Journal Logo


Association of Military Training with Oxidative Stress and Overreaching


Author Information
Medicine & Science in Sports & Exercise: August 2011 - Volume 43 - Issue 8 - p 1552-1560
doi: 10.1249/MSS.0b013e3182106d81
  • Free


In competitive sports and military training, the goal of the training is to improve performance. However, both prolonged aerobic and strength training with high intensity or volume together with inadequate recovery could lead to nonfunctional overreaching (OR) resulting in stagnation or decrements in performance capacity (16). OR could also lead to the development of overtraining syndrome (OTS), both of which are characterized by an unexpected decrease or at least stagnation in performance despite increased or sustained training load. The time required for the recovery from OR takes several days to several weeks, whereas the development of OTS and recovery from it may take a much longer time.

Several factors and symptoms have been linked to OTS. However, there are no reliable and specific markers of impending OTS nor knowledge on how to identify individuals who are more vulnerable to OR. Moreover, the precise mechanisms leading to OR and OTS have not yet been delineated. Recently, impaired antioxidant capacity and increased oxidative stress, a state in which the production of reactive oxygen species (ROS) overwhelms antioxidant defenses, were found to be related to OTS in elite athletes (28). Animal (18,36) and human studies have provided evidence that a progressively increased training volume, which causes symptoms of OR or OTS, can increase oxidative stress and attenuate antioxidant capacity (15,35). Whereas a single bout of physical exercise of sufficient intensity and duration generates ROS, both aerobic (21) and anaerobic training (2) have been shown to enhance antioxidant status and decrease the generation of ROS. In contrast, overload training can lead to an impaired antioxidant defense and lack of expected adaptations to training (25) and distortion of the redox balance (8). Extreme increases in training volumes may lead to a substantial rise in inflammation and apoptosis markers (7). Therefore, overload training may induce inflammation, which is also associated with increased oxidative stress.

We hypothesized that increased oxidative stress and disrupted redox balance in response to heavy physical training may be predisposing factors and markers for OR. For this purpose, we aimed to study the association of oxidative stress markers and antioxidant status with physical activity during an 8-wk military basic training (BT) period. BT comprises the first 8 wk of military service and includes both endurance and strength types of training. A particular aim was to evaluate whether the levels of oxidative stress markers and antioxidant status and physical activity differ between OR and non-OR subjects.



Male conscripts (N = 35, age 19.6 ± 0.3 yr) participated in the present study. They were a part of a larger group of 60 soldiers who initially volunteered to participate and fulfilled selection criteria (29). All subjects were fully informed of the experimental protocol and gave their written consent to participate in the study. They were also advised of their right to withdraw from the investigation at any time. The study protocol was approved by the Finnish Defence Forces and the ethical committees of the University of Jyväskylä and the Kainuu region of Finland. The study took place during winter in Finland, when daily outdoor temperatures ranged from −31°C to +1°C, with an average of −13°C (data from the local weather station). The overall physical load of the 8-wk BT period was set according to the standard basic program (31). It included a total of 300 h of military training, of which 100 h was military-related physical training, 33 h was sports-related physical training such as combat training and marching, and the rest of the hours were other military training such as shooting, material handling, skill training, and general military education (23). The intensity level of physical activity in the daily program was planned to be low in the first week of BT and increased gradually thereafter. The conscripts slept from 10:00 p.m. to 5:45 a.m. in dormitory-type rooms at the garrison and marched four times per day (approximately 5 km in total) to dining. The BT schedule included four longer (from 2 to 8 h) marching exercises with a combat gear and two overnight field exercises from 1 to 3 d. Food and water intake were in accordance with the standard army meal, and water intake was not restricted. However, subjects were not permitted to use any extra nutritional supplements throughout the study.

Experimental protocol.

Performance tests (maximal aerobic uptake (V˙O2max) and submaximal exercise tests) were performed, and oxidative stress and antioxidant status were determined three times, and psychological markers (questionnaires) were determined five times during the 8-wk training period. In addition, acute responses to a submaximal exercise were studied. For the submaximal exercise, a marching test with a 20-kg backpack was selected because it closely resembles the routine activities of conscripts and enables the test to be performed for a large group of subjects. The experimental protocol is presented in Table 1.

Experimental design during the 8-wk military BT period.

Performance tests.

