This research project applied a crossover study design (see Figure 1). Participants were randomly assigned to two experimental groups. Before the start of the study, athletes from both groups were familiarized with CT procedures. While experimental group I performed power training before endurance training (PE) in one exercise session, group II followed the sequencing scheme endurance training before power training (EP). After a wash-out period of 2 weeks with regular and sport-specific exercise routines but without CT and a day off prior to session 2, group I followed the sequencing scheme of group II and vice versa. The training intensity was not quantified during wash-out phase. Over a period of four nights, athletes were monitored for changes in capillary immunological blood markers. For this purpose, blood samples were collected in the evening and morning before (BL; BL_pre, BL_post) and after (INT; INT_pre, INT_post) the intervention, as well as immediately following the CT exercise session (Ex_post). Because of the COVID-19 restrictions, participants had to arrive in three groups between 7:00 a.m. and 11:30 a.m. The BL_pre-measurements were scheduled between 5:00 p.m. and 7:00 p.m. and INT_pre 6 h after the training sessions.

Our participants were recruited from a youth academy judo performance center. Twenty male judo athletes aged 16.0 ± 1.8 years (body height: 171.2 ± 8.0 cm, body mass: 64.7 ± 11.6 kg) were recruited according to the conceptual model of long-term athlete development (as described by Granacher et al. (2016)). Participants’ chronological age ranged from 13 to 19 years. The maturity status was assessed using Tanner stages and all athletes were classified (Tanner stages 3–4). The training and performance calibre of the recruited athletes was highly-trained, national level (Tier 3) (McKay et al., 2022). Thus, the enrolled athletes regularly participated in national and/or international competitions. The participating athletes performed two daily training sessions on 5 days per week. Each training session lasted one to 2 h. Additionally, individuals with acute injuries or those reporting infectious diseases before or at any point during the experimental period were excluded from study participation. Due to the complexity of our statistical approach, statistical power was not conducted a priori. Participants and their legal representatives were informed about the experimental procedures, risks and benefits of the study. Written informed consent was obtained from the participating athletes and their legal representatives. This study was approved by the University Human Ethics Committee of the Potsdam University, Germany.

All athletes performed a warm-up routine with exercises from the FIFA 11+ program. Afterwards, CT was started in the group specific order. The muscle power exercise routine was performed on a leg-press machine (SCHNELL, Peutenhausen, Germany) with four sets of eight repetitions at 30%–40% of the individual’s 1-repetition maximum. Repetitions were carried out as fast (explosive) as possible. A 4 min rest between the sets was allowed. The exercise protocol is in line with the recommendations from the national strength and conditioning association (Bergeron et al., 2015). The Special-Judo-Fitness-Test (Sterkowicz et al., 1999) was used as sport-specific anaerobic endurance exercise. For this purpose, two opponents were placed 12 m apart and the athletes had to be thrown as often as possible with the ippon-seoinage technique (Franchini et al., 2011b). Ippon-seoinage is one of the many throwing techniques (nage-waza) in judo and the term can be translated to “shoulder throw.” Participants performed four rounds consisting of three sets with 15, 30, and 30 s duration with a break of 10 s between the sets and 4 min between the rounds. The total work time for each exercise was approximately 25 min.

WBC, GRAN, LYM, monocytes (MON), hemoglobin (HGB), hematocrit (HCT) and platelets (PLT) were measured using a 20 µL capillary blood sample taken from the earlobe. Analysis was performed using a medonic hematology system (Medonic M32, Boule Medical AB, Spånga, Sweden). Granulocytes are primarily involved in the acute inflammatory response and serve as the first line of defense against pathogens. Lymphocytes are central to the adaptive immune system, with T cells playing roles in cell-mediated immunity and B cells responsible for producing antibodies that target specific antigens. This specificity allows for a long-lasting immune memory against previously encountered pathogens. Monocytes, which can differentiate into macrophages and dendritic cells upon migration into tissues, are essential for phagocytosis, antigen presentation, and the regulation of immune responses. They act as a bridge between the innate and adaptive immune systems, helping to initiate and shape the immune response. Taking changes in plasma volume following exercise into account, capillary blood markers were corrected with respect to Dill and Costill (Dill and Costill, 1974) using the following formula: P l a s m a v o l u m e c h a n g e % = 100 x H G B p r e H G B p o s t x 1 − 0 , 01 * H C T p o s t 1 − 0 , 01 * H C T p r e − 100 C o r r e c t e d   v a l u e s = n o t   c o r r e t e d   v a l u e s   x   100 − P l a s m a   v o l u m e   c h a n g e   % 100

Corrected values were used to calculate the systemic inflammation index (SII). Therefore, the following formula (Hu et al., 2014) was used: S I I = P L T * G R A N L Y M

