Association of Diabetes and Hemodialysis with Ankle Pressure and Ankle-Brachial Index in Japanese Patients with Critical Limb Ischemia

Clinical Care/Education/Nutrition/Psychosocial Research ORIGINAL ARTICLE

Association of Diabetes and Hemodialysis With Ankle Pressure and Ankle-Brachial Index in Japanese Patients With Critical Limb Ischemia

1 1 MITSUYOSHI TAKAHARA, MD TAKA-AKI MATSUOKA, MD, PHD On the other hand, it has often been 1 2 HIDEAKI KANETO, MD, PHD MASAHIKO IKEDA, MD, PHD 2 1 pointed out that patients with diabetes OSAMU IIDA, MD IICHIRO SHIMOMURA, MD, PHD 1 andrenalfailurewouldhavetheirankle NAOTO KATAKAMI, MD, PHD systolic blood pressure and ABI falsely elevated because of their noncompressible fi OBJECTIVEd vessels caused by vascular calci cation To investigate whether diabetes and regular hemodialysis are associated with (2,8). As previous studies showed, ele- false elevation of ankle systolic blood pressure and ankle-brachial systolic pressure index (ABI) because of their arterial calcification in patients with critical limb ischemia (CLI). vated ABI indicates a clinically important characteristic and has attracted an RESEARCH DESIGN AND METHODSdWe recruited 269 Japanese patients who un- increasing attention in clinical practice derwent endovascular therapy for CLI. Ankle systolic blood pressure and ABI were assessed (9–11). before endovascular therapy. Arterial stenosis and calcification were evaluated angiographically. No clinical data are so far available, We investigated the associations among clinical comorbidities, arterial calcification, and mea- however, about whether these falsely surements of ankle systolic blood pressure and ABI. elevated measurements could be simi- RESULTSdAnkle systolic blood pressure was 85 6 56 mmHg, and ABI was 0.59 6 0.37. larly observed in patients with critical Arterial calcification was observed in 69% of the patients. The prevalence of diabetes and regular limb ischemia (CLI). CLI is a manifestation hemodialysis was 71 and 47%. Diabetes and regular hemodialysis were both significantly asso- of PAD that describes patients with chronic ciated with the presence of arterial calcification; their adjusted odds ratios were 2.33 (P =0.01) ischemic rest pain or with ischemic skin and 7.40 (P , 0.01), respectively. However, there was no significant difference in ankle systolic lesions, either ulcers or gangrene. It is blood pressure or ABI level between those with and without these comorbidities. Furthermore, associated with an extremely poor prog- fi the presence of arterial calci cation was not associated with ankle systolic blood pressure or ABI nosis for both survival and limb salvage; level, whereas arterial stenoses of all segments in the lower body had independent associations its prompt diagnosis and following re- with reduced ankle systolic blood pressure and ABI level. vascularization are therefore important. CONCLUSIONSdDiabetes and regular hemodialysis were significantly associated with ar- Although TransAtlantic Inter-Society terial calcification, but not with elevated measurements of ankle systolic blood pressure or ABI, in Consensus (TASC) (12), a worldwide CLI patients. clinical guideline for PAD, and its revised consensus TASC II (1) acknowledge the validity of using ankle systolic blood pressure and ABI in the detection of CLI, little is known about the distribution of nkle systolic blood pressure and its in identifying healthy individuals (1). Be- these measurements and the influence of Aratio to brachial systolic blood pres- cause of their predictive capability and non- comorbidities on these measurements in sure, that is, ankle-brachial systolic invasiveness, these measurements are CLI patients. A previous clinical study fl pressure index (ABI), re ect arterial he- recommended for screening and determin- disclosedinpatientcharacteristicsthat modynamics in lower extremity. A re- ing PAD in many clinical settings (2). about a half of the recruited CLI patients duced value of these measurements is Furthermore, ABI is also known as an inde- with ischemic skin lesions had ankle sys- fi caused by hemodynamically signi cant pendent prognostic risk