The valid and reliable Wingate Anaerobic Test (WAnT) has received much attention among exercise physiologists and has been used extensively for the assessment of anaerobic muscle performance and the evaluation of responses to supramaximal exercise (3). The WAnT consists of 30-s, all-out supramaximal cycling against a high resistance set relative to the subject's body mass. The test-retest reliability of mean power output (MP) in the WAnT has been evaluated extensively and has been found to be very reliable (test-retest reliability of MP 0.90-0.97 in children, young adults, the elderly, healthy and disabled subjects, athletes, and nonathletes) (1-3,8,10,19).
Blood lactate concentration ([La]) before and after exercise is often used in the laboratory or in field situations to assess the contribution of anaerobic glycolysis to the energy requirements associated with supramaximal exercise (see 17 for review). Peak [La] ([La]pk) following 60 s of supramaximal exercise has been used as an indicator of the glycolytic system's training status(17,24). [La]pk usually appears 5-8 min into recovery (12). [La]pk following WAnT has been reported to occur following 5-10 min of recovery (5,6,13).
Pearson correlation coefficients between [La]pk following maximal treadmill exercise and track running times of 100, 200, and 400 m were reported to be −0.65, −0.78 and, −0.79, respectively, in college students of unspecified fitness levels (12). However, Ohkuwa et al. (18) reported a very low and statistically insignificant correlation (r = −0.38) between [La]pk and 100-m sprint performance in competitive sprinters. Furthermore, the reliability of [La]pk was similar to the reliability of the preceding performance (17,24).
The test-retest reliability of [La]pk following supramaximal exercise has been evaluated to a limited extent. Fujitsuka et al. (12) reported a value of 11% for the coefficient of variation (CV) of [La]pk following 60 s of exhaustive treadmill running in young nonathletic healthy males. Coggan and Costill (4) reported CV values of 11.0 and 13.8% for all-out, isokinetic cycle ergometry lasting 30 and 60 s, respectively. It should be noted that the WAnT is a constant-resistance variable-velocity test, whereas the isokinetic test is a constant-velocity variable-resistance test. Therefore, differences between the tests may exist in the motor unit recruitment pattern. We are aware of no study which specifically investigated the test-retest reliability of the [La]pk response to the WAnT or other constant-resistance tests.
[La] may be affected by a change in plasma volume (ΔPV). A large decrease in ΔPV results in an increase in [La] values, which do not necessarily reflect the involved glycolytic activity. Following performance of an intensive exercise, there is a reduction in PV (20,24), reflected by a higher blood hematocrit (24). Rotstein et al. (20) reported a ΔPV of −10.4% 4 min following the WAnT, compared with control pre-WAnT values. Hebestreit et al. (16) reported a 16.9% and 13.0% decline in PV 3 and 10 min following the WAnT, respectively. Thus, variations in ΔPV between tests might possibly explain test-retest variability in [La]pk following the WAnT or similar exercise performances. The reliability of HR responses to aerobic exercise is high and well documented (25). The reliability of HR response to anaerobic exercise, on the other hand, is not clear. Therefore, the aim of the present study was to evaluate the test-retest reliability of the [La]pk, heart rate (HR), and ΔPV response to the WAnT.
Subjects. Twenty-nine subjects (15 male and 14 female) participated in the present study. All subjects were healthy nonsmokers, sedentary to very active. Subject characteristics are summarized in Table 1. An informed consent, in accordance with the Helsinki convention, was obtained from all subjects after the test procedures were thoroughly explained.
Protocol and study design. On the first visit to the laboratory, height, weight, and skin-fold thickness at four sites (biceps, triceps, subscapular, and suprailiac) were measured in each subject for the assessment of percent body fat (9). Before the first visit, each subject underwent a physical examination. The subjects were instructed to consume their normal diet on the day before the test. Additionally, subjects were asked not to exercise on the day of the test, to refrain from consuming caffeine, alcohol, and medications at least 12 h before the test and from eating at least 2 h before the test. Twenty-four hour recall of dietary and activity questionnaires were filled out by the subjects to substantiate their compliance with these instructions.
