Introduction

As technologies and research into the treatment of cardiovascular disease and peripheral vascular disease (PVD) advance, applications of treatment regimens not only reduce morbidity and mortality, but they also can lead to undesirable, unintentional consequences—the development of a population of patients who no longer respond to the therapies that initially prolonged their lives, increased their functional capacity, or increased their quality of life. Neuromodulation via spinal cord stimulation (SCS) may be a partial answer to the challenge being faced for patients with refractory angina (RA) or irreparable PVD. It has not been Food and Drug Administration (FDA)–approved for use in either RA or PVD as of yet in the United States. Originally used to treat a patient with cancer pain by Shealy, Mortimer, and Reswick in 1967,¹ SCS therapy has been extended to many other areas with varying amounts of success. Not everyone is a candidate for this technique. But appropriately screened patients who meet appropriate diagnostic and psychological criteria may benefit from SCS, which may alleviate chronic pain while adding an anti-ischemic benefit so patients’ quality of life and functional capacity improve.

This chapter discusses RA in its first section and PVD in its second. After the section on epidemiology, each section contains these parts: establishing diagnosis, anatomy, indications, basic science mechanisms, guidelines, equipment and techniques, outcomes, and complications. Ideally this chapter will provoke further thought and research into this novel application of neuromodulation, leading to eventual FDA approval for using SCS in RA and expanded use in PVD.

Epidemiology: Coronary Artery Disease and Peripheral Vascular Disease

Coronary artery disease (CAD) remains the number one cause of morbidity and mortality in the United States, yet the prevalence of RA in the population has not been well-defined in the literature. Several estimates have been proposed. According to the American Heart Association’s Heart Disease and Stroke Statistics—2010 Update, approximately 10.2 million people suffer from angina pectoris. Currently the annual incidence is estimated to be approximately 500,000.² In 1999 Mukherjee and associates³ attempted to estimate the incidence of RA. Their approximation of 12% of the total population of those with angina is based on the percentage of people undergoing angiography at tertiary referral centers who ultimately are not eligible for percutaneous or coronary artery bypass graft (CABG) revascularization procedures.³ This percentage is also endorsed by Mannheimer and associates,⁴ who remarked on the paucity of data but noted that one study they reviewed had a prevalence of 5% to 15% of the population as having RA. In 2002 Holmes⁵ noted that approximately 2.4 million people in the United States suffer from CAD untreatable by either percutaneous revascularization or CABG. With a large and ever-expanding population, the demand for novel treatments will certainly climb. This demand will undoubtedly be mirrored in the population that suffers from PVD.

PVD likely affects more than 10 million people in the United States according to Vallejo et al estimates in 2006, and it affects 12% to 20% of the population aged 65 and older.⁶ Of the people who develop PVD with signs of intermittent claudication, approximately 20% are believed to develop chronic critical limb ischemia (CLI); 25% of CLI sufferers require an amputation.⁷ The mortality for this population also is quite high, ranging from 25% to 30% at 2 years and increasing to 50% to 75% at 5 years after the onset of CLI. Although the amputation rate and mortality are significant in this subpopulation, the implementation of SCS shows promise in alleviating some of the suffering and morbidity and mortality of this disease.

Establishing Diagnosis: Angina and Role of Spinal Cord Stimulation

An imbalance between myocardial oxygen delivery and its consumption rapidly results in ischemic conditions within the heart that may progress to infarction if not corrected. This imbalance may or may not be associated with symptoms of angina pectoris because some patients develop silent ischemia. Besides being a harbinger of a potentially serious condition, angina not only presents significant acute or chronic discomfort, but it also typically causes psychological stress that may impact a person’s quality of life and functionality.

As defined by the American Heart Association in 1999,⁸ angina comprises a clinical syndrome of pain and discomfort in the chest, jaw, shoulder, back, or arm that may be exacerbated by physical exertion or emotional stress. Although angina typically is associated with CAD involving the epicardial vessels and subsequent ischemia, a more robust differential diagnosis of angina must be considered from both a cardiac and noncardiac standpoint. Cardiovascular origins of angina may also include valvular heart disease, severe hypertension, hypertrophic cardiomyopathies, acute aortic dissection, acute pericarditis, severe aortic stenosis, coronary vasospasm, and cardiac syndrome X (CSX).^8–10 CSX patients present similarly to those with CAD-related angina. Although patients with CSX tend to feel pain with exertion and their electrocardiograms (ECGs) may show ST-segment depression during exercise stress tests that induce angina, these patients do not have signs of obvious CAD on angiography; in fact, in some patients, evidence suggests that ischemia may not be a causative factor.¹¹ Proposed mechanisms for CSX include estrogen deficiency, abnormal function and distribution of adenosine receptors, and coronary microvascular dysfunction.^9,11,12 Noncardiac origins of chest pain include trauma and esophageal conditions, including reflux and motility disorders, biliary colic, costochondritis, and pulmonary embolism or pulmonary hypertension.

In addition, angina may be classified as stable or unstable. Stable angina typically arises during physical or emotional stress secondary to severe stenotic lesions affecting more than 70% of the affected coronary artery lumen of one or more arteries that causes myocardial ischemia.¹³ Conversely, unstable angina involves acute formation of a thrombosis within an already stenotic coronary vessel, which may or may not immediately lead to a myocardial infarction (MI). Unstable angina is of greatest concern since it predicts an elevated short-term risk of a cardiac event⁸ and necessitates immediate coronary revascularization to alleviate. Unstable angina is beyond the scope of this chapter; however, episodes of unstable angina pain manifest even in patients with spinal cord stimulators for RA, as discussed in the following paragraphs.