An incremental test until voluntary exhaustion to determine maximal oxygen uptake (V˙O2max) was carried out on a treadmill at baseline and on weeks 5 and 8 as described earlier (29). The 45-min submaximal exercise test was performed on an outdoor track at the 70% level of the subject's individual maximal workload in the beginning (baseline) of BT and on weeks 4 and 7. The follow-up tests for each subject were always performed at the same time of day (V˙O2max test between 8:30 a.m. and 5:00 p.m., submaximal exercise between 9:00 a.m. and 12:00 noon) with the subjects having consumed a similar diet before each test and with the same test protocol. The hormonal responses to submaximal exercise have previously been reported (30).

Body composition.

Body composition measurements (body mass (BM), fat-free mass, fat mass (FM), and percentage of body fat) were performed using eight-point bioelectrical impedance (InBody720; Biospace Co., Ltd., Seoul, Korea). For each subject, the repeated measurements were performed at the same time between 6:00 and 7:00 a.m. after an overnight fast and after voiding, with no exercise for 12 h before the test. The physical activities in the daily program were planned to be of a low intensity on the day preceding each measurement, and fluid status was estimated to be balanced on the basis of the dietary records of the subjects (29). The subjects were barefoot and wore T-shirts and trousers. Body height was measured to the nearest 0.5 cm using a wall-mounted stadiometer. Body mass index (BMI) was calculated as BM (kg) divided by height (m) squared.

Blood samples and analyses.

Blood samples from an antecubital vein and fingertip were drawn 2 h after a light breakfast at rest, before submaximal exercise, and immediately after exercise. Circadian variability in blood parameters was minimized by collecting individual preexercise and postexercise samples at the same time of day between 9:00 a.m. and 12:00 noon preceded by similar patterns of food ingestion. Subjects were instructed not to ingest alcohol, coffee, tea, chocolate, or cola drinks since the evening before the measurements.

Blood samples were centrifuged at 1200g and 4°C for 15 min immediately after collection to separate the plasma. Plasma samples were stored in multiple portions at −80°C until analysis. Protein carbonyls, markers of protein oxidative damage, were measured using an ELISA method as previously described (19). Nitrotyrosine concentrations were determined with the ELISA method using a commercial kit (Hycult Biotechnology BV, Uden, The Netherlands). Lipid peroxidation marker total malondialdehyde in plasma was measured according to the method of Gérard-Monnier et al. (9). Oxygen radical absorbance capacity (ORAC) was used for the measurement of antioxidant capacity, which was performed using a multiwell plate reader according to the methods described previously (13). The maximal intra-assay coefficients of variation for protein carbonyls, nitrotyrosine, malondialdehyde, and ORAC were 5.9%, 10.0%, 6.2%, and 8.1%, respectively, and the maximal interassay coefficients of variation were 9.2%, 12.7%, 10.6%, and 11.3%, respectively. Protein carbonyls and ORAC measurements were performed in triplicate. Total glutathione (TGSH) and oxidative stress marker oxidized glutathione (GSSG) concentrations were determined spectrophotometrically as described previously (24). The tissues were deproteinized with metaphosphorous acid for TGSH and GSSG analysis. The maximal intra-assay and interassay coeffcients of variance were 4.8% and 6.4%. Hemoglobin and hematocrit were analyzed using a Sysmex KX-21N analyzer (Sysmex Co., Kobe, Japan). Blood lactate was analyzed from fingertip blood sample using Lactate Pro® analyzer (ARKRAY, Kyoto, Japan). Submaximal exercise-induced changes in plasma volume were calculated from changes in hemoglobin and hematocrit (6), and malondialdehyde, nitrotyrosine, and ORAC postexercise values were reported adjusted for these changes. Plasma protein carbonyl results were expressed in nanomoles per milligram of protein. The BT and/or exercise-induced relative changes are expressed as percent change.

Questionnaires and sick leave.

The subjects rated how severely they experienced a list of symptoms during the last week using a five-point Likert scale: 1, not at all; 2, 1 d; 3, 2-3 d; 4, 4-5 d; and 5, 6-7 d. The somatic symptoms were subjective ratings of well-being, upper respiratory track infections, flu-like symptoms, digestive disorders and reduced appetite, musculoskeletal and physical complaints, and sleep difficulties (30). The sum of the symptoms associated with OTS was determined as the sum of the symptom scores. In addition, the question "do you feel physically or mentally overloaded?" was asked. Sick leave was defined as an attendance/nonattendance in daily service due to illnesses or injuries that were evaluated by a physician.

Physical activity.

