- Email: [email protected]
Epidural Spinal Cord Electrical Stimulation in Diabetic Critical Lower Limb Ischemia
Epidural Spinal Cord Electrical Stimulation in Diabetic Critical Lower Limb Ischemia Ioannis E. Petrakis Vincenzo Sciacca
ABSTRACT Spinal cord stimulation (SCS) has been suggested to improve microcirculatory blood flow to relieve ischemic pain and to reduce amputation rate in patients with peripheral arterial occlusive disease (PAOD). The aim of this study was to evaluate the specific prognostic parameters in the prediction of successful SCS, in diabetic patients, performing a retrospective data analysis. To perform this evaluation, 64 diabetic patients (39 men, 25 women; mean age, 69 years) classified as Fontaine’s stage III and IV, with PAOD, were treated with SCS for rest pain and trophic lesions with dry gangrene, after failed conservative or surgical treatment. In clinical controls, pedal transcutaneous oxygen tension (TcPO2), ankle/brachial blood pressure index (ABI), and toe pressure Doppler measurements were utilized to select and followup the patients. After 58 months of follow-up (range, 20–128 months), pain relief greater than 75% and limb salvage were achieved in 38 diabetic patients. A partial success was obtained in nine patients with pain relief greater than 50% and limb salvage for at least 6 months. The method failed in 17 patients or the device was removed due to technical problems, and the limb was amputated in these patients. TcPO2 was assessed on the dorsum of the foot. Clinical improvement
and SCS success were associated with increase of TcPO2, before and after implantation. Limb salvage was achieved in the patients who had significant TcPO2 increase within the 2 weeks of the testing period, independently of the stage of the disease. A TcPO2 increase of more than 50% in the first 2 months after implantation was predictive of success, and was related to the presence of adequate paresthesias in the painful area during the trial period. TcPO2 significantly increased after longterm follow-up in all patients with limb salvage (from 22.1 to 43.1 mm Hg in the rest pain patients, from 15.8 to 36.4 mm Hg in those with trophic lesions of less than 3 cm2, and from 12.1 to 28.1 in those with trophic lesions of greater than 3 cm2, (p ⬍ 0.01). ABI did not changed under stimulation. In diabetic patients with PAOD, the SCS increases the skin blood flow, is associated with significant pain relief, and could be proven an excellent alternative therapy, improving the life quality. Significant TcPO2 increase within the 2-week test period, is a predictive index of therapy success and should be considered before the final decision in terms of cost effectiveness, before the permanent implantation. (Journal of Diabetes and Its Complications 13; 5/6: 293–299, 1999.) 2000 Elsevier Science Inc.
INTRODUCTION
First Department of General Surgery, Policlinico Umberto I, University of Rome, La Sapienza, Rome, Italy Reprint requests to be sent to: Dr. Ioannis E. PETRAKIS, Via Roma 95 Vasanello (VT), I-01030 Italy. Journal of Diabetes and Its Complications 1999; 13:293–299 2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
S
pinal cord stimulation (SCS) has been suggested to improve microcirculatory blood flow, to relieve ischemic pain and reduce amputation rate in patients with severe peripheral arterial occlusive disease (PAOD). Vascular reconstruction is the treatment of choice for diabetic patients 1056-8727/99/$–see front matter PII S1056-8727(99)00061-6
294
PETRAKIS AND SCIACCA
with severe PAOD classified as Fontaine’s stages IIb (claudication free interval less than 50 min), III (chronic ischemic rest pain), and IV (ischemic pain ulcers and/ or dry gangrene).1 Advances in endovascular treatment and vascular surgery have resulted in increased limb salvage rates. Despite this undeniable progress, the number of patients with nonreconstructible lesions remains high.2 Ideal treatment in these stages allows the patients to retain his limb with no or tolerable pain and to regain a satisfactory level of independence. Spinal cord stimulation was used initially, two decades ago, into clinical practice to manage patients with chronic intractable pain or to improve motor function in patients with partial spinal cord lesion.3,4,5 Several authors observed a marked improvement in lower limb blood flow in a group of patients who were being treated with spinal cord stimulation for pain related to multiple sclerosis. Based on these observations SCS has attracted greater interest in the treatment of ischemic rest pain especially in diabetic patients.6 Some authors have reported significant pain relief and healing of ischemic ulcers in both diabetics and patients with end-stage vascular disease who are receiving SCS.6–13 Jacobs et al.,14,15 found that the number of capillaries perfused and the red blood cell velocity were significantly increased by SCS. Several non-invasive and scarcely invasive techniques (Doppler, rheography, plethysmography, thermography, transcutaneous oxygen tension, 201TI muscle scintigraphy, xenon133 clearance) have been applied in the effort to quantify the SCS effect on blood flow.16–19 However, the purpose of this study was to evaluate the specific clinical and laboratory parameters related to the transcutaneous oxygen tension (TcPO2) changes in the prediction of successful SCS in a retrospective analysis of the obtained data, during the follow-up of our implanted diabetic patients for nonreconstructible PAOD of the lower limbs. MATERIALS AND METHODS Sixty-four diabetic patients with nonreconstructible PAOD of the lower limbs treated since 1982 were involved in the present study. They were 39 men and 25 women with mean age 69 years (range, 48–84 years) and mean duration of diabetes 24 years (range, 17–38 years). The indication to SCS was based on clinical and laboratory tests. There were 14 patients with at list 1-month duration rest pain, 28 patients with trophic lesions less than 3 cm2, and 22 patients with trophic lesions greater than 3 cm2 (dry gangrene of one or two toes). Clinical diagnosis was confirmed by an ankle/ brachial blood pressure index (ABI) of less than 0,20 ⫾ 0,08 and toe blood pressure of 30 mm Hg or less. Intraarterial subtraction or intraoperative arteriograms were obtained from all diabetic patients and showed occluded femoral and tibial vessels, unsuitable for a by-
J Diab Comp 1999; 13:293–299
pass procedure or angioplasty. Transcutaneous oxygen tension (TcPO2) at the dorsum of the foot, was measured to evaluate the reduction in skin circulation.16–19 The intensity of symptoms was reported to be severe in all cases. The duration of vascular history ranged from 6 months to more than 6 years. Prior to implantation, all patients had received conservative treatment, including management of risk factors, pain therapy, and surgical foot care. Failed bypass procedures in the involved leg were noted in 28 patients, 4 had pervious lumbar sympathectomy before being referred to our institution. Atherosclerotic disease with unsuitable vessels after exploration of the tibial or/and popliteal artery was found in 26 patients, while 6 of them had already a major contraleral amputation above the knee with severe ischemia and pain in the other limb. Absence of reconstructible arteries at angiography was found in 10 patients. Significant heart failure, severe pulmonary or renal insufficiency, unstable angina, noncooperative patient, and short life expectancy were contraindications to SCS therapy. Other contraindications for SCS were spine disease, limb gangrene with osteomyelitis and wet gangrene with lesions larger than 3 cm2. To benefit from SCS therapy the patient must be able to understand the principles of treatment: the stimulation produce paresthesias, which the patient must be willing to accept. These selection criteria include thorough vascular evaluation, failure of previous attempts to conservative and surgical therapy, adequate professional equipment, experience, and follow-up facilities. Diabetic patients who died from other causes during the followup, were excluded from the study. All patients were submitted to temporary implantation, while only in those with significant pain relief the permanent device was inserted, and were considered in this study. Analysis of pain relief and measurements of ankle pressure, with calculation of ABI, and TcPO2, (mm Hg) respectively, were performed before implantation, 2 weeks after the temporary implantation, two months after permanent implantation and after 6 months. The pain relief and the success of the method was based on intake of analgesic drugs and patients self pain evaluation using the following visual analogue scale (VAS): (A) Success was defined as pain relief of greater than 75% and limb salvage or minor amputation (toe). (B) Partial success was defined as limb salvage associated with pain relief evaluated between 50% and 70% or as temporary benefit (for a minimum of 6 months). (C) Failure was defined as pain relief evaluated as less than 50% and resulting in limb loss within the first 6 months or technical problems resulting in dislocation of the electrode, during the permanent implantation procedure, or during the follow-up, or infected electrode, or generator which required removal of the device. The severity of PAOD was accurately determined us-
J Diab Comp 1999; 13:293–299
295
SCS IN DIABETIC ISCHEMIA
ing the TcPO2 measurement, which is known to correlated with the clinical stage of the disease.16–19 The TcPO2 was measured by Clark’s electrode and seems to be a reliable noninvasive index of tissue perfusion, useful for assessing peripheral arterial occlusive disease.16–19 TcPO2 measurements were performed with the patients in supine position and the electrode, attached to the dorsum of the foot, was warmed to 44⬚C, causing maximum vasodilation of the underlying skin circulation. Systolic arterial ankle and toe pressures (in mm Hg) were measured using bi-directional Doppler ultrasonography in the posterior and anterior tibial arteries, respectively, and then in the digital arteries. The ABI was determined by dividing systolic ankle pressure by systolic brachial artery pressure.