To improve clinical classification of angina, the Canadian Cardiovascular Society (CCS)^8,14 modified the New York Heart Association’s (NYHA) Functional Classification (Box 11-1). Campeau and Letter¹⁴ developed four classes—I through IV. In class I angina patients perceive angina only with strenuous activity. Class II patients experience only slight limitation of activity secondary to angina. Class III patients experience marked limitation of normal activity, and in class IV they experience angina with any activity, including being at rest. MI falls within classes III and IV.

RA is marked by severe chronic chest pain for more than 3 months secondary to coronary insufficiency; it occurs in patients who have failed to obtain or undergo appropriate control via other modalities, including medical therapy, percutaneous revascularization, and CABG, yet who continue to have a reversible ischemia.^9,15,16 RA occurs when all reversible causes for ischemia have been ruled out. SCS has been used successfully to target RA—decreasing the frequency and severity of the episodes and improving patient functionality and quality of life.

Anatomy: Pain Pathway for Angina

Signals for angina pain are initiated at both the chemosensitive and mechanoreceptive nociceptors in the adventitia of the coronary arteries and myocardium.¹⁶ Because the majority of these nerves are slow-conducting C fibers, the predominant pain experienced is of a dull, aching, heavy, and squeezing type.⁸ Aδ fibers that carry stabbing and sharp pain are typically not involved in angina. On their activation these nociceptors release a variety of chemical mediators, including adenosine, bradykinin, prostaglandins, and others, which initiate signals in the sympathetic and parasympathetic (vagal) afferent pathways to dorsal spinal cord and parasympathetic ganglia located from C7 to T5.^17,18 The pain experienced by the patient during an episode of angina is related to the convergence of common pathways at the dorsal spinal cord between C7 and T5, where afferent myocardial inputs and cutaneous nociceptors converge on the same interneurons at the same level within the spinal cord.¹⁶ Thus the pain perceived by the patient is distributed within the dermatome from where the cutaneous afferents converge on the same spinal segment as from the heart.

Indications: Spinal Cord Stimulation and the Alternative Therapies Available for Refractory Angina

As previously mentioned, neuromodulation has treated RA successfully; however, before choosing SCS as a treatment modality, the patient must be properly assessed, and all other modalities pursued. SCS provides analgesia and anti-ischemic effects, but it does not cure advanced cardiovascular disease. This section briefly reviews therapies for angina and establishes a framework within which SCS should be considered as an alternative.

As medical, interventional, and operative treatment modalities for angina and occlusive vascular disease have developed along with an aging population, so has the portion of the population with angina that lacks viable options for further improvement, despite optimization from one of the novel or improved therapies. This subgroup contains patients with RA who typically have failed either percutaneous coronary interventions (PCIs) or CABG, or who have not been candidates for either procedure because of poor coronary anatomy, prior surgical repairs not amenable to further manipulation, impaired left ventricular function, co-morbid noncardiac disease compounding their cardiovascular status, or advanced age.^19,20 Jolicoeur and associates¹⁹ also mentioned that a person with RA may fall into the category of “nonrevascularization” secondary to either their geographical location where practitioners may not have the expertise or the patients lack of interest in pursuing a particular therapeutic path.

Multiple pharmacological and nonpharmacological (mechanical intervention) therapies for RA are reviewed in the following paragraphs. The reviews are based primarily on Jolicoeur and colleagues’ report of the working group on clinical and research issues regarding chronic advanced CAD.¹⁹ Although their reviews are not comprehensive, they highlight the current management options for RA.

Pharmacological agents mitigate angina symptoms by a variety of mechanisms so the patient has reduced myocardial oxygen demand and increased supply via a reduction in heart rate (HR), decreased afterload, and decreased contractility. Agents with antianginal properties that are helpful in accomplishing these changes include β-blockers, nitrates, calcium channel blockers, and opioids. Each provides angina relief via different mechanisms, but each is not without side effects that lead to dose limitation or intolerance. Antithromboembolic agents such as aspirin and clopidogrel (Plavix) target the platelets; whereas antihyperlipidemics such as statins promote vessel patency, thus decreasing risk for myocardial ischemia and resultant angina. Ranolazine (Ranexa) is a relatively new drug with direct antianginal and anti-ischemic properties. A debate as to its mechanism of action exists; however, it permits an increase in exercise performance without changing HR or blood pressure.²¹ In addition, ranolazine improves angina and exercise threshold when used as a monotherapy and in combination with more traditional therapies listed previously.^21,22

Mechanical interventions are also available for RA: percutaneous stent placement, transmyocardial laser revascularization, and CABG. Despite these interventions, people continue to have angina symptoms, and some people are not candidates for such therapies. Chronic total occlusion (CTO) recanalization via the percutaneous approach has been conducted after the development of chronic occlusions (greater than 3 months) with a substantial improvement in 10-year survival.¹⁹ The reported success of revascularization achieved by CTO recanalization remains a stable 71% despite attempts at canalizing longer lesions, whereas reported complications of the procedure range from 3.8% to 5.1%.^23–25 Technical advances may eventually make this technique more effective and safe.¹⁹

Another mechanical technique sometimes used is enhanced external counterpulsation (EECP), approved by the FDA in 1995 for angina. In EECP three bilateral lower extremity cuffs are inflated sequentially during diastole to provide increased venous return and diastolic augmentation.²⁶ Currently this technique requires 35 consecutive days of 1-hour sessions to show some lasting benefit of decreased angina symptoms and a longer time to greater-than–1-mm ST depression on an exercise stress test.²⁶ EECP has not been studied in a large, randomized controlled setting; and it has side effects, including edema, bruising, and pain in the lower extremities. According to Soran and associates,²⁷ patients undergoing EECP noted a substantial improvement in the quality of life; 72% improved from severe to mild or moderate angina, 52% stopped nitroglycerin use, and at 2 years 55% of the patients reported a maintained decrease in their angina.