Physical activity (PA) was measured from 18 subjects with a customized version of the Polar AW200 Activity monitor (Polar Electro Oy, Kempele, Finland) that was worn on the nondominant wrist. AW200 has been found useful and accurate for the measurement of energy expenditure during long-term exercise (3). AW200 contains a uniaxial accelerometer. The acceleration signal is band-pass filtered (0.3-3.0 Hz), and the device has reduced sensitivity to repeated low-intensity hand movements. The device counts hand movements if acceleration exceeds 1.0 m·s−2 (12). Epoch length was set at 1 min, and a curvilinear equation was used to transform activity counts to metabolic equivalents (1-16 METs), which were further adjusted by body height. Among the same subjects, AW200 weekly physical activity energy expenditure correlated well (r = 0.78) to that obtained with doubly labeled water (unpublished data, see Tanskanen et al. (29)). Daytime was determined from 6:00 a.m. to 9:00 p.m., and nighttime, from 9:00 p.m. to 6:00 a.m. Periods that contained no single movement during 30 min in daytime or during 6 h or more in nighttime were classified as nonwear time and excluded from analysis. PA by each minute was classifies into 1) no activity (REST), ≤1.0 MET; 2) very light to light activity (VLPA), 1.0-3.9 METs; or 3) moderate to vigorous physical activity (MVPA), ≥4 METs. REST (during daytime) or sleep (during nighttime) was selected if the accelerometer showed no hand movements within >50% of a 10-min moving window. Each subject's daily and nightly data were included for analysis only if activity recording covered >80% of daytime and 90% of nighttime. Measures of PA were adjusted by the proportion of time recorded. The activity watches were collected every evening between 9:00 and 10:00 p.m. for data download and redistributed within 30 min. Out of a total of 41 d, we included days and nights when at least two-thirds of subjects had enough data. In all, 33 d and 28 nights were included in the analysis. Data were pooled into 2-wk periods: weeks 1-2, 3-4, 5-6, and 7-8.

Criteria for OR.

In this study, subjects had to fulfill three of five criteria to be classified as OR subjects (30). Criteria 1 was a reduced V˙O2max of >5% (11,26) or nonperformance of the test because of illness. Absence from thetest was set as an additional criterion because all three V˙O2max tests were completed by a total of 21 of 35 conscripts. It was assumed that illness itself reduces performance. Furthermore, overload training has been reported as a risk factor for upper respiratory track infections (27,34). Criteria 2 was an increase in mean RPE during the submaximal exercise >1.0 (11) from the lowest value at week 1 or 4 until the end of BT. Criteria 3 was an increase in somatic symptoms of OTS (16,32) >15% from weeks 4 to 7 remaining the same or increasing from weeks 7 to 8. Subjects were divided into tertiles on the basis of an increase in somatic symptoms of OTS from weeks 4 to 7; 15% was the cutoff for the upper third. Criteria 4 was admitted feeling physically or mentally overloaded (16,33,34) at week 7 or 8. Criteria 5 was a sick leave >10% of daily service. Subjects were divided into tertiles on the basis of sick leave during BT; 10% was the cutoff for the upper third.

Statistical analyses.

Statistical analyses were performed using SPSS (Version 16.0.1. 2005; SPSS, Inc., Chicago, IL). The level of statistical significance was set at P < 0.05. Assumptions for normality were not met for protein carbonyls, malondialdehyde, and nitrotyrosine, and data were log-transformed before statistical analysis. The untransformed values are shown in the text, tables, and figures for more meaningful comparison, except for nitrotyrosine. Responses to submaximal exercise (before and after) and BT period (baseline and weeks 4 and 7) were assessed using repeated-measures ANOVA. Mixed-design factorial ANOVA (group (non-OR vs OR) × exercise × BT) was used to identify differences between and within the OR and non-OR subjects. Bonferroni as the post hoc analysis was used to identify significant differences. In addition, the effect of exercise and BT and their interaction were calculated. Pearson product-moment correlations were used to observe associations between variables. All data are presented as mean ± SD.


V˙O2max and body composition.

There was a main effect of BT for V˙O2max (P < 0.001) with an increase in V˙O2max after 5 wk of BT (P < 0.001). From weeks 5 to 8, V˙O2max did not change significantly (baseline = 45 ± 7 mL·kg−1·min−1, week 4 = 49 ± 5 mL·kg−1·min−1, week 7 = 49 ± 5 mL·kg−1·min−1). The initial mean body height was 178.2 ± 7.9 cm, BMI was 24.7 ± 4.5 kg·m−2, and percentage of body fat was 19.3% ± 7.5%. Both BM and FM exhibited a main effect of BT (P < 0.001), decreasing from weeks 4 to 7 (P < 0.001 and P < 0.01, respectively), with lower values compared with the pre-BT values (P < 0.001; BM: baseline = 78.7 ± 17.7 kg, week 4 = 78.5 ± 16.9 kg, week 7 = 77.0 ± 15.9 kg; FM: baseline = 16.1 ± 10.7 kg, week 4 = 15.6 ± 10.7 kg, week 7 = 14.0 ± 9.1 kg). In contrast, the mean fat-free mass was higher after 7 wk of BT compared with baseline (P < 0.05; baseline = 62.6 ± 9.4 kg, week 4 = 62.9 ± 8.8 kg, week 7 = 63.0 ± 9.3 kg).