Furthermore, the systemic inflammatory response index (SIRI) was calculated with reference to (Yi et al., 2021; Bai et al., 2024): S I R I = G R A N * M O N L Y M

The raw data is included in Supplementary Material 1 and plasma-volume corrected data are presented in Figure 2. The descriptive statistics can be found in Supplementary Material 2. To account for repeated measurements and randomly missing data, order (PE, EP) ^* time (BL, INT) interactions were analyzed using a generalized estimating equation (GEE) approach. GEE is known to produce robust parONameter estimates and standard errors (Zeger and Liang, 1986; Burton et al., 1998). Plasma volume-corrected evening to morning change-scores for BL and INT were reported as marginal means and 95% confidence intervals. BL was used as baseline. Change-scores of INT were referenced to BL change scores using z-transformation. Capillary blood markers were entered separately as dependent variables. Order, time and order ^* time were used as independent variables. Within subjects, dependencies were modeled as first-order autoregressive and a Gaussian link function was used. With reference to Fritz et al. (2012), effect sizes were classified as large, medium and small effects (0.5; 0.3; 0.1). The significance level was set to α = 5%. To account for multiple comparison, we reported Bonferroni-Holm corrected p-values. Data analyses was performed using R version 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria) and “geepack” package (Yan, 2002; Yan and Fine, 2004; Halekoh et al., 2006).

The present study explored the sequencing effects of CT either applied as PE or EP on immunological recovery in young male judo athletes. The key findings revealed significant order-by-time interactions for WBC, GRAN, SII and SIRI. However, after Bonferroni Holm correction, only GRAN and SIRI remained significantly affected. These results suggest that the sequence of power and endurance exercises within CT plays a crucial role in influencing immunological regulation overnight.

Strength training is known to result in an increased release of interleukin-6 (IL-6) from muscle fibers, which facilitates immunological regulation for the repair of damaged muscle tissue through satellite cells (Izquierdo et al., 2009; Ihalainen et al., 2017). The increase in IL-6 is dependent on the type of mechanical load. Ihalainen et al. (2014) demonstrated that an accentuated eccentric load, as in hypertrophy training, leads to a greater release of IL-6 and cortisol compared to maximal strength training within a 30 min period post training. Intense endurance training induces an increase in cortisol levels, which rise several hours post-exercise, causing a delayed immunological stress response, characterized by lymphopenia and neutrophilia (Gabriel and Kindermann, 1997). Although the total load is the same in both CT sequencing schemes, significant differences in overnight immunological regulation are evident. The PE sequence resulted in a stronger immunological response overnight. This varying response is likely due to the restoration of immunological homeostasis during the night. The underlying mechanism for the different stress responses during the day and the resulting overnight regulation remains unknown. Due to the different immunological effects of EP and PE within CT, EP appears to cause a smaller disruption of immunological homeostasis compared to PE. CT order influences immunological systemic recovery overnight, which highlights its importance for health prevention. Therefore, EP should be preferred, especially in the presence of mild upper respiratory symptoms.

CT and the sequencing of CT influenced myeloid cells (GRAN and MON) but had no effect on lymphoid cells (LYM). This suggests that CT impacts the immunological regulation of the innate immune system but not the adaptive immune system. GRAN play a crucial role in countering bacterial infections. Although GRAN, such as neutrophil granulocytes, operate through phagocytosis, they are not classified as antigen-presenting cells (APCs), which bridge the innate and adaptive immune systems by facilitating T-cell activation. Monocytes, in contrast, are highly dynamic and play a dual role in immune regulation and tissue repair. Exercise-induced mobilization of monocytes often results in their infiltration into skeletal muscle, where they differentiate into tissue-resident macrophages that support repair and regeneration, particularly after intense exercise causing muscle damage (Peake et al., 2017). Monocytes also differentiate into dendritic cells, which, along with macrophages, are among the most important antigen-presenting cells (APCs) responsible for activating the adaptive immune system. In our study, the observed effect of CT on GRAN but not MON suggests that CT primarily disrupts immunological homeostasis within the innate immune system, without significantly affecting the complex, context-dependent roles of monocytes in inflammation and tissue repair.

The observed order-by-time interaction in GRAN may be linked to altered endocrine responses after the training session. Notably, the power-endurance CT showed a higher increase in GRAN on the evening after the exercise compared to endurance-power CT. This differential response might be indicative of varying physiological stressors imposed by different exercise sequences, impacting the immune system differently (Bessa et al., 2016). Furthermore, the CT order affects immunological systemic recovery overnight, which highlights its importance for health prevention.

