Intermittent claudication as a manifestation of silent myocardial ischemia: A pilot study

Intermittent claudication as a manifestation of silent myocardial ischemia: A pilot study A-M Salmasi, PhD, A. Nicolaides, MS, FRCS, FRCSE, A. AI-Katoubi, MD, FRCR, T. N. Sonecha, MRCP, DTM&H, P. R. Taylor, ChM, FRCS, S. Serenkuma, BSc, and H. H. G. Eastcott, MS, FRCS, FACS, London, United Kingdom One hundred consecutive patients with intermittent claudication were screened noninvasively with electrocardiography chest wall mapping stress test and transcutaneous aortovelography during bicycle ergometry. Electrocardiographic chest wall stress testing indicated three-vessel coronary disease in 25 patients and left anterior descending plus circumflex (left main stem equivalent) disease in seven. In these 32 patients transcutaneous aortovelography demonstrated a decrease in stroke distance (an index of cardiac stroke volume) (median, -28%; 90% range, +5% to -48%), and coronary angiography confirmed the presence and severity of the disease. The claudication distance ranged between 50 and 250 meters. After myocardial revascularization or medical therapy a significant increase occurred in the stroke distance after exercise (median, + 20; 90% range, + 40% to -25%); also a significant increase in the postexercise pressure index and a reduction in the recovery time (p < 0.01). No change occurred in the ankle/pressure index at rest. Twelve patients were able to walk without being limited by claudication; 15 reported improvement with a two to tenfold increase in claudication distance. No change occurred in three. The results indicate that silent myocardial ischemia is a common finding in patients with intermittent claudication. It produces left ventricular dysfunction and a decrease in stroke volume leading to a large fall in ankle pressure and early onset of claudication during exercise. Niltrates and myocardial revascttlarization tend to reverse this. (J VAsc SuRG 1991;14:76-86.) Patients with lower limb atherosclerotic disease have a high prevalence of obstructive coronary artery disease. >s When routine coronary angiography was performed in one series only 18% were found to have normal coronary arteries, and approximately 34% had severe three-vessel coronary artery disease, s Severe concomitant coronary disease, often undetected, accounts for the overall perioperative mortality rate of 2% to 5% and the 3-year mortality rate of 30% to 40% reported in patients undergoing peripheral vascular reconstruction. 6-1 Because patients with peripheral vascular disease are often limited by claudication, the treadmill exercise test is not ideally suited for the evaluation of myocardial ischemia. These patients are often stopped by claudication before an adequate heart rate necessary to unmask myocardial ischemia is attained. From the Vascular Section, Academic Surgical Unit, Irvinc Laboratory for Cardiovascular Investigation and Research, Vascular Unit, Department of Cardiac Sciences, and Department of Radiology, St. Mary's Hospital Medical School, London. Presented at the Forty-fourxh Annual Meeting of the Sociew for Vascular Surgery, Los Angeles, Calif., June 4-6, 1990. Reprint requests: Andrew N. Nicolaides, MS, Vascular Section, St. Mary's Hospital Medical School, London, United Kingdom 24/6/25531 Dipyridamole thallium scintigraphy and Holter monitoring have been advocated as alternative noninvasive methods to obviate this difficulty, n-i4 Frequency modulated Holter monitoring can detect episodes of silent myocardial ischemia, a marker for perioperative and postoperative ischemic cardiac events, la,14 and can be used before peripheral vascular surgery to identify patients at high risk. ~-~8 The reported prevalence of silent myocardial ischemia is 12% to 15% in middle-aged asymptomatic men with coronary risk factors and 35% in diabetic patients? 6,17,19 The prevalence of silent myocardial ischemia in patients with claudication would be expected to be high also. However, accurate information on the latter group is not yet available. During the last 7 years our team has developed and validated two further noninvasive tests for the detection of silent myocardial ischemia and assessment of left ventricular function: electrocardiographic (ECG) chest wall mapping stress test with bicycle ergometry and transcutaneous aortovelography (TAV). 2-22"24a~ Electrocardiographic stress testing with bicycle ergometry can produce an adequate heart rate response in two thirds of patients with claudication, 2~ so that the more expensive intrave- 76

Volume 14 Number 1 3~y 1991 Silent myocardial ischemia in patients with intermittent claudications 77 ::::::::::::::::::::::....4!i, I ed2 ed3 ed4 Fig. 1. Position of electrodes on front and back of thorax. nous dipyridamole thallium test is indicated only in the remaining third. Bicycle ergometry with ECG chest wall mapping not only can detect myocardial ischemia but can determine the distribution of significant (->50% diameter stenosis) disease in individual coronary artery territories. 2,23 Transcutaneous aortovelography is an ultrasonic technique that can demonstrate changes in stroke volume as a result of infusion of inotropes, changes in preload, or exercise.22, 24-29 The detection of severe three-vessel coronary disease in patients with intermittent claudication often leads to the recommendation that myocardial revascularization be considered before peripheral vascular reconstruction to minimize perioperative cardiac risk. After coronary artery bypass surgery many patients find that their clandication distance improves to such an extent that the initially proposed vascular operation is no longer warranted. The purpose of our study was to determine the prevalence of silent myocardial ischemia with consequent left ventricular dysfunction in patients with intermittent claudication and to document the effects of therapeutic intervention designed to improve left ventricular performance on the lower limb arterial hemodynamics and claudication. MATERIAL AND METHODS In part I of the study 100 patients (85 men and 15 women) under the age of 75 being treated for intermittent claudication were studied. Their mean age was 61.9 years (SD + 6.3 and range 54 to 74). None of the patients had previous peripheral arterial reconstruction or coronary bypass grafting. Coronary artery disease was suspected in 21 patients because of a history of angina in the past (12 patients), previous myocardial infarction (14 patients), and/or an