P (patients)
I (intervention)
C (comparator group)
O (outcomes measured)
Patients with limb-threatening ischmia
Non-operative medical therapies
Amputation
Graft patency, major limb amputation, quality of life, and death
Critical Limb Ischemia
The term critical limb ischemia is actually a misnomer because it implies a specific hemodynamic cut-off and a critical situation; i.e., without urgent intervention, limb loss will inevitably result. Although that may be true for a subset of CLI patients, for many it is not. Paramount to addressing concerns for treatment options is to understand the natural history of the disease [1, 2]. What happens when patients with CLI are not revascularized or have their limb amputated? In the Biancari meta-analysis of 83,000 patients, 20 % of patients were treated without surgery [1]. Of those patients, 1-year survival was 75.4 % and amputation-free survival was 51.4 %. In the BASIL trial, which included patients with “severe PAD” but who did not meet stringent criteria for CLI, at 5 years 37 % had died [3].
In the placebo arm of the Circulase trial, which enrolled patients with hemodynamic and clinical criteria for CLI, the amputation rate at 6 months was only 13 % [4]. With wound care alone in 149 patients with CLI, Marston and colleagues reported a 52 % wound healing rate with a 23 % major limb amputation rate at 12 months [5]. A more recent study from Sweden evaluated over six-hundred patients with diabetic foot ulcers who had either a systolic TP < 45 mmHg or an AP < 80 mmHg and who did not undergo endovascular or surgical revascularization. They reported that 50 % healed primarily with standard wound care or after minor amputation; 17 % healed a major amputation; and 33 % died with limbs intact but with unhealed wounds [6]. These considerations strongly suggest that care for patients with severe PAD and wounds must be individualized and that the term CLI may imply a false sense of doom for many patients. For these and many other reasons, we prefer the terms threatened limb and limb-threatening ischemia.
Surgical Intervention in Limb-Threatening Ischemia: Predictors of Surgical Failure and How to Define Successful Outcomes
Limb-threatening ischemia is a disease spectrum and patients suffering from it comprise a heterogeneous group. Determining treatment for any given patient should be individualized and based on their associated comorbidities and overall goal for mobility. Below, we discuss relevant studies that help define the probability of successful treatment for such patients and associated algorithms. The majority of studies evaluated and discussed here regularly define primary success as a composite of mortality and limb loss ± vessel/bypass patency. Unless otherwise noted, these studies are retrospective analyses with multivariate Cox models to estimate hazards ratios.
Demographic and Patient Risk Factors
In our analysis, 9 relevant studies were included in this section. Overall, the predictors for worse outcome (graft failure, major amputation, poor quality of life, death) were dialysis dependent renal failure, functional dependence, tissue loss and age >80 years. These factors had the highest predictive value and were the most commonly reported factors that contribute to poor outcomes. Other key factors included low serum albumin, current smoking, atrial fibrillation, prior amputation, emergency operation, CHF, CAD, no statin usage, and need for multiple vein segments for bypass. Specific to patients with diabetes and PAD, the prospective EURODIALE trial, which was published prior to our 5-year search window, found that wound size, degree of PAD, ESRD and age were predictive of wound healing in diabetic patients [7].
Multiple groups have developed models to predict surgical outcomes based on a variety of combinations of these characteristics and risk factors. Meltzer et al. evaluated NSQIP data and developed a composite index called a “CRAB” (Comprehensive Risk Assessment for Bypass) score to predict 30-day perioperative mortality and major morbidity (sepsis, MI, stroke, pneumonia, PE, renal failure, graft failure, major wound infection) in patients with PAD undergoing revascularization surgery. This index is based on age, tissue loss, dialysis dependence, functional status, emergency operation and prior amputation status. The calculated scores categorize patients into low-, medium- and high-risk patient groups, which have correspondingly 5, 14 and 25 % risks of 30-day mortality and major morbidity [8]. Another risk assessment tool, the PIII CLI risk score, predicts amputation-free survival (AFS) in patients undergoing infrainguinal bypass [9]. In decreasing order of importance, this system categorizes patients according to dialysis dependence, tissue loss, age and CAD. With these four characteristics patients were determined to be high, moderate and low risk, stratifications, respectively associated with AFS rates of 45 %, 64 % and 88 %. The looming question raised by these data is whether surgery (endovascular or open) would or should be denied to some selected surgical candidates based only on their predicted failure risk.
