. If present for >2 weeks and combined with hemodynamic evidence of severely impaired perfusion (eg, absolute AP <50 mm Hg, absolute TP <30 mm Hg), it is diagnostic of CLTI.²³⁰

Ischemic ulceration is frequently located on the toes and forefoot, but other areas may be affected in patients with diabetic neuropathy, altered biomechanics, or foot deformity. Gangrene usually occurs on the forefoot. A range of perfusion deficits may be limb threatening in different scenarios of tissue loss and concomitant infection (Section 1). Thus, all patients presenting with signs or symptoms of suspected CLTI should undergo a complete vascular assessment.

In addition to a carefully documented history of presenting limb complaints, it is important to record details of cardiovascular risk factors, drug history, and previous vascular and endovascular revascularization procedures and amputations.^230,231 Assessment of frailty, functional status, and HRQL is also important.^232,233

Physical examination

All patients with suspected CLTI should undergo a complete physical examination.^234,235 Palpation of lower limb pulses can help determine the likely presence and distribution of arterial disease.^236–240 Although they can be nonspecific, features such as coolness, dry skin, muscle atrophy, hair loss, and dystrophic toenails are frequently observed in patients with PAD. Buerger sign, pallor of the foot on elevation and rubor (so-called sunset foot) on dependency, is usually present in CLTI. The capillary refill time will usually exceed 5 seconds, especially when the patient is lying supine or the leg is elevated.²³⁹ It is important not to examine the patient with suspected CLTI sitting in a chair with the leg hanging down as that may lead to false reassurance regarding the perfusion of the foot.

Many patients with CLTI, especially those with DM, have “glove and stocking”²³⁹ sensory, motor, and autonomic neuropathy that may be asymptomatic or be associated with tingling, numbness, weakness, and burning pain in the feet and ankles. The presence of such neuropathy is a major risk factor for tissue loss and should be carefully sought and evaluated using monofilaments and, if available, a tuning fork (loss of vibration sense is an early feature).^241–244 Neuropathy often leads to abnormal foot biomechanics and deformity, and neuropathic (neuroischemic) ulcers often occur at sites of abnormal pressure (load bearing). In patients with suspected CLTI who have a foot ulcer, a probe-to-bone test should be performed to assess depth and the probability of underlying osteomyelitis.^245,246

Noninvasive hemodynamic tests

Other methods for noninvasive diagnosis of CLTI

Alternative noninvasive testing methods can also be used to assist in the diagnosis of CLTI (Table 3.1). Whereas each method has its own advantages and limitations, depending on local availability and expertise, they can be used to augment APs and TPs and indices. Segmental pressures can provide information on anatomic localization of lower limb vascular disease in patients with CLTI but are used infrequently today, at least in HICs. Several other noninvasive tests, including laser Doppler flowmetry, TcPO₂, skin perfusion pressure, and plethysmography, have been used to evaluate limb perfusion.^16,253 However, these tests can be influenced by a variety of confounding factors and are not used routinely in most vascular laboratories around the world.

Wound and tissue loss classification systems

A number of limb and wound classification systems have been developed to try to improve clinical decision-making and clinical outcomes.^254–256 The WIfI system¹⁰ is based on three key factors: wound, ischemia, and foot infection (Tables 3.2-3.5). WIfI correlates with limb salvage, amputation risk, and wound healing and can identify patients who are likely to benefit from revascularization.^68,69

A limb-staging classification system, such as WIfI, should be used in all patients presenting with suspected CLTI (Tables 3.2-3.5). Limb staging should be repeated after vascular intervention, foot surgery, or treatment of infection and whenever there is suspected clinical deterioration.

Imaging of vascular anatomy

Vascular imaging should be performed in all patients with suspected CLTI (Table 3.6) to determine the presence, extent, and severity of arterial disease and to help inform decisions about revascularization. Although there have been huge advances in imaging techniques in recent years, access to these latest modalities, and so practice, varies considerably between and even within countries.

In patients with CLTI who are candidates for revascularization (Section 6), imaging should allow complete anatomic staging using, for example, GLASS (Section 5). Adequate imaging of the tibial and pedal vessels is of critical importance, particularly in planning intervention in patients with tissue loss. History and physical examination often help guide the optimal imaging approach. For those with tibial disease, particularly in the setting of tissue loss, computed tomography angiography (CTA) and magnetic resonance angiography (MRA) may offer useful information but may fail to completely image the ankle and foot vessels with sufficient resolution for procedural planning. Many vascular specialists believe that digital subtraction angiography (DSA) remains the “gold standard.” CTA offers more precise quantification of arterial calcification compared with MRA and DSA. Selective intra-arterial dual-energy CTA combines the low contrast material dose of conventional angiography with computed tomography; if it is available, it may allow crural artery visualization in patients with renal insufficiency.²⁵⁷ This technology is in evolution and not routinely available.

Antithrombotic therapy.

Antiplatelet agents are strongly recommended for all patients with symptomatic PAD to reduce the risk of major adverse cardiovascular events (MACE).^33,34,291 The Antithrombotic Trialists’ Collaboration performed a meta-analysis of antiplatelet agent trials before 1997.³³ It included 135,000 patients with cerebrovascular disease, coronary disease, or PAD (IC) who were treated with antiplatelet agents and 77,000 control patients. The antiplatelet therapy group had a 22% reduction in MACEs, and 75 to 150 mg of aspirin per day was as effective as higher doses but with a lower risk of bleeding.³³ A more recent meta-analysis studied the specific benefit of aspirin in 16 secondary prevention trials comprising 17,000 patients.³⁴ This study confirmed the benefit of antiplatelet agents with an 18.2% reduction in MACEs in both men and women. The Critical Leg Ischaemia Prevention Study (CLIPS) group compared the benefit of 100 mg of aspirin per day in 185 patients with symptoms of PAD and an ABI <0.85 or a TBI <0.6 with placebo and reported a 64% risk reduction in vascular events compared with a 24% reduction in the placebo group.²⁹¹

However, there is a growing body of literature indicating that alternatives to aspirin, such as ticlopidine, dipyridamole, and clopidogrel, may be more effective.^35,292–294 The Clopidogrel versus Aspirin in Patients at Risk for Ischaemic Events (CAPRIE) trial, although not specifically designed to address CLTI, compared 75 mg of clopidogrel per day with 325 mg of aspirin per day in patients with PAD. Researchers noted an 8.7% decrease in MACEs with clopidogrel compared with aspirin. There was no significant difference in bleeding risks between the two agents.³⁵

Other antiplatelet agents, such as ticagrelor and vorapaxar, have also been shown to reduce MACEs in patients with PAD.^292–294 However, benefit over clopidogrel has not been demonstrated.^36,294–298 The Examining Use of Ticagrelor in Peripheral Artery Disease (EUCLID) trial compared ticagrelor with clopidogrel in 13,885 patients with symptomatic PAD and an ABI ≤0.8.³⁶ Although both drugs had a similar safety profile, ticagrelor was not superior to clopidogrel. The Trial to Assess the Effects of Vorapaxar in Preventing Heart Attack and Stroke in Patients with Atherosclerosis-Thrombolysis in Myocardial Infarction 50 (TRA2P-TIMI 50) examined the effects of the protease-activated receptor 1 antagonist vorapaxar on secondary prevention of ischemia events in patients with stable atherosclerosis, including symptomatic PAD.²⁹⁵ Acute limb ischemia, a prespecified study end point, was reduced by 41% among the PAD cohort.²⁹⁸ However, vorapaxar has been associated with an increase in intracranial hemorrhage in patients who have had a prior stroke or transient ischemic attack.²⁹⁶ In a meta-analysis, vorapaxar added to aspirin yielded little improvement in the reduction of MACEs in patients with atherosclerosis and was associated with a slightly higher incidence of intracranial hemorrhage.²⁹⁴ Finally, a meta-analysis that reviewed the use of ticagrelor, ticlopidine, aspirin, cilostazol, picotamide, vorapaxar, and clopidogrel as single antiplatelet therapy or dual antiplatelet therapy (DAPT) in patients with PAD found that clopidogrel monotherapy resulted in the best overall safety and efficacy (reduction of MACEs).²⁹⁷

The long-term use of DAPT or systemic anticoagulation with vitamin K antagonists is not indicated for PAD.^299,300 The role of direct oral anticoagulants is currently the subject of intense investigation. The Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) trial, a multicenter randomized trial of 7470 individuals with stable, mild to moderate PAD, found that low-dose rivaroxaban (an oral factor Xa inhibitor) in combination with aspirin reduced MACEs (death, myocardial infarction, or stroke) and major adverse limb events (MALEs) compared with aspirin alone.³⁷ Patients who had previous lower extremity revascularization, amputation, or history of IC and ABI of <0.9 and documented peripheral stenosis of >50% or carotid stenosis of >50% were included in the study. Overall, 8.5% of study patients had an ABI of <0.7. In this population, there was a significant reduction in MALEs, major amputation, and acute limb ischemia compared with aspirin alone.³⁰¹ This drug combination was associated with a small but statistically significant increase in clinically relevant bleeding. Whereas the study results are promising, the benefits and risks of the low-dose rivaroxaban and low-dose aspirin combination in patients with CLTI have not yet been adequately defined. In addition, this drug combination is not globally available at this time.

The ongoing VOYAGER trial (ClinicalTrials.gov identifier {"type":"clinical-trial","attrs":{"text":"NCT02504216","term_id":"NCT02504216"}}NCT02504216) is comparing the same two antithrombotic regimens in PAD patients undergoing peripheral revascularization.³⁰²

Lipid-lowering therapy.

The Heart Protection Study (HPS) evaluated the effect of blood lipid lowering on cardiovascular events in PAD and included patients with CLTI.⁴⁰ Other studies, although similar, limited inclusion to patients with IC.⁴¹ The HPS included 20,536 high-risk individuals with a total cholesterol concentration of at least 135 mg/dL (3.5 mmol/L). The participants were randomized to 40 mg/d of simvastatin or a placebo. In the simvastatin group, there was a 25% (95% CI, 16%−33%) relative risk (RR) reduction in the first major vascular event among patients who had no history of a coronary event at baseline.⁴⁰ In addition, lipid lowering was shown to be most effective in patients with a blood cholesterol concentration >135 mg/dL (> 3.5 mmol/L). There was also a significant reduction in cardiovascular events (P < .0001) among a subgroup of individuals with PAD.

A Cochrane review evaluated 18 lipid-lowering trials comprising 10,049 PAD patients.^39,42 Whereas the majority had IC and only some trials included CLTI, the results appear relevant to the CLTI population. Only one study showed a negative effect of lipid lowering. When this study was excluded, analysis showed that lipid-lowering therapy significantly reduced the risk of total cardiovascular events in PAD (OR, 0.74; CI, 0.55–0.98).⁴² This was primarily due to a positive effect on total coronary events (OR, 0.76; CI, 0.67–0.87).

The impact of statin agents may extend beyond their lipid-lowering effect by reducing inflammation in patients with PAD.^303,304 An individual-patient data meta-analysis of 54 prospective cohort studies demonstrated that inflammatory biomarkers independently predict vascular risk with a magnitude of effect at least as large as that of blood pressure or cholesterol.³⁰⁵ Even after adjustment for age, sex, and traditional risk factors, patients with PAD are known to have increased levels of inflammatory cytokines, acute phase reactants, and soluble adhesion molecules.³⁰⁶ However, although the attributable vascular risk associated with inflammation is large and animal models using targeted anti-inflammatory therapies have shown promise, it remains unknown whether inhibiting inflammation alone will lower vascular event rates.

