Angiology
http://ang.sagepub.com/
Comparison of Three Progressive Exercise Protocols in Peripheral Vascular Occlusive Disease Andrew W. Gardner, James S. Skinner, Natalie R. Vaughan, Cedric X. Bryant and L. Kent Smith ANGIOLOGY 1992 43: 661 DOI: 10.1177/000331979204300806 The online version of this article can be found at: http://ang.sagepub.com/content/43/8/661
Published by: http://www.sagepublications.com
Additional services and information for Angiology can be found at: Email Alerts: http://ang.sagepub.com/cgi/alerts Subscriptions: http://ang.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://ang.sagepub.com/content/43/8/661.refs.html
>> Version of Record - Aug 1, 1992 What is This?
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
Comparison of Three Progressive Exercise Protocols in Peripheral Vascular Occlusive Disease Andrew W. Gardner, Ph.D. James S. Skinner, Ph.D. Natalie R. Vaughan, B.S. Cedric X. Bryant, Ph.D. and L. Kent Smith, M.D., M.P.H.*
TEMPE and PHOENIX, ARIZONA
Abstract
Although claudication pain and hemodynamic responses to exercise are usually clinically assessed via graded treadmill walking, measuring these responses to other commonly performed tasks may yield a more nearly complete evaluation of peripheral vascular occlusive disease. Thus, the purpose of this study was twofold: (1) to determine the reliability of claudication and hemodynamic responses to level walking and stairclimbing and (2) to compare these responses with those obtained with graded walking at similar oxygen consumption. Ten patients with stable claudication symptoms performed graded walking, level walking, and stairclimbing progressive protocols with respective increases in grade, walking speed, and stepping rate on a modified stairclimbing device every two minutes. Similar peak oxygen consumption (13.60 to 14.18 mL/kg/min) was attained with the three protocols (P = NS). Reliability coefficients for the times to onset and to maximal claudication pain during level walking (R = 0.95 and 0.95, respectively) and during stairclimbing (R = 0.92 and 0.82, respectively) were similar to those previously obtained during graded walking. Reliability coefficients for foot transcutaneous oxygen tension during and following level walking (R = 0.78 to 0.96) and stairclimbing (R = 0.65 to 0.98) and for ankle systolic blood pressure following level walking (R = 0.95 to 0.97) and stairclimbing (R = 0.90 to 0.98) were also similar to those previously found with graded walking. Additionally, claudication and hemodynamic measurements were similar among the three exercise protocols. Thus, because From the Exercise and Sport Research Institute, Arizona State University, Tempe, and the *Arizona Heart Institute, Phoenix, Arizona This study was financially supported by grants from the Rockport Walking Institute, Marlboro, Massachusetts, and StairMaster Sports/Medical Products, Inc., Kirkland, Washington
661 Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
662
graded walking, level walking, and stairclimbing progressive exercise protocols yield reliable and similar information about the hemodynamic severity of peripheral vascular occlusive disease, only one is needed for evaluation. Introduction The times to onset of claudication pain (CPT) and to maximal pain (MPT) have been used to evaluate the functional hemodynamic severity of peripheral vascular occlusive disease. 1-4 To provide further information about the circulatory impairment of the lower extremities during normal and hyperemic conditions, measurements of foot transcutaneous oxygen tension (TcP02) before, during and after exercise,5-9 and measurements of ankle systolic blood pressure (SBP) before and after exercise 10-12 are routinely taken. Because ankle SBP is dependent on mean systemic pressure as well as peripheral flow, it is frequently expressed relative to brachial systolic pressure, termed the ankle/brachial index (ABI). 13-15 Although claudication pain impairs many daily activities, most investigations have measured its rate of development only during graded walking. In contrast, little is known about the reliability and response of claudication pain during other frequently performed activities, such as level walking and stairclimbing. Even less is known about the reliability and response of hemodynamic measurements taken from the claudicating lower extremities during and following these activities. Thus, the purpose of this study was twofold: (1) to determine the reliability of claudication and hemodynamic responses to level walking and stairclimbing and (2) to compare these responses with those obtained with graded walking at similar oxygen consumption (V02) .
Materials and Methods
Recruitment Procedures Participants were recruited by posting flyers throughout the Phoenix metropolitan area and by mailing letters to patients of the Arizona Heart Institute. To obtain a sample of patients who were limited by their claudication, people were excluded for the following reasons: (1) lower extremity pain while at rest, (2) ST segment depression greater than 1 mm either at rest or during exercise, (3) exercise tolerance limited by factors other than claudication (eg, dyspnea, fatigue, dizziness), (4) exercise-induced lower extremity pain not of cardiovascular origin, (5) current use of medication for their claudication, and (6) diabetes. Sample Characteristics Of 34 consecutive people who were screened, 10 qualified for the study (8 men and 2 women). Three patients had had surgery in iliac (N =1 ) and femoral (N = 2) arteries two to six years prior to this study. Two patients had unilateral disease with stenoses present in the iliac (I limb) and femoral (1 limb) arteries. The remaining 8 patients had bilateral disease, and most of them had multiple stenoses present in each limb. The number of limbs with stenoses in the following arteries were: iliac (3 limbs), femoral (9), popliteal (10), and tibial (5). The stenotic locations were determined from previous arteriograms and were functionally assessed at the time of investigation via segmental blood pressure measurements. Three of the patients were smokers, but they refrained from smoking on the days of testing until completion of the proce-
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
663
weight, and hemodynamic variables of the group measured during severely diseased lower limb and the right upper limb are displayed in supine Table I. The procedures used in this study were approved by the institutional review board for research involving humans at Arizona State University. Written informed consent was obtained from each patient prior to investigation.
dures. The
mean
rest in the
age,
more
TABLE I 10 Claudicant Patients; Values Are Means + Standard Errors (MISE)
Clinical Characteristics
of
Stairclimbing Progressive Exercise Protocols Each patient performed one graded walking test (2 mph, 0 % grade with 2 % increase every two minutes), two level walking tests (2 mph, 0% grade with 0.3 mph increase every two minutes) and two stairclimbing tests (12 steps/minute with an increase of 4 steps/minute every two minutes on a modified stairclimbing device; step height was approximately 8 inches). One week separated each of the five tests. Only one graded test was done because the reliability of the claudication and hemodynamic responses during and following this protocol has been previously determined.’6 The graded protocol was performed first to measure the V02 obtained from each stage so that comparable values could be obtained during the level walking and stairclimbing protocols. The increment in walking speed (0.3 mph) and step rate (4 steps/minute) during level walking and stairclimbing, respectively, were selected because unpublished data from our laboratory demonstrated that the increases inV02were similar to those obtained by the 2 % grade increment. Because claudication pain is less reliable when claudicants hold onto the handrail as they walk,&dquo; support was allowed only for brief moments to maintain balance. During stairclimbing, patients were permitted to lightly touch the handrail to maintain balance.
