Subjects

The sample comprised 14 male handball players (Age: 22.41 ± 2.27 years; height: 1.84 ± 0.06 cm; body mass: 87.28 ± 10.11 kg), who had experience with this sport for approximately five years. All players were trained twice a day (90-120 min per session), five days per week (depending on the number of matches per week). The training sessions consisted of specific conditioning work, weight training, handball drills, sprints, tactics, and intermittent running exercises.

The exclusion criteria were use of drugs/tobacco, existence of oral disease and any type of physical injury. Thus, the saliva samples of one athlete and isokinetic evaluation of two athletes were properly excluded, because one player had an oral injury during the match, and two players were injured one day before the isokinetic test.

Experimental design

The experiment was divided in two phases. In the first stage, the participants’ isokinetic concentric strength of the dominant and non-dominant lower limb was assessed on the isokinetic dynamometer (Biodex Medical Systems Inc., NY, USA). After they performed a 5 min warm-up on a treadmill (Movement LX 150), the athletes were positioned seated on the isokinetic dynamometer with their hips flexed at approximately 85°, and trunk and thigh stabilized, to avoid compensatory movements. The dynamometer axis was visually aligned with lateral femoral condyle, while the knees were flexed at 90°. The length of the lever arm was individually determined using the length of each athlete's lower leg. The knee was tested on a range of motion from 0° to 90° of knee flexion, with full knee extension considered 0°. As part of the familiarization process, the athletes performed three repetitions in the equipment before starting the test.^2,18

The test protocol was based on concentric isokinetic repetition of flexion and extension with 60 s recovery between sets, which were at angular velocity of 60°/s (five repetitions) and 180°/s (30 repetitions). The evaluated isokinetic parameters were peak torque (PT) of the hamstring (H) and quadriceps (Q) muscles normalized by body mass, fatigue index (FI), and H/Q peak torque ratio (H/Q ratio).^1,2,18

The second phase started after 48 h, where the biochemical changes of saliva were verified in response to a simulated handball match. First, the athletes performed a 20 min warm-up, which included stretching, running, and sport-specific exercises. The match was composed of two halves of 30 min with 5 min interval for recovering and hydration ad libitum. In addition, all official rules were maintained and no player was substituted during the match.

The match intensity was assessed by exercise duration (minutes) multiplied by training load, using the Borg CR-10 scale,¹⁹ recorded 30 min after match. The players answered a simple question: “How was your workout?”, and for this, a chart was shown that outlined the full RPE scale with the appropriate explanations.

All subjects were informed by an information document in advance to abstain from food and caffeine products for at least 2 h prior to the saliva collection. The athletes were required to rinse out their mouths with distilled water, then remained seated with eyes open, head tilted slightly forward, and avoiding orofacial movements.^3,8 The collection was performed approximately 10 min before the pre-match warm-up (∼16:00 h) and post-match at 5 and 120 min.

Unstimulated saliva was collected into a 2 ml sterile tube for approximately 10min per athlete. The samples were immediately cooled down at 18°C and centrifuged at 6500 rpm for 30 min. The supernatant was stored at −20°C.

Defrosted samples were centrifuged again for 5 min at 6500 rpm. After this process, 15 μl of salivary supernatant was deposited on a calcium fluoride (CaF₂) window and dehydrated for 30 min (Eppendorf Concentrator 5301).

Identification of the infrared bands of cortisol and SIgA was done by the acquisition of the spectra of pure substances, purchased from Sigma-Aldrich^® and In vivogen companies, respectively. For this, 15 μl of cortisol in methanol (0.015 mg/μl) and 15 μl of human SIgA in saline (0.015 mg/μl) were deposited on CaF₂ window and lyophilized.

Infrared spectra were collected by a Spectrum 400 spectrophotometer coupled to a microscope (Perkin-Elmer, Spotlight 400, USA), controlled by a computer using Spotlight 400 Software. Spectra were recorded in the spectral region 4000 to 750 cm¹, with 32 scans and a resolution of 4 cm⁻¹. The measurements were performed along the thin film in eight random points, formed on the CaF₂ surface.²⁰ The total spectra collected were 312 for saliva samples.

Statistical analyses

The means and standard deviation of the data were calculated. The distribution of the areas calculated was assumed to be Gaussian, and this assumption/hypothesis was verified using the Kolmogorov-Smirnov test for normal distributions. The one-way ANOVA test was used to compare the group means (Spectra of the saliva sample collected pre-match, post-match and two hours post-match).^21,22 The significance level was set at 0.05. Pearson's coefficient was used for data correlation.

The FT-IR spectra of saliva sample and pure substances (cortisol and human SIgA) were standardized, baseline correction, and normalization (0-1). Then, the areas were calculated by integration of spectral region as follows: region 1 (1180-955 cm⁻¹), region 2 (1484-1191 cm⁻¹), and region 3 (1584-1489 cm⁻¹). For this procedure, the software Excel 2007, Origin V8.5, and OPUS V4.2 was used.