with the help of a portable computer, according to the patient clinical state. Statistical Analysis. The TcPO2 data were calculated as mean values and standard deviations. Paired t tests, with appropriate correction to reduce the type-I errors were performed and values of p ⬍ 0.05 were considered statistically significant. RESULTS The diabetic patients, during the 2-week testing period before the permanent device implantation, experienced pain relief ranging from significant to moderate and were followed postoperatively for an average of 58 months (range, 20–128 months). Among the 14 diabetic patients with rest pain, longterm limb salvage and pain relief greater than 75% were obtained in 11 (79%); while a partial success was achieved in one patient (7%, pain relief greater than 50% and limb salvage for at least 6 months). In one patient, SCS failed to salvage the limb and within the first 6 months, after implantation, the patient underwent a major amputation (Table 1). In one patient, the generator was infected and removed with failure of the SCS therapy. Among the 28 diabetic patients with trophic lesions less than 3 cm2 long-term limb salvage and pain relief greater than 75% were obtained in 19 patients (68%), with simultaneous healing of trophic lesions. In 4 diabetic patients (14%) an amelioration of the ischemic pain greater than 50% initially was found, however, it diminished with time, and after a 12-month period (range, 6–22 months), the pain control and the clinical improvement were reduced, and the use of analgesic therapy was necessary. Severe ischemic pain developed in three of them (pain control less than 50%) within the first 6 months after implantation, leading to major amputations (Table 1). In two patients the electrode was either dislocated or infected, and the device was removed, with failure of the SCS therapy.
Implantation Technique. A quadripolar electrode (Pisces-Quad 3487A, Medtronic, Minneapolis, Minn.) was placed in the epidural space by percutaneous lumbar puncture between L2 and L3 or L3 and L4. The electrode was positioned mostly in the midline under fluoroscopic guidance up to the level T10-11. Connecting a portable stimulator to the electrode allowed intraoperative test stimulation producing comfortable paresthesias in the painful foot or limb. The clinical effects were tested during a trial period (2 weeks) and in case that the patient had a significant pain relief, an implantable pulse generator (Itrel II IPG, Medtronic Inc. Minneapolis USA) was placed in an abdominal subcutaneous pocket. Various settings of the active quadripolar extremities of the electrode were studied to evaluate the most appropriate combination for the patient pain relief. Bipolar stimulation was assessed by the presence of paresthesias and a feeling of warmth in the painful area. The setting parameters were pulse amplitude between 1.0 and 5.0 V, frequency between 40 and 120 pps and pulse width from 150 to 450 sec. Stimulation was continuous 24 hours a day to obtain maximum information on the clinical outcome. During follow up the parameters of stimulation can be reset
TABLE 1. CLINICAL RESULTS AND PAIN RELIEF AFTER SPINAL CORD STIMULATION (SCS) TREATMENT SCS Therapy Success
Rest pain, 39 patients Trophic lesions ⬍3 cm2, 58 patients Trophic lesions ⬎3 cm2, 53 patients Overall series
Partial Success
Failure
Nⴗ
%
Nⴗ
%
Nⴗ
%
11 19 8 38
(79%) (68%) (37%) (59%)
1 4 4 9
(7%) (14%) (18%) (14%)
2 5 10 17
(14%) (18%) (45%) (27%)
Note: The diabetic patients with therapy success achieved a pain relief ⬎75% and long-term limb salvage, while these with partial success achieved a pain control ⬎50% with or without use of analgesics and limb salvage for at least 6 months. Patients with SCS failure (due also to technical problems) were submitted in major amputations after a short period of time from the implantation.
296
PETRAKIS AND SCIACCA
J Diab Comp 1999; 13:293–299
TABLE 2. MEAN (SD) OF TcPO2 VALUE CHANGES IN THE DIABETIC PATIENTS WITH LIMB SALVAGE FOR AT LEAST 6 MONTHS (THERAPY SUCCESS AND PARTIAL SUCCESS) BEFORE TREATMENT, AFTER TEMPORARY IMPLANTATION, 2 MONTHS AFTER PERMANENT IMPLANTATION, AND AFTER LONG-TERM FOLLOW-UP TcPO2 Changes, mm Hg Rest pain patients Trophic lesions ⬍3 cm2 patients Trophic lesions ⬎3 cm2 patients
Before Treatment
After Temporary Implantation
p Value
2 Months
Long Term
p Value
22.1 (5.9) 15.8 (6.6) 12.1 (5.1)
30.5 (6.6) 22.7 (5.3) 17.36 (4.7)
⫽0.04 ⫽0.03 ⫽0.03
34 (5.4) 24.2 (5.2) 19.1 (4.3)
43.1 (5.7) 36.4 (6.6) 28.1 (6.8)
⬍0.01 ⬍0.02 ⬍0.01
Among the 22 diabetic patients with trophic lesions greater than 3 cm2, pain relief greater than 75% with lesion healing was obtained in 8 patients (40%), while in 4 patients (20%, two of them underwent minor amputations), a partial success, with pain relief greater than 50% and limb salvage for at least 6 months (mean, 9 months; range, 6–12 months) was achieved. After this period, the ischemic pain recurred and major amputations were unavoidable. Eight patients treated with SCS, initially presented a pain relief less than 50%, but after 2 months, analgesic medication was necessary for the pain control, however, within the first 6 months from the device implantation all these patients were submitted to major amputations, because severe ischemic pain recurred (Table 1). In two patients, the electrode was either dislocated or infected, and was removed, with failure of the SCS therapy. TcPO2 foot values increased from 22.1 ⫾ 5.9 mm Hg to 43.1 ⫾ 5.7 mm Hg (p ⬍ 0.01) in the patients with rest pain and limb salvage. In the patients with trophic lesions less than 3 cm2 and limb salvage TcPO2 mean foot values showed an average improvement from 15.8 ⫾ 6.6 to 36.4 ⫾ 6.8 mm Hg (p ⬍ 0.02). In the patients with trophic lesions greater than 3 cm2 (with dry gangrene) and limb salvage TcPO2 mean foot values showed an average improvement from 12.1 ⫾ 5.1 to 28.1 ⫾ 6.8 mm Hg (p ⬍ 0.01), (Table 2). TcPO2 changes are related with the presence of adequate paresthesias in the painful area during the trial period. Major changes in TcPO2 values were achieved within the first 2 weeks after temporary implantation, while an increasing of more than 50% in the first 2 months
after implantation was revealed in the patients with limb salvage, (Table 2). Changes of TcPO2 in all amputated patients where SCS failed to save the limb, were not significant even after 2 or 4 weeks from the device implantation (Table 3). No significant modifications were observed in the ABI or toe pressure values before permanent implantation (test period) and during the follow-up. In our casuistic we report only one dislocation of the electrode, during the permanent implantation procedure, in two patients the electrode or the generator was infected and required removal of the device. During our follow up, 1 electrode had been dislocated in the early post-operative period. Limb salvage, for at least 6 months (SCS therapy success and partial success), was achieved in 47 patients (73%) while in 17 patients (25%) major amputations within the first six months after implantation, due to SCS failure or technical problems with device removal, were unavoidable. After a mean period of 3 years 8 patients were submitted to substitution of exhausted generator. DISCUSSION Peripheral angiopathy and neuropathy is a common long-term diabetes complication.20,21 About 7.5% of unselected adults attending a hospital diabetic clinic have painful neuropathic symptoms, mainly in the lower limbs.22 Pain varies from mild paresthesias in a few toes to severe unremitting pain in both legs.22,23 Nighttime exacerbation of the pain plus contact hypersensitivity to bed-cloths results in loss of sleep, and pain in diabetic neuropathy can be disabling.23 The cause of
TABLE 3. MEAN (SD) of TcPO2 VALUE CHANGES BEFORE TREATMENT, AFTER TEMPORARY IMPLATATION AND 2 MONTHS AFTER PERMANENT IMPLANTATION, IN THE DIABETIC PATIENTS WITH FAILURE OF THE SCS TO SALVAGE THE LIMB FOR AT LEAST 6 MONTHS TcPO2 Changes, mm Hg Rest pain patients Trophic lesions ⬍3 cm2 patients Trophic lesions ⬎3 cm2 patients * Not significant.