Walking
and
Measurements Obtained Supine Rest. Patients rested supine for twenty minutes, after which hemodynamic measurements were recorded as previously described. 16-18 Foot TcP02 was measured on the dorsum of the foot over the third cuneiform bone with a polarographic electrode and a transcutaneous oxygen monitor. Ankle SBP was measured in the posterior tibial and dorsalis pedis arteries of both lower extremities by a bidirectional Doppler and bidirectional 8 MHZ probe. The artery yielding the higher value in the more severely diseased lower extremity was used for the ankle SBP’° Systolic and diastolic blood pressure (DBP) were also obtained in the brachial artery of the right upper limb by the auscultatory method. ABI was calculated, and heart rate was obtained from
an
electrocardiograph.
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
664 Exercise. V02 was measured on-line every fifteen seconds as patients breathed through a mouthpiece and a Daniels valve into a mixing chamber. A ventilation measurement module was used to measure inspiratory ventilation, and gas analyzers were used to measure oxygen and carbon dioxide content in the expired air. Prior to each test, ventilation was calibrated with a 3-liter syringe. Gas analyzers were calibrated with known gas concentrations and room air. Patients indicated when the onset of claudication pain occurred (CPT) and then proceeded to walk to maximal pain (MPT). Foot TCPO, and heart rate were continuously monitored during exercise, and values were recorded at CPT and MPT. Brachial pressures were taken during the final minute of each exercise stage. Supine Recovery. Foot TcP02 and heart rate were continuously monitored for fifteen minutes following exercise, and values were recorded at minutes 1, 2, 4, 6, 8, 10, 12, and 15. Ankle and brachial pressures were also measured, and ABI was calculated at each of these eight occasions.
Statistical Analyses The reliability coefficient (R)’9 and the coefficient of variation (CV)2° were determined for each variable before, during, and after exercise. Dependent t tests were used to compare peak V02 and claudication values between the two level walking tests and between the two stairclimbing tests. Two-factor, repeated measures ANOVAs (two tests x occasion of measurement within each test) were used to assess the hemodynamic variables measured before, during, and after level walking and stairclimbing. If similar results were found between the two tests of level walking and of stairclimbing, values were averaged and compared with those of graded walking by one-factor, repeated measures ANOVAs (protocol) for peak VOZ and claudication values, and by two-factor, repeated measures ANOVAs (protocol x occasion of measurement) for each hemodynamic variable. Tukey post-hoc comparisons were used to locate mean differences whenever a protocol effect, an occasion effect, or an interaction was found . 21 Statistical significance was set at P < 0.05 . Measurements are presented as means ± standard errors.
Results
Oxygen Consumption Reliability. PeakV02was reliable during level walking and stairclimbing, for values between the two level walking tests ( 13 .58 ± 0.63 and 13.61 ± 0.54 mL/kg/min) and between the two stairclimbing tests (14.11 ±0.81 and 13.33 ±0.57 mL/kg/min) were not different (P=NS). Furthermore, the R values were high for these protocols (0.87 and 0.89, respectively), and the CV values were low (7.3% and 7.8%, respectively). Comparison Among Protocols. As noted in Figure 1, the rate of increase in V02 during the graded protocol was lower than that predicted by a linear model published by the American College of Sports Medicine (ACSM) . Zz Given this result, the increments of V02 during the subsequent level walking and stairclimbing protocols were chosen to match the actual measured value rather than the predicted value. Measured peak V02 of level walking and stairclimbing was similar to that during graded walking ( 14 .18 ± 0. 77 mL/kg/min; P=NS). However, since the predicted peak V02 of graded walking was higher than that actually obtained, some energy may have been derived anaerobically. Consequently, the peak V02 during
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
665
walking and stairclimbing was matched to the predicted peak V02 of graded walking in a subsequent analysis by a repeated measures analysis of covariance (ANCOVA). 21
level
FIG. 1. Measured oxygen consumption (V02) of each the graded walking progressive protocol. Also shown is the predicted V02 (PV02) based on the equation of the American College of Sports Medicine.&dquo; Measured V02 during the level walking and stairclimbing protocols was similar to values depicted during the graded
patient (Px) during
protocol.
Claudication Pain Reliability. CPT and MPT were similar between the two level walking tests (P NS) and between the two stairclimbing tests (P = NS) (Table II) . The R values between the two tests of level walking and of stairclimbing were similar to those previously obtained with the graded protocol (0.89 and 0.93 for CPT and MPT, respectively).’6 The CV values for CPT and MPT during level walking and stairclimbing were also similar to those previously obtained during graded walking (15.8% and 12.6% for onset and maximal pain time, respectively).’6 Comparison Among Protocols. The average times to CPT and MPT between the two level walking tests and the two stairclimbing tests were similar to values measured during graded walking (P=NS) (Table II) . When values were compared at the same predicted peak VO2, adjusted means for CPT (233 .1, 253 .4, 248.6) and to MPT (441.1, 436.2, 433.7) were also similar among the graded walking, level walking, and stairclimbing protocols, respectively =
(P = NS) . TABLE II Mo.~M/ Pain Claudication Pain C/aM~c~oAt P~n n/Mc Time (CPT) P~n Time 7wM’ (MPT) ~CPT) and Maximal (mop7) During Graded Walking (G), Level Walking (L), and Stairclimbing Progressive Protocols, Values Are Means 1: Standard Errors, Reliability Coefficients (R), and Coefficients of Variation (CV)
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
666
Hemodynamic
Variables
The R and CV values of the hemodynamic measurements before, during, and after level and stairclimbing are shown in Tables III and IV, respectively. The reliability values were similar to those previously obtained with graded walking.’6
Reliability.
TABLE IV
TABLE III
sive Protocols
Coefficients of Variation (%) of Each Hemodynamic Variable Measured Before, During, and After Level Walking (L) and Stairclimbing (S) Progressive Protocols
*TcP02=transcutaneous oxygen tension, SBP=systolic blood pressure, ABI=ankle/brachial SBP index, DBP=diastolic blood pressure. tCPT=claudication pain time, MPT=maximal pain time.
