Sagittal spinal morphotype assessment in 8 to 15 years old Inline Hockey players

Background Physiological sagittal spinal curvatures play an important role in health and performance in sports. For that reason, several scientific studies have assessed spinal morphology in young athletes. However, to our knowledge, no study has assessed the implications of Inline Hockey (IH) practice on sagittal integrative spinal morphotype in adolescent players. Objectives The aims of the present study were to describe habitual sagittal spinal posture in young federated IH players and its relationship with training load and to determine the sagittal integrative spinal morphotype in these players. Methods An observational analysis was developed to describe the sagittal spinal morphotype in young federated IH players. A total of 74 IH players from the Technification Plan organized by the Skating Federation of the Valencian Community (aged from 8 to 15 years) participated in the study. Thoracic and lumbar curvatures of the spine were measured in a relaxed standing position (SP), in a slump sitting position (SSP) and in maximum flexion of the trunk (MFT) to determine the “Sagittal Integrative Morphotype” of all players. An unilevel inclinometer was used to quantify the sagittal spinal curvatures. The Hip Joint Angle test was used to quantify the Lumbo-Horizontal angle in flexion (L-H fx) of all participants with a goniometer. Results When thoracic curvature was analyzed according to normality references, it was found that 64.9% of IH players had thoracic hyperkyphosis in a SSP, while 60.8% and 74.3% of players were classified as normal in a SP and in MFT, respectively. As for the lumbar curve, 89.2% in a SP and 55.4% in MFT were normal, whereas 68.9% of IH players presented lumbar hyperkyphosis in a SSP. Regarding the “Sagittal Integrative Morphotype,” only 17.6% of players were classified as “Normal” in the three measured positions for the thoracic curve, while 37.8% had “Thoracic Hyperkyphosis” and 41.8% presented “Functional Thoracic Hyperkyphosis.” As for the “Sagittal Integrative Lumbar Morphotype,” only 23% of athletes had a normal curve in the three positions, whereas 66.2% presented “Functional Lumbar Hyperkyphosis.” When the L-H fx was evaluated, the results showed that only 16.2% of the athletes were classified as normal. Conclusions Federative IH practice seems to cause specific adaptations in spinal sagittal morphotype. Taking into account the “Sagittal Integrative Morphotype” only 17.6% IH players presented “Normal Morphotype” with a normal thoracic kyphosis in the three measured positions, while only 23% IH players presented “Normal Morphotype” with a normal lumbar curvature in the three assessed positions. Furthermore, only 16.2% of IH players showed normal pelvic tilt. Exercise programs to prevent or rehabilitate these imbalances in young IH players are needed.


INTRODUCTION
Physiological sagittal spinal curvatures play an important role in health and performance in sports since the distribution of mechanical strains greatly affects the structures of the spine and can influence athletes' stability as well as result in overuse injuries to the spine . Hence, these curvatures should be neither reduced nor excessive in order to maintain a physiological, harmonic and balanced posture. In this sense, to have sagittal spinal curvatures within the normal ranges could favor the athlete's trunk mobility as well as improve a player' stability due to the lower center of gravity and the better distribution of the load (Ackland, Elliott & Bloomfield, 2009).
For those reasons, several experts in the analysis of the locomotor system recommend the assessment of sagittal spinal curvatures to describe the sagittal morphotype of spine in sports (Sainz de Baranda, Rodríguez-García & Santonja, 2010;Sainz de Baranda, Santonja & Rodríguez-Iniesta, 2009;Santonja & Pastor, 2000;Sanz-Mengibar, Sainz-de-Baranda & Santonja-Medina, 2018). In fact, this knowledge could contribute to the development of more effective preventive interventions to be adopted by a multidisciplinary professional team. Specifically, for the assessment of the sagittal plane of the spine, it is recommended to evaluate the thoracic and lumbar curves in a SP, in a SSP and in MFT to finally establish the "Sagittal Integrative Morphotype" of the spine (Santonja, 1996).
In addition, as the spine of an adolescent is in a maturation period, it shows changes in posture and balance of its curvatures during growth and it is more vulnerable (Sainz de . Thus, sports professionals should be aware of the loads and overloads inherent in sport and training and its impact on the young athlete's spine (Sainz de Baranda, Rodríguez-García & Santonja, 2010).
The participation in Inline roller hockey or IH among adolescents has increased in the past few years thanks to the popularity of inline-skating. Since its introduction in 2000 in Spain, IH has been one of the fastest growing sports in the different federative categories. Over the 2005/06 and 2016/17 seasons, there were 3,100 and 5,234 licenses, respectively, which is an increase of almost 60% in the number of licenses within the last 10 years (Real Federación Española de Patinaje, 2018).
In order to find out how IH can affect young players' spine as well as to help sport professionals to plan specific preventive interventions, the current investigation was carried out. The aims of the present study were: (1) to describe habitual sagittal spinal posture in young federated IH players and its relationship to training load, and (2) to determine the "Sagittal Integrative Spinal Morphotype" in these players. Our hypothesis is that there is a special adaptation of the spine to the specific requirements of IH in young players.

