PERIPHERAL ARTERIAL occlusive disease

Transcription

1 Dialysis Therapies Peripheral Arterial Occlusive Disease Is More Prevalent in Patients With Hemodialysis: Comparison With the Findings of Multidetector-Row Computed Tomography Koji Okamoto, MD, Machiko Oka, MD, Kyoko Maesato, MD, Ryota Ikee, MD, PhD, Tsutomu Mano, MD, Hidekazu Moriya, MD, Takayasu Ohtake, MD, PhD, and Shuzo Kobayashi, MD, PhD Background: Peripheral arterial occlusive disease (PAOD) influences the mortality of patients on hemodialysis therapy. Although the ankle-brachial pressure index (ABI) is used widely to detect PAOD, it yields false-negative results because of calcifications of vascular walls. To more accurately assess the prevalence of PAOD, we investigated which noninvasive method, among ABI, toe-brachial pressure index, transcutaneous PO 2, and skin perfusion pressure (SPP), had superior sensitivity and specificity to the others. Methods: Multidetector-row computed tomography was performed in 36 hemodialysis patients. We then compared the 4 noninvasive methods with findings of multidetector-row computed tomography and calculated the sensitivity and specificity of each method by means of receiver operating characteristic analysis. Irrespective of symptoms, PAOD is defined as the presence of complete obstruction in the case of lesions below the knee or more than 75% stenosis for lesions above the knee. Results: Seven of 36 patients (19.4%) had an ABI less than 0.9. Sensitivity of the ABI was only 29.9%, whereas an SPP set at 50 mm Hg was more accurate, with sensitivity of 84.9% and specificity of 76.9%. A total of 41.4% of patients had an SPP less than 50 mm Hg. For lesions located above the knee, toe-brachial pressure index provided sensitivity of 91.7% and specificity of 86.7%. Conclusion: SPP is the most useful tool for detecting PAOD in hemodialysis patients, with accuracy of 84.9%. There is a strong possibility that more patients than previously expected have early PAOD. Am J Kidney Dis 48: by the National Kidney Foundation, Inc. INDEX WORDS: Skin perfusion pressure; multidetector-row computed tomography; peripheral arterial occlusive disease; hemodialysis (HD); diabetes mellitus. PERIPHERAL ARTERIAL occlusive disease (PAOD) influences the mortality of patients on hemodialysis (HD) therapy. 1 Early detection of PAOD is extremely important to improve the prognosis of these patients. In this regard, various noninvasive methods have been applied to the diagnosis of PAOD in patients. 2-4 The anklebrachial blood pressure index (ABI) correlated well with severity of PAOD symptoms and angiographic findings. 4-7 Therefore, the ABI is used as a standard tool for classifying the severity of PAOD or assessment of the presence of critical limb ischemia. 8,9 However, arterial rigidity associated with medial calcification may interfere with ABI measurement. In some cases, it may be difficult to detect isolated obstruction in 1 or even 2 of the 3 branches of the popliteal artery between the knee and ankle or obstruction in more distal vessels. 4,7 These false-negative results occur in 17% to 24% of diabetic limbs and also were reported in patients undergoing HD or in other elderly patients. 4,10,11 Although the usefulness of the toebrachial pressure index (TBI) or transcutaneous oxygen tension (tcpo 2 ) has been reported, 12 Castronuovo et al 13 found that skin perfusion pressure (SPP) is an objective and noninvasive method that can be used to diagnose critical limb ischemia with accuracy of approximately 80% in non-hd patients. In diabetic feet, SPP correlated well with TBI. 14 Unfortunately, it remains unclear which of the noninvasive methods, among ABI, TBI, tcpo 2, and SPP, is superior to the others, particularly in HD patients. The reason that ABI yields false-negative results in HD From the Department of Nephrology and Kidney and Dialysis Center, Shonan Kamakura General Hospital, Kamakura, Japan. Received January 17, 2006; accepted in revised form April 11, Originally published online as doi: /j.ajkd on June 20, Support: None. Potential conflicts of interest: None. Address reprint requests to Shuzo Kobayashi, MD, PhD, Vice President and Chairman, Department of Nephrology, and Kidney and Dialysis Center, Shonan Kamakura General Hospital, , Yamazaki, Kamakura , Japan. shuzo@shonankamakura.or.jp 2006 by the National Kidney Foundation, Inc /06/ $32.00/0 doi: /j.ajkd American Journal of Kidney Diseases, Vol 48, No 2 (August), 2006: pp

2 270 patients may occur in part because lesions in these patients often have arterial calcification in addition to their location in the more distal parts of the lower-limb arteries. In the present study, we compare the validity of various noninvasive measurements (ABI, TBI, tcpo 2, and SPP) in HD patients and assess sensitivity or specificity in comparison to multidetectorrow computed tomography (MDCT). In addition, we define the reference range of SPP to detect PAOD in HD patients. METHODS This cross-sectional study was performed in a single center. Subjects included 140 HD patients (94 men, 46 women) with a mean age of 67 years (range, 35 to 89 years). Patients were excluded if they had a previous history of amputation. Fifty-six patients with diabetes mellitus were included. We performed MDCT in 36 patients who gave informed consent. To study the