Spinal Cord (1999) 7, 9 ã 1999 nternational Medical Society of Paraplegia All rights reserved 9/99 $. http://www.stockton-press.co.uk/sc Reproducibility of the Adapted Leger and Boucher est for wheelchair-dependent athletes M Poulain*,1, A Vinet, PL Bernard and A Varray 1 1 Laboratory `Sports, ntervention, Optimization', School of Sports Sciences and Physical Education, 7 avenue du Pic St Loup, 9 Montpellier, France; Laboratory `Sports, Health, Development', School of Sports Sciences and Physical Education, 7 avenue du Pic St Loup, 9 Montpellier, France; Laboratory of Biomechanics and Exercise Biology, UFR SAPS, 1 route de Grenoble, F-, Nice, France Study design: his study analyzed the reproducibility of a eld test. n a previous study, we showed that this test, the Adapted Leger and Boucher est (ALB), was progressive and maximal. 1 he Leger and Boucher predictive equation for able-bodied subjects was not accurate for WD athletes, however, and a new predictive equation is needed. Objectives: o determine the reproducibility of an adapted incremental eld test for wheelchair-dependent (WD) athletes. Setting: France at Montpellier. Methods: he proposed protocol was conducted on a m track. Eight male paraplegics (mean age:.8+.1 years) performed the test three times in the same conditions, ie same time of day, same wheelchair, same material. Maximal heart rate (HRmax) and maximal speed (Smax) were measured. Results: We found no signi cant di erences (P.) between tests for either variable. he Bland and Altman graphic analyses showed a good reproducibility for both variables. Lastly, the reproducibility coe cients of HRmax and Smax were very low (% and 1%, respectively). Conclusion: he ALB is reproducible concerning measurements of HRmax and Smax. A valid predictive equation of maximal oxygen uptake from the Smax is now needed for WD athletes during this eld test. Keywords: paraplegia; incremental eld test; reproducibility; maximal speed; Leger and Boucher est ntroduction Research in athletics is focused on the assessment of the factors that constitute performance. Based on the results of the Paralympic Games, one can conclude that wheelchair athletics is expanding and that these athletes would bene t from data geared toward optimizing their performance. For example, today's competitive wheelchair athletes need training programs that are based on regular evaluation of their aerobic metabolism in the wheelchair. Laboratory procedures for cardiorespiratory tness assessment are now wellestablished, with di erent types of ergometer adapted for wheelchair propulsion. Wheelchair ergometers, however, do not accurately reproduce wheelchair propulsion conditions such as rolling resistance. Field testing has been shown to provide an accurate estimation of aerobic metabolism in practice conditions and is easy to conduct in able-bodied (AB) subjects. he di erent eld tests are used regularly to evaluate maximal oxygen uptake (VO max) in order to adapt training programs. he -min test of Cooper *Correspondence: M Poulain, Cinique du Sou e `la Solane', 19 rue des Casteillets, Osse ja, France was validated 7,8 in paraplegic subjects, but its nonprogressive character requires learning and the ability to establish a steady pace. Moreover, the subjects were evaluated with the same standardized wheelchair. Recently, we validated the Adapted Leger and Boucher est (ALB) in wheelchair-dependent (WD) athletes 1 as a progressive and maximal test. However, the predictive equation, based only on maximal speed (Smax) for AB subjects, was not valid. he assumption was therefore made that other parameters, in addition to Smax, would be needed for a valid equation for WD athletes. he next step was thus to assess the reproducibility of this test in terms of: (1) maximal heart rate (HRmax), to verify the reproducibility of the maximal character of the test, and () maximal speed (Smax), to predict maximal oxygen uptake (VO max). Methods Subjects Eight male WD athletes, aged.8+.1 years (range to 9 years) participated in this study after giving
