The midpoint between ventilatory thresholds approaches maximal lactate steady state intensity in amateur cyclists

Abstract

The aim was to determine whether the midpoint between ventilatory thresholds (MPVT) corresponds to maximal lactate steady state (MLSS). Twelve amateur cyclists (21.0 ± 2.6 years old; 72.2 ± 9.0 kg; 179.8 ± 7.5 cm) performed an incremental test (25 W·min^-1) until exhaustion and several constant load tests of 30 minutes to determine MLSS, on different occasions. Using MLSS determination as the reference method, the agreement with five other parameters (MPVT; first and second ventilatory thresholds: VT1 and VT2; respiratory exchange ratio equal to 1: RER = 1.00; and Maximum) was analysed by the Bland-Altman method. The difference between workload at MLSS and VT1, VT2, RER=1.00 and Maximum was 31.1 ± 20.0, -86.0 ± 18.3, -63.6 ± 26.3 and -192.3 ± 48.6 W, respectively. MLSS was underestimated from VT1 and overestimated from VT2, RER = 1.00 and Maximum. The smallest difference (-27.5 ± 15.1 W) between workload at MLSS and MPVT was in better agreement than other analysed parameters of intensity in cycling. The main finding is that MPVT approached the workload at MLSS in amateur cyclists, and can be used to estimate maximal steady state.

INTRODUCTION

The exercise intensity eliciting maximal steady state blood lactate concentration (MLSS; maximal lactate steady state) is a reliable index of endurance capacity [1, 2], i.e. the physiological ability to tolerate long lasting exercises at a higher aerobic rate without intramuscular and blood acid-base perturbations [3–5]. Mean metabolic rate at a workload corresponding to MLSS is about 70-75% of maximum oxygen uptake (VO_2max) for cyclists [4], which does not differ from 75 ± 5% VO_2max reported in other sports modalities [2, 6]. Indeed, MLSS relative to workload at VO_2max (65-70%) is more independent of motor task performance than blood lactate concentration ([La^-]), which ranges from 2 to 8 mmol ∙ L^-1 at MLSS, and relates to the amount of muscle mass engaged in exercise [2, 7, 8].

Concerning the methodological aspects of MLSS assessment, the gold standard protocol requires three to four 30-minute tests with exercise intensity ranging from 60 to 80% VO_2max [2]. By applying this protocol, the velocity or workload at MLSS is defined as the highest exercise intensity attained without blood lactate concentration changes above 1 mmol.L^-1, during the final 20 minutes of the bout [9]. However, its usefulness is limited by the need for time-consuming tests [5, 10]. The practical disadvantage of numerous tests for direct MLSS assessment has motivated studies to investigate time saving, less expensive and non-invasive procedures. Many attempts have related protocols of aerobic capacity evaluation, such as critical velocity, to the velocity at MLSS, reporting good relationships between these indexes [11–13]. Even other remarkable indexes of endurance capacity such as critical power (CP) [1, 3], ventilatory (VT) or lactate threshold (LT) have evidenced similarities to the velocity or workload at MLSS [1, 14–20], but none of them confirmed that the physiological responses encompassed by MLSS could be exchanged for these indexes.

Despite CP being recognized as the exercise intensity near to MLSS, the metabolic correspondence between these two points has not been demonstrated [21, 22]. Indeed if, as seems probable, intensity corresponding to MLSS in different sports modalities lies above VT₁ (or LT), and below the point where respiratory compensation for metabolic acidosis starts (RCP, or VT₂) [1, 14, 21, 23–25], it would be expected that an intermediate intensity might be the nearest to MLSS. For this reason an intensity corresponding to 3.5 mmol ∙ L^-1 [26], referred to as the individual anaerobic threshold (IAT) [17, 20, 27], or the intensity corresponding to a respiratory exchange ratio (RER) equal to 1 [25, 28], has been proposed as an indicator of MLSS.