sensitivity or specificity of various measurements (ABI, TBI, tcpo 2, and SPP), PAOD is defined as the presence of complete obstruction in the case of lesions below the knee (BK) or stenosis of more than 75% in the case of lesions above the knee (AK) on MDCT. ABI was measured in all patients by using an ABI-form (Colin, Tokyo, Japan) that simultaneously measured unilateral brachial pressure in the arm without an arteriovenous fistula and ankle blood pressure by using an oscillometric method. ABI was calculated as the ratio of ankle systolic pressure divided by brachial systolic pressure. TBI was measured by using photoplethysmography A pneumatic cuff was wrapped around the proximal phalanx of the great toe and a photoelectric probe was placed distal to the cuff. TBI was calculated as the ratio of this pressure divided by brachial systolic pressure. tcpo 2 was measured at the dorsum of the foot in the supine position by using a PO 2 monitor (Kohken Medical, Tokyo, Japan). Electrode temperature was set at 44 C at a constant room temperature of 23 C. 18 To measure SPP, we used a Laser Dopp PV2000 (Kaneka, Osaka, Japan) according to the method described by Castronuovo et al. 13 Briefly, the laser Doppler skin perfusion pressure transducer consists of a laser Doppler probe secured within the bladder of a blood pressure cuff that contains a transparent polyvinylchloride window so that microcirculatory perfusion measurements can be made during cuff deflation. To define a reference range for SPP, we measured it in 30 healthy individuals (mean age, years; range, 22 to 57 years) who were free of diabetes mellitus, hypertension, hyperlipidemia, or other disorders. To confirm reproducibility, SPP was measured repeatedly in 5 of these 30 healthy volunteers who were nonsmokers. Metatarsal SPP of these 5 healthy subjects aged 22 to 50 years (4 men, 1 woman) was measured 3 times by 3 different operators. Measurement of SPP and tcpo 2 was performed within 1 hour after an HD session, whereas measurement of ABI and OKAMOTO ET AL TBI was performed within 6 hours after an HD session. All 4 tests were performed on the same day after an HD session. Computed Tomographic Angiography In patients who had urine volume greater than approximately 300 ml/d, MDCT was performed immediately before HD, whereas in the others, MDCT was performed on the day of a mid-hd session. MDCT was performed with an 8-detector row computed tomographic (CT) scanner (Light Speed Ultra; GE Medical System, Tokyo, Japan). The scanning range was planned with a scout view and included the entire vascular tree from the abdominal aorta to the toes. First, a single nonenhanced low-dose scan (40 ma) of the abdominal aorta was obtained at the level of the origin of the renal arteries. On this image, a 10- to 15-mm 2 region of interest was placed in the lumen of the aorta. This region of interest was used to help optimize intraluminal contrast enhancement. Delay time from contrast material injection to scanning was determined individually and automatically for each patient by using a bolus-tracking technique. A total of 100 ml of contrast media (Iopamiron 370, Tokyo, Japan) was administered with an automated injector (dual shot; Nemoto, Tokyo, Japan) at a flow rate of 3 ml/s through a 16- to 20-G needle that was placed in a superficial vein or blood access. At the same flow rate, 20 ml of saline was injected intravenously after contrast material administration. After peak concentration was reached in the region of interest placed in the aorta, CT scanning was initiated automatically 3 seconds later. During this 5-second period, a breathhold command was given to the patient. Data acquisition was performed in the craniocaudal direction by using 1.25-mm thick sections, table feed of 0.3 mm/rotation, and a 0.5- second gantry rotation period. Tube voltage setting was 120 kv, and tube current was 300 ma. All abdominal scans were obtained during breath holds. Images of the legs were reconstructed with a slice width of 1.25 mm and interval of 1.25 mm. CT data were transferred to an Advantage Windows workstation (Amin Co, Ltd, Tokyo, Japan) to create volume-rendered images of vessels of the lower limbs. All reconstructions were performed by the same technician, who had 2 years of experience in 3-dimensional postprocessing techniques. Display parameters, including width, level, attenuation, and brightness, were visually optimized with the first volume-rendering process and saved as preset values for all subsequent vessel reconstructions in the same patient. We defined the area from the iliac artery to the knee joint as above the knee (AK) and everything distal to the knee joint as below the knee (BK). For AK vessels, stenosis occupying more than 75% of the vascular lumen was defined as significant PAOD. In the case of BK vessels, complete obstruction with blocking of contrast media was defined as significant. MDCT images were analyzed in 36 patients (72 limbs) who gave informed consent. Coronary artery disease is defined as a previous history of angina pectoris or myocardial infarction. Cerebrovascular disease is defined as a previous history of clinical cerebral infarction or hemorrhage. Duration of renal insufficiency is defined as the period from first detection of a serum creatinine level greater than 1.3 mg/dl ( 115 mol/l).

