Patellar and Achilles Tendon Thickness Differences among Athletes with Different Numbers of Meals per Day: A Cross-Sectional Study

objective of this study is to analyse differences in the thickness of the patellar (PT) and Achilles tendons (AT) among athletes with different number of meals per day. The design is a cross-sectional, observational study. A total of thirty-six male athletes (with mean age groups ranging from 31 to 40) were recruited and divided into three groups based on the number of daily meals they had (3, 4 or 5 meals). PT and AT were assessed by ultrasound. There were statistically signi cant differences in PT when comparing groups 1 and 3, at both longitudinal (p< 0.03) and transversal (p< 0.002)planes. There were no differences when comparing groups 1 and 2 or groups 2 and 3. There was a negative correlation between the number of meals per day and tendon thicknesses in both PT (longitudinal plane: r = 0.384;p= 0.02/transversal plane: r

cant differences in PT when comparing groups 1 and 3, at both longitudinal (p< 0.03) and transversal (p< 0.002)planes. There were no differences when comparing groups 1 and 2 or groups 2 and 3. There was a negative correlation between the number of meals per day and tendon thicknesses in both PT (longitudinal plane: r = 0.384;p= 0.02/transversal plane: r = 0.406;p= 0.01) and AT (transversal plane: r = 0.386;p= 0.02). In conclusion, there were patellar tendon thickness differences between participants and the number of daily meals could play a key role in tendon thickness, healing and performance. Keywords: Achilles tendon; patellar tendon; ultrasonography; runners; metabolic health; metabolic education 1. Introduction The Achilles (AT) and patellar (PT) tendons are commonly altered among athletes, and particularly in runners [1]. Running is one of the most popular sports activities and therefore the number of studies focus on runners and related injuries are increasing [2,3]. The overall incidence of injury to runners ranges from 18.2% to 94.4% [4]. The Achilles tendinitis present has an incidence of9.1–10.9% in runners, and patellar tendinitis has an incidence of 5.5–22.7%, being among the most frequent injuries in sports [4,5]. It is thought that high loading rates, when combined with suboptimal technique, may pro- duce aberrant strain to lower extremity tendons, contributing to microtrauma that may trigger tendinopathies [6,7]. Furthermore, tendinopathy and tendon rupture are asso- ciated with obesity and associated metabolic conditions, such as insulin resistance and dyslipidaemia [8]. Ultrasound imaging (US) is frequently used to assess tendon morphology, and it is considered to be as reliable as magnetic resonance imaging [9]. The normal ultrasound characteristic of PT and AT is well established in a healthy population, but not in athletes; nevertheless, several abnormalities have been shown to be common in both populations [6]. Tendon changes can be diagnosed before they become symptomatic with US, possibly being re ective of adaptations or maladaptations [10,11]. Additionally, identifying other Int. J. Environ. Res. Public Health2022,19, 2468.

to be common in both populations [6]. Tendon changes can be diagnosed before they become symptomatic with US, possibly being re ective of adaptations or maladaptations [10,11]. Additionally, identifying other Int. J. Environ. Res. Public Health2022,19, 2468.

Int. J. Environ. Res. Public Health2022,19, 2468 2 of 10 factors that may produce tendon changes would help in detecting presymptomatic athletes and would enable the introduction of modi cations to their training regimen [6,12]. One of these factors could be the number of meals per day in runners. In this regard, there is evi- dence showing that high levels of glucose provoke higher insulin levels and a consecutive glucose disbalance. Hypoglycaemia can induce a rebound effect on the level of gluco- neogenesis induced by glucocorticoids, which in excess may affect tendon structure [8]. Glucocorticoid release is regulated by the hypothalamic–pituitary–adrenal (HPA) axis, and its function can be summarized in two ways. First, an acute activation of the HPA axis generates a glucocorticoid peak, which increases gluconeogenesis in the liver and inhibits insulin production [13]. If the subject presents proper energy reserve and distribution, they will be able to respond adequately to the stress factors which caused the stress response and activation of the sympathetic nervous system and the HPA axis [14]. Second, when the HPA axis loses its rhythm and is chronically activated, cortisol release is constant, and this could provoke maladaptive responses leading to insulin resistance, a condition that could cause an increase in and the acceleration of the generation of cross links [15]. The opposite is also a possible cause of collagen tissue atrophy and possible injuries. Insulin resistance and HYPER-insulinemia induce the overexpression of undercarboxylated osteocalcin and loss of bone density and collagen tissue damage by activation of RANKL in bones and increased uptake of glucose in collagen tissue, respectively [16]. The increase in intracellular glucose activates the polyol pathway that induces the production of sorbitol and fructose under severe oxidative stress [17]. The latter could be responsible for mitochondrial damage and tissue breakdown [17]. The composition of food has been extensively researched and understood. However, meal frequency and timing are important aspects of nutrition, and how they impact health, performance and healing needs to be further investigated [18]. To the best of our knowledge, there have not been studies investigating to what extent and if

