Evaluation of Thorax Diameter Changes through Trunk Morphology and Different Running Intensities

Featured Application

This article is the first step of a recent investigation line which gives a new perspective to the physical condition evaluation using the quantification of the thoracic expansion. Knowing how much the thorax is expanded during different exercise intensities could be used to classify different persons for making exercise recommendations of the respiratory muscles, for evaluating the predisposition for endurance sports or as an evaluation for breathing rehabilitation processes.

Abstract

During breathing, the human thorax is expanded or contracted during inspiration and expiration. The morphology of the thorax seems to be determinant for endurance efforts. This study aims to analyse the variation of the thorax cross-sections during running exercises at different intensities and the influence of thorax morphology. Twenty-two athletes were captured using a motion capture system (13 reflective markers) while they performed an incremental running test. Three captures of each subject were performed at different intensities (45%, 70%, and 85% of HRR (heart rate reserve)) and three morphology groups were made by splitting their thoracic index. The results showed a significant increase in the anteroposterior and mediolateral cross-sections when the intensity of the exercise is also increased. No differences were found in the cross-sections due to the morphology of the thorax. However, subjects with a deeper thorax showed a different behaviour as they increased their anteroposterior cross-section during higher intensities, while flatter thorax subjects did not perform in the same way. This could be explained as compensation for the morphological disadvantage that a deeper ribcage shape suffers when developing endurance efforts. More investigations regarding thorax kinematics are needed for a better understanding of breathing disorders and physical activities.

1. Introduction

During quiet breathing, the human thorax seems to be static, but during exercise, it expands or contracts through inspiration and expiration [1,2]. The increase and decrease of the thoracic volumes are due to the movements of the costal and sternum bones produced by the synergic contraction of the intercostal, abdominal, and diaphragm muscles [3,4]. Some authors have documented that the thoracic volume is one of the key factors in determining the correct functioning of the respiratory system [5,6].

Not all individuals have the same respiratory pattern: women exhibit a breathing pattern with a greater increase in the anteroposterior cross-section in the upper part and a more thoracic breathing character; in men, it is mixed (both in the upper and lower parts); in children, it is abdominal; and in the elderly, it is characterized by spinal curvature, leading to increased abdominal breathing [3,7]. Moreover, the BMI (body mass index) modifies the breathing pattern and may even reach hypoventilation-obesity syndrome [8].

Traditionally, the spirometry technique and expiration gas analyser are methods used for measuring respiratory variables [9,10]. However, these techniques are unable to detect changes in the kinematics on the chest wall and also impede normal breathing activity because of the mask and tubes that are required [6,10,11,12,13,14]. Espinosa and Sánchez-Lafuente (2002) considered that physical condition evaluation will have higher validity when the evaluation of adaptation of the organism to a specific physical activity enables reproducing the technique of a movement more specifically [14]. The analysis of chest wall movements during breathing and thoracoabdominal volume calculation allows pathologies to be viewed from another perspective, providing the latest information for medical treatments or rehabilitation [15]. Developing methods for evaluating the physical condition and health of athletes based on the thoracic range of movement (ROM) in specific situations is a promising research topic [14,15,16,17,18,19].

Some studies have demonstrated that the cross-sections could be simple and effective variables for analysing the thoracic ROM in different body postures or from maximum inspiration to maximum expiration [20,21,22,23]. Meanwhile, handbooks of physiology have reported the behaviour of the three cross-sections of the thorax during breathing, but to date, no study has measured them during high-intensity exercise [3]. We contend that the use of cross-sections for evaluating exercise intensity levels or establishing the relationship between them has potential as a new method for evaluating the functioning of the thorax and quantifying fitness status.

One of the most frequent non-invasive techniques is the use of optoelectronic systems, which register and represent the position of reflective markers dispersed over different body parts. The use of this technique for analysing breathing patterns started in 1994 with the first study that used four cameras and thirty-two reflective markers for estimating the thorax volume [24]. This technique has evolved to create optoelectronic plethysmography (OEP) [9,15,17,18,19,25], which detects every single movement of the thorax [11]. The OEP uses infrared cameras and a geometrical model to calculate breathing variables [4,9,11,15,16,17,18,19,26,27] and to analyse respiratory pathologies and other physiological disorders [6,13,16,18,28,29].