Before choosing SCS as a treatment modality for RA, the patient must be properly assessed according to various diagnostic algorithms, ideally ones that have been standardized. One such algorithm for assessing a person’s appropriateness for SCS was proposed by the European Society of Cardiology.²⁰ First a team of cardiologists and cardiac surgeons should determine if a patient’s angina is of ischemic origin and evaluate him or her for revascularization. A recent angiogram should be used to rule out newly treatable coronary pathology. In addition, all other forms of chest pain should be eliminated. For example, the differential diagnosis of chest pain should include noncardiac origins of chest pain, including esophageal pain, gastroesophageal reflux, musculoskeletal pain, costochondritis, anemia, uncontrolled hypertension, atrial fibrillation, and thyroid disorder.^19,20 Once these criteria have been met, the patient’s medical therapy must be optimized. After an appropriate diagnosis has been established and subsequent medical and surgical optimization has occurred, the focus shifts to psychological evaluation to determine if the patient’s perception of pain has a significant co-morbidity of depression or anxiety that should be treated before SCS implantation or if the person should be evaluated on the basis of compliance with other treatments. Finally, risk factor management and cardiac rehabilitation must continue.

Basic Science: Mechanism of Action of SCS for Angina

SCS for angina enables both antianginal and anti-ischemic effects to be integrated to achieve pain relief. The severity of the angina decreases; and cardiac function improves, as evidenced by the following: decreased need for short-acting oral nitrates and 24-hour cardiac monitoring and markers of functional status and increased perceived quality of life.²⁸ Multiple mechanisms have been proposed to explain these effects, a number of which are outlined here. note: What follows is not an exhaustive review of the literature on this topic.

Given the benefits of the antianginal and anti-ischemic effects of SCS, one should also consider its safety for patients with RA. An early concern was that SCS could mask angina pain associated with myocardial ischemia and eliminate a patient’s subjective sensation of RA. This concern has not been borne out in several studies.²⁸ Instead, SCS merely raises the patient’s threshold for sensing angina, thus permitting the patient to increase exercise capacity. Investigators have not noted any resultant increase in morbidity and mortality with SCS.

The following mechanisms have been suggested.

Guidelines: Patient Selection for Spinal Cord Stimulation in Refractory Angina

Careful consideration of a patient’s candidacy for implantation involves a multidisciplinary approach to assess the patient’s cardiovascular status and other issues typically faced with SCS implantation. The evaluation team must include at a minimum the following specialists: a pain physician and/or a neurosurgeon, an anesthesiologist, a cardiologist, and a psychologist.^28,51 De Vries and associates²⁸ suggest a series of inclusion and exclusion criteria to determine patients’ suitability for SCS in RA (Box 11-2).

The cardiologist’s evaluation is the most critical one, because the pain physician should operate under the assumption that the angina is not refractory until proven otherwise. The cardiologist evaluates the patient’s cardiac status to determine whether the patient suffers from RA or a nonoptimized cardiac condition. If the cardiologist finds the patient to be an SCS candidate, he or she can also comment on the patient’s cardiac stability in terms of blood pressure, ejection fraction, and other parameters that aid the anesthesiologist in determining the risk and tolerability for the patient to undergo anesthesia. The cardiologist’s other significant contribution is to optimize the patient’s drugs that impact bleeding, including aspirin, warfarin (Coumadin), and medications that alter platelet function⁵¹ (Table 11-1). Use of blood thinners must be stopped secondary to the risk of surgical bleeding and, more important, the risk of epidural hematoma that may lead to paralysis. The cardiologist must determine the feasibility of stopping these medications for the period of device implantation. The patient may remain on baby aspirin but should be off warfarin for a minimum of 3 days with an international normalized ratio (INR) check before the time of trial or implantation. Clopidogrel and associated drugs should be stopped for 10 days. If there is a question about continued alteration of platelet function caused by clopidogrel, a platelet function assay should be considered. Because placement of an SCS device is an elective procedure, the injunction, “Do no harm to the patient,” takes precedence. If taking a patient off these medications causes excessive risk, the procedure should not be attempted. If it is a concern that the patient is off medications, he or she may be a candidate for hospital admission for initiation of heparin therapy, which may be stopped 6 hours before surgery and restarted 6 hours afterward. Low–molecular weight heparin should be avoided secondary to the increased risk of epidural bleeding after SCS lead placement for patients on this therapy.

Table 11-1 Drug Recommendations

INR, International normalized ratio; NSAID, nonsteroidal antiinflammatory drug.

From Deer, T.R (ed): Spinal cord stimulation for the treatment of angina and peripheral vascular disease. Curr Pain Headache Rep 2009, pp. 20.

