*Article* **Biological, Psychological, and Physical Performance Variations in Football Players during the COVID-19 Lockdown: A Prospective Cohort Study**

**Giulia My <sup>1</sup> , Santo Marsigliante <sup>1</sup> , Antonino Bianco <sup>2</sup> , Daniele Zangla <sup>2</sup> , Carlos Marques da Silva <sup>3</sup> and Antonella Muscella 1,\***


**Abstract:** This prospective cohort study aimed to evaluate whether COVID-19 lockdown caused biological, psychological, and/or physical performance variations in footballers. We compared the 2018/2019 and 2019/2020 seasons evaluating the plasma volume, hematological parameters, iron/ferritin, creatine kinase, vitamin D, cortisol, testosterone, and physiological state of players of the Italian football major league (Serie A). Measurements were performed before the preparatory period (T0), at the beginning (T1) and in the middle (T2) of the championship, and in March (T3) and at the end of season (T4). The results showed that in the 2019/2020 season affected by the lockdown, the weight, BMI, and fat mass percentage were higher than in the previous season. Hematocrit, hemoglobin, red blood cells, and ferritin decreased during both seasons, more significantly than in the regular season. During both seasons, creatine kinase increased from T2 whilst iron concentrations decreased in T3. Testosterone increased in both seasons from T0 to T3 and returned to initial levels at T4; cortisol increased in T2 and T3 during the 2018/2019 season but not during the COVID-19 season. Physical performance tests revealed differences associated with lockdown. Thus, although from a medical point of view, none of the evaluated changes between the two seasons were clinically relevant, training at home during lockdown did not allow the players to maintain the jumping power levels typical of a competitive period.

**Keywords:** COVID-19 lockdown; hematological parameters; psychological stress; cortisol; testosterone; physical performance; football; Serie A

#### **1. Introduction**

On 11 March 2020, the World Health Organization (WHO) officially declared a pandemic status, caused by the new coronavirus (SARS-CoV-2). The official name given by the World Health Organization to the syndrome caused by the virus is COVID-19 (short for Coronavirus Disease—2019). The pandemic of the viral disease caused by the new coronavirus is still ongoing and returning to normal activities is still a challenge [1].

The need to reduce the risk of disease transmission has also had a huge impact on sport and exercise in general. Team sports activities have been suspended, starting from the football championships up to the pinnacle of sporting excellence, the 2020 Olympic Sports Games. It emerged, in fact, that even elite athletes, who are already normally under competitive stress, are affected by physical and mental conditions consequences of the COVID-19 pandemic [2]. This pandemic-induced mental stress for elite athletes originated

**Citation:** My, G.; Marsigliante, S.; Bianco, A.; Zangla, D.; Silva, C.M.d.; Muscella, A. Biological, Psychological, and Physical Performance Variations in Football Players during the COVID-19 Lockdown: A Prospective Cohort Study. *Int. J. Environ. Res. Public Health* **2022**, *19*, 2739. https:// doi.org/10.3390/ijerph19052739

Received: 2 February 2022 Accepted: 24 February 2022 Published: 26 February 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

as early as the cancellation or postponement of matches, the ban on training, and the frequent removal and placement of blocks, also generating uncertainty for their athletic career [3,4].

After lockdown, football leagues in Europe have faced a congested schedule with multiple matches per week and short recovery periods to complete the season. This hampered the adequate preparation of players [5], leading to lower physical performance [6] and increased mental stress [7] in the season continuation after the COVID-19 lockdown. The incidence of injuries does not seem to have changed significantly with the return-to-play after the first COVID-19 lockdown in Italian professional soccer players; however, the schedule congestion and changes to the pace of the game seem to have revised the epidemiological data to date [5]. To minimize overtraining and/or injury risks in such periods, football players need the best individual assessment of their health.

Therefore, the purpose of this study was to evaluate any changes in the metabolic, muscle, and hormonal responses of elite-level football players during two consecutive seasons: the pre-pandemic season 2018/2019 and the following one (2019/2020) affected by the COVID-19 pandemic.