In numerous studies, the results of the OEP technique have been compared with spirometry, validating the OEP technique as a gold standard for analysing breathing patterns [6,13,15,17,18,19,21,26,27]. Because of the complexity of the thorax, with OEP methodology, the position of the markers is crucial to compute the variables required [15,17,18,19]. The number of markers varies among different studies from 89 to 12 markers [4,17,18,19,28,30]. Recently, a 12-marker model has shown equivalent results to the spirometry technique when analysing breathing patterns, showing that models based on a reduced number of markers facilitate the test and offer satisfactory results [31].

Moreover, morphology has been observed to affect respiratory functioning [32]. Studies in anthropology show that the morphology of the thorax varies from one type of hominid to another depending on their lifestyle demands, concluding that running performance also depends on trunk shape, which encompasses important aspects such as the width of the pelvis, trunk flexion angle, lumbar lordosis, hip flexion, and breathing mechanics [32,33]. Following these findings, more knowledge about the influence of thorax morphology is needed as a way to have clues about the sporting predisposition of the subjects.

The analysis of breathing in sports also suffers from a lack of information about the change of the thorax volume throughout exercise. Such information would improve our knowledge about human body responses to physical activity and how morphology affects their predisposition to obtain a better performance during running. Despite the validation of the OEP technique as a reliable method for analysing kinematic variables, the principal cross-sections of the thorax have not been observed during exercise at different intensities, nor has there been an analysis of how the morphology of each subject may modify those variations. The objective of the present study was to analyse the variation of the thorax cross-sections during running exercises at different intensities and the influence of thorax morphology. The hypotheses were (1) that the increase of the cross-sections of the thorax would be greater when the exercise intensity increases and (2) that flatter trunks would experience a greater change in breathing patterns.

2. Materials and Methods

2.1. Data Acquisition Technique

An optoelectronic system (VICON^®, (Oxford Metrics Ltd., Oxford, UK)) comprising six Full HD video cameras (three at the front, two at the back, and one on the left side of the subject) working at 120 Hz, was used for determining the 3D position of a set of 13 reflective markers attached to the skin of the subjects using specific anatomic landmarks (Figure 1 and

Table 1. Markers model description.

Marker Name	Marker Location
T1	T1 spinal apophysis
T5	T5 spinal apophysis
T7	T7 spinal apophysis
RCO1	Anterior part of the right first rib
LCO1	Anterior part of the left first rib
CLAV	Jugular suprasternal notch
STRN	Xiphoid process
TOAC5R	Mammillary line of the right fifth rib
TOAC5L	Mammillary line of the left fifth rib
TOLR	Lateral part of the right tenth rib
TOLL	Lateral part of the left tenth rib
TOAUCR	Anterior inferior part of the right tenth rib
TOAUCL	Anterior inferior part of the left tenth rib

). Following the guidelines of the VICON^® System, the cameras were calibrated, obtaining an error of less than 1% and a reliability on the determination of the position of the reflective markers of each camera of less than ±2 mm. The 3D data of the markers were interpolated and smoothed using Woltring’s method (mean square error of 4 mm²). A custom-made code on Body Builder (VICON^®) using programming language was used for determining the variables.

2.2. Mechanical Variables

The analysed variables were the cross-section distances of the three thorax axes ─anteroposterior, mediolateral, and craniocaudal—following Hamill et al. (2003) at different running intensities [34]. The running intensities were measured using the percentage of the heart rate reserve (HRR) (45%, 70%, and 85% of HRR). These three intensities correspond with the cardiorespiratory phase 1 [i.e., light intensity, below the ventilatory threshold (VT)], phase 2 [i.e., moderate intensity, between the VT and the respiratory compensation threshold (RCT)], and phase 3 [i.e., high intensity, above the RCT]. The expansion of the thorax was estimated by analysing thorax cross-sections on the dynamic captures.