The landmark Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) examined the use of intensive statin therapy (rosuvastatin 20 mg daily vs placebo) in a primary prevention trial.^307,308 In total, there were 17,802 individuals who had low levels of LDL-C but an elevated vascular risk based on a proinflammatory biomarker (high levels of high-sensitivity C-reactive protein). Investigators demonstrated a 44% reduction in major vascular events, including a 54% reduction in myocardial infarction, a 48% reduction in stroke, a 46% reduction in arterial revascularization, a 43% reduction in deep venous thrombosis or pulmonary embolism, and a 20% reduction in mortality. The greatest absolute risk and the greatest absolute risk reduction were observed among those with the highest levels of high-sensitivity C-reactive protein. There are now multiple studies showing a decrease in cardiovascular events in patients with established atherosclerosis treated with intensive statin therapy.^{43,224,309,310} A large retrospective cohort study from the U.S. Veterans Affairs population demonstrated reduced mortality and major amputation rates among patients with established PAD receiving intensive-dose statins.³¹¹ Statin therapy can be associated with muscle aching, the most common adverse effect limiting its use. In the setting of this complication, statin dose can be lowered to the maximum tolerated dose, and a second nonstatin cholesterol-lowering drug can be added to reduce cholesterol levels even further.

Recent (2013, 2018) American College of Cardiology/American Heart Association guidelines on treatment of blood cholesterol recommend the use of moderate- to high-intensity statins for all individuals with established atherosclerotic CVD including PAD.^312,313 Both rosuvastatin (20–40 mg) and atorvastatin (40–80 mg) have been shown to be effective.³¹⁰ The 2018 guideline describes “very high risk” individuals to include those with symptomatic PAD and at least one other high-risk condition (age ≥65 years, familial hypercholesterolemia, history of coronary revascularization, DM, hypertension, CKD, current smoking, congestive heart failure)da categorization that applies to the overwhelming majority of patients with CLTI. For this population, high-intensity/maximally tolerated statin dosing is recommended, and if on-treatment LDL-C levels remain ≥70 mg/dL (1.8 mmol/L), the addition of ezetimibe is considered reasonable.³¹³

New lipid-lowering agents have entered the armamentarium. Proprotein convertase subtilisin/kexin type 9 (PCSK9) directs the degradation of LDL receptors in the liver and has become a drug target. The Further Cardiovascular Outcomes Research with PCSK9 Inhibition in Subjects with Elevated Risk (FOURIER) RCT demonstrated an additional benefit of evolocumab (a PCSK9 inhibitor) in reducing MACEs in PAD patients already receiving statin therapy.³¹⁴ The composite end point of cardiovascular death, myocardial infarction, stroke, hospital admission for unstable angina, or coronary revascularization was statistically reduced in PAD patients treated with the PCSK9 inhibitor evolocumab (hazard ratio [HR], 0.79; P = .0040). There was also a reduction in the risk of MALEs, including acute limb ischemia and major amputation. Further studies will be needed in PAD subpopulations including CLTI.

Further studies of these agents are desirable in high-risk PAD subpopulations including CLTI.

Management of hypertension.

It is universally accepted that control of hypertension reduces MACEs in patients with PAD. The International Verapamil-SR/Trandolapril Study (INVEST) analyzed the impact of control of hypertension on all-cause death, nonfatal myocardial infarction, and nonfatal stroke in 22,576 hypertensive patients with stable coronary artery disease (CAD), of whom 2699 also had PAD.⁴⁶ PAD patients had a significantly higher incidence of sustaining a primary end point MACE compared with those without PAD (16.3% vs 9.2%). In addition, among those with PAD, a MACE was less likely to occur in patients with systolic blood pressure <145 mm Hg and diastolic pressures <90 mm Hg. Further reduction of blood pressure to below 130 mm Hg systolic and 80 mm Hg diastolic provides even greater protection from cardiovascular events.⁴⁸ The Systolic Blood Pressure Intervention Trial (SPRINT) compared blood pressure control with a systolic pressure of 120 mm Hg (intensive control) or 140 mm Hg (standard control) in 2510 patients with a mean age of 79.9 years observed for a mean of 3.14 years.³¹⁵ The study documented a significantly lower incidence of composite cardiovascular events of death with intensive control. However, intensive blood pressure control may result in greater morbidity associated with periods of clinically significant hypotension.^45,47 Optimal blood pressure control for patients with CLTI has not been established, and although maintaining systolic pressure <140 mm Hg and diastolic pressure <90 mm Hg is important, lower pressures may be beneficial to further reduce MACEs.

The first-line category of oral antihypertensive does not appear to be of significance. Angiotensin-converting enzyme inhibitors (ACEIs), calcium channel blockers, and diuretics, when successful in lowering blood pressure to target, reduce cardiovascular events to a similar extent.^316,317 Although the Ongoing Telmisartan Alone and in Combination with Ramipril Global Endpoint Trial (ONTARGET) and Heart Outcomes Prevention Evaluation (HOPE) study suggested that in the absence of heart failure, monotherapy with an ACEI (ramipril) reduces the rate of MACEs in high-risk patients, there is recent evidence to suggest that this class of drug may result in a higher amputation rate for patients with CLTI.³¹⁸ In an analysis of the Medicare database for 2007 to 2008, there were 22,954 patients who underwent lower extremity revascularization. Of these, 64.6% were treated for CLTI. Compared with those not taking an ACEI, patients who presented with rest pain and were taking an ACEI after the index procedure had a higher risk of amputation. Other studies have not noted an increased risk of amputation associated with ACEIs but have suggested an increased rate of reintervention. A propensity score-matched cohort study of 17,495 Danish patients compared those receiving ACEIs with those who were not after vascular reconstruction. Observed for a mean of 1.6 years, the patients treated with ACEIs had a lower all-cause mortality (20.4% vs 24.9%) but underwent more reintervention (24% vs 23.1%).³¹⁹ Using the same general methodology, these investigators found that the use of beta blockers after primary vascular reconstruction was associated with a decrease in the incidence of major amputation but a higher rate of myocardial infarction and stroke without an increase in all-cause mortality.³²⁰

Globally, adequate control of hypertension remains a significant challenge. In LMICs, the availability of oral antihypertensives is limited and costs are high, resulting in poor overall blood pressure control. Strategies are urgently required to improve availability and affordability of drugs so that vascular specialists can treat their patients to target.³²¹

There have been concerns that drugs reducing heart rate and blood pressure will worsen ischemia in patients with PAD. Although beta blockade has not been directly evaluated in CLTI, it has been the subject of several clinical trials in IC and has been shown to be effective in lowering blood pressure without worsening symptoms.^322,323

Management of diabetes.

Type 2 DM is a significant risk factor for PAD,^324,325 and the extent of vascular disease appears related to the duration and severity of hyperglycemia. Glycemic control is therefore essential in all diabetic patients with PAD. Metformin monotherapy is generally recognized as the best initial oral hypoglycemic agent. When additional therapy is needed, any other class of oral hypoglycemic agent, including sulfonylurea, thiazolidinedione, dipeptidyl peptidase 4 inhibitor, or α-glucosidase, can be added with equal effectiveness.⁵⁴

Sodium-glucose co-transporter 2 (SGLT-2) inhibitors are a newer class of agents that have been associated with beneficial effects on cardiovascular complications, renal disease, and mortality in type 2 diabetics. However, one large trial (10,142 subjects) demonstrated an approximately 2-fold increased risk of lower limb amputations associated with the use of canaglifozin, an SGLT-2 inhibitor, prompting a “blackbox” warning.^326–328 The mechanism is unclear and may be generically related to diuretic actions in this population.³²⁹ Caution is advised in the use of this agent in diabetic patients with advanced PAD and/or CLTI.

Whereas there are some data to suggest that the dipeptidyl peptidase 4 inhibitors may reduce the risks of myocardial infarction and stroke, the impact on PAD in patients with CLTI has not yet been defined.³³⁰ The goal for most adults with DM is to maintain a glycosylated hemoglobin A1_c level of <7% (equivalent to International Federation of Clinical Chemistry units of 53 mmol/mol).^49–52 However, less stringent goals (eg, hemoglobin A1_c level <8%) may be appropriate for individuals with advanced vascular complications or limited life expectancy.⁵³

Type 2 DM patients with abnormal renal function treated with metformin may be at higher risk for contrast-induced nephropathy and lactic acidosis. Whereas the matter is the subject of continued debate, it is reasonable to withhold metformin for 24 to 48 hours after the administration of an iodinated contrast agent.^{55–57,270,271}

Lifestyle modifications.

In addition to controlling risk factors as discussed, it is important to encourage CLTI patients to adopt a healthier lifestyle. Stopping smoking (tobacco and other recreational drugs) completely and permanently, adopting a healthy diet and weight control, and regular exercise must be stressed as extremely important for both life and limb.^331,332

Tobacco.

The adverse impact of tobacco use on cardiovascular health has been well established. Despite the use of best medical therapy, male and female smokers (even those smoking 1–10 cigarettes per day) have a significantly higher rate of disease progression and MACEs.^58–60 Thus, all patients presenting with CLTI should be asked about smoking and referred to a smoking cessation program if they are still smoking. To encourage compliance with advice to stop smoking, patients should be challenged about smoking at every medical encounter.^61,62 The safety of electronic cigarettes has not been established, including for patients with PAD, and until more evidence becomes available should not be considered in patients with CLTI.³³³

Diet and exercise.

Although diet and exercise have not been specifically evaluated in CLTI, there is compelling evidence that they affect the progression of atherosclerosis. Diets that are high in carbohydrates and saturated fats are associated with a higher risk of MACEs.³³⁴ A diet that reduces the intake of saturated fats and increases the intake of monounsaturated fats, omega-3 fatty acids, antioxidants, and other natural plant sterols and stanols is associated with a reduction in plaque burden and MACEs.^335–337 Patients should be encouraged to adopt a low-fat or Mediterranean diet.³³⁸ Unfortunately, fruits and vegetables are not always available or affordable, especially in LMICs.³³⁹

Although CLTI studies are not available, numerous trials have confirmed the benefits of supervised exercise in IC.³⁴⁰ Exercise-based cardiac rehabilitation reduces the risk of subsequent myocardial infarction and cardiovascular mortality.³⁴¹ It therefore seems reasonable to suggest that a postrevascularization walking-based exercise program would also benefit CLTI patients who are cleared for full weight-bearing.

Assumptions and approach.

As CLTI is usually the result of complex multilevel occlusive disease, certain simplifying assumptions are required to develop a usable anatomic staging system (Table 5.1). First, because existing schemes for AI disease appear adequate, the focus of GLASS is on infrainguinal disease (a simplified inflow disease scheme is presented in Table 5.2). In GLASS, the CFA and PFA are seen as inflow arteries, and the infrainguinal system begins at the origin of the SFA. This is justified by the distinct approaches used in the treatment of CFA and PFA disease (Section 6) and long-term results that are similar to those for AI interventions.

