Journal of Vascular Surgery
Volume 33, Issue 4 , Pages 721-727, April 2001

Pressure measurements at rest and after heavy exercise to detect moderate arterial lesions in athletes☆☆☆

Angers and Lyon, France

From the Laboratoire de physiologie et d'explorations vasculaires,a and the Laboratoire de statistique medicale,b Centre Hospitalier Universitaire, and the Service de chirurgie vasculaire, Hopital E Herriot.c

Received 2 May 2000; accepted 27 October 2000.

Article Outline

Abstract 

Purpose: This study defined how ankle arterial blood pressure measurements should be analyzed for the detection of moderate arterial disease (asymptomatic while walking). We used external iliac artery endofibrosis as a unique model of an isolated moderate arterial lesion, the role of which in exercise-related pain can be surgically proven. Methods: Patients who were ambulatory in our institutional referral center were studied. Brachial pressures, ankle pressures, and heart rate were measured simultaneously on all four limbs at rest and after maximal exercise in 108 healthy athletes and 78 patients (among 89 athletes referred for suspicion of endofibrosis) with confirmed or excluded external iliac endofibrosis. For these 78 patients, we calculated systolic ankle pressure change, ankle/brachial index, and deviation from the ankle/brachial index to heart rate regression line (DAHR) that was defined in the 108 healthy athletes. Results: In patients with endofibrosis, ankle/brachial index and ankle pressure were normal at rest. One minute after exercise, areas (mean ± SE of area) under the receiver operating characteristics curve for the diagnosis of endofibrosis were 0.91 ± 0.02, 0.91 ± 0.03, 0.95 ± 0.02, and 0.96 ± 0.02 for ankle pressure, pressure change, ankle/brachial index, and DAHR, respectively. For all criteria, area decreased with time in the recovery period. Conclusion: After heavy-load exercise, the ankle/brachial index at minute 1 should be used rather than the systolic ankle pressure value or ankle pressure change as a means of improving the efficacy of the detection of endofibrosis in athletes. A 0.66 value of the index at minute 1 after maximal exercise seems an optimal cutoff point for clinical use, providing a 90% sensitivity rate and 87% specificity rate in the diagnosis of moderate arterial lesions. At rest and after 1 minute of recovery, the ankle/brachial index to heart rate relationship should be considered to be an efficient tool for analyzing the results of pressures measurements and improving detection efficiency. (J Vasc Surg 2001;33:721-7.)

 

Minor peripheral arterial lesions remain asymptomatic in sedentary or moderately active subjects. However, when the intensity of exercise is increased, mild-to-moderate arterial lesions can be responsible for exercise-related lower limb pain. In young subjects1 and in athletes of all ages,2 a long delay usually occurs before the vascular origin of the pain is suspected. Furthermore, once suspected, when moderate arterial lesions are revealed by means of investigation, the causal relationship between these arterial lesions and the exercise-induced pain is difficult to prove, because: (1) Doppler and pressure measurements at rest are often within normal limits; (2) a walking test may fail to reproduce the symptoms that appear during higher intensity exercise; (3) objective criteria to be used after heavy-load exercise are lacking; and (4) these moderate lesions are usually not treated surgically at this stage. Thus, recovery from treated lesions, as proof of the causal relationship between the symptoms and the lesions, may not occur.

A typical example of claudication during intense exercise caused by moderate arterial stenosis is external iliac artery endofibrosis (EIAE).2, 3, 4 EIAE affects mainly highly trained cyclists and is characterized by the progressive thickening of the endothelial wall of the artery with fibrotic material. Although it is generally focused on the first centimeters of the left iliac artery, localization on other sites (femoral artery) or bilateral lesions have been reported.2, 3, 4 To date, no etiology has been demonstrated. This disease is a unique model of an isolated moderate lesion that can be improved by means of surgery, allowing the confirmation of the causal relationship between the moderate arterial lesion and the exercise-related symptoms.