The cross-section variables that were estimated were the results of the following operations:

-

Craniocaudal thorax cross-section (CTC): distance between SMP and IMP points ((SMP: superior middle point: middle point between CLAV, T1, LCO1, and RCO1); (IMP: inferior middle point: middle point between STRN, TOAUCL, TOAUCR, TOLL, TOLR, and T7)).

-

Anteroposterior thorax cross-section (ATC): distance between AMP and PMP points ((AMP: anterior middle point: middle point between CLAV, STRN, TOAC5L, and TOAC5R); (PMP: posterior middle point: T5)).

-

Mediolateral thorax cross-section (MTC): distance between TOLR and TOLL.

All these variables are shown in Figure 1.

2.3. Sample

Twenty-two healthy Caucasian males, amateur athletes of different disciplines (eight from athletics, eight from soccer, and six from basketball), all undergraduate students of the Sports Science Faculty, participated in the study (see

Table 2. Characteristics of the study participants.

Descriptives (N = 22)
	Age (Years)	Weight (kg)	Height (m)	RHR (bpm)	HRmax (bpm)
Mean	20.86	69.31	1.77	63.91	197.95
Standard deviation	2.75	6.21	0.06	7.93	1.84
Minimum	18	57.40	1.65	53	192
Maximum	29	81.00	1.90	86	199

RHR: resting heart rate; HRmax: maximum heart rate.

). Approval was obtained from the ethics committee of the university. The procedures used in this study adhere to the tenets of the Declaration of Helsinki. All the participants signed an informed consent document accepting the possible risks of the test. The inclusion criteria were to have not suffered any injury in the past six months and to be undertaking regular exercise, including long-distance running (covering 20 km per week). The participants could not contribute to the study if they fulfilled any exclusion criteria, such as having suffered respiratory pathologies or other conditions that make it dangerous or impossible to perform an incremental running test.

2.4. Experimental Design

The running test was a progressive incremental test on a treadmill device (Telju JT4100-Litin-035, Toledo, Spain). The test started at 7 km/h with a 1% slope, and the velocity was increased 0.5 km/h every 30 s. The subject’s maximum heart rate was previously estimated before the test and monitored continuously with a pulsometer (Polar Ceinture H10+, Polar Electro OY, Kempele, Finland). The VICON^® captures were taken when the heart rate reached 45%, 70%, and 85% of the HRR (20 s for each capture).

2.5. Statistical Variables Definition

One way of analysing the flatness or depth of the thorax is using the thoracic index. This calculation consists of dividing the mediolateral cross-section by the anteroposterior one, so it reflects a percentage variable [35]. A higher thoracic index reflects a flatter thorax, while a lower one exhibits a more rounded ribcage. This index is a “useful tool for evaluating coastal mobility as a component of respiratory mechanics” [36]. Three morphology groups were made according to percentile calculations of the thoracic index of the total sample (Group 1: percentile 0 to percentile 33; Group 2: percentile 33 to percentile 67; Group 3: percentile 67 to percentile 100). Group 1 corresponds to subjects with a flatter thorax, group 3 to the deeper ones, and group 2 to medium thoraxes. To evaluate the thoracic cross-sections from each capture of every subject, ten intervals of one second (120 frames) were analysed. In order to filter the outliers’ data, the maximum and minimum of each thoracic cross-section were determined using the percentiles 95 and 5 of the sample of data recorded during each of the intervals. To avoid the effect of the anthropometry of the subject, we used the increased percentage (IP = ((maximum cross section-minimum cross-section)/(minimum cross-section)) * 100). The mean IP of ten intervals for each running intensity and cross-section were determined. The independent variables were the running intensities and the groups of trunk morphology. The dependent variables were the IP of the different thorax cross-sections.

