Group Heart Rate Not an Adequate Predictor of Power Output for Individual Cyclists Jason P. McCarthy and Frank B. Wyatt Dept of Health, Human Performance, and Recreation, Baylor University, Waco TX Introduction: Heart rate (HR) increases as a result of producing more work (i.e. power output, PO) while cycling. The purpose of this study was to determine if HR (bpm) is an adequate predictor of PO (W) for trained cyclists. Methods: Male USCF Category IV-I cyclists (n=16) completed cycle ergometry tests to volitional fatigue using their own bicycles on a PowertrainÆÊ cycle ergometer. HR was measured each minute during the three-minute stages of increasing power maintained by the cyclist (50W increases to 350W, then 30W increases to volitional fatigue). Statistics: Descriptive statistics of mean and standard deviation were measured to establish sample demographics. Pearson Product Moment R Correlation Coefficient was used to determine association between HR and PO. Linear regression analysis was performed to determine prediction level and to establish a regression equation for each individual cyclist and for the group based on HR per minute averages across all 16 subjects. Standard error of the estimate was calculated to establish variance between the group prediction equation and the independent variable heart rate for each subject. Results: All 16 cyclists (age 25.8¡Ó6y, body fat 9.4¡Ó 3.6%, height 178¡Ó 5cm, weight 74.5¡Ó 5.7kg, Peak PO 396¡Ó 35W) showed a positive correlation predicting PO from HR. The range for prediction of PO from HR for each individual was an r2 = 0.96 to 0.99 (p<0.0001.) Taking the average HR per minute across all cyclists produced a correlation to PO at r2 = 0.986 (p<0.0001.) Although the average prediction equation for PO based on HR for all 16 cyclists is PO = 3.503*HR ¡V 291.94, the prediction error for each individual cyclist is not adequate for estimating performance (mean prediction error of -3W and a standard deviation of 39W.) Conclusion: It was determined that although each individual cyclist¡¦s HR can accurately predict PO; one prediction equation for the group did not show significant prediction results. In conclusion, statistical analysis of the data shows cyclists can use individual equations for predicting PO from HR for training regimens, but one equation for all cyclists does not accurately predict PO from HR. References: 1. Arts F. et al. Int J Sports Med. 15:228-231,1994. 2. Jeudendrup A. et al. J Sports Sci 16:S91-S99,1988. 3. Liedl M. et al. Med Sci Sports Exerc. 31:1472-1477. 4. Palmer G. et al. IOC World Congress Sport Sci (5th:1999; Sydney Au). 5. Wilmore J.H. et al. Physician Sports Med. 14(3):133-137;140-143. Comparison of the Changes in Muscular Strength and Body Composition resulting from Resistance Training and Consumption of different Protein Supplements Darren G. Candow and Darren G. Burke College of Kinesiology, University of Saskatchewan, Saskatoon, SK, Canada, and Department of Human Kinetics, St. Francis Xavier University, Antigonish, NS, Canada Introduction: The protein needs of athletes engaged in resistance training exercise are often greater than the recommended daily amounts. Previously, weight training combined with a powder supplement containing whey protein and creatine has been found to yield greater changes in lean tissue mass and 1RM strength than the isocaloric consumption of whey protein alone or placebo. The purpose of this study was to compare the effect of three different protein-based supplements on muscular strength and body composition when combined with 6-weeks of resistance training exercise. Methods: Thirty-nine trained males (18-33y) were randomly assigned (double-blind) into three supplement groups. The different supplements were provided in bar form and included the following: MP (milk protein bar: 255 kcal, 34-g [whey + casein protein], 3X/d); SY (soy protein bar: 225 kcal, 14-g soy protein, 3X/d); and PC (protein + creatine bar: 290 kcal, 35-g [whey + casein protein], 3-g creatine, 3X/d). All subjects followed the same whole-body resistance training program for 6-weeks and were tested prior to the study, at the end of two weeks, at the end of four weeks, and at the end of the study for body weight, muscular strength (1RM bench press and squat), muscular endurance, and body composition (lean tissue mass and % fat) using dual energy x-ray absorptiometry (DXA). Statistics: Data were analyzed by two-way (supplement x time) repeated measures ANOVA, and whenever significance was evident, Tukey post hoc tests were performed for pairwise