Warm-up is a fundamental moment in which both recreational and professional athletes prepare themselves physically and psychologically to perform intense physical activity. It has been observed that various types of warm-ups can reduce the risk of injuries in the lower limbs (Herman et al., 2012; Behm et al., 2016). The implementation of stretching exercises was explored in the past, with controversial findings, describing whether static or dynamic stretching was useful or not in preparing the muscles for subsequent activity with the potential to reduce the incidence of injuries (Herman et al., 2012; Behm et al., 2016; Opplert and Babault, 2018; Chaabene et al., 2019).

Change of direction (COD) exercises require a high physical demand due to the forces that converge on the different anatomical structures in the area of the knee both for the joint (Donelon et al., 2020) and for the muscles (Kekelekis et al., 2022). However, they are key movements to master for athletes to reach a high peak of performance, especially in team sports (Loturco et al., 2022), where performing them with high efficacy is considered an indicator of talent (Scanlan et al., 2021). The demands on the knee, when performing a COD exercise, highly depend on the velocity and angle of the action, with 90 degree COD determining higher braking and propulsive forces than other angles (Schot et al., 1995). Therefore, it is fundamental to choose the right warm-up for athletes before they approach physical activities involving COD exercises, considering that this maneuver has been considered one of the main actions resulting in non-contact injuries of the anterior cruciate ligament in different sports (Waldén et al., 2015; Montgomery et al., 2018).

Nowadays, innovative technologies provide researchers and clinicians new tools to use and new data to analyze, allowing an improvement in the understanding of human physiology both under static (Singla and Veqar, 2014) and dynamic (Roggio et al., 2021) conditions. One of the most interesting and promising techniques in this field is infrared thermography (IRT), which is used in different kinds of evaluations, from clinical medicine to sports and exercise physiology (Hildebrandt et al., 2010; Ring and Ammer, 2012). IRT evaluations are easy to perform, and the technique is non-harmful, highly reproducible (Zaproudina et al., 2008; da Silva et al., 2022), and reliable under different conditions when performed following specific protocols for thermal data acquisition (Moreira et al., 2017). IRT cameras use a sophisticated form of the radiation thermometer to detect heat radiation emitted by the human body within an electromagnetic spectrum that is not visible to the human eye (Costa et al., 2016). By identifying the surface temperature, these cameras provide insights into the blood flow and muscle metabolism, which, in turn, enhance the convection of heat across various regions of the body (del et al., 2017). The outcome measurement evaluated by this technique is the skin temperature (Tsk) distribution of the region of interest (ROI) framed by the camera (Jones and Plassmann, 2002). Studies support the application of IRT in the detection of Tsk changes after different kinds of exercises (Fernandes et al., 2016; Ferreira-Júnior et al., 2021; Alburquerque et al., 2022), as an imaging strategy to reduce the incidence of muscle injuries (Côrte et al., 2019; Gómez-Carmona et al., 2020) and to monitor the fatigue after specific warm-up exercises (Hadžić et al., 2019).

Past studies investigated the effects of static or dynamic stretching on COD performance (Beckett et al., 2009; Wong et al., 2011; Turki et al., 2019) in terms of velocity, power, and biomechanics improvement. To date, the thermal response of the anatomical structures in the area of the knee after a COD exercise, anticipated by a static or dynamic stretching warm-up, was never explored. These structures are fundamental while practicing different athletic movements, from individual to team sports. To optimize the warm-up routines, considering the thermal response of the knee is fundamental to improve the wellness of athletes and help coaches in tailoring to their training interventions. Moreover, this study will provide researchers with further insights into the thermoregulatory response of the knee and new perspectives on how to adjust the warm-up interventions to better suit the physical demands of exercise. Hence, the aim of this study was to investigate the efficacy of static and dynamic stretching warm-ups in preparing the knee for a 90° COD exercise, utilizing IRT. Additionally, the study aimed to analyze the thermal responses in healthy young adults, differentiating between genders and types of stretching.