#### **2. Materials and Methods**

#### *2.1. Design*

This was a prospective cohort study performed in a professional elite football club in the Italian Premier League during the 2018–2019 and 2019–2020 seasons up until after the lockdown due to the COVID-19 pandemic (Figure 1). All players were evaluated five times during the study: i.e., T0, T1, T2, T3, and T4. As shown in Figure 1, T0 was scheduled before the start of the preparatory period (week 0; middle July); T1 was about at the beginning of the championship (week 14; October); T2 was programmed in the middle of the championship (week 25; January); T3 was in March (week 35); and T4 was at the end of the season (week 51; June).

**Figure 1.** A timeline of the 2019/2020 Serie A season surrounding the impact of the COVID-19 pandemic lockdown.

#### *2.2. Participants*

Twenty-four players (aged 22–35 years) from an Italian Serie A football team, participating in both the 2018/2019 and 2019/2020 season, were recruited. As exclusion criteria, samples of blood were not taken from the player if they had any injury during both seasons; players that tested positive to COVID were also excluded. Goalkeepers were excluded due to their specific role in the team. Athletes who attended less than 85% of the scheduled training sessions and matches were also excluded from the study. In total, 17 players were eligible for inclusion in this study.

#### *2.3. Training Program*

Table 1 shows the training program followed by the team during the two seasons, differentiated according to the number of games played during the week (one or two). This

program was the same in both seasons, excluding the COVID-19 lockdown, imposed in Italy from 9 March, 2020 to 3 May, 2020.

**Table 1.** Weekly training schedule, divided into 2018/2019 season, 2019/2020 season (COVID-19), and lockdown period.


The number and duration of training sessions throughout the study were the same for all players. The duration of each training session was 90 min for all players. All training sessions were preceded by a standardized warm-up of 5–15 min.

For weeks where only one match was played on Sunday, the training protocol included sessions on Tuesdays, Wednesdays, Thursdays, Fridays, and Saturdays. Monday was instead set as a day off.

In the weeks in which two games were played (e.g., Wednesday and Sunday), the protocol included training sessions on Mondays, Tuesdays, Thursdays, and Fridays.

During training, football players reach average heart rate values (HR) of 146 beats/min corresponding to approximately 87–97% of the maximum heart rate.

Home-based training during lockdown was performed to maintain players' physical performance levels by programs individually provided by the team's coaches.

#### *2.4. Anthropometric Evaluation*

The height of the participants was measured with a Seca stadiometer to the nearest 0.1 cm, while the weight was measured with an Omron balance to the nearest 0.5 kg. Anthropometric-determined measurements included: height (m), weight (kg), body mass index (BMI (kg/m<sup>2</sup> ) = weight/height<sup>2</sup> ), percentage of body fat (BFP, %), and fat-free mass (FFM, kg). In particular, the percentage of body fat was estimated, following the measurement of three skin folds (chest, abdomen, and quadriceps) with a GIMA mechanical skinfold meter, using the formula developed by Jackson–Pollock [8]. The percentage of fat-free mass was measured using a bioimpedance analyzer (BIA-AKERN EFG). At each timepoint, anthropometric assessments were also carried out in the early morning, always before each workout.

#### *2.5. Blood Parameters*

Venous blood samples were taken following fasting in the early morning (8.00 am) following a day off. Blood (10 mL) was collected in vacutainer tubes, using an anticoagulant. The freshly drawn blood was immediately centrifuged at 3000 r·min<sup>1</sup> (825 g) for 10 min to remove the plasma. Analyses were performed using a coulter blood counter (Model

S-plus II, Coulter Electronics inc., Hialeah, FL, USA) and yielded values for hematocrit (Ht), hemoglobin (Hb), red blood cells (RBC), serum iron, and ferritin.

Percentage changes in plasma volume during the study period were assessed by the method described by Saidi et al. (2019) [9].

For the total 25(OH)D measurement, an Abbott Architect 25-OH D reagent on an i2000 Architect analyzer (Abbott Laboratories, Abbott Park, IL 60064, USA) was used with a chemiluminescent competitive delayed phase immunoassay (Chemiflex) standardized according to the NIST SRM 2972 (National Institute of Standard and Technology Standard Reference Material 2972). As previously reported [10], serum testosterone and cortisol were analyzed by the IMMULITE 2000 Immunoassay System (Medical Systems).

Intra- and inter-assay coefficients of variance for cortisol were 4.6% and 7.6%, respectively. The intra- and inter-assay coefficients of variance for testosterone were 3.7% and 5.6%, respectively. Serum testosterone and cortisol reference ranges were 10–75 ng/dL and 7–25 g/dL, respectively.