MATERIALS AND METHODS
In order to confirm or rule out our hypothesis, an observational analysis was developed to describe the sagittal spinal morphotype in young federated IH players.
The study was approved by the Ethics and Research Committee of the University of Murcia (Spain) (ID: 1702/2017).

Participants
The subject population was selected through a convenience sample from the Technification Plan organized by the Skating Federation of the Valencian Community in the season 2016-2017, in which the best IH players of the Valencian Community took part in (Castellón de la Plana, Region of Valencia, Spain). A total of 90 IH players from the Skating Federation of the Valencian Community were selected to participate in this study.
Following the inclusion criteria, those who were from 8 to 15 years old and were playing within the Spanish Federative Categories of "Benjamín" (U11) "Alevín" (U13) and "Infantil" (U16) were included in the study (n = 77), whereas goalkeepers and players who belonged to the U17 team were not included (n = 15). However, those who had previously received treatment for any frontal or sagittal plane-related pathology by the use of a corset or specific kinesiotherapy or those who presented specific symptoms or musculoskeletal limitations to perform the tests correctly were excluded (n = 1).
Finally, a total of 74 IH players U16 participated in the study (Table 1).

Procedure
The study was conducted in the season 2016-2017. According to the Declaration of Helsinki, the procedures and potential risks were explained to IH players, parents and coaches prior to participation and legal tutors expressed written consent. "Sagittal Integrative Morphotype," as well as Lumbo-Horizontal angle in flexion (L-H fx) of all participants, were assessed. In addition, participants completed an ad hoc questionnaire about their sport-related background (federative category, current competitive level, tactical position, stick length, dominant leg (defined as the participant's preferred kicking leg)), anthropometric characteristics (age, weight, height and body mass index), regular training workload (years of sport experience, training months per year, training days per week, training hours per week, current competitive level) as well as about prior and current musculoskeletal injuries and treatment.
According to Wojtys et al. (2000) and Sainz de Baranda, Rodríguez-García & Santonja (2010), "training hours per year" and "training volume" were calculated. "Training hours per year" was equal to training hours per week × 4 weeks per month × months per year. "Training volume" was equal to years of sport experience × training hours per year. Two groups were established based on those who had trained more or less than 160 h per year. The players were also divided according to the accumulated training workload in more or less than 480 h of training.

Sagittal spinal morphotype assessment
Data from each IH player were taken during the same assessment session and with the same temperature (25 C). All the measurements were performed by the same Sport Science expert and participants were assessed wearing undergarments and barefoot. Athletes did not perform warm-up or stretching exercises before or during the measurement in order to achieve real clinical conditions (Aalto et al., 2005;Cejudo, 2015;Ginés-Díaz et al., 2019). An unilevel inclinometer (ISOMED, Inc., Portland, OR, USA) was used to quantify the sagittal spinal curvatures by providing considerable reproducibility and validity, with a good correlation with the radiographic measurement (Mayer et al., 1984;Saur et al., 1996), and according to the methodology described by Santonja (1996), which has been used in previous studies (Ginés-Díaz et al., 2019;Sainz de Baranda, Santonja & Rodríguez-Iniesta, 2009;Sanz-Mengibar, Sainz-de-Baranda & Santonja-Medina, 2018). A goniometer provided with a spirit level system was used to quantify the L-H fx (Sainz de Baranda et al., 2014;Santonja, Andújar & González-Moro, 1994).
Intra-tester reliability of thoracic and lumbar curvatures and pelvic tilt was calculated in a previous pilot study. A priori reliability was established by the primary investigator in a sample of convenience (university students; n = 12, ranged from age = 20-22 years; (age: 25.8 + 0.9 years; height: 1.71 + 0.09 m; body mass: 72.05 ± 9.28 kg) measured on two occasions by the same tester in a single session. Intra-class correlation coefficients (ICC) with 95% confidence intervals (CI) were calculated. The ICC and the minimal detectable change at a 95% confidence interval (MDC95) values for all measures ranged from 0.88 to 0.97 (95% CI [0.70-0.99]) and 0.53 to 1.1 respectively (95% CI [0.38 -1.6 ]).