be responsible for mitochondrial damage and tissue breakdown [17]. The composition of food has been extensively researched and understood. However, meal frequency and timing are important aspects of nutrition, and how they impact health, performance and healing needs to be further investigated [18]. To the best of our knowledge, there have not been studies investigating to what extent and if the number of meals per day correlate with PT and AT thicknesses in runners. Our hypothesis is that differences in the number of meals per day could explain morphological changes in tendons, being the trigger factor which leads to a metabolic disorder. Hence, the aim of the study was to analyse differences in PT and AT thicknesses between groups with different meal frequencies (3,4 or 5). The second outcome is to study if meal frequency could be related with thickness of the PT and AT. 2. Material and Methods This study was conducted in full accordance with the provisions of the Declara- tion of Helsinki regarding ethical principles for medical research involving human sub- jects, and was approved by the Medical Ethics Committee of the University of Malaga (CEUMA122017H), Spain. This was a cross-sectional, observational study. We used the STROBE-statement, which was developed to scienti cally strengthen the impact of observational studies in epidemiology [19]. A total of thirty-six male athletes were recruited from different athletic clubs and volunteered to participate in our study. All subjects were at least 18 years old and were able to follow the study instructions. Informed consent was obtained in every case. The inclusion criteria were: (1) The runners had run at least 50 km per week during the previous six consecutive months, (2) at a pace of no more than six minutes per km [20]; The exclusion criteria were: (1) osteo-degenerative disease, (2) metabolic disease, (3) neurological problems or surgical intervention in the lower limb, (4) processes of an infectious, (5) cancer, (6) cogni- tive impairment, (7) musculoskeletal injuries of the lower limbs in the last three months, and (8) use of corticosteroids.

[20]; The exclusion criteria were: (1) osteo-degenerative disease, (2) metabolic disease, (3) neurological problems or surgical intervention in the lower limb, (4) processes of an infectious, (5) cancer, (6) cogni- tive impairment, (7) musculoskeletal injuries of the lower limbs in the last three months, and (8) use of corticosteroids.

Int. J. Environ. Res. Public Health2022,19, 2468 3 of 10 2.1. Sample Size The sample size was determined by application of the EPIDAT program, using the criterion of AT with a detectable difference of the mean of 0.818 to evaluate the statistical power [21]. The study was designed to detect changes exceeding 0.8 (large effect size) for a variation of the sample according to the above classi cation, with a type I error of 0.05 and a type II error of 0.2. This calculation produced a necessary sample size of 18 subjects in each group. 2.2. Study Protocol The subjects were evaluated at the podiatric healthcare teaching unit of the University. In a single session, the relevant study data were obtained—age, BMI, and number of meals per day, together with the ultrasound measurements, performed by an expert in musculoskeletal ultrasound imaging with 8 years of experience and a second examiner assistant, who placed a shield on the ultrasound screen to assure the blinding, taking all the data from the screen. The participants were divided into three groups, based on the number of meals per day: Group 1: 3 meals; Group 2: 4 meals; Group 3: 5 meals. The intraclass correlation coef cient of ultrasound examiner was used with 10 participants to evaluate the reproducibility of measurement of the thickness and cross-sectional area of the AT and the thickness of the AT and PT, with a sample of ten subjects measured at baseline and after 24 h. They were estimated by calculating the ICC for the second author, using a one-way random effect model, and found to be excellent (ICC 0.94: 95% CI: 0.90–0.96) at 0, and 0.87 (95% CI: 0.80–0.92) at 60. 2.3. Ultrasound Measurements A diagnostic ultrasound unit, Sonosite M-turbo (GE Healthcare, Wauwatosa, WI, USA) with a dynamic range of up to 165 dB, was used. In addition, a 6–13 MHz linear transducer with 196 piezoelectric crystals tted with a speci c ultrasound system, SonoMB ® multi- beam imaging, was employed to increase resolution and improve the visualisation of subtle physiological and tissue differences. These images were captured as