All subjects performed two WAnTs with identical resistance loads, 3-7 d apart, and at a similar time of the day. The WAnT was performed on a computerized cycle ergometer(Fleisch Metabo Ergostat Universel, Lausanne, Switzerland) (see Fig. 1 for protocol). The subjects warmed-up for 5 min at a pedaling rate of 60-70 RPM against a resistance equal to 20% of that calculated for the subsequent WAnT. Two unloaded 5-s sprints were performed at the end of the 3rd and 5th min of the warm-up period. The maximal pedaling rate (RPMmax) attained during the sprints was recorded. Following a 10 min-rest, subjects performed the WAnT against a resistance of 0.052 kg·kg−1 body mass for male subjects (5.13 J pedal rev·kg−1 body mass) and 0.047 kg·kg−1 body mass for female subjects (4.62 J pedal rev·kg−1 body mass). The subjects were instructed to pedal as fast as possible from the onset of the test. The resistance was applied when 75% of the previously recorded RPMmax was attained (Fig. 1). The subjects were verbally encouraged to maintain as high a pedaling rate as possible throughout the 30-s test duration. Pedal revolutions were monitored at a resolution of 0.025 revolution and recorded at 1-s intervals. Mean power (MP) was the average power generated throughout the 30-s test. HR was monitored throughout the warm-up, the test, and the recovery period, using a Sport-Tester heart-rate monitor (Polar Electro, Kempele, Finland). It should be noted that this system calculates a moving average of the heart rate. Therefore, in nonsteadystate conditions, as in the present study, HR values, especially immediately after the exercise, may underestimate peak values. HR was recorded before the warm-up, before the WAnT and at 3, 5, 7, and 9 min of recovery.
Capillary blood samples from all subjects were drawn from the fingertip of a prewarmed hand (in a water container at 38 ± 1°C) for lactate analysis before the warm-up, immediately before the test, and at 3, 5, 7, and 9 min following warm-up. Additionally, duplicate blood samples were drawn into heparnized-capillary tubes (25 μL) for hematocrit (Hct) determination from a subset of 19 subjects, before the warm-up, before the WAnT, and 3 and 13 min following the WAnT.
Blood analysis. Whole blood lactate was analyzed using the YSI Sport 1500 L-lactate analyzer (Yellow Springs Instruments, Yellow Springs, OH) with a cell lysing agent. Hct was visually determined with a Hct meter following centrifuging at 3000 × g for 10 min. ΔPV was calculated as a percentage value using the equation provided by van Beaumont (23). Because pre-WAnT Hct values were not significantly different from those obtained before the warm-up (15 min pre-WAnT,Fig. 1), ΔPV during recovery was calculated relative to Hct values measured immediately before the WAnT (Fig. 1).
Statistical analysis. Results of male and female subjects were pooled. Differences in [La]pk and HRpk between test 1 and test 2 were analyzed with a repeated measures ANOVA. The test-retest reliability for [La]pk, MP, ΔPV, and peak HR (HRpk) was assessed using the intraclass correlation coefficient, as described by Thomas and Nelson (22). In addition, Pearson correlation coefficients were calculated for comparison with other studies. Changes in [La] over time following test 1 and test 2 (time effect), and differences between tests (order effect) were analyzed with repeated measures ANOVA. Data are presented as means ± SE (HR, [La], MP, Hct, ΔPV) and means ± SD (subjects' characteristics). Differences were considered statistically significant at P < 0.05. All analyses were carried out using Systat™ Statistical package (Systat Inc., Evanston, IL).
Mean(± SE) MP, HRpk, and [La]pk are presented in Table 2. No significant differences were observed between test 1 and test 2 in any of the variables. Figures 2 and 3 illustrate the test-retest relationships for MP, HRpk, [La]pk, and ΔPV. High intraclass correlation coefficients values (R > 0.88; P < 0.001) were found for MP, HRpk, [La]pk, and ΔPV. The dynamics of [La] and HR changes from rest until the end of the WAnT and throughout recovery from the WAnT are presented in Fig. 4. [La] and HR increased significantly following the WAnT [time effect: [F(5,46 = 208.4,P < 0.0001] and [F (5,47) = 134.4, P< 0.0001], for [La] and HR, respectively], and remained significantly higher than the pre-WAnT values (P < 0.001) throughout the 9th min of recovery. The repeated measures ANOVA revealed no significant differences between test 1 and test 2, and no interaction between test and order, in either [La] or in HR responses.