Standing position
To assess the SP, the participant was standing and relaxed (Ginés-Díaz et al., 2019;Sanz-Mengibar, Sainz-de-Baranda & Santonja-Medina, 2018). The inclinometer was placed at the first mark (T1) and calibrated to 0 , then the curvature was profiled until maximum angulation of thoracic curvature was reached and the angle was recorded. Subsequently, the inclinometer was calibrated to 0 again at this point and the lumbar curvature was profiled until the maximum angle was reached and recorded.

Slump sitting position
To measure the SSP, the participant was sitting on the stretcher in a relaxed posture with the forearms resting on the thighs, knees flexed and without feet support (Ginés-Díaz et al., 2019;Sanz-Mengibar, Sainz-de-Baranda & Santonja-Medina, 2018). First, the inclinometer was placed at the first mark (T1) and it was calibrated to 0 . Then, the inclinometer would be placed on the second mark (T12) and the grades for the thoracic curve would be recorded. After that, the inclinometer was calibrated to 0 again on this mark and then the inclinometer was placed on the third mark (L5-S1) in order to record the lumbar curve angle.
However, the same procedure as in SP was used when it was observed that participants kept their lumbar lordosis in this position.
Maximum flexion of the trunk in a Toe-Touch test position Firstly, participants were standing on a box 36 cm high with their feet bare and hip-width apart. They were asked to flex the trunk as far as possible, while knees, arms and fingers were fully extended.
The athlete had to keep the MFT for 6-8 s while sagittal spinal curvatures were measured following the same procedure as in the SSP (Sainz de Baranda et al., 2014). References of normality for thoracic and lumbar curves The references of normality for thoracic and lumbar curves in each assessed position are described in Table 2.

Sagittal integrative morphotype diagnosis
Tables 3 and 4 detail the different categories and subcategories for the integrative diagnosis of the sagittal integrative thoracic and lumbar morphotype, respectively.

Hip joint angle test: L-H fx
The HJA is a field-based test that might be proposed as an alternative to the Passive Straight Leg Raise test or the SRT for the assessment of hamstrings flexibility. The score achieved in this test is not negatively influenced by the pelvic position or stability and only one examiner and an inexpensive gravity goniometer are required (Ayala et al., 2013;Sainz de Baranda et al., 2014). The HJA test can be measured at the end point of   In the current study, the L-H fx was measured with a goniometer while the subject was performing a MFT in a horizontal position (Sainz de Baranda et al., 2014;Santonja, 1996;Santonja, Ferrer & Andújar, 1994). The branches of the goniometer were aligned with the horizontal line and the spinous processes of L4-S2 in order to record the angle between the two references, however, the supplementary angle was used for the data analysis ( Fig. 2). Pelvic tilt data were classified as normal (100 ), mild posterior pelvic tilt (101-110 ) and moderate posterior pelvic tilt (>110 ).

Statistical analysis
Descriptive statistics including means and standard deviations, minimum and maximum were calculated for each variable: age, weight, height, BMI, training hours per year, training volume, sagittal spinal angles in SP, SSP and MFT and sagittal pelvic disposition.
Prior to the statistical analysis, the distribution of raw data sets was checked using the Kolmogorov-Smirnov test to determine normal distribution. The results demonstrated that the sagittal spinal variables were normally-distributed (p > 0.05). Therefore, parametric analyses were carried out in order to compare sagittal spinal mean angles by competition categories and training workload. Pairwise comparison of means (Student t-test for independent samples) was used to examine the differences between groups of training hours per year and training volume in relation to sagittal spinal and pelvic disposition angles. One-Way Analysis of Variance was carried out to analyze sagittal spinal mean angles between competition categories. Furthermore, the absolute and relative frequency of athletes in each category of spinal morphotype and pelvic disposition were also calculated. Likewise, it was also calculated the absolute and relative frequency of players in each category and subcategory according to their "Sagittal Integrative Morphotype." The analysis was performed using SPSS version 23.0 (SPSS Inc., Chicago, IL, USA). The level of significance (a) was set at 0.05; therefore, p values less than 0.05 were considered to be statistically significant.