abnormal conventional resting electrocardiogram (18 patients). All patients underwent ECG chest wall mapping stress test, z m'23 TAV, 29-31 ankle/brachial systolic pressure index at rest and after a standard 1- minute exercise test. 32'33 Patients considered to be at high risk because of a positive TAV test in the presence of three-vessel coronary disease or left anterior descending (LAD) and circumflex (left main stem equivalent) disease indicated by ECG chest wall mapping stress test, were selected for part II of the study and underwent coronary angiography to confirm the findings. They were referred back to the referring physician or cardiologist who decided on the form of intervention: myocardial revascularization or medical therapy. The ECG chest wall mapping stress test, TAV, and 1-minute treadmill test were repeated 6 weeks later. ECG chest wall mapping exercise test Skin preparation and position of electrodes. The chest wall was shaved when necessary and the skin cleaned with 70% alcohol to ensure good contact with the disposable press stud electrodes; these were positioned in four vcrtical rows (Fig. 1). Row A was along a line going through the right sternoclavicular joint, B through the left sternoclavicular joint, C through the left anterior axillary line, and D at the back 6 cm medial to the posterior axillary line. Electrode C3 was on the anterior axillary line at the fifth intercostal space corresponding to the conventional Vs position: this determined the horizontal levels of A3, B3, and D3. The distance between B3 and C3 was used to determine the distance between the horizontal rows. A reference lead was placed at the right infraclavicular fossa and an earth lead was placed in the left midaxillary line just above the iliac crest. Exercise. Exercise was performed on a Kern 2 bicycle ergometer (Mijnhardt, Holland) at a constant speed of 60 rpm. The starting work load was 50 W, which was increased by 25 W every 2 minutes up to

78 Salmasi, Nicolaides, Al-Katoubi et al. journal of VASCULAR SURGERY 2 F cm sec -I IOO[ 0 L. b I*-Tf--~ _~ ~ = _-- =.".,-......... ---'...-,................ Fig. 2. Typical recordings by transcutaneous aortovelography (a) at rest and (b) at the end of standard exercise in a patient demonstrating an increase in Sd. Tf, Flow time; Pv, peak velocity; Sd, stroke distance (1/2 Pv Tf). it b.i a _a 2 ~---. ~ ~- -~--:- -_=--" = ~5~ ~ _=.:k= g- ~ ~_. _ ~- _% Fig. 3. Typical recordings by transcutaneous aortovelography (a) at rest and (b) at the end of standard exercise in a patient with coronary artery disease. There is a decrease in Sd associated with ST segment depression on the ECG. 100 W and then by 25 W every minute tmtil the end point, determined by the maximum predicted heart rate for the patient's age. Indications for stopping the exercise were the development of severe chest pain, ST segment depression >3 mm, dyspnea, exhaustion, threatening arrhythmias, a decrease in systolic blood pressure, or claudication. Electrocardiographic recording and blood pressure monitoring. Recordings from the chest leads were obtained at rest and every minute during and after exercise until the heart rate and ECG changes had reverted to the preexercise state. The 12 leads were connected to a standard three-channel ECG recorder via a 4-way switch, which allowed simultaneous recordings from any three leads in a vertical row. Blood pressure was measured at rest and every minute throughout the test with a sphygmomanometer. Isopotential chest wall mapping. Changes in ST segment were mapped as isopotential areas of 1 mm, 2 mm, or 3 mm for every minute of the test. The presence of Q waves at rest was also mapped. The ST segment was considered to be depressed if it was -> 1 mm below the isoelectric line and its duration after the J-point was ->0.8 second. A Q wave was considered to be significant if its duration was ->0.04 second or its amplitude was ->25% of the amplitude of the R wave. Transcutaneous aortovelography The transcutaneous aortovelograph in this study was a commercial prototype (type 1006) based on the machine developed by Light. = Preexercise recordings were obtained after at least 5 minutes' supine rest, with the subject's shoulder and head slightly elevated and the legs horizontal. The transducer was placed on the suprasternal notch and directed downward and laterally in a direction tangential to the distal part of the aortic arch. Its position was adjusted until the highest pitched-sound was heard on the loudspeaker, and signals with maximum peak frequency were recorded. Because of the variability in blood velocities after meals, all recordings were made in the morning at least 3 hours after the intake of food or hot fluids. Exercise. Supine exercise was carried out on a bicycle ergometer (Kem 2), which was attached to the end of a couch (Mijnhardt, Holland). The axle height was 10 cm above couch level. The same standard protocol with stepwise increase in work toad as described for ECG chest wall mapping was used. The TAV recording was repeated when the subjects stopped exercising, having reached their maximum. The ECG was monitored continuously, and blood pressure was recorded at I-minute intervals during exercise, at peak exercise, and during the recovery phase until they returned to preexercise levels. Figs. 2 and 3 illustrate typical recordings at rest

Volume 14 Number 1 July 1991 Silent myocardial ischemia in patients with intermittent claudications 79 and at maximal exercise. The outline (envelope) of the dark complexes, which are spectrograms of the Doppler frequency.,shifts, indicates instantaneous velocity of mainstream blood flow throughout systole. Interpretation of TAV signals. The terminology introduced by Mowat et al.31 for derived information has been followed: the distance traveled by the blood in each cycle is given by the area of each systolic complex and has been termed the "stroke distance." This is an index of stroke volume. Five consecutive complexes from each recording were measured and analyzed on a digitizer-microprocessor programmed and averaged to derive peak velocity (Vp), stroke distance (Sd), and flow time (Tf) (Fig. 2). Stroke distance was calculated from 1/2 Vp x Tf. For simplicity, a triangular approximation to each complex was drawn and the vertexes of circumscribed triangles were digi6zed. Whereas this approximation introduces positive errors of some 10% to 25%, in Vp, Tf, and area, such an error is normally constant in any one subject. 