factor for fatal and tolic blood pressure higher than 70 mmHg arterial stenosis and therefore indicates nonfatal cardiovascular disease and all- (13). Elevation of these measurements the presence of peripheral arterial disease cause mortality in a general population might be more common in CLI patients # – (PAD). ABI 0.90 in symptomatic indi- (3 5), as well as in diabetic patients (6) than expected. viduals, for example, is approximately andinpatientswithend-stagekidneydisease We hypothesized that in CLI patients, 95% sensitive in detecting arteriogram- receiving hemodialysis (7). Their usefulness diabetes and end-stage renal disease were fi positive PAD and almost 100% speci c in clinical practice is now widely recognized. associated with arterial calcification and ccccccccccccccccccccccccccccccccccccccccccccccccc consequently with falsely elevated ankle From the 1Department of Metabolic Medicine, Osaka University Graduate School of Medicine, Osaka, Japan; systolic blood pressure and ABI. To verify and the 2Cardiovascular Division, Kansai Rosai Hospital, Hyogo, Japan. the hypothesis, we investigated in the Corresponding author: Hideaki Kaneto, [email protected]. current study whether these clinical co- Received 25 August 2011 and accepted 25 April 2012. morbidities had significant associations DOI: 10.2337/dc11-1636 © 2012 by the American Diabetes Association. Readers may use this article as long as the work is properly with ankle systolic blood pressure and cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/ ABI, as well as arterial calcification, in CLI licenses/by-nc-nd/3.0/ for details. patients. care.diabetesjournals.org DIABETES CARE 1 Diabetes Care Publish Ahead of Print, published online June 20, 2012 Diabetes and ABI in CLI patients

RESEARCH DESIGN AND We obtained writteninformedconsent duration and its quartiles were 20 (10: METHODS from all the recruited patients. 27) years. Regular hemodialysis was received by 126 patients (47%), with their Study population and definitions Statistical analyses median dialysis vintage 5 (2: 10) years. We recruited 269 consecutive Japanese Data are given as mean 6 SD for contin- One hundred ninety three patients patients with CLI who underwent angio- uous variables and as percentages for di- (72%) had antihypertensive treatment. graphical examinations in the lower body chotomous variables if not mentioned The prevalence of dyslipidemia, smoking, including tibial segments as well as the otherwise. and foot ulceration was 77 (n = 206), 67 measurement of ankle systolic blood pres- We first assessed the distributions of (n = 181), and 84% (n = 227), respectively. sure and ABI and were treated with sub- ankle systolic blood pressure and ABI in Ankle systolic blood pressure was 85 6 56 sequent endovascular therapy in Kansai the recruited CLI patients. We also ex- mmHg and ABI was 0.59 6 0.37, whereas Rosai Hospital, Hyogo, Japan. The hospi- amined the prevalence of ankle systolic brachial systolic blood pressure was tal was positioned as the regional core blood pressure $250 mmHg and ABI 143 6 27 mmHg. Ankle systolic blood hospital supporting community medicine .1.40, suggesting typical noncompress- pressure was significantly associated with in Hyogo Prefecture, and patients with ible measurements (1), as well as that of brachial systolic blood pressure (Pearson’s chronic ischemic rest pain and/or foot ulcer ankle systolic blood pressure $70 mmHg r = 0.36, P , 0.01), whereas ABI was not or gangrene were referred to the cardiovas- and ABI .0.90, suggesting false eleva- (Pearson’s r =0.08,P = 0.18). A total of 171 cular division of the hospital. tions in CLI patients (1). patients (64%) presented ankle systolic All the referred patients were evaluated We subsequently assessed the follow- blood pressure $70 mmHg and 60 pa- for limb ischemia by angiography. The ing three associations in the recruited CLI tients (22%) had ABI more than 0.90. ABI diagnosis and management of CLI were patients: 1) the association of clinical char- .1.40 was observed in only one patient compliant with TASC (12) or its revised acteristics with ankle systolic blood pres- (0.4%), and the