Even if a patient is cleared from a cardiac standpoint, the implanting physician must evaluate him or her to determine tolerance for such a device, comprehension of its function, and understanding of how to use the available programs to self-treat. If the patient fails to understand the purpose and functionality of SCS, even after appropriate training, he or she should be excluded from implantation. Also important are the patient’s expectations about pain relief. To avoid unrealistic expectations and eventual failure of SCS therapy, the physician should point out that the SCS technique may reduce but probably not eliminate the pain. In addition, a psychological evaluation is imperative to elucidate personality disorders or psychological barriers to having or using such a device.

Equipment and Technique for Spinal Cord Stimulation in Refractory Angina

With implantation of SCS in any patient, the most critical step is placement of the SCS leads at the proper level and location to achieve paresthesias over the area the patient experiences the angina. Typically the tip of the stimulator lead should be placed, under fluoroscopic guidance, at the C7 level in the dorsal epidural space since the afferent fibers are located from approximately T1 to T4. Most successful placement involves stimulation between T1 and T2.^46,48,52 For percutaneous placement the 14-gauge Tuohy needle usually used for insertion should be introduced at approximately the T4-T5 to T6-T7 levels, depending on the patient’s anatomy. The further down the vertebral column a needle is introduced, the more difficult it may become to advance the percutaneous stimulator lead. Because the most common cause of failure is improper position of the stimulator electrode,^9,51 care should be taken to direct one stimulator lead slightly to the left of midline and one at midline to capture most of the angina pain.

Other considerations exist. One is whether the patient has a pacemaker; an automatic implantable cardioverter-defibrillator (AICD) is a contraindication to SCS. SCS compatibility with the pacemaker should be confirmed before implantation, and the pacemaker should be interrogated before and after implantation of the final device. Because of the acuity of this patient population, another consideration is how to trial the patient. If the patient requires titration on and off of blood thinners and/or has difficult lead placement, he or she may be a candidate for implanting trial leads so the leads don’t have to be reintroduced if the trial is successful; instead the patient may go directly to generator placement. Deer⁵¹ points out that no outcome studies have been published to elucidate a better trial technique.

Another consideration is careful review with the patient of pocket placement before implantation. Available sites include the low back above the beltline, buttock, lateral abdominal wall, and anterior chest wall on the right side, opposite to where a pacemaker would be placed. Patient involvement in the selection of the location is critical, not only to promote a sense of autonomy, but also because the patient’s level of function and occupation may suggest the optimal location. From a practical standpoint an object of similar size and shape may be used to demonstrate the future presence of the device at a location and aid the patient in the decision process.

Outcomes for Spinal Cord Stimulation and Refractory Angina

Since SCS was introduced in 1987 as a treatment modality for RA, no large or long-term randomized controlled clinical trials have been conducted to examine its efficacy, likely because of the cost. Instead, several observational, retrospective, and small randomized controlled trials (RCTs) have been done, each measuring different primary and secondary endpoints, excluding any common standardized outcome measures. In addition, as Mannheimer and associates⁵³ pointed out in their study, patients enrolled for SCS caused by RA must be considered in relation to the medical knowledge and technology at that time. Studies conducted over too long a period potentially lead to differences between early and late enrollees secondary to advancing surgical, interventional, and anesthetic techniques.⁵³

Most of the outcomes literature comes from European studies, which focused on the following measures: exercise capacity, ischemic burden, nitrate drug consumption, functional class of angina, health-related quality of life, adverse events and SCS-related complications, cost, reduced frequency and severity of angina, change in pain scores, sleep, and morbidity and mortality. To put many of these outcome measures into perspective, Taylor and associates⁵⁴ prepared a systematic literature review and meta-analysis of SCS use for RA as of February 2008. Following the methods outlined in the Cochrane Handbook for Systematic Reviews of Interventions, the team ultimately examined 11 papers involving seven RCTs with similar inclusion criteria: all patients developed RA despite medical optimization, they were not candidates for revascularization, and they were included in NYHA angina classes III and IV. All seven trials were short term, except for the electrical stimulation versus coronary artery bypass surgery study (ESBY) in severe angina pectoris conducted by Mannheimer and associates⁵³ and the open label, single-centre, randomized trial of spinal cord stimulation versus percutaneous myocardial laser revascularization in patients with refractory angina pectoris (SPiRiT trial) conducted by McNab and associates⁵⁵ that followed up beyond 1 year.⁵⁵

Taylor and colleagues⁵⁴ examined seven specific outcome categories across the seven RCTs: exercise capacity, ischemic burden, nitrate drug consumption, functional class of angina, health-related quality of life, adverse events and SCS-related complications, and cost. The RCTs reviewed included one that examined SCS implantation vs. no implantation conducted by de Jongste et al in 1994.⁵⁶ Four of the RCTs involved an SCS on state vs. an SCS off state with all patients in the study being implanted.^52,57–59 The ESBY study compared CABG to SCS.⁵³ The SPiRiT trial examined percutaneous myocardial laser revascularization vs. SCS.⁵⁵ Much of the following discussion is based on Taylor and associates’ findings with respect to these investigations. Adverse events and SCS-related complications are addressed in a different section, and cost is examined last within this section from the perspective of investigators other than Taylor and colleagues.