24 Therefore both resting and exercising values are similarly affected. Functional changes with exercise, the main object of this study, should thus be substantially correct. Because only relatively small changes in aortic cross-sectional area of flow of blood lost to head and arms are likely to occur vcith exercise, the percentage change in stroke distance (%ASd) in our subjects should also be close to the percentage change in their stroke volume. The TAV test was considered normal (negative) if, as a result of exercise there was an increase in Sd equal to or greater than 6%. It was considered abnormal (positive) if there was an increase less than 6% or a decrease fi3rm the resting value. One-minute treadmill test After resting in the,;upine position for 10 to 15 minutes the arm and ankle arterial systolic pressures were measured by means of Doppler ultrasonography and a pressure cuff. The patient then walked on a treadmill (4 kin/hour, 10% slope) for 1 minute and returned to the couch. The ankle pressure was measured at 30 seconds and subsequently every minute until the pressure returned to the preexercise level. The recovery time (RT) was defined as the time required for the postexercise ankle pressure to return to the preexercise level. Coronary angiography Left ventriculography and coronary angiography were performed according to the technique of Judkins 34 of Sones and Shirey. 3s A coronary lesion was considered to 'be hemodynamically significant if Table I. Results of ECG chest wall mapping stress test and TAV in 100 consecutive patients with intermittent claudication No. of ECG stress test TA V patients -- - 52 - + 1 + - 6 + + 41 TAV, Transcutaneous aortovelography; -, negative or inconclusive (see Table II); +, positive. Table II. The distribution of coronary artery territories with significant disease (>50% diameter stenosis) as indicated by the ECG chest wall mapping stress test in the 100 patients with intermittent claudication Test findings Negative test Inconclusive Total One-vessel disease LAD Cx R Two-vessel disease R+ Cx R+ LAD LAD + Cx Three-vessel disease Total No. of patients 36 17 53 5 4 7 26 47 LAD, Left anterior descending coronary artery; Cx, circumflex coronary artery; R, right coronary arteu. there was occlusion or a stenosis equal to or greater than 50% in diameter. The members of the team who interpreted the angiograms were not aware of the electrocardiographic findings. RESULTS Part I The results of the ECG chest wall mapping stress test and TAV are shown in Table I. Both tests were negative in 52 and both positive in 41. The TAV test alone was positive in one, and the stress test alone was positive in six. The number of territories indicated as having significant coronary artery disease by the ECG chest wall mapping stress test are shown in Table II. The ECG chest wall mapping was truly negative in 36 of the 53 because 85% of the target heart rate was reached. It was considered to be "inconclusive" in the remaining 17 because of an inadequate heart rate response.

80 Salmasi, Nicolaides, Al-I~toubi et al. Journal of VASCULAR SURGERY Table III. The distribution of coronary artery territories with significant disease (->50% diameter stenosis) as indicated by the ECG chest wall mapping stress test in the 41 patients in whom both the latter test and the TAV test were positive Test findi~zgs No. of patients Two-vessel disease R+Cx 5 R+ LAD 4 LAD + Cx 7 ~ Three-vessel disease 25 ~ LAD, Left anterior descending coronatw artery; Cx, circumflex coronary artery; R, right coronary artery. *Paticnts who underwent angiography. Pressure Index at rest 1-0- 0.8 0.6 0.4 Table III shows the number of territories with significant coronary disease in the 41 patients in whom both the ECG stress test and TAV were positive. It also shows the findings on the basis of which the 32 patients who underwent coronary angiography were selected. Part II Coronary angiography confirmed the severi~ of coronary artery disease in the 32 patients selected for part II of the study (Table III). The correlation between ECG chest wall mapping stress test and coronary angiography is shown in Tables IV and V. The claudication distance of the 32 patients was between 50 and 250 meters. Thirteen of the 32 patients subsequently underwent a coronary artery bypass operation. Seventeen of the remaining 19 were treated medically, 6 because they were biologically and chronologically older than 70 and 13 because the patients declined or the physician in charge did not support the recommendation for a coronary bypass operation. Two were referred elsewhere and were lost to follow-up. Medical therapy consisted of maximum "antianginal therapy" of calcium antagonists in combination with nitrates short of producing side effects. Transdermal nitrate patches (10 mg) during the day was the preferred method of nitrate administration. Eight patients requiring beta adrenoceptor blocking agents were maintained on a cardioselective preparation. Six weeks after myocardial revascularization or medical therapy there was improvement in the clandication distance. Twelve were able to walk without being limited by clandication. Fifteen reported improvement with a two to tenfold increase in claudication distance. No change occurred in the remaining 0-2 0~ I Before After Fig. 4. The ankle pressure index at rest in 30 patients before and after. The results are presented as median and 90% tolerance levels (90% range), p > 0.05, Wilcoxon rank sum test. three patients. The pressure index (before and after exercise), pressure recovery time (time taken for pressure index to return to preexercise level), and change in stroke distance before and after myocardial revascularization or medical therapy are shown in Figs. 4-7. Because the results are not normally distributed they are shown as median and 90% tolerance levels (90% range). They indicate that therapy (surgical or medical) for myocardial ischemia had no effect on the pressure index at rest, but an increase in prcssure index occurred at the end of i minute of exercise accompanied by a shorter recovery time. Also, after an increase occurred in stroke volume as a result of exercise. Ischcmic ST segment response during the follow-up ECG chest wall mapping stress test was no longer evident in 11 of the 13 patients who underwent corrective coronary artery surgery. Marked improvement was apparent in the remaining two and in those treated medically. No change occurred in the presence of Q waves after medical or myocardial revascularization.