For GLASS to be useful in everyday clinical practice and to form the basis of practice-changing research, it is important that it does not rely on complex methods of lesion characterization. With regard to vessel calcification, GLASS adopts a dichotomous subjective scale in which severe calcification (eg, >50% of circumference; diffuse, bulky, or “coral reef” plaques) increases the within-segment grade by one numeric level. This is a subjective determination made by the treating physician that the severity of calcification significantly increases technical complexity (and expected technical failure rates) for endovascular intervention. Alternative approaches for quantifying arterial calcification in PAD have been suggested but are more complex, and none of these has been validated for discriminating clinical outcomes.^346,347 With regard to IM disease, GLASS employs a three-level modifier (Fig 5.1) to describe the status of arteries crossing the ankle (including the terminal divisions of the peroneal artery) and the pedal arch. Currently, the IM disease modifier is not considered within the primary assignment of limb stages in GLASS, given the absence of strong evidence on how it affects treatment outcomes. It should, however, be captured in future studies to better define how to incorporate pedal outflow disease into anatomic staging in CLTI.

Open in a separate window

Fig 5.1.

Inframalleolar (IM)/pedal disease descriptor in Global Limb Anatomic Staging System (GLASS). Representative angiograms of P0 (left), P1 (middle), and P2 (right) patterns of disease.

GLASS also makes the following assumptions:

• Restoring durable (pulsatile) in-line flow to the affected part, particularly in patients with tissue loss, is a primary goal of revascularization in CLTI.
• Using high-quality imaging (Section 3), the vascular specialist chooses and defines a TAP that is most likely to achieve that in-line flow.
• The TAP will usually involve the least diseased IP artery.
• Other IP arteries (not selected for the TAP) are equally diseased or more so.

In addition, although it is an important research question, the current version of GLASS does not consider multivessel IP revascularization because evidence of its role is still lacking. Where the clinician is considering such revascularization, GLASS staging is based on the primary IP target, as defined by the clinician before the intervention.

In defining infrainguinal anatomic stages (I-III), GLASS combines grades (0–4) for the FP (origin of the SFA to the origin of the anterior tibial [AT] artery; Fig 5.2) and IP (origin of the tibioperoneal trunk and the AT artery to the malleoli; Fig 5.3) segments in series. Stages were developed to correlate with estimated LBP, defined as maintenance of in-line flow through the entire length of the TAP, from the SFA origin to the malleoli. LBP is considered to be lost when any one of the following occurs:

1. Anatomic failure: occlusion, critical stenosis, or reintervention affecting any portion of the defined TAP; or
2. Hemodynamic failure: a significant drop in ABI (≥0.15) or TBI (≥0.10), or identification of ≥50% stenosis in the TAP, in the presence of recurrent or unresolved clinical symptoms (eg, rest pain, worsening or persistent tissue loss).

Open in a separate window

Fig 5.2.

Femoropopliteal (FP) disease grading in Global Limb Anatomic Staging System (GLASS). Trifurcation is defined as the termination of the popliteal artery at the confluence of the anterior tibial (AT) artery and tibioperoneal trunk. CFA, Common femoral artery; CTO, chronic total occlusion; DFA, deep femoral artery; Pop, popliteal; SFA, superficial femoral artery.

Open in a separate window

Fig 5.3.

Infrapopliteal (IP) disease grading in Global Limb Anatomic Staging System (GLASS). AT, Anterior tibial; CTO, chronic total occlusion; TP, tibioperoneal.

LBP is an important new concept allowing more direct comparison between revascularization approaches in CLTI. Estimating LBP after surgical or endovascular intervention is central to the development of EBR (Section 6). The writing group defined three GLASS stages based on the likelihood of immediate technical failure (ITF)³⁴⁷ and 1-year LBP after endovascular intervention of the selected TAP. GLASS stages for the limb thus reflect a gradient of infrainguinal disease complexity:

• Stage I: low-complexity disease: expected ITF < 10% and 1-year LBP > 70%
• Stage II: intermediate-complexity disease: expected ITF < 20% and 1-year LBP 50% to 70%
• Stage III: high-complexity disease: expected ITF > 20%; or 1-year LBP < 50%

Consensus process and assignment of limb stages.

To assign GLASS stages (I-III) in the two-dimensional matrix shown in Table 5.3, a multinational, multispecialty group of vascular specialists (GVG writing group and invited external experts) as well as evidence summaries⁷ and other published material^{79,160,348–404} were surveyed. Representative examples of GLASS stage I to stage III disease are illustrated in the angiograms depicted in Figs 5.4 to 5.6. Table 5.4 provides a descriptive summary of the three GLASS stages.

Open in a separate window

Fig 5.4.

Representative angiograms of Global Limb Anatomic Staging System (GLASS) stage I disease patterns. The target arterial path (TAP) is outlined in yellow. Left panel, TAP includes the anterior tibial (AT) artery. Femoropopliteal (FP) grade is 0. Infrapopliteal (IP) grade is 2 (3-cm chronic total occlusion; chronic total occlusion of AT artery and total length of disease <10 cm). Right panel, TAP includes the peroneal artery. FP grade is 2 (chronic total occlusion <10 cm; total length of disease <²/₃). IP grade is 0.

Open in a separate window

Fig 5.6.

Representative angiograms of Global Limb Anatomic Staging System (GLASS) stage III disease patterns. The target arterial path (TAP) is outlined in yellow. Left panel, TAP includes the peroneal artery. Femoropopliteal (FP) grade is 4 (superficial femoral artery [SFA] disease length, 10–20 cm; popliteal stenosis <5 cm; heavily calcified). Infrapopliteal (IP) grade is 2 (stenosis of tibioperoneal trunk and proximal peroneal <10 cm). Right panel, TAP includes the anterior tibial (AT) artery. FP grade is 4 (popliteal chronic total occlusion extending into trifurcation). IP grade is 3 (chronic total occlusion of target artery origin).

Managing CLTI with GLASS.

Use of the GLASS system involves the following steps (Fig 5.7):

1. Obtain high-quality angiographic imaging to includethe ankle and foot (Section 3).
2. Identify the TAP.
3. Determine the FP GLASS grade (0–4) (Fig 5.2).
4. Determine the IP GLASS grade (0–4) (Fig 5.3).
5. Decide whether there is severe calcification (eg, >50% of circumference; diffuse, bulky, or coral reef plaques likely to compromise endovascular outcomes) within the FP and IP segments of the TAP. If present, increase the segment grade by one.
6. Combine FP and IP grades to determine the overall GLASS stage (Table 5.3).
7. Use the pedal modifier (P0, P1, or P2) to describe the status of IM arteries.

Open in a separate window

Fig 5.7.

Flow chart illustrating application of Global Limb Anatomic Staging System (GLASS) to stage infrainguinal disease pattern in chronic limb-threatening ischemia (CLTI). FP, Femoropopliteal; IP, infrapopliteal; PLAN, patient risk estimation, limb staging, anatomic pattern of disease; TAP, target arterial path; WIfI, Wound, Ischemia, and foot Infection.

For the individual patient with CLTI, an EBR strategy (Section 6) is based on the full integration of

1. estimated patient risk and long-term survival;
2. severity of limb threat (eg, using WIfI) (Sections 1 and 3); and
3. anatomic pattern and severity of disease in the affected limb (eg, GLASS).

PLAN: Patient risk estimation.

The first step involves assessing the patient for candidacy for limb salvage, periprocedural risk, and life expectancy.

CLTI is associated with advanced age, multiple comorbidities, and frailty. The goals of treatment include relief of pain, healing of wounds, and preservation of a functional limb. However, revascularization may incur significant morbidity and mortality, requiring multiple hospitalizations, prolonged outpatient care, and thus considerable health and social care costs. Whereas the majority of patients with CLTI should be considered candidates for limb salvage, some may be appropriately treated with primary amputation or palliation after shared decision-making. Patients, families, and caregivers should have access to appropriate expertise in making these challenging decisions. Although maintenance of independent ambulatory status is an important goal, predicting functional outcomes after revascularization may be challenging, particularly in patients who are severely deconditioned. Palliative care consultants, where available, may be a valuable resource to optimize symptom management in patients with limited goals of care.

Palliative therapy should rarely include revascularization except in special circumstances, such as

• treatment of hemodynamically significant inflow disease, if needed to improve the likelihood of a successful amputation at the most distal possible level; and
• relief of intractable pain or to improve wound healing after shared decision-making with the patient, family, and vascular treatment team.

Estimation of operative risk and life expectancy plays a critical role in EBR. Tradeoffs between risk, invasiveness, hemodynamic gain, and anatomic durability of the vascular intervention are commonly made in everyday practice. Risk stratification tools can assist by providing objective criteria for such decisions. Multiple tools have been developed and applied to the CLTI population (Table 6.1).^{63–67,225,409–412} End points modeled have included all-cause mortality, major amputation, AFS, and perioperative events. The list of predictors identified in these models includes advanced age (>75 or 80 years), CKD, CAD, congestive heart failure, DM, smoking, cerebrovascular disease, tissue loss, BMI, dementia, and functional status. Frailty, a recently identified functional measure, is also of clear importance in the CLTI population.^413,414 Patients with ESRD are at the highest risk in many reports and yet have been specifically excluded in some CLTI studies.^415,416 All of these tools have been developed retrospectively using data from patients who have undergone revascularization, thereby excluding those who were managed conservatively or selected for primary amputation. Whereas some were validated in external data sets of similar patients, none has been prospectively tested across the spectrum of CLTI presenting for initial evaluation and treatment. As such, no specific tool and model can be recommended in preference to others.

Specific recommendations about preoperative cardiac and anesthetic evaluation before limb revascularization are beyond the scope of this document. The reader is referred to Section 4 and to other published guidelines.^417,418

PLAN: Limb staging.

CLTI patients present with a broad spectrum of disease severity. Staging of the limb is central to EBR (Section 3), and use of the SVS Threatened Limb Classification System (WIfI) is recommended (Section 1).^{10,68–72,171} This is the only system that fully integrates wound severity, ischemia, and infection to stage CLTI.

The severity of ischemia and the benefits of revascularization do not map in an exclusively concordant fashion with amputation risk across the spectrum of CLTI, as expressed in the original WIfI consensus document.¹⁰ Expert opinion, now supported by reports from institutional series,^69,70,72 suggests that the presumed benefit of revascularization in CLTI is linked to both the severity of ischemia and the degree of limb threat (Fig 6.3). All symptomatic patients who have severe (eg, WIfI grade 3) ischemia should undergo attempted revascularization, presuming they are appropriate candidates for limb salvage.⁵ In settings of advanced tissue loss or infection (eg, WIfI stage 4 limbs), revascularization may also be of benefit in the presence of moderate ischemia (eg, WIfI ischemia grades 1 and 2). Conversely, patients with lesser degrees of tissue loss or infection (eg, WIfI stages 1 to 3) and mild to moderate ischemia are often successfully treated with infection control and wound and podiatric care. Revascularization may be considered selectively in these patients if their wounds fail to progress (or regress) despite appropriate limb care after 4 to 6 weeks or if they have signs or symptoms of clinical deterioration. In such cases, all elements of the initial staging and treatment plan, including treatment of underlying moderate ischemia, should be re-evaluated. Whenever possible, the limb should be restaged after surgical drainage or débridement and after the infective component is stabilized. During the course of treatment, periodic restaging of the limb is important in guiding subsequent decisions, particularly when there is lack of progress in healing or any deterioration of symptoms.