rest had ABI below 1.40. consensus, TASC II (1). Once patients sure and ABI, 2) the association of clinical No patients had their ankle systolic blood were diagnosed as CLI, endovascular ther- characteristics with arterial calcification, pressure 250 mmHg or more. apy was used as the first-line procedure for and finally 3) the association of angio- revascularization, within the recommenda- graphic observations, including calcifica- Association of clinical characteristics tions in TASC (II). The indication of endo- tion as well as arterial stenosis, with ankle with ABI and ankle systolic vascular therapy was judged by consensus systolic blood pressure and ABI. Measure- blood pressure among vascular specialists including vas- ments of ankle systolic blood pressure and Table 1 shows the association of clinical cular surgeons. Patients were excluded if ABI were assessed as continuous depen- characteristics with ABI. Patients with di- they were considered to be poor candi- dent variables. Arterial calcification was abetes and regular hemodialysis had dates for angiography and subsequent re- treated as a dichotomous variable. slightly, but not significantly, elevated vascularization as a result of severe In unadjusted analyses for continu- ABI compared with those without these comorbidities including hemodynamic ous outcomes, the significant difference comorbidities; the differences and 95% risk or difficulty having supine rest during between patients with and without each CI were only 0.04 [20.05, 0.14] and the intervention or they refused the inter- comorbidity was assessed by the use of 0.06 [20.02, 0.15], respectively. Neither vention. Ankle systolic blood pressure and unpaired t test. Because ankle systolic did other clinical characteristics have any ABI were evaluated in all the recruited pa- blood pressure was expected to be significant association with ABI (Table 1). tients before endovascular therapy with strongly influenced by systemic blood Similar findings were observed when the use of the automated oscillometric de- pressure, we also demonstrated its dif- we investigated the association of diabetes vice provided by OMRON COLIN Co., ference adjusted for measurements of and regular hemodialysis with ankle sys- Ltd. (Tokyo, Japan). Its validity has been brachial systolic blood pressure by tolic blood pressure. As shown in Table 2, reported elsewhere (14,15). Arterial ste- ANCOVA. The association with dichoto- patients with diabetes and regular hemo- noses of the lower body and calcification mous outcomes was analyzed in a dialysis had slightly, but not significantly, of tibial arteries were assessed angiograph- univariate logistic regression model. higher ankle systolic blood pressure com- ically. Calcified lesions were defined Explanatory variables were subsequently pared with those without these comor- quantitatively as obvious densities noted all together entered into multivariate bidities; their adjusted differences and within the apparent vascular wall in the model to reveal the independent influence 95% CI were 5 [29, 19] and 11 [22, angiogram (16). on the outcomes. 24] mmHg, respectively. Neither did Diabetes was determined when patients A P value ,0.05 was considered statis- any other comorbidity provide significant had been treated for it or when they met the tically significant. Differences in continu- difference (Table 2). following criteria: fasting plasma glucose ous dependent variables and odds ratios level $7.0 mmol/L, casual plasma glucose (OR) of dichotomous dependent variables, Association of clinical characteristics $ $ fi fi level 11.1 mmol/L, or HbA1c level 6.5% as well as their 95% con dence intervals with arterial calci cation (17). Dyslipidemia was defined as serum (CI), are reported. Statistical analyses Calci fication of tibial arteries was found LDL cholesterol $2.6mmol/LorHDL were performed using IBM SPSS Statistics in 186 patients (69%), and we assessed cholesterol ,1.0 mmol/L or triglycerides Version 19 (SPSS, an IBM company). the association of clinical characteristics $1.7 mmol/L or having been treated for with the arterial calcification. As shown in dyslipidemia. We performed the current RESULTSdThe recruited patients were Table 3, diabetes and regular hemodialy- study in accordance with the declaration 71 6 11 years old, and 184 of 269 (68%) sis, as well as male sex, had a significant of Helsinki, and it was approved by the were male. A total of 191 patients (71%) association with arterial calcification in the ethics committee of Kansai Rosai Hospital. had diabetes, and their median diabetic univariate model. These two variables