comparisons. Results: There were no significant differences among groups at baseline for any measure (p<0.05). All groups demonstrated significant increases in 1RM bench press and squat, but only MP and PC experienced a significant increase in lean tissue mass (p<0.05). Post hoc analysis indicated that the PC group had greater increases in body weight (+3.2 %), 1RM bench press (+19.5 %), muscular endurance (+39.8%), lean tissue mass (+5.2 %), and a greater decrease in percentage fat (-6.7%) as compared to the other two groups. Discussion: These findings indicate that supplement bars containing 34-g milk protein, 35-g milk protein + 3-g creatine, or 14-g soy protein combined with resistance training lead to significant increases in muscular strength, and that supplement bars containing milk protein also lead to significant increases in lean tissue mass. These positive changes in muscular strength and body composition are further augmented when 35-g milk protein is combined with 3-g creatine (PC) 3X/d. References: 1. Lemon P. et al. J Appl Physiol. 73: 767-775, 1992. 2. Tarnopolsky, M. et al. J Appl Physiol. 73: 1986-1995, 1992. 3. Burke, D. et al. Int J Sport Nutr Exerc Metab. 11: 384-399, 2001. Validity and Reliability of the AccusportÆÊ Blood Lactate Analyzer Robert A. Robergs, G. Juranovich, P.A.J. Hope and M.J. Newton Center For Exercise and Applied Human Physiology, Exercise Science Program, University of New Mexico, USA, and Sports Science Program, School of Biomedical and Sports Science, Edith Cowan University, Australia Introduction: Research of the validity and reliability of the AccusportÆÊ Blood Lactate Analyser (ALA) has produced conflicting results. Methods: We studied the validity of the ALA using each of whole blood, plasma and saline prepared with known concentrations of lactate. In addition, the reliability of the ALA for whole blood lactate across 15 samples and between two units, and the variability in results with different sample volumes (10, 15, 20, 25 and 30 £gL) were determined. The range of the 19 lactate samples for each of blood, plasma and saline were 1.4 to 29.4, 1.7 to 22.9, and 0 to 30 mmol/L, respectively. Blood, plasma and saline samples were assayed for lactate using the ALA and the criterion method of enzymatic spectrophotometry (ES). Sample volumes between 20-30 £gL were shown to be adequate. Statistics: Accuracy of the ALA compared to results from enzymatic spectrophotometry of the same samples was assessed by; 1) correlation, 2) standard error of the estimate, 3) mean raw score residual error, 4) two-segment regression correlation, residuals and standard error of estimate, and 5) limits of agreement as proposed by Bland and Altman. Reliability of the ALA was assessed by 1) measurement (typical or technical) error (ME = SD of repeat measures), 2) the relative coefficient of variation (CV = [ME/mean]*100), and 3) reliability limits of agreement (RLA = ¡Ó2.77*ME). Results: Compared to ES, lactate assay by ALA produced two function linear relationships for each of whole blood, plasma and saline to upper limits of 13.1, 10.9, and 17.3 mmol/L, respectively. The ALA underestimated lactate from blood, plasma, and saline by 0.88¡Ó0.55, 1.54¡Ó1.44, and 0.85¡Ó0.66 mmol/L, respectively. Correlation and standard error of estimate statistics for the first linear segments for blood, plasma and saline were 0.97 (p<0.05) and 0.95 mmol/L, 0.973 (p<0.05) and 1.13 mmol/L, and 0.99 (p<0.05) and 0.65 mmol/L, respectively. There was no difference in lactate between two units of the ALA for an 11.5 mmol/L sample (8.69¡Ó0.35 vs. 8.72¡Ó0.42 mmol/L). Reliability of the ALA for blood lactate of 2.01, 7.1, and 14.7 mmol/L resulted in coefficients of variation of 15.8, 5.5, and 4.0 %, respectively. Discussion: The results indicate that the ALA is not valid for blood lactates ranging from 1.4 to 13.1 mmol/L, and values > 13.1 mmol/L exceed the range of the instrument. The ALA is more accurate for saline than blood samples. The reliability is poor for low, moderate and high blood lactate readings, suggesting that the unit should not be used for research. Nevertheless, use of regression corrected values from the ALA produces acceptable results for whole blood lactate values <7 mmol/L. References: 1. Bland JM, Altman DG. Lancet 8: 307-310, 1986. 2. Fell JW, et al. Int J Sports Med 19: 199-204, 1998. 3. Hopkins WG. www.sportsi.org/resource/stats/precision.html 4. Pinnington H, Dawson B. J Sci Med Sports 4(1): 129-138, 2001. 5. Simmons DB. et al. J Equine Vet Sci 19:402-407, 1999.
© Copyright 2001 All rights reserved.