Sagittal thoracic and lumbar morphotype & pelvic disposition
The means and standard deviations values for spinal curves in each of the three positions and for values of pelvic disposition are shown in Table 5 according to competition categories and training workload.
With regard to "training hours per year," it was found that those who trained >160 h per year had higher dorsal kyphosis in SP (t (72) = −2.051; p = 0.044; d = 0.63) and in SSP (t (72) = −2.694; p = 0.009; d = 0.48) than those who had less training load per year. In addition, it was observed a tendency toward signification for dorsal and lumbar kyphosis in MFT (p = 0.065 and p = 0.067, respectively), presenting a greater dorsal kyphosis and a less pronounced lumbar kyphosis those who trained >160 h per year. Nevertheless, no statistically significant differences were found when spinal curves in each position were analyzed depending on the "training volume." Taking into account the references of normality presented in Table 2, it is observed that IH players have an angular mean above normal in SSP for both the dorsal and lumbar curvature. Table 6 shows the percentage and frequency of athletes within each category by assessment position for each spinal curvature and for pelvic disposition.
As for the relaxed SP, the results showed that 60.8% of the athletes presented normal kyphosis, 37.8% had hyperkyphosis, and 1.4% had rectification (hypo-or reduced kyphosis) for the thoracic curve, while 89.2% of the athletes were classified as normal, 1.4% had hyperlordosis and 9.5% presented rectification (hypo-or reduced lordosis) for the lumbar curvature.
With regard to the SSP, the results showed that 35.1% of the athletes presented normal kyphosis, 64.9% had hyperkyphosis, and 1.4% had hypokyphosis for the thoracic curve. On the other hand, 31.1% were within normal ranges, 68.9% had hyperkyphosis and 0% presented hypokyphosis for the lumbar curve.
In a MFT, 74.3% of the athletes presented normal kyphosis, 17.6% had hyperkyphosis, and 8.1% had hypokyphosis for the thoracic curve. As for the lumbar curvature, the results showed that 55.4% had a normal lumbar curve, 44.6% had hyperkyphosis and 0% presented hypokyphosis.
When the L-H fx was evaluated, the results showed that only 16.2% of the athletes were classified as normal, whereas most of IH players were categorized in a posterior pelvic tilt (41.9% with a mild posterior pelvic tilt and 41.9% with a moderate posterior pelvic tilt).    Notes: a n: number of cases; %: number of cases with respect to the total IH players. * Classification of thoracic integrative morphotype according to thoracic values in SP, SSP and in MFT (Santonja, 1996).

Sagittal integrative spinal morphotype
The values for the sagittal morphotype of the spine integrating the three assessed positions (SP, SSP and MFT) can be observed in Tables 6 and 7. Both tables show the frequency of IH players in each category according to the integrative diagnosis of the sagittal spinal morphotype (Santonja, 1996).
With regard to sagittal thoracic morphotype, only 13 IH players presented "Normal Morphotype" with a normal kyphosis in the 3 measurement positions. Thirty-one IH players adopted a normal kyphosis in a relaxed SP, but with an increased kyphosis (hyperkyphosis) in a SSP (static) or in MTF (dynamic), and they were diagnosed with "Functional Thoracic Hyperkyphosis." Twenty-eight IH players were diagnosed with "Hyperkyphosis" because they adopted a hyperkyphotic curvature in a SP and in a SSP (static) or in MFT (dynamic). When a player presented a hyperkyphotic morphotype in the three positions he was categorized as "Total Hyperkyphosis." Only one IH player was diagnosed with "Hypomobile Kyphosis" (adopted a normal kyphosis in a relaxed SP and in a SSP, but presented a hypokyphosis in MFT), and another player with "Hypokyphosis or Hypokyphotic Attitude" (adopted a normal kyphosis in a SSP and in MFT, while a hypokyphosis is presented in a relaxed SP) ( Table 7).
With regard to the sagittal integrative lumbar morphotype (Table 8), only 17 IH players presented "Normal morphotype" with a normal lumbar curvature in the three assessed positions. Forty-nine IH players adopted a normal kyphosis in a relaxed SP, but with an increased kyphosis (hyperkyphosis) in a SSP (static) (n = 15) or in a MFT (dynamic) (n = 4), or in both positions (total) (n = 30), and they were diagnosed with "Functional Lumbar Hyperkyphosis." Five IH players were diagnosed with "Structured Lumbar Kyphosis" because they presented a hypolordosis or kyphosis in a SP and a hyperkyphosis in a SSP and in MFT. Only two IH players were diagnosed with "Hypolordosis" (with a hypolordosis in a relaxed SP, but a normal lordosis in a SSP and in MFT). Finally, another IH player was diagnosed with "Lumbar Hypermobility." No players presented the morphotype "Hyperlordotic Attitude" or "Structured Hyperlordosis."