Wound and Anatomic Risk Factors
Specific anatomic risk factors are also predictive of poor surgical outcomes. As discussed, the EURODIALE trial found baseline ulcer size to be a key outcome predictor in patients with diabetes [6]. More recently, Lida and colleagues retrospectively evaluated 1057 limbs in 884 patients that had endovascular procedures (stent or angioplasty) between 2004 and 2010 with 2-year follow-up. They found that calcified lesions, target vessels <3.0 mm, lesions >30 cm in length and no below-knee run-off were all associated with worse outcomes. High risk-patients by their model had a 70 % chance (about 2X increased probability) of major amputation or reintervention compared to their low-risk group [10].
Non-surgical Treatment Options for Limb Threatening Ischemia
Risk Factor Optimization and Pharmaceutical Treatments
There are many excellent reviews and large randomized controlled trials that define beneficial pharmaceutical practices for treating patients with PAD. In summary, anti-platelet agents, statin therapy, blood pressure control, diabetes control and smoking cessation all have high-quality data that support their efficacy. However, studies that address whether pharmaceutical intervention specifically influences outcomes, particularly wound healing and AFS in “CLI” patients are much more limited.
Statins
Statin therapy was evaluated in a retrospective non-randomized manner in Japan. Tomoi et al. reviewed 812 patients with CLI treated by angioplasty and then discharged on a statin (169) or without (643). Survival, freedom from re-intervention and freedom from amputation were significantly higher in those treated with statins [13].
Cilostazol
A retrospective analysis evaluated 63 patients with CLI 3 months after revascularization with infrapopliteal angioplasty. In the 32 treated with cilostazol (compared to no cilostazol), re-occlusion was significantly lower (20.5 % vs 43.6 %, p = 0.015) [14]. In another small study, Cilostazol was found to significantly increase skin perfusion pressure in patients with CLI [15].
Propionyl-L-Carnitine
Propionyl-L-Carnitine is an over-the-counter anti-oxidant thought to improve endothelial function and microcirculation. One study evaluated 48 patients randomized to placebo vs treatment [16]. They found a significant increase in endothelium responsiveness and reduction in pain in treated versus the control group.
Steroids
The hypothesis that steroids are effective in the treatment has long been shown to be incorrect. A recent meta-analysis included 3 placebo-controlled trials with 109 total patients. They found no benefit for testosterone over placebo [17].
Prostanoids
Prostanoids are a class of bioactive molecules comprised of prostacyclins, prostaglandins and thromboxanes. They signal through autocrine and paracrine pathways to help regulate very diverse pathways including labor and delivery, bronchoconstriction, vasoconstriction, vasodilation, pain and segments of immune and inflammatory pathways including platelet activation and inhibition. Formulations of PGE1 and PGI2 are the two key molecules that have been used and studied the most for treatment of CLI and PAD. PGE1 is a vasodilator that is used to treat erectile dysfunction, maintain patent ductus arteriosus and promote peripheral vascular vasodilation. PGI2 inhibits platelet activation and has vasodilatory effects. Enthusiasm in many parts of the world for the use of prostanoids in CLI has been great, but the supporting data are limited. Ruffolo and colleagues published a recent meta-analysis that included 20 RCTs and found moderately favorable results for ulcer healing (RR 1.54) and pain relief (RR 1.32) [18]. A more recent meta-analysis evaluated prostanoids in PAD and included 2773 patients from 18 RCTs [19]. They concluded that data were poor in quality, and although some studies showed moderate benefit, side-effects (headaches, nausea, pain, diarrhea) led to patient withdrawal, and no definitive evidence supported using prostanoids in patients with PAD. The Scottish-Finnish trial, the most recently published placebo-controlled RCT, was felt to be adequately powered with 111 patients, but could not detect a difference between ulcer healing or amputation rate at 6 months between the treatment arm receiving taprostene (PGI2) versus placebo [20]. These data echoed the earlier Circulase trial, which found no benefit for lipo-ecraprost (PGE1); the latter study was terminated after 383 patients had been enrolled because interim analysis showed no benefit [4].
Wound Care
As previously discussed, patients with critical limb ischemia are not all critical. In addition, the term was never intended to be applied to patients with diabetes, who now constitute the largest group of patients presenting with threatened limbs due to the global epidemic of diabetes. Local wound care alone (albeit with significant effort and at a facility that can provide a high level of care) results in approximately a 50 % wound-healing rate at 1 year and a major amputation rate of only 25 %. These data are reinforced in retrospective reviews as well as the placebo arms of various studies cited above. These numbers may seem surprisingly positive, but are difficult to generalize because patient populations and severity of CLI within these and other studies are not uniform. The SVS WIfI (Wound, Ischemia, foot Infection) lower extremity threatened limb classification should help better stratify amputation risk and allow more meaningful inter-study comparisons through a better patient classification system [21]. The underlying premise of SVS WIfI classification is that amputation risk increases as the disease burden or limb stage presentation progresses from stage 1 (very low risk) to stage 4 (high-risk). Ischemia is one component of this system, along with wound size and complexity, and the presence and severity of infection.
