Resistance training (RT), besides providing gains in muscle strength, size, power and local endurance, may help avoid injury (Faude et al. 2009). However, to fulfill these objectives of RT, exercise performance needs to be optimized by objectively determining the ideal training load for each individual (Gonzalez-Badillo and Sanchez-Medina, 2010). One way to do this is to calculate a relative load percentage of the one-repetition maximum (1RM), or maximal strength (Fry, 2004). Another way to determine the ideal exercise intensity is the use of movement velocity as an indicator of relative load in a given exercise, thus avoiding the need for a 1RM test (Gonzalez-Badillo and Sanchez-Medina, 2010). As opposed to resistance exercise, in endurance sports such as running, cycling or swimming, the first and second thresholds of the triphasic model first described by Skinner and McLellan (1980) are often used as indicators of the training load and may be determined using several methods (Bosquet et al., 2002). In addition, a wide range of intensities is used in exercise prescription models for both healthy subjects and patients and these intensities are based on maximal heart rate (HR) and oxygen uptake ([VO.sub.2]) (Hofmann and Tschakert, 2011). This generates heterogeneous cardiovascular and metabolic responses, which makes it difficult to define adequate work zones and leads to errors. A solution to this problem could be the selection of training loads based on markers of submaximal exercise (thresholds or cutoffs) (Hofmann and Tschakert, 2011). One such method is to locate the lactate threshold (LT2), defined as the first significant sustained increase produced in blood lactate concentration above resting levels during incremental exercise (Meyer et al., 2005). The LT2 is considered a valid indicator of endurance performance (Faude et al., 2009) and is associated with a low/moderate exercise intensity in high-level endurance
athletes (Lucia et al., 2001). In effect, this intensity reflected by the LT2 has been also described to help improve cardiorespiratory fitness in recreational sports (Faude et al., 2009), healthy adults and elderly persons (Ratamess et al., 2009; Simoes et al., 2010; de Sousa et al., 2012), and patients (Meyer et al., 2005). The LT2 marks the upper limit of almost exclusively aerobic work, as an increase in blood lactate is related to a predominant anaerobic metabolism (Meyer et al., 2005). Thus, working at an individual's LT2 has the benefit that exercise can be performed over several hours (Faude et al., 2009). In addition, working at the LT2 could also help improve local muscular endurance through aerobic metabolism (Ratamess et al., 2009; Simoes et al., 2010; de Sousa el al., 2012). Despite these implications of the LT2, its use as an indicator of exercise intensity in RT has been little addressed (Barros et al., 2004; de Sousa et al., 2011; 2012; Moreira et al., 2008; Oliveira et al., 2006; Simoes et al., 2010; 2013). A possible reason for this could be that RT conducted at low or moderate loads (40%-70% 1RM) is dependent on anaerobic metabolism with muscle glycogen used as the main energy source (Collins et al., 1989) and it is difficult to determine the LT2 during RT, since blood lactate levels are elevated from the onset of exercise. In addition, RT induces the recruitment of fast twitch muscle fibers reducing the activity of oxidative enzymes (Dudley, 1988) and may also lead to a decrease in mitochondrial volume and capillary density, and thus limit aerobic performance (Kraemer et al., 1988). Such a predominance of anaerobic work will also impair the detection of the LT2. To address these difficulties in LT2 detection, a new method has been recently introduced that periodizes intensities in RT at the LT2 intensity (de Sousa et al., 2011). To develop this method, the authors identified the LT2 in the leg press (LP) (Barros et al., 2004; Moreira et al., 2008; Oliveira et al., 2006; Simoes et al., 2010; 2013), biceps curl (elbow flexion) (Barros et al., 2004), and bench press (Oliveira et al., 2006; Moreira et al., 2008), and examined responses to constant-load exercise at the LT2 intensity (de Sousa et al., 2012). The findings of this last study indicate that cardiorespiratory and metabolic variables stabilize when the intensity is ~ 30% 1RM. Compared to LP, the leg movement in HS probably requires greater muscular activity. In a study comparing LP and squats, the latter were found to elicit greater muscle activation and achieved greater strength gains (Escamilla et al., 2001). According to Garhammer (1981), free-weight resistance exercise like squat shows more neuromuscular specificity and may be appropriate for transfer to sports activities. However, the literature lacks similar studies examining the LT2 in other forms of resistance exercise such as half squat (HS). This type of exercise is especially suitable for improving patterns of intra- and intermuscular coordination and resembles specific tasks involving an individual's own body weight (Ratamess et al., 2009), which are, in turn, essential for executing activities of daily living.
