During all the training sessions, HR was assessed in each subject (Polar Team System; Polar Electro Oy), and data were downloaded on a portable personal computer and analyzed using a dedicated software (Polar ProTrainer 5; Polar Electro Oy) and an electronic spreadsheet (Excel; Microsoft Corporation, Redmond, WA).
As a performance assessment, the times obtained during 5000- and 10,000-m track tests (tartan surface) were considered. These tests were conducted 4-5 d after the end of the training program at least 3 d apart to avoid cumulative fatigue between tests. Performance tests were individual time trials with running times assessed with a hand stopwatch (HS-3V-1B; Casio, Japan). Before each race, subjects performed a standardized warm-up protocol consisting of 15 min of slow running (self-selected pace) and three to four strides at running pace.
The TRIMP method considers the ΔHR (HRexercise − HRrest /HRmaximal − HRrest) as the main exercise variable (2). The duration of any specific training session is multiplied by the average ΔHR achieved during that session. To avoid giving a disproportionate importance to long-duration activity at low ΔHR levels compared with intense but short-duration activity, the ΔHR is weighted by a multiplying factor (y) to reflect the intensity of effort. This y factor is based on the exponential rise of blood lactate levels with the fractional elevation of exercise above HRrest (2). This factor served to equate the TRIMP scores of exercises of long duration and light HR with exercises of short duration and high HR. Thus, overall:
where y is a nonlinear coefficient given by the equation, y = 0.64e 1.92 x, with e = base of the Napierian logarithms, x = ΔHR, and the constants b = 0.64 and c = 1.92 (for males).
However, the TRIMP method as proposed by Banister (2) (TRIMPBan) uses the mean exercise HR during an exercise bout, and the multiplying factor y is computed using two constants in the equation (b, c) that are equal for all subjects. The use of the mean exercise HR and the same multiplying factor y potentially fails to reflect the individual physiological demands of each training session. To address this issue, we introduced an individual weighting factor (yi) for each subject. This yi reflects the profile of a typical blood lactate response curve to increasing exercise intensity. Individual yi values were calculated for each subject with the best-fitting method using exponential models (Fig. 1). Thus, as exercise intensity increases, as indicated by the HR response, recorded as one average value every 5 s, the weighting factor yi increases exponentially. Thus, during each training session, a TRIMPi can be calculated at any time as the area under the curve represented by the pseudo integral of all ΔHR data points. In this investigation, 320-384 training sessions were analyzed.
The results are expressed as means ± SD and 95% confidence intervals (95% CI). Before using the parametric tests, the assumption of normality was verified using the Shapiro-Wilk W-test. Pearson's product moment correlation coefficients were used to examine correlations between variables. To assess the strength of the correlation coefficients, the effect size (ES) was calculated according to Cohen (4). ES of 0.8 or greater, around 0.5, and 0.2 or less were considered as large, moderate, and small, respectively. Student's paired t-test was used to determine any significant differences in physiological variables before and after training. Significance was set at P < 0.05. The SPSS statistical software package (Version 13.0.1 for Windows; SPSS, Inc., Chicago, IL) was used for all statistical analyses.
Adherence to the specific training contents within each training session and compliance with the program was 100%, as inferred from the HR monitoring device that recorded, in addition to HR, the date of each training session.
Peak velocity during the treadmill test was 18.0 ± 1.2 km·h−1 at baseline and 18.9 ± 1.2 km·h−1 at the end of the training program (P < 0.01). Corresponding peak lactate values were 9.4 ± 1.7 and 9.0 ± 1.3 mmol·L−1 (not significant) before and after training, respectively.
Speed at 2 mmol·L−1 (+21.3 ± 5.2%, P < 0.001) and 4 mmol·L−1 (+10.6 ± 2.4%, P < 0.01) significantly increased after training (Table 3). Speed improvements (%) at 2 mmol·L−1 (r = 0.87, P = 0.005; 95% CI = 0.97-0.41; ES = 0.98) and 4 mmol·L−1 (r = 0.74, P = 0.04; 95% CI = 0.95-0.07; ES = 0.92) were significantly related to weekly TRIMPi sum (Fig. 2). In addition, there was a significant inverse relationship between TRIMPi and both 5000-m (r = −0.77, P = 0.02; 95% CI = −0.95 to −0.15; ES = 0.97) and 10,000-m track performance (r = −0.82, P = 0.01; 95% CI = −0.96 to −0.27; ES = 0.99; Fig. 3).