Isokinetic muscle performance

The results from isokinetic tests are in Table 1. The statistics analysis showed no significant difference between the non-dominant (ND) and dominant (D) lower limb from 60°/s and 180°/s, in both extension and flexion, for variables of PT and FI. As expected, the H/Q ratio at 60°/s was lower than at 180°/s.

Table 1 Peak Torque values normalized by body mass, Fatiguez Index and Hamstrings Muscles Quadriceps Muscles Peak Torque Ratio ratio to 60 and 180°/s non-dominant and dominant lower limb of athletes (n = 12). 

Values are presented as mean ± SD;

PT = peak torque normalized by body mass; H = hamstrings muscles; Q = quadriceps muscles; FI = fatigue index; and H/Q ratio = H/Q peak torque ratio.

Analysis of pure substances of cortisol and human SIgA by FT-IR

The average FT-IR spectra of pure substances of cortisol and human SIgA are shown in Fig. 1. The spectra were spread in three regions: (1) 1180-955 cm⁻¹, the bands are assignment mostly for cortisol; (2) 1484-1191 cm⁻¹, cortisol and human SIgA have similar contribution; and (3) 1584-1489 cm⁻¹ has strong contribution of SIgA compared to cortisol.

Fig. 1 Average Fourier transform infrared spectroscopy spectra of cortisol and human salivary immunoglobulin A (pure substances). 

Training load and salivary immune-endocrine responses in handball players by FT-IR

The training load reported by the athletes was 15%, very difficult; 62%, difficult; and 23%, moderate, respectively. Then physical effort of the game was considerably intense.

Fig. 2 shows the average FT-IR spectra of the saliva sample collected pre-match (Pre), post-match (Post), and two hours post-match (Post 2 h). The results showed difference after exercise that could be attributed by biochemical changes in saliva, especially in region 1.

Fig. 2 Average Fourier transform infrared spectroscopy spectra of saliva sample obtained before match, after match, and after 2 h recovery. 

In order to visualize these differences statistically, the integration area was performed in the three regions (Fig. 3). There was no significant difference between the saliva sample collected before, after match, and after two h of recovery, but an increase of bands intensities in regions 1 and 2 was observed as a function of physical effort.

Fig. 3 Means, standard deviation, integrated area of spectra of saliva sample collected before, after match, and after 2 h of recovery for three spectral regions (region 1: 1180-955 cm⁻¹; region 2: 1484-1191 cm⁻¹; region 3: 1584-1489 cm⁻¹). 

Region 1 had the highest variation of intensity after the match although without statistic relevance. These results could be explained by the player position due to difference in physiological demands during the handball match,²³ increasing SD of the area.

In order to take into account the player position and the results of region 1 assigned to cortisol, Fig. 4 shows the differences between pre and post-match as function of the position, such as backcourt, goalkeeper, pivot, and wings. Wings and backcourts had the highest percentages of variation compared to goalkeepers and pivots (Fig. 4).

Fig. 4 Variation of salivary cortisol before and after handball match of players by playing positions. 

Isokinetic muscle performance

There are few studies reporting isokinetic parameters of handball players’ knees. Thus, the results presented in this studied can serve for future comparisons between athletes from the same sport.

The peak torque comparison between D and ND lower limb of athletes were not statistically relevant. Similar results were found in volleyball and soccer players, where the athletes also did not present a difference between D and ND limb.² These results can be explained by training specificity, because the athletes are submitted to training programs involving strength, resistance, balance, and flexibility, which could correct some muscular deficit. This training program is important to improve sport performance and prevent injury.^24,25

Isokinetic evaluation of the H/Q ratio provides a quantitative measurement of torque from agonist and antagonist muscle contraction surrounding the knee joint. The athletes had values of H/Q ratio at 60°/s and 180°/s for D and ND limb in the range described in the literature.^26,27 Although it is not specific for handball athletes, H/Q ratio is typically in the range of 50-80% with an increased values at faster speeds.^26,27 According to Andrade et al.,¹ H/Q ratio higher than 60% at 180°/s is a good marker because only at high test speed are the similar results to sport tasks.

The FI values for agonist/antagonist musculature of D and ND lower limb were not significantly different. All results were below 50%, which represents the ability of the muscle to maintain at least 50% of job generation, during 30 repetitions.²⁸ This reduces injury risks in periods of training and competition, because there is a similarity between flexor and extensor musculature to resist fatigue.

Analysis of pure substances of cortisol and human SIgA by FT-IR

The infrared spectra of pure substance of cortisol and human SIgA were performed in order to show the main bands in the infrared region. These biomarkers are widely used to investigate different physiological responses in collective sports modalities.^3,29 Thus, the immune-endocrine response in handball players was examined by these two salivary biomarkers.