2 DIABETES CARE care.diabetesjournals.org Takahara and Associates

Table 1dAssociation of clinical characteristics with ABI CONCLUSIONSdAnkle systolic blood pressure and ABI can noninvasively assess ABI difference P value hemodynamics in the lower limb and are now widely used in clinical practice to Male (vs. female) +0.03 [20.06, 0.13] 0.47 screen and determine PAD (1). These mea- Age $70 years 20.03 [20.12, 0.06] 0.49 surements are also used in the detection Diabetes +0.04 [20.05, 0.14] 0.37 of CLI; when patients had unhealing foot Antihypertensive treatment +0.03 [20.07, 0.13] 0.56 ulceration or rest pain, one can suspect Dyslipidemia 20.07 [20.17, 0.04] 0.20 it of resulting from lower-limb ischemia Smoking habit 20.07 20.17, 0.02] 0.13 if these measurements are reduced. The Regular hemodialysis +0.06 [20.02, 0.15] 0.16 current study demonstrated that arterial Data are unadjusted differences in ABI and their 95% CI. The differences in ABI were assessed between stenosis in each segment of the lower patients with and without each clinical characteristic, whose statistical significance was evaluated by unpaired body was independently associated with t test. a reduced ankle systolic blood pressure andABIlevelinCLIpatients.Thesefindings were also significantly associated with arterial Association of angiographic suggest that accumulating arterial ste- calcification in a multivariate model, in findings with ABI and ankle noses of the leg would additively reduce which adjusted ORs were 2.33 [1.24, 4.36] systolic blood pressure these measurements and that smaller val- (P = 0.01) and 7.40 [3.72, 14.7] (P , 0.01), We finally investigated the association of ues of the measurements would clinically fl respectively. Other variables had no signifi- angiographic characteristics with ABI and re ect more widely distributed arterial ste- cant association with the arterial calcification. ankle systolic blood pressure. The steno- noses. We subsequently performed category sis of iliac, femoral, and tibial arteries was On the other hand, it has been analysis to assess the effect of diabetic found in 45, 171, and 222 patients (17, pointed out that these measurements duration and dialysis vintage on the out- 64, and 83%, respectively), and patients can be falsely elevated in patients with come. In brief, diabetic patients were with arterial stenosis in each segment had diabetes and end-stage renal disease be- categorized into two groups with the use lower levels of ABI and ankle systolic cause of their noncompressible vessels fi of the median values of diabetic duration blood pressure than those without the caused by vascular calci cation (2,8). (20 years) and were compared with non- stenosis. As shown in Table 4, multivariate However, it still remains unrevealed diabetic population in the multivariate analysis revealed that all of these arterial whether these false elevations could be model. Similarly, patients receiving regu- stenoses were independently associated similarlyobservedinCLIpatients.Fur- lar hemodialysis were classified with the with reduced ABI and ankle systolic blood thermore, although it is often mentioned median dialysis vintage (5 years) and were pressure. The adjusted reduction by each that ankle systolic blood pressure in CLI compared with nondialysis receivers. As a segmental stenosis ranged from 20.13 to patients is typically lower than 70 mmHg, result, longstanding diabetes and hemo- 20.28 in ABI and from 218 mmHg to its true distribution in CLI patients has dialysis were both independently associ- 240 mmHg in ankle systolic blood pres- been seldom reviewed and so far is not ated with an increased risk of arterial sure (Table 4). fully