3 PERIPHERAL ARTERIAL OCCLUSIVE DISEASE IN HD PATIENTS 271 Table 1. Basic Characteristics of 140 Patients and 36 Patients Who Underwent MDCT All Patients (n 140) Patients Who Underwent MDCT (n 36) Age (y) (33 88) Duration of HD (y) (0.5 40) Duration of renal insufficiency (y) * (1 40) ABI (20.0) 7 (19.4) Claudication ( 300 m) 12 (8.6) 6 (16.7) Coldness 70 (50.0) 19 (52.8) Fatigue * 14 (38.9) Roll nail * 10 (27.8) Pain 3 (21) 2 (5.6) Numbness * 7 (19.4) Diabetes mellitus 56 (40.0) 15 (41.7) Coronary disease 35 (25.0) 7 (19.4) Cerebrovascular disease 28 (20.0) 2 (5.6) Hypertension 120 (85.7) 20 (55.6) NOTE. Values expressed as mean SD (range) or number (percent). Coronary disease indicates previous history of angina pectoris or myocardial infarction, and cerebrovascular disease indicates previous history of clinical cerebral infarction or hemorrhage. Values expressed as mean 1 SD (range). *Undetermined. Statistical Analysis Data are presented as mean SD. Unpaired t-test was used for between-group comparisons. Prevalence data were analyzed by means of chi-square or Fisher exact test, as appropriate. Receiver operating characteristic curve analysis was performed to assess sensitivity or specificity (SPSS II, version 11.0, SPSS, Tokyo, Japan). To compare significance between more than 2 groups, 1-way analysis of variance was performed, followed by the Student-Newmann-Keuls test. Average SPP of each subject measured by the same operator was subjected to 2-way analysis of variance to assess reproducibility. To analyze the risk factor for extension of arterial stenosis, we used probit regression analysis. The dependent factor was the length of the stenotic lesions classified as follows: below-the-ankle (BA) stenosis equals 1, BK and BA stenosis equal 2, and AK, BK, and BA stenosis equal 3. Sex and the existence of coronary disease were qualitative factors. Statistical significance is defined as P less than RESULTS Of 140 HD patients, 36 patients gave informed consent. Basic characteristics of these 36 patients who underwent MDCT are listed in Table 1. The prevalence of diabetes mellitus was 41.7%. As shown in Fig 1, 12 limbs of the 36 HD patients had AK stenosis, 23 limbs had BK stenosis, and 11 limbs had BA stenosis. All stenotic lesions Fig 1. Distribution of arterial stenosis in lower limbs.