Region 1 (1180-955 cm⁻¹), referring to stretching of CO (carbonoxygen bond), CC (carboncarbon bond), angular bending of COH (carbonoxygenhydrogen bond), COC (carbon-oxygen-carbon bond) and symmetric stretching vibrations of the PO (phosphoroxygen double bond) for phosphate group, is largely attributed to the cortisol molecule. This hormone is formed by C₂₁H₃₀O₅ containing three six-membered rings and one cyclopentane ring with ketone oxygens at C-3 (carbon-3) and C-20 (carbon-20), the double bond between C-4 (carbon-4) and C-5 (carbon-5), the b-hydroxyl group at C-11 (carbon-11), the two-carbon chain in b-orientation, the hydroxyl group in at C-17 (carbon-17), and b-methyl groups at C-18 (carbon-18) and C-19 (carbon-19).¹⁵

Region 2 (1484-1191 cm⁻¹), which corresponds to vibrational modes of CH₂/CH₃, CO (carbonoxygen double bond) of COO⁻ (carboxyl group) symmetric (stretching), amide III (protein), and asymmetric stretching bond (PO) of phosphate group, showed bands of pure substances of cortisol and human SIgA.

Region 3 (1584-1489 cm⁻¹), assigned to tyrosine band, amide II and α-helix (amino acids and proteins), showed only one band of SIgA. This protein consists of four heavy chains and two light chains, adopting a characteristic “immunoglobulin fold”, comprised of various amino acid residues of β antiparallel-sheet, around an internal disulfide bridge.³⁰

Training load and salivary immune-endocrine responses in handball players by FT-IR

Training load was used as an indicator of exercise intensity. This method has largely been used as a useful tool for prescribing exercise intensity based on its relationship with physiological indicators of exercise stress. In addition, it does not appear to be affected by exercise modality or training state.³¹

In this study, the players indicated that the match intensity was difficult or very difficult. The goalkeepers reported the physical effort intensity as moderate, which can be explained by faster recovery intervals of this position.²³

The effects of a handball match on immune-endocrine system by infrared analysis of the salivary biomarkers (cortisol and human SIgA) were examined. The results showed differences in band intensities for spectra of saliva sample collected before, after match, and after 2 h of recovery, especially in regions 1 and 2. However, the comparison of the mean values of the area of region 1 was not statistically relevant.

This shows that the stress imposed by the match was not enough to significantly increase cortisol levels. Handball is considered a complex and highly demanding intermittent sport, because it involves multiple high-intensity and low-intensity activities, such as turns, stops, jumps, throws, changes of direction, and one-on-one situations in offensive and defensive play.²³ Thus, the recovery periods during the handball match and physiological adaptations resulting from physical training avoid high stress levels.³² These finding corroborate with Moreira et al.³³ who did not observe any significant differences in salivary cortisol by enzyme-linked immunosorbent assays (ELISA), after a competitive training match in male professional soccer players.

In addition, cortisol is a glucocorticoid produced in the adrenal glands that are controlled by the hypothalamus-pituitary-adrenal axis in response to stress, which could be related to emotional, physiological, and behavioral dimensions.³⁴ In this sense, the psychological factor as an additional stressor in sports also helps to explain the results. Some studies, for example, showed that levels of cortisol after an official match are considerably higher compared to a simulated match.^3,35

The individual variation of salivary cortisol showed that backcourt players have higher stress levels than other playing positions. Póvoas et al.²³ showed that these player positions require more physically demanding play per game than other positions. This variation of the cortisol values as function of playing positions was correlated with the training load (r = 0.73, p < 0.05).

For the mucosal parameters, there were no differences in band intensities in the region assigned to SIgA for saliva samples collected before, after match, and after 2 h of recovery. These results suggest that the simulated match also did not change SIgA levels in handball players.

As previously discussion, the physical demands (intermittent nature) of a handball match, which includes periods of recovery between efforts and relatively short durations of action,²³ could explain these findings in mucosal parameters. In addition, the psychological factor helps to explain this immune response because it was a simulated match.³

Some authors report that the influence of exercise on the immune response depends on exercise intensity and duration.^36,37 However, the lack of information about the physiological parameters during the match challenge the statement, for example, heart rate and lactate. In addition, time motion analysis could reveal differences in the number and intensity of actions and distance covered by the players, which could help to explain the results obtained. However, some of these measurements are expensive and difficult to monitor during a handball match.

It is noteworthy that the main limitation of this study is the lack of measurements of cortisol and SIgA concentrations using ELISA. These measurements are important to confirm the infrared results because the bands observed in this infrared region are not exclusively for cortisol and SIgA, because other salivary biomarkers can change in response to stress and could have similar chemical structure, showing similar bands in regions of SIgA and cortisol (e.g., α-amylase and testosterone). Despite these limitations, FT-IR spectroscopy produced promising results with quantitative analysis of stress biomarkers and provided real-time information without the use of any reagents. This decreases the price of the analyses and precise monitoring of players that belong to teams without financial support, to help them avoid excessive training loads during the preparation to decisive matches.

In summary, the results of this study showed that the handball players had good performance of the lower limb muscles with balanced H/Q, no strength difference between D and ND limbs, and appropriate IF percentage. The results of infrared analysis showed the main absorption bands for cortisol in region 1180-955 cm⁻¹ and for human SIgA in region 1584-1489 cm⁻¹. There were no significant spectral differences to saliva sample collected pre and post-match, as well as after 2 h of recovery. In addition, the playing position as function of cortisol absorption band could be positively correlated to the training load, showing the playing position that require higher degrees of stress during a simulated handball match.