DISCUSSION
This study was undertaken to investigate the sagittal spinal curvatures of the thoracic and lumbar spine and its relationship to training load, and to describe the "Sagittal Integrative Morphotype" in young federated IH players. IH has specific requirements for players, which include trunk forward bending and concrete movements such as skating, generating physical needs that do not occur in other sports. These physical demands make IH players susceptible to certain postural and structural adaptations that can result in subsequent injuries and back pain. Previous studies have shown that those specific and repetitive movements and postures of each sport influence spinal curvatures (Rajabi et al., 2008;Uetake et al., 1998;Wodecki et al., 2002) and for that reason, several studies agree on the importance of a initial postural evaluation in order to identify spinal deformities and sagittal imbalances. Sagittal curvatures are geometric parameters which influence mechanical properties of the spine during compressive loading Keller et al., 2005). Sagittal alignment influences postural loading and the load balance of the intervertebral disc, therefore, abnormal spinal curvatures cause increased forces to act upon the intervertebral discs . Alterations in spinal curvatures may potentially influence the development of lower back pain Smith, O'Sullivan & Straker, 2008), which is a common pathology among athletes (Kameyama et al., 1995).
The most reliable technique to quantify kyphosis and lordosis is the conventional spinal X-ray method. There are other methods free of ionizing radiation that assess the curvatures of the spine in the sagittal plane, for instance, the inclinometer provides a noninvasive evaluation with good reproducibility, reliability and correlation with the radiographic measurement The integrative diagnosis of the sagittal morphotype of the spine was defined by Santonja (1996) and adds the assessment of the sagittal curvatures during MFT and in a SSP (Sainz de Baranda, Rodríguez-García & Santonja, 2010; Sainz de Baranda, Santonja & Rodríguez-Iniesta, 2009;Santonja, 1996; Sanz-Mengibar, Sainz-de-Baranda & Santonja-Medina, 2018) to the classical quantification of the thoracic and lumbar curves in a relaxed SP in order to perform a more accurate diagnosis.