2.6. Statistical Analysis

A two-factor ANOVA was conducted (thoracic morphology factor with three levels: group 1, 2, and 3; and running intensity factor with three levels: 45% HRR, 70% HRR, and 85% HRR) with each of the three analysed cross-sections. The software used was Jamovi (2.2.5. version). GPower (3.1. version) software was used to estimate the sample needed, a minimum sample of 24 subjects was desirable for this type of variance analysis establishing a power of 0.9 and an effect size (f) of 0.35. The number of subjects recruited was 27 although 5 of them could not be used for the analysis because of data acquisition problems. Only the differences reaching above the effect size established were considered. Levene and Greenhouse–Geisser tests were performed previously. The homogeneity of the Variance Test (Levene’s) showed a homogeneous distribution of the data (p > 0.05), except for ATC_85% and MTC_45% variables. As the majority of the variables showed a normal distribution of data, parametric statistical tests were used. The size of the main effects was analysed using Generalized Squared Eta (ŋ²), with threshold values for small, medium, and large effects being 0.01, 0.06, and 0.14, respectively [37]. Scheffe post hoc test was used for analysing differences among groups. The significance level established for the significant difference in the main effects and the post hoc comparisons was established at p > 0.05.

3. Results

3.1. Craniocaudal Thorax Cross-Section

The effect of the exercise intensity (intrasubject variable) on the craniocaudal cross-section showed no significant results (F2,10 = 3.665, p = 0.104, ŋ² = 0.135). However, a trend of significant statistical increase in the thorax IP was found from 45% to 85% HRR, showing an increase of 0.904% (p = 0.057) (

Table 3. IP results of the craniocaudal thorax cross-section (CTC) among different exercise intensities and divided by morphology groups.

Estimated Marginal Means—CTC * Morphology Group
Morphology Group	CTC	Mean	SE	95% Confidence Interval
				Lower	Upper
1	CTC_45%	2.45	0.536	1.0703	3.83
	CTC_70%	2.22	0.9	−0.0922	4.53
	CTC_85%	3.06	0.997	0.4994	5.62
2	CTC_45%	2.27	0.438	1.1452	3.4
	CTC_70%	2.91	0.735	1.0172	4.79
	CTC_85%	3.24	0.814	1.1442	5.33
3	CTC_45%	1.39	0.438	0.2649	2.52
	CTC_70%	1.91	0.735	0.0183	3.8
	CTC_85%	2.52	0.814	0.4298	4.61
Total	CTC_45%	1.98	0.287	1.31	2.66
	CTC_70%	2.36	0.415	1.38	3.34
	CTC_85%	2.92	0.438	1.89	3.96

SE: standard error; CTC_45%: IP of craniocaudal thorax cross-section at a running intensity of 45% HRR; CTC _70%: IP of craniocaudal thorax cross-section at a running intensity of 70% HRR; CTC _85%: IP of craniocaudal thorax cross-section at a running intensity of 85% HRR. * Statistical difference between IP at 45% HRR and 70% HRR. # Statistical difference between IP at 70% HRR and 85% HRR. ¤ Statistical difference between IP at 45% HRR and 85% HRR.

and Figure S1). The effect of the morphology (inter-subject variable) was not significant (F2,5 = 0.539, p = 0.614, ŋ² = 0.145).

3.2. Anteroposterior Thorax Cross-Section

The effect of the exercise intensity in the anteroposterior cross-section IP was significant (F2,32 = 52.340, p < 0.001, ŋ² = 0.469). The runners’ thoracic expansion increased significantly when the HRR increased from 45% to 85% HRR (2.963%, p < 0.001) and from 70% to 85% HRR (2.268%, p < 0.001), reaching a large effect size. The morphology factor did not show a significant effect on the changes of the anteroposterior cross-section (F2,16 = 0.737, p = 0.494, ŋ² = 0.063) as there were no differences between morphology groups when comparing them in the same intensity (

Table 4. IP results of the anteroposterior thorax cross-section (ATC) among different exercise intensities and divided by morphology groups.