Normal iron storage (ferritin > 110 µg L−<sup>1</sup> , Hb > 14 g dL−<sup>1</sup> ), iron depletion (ferritin < 30 µg l−<sup>1</sup> , Hb > 14 g dL−<sup>1</sup> ), iron deficiency (ferritin < 12 µg L−<sup>1</sup> , Hb > 14 g dL−<sup>1</sup> ), and iron deficiency anemia (ferritin < 10 µg L−<sup>1</sup> , Hb < 14 g dL−<sup>1</sup> ) were defined according to population references for iron status measures in males, 24.25.

#### *2.6. Physical Performances*

To obtain information regarding the physiological status of youth players, we used tests that have been frequently used in similar studies: countermovement jumps test (CMJ) and Mognoni test.

To minimize any effects of diurnal variation, the three testing sessions were conducted within 2 h of the same time of the day.

Then, each player performed maximal CMJ on the contact platform from a standing position and with the hands on the hips. At the start, the subjects made a preparatory movement: from the extended leg position, they made a rapid bending of the knees until reaching the 90◦ angle, keeping the heels in contact with the ground and the trunk erect. After the jump, keeping the hands on the hips, the fall was performed with the knees extended, on the tip of the toes with subsequent cushioning to avoid trauma.

The ground reaction force generated during these vertical jumps was estimated with an ergo jump (Opto Jump Microgate, Bolzano, Italy). The height of the jump (cm) was the maximal height reached during the flight phase.

The Mognoni test is a simple method to evaluate the speed at which the athlete reaches OBLA (Onset of Blood Lactate Accumulation).

The test execution protocol provides that the subjects must travel 1350 m in 6 min, maintaining a constant speed of 13.5 km/h [11]. In the field version of this test used in the present study, pins were placed in the path, at regular 50 m intervals, causing players to hear a sound that informed them when the transition at each pin should take place. Immediately upon completion of the 6 min run, the capillary blood lactate concentration was measured from the earlobe with a portable lactate analyzer (Lactate Plus; Nova Biomedical, Waltham, MA, USA): the lower the value of lactate after the test, the better the aerobic fitness level [11].

#### *2.7. Statistical Analysis*

Results obtained were stored in Microsoft Office Excel 2016 and statistically analyzed by GraphPad PRISM 5 software (GraphPad Software). All variables used in this study were checked for the normality of distribution before the analyses (Kolmogorov–Smirnov tests). Student's paired *t*-test and Spearman correlation were used. *p* < 0.05 was accepted as a level of statistical significance. All data obtained from the study were expressed as mean ± standard deviation.

#### **3. Results**

#### *3.1. Anthropometric Characteristics of Football Players*

The anthropometric characteristics of the players (weight, height, body mass index, percentage of body fat BFP, and fat-free mass FFM) measured during the two seasons analyzed are shown in Table 2.

**Table 2.** Anthropometric characteristics of football players during different times during the 2018/2019 and 2019/2020 seasons.


\* Statistical difference from T0. # Statistical difference from T1. § Statistical difference from T2. ◦ Statistical difference from T3. <sup>A</sup> Statistical difference from 2018/2019 season.

The reported values show some statistical differences (*p* < 0.05) between the various points of the season. In addition, in the 2019/2020 season (COVID-19), there were higher weight values (*p* = 0.01) and BMIs (*p* = 0.03) and lower percentages of FFM (*p* = 0.01), due to the forced stop period.

Exercise is known to affect hematological variables: some studies have reported a stimulation of erythrocytosis with a consequent reduction in the values of HT and HB in athletes [12]. However, it is also known that excessive physical exercise induces the physical destruction of red blood cells, also causing decreases in HT and HB [13,14]. Thus, we monitored these hematological variables, and the values measured were within the physiological ranges (4.2–5.6 <sup>×</sup> <sup>10</sup>12/L for RBC; 13.0–17.5 g/dL for HB; 37–54% for HT). However, decrements in HB, RBC, and HT during both seasons (Figure 2), mostly significant in the 2018/2019 regular season were observed (Table 3).