*TcPO,=transcutaneous oxygen tension, SBP=systolic blood pressure, ABI=ankle/brachial SBP index, DBP=diastolic blood pressure. tCPT=claudication pain time, MPT=maximal pain time.
Reliability Coefficients of Each Hemodynamic Variable Measured Before, During, and After Level Walking (L) and Stairclimbing (S) Progres-
Comparison Among Protocols. Foot TcP02 (Figure 2), ankle SBP (Figure 3), ABI (Figure 4), and heart rate (data not shown) values were not different among the three protocols. Brachial pressures during graded walking were 2-10 mm Hg higher at maximal pain than during the other protocols (P < 0.05). However, when the exercise brachial pressures were adjusted for a given predicted peak V02, differences among the protocols no longer remained (P=NS). Within each protocol, the hemodynamic variables measured from the claudicating limb demonstrated a significant change (P < 0.05) during and following exercise (occasion of meadecreased while surement main effect). Foot TCP02 increased as patients stood (P < 0.05), they walked to CPT (P < 0.05) with no further reduction to MPT (P=NS), and was lower than resting baseline during the first twelve minutes of recovery (P < 0.05). Ankle SBP and ABI were lower than resting baseline during the first ten minutes of recovery (P < 0.05).
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
667
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
668 Discussion
Oxygen Consumption An objective of the present investigation was to match the V02 during each exercise stage of level walking and stairclimbing with that obtained during graded walking. In this manner, claudication and hemodynamic responses to the three exercise protocols could be compared at similar metabolic requirements. Although similar values were achieved, note that values during the first exercise stage of graded walking were higher than those predicted.22 This indicates that the claudicants were metabolically uneconomical while they walked. Furthermore, the rate of increase in V02 between stages was approximately 1.1 mL/kg/min, which is only about half the predicted value. This suggests that the claudicants were either unable to meet the increased energy requirement through aerobic metabolism or that they were slow in doing so, owing to inadequate perfusion of the working lower limb muscles. Because the graded protocol had the highest predicted energy requirement, and because the claudication and hemodynamic measurements taken from the claudicating limb were similar among the three protocols, it could be interpreted that relatively less hemodynamic stress was placed on the limb during this protocol. Thus, the data were also analyzed at a given predicted peak VOZ via ANCOVA statistical procedures. This analysis reduced the claudication values during graded walking and increased the values during level walking and stairclimbing. However, none of the results from the original analyses changed, except that brachial pressures were no longer lowest during stairclimbing. Claudication Pain The reliability of claudication pain during level walking and stairclimbing were similar to those previously reported for graded walking.’6 Furthermore, the claudication measures were similar when patients performed the three protocols, suggesting that similar information was obtained about the circulatory impairment of the claudicating limb. It is difficult to compare the results of the present investigation with those of previous studies because of methodological differences. For example, Bonde-Petersen23 and Hillestad24 used different speeds and grades than the present study used. Bonde-Petersen23 found that maximal walking distance was reduced more by increasing the grade from 0 ° to 5 ° at constant speeds of 4.0 and 4.9 km/hr than by increasing speed from 2.4 to 4.9 km/hr at a constant incline of 5 ° . This result was expected, however, since the increment inV02was greater with the alteration in grade (approximately 17.5 mL/kg/min) than with the change in speed (approximately 10.5 mL/kg/min). No measure of claudication pain distance was reported under either condition. Hillestad 24 found that an increase in speed from 2 to 4 km/hr at 0 % grade decreased maximal walking distance. Grade was not increased, and thus, its effect on distance could not be compared. No data on claudication responses during stairclimbing have been previously
reported. Hemodynamic Variables During Exercise. The response of each hemodynamic variable measured during exercise was similar among the three protocols. The reduction in foot TcP02 during exercise in claudicants is due to the vasodilation in the active musculature, thus compromising the delivery of oxygen to peripheral tissue.II.18 The similar reduction in foot TCPO, during each protocol to
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
669
pain suggests an equivalent amount of vasodilation. Although not statistithere cally significant, appears to be a tendency for foot TCPO, to be higher at CPT and MPT during stairclimbing than during either walking protocol. The hemodynamic significance of this trend is unclear but may be due to the lower rate of contraction as compared with walking and to the longer periods of standing while exercising.
maximal tolerable
The response of each hemodynamic variable measured during recovery was similar among the three exercise protocols. Thus, each demonstrated inadequate circulatory responses of the claudicating limbs to exercise. The peripheral circulatory variables (foot TcP02, ankle SBP, and ABI) were lower during the first ten to twelve minutes than those obtained at supine rest. In addition to a decrease in perfusion pressure, due to a reduction in cardiac output from exercise to recovery, the impairment of peripheral blood flow during recovery is also a result of the vasodilatation still present in previously active, ischemic lower limb muscles. Because arterial stenoses and occlusions diminish the peripheral circulation to more distal regions of the lower limb, blood flow to both muscular and cutaneous tissue cannot be met under a hyperemic condition in muscle.&dquo;&dquo;’ In less diseased or healthy individuals, however, the increased conductance in muscle following exercise does not occur at the expense of the peripheral circulation.5,18 As the vasodilatory stimulus gradually subsides, the peripheral circulation is restored toward baseline, manifested by an increase in foot TcP02, ankle SBP, and ABI.
During Supine Recovery.
Utility of the Three Progressive Protocols In considering the clinical utility of each protocol, level walking is probably the least feasible because some patients may complete the exercise stage associated with their maximal walking speed (3-4 mph) before MPT is attained. This problem would probably occur only in patients who are mildly diseased, and in such cases, maintaining their maximal speed would eventually lead to maximal pain. A second disadvantage to this protocol is that the safety of some patients may be compromised, especially while they are experiencing claudication pain, when limping is commonly observed. This would be particularly true in patients who are unable to walk fast (eg, severely diseased and frail claudicants). Although the safety of the level walking protocol might be enhanced by allowing patients to hold onto the handrail, this would not be advisable because the reliability of the claudication and hemodynamic responses would be reduced.’6 Thus, the level walking protocol appears to be clinically useful only in claudicants with moderate disease severity. Even in these patients, this protocol does not offer any distinct advantage to the more traditional graded protocol. The stairclimbing protocol does not have the limitations that are associated with the level walking protocol. Relatively large increases in energy cost can be achieved with each progressive exercise stage. Consequently, MPT would likely be attained before the highest possible power output, even in mildly diseased claudicants. Furthermore, there is less risk of falling because the patient can choose when to discontinue exercise. However, the stairclimbing protocol does have limitations of its own. First, the task of stairclimbing is more novel than walking on a treadmill for most patients, and second, handrail support (lightly touching the front handrail) appears to be necessary to maintain balance. As a result, more time is required to practice the exercise task, and periodic reminders to not grasp the handrails during exercise
Clinical
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
670 needed. Great care was taken to make with their fingers extended.
are
sure
that
patients only lightly
touched the handrail
Conclusions The main findings of this investigation were that: (1) the reliability values of the claudication and hemodynamic variables during and following level walking and stairclimbing were similar to those previously obtained with graded walking, and (2) the claudication and hemodynamic responses to the three protocols were similar. Thus, because graded walking, level walking, and stairclimbing progressive exercise protocols yield reliable and similar information about the hemodynamic severity of peripheral vascular occlusive disease, only one is needed for evaluation.