Volume 14 Number 1 luly 1991 Silent myocardial ischemia in patients with intermittent claudications 81 Table IV. Number of diseased coronary arteries in the 32 patients subjected to coronary angiography Number of diseased coronary arteries o1"1 angiography ~ (->S0% stenosis) ~Coronary angiography is the independent variable Number of diseased coronary arteries indicated by ECG chest wall mapping stress test 0 1 2 3 Total 2 2 2 5 3 8 3 22 22 Total 7 25 32 Table V. Findings in individual coronary artery territories in the 32 patients subjected to coronary angiography Significant stenosis (->50%) on coronary ECG chest wall mapping stress test None LAD Cx R None 7 1 3 1 LAD 29 angiography ~ Cx 26 R 2 2 25 LAD, Left anterior descending coronary, artery; Cx, circumflex coronary, artery; R, right coronary artery. ~Coronary angiography is the independent variable. DISCUSSION Patients undergoing peripheral vascular reconstruction have an overa[~ perioperative mortality rate of 2% to 5% and a 3-year mortality rate of 30% to 40%; most deaths were due to myocardial infarction. 6-1 In contrast, the management of ischemic heart disease, in terms of short-term and long-term results, has improved dramatically in the last 15 years. The overall perioperative mortality rate of aortocoronary bypass in many centers is now less than 2%, and the 5-year survival rate is greater than 90%. 36-3s A question often asked :is why, despite the advent of aortocoronary bypass surgery, advances in vascular surgical and anesthetic techniques, and availability of intensive care units for postoperative care, the same degree of success has not been attained in patients undergoing peripheral vascular reconstructive surgery. This lack of success is because without the availability ofnoninvasivc cardiac investigations, the presence of concomitant coronary artery disease has often been overlooked by the vascular surgeon,-and the cardiac status of the patient has been less than adequately evaluated before peripheral vascular reconstructive procedures. The incidence of severe correctable coronary artery disease in candidates for peripheral vascular surgery is high in all age groups. It is 22% in those younger than 50 years rising to 42% in those older than 70 years. 5 This, taken together with the increased survival achieved when vascular reconstruction is performed after myocardial revascularization, 39 has reinforced the need for adequate cardiac evaluation. Because it is not practical to undertake routine coronary angiography in all patients with claudication, various noninvasivc screening tests have been suggested. Three relatively new noninvasive tests have been used in this study: ECG chest wall mapping stress test, TAV, and 1-minute treadmill test, the first two for,evaluation of cardiac function and the lattcr for assessment of lower limb ischemia. The principle of ECG chest wall mapping stress test is that the ECG changes monitored by electrodes on the chest wall reflect changes in the part of thc myocardium that lies immediately beneath it. Initially described by Fox et al.40-42 and subsequently modified by Salmasi et al.,20 this test has been found to havc the ability to detect the presence of significant (->80% stenosis) disease in individual coronary artery

82 Salmasi, Nicolaides, Al-Katoubi et al. Journal of VASCULAR SURGERY 16-0"8 14 12 0.6 Pressure Index after exercise 0.4 Recovery Time in minutes 10 0"2 l ~ ' 0 Before After Fig. 5. The ankle pressure index 30 seconds after the end of exercise in 30 patients before and after. The results are presented as median and 90% tolerance levels (90% range), p < 0.01, Wilcoxon rank sum test. territories. In our original study involving patients with chest pain, the presence or absence of disease was detected with a sensitivity and specificity of 87% and 85%, respectively, for the territory of LAD and diagonal artery, 74% and 85% for the right coronary artery, and 85% and 80% for the circumflex artery. 2 Two factors played an important part--namely the position of the leads on the chest wall that provided spatial information and the use of independent electrocardiographic criteria of coronary disease: the presence of Q waves at rest and ST segment depression as a result of exercise. Subsequently, when ECG chest wall mapping with bicycle ergometry was used to study patients with claudication it provided meaningf-ul results in nearly two thirds because of an adequate increase in heart rate. = The change in stroke volume in response to exercise, which can be easily measured with TAV is an indication of global left ventricular function. Tramcutaneous aortovelography initially developed by Light 23 in 1969 has been validated by comparison to simultaneous invasive measurements performed dur- 1 m I Before J After treatmenl Fig. 6. The recovery time in 30 patients before and after. The results are presented as median and 90% tolerance levels (90% range).p < 0.01, Wilcoxon rank sum test. ing cardiac catheterization. The reproducibility of the percentage change in stroke distance (%A Sd) was found to be high (coefficient of variation: 3% to 15%).41 A 9% to 48% increase in Sd has been shown to be induced by exercise in normal subjects and no change or a decrease up to 60% in patients with symptomatic ischemic heart disease associated with left ventricular dysfunction. 26,~7 Despite the decrease in Sd (stroke volume) most patients with ischemic heart disease maintain their cardiac output and blood pressure by an increase in the heart rate and peripheral vasoconstriction. 27 The 1-minute treadmill test was introduced by Laing and Greenhalgh in the early 1980s. 32'33 Its main advantage over the conventional test of exercising the patient up to claudication or a maximum of 5 minutes is that the decrease in ankle pressure measured at the end of this standard exercise is proportional to the severity of peripheral arterial dis-