According to Taylor and associates, investigators in six RCTs studied exercise capacity. However, investigators in only three RCTs found statistical significance between baselines and follow-up with the SCS out of four studies that looked at change in exercise capacity within the SCS group. The authors pointed out that in the fourth, the ESBY study, the investigators turned off the stimulator while the patients were undergoing evaluation for exercise capacity; thus the patients did not demonstrate any significant change in work capacity. They stated that a pooled analysis demonstrated an improvement in exercise capacity in the SCS-on vs. SCS-off state (p = 0.03) but noted no difference when compared to CABG or percutaneous myocardial laser revascularization (PMR) to SCS and exercise capacity.⁵⁴

Next the authors evaluated ischemic burden in four of the studies they examined. This burden was measured in these studies by evaluating the patient’s 24- or 48-hour ECG monitoring for the frequency and magnitude of ST depression.⁵⁴ No statistical difference with regard to ischemic burden was noted between SCS and CABG (p = 0.44), and only a trend favored SCS when compared to no SCS or nonoperating SCS (p = 0.12).

When the authors examined the third category, nitrate consumption, they reported that three RCTs individually noted a statistically significant reduction in short-acting nitrate use when comparing SCS patients to controls. However, when pooling the data, this decrease in nitrate consumption within the SCS group was not borne out. Furthermore, when they examined 6-month and 2-year follow-up data, Taylor and associates reported no difference in short-acting nitrate consumption between the SCS and the CABG groups.

Only two RCTs evaluated a change functional class of angina by using the CCS angina classification (similar to NYHA). At 3 and 12 months, McNab and associates⁵⁵ reported an improvement in CCS class (lower) of p = 0.049 and p = 0.093, respectively, when compared to patients receiving PMR. But a decrease in CCS class was only statistically significant at 3 months.⁵⁵ The other study by Eddicks and associates⁵⁸ was statistically significant with p = 0.002, with a mean decrease in CCS functional class of 1.6.

Quality of life was ascertained by five of the studies analyzed by Taylor and associates. Four of the studies demonstrated improved quality of life when comparing the SCS to either the no implant state or the SCS off, but no difference when compared to either CABG or PMR.

Taylor and colleagues concluded that SCS is a viable option for patients with RA citing the significant improvement in both exercise capacity and health related quality of life in most of the RCTs they evaluated.⁵⁴ Although they noted that none of the parameters differed tremendously from CABG or PMR, it must be pointed out that these techniques have not been very beneficial for patients with RA. Not one of the randomized controlled trials in the literature use change in pain (any scale) as a primary outcome measure. Instead, all primary outcomes are linked to either exercise capacity in five studies, and ischemic burden in one study. The focus on these outcome measures points to the importance of SCS use for RA to improve primarily function and quality of life as opposed only to pain.

Before the review by Taylor and associates,⁵⁴ the American College of Cardiology and American Heart Association 2002 Guideline Update on Chronic Stable Angina classified SCS as class IIb (usefulness/efficacy is less well established by evidence/opinion).⁶⁰ Up to the time the guidelines were written, only two small RCTs had been conducted. Since that time, no multiple large-scale RCTs have been conducted demonstrating intermediate and long-term benefit that would elevate the use of SCS to level IA data.

A recent review by Simpson and associates⁶¹ out of the United Kingdom evaluated clinical and cost-effectiveness of SCS for both neuropathic and ischemic pain. Their review uses four of the same studies analyzed by Taylor and associates. From a clinical standpoint Simpson and associates come to a similar conclusion. In addition, they went on to assess the cost-effectiveness of SCS as a treatment. Because the paucity of data available to determine comparative efficacy to other treatments, including medical management, CABG, or PCI, Simpson and associates⁶¹ conducted a threshold analysis to determine what the cost benefit would be with the choice of SCS instead of one of the other treatments. Each analysis suggested that each comparison examined would require the patient receiving SCS to live longer to achieve benefit for a life year group. Besides a lack of evidence for comparison, the comparison does not take into account the fact that the suggested use of SCS is for patients who would not be candidates for one of the other interventions on account of the fact that their RA is not amenable to other such treatments.

In addition, several studies by investigators other than Taylor and colleagues have demonstrated cost-effectiveness with the use of SCS. In a study by Rasmussen and associates,⁶² use of the device was shown to save approximately 30% on medical costs because of decreased invasive testing alone. In addition, in a retrospective study by Yu and associates⁶³ the cost of the total SCS procedure was recuperated within 16 months, which they point out was less than 40% of the life span of the device as of 2004. Murray and associates⁴⁹ noted that rehospitalization and length of stay after SCS when compared to revascularization was significantly lower (p = 0.002). All of these studies have been observational or retrospective; none have been prospective RCTs.

After approximately a decade of use, a study by TenVaarwerk and colleagues⁶⁴ demonstrated in a retrospective clinical outcome study that the use of SCS in RA did not lead to any increase in the rate of adverse events when compared to populations that did not undergo SCS implantation. This retrospective study evaluated 517 patients by questionnaire at 14 centers in Europe. They concluded that the patients died from either events unrelated to heart disease or coronary events linked to lower ejection fraction and the severity of CAD, not the presence of the SCS. In addition, when compared to patients receiving medical management, there was no change in mortality.

Complications for Spinal Cord Stimulation and Refractory Angina

Whenever SCS is implanted, several complications are possible—from the most common, lead migration, to a variety of others, including loss of paresthesia, electrode failure, premature battery exhaustion, and infection. Buchser and Durrer⁶⁵ stated that the combined rate of complication is 6.8%. Simpson and associates⁶¹ reported that four studies totaling 403 patients showed a 1% incidence for device removal secondary to infection and a 5%- to 38%-range for device complications in studies with follow-ups greater than 2 months. They suggested such a range for device complications because of the variable sizes of studies, different follow-up periods, or clinical circumstances.