The measurement of arterial pressures at rest is usually performed as a noninvasive method of testing the arterial origin of the leg pain.5, 6 In moderate lesions, absolute systolic ankle pressure (ASBP) and ankle/brachial index (ABI) results can remain within the reference range, but their diagnostic performance can be increased by increasing blood flow and thereby increasing the pressure drop across the stenosis.7, 8 However, walking tests seem insufficient for moderate lesions revealed only during high-intensity exercise.9, 10 It has recently been suggested that ASBP alone and not ABI should be considered in the diagnosis of arterial lesions after exercise tests.11 However, after heavy-load exercise in healthy subjects, the change in ASBP is not linearly related to work load, and ABI may decrease.10, 12, 13 Although a wide variability exists in ASBP changes, some normalization is achieved through the calculation of ABI, or the ABI to heart rate (HR) relationship during heavy-load exercise.12, 14, 15 We compared different methods of interpretation of arterial pressure measurements before and after heavy-load exercise to diagnose EIAE in the athlete and define the optimal cutoff point to be used in this population.

We measured ankle and arm pressures and HR at rest and after maximal incremental exercise in two different protocols. Healthy athletes were used to define the normal ABI to HR relationship. In another group of athletes who were referred with suspected EIAE, we calculated absolute ASBP, ASBP changes from resting values, ABI, and deviation from the normal ABI to HR relationship (DAHR) as defined in the first protocol. We analyzed whether the causal relationship between moderate arterial lesions and heavy-load exercise-related lower limb pain can be proved with arterial pressure measurements.

Back to Article Outline

Material and methods 

Protocol 1 

One hundred eight healthy athletes (81 men, 27 women) were referred to the laboratory for physiologic evaluation of exercise performance. The mean age of the athletes was 25 ± 12 years, their mean weight was 65 ± 10 kg, and their mean height was 172 ± 8 cm (mean ± SD). All were trained athletes who did not smoke and had no symptoms. The results of a physical examination were normal at rest. All subjects were in good health, with no personal nor familial history of arterial disease or claudication and no known risk factors for atherosclerosis. The subjects were thoroughly informed about the protocol and gave their consent to the study, which was approved by our institutional ethics committee.

Protocol 2 

Since 1993, 89 highly trained athletes who did not smoke and had no known vascular risk factors (85 men, 4 women) and who were referred for suspected endofibrosis were examined at rest and after exercise. Many of these athletes were referred to the laboratory after multiple investigations (sometimes including Doppler tests at rest, vascular ultrasound scanning imaging, or both) had been performed and/or various treatments or physiotherapy had been attempted. The usual complaint was a subjective sensation of a swollen thigh or paralysis during sprinting or uphill climbing, which improved when the intensity of exercise decreased and reappeared for an almost constant level of exercise intensity. The mean age of the athletes was 27 ± 7 years, the mean height was 176 ± 5.0 cm, and the mean weight was 66.8 ± 4.0 kg (mean ± SD). Ultrasound scanning imaging of the iliac and femoral arteries was performed systematically by trained operators, as described earlier.16 Arteriography was not systematically performed, but proposed only when history, clinical examination at rest or after exercise with a prolonged arterial bruit and typical clinical pain, or Doppler or color Doppler imaging was considered to be convincing for the diagnosis of EIAE. For patients considered free of EIAE, follow-up was performed by calling the patients or their usual physician 3 to 6 months after the visit. In these patients, when the symptoms persisted, new evaluations were proposed. Eight of the subjects initially considered to be free of EIAE were eventually referred to arteriography after two or more investigations. Because none of these four methods of detection (including arteriography2, 9, 10) is reported to have a 100% sensitivity rate, surgery was proposed when three of the four investigations (history, examination, ultrasound scanning, arteriography) had results that were considered abnormal. Information about recovery in the athletes who underwent surgery was attained by direct contact with the patients or their usual physician 6 months or longer after surgery.

Only the patients who had undergone surgery and were able to return to competition with absolutely no symptoms were included in the diseased group (EIAE, n = 38). Among the patients considered to be free of EIAE, those who had been treated successfully for another disease of nonarterial origin and were subsequently symptom free were included in the group of nonarterial-diseased patients (NADPs, n = 40). The final diagnosis of these patients is presented in Table I.