Open in a separate window

Fig 6.3.

The benefit of performing revascularization in chronic limb-threatening ischemia (CLTI) increases with degree of ischemia and with the severity of limb threat (Wound, Ischemia, and foot Infection [WIfI] stage). WIfI stage 1 limbs do not have advanced ischemia grades, denoted as not applicable (N/A).

WIfI also provides a useful and necessary tool through which one can compare and contrast the quality of different revascularization strategies in CLTI. This has become an issue of critical importance as an ever-increasing array of technologies and treatment strategies are being used. The magnitude and durability of increased perfusion required to resolve the clinical situation, and to maintain satisfactory limb health (eg, preservation of a functional foot, freedom from recurrent CLTI), will vary considerably across the spectrum. The extent of benefit for revascularization (Fig 6.3) is also linked to anatomic durability of the selected intervention. These concepts are central to PLAN and to the development of EBR strategies in CLTI.

PLAN: Anatomic pattern of disease (and conduit availability).

Although secondary to the broader context of patient risk and limb threat severity, the anatomic pattern of arterial occlusive disease is a dominant consideration in EBR. The overall pattern and severity of disease in the limb (eg, as described by GLASS; Section 4) help define the optimal strategy for vascular intervention. Furthermore, the availability and quality of autologous vein conduit (especially the great saphenous vein [GSV]) are key considerations for bypass surgery and should be defined before revascularization decisions are taken in average-risk patients.^13,77,79

“No-option” anatomy.

The majority of CLTI patients are anatomically suitable for revascularization, and establishing direct in-line flow to the foot is the primary technical goal. One important exception is ischemic rest pain, for which correction of inflow disease alone or treatment of FP disease even without continuous tibial runoff to the foot may provide relief of symptoms. This may also be the case in patients presenting with minor degrees of tissue loss (eg, WIfI stage 2). Thus, the definition of a no-option anatomic pattern of disease is dependent on clinical context. Lack of a target artery crossing the ankle and absence of a suitable pedal or plantar artery target (eg, GLASS P2 modifier) may be considered no-option disease patterns in patients with advanced CLTI (eg, WIfI stages 3 and 4). Angiography may occasionally fail to detect a patent distal artery target, and there are reports of successful tibial and pedal bypass grafting based on exploration of an artery identified on Doppler ultrasound examination that was not identified on contrast arteriography.^419,420 Careful selection and experienced surgical judgment are required before proceeding to surgery in such instances.

EBR strategies in CLTI.

The technical options for treating complex patterns of disease in a minimally invasive fashion have increased markedly in recent years and led some to advocate an “endovascular-first” approach for most or all patients with CLTI, reserving bypass surgery as a secondary option. However, existing evidence argues strongly for a selective revascularization algorithm based on specific clinical and anatomic scenarios, as described here. Currently enrolling RCTs are eagerly awaited to provide higher quality data in support of EBR in patients with CLTI.^13–15

The Bypass vs Angioplasty in Severe Ischaemia of the Leg (BASIL) trial (now called BASIL-1) remains the only multi-center RCT to have directly compared an endovascular-first with a bypass surgery-first strategy in limb-threatening ischemia due to infrainguinal disease.^159,421 BASIL was conducted across 27 hospitals in the United Kingdom and enrolled 452 participants between 1999 and 2004. All but six patients in the endovascular arm received plain balloon angioplasty (PBA) alone; approximately 25% of the bypasses were prosthetic; around one-third of the procedures were IP; and just more than 50% of patients were observed for >5 years. Considering the follow-up period as a whole, an intention-to-treat analysis showed no significant difference between the two arms in terms of AFS and overall survival. However, for the approximately 70% of patients who lived for >2 years, HRs for overall survival (0.65; P = .009) and AFS (0.85; P = .108) were better for those treated initially with bypass surgery. An analysis by treatment received showed that prosthetic bypasses performed very poorly (worse than PBA) and that patients having bypass after failed PBA had a highly significantly worse AFS and overall survival compared with those patients who received bypass as their first allocated treatment.¹⁶⁰

A systematic review comparing open and endovascular treatments for CLTI found only nine studies meeting standard criteria, three of which were RCTs (among which only BASIL met all of the study quality benchmarks).⁶ Researchers concluded that low-quality evidence (due to heterogeneity and imprecision) suggested similar mortality and amputation outcomes but better expected patency for bypass surgery. Other comparative reviews have yielded broadly similar conclusions.^{227,422–425} OPGs for endovascular interventions in CLTI based on open surgical data from high-quality sources have been suggested and provide minimum standards of safety and efficacy until direct comparative data become available.¹⁶²

To obtain updated data on outcomes after endovascular and open bypass surgery in CLTI, a review was conducted of comparative studies and noncomparative studies that met more inclusive criteria.⁷ These criteria included prospective study design, 50 or more patients with critical or severe limb ischemia (Rutherford class 4–6 definition), infrainguinal procedure, minimum follow-up of 1 year, at least 50 procedures of each subtype (endovascular or open), and adequate anatomic description of lesion location and types of subinterventions (eg, percutaneous transluminal angioplasty, stent, atherectomy) employed. In total, 44 studies enrolling 8602 patients were reviewed in detail and results tabulated to display outcomes across anatomic subsets and from 30 days to 5-year follow-up intervals. Most of the studies were assessed as having moderate to high risk of bias, and the study quality was variable.

Review of the attributes of these studies revealed several notable limitations: few studies of SFA intervention were included because of inadequate numbers of CLTI patients (vs those with IC); the majority of FP bypass studies included prosthetic grafts; and although a good number of studies (20) addressed endovascular intervention for IP disease, the severity of disease was generally mild to moderate (GLASS IP grades 1 and 2), with no studies including GLASS IP grade 4 disease. Thus, the current state of evidence in CLTI remains severely limited, particularly for assessing endovascular outcomes in commonly encountered, complex (especially distal) disease patterns. Caveats aside, the compendium of data suggests similar mortality, amputation, and AFS rates for endovascular and bypass surgery at 1 year, with improved patency for bypass using vein compared with endovascular interventions or prosthetic bypass grafts at 1 year and beyond.

Additional evidence, including a larger body of retrospective studies and registries, provides further insights into specific factors associated with inferior outcomes for individual techniques and informs current vascular practice.^{79,365,366,369,372,373,376,385,391,393,395,402,407,426–438} Surgical bypass with nonautologous conduits to IP targets in CLTI performs poorly. Similarly, patency rates for endovascular intervention are poor in settings of diffuse tibial disease and popliteal and trifurcation occlusions and are diminished in small, diffusely diseased or heavily calcified FP arteries. Several studies suggest that endovascular outcomes for advanced tissue loss (eg, gangrene, WIfI stage 4, WIfI ischemia grade 3, or foot infection grades 2 and 3) are inferior, with high early rates of major amputation.^171,439 Patients with ESRD experience higher rates of limb loss across all interventions. These factors must be carefully considered in each individual case, evaluating the available treatment options against the patient risk, limb stage, functional status, and presumptive importance of a hemodynamically durable intervention for resolving the clinical scenario at hand.

Finally, a nonselective endovascular-first approach carries some risk of both clinically ineffective and costineffective treatment and potential for harm. Whereas a significant percentage of CLTI patients are appropriate candidates for endovascular intervention, those with severe anatomic patterns and higher stages of limb threat may not be well served by a nonselective approach for several reasons. First, ineffective revascularization can lead to poor symptom relief, limited durability of benefit, delayed wound healing, inadequate clearance of infection, or progression of tissue loss in the foot. There are both patient and system costs to inadequately treated CLTI. Another important consideration is the potential effect of endovascular failures on the outcomes of secondary bypass surgery in CLTI. Although data in this regard are limited, several multi-center data sets including BASIL¹⁶⁰ and large regional registries^440,441 suggest that the outcomes of bypass surgery in patients who have undergone failed endovascular interventions are significantly inferior to those in patients who underwent primary bypass surgery. The inferior outcomes associated with “secondary bypass” are similar whether the initial failure was percutaneous or a prior bypass graft. This may be a particularly high penalty to pay if clinical success of the initial procedure was short-lived. These studies cannot establish causality vs association, but they strongly suggest that the success of the initial vascular intervention is of importance in CLTI and that endovascular failure, like open bypass failure, carries consequences. Thus, an important consideration is to avoid risking potential loss of bypass targets in performing endovascular interventions. Conversely, surgical bypass may incur significant morbidity and mortality despite the potential attractiveness of greater durability. Factors that may increase the risk of wound complications, graft failure, or other major postoperative complications must be carefully weighed. These considerations informed the consensus recommendations on specific EBR strategies.

EBR: Treatment of inflow disease.

Inflow disease is defined here as proximal to the origin of the SFA and meeting one or more of the following criteria:

• absent femoral pulse
• blunted CFA waveform on Doppler ultrasound
• >50% stenosis by angiography in the aorto-iliac arteries or CFA
• aorta to CFA systolic pressure gradient >10 mm Hg at rest

The decision to perform staged vs multilevel revascularization for patients with combined inflow and outflow disease is individualized on the basis of severity of limb threat (especially presence of tissue loss), anatomic complexity, and patient risk. In settings of rest pain and minor tissue loss, inflow correction alone may suffice to achieve the desired clinical outcome. As procedural complexity increases, perioperative morbidity and mortality rise as well. Most patterns of AI disease may be successfully treated using an endovascular approach, frequently employing bare-metal or covered stents.^82–84 Surgery is often reserved for extensive occlusions or after failure of endovascular procedures. The choice of an open surgical inflow procedure should be based on patient risk, anatomic pattern of disease, and other clinical factors. Direct anatomic bypass (eg, aortofemoral) grafting may be preferred to extra-anatomic reconstruction in average-risk patients with severe ischemia (WIfI ischemia grades 2 and 3) because of greater anatomic and hemodynamic durability.^85–87

CFA endarterectomy can be performed with low morbidity and excellent long-term durability.^88,89 It remains the optimal approach to treatment of hemodynamically significant CFA disease, which often includes bulky calcific plaque. In some cases, femoral interposition grafting may be preferred. In all cases, durable in-line PFA flow should be maximized. CFA endarterectomy may be combined with proximal intervention to treat combined disease in a “hybrid” fashion.⁹⁰ Although long-term outcome data are sparse, reports suggest that endovascular treatment of CFA disease may be a safe alternative in selected patients (eg, high surgical risk, hostile groin anatomy).^91–94

Surgical treatment (eg, profundaplasty or bypass grafting) of PFA disease is an important component of CLTI revascularization with a major impact on the long-term prognosis for the limb. The indications for and optimal approaches to treatment of nonorificial (ie, not in continuity with the CFA) or long-segment PFA disease are not established. There is limited evidence regarding the use of endovascular interventions for PFA disease. However, it may be considered a secondary approach in settings of hostile groin anatomy or in other high-risk circumstances.

EBR: Treatment of infrainguinal disease in average-risk patients.

Outflow (infrainguinal) disease starts at the SFA origin (Section 5). An average-risk patient is defined as one in whom the anticipated periprocedural mortality is <5% and the anticipated 2-year survival is >50% (Recommendation 6.4). These patients are potential surgical or endovascular candidates, depending on individual clinical and anatomic factors.