Although the midpoint between the ventilatory thresholds (VT₁ and VT₂) from a progressive ramp protocol would correspond to MLSS, and one maximal aerobic test would be enough to locate MLSS, as far as we have been able to ascertain this has not yet been explored. Therefore, the aim of the present study was to verify whether the intensity corresponding to the midpoint between the ventilatory thresholds (MP_VT) corresponds to MLSS intensity among an amateur group of cyclists. We hypothesized that the power output corresponding to MP_VT, determined during a single maximal incremental test, would allow easier calculation of MLSS power output.

MATERIALS AND METHODS

Subjects

Twelve amateur road cyclists (elite-sub23 category) were selected for this investigation (21.0±2.6 years, 179.8±7.5 cm, 72.2±9.0 kg). A physical examination before the start of the study was carried out to ensure that each participant was in good health. The benefits and risks of the protocol were explained, and the subjects signed an informed consent form, following approval from the ethical committee of the Technical University of Madrid.

Procedures

Each subject carried out an incremental test during the first visit. Several constant load tests of 30 minutes were performed thereafter (48 h) in order to determine the intensity corresponding to MLSS. These steady state tests were carried out with a 48 h interval between them. Each cyclist performed all tests at the same time of day under similar environmental conditions (22.8±0.6ºC and 62.4±4.4% relative humidity). Subjects were asked to refrain from hard physical work and consumption of any medication or stimulants for at least 24 h before each experimental session. During the tests, subjects adopted the conventional upright cycling posture. This posture is characterized by a trunk inclination of ~75º and by the subject placing their hands on the handlebars with elbows slightly bent (~10º). Before the tests, each cyclist adjusted the corresponding cycle ergometer and used their own clip-on pedals [29, 30].

Gas exchange data were collected continuously during each test using an automated breath-by-breath system (Jaeger Oxycon Pro gas analyser, Erich Jaeger, Viasys Healthcare, Germany). The following variables were recorded during the tests: oxygen uptake (VO₂), carbon dioxide output (VCO₂), respiratory exchange ratio (RER), ventilation (V_E), respiratory rate (RR), the end tidal partial pressures of O₂ (PETO₂) and CO₂ (PETCO₂), and the respiratory equivalents of O₂ (VE·VO₂^-1) and CO₂ (VE·VCO₂^-1). A 12-lead electrocardiogram (ECG; Viasys Healthcare, Germany) was continuously recorded during the tests to determine heart rate (HR) [31, 32].

Maximal incremental test

A continuous incremental cycling test was used to determine maximal oxygen uptake (VO_2max) and ventilatory thresholds (VT). The test was performed on a conventional cycle ergometer (Jaeger ER800, Erich Jaeger, Germany). After a 3-min warm-up at 50 W, the workload was increased by 5 W every 12 s (25 W·min^-1) until exhaustion. Subjects were allowed to choose their preferred cadence within the 70-90 rpm range. Verbal encouragement was provided to ensure that maximal effort was reached. All the subjects had previous experience with this type of protocol, which has been used for the physiological evaluation of professional cyclists in several previous studies [30, 33–35] and is reliable for the detection of the VT [32]. At least two of the following criteria were required for the attainment of VO_2max: a plateau in VO₂ values despite increasing workload, RER≥1.1, or the attainment of 95% of the age-predicted maximum heart rate (HR_max) [30, 36].

The maximum 15 s average value of VO₂ attained during the test was reported as VO₂max, and the maximum workload achieved during the last stage of the progressive test was identified as the Maximum [25]. The first and the second VT (VT₁ and VT₂, respectively) were set at the points of maximum agreement of the most common methods of assessment [37]. Briefly, VT₁ was calculated 1) according to the V-slope method [31], where VT₁ is the break point of the VCO₂-VO₂ relationship, 2) as the first exponential increment in ventilation [38], and 3) as the first rise in V_E·VO₂^-1 without increments in V_E·VCO₂^-1 [39]. VT₂ was determined as the second rise in ventilation [38] and as the intensity that accompanied a second rise in V_E·VO₂^-1 with a concurrent rise in V_E·VCO₂^-1 [39]. All tests were evaluated by two researchers in a double blind process. The coefficient of variation between the assessments of these two researchers and those of a highly experienced expert was 1.3%.