Oxidative stress markers and antioxidant capacity.

There was a main effect of exercise for TGSH (P < 0.001), GSSG (P < 0.001), and the GSSG/TGSH ratio (P < 0.001); a main effect of BT for TGSH (P < 0.05) and ORAC (P < 0.01); and an exercise × BT interaction for TGSH (P < 0.05), GSSG (P < 0.001), the GSSG/TGSH ratio (P < 0.01), and ORAC (P < 0.05). TGSH and ORAC (Fig. 1) and protein carbonyls, malondialdehyde, and nitrotyrosine at rest remained unchanged during BT. After 4 wk of BT, a decrease at rest in GSSG (P < 0.01) and an increase in GSSG and GSSG/TGSH ratio at the latter part of BT were observed (Fig. 1). At every time point, submaximal exercise induced a decrease in TGSH and an increase in GSSG and GSSG/TGSH ratio (Fig. 1). However, a significant submaximal exercise-induced decrease in ORAC was observed only at week 4 (Fig. 1). In addition, at week 4, TGSH and ORAC were lower after submaximal exercise than at baseline (TGSH, P < 0.01; ORAC, P < 0.01) and week 7 (TGSH, P < 0.001; ORAC, P < 0.001). GSSG and the GSSG/TGSH ratio after submaximal exercise at week 4 (GSSG, P < 0.05; GSSG/TGSH, P < 0.01) and GSSG at week 7 (P < 0.001) were higher than at baseline (Fig. 1). No significant changes were observed because of the submaximal exercise in protein carbonyls, malondialdehyde, and nitrotyrosine.

Mean ± SD plasma TGSH (A), GSSG (B), GSSG/TGSH (C), and ORAC (D) concentrations before and after the 45-min submaximal exercise at baseline and at weeks 4 and 7 of BT. Difference compared with before exercise: ***P < 0.001, **P < 0.01; baseline: $$$P < 0.001, $$P < 0.01, ∧P < 0.05; week 4: ∧∧∧P < 0.001, ∧∧P < 0.01. N = 35.

Differences between non-OR and OR subjects.

At least three of the five criteria were detected in 31% (n = 11 of 35) of the subjects, all of whom were classified as OR subjects. The remaining 69% (n = 24) were classified as non-OR subjects. From OR subjects, nine subjects fulfilled criteria 1; six, criteria 2; eight, criteria 3; nine, criteria 4; and six, criteria 5. There was a group × exercise × BT interaction only for GSSG (P < 0.001) and GSSG/TGSH ratio (P < 0.05). At baseline, GSSG (P < 0.01) and the GSSG/TGSH ratio (P < 0.05) were higher in OR than those in non-OR at rest (Figs. 2A, C, respectively). In OR, no response of GSSG and GSSG/TGSH to acute submaximal exercise was seen, in contrast to non-OR subjects (Figs. 2B, D, respectively). Furthermore, OR subjects had higher malondialdehyde at baseline (P < 0.05) than non-OR (Table 2). V˙O2max and body composition did not differ between the groups.

Mean ± SD plasma GSSG, GSSG/TGSH, and ORAC concentrations at rest (A, C, E) and changes (%) due to the 45-min submaximal exercise (B, D, F) at baseline and at weeks 4 and 7 of BT among non-OR (n = 24) and OR (n = 11) subjects. Difference compared with non-OR: ##P < 0.01, #P < 0.05; baseline: $$P < 0.01, $P < 0.05; week 4: ∧P < 0.05. Significant change due to the exercise: ***P < 0.001, *P < 0.05.
TGSH and oxidative stress markers among non-OR (n = 24) and OR (n = 11) subjects at rest and relative change due to the submaximal exercise (Δ% ex.).