Fig 6.4 provides a summary of preferred infrainguinal revascularization strategies for an average-risk patient with available vein conduit based on the presenting combination of limb stage (WIfI) and anatomic pattern of disease (GLASS). Open bypass surgery and endovascular therapy have complementary roles, with notable lack of consensus across the intermediate ranges of clinical and anatomic complexity. Comparative effectiveness studies employing these staging schemes are urgently needed to improve the quality of evidence for interventions in specific clinical scenarios.

Open in a separate window

Fig 6.4.

Preferred initial revascularization strategy for infrainguinal disease in average-risk patients with suitable autologous vein conduit available for bypass. Revascularization is considered rarely indicated in limbs at low risk (Wound, Ischemia, and foot Infection [WIfI] stage 1). Anatomic stage (y-axis) is determined by the Global Limb Anatomic Staging System (GLASS); limb risk (x-axis) is determined by WIfI staging. The dark gray shading indicates scenarios with least consensus (assumptions—inflow disease either is not significant or is corrected; absence of severe pedal disease, ie, no GLASS P2 modifier).

Patients lacking adequate autologous (GSV) conduit must be considered separately as this is a critical factor in determining the likely success and durability of bypass surgery. For those with no suitable venous conduit, prosthetic or venous allografts are the only options. Given the inferior performance of these conduits in CLTI, endovascular intervention is preferred when possible.¹⁶⁰ Use of prosthetic or biologic conduits (eg, cryopreserved vein allografts) for infrainguinal bypass in CLTI may be reasonable in highly selected cases, such as in patients with untreatable anatomy for endovascular intervention or prior endovascular failure, with acceptable runoff, and in patients who are able to tolerate aggressive antithrombotic therapy.

In many patients lacking GSV, arm/spliced vein bypass conduits may be an option. However, the results of arm/spliced vein bypass are highly dependent on the operator’s training and experience. The determination of when and how to employ these alternative vein conduits is surgeon specific. In general, large single-center and multicenter reports demonstrate that arm and spliced vein bypasses perform better than nonautologous grafts to distal targets and are inferior to autologous GSV conduits.^7,79,442,443 However, these higher risk vein grafts require closer surveillance and more reinterventions to maintain primary assisted patency.⁴⁴⁴

EBR: Treatment of infrainguinal disease in high-risk patients.

A high-risk patient is defined as one in whom the anticipated perioperative mortality is >5% or the anticipated 2-year survival is <50%. Because endovascular intervention can be performed with reduced morbidity, it may often be preferred in high-risk patients who are otherwise candidates for functional limb salvage. Shared decision-making is of great importance in high-risk patients to allow the patient, family, and other stakeholders to express value judgments on the tradeoffs between risk and effectiveness in relation to the desired goals.

EBR: Infra-malleolar disease.

Severe IM disease creates a major challenge to effective revascularization.⁴⁰⁷ The P2 modifier in GLASS describes the circumstance in which no named artery crosses the ankle into the foot and there is no suitable target for bypass surgery. Although technically successful endovascular interventions in the pedal arch have been reported, their durability and hemodynamic and clinical effectiveness remain unknown.⁴³⁸ Diabetic patients often have a segment of preserved pedal artery that may be a target for bypass. Open bypass surgery has also been successfully employed to tarsal and plantar arteries, but again, techniques and outcomes are not established. Given the technical difficulty and the likely reduced hemodynamic impact and durability, the appropriate role for interventions at this level is not determined. The impact of IM disease on the success of proximal revascularization, whether open or endovascular, is likewise unknown. Although the presence of an intact pedal arch appears important for both, clinical success may still be attained in the presence of significant IM disease. The severity of limb threat (tissue loss or infection) is likely to be a critical modifier of the relationship between IM disease severity and postprocedural clinical outcomes.

EBR: Role of angiosome-guided revascularization.

Whereas few would argue about the desirability of maximizing perfusion at the site of tissue loss, there is considerable debate about the utility of angiosome-guided revascularization.^445,446 First, unambiguous assignment of foot wounds to an individual angiosome is possible in only a minority of cases.⁴⁴⁷ Toe lesions, which typically represent more than half of the lesions encountered, have a dual blood supply (AT and PT), although for more proximal foot lesions, unique angiosome assignment may be achieved in up to 75% to 80% of patients. Then there is the practical question of whether the desired target artery for the angiosome is available and the comparative hemodynamic and clinical effectiveness of “direct” vs “indirect” revascularization. Tibial and peroneal bypasses perform equally well for limb salvage, and DP bypass can be effective for some hindfoot lesions.⁴⁴⁸ Systematic reviews have yielded conflicting results,^96–99 and data are inextricably confounded by the quality of the pedal arch and the nature of the revascularization performed.^95,449 Whereas wound healing may be improved when direct revascularization is achievable, major amputation rates and patency are not consistently different. To date, none of the analyses take into account the confounding effect of limb staging, for example, using WIfI. In summary, angiosome-guided revascularization may be of importance in the setting of endovascular intervention for midfoot and hindfoot lesions but is likely to be irrelevant for ischemic rest pain and of marginal value for most forefoot lesions and minor ulcers. The role of multivessel (tibial) revascularization is also currently unknown. However, it may be reasonable in selected patients with advanced limb threat (eg, WIfI stages 3 and 4) undergoing endovascular therapy if it can be safely accomplished without risking loss of a bypass target or compromising runoff to the foot.

EBR: Preferred endovascular techniques for infrainguinal disease.

PBA, drug-coated balloon (DCB) angioplasty, stent placement (bare-metal stent, drug-eluting stent [DES], or covered stent), and atherectomy may all be reasonable options in specific circumstances and lesion anatomies. However, unfortunately, there are few high-quality comparative data to guide the choice of a specific endovascular approach in CLTI.^{7,380,387–389,396,450–455}

PBA may be inferior to DCB angioplasty and stents for the treatment of intermediate-length SFA disease (FP grades 2–4) in patients with IC and possibly rest pain.^100–103 However, there are inadequate data to support a preferred endovascular approach for FP disease in CLTI.

PBA remains a reasonable primary endovascular approach for anatomically suitable IP disease as current evidence is inadequate to support other, more expensive techniques. Atherectomy is not superior to PBA and is associated with greatly increased costs.⁴⁵³ Combination approaches, such as atherectomy followed by DCB angioplasty, add significant cost and lack high-quality comparative data. Several modestsized trials suggest potential short-term benefit for DESs in short (ie, <3 cm) tibial lesions, but one cannot generalize these data to the population of CLTI patients as a whole, who typically present with much more extensive disease.^7,456 DES may be a preferred endovascular “bailout” after technical complications (eg, dissection) or failed PBA for short, proximal IP lesions. Although early studies suggested a potential advantage for DCBs in tibial arteries, an RCT showed no benefit of DCB angioplasty over PBA, with a nonsignificant higher rate of amputations in the DCB angioplasty group.³⁹⁶ The results of further, ongoing studies are awaited. In summary, PBA currently remains the standard of care for the endovascular treatment of IP disease in CLTI.

Technical advances in endovascular intervention include improved wires, low-profile catheters, and retrograde access to allow treatment of complex disease patterns down to the distal calf and foot. Specialized catheters may facilitate crossing of difficult chronic total occlusions and ensure re-entry into the true lumen. Retrograde access techniques using either fluoroscopic or ultrasound guidance may increase the ability to cross chronic total occlusions at the IP and popliteal levels. The “pedal loop technique” has been described to achieve complete arch reconstitution in the presence of IM disease, and some reports suggest that it may be of value in highly selected patients.^438,457 The clinical efficacy of these techniques remains to be defined in CLTI as hemodynamic durability remains the primary limitation of endovascular interventions in high-complexity target path anatomy.

Interventional nonrevascularization treatments

Lumbar sympathectomy (LS)

Intermittent pneumatic compression (IPC)

Guidelines on nonrevascularization interventions

The TransAtlantic Inter-Society Consensus II (TASC II) document on the management of PAD concluded that there is low-level evidence available for the recommendation of SCS.¹⁵⁶ Likewise, guidelines from the ESVS state that the benefit of SCS is unproven, with insufficient evidence to recommend its use in the treatment of CLTI.⁴⁹⁶

Although the TASC II document did not include LS in the treatment of CLTI, it did mention its potential role in the management of complex regional pain syndrome.¹⁵⁶ The ESVS guidelines conclude that LS should not be considered an option to prevent amputation but can be considered in patients who are not amenable to revascularization to relieve symptoms.⁴⁹⁶ The American Heart Association’s guidelines on the management of PAD do not mention LS.⁴⁹⁶ Finally, the international guidelines make no reference to IPC at all.

Pharmacotherapy

Vasodilators

Because vasodilators can cause shunting of blood away from ischemic areas to nonischemic areas, they are of no value to patients with CLTI.¹⁵⁶

Defibrinating agents

Two small RCTs compared ancrod, a defibrinating agent, with placebo in CLTI.^506,507 Although one study showed positive changes in APs and TPs, both studies failed to demonstrate any improvements in clinical outcome.

Hyperbaric oxygen therapy (HBOT)

There are numerous plausible mechanisms for HBOT to have a therapeutic role in CLTI. These include increased oxygen transport capacity of plasma (independent of red blood corpuscle number and function), improved function of the leukocyte oxygen-dependent peroxidase system, reduced tissue edema due to the osmotic effect of oxygen, stimulation of progenitor stem cell mobilization and angiogenesis, and improved fibroblast function.⁵⁰⁸ If there is superimposed infection, HBOT also inhibits bacterial growth (particularly anaerobes), generates free radicals that destroy bacterial cellular structures, and improves the oxygen-dependent transport of antibiotics.⁵⁰⁹

In 2015, a Cochrane review of the role of HBOT in healing of chronic wounds was published,¹¹⁰ involving 12 trials and 577 patients. Ten of the 12 trials studied the effect of HBOT on ulcer healing in patients with diabetes. The 2015 review concluded that HBOT increased the rate of ulcer healing in DFUs at 6 weeks but not at longer term follow-up, with no significant difference in the risk of major amputation.¹¹⁰

Three other studies involved patients with ischemic ulcers, but each study used varying definitions of ischemia.^510–512 Abidia et al⁵¹¹ randomized 18 patients with an ABI of <0.8 or TBI of <0.7 and found improvement in wound healing in the treatment group. Löndahl et al⁵¹² randomized 94 patients with adequate distal perfusion or nonreconstructible arterial disease. They found that 57% of patients had a TP of <60 mm Hg (median, 52 mm Hg). Complete ulcer healing occurred in 52% of the patients treated with HBOT compared with 29% of controls at 12 months (P < .02). Stratification based on TPs did not appear to affect healing rates. A subsequent publication by this group demonstrated that preintervention TcPO₂ correlated with ulcer healing and that individuals with a TcP_O2 of <25 mm Hg did not heal.⁵¹³ There was no significant difference in major amputations between the two groups, with three amputations in the HBOT cohort and one in the control cohort.