Determination of MLSS

Constant load tests of 30 min were carried out to determine MLSS. These were performed on a road bicycle fitted with an SRM powermeter (Schoberer Rad Messtechnik SRM, Jülich, Germany). The bicycle was then mounted on a Tacx CycleForce Grand Excel ergometer (Technische Industrie Tacx BV, Netherlands). This ergometer was not used for analysis purposes but only as a platform on which to mount the test rig. Participants were allowed to use their own pedals and saddle. Height and reach were adjusted to match the participant’s own bicycle as closely as possible.

The first constant workload trial was performed at an intensity corresponding to MP_VT, previously calculated in the maximal incremental test [(workload at VT₁ + workload at VT₂)·2^-1]. Another 30 min test was performed at a higher intensity with an increase of 5% of maximum load 48 h later if, during the first test, lactate concentration [La^-] remained steady or decreased. Subsequent 30 min constant tests were increased by an additional 5% of the previous intensity until no lactate steady state could be maintained. Inversely, if [La^-] increased continuously or the exercise was interrupted due to the subject’s fatigue during the first 30 min test, the workload was decreased by 5% of Maximum for each test until a steady state could be maintained. MLSS was defined as the highest workload that could be maintained with an increase in [La^-] lower than 1.0 mmol·L^-1 during the final 20 min of the constant load tests [5, 28, 40–42].

Blood samples

Before each test, an 18G catheter was inserted into a forearm vein for venous blood sampling. Samples were drawn prior to and during exercise at different moments in order to determine [La^-] every 2 min and at the moment when maximal effort was deemed to have been reached in the incremental test, every 5 min throughout the steady state tests (0, 5, 10, 15, 20, 25 and 30 min) and at exercise termination if the test could not be maintained. [La^-] was analysed by an enzymatic method (YSI 1500, Yellow Springs Instruments Co., Ohio, USA).

Statistical analysis

All data are reported as mean (±SD). One way ANOVA was used to examine the differences between the values obtained at the different points in the incremental test (VT₁, MP_VT, VT₂, RER=1.00 and Maximum) with the values at MLSS. Multiple comparisons were made using the Bonferroni post hoc test. The coefficient of variability (CV%), standard error of the mean and Pearson’s correlation coefficient were calculated to evaluate the workload differences between MLSS and the different points. Bland–Altman plots [43] were drawn to establish the limits of agreement for the five points of the incremental test plotted against MLSS. Bland-Altman plots were also used to compare VO₂, VE, HR and [La^-] assessed using MP_VT and MLSS. Linear regression analysis and correlation coefficients were calculated and included in the plots. All analyses were carried out with SPSS version 19 (Chicago, Illinois, USA), and the level of statistical significance was set at p<0.05 for all analyses.

RESULTS

The mean value of workload at MLSS was 284±30 W, within a range from 236 to 323 W. Table 1 shows the results and the differences found between the physiological parameters from the incremental test (VT₁, MP_VT, VT₂, RER=1.00 and Maximum) and MLSS. The workload at MLSS was not different from VT₁ or MP_VT for absolute values (W), those relative to body mass (W.kg^-1) or those relative to maximum values, but was located closer to MP_VT than VT₁. The VO₂ at MLSS was not different from VO₂ at MP_VT, VT₂ or RER=1.00, comparing the values in absolute terms, relative to body mass, as well as relative to maximum VO₂. Otherwise, VO₂ values for MLSS and MP_VT parameters were the closest. Similarly, HR at MLSS did not differ significantly from HR at MP_VT, VT₂ and RER=1.00. The values for other parameters at MLSS (VE, RR, VE·VO₂^-1, PETO₂ and [La^-]) were similar to those at VT₂ and RER=1.00.