Table I. Frequency of the different final diagnoses suspected in the 40 athletes included in the group of patients without arterial disease
Abnormal bicycle position4
Atypical sciatica6
Venous insufficiency2
Pubalgia4
Musculotendinous lesions15
Uncertain diagnosis or information missing9
In 11 subjects, the presence or absence of a causal relationship between arterial lesions and claudication was unclear, and they were excluded from further analysis to avoid inadequate classification. Of these 11 patients, two were lost to follow-up, and nine still had symptoms at follow-up. Therefore, of the initial 89 patients who were referred, only the 78 patients who were either able to return to competition absolutely symptom free after surgery (n = 38, EIAE) or who had recovered from a disease of nonarterial origin (n = 40, NADP) were included. Thus, in this retrospective analysis of patients who originally had unexplained lower limb pain during heavy exercise, we classified subjects based on the ultimate evidence regarding the link between their pain and the final diagnosis.

Among the 38 patients who had successfully recovered from surgery, endofibrosis was confirmed by means of a histologic examination in each case: 3 subjects had bilateral lesions, 6 subjects had right iliac stenosis, and the other 29 subjects had left iliac lesions. The frequency for each range of severity of arterial stenosis in this group of patients is reported in Fig 1 (for patients with bilateral lesions, the stenosis is quantified for the more severe lesion, assumed to be the first symptomatic leg during exercise).

  • View full-size image.
  • Fig. 1. 

    Histogram of the estimated severity of stenosis in endofibrosis while artery is submitted (echography or arteriography) or not submitted (histology or angioscopy) to intraluminal arterial pressure. A zero value is noted when examination was reported to be normal.

Percent stenosis was calculated as the ratio of (1) the arterial lumen to the arterial wall on a longitudinal scan for ultrasound scanning examinations; (2) the arterial diameter at the level of the stenosis to distal external iliac arterial diameter on a transverse view for arteriography; and (3) the lumen surface to the arterial surface on either a histologic cross section of the artery when available or angioscopic or surgical direct view. During surgery or with histology pieces (endofibrosis being a distensible lesion), the stenosis appears much more severe than with ultrasound scanning or arteriography, because the artery is not distended by intraluminal arterial pressure.

In both protocols, systolic arterial pressure was measured with four synchronized automatic sphygmomanometers (Dynamap Criticon; Johnson and Johnson, Tampa, Fla). The system allowed for automated measurements at 1-minute intervals. For this protocol, all values were automatically recorded on paper and then transferred to a computer for ABI calculation and analysis. The accuracy of this system has been demonstrated.17 Fifteencentimeter large pressure cuffs were positioned on both the upper arm and ankle arteries. Measurements were performed simultaneously on the four limbs, with the patient in the recumbent position in an air conditioned room (21° ± 2°C). One set of measurements was performed at rest. Subjects were then asked to cycle on an ergometer (PPG Hellige, Keiper/dynavit, Germany or VP100, Techmachine, France) until they were exhausted or until their lower limb pain became intolerable (in subjects referred for suspected EIAE). The initial workload was 100 W, and the workload was increased by 50 W to 300 W and then by 30 W every 3 minutes. After exercise, measurements were taken at minute 1 and repeated every minute for 10 minutes. For each measurement, ABI was calculated as the ratio of ankle systolic blood pressure and the mean of brachial systolic blood pressures.

In healthy subjects, we analyzed the regression line of the ABI-HR relationship to be used in protocol 2. In subjects referred for suspected EIAE, the sensitivity and specificity rates of different criteria in the diagnosis of EIAE were studied. The four criteria were (1) the preexercise and postexercise ASBP; (2) the preexercise to postexercise ASBP changes; (3) the preexercise and postexercise absolute value for ABI; and (4) the deviation from the ABI-HR relationship (DAHR). DAHR was calculated as the difference between the measured ABI and the ABI calculated with measured HR by using the formula resulting from protocol 1. In subjects in whom multiple investigations in the diagnostic process were performed, only the first exercise test was used for the analysis.