No significant relationships were observed between TRIMPBan and percentage of speed improvements at 2 mmol·L−1 (r = 0.61, P = 0.11; 95% CI = −0.91 to −0.17; ES = 0.98) and 4 mmol·L−1 (r = 0.59, P = 0.12; 95% CI = −0.91 to −0.19; ES = 0.96). Again, the TRIMPBan showed no relationship with 5000-m (r = −0.41, P = 0.31; 95% CI = −0.86 to −0.31; ES = 0.96) and 10,000-m track performances (r = −0.54, P = 0.16; 95% CI = −0.90 to −0.26; ES = 0.99). The weekly TRIMPi (671 ± 94 AU) was significantly (P < 0.01) higher than TRIMPBan (565 ± 59 AU; r = 0.78, P = 0.02; Fig. 4).
The main and novel finding of this study is the demonstration that TRIMPi is a valid tool in improving fitness (speed at 2 and 4 mmol·L−1) and performance (i.e., 5000- and 10,000-m races) in recreational LDR and is more valuable than the original method on the basis of average-based group values.
The TRIMP method is an integrated measure of TL, which permits to account for intensity and volume effects on the biological and physiological systems of the athletes (2). Several HR-based methods have been proposed in an attempt to quantify the individual TL. Recently, Lucia et al. (20) proposed to evaluate the individual responses to TL using the time spent in three different HR zones determined according to the ventilatory threshold and respiratory compensation points. To account for differences in time spent at a given load due to exercise intensity, the individual TL response was considered as the sum of the time spent in three HR zones multiplied by intensity-related coefficients (i.e., 1 to 3 from lower to higher intensities) (20). This method represents a modified version of the original model proposed by Edwards (6) in which arbitrary, not physiologically related, HR zones were used to describe internal TL. Recently, Esteve-Lanao et al. (7) used the method of Lucia et al. (20) to monitor the TL in subelite distance runners during the preparatory phase of the competitive season. However, this HR-based zone technique failed to show any relationship with running performance over distances similar to those considered in the present study (i.e., 4- and 10-km races). This result was confirmed in our study when the TRIMPBan was used for comparison.
The present study is the first to show a significant and strong relationship between a measure of training dose and race performance in recreational LDR. Indeed, the faster the time in the 5000- and 10,000-m performance tests, the greater the mean weekly TRIMPi value was.
These findings suggest that when dealing with subelite or recreational endurance runners, attention should be paid in choosing the right method to monitor internal TL (16). Specifically, only methods that show an association (i.e., convergent construct validity) (33) between internal TL and performance and/or fitness variables should be considered.
In light of our results, the internal TL should be detected using individual physiological characteristics (i.e., HR and blood lactate profiles) rather than average exercise values (2). Indeed, the TRIMPi showed to reflect the actual adaptation to TL, being strongly related to both physiological and performance variables (32). Thus, the TRIMPi seems to provide an accurate feedback of the training adaptation, a valid tracking of the progression of TL during the preparation, and a good predictor of performance.
This is in line with the study of Stagno et al. (31) who proposed TRIMP individualization in elite hockey players. However, the TRIMPi used in the present study differs from that used by Stagno et al. (31). Indeed, for TRIMPi calculation, we introduced the weighting factor (y) determined for each single runner and not derived from a group average.
Our approach is different from that of previous studies that used training session average values of HR and allowed more HR values to be considered in calculating TRIMP. In addition, the calculation or the weighting factor in our study was based on individual physiological response and not on standard population- and gender-dependent coefficients as reported by others (2,31). The above-mentioned differences in TRIMP calculation may well explain the lack of sensitivity (i.e., relationship with fitness variations) and validity (i.e., association with performance) observed in the present study when we used the early method (2). Anyway, the ES of the mentioned relationships was shown to be large in our study.