TABLE 1.

Mean ± SD for variables obtained at each point during the incremental test and mean last 20 minute values at MLSS.

	MLSS	VT1	MPVT	VT2	RER=1.00	Maximum
Workload (W) Workload (W·kg-1) %Workloadmax	284 ± 30 4.0 ± 0.4 60.0 ± 5.6	253 ± 37 3.5 ± 0.5 53.3 ± 6.3	311 ± 32 4.3 ± 0.5 65.8 ± 5.6	370 ± 32a 5.2 ± 0.6a 78.3 ± 6.3a	347 ± 35a 4.9 ± 0.6a 73.8 ± 8.6a	476 ± 62a 6.6 ± 0.5a 100.0a
VO2 (mL·min-1) VO2 (mL·min-1·kg-1) %VO2max VCO2 (mL·min-1) VE (L·min-1)	4225 ± 414 58.8 ± 4.7 81.8 ± 7.0 3821 ± 404 121 ± 12	3244 ± 464a 45.2 ± 6.5a 62.7 ± 7.6a 2802 ± 458a 72 ± 15a	4052 ± 362 56.6 ± 6.3 78.5 ± 5.9 3712 ± 415 99 ± 11a	4588 ± 364 64.2 ± 7.3 88.9 ± 4.8 4534 ± 416a 123 ± 12	4367 ± 386 61.9 ± 8.4 87.2 ± 4.8 4384 ± 389 119 ± 12	5175 ± 474a 72.4 ± 8.7a 100.0a 5173 ± 256a 176 ± 12a
HR (beats·min-1) RR (breaths·min-1) PETO2 (kPa) PETCO2 (kPa)	175 ± 8 47 ± 7 13.5 ± 0.5 4.7 ± 0.5	155 ± 14a 31 ± 6a 12.2 ± 0.5a 5.6 ± 0.5a	171 ± 10 37 ± 5a 12.8 ± 0.4a 5.3 ± 0.4	183 ± 10 41 ± 5 13.2 ± 0.5 5.3 ± 0.5	179 ± 8 41 ± 6 13.2 ± 0.4 5.3 ± 0.4	194 ± 6a 59 ± 6a 14.2 ± 0.5a 4.4 ± 0.5
VE·VO2-1 VE·VCO2-1 [La-] (mmol·L-1)	28.5 ± 2.7 31.6 ± 3.8 3.60 ± 0.81	21.6 ± 2.3a 24.9 ± 2.3a 1.32 ± 0.43a	23.8 ± 2.1a 26.0 ± 1.8a 1.90 ± 0.69a	26.2 ± 2.6 26.5 ± 2.3a 3.40 ± 1.27	26.5 ± 2.1 26.4 ± 2.1a 2.76 ± 1.25	34.1 ± 2.9a 33.6 ± 2.2 8.10 ± 1.90a

Note: VT1, first ventilatory threshold; MPVT, midpoint between the ventilatory thresholds; VT2, second ventilatory threshold; RER=1.00, respiratory exchange ratio equal to 1; MLSS, maximal lactate steady state; %Workload_max, percentage of maximal workload; VO₂, oxygen uptake; %VO2max, percentage of maximal oxygen uptake; VCO₂, carbon dioxide production; VE, ventilation; HR, heart rate; RR, respiratory rate; PETO₂, end tidal partial pressure of oxygen; PETCO₂, end tidal partial pressure of carbon dioxide.

^aSignificantly different from MLSS.