Reference values and categories for thoracic curvature in previous studies
In the current study, mean thoracic curvature value was 38.5 , 45.2 and 53.7 in a relaxed SP, in a SSP and in MFT, respectively.
Wojtys et al. (2000) found similar values (a mean of 38.1 ) when they studied the thoracic curve in 189 ice-hockey players (aged between 8 and 18 years) in a relaxed SP. In sports like basketball, handball, volleyball and female artistic gymnast, some studies  have found similar or lower angular values; while other studies found higher values in swimmers, runners, tennis players, trampoline gymnasts, male artistic gymnasts, cross-country skiers, and paddlers (Table 9). On the other hand, it is interesting to note that in sports related to dancing abilities, thoracic angular values tend to be much lower than in other types of sports (Gómez-Lozano, 2007;Gómez-Lozano et al., 2013;Nilsson, Wykman & Leanderson, 1993).
When the results were analyzed by "training hours per year," it was found that players who trained more than 160 h per year presented a significantly higher dorsal kyphosis in SP and SSP and a tendency toward signification in MFT than those who had less training load. Furthermore, it was observed that U16 had a significant higher dorsal kyphosis than U11 in MFT. In this sense, varies studies have found a relationship between sports training and variations in the sagittal spinal curvatures of adolescent athletes (Ferreira-Guedes & Amado-João, 2014; Grabara, 2015;Hecimovich & Stomski, 2016;Sainz de Baranda, Santonja & Rodriguez-Iniesta, 2010). Wojtys et al. (2000) reported an increase in sagittal spinal curvature in adolescents who participated in ice hockey and exceeded 400 h of training per year. Intense physical training combined with a developing spine, where loads are transferred from the upper to the lower extremities, leads to an overload of the spinal structures that could influence the deformation of the spine.
Inline hockey is a very fast paced game, which is characterized by high intensity intermittent skating, rapid changes in velocity and duration, frequent body contact, and the execution of a wide variety of technical skills (Flik, Lyman & Marx, 2005;Mölsä et al., 2003). IH implicates adapting the body to a hard physical effort and to the required posture for that sport. As a result, athletes commonly present postures that are related to the most common sports abilities in each discipline (Rajabi et al., 2008;Usabiaga et al., 1997). As Wojtys et al. (2000) stated, specific postures and actions which take place in IH practice might modify the sagittal spinal curvatures by altering spine's exposure to certain mechanical loads during the athlete's growth. In adolescents, Ferreira-Guedes & Amado-João (2014) note that "these biomechanical compensations may influence the growth processes and lead to the development of various postural patterns due to the immaturity of their musculoskeletal structures. At first, the postural compensations are adaptive, but later they can become permanent and even predispose young athletes to injuries." So, the results of the current study could confirm that sagittal curvatures of the spine can be modified with regular IH training as previously described in other sports (Ginés-Díaz et al., 2019;Sainz de Baranda, Santonja & Rodríguez-Iniesta, 2009;Sanz-Mengibar, Sainz-de-Baranda & Santonja-Medina, 2018;Uetake et al., 1998).
In the current study, most of IH players had normal angular values in a relaxed SP (n = 45/74, 60.8%) and in a MFT (n = 55/74, 74.3%) for the thoracic curve (Fig. 3). However, in the SSP there was a higher percentage of IH players with an increased thoracic curvature or hyperkyphosis (n = 48/74, 64.9%).
In another study, Muyor, López-Miñarro & Alacid (2011b) reported that elite cyclists showed a statistically higher thoracic hyperkyphosis than non-athlete subjects. These authors justified their findings with specific sport adaptations. In this sense, Grabara & Hadzik (2009) found that a kyphotic posture tended to be more frequent and the lordotic one less frequent in volleyball players than in untrained subjects. The authors attributed that finding to the typical volleyball posture consisting of forwarding bending with rounded back as well as the arms and shoulders protruding. Wojtys et al. (2000) reported that high-intensity training increases the risk of developing adolescent hyperkyphosis. In this sense, Alricsson & Werner (2006) found that after 5 years of intensive training the skiers increased their thoracic kyphosis but no change in lumbar lordosis was noticed.
Other studies found lower percentages of thoracic hyperkyphosis. For instance, Muyor et al. (2013) found 37.5% and 6.2% of thoracic hyperkyphosis in 24 male and 16 female elite adolescent tennis players, respectively. López-Miñarro et al. (2008)  The results showed that 25% and 45% of kayak and canoe athletes, respectively, had thoracic hyperkyphosis in MFT. In the same study, these authors found that 82% and 95% of kayak and canoe athletes, respectively, had thoracic hyperkyphosis in a SSP. Pastor et al. (2002) observed only 24.7% of the morphotypes within normality in swimmers, 29.4% of the morphotypes with mild kyphosis and 45.9% with moderate kyphosis. The same author performed a radiological study in the position of Sit and Reach test (SRT) and observed a higher percentage of moderate and marked thoracic curves (p < 0.05) and a significant tendency to increase the number of vertebral wedges as the value of kyphosis and age increased. In addition, these wedges were related to the dynamic thoracic kyphosis, since the swimmers with more thoraco-lumbar wedges presented higher values of dynamic thoracic kyphosis (p < 0.05).
In contrast, Gómez-Lozano (2007) only observed 6.1% and 3% of misaligned morphotypes in classic and Spanish dancers, respectively, as only some mild hyperkyphotic attitudes were diagnosed in this posture.
It must be pointed out that not all studies use the same spinal assessment protocol and the same references to categorize sagittal spinal angular values. In this sense, Grabara (2016a) established that values above 35 are considered thoracic hyperkyphosis, thoracic normality is considered from 25 to 35 and thoracic hypokyphosis is accepted when the value is lower than 25 . Muyor et al. (2013) used the references of normality proposed by Mejia et al. (1996) andTüzün et al. (1999), where the values between 20 and 45 are accepted as neutral thoracic kyphosis, values below 20 are considered thoracic hypokyphosis, and values above 45 are considered thoracic hyperkyphosis. However, Pastor et al. (2002) and Gómez-Lozano (2007) used the same reference values as in the current study.