Analysis of results 

Linear regression analysis in protocol 1 by means of the least squares method and statistical analysis of protocol 2 were performed with statistical software (SPSS 6.1.2, SPSS, Chicago, Ill). A 2-tailed Mann-Whitney test was used as a means of testing whether differences existed in ABI values at rest between healthy legs and legs with EIAE. The receiver operating characteristic (ROC) of the different criteria was studied from protocol 2. On the ROC curve, the distance from the 100% sensitivity and specificity rates angle was calculated for different values of each criterion; its lowest value was considered to be the best acceptable compromise of sensitivity to specificity for the clinical use. Statistical analysis for area under the ROC curves (area ± SE of area) was performed with the method described by Hanley and McNeil.18, 19 In this method, the closer the area is to 1.00, the better the diagnostic performance of the test; 0.50 would result from a random choice. For all statistical tests, a 2-tailed probability level of P less than .05 was used to indicate statistical significance.

Back to Article Outline

Results 

Results (mean ± SD) for ASBP, ASBP-changes, ABI, and DAHR in healthy athletes and athletes in the EIAE or NADP groups at rest and every minute after exercise are presented in Fig 2.

  • View full-size image.
  • Fig. 2. 

    Mean ± SD at rest and after incremental maximal exercise of ankle systolic blood pressure (ASBP ), absolute pressure drop from rest (pressure change ), ankle to brachial index (ABI ), and deviation from normal ABI to heart rate relationship (DAHR ), in healthy athletes (hollow circles ) and in diseased leg (filled squares ) and nondiseased leg (hollow squares ) of subjects referred for suspected external iliac artery endofibrosis.

Protocol 1 

The ABI was 1.12 ± 0.09 and the HR was 64 ± 11 beats/min–1 at rest in the 108 healthy subjects (mean ± SD). Results for the ABI to HR relationship in protocol 1 are presented in Fig 3.

  • View full-size image.
  • Fig. 3. 

    ABI to HR relationship in 108 healthy athletes (filled circles ) and in diseased leg in EIAE subjects (hollow circles ) at rest and after incremental exercise on cycle ergometer.

Linear regression analysis of the relation resulted in this equation in this population of athletes: ABI = –0.585 × (HR/100) + 1.466, r 2 = 0.579, P < .0001.

Protocol 2 

At rest 

In spite of the presence of the arterial stenosis, ASBP was not significantly decreased in diseased legs, compared with healthy legs (149 ± 16 vs 150 ± 15 mm Hg, respectively), whereas small but significant differences were found in ABI (1.11 ± 0.07 vs 1.14 ± 0.07; P < .01) and DAHR (–0.02 ± 0.07 vs 0.03 ± 0.10; P < .001). ASBP at rest was shown by means of ROC curve analysis to provide no information on the presence or absence of an arterial lesion, because the area under the curve (0.49 ± 0.05) was not different from the results of a random choice. Areas at rest for ABI (0.62 ± 0.05) or DAHR (0.69 ± 0.02) were significantly larger than those for ASBP (P < .05 and P < .0005, respectively), but still reflected very low diagnostic efficiency.

One minute after exercise 

A wide variability exists in the absolute values of ASBP 1 minute after exercise in both healthy and affected legs. Paradoxically, values as low as 73 mm Hg were found in healthy legs, and values as high as 163 mm Hg were found in diseased legs. After exercise, the ABI decreased significantly in both healthy legs and in legs with EIAE. At the first minute of recovery, the area under the ROC curve was 0.91 ± –0.03 for ASBP-changes and 0.91 ± 0.02 for ASBP. The area was significantly greater for ABI, 0.95 ± 0.02, as compared with ASBP-changes (P < .05, r = 0.80) or the ASBP value (P < .02, r = 0.85). For the DAHR criterion, although the area was larger for DAHR (0.96 ± 0.02) than for the other criteria, comparison with ASBP (P = .10, r = 0.39) or ASBP-changes (P = .11, r = 0.34) showed no significant difference, because a low correlation exists with these methods. Values for 100% sensitivity or specificity rate in the diagnosis of EIAE are reported in Table II.