Volume 14 Number 1 luly 1991 Silent myocardial ischemia in patients with intermittent claudications 83 ease; 43 also, almost every patient can exercise for 1 minute. This test has been found valuable in following progression of disease by sequential studies? 3 The results of our study indicate that in patients with claudication the ECG chest wall mapping exercise test used in,combination with TAV provides a good screening method for the selection of those who develop silent myocadial ischemia and left ventricular dysfunction. The patients in part II of the study had various grades of lower limb atherosclerotic disease. However, often the leg symptoms were disproportionately severe for the angiographic findings. Many had limiting claudication at 50 meters despite only superficial femoral artery occlusion with a normal aortoiliac segment, normal deep and popliteal arteries with three-vessel runoff. Our findings suggest that in many of these patients the symptoms of severe claudication are a manifestation of myocardial ischernia and a decrease in stroke volume. The results also suggest that of the silent myocardial ischemia can increase the claudication distance by improving left ventricular performance. This is supported by a higher ankle/pressure index and a shorter recovery time after therapy. The incidence of silent myocardial ischemia found in our patients is unexpectedly high and needs to be confirmed by Holter monitoring, which will indicate whether it occurs during routine daily activity also or just on bicycle ergometry. The incidence of silent myocardial ischemia is high probably because claudication is the presenting symptom, which preselects them for referral to the vascular surgeon rather than the cardiologist. Our findings reinforce the view that intermittent claudication is a marker of coronary artery disease, and that claudication is an indication to refer the patient to a cardiologist, not necessarily for routine coronary angiography but for noninvasive screening so that a subset of patients at high risk and requiring cardiac therapy in the first instance rather than a peripheral vascular operation can be identified. The effects of cardiac therapy on claudication shown by our study should be interpreted with caution. Vascular surgeons are aware that claudication is a variable symptom irrespective of therapy; also, that the symptoms of claudication tend to improve with time as a result of the development of colateral circulation and that patients are highly susceptible to placebo effect. The present study should be regarded as a pilot investigation. The observations need to be confirmed by a randomized placebo-controlled double blind study. Two separate groups of patients with claudication should be randomized: patients with se- 50-25 %ASd 0-25 / -50 - I Before After Fig. 7. The percentage change in stroke distance (%a Sd) in 30 patients before and after. The results are presented as median and 90% tolerance levels (90% range). p < 0.01, Wilcoxon rank sum test. vere coronary artery disease and silent myocardial ischemia producing left ventricular dysfunction, and patients without concomitant coronary artery disease. Only in this way would it be possible to determine whether the effect of medications such as calcium antagonists and nitrates on claudication is from diminishing myocardial ischemia and improving left ventricular performance rather than from having a direct action on the perpiheral arterial system. We are grateful to the Cardiovascular Disease Educational and Research Trust for a grant to T.N.S. and Professor H.A.F. Dudley, Professor P. Guillou and Dr. R. Foalc for 'their support. REFERENCES 1. Cutler BS, Wheeler HB, Paraskos JA, et al. Assessment of operative risk with electrocardiographic exercise testing in I

84 Salmasi, Nicolaides, Al-Katoubi et al. Journal of VASCULAR SURGERY patients with peripheral vascular disease. Am J Surg 1979; 137:484-90. 2. McCabe CI, Rerdy NC, Abbott WM, et aft. The value of electrocardiogram monitoring during treadmill testing for peripheral vascular disease. Surgery 1981;89:183-6. 3. Kanazawa M, Rose H, Vyden JK, et al. The importance of cardiac monitoring in treadmill claudication testing. Prac Cardiol 198i;7:48-52. 4. Vecht RJ, Nicolaides AN, Brandao E, et al. Resting and treadmill electrocardiographic findings in patients with intermittent daudication. Int Angiol I982; 1:119-23. 5. Hertzer NR, Bevan EG, Young JR, et al. Coronary artery disease in peripheral vascular patients. A classification of 1,000 coronary angiograms and results of surgical management. Ann Surg 1984;199:223-33. 6. Crawford ES, Bomberger RA, Gkeser DH, et al. Aortoiliac occlusive disease: factors influencing survival and function following reconstructive operation over a twenty-five year period. Surgery 1981;40:1055-67. 7. Hertzer NR. Fatal myocardial infarction following lower extremity revascularization. 273 patients followed 6 to 11 postoperative years. Ann Surg i981;i93:492-8. 