Acknowledgments Appreciation is extended to Craig Phelps, D.O.; Joel Sellers, D.O.; Scott Steingard, D.O.; and Larry Ashker, D.O., of the Center for Sports Medicine and Orthopedics in Phoenix for their medical supervision of the treadmill tests. We are also grateful to Andy Chittick of the Arizona Heart Institute for his assistance in patient recruitment and the loan of a bidirectional Doppler, bidirectional probe, and supplies. The authors are grateful to Novametrix Medical Systems, Inc. (Wallingford, CT) for the loan of a transcutaneous oxygen monitor and supplies. This study was financially supported by grants from The Rockport Walking Institute (Marlboro, MA) and Randal Sports/Medical Products, Inc. (Kirkland, WA). The authors are grateful to Novametrix Medical Systems, Inc. (Wallingford, CT) for the loan of a transcutaneous oxygen monitor and supplies. Andrew W. Gardner, Ph. D. University of Vermont Department of Medicine Metabolic Unit, Given C-332 Burlington, VT 05405
References 1.
2.
3.
4.
5.
Clyne CAC, Tripolitis A, Jamieson CW, et al: The reproducibility of the treadmill walking test for claudication. Surg Gynecol Obstet 149:727-728, 1979. Harris KA, Meads GE: A single treadmill exercise test does not accurately quantitate claudication. Can J Surg 30:446-448, 1987. Hiatt WR, Nawaz D, Regensteiner JG, et al: The evaluation of exercise performance in patients with peripheral vascular disease. J Cardiopulmonary Rehabil 12:525-532, 1988. Siggaard-Andersen J, Bonde-Petersen F: Intermittent claudication: A comparison between subjective and measured claudication walking distance. Angiology 19:426-434, 1968. Byrne P, Provan JL, Ameli FM, et al: The use of transcutaneous oxygen tension measurements in the diagnosis of peripheral vascular insufficiency. Ann Surg 200:159-165, 1984.
6. Christensen KS, Larsen JF, Klaerke M: Transcutaneous oxygen tension response to exercise in health and in occlusive arterial disease. Acta Chir Scand
152:657-660, 1986. 7. Holdich TAH, Reddy PJ, Walker RT, et al: Transcutaneous oxygen tension during exercise in patients with claudication. Br Med J 292:1625-1628, 1986. 8. Matsen FA III, Wyss CR, Simmons CW, et al: The effect of exercise upon cutaneous oxygen delivery in the extremities of patients with claudication and in a human laboratory model of claudication. Surg Gynecol Obstet 158:522-528, 1984. 9. Wyss CR, Matsen FA III, Simmons CW, et al: Transcutaneous oxygen tension measurements on limbs of diabetic and nondiabetic patients with peripheral vascular disease. Surgery 95:339-345, 1984. 10. Carter SA: Response of ankle systolic pressure to leg exercise in mild or questionable arterial disease. N
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
671 Engl J Med 287:578-582, 1972. 11. Strandness DE Jr, Bell JW: An evaluation of the hemodynamic response of the claudicating extremity to exercise. Surg Gynecol Obstet 119:1237-1242, 1964. 12. Sumner DS, Strandness DE Jr: The relationship between calf blood flow and ankle blood pressure in patients with intermittent claudication. Surgery 65: 763-771, 1969. 13. Baker JD, Dix D: Variability of Doppler ankle pressures with arterial occlusive disease: An evaluation of ankle index and brachial-ankle pressure gradient. Surgery 89:134-137, 1981. 14. Yao ST: Haemodynamic studies in peripheral arterial disease. Br J Surg 57:761-766, 1970. 15. Yao ST, Needham TN, Gourmoos C, et al: A comparative study of strain-gauge plethysmography and Doppler ultrasound in the assessment of occlusive arterial disease of the lower extremities. Surgery 71:4-9, 1972. 16. Gardner AW, Skinner JS, Cantwell BW, et al: Progressive vs single-stage tests for evaluation of claudication. Med Sci Sports Exerc 23:402-408, 1991. 17. Gardner AW, Skinner JS, Smith LK: Effects of handrail support on claudication and hemodynamic responses to single-stage and progressive treadmill protocols in peripheral vascular occlusive disease. Am
J Cardiol 68:99-105, 1991. 18. Gardner AW, Skinner JS, Cantwell
BW, et al: Relationbetween foot transcutaneous oxygen tension and ankle systolic blood pressure at rest and following exercise. Angiology 42:481-490, 1991. Safrit MJ: Evaluation in Physical Education, ed 2. Englewood Cliffs, NJ: Prentice-Hall, 1981, pp 96104. SPSS Inc: SPSS-X User’s Guide, ed 2. Chicago: SPSS Inc, 1986, p 110. Glass GV, Hopkins KD: Statistical Methods in Education and Psychology, ed 2. Englewood Cliffs, NJ: Prentice-Hall, 1984, pp 371-372 and 492-507. American College of Sports Medicine: Guidelines for Exercise Testing and Prescription, ed 2. Philadelphia: Lea & Febiger, 1980, pp 137-142. Bonde-Petersen F: The effects of varying walking speeds when measuring the claudication distance on horizontal and sloping levels. Acta Chir Scand 133:627-630, 1967. Hillestad LF: The peripheral blood flow in intermittent claudication: IV. The significance of the claudication distance. Acta Med Scand 173:467-478, 1963. Provan JL, Moreau P, MacNab I: Pitfalls in the diagnosis of leg pain. Can Med Assoc J 121:167-172, 1979.
ship 19.
20. 21.
22.
23.
24.
25.