#### *3.2. Vitamin D–Iron–Ferritin*

In both seasons, vitamin D significantly decreased in T3 (<30 ng/dL; *p* < 0.05); in the 2019/2020 season (COVID-19), only vitamin D levels were also low in T2 (Figure 3 and Table 4). Cross-sectional studies indicated an association between low vitamin D concentration and low iron state [15], but we found no association (*p* > 0.05 by Spearman's rank correlation). Ferritin levels decreased in T1 and T2 during the 2019/2020; however, a major decrement was observed in T2-T4 periods during the 2018/2019 season. Iron concentrations decreased in T3 in both seasons (Figure 3 and Table 4).

#### *3.3. CK–Cortisol–Testosterone–T/C Ratio*

During both seasons, CK increased starting from T2 to the end of the seasons (Figure 4A,B and Table 5). As reported previously [10], cortisol concentrations increased significantly in T2 and T3 during the 2018/2019 season; nevertheless, such an increment was not found during the 2019/2020 (COVID-19) period (Figure 4C,D and Table 5).

**Figure 2.** The effects of training on serum red blood cells (**A**,**B**), hematocrit (HT) percentage (**C**,**D**), and hemoglobin concentration (**E**,**F**) in football players during the 2018/2019 and 2019/2020 seasons. Box and whiskers representation of red blood cells, hemoglobin concentration, and hematocrit (HT) percentage evaluated five times (T0, T1, T2, T3, and T4) during the seasons. In this representation, the central box covers the middle 50% of the data values, between the upper and lower quartiles. The bars extend out to the extremes, while the central line is at the median. *p*-values were obtained by Student's paired *t*-test between each timepoint and T0. \* *p* < 0.05.


**Table 3.** Differences (∆) in red blood cells and hemoglobin concentration, hematocrit percentage, and plasma volume between each timepoint in football players for 2018/2019 and 2019/2020. *p*-values < 0.05 obtained by *t*-test show statistical differences in ∆ values.

**Table 4.** Differences (∆) in ferritin, iron, and vitamin D concentration between each timepoint in football players for 2018/2019 and 2019/2020. *p*-values < 0.05 obtained by *t*-test show statistical differences in ∆ values.


**Figure 3.** The effects of training on serum ferritin (**A**,**B**), iron (**C**,**D**), and vitamin D (**E**,**F**), concentration in football players during the 2018/2019 and 2019/2020 seasons. Box and whiskers representation of serum ferritin, iron, and vitamin D evaluated five times (T0, T1, T2, T3, and T4) during the seasons. In this representation, the central box covers the middle 50% of the data values, between the upper and lower quartiles. The bars extend out to the extremes, while the central line is at the median. *p*-values were obtained by Student's paired *t*-test between each timepoint and T0. \* *p* < 0.05.

**Figure 4.** The effects of training on serum CK (**A**,**B**), cortisol (**C**,**D**), testosterone concentration (**E**,**F**), and T/C ratio (**G**,**H**), in football players during the 2018/2019 and 2019/2020 seasons. Box and whiskers representation of serum CK, cortisol, and testosterone concentration and T/C ratio evaluated five times (T0, T1, T2, T3, and T4) during the seasons. In this representation, the central box covers the middle 50% of the data values, between the upper and lower quartiles. The bars extend out to the extremes, while the central line is at the median. *p*-values were obtained by Student's paired *t*-test between each timepoint and T0. \* *p* < 0.05.


**Table 5.** Differences (∆) in CK, cortisol, and testosterone concentration and T/C ratio between each timepoint in football players all for 2018/2019 and 2019/2020. *p*-values < 0.05 obtained by *t*-test show statistical differences in ∆ values.

Testosterone concentration increased in both seasons from T0 to T3 and, at the end of the season, it decreased toward initial levels; although such a decrement was statistically significant in both seasons (*p* < 0.05), it was higher in the 2018/2019 season (Figure 4E,F).

The testosterone to cortisol ratio (T/C) increased only in T1 of the 2018/2019 season; in all other periods, it did not show significant changes (Figure 4G,H).

#### *3.4. Electrolytes*

The clinical chemistries shown in Table 6 represent common clinical chemistries used to monitor clinical aspects of electrolytes and metabolism.

**Table 6.** Differences (∆) in electrolytes concentration between each timepoint in football players for 2018/2019 and 2019/2020 seasons. *p*-values < 0.05 obtained by *t*-test show statistical differences in ∆ values.