This pre–post-observational study involved 85 healthy young adults (46 men [M] and 39 women [F]) aged 20–31 years and evaluated the Tsk of the knees with IRT at four time points before and after different types of exercises. The study was approved by the Research Center on Motor Activities (CRAM) Scientific Committee (Protocol no. CRAM-020-2021, 20/12/2021) in accordance with the Declaration of Helsinki. Informed consent was obtained from all subjects before participation.

The characteristics of the participants, in terms of means and standard deviations for anthropometrics, BMI, age, and temperature at baseline, are presented in Table 1. The IRT results of all the ROIs examined in both the anterior and posterior coronal planes in means, divided by gender, and type of stretching, for the right knee, are presented in Table 2.

All data respected the assumption of normality and homogeneity in the Shapiro–Wilk test and Levene’s test, respectively. The temperature of all four ROIs analyzed was higher in men at each of the time points T0, T1, T2, and T3 for both the right and left knees. The comparison between the right and left knees showed no significant differences, and the repeated-measure ANOVA results between the right and left knees were not significant. Hence, only the results for the right knee are presented. The general trend of temperature was a slight decrease in temperature at T1, followed by a higher decrease at T2 and a recovery of temperature, near or superior to the baseline at T3. To identify statistical significance, a mixed ANOVA was used to analyze the four ROIs individually and for gender separately. The results of the mixed ANOVA are presented in Table 3.

The type of stretching did not show significant results in the four ROIs examined, with p-values >0.05. However, the variable stretching/time points showed significant results in every ROI aside from the AUH. To spot the significance through pairwise comparison, the Bonferroni–Holm post hoc test was used, highlighting various significant meaningful comparisons, as shown in Table 4. A significant difference was found between T1 static and T1 dynamic in the ALH for both men and women and the PUH only in men. The dynamic stretching was not significant in all the ROIs examined, while the static stretching elicited a decrease in temperature in the ALH aside from gender and in the PUH only in men. Significant differences were spotted between T0 and T2 in the dynamic group in the ALH for men and in the static group for both men and women in the ALH and PLH and only in men in the PUH. Figure 2 presents the variation in temperature in time points for stretching and gender in the ROIs examined.

In this study, the thermal response of the knee after a static or dynamic warm-up and a 90° COD exercise was investigated in a cohort of healthy, young male and female adults. It was found that men presented a higher temperature at the baseline than women. The evaluation of time points showed a general trend of a slight decrease in temperature after warm-up, followed by a higher decrease after the 90° COD exercise and an increase in temperature with a return to baseline or slightly higher at rest. The static stretching warm-up elicited this phenomenon in the ALH in both men and women and in the PUH only in men; instead, the dynamic warm-up did not show significant results in all the ROIs examined. Moreover, the comparison between the static and dynamic groups at T1 showed significant results in the ALH aside from gender and in the PUH in men.

The Tsk is influenced by many factors (Fernández-Cuevas et al., 2015), and differences in temperature at rest between men and women in different body regions have already been investigated in different studies (Chudecka and Lubkowska, 2015; Salamunes et al., 2017). Our study aligns with a previous study that found that men have a higher Tsk than women in the lower limbs (Neves et al., 2017). Women have higher subcutaneous fat percentages than men (Karastergiou et al., 2012), and this characteristic affects the Tsk detection through IRT (Reis et al., 2023). Moreover, the metabolic rate is also higher in men than women (Lazzer et al., 2010), fostering this thermal difference in areas of the body where men typically have more muscle tissue. Marins et al. (2014) also found that there are significant differences in the temperature of the lower limbs between men and women, especially in the calves and ventral and posterior thighs. Thus, it is essential to consider gender when performing IRT acquisition regarding lower limbs.