One study randomized 70 patients with DFUs to either HBOT or standard care.⁵¹⁰ The mean ABI and TcP_O2 were 0.65 and 23 mm Hg in the HBOT cohort and 0.64 and 21 mm Hg in the non-HBOT group. All patients with an ABI <0.9 or TcPO₂ <50 mm Hg were considered ischemic, underwent an iloprost infusion, and were examined for possible revascularization. Thirteen patients in each group underwent a revascularization procedure. At the completion of the therapy, resting TcPO₂ increased by a mean of 12.1 in the HBOT group and 5.0 in the control group (P< .0002). There was a significant reduction in major amputations in the HBOT group (P< .016).⁵¹⁰

A large longitudinal cohort study using data from a wound healing group in the United States⁶¹ included patients with DFUs and adequate foot perfusion as determined by clinicians. A total of 793 patients underwent HBOT. Propensity scoring was used to compensate for the lack of randomization. The study found that individuals treated with HBOT were less likely to have healing of ulcers (HR, 0.68; 95% CI, 0.63–0.73) and more likely to undergo lower limb amputation (HR, 2.37; 95% CI, 1.84–3.04).⁵¹⁴

A subsequent multicenter RCT (Does Applying More Oxygen Cure Lower Extremity Sores? [DAMO₂CLES]) undertaken in 25 hospitals in the Netherlands and Belgium randomized 120 patients with an ischemic foot wound and diabetes to standard care with or without a course of HBOT. Ischemia was defined as AP <70 mm Hg, TP <50 mm Hg, or TcPO₂ <40 mm Hg. All patients were assessed for revascularization, and when applicable, this was generally performed before HBOT. Primary outcomes were limb salvage, wound healing at 12 months, and time to wound healing. Mortality and AFS were also analyzed. Limb salvage (47/60 in the standard care cohort and 53/60 in the standard care with HBOT cohort), index wound healing at 12 months (28/60 in the standard care cohort vs 30 in the standard care with HBOT cohort), and AFS (41/60 in the standard care cohort vs 49 in the standard care with HBT cohort) were not significantly different between the two groups. A high proportion (35%) of those allocated to HBOT were unable to undergo HBOT or did not complete at least 30 treatments, mostly for medical comorbidities or logistical reasons, reinforcing the significant medical comorbidities present in these patients.¹¹²

Overall, whereas controversy remains, there may be a role for the use of HBOT to accelerate ulcer healing in diabetic patients with nonhealing neuropathic ulcers and low-grade ischemia who have failed to respond to conventional wound care. However, HBOT does not prevent major limb amputation and should not be used as an alternative to revascularization in patients with CLTI.

Guidelines on nonrevascularization pharmacotherapy

The TASC II document notes that although previous studies with prostanoids in CLTI suggested improved healing of ischemic ulcers and reduction in amputation, trials do not demonstrate a benefit for prostanoids in promoting AFS.¹⁵⁶ The current PAD guidelines and recommendations of the American College of Cardiology Foundation and the American Heart Association state that parenteral administration of PGE1 or PGE2 may be considered to reduce pain and to improve ulcer healing in CLI but that the beneficial effect is likely to occur only in a small subset of patients.⁵¹⁵

Finally, international guidelines do not address vasoactive drugs, vasodilators, or defibrinating agents. However, the TASC II guideline advocated for considering HBOT in selected patients who have not responded to revascularization.¹⁵⁶

Trials of gene and stem cell therapy in CLTI

Safety of therapeutic angiogenesis

Early concerns about off-target angiogenesis and the potential for progression of diabetic proliferative retinopathy or occult tumor growth previously resulted in significant restrictions in the inclusion and exclusion criteria for entry into these studies. As early studies demonstrated an acceptable safety record for this therapy and potential concerns about off-target angiogenic complications lessened, these restrictions have since decreased.

Unanswered questions in the field

Minor amputations.

Minor amputations of the foot include digital and ray amputation of the toe, transmetatarsal amputation of the forefoot, and Lisfranc and Chopart amputations of the midfoot. Each of these can be useful to preserve foot function in appropriately selected patients. Although there is a significant risk of need for reamputation at a higher level in diabetics, the use of minor amputations, including single-digit and ray amputations, can preserve foot function in the majority of patients.^538–540 There are some instances in which transmetatarsal amputation may be a better first procedure, including necrosis of the great toe requiring long ray amputation or ray amputation of the first and fifth toes, but ensuring adequate distal perfusion and appropriate offloading of the forefoot are the major principles for preservation of foot function.^114,541

There are, however, situations in which an aggressive attempt at limb salvage would be unlikely to succeed, would pose too great a physiologic stress on the patient, or would be of limited value because of other causes of limb dysfunction. For these patients, major amputation may be considered a reasonable option. Because a well-planned primary amputation can often result in a high likelihood of independent ambulation for many patients, this procedure should not be considered a failure of vascular surgery. Rather, it should be viewed as another path to the goal of preserving the walking ability in carefully selected patients or for resolution of ischemic pain, ulceration, and infection.

Primary amputation.

Primary amputation in patients with CLTI is defined as lower extremity amputation without an antecedent open or endovascular attempt at limb salvage. There are four major goals of primary amputation for patients with CLTI: (1) relief of ischemic pain; (2) removal of all lower extremity diseased, necrotic, or grossly infected tissues; (3) achievement of primary healing; and (4) preservation of independent ambulatory ability for patients who are capable. In addition, there are five major indications for primary amputation.

1. Nonreconstructible arterial disease, as confirmed by clear distal imaging studies that fail to identify patent distal vessels needed for a successful intervention. In the setting of severe distal ischemia, in particular in association with ischemic ulceration, gangrene, or infection, the inability to improve straight-line distal perfusion often results in major amputation even with a patent bypass graft. Bypasses to arteries that do not have at least large, angiographically apparent collateral vessel outflow provide little additional flow to the foot for distal limb salvage.⁵⁴² Patients without any appropriate targets for successful distal revascularization are frequently better served with a primary major amputation.
2. Destruction of the major weight-bearing portions ofthe foot, rendering it incompatible with ambulation. The weight-bearing portions of the foot consist of the calcaneus, the first and fifth metatarsal heads, and a functional arch. Patients with gross destruction of the calcaneus and overlying skin should be considered for primary amputation because a functional foot can infrequently be salvaged. After aggressive heel ulcer excision and extensive calcanectomy, complete wound healing is infrequent and chronic pain is common.^543,544
3. Nonfunctional lower extremity due to paralysis or unremediable flexion contractures. These patients are unlikely to benefit from attempts at revascularization, and there will be little change in quality of life despite a successful intervention.
4. Severe comorbid conditions or limited life expectancy due to a terminal illness. The goal of treatment for these patients is relief from ischemic pain, if present, and an improvement in the remaining quality of life. Extensive distal revascularization, prolonged hospitalization, and protracted recovery should be avoided. Assessment of the patient’s frailty may be of value to determine whether primary major amputation is more appropriate than distal revascularization.^545,546
5. Multiple surgical procedures needed to restore a viable lower extremity. As the technology and techniques of vascular surgery have improved, surgeons have advanced beyond revascularization to complex vascular and soft tissue reconstruction. This approach usually involves multiple surgical procedures to increase distal flow, removal of all necrotic tissue, and reconstruction of these areas with free flaps. The course of treatment is prolonged, involving multiple returns to the operating room, long periods of inactivity, and a difficult recovery. For these patients, if multiple procedures with high morbidity are required, primary amputation should be strongly considered to permit early ambulation. A detailed discussion with the patient to develop a comprehensive treatment plan with shared decision-making is important for such advanced vascular disease.

For all patients considered for primary amputation, also consider revascularization to improve inflow in an attempt to reduce the level of the amputation.^118,119 For example, those patients with extensive infrainguinal arterial occlusion, including the common and proximal PFA, might benefit from restoration of flow into the deep femoral system to reduce the amputation level from the upper thigh to the level of the knee. In such cases, despite some additional risk, proximal revascularization has the potential to offer a tangible and significant benefit to the patient.

Secondary amputation.

For those in whom one or more attempts at revascularization have failed and the likelihood of a successful and durable redo procedure is limited, major amputation with a goal of rehabilitation to independent ambulation should be considered.

Level of amputation.

Selecting the level of amputation that will heal primarily is critical to successful prosthetic rehabilitation and maximal functional mobility. Thus, a great deal of consideration must go into selecting the initial level of amputation. Preoperative tissue perfusion assessment can make it possible to lower the level of amputation, although there is no accurate method to predict the optimal level of amputation.⁵⁴⁷ In addition, whereas assessment of preoperative tissue perfusion can aid in decision-making, it still remains largely a clinical decision. Many techniques to evaluate tissue perfusion have been tried, including laser Doppler flowmetry, thermography, skin perfusion pressure, fluorometric quantification of a fluorescein dye, TcPO₂, and indocyanine green fluorescence angiography. In particular, TcPO₂ has been extensively evaluated, and it has been shown that wound complications increase as TcPO₂ levels fall below 40 mm Hg.⁵⁴⁷ Currently, there is still no single definitive method of evaluating tissue perfusion that can accurately predict the wound healing potential or failure at the site of amputation.

Healing rates of amputations and reamputations.

Achieving primary healing is challenging in ischemic lower limbs, and it is difficult to predict early failure (Table 9.1). Multiple débridements and reamputations are required in 4% to 40% of patients, depending on the level of amputation.^548–550 Likewise, readmission rates of 20% have been reported even after minor amputations (toe and distal forefoot), with the majority of reamputations occurring within 1 month.^548–550 Reported long-term healing rates after transmetatarsal amputations are approximately 53%.⁵⁵¹ These amputations should not be offered to patients who have poor rehabilitation potential.

The role of partial foot or midfoot (eg, Lisfranc, Chopart) amputations remains controversial. Prosthetic specialists discourage the use of these procedures as they have higher rates of delayed healing, require more revisions, and develop deformities and ulcers, and patients often struggle to achieve their full rehabilitation potential. Conversely, these amputations preserve a weight-bearing heel and allow amputees the ability to mobilize for short distances without prostheses.⁵⁵²

Transtibial amputation (below-knee amputation [BKA]) and transfemoral amputation (above-knee amputation [AKA]) are performed with an almost equal frequency in patients with CLTI. Reports have shown primary healing rates for BKA of approximately 60%, with 15% leading to a transfemoral amputation.^121,548 The transfemoral amputation has the highest probability of successful primary healing and therefore has been the amputation of choice in individuals who are less likely to ambulate with a prosthesis.

Recent data from the American College of Surgeons National Surgical Quality Improvement Program show improved results with a 12.6% early failure rate for BKA compared with 8.1% for AKA.⁵⁵³ A similar trend is found in data from the National Vascular Registry of the United Kingdom, which show that one in eight AKAs and one in six BKAs remain unhealed at 30 days.⁵⁵⁴

Knee disarticulation.

The biomechanical advantages of a knee disarticulation or through-knee amputation (TKA) compared with an AKA are well recognized, although it remains an infrequently performed amputation. A well-performed TKA offers healing rates that are comparable to those of AKA and provides bedridden and wheelchair-bound patients with a higher level of mobilization and transfer, counterbalance, and reduced potential for contractures. Even in patients who have rehabilitation potential, the current prosthetic technology permits excellent functional mobility, making TKA a good amputation choice when a BKA is unlikely to heal. The aesthetic disadvantage of a TKA is that the prosthetic knee will be marginally distal to the normal contralateral knee in a sitting position.

Mortality.