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
http://ang.sagepub.com/
Comparison of Three Progressive Exercise Protocols in Peripheral Vascular Occlusive Disease Andrew W. Gardner, James S. Skinner, Natalie R. Vaughan, Cedric X. Bryant and L. Kent Smith ANGIOLOGY 1992 43: 661 DOI: 10.1177/000331979204300806 The online version of this article can be found at: http://ang.sagepub.com/content/43/8/661
Published by: http://www.sagepublications.com
Additional services and information for Angiology can be found at: Email Alerts: http://ang.sagepub.com/cgi/alerts Subscriptions: http://ang.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://ang.sagepub.com/content/43/8/661.refs.html
>> Version of Record - Aug 1, 1992 What is This?
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
Comparison of Three Progressive Exercise Protocols in Peripheral Vascular Occlusive Disease Andrew W. Gardner, Ph.D. James S. Skinner, Ph.D. Natalie R. Vaughan, B.S. Cedric X. Bryant, Ph.D. and L. Kent Smith, M.D., M.P.H.*
TEMPE and PHOENIX, ARIZONA
Abstract
Although claudication pain and hemodynamic responses to exercise are usually clinically assessed via graded treadmill walking, measuring these responses to other commonly performed tasks may yield a more nearly complete evaluation of peripheral vascular occlusive disease. Thus, the purpose of this study was twofold: (1) to determine the reliability of claudication and hemodynamic responses to level walking and stairclimbing and (2) to compare these responses with those obtained with graded walking at similar oxygen consumption. Ten patients with stable claudication symptoms performed graded walking, level walking, and stairclimbing progressive protocols with respective increases in grade, walking speed, and stepping rate on a modified stairclimbing device every two minutes. Similar peak oxygen consumption (13.60 to 14.18 mL/kg/min) was attained with the three protocols (P = NS). Reliability coefficients for the times to onset and to maximal claudication pain during level walking (R = 0.95 and 0.95, respectively) and during stairclimbing (R = 0.92 and 0.82, respectively) were similar to those previously obtained during graded walking. Reliability coefficients for foot transcutaneous oxygen tension during and following level walking (R = 0.78 to 0.96) and stairclimbing (R = 0.65 to 0.98) and for ankle systolic blood pressure following level walking (R = 0.95 to 0.97) and stairclimbing (R = 0.90 to 0.98) were also similar to those previously found with graded walking. Additionally, claudication and hemodynamic measurements were similar among the three exercise protocols. Thus, because From the Exercise and Sport Research Institute, Arizona State University, Tempe, and the *Arizona Heart Institute, Phoenix, Arizona This study was financially supported by grants from the Rockport Walking Institute, Marlboro, Massachusetts, and StairMaster Sports/Medical Products, Inc., Kirkland, Washington
661 Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
662
graded walking, level walking, and stairclimbing progressive exercise protocols yield reliable and similar information about the hemodynamic severity of peripheral vascular occlusive disease, only one is needed for evaluation. Introduction The times to onset of claudication pain (CPT) and to maximal pain (MPT) have been used to evaluate the functional hemodynamic severity of peripheral vascular occlusive disease. 1-4 To provide further information about the circulatory impairment of the lower extremities during normal and hyperemic conditions, measurements of foot transcutaneous oxygen tension (TcP02) before, during and after exercise,5-9 and measurements of ankle systolic blood pressure (SBP) before and after exercise 10-12 are routinely taken. Because ankle SBP is dependent on mean systemic pressure as well as peripheral flow, it is frequently expressed relative to brachial systolic pressure, termed the ankle/brachial index (ABI). 13-15 Although claudication pain impairs many daily activities, most investigations have measured its rate of development only during graded walking. In contrast, little is known about the reliability and response of claudication pain during other frequently performed activities, such as level walking and stairclimbing. Even less is known about the reliability and response of hemodynamic measurements taken from the claudicating lower extremities during and following these activities. Thus, the purpose of this study was twofold: (1) to determine the reliability of claudication and hemodynamic responses to level walking and stairclimbing and (2) to compare these responses with those obtained with graded walking at similar oxygen consumption (V02) .
Materials and Methods
Recruitment Procedures Participants were recruited by posting flyers throughout the Phoenix metropolitan area and by mailing letters to patients of the Arizona Heart Institute. To obtain a sample of patients who were limited by their claudication, people were excluded for the following reasons: (1) lower extremity pain while at rest, (2) ST segment depression greater than 1 mm either at rest or during exercise, (3) exercise tolerance limited by factors other than claudication (eg, dyspnea, fatigue, dizziness), (4) exercise-induced lower extremity pain not of cardiovascular origin, (5) current use of medication for their claudication, and (6) diabetes. Sample Characteristics Of 34 consecutive people who were screened, 10 qualified for the study (8 men and 2 women). Three patients had had surgery in iliac (N =1 ) and femoral (N = 2) arteries two to six years prior to this study. Two patients had unilateral disease with stenoses present in the iliac (I limb) and femoral (1 limb) arteries. The remaining 8 patients had bilateral disease, and most of them had multiple stenoses present in each limb. The number of limbs with stenoses in the following arteries were: iliac (3 limbs), femoral (9), popliteal (10), and tibial (5). The stenotic locations were determined from previous arteriograms and were functionally assessed at the time of investigation via segmental blood pressure measurements. Three of the patients were smokers, but they refrained from smoking on the days of testing until completion of the proce-
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
663
weight, and hemodynamic variables of the group measured during severely diseased lower limb and the right upper limb are displayed in supine Table I. The procedures used in this study were approved by the institutional review board for research involving humans at Arizona State University. Written informed consent was obtained from each patient prior to investigation.
dures. The
mean
rest in the
age,
more
TABLE I 10 Claudicant Patients; Values Are Means + Standard Errors (MISE)
Clinical Characteristics
of
Stairclimbing Progressive Exercise Protocols Each patient performed one graded walking test (2 mph, 0 % grade with 2 % increase every two minutes), two level walking tests (2 mph, 0% grade with 0.3 mph increase every two minutes) and two stairclimbing tests (12 steps/minute with an increase of 4 steps/minute every two minutes on a modified stairclimbing device; step height was approximately 8 inches). One week separated each of the five tests. Only one graded test was done because the reliability of the claudication and hemodynamic responses during and following this protocol has been previously determined.’6 The graded protocol was performed first to measure the V02 obtained from each stage so that comparable values could be obtained during the level walking and stairclimbing protocols. The increment in walking speed (0.3 mph) and step rate (4 steps/minute) during level walking and stairclimbing, respectively, were selected because unpublished data from our laboratory demonstrated that the increases inV02were similar to those obtained by the 2 % grade increment. Because claudication pain is less reliable when claudicants hold onto the handrail as they walk,&dquo; support was allowed only for brief moments to maintain balance. During stairclimbing, patients were permitted to lightly touch the handrail to maintain balance.