The general trend of our results showed a slight decrease in Tsk after warm-up exercise and a higher reduction after a more intense COD exercise, followed by an increase in temperature after 20 min of rest, especially in the static group. This temperature trend after acute exercise is in line with different study results (Torii et al., 1992; Fernández-Cuevas et al., 2014; Chudecka et al., 2015; Moreira-Marconi et al., 2019; Lino-Samaniego et al., 2022; Sillero-Quintana et al., 2022), which highlighted a decrease in temperature immediately post-exercise and an increase during the rest period, pursuant to different types of physical activities. When administering incremental exercise, the temperature decreases at the start, with a progressive decrease until exhaustion, and no increase is observed until the end of exercise during rest (Hillen et al., 2020). IRT allows us to detect Tsk, which is the result of the heat transfer of deeper tissue to the upper layers and blood flow (Ammer and Formenti, 2016). The decrease in temperature after acute exercise is caused by the physiological vasodilatation response due to the skin that reflexively contracts as blood is redistributed from non-working tissues to the working skeletal muscle (Charkoudian, 2010). Another physiological reason behind this trend is the start of the thermoregulatory response, particularly sweating, that causes the Tsk to decrease during the first few minutes of exercise compared to thermograms taken at other times; subsequently, during exercise, the Tsk will increase progressively (Fernandes Ade et al., 2014).

We did not find any statistical difference in temperature between the right and left knees. This could be explained by the nature of the exercise that made both legs physically active, considering that temperature follows the same trend in the non-exercised leg (Ferreira et al., 2008).

The physical demand of a sharp 90° COD exercise requires high muscle activation on the biceps femoris and vastus lateralis, while the quadriceps is fundamental for the deceleration phase (Dos’Santos et al., 2018). Moreover, the hamstrings do the essential job of stabilizing the knee to protect it from abnormal movement that can lead to an injury (Neptune et al., 1999; Besier et al., 2003).

Williams et al. (2015) found that static stretching can elicit differences in the quadriceps torque output, analyzing their sample with surface electromyography, and another study (Amiri-Khorasani and Kellis, 2013) found that dynamic stretching can elicit significant differences in electromyographic activity of the rectus femoris and vastus medialis and lateralis.

Conversely, we found that neither static nor dynamic stretching elicited a change in temperature in the AUH. This may be due to the intensity of the warm-up, which was not high enough to prepare one of the biggest muscles in the body for activity in terms of temperature. Generally, static stretching focused on the quadriceps can elicit positive changes in it (Williams et al., 2015); however, the static stretching in our study was focused not only on the quadriceps but also on all the main muscle groups of the legs that contribute to the stabilization of the knee. A lowering of Tsk can be interpreted as a redirection in blood flow from the skin vessels to the muscles (Tanda, 2015), suggesting that muscles might be more ready to perform the exercise. Differently from our results, de Oliveira et al. (2018) observed an increase in muscle temperature of the hamstrings after 180 s of static stretching. However, their study was performed in laboratory settings, while our study focused on testing a stretching warm-up routine closer to that performed before training. Moreover, the authors evaluated all the posterior sections of the thigh, while we analyzed only the insertion of the hamstring close to the knee.

This study has some limitations. First, we examined a sample of people who perform physical activity as a recreational activity, so these findings should be interpreted carefully when comparing with other populations such as athletes. Second, we used the BMI instead of the body fat percentage as an exclusion criterion, which can impact the IRT results. The observed differences in thermal response between men and women suggest that customized warm-up routines might be beneficial for sport practitioners. Thus, incorporating tailored stretching exercises could be more effective in preparing the muscles, potentially reducing injury risk during high-intensity maneuvers like 90° COD exercises. Future studies should evaluate the knee thermal response to stretching in professional athletes and change of direction exercise during a training period and explore possible connections between abnormal thermal response to exercise and the rates of injuries.

This study presents an exploration into the effects of two distinct stretching warm-ups and a 90° COD exercise on the thermal response of the upper and lower regions of the knee, utilizing IRT. We found that men exhibited a higher Tsk than women. Notably, the static stretching warm-up elicited a significant reduction in Tsk compared to dynamic stretching, particularly in the anterior lower and posterior upper regions of the knee for men and only in the anterior lower region for women. These findings provide valuable practical information for trainers in determining the most suitable stretching warm-up for their athletes tailored to the physical demands required by specific exercises. Coaches can use this information to customize warm-up protocols that optimize knee temperature, preparing their athletes with more precision for subsequent performances, attempting to reduce the risk of potential injuries. Moreover, the results of this study may be useful for researchers investigating the thermal responses of the knee after different kinds of moderate and more intensive exercises.