Estimated Marginal Means—ATC * Morphology Group
Morphology Group	ATC	Mean	SE	95% Confidence Interval
				Lower	Upper
1	ATC_45%	4.14 ¤	0.523	3.04	5.25
	ATC_70%	5.41	0.52	4.3	6.51
	ATC_85%	7.29	0.597	6.03	8.56
2	ATC_45%	5.26 ¤	0.619	3.95	6.57
	ATC_70%	5.9	0.616	4.6	7.21
	ATC_85%	8.25	0.706	6.75	9.74
3	ATC_45%	4.83 ¤	0.523	3.72	5.94
	ATC_70%	5.01 #	0.52	3.9	6.11
	ATC_85%	7.58	0.597	6.32	8.85
Total	ATC_45%	4.69 ¤	0.318	4.02	5.36
	ATC_70%	5.39 #	0.309	4.74	6.04
	ATC_85%	7.65	0.353	6.91	8.39

Key: SE: standard error; ATC_45%: IP of anteroposterior thorax cross-section at a running intensity of 45% HRR; ATC _70%: IP of anteroposterior thorax cross-section at a running intensity of 70% HRR; ATC _85%: IP of anteroposterior thorax cross-section at a running intensity of 85% HRR. * Statistical difference between IP at 45% HRR and 70% HRR. # Statistical difference between IP at 70% HRR and 85% HRR. ¤ Statistical difference between IP at 45% HRR and 85% HRR.

and Figure S2).

Interestingly, when pair comparisons were analysed group by group, some significant differences were found: subjects with a deeper thorax (group 3) had similar behaviour to the total group; however, the subjects of group 1 (flatter thorax) and group 2 experienced only a significant increase from 45% to 85% HRR (3.147%, p < 0.001 and 2.986%, p = 0.006, respectively).

3.3. Mediolateral Thorax Cross-Section

The mediolateral cross-section IP showed a statistically significant effect of the exercise intensity when comparing the subjects as a whole without dividing them by morphology groups (intra-subject effect; F2,38 = 35.789, p < 0.001, ŋ² = 0.389), reaching a large effect size. From 45% HRR to 70% HRR, an increase of 1.369% in the mediolateral thorax cross-section was found (p < 0.001). Also, an increase of 2.026% was observed from 45% to 85% HRR (p < 0.001). Plus, a trend of significant difference was found (0.657%, p = 0.060) between 70% and 85% HRR. The main effect of the morphology groups was not significant as there were no differences when comparing them in the same running intensity. The data from the mediolateral cross-section IP are shown in

Table 5. IP results of the mediolateral thorax cross-section (MTC) among different exercise intensities and divided by morphology groups.

Estimated Marginal Means—MTC * Morphology Group
Morphology Group	MTC	Mean	SE	95% Confidence Interval
				Lower	Upper
1	MTC_45%	3.55 ¤	0.383	2.75	4.35
	MTC_70%	5.01	0.419	4.13	5.89
	MTC_85%	5.63	0.481	4.63	6.64
2	MTC_45%	3.31 ¤	0.383	2.51	4.11
	MTC_70%	4.84	0.419	3.97	5.72
	MTC_85%	5.41	0.481	4.4	6.42
3	MTC_45%	3.18 ¤	0.358	2.43	3.93
	MTC_70%	4.29	0.392	3.47	5.11
	MTC_85%	5.07	0.45	4.13	6.01
Total	MTC_45%	3.34 ¤*	0.208	2.91	3.77
	MTC_70%	4.7	0.235	4.21	5.18
	MTC_85%	5.36	0.263	4.81	5.91

Key: SE: standard error; MTC_45%: IP of mediolateral thorax cross-section at a running intensity of 45% HRR; MTC_70%: IP of mediolateral thorax cross-section at a running intensity of 70% HRR; MTC_85%: IP of mediolateral thorax cross-section at a running intensity of 85% HRR. * Statistical difference between IP at 45% HRR and 70% HRR. # Statistical difference between IP at 70% HRR and 85% HRR. ¤ Statistical difference between IP at 45% HRR and 85% HRR.

and Figure S3. However, when pair comparisons were conducted for each morphology group, all groups had the same behaviour, showing a significant increase in IP from 45% to 85% HRR (IP = 2.084%, p = 0.027; IP= 2.103%, p = 0.025 and IP = 1.892%, p = 0.034, respectively).