Walking
and
Measurements Obtained Supine Rest. Patients rested supine for twenty minutes, after which hemodynamic measurements were recorded as previously described. 16-18 Foot TcP02 was measured on the dorsum of the foot over the third cuneiform bone with a polarographic electrode and a transcutaneous oxygen monitor. Ankle SBP was measured in the posterior tibial and dorsalis pedis arteries of both lower extremities by a bidirectional Doppler and bidirectional 8 MHZ probe. The artery yielding the higher value in the more severely diseased lower extremity was used for the ankle SBP’° Systolic and diastolic blood pressure (DBP) were also obtained in the brachial artery of the right upper limb by the auscultatory method. ABI was calculated, and heart rate was obtained from
an
electrocardiograph.
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
664 Exercise. V02 was measured on-line every fifteen seconds as patients breathed through a mouthpiece and a Daniels valve into a mixing chamber. A ventilation measurement module was used to measure inspiratory ventilation, and gas analyzers were used to measure oxygen and carbon dioxide content in the expired air. Prior to each test, ventilation was calibrated with a 3-liter syringe. Gas analyzers were calibrated with known gas concentrations and room air. Patients indicated when the onset of claudication pain occurred (CPT) and then proceeded to walk to maximal pain (MPT). Foot TCPO, and heart rate were continuously monitored during exercise, and values were recorded at CPT and MPT. Brachial pressures were taken during the final minute of each exercise stage. Supine Recovery. Foot TcP02 and heart rate were continuously monitored for fifteen minutes following exercise, and values were recorded at minutes 1, 2, 4, 6, 8, 10, 12, and 15. Ankle and brachial pressures were also measured, and ABI was calculated at each of these eight occasions.
Statistical Analyses The reliability coefficient (R)’9 and the coefficient of variation (CV)2° were determined for each variable before, during, and after exercise. Dependent t tests were used to compare peak V02 and claudication values between the two level walking tests and between the two stairclimbing tests. Two-factor, repeated measures ANOVAs (two tests x occasion of measurement within each test) were used to assess the hemodynamic variables measured before, during, and after level walking and stairclimbing. If similar results were found between the two tests of level walking and of stairclimbing, values were averaged and compared with those of graded walking by one-factor, repeated measures ANOVAs (protocol) for peak VOZ and claudication values, and by two-factor, repeated measures ANOVAs (protocol x occasion of measurement) for each hemodynamic variable. Tukey post-hoc comparisons were used to locate mean differences whenever a protocol effect, an occasion effect, or an interaction was found . 21 Statistical significance was set at P < 0.05 . Measurements are presented as means ± standard errors.
Results
Oxygen Consumption Reliability. PeakV02was reliable during level walking and stairclimbing, for values between the two level walking tests ( 13 .58 ± 0.63 and 13.61 ± 0.54 mL/kg/min) and between the two stairclimbing tests (14.11 ±0.81 and 13.33 ±0.57 mL/kg/min) were not different (P=NS). Furthermore, the R values were high for these protocols (0.87 and 0.89, respectively), and the CV values were low (7.3% and 7.8%, respectively). Comparison Among Protocols. As noted in Figure 1, the rate of increase in V02 during the graded protocol was lower than that predicted by a linear model published by the American College of Sports Medicine (ACSM) . Zz Given this result, the increments of V02 during the subsequent level walking and stairclimbing protocols were chosen to match the actual measured value rather than the predicted value. Measured peak V02 of level walking and stairclimbing was similar to that during graded walking ( 14 .18 ± 0. 77 mL/kg/min; P=NS). However, since the predicted peak V02 of graded walking was higher than that actually obtained, some energy may have been derived anaerobically. Consequently, the peak V02 during
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
665
walking and stairclimbing was matched to the predicted peak V02 of graded walking in a subsequent analysis by a repeated measures analysis of covariance (ANCOVA). 21
level
FIG. 1. Measured oxygen consumption (V02) of each the graded walking progressive protocol. Also shown is the predicted V02 (PV02) based on the equation of the American College of Sports Medicine.&dquo; Measured V02 during the level walking and stairclimbing protocols was similar to values depicted during the graded
patient (Px) during
protocol.
Claudication Pain Reliability. CPT and MPT were similar between the two level walking tests (P NS) and between the two stairclimbing tests (P = NS) (Table II) . The R values between the two tests of level walking and of stairclimbing were similar to those previously obtained with the graded protocol (0.89 and 0.93 for CPT and MPT, respectively).’6 The CV values for CPT and MPT during level walking and stairclimbing were also similar to those previously obtained during graded walking (15.8% and 12.6% for onset and maximal pain time, respectively).’6 Comparison Among Protocols. The average times to CPT and MPT between the two level walking tests and the two stairclimbing tests were similar to values measured during graded walking (P=NS) (Table II) . When values were compared at the same predicted peak VO2, adjusted means for CPT (233 .1, 253 .4, 248.6) and to MPT (441.1, 436.2, 433.7) were also similar among the graded walking, level walking, and stairclimbing protocols, respectively =
(P = NS) . TABLE II Mo.~M/ Pain Claudication Pain C/aM~c~oAt P~n n/Mc Time (CPT) P~n Time 7wM’ (MPT) ~CPT) and Maximal (mop7) During Graded Walking (G), Level Walking (L), and Stairclimbing Progressive Protocols, Values Are Means 1: Standard Errors, Reliability Coefficients (R), and Coefficients of Variation (CV)
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
666
Hemodynamic
Variables
The R and CV values of the hemodynamic measurements before, during, and after level and stairclimbing are shown in Tables III and IV, respectively. The reliability values were similar to those previously obtained with graded walking.’6
Reliability.
TABLE IV
TABLE III
sive Protocols
Coefficients of Variation (%) of Each Hemodynamic Variable Measured Before, During, and After Level Walking (L) and Stairclimbing (S) Progressive Protocols
*TcP02=transcutaneous oxygen tension, SBP=systolic blood pressure, ABI=ankle/brachial SBP index, DBP=diastolic blood pressure. tCPT=claudication pain time, MPT=maximal pain time.