Table II. Full range of the different criteria for the diagnosis of moderate lower extremity arterial disease, resulting in 100% sensitivity or specificity rate at the first minute of recovery from incremental heavy load exercise
% Specificity for 100% sensitivityValue% Sensitivity for 100% specificityValue
Ankle pressure change39+15 mm Hg24–90 mm Hg
Absolute ankle pressure44163 mm Hg3272 mm Hg
ABI520.78510.43
Deviation from the ABI-HR relationship16+0.2019.5–0.40
The optimal ASBP-change was a 30–mm Hg decrease that provided an 83% sensitivity rate and 89% specificity rate in the diagnosis of endofibrosis. The optimal cutoff point for an ASBP of 121 mm Hg provided an 85% sensitivity rate and 86% specificity rate in the detection of EIAE in our population. Finally, sensitivity/specificity rates were 90%/86% and 93%/90% when the normal limits of ABI and DAHR were 0.66 and –0.10, respectively.

More than 1 minute of recovery 

During the whole recovery period, the diagnostic performance of all decreased with time (Fig 4), whereas the cutoff point between normal and abnormal results slowly returned to resting value (Fig 5).

  • View full-size image.
  • Fig. 4. 

    Evolution of diagnostic efficiency (area under ROC curve) of the different diagnostic criteria at rest and with time in recovery period from maximal exercise (0.50 would be the result of a random choice, 1.00 the result of an ideal test). a and b show statistically significant differences with both ankle pressure and pressure change for ankle to brachial index (ABI ) and deviation from normal ABI-HR relationship, respectively. DAHR, Deviation from normal ABI to HR relationship; ROC, receiver operating characteristic.

  • View full-size image.
  • Fig. 5. 

    Examples of determination of optimal cutoff point from calculation of the distance of ROC curve to the 100%/100% sensitivity/specificity rate angle for different values of ABI in the discrimination of normal and abnormal results in diagnosis of EIAE. The closer to zero the lower part of the curve, the better the diagnostic performance for the diagnosis of EIAE. The corresponding value on the x-axis is cutoff point to be used. As shown in this example, at different times in recovery period, the diagnostic performance of ABI decreases with time, whereas cutoff point increases toward resting values.

After the first minute of recovery, DAHR provided the larger area under the ROC curve. The difference was significant, with both ASBP and ASBP-changes showing the lowest results (Fig 4).

Back to Article Outline

Discussion 

In our study, the ABI at rest was in the range of normal reported values20, 21 in both the nonaffected and the affected legs of patients with EIAE and in the NADP group. This is not surprising, because moderate arterial stenosis of the iliac artery is probably not severe enough to induce a pressure drop at rest. However, athletes with a low resting HR are expected to have higher ABI values than sedentary people, because an inverse relationship between ABI and HR at rest was reported.22

Contradictory reports exist in the literature about ASBP response in healthy subjects after exercise,12, 13, 14, 23, 24, 25, 26, 27 and when a decrease of ASBP is found, this pressure drop is often suggested to be very short-lived as compared with that in diseased subjects. In such a case, the diagnostic performance of ASBP or ASBP-changes should increase with time in the recovery period (ie, early normalization in healthy subjects, persistent pressure drop in patients with the disease). This was not the case in our study. In a recent consensus conference,11 it was suggested that ASBP, and not ABI after exercise, should be considered when analyzing the results of exercise tests in the diagnosis of lower extremity arterial disease. Although this is probably true during moderate exercise (eg, walking), it is probably inapplicable to incremental bicycle exercise. In our population, ASBP and ASBP-changes from resting value appear to be among the least sensitive criteria for the diagnosis of EIAE.

Although ASBP changes vary from one subject to another, some normalization is achieved through the calculation of ABI.12 Recent studies have found a decrease of ABI after both treadmill and cycle heavy-load exercise in healthy subjects,13, 14, 15 but the normal limit was not defined. In the current study, ABI measurement at minute 1 provided a higher performance in the diagnosis of EIAE than did ASBP or ASBP-changes, but the diagnostic performance of ABI further decreased with time in the recovery period. This confirms the previously reported suggestion by Chevalier et al9, 28 that in patients with EIAE exercise should be maximal to attain very high blood flows and that measurements should be performed early in the recovery period. By using a 0.5 normal limit,10, 28 they reported an 80% to 85% sensitivity rate in the diagnosis of endofibrosis, but did not report a specificity rate. The 0.66 value found in the current study is close to this value. The difference may result from the type of exercise protocol used or, more likely, from the delay from the end of exercise.