4. Discussion

In the present study, the variations in thorax expansion through different running intensities and various morphology groups were evaluated using thorax cross-sections recorded by a motion capture system. Our results have partially confirmed the first hypotheses of the study as it was found that the three thorax cross-sections increased when the intensity of the exercise was raised, especially the mediolateral and anteroposterior cross-sections. However, there were no significant differences in thorax expansion in any of the cross-sections between morphology groups when compared at the same running intensity. Although the effect of the thorax morphology was not significant, some influence of it was seen in the changes in the anteroposterior cross-section. The subjects with a flatter thorax (group 1 and group 2) behaved differently than group 3 (deeper thorax) because they did not increase the anteroposterior cross-section when they passed from medium to highest intensity.

It is important to highlight the relevance of our research as a new easy-to-use and ecological approach. This method has been developed for assessing breathing mechanics with applications to sports performance and the assessment of patients with respiratory diseases. In addition, it is important to mention that the results obtained in this study are related to the sample characteristics: young males with low BMI.

We have found that the three cross-sections had a larger expansion of the thorax when subjects were running at an intensity of 85% HRR and a smaller expansion at 45% HRR. These results agree with those obtained by Vogiatzis et al. (2005), where they used OEP during cycling at different intensities, concluding that inspiratory volume increases with the increase of intensity [38]. We have found that the IP of the craniocaudal and anteroposterior thorax cross-sections showed a smaller change from 45% to 70% HRR than from 70% to 85% HRR. However, in the mediolateral cross-section, a bigger change took place from 45% to 70% HRR compared to the change between 70% and 85% HRR, reflecting a quicker change during low- to mid-intensity exercise than during the highest intensity. These results are expected from a physiological point of view [3]. During low- to mid-intensity exercise, the elevation of the lower ribs produced by the contraction of the diaphragm increases the mediolateral cross-section of the thorax, followed by the elevation of the sternum and the upper ribs during high-intensity exercise enhancing the anteroposterior cross-section [3]. Kaneko and Horie (2012) documented differences between the movement of the thorax during quiet breathing and deep breathing in different positions (sitting and standing) without performing any effort [9]. They observed that the expansion of the thorax varies when the breathing pattern is modified [9]. It has also been demonstrated that body posture affects thorax volume variations during breathing due to the position of the ribs, the diaphragm, etc. [28]. This can be the reason for the discrepancies between our data and those obtained by other authors analysing breathing thorax deformation in still postures such as sitting or lying [28].

In the present experiment, we have observed that higher changes of IP on thorax cross-sections happened in the anteroposterior, followed by the mediolateral and the craniocaudal cross-sections. This is partially in agreement with the literature as it is known that bigger changes in the volume of the thorax are produced in the craniocaudal cross-section followed by the anteroposterior and mediolateral; all these changes are caused mainly by the diaphragm and the breathing muscles [3]. The diaphragm is a dome-shaped muscle that closes the thorax from the bottom [39]. In breathing, diaphragm contraction pushes the ribs outwards, generating an expansion of the thorax on the anteroposterior and mediolateral cross-sections, making those cross-sections longer and increasing the ribcage volume [39]. When the diaphragm is contracted, the dome’s shape is flattened, increasing thorax volume for the expansion of the lungs, and enhancing the craniocaudal cross-section [39]. The rise in the craniocaudal cross-section cannot be measured completely with the technique used in the present work because the reflective markers were placed on osseous anatomic locations. This might have influenced our lower results in the craniocaudal cross-section compared with the literature.