*TcPO,=transcutaneous oxygen tension, SBP=systolic blood pressure, ABI=ankle/brachial SBP index, DBP=diastolic blood pressure. tCPT=claudication pain time, MPT=maximal pain time.
Reliability Coefficients of Each Hemodynamic Variable Measured Before, During, and After Level Walking (L) and Stairclimbing (S) Progres-
Comparison Among Protocols. Foot TcP02 (Figure 2), ankle SBP (Figure 3), ABI (Figure 4), and heart rate (data not shown) values were not different among the three protocols. Brachial pressures during graded walking were 2-10 mm Hg higher at maximal pain than during the other protocols (P < 0.05). However, when the exercise brachial pressures were adjusted for a given predicted peak V02, differences among the protocols no longer remained (P=NS). Within each protocol, the hemodynamic variables measured from the claudicating limb demonstrated a significant change (P < 0.05) during and following exercise (occasion of meadecreased while surement main effect). Foot TCP02 increased as patients stood (P < 0.05), they walked to CPT (P < 0.05) with no further reduction to MPT (P=NS), and was lower than resting baseline during the first twelve minutes of recovery (P < 0.05). Ankle SBP and ABI were lower than resting baseline during the first ten minutes of recovery (P < 0.05).
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
667
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
668 Discussion
Oxygen Consumption An objective of the present investigation was to match the V02 during each exercise stage of level walking and stairclimbing with that obtained during graded walking. In this manner, claudication and hemodynamic responses to the three exercise protocols could be compared at similar metabolic requirements. Although similar values were achieved, note that values during the first exercise stage of graded walking were higher than those predicted.22 This indicates that the claudicants were metabolically uneconomical while they walked. Furthermore, the rate of increase in V02 between stages was approximately 1.1 mL/kg/min, which is only about half the predicted value. This suggests that the claudicants were either unable to meet the increased energy requirement through aerobic metabolism or that they were slow in doing so, owing to inadequate perfusion of the working lower limb muscles. Because the graded protocol had the highest predicted energy requirement, and because the claudication and hemodynamic measurements taken from the claudicating limb were similar among the three protocols, it could be interpreted that relatively less hemodynamic stress was placed on the limb during this protocol. Thus, the data were also analyzed at a given predicted peak VOZ via ANCOVA statistical procedures. This analysis reduced the claudication values during graded walking and increased the values during level walking and stairclimbing. However, none of the results from the original analyses changed, except that brachial pressures were no longer lowest during stairclimbing. Claudication Pain The reliability of claudication pain during level walking and stairclimbing were similar to those previously reported for graded walking.’6 Furthermore, the claudication measures were similar when patients performed the three protocols, suggesting that similar information was obtained about the circulatory impairment of the claudicating limb. It is difficult to compare the results of the present investigation with those of previous studies because of methodological differences. For example, Bonde-Petersen23 and Hillestad24 used different speeds and grades than the present study used. Bonde-Petersen23 found that maximal walking distance was reduced more by increasing the grade from 0 ° to 5 ° at constant speeds of 4.0 and 4.9 km/hr than by increasing speed from 2.4 to 4.9 km/hr at a constant incline of 5 ° . This result was expected, however, since the increment inV02was greater with the alteration in grade (approximately 17.5 mL/kg/min) than with the change in speed (approximately 10.5 mL/kg/min). No measure of claudication pain distance was reported under either condition. Hillestad 24 found that an increase in speed from 2 to 4 km/hr at 0 % grade decreased maximal walking distance. Grade was not increased, and thus, its effect on distance could not be compared. No data on claudication responses during stairclimbing have been previously
reported. Hemodynamic Variables During Exercise. The response of each hemodynamic variable measured during exercise was similar among the three protocols. The reduction in foot TcP02 during exercise in claudicants is due to the vasodilation in the active musculature, thus compromising the delivery of oxygen to peripheral tissue.II.18 The similar reduction in foot TCPO, during each protocol to
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
669
pain suggests an equivalent amount of vasodilation. Although not statistithere cally significant, appears to be a tendency for foot TCPO, to be higher at CPT and MPT during stairclimbing than during either walking protocol. The hemodynamic significance of this trend is unclear but may be due to the lower rate of contraction as compared with walking and to the longer periods of standing while exercising.
maximal tolerable
The response of each hemodynamic variable measured during recovery was similar among the three exercise protocols. Thus, each demonstrated inadequate circulatory responses of the claudicating limbs to exercise. The peripheral circulatory variables (foot TcP02, ankle SBP, and ABI) were lower during the first ten to twelve minutes than those obtained at supine rest. In addition to a decrease in perfusion pressure, due to a reduction in cardiac output from exercise to recovery, the impairment of peripheral blood flow during recovery is also a result of the vasodilatation still present in previously active, ischemic lower limb muscles. Because arterial stenoses and occlusions diminish the peripheral circulation to more distal regions of the lower limb, blood flow to both muscular and cutaneous tissue cannot be met under a hyperemic condition in muscle.&dquo;&dquo;’ In less diseased or healthy individuals, however, the increased conductance in muscle following exercise does not occur at the expense of the peripheral circulation.5,18 As the vasodilatory stimulus gradually subsides, the peripheral circulation is restored toward baseline, manifested by an increase in foot TcP02, ankle SBP, and ABI.
During Supine Recovery.
Utility of the Three Progressive Protocols In considering the clinical utility of each protocol, level walking is probably the least feasible because some patients may complete the exercise stage associated with their maximal walking speed (3-4 mph) before MPT is attained. This problem would probably occur only in patients who are mildly diseased, and in such cases, maintaining their maximal speed would eventually lead to maximal pain. A second disadvantage to this protocol is that the safety of some patients may be compromised, especially while they are experiencing claudication pain, when limping is commonly observed. This would be particularly true in patients who are unable to walk fast (eg, severely diseased and frail claudicants). Although the safety of the level walking protocol might be enhanced by allowing patients to hold onto the handrail, this would not be advisable because the reliability of the claudication and hemodynamic responses would be reduced.’6 Thus, the level walking protocol appears to be clinically useful only in claudicants with moderate disease severity. Even in these patients, this protocol does not offer any distinct advantage to the more traditional graded protocol. The stairclimbing protocol does not have the limitations that are associated with the level walking protocol. Relatively large increases in energy cost can be achieved with each progressive exercise stage. Consequently, MPT would likely be attained before the highest possible power output, even in mildly diseased claudicants. Furthermore, there is less risk of falling because the patient can choose when to discontinue exercise. However, the stairclimbing protocol does have limitations of its own. First, the task of stairclimbing is more novel than walking on a treadmill for most patients, and second, handrail support (lightly touching the front handrail) appears to be necessary to maintain balance. As a result, more time is required to practice the exercise task, and periodic reminders to not grasp the handrails during exercise
Clinical
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
670 needed. Great care was taken to make with their fingers extended.
are
sure
that
patients only lightly
touched the handrail
Conclusions The main findings of this investigation were that: (1) the reliability values of the claudication and hemodynamic variables during and following level walking and stairclimbing were similar to those previously obtained with graded walking, and (2) the claudication and hemodynamic responses to the three protocols were similar. Thus, because graded walking, level walking, and stairclimbing progressive exercise protocols yield reliable and similar information about the hemodynamic severity of peripheral vascular occlusive disease, only one is needed for evaluation.