Establishing Diagnosis: Peripheral Vascular Disease and Role of Spinal Cord Stimulation

PVD most commonly develops secondary to advancing atherosclerosis; however, a variety of other conditions lead to the final common pathway, but for brevity’s sake they are not discussed here.⁶⁶ If left unchecked, PVD progresses to CLI and eventual infarction secondary to a decrease in blood flow rate necessary to maintain tissue metabolic function. Patients suffering from PVD typically first notice symptoms of vascular claudication or intermittent pain with ambulation that resolves at rest. If left untreated, the disease progresses to infarction and gangrene, ultimately requiring amputation if other measures fail.

Establishing a CLI diagnosis is critical for selecting patients for various therapies. The second European consensus conference on chronic CLI recommended two ways to identify and define CLI. First, in patients with and without diabetes CLI may be defined by either of two criteria: (1) persistently recurring ischemic rest pain requiring regular adequate analgesia for more than 2 weeks, with an ankle systolic pressure ≤50 mm Hg and/or a toe systolic pressure of ≤30 mm Hg; or (2) ulceration or gangrene of the foot or toes, with an ankle systolic pressure ≤50 mm Hg and/or a toe systolic pressure of ≤30 mm Hg.⁷ Second, the conference recommended a more precise way to identify and define CLI (i.e., use angiography to identify large-vessel disease, take toe arterial pressure, and use methods such as transcutaneous partial pressure of oxygen (TcpO₂) to delineate local microcirculation.⁷ It is important to note that not every patient with CLI falls within these definitions, despite having severe disease. Best clinical judgment must be used to evaluate all patients for extent of disease.

For those patients with concomitant diabetes, a few important factors must be considered during patient evaluation. First, because patients with diabetes may also suffer from painful diabetic neuropathy, their neuropathic pain should be distinguished from their CLI pain at rest. In addition, as the consensus conference points out, because patients with diabetes may have falsely high ankle systolic pressures, pressure readings with a toe cuff should be taken.⁷ Patients with diabetes may also lack palpable pulses or be so weak that observer variability may affect diagnosis. Thus the consensus conference suggested evaluating absolute pressure at the ankle or toe as a more appropriate method of determining extent of disease than the standard ankle/arm pressure index.

Two classification systems, Fontaine stages and Rutherford categories, have been developed to help determine the extent of CLI in PVD (Table 11-2). The Fontaine classification helps assess limb ischemia and the severity of PVD. In stage I the patient is asymptomatic; in stage II the patient demonstrates intermittent claudication; in stage III the patient develops pain at rest; and in stage IV the patient has evidence of tissue loss, including ulcers and gangrene, and pain at rest.^10,67 Fontaine stages III and IV (Rutherford categories 4, 5, and 6), along with the associated blood pressure criteria noted later in the paragraph, define CLI.^28,66 Although these definitions help categorize the extent of CLI, they do not aid in prognosis for the affected limb. According to De Vries and associates,²⁸ no consensus currently exists as to how to most accurately determine prognosis. Patients with Fontaine stages III or IV disease have affected macrocirculation and microcirculation, each of which is evaluated differently. According to Jacobs and Jorning, ⁶⁸ macrocirculation is best evaluated by systolic ankle/arm pressure measurements at rest and after treadmill exercise. Microcirculatory (cutaneous) blood flow may be evaluated by tissue oxygen pressure measurement, laser Doppler flowmetry, and radioisotope clearance.^28,68 Ubbink and associates⁶⁹ evaluated prognostic capabilities of these methods. They determined that microcirculatory classification predicted the need for eventual amputation; but Fontaine stage, ankle blood pressure, or diabetes did not. The combination of toe blood pressure of 38 mm Hg and TcpO₂ of 35 mm Hg supine was of good prognostic value.^28,70

Anatomy: Pain Pathway for Peripheral Vascular Disease

Pain within the lower-extremity distribution is believed to be transmitted by a combination of Aδ and C fibers of the somatic nervous system. The pain is also likely to be partially mediated by the SNS. In addition, the pain experienced may have both ischemic and neuropathic origins.

Indications: Spinal Cord Stimulation and the Alternative Therapies Available for Peripheral Vascular Disease

A variety of modalities have been used to treat and alleviate the symptoms of CLI in PVD. As in the case of angina, medical, interventional, and surgical techniques are used to curtail the progression of the disease. SCS is an alternative for patients with inoperable lesions who are facing the prospect of amputation.

Conservative therapies include a variety of pharmacological agents, but options are sparser than for the treatment of angina. As of 2005 several drugs were under investigation to treat CLI. Pentoxifylline (Trental), a xanthine derivative with vasodilator and hemorrheologic properties, did not demonstrate any significant benefit for patients with CLI. Cilostazol, approved for intermittent claudication patients, as of 2005 had not been demonstrated as effective in patients with CLI.⁶⁶ According to Hirsch and associates,⁶⁶ three prostaglandins (PGs)—PGE-1, iloprost, and ciprostene—with vasodilatory properties have undergone multiple trials in patients with inoperable CLI without success. More traditional conservative therapies, including analgesics, vasodilators, or anticoagulants, either slightly limit the progression or aid in alleviating the symptoms of the disease.⁷¹

Many patients with CLI are candidates for revascularization procedures before being considered for SCS. Approximately 5% of the patients with intermittent claudication develop CLI, and approximately half of those require revascularization.⁶⁶ Revascularization may be conducted by either endovascular stenting or surgical arterial vascular reconstruction. According to Hirsh and associates,⁶⁶ the mortality associated with surgical revascularization ranges from 0% to 6%, whereas the mortality for amputation is 4% to 30% during the first 30 days. Furthermore, if patients do undergo amputation, they are at greater risk for other complications, especially with decreased mobility in the elderly population. Thus, if at all possible, revascularization by one technique or another should be considered as the primary mode of treatment for patients with CLI. Not all of them are candidates for revascularization for one reason or another, but they may benefit from SCS implantation.