Survival after major lower limb amputation is poor, as seen in a systematic review that reported 30-day postoperative mortality rates of 4% to 22%.⁵⁵⁵ Even after minor amputations, the 1-year and 5-year mortality rates are reported to be 16% and 25%, respectively, for those with limb ischemia.⁵⁵⁶ Mortality rates for minor amputations are higher in diabetics, with type 2 diabetics having a 5-year mortality of >50%.⁵⁵⁷ The 5-year mortality after major amputations varies from 30% to 70% and is significantly worse for AKA than for BKA.^558,559 The mortality is even higher in bilateral lower limb amputees, with a 5-year survival rate of <40%.⁵⁶⁰ These mortality rates demonstrate the high rate of comorbidities and the frailty of this group of patients.

In patients with diabetes who have had major amputations, survival is often worse than in some malignant diseases. Survival rates have been reported as 78% at 1 year, 61% at 3 years, 44% at 5 years, and 19% at 10 years.⁵⁶¹

In 2010, recognizing the need to do more to reduce perioperative mortality, the Vascular Society of Great Britain and Ireland introduced a quality improvement framework to reduce mortality from amputation surgery to <5% by 2015, which was later revised to <10% in 2016.⁵⁶² Recent data from the United Kingdom’s National Vascular Registry showed mortality rates of 11.6% for AKA and 6.1% for BKA by establishment of dedicated multidisciplinary amputation services that provided expeditious and comprehensive preoperative and postoperative care.⁵⁵⁰ These rates are similar to results from the American College of Surgeons National Surgical Quality Improvement Program of 12.7% for AKA and 6.5% for BKA, with an overall 9.1% mortality of 6389 patients studied.⁵⁶³

Fate of contralateral limb after lower extremity amputation.

Published reports of the risk of contralateral amputation vary from 2.2% to 44%, with a lower risk if the index amputation is a minor amputation.¹²⁴ In most patients, the reason for contralateral amputation is disease progression, although the medical management of unilateral amputees can also be suboptimal, with one-third of patients not prescribed a statin and an antiplatelet agent.¹²³ Continued follow-up of these patients at least yearly after amputation with attention to the contralateral limb is important.¹²⁴

Prosthetic rehabilitation, mobility, and quality of life.

When an amputation is inevitable, and whenever possible, a prosthetic specialist should be involved in decision-making with the surgical team regarding the optimal level of amputation that will ensure the best opportunity for healing, survival, and maximum functional mobility. Advances in prosthetics have resulted in a prosthesis for every stump. However, to use the prosthesis effectively, the stump must be created to truly function as a dynamic sensorimotor end organ and not simply as an inert filler in the socket.

Muscle-stabilizing procedures can help create a stump with its proprioception intact and any of the procedures can be used, including myoplasty, myodesis, and osteomyoplasty. The stump evolves with time, and the prosthetic requirements continue to change. The patient requires regular adjustments in the prosthesis and often complete revisions. A poorly fitting prosthesis can be as disabling as the actual amputation.

The quality of life after amputation is significantly influenced by pain, social isolation, depression, and the patient’s lifestyle before amputation. Mobility has a direct effect on quality of life. It is a key determinant to the social reintegration of the amputee and has a beneficial effect on late mortality.

Energy expenditures of ambulation increase with ascending levels of amputation. Energy consumption during ambulation is increased by 10% to 40% after BKA and by 50% to 70% after AKA.⁵⁶⁴ The potential for rehabilitation is better with BKA than with AKA. Therefore, it is worthwhile to try to salvage a BKA in a patient who has the potential to ambulate fully. In studies involving >100 patients, ambulatory status at 6 to 12 months after amputation varies from 16% to 74%.⁵⁵¹ At 2 years, only 40% of BKA patients achieve full mobility.¹²¹

Maintaining ambulation is one of the most important factors in preserving independence. A significant amount of evidence is available to suggest that early postsurgical prosthetic fitting leads to early mobility.⁵⁶⁵ However, to achieve and to maintain daily functional ambulation, multidisciplinary inputs are needed from physiotherapists, occupational therapists, prosthetists, social workers, recreational therapist nurses, psychologists, and the surgeon. Despite initial successful prosthetic rehabilitation, prosthetic use deteriorates over time, and most patients eventually become household walkers only.⁵⁶⁶

Medical therapies

All patients who have undergone revascularization for CLTI should continue with best medical therapies to slow the progression of atherosclerosis and mitigate the adverse impact of risk factors as recommended in Section 4. In addition, the role of specific pharmacotherapy for maintaining the benefits of revascularization has been the subject of a number of studies.

Surveillance and reintervention

Objective performance goals OPGs

The SVS Critical Limb Ischemia Working Group developed a standardized set of outcome measurements, OPGs, derived from CLTI patients undergoing open bypass in several RCTs.¹⁶² The OPGs include major adverse limb events (MALE) and postoperative death as a measure of early safety and AFS to define longer term clinical effectiveness. Additional safety and efficacy OPGs were created for specific outcome variables of interest, and risk-stratified guidelines based on clinical, anatomic, and conduit criteria were identified for defined subgroups. The main aim of the OPG initiative was to establish benchmark values against which novel endovascular therapies could be initially evaluated without undertaking full RCTs. However, without good-quality RCTs, OPGs cannot be refreshed and, over time, will increasingly come to rely on historical controls. As such, RCTs are still required to determine both the clinical effectiveness and cost-effectiveness once safety and efficacy OPGs have been met.

RCTs

An appropriately designed RCT remains the optimal means of providing critical confirmatory evidence before the widespread adoption of novel interventions.^629–631 The paucity of such studies in CLTI,^13–15,632 however, underscores the many challenges that aspiring investigators face, particularly in trying to complete trials on time and on budget.

Outcomes

Sample and effect size

CLTI patients who are entered into the “nonactive treatment” (placebo) group in RCTs often have outcomes that are better than expected compared with similar patients who are treated outside of research conditions. This makes it more difficult to demonstrate differences in clinical effectiveness and cost-effectiveness among the comparator interventions for CLTI. As a result, researchers must avoid the potential pitfall of basing the power calculation for their trial on an unrealistically large effect size. It is widely agreed that it is poor science and unethical to embark on a trial when there is no realistic prospect of answering the question being posed. An overpowered trial is equally undesirable as it is a misuse of resources, and patients may be disadvantaged by continuing to receive a treatment that is likely of little or no value or even potentially harmful to them. Despite this understanding, the CLTI literature is characterized by studies that present highly questionable, post hoc, subgroup analyses. To guard against this, all CLTI protocols along with full statistical analysis plans should be published in peer-reviewed journals to allow independent, public, and transparent scrutiny.

Beyond the pivotal RCT

Given the challenges inherent in evaluating the wide array of novel endovascular modalities for CLTI, comparative trials of varying size and scope can be effective in establishing the utility of a particular technique, device, or overall revascularization strategy. As described within the OPGs, focused superiority or noninferiority RCTs can also be used to test a novel intervention against more established alternatives, and the safety and efficacy of new technologies can be effectively studied in a timely fashion. However, once the pivotal RCTs have been successfully completed, it is important that ongoing surveillance be rigorously undertaken with the use of well-designed, large, prospective, observational studies, including disease- or procedure-based national registries. Of note, some countries require manufacturers and importers to submit reports of device-related deaths, serious injuries, or malfunctions to the appropriate regulatory bodies.

Also important is cooperation among publicly and industry-funded investigators in designing and performing RCTs. Currently, this is happening with the BASIL and BEST-CLI trials, which will serve to facilitate subsequent individual patient data analyses, meta-analyses, and subgroup analyses. Ultimately, this type of data sharing will provide a powerful framework for refining OPGs and validating the use of tools to better define patient, limb, lesion, and anatomic risk in CLTI patients, such as WIfI and GLASS.

Center of Excellence.

In 2010, building on the work of the International Working Group on the Diabetic Foot, three tiers of care were proposed for an amputation prevention team—basic, intermediate, and Center of Excellence (Table 12.1).⁶⁴⁹ The basic model of care is performed in an office setting with a general practitioner, internist, or endocrinologist and a specialist nurse. An intermediate model of care is set in a hospital or multidisciplinary clinic and consists of various specialists to heal wounds and to prevent limb loss. This model is similar to a wound care center in the United States or a diabetic foot clinic in Europe. A Center of Excellence model is typically found in a tertiary care hospital with a predetermined team of specialists operating under clinical practice pathways, policies, and procedures. The Center of Excellence has advanced diagnostics and can intervene rapidly to prevent limb loss.

Currently, in many countries, there are no criteria required to designate oneself a Center of Excellence for health care. Anyone or any institution can use the terminology, and doing so does not guarantee that excellent care is being delivered. Based on experience in creating Centers of Excellence, a set of criteria are proposed to determine Center of Excellence designation in CLTI and amputation prevention, as outlined in Table 12.2.

Team setting, components, and function.

No single specialist possesses all the necessary skills to manage diabetic foot syndrome. Therefore, it is important to create a team of specialists with the required skills. Whereas some of the services required to treat CLTI and to prevent amputation can be performed in the outpatient setting, many needed services are intensive and require access to an acute care hospital.

An understanding of the natural history of amputation in diabetes can assist in determining how to build an effective team (Fig 12.1).⁶⁵⁰ Diabetes leads to peripheral neuropathy, although the timing of its onset is related to long-term control of blood glucose level. Peripheral neuropathy leads to unfelt repetitive trauma and in combination with foot deformity causes DFU.⁶⁵¹ Approximately half of these patients have significant PAD with their DFU. Still, more often than not, infection serves as the final event leading up to the amputation.⁶⁵²

Open in a separate window

Fig 12.1.

The elevating risk of the “stairway to an amputation” or the natural history of diabetes-related amputations. (Adapted from Rogers LC, Armstrong DG. Podiatry care. In: Cronenwett JL, Johnston KW, editors. Rutherford’s vascular surgery. 7th ed. Philadelphia: Saunders Elsevier; 2010. p. 1747–60.)

Fitzgerald et al⁶¹⁰ described the seven essential skills for limb salvage teams. These were modified to identify nine skills needed for the comprehensive management of diabetic foot. Table 12.3 lists the essentials skills as well as the type of specialist who should be added to the team to complete a given task. The simplest method to construct a team for a Center of Excellence is to ensure that each of these skills is covered by an expert on the team. In addition, several authors have described an irreducible minimum to the team that includes vascular surgery and surgical podiatry. These two specialties have been nicknamed the “toe and flow” team.^610,649

Team-driven protocols.

It is simply not enough to have a designated team. The team must be used in an effective manner, and outcomes should be monitored in a structured fashion. Fig 12.2 illustrates a useful pathway in setting up the structure of the team, establishing goals, and ensuring that the goals are met. Published CPGs from medical and surgical societies establish best practices, but they are not always feasible for practice in all settings. Current CPGs exist for PAD in diabetes, diabetic foot infections, DFUs, offloading of DFUs, inpatient management of the diabetic foot and the Charcot foot, and prevention of diabetic foot problems.^{158,653–658} Whereas these CPGs can serve as a template, localities are encouraged to create their own clinical practice pathways specific to the facility or system in which they practice.

Open in a separate window

Fig 12.2.

A schematic on how to organize the diabetic foot care within a multidisciplinary team.