Table 3 presents results for the hematocrit (Hct) values measured at rest (0 min) and following test 1 and test 2 (3 and 13 min), as well as the corresponding calculated ΔPV. The changes in Hct values over time were statistically significant [F(3,96) = 164.8, P < 0.0001]. The corresponding calculated ΔPV were also statistically significant (F(3,96 = 82.1, P < 0.0001). The repeated measures ANOVA revealed no significant differences between test 1 and test 2 and no test-order interaction, in either Hct or the ΔPV responses.
The main findings of the present study are that [La]pk, HRpk and ΔPV following the WAnT are reliable measures. The intraclass coefficients for [La]pk, HRpk and ΔPV (r = 0.926, 0.941 and 0.878, respectively) appear lower than the value calculated for MP (0.982). That is, the coefficients for the physiological responses to the WAnT ([La]pk, HRpk, and ΔPV) appear lower than the coefficient for the performance(power output, MP), which elicited the response. It should be noted that the intraclass correlation is the recommended procedure for assessing test-retest reliability (22). However, most studies have employed the Pearson correlation procedure. Therefore, the results of the present study were also analyzed with the Pearson correlation procedure. The Pearson correlation coefficient for [La]pk(r = 0.86, between tests 1 and test 2) obtained in the present study is similar to the value reported by Crielaard et al. (7) following a 30 s all-out, isokinetic cycling test (r = 0.87).
The values obtained for [La]pk are within the range of values reported in the literature following WAnT (11,13,14,24) but lower than [La]pk values measured, for example, following 100- to 400-m sprints (12). The coefficient of variation values (CV) assessed in the present study for [La]pk(17.7 and 17.8%, test 1 and test 2, respectively) are within the range of values reported in the literature (4,12). Specifically, our value appears higher than the coefficient reported by Fujitsuka et al. (12) following supramaximal treadmill running (7.2%) but lower than the values reported by Graham and Andrew (15) following a progressive treadmill test to V˙O2max (21%). The apparent differences between the values may reflect the glycolytic vs aerobic contribution to the exercise performed, the type of task (progressive vs fixed-resistance protocol), mode of exercise (running vs pedaling), and the choice of subjects (highly trained, moderately active, or sedentary). The high test-retest reliability observed in the present study may be related, in part, to the fact that the subjects were of diverse activity and fitness levels, and to the pooling of data from several groups, resulting in a large range of [La]pk. When the subjects in our study were divided by gender, test-retest coefficients of r = 0.950 and 0.827, for male and female subjects, respectively, were obtained. Thus, the test-retest reliability of the physiological responses to the WAnT needs to be investigated further in homogeneous groups of subjects (e.g., similar fitness level).
Among the WAnT indices, it is widely accepted that the total work performed, as reflected by MP, depicts the glycolytic contribution to the WAnT performance (3). Therefore, we tested the correlation between MP and [La]pk. The interclass correlation coefficient value (for 15 subjects who had complete data, including ΔPV) was 0.478 (r2 = 0.228; SEE = 1.635 mM). This correlation value is higher than those reported by Goslin and Graham (14)(r = 0.14) and by Dotan et al. (unpublished) (r = 0.20) but lower than the values found by Tamayo et al. (21) (r = 0.60). Other factors that may contribute to the unexplained [La]pk variance may include the variability in the subjects' fitness level and muscle fiber composition.
The relatively low correlation coefficient obtained between MP and [La]pk in the present study and the relatively large [La]pk variance emphasize that many factors may determine [La]pk including aerobic and phosphagenic contribution to the WAnT performance, mechanical efficiency, as well as the time and energy required to attain the prescribed RPM. More research is needed to explain the variance of [La]pk following the WAnT.
The ΔPV following the WAnT observed in the present study at 3 min of recovery (−14.1%, for male subjects) is higher than the values reported among men by Rotstein et al. (20) at 4 min of recovery from the WAnT (−10.4%) but is somewhat lower than the values observed by hebestreit et al. (16) at 3 min following the WAnT (−16.9%). Extending the data reported by Hebestreit et al. (16), our results demonstrate that by 13 min following the WAnT, PV had still not returned to preexercise levels. The reliability coefficient of ΔPV of all the subjects(r = 0.878, P < 0.025; CV = 18.3%) was slightly lower than that of[La]pk. Nevertheless, ΔPV did not significantly explain any portion of the variability in [La]pk.