4 272 OKAMOTO ET AL Table 2. Inspection Modality, Symptoms, and Previous History as a Predictor of Arterial Stenosis on MDCT in the Serial Portion of the Leg in 72 Limbs in 36 Patients No Stenosis AK Stenosis BK BA Stenosis No Stenosis v AK BK BA Stenosis AK Stenosis v BK BA Stenosis No. of limbs Age (y) NS NS Duration of HD (y) NS NS Duration of renal insufficiency (y) NS NS ABI TBI SPP tcpo NS Claudication 800 m (%) Coldness (%) NS NS Fatigue (%) Roll nail (%) NS NS Pain (%) NS NS Ulcer (%) NS NS Numbness (%) NS 0.05 Coronary disease (%) Cerebrovascular disease (%) NS NS Diabetes mellitus (%) NS NS Hypertension (%) NS NS NOTE. N 72. No stenosis indicates no arterial stenosis on MDCT, coronary disease indicates previous history of angina pectoris or myocardial infarction, and cerebrovascular disease indicates previous history of clinical cerebral infarction or hemorrhage. Abbreviation: NS, not significant. were located in the distal portion of the popliteal artery. There were no significant differences in age or duration of HD therapy between the groups with and without PAOD or between the AK and BK plus BA stenosis groups. There were significant differences in ABI, TBI, SPP, and tcpo 2 values, as well as differences in previous histories of cardiovascular disease, claudication, and fatigue, between the 2 groups (Table 2). Thirteen of 72 limbs examined showed an ABI less than 0.9. Receiver operating characteristic analysis showed that sensitivity of an ABI set at 0.9 for detecting PAOD was only 29.9%, although specificity was 100.0% (positive predictive value, 1.000; negative predictive value, 0.475; Fig 2). For AK lesions, sensitivity increased to 75.0% with specificity of 94.2% (positive predictive value, 0.692; negative predictive value, 0.949; Fig 3). Fifteen of the 72 limbs showed a TBI less than 0.6. Likewise, receiver operating characteristic analysis showed that sensitivity of a TBI set at 0.6 for detecting PAOD was only 45.2%, although specificity was 100% (positive predictive value, 1.000; negative predictive value, 0.528; Fig 2). For AK lesions, sensitivity of TBI increased to 91.7%, with specificity of 86.7% (positive predictive value, 0.579; negative predictive value, 0.660; Fig 3). Thirty-five of 72 limbs showed a tcpo 2 less than 50 mm Hg. Sensitivity of a tcpo 2 of 50 mm Hg for detecting PAOD was 61.1%, and specificity was 70% (positive predictive value, 0.765; negative predictive value, 0.526; Fig 2). To detect PAOD located AK, tcpo 2 was not significant (Fig 3). Forty of 72 limbs showed an SPP less than 50 mm Hg. Sensitivity of SPP of 50 mm Hg for detecting PAOD was 84.9%, and specificity was 76.6% (positive predictive value, 0.888; negative predictive value, 0.734; Fig 2). If the cutoff value for SPP was set at 45 mm Hg, sensitivity decreased to 78.6%, but specificity improved to 91.6%. For AK lesions, sensitivity of SPP set at 50 mm Hg was 100.0%, and specificity was 48.9% (positive predictive value, 0.300; negative predictive value, 1.000; Fig 3). Conversely, for

5 PERIPHERAL ARTERIAL OCCLUSIVE DISEASE IN HD PATIENTS 273 Fig 2. Receiver operating characteristic analysis of modality as a predictor of BK stenosis on MDCT. BK lesions, SPP greater than 50 mm Hg showed 84.8% sensitivity and 59.5% specificity. As shown in Fig 4, the 4 kinds of noninvasive tests showed value distribution according to the presence or absence of limb stenosis that allowed the findings of MDCT to be recognized. Next, we tried to measure ABI, TBI, tcpo 2, and SPP in all 280 limbs of the 140 HD patients. Although we could not measure SPP Fig 3. Receiver operating characteristic analysis of modality as a predictor of AK stenosis on MDCT.

6 274 OKAMOTO ET AL Fig 4. Four kinds of noninvasive measurements show the distribution of each value according to the presence or absence of limb stenosis to recognize findings of MDCT at a glance. Abbreviation: No stenosis, no arterial stenosis on MDCT. in 14 limbs because of involuntary movement and variations in cuff pressure, 110 of the remaining 266 limbs (41.4%) showed SPP less than 50 mm Hg (Fig 5). Factors independently associated with longer PAOD lesions were examined by means of probit regression analysis. In patients without diabetes mellitus, durations of hypertension and HD therapy were selected as independent risk factors for PAOD. In patients with diabetes mellitus, duration of renal insufficiency was selected, as listed in Table 3. Repeated measurement of SPP in 5 healthy volunteers by 3 well-trained operators showed that intraobserver variance was 4.81% and interobserver variance was 2.79%. In 30 healthy Fig 5. One hundred ten of 266 limbs (41.4%) showed an SPP less than 50 mm Hg.