understood. According to a very calcification (P , 0.01). The adjusted OR On the other hand, arterial calcifica- few previous clinical trials, which dis- [95% CI] for diabetes with duration $20 tion had no significant association with closed its distribution in their patient years was 4.65 [2.03, 10.6], whereas that either ABI (P = 0.89) or ankle systolic characteristics, elevated ankle systolic for diabetic duration ,20 years was 1.57 blood pressure (P = 0.56) (Table 4). Nei- blood pressure might be more common [0.78, 3.15]. On the other hand, the adjusted ther was its association with ankle systolic than expected (13). To clarify these is- OR for regular hemodialysis with vintage blood pressure statistically significant af- sues, we limited the study population to ,5 years and $5 years was 4.45 [1.90, ter additional adjustment for brachial sys- CLI patients and investigated the distri- 10.4] and 10.1 [3.61, 28.3], respectively. tolic blood pressure (P = 0.96). bution of these measurements and the association of clinical characteristics with these measurements. We hypothesized Table 2dAssociation of clinical characteristics with ankle systolic blood pressure that diabetes and regular hemodialysis would increase the risk of arterial calcification and would consequently be associ- Difference adjusted ated with a false elevation of ankle systolic Unadjusted for brachial systolic blood pressure and ABI in CLI patients. difference (mmHg) P value pressure (mmHg) P value The current study found that mean Male (vs. female) +2 [212, 17] 0.76 +4 [29, 18] 0.52 measurements of ankle systolic blood Age $70 years 28[222, 6] 0.25 24[217, 8] 0.49 pressure and ABI were as high as 85 Diabetes +4 [211, 19] 0.63 +5 [29, 19] 0.45 mmHg and 0.59 and that a substantial Antihypertensive treatment +8 [27, 23] 0.28 +2 [212, 16] 0.80 part of the patients had ankle systolic $ Dyslipidemia 29[225, 7] 0.29 28[223, 7] 0.28 blood pressure 70 mmHg and/or ABI . Smoking habit 212 [226, 3] 0.11 211 [224, 23] 0.11 0.90. However, in contrast with our ini- Regular hemodialysis +10 [23, 24] 0.14 +11 [22, 24] 0.09 tial hypothesis, neither diabetes nor reg- fi Data are differences in ankle blood pressure and their 95% CI. The unadjusted differences in ankle pressure ular hemodialysis had any signi cant were assessed between patients with and without each clinical characteristic, whose statistical significance association with these elevated measure- was evaluated by unpaired t test. The adjusted differences were assessed with the use of ANCOVA. ments, although these two comorbidities, care.diabetesjournals.org DIABETES CARE 3 Diabetes and ABI in CLI patients

Table 3dAssociation of clinical characteristics with arterial calcification the recruited patients, it would be of no surprise if patients with calcified tibial ar- Unadjusted OR P value Adjusted OR P value teries were very likely to have other arteries, including those in the upper Male (vs. female) 2.31 [1.34, 3.98] ,0.01 2.11 [0.98, 4.53] 0.06 limb, similarly calcified (24–26). It might Age $70 years 0.87 [0.52, 1.48] 0.62 1.09 [0.59, 2.04] 0.78 be that CLI patients whose ankle systolic Diabetes 2.63 [1.51, 4.57] ,0.01 2.28 [1.21, 4.30] 0.01 blood pressure was falsely evaluated be- Antihypertensive treatment 1.14 [0.65, 2.02] 0.65 1.14 [0.58, 2.22] 0.70 cause of calcified tibial arteries were more Systolic blood pressure (mmHg) 0.99 [0.98, 1.00] 0.13 0.99 [0.98, 1.00] 0.18 likely to have falsely elevated brachial Dyslipidemia 0.63 [0.33, 1.21] 0.17 0.66 [0.32, 1.39] 0.28 systolic blood pressure as a result of cal- Smoking habit 1.07 [0.62, 1.85] 0.81 0.75 [0.34, 1.65] 0.47 cified brachial arteries (27–31). The Regular hemodialysis 8.34 [4.30, 16.1] ,0.01 7.39 [3.72, 14.7] ,0.01 coexisting false elevation of brachial sys- Data are OR for arterial calcification and their 95% CI. Adjusted OR and their 95% CI were obtained in tolic blood pressure would cancel the multivariate model into which all the variables listed in the table were entered. false increase in ABI caused by tibial calcification, and this is one possible explanation of no significant association especially with a long duration and