The magnitude of TL experienced by LDR had a higher effect over lower exercise intensities (i.e., speed at 2 mmol·L−1 and 10,000-m time). This suggests that in the medium to low weekly mileage runners, higher TL mainly affect long-term endurance capability (14,21). This contrasts with the findings reported by Esteve-Lanao et al (7) in subelite distance runners, in whom a volume effect of low training intensity (i.e., lower than ventilatory threshold) over endurance performance (i.e., 4- and 10-km races) was observed. This finding may be used as evidence for training periodization in endurance running, suggesting the magnitude of the individual TL response as the main factor for performance and physiological improvements in the early stages of the season (i.e., preparation phase) in recreational endurance runners. In this context, it would be of interest for coaches and fitness trainers to know what could be considered the critical value of TL to control the training process. This estimation may be obtained using the equations that link TRIMPi with changes in aerobic fitness (i.e., y = 0 in the TRIMPi vs performance and speed at 2 mmol·L−1 relationships). This procedure showed that to maintain improvements in endurance fitness (i.e., speed at 2 mmol·L−1) after 8 wk of training, LDR should accumulate a mean weekly TRIMPi of approximately 210 AU. However, these are only indicative data obtained through regression analyses, and consequently, training studies (i.e., maintenance-training prescription) should be performed to grant conclusive inferences.
In subelite endurance-trained runners who experienced improvement in long-distance performance variables, Esteve-Lanao et al. (7,8) reported weekly TRIMP in the range of 360-495 AU. From these studies (7,8), apparently weekly TRIMP higher than 360 AU should be achieved to obtain endurance performance improvement. However, the method used in one study (7), in which linear models have been used to account for individual difference in aerobic fitness and TL adaptations, was not related with performance enhancements, whereas we found a strong relationship between TRIMPi and performance. This concept is further emphasized by the results on the comparison between TRIMPi and TRIMPBan.
Although comparisons between studies that attempted to assess TL should be made with caution; nevertheless, the findings of the present study would indicate that attention should be paid when using linear models and average physiological variables (2,7) with the aim to track individual adaptation to training and related performance.
Recreational long-distance running is a worldwide exercise activity that celebrates its popularity with mass participation in city marathons. In runners lay journals, running experts provide information regarding proper preparation and race strategies in the attempt to help runners achieve their individual goals. In light of the present study, the TRIMPi might be considered as an accessible, valid, and low-cost method to monitor individual responses to TL and, consequently, to optimize training progresses while avoiding a possible state of overreaching or overtraining (12,25).
The main limitation of the present investigation is the small sample size, which is a common characteristic of studies carried out in athletes. This limitation is compensated, in part, by the strong consistency of our observations, which led to statistically significant results. Second, pre- versus posttraining performance assessments (i.e., 5000 and 10,000 m) were not performed as a part of this investigation according to a previous study (7). However, we believe that performance after a structured training program would have not resulted in a performance improvement. Finally, our findings ensue from a relative short period of training. Future studies using the TRIMPi over more prolonged training periods and, possibly, in larger populations with frequent performance assessment are clearly warranted.
In conclusion, the results of this study indicate that TRIMPi is a valid method in tracking fitness and performance in LDR and is more valuable than TRIMP methods on the basis of average group values of HR. TRIMPi correlated with race performance. The results suggest that TRIMPi can be used by coaches to quantify TL and to tailor training sessions on an individual basis in LDR.
This study was supported, in part, by Agenzia Spaziale Italiana funds (Disturbi del Controllo Motorio e Cardiorespiratorio) granted to F. Iellamo.
No disclosure of funding received for this work from the National Institutes of Health, Wellcome Trust, Howard Hughes Medical Institute, and others.
The results of the present study do not constitute endorsement by ACSM.
1. Achten J, Jeukendrup A. Heart rate monitoring: applications and limitations. Sports Med
2. Banister EW. Modeling elite athletic performance. In: Green H, McDougal J, Wenger H, editors. Physiological Testing of Elite Athletes
. Champaign (IL): Human Kinetics; 1991. pp. 403-4244.
3. Banister EW, Calvert TW. Planning for future performance: implications for long term training. Can J Appl Sport Sci
4. Cohen J. Statistical Power Analysis for the Behavioral Sciences
. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988. pp. 567.
5. Edwards JE, Lindeman AK, Mikesky AE, Stager JM. Energy balance in highly trained female endurance
runners. Med Sci Sports Exerc
6. Edwards S. High performance training and racing. In: Edwards S, editor. The Heart Rate Monitor Book
. Sacramento (CA): Feet Fleet Press; 1993. pp. 113-23.