7 PERIPHERAL ARTERIAL OCCLUSIVE DISEASE IN HD PATIENTS 275 Table 3. Probit Regression Analysis of Factors That Affect Artery Stenosis Progression in Patients With and Without Diabetes Mellitus Variables No Diabetes Mellitus Diabetes Mellitus Sex ( ) ( ) Age ( ) ( ) Duration of diabetes mellitus ( ) Duration of hypertension 1.039* ( ) ( ) Duration of renal insufficiency ( ) ( ) Duration of HD 1.080* ( ) ( ) P NOTE. The dependent factor is the semi-quantity of length of stenosis lesion: BA stenosis equals 1, BK plus BA stenosis equals 2, and AK, BK, and BA stenosis equals 3. Sex is a qualitative factor. *P 0.05 (inadequate model). P individuals, the reference range of SPP was determined to be mm Hg. DISCUSSION PAOD recently has attracted considerable attention as a risk factor for adverse outcomes in HD patients. The incidence of nontraumatic lower-extremity amputation in HD patients is approximately 10 times greater than that in non-hd patients. 19 PAOD causes significant morbidity and mortality in HD patients, and death occurs within 2 years after amputation in 73% of HD patients who lose a limb versus 21% of amputees without end-stage renal disease. 19 To improve outcomes for HD patients, it is extremely important to devise intervention strategies for PAOD. Because the outcome for patients with foot gangrene is poor, it is important to detect PAOD as early as possible. However, current noninvasive tests, such as ABI, are inadequate for the diagnosis of PAOD, 20,21 particularly in HD patients. Because of the high prevalence of stenoses at more distal locations associated with arterial calcification, systolic blood pressure measured by using a cuff gives a reading higher than the actual pressure. Sensitivity of an ABI less than 0.9, which is the standard cutoff value for possible diagnosis of PAOD, 22 was only 29.9% in the present study. The distribution curve shows that the percentage of HD patients with an ABI less than 0.9 was 16.7% in our hospital, whereas the percentage with an ABI greater than 1.3 considered to be patients with calcified lesions was 45%. Conversely, the percentage of patients who had an SPP less than 50 mm Hg was 41.4%. Although there are other methods, such as TBI or tcpo 2, it was unclear which test is superior to the others. In the present study, we show that SPP is a more sensitive and specific method for detecting PAOD in patients with HD, possibly because SPP shows the final pathway of capillary flow through the skin with the laser Doppler probe. Measurement of SPP by using the laser Doppler method was developed originally by Castronuovo et al, 13 who concluded that SPP measurement is an objective noninvasive method that can be used to diagnose critical limb ischemia with accuracy of approximately 80%. Moreover, in a comparison of healing outcome with SPP in patients managed with local débridement, minor amputation, or both, 30 mm Hg of SPP is the prediction factor of healing of the amputation edge. 23,24 We show that SPP of 50 mm Hg is the cutoff value for detecting PAOD in HD patients with high accuracy (sensitivity, 84.9%; specificity, 76.9%). However, in case of AK lesions, TBI may be superior to the others. Based on the reference range of SPP (50 mm Hg), the prevalence of PAOD in our patients was estimated to be 41.4%, a considerably greater rate than that of 13.5% in a previous report. 25 Of patients with an SPP less than 50 mm Hg, 50.0% (55 of 110 limbs) had no symptoms or only foot coldness or numbness (Fontaine I). Although we cannot directly compare MDCT findings with angiography, we believe that MDCT provides clear images, and its delineation of arterial lesions is useful to avoid false-negative results. For our definition of PAOD, we adopted complete obstruction in the case of BK lesions because it may be difficult to detect lesions in the more