8. Jamieson WILE, Jancesz TM, Migagishima RT, Geran A. Influence of ischemic heart disease on early and late mortaliq, after surgery for peripheral occlusive vascular disease. Circulation 1982;66:92-7. 9. Burnham NR, Johnson G, Gurri IA. Mortality risks for survivors of vascular reconstructive procedures. Surge D, 1982; 92:1072-6. i0. Bergan JJ, Veith FJ, Bernhard VM, et al. Randomization of autogenous vein and polytetrafluoroethylene grafts in femoro-distal reconstruction. Surgery 1982;92:921-30. ii. Boucher CA, Brewster DC, Darling RC, et al. Determination of cardiac risk by dipyridamole thallium imaging before peripheral vascular surgery. N Engl J Med 1985;312:389-94. 12. McPhail NV, Ruddy TD, Calvin JE, Davies RA, Barber GG. A comparison of dipyridamole-thallium imaging and exercise testing in the perdiction of postoperative cardiac complications in patients requiring arterial reconstruction. J VASC SURG I989;i0:51-6. 13. Raby KE, Goldman L, Creager MA, et al. Correlation between pre-operative ischemia and major cardiac events after peripheral vascular surgery. N Engl J Med i989;321:1296-300. 14. Ouyung P, Gerstenblith G, Furman WR, et al. Frequency mad significance of early postoperative silent myocardial ischemia in patients having peripheral vascular surgery. Am J Cardiol 1989;64:113-6. 15. Gottlieb SO. Silent ischemia: the American viewpoint. Cardiol Prac 1990;8:16-24. 16. Gordon DJ, Ekelund LG, Karon JM, et at. Predictive value of the exercise tolerance test for mortality in North American men: the lipid research clinics mortality follow-up study. Circulation 1986;74:252-61. 17. Multiple risk factor intervention trial research group: exercise electrocardiogram and coronary heart disease mortality in the multiple risk factor intervention trial. Am J Cardiol 1985; 55:16-24. 18. Giagnoni E, Secchi MB, Wu SC, et al. Prognostic value of exercise EKG testing in asymptomatic normotensive subjects: a prospective matched study. N Engl J Med 1983;309: 1085-8. 19. Chiavrello M, Indolti C, Cotecchia MR, Sifola C, Romano M, Condorelli M. Asymptomatic transient ST changes during ambulatory ECG monitoring in diabetic patients. Am Heart J 1985;I10:529-34. 20. Sahnasi A-M, Nicolaides AN, Vecht RJ. Electrocardiographic chest wall mapping in the diagnosis of coronary artery disease. Br Med J 1983;2:9-12. 21. Sonecha TN, Nicolaides AN, Salmasi A-M, Nicolaides EP, Papadakis K, Nissiotis T. Noninvasive detection of coronary artery disease in patients presenting with dandication. J Intern Angiol 1990;9:79-83. 22. Light LH. Noninvasive ultrasonic technique for observing flow in the human aorta. Nature i969;244:1119-21. 23. Salmasi A-M, Salmasi SN, Nicolaides AN. The value of exercise-induced U-wave inversion on ECG chest wall mapping in the identification of individual coronary arterial lesions. Eur Heart J 1985;6:437-43. 24. Sequeira RF, Light LH, Cross G, Raftery EB. Transcutaneous aortovelography. A quantitative evaluation. Br Heart J 1976;38:443-50. 25. Buchtal A, Hanson GC, Peisach AR. Transcutaneous aortovelography, potentially useful technique in management of critically ill patients. Br Heart J i976;38:451-6. 26. Salmasi A-M, Salmasi SN, Dore C, Nicolaides AN. Noninvasive assessment of changes in aortic blood velocity and its derivatives with exercise in normal subjects by Doppler ultrasound. J Cardiovasc Surg 1987;28:321-7. 27. Salmasi A-M, Salmasi SN, Nicolaides AN, et al. Noninvasive assessment of left ventricular function in coronary artery disease by Doppler stress testing. J Cardiovasc Surg 1987; 28:313-20. 28. Bryg RJ, Labovitz AJ, Mehdirad AA, Williams GA, Chaitman BR. Effect of coronary artery disease on Doppler-derived parameters of aortic flow during upright exercise. Am J Cardiol 1986;58:14-9. 29. Salmasi A-M, Salmasi SN, Nicolaides AN. Assessment of the effect of coronary artery bypass grafting on left ventricular performance by Doppler measurement of the aortic blood velocity during exercise. J Cardiovasc Surg 1988;29:89-94. 30. Nicolaides AN, Foale R, Cunningham D, Sonecha T, Salmasi A-M. Preoperative evaluation and risk minimization in the patient requiring vascular surged,. In: Veith FJ, ed. Current critical problems in vascular surgery. St. Louis: Qualiw Medical Publishing, Inc, 1989:415-29. 31. Mowat DHR, Hites NE, Rawley JM. Aortic blood velocity measurement in healthy adults using a simple ultrasound technique. Cardiovasc Res 1983;17:75-80. 32. Laing SP, Greenhalgh RM. Standard exercise test to assess peripheral arterial disease. Br Med J 1980;280:13-6. 33. Laing S, Greenhalgh RM. The detection and progression of asymptomatic peripheral arterial disease. Br J Surg 1983;70: 628-30. 34. Judldns MP. Selective coronary arteriography, A direct percutaneous transfemoral technique. Radiology 1967;89:815-24. 35. Sones FM, Sirey EK. Cine coronary arteriography. Mod Cone Cardiovasc Dis 1962;31:735-8. 36. Julian DG. Coronary artery bypass surgery: the European prospective randomized study. Implications for management after infarction. Tex Heart Inst J 1982;9:483-90. 37. CASS principal investigators and their associates. Coronary artery surgery study (CASS): A randomized trial of coronary artery bypass surgery. Survival data. Circulation 1983;68: 939-60. 38. Loop FD, Cosgrove DM, Lytle BW, et al. Life expectancy after coronary artery surgery. Am J Surg 1981;141:665-71. 39. Hertzer NR, Young JR, Bevan EG, et al. Late results of coronary bypass ha patients presenting with peripheral vascular disease. Five-year survival according to age and clinical cardiac status. Cleve Clin Q 1986;53:133-43.