Basic Science: Mechanism of Action for Peripheral Vascular Disease

PVD develops through a process of slowly narrowing blood vessels, especially in the lower extremities. Eventually decreased vascular patency progresses to distal tissue ischemia and resultant limb pain. This limb pain is likely a combination of nociceptive pain from the ischemic process and neuropathic pain. Neuromodulation has been used successfully to relieve pain from tissue ischemia; but, more important, it promotes enhanced blood flow via vasodilation to the extremities.^72,73 Promotion of blood flow was first demonstrated by Cook and associates⁷² in 1976, who observed ulcer healing in patients with lower extremity PVD and attributed the improvement to increased vascular flow.

Those who suffer from PVD typically present with a history of claudication after having walked a fixed distance. As the disease progresses, claudication occurs even at rest; a condition known as CLI. CLI can be followed by infarction and the need for amputation. The ischemia that develops in the limbs occurs during a switch from aerobic to anaerobic metabolism.

As is the case of using neuromodulation to treat patients with RA, the mechanism of action for treating ischemic limb pain has not been fully elucidated. Despite this, two significant plausible mechanisms have been proposed.³¹ First, stimulation activates sensory fibers via an antidromic mechanism, resulting in a release of vasodilators. Second, neuromodulation decreases sympathetic outflow, resulting in vasodilation of distal arterial vessels.

Guidelines: Patient Selection for Spinal Cord Stimulation in Peripheral Vascular Disease

Although pain relief and limb salvage are the primary goals of SCS in PVD, patients must be selected carefully to maximize the likelihood of success and minimize inappropriate use of the device and health care resources. Patients deemed candidates for this modality should meet several criteria: have a CLI diagnosis (Fontaine stage III or IV) that is currently not evolving rapidly over a period of days or weeks,²⁸ have failed conservative measures, have been considered unsuitable for endovascular or surgical revascularization, or have already failed such interventions. Furthermore, De Vries and associates²⁸ suggest that ulcers or other skin lesions should not exceed 3 cm² since lesions beyond this size suggest a more advanced stage of disease that typically cannot be reversed with SCS. Patients should also meet macrovascular and microvascular criteria, especially the latter. If patients have a Doppler ankle systolic pressure ≤50 mm Hg or an ankle/brachial index ≤35 mm Hg and TcpO₂ between 10 and 35 mm Hg,⁹ they can be included. note: Patients with TcpO₂ ≤10 mm Hg are at greater risk for impending amputation because of significantly decreased microvascular reserve, whereas patients with TcpO₂ ≥30 mm Hg are likely to improve without the SCS.⁸³

Exclusion criteria should be considered in the decision as well. Patients not suitable for implantation include those without at-rest pain, gangrene, or ulceration; and those with infection, cancer, a vascular disease other than atherosclerosis, and a psychological or social incompetence, as outlined in Box 11-3.

Equipment and Technique for Spinal Cord Stimulation in Peripheral Vascular Disease

SCS for PVD requires epidural placement of the device leads at a level that will enable the patient to feel paresthesias over the same area as their pathology. Usually leads are introduced epidurally at a level two to three below the final resting place of the tip of the lead.²⁸ For example, to cover lower-extremity PVD, placement typically is targeted at the T10 level initially and adjusted from there (see Table 11-3 for the complete listing suggested by Deer). Again, placement should be attempted at a 45-degree angle to minimize the risk of dural puncture with the Tuohy needle and to allow a smooth insertion of the stimulator lead, both for the trial and for the placement of the permanent lead. The patient should be sedated lightly enough to be aware of paresthesias as the lead is advanced. By confirming paresthesias of the affected areas, the patient can aid the physician in determining proper placement of the lead. Once paresthesias are confirmed, the lead may be tunneled laterally to minimize movement and decrease infection risk during the trial period.

Table 11-3 Lead Placement

From Deer TR, editor: Spinal cord stimulation for the treatment of angina and peripheral vascular disease, Curr Pain Headache Rep 13(1):19, 2009.

To determine if the technique will be successful, Petrakis and Sciacca⁷⁰ suggested that a trial time of up 2 weeks be conducted. Their suggestion is based on the finding that limb salvage was achieved in their patients who demonstrated a significant increase in their TcpO₂ measured at the foot within 2 weeks of the trial and a 50% increase in TcpO₂ after 2 months of implantation.

Drug regimens that impact blood clotting before trial and implantation should be optimized, as is also the case for SCS in RA.