Reference values and categories of lumbar curvature
Regarding the lumbar curvature, in the current study mean values were 28.7 , 20.3 and 31.5 in a relaxed SP, in a SSP and in MFT, respectively.
In comparison with our results, Wojtys et al. (2000) found an average value much higher (44.5 ) for the lumbar curvature in a relaxed SP in 189 ice-hockey players aged from 8 to 18 years old. In all the sports participants whose spinal morphotype has been assessed (Table 10) When the results were compared by sex, it was observed a greater lumbar lordosis in girls (40.31 ± 10 ) than in boys (32.06 ± 7.7 ).
When Ohlén, Wredmark & Spangfort (1989) assessed the spinal morphotype, it was found a mean value of 35.6 ± 7.8 for the lumbar curve with a Debrunner's cifometer and value of 35.2 ± 6.9 with an inclinometer in 64 artistic gymnasts. 20% of the gymnasts manifested lower back pain. When values for the lumbar curve were compared between the gymnasts with pain (40.6 ± 7.9 ) and the asymptomatic gymnasts (35.4 ± 7.2 ), it was observed that the mean lordotic value was higher in gymnasts with back pain. In addition, the authors found a significant correlation between back pain and a lumbar lordosis greater than 41 .
Martínez-Gallego ( Conesa Ros (2015) observed mean value of 32.9 ± 8.5 in a group of competitive esthetic gymnasts. In addition, the author observed that a lumbar lordosis tended to increase with age. Thus, the group of competitive esthetic gymnasts under 11 years old had a lumbar value of 28 ± 6.8 and the group over 15 years old had a mean lumbar value of 36.4 ± 9.2 . The group of competitive rhythmic gymnasts under 11 had a mean lumbar lordosis of 33.8 ± 9.4 , while the group over 15 years old had a mean lumbar value of 39.2 ± 8.6 . This evolution of lordosis with age has also been found in previous studies carried out with school-aged children (Cil et al., 2005;Murray & Bulstrode, 1996;Voutsinas & MacEwen, 1986).
Sainz de Baranda, Santonja & Rodríguez-Iniesta (2009) observed a mean value of 17.4 ± 9.6 for the lumbar curve in gymnasts of the Trampoline modality. When the   results were compared by sex, a significantly greater lumbar kyphosis was observed in males (21 ± 7.9 ) than in female gymnasts (14 ± 10 ) (p < 0.004). Conesa Ros (2015) and Martínez-Gallego (2004) showed how sports practice can influence or can be related to a higher angular value for the lumbar curve in their studies with esthetic and rhythmic gymnasts. In this sense, both rhythmic and esthetic gymnasts (16.7 ± 6.6 and 15.9 ± 8.1 , respectively) had a significantly greater lumbar kyphosis than the control group (13.8 ± 7.7 ) (p = 0.033).
The incorrect alignment of the lumbar spine found in the SSP in the three modalities of gymnastics could be due to repetitive hyperflexions and hyperextensions of the trunk which are performed in gymnastics. Thus, these movements could come to a hypermobile lumbar curve.
López -Miñarro et al. (2008) found angular lumbar values lower than 20 in 43 infantile paddlers (23 kayakers and 20 canoeists), and no significant differences were found between kayakers and canoeists. Likewise, when López-Miñarro et al. (2008 assessed the sagittal spinal curves of 130 canoeists (aged from 15 to 20 years old), the authors found values lower than 20 for the lumbar curvature, with no significant differences regarding gender.
It was also found that 68.9% and 44.6% of the IH players had lumbar hyperkyphosis in a SSP and in a MFT, respectively. Some previous studies found higher percentages of hyperkyphosis for the lumbar curvature in these positions, possibly due to the practice of the sport in a sitting position or the repetition of technical gestures with a maximal ROM in the lumbar spine and lower limb. In this sense, López-Miñarro et al. (2008) reported that around 90% of paddlers had lumbar hyperkyphosis in MFT. In addition, these authors found lumbar hyperkyphosis in around 75% of kayak and canoe athletes in a SSP.
It must be pointed out that not all studies use the same spinal assessment protocol and the same references to categorize sagittal spinal angular values. In this sense, for the relaxed SP, Grabara (2016b) established that values above 35 are considered lumbar hyperlordosis, a neutral lumbar spine is considered from 25 to 35 and lumbar hypolordosis is accepted when the value is lower than 25 . Muyor et al. (2013) used the references of normality proposed by Tüzün et al. (1999), where the values between 20 and 40 are accepted as a neutral lumbar spine, values below 20 are considered as an hypolordotic lumbar spine, and values above 40 are considered hyperlordosis.
With respect to the MFT, Pastor et al. (2002) established that values below 22 are considered as normal lumbar kyphosis and values between 22 and 29 are considered as lumbar hyperkyphosis. As for the SSP, the author established that values below 14 were accepted as normal lumbar kyphosis and values between 14 and 21 were considered lumbar hyperkyphosis.
In this sense, as Purcell & Micheli (2009) stated, repetitive flexo-extension and torsion movements because of technical-tactical actions in IH can result in overuse injuries to the spine. In fact, these repetitive movements with an imbalanced spinal posture (e.g., hyperkyphotic position with the trunk bent forward) are particularly worrisome in young IH players. Therefore, the position of the lumbar spine while sitting or trunk forward bending should be trained for a better alignment through pelvic proprioceptive exercises or trunk muscles strengthening.
Furthermore, it is important to highlight the fact that 6.8% of the IH players were diagnosed with "Structured Lumbar Kyphosis." In this sense, it has to be pointed out that the high prevalence of posterior pelvic tilt found among players could have led to a misaligned lumbar spine during MFT probably due to the hamstrings tightness that can make the pelvis lose its horizontality and adopt a posterior pelvic tilt (López-Miñarro et al., 2012;Santonja, Andújar & González-Moro, 1994). Since the trunk forward bending while standing is the basic posture in IH, the players should train their hamstrings flexibility in order to keep a neutral lumbar spine when their trunk is bent forward. In addition, good pelvic proprioception would be necessary to keep a neutral pelvic tilt while IH players have to stay in constant quadruple flexion (ankle, knee, hip and trunk). It is important to note that if sagittal spinal assessment had been only carried out in SP, the results of this study would have shown that most of IH players were within the normal ranges for both curves (60.8% of the athletes presented normal kyphosis and the 89.2% of the athletes presented normal lordosis in the SP). However, taking into account the "Sagittal Integrative Morphotype" it was determined that IH players had a misaligned sagittal spine. These results show how important is to include the assessment of the three positions as part of the protocol in order to define "Sagittal Integrative Morphotype" and so as to establish a correct diagnostic (Sanz-Mengibar, Sainz-de-Baranda & Santonja-Medina, 2018). Therefore, an incorrect sagittal spinal assessment leads to misclassification of the athletes' morphotypes, generating negative consequences, not only in terms of deformity and pain but also in preventive and rehabilitative terms (Ginés-Díaz et al., 2019).
Some limitations of the present study must be reported. The age distribution of the participants was relatively limited and the sample size was relatively small. Furthermore, only male IH players were assessed. In addition, patients with radiographic evidence deformity may present a normal spinal morphotype. Future studies which include a larger sample should investigate the association between sagittal spinal morphotype and back pain or the ratio of injury. Furthermore, prospective investigations in order to study how sagittal spinal curves develop with age and practice are needed.