M-turbo (GE Healthcare, Wauwatosa, WI, USA) with a dynamic range of up to 165 dB, was used. In addition, a 6–13 MHz linear transducer with 196 piezoelectric crystals tted with a speci c ultrasound system, SonoMB ® multi- beam imaging, was employed to increase resolution and improve the visualisation of subtle physiological and tissue differences. These images were captured as grey-scale images with 256 levels. All participants received standardised ultrasonography in the dominant lower limb. The ultrasound images were obtained by a single examiner, a quali ed physiotherapist with 9 years of experience with musculoskeletal ultrasound imaging. The ultrasound examiner was blind to the characteristics of the participants. Simultaneously, the data were recorded by a research assistant, who was blind to the characteristics. Each participant was issued an identi cation number, which was the only information provided to the examiners. All ultrasound measures are expressed in millimetres [22]. 2.4. Achilles Tendon The participants adopted a prone position, and the AT was scanned both longitudinally and transversely, with the transducer placed on the AT. The thickness of the tendon was measured at the level of the medial malleolus, in order to standardise the measurements, as was shown in previous studies [23]. The thickness of the AT was measured by its maximum anteroposterior diameter (Figure) [22].

Int. J. Environ. Res. Public Health2022,19, 2468 4 of 10Int. J. Environ. Res. Public Health 2022, 19, x FOR PEER REVIEW 4 of 10 Figure 1. AT measurement in longitudinal (left) and transversal (right) axis. 2.5. Patellar tendon The participants adopted a supine position. The placement of the transductor and subsequently the measurements were taken 2 cm distal to the end of the patella, and both longitudinal andtransversal axes were measured [24]. The thickness of the PT was measured by its maximum anteroposterior diameter (Figure 2) [22]. Figure 2. PT measurement in longitudinal (left) and transversal (right) axis. 2.6. Statistical Analysis Normality for all US variables was explored using the Shapiro–Wilk test for the three groups of participants. To determine between‐groups differences for all the out‐ come measurements, one‐way ANOVA test was calculated with Tukey post‐hoc estima‐ tion [22]. A p‐value less than 0.05 was considered statistically significant. To calculate the intra‐rater reliability of all the US variables, three measurements of each one were col‐ lected, and a two‐way mixed (3,1), consistency, intraclass correlation coefficient (ICC) was then calculated. A reliability coefficient < 0.50 was an indication of “poor” reliability; “moderate” between 0.50 and 0.75; “good” between 0.76 and 0.90; and “excellent” over 0.90 [25]. The standard error of measurement (SEM) and the minimal detectable change at 95% confidence of interval (MCD95) were also obtained. To determine the correlations between tendon thicknesses (PT and AT) and the number of meals per day, a Pearson correlation coefficient was calculated for a normal data distribution, or a Spearman’s co‐ efficient in the case of absence of normality. Weak correlation was defined as values between 0.3 and 0.5; between 0.5 and 0.7 correlation was considered moderate; and strong was considered greater than 0.7. For all the US calculations, three measurements were taken by the examiner, and an average of three was used for the statistical analysis for each angle. An interval of 1 min was provided between measures, and the patient was encouraged to move freely. Pa‐ tients were then repositioned and the second and third sets of measurements were suc‐ cessively taken.