Before Treatment
After Temporary Implantation
p Value
2 Months
p Value
20.1 (4.6) 15.1 (5.1) 9.1 (5.6)
23.4 (4.4) 17.4 (4.8) 10.6 (4.9)
NS* NS NS
26.2 (5.2) 18 (3.7) 12.2 (5.8)
NS NS NS
J Diab Comp 1999; 13:293–299
chronic sensory-motor diabetic neuropathy or indeed neuropathic pain is not known although metabolic and microvascular systems may be involved.24,25 While the search for potential therapeutic agents to halt or reverse the neuropathic process continues,26 current treatment is largely aimed at relieving painful symptoms. The SCS offers a new and effective way of relieving chronic diabetic neuropathic and ischemic pain and improves exercise tolerance. The contemporary literature reports demonstrate the efficacy of SCS in relieving ischemic pain.27–29 Some authors showed that in PAOD patients, SCS recruits small capillaries not ordinarily perfuse, enhancing skin blood flow, improvements that may explain the beneficial clinical effects experienced by these patients.15,16 However, their data and the use of SCS in the treatment of POAD were recently criticized, maybe because there are not standard indications for the SCS implantation to obtain the best results. Thus contrary to the recommendations of other authors in our opinion SCS should not be used to treat patients affected by extensive gangrenous lesions of the foot (classified in Fontaine’s IVb stage). For these patients, nonresponders to medical treatment neither amenable to vascular reconstructions, the recommended treatment is primary amputation. In the presence of severe nociceptive somatic pain the analgesia associated with SCS is less effective than obtained with epidural anesthetics with or without opiates. Concerning patients affected by only claudication, we recommend that device implantation is indicated if pain-free walking interval is less than 50 meters (Fontaine’s IIb stage patients) and if the other therapeutic options have definitely been excluded. The presence of diabetes does not represent a contraindication to SCS use. The exact underlying mechanism of how SCS influences chronic and ischemic pain remain unclear. The gate-control theory of pain, proposed by Melzack and Wall30 in 1965, has been the one widely used to explain the action of SCS. This theory postulated that an inhibitory mechanism in the dorsal horn is activated by the recruitment of large-diameter fibers. The substantia gelatinosa activation induces inhibition of second order neurons processing nociceptive information. Presently, the mechanisms proposed to be responsible for pain relief are neurophysiologic and neurochemical in nature. The neurophysiology of pain relief under SCS varies from simple blocking of pain transmission by a direct effect on spinothalamic tracts, segmental inhibition via coarse fiber activation, effects on central sympathetic system, and brain stem loops to thalamocortical mechanisms.31 The inhibitory effects of SCS on the transmission of nociceptive impulses may be exerted segmentally in the spinal cord and/or at a supraspinal level.32 Clinical observations indicate that the mechanisms involved in the stimulation-induced relief ischemic pain are different from those related to relief of
SCS IN DIABETIC ISCHEMIA
297
others types of pain.33 Indeed, nociceptive pain is more resistant to SCS, and significant pain relief is almost never obtained before a couple of days. In contrast, neuropathic pain of peripheral origin responds well and directly to SCS. Both components, nociceptive as well as neuropathic are present in ischemic pain. Several authors have postulated that the principal factor in the relief of ischemic pain is the inhibition of the pain signal per se, leading to a decrease in sympathetic activity and improved skin microcirculation.8,34 Another hypothesis is that SCS depresses autonomic sympathetic activity.33,35 However, many authors have reported success even after sympathectomy.8,14 SCS has been shown to be useful in vasospastic disease and reflex sympathetic dystrophy.34 It is necessary that during the testing period of about 2 weeks, in addition to pain control, we assess carefully the effect on peripheral blood flow to ascertain if the warmth feeling and the paresthesias reported by the patient are related to increased skin microcirculation. This assessment should include determination of the pain-free walking interval under standard conditions (treadmill), confirmation of ulcer healing (surface measurements), and verification of improved blood flow. The decrease of ischemic pain with SCS is probably secondary to the positive effects on microcirculation rather than vice versa. Pain relief may also result in attenuation of sympathetic activity with vasodilatation, leading to further pain relief.32 The most useful instrumental technique for peripheral vascular screening during SCS, TcPO2, a noninvasive method, is suitable for accessing skin circulation8,14,15–19,35,36 and has been used to evaluate microcirculatory changes induced by SCS. In PAOD, the aim of the SCS is not only to achieve effective analgesia (which might be obtained by others less expensive techniques), but also to promote the trophic-functional recovery of the body segment affected by an advanced ischemic process. Furthermore, in agreement with other authors, SCS should be considered in patients with neuropathic pain who do not respond to conventional treatment,37 but we believe that diabetic patients with severe neuropathic pain and advanced PAOD should be excluded. In fact, diabetic patients with neuropathy and extended lesions (greater than 3 cm2), in our experience, after stimulation have limited or no TcPO2 increase and the method fails in the most of the cases. In this study, we found that all 64 implanted diabetic patients, during the test-period stimulation, noted immediate significant pain relief, while only in those patients with limb salvage, TcPO2 value changes showed a significant increase following stimulation. Two months after implantation a TcPO2 increase of more than 50% was achieved only in the patients with limb salvage and was predictive of SCS success. Moreover, we demonstrated that the differences of the TcPO2 values before
298
PETRAKIS AND SCIACCA