Because an inverse ABI-workload relationship exists, some of the differences observed may result from differences in the training status or workload sustained. The ABI difference between trained and untrained subjects after maximal cycle exercise results in the same ABI-HR relationship.15 The ABI-HR relationship was used in an attempt to control for exercise or training-status differences between the subjects and take into account the differences in the recovery speed or hemodynamic profiles of the subjects. Studying the ABI-HR relationship in the current study provided the best diagnostic performance both at rest and during the whole recovery period, except at minute 1, because the correlation coefficient with other methods was low.

All the EIAE subjects recovered completely, and the stenosis was confirmed during surgery, excluding any possible false-positive results in the diseased legs. Nevertheless, it should be noted that in some cases the diagnosis was only ascertained by means of angioscopy immediately before surgery. For ethical reasons, arteriography, not being the gold standard10, 28 in EIAE and being an invasive investigation, was not performed systematically in the NADP group in the current study. Therefore, it could be suggested that some of the NADPs had false-negative results for EIAE. If so, the absolute value to be used for the cutoff point would be changed, but because a physiologic decrease of ABI is found in healthy subjects, the low diagnostic performance of resting values, the need for intense exercise to attain high blood flow through those moderate lesions, and the decrease of diagnostic performance of each criterion with time would all remain unchanged. Because of the lack of symptoms at follow-up and the unlikelihood of spontaneous disappearance of a vascular lesion, we think that inadequate negative classification was unlikely. Consistently, three of the nine patients who had symptoms on follow-up were patients in whom EIAE was suspected and who had refused surgery.

We conclude that, in young athletes complaining from exercise-related pain, when searching for moderate arterial lesions such as endofibrosis (1) resting ASBP is a very poor indicator of the presence of an arterial stenosis; (2) the diagnostic performance of pressure measurements increases when heavy-load exercise is performed to attain high blood flows through the lesions; (3) measurements should be performed as early as possible in the recovery period from exercise because the diagnostic efficiency of all interpretation methods of pressure measurement decreases along with blood flow in the recovery period; (4) the diagnostic performance of ASBP at rest and 1 minute after exercise can be improved by referring ankle pressure to the simultaneous arm pressure measurement (ABI); (5) with incremental maximal bicycle exercise, an ABI value of 0.66 at 1 minute of recovery is the optimal cutoff point for clinical use; and (6) at-rest and after the first minutes of the recovery period, the ABI-HR relationship is an easy, efficient tool to analyze the results of pressure measurements after exercise. Whether the results of the current study can be extended to subjects with more severe lesions, patients of all ages, exercise training or vascular pathology remains a subject for future investigations.

Back to Article Outline

Acknowledgements 

We thank Dr La Combe and Dr Charkoudian for their help in reviewing the grammar and style of the manuscript and Dr Loire (anatomopathologist), Dr L'hoste (radiologist), and the members of the Department of Surgery in Lyon for their help in the analysis of the lesions.