ncremental field test for wheelchair-dependent athletes MPoulainet al able 1 Subjects (n=8) 1 7 8 Mean SD General characteristics of WD athletes Age (years 1 8 9.8.1 Height (cm) 171 17 18 17 17 17 17 17 17.7.7 Date of injury 198 198 198 1988 1988 198 198 198 Lesion level (-L)* 8 L 9 8 1 Lesion type (-C-S-F)** /F C/F /S Sport (F--){ F raining volume (h week 1 ) Level{ h 7 min h min Marathon R N R R R *=thoracic, L=lumbar; **=incomplete, C=complete, S=spastic, F= accid; {F=track and eld, =tennis, =table tennis; {nternational classi cation, R=regional, N=national, =international informed written consent. All were paraplegics with lesion levels of L, stabilized from their injuries, and were classed according to the SMG classi cation. 9 Furthermore, all had spent a minimum of years in a wheelchair. he athletes practiced track and eld (marathon), tennis (regional, national and international competitors) or table tennis (national competitor). hey had acquired good control of their wheelchairs, which was essential for the test, and they were used to all-out e ort. he general characteristics of the subjects are shown in able 1. Protocol Each subject performed the ALB on three separate occasions. he tests were scheduled for the same time of day, with the same material, and in veri ed comparable weather conditions, especially concerning wind speed and temperature. For the analysis of short-term reproducibility (SR), we established a minimal interval of days and a maximal interval of 1 days between the rst and second tests. For the analysis of long-term reproducibility, the intervals ranged from a minimal period of 8 days to a maximal of 1 month between the rst and third tests (LR). his maximal period was chosen to avoid the risk of any signi cant modi cation in the subject's physical tness. he ALB was conducted on a m tartan eld marked-o every m with pylons. he wind speed was measured with an anemometer prior to testing and had to be less than m s 71. he test protocol required each subject to propel his wheelchair as far as possible following the rhythm imposed by means of an audiotape that provided feedback to the subjects. he turning speed of the tape recorder was checked prior to the start of each test to ensure that any deviation waslessthan1smin 71. 1 he athletes adjusted their speed as necessary when crossing a pylon at the sound signal. here was only a general warm-up, which was included in the test. he initial speed was km h 71 able ndividual HRmax and variability coe cients for the three tests HRmax (beats.min 1 ) Var coef Subjects First test Second test hird test HRmax 1 7 8 Mean SD 17 19 171 171 19 181.1 1. 17 187 18 17 17 177 19 179.8 1.9 Var coef=variability coe cient 17 18 18 17 18 19 188 181. 8....... for 1 min; thereafter it was increased by 1 km h 71 every minute until exhaustion, which was de ned as the inability to maintain the required speed with a -m distance behind the appropriate pylon at the sound signal and the inability to catch up at the next pylon. he number of executed laps, the number of crossed pylons and the duration of the test were noted. he last speed to be reached and maintained 1 min corresponded to the Smax de ned as the speed necessary to attain VO max.,11 hroughout the test, HR was measured every 1 s with a cardiofrequency meter (Sport ester PE, Polar, Finland). he theoretical maximal HR [1 (.age)] was calculated to determine whether subjects reached their maximal values. Statistical analysis Speci c contrasts were calculated to determine whether di erences existed between the theoretical HRmax and the HRmax obtained at the three tests. Results were considered to be signi cant when a risk was less than.. A one-way repeated-measures analysis of
ncremental field test for wheelchair-dependent athletes M Poulain et al 1 variance (MANOVA), realized with SYSA., was used to examine the di erences between means (SD) of both Smax and HRmax and temperature. he Bland and Altman procedure and the calculation of reproducibility coe cients (di erence/mean) were used to test the reproducibility of the ALB for HRmax and Smax. For the Bland and Altman procedure, we plotted the di erence of two measurements, ie between the rst and the second test (SR) and the rst and the third test (LR), for each subject against his mean value. Results are reported as mean+standard deviation. Figure 1 Maximal speed (Smax in km h 71 ) obtained during the three tests in each subject
ncremental field test for wheelchair-dependent athletes MPoulainet al Results he values of theoretical HRmax ranged from 18. to 19. beats min 71. he values of HRmax (able ) of the ALB ranged from 18 to beats min 71. he mean value was 18.8+9.98, which represents 9.%+. of theoretical HRmax. he theoretical HRmax compared with the HRmax obtained during the tests presented no signi cant di erence (F=.9, P=.1). he mean temperature in degrees Celsius was.18+.9 and there was no signi cant di erence between the three tests (F=., P=.9). No signi cant di erences were observed between the Figure Maximal heart rate (HRmax in beats min 71 ) measured during the three tests in each subject