vin- Although these measurements provide between tibial calcification and increased tage, significantly increased the risk of ar- similar results in some articles (15,21,22), ABI level. terial calcification. The lack of significant it is possible that these three methods pro- Some may argue that these hypothe- association was also observed between ar- vide different results in calcified arteries. ses do not explain the absence of a sig- terial calcification and elevated ankle sys- The oscillometric technique might be less nificant association between tibial tolic blood pressure and ABI levels in the subject to arterial calcification, whereas calcification and ankle systolic blood population. These findings suggest that use of different methods might provide pressure measurements. However, in both diabetes and regular hemodialysis higher pressure measurements in calcified clinical practice, brachial blood pressure are associated with arterial calcification arteries. is used as a substitute for systemic blood as reported previously (18,19), but that Another possible explanation of the pressure; patients’ systemic blood pres- thepresenceofarterialcalcification is different findings in our current study sure is generally evaluated and controlled notassociatedwithelevatedanklesys- from those in previous reports would be on the basis of the measurements of bra- tolic blood pressure or ABI in CLI patients. systemic distribution of noncompressible chial blood pressure (32–34). No one ever Our current findings were in contrast vessels in CLI patients. Previous studies knows in clinical practice whether their with those in some previous literature often recruited PAD patients, including elevated brachial pressure reflects the (20), which demonstrated that diabetes those with only intermittent claudication presence of true hypertension or is caused was a dominant risk factor for elevated and with no symptoms, or patients sus- by calcification of brachial arteries (27–31). ABI levels. There would be two possible pected of PAD (20), whereas we limited Patients with truly elevated brachial explanations of this discrepancy between our study population to CLI patients. It is pressure would be more likely to have our findings and theirs. One possible ex- well known that patients with CLI suffer their ankle pressure elevated by their hy- planation would be the way of ABI mea- more progressed atherosclerosis and pertension too. On the other hand, as dis- surement. In that previous study, they more severe arterial calcification com- cussed above, patients with falsely measured ABI by detecting the signals paredwithpatientswithotherPAD elevated brachial pressure by arterial cal- with photoplethysmography at the (23). In the current study population, cification would be more likely to suffer third finger and great toe, whereas in the prevalence of tibial calcification widely distributed calcification in the sys- the current study we measured ABI with had a considerable overlap (;90%) with temic artery (27) and therefore to have the use of the automated oscillometric that of the arterial calcification elsewhere their ankle pressure falsely elevated. As a device. There are three common methods in the lower body (data not shown). Al- result, it would be clinically impossible to of assessing ABI: oscillometric, Doppler, though we did not have any data about distinguish the false elevation of ankle and photoplethysmographic techniques. arterial calcification of the upper limb in pressure induced by arterial calcification from the true elevation as a result of hypertension. This would possibly be why Table 4dAssociation of angiographic characteristics with ABI and ankle systolic we failed to observe significant difference blood pressure in ankle systolic blood pressure between those with and without calcified tibial arteries. Adjusted difference In conclusion, both diabetes and reg- Adjusted in ankle systolic fi P P ular hemodialysis were signi cantly asso- difference in ABI value blood pressure (mmHg) value ciated with arterial calcification, but not Stenosis with elevated measurements of ankle Iliac 20.17 [20.29, 20.06] ,0.01 223 [240, 25] 0.01 systolic blood pressure or ABI, in CLI Femoral 20.28 [20.36, 20.19] ,0.01 240 [253, 226] ,0.01 patients. Furthermore, these noninvasive Tibial 20.13 [20.23, 20.02] 0.02 218 [235, 22] 0.03 measurements were not affected by the Calcification 20.01 [20.09, +0.08] 0.89 24[218, +10] 0.56 presence of calcification in tibial arteries Data are adjusted differences in ABI and ankle pressure and their 95% CI. Adjusted OR and their 95% CI were in these patients, whereas accumulating obtained in multivariate model into which all the variables listed in the table were entered. arterial stenoses in the lower body had