and after implantation are more important for the SCS success and limb salvage, rather than the stage of the disease. Little or no TcPO2 changes in diabetic patients after implantation, independently from the stage of the disease, are predictive of SCS failure and should be taken in consideration prior to implantation, in terms of cost effectiveness. After SCS, TcPO2 on the dorsum of the foot significantly increased, whereas the ABI did not change. This is not surprising because patients treated with SCS had crural vessels that were unsuitable for reconstruction. The changes in TcPO2 related to increase skin perfusion were not result of improved arterial inflow. The feeling of warmth in the painful area under SCS was related to increase in TcPO2. TcPO2 measurement is able to evaluate the effects of SCS on microcirculatory blood flow and represents a predictive factor of guarantee success as an increase in TcPO2 greater than 50% in the first 2 months correlates with a positive result. Although the initial pain control after SCS, the current study suggests 2 weeks of temporary implantation to evaluate the paresthesias, the warmth feeling and the TcPO2 changes of the limb. Thus in case of limited TcPO2 changes in PAOD diabetic patients with crural anelastic vessels without any compliance the method fails. This is important for the final decision, in terms of cost effectiveness, in all the cases where the failure of the SCS is quite sure after a short period of time. In conclusion, in PAOD diabetic patients with or without ischemic foot lesions, after failed medical or surgical therapy, limb salvage was achieved after SCS therapy, in those with significant TcPO2 increase. Except the pain control, clinical improvement was associated with a significant increase in TcPO2 within the first post-implanted period. Effects of SCS on macrocirculation were not significant. Relevant research data are needed to show how SCS affects pain and to explain the effects on the vascular system. We suggest that SCS is promising in diabetic patients with PAOD when previous therapies have failed. However, a 2-week test period is important to appropriately select those patients with significant TcPO2 increase, paresthesias and clinical warmth feeling, for permanent implantation. REFERENCES 1. Friedman SG, Kerner BA, Friedman MS, Moccio CG: Limb salvage in elderly patients: Is aggressive surgical therapy warranted? J Cardivasc Surg 30:848–851, 1989. 2. Gregg RO: Bypass or amputation: Concomitant review of bypass arterial grafting and major amputation. Am J Surg 149:397–402, 1985. 3. Shealy CN, Mortimer JT, Reswick JB: Electrical inhibition of pain by stimulation of the dorsal columns: Preliminary clinical report. Anesth Analg 46:489–491, 1967. 4. Meglio M, Cioni B, Rossi GF: Spinal cord stimulation
J Diab Comp 1999; 13:293–299
in management of chronic pain: A 9-years experience. J Neurosurg 70:519–524, 1989. 5. Cook AW: Electrical stimulation in multiple sclerosis. Med Biol Eng Comput 18:48–56, 1980. 6. Augustinsson LE: Epidural spinal electrical stimulation in peripheral vascular disease. Pace 10:205–206, 1987. 7. Broggi G, Servello D, Franzini A: Spinal cord stimulation for treatment of peripheral vascular disease. Appl Neurofisiol 50:439–441, 1987. 8. Broseta I, Barbera I, De Vera JA: Spinal cord Stimulation in peripheral arterial disease. J Neurosurg 64:71–80, 1986. 9. Dooley D, Kasprak M: Modification of blood flow to the extremities by electrical stimulation of the nervous sistem. South Med J 69:1309–1311, 1976. 10. Fiume D: Spinal cord stimulation in peripheral vascular pain. Appl Neurophisiol 46:290–294, 1983. 11. Groth KE: Spinal cord stimulation for the treatment of peripheral vascular disease. European Multicenter Study, in Fields H (ed.), Advances in Pain Research and Therapy. New York, Raven, 1985, pp. 861–870. 12. Horsch S, Claeys L: Epidural spinal cord stimulation in the treatment of severe peripheral arterial occlusive disease. Ann Vasc Surg 8:468–474, 1994. 13. Visconti W, Fontana P, Buonocuore N, Grillo N, Seno S: Spinal cord stimulation in advanced intractable chronic obliterant arteriopathies. Minerva Cardioangiol 44:19–27, 1996. 14. Jacobs MHJM, Joming PJG, Soures RJ, Kitslaar PJEHM, Slaaf DW, Reneman RS: Epidural spinal cord electrical stimulation improves microvascular blood flow in severe limb ischemia. Ann Surg 207:179–183, 1988. 15. Jacobs MJ, Joming PJ, Beckers RCY, Ubbink DT, van Kleef M: Foot salvage and improvement of microvascular blood flow as a result of epidural spinal cord electrical stimulation. J Vasc Surg 12:354–360, 1990. 16. Franzeck UK, Talke P, Bernstein EF, Colbranson FL, Fronek A: Transcutaneous PO2 measurements in health and peripheral arterial occlusive disease. Surgery 91: 156–163, 1982. 17. Hauser CL, Shoemaker WC: Use of transcutaneous PO2 regional perfusion index to quantify tissue perfusion in peripheral vascular disease. Ann Surg 197:337–343, 1983. 18. Jaszezak P, Poulsen J: Estimation of blood flow in transcutaneous PO2 measurements. Acta Anaesthisiol Scand 27:174–180, 1983. 19. Sciacca V, Mignoli A, di Marzo L, Maggiore C, Fiume D, Cavallaro A: Predictive value of transcutaneous oxygen tension measurements in the indication for spinal cord stimulation in patients with peripheral vascular disease: Preliminary results. Vasc Surg 3:128–132, 1989. 20. Clyne CAC, Ryas J, Webster JHH, et al.: Oxygen tension on the skin of ischermic legs. Am J Surg 143:315–318, 1982. 21. Tesfaye S, Malik R, Ward JD: Vascular factors in diabetic neuropathy. Diabetologia 37:847–854, 1994. 22. Chen AW, MacFarlane IA, Bowsher DR, Wells JC, Bessex C, Griffiths K: Chronic pain in patients with diabetes mellitus; comparison with non-diabetic population. Pain Clin 3:147–159, 1990.
J Diab Comp 1999; 13:293–299
23. Watkins PJ: Pain and diabetic neuropathy. BMJ 288:168– 169, 1984. 24. Editorial. Pain perception in diabetic neuropathy. Lancet i:83–85, 1985. 25. Chen AW, Mac Farlane IA, Bowsher DR, Wells JDC: Does acute hyperglycaemia influence heat pain thresholds? J Neurol Neurosurg Psychiatry 51:688–690, 1988. 26. Cameron NE, Cotter MA: Potential therapeutic approaches to the treatment or prevention of diabetic neuropathy: evidence from experimental studies. Diabetic Med 10:593–605, 1993. 27. Tesfaye S: Diabetic neuropathy: Current treatment and potential therapeutic approaches. Diab Nutr Metab 7: 375–379, 1994. 28. Augustinsson LE, Holm J, Carlsson CA, Jivegard L: Epidural electrical stimulation in severe ischemia: Evidence of pain relief, increased blood flow and a possible limb-saving effect. Ann Surg 1:354–349, 1985. 29. Jivegard LE, Augustinsson LE, Holm J, Risberg B, Ortenwall P: Effects of spinal cord stimulation (SCS) in patients with inoperable severe lower limb ischemia: A Prospective randomized controlled study. Eur J Vasc Endov Surg 9:421–425, 1995. 30. Metzack R, Wall PD: Pain mechanisms: A new theory. Science 150:1971–1979, 1965.
SCS IN DIABETIC ISCHEMIA
299
31. Gybeis JM, Kupers R: Central and peripheral electrical stimulation of the nervous system in the treatment of chronic pain. Acta Neurochir 38 (suppl):65–75, 1987. 32. Linderoth B: Dorsal column stimulation and pain: Experimental study of putative neurochemical and neurophysiological mechanisms. Thesis. Stockholm, 1992, pp. 11–19. 33. Meyerson BA: Electrical stimulation of the spinal cord and brain, in Bonica JJ, Loeser JD, Chapman RC, et al. (eds.) The Management of Pain. Philadelphia, Lea and Febiger, 1990, pp. 1862–1877. 34. Hosobuchi Y: Treatment of ischemic pain by neurostimulation, in Lipton S, Tunks E, Zoopi M (eds.), Advances in Pain Research and Therapy, Vol, 13. New York, Raven, 1990, pp. 223–226. 35. Cina C, Katsamouris A, Megerman J, Brewster DC, Strayhorn EC, Robison JG, Abbott WM: Utility of transcutaneous oxygen tension measurements in peripheral arterial occlusive disease. J Vasc Surg 1:362–371, 1984. 36. Petrakis I, Sciacca V: Transcutaneous oxygen tension (TcPO2) in the testing period of spinal cord stimulation (SCS) in critical limb ischemia of the lower extremities. Int Surg 84:122–128, 1999. 37. Tesfaye S, Watt J, Benbow JS, Pang AK, Miles JC, MacFarlane IA: Electrical spinal-cord stimulation for painful diabetic peripheral neuropathy. Lancet 348:1696–1701, 1996.