The Bland-Altman agreement analysis for workload intensity at MLSS with workload at MP_VT, VT1, VT₂, RER=1.00, and Maximum are shown in Figure 1. The mean difference between workload at MLSS and at VT₁ (Fig. 1A), VT₂ (Fig. 1C), RER=1.00 (Fig. 1D) and Maximum (Fig. 1E) was 31.1±20.0 W (range: 18.3 to 43.8 W), -86.0±18.3 W (range: -74.4 to -97.7 W), -63.6±26.3 W (range: -49.3 to -86.9 W) and -192.3±48.6 W (range: -161.4 to -223.2 W), respectively. Thus, VT₁ underestimated MLSS while VT₂, RER=1.00 and Maximum overestimated it. The mean difference of -27.5±15.1 W (range: -17.9 to -37.1 W) between workload at MLSS and at MP_VT was the smallest difference among analysed workload intensities (Figure 1B).

FIG. 1.

Bland-Altman plots comparing workload for VT₁-MLSS (A), MP_VT-MLSS (B), VT₂-MLSS (C), RER=1.00-MLSS (D) and Maximum- MLSS (E). Linear regression analysis and correlation coefficient (r) are included in the plots.

Workload corresponding to MLSS was significantly correlated with all points (Table 2). Workload at MP_VT was highly correlated with MLSS (r=0.885, p<0.05), and the standard error of the mean was the lowest between MLSS and the different points of the incremental test (MLSS- MP_VT: 4.3 W; Table 2). Moreover, %CV of MLSS-MP_VT was 6.6%.

TABLE 2.

Correlation coefficient (r), standard error of the mean and coefficient of variability (CV%).

	R	Standard error of the mean (W)	%CV
MLSS-VT1	0.836*	5.8	8.5
MLSS-MPVT	0.885*	4.3	6.6
MLSS-VT2	0.827*	5.3	18.7
MLSS-RER=1.00	0.730*	8.2	14.3
MLSS-Maximum	0.653*	14.0	35.5

Note: VT1, first ventilatory threshold; MPVT, midpoint between the ventilatory thresholds; VT2, second ventilatory threshold; RER=1.00, respiratory exchange ratio equal to 1; MLSS, maximal lactate steady state.

^*Indicates significant correlation (p<0.05).

The mean differences in VO₂ (173.2±282.8 mL·min^-1), V_E (22.1±12.0 L·min^-1), HR (4±11 beats·min^-1) and [La^-] (1.7±0.9 mmol·L^-1) between MLSS and MP_VT are shown in Figure 2. No significant correlations were found between MLSS and MP_VT in VO₂, V_E, HR and [La^-].

FIG. 2.

Bland-Altman analyses between MLSS and MP_VT: VO₂ (A), VE (B), HR (C) and [La^-] (D). Linear regression analysis and correlation coefficient (r) are included in the plots.

DISCUSSION

The main finding of this study was to locate the workload corresponding to MLSS in well-trained amateur cyclists close to the midpoint between the ventilatory thresholds. This intensity is the closest to MLSS, as the average mean difference was -27.5±15.1 W, whereas the intensities at VT₁, VT₂, RER=1.00 and Maximum are further away and these points cannot be taken as indicators of intensity at MLSS.

The role of MLSS as an index of aerobic endurance [4, 18, 44–46] and as a training stimulus to improve this ability [10, 47, 48] has motivated the search for a single assessment protocol [45, 49–51], since the gold standard protocol comprises the performance of an incremental test followed by successive constant intensity tests [5, 9, 52]. Although some of the defined points during an incremental test have been proposed as intensities that indicate MLSS, which would permit its determination with one single test [1, 15, 19, 25, 27, 28, 52–55], none of these studies is definitive, and the challenge remains to be able to determine this intensity with just one test.