Acknowledgments Appreciation is extended to Craig Phelps, D.O.; Joel Sellers, D.O.; Scott Steingard, D.O.; and Larry Ashker, D.O., of the Center for Sports Medicine and Orthopedics in Phoenix for their medical supervision of the treadmill tests. We are also grateful to Andy Chittick of the Arizona Heart Institute for his assistance in patient recruitment and the loan of a bidirectional Doppler, bidirectional probe, and supplies. The authors are grateful to Novametrix Medical Systems, Inc. (Wallingford, CT) for the loan of a transcutaneous oxygen monitor and supplies. This study was financially supported by grants from The Rockport Walking Institute (Marlboro, MA) and Randal Sports/Medical Products, Inc. (Kirkland, WA). The authors are grateful to Novametrix Medical Systems, Inc. (Wallingford, CT) for the loan of a transcutaneous oxygen monitor and supplies. Andrew W. Gardner, Ph. D. University of Vermont Department of Medicine Metabolic Unit, Given C-332 Burlington, VT 05405
References 1.
2.
3.
4.
5.
Clyne CAC, Tripolitis A, Jamieson CW, et al: The reproducibility of the treadmill walking test for claudication. Surg Gynecol Obstet 149:727-728, 1979. Harris KA, Meads GE: A single treadmill exercise test does not accurately quantitate claudication. Can J Surg 30:446-448, 1987. Hiatt WR, Nawaz D, Regensteiner JG, et al: The evaluation of exercise performance in patients with peripheral vascular disease. J Cardiopulmonary Rehabil 12:525-532, 1988. Siggaard-Andersen J, Bonde-Petersen F: Intermittent claudication: A comparison between subjective and measured claudication walking distance. Angiology 19:426-434, 1968. Byrne P, Provan JL, Ameli FM, et al: The use of transcutaneous oxygen tension measurements in the diagnosis of peripheral vascular insufficiency. Ann Surg 200:159-165, 1984.
6. Christensen KS, Larsen JF, Klaerke M: Transcutaneous oxygen tension response to exercise in health and in occlusive arterial disease. Acta Chir Scand
152:657-660, 1986. 7. Holdich TAH, Reddy PJ, Walker RT, et al: Transcutaneous oxygen tension during exercise in patients with claudication. Br Med J 292:1625-1628, 1986. 8. Matsen FA III, Wyss CR, Simmons CW, et al: The effect of exercise upon cutaneous oxygen delivery in the extremities of patients with claudication and in a human laboratory model of claudication. Surg Gynecol Obstet 158:522-528, 1984. 9. Wyss CR, Matsen FA III, Simmons CW, et al: Transcutaneous oxygen tension measurements on limbs of diabetic and nondiabetic patients with peripheral vascular disease. Surgery 95:339-345, 1984. 10. Carter SA: Response of ankle systolic pressure to leg exercise in mild or questionable arterial disease. N
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
671 Engl J Med 287:578-582, 1972. 11. Strandness DE Jr, Bell JW: An evaluation of the hemodynamic response of the claudicating extremity to exercise. Surg Gynecol Obstet 119:1237-1242, 1964. 12. Sumner DS, Strandness DE Jr: The relationship between calf blood flow and ankle blood pressure in patients with intermittent claudication. Surgery 65: 763-771, 1969. 13. Baker JD, Dix D: Variability of Doppler ankle pressures with arterial occlusive disease: An evaluation of ankle index and brachial-ankle pressure gradient. Surgery 89:134-137, 1981. 14. Yao ST: Haemodynamic studies in peripheral arterial disease. Br J Surg 57:761-766, 1970. 15. Yao ST, Needham TN, Gourmoos C, et al: A comparative study of strain-gauge plethysmography and Doppler ultrasound in the assessment of occlusive arterial disease of the lower extremities. Surgery 71:4-9, 1972. 16. Gardner AW, Skinner JS, Cantwell BW, et al: Progressive vs single-stage tests for evaluation of claudication. Med Sci Sports Exerc 23:402-408, 1991. 17. Gardner AW, Skinner JS, Smith LK: Effects of handrail support on claudication and hemodynamic responses to single-stage and progressive treadmill protocols in peripheral vascular occlusive disease. Am
J Cardiol 68:99-105, 1991. 18. Gardner AW, Skinner JS, Cantwell
BW, et al: Relationbetween foot transcutaneous oxygen tension and ankle systolic blood pressure at rest and following exercise. Angiology 42:481-490, 1991. Safrit MJ: Evaluation in Physical Education, ed 2. Englewood Cliffs, NJ: Prentice-Hall, 1981, pp 96104. SPSS Inc: SPSS-X User’s Guide, ed 2. Chicago: SPSS Inc, 1986, p 110. Glass GV, Hopkins KD: Statistical Methods in Education and Psychology, ed 2. Englewood Cliffs, NJ: Prentice-Hall, 1984, pp 371-372 and 492-507. American College of Sports Medicine: Guidelines for Exercise Testing and Prescription, ed 2. Philadelphia: Lea & Febiger, 1980, pp 137-142. Bonde-Petersen F: The effects of varying walking speeds when measuring the claudication distance on horizontal and sloping levels. Acta Chir Scand 133:627-630, 1967. Hillestad LF: The peripheral blood flow in intermittent claudication: IV. The significance of the claudication distance. Acta Med Scand 173:467-478, 1963. Provan JL, Moreau P, MacNab I: Pitfalls in the diagnosis of leg pain. Can Med Assoc J 121:167-172, 1979.
ship 19.
20. 21.
22.
23.
24.
25.
Downloaded from ang.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on November 4, 2014