CONCLUSIONS
Federative IH practice seems to cause specific adaptations in spinal sagittal morphotype in young players. The findings reported in the present study suggest an association between exposure to IH athletic training, age and increased thoracic kyphosis.
The most prevalent sagittal spinal misalignments in young federated male IH players were the thoracic hyperkyphosis (64.9%) and the lumbar hyperkyphosis (68.9%) in a SSP and the lumbar hyperkyphosis (44.6%) in a MFT.
However, taking into account the "Sagittal Integrative Morphotype" only 13 (17.6%) IH players presented "Normal Morphotype" with a normal thoracic kyphosis in the three measured positions. While only 17 (23%) IH players presented "Normal Lumbar Morphotype" with a normal lumbar curvature in the three assessed positions.
For the thoracic curvature, 18.9% of the IH players presented "Total Functional Hyperkyphosis," 17.6% of the players presented a "Static Functional Hyperkyphosis" and 16.2% of the players had a "Total Hyperkyphosis". Whereas for the lumbar curvature a 40.5% of the players presented "Total Functional Hyperkyphosis" and the 20.3% of players were diagnosed with "Static Functional Hyperkyphosis". Furthermore, only 16.2% of IH players showed normal pelvic titl.

Clinical implications
It is important to assess the "Sagittal Integrative Spinal Morphotype" in sports for the preemptive care of spinal deformities from the earliest stages. The assessment of sagittal spinal curvatures in the three described positions gives a new perspective for the diagnosis of sagittal spinal morphotype, which may reduce the number of players wrongly classified as normal. It is important to note that the protocol is easy to apply due to the low-cost instruments used and it only requires an examiner.
Exercise programs to prevent or rehabilitate these imbalances in young IH players are needed. Pelvic proprioceptive exercises, trunk muscles strengthening, and flexibility training could be included as a part of preventive programs. This manuscript creates a paradigm for future studies about associated risk factors to develop unbalanced sagittal spines in IH players.

ADDITIONAL INFORMATION AND DECLARATIONS Funding
The research was supported by the Spanish Ministry of Economy, Industry and Competitiveness (Grant Number: DEP2010-21793); and the Spanish Ministry of Education, Culture and Sports for the Training of University Teaching Staff (Grant Numbers: FPU15/05200 and FPU18/00702). This research is part of the project entitled "Study of injury risk in young athletes through artificial intelligence networks," funded by the Spanish Ministry of Science and Innovation (reference: DEP2017-88775-P). This study was carried out during the research stay at the Facoltà di Scienze Motorie of Università degli Studi di Urbino "Carlo Bo" from 29/10/2018 to 03/02/2019 funded by Erasmus Teaching. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.