The present study has identified MP_VT as the nearest intensity to MLSS, as the rest of the points determined were further away. The difference between MP_VT and MLSS could be taken as a reference for determining MLSS. Bearing in mind that the maximal test was performed on a different ergometer from the one used in the constant load tests, the difference in the load between MP_VT and MLSS could be attributed to this circumstance, decreasing the internal validity of the study. However, external validity increases, as the data obtained in the laboratory can be transferred to training sessions, using a portable ergometer such as the SRM system. We suggest that a difference of 27 W in training intensity is probably realistic for amateur cyclists and easy to adjust with a field test, such as a 40 km time trial. In any case, the variability of MLSS-MP_VT is low and in line with the results of Hauser et al. [56], who reported variability values of 3% for MLSS power. Furthermore, our results suggest an easy approach to determine MLSS, since the mean difference between workload at MLSS and at MP_VT is reduced to -3.2±12.4 W by subtracting 27 W from MP_VT. Only one subject shows a difference greater than 20 W, while the variability is 2.4%.

The difference between RER=1.00 and MLSS reported by Laplaud et al. [28] was 3.6±8.1 W, which is less than the difference observed in the present study, although they did not report the intensity at RER=1.00 as MP_VT, despite this point coinciding with the mean value between ventilatory thresholds. The estimation of MLSS from the intensity at RER=1.00, VT₁ and VT₂, determined during a single maximal incremental field test in well-trained long and middle distance male runners, showed a better agreement between velocity at MLSS and RER=1.00, than with velocity at VT₁ and VT₂ [25]. Again, RER=1.00 coincided with the velocity at MP_VT. The level of training could affect exercise intensity where RER=1.00 during a maximal test. In highly trained athletes this is near to the VT₂, as shown in our results, due to greater energy production from lipid metabolism [28, 57, 58]. Thus, in well-trained athletes it is not advisable to consider that the intensity at RER=1.00 represents MLSS. Workload at VT₁ seems to underestimate the intensity at MLSS [1, 28, 59], as even though at this intensity lactate concentration is steady [53], it does not match the maximum steady state level. Our results showed that MLSS was also above VT₁. Conversely, VT₂ overestimates MLSS [1, 28], although both are physiologically related [1]. By definition, MLSS should be between the two ventilatory thresholds, as reported by Benito et al. [60], observing a steady state [La^-] for exercise intensity at MP_VT.

Different lactate threshold methods have been proposed for estimating MLSS, such as the IAT, or anaerobic thresholds at fixed [La^-] of 3.5 and 4 mmol ∙ L^-1 [16, 19, 61, 62], but none of these methodologies have presented conclusive results. The differences in test protocols used in the original threshold investigations, the large individual differences shown by the lactate thresholds, and [La^-] applied as references for the anaerobic threshold explain the discrepancies [26, 63, 64]. Probably, lactate threshold methods do not accurately estimate the intensity corresponding to MLSS, as the correspondence found between the fixed lactate concentrations and the intensity corresponding to MLSS may be due to a mere coincidence, and an overall interpretation of the result neglecting the individual differences [48, 54, 64]. The range of [La^-] at MLSS [2, 7] and high day-to-day variability for lactate at MLSS [56] support the coincidental similarity between a given lactate value from incremental and constant intensity exercise. Therefore a comparison of lactate “intensities” should be avoided, being more adequate using power or workload parameters. Indeed, differences in physiological profile during exercise at constant intensity (and steady metabolic rate) from a non-constant and increasing exercise rate have been well documented [1, 65] and likewise shown by our results.

CONCLUSIONS

The main conclusions of the study were: (a) the workload corresponding to MLSS in amateur cyclists is located at a point which is near to the MP_VT, being the nearest intensity, while VT₁, VT₂, RER=1.00 and Maximum cannot be taken as indexes of MLSS; and (b) MLSS could be determined with a single maximum incremental test, as it is located at a workload fairly close to the MP_VT, or even just below it. Further information is required to confirm that the MP_VT is a good estimator of MLSS, focusing on a broad sample of elite cyclists from different specialties, and non-elite and elite endurance athletes from other cyclic sports.

Conflict of interests

The authors declared no conflict of interests regarding the publication of this manuscript.