ABSTRACT Spinal cord stimulation (SCS) has been suggested to improve microcirculatory blood flow to relieve ischemic pain and to reduce amputation rate in patients with peripheral arterial occlusive disease (PAOD). The aim of this study was to evaluate the specific prognostic parameters in the prediction of successful SCS, in diabetic patients, performing a retrospective data analysis. To perform this evaluation, 64 diabetic patients (39 men, 25 women; mean age, 69 years) classified as Fontaine’s stage III and IV, with PAOD, were treated with SCS for rest pain and trophic lesions with dry gangrene, after failed conservative or surgical treatment. In clinical controls, pedal transcutaneous oxygen tension (TcPO2), ankle/brachial blood pressure index (ABI), and toe pressure Doppler measurements were utilized to select and followup the patients. After 58 months of follow-up (range, 20–128 months), pain relief greater than 75% and limb salvage were achieved in 38 diabetic patients. A partial success was obtained in nine patients with pain relief greater than 50% and limb salvage for at least 6 months. The method failed in 17 patients or the device was removed due to technical problems, and the limb was amputated in these patients. TcPO2 was assessed on the dorsum of the foot. Clinical improvement
and SCS success were associated with increase of TcPO2, before and after implantation. Limb salvage was achieved in the patients who had significant TcPO2 increase within the 2 weeks of the testing period, independently of the stage of the disease. A TcPO2 increase of more than 50% in the first 2 months after implantation was predictive of success, and was related to the presence of adequate paresthesias in the painful area during the trial period. TcPO2 significantly increased after longterm follow-up in all patients with limb salvage (from 22.1 to 43.1 mm Hg in the rest pain patients, from 15.8 to 36.4 mm Hg in those with trophic lesions of less than 3 cm2, and from 12.1 to 28.1 in those with trophic lesions of greater than 3 cm2, (p ⬍ 0.01). ABI did not changed under stimulation. In diabetic patients with PAOD, the SCS increases the skin blood flow, is associated with significant pain relief, and could be proven an excellent alternative therapy, improving the life quality. Significant TcPO2 increase within the 2-week test period, is a predictive index of therapy success and should be considered before the final decision in terms of cost effectiveness, before the permanent implantation. (Journal of Diabetes and Its Complications 13; 5/6: 293–299, 1999.) 2000 Elsevier Science Inc.
INTRODUCTION
First Department of General Surgery, Policlinico Umberto I, University of Rome, La Sapienza, Rome, Italy Reprint requests to be sent to: Dr. Ioannis E. PETRAKIS, Via Roma 95 Vasanello (VT), I-01030 Italy. Journal of Diabetes and Its Complications 1999; 13:293–299 2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
S
pinal cord stimulation (SCS) has been suggested to improve microcirculatory blood flow, to relieve ischemic pain and reduce amputation rate in patients with severe peripheral arterial occlusive disease (PAOD). Vascular reconstruction is the treatment of choice for diabetic patients 1056-8727/99/$–see front matter PII S1056-8727(99)00061-6
294
PETRAKIS AND SCIACCA
with severe PAOD classified as Fontaine’s stages IIb (claudication free interval less than 50 min), III (chronic ischemic rest pain), and IV (ischemic pain ulcers and/ or dry gangrene).1 Advances in endovascular treatment and vascular surgery have resulted in increased limb salvage rates. Despite this undeniable progress, the number of patients with nonreconstructible lesions remains high.2 Ideal treatment in these stages allows the patients to retain his limb with no or tolerable pain and to regain a satisfactory level of independence. Spinal cord stimulation was used initially, two decades ago, into clinical practice to manage patients with chronic intractable pain or to improve motor function in patients with partial spinal cord lesion.3,4,5 Several authors observed a marked improvement in lower limb blood flow in a group of patients who were being treated with spinal cord stimulation for pain related to multiple sclerosis. Based on these observations SCS has attracted greater interest in the treatment of ischemic rest pain especially in diabetic patients.6 Some authors have reported significant pain relief and healing of ischemic ulcers in both diabetics and patients with end-stage vascular disease who are receiving SCS.6–13 Jacobs et al.,14,15 found that the number of capillaries perfused and the red blood cell velocity were significantly increased by SCS. Several non-invasive and scarcely invasive techniques (Doppler, rheography, plethysmography, thermography, transcutaneous oxygen tension, 201TI muscle scintigraphy, xenon133 clearance) have been applied in the effort to quantify the SCS effect on blood flow.16–19 However, the purpose of this study was to evaluate the specific clinical and laboratory parameters related to the transcutaneous oxygen tension (TcPO2) changes in the prediction of successful SCS in a retrospective analysis of the obtained data, during the follow-up of our implanted diabetic patients for nonreconstructible PAOD of the lower limbs. MATERIALS AND METHODS Sixty-four diabetic patients with nonreconstructible PAOD of the lower limbs treated since 1982 were involved in the present study. They were 39 men and 25 women with mean age 69 years (range, 48–84 years) and mean duration of diabetes 24 years (range, 17–38 years). The indication to SCS was based on clinical and laboratory tests. There were 14 patients with at list 1-month duration rest pain, 28 patients with trophic lesions less than 3 cm2, and 22 patients with trophic lesions greater than 3 cm2 (dry gangrene of one or two toes). Clinical diagnosis was confirmed by an ankle/ brachial blood pressure index (ABI) of less than 0,20 ⫾ 0,08 and toe blood pressure of 30 mm Hg or less. Intraarterial subtraction or intraoperative arteriograms were obtained from all diabetic patients and showed occluded femoral and tibial vessels, unsuitable for a by-
J Diab Comp 1999; 13:293–299
pass procedure or angioplasty. Transcutaneous oxygen tension (TcPO2) at the dorsum of the foot, was measured to evaluate the reduction in skin circulation.16–19 The intensity of symptoms was reported to be severe in all cases. The duration of vascular history ranged from 6 months to more than 6 years. Prior to implantation, all patients had received conservative treatment, including management of risk factors, pain therapy, and surgical foot care. Failed bypass procedures in the involved leg were noted in 28 patients, 4 had pervious lumbar sympathectomy before being referred to our institution. Atherosclerotic disease with unsuitable vessels after exploration of the tibial or/and popliteal artery was found in 26 patients, while 6 of them had already a major contraleral amputation above the knee with severe ischemia and pain in the other limb. Absence of reconstructible arteries at angiography was found in 10 patients. Significant heart failure, severe pulmonary or renal insufficiency, unstable angina, noncooperative patient, and short life expectancy were contraindications to SCS therapy. Other contraindications for SCS were spine disease, limb gangrene with osteomyelitis and wet gangrene with lesions larger than 3 cm2. To benefit from SCS therapy the patient must be able to understand the principles of treatment: the stimulation produce paresthesias, which the patient must be willing to accept. These selection criteria include thorough vascular evaluation, failure of previous attempts to conservative and surgical therapy, adequate professional equipment, experience, and follow-up facilities. Diabetic patients who died from other causes during the followup, were excluded from the study. All patients were submitted to temporary implantation, while only in those with significant pain relief the permanent device was inserted, and were considered in this study. Analysis of pain relief and measurements of ankle pressure, with calculation of ABI, and TcPO2, (mm Hg) respectively, were performed before implantation, 2 weeks after the temporary implantation, two months after permanent implantation and after 6 months. The pain relief and the success of the method was based on intake of analgesic drugs and patients self pain evaluation using the following visual analogue scale (VAS): (A) Success was defined as pain relief of greater than 75% and limb salvage or minor amputation (toe). (B) Partial success was defined as limb salvage associated with pain relief evaluated between 50% and 70% or as temporary benefit (for a minimum of 6 months). (C) Failure was defined as pain relief evaluated as less than 50% and resulting in limb loss within the first 6 months or technical problems resulting in dislocation of the electrode, during the permanent implantation procedure, or during the follow-up, or infected electrode, or generator which required removal of the device. The severity of PAOD was accurately determined us-
J Diab Comp 1999; 13:293–299
295
SCS IN DIABETIC ISCHEMIA
ing the TcPO2 measurement, which is known to correlated with the clinical stage of the disease.16–19 The TcPO2 was measured by Clark’s electrode and seems to be a reliable noninvasive index of tissue perfusion, useful for assessing peripheral arterial occlusive disease.16–19 TcPO2 measurements were performed with the patients in supine position and the electrode, attached to the dorsum of the foot, was warmed to 44⬚C, causing maximum vasodilation of the underlying skin circulation. Systolic arterial ankle and toe pressures (in mm Hg) were measured using bi-directional Doppler ultrasonography in the posterior and anterior tibial arteries, respectively, and then in the digital arteries. The ABI was determined by dividing systolic ankle pressure by systolic brachial artery pressure.