Volume 14 Number 1 July 1991 Silent myocardial ischemia in patients with intermittent claudications 85 40. Fox K, Selwyn A, ShilSngford J. Precordial electrocardiographic mapping after exercise in the diagnosis of coronary artery disease. Am J Cardiol 1979;43:541-6. 4i. Fox K, Selwyn A, Oakley D, et al. Relation between the precordial projection of ST segment changes after exercise and coronary angiographic findings. Am J Cardiol 1979; 44:1068-75. 42. Fox KM, Selwyn AP, Shillingford JP. Projection of electrocardiographic signs in precordial maps after exercise in patients with ischemic heart disease. Br Heart J 1979;42:416-21. 43. Yao JST, Hobbs JT, Ir:ine WT. Ankle systolic pressure measurements in arterial disease affecting the lower extremities. Br J Surg 1969;56:675-9. DISCUSSION Dr. Craig MiUe, r (Stanford, Calif.). This work by Nicolaides et al. obviously was very carefully performed, but I am not so sure it was as carefully analyzed or interpreted. I have a series of points I would like to make and some questions for Mr. Nicolaides. The recurring question as I read the manuscript was the following: What is really new in this paper? It does, however, reaffirm two old clinical notions that are always deserving of reemphasis (1) In patients with claudication that is out of proportion to the pathoanatomic extent of disease, the symptoms may well be a result of inadequate cardiac reserve during exercise, something that we tend to forget over time; and (2) The authors suggested that claudication should really be a signal to refer the patient to a cardiologist. This point is well taken, and I agree wholeheartedly. These patients should be referred to a seasoned cardiologist with sound clinical judgment to evaluate them for both the presence of coronary disease as well as other forms of heart disease. Otherwise, what is really new here? Let us examine the two basic aims of their study. Number one, the authors wanted to determine the prevalence of silent myocardial ischemia with consequent left ventricular dysfimction in patients with intermittent claudication. Number two, they attempted to document the effects of therapeutic intervention designed to improve left ventricular performance on lower limb hemodynamics and claudication symptoms. Did the authors accomplish goal number one? I do not think so for many reasons. This is a highly selected patient population; coronary disease was already suspected on clinical grounds in 53 of the 100 patients. The authors did not provide any concordance information between symptomatic and asymptomatic cohorts. I do not think the authors can ask us to accept a priori the high specificity and sensitivity rates you saw on the screen that were obtained in patients with chest pain. There was no control or gold standard test, such as coronary arteriography, performed in 70 of the 100 patients. I also do not think the authors can justify the claim that ECG chest wall mapping was "truly negative" in 36 of 53 patients who had a negative test outcome simply because the targeted heart rate was achieved. How do they know that these were not falsenegative tests, which all of you realize is the biggest problem with conventional stress testing. The authors furthermore did not discimrinatc between silent and symptomatic ischemia. Finally, as a tangent in the context of an isolated screening test, I would hope these data would be corrob- orated with FM Holter monitoring for ST segment depression, which I personally believe may well be the screening test of choice today. In terms of accomplishing goal number two, my verdict is "maybe." But again I have several questions. I do not think the authors can discriminate between cause and effect; that is, were the observed changes caused by improved left ventricular systolic performance, a change in left ventricular diastolic function, or perhaps caused by a direct effect of the drugs administered themselves (probably including ACE inhibitors and/or calcium channel blockers)? Many other points deserve mention: One should be cautious in drawing sweeping, general conclusions from such a small sample (only 30 patients). What about the heterogeneous patient subsets and modes of therapy (medical vs surgical)? The study was not randomized, and the medical limb was not placebo controlled. Even though nonparametric methods were used to test for statistical significance, the 90% confidence levels between the different groups were largely overlapping. Finally, I am afraid I cannot agree that their index-- percent change in stroke distance--is really a good indicator of left ventricular systolic mechanics. It is an ejection phase index; it correlates with stroke volume, but all ejection phase indexes are highly dependent on left ventricular loading conditions, patxicularly afterload (mean or peak left ventricular systolic wall stress). Were the changes (defined as % A SD) simply a result of change in left ventricular afterload between the two examinations? I would be interested in the data normalized for mean systemic blood pressure, which was measured. Lastly, all of us should consider more frequently that left ventricular lusitropic dysfunction--or impaired diastolic left ventricular filling--occurs, which can limit exercise hemodynamic reserve in patients with coronary disease, as well as in older patients without coronary artery disease. Left ventricnlar hypertrophy and aging can cause left ventricular diastolic abnormalities, which can impair cardiac functional reserve. Dr. Ivan Crosby (Winston-Salem, N.C.). If you pursue the concept that Nicolaides et al. have espoused today it opens up Pandora's box in terms of the management of peripheral vascular disease in patients with coronary disease who may be asymptomatic in either the coronary or the peripheral vascular area, and I know that everyone in this country has struggled with Medicare precertification, and the like, in patients that you arc operating on for one area of vascular insufficiency.