The clinical practice pathways are used to identify the team structure and patient flow, when to engage various members, and what to do if the patient is not improving as expected. Policies and procedures are then created to assist providers and staff in complying with the pathway. Quality assurance goals are also created for measurable policies and procedures. Certain outcomes are self-explanatory, such as limb salvage rate, whereas others should be followed to ensure the quality of care delivered by the Center of Excellence. These can include the high-low amputation ratio,⁶⁵⁹ median days to heal for foot wounds, healing percentage, and quality of life measures. Table 12.4 lists the most important measurable outcomes for limb salvage and their calculation. These data may not always be easy to track. Existing electronic health record systems are lacking in their ability to track and to report most of these or other custom measures. Centers of Excellence often resort to developing their own software or keeping track of data manually in spreadsheets.

Finally, performance improvement plans must be drafted and initiated when the quality assurance goals are not met. Fig 12.3 shows an example of how this system would be applied to vascular disease screening in DFUs.

Open in a separate window

Fig 12.3.

An example of using the organized care model for peripheral artery disease (PAD) screening in diabetic foot ulcers (DFUs). CPG, Clinical practice guideline; CPP, clinical practice pathway; P&P, policies and procedures; PI, performance improvement; QA, quality assurance.

Team impact.

In 2005, the World Health Organization and the International Diabetes Federation declared that up to 80% of diabetes-related amputations are preventable.^660,661 Currently, the only intervention to address this has been the formation of multidisciplinary teams to prevent unnecessary amputations. In fact, the multidisciplinary team to prevent diabetes-related amputations dates back to at least 1934, when Elliott P. Joslin, an endocrinologist in Boston, established his team to treat diabetic gangrene.⁶⁶²

In the United States, an organized team in a public hospital reduced lower extremity amputations 72% during 2 years. In the Veterans Affairs medical centers, several factors were significant in the reduction of lower extremity amputations, including use of a specialized team and establishment of a high-risk foot clinic.^663,664 In a military medical center, amputations were reduced by 82% as a result of a specialized limb preservation service.⁶⁶⁵ Another report showed a reduction improvement in diabetes-related foot outcomes with an integrated interdisciplinary team in a large academic medical center.⁶⁶⁶ In several other studies, adding podiatry to the team was found to be helpful in reducing amputations and significantly reducing the cost associated with diabetic foot.^{641,663,667,668}

The impact of a limb salvage team is not limited to any geographic area. In The Netherlands, investigators reported a 34% nationwide reduction in amputations after setting up multidisciplinary teams.⁶⁶⁹ In Brazil, the establishment of >20 interdisciplinary foot clinics nationwide is leading to improved care.⁶⁷⁰ In Italy, investigators reported a reduction in hospitalizations and amputations in the diabetic foot after implementing a multidisciplinary referral team.^670,671 In Spain, a multidisciplinary foot team reduced amputations during 3 years compared with the previous 6 years.⁶⁷² The United Kingdom has also seen reduced amputations secondary to better-organized diabetic foot care with specialized clinics that follow multidisciplinary care pathways and protocols.^673,674 Lastly, in Finland, a decrease in major amputations was correlated with rising interest in limb salvage and an increase in distal vascular procedures.⁶⁷⁵ In a subsequent study, researchers reported a reduction in amputations and length of stay when inpatient care was reorganized.⁶⁷⁶

Definition and classification.

Clinical criteria, history, and examination are the mainstays of CLTI diagnosis across the world, with the use of adjunctive hemodynamic and perfusion measurements appearing to be highly variable. ABI testing was used by all except one respondent. However, although all respondents regularly dealt with diabetic vascular disease and the acknowledged limitations of APs in that setting, only two used TPs; none used TcPO₂ routinely. All (except one who exclusively used WIfI) used either the Fontaine or Rutherford classification for staging, approximately in equal numbers. About one-third of respondents described employing WIfI in addition to another clinical classification system. In summary, therefore, across most of the world, there appears to be limited adherence to any one published definition or staging system for CLTI.

Epidemiology and risk factors.

Although accurate country-specific epidemiologic data are sparse, there seems little doubt that the increasing prevalence of DM (Fig 13.1) together with the growing use of tobacco and population aging is resulting in a significant increase in CLTI and amputations across much of the world, especially in LMICs.⁶⁷⁷

Open in a separate window

Fig 13.1.

International Diabetes Federation global diabetes projections. (From the International Diabetes Federation. IDF diabetes atlas. 7th ed. Brussels, Belgium: International Diabetes Federation; 2015.)

In 2013, Fowkes et al¹ undertook a meta-analysis of 34 studies to compare the prevalence and risk factors between HICs and LMICs. This is well outlined in Section 2 of this document, but it is worth recalling some of the key presented data. They concluded, “Globally, 202 million people were living with peripheral artery disease in 2010, 69.7% of them in LMIC, including 54.8 million in Southeast Asia and 45.9 million in the western Pacific Region. During the preceding decade, the number of individuals with peripheral artery disease increased by 28.7% in LMIC and 13.1% in HIC. Also of note is the percentage of increase of PAD is higher in women than men in LMIC which is opposite of HIC.” The increase in PAD burden observed in women and in the younger, economically productive age groups is especially worrisome (Table 13.2).

The data on country-specific incidence of PAD and CLTI are sparse in these LMICs, unlike in HICs. There are no relevant epidemiologic data from large regions, but the updated data from Abbas are tabulated for perspective, reflecting PAD in diabetics in sub-Saharan Africa (Table 13.3).

Lacking firm epidemiologic data, recent estimates of CLTI prevalence have used extrapolations from demographic and other available disease prevalence data, yielding global estimates of between 20 and 40 million individuals afflicted. About two-thirds of these are projected to be in LMICs. Unfortunately, documented data to support this are difficult to find in any indexed, peer-reviewed journals.

According to the survey respondents, the risk factors for CLTI in their regions are largely as expected, but DM is a predominant cause, more than in HICs. The prevalence reported by respondents varied from 40% to 90%. Interestingly, a cultural preference for walking barefoot or a lack of appropriate footwear is a significant problem in some countries. Approximately 60% to 80% of all the PAD patients seen by the respondents present with CLTI. The average age was around 65 years, and about 70% were men. Most respondents reported that 70% to 100% of CLTI patients presented with tissue loss; in three countries, it was <50%. Primary amputation was performed in 10% to 40% of CLTI patients, this being mainly (25%−90%) because of delayed presentation or referral. Only two countries reported a primary amputation rate of <10%. Postprocedural amputation rates were reported at around 5% to 10%, although two countries reported much higher rates (60%−70%) because of late presentation or aggressive disease patterns encountered.

Diagnostic evaluation.

DUS appears to be used almost universally, although three respondents preferred to proceed directly to other imaging modalities. Only five respondents performed DSA as their primary imaging modality. The remainder opted for MRA and CTA in about equal numbers. In patients with renal impairment, DSA was preferred by most, with half opting for iodinated contrast agents with appropriate renal protection measures and the other half favoring CO₂ angiography. Two respondents performed only noninvasive testing in such patients before intervention.

Medical and noninterventional management (with or without revascularization).

Respondents reported widespread routine use of antiplatelet and lipid-lowering agents. ACEIs, vasoactive drugs (such as cilostazol and pentoxifylline), and anticoagulants were used selectively. IV prostanoids and vasodilators were used by some as adjuncts to revascularization and in those with nonreconstructible disease. Use of arterial assist devices (compression pump), HBOT, and SCS was uncommon. Lumbar sympathectomy was performed by a third of respondents, possible in patients with Buerger’s disease (not specified).

Postprocedural surveillance and follow-up.

All the respondents said they had defined follow-up protocols for patients undergoing infrainguinal revascularization. All patients (surgical and endovascular) are observed at least every 3 months for a year and then at variable intervals thereafter. Clinical evaluation and ABI are the mainstays of surveillance. Use of other noninvasive methods (PVR, DUS) is variable. Specific protocols for vein and prosthetic bypass grafts seem to be standardized per available data in a minority of centers. Approach to surveillance-detected lesions is similarly variable but mostly dictated by the patient’s symptoms rather than by the result of physiologic testing. Arteriography is reserved for clinically significant lesions. Postprocedural drug therapy, for example, with antiplatelet and lipid-lowering agents, appears concordant with current published recommendations. Because most CLTI patients had tissue loss, almost all the centers provided intensive wound services within their department as part of a multidisciplinary team approach. Nearly all agreed that wound infection is a significant determinant of outcome after revascularization and possible cause of amputation even after successful revascularization.

Health economics.

CLTI has a serious adverse economic impact on patients, their families, and wider communities right across the world but especially so in LMICs. Although these countries are often grouped together, the division between middle income and lower income is variable and imprecise. Furthermore, there is often considerable inequality within each LMIC, and respondents reported that most patients with CLTI (30%−90%) appear to come from the poor socioeconomic backgrounds. The following data from the Indian National Sample Survey Office could represent the situation in many LMICs⁶⁷⁸:

1. Only 18% of the urban population and 14% the of ruralpopulation are covered by some form of health insurance.
2. Governmental health expenditure is <2% of gross domestic product overall.
3. People in villages mainly depend on “household income or savings” (68%) and “borrowings” (25%) to fund hospitalization expenses.
4. Around 1% of the poor in rural areas have to sell theirphysical assets to meet health expenditure, and >5% seek help of friends and relatives. This is also in line with earlier studies showing that millions are pushed into poverty each year by medical expenditure and that such expenses are among the leading causes of indebtedness among the poor.
5. In cities, people rely much more on their income or savings (75%) than on borrowings (18%) to fund their treatment. Previous studies have repeatedly shown that India has one of the most privatized health care systems in the world, with out-ofpocket expenses accounting for the bulk of medical spending.

In India, the cost of IP bypass is U.S. $1500 to $3000, and costs of balloon angioplasty are similar. The use of a stent or DCB would add another U.S. $500 to $1000, and wound care adds at least U.S. $500. Such out-of-pocket expenses are probably unaffordable for most CLTI patients. Importantly, these costs depend on recycling of single-use devices like sheaths, angioplasty balloons, and guidewires. Without such practice, the cost would increase by at least 50%, and far fewer patients, especially poorer ones, would have access to treatment, resulting in much greater loss of life and limb. Recycling of single-use devices (not just vascular devices) is common in Asia, Africa, Latin America, and eastern Europe, and proper regulation of the practice, including appropriate consent procedures, is important to mitigate patient harm.⁶⁷⁹

Summary of global perspectives.

Based on the responses to the questionnaire and the limited published and unpublished data at times, we can draw the following conclusions.

1. CLTI is a significant and increasing global problem, especially in LMICs, where the incidence in women appears to be rising more quickly than in men.
2. Diabetes and unabated smoking are the major causes of CLTI globally.
3. Although vascular specialists try to follow the published evidence base, economic and social constraints mean that the approach to CLTI must to tailored to the working environment.
4. CLTI and diabetic foot problems are associated with high amputation rates in LMICs because of delayed presentation and referral and limited access to affordable care.
5. Economic constraints are an important limitation in the adoption of advanced vascular technologies, and practical issues such as recycling of single-use devices require oversight from a public health perspective.
6. Few countries maintain national registries or other CLTI data sets.
7. Most countries do not have a standardized approach to CLTI, with considerable locoregional variation in practice.
8. Most countries do not have well-organized and supported vascular societies where best practice and research can be shared and disseminated.

The Steering Committee wishes to acknowledge the outstanding administrative support of Ms Patricia Burton and Ms Kristin Hitchcock from the Society for Vascular Surgery throughout the course of the global guidelines project.