with the help of a portable computer, according to the patient clinical state. Statistical Analysis. The TcPO2 data were calculated as mean values and standard deviations. Paired t tests, with appropriate correction to reduce the type-I errors were performed and values of p ⬍ 0.05 were considered statistically significant. RESULTS The diabetic patients, during the 2-week testing period before the permanent device implantation, experienced pain relief ranging from significant to moderate and were followed postoperatively for an average of 58 months (range, 20–128 months). Among the 14 diabetic patients with rest pain, longterm limb salvage and pain relief greater than 75% were obtained in 11 (79%); while a partial success was achieved in one patient (7%, pain relief greater than 50% and limb salvage for at least 6 months). In one patient, SCS failed to salvage the limb and within the first 6 months, after implantation, the patient underwent a major amputation (Table 1). In one patient, the generator was infected and removed with failure of the SCS therapy. Among the 28 diabetic patients with trophic lesions less than 3 cm2 long-term limb salvage and pain relief greater than 75% were obtained in 19 patients (68%), with simultaneous healing of trophic lesions. In 4 diabetic patients (14%) an amelioration of the ischemic pain greater than 50% initially was found, however, it diminished with time, and after a 12-month period (range, 6–22 months), the pain control and the clinical improvement were reduced, and the use of analgesic therapy was necessary. Severe ischemic pain developed in three of them (pain control less than 50%) within the first 6 months after implantation, leading to major amputations (Table 1). In two patients the electrode was either dislocated or infected, and the device was removed, with failure of the SCS therapy.
Implantation Technique. A quadripolar electrode (Pisces-Quad 3487A, Medtronic, Minneapolis, Minn.) was placed in the epidural space by percutaneous lumbar puncture between L2 and L3 or L3 and L4. The electrode was positioned mostly in the midline under fluoroscopic guidance up to the level T10-11. Connecting a portable stimulator to the electrode allowed intraoperative test stimulation producing comfortable paresthesias in the painful foot or limb. The clinical effects were tested during a trial period (2 weeks) and in case that the patient had a significant pain relief, an implantable pulse generator (Itrel II IPG, Medtronic Inc. Minneapolis USA) was placed in an abdominal subcutaneous pocket. Various settings of the active quadripolar extremities of the electrode were studied to evaluate the most appropriate combination for the patient pain relief. Bipolar stimulation was assessed by the presence of paresthesias and a feeling of warmth in the painful area. The setting parameters were pulse amplitude between 1.0 and 5.0 V, frequency between 40 and 120 pps and pulse width from 150 to 450 sec. Stimulation was continuous 24 hours a day to obtain maximum information on the clinical outcome. During follow up the parameters of stimulation can be reset
TABLE 1. CLINICAL RESULTS AND PAIN RELIEF AFTER SPINAL CORD STIMULATION (SCS) TREATMENT SCS Therapy Success
Rest pain, 39 patients Trophic lesions ⬍3 cm2, 58 patients Trophic lesions ⬎3 cm2, 53 patients Overall series
Partial Success
Failure
Nⴗ
%
Nⴗ
%
Nⴗ
%
11 19 8 38
(79%) (68%) (37%) (59%)
1 4 4 9
(7%) (14%) (18%) (14%)
2 5 10 17
(14%) (18%) (45%) (27%)
Note: The diabetic patients with therapy success achieved a pain relief ⬎75% and long-term limb salvage, while these with partial success achieved a pain control ⬎50% with or without use of analgesics and limb salvage for at least 6 months. Patients with SCS failure (due also to technical problems) were submitted in major amputations after a short period of time from the implantation.
296
PETRAKIS AND SCIACCA
J Diab Comp 1999; 13:293–299
TABLE 2. MEAN (SD) OF TcPO2 VALUE CHANGES IN THE DIABETIC PATIENTS WITH LIMB SALVAGE FOR AT LEAST 6 MONTHS (THERAPY SUCCESS AND PARTIAL SUCCESS) BEFORE TREATMENT, AFTER TEMPORARY IMPLANTATION, 2 MONTHS AFTER PERMANENT IMPLANTATION, AND AFTER LONG-TERM FOLLOW-UP TcPO2 Changes, mm Hg Rest pain patients Trophic lesions ⬍3 cm2 patients Trophic lesions ⬎3 cm2 patients
Before Treatment
After Temporary Implantation
p Value
2 Months
Long Term
p Value
22.1 (5.9) 15.8 (6.6) 12.1 (5.1)
30.5 (6.6) 22.7 (5.3) 17.36 (4.7)
⫽0.04 ⫽0.03 ⫽0.03
34 (5.4) 24.2 (5.2) 19.1 (4.3)
43.1 (5.7) 36.4 (6.6) 28.1 (6.8)
⬍0.01 ⬍0.02 ⬍0.01
Among the 22 diabetic patients with trophic lesions greater than 3 cm2, pain relief greater than 75% with lesion healing was obtained in 8 patients (40%), while in 4 patients (20%, two of them underwent minor amputations), a partial success, with pain relief greater than 50% and limb salvage for at least 6 months (mean, 9 months; range, 6–12 months) was achieved. After this period, the ischemic pain recurred and major amputations were unavoidable. Eight patients treated with SCS, initially presented a pain relief less than 50%, but after 2 months, analgesic medication was necessary for the pain control, however, within the first 6 months from the device implantation all these patients were submitted to major amputations, because severe ischemic pain recurred (Table 1). In two patients, the electrode was either dislocated or infected, and was removed, with failure of the SCS therapy. TcPO2 foot values increased from 22.1 ⫾ 5.9 mm Hg to 43.1 ⫾ 5.7 mm Hg (p ⬍ 0.01) in the patients with rest pain and limb salvage. In the patients with trophic lesions less than 3 cm2 and limb salvage TcPO2 mean foot values showed an average improvement from 15.8 ⫾ 6.6 to 36.4 ⫾ 6.8 mm Hg (p ⬍ 0.02). In the patients with trophic lesions greater than 3 cm2 (with dry gangrene) and limb salvage TcPO2 mean foot values showed an average improvement from 12.1 ⫾ 5.1 to 28.1 ⫾ 6.8 mm Hg (p ⬍ 0.01), (Table 2). TcPO2 changes are related with the presence of adequate paresthesias in the painful area during the trial period. Major changes in TcPO2 values were achieved within the first 2 weeks after temporary implantation, while an increasing of more than 50% in the first 2 months
after implantation was revealed in the patients with limb salvage, (Table 2). Changes of TcPO2 in all amputated patients where SCS failed to save the limb, were not significant even after 2 or 4 weeks from the device implantation (Table 3). No significant modifications were observed in the ABI or toe pressure values before permanent implantation (test period) and during the follow-up. In our casuistic we report only one dislocation of the electrode, during the permanent implantation procedure, in two patients the electrode or the generator was infected and required removal of the device. During our follow up, 1 electrode had been dislocated in the early post-operative period. Limb salvage, for at least 6 months (SCS therapy success and partial success), was achieved in 47 patients (73%) while in 17 patients (25%) major amputations within the first six months after implantation, due to SCS failure or technical problems with device removal, were unavoidable. After a mean period of 3 years 8 patients were submitted to substitution of exhausted generator. DISCUSSION Peripheral angiopathy and neuropathy is a common long-term diabetes complication.20,21 About 7.5% of unselected adults attending a hospital diabetic clinic have painful neuropathic symptoms, mainly in the lower limbs.22 Pain varies from mild paresthesias in a few toes to severe unremitting pain in both legs.22,23 Nighttime exacerbation of the pain plus contact hypersensitivity to bed-cloths results in loss of sleep, and pain in diabetic neuropathy can be disabling.23 The cause of
TABLE 3. MEAN (SD) of TcPO2 VALUE CHANGES BEFORE TREATMENT, AFTER TEMPORARY IMPLATATION AND 2 MONTHS AFTER PERMANENT IMPLANTATION, IN THE DIABETIC PATIENTS WITH FAILURE OF THE SCS TO SALVAGE THE LIMB FOR AT LEAST 6 MONTHS TcPO2 Changes, mm Hg Rest pain patients Trophic lesions ⬍3 cm2 patients Trophic lesions ⬎3 cm2 patients * Not significant.