Cell Therapy
Cell therapy for patients with CLI is a promising field for clinical and basic science research. The treatment strategy is to harvest autologous stem cells and deliver them back to patients; the goal of the therapy is the development of improved and new blood vessels in addition to wound healing in diseased limbs. We reviewed 22 studies published in the past 5 years and reference the largest 5 studies with 50–100 patients per study [5, 22–25]. All 22 studies had encouraging results, although the data from these studies are highly heterogeneous. Various cell types are used in the treatment; multiple mechanisms for harvesting cells are employed; delivery back to the patient is intravenous, intramuscular or intra-arterial. Because of this heterogeneity, the current meta-analyses on the topic are limited and add little more than the cumulative picture of the overall positive results in several independent small studies.
After evaluating 22 new trials combined with published meta-analyses, we believe these data and the various modalities employed in cellular treatment strategies are the most promising area of novel treatments for patients with limb-threatening ischemia and no surgical options (and should likely expand to those with surgical options). Important questions remain to be answered: Which bone marrow cell type and which harvesting and delivering modality are most beneficial? A key impediment to answering these questions and, more importantly, expanding the use of cell therapy is the availability of technology and expertise.
Gene Therapy
Gene therapy has used various strategies to deliver genetic material that can be translated into pro-angiogenic gene-products (VEGF, FGF, HGF, Del-1, HIF-1). There are two recent relevant meta-analyses. Hammer and colleagues evaluated 1494 patients from 12 RCTs [26]. They found no benefit or harm for gene therapy for all-cause mortality, amputations or wound healing. Another meta-analysis with slightly different criteria included 543 patients from 6 RCTs found no benefit for amputation rate, but found an increased rate of non-serious adverse events (edema, hypotension and proteinuria) [27]. These data did show however, that serious adverse events such as mortality, malignancy and retinopathy did not increase in treated patients compared to placebo-controlled patients. Similar to evaluating data from cell therapy studies, each study had slightly different delivery strategies (viral, plasmid, naked DNA) and different angiogenic factors, greatly undermineing the validity of the meta-analyses. Six additional studies were completed in the past 5 years that are small in number and add no more conclusive evidence.
Early studies in this field of research found limited benefit and significant side effects for these pro-angiogenic gene therapies, which has mitigated enthusiasm for pursuing this avenue of research. However, there are ongoing trials pursuing different targets with similar treatment strategies. The majority of new studies are focusing on delivery of HGF or FGF via plasmid-based systems. All have very small numbers of patients and show mildly encouraging to equivocal results. The logic of genetically promoting blood vessel development is enticing and sensible. However, these data indicate that our basic science knowledge of how to achieve that end is either inadequate or is not being employed.
Spinal Cord Stimulation
Spinal cord stimulation (SCS) consists of an implantable device that delivers pulsed electrical signals to the spine via probes inserted into the epidural space. A recent meta-analysis included 450 patients from 6 trials evaluating spinal cord stimulation (SCS) for CLI patients. Limb salvage after 12 months was significantly higher, decrease in Fontaine score significantly better and pain was significantly reduced in the intervention arm. However, re-implantion was required in 15 % of patients secondary to complications and infection occurred in 3 % [28].
These data are not randomized or controlled. As such, even the moderate benefits seen garner little support to recommend SCS. Compared to the risks, SCS implantation is not supported.
Hyperbaric Oxygen
The theoretical benefit of hyperbaric oxygen therapy (HBOT) is that under increased pressure and with a higher concentration of oxygen in the air, more oxygen should be delivered to blood-deprived tissue. The evidence for the effect of HBOT on PAD or CLI is of low quality and inconsistent. Kranke et al. published a meta-analysis in 2012 and were able to combine 118 patients from 3 RCTs [29]. They found HBOT significantly prevented major amputations at 1 year (RR 0.31, CI .13–.71), but had insufficient data regarding wound healing. A more recent meta-analysis of two adequate RCTs evaluating HBOT in patients with ischemic diabetic foot ulcers reported a non-significant improved ulcer healing at 1 year but no difference in amputation [30]. A recent RCT that included 75 patients with diabetic foot ulcers showed the beneficial effect of HBOT diminished with worsening ABI, toe pressure and TcPO2 [31]. Those with TcPO2 < 25 mmHg on the dorsum of foot had no benefit. These data could explain the conflicting data from multiple previous trials; as the patient populations had increasingly severe PAD, outcomes were worse.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