4 DIABETES CARE care.diabetesjournals.org Takahara and Associates additive effects on reduced levels of these factors of PAD among patients with ele- 22. Harrison ML, Lin HF, Blakely DW, measurements. vated ABI. Eur J Vasc Endovasc Surg 2008; Tanaka H. Preliminary assessment of an 35:709–714 automatic screening device for peripheral 10. Ix JH, Katz R, Peralta CA, et al. A high arterial disease using ankle-brachial and ankle brachial index is associated with toe-brachial indices. Blood Press Monit AcknowledgmentsdNo potential conflicts of greater left ventricular mass MESA (Multi- 2011;16:138–141 interest relevant to this article were reported. Ethnic Study of Atherosclerosis). J Am 23. Guzman RJ, Brinkley DM, Schumacher PM, M.T. and H.K. researched data and wrote Coll Cardiol 2010;55:342–349 Donahue RM, Beavers H, Qin X. Tibial the manuscript. O.I., N.K., T.M., and M.I. 11. Suominen V, Uurto I, Saarinen J, artery calcification as a marker of amputa- contributed to the discussion. I.S. reviewed Venermo M, Salenius J. PAD as a risk tion risk in patients with peripheral arterial and edited the manuscript. H.K. is the guar- factor for mortality among patients with disease. J Am Coll Cardiol 2008;51:1967– antor of this work and, as such, had full access elevated ABIda clinical study. Eur J Vasc 1974 to all the data in the study and takes re- Endovasc Surg 2010;39:316–322 24. Allison MA, Criqui MH, Wright CM. Pat- sponsibility for the integrity of the data and the 12. Dormandy JA, Rutherford RB. Manage- terns and risk factors for systemic calcified accuracy of the data analysis. ment of peripheral arterial disease (PAD). atherosclerosis. Arterioscler Thromb Vasc The authors thank Shinsuke Nanto (De- TASC Working Group. TransAtlantic Biol 2004;24:331–336 partment of Advanced Cardiovascular Ther- Inter-Society Consensus (TASC). J Vasc 25. Kuller LH, Matthews KA, Sutton-Tyrrell K, aputics, Osaka University Graduate School of Surg 2000;31:S1–S296 Edmundowicz D, Bunker CH. Coronary Medicine) and Masaaki Uematsu (Cardio- 13. Nehler MR, Brass EP, Anthony R, et al.; and aortic calcification among women vascular Division, Kansai Rosai Hospital) for Circulase investigators. Adjunctive paren- 8 years after menopause and their pre- help with preparing the manuscript. teral therapy with lipo-ecraprost, a prosta- menopausal risk factors : the healthy glandin E1 analog, in patients with critical women study. Arterioscler Thromb Vasc limb ischemia undergoing distal revascu- Biol 1999;19:2189–2198 References larization does not improve 6-month out- 26. Odink AE, van der Lugt A, Hofman A, 1. Norgren L, Hiatt WR, Dormandy JA, et al.; comes. J Vasc Surg 2007;45:953–960 et al. Association between calcification in TASC II Working Group. Inter-Society 14. Koji Y, Tomiyama H, Ichihashi H, et al. the coronary arteries, aortic arch and ca- Consensus for the Management of Periph- Comparison of ankle-brachial pressure rotid arteries: the Rotterdam study. Ath- eral Arterial Disease (TASC II). Eur J Vasc index and pulse wave velocity as markers erosclerosis 2007;193:408–413 Endovasc Surg 2007;33(Suppl. 