Before Treatment
After Temporary Implantation
p Value
2 Months
p Value
20.1 (4.6) 15.1 (5.1) 9.1 (5.6)
23.4 (4.4) 17.4 (4.8) 10.6 (4.9)
NS* NS NS
26.2 (5.2) 18 (3.7) 12.2 (5.8)
NS NS NS
J Diab Comp 1999; 13:293–299
chronic sensory-motor diabetic neuropathy or indeed neuropathic pain is not known although metabolic and microvascular systems may be involved.24,25 While the search for potential therapeutic agents to halt or reverse the neuropathic process continues,26 current treatment is largely aimed at relieving painful symptoms. The SCS offers a new and effective way of relieving chronic diabetic neuropathic and ischemic pain and improves exercise tolerance. The contemporary literature reports demonstrate the efficacy of SCS in relieving ischemic pain.27–29 Some authors showed that in PAOD patients, SCS recruits small capillaries not ordinarily perfuse, enhancing skin blood flow, improvements that may explain the beneficial clinical effects experienced by these patients.15,16 However, their data and the use of SCS in the treatment of POAD were recently criticized, maybe because there are not standard indications for the SCS implantation to obtain the best results. Thus contrary to the recommendations of other authors in our opinion SCS should not be used to treat patients affected by extensive gangrenous lesions of the foot (classified in Fontaine’s IVb stage). For these patients, nonresponders to medical treatment neither amenable to vascular reconstructions, the recommended treatment is primary amputation. In the presence of severe nociceptive somatic pain the analgesia associated with SCS is less effective than obtained with epidural anesthetics with or without opiates. Concerning patients affected by only claudication, we recommend that device implantation is indicated if pain-free walking interval is less than 50 meters (Fontaine’s IIb stage patients) and if the other therapeutic options have definitely been excluded. The presence of diabetes does not represent a contraindication to SCS use. The exact underlying mechanism of how SCS influences chronic and ischemic pain remain unclear. The gate-control theory of pain, proposed by Melzack and Wall30 in 1965, has been the one widely used to explain the action of SCS. This theory postulated that an inhibitory mechanism in the dorsal horn is activated by the recruitment of large-diameter fibers. The substantia gelatinosa activation induces inhibition of second order neurons processing nociceptive information. Presently, the mechanisms proposed to be responsible for pain relief are neurophysiologic and neurochemical in nature. The neurophysiology of pain relief under SCS varies from simple blocking of pain transmission by a direct effect on spinothalamic tracts, segmental inhibition via coarse fiber activation, effects on central sympathetic system, and brain stem loops to thalamocortical mechanisms.31 The inhibitory effects of SCS on the transmission of nociceptive impulses may be exerted segmentally in the spinal cord and/or at a supraspinal level.32 Clinical observations indicate that the mechanisms involved in the stimulation-induced relief ischemic pain are different from those related to relief of
SCS IN DIABETIC ISCHEMIA
297
others types of pain.33 Indeed, nociceptive pain is more resistant to SCS, and significant pain relief is almost never obtained before a couple of days. In contrast, neuropathic pain of peripheral origin responds well and directly to SCS. Both components, nociceptive as well as neuropathic are present in ischemic pain. Several authors have postulated that the principal factor in the relief of ischemic pain is the inhibition of the pain signal per se, leading to a decrease in sympathetic activity and improved skin microcirculation.8,34 Another hypothesis is that SCS depresses autonomic sympathetic activity.33,35 However, many authors have reported success even after sympathectomy.8,14 SCS has been shown to be useful in vasospastic disease and reflex sympathetic dystrophy.34 It is necessary that during the testing period of about 2 weeks, in addition to pain control, we assess carefully the effect on peripheral blood flow to ascertain if the warmth feeling and the paresthesias reported by the patient are related to increased skin microcirculation. This assessment should include determination of the pain-free walking interval under standard conditions (treadmill), confirmation of ulcer healing (surface measurements), and verification of improved blood flow. The decrease of ischemic pain with SCS is probably secondary to the positive effects on microcirculation rather than vice versa. Pain relief may also result in attenuation of sympathetic activity with vasodilatation, leading to further pain relief.32 The most useful instrumental technique for peripheral vascular screening during SCS, TcPO2, a noninvasive method, is suitable for accessing skin circulation8,14,15–19,35,36 and has been used to evaluate microcirculatory changes induced by SCS. In PAOD, the aim of the SCS is not only to achieve effective analgesia (which might be obtained by others less expensive techniques), but also to promote the trophic-functional recovery of the body segment affected by an advanced ischemic process. Furthermore, in agreement with other authors, SCS should be considered in patients with neuropathic pain who do not respond to conventional treatment,37 but we believe that diabetic patients with severe neuropathic pain and advanced PAOD should be excluded. In fact, diabetic patients with neuropathy and extended lesions (greater than 3 cm2), in our experience, after stimulation have limited or no TcPO2 increase and the method fails in the most of the cases. In this study, we found that all 64 implanted diabetic patients, during the test-period stimulation, noted immediate significant pain relief, while only in those patients with limb salvage, TcPO2 value changes showed a significant increase following stimulation. Two months after implantation a TcPO2 increase of more than 50% was achieved only in the patients with limb salvage and was predictive of SCS success. Moreover, we demonstrated that the differences of the TcPO2 values before
298
PETRAKIS AND SCIACCA