The anteroposterior and the mediolateral cross-sections between maximum inspiration and expiration were evaluated by Sarro et al. in 2018 who mainly dismissed the craniocaudal cross-section, finding changes of 2.51–2.79 cm (±1.03 cm) on the anteroposterior cross-section and 2.08–2.36 cm (±0.75 cm) on the mediolateral [21]. Interestingly, our results, which have been expressed as a normalized increment, agree with the earlier data in the sense that the anteroposterior cross-section increases more than the mediolateral. However, if we express them in length units, we can see a smaller increase of 0.97–1.55 cm (± 0.50 cm) on the anteroposterior and 1.00–1.61 cm (± 0.61 cm) on the mediolateral cross-section. The expansion of the thorax increases when the demand for exercise is enhanced but does not reach an expansion similar to that produced when a maximum inspiration and exhale occur. This difference could be explained by the breathing frequency behaviour during high-intensity exercise (67–73 breaths/min at the peak) [40]. In this sense, Nicolò et al. (2020) demonstrated that breathing frequency increases with the intensity of activity, allowing enhanced air volume and oxygen exchange during exercise [40]. When breathing frequency rises, the thoracic ROM oscillates less because inspiration and expiration times are reduced [3]. In addition, during high-intensity exercises, the expiration is not forced due to the high breathing frequency that is needed, so the residual volume (air remaining inside the lungs when the usual expiration is done) remains inside the lungs [3]. This can be the reason for the reduced increase of the thorax volume found in this work when compared with the results from studies that do not observe high-intensity activities or even analyse deep inspirations and expirations.

Unfortunately, most of the results found in the literature about breathing and exercise have obtained changes in breathing volume or other physiological variables which cannot be compared quantitatively with results in this investigation. To the best of our knowledge, there is no information about the influence of high-intensity running on thorax cross-sections using a motion capture system. Therefore, the data presented in our study about breathing mechanics have great potential for measuring the quantification of fitness status and different respiratory disease symptoms.

The only changes between morphology groups occurred on the anteroposterior cross-section, agreeing partially with the results obtained from other studies which conclude that the respiratory biomechanics are altered depending on the depth or flatness of the thorax [33]. However, in our study, deeper thoraxes showed a higher increase in the anteroposterior cross-section when reaching high intensities.

Further, Bastir et al. (2022) and Gómez-Recio et al. (2022) determined a statistical relationship between trunk morphology and locomotor performance during running in an incremental test when the subjects exceeded the anaerobic threshold, finding that a smaller anteroposterior cross-section seems to be positive for better performance during endurance exercise [32,33]. However, Gómez-Recio et al. in 2022 conducted 3D shape analyses and no explicit information about thorax cross-sections can be compared with data containing size in this study. In the future, 3D sizes and shape analyses should be conducted in the context of exercise [33].

This difference among the anteroposterior cross-section of different morphology groups when reaching high-intensity efforts could be explained as a factor used by the deeper thoraxes to combat their morphological disadvantage. Flatter thoraxes have shown a better predisposition for performing endurance efforts, so deeper thoraxes need to expand their anteroposterior cross-section in a way to increase their air volume to perform exercises such as running, cycling, etc.

However, there are some limitations in this study. No physiological variables related to the quality of breathing were analysed. The intensity of the exercise was assessed with a heart rate pulsometer without considering the volume of oxygen consumed. Also, the difficulty of using this technique with female subjects justifies the absence of females in the sample but gives the results a lack of external validity. A method that could be used on females needs to be developed. More investigations into the kinematics of the thorax are needed, especially on the craniocaudal cross-section, for a better understanding of breathing disorders, physical condition evaluations, and breathing diseases. Apart from that, the sample of this study is limited so a bigger one is needed to be recruited for future studies, especially if the effects of morphology need to be evaluated. In this way, the kinematics of the thorax may develop into a parameter in the assessment of quality of life and/or sports performance.

5. Conclusions

In conclusion, we consider that the objectives proposed in this work have been achieved and the results have allowed us to confirm the first hypothesis partially (the increase of the cross-sections of the thorax would be greater when the exercise intensity is increased) and reject the second one (flatter trunks would experience a greater change in breathing patterns). The main conclusions are the following: (1) thorax expansion is produced on the anteroposterior and mediolateral cross-sections when running intensity rises, but (2) the craniocaudal cross-section of the thorax is not affected when it is observed using only osseous-located markers (although the craniocaudal cross-sections increases by the flattening of the diaphragm) and (3) the influence of the morphology of the thorax on its expansion during high-intensity exercise was not significant, but (4) a possible means of combatting the disadvantage during breathing was found in subjects with a deeper thorax when reaching the highest intensities.