Back to Article Outline

References 

  1. Sise J, Shackford SR, Rowley WR, Pistone FJ. Claudication in young adults: a frequently delayed diagnosis. J Vasc Surg. 1989;10:68–74
  2. Abraham P, Chevalier JM, Saumet JL. External iliac artery endofibrosis in athletes. Sports Med. 1997;24:221–226
  3. Abraham P, Chevalier JM, Loire R, Saumet JL. External iliac artery endofibrosis in a young athlete. Circulation. 1999;100:38e
  4. Mosimann R, Walder J, Van Melle G. Stenotic intimal thickening of the external iliac artery; illness of the competition cyclists?. Vasc Surg. 1985;19:258–263
  5. Carter SA. Indirect systolic pressures and pulse waves in arterial occlusive disease of the lower extremities. Circulation. 1968;37:624–637
  6. Newmann AB, Sutton-tyrell K, Rutan GH, Locher J, Kuller LH. Lower extremity arterial disease in elderly subjects with systolic hypertension. J Clin Epidemiol. 1991;44:15–20
  7. Carter SA. Response of ankle systolic pressure to leg exercise in mild or questionable arterial disease. New Engl J Med. 1972;287:578–582
  8. Ouriel K, McDonnell AE, Metz CE, Zarins K. A critical evaluation of stress testing in the diagnosis of peripheral vascular disease. Surgery. 1982;91:686–693
  9. Chevalier JM, Enon B, Walder J, Barral X, Pillet J, Megret A, et al.  Endofibrosis of the external iliac artery in bicycle racers: an unrecognized pathological state. Ann Vasc Surg. 1986;1:297–303
  10. Chevalier JM, L'Hoste P, Bouvat E, Megret A, Saint-Andre JP, Ruault P. L'endofibrose artérielle du sportif de haut niveau; formes inhabituelles. Journal de Traumatologie du Sport. 1991;8:176–181
  11. Weitz JI, Byrne J, Clagett GP, Farkouh ME, Porter JM, Sackett DL, et al.  Diagnosis and treatment of chronic arterial insufficiency of the lower extremities: a critical review. Circulation. 1996;94:3026–3049
  12. Desvaux B, Abraham P, Colin D, Leftheriotis G, Saumet JL. Ankle systolic blood pressure following sub-maximal and maximal exercises in healthy young men. J Sports Med Phys Fitness. 1995;35:127–130
  13. De Cossard L, Marples M, Cloherty G, Marcuson RW. The monarch bicycle ergometer in screening arterial patients. Br J Surg. 1982;9:543–544
  14. Abraham P, Desvaux B, Saumet JL. Ankle-brachial index after maximum exercise in treadmill and cycle ergometers in athletes. Clin Physiol. 1998;18:321–326
  15. Desvaux B, Abraham P, Colin D, Leftheriotis G, Saumet JL. Ankle to arm index following maximal exercise in normal subjects and athletes. Med Sci Sports Exerc. 1996;28:836–839
  16. Abraham P, Leftheriotis G, Bourre Y, Chevalier JM, Saumet JL. Echography of external iliac artery endofibrosis in cyclists. Am J Sports Med. 1993;21:861–863
  17. Mundt K, Chambless LE, Burnham CB. Measuring ankle systolic blood pressure, validation of the dynamap 1846SX. Angiology. 1992;43:555–564
  18. Hanley JA, McNeil BJ. Method of comparing the area under receiver operating characteristic curve derived from the same cases. Radiology. 1983;148:839–843
  19. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143:29–36
  20. Fronek A, Johansen KH, Dilley RB, Bernstein EF. Non-invasive physiologic tests in the diagnosis and characterization of peripheral arterial occlusive disease. Am J Surg. 1973;126:205–214
  21. Hirai M, Schoop W. Clinical insufficience of Doppler velocity and blood pressure measurements in peripheral arterial occlusive disease. Angiology. 1984;85:45–53
  22. Abraham P, Desvaux B, Colin D, Leftheriotis G, Saumet JL. Heart rate corrected ankle-to-arm index in the diagnosis of moderate lower extremity arterial disease. Angiology. 1995;46:673–677
  23. King LT, Strandness JR, Bell JW. The hemodynamic response of the lower extremities to exercise. J Surg Res. 1965;5:167–171
  24. Stahler C, Strandness JR. Ankle blood pressure response to graded treadmill exercise. Angiology. 1967;18:237–241
  25. Baker JD. Post-stress doppler ankle pressures: a comparison of treadmill exercise with two other methods of induced hyperaemia. Arch Surg. 1978;113:1171–1173
  26. Laing SP, Greenhalgh RM. Standard exercise test to assess peripheral arterial disease. BMJ. 1980;1:13–16
  27. Yao ST. New techniques in objective arterial evaluation. Arch Surg. 1973;106:600–604
  28. Beck F. Arterial endofibrosis: typical and non-typical forms (a review of 164 cases)[dissertation]. In: Lyon, France:: University of Lyon; 1995;p. 1

☆ Competition of interest: nil.

☆☆ Reprint requests: Prof J. L. Saumet, Laboratoire de physiologie et d'explorations vasculaires, Centre hospitalier universitaire, 4 rue Larrey 49033 Angers Cedex 01 France (e-mail: jean-louis.saumet@univ-angers.fr ).

PII: S0741-5214(01)16229-4

doi:10.1067/mva.2001.112802

Journal of Vascular Surgery
Volume 33, Issue 4 , Pages 721-727, April 2001

Article Tools

* Image Usage & Resolution