and after implantation are more important for the SCS success and limb salvage, rather than the stage of the disease. Little or no TcPO2 changes in diabetic patients after implantation, independently from the stage of the disease, are predictive of SCS failure and should be taken in consideration prior to implantation, in terms of cost effectiveness. After SCS, TcPO2 on the dorsum of the foot significantly increased, whereas the ABI did not change. This is not surprising because patients treated with SCS had crural vessels that were unsuitable for reconstruction. The changes in TcPO2 related to increase skin perfusion were not result of improved arterial inflow. The feeling of warmth in the painful area under SCS was related to increase in TcPO2. TcPO2 measurement is able to evaluate the effects of SCS on microcirculatory blood flow and represents a predictive factor of guarantee success as an increase in TcPO2 greater than 50% in the first 2 months correlates with a positive result. Although the initial pain control after SCS, the current study suggests 2 weeks of temporary implantation to evaluate the paresthesias, the warmth feeling and the TcPO2 changes of the limb. Thus in case of limited TcPO2 changes in PAOD diabetic patients with crural anelastic vessels without any compliance the method fails. This is important for the final decision, in terms of cost effectiveness, in all the cases where the failure of the SCS is quite sure after a short period of time. In conclusion, in PAOD diabetic patients with or without ischemic foot lesions, after failed medical or surgical therapy, limb salvage was achieved after SCS therapy, in those with significant TcPO2 increase. Except the pain control, clinical improvement was associated with a significant increase in TcPO2 within the first post-implanted period. Effects of SCS on macrocirculation were not significant. Relevant research data are needed to show how SCS affects pain and to explain the effects on the vascular system. We suggest that SCS is promising in diabetic patients with PAOD when previous therapies have failed. However, a 2-week test period is important to appropriately select those patients with significant TcPO2 increase, paresthesias and clinical warmth feeling, for permanent implantation. REFERENCES 1. Friedman SG, Kerner BA, Friedman MS, Moccio CG: Limb salvage in elderly patients: Is aggressive surgical therapy warranted? J Cardivasc Surg 30:848–851, 1989. 2. Gregg RO: Bypass or amputation: Concomitant review of bypass arterial grafting and major amputation. Am J Surg 149:397–402, 1985. 3. Shealy CN, Mortimer JT, Reswick JB: Electrical inhibition of pain by stimulation of the dorsal columns: Preliminary clinical report. Anesth Analg 46:489–491, 1967. 4. Meglio M, Cioni B, Rossi GF: Spinal cord stimulation
J Diab Comp 1999; 13:293–299
in management of chronic pain: A 9-years experience. J Neurosurg 70:519–524, 1989. 5. Cook AW: Electrical stimulation in multiple sclerosis. Med Biol Eng Comput 18:48–56, 1980. 6. Augustinsson LE: Epidural spinal electrical stimulation in peripheral vascular disease. Pace 10:205–206, 1987. 7. Broggi G, Servello D, Franzini A: Spinal cord stimulation for treatment of peripheral vascular disease. Appl Neurofisiol 50:439–441, 1987. 8. Broseta I, Barbera I, De Vera JA: Spinal cord Stimulation in peripheral arterial disease. J Neurosurg 64:71–80, 1986. 9. Dooley D, Kasprak M: Modification of blood flow to the extremities by electrical stimulation of the nervous sistem. South Med J 69:1309–1311, 1976. 10. Fiume D: Spinal cord stimulation in peripheral vascular pain. Appl Neurophisiol 46:290–294, 1983. 11. Groth KE: Spinal cord stimulation for the treatment of peripheral vascular disease. European Multicenter Study, in Fields H (ed.), Advances in Pain Research and Therapy. New York, Raven, 1985, pp. 861–870. 12. Horsch S, Claeys L: Epidural spinal cord stimulation in the treatment of severe peripheral arterial occlusive disease. Ann Vasc Surg 8:468–474, 1994. 13. Visconti W, Fontana P, Buonocuore N, Grillo N, Seno S: Spinal cord stimulation in advanced intractable chronic obliterant arteriopathies. Minerva Cardioangiol 44:19–27, 1996. 14. Jacobs MHJM, Joming PJG, Soures RJ, Kitslaar PJEHM, Slaaf DW, Reneman RS: Epidural spinal cord electrical stimulation improves microvascular blood flow in severe limb ischemia. Ann Surg 207:179–183, 1988. 15. Jacobs MJ, Joming PJ, Beckers RCY, Ubbink DT, van Kleef M: Foot salvage and improvement of microvascular blood flow as a result of epidural spinal cord electrical stimulation. J Vasc Surg 12:354–360, 1990. 16. Franzeck UK, Talke P, Bernstein EF, Colbranson FL, Fronek A: Transcutaneous PO2 measurements in health and peripheral arterial occlusive disease. Surgery 91: 156–163, 1982. 17. Hauser CL, Shoemaker WC: Use of transcutaneous PO2 regional perfusion index to quantify tissue perfusion in peripheral vascular disease. Ann Surg 197:337–343, 1983. 18. Jaszezak P, Poulsen J: Estimation of blood flow in transcutaneous PO2 measurements. Acta Anaesthisiol Scand 27:174–180, 1983. 19. Sciacca V, Mignoli A, di Marzo L, Maggiore C, Fiume D, Cavallaro A: Predictive value of transcutaneous oxygen tension measurements in the indication for spinal cord stimulation in patients with peripheral vascular disease: Preliminary results. Vasc Surg 3:128–132, 1989. 20. Clyne CAC, Ryas J, Webster JHH, et al.: Oxygen tension on the skin of ischermic legs. Am J Surg 143:315–318, 1982. 21. Tesfaye S, Malik R, Ward JD: Vascular factors in diabetic neuropathy. Diabetologia 37:847–854, 1994. 22. Chen AW, MacFarlane IA, Bowsher DR, Wells JC, Bessex C, Griffiths K: Chronic pain in patients with diabetes mellitus; comparison with non-diabetic population. Pain Clin 3:147–159, 1990.
J Diab Comp 1999; 13:293–299
23. Watkins PJ: Pain and diabetic neuropathy. BMJ 288:168– 169, 1984. 24. Editorial. Pain perception in diabetic neuropathy. Lancet i:83–85, 1985. 25. Chen AW, Mac Farlane IA, Bowsher DR, Wells JDC: Does acute hyperglycaemia influence heat pain thresholds? J Neurol Neurosurg Psychiatry 51:688–690, 1988. 26. Cameron NE, Cotter MA: Potential therapeutic approaches to the treatment or prevention of diabetic neuropathy: evidence from experimental studies. Diabetic Med 10:593–605, 1993. 27. Tesfaye S: Diabetic neuropathy: Current treatment and potential therapeutic approaches. Diab Nutr Metab 7: 375–379, 1994. 28. Augustinsson LE, Holm J, Carlsson CA, Jivegard L: Epidural electrical stimulation in severe ischemia: Evidence of pain relief, increased blood flow and a possible limb-saving effect. Ann Surg 1:354–349, 1985. 29. Jivegard LE, Augustinsson LE, Holm J, Risberg B, Ortenwall P: Effects of spinal cord stimulation (SCS) in patients with inoperable severe lower limb ischemia: A Prospective randomized controlled study. Eur J Vasc Endov Surg 9:421–425, 1995. 30. Metzack R, Wall PD: Pain mechanisms: A new theory. Science 150:1971–1979, 1965.
SCS IN DIABETIC ISCHEMIA
299
31. Gybeis JM, Kupers R: Central and peripheral electrical stimulation of the nervous system in the treatment of chronic pain. Acta Neurochir 38 (suppl):65–75, 1987. 32. Linderoth B: Dorsal column stimulation and pain: Experimental study of putative neurochemical and neurophysiological mechanisms. Thesis. Stockholm, 1992, pp. 11–19. 33. Meyerson BA: Electrical stimulation of the spinal cord and brain, in Bonica JJ, Loeser JD, Chapman RC, et al. (eds.) The Management of Pain. Philadelphia, Lea and Febiger, 1990, pp. 1862–1877. 34. Hosobuchi Y: Treatment of ischemic pain by neurostimulation, in Lipton S, Tunks E, Zoopi M (eds.), Advances in Pain Research and Therapy, Vol, 13. New York, Raven, 1990, pp. 223–226. 35. Cina C, Katsamouris A, Megerman J, Brewster DC, Strayhorn EC, Robison JG, Abbott WM: Utility of transcutaneous oxygen tension measurements in peripheral arterial occlusive disease. J Vasc Surg 1:362–371, 1984. 36. Petrakis I, Sciacca V: Transcutaneous oxygen tension (TcPO2) in the testing period of spinal cord stimulation (SCS) in critical limb ischemia of the lower extremities. Int Surg 84:122–128, 1999. 37. Tesfaye S, Watt J, Benbow JS, Pang AK, Miles JC, MacFarlane IA: Electrical spinal-cord stimulation for painful diabetic peripheral neuropathy. Lancet 348:1696–1701, 1996.