The anthropometric profile and body composition of youth soccer goalkeepers after the COVID-19 pandemic, according to the maturity offset

In July 2022, forty-two young male goalkeepers (age 14.15 ± 2.99) were recruited from different soccer teams during a training camp dedicated to youth goalkeepers. Participants were classified into three subgroups: pre-PHV, circa-PHV, and post-PHV. All participants were involved in soccer as goalkeepers for at least one year and performed 3–5 training sessions per week (n = 3.6 ± 1.08). The study was conducted in accordance with the Declaration of Helsinki. All individuals gave their assent to voluntarily take part in the study and delivered written informed consent to parents for permission to participate, while the technical staff approved the conduct of the study. Nonetheless, the present data arose as a condition of the monitoring procedures performed during the training camp. Because of the retrospective nature of the analyses without interfering with the training routine, usual appropriate ethics committee clearance was not required [14]; nevertheless, all physical performance data were anonymized before analyses to ensure player confidentiality.

All the anthropometric measurements were performed by a certified specialist (i.e., a level 1 certification of the International Society of the advancement of Kinanthropometry, ISAK). Participants wore light clothing and had fasted for at least 12 h before the assessments. Height and sitting height were measured to the nearest 0.1 cm using a stadiometer (GPM, Zurich, Switzerland) and a 30 × 40 × 50 cm high wooden anthropometric box and sitting height to height ratio (SH/Hr) was obtained. Body weight was measured to the nearest 0.1 kg using a stadiometer with a balance-beam scale (SECA 770, Seca, Hamburg, Germany). Waist circumference (WC) was measured to the nearest 0.1 cm using a Cescorf anthropometric tape (Cescorf, Porto Alegre, Brazil). The percentage of fat mass (%FM) was calculated according to Brambilla [15] and fat mass in kilograms (FM) was derived. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height, expressed in meters. Thus, the BMI > 85th percentile was used to classify the participants as normal weight and overweight/obese [16]. Fat mass index (FMI) and fat-free mass index (FFMI) were calculated as fat mass and fat-free mass weight in kilograms, respectively, divided by the square of height, expressed in meters. According to McCarthy’s waist circumference cut-points, participants with a WC > 90th percentile were considered to have abdominal obesity [17]. Waist-to-height ratio (W/Hr) > 0.5 identified those participants at higher cardiometabolic risk [18].

The maturity offset was defined according to the Mirwald equation based on the distance from peak height velocity (PHV) in years (YPHV), and derived from height, sitting height, and leg length as somatic dimension, the chronological age, and their interaction [19]. Thus, participants were classified into three subgroups according to their YPHV: pre-PHV (offset <  − 1 year), circa-PHV (≤ ± 1 year), and post-PHV (offset >  + 1 year), as a previous study reported [8]. Considering that the YPHV may differ significantly between participants, they were further divided into three subgroups according to maturity offset velocity (early, average, late maturer). The predicted adult height (PAH) and the percentage of predicted adult height (%PAH) were estimated according to the Sherar method [20]. Finally, to evidence the occurrence of a selection bias based on the relative age effect, the date of birth was used to define the quartile of birth and months from one to six and from seven to twelve, respectively, were merged into two categories (first/second vs. third/fourth) [21].

Regarding the previously evaluated sample, information on sample recruitment, anthropometry, body composition and maturity assessment can be found elsewhere [8] and we followed the same procedure, as reported in the present study.

The Shapiro–Wilk test was performed to check the normality of the data. When data were normally distributed, the analysis of variance (ANOVA) was used to determine the differences between the maturity status subgroup (defined as fixed factor) in body composition and anthropometric parameters (defined as dependent variables), and partial eta squared (η²_p) was calculated to indicate the effect-size (small = 0.01, medium = 0.06; large = 0.14). When the analysis of variance (ANOVA) showed significant results, Tukey’s post hoc test was used to confirm where the difference occurred. Pearson’s correlation coefficient was used to determine the extent of correlation between YPHV and anthropometric measures. The magnitude of correlations was considered as: r = 0.00–0.09, negligible; r = 0.10–0.39, weak; r = 0.40–0.69, moderate; r = 0.70–0.89, strong; r = 0.90–1.00, very strong [22]. When data were not normally distributed, the Kruskal–Wallis with Dunn’s post-hoc test was used instead of ANOVA and Tukey’s post hoc, respectively, and Spearman’s Rho was used instead of Pearson’s. Considering we did not meet the conditions to run the Pearson chi-square (χ²) test for independence to check the association between maturity status and the identified categories, the number of individuals and frequencies were calculated, as well as the odds ratio for each pair of subgroups. The comparison against previously published values (July 2019 Vs. July 2022) was computed using the One sample t test or the Wilcoxon signed-rank test when data were not normally distributed. Descriptive data are presented as mean ± standard deviations, while categorical data as frequency. The statistical significance was set at < 0.05.

Table 1 shows descriptive statistics for anthropometric and body composition variables and subgroup comparisons for maturity offset. Age, height, weight, sitting height, the percentage of predicted adult height (%PAH), and %FM were significantly different between the three subgroups. Sitting height to height ratio (SH/Hr), BMI, FM, and FFMI significantly differed between pre-PHV vs. post-PHV subgroups. The WC was significantly different between pre-PHV vs. circa-PHV and post-PHV subgroups. The FMI was significantly different between circa-PHV vs. post-PHV and pre-PHV vs. post-PHV subgroups. Finally, the predicted adult height (PAH) and the W/Hr were not different between subgroups.

Table 2 shows the correlation between YPHV and the anthropometric and body composition values. Briefly, a positive, very strong correlation was found with height, sitting height, and %PAH; A positive, strong correlation was found with weight; a positive, moderate correlation was found with WC, BMI, and FFMI; a positive, weak correlation was found with SH/Hr and FM; A negative, very strong and a negative, moderate correlation were found with %FM and FMI, respectively. Finally, no correlations were found with W/Hr and PAH.

Figure 1 shows the number of individuals and frequencies for each category while Table 3 shows the odds ratio and 95% confidence interval for each pair of subgroups.

In previously published data, there were twelve pre-PHV, fourteen circa-PHV, and sixteen post-PHV goalkeepers [8]. In the present study, we found sixteen pre-PHV, seven circa-PHV, and nineteen post-PHV goalkeepers. We also found lower SH/Hr and W/Hr in the pre-PHV subgroup (W = 8.0, p = 0.007; W = 17.0, p = 0.014, respectively) and higher YPHV, age, %PAH, and WC in the post-PHV subgroup (W = 160, p = 0.042; t = 2.768 (19), p = 0.012; W = 185, p = 0.003; t = 1.911 (19), p = 0.036, respectively) compared to previously published values. Table 4 shows the comparison between the current and the previous sample, while the descriptive data of the anthropometric and maturity characteristics of the latter were published elsewhere [8].