The reliability of a variable (e.g. [La]pk) upon a repeated measurement has implications for the ability to compare results on the variable across subjects, over time, or consequent to a particular treatment. In conclusion, the results of the current study indicate that [La]pk, HRpk, and ΔPV following the WAnT performance are reliable measures and can be used for comparisons between subjects and treatments and as characteristic of the response to the WAnT.
1. Bar-Or, O., R. Dotan, and O. Inbar. A 30 second all-out ergometric test: its reliability and validity for anaerobic capacities. Israel J. Med. Sci.
2. Bar-Or, O. Pediatric Sports Medicine for the Practitioner.
New York: Springer-Verlag, 1983, pp. 323-325.
3. Bar-Or, O. The Wingate Anaerobic Test: an update on methodology, reliability and validity. Sports Med.
4.Coggan, A. R., and D. L. Costill. Biological and technological variability of three anaerobic ergometer tests. Int. J. Sports Med.
5. Collomp, K. R., S. B. Ahmadi, M. A. Audran, et al. Effects of caffeine ingestion on performance and anaerobic metabolism during the Wingate test.Int. J. Sports Med.
6. Collomp, K. R., S. B. Ahmadi, C. F. Caillaud, et al. Effects of benzodiazepine during a Wingate test: interaction with caffeine. Med. Sci. Sports Exerc.
7. Crielaard, J. M., P. Ledent, M. Grosjean, et al. Evaluation en laboratoire de la capacite anaerobic lactique: mise au point d'un test. Med. Sport
8. Dotan, R., and O. Bar-Or. Load optimization for the Wingate Anaerobic Test. Eur. J. Appl. Physiol.
9. Durnin, J. V. G. A., and J. Womersley. Body fat assessed from total body density and its estimation from skinfold thickness measurements on 481 men and women aged 13-72 years. Br. J. Nutr.
10. Evans, J. A., and H. A. Quinney. Determination of resistance settings for anaerobic power testing. Can. J. Appl. Physiol.
11. Froese, E. A., and M. E. Houston. Performance during the Wingate anaerobic test and muscle morphology in males and females. Int. J. Sports Med.
12. Fujitsuka, N., T. Yamamoto, T. Okhuwa, et al. Peak blood lactate after short periods of maximal treadmill running. Eur. J. Appl. Physiol.
13. Gokbel, H., C. Dolek., C. S. Bediz, et al. The relationship of lactic acid and total testosterone levels after the Wingate test. Turk. J. Med. Sci.
14.Goslin, B. R., and T. E. Graham. A comparison of "anaerobic" components of O2
debt and the Wingate Test. Can. J. Appl. Sport Sci.
15. Graham, T. E., and G. M. Andrew. The variability of repeated measurement of oxygen debt in man following a maximal treadmill exercise. Med. Sci. Sports Exerc.
16. Hebestreit, H., F. Meyer, Htay-Htay, et al. Plasma metabolites, volume and electrolytes following 30 s, volume and electrolytes following 30 s, high intensity exercise in boys and men. Eur. J. Appl. Physiol.
17.Jacobs, I. Blood lactate: implications for training and sports performance.Sports Med.
18. Okhuwa, T., Y. Kato, K. Katsumata, et al. Blood lactate and glycerol after 400 m and 3000 m runs in sprint and long distance runners. Eur. J. Appl. Physiol.
19. Patton, J. F., M. M. Murphy, and F. A. Frederick. Maximal power outputs during the Wingate Anaerobic Test. Int. J. Sports Med.
20. Rotstein, A., O. Bar-Or, and R. Dlin. Hemoglobin, hematocrit and calculated plasma volume changes induced by a short, supramaximal task. Int. J. Sports Med.
21. Tamayo, M., A. Sucec, W. Phillips, et al. The Wingate anaerobic power test, peak blood lactate, and maximal oxygen debt in elite volleyball players: a validation study. Med. Sci. Sports Exerc.
22. Thomas, J. R., and J. K. Nelson. Research Methods in Physical Activity,
2nd Ed. Champaign, IL: Human Kinetics Books, 1990, pp. 350-351.
23. van Beaumont, W. Evaluation of hemoconcentration from hematocrit measurements. J. Appl. Physiol.
24.Vandewalle, H., G. Peres, and H. Monod. Standard anaerobic exercise tests. Sports Med.
25. Whipp, B. J., and S. A. Ward. Cardiopulmonary coupling during exercise. J. Exp. Biol.