These findings align with prior work of (Cadore et al., 2014) who delved into the immune responses to CT. The study supports the notion that understanding the order of power and endurance exercises is paramount for optimizing the immune response in athletes. The present study builds upon this foundation by specifically focusing on young male athletes and emphasizing the importance of immunological recovery overnight.

An examination of the raw data in Figure 2 reveals that values at the INT_post time point return to levels observed at BL_post. While this may suggest that the CT sequencing effect has only a short-term impact, considering the findings of Steidten et al. (2021), it is expected that differential overnight immunological regulation could result in a reduction after three consecutive days of CT. This reduction, likely influenced by varying levels of immune system strain, may also exhibit an interaction effect. However, based on the present data, this remains speculative.

Considering the intricate relationship between sleep and immune function, the observed changes in immunological markers may also be influenced by sleep behavior. Sleep quality and duration are known to impact immune parameters (Prather et al., 2015; Besedovsky et al., 2019). Therefore, quantitative variations in immunological markers observed in the study could be attributed, at least in part, to differences in sleep patterns among participants.

This aligns with the sleep-immune interactions highlighted by Steidten et al. (2021). The study emphasizes the need for considering sleep behavior in investigations related to immune responses in athletes. Future research should delve deeper into the interplay between exercise sequencing, sleep, and immunological recovery, as understanding these multifaceted relationships is crucial for optimizing training protocols.

In addition to sleep, recent research has highlighted the impact of dietary intake and meal timing on our immune system. Jordan et al. (2019) have indicated that fasting periods, specifically those lasting around 16 h, can prompt monocytes to migrate to the bone marrow, where they undergo regeneration. This finding suggests a link between dietary habits and immune cell dynamics and the need for standardized nutritional patterns. However, in this study, caloric intake and meal timing were not standardized due to the challenges posed by the COVID-19 pandemic and the associated individual measurements. Consequently, the results of this study may be influenced by the variations in meal times and caloric intake among participants.

The implications of these findings for training strategies applied in young male judo athletes are substantial. Given the observed differences in immunological responses based on the order of power and endurance exercises, coaches and strength and conditioning specialists should tailor training programs to prioritize immunological recovery. Specifically, when immunological stress is already present, the study suggests that the EP sequencing scheme may be the preferred option, as it results in less immune system disruption compared to the PE sequencing scheme. However, for healthy athletes without signs of immune stress or upper respiratory symptoms, both sequencing options (EP and PE) allow for effective immunological recovery overnight, returning to baseline levels by the morning. This means that, under normal circumstances, healthy athletes are unlikely to experience significant immune differences the following day. On the other hand, when signs of mild upper respiratory symptoms, such as a sore throat or runny nose, are present, the study recommends choosing the endurance-first sequencing scheme (EP) to minimize the immunological burden and ensure better recovery. This consideration is crucial in optimizing training and recovery strategies to support long-term athlete health and performance.

Moreover, incorporating interventions to enhance sleep quality and duration may further contribute to optimal immunological recovery in young athletes. This aligns with the broader context of ensuring the holistic development of young sports talents, as emphasized in the introduction.

Despite the insights provided by this study, it is essential to acknowledge its limitations. The small sample size and the focus on male judo athletes from a national training center may limit the generalizability of the findings. Previous research indicates that women typically perform more repetitions at submaximal loads, such as 80% of 1-RM, compared to men. This is largely due to gender-specific differences in muscle fatigue resistance and fiber composition, with women often exhibiting a greater proportion of Type I muscle fibers, which are more resistant to fatigue (Falch et al., 2023). To ensure more consistent and interpretable results, we focused on male athletes. However, similar results are expected for females.

A classic cross-over design was not feasible in our study due to practical limitations, meaning that we could not fully assess intersubject variability as no participant could complete both sequences (A,B and B,A; respectively EP-PE and PE-EP). Consequently, we opted for a between-subject design. Given these conditions, it was predetermined that a pooled analysis using a GEE model represented one of the most appropriate methodological choices. Statistical power is influenced by both sample size and the number of repeated measures. However, our results can only be interpreted in the context of the overall group of participants. An interpretation regarding intrasubject effects is not possible based on our analysis.

He Generalized Estimating Equations (GEE) method does not require a specific distribution of outcome measures. GEE is suitable for small sample sizes (n = 8) and repeated measures (n = 6), achieving approximately 80% power (Ma et al., 2012). A post hoc power analysis was performed for WBC with an order × time interaction effect of 0.92 and a Type I error rate of 0.05. The results indicated a statistical power of 100%.

Future research should aim for larger and more diverse samples to strengthen the external validity of the results.

Additionally, further investigations should explore the underlying mechanisms of altered immune responses and regulation overnight. Longitudinal studies tracking sleep patterns, hormonal fluctuations, and immune responses over an extended period would offer a more comprehensive understanding of these complex interactions.