For all the US calculations, three measurements were taken by the examiner, and an average of three was used for the statistical analysis for each angle. An interval of 1 min was provided between measures, and the patient was encouraged to move freely. Pa‐ tients were then repositioned and the second and third sets of measurements were suc‐ cessively taken. 3.Results Figure 1.AT measurement in longitudinal (left) and transversal (right) axis. 2.5. Patellar Tendon The participants adopted a supine position. The placement of the transductor and subsequently the measurements were taken 2 cm distal to the end of the patella, and both longitudinal and transversal axes were measured [24]. The thickness of the PT was measured by its maximum anteroposterior diameter (Figure) [22].Int. J. Environ. Res. Public Health 2022, 19, x FOR PEER REVIEW 4 of 10 Figure 1. AT measurement in longitudinal (left) and transversal (right) axis. 2.5. Patellar tendon The participants adopted a supine position. The placement of the transductor and subsequently the measurements were taken 2 cm distal to the end of the patella, and both longitudinal andtransversal axes were measured [24]. The thickness of the PT was measured by its maximum anteroposterior diameter (Figure 2) [22]. Figure 2. PT measurement in longitudinal (left) and transversal (right) axis. 2.6. Statistical Analysis Normality for all US variables was explored using the Shapiro–Wilk test for the three groups of participants. To determine between‐groups differences for all the out‐ come measurements, one‐way ANOVA test was calculated with Tukey post‐hoc estima‐ tion [22]. A p‐value less than 0.05 was considered statistically significant. To calculate the intra‐rater reliability of all the US variables, three measurements of each one were col‐ lected, and a two‐way mixed (3,1), consistency, intraclass correlation coefficient (ICC) was then calculated. A reliability coefficient < 0.50 was an indication of “poor” reliability; “moderate” between 0.50 and 0.75; “good” between 0.76 and 0.90; and “excellent” over 0.90 [25]. The standard error of measurement (SEM) and the minimal detectable change at 95% confidence of interval (MCD95) were also obtained. To determine the correlations between tendon thicknesses (PT and AT) and the

then calculated. A reliability coefficient < 0.50 was an indication of “poor” reliability; “moderate” between 0.50 and 0.75; “good” between 0.76 and 0.90; and “excellent” over 0.90 [25]. The standard error of measurement (SEM) and the minimal detectable change at 95% confidence of interval (MCD95) were also obtained. To determine the correlations between tendon thicknesses (PT and AT) and the number of meals per day, a Pearson correlation coefficient was calculated for a normal data distribution, or a Spearman’s co‐ efficient in the case of absence of normality. Weak correlation was defined as values between 0.3 and 0.5; between 0.5 and 0.7 correlation was considered moderate; and strong was considered greater than 0.7. For all the US calculations, three measurements were taken by the examiner, and an average of three was used for the statistical analysis for each angle. An interval of 1 min was provided between measures, and the patient was encouraged to move freely. Pa‐ tients were then repositioned and the second and third sets of measurements were suc‐ cessively taken. 3.Results Figure 2.PT measurement in longitudinal (left) and transversal (right) axis. 2.6. Statistical Analysis Normality for all US variables was explored using the Shapiro–Wilk test for the three groups of participants. To determine between-groups differences for all the outcome mea- surements, one-way ANOVA test was calculated with Tukey post-hoc estimation [22]. A p-value less than 0.05 was considered statistically signi cant. To calculate the intra-rater reliability of all the US variables, three measurements of each one were collected, and a two-way mixed (3,1), consistency, intraclass correlation coef cient (ICC) was then calcu- lated. A reliability coef cient < 0.50 was an indication of “poor” reliability; “moderate” between 0.50 and 0.75; “good” between 0.76 and 0.90; and “excellent” over 0.90 [25]. The standard error of measurement (SEM) and the minimal detectable change at 95% con dence of interval (MCD95) were also obtained. To determine the correlations between tendon thicknesses (PT and AT) and the number of meals per day, a Pearson correlation coef cient was calculated for a normal data distribution, or a Spearman's coef cient in the

over 0.90 [25]. The standard error of measurement (SEM) and the minimal detectable change at 95% con dence of interval (MCD95) were also obtained. To determine the correlations between tendon thicknesses (PT and AT) and the number of meals per day, a Pearson correlation coef cient was calculated for a normal data distribution, or a Spearman's coef cient in the case of absence of normality. Weak correlation was de ned as values between 0.3 and 0.5; be- tween 0.5 and 0.7 correlation was considered moderate; and strong was considered greater than 0.7. For all the US calculations, three measurements were taken by the examiner, and an average of three was used for the statistical analysis for each angle. An interval of 1 min was provided between measures, and the patient was encouraged to move freely. Patients were then repositioned and the second and third sets of measurements were successively taken. 3. Results The recruitment included a total of 42 participants, with six participants who did not t the inclusion criteria because of the presence of pain. A nal number of 36 participants were enrolled in the study (see ow Figure).

This study investigates tendon thickness differences related to meal frequency in athletes.