7. Esteve-Lanao J, San Juan AF, Earnest CP, Foster C, Lucia A. How do endurance
runners actually train? Relationship with competition performance. Med Sci Sports Exerc
. 2005;37(3): 496-504.
8. Esteve-Lanao J, Foster C, Seiler S, Lucia A. Impact of training intensity distribution on performance in endurance
athletes. J Strength Cond Res
9. Foster C. Monitoring training in athletes with reference to overtraining syndrome. Med Sci Sports Exerc
10. Foster C, Florhaug JA, Franklin J, et al. A new approach to monitoring exercise training. J Strength Cond Res
11. Foster C, Hoyos J, Earnest C, Lucia A. Regulation of energy expenditure during prolonged athletic competition. Med Sci Sports Exerc
12. Foster C, Lehmann M. Overtraining syndrome. In: Guten G, editor. Running Injuries
. Orlando (FL): W.B. Saunders, Co.; 1997. pp. 173-88.
13. Fry RW, Morton AR, Keast D. Periodisation of training stress-a review. Can J Sport Sci
14. Heck H, Mader A, Hess G, Mucke S, Muller R, Hollmann W. Justification of the 4-mmol/L lactate threshold
. Int J Sports Med
15. Hughson RL, Weisiger KH, Swanson GD. Blood lactate concentration increases as a continuous function in progressive exercise. J Appl Physiol
16. Impellizzeri FM, Rampinini E, Coutts AJ, Sassi A, Marcora SM. Use of RPE-based training load
in soccer. Med Sci Sports Exerc
17. Impellizzeri FM, Marcora SM, Castagna C, et al. Physiological and performance effects of generic versus specific aerobic training in soccer players. Int J Sports Med
18. Krustrup P, Mohr M, Amstrup T, et al. The yo-yo intermittent recovery test: physiological response, reliability, and validity. Med Sci Sports Exerc
19. Lacour JR, Padilla-Magunacelaya S, Barthelemy JC, Dormois D. The energetics of middle distance running. Eur J Appl Physiol Occup Physiol
20. Lucia A, Hoyos J, Santalla A, Earnest C, Chicharro JL. Tour de France versus Vuelta a España: which is harder? Med Sci Sports Exerc
21. Mader A, Heck H. A theory of the metabolic origin of "anaerobic threshold". Int J Sports Med
. 1986;7(suppl 1):45-65.
22. Morton RH, Fitz-Clarke JR, Banister EW. Modeling human performance in running. J Appl Physiol
23. Mujika I. The influence of training characteristics and tapering on the adaptation in highly trained individuals: a review. Int J Sports Med
24. Mujika I, Padilla S. Scientific bases for precompetition tapering strategies. Med Sci Sports Exerc
25. Mujika I, Padilla S, Pyne D, Busso T. Physiological changes associated with the pre-event taper in athletes. Sports Med
26. Padilla S, Mujika I, Orbananos J, Angulo F. Exercise intensity during competition time trials in professional road cycling. Med Sci Sports Exerc
27. Padilla S, Mujika I, Orbananos J, Santisteban J, Angulo F, Jose Goiriena J. Exercise intensity and load during mass-start stage races in professional road cycling. Med Sci Sports Exerc
28. Padilla S, Mujika I, Santisteban J, Impellizzeri FM, Goiriena JJ. Exercise intensity and load during uphill cycling in professional 3-week races. Eur J Appl Physiol
29. Rodriguez-Marroyo JA, Garcia Lopez J, Juneau CE, Villa JG. Workload demands in professional multi-stage cycling races of varying duration. Br J Sports Med
30. Smith DJ. A framework for understanding the training process leading to elite performance. Sports Med
31. Stagno KM, Thatcher R, van Someren KA. A modified TRIMP to quantify the in-season training load
of team sport players. J Sports Sci
32. Tanaka K, Matsuura Y. Marathon performance, anaerobic threshold, and onset of blood lactate accumulation. J Appl Physiol
33. Thomas JR, Nelson JK, Silverman J. Research Methods in Physical Activity
. 5th ed. Champaign (IL): Human Kinetics; 2005. pp. 10-4.
Keywords:©2009The American College of Sports Medicine
TRAINING LOAD; HR; LACTATE THRESHOLD; ENDURANCE