8 276 distal arteries of the lower limb. Therefore, lesions with mild stenosis may have a greater prevalence. The rate of consent to MDCT was low in this study because it is not covered by the national health scheme; thus, the present study may have limitations regarding patient selection. In patients with diabetes mellitus, duration of renal dysfunction was the risk factor for progressive artery stenosis. In patients without diabetes mellitus, durations of hypertension and HD therapy were independent risk factors. However, there was no difference in prevalence of PAOD between patients with and without diabetes mellitus; therefore, we need to carefully monitor patients on HD therapy even if they do not have diabetes mellitus. In conclusion, we compared various noninvasive measurements of limb blood flow to assess their sensitivity and specificity for detecting PAOD in HD patients. We conclude that SPP is the most useful tool with respect to both sensitivity and specificity for early detection of PAOD in these patients. There also is a strong possibility that many HD patients have early PAOD. REFERENCES 1. Jaar BG, Astor BC, Berns JS, Powe NR: Predictors of amputation and survival following lower extremity revascularization in hemodialysis patients. Kidney Int 65: , Sumner R: Noninvasive assessment of peripheral arterial occlusive disease, in Rutherford RB (ed): Vascular Disease (ed 3). Philadelphia, PA, Saunders, 1989, pp Prineas RJ, Harland WR, Janzon L, Kannel W: Recommendations for use of non-invasive methods to detect atherosclerotic peripheral arterial disease in population studies. American Heart Association Council on Epidemiology. Circulation 65:1561A-1566A, Carter SA: Role of pressure measurements in vascular disease, in Berstein EF (ed): Vascular Diagnosis (ed 4). St Louis, MO, Mosby, 1993, pp Yao ST, Hobbs JT, Irvine WT: Ankle systolic pressure measurements in arterial disease affecting the lower extremities. Br J Surg 56: , Carter SA, Lezack JD: Digital systolic pressures in the lower limb in arterial disease. Circulation 43: , Carter SA: Clinical measurement of systolic pressures in limbs with arterial occlusive disease. JAMA 207: , Rutherford RB, Baker JD, Ernst C, et al: Recommended standards for reports dealing with lower extremity ischemia: Revised version. J Vasc Surg 26: , European Working Group on Chronic Critical Leg Ischemia: Second European Consensus Document on Chronic OKAMOTO ET AL Critical Leg Ischemia. Circulation 84:Siv1-Siv26, 1991 (suppl 4) 10. Larsson JAJ, Castenfors J, Agardh CD, Stenstrom A: Distal blood pressure as a predictor for the level of amputation in diabetic patients with foot ulcer. Foot Ankle 14: , Carter SA: Ankle and toe systolic pressures comparison of value and limitations in arterial occlusive disease. Int Angiol 11: , Holstein P: Ischaemic wound complications in aboveknee amputations in relation to the skin perfusion pressure. Prosthet Orthot Int 4:81-86, Castronuovo JJ Jr, Adera HM, Smiell JM, Price RM: Skin perfusion pressure measurement is valuable in the diagnosis of critical limb ischemia. J Vasc Surg 26: , Tsai FW, Tulsyan N, Jones DN, Abdel-Al N, Castronuovo JJ Jr, Carter SA: Skin perfusion pressure of the foot is a good substitute for toe pressure in the assessment of limb ischemia. J Vasc Surg 32:32-36, Fronek A, Blazek V, Curran B: Toe pressure determination by audiophotoplethysmography. J Vasc Surg 20: , Duprez D, Missault L, Van Wassenhove A, Clement DL: Comparison between ankle and toe index in patients with peripheral arterial disease. Int Angiol 6: , Bone GE, Pomajzl MJ: Toe blood pressure by photoplethysmography: An index of healing in forefoot amputation. Surgery 89: , Franzeck UK, Talke P, Bernstein EF, Golbranson FL, Fronek A: Transcutaneous PO 2 measurements in health and peripheral arterial occlusive disease. Surgery 91: , O Hare A: Lower-extremity peripheral arterial disease among patients with end-stage renal disease. J Am Soc Nephrol 12: , McLafferty RB, Moneta GL, Taylor LM Jr, Porter JM: Ability of ankle-brachial index to detect lower-extremity atherosclerotic disease progression. Arch Surg 132: ; discussion , Leskinen Y, Salenius JP, Lehtimaki T, Huhtala H, Saha H: The prevalence of peripheral arterial disease and medial arterial calcification in patients with chronic renal failure: Requirements for diagnostics. Am J Kidney Dis 40: , TransAtlantic Inter-Society Consensus (TASC): Management of peripheral arterial disease (PAD). Eur J Vasc Endovasc Surg 31:S1-S278, 2000 (suppl 1) 23. Faris I, Duncan H: Skin perfusion pressure in the prediction of healing in diabetic patients with ulcers or gangrene of the foot. J Vasc Surg 2: , Ikuo STY, Jun T, Toshiki N, Hiroyuki I, Minoru H, Takashi O: Noninvasive diagnostic modality for chronic arterial occlusive disease. J Jpn Coll Angiol 43: , Nakamura S, Sasaki O, Nakahama H, Inenaga T, Kawano Y: Clinical characteristics and survival in end-stage renal disease patients with arteriosclerosis obliterans. Am J Nephrol 22: , 2002