ncremental field test for wheelchair-dependent athletes M Poulain et al three tests for either Smax or HRmax (F=.8, P=.9 and F=.1, P=.87, respectively). Figures 1 and present the maximal, data for all subjects. he Bland and Altman procedure (Figures and ) de ned the short term (SR) and the long term (LR) reproducibility. his analysis indicated good reproducibility for Smax and HRmax between the three tests in terms of the means of di erence, with no problem of proportionality between the values of the variables and their di erences. he mean di erences of Smax were 7.+.8 km h 71 (SR) and 7.+.7 km h 71 (LR) (able ). For HRmax, the mean di erences were 1.+11. beats min 71 (SR) and 7.+7.7 beats min 71 (LR). he reproducibility coe cients of these two variables are presented in ables and. he coe cients were only % for Smax (extreme values=% to %) and 1% (extreme values=% to %) for HRmax (able ). Discussion Eight WD athletes performed the ALB three times. HRmax and Smax did not signi cantly di er between the three tests. he mean values of HRmax were comparable with values reported in previous studies of maximal wheelchair exercise 1,1 and close to those reported by Cooper et al 1 (187 beats min 71 ) in spinal Figure Comparison of maximal speed (Smax in km h 71 ) for rst and second, and rst and third tests. (a) Comparison of Smax for rst and second tests (SR). (b) Comparison of Smax for rst and third tests (LR) Figure Comparison of maximal heart rate (HRmax in beats min 71 ) for rst and second, and rst and third tests. (a) Comparison of HRmax for rst and second tests (SR). (b) Comparison of HRmax for rst and third tests (LR)
ncremental field test for wheelchair-dependent athletes MPoulainet al able ndividual Smax and variability coe cients for the three tests Smax (km.min 1 ) Var coef Subjects First test Second test hird test Smax 1 7 8 Mean SD 1 1 11. 1..7. Var coef=variability coe cient.7..... cord-injury athletes. All subjects reached their HRmax, which con rmed that the ALB is a maximal test. As often reported, the individual values overall re ected the principle of inverse relationship between HRmax and lesion level inferior to. 1 Nevertheless, some of our subjects presented the contrary. hese unusual values may have been due to the di erent neurological and neuro-anatomical factors that a ect medullary functioning, such as complete or incomplete and spastic or accid lesions. 17 Age, athletic activity, training volume, and the number of years of athletic training can also strongly in uence HRmax and, although high and complete paraplegia presents lower HRmax, this last may increase with athletic activity and training volume. 1,18 Determining the reproducibility of a test is one of the steps of validation. HRmax was chosen to con rm the reproducibility of the maximal character of the ABL. Smax was used as a potential predictive variable of VO max in WD athletes, as was done in able-bodied subjects during performance of the original Leger and Boucher test. he eld test of WD athletes proposed in the present study showed good reproducibility for both variables, and the simplicity of the protocol may explain this. he ALB is an incremental and continuous eld test, and the imposed pace facilitates test performance for the subjects. Although the subjects performed the test three times, which may have induced a learning e ect, the Smax variability between tests was less than 1kmh 71 for all athletes. hese stable low values of variability therefore indicate that there was no bias due to learning e ect in the assessment of maximal variables. Subjects did not need to employ a particular technique to propel their wheelchairs during testing. All subjects had been wheelchair-dependent for more than years, and had acquired good control of them. t therefore seems likely that this population's experience and ease with their personal wheelchairs also had an in uence on the reproducibility coe cients and physiological responses that we obtained. Lastly, the good reproducibility may in part be explained by the use of personal general-purpose wheelchairs rather than a standardized chair. ndeed, each wheelchair was speci cally adjusted to each paraplegic, and thus was a function of very speci c characteristics, such as lesion level, lesion character, length of arm, athletic activity, and so on. 19 1 he di erences in