1):S1–S75 of the presence of coronary artery disease 27. Sheckman P, Klassen G. Pseudohypertension 2. Jude EB, Eleftheriadou I, Tentolouris N. in subjects with a high risk of atheroscle- secondary to a noncompressible brachial Peripheral arterial disease in diabetesda rotic cardiovascular disease. Am J Cardiol artery. Can Med Assoc J 1974;111:1227– review. Diabet Med 2010;27:4–14 2004;94:868–872 1228 3. Heald CL, Fowkes FG, Murray GD, Price 15. Richart T, Kuznetsova T, Wizner B, 28. Foran TG, Sheahan NF, Cunningham C, JF; Ankle Brachial Index Collaboration. Struijker-Boudier HA, Staessen JA. Valida- Feely J. Pseudo-hypertension and arterial Risk of mortality and cardiovascular dis- tion of automated oscillometric versus stiffness: a review. Physiol Meas 2004;25: ease associated with the ankle-brachial manual measurement of the ankle-brachial R21–R33 index: Systematic review. Atherosclerosis index. Hypertens Res 2009;32:884–888 29.LittenbergB,WolfbergC.Pseudohy- 2006;189:61–69 16. SogaY,IidaO,HiranoK,SuzukiK,YokoiH, pertension masquerading as malignant 4. Leng GC, Fowkes FG, Lee AJ, Dunbar J, Nobuyoshi M. Restenosis after stent im- hypertension. Case report and review of Housley E, Ruckley CV. Use of ankle plantation for superficial femoral artery the literature. Am J Med 1988;84:539–542 brachial pressure index to predict car- disease in patients treated with cilostazol. 30. Taguchi JT, Suwangool P. “Pipe-stem” diovascular events and death: a cohort Catheter Cardiovasc Interv 2012;79:541–548 brachial arteries. A cause of pseudohy- study. BMJ 1996;313:1440–1444 17. Seino Y, Nanjo K, Tajima N, et al. Report pertension. JAMA 1974;228:733 5. Fowkes FG, Murray GD, Butcher I, et al.; of the Committee on the classification and 31. Kuriyama S, Tomonari H, Utsunomiya Y, Ankle Brachial Index Collaboration. Ankle diagnostic criteria of diabetes mellitus: Kinoshita N, Matsumoto H. Pseudohy- brachial index combined with Framingham The Committee of the Japan Diabetes pertension in hemodialyzed arteriosclerotic Risk Score to predict cardiovascular events Society on the diagnostic criteria of di- patients. Nephron 1994;66:479–480 and mortality: a meta-analysis. JAMA 2008; abetes mellitus. Diabetol Int 2010;1:2–20 32. American Diabetes Association. Standards 300:197–208 18. Potier L, Abi Khalil C, Mohammedi K, of medical care in diabetesd2011. Di- 6. Mehler PS, Coll JR, Estacio R, Esler A, Roussel R. Use and utility of ankle brachial abetes Care 2011;34(Suppl. 1):S11–S61 Schrier RW, Hiatt WR. Intensive blood index in patients with diabetes. Eur J Vasc 33. Mancia G, Laurent S, Agabiti-Rosei E, pressure control reduces the risk of car- Endovasc Surg 2011;41:110–116 et al.; European Society of Hypertension. diovascular events in patients with periph- 19. Raggi P, Boulay A, Chasan-Taber S, et al. Reappraisal of European guidelines on eral arterial disease and type 2diabetes. Cardiac calcification in adult hemodialy- hypertension management: a European Circulation 2003;107:753–756 sis patients. A link between end-stage re- Society of Hypertension Task Force 7. Kitahara T, Ono K, Tsuchida A, et al. nal disease and cardiovascular disease? document. J Hypertens 2009;27:2121– Impact of brachial-ankle pulse wave ve- J Am Coll Cardiol 2002;39:695–701 2158 locity and ankle-brachial blood pressure 20. Aboyans V, Ho E, Denenberg JO, Ho LA, 34. Chobanian AV, Bakris GL, Black HR, et al.; index on mortality in hemodialysis pa- Natarajan L, Criqui MH. The association Joint National Committee on Prevention, tients. Am J Kidney Dis 2005;46:688–696 between elevated ankle systolic pressures Detection, Evaluation, and Treatment of 8. Leskinen Y, Salenius JP, Lehtimäki T, and peripheral occlusive arterial disease in High Blood Pressure. National Heart, Lung, Huhtala H, Saha H. The prevalence of diabetic and nondiabetic subjects. J Vasc and Blood Institute; National High Blood peripheral arterial disease and medial ar- Surg 2008;48:1197–1203 Pressure Education Program Coordinating terial calcification in patients with chronic 21. Beckman JA, Higgins CO, Gerhard- Committee. Seventh report of the Joint renal failure: requirements for diagnostics. Herman M. Automated oscillometric de- National Committee on Prevention, Detec- Am J Kidney Dis 2002;40:472–479 termination of the ankle-brachial index tion, Evaluation, and Treatment of High 9. Suominen V, Rantanen T, Venermo M, provides accuracy necessary for office Blood Pressure. Hypertension 2003;42: Saarinen J, Salenius J. Prevalence and risk practice. Hypertension 2006;47:35–38 1206–1252 care.diabetesjournals.org DIABETES CARE 5