The aim of this study was thus to determine the exercise intensity corresponding to LT2 in a HS incremental resistance test. Once the LT2 had been established, we assessed the effects on metabolic and cardiorespiratory variables of an exercise protocol conducted at a constant exercise intensity equivalent to the LT2. The working hypothesis was that stable acute responses would persist over time.
Experimental approach to the problem
Each participant completed three HS exercise sessions, each separated by a rest period of 48 hours. In the first session, the individual's 1RM was determined. In the second session, LT2 was determined in an incremental test, in which the load was increased in small percentage steps relative to the 1RM. Then, using the load corresponding to the LT2, an exercise protocol was designed for each subject to include periods of work and rest; this protocol was executed for approximately 31 min in the third session. Participants were instructed on the following established protocol for this exercise: 1) each repetition should last 2 s, 1 s for the eccentric and 1 s for the concentric phase (except the 1RM test); these phases were identified through visual and acoustic signals emitted by a metronome (except the 1RM test), 2) for the HS movement, the knees should be flexed to an angle of 90[degrees] (monitored using a goniometer).
24 healthy young men (age 21.5 [+ or -] 1.8 years; height 180.1 [+ or -] 5.2 cm; weight 81.9 [+ or -] 8.7 kg) were recruited among the undergraduates of the degree course in Physical Activity and Sport Sciences. Subjects were excluded if they were elite athletes or took any sort of medication or performance-enhancing drugs. The subjects were required to obtain a 1RM test result of at least 150 kg in the HS (mean 1RM = 203.7 [+ or -] 42.2 kg) (Table 1). The minimum experience in resistance training of the study participants was 6 months and they were all accustomed to HS exercise.
Participants were informed of the experimental procedures and signed a consent form before performing the tests. During the study, subjects refrained from participating in other strength or endurance sports. They were also instructed not to eat, smoke or drink tea or coffee in the two hours prior to the tests (they were allowed some water). Participants performed all tests at the same time each day. The study protocol received institutional review board approval (Department of Physical Activity and Sport Sciences) and was conducted according to the tenets of the Declaration of Helsinki.
Session 1: one-repetition maximum
The 1RM for the HS was determined using the method described by Baechle et al. (2000). After a general warm-up (10 min of stretching and joint mobility exercises), subjects undertook a HS warm-up protocol (1 set of 3-5 repetitions lifting a weight equivalent to their body weight). After a 2-min recovery period, the participants started the 1RM test in which increasing weights were lifted in 3 to 5 exercise sets of 1 repetition each. In the first set, the load used was 50% of the estimated 1RM; in the second and third sets, it was 70% and 90% of this value, respectively. For subsequent sets, the weight was adjusted until the 1RM, recorded as the heaviest load lifted by the subject in one repetition. Rest intervals between sets were 4 min.
Session 2: graded incremental test
The incremental resistance test (IRT) used was that described by de Sousa et al. (2012), Garnacho-Castano et al. (2015a) and (2015b) in which the starting load is 10% 1RM and this is increased in small steps until 40% 1RM (20, 25, 30, 35, 40% 1RM). Subsequently, the load lifted was increased in steps of 10% of 1RM until exhaustion. This enables detection of the point at which blood lactate starts to rise from resting levels. The test was preceded by a general warm-up consisting of 5 min of gentle running followed by 5 min of ballistic stretching and upper and lower extremity joint mobility exercises. Each step of the test performed at the different loads (% 1RM) lasted 1 min, and included 30 repetitions of 2 s each (1 s eccentric, 1 s concentric). Between each load increase, blood samples were collected during a 2-min recovery period. These recovery periods were necessary for several reasons: 1) so that the weights on the barbell could be replaced increasing the workload by 5% of the 1RM each time, 2) so that blood samples could be obtained by finger pricking for lactate determination, 3) to more easily detect the workload at which blood lactate starts to rise. Using smaller stepwise increases in workloads without rest periods makes it more...