HRmax between tests were greater than those observed in Smax, ie, four subjects presented standard deviations of more than beats min. 71 he greatest variability was observed for SR. his variability was not due to wind speed because this parameter was carefully controlled and, moreover, there was no proportionality between SD and wind speed. Stress, however, may explain this relatively high variability. We noted that some of the subjects began the rst test session with resting HR values that had begun to increase. t seems quite likely that lack of familiarity with the test, and thus uncertainty about personal performance capabilities, may have been experienced as stressful. he resulting higher resting HR would have in uenced the HR/VO relationship. However, it is important to note that this greater HRmax variability between the rst and second tests had no e ect on Smax variability. Stress therefore did not a ect Smax and di erences in HRmax between tests would have no in uence on the prediction of VO max. Conclusion Maximal values of heart rate and speed were reached with the adapted Leger and Boucher test for wheelchair-dependent athletes. Moreover, short- and longterm reproducibility were veri ed. he advantages of this eld test include the simplicity of its protocol for both athlete and trainer, and its modest time and equipment requirements. he test uses only two variables, HRmax and Smax, and can be administered by trainers, athletics clubs and rehabilitation centers to adapt training programs with the goal of improving physical tness and the athletic performances of paraplegics. he next step will be to de ne a predictive equation for VO max for this population. his equation will need to take into account variables such as the energy cost of wheelchair propulsion. References 1 Vinet A et al. Validation of an incremental eld test for the direct assessment of peak oxygen uptake in wheelchair-dependent athletes. Spinal Cord 199; : 88 9. Le ger L, Boucher R. An indirect continuous running multistage eld test: the University of Montral track test. CanJofApplSpt Sci 198; : 77 8. Bar-Or O and Zwiren LD. Maximal oxygen consumption test during arm exercise-reliability and validity. J of Appl Physiol 197; 8:. Glaser RM. Arm exercise training for wheelchair users. Med Sci Sports Exerc 1989; 1: 19 18.
ncremental field test for wheelchair-dependent athletes M Poulain et al Gass GC, Camp EM. Physiological characteristics of trained Australian paraplegic and tetraplegic subjects. Med Sci Sports Exerc 1979; 11:. Lundberg A. Wheelchair driving: evaluation of a new training out t. Scan J Rehabil Med 198; : 7 7. 7 Franklin BA et al. Field test estimation of maximal oxygen consumption in wheelchair users. Arch Phys Med Rehabil 199; 71: 7 78. 8 Rhodes EC, McKenzie CD, Coutts KD, Rogers AR. A eld test for the prediction of aerobic capacity in male paraplegics and quadraplegics. CanJApplSptSci1981; : 18 18. 9 MacCann B. Classi ction of the locomoter disabled for competitive sports: theory and practice. nt J Sports Med 198; : 17 17. 1 Le ger L, Rouillard M. Speed reliability of cassette and tape players. CanJApplSptSci198; 8: 7 8. 11 SjoÈ din B, Svendenhag J. Applied physiology of marathon running. Sports Med 198; : 8 99. Lange-Anderson K, Shepard RJ, Denolin H. Fundamentals of Exercise esting. Geneva, 1971; World Health Organization. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. he Lancet 198; 1: 7 1. 1 Cooper RA et al. Maximal exercise responses of paraplegic wheelchair road racers. Paraplegia 199; : 7 81. 1 Coutts K, Rhodes EC, McKenzie DC. Maximal exercise responses to tetraplegics and paraplegics. J Appl Physiol 198; : 79 8. 1 Davis GM. Exercise capacity of individuals with paraplegia. Med Sci Sports Exerc 199; :. 17 VanLoanMD,McCluerS,LoftinJM,BoileauRA.Comparison of physiological responses to maximal arm exercise among ablebodied, paraplegics and quadraplegics. Paraplegia 1987; : 97. 18 Veeger HEJ et al. Peak oxygen uptake and maximal power output of Olympic wheelchair-dependent athletes. Med Sci Sports Exerc 1991; : 1 9. 19 Bambhani N, Ertksson P, Steadward D. Reliability of peak physiological responses during wheelchair ergometry in persons with spinal cord injury. Arch Phys Med Rehabil 1991; 7: 9. Bernard PL. Ergonomie du deâ placement en fauteuil roulant chez le traumatiseâ meâ dullaire. Revue SAPS 199; 9: 77 9. 1 Van Der Woude LHV, Veeger DJ, Rozendal RH, Sargeant J. Seat height in handrim wheelchair propulsion. J Rehabil Research Dev 1989; : 1.