Outcomes for Spinal Cord Stimulation and Peripheral Vascular Disease

Recently Ubbink and Vermeulen⁶⁵ reviewed the available literature on SCS for nonreconstructible chronic CLI for the Cochrane Collaboration. They conducted a meta-analysis based on 10 papers encompassing six randomized or controlled trials that met their selection criteria. The investigators picked studies that involved patients of advanced age and CLI secondary to arthrosclerosis that was considered nonreconstructible. The randomized trials picked by Ubbink and Vermeulen⁶⁵ included those by Suy and associates, Jivegard and associates, Claeys and Horsch, Klomp and associates, and Spincemaille and associates; the controlled trial was by Amann and associates.^84–93 Combined, these studies pooled 444 patients between 1994 and 2006. The primary outcome noted in all studies was limb salvage (defined as no major amputation of foot or higher after 12 months).⁹⁴ Limb amputation leads to significant further morbidity caused by decreased mobility in the elderly population, which translates to an overall increased mortality. Secondary end points reviewed included pain relief, clinical improvement, change in macrocirculation and microcirculation, quality of life, SCS complications (see next section), and costs.

Ubbink and Vermeulen⁸³ focused on limb salvage, the primary end point of the reviewed studies first. They noted that the overall amputation prevalence was 50% no matter what treatment was provided. However, they did note that all the studies that reviewed the SCS groups demonstrated a tendency toward improved limb salvage, especially if the patients were selected based on initial TcpO₂. In addition, normotensive patients with SCS implanted had a lower amputation rate.⁸⁶ They also proposed that, on the basis of their meta-analysis of pooling the results after 12 months, the number needed to treat to prevent one major amputation was nine.⁸³

In terms of the secondary outcome of pain relief, Ubbink and Vermeulen⁸³ noted that data from the selected studies could not be pooled secondary to lack of standard deviations.⁸³ They did point out that with two studies pain was significantly better at 3 and 12 months with the SCS groups than with the medically treated groups by Jivegard and associates⁸⁶ and Spincemaille and colleagues,⁹¹ respectively. In addition, a study by Spincemaille and colleagues⁹⁰ noted that patients with SCS implantation for CLI used significantly less opioid and nonopioid analgesics, suggesting lower pain levels.

The next outcome examined was clinical improvement. The studies by Suy and associates⁹³ and Claeys and Horsch⁸⁵ showed clinical stage (measured by use of the Fontaine stages) improvement from CLI to intermittent claudication (p = 0.0014) when compared to those receiving medical treatment alone.⁶⁵ When Ubbink and Vermeulen⁶⁵ pooled the data, they showed that the number of patients who needed treatment to convert from rest pain to intermittent claudication was three. Two of the reviewed studies evaluated wound healing; but, when the data were pooled, no significant differences between the groups were determined. In addition, no significant differences were noted between patients with and without diabetes.⁸⁵

To determine the effects of SCS on circulation, investigators have looked for improved macrocirculation (measured by the ankle/brachial pressure index [ABPI]) and microcirculation. In the Cochrane review SCS-treated patients in Claeys’ study demonstrated improved ABPI of 10% with SCS and a decrease of 17% in the medically treated group (p < 0.02), whereas Jivegard and associates did not demonstrate a difference.^85,86 Ubbink and Vermeulen⁹⁴ reported that they were not able to pool the data for these studies. When evaluating the studies for an improvement in microcirculation (change in TcpO₂), no determination of significant improvement in TcpO₂ values over medical management could be made. However, when looking at individual studies, significant increases in microcirculation were observed (Claeys: p < 0.001; Ubbink and associates: p < 0.05).^85,95

Investigators in each of the studies evaluated quality of life in only a few of the studies. Unfortunately their different methods of evaluation of outcomes prevented overall compilation and analysis of data. When looking at the reviewed studies individually, the study by Amann and colleagues⁸⁴ demonstrated no overall decrease in quality of life. Spincemaille and colleagues⁹⁰ demonstrated quality of life significantly improved in patients treated with SCS (p < 0.01). In addition, some of the study groups evaluated only the SCS arm and not the conservatively treated one. Spincemaille and colleagues⁹² used the Nottingham Health Profile, which demonstrated improved quality of life in both the conservative and SCS groups, but the mobility score was significantly improved in the SCS group (p < 0.01).

Only one study reviewed,⁹⁰ done in Europe, compared costs between the two groups at 2 years. The SCS group cost substantially more than the conservative group (p < 0.009), even when adjusted for mortality (p < 0.002).

Complications for Spinal Cord Stimulation and Peripheral Vascular Disease

In Ubbink and Vermeulen’s review,⁹⁴ they noted no differences between the mortality of medically managed and SCS groups. SCS implantation complications in patients with PVD included implantation difficulties (inability to place lead) approximately 8% of the time, increasing to 12% in multicenter trials. This level of difficulty suggests the need for better-trained specialists in this technique. Other complications included battery failure, infection of lead or generator pocket, and lead fracture—all of which may potentially occur with any patient undergoing an SCS implantation.

Conclusions

If patients are selected carefully, SCS appears to be an excellent modality with multifaceted results for treating RA and/or PVD. Patients with RA not only gain pain relief, but they also improve cardiac outcome as their capacity for exercise increases. This may lead to an improved functional capacity and increased quality of life. Furthermore, SCS is advantageous because it does not mask the symptoms of acute angina from a new event or chronic angina pain. As the population of angina patients who fail a multitude of other medical and interventional approaches grows, the role of SCS will likely become more prominent. Similarly, for equally debilitating PVD, SCS promises improved pain relief, clinical improvement of disease, improved limb salvage, and better quality of life. These outcomes occur with relatively low complication rates and reasonable cost benefit.

For both disease states more research must be pursued to develop standardized patient-inclusion criteria so the possible outcomes of this expensive, highly specialized modality can be optimized and long-term efficacy may be confirmed.