86 Salmasi, Nicolaides, Al-Katoubi et al. Journal of VASCULAR SURGERY If you were to extrapolate your results, Mr. Nicolaides, I wonder if you could give us some information on how you manage two groups of patients all with proven peripheral vascular disease? The first group has historic evidence of coronary disease, either with angina or a history of myocardial infarction, and the second group has unstable angina or had an acute coronary insufficiency episode or a recent acute myocardial infarction. In the management of patients with coexisting system disease, namely coronary and peripheral, do you in your institution see any role for preoperative percutaneous transluminal coronary angioplasty? And, do you manage patients differently with these coexisting problems, if they have just claudication versus threatened tissue loss? The concept of 70 years of age being a cut off for coronary disease is intriguing in this country for the management by surgery. How do you manage the "arteriopath"-- people with a viable brain, who have severe carotid disease, coronary disease, and peripheral vascular disease? Are they just routinely assigned to medical therapy? I was not sure from your presentation as to how many of these 100 patients, if any, were actually operated on or treated for their peripheral vascular disease. Dr. Wiley Barker (Los Angeles, Calif.). I want to speak about a group of patients not so well documented as those just presented. In years past, when we did not have such accurate noninvasive evaluation available, the diagnosis of acute occlusion was at times in doubt. An episode clearly typical of an embolism was promptly followed by an evaluation of the heart, but a more gradual thrombosis was considered a local problem. Immediate intervention at times led to serious consequences when it became apparent that the thrombosis was often related to a preceding episode of myocardial ischemia and a diminished cardiac output. I learned to look to the patient's heart in all such cases before proceeding with other diagnostic or therapeutic endeavors. I believe this was a manifestation of the same phenomenon that Mr. Nicolaides has described so nicely with regard to claudication. Dr. William Fry (Ann Arbor, Mich.). I notice that on one of the slides that you presented there was ST depression before you started exercise that only amplified with exercise. You might point out to us what your measurements are on your ECGs related to your exercise as to positivity or negativity. Mr. Nicolaides. Perhaps I can start from the last question and say any patient who has ST segment depression before exercise does not get exercise. I think on that particular patient, the ST segment was less than 1 mm; therefore it was not considered significant. Dr. Miller has posed a number of questions, that have also occurred to us, because this work poses more questions than answers. I will answer some of the points. I will start by saying it is not possible in a 10-minute presentation to satisfy all the questions that might occur in the audience, but let us take it one by one. The first question Dr. Miller has posed is what is new. Well, what is new is to put it together and present the concept that intermittent claudication can be precipitated by silent myocardial ischemia. Here we have a noninvasive screening method by combining two tests that are very inexpensive--much cheaper than an intravenous thallium dipyridamole test--that can do a very good job in identifying patients who have ischemia as a result of exercise and who develop left ventricular dysfimction. What else is new? It is possible to identify the subgroup that has this problem. Stroke volume is a far more sensitive index of left ventricular dysfunction than ejection fraction, because the ejection fraction will increase as the volume of the heart increases during exercise. It will decrease in a much smaller number of patients than the stroke volume would decrease in, and we have shown in previous publications that there is a correlation between the ejection fraction, certainly at rest, in the damaged heart and the fall in stroke volume with exercise. What has been presented should be treated as a pilot study. We now need a proper randomized study, placebo controlled, in which we take a number of patients with claudication without any evidence of cardiac disease to see if nitrates have an effect on the claudication. The second group should be a group with claudication in which we have proved, by these tests and by coronary angiography, that there is severe coronary disease and left ventricular dysfimction. They should also be randomized into placebo or nitrates groups. Now, as far as the second commentator is concerned, the question asked was how would I treat various groups of patients--the people who have historic angina and now with dandication they no longer get angina, and the patient who has unstable angina. I would definitely investigate their coronaries. If they have severe three-vessel coronary disease, it has been our practice for the last 5 or 7 years to recommend that these patients have a coronary bypass operation first, before any operation on the legs is considered, such as an aortobifemoral reconstruction. To our surprise--and this is what stimulated the study--many of these patients would come back a month or 2 after their coronary bypass, and they would say, "I don't want an operation on my legs anymore. I can now walk very well." It was only when we investigated some of these patients that we began to realize that these people with better left ventricular fimction that had improved as a result of the coronary bypass had their claudication improved also. What about tissue loss? Of course, if there is tissue loss, at the top of the list is to save the leg. Of course under such circumstances we would go for an operation on the lower limb. We would cover the operation with beta blockers and nitrates. These days we have good anesthesiologists, and it is not a problem. What about the age of 70? We go on biologic age. The 75-year-old who looks 65 is treated as if he is 65. Finally, I thank Dr. Barker for his kind remarks.