In response to BT at rest, OR subjects exhibited decreased GSSG (P < 0.01) and a trend of decreased GSSG/TGSH ratio (P = 0.058) from baseline to week 4 (Figs. 2A, C). Compared with baseline, OR had a higher relative increase in GSSG due to submaximal exercise both after 4 (P < 0.01) and 7 (P < 0.05) wk of BT, and in the GSSG/TGSH ratio, after 4 wk of BT (P < 0.05) (Figs. 2B, D). Among the non-OR subjects, exercise-induced decrease in TGSH was lower after 7 wk of BT than after 4 wk of BT (P < 0.01) (Table 2). However, only the OR subjects exhibited a significant exercise-induced decrease in ORAC after 4 wk of BT (P < 0.05), which was significantly lower compared with week 7 (P < 0.05) (Fig. 2F). There were no differences between the groups either in TGSH, ORAC, protein carbonyls, and nitrotyrosine at rest or in the relative responses to exercise at baseline or during BT (Table 2). The relative changes at rest during BT in TSGH, GSSG, GSSG/TGSH ratio, ORAC, protein carbonyls, malondialdehyde, or nitrotyrosine also did not differ between the groups. However, OR had a higher incidence of sick leave only during the last week of BT compared with the non-OR subjects (mean difference = 5.5%, 95% confidence interval = 1.8%-9.0%, P < 0.004).


During the entire BT period, the average daytime MVPA was 2 h 7 min ± 24 min; VLPA, 11 h 36 min ± 30 min; and REST, 30 ± 18 min. During the nighttime between 9:00 p.m. and 6:00 a.m., the average MVPA was 5 ± 2 min; VLPA, 1 h ± 18 min; and sleeping time, 7 h 30 min ± 18 min. There was a main effect of BT for daytime MVPA (P < 0.001), VLPA (P < 0.05), REST (P < 0.001), and nighttime MVPA (P < 0.001). Daytime MVPA was higher (P < 0.001) (Fig. 3), and VLPA (P < 0.001-0.05) (Fig. 3) and REST were lower (P < 0.001) during the latter part of BT compared with weeks 1-2 (REST: weeks 1-2 = 2 h ± 6 min, weeks 3-4 = 24 ± 12 min, weeks 5-6 = 30 ± 24 min, weeks 7-8 = 30 ± 24 min). However, nighttime MVPA was higher during the first 4 wk of BT compared with the latter half of BT (P < 0.01-0.05) (Fig. 3). Nighttime VLPA and sleeping time remained the same during the entire BT.

Mean ± SD daytime VLPA (A) and MVPA (B) and nighttime MVPA (C) during the BT period among non-OR and OR subjects. Difference compared with non-OR: #P < 0.05; weeks 1-2: $$$P < 0.001, $$P < 0.01, $P < 0.05; weeks 3-4: ∧∧∧P < 0.001, ∧∧P < 0.01, weeks 5-6: ¤¤¤P < 0.001; ¤P < 0.05.

PA among non-OR and OR subjects.

PA was monitored from nine OR and nine non-OR subjects. These subjects did not differ in incidence of sick leave during BT; thus, there was no need to take into account the influence of sick leave on physical activity. There were main effects of group for daytime VLPA (P < 0.05) and nighttime MVPA (P < 0.05). OR had higher nighttime MVPA compared with non-OR during weeks 3-4 (P < 0.05) and 5-6 (P < 0.05) (Fig. 3) and less daytime VLPA during weeks 1-2 (P < 0.05) and 5-6 (P < 0.05) (Fig. 3). However, OR had higher daytime MVPA compared with non-OR during weeks 1-2 (P < 0.05) and 5-6 (P < 0.05) (Fig. 3). There were no differences between the groups in REST, nighttime VLPA, and sleeping time. However, REST was less during weeks 5-6 than 7-8 (P < 0.01; weeks 5-6 = 36 ± 24 min, weeks 7-8 = 42 ± 24 min) in OR but not in non-OR.


In this study, we showed that increased oxidative stress is associated with OR. To our knowledge, this is the first study showing that increased oxidative stress before the strenuous 8-wk combined aerobic and strength training period was related to the incidence of OR during the training period. Oxidative stress markers, including GSSG, the GSSG/TGSH ratio, and malondialdehyde at rest, were higher among OR than in non-OR subjects. In addition, OR subjects had no increase in GSSG and GSSG/TGSH ratio in the submaximal exercise, which might be a consequence of high resting levels. Recently, we have reported that impaired antioxidant capacity and increased oxidative stress among athletes at the state of OTS compared to non-OTS athletes (28). However, whether increased oxidative stress is a cause for OTS or an outcome of a training protocol with inadequate recovery remains unclear.

The average training load during the 8-wk BT was 2 h·d−1 with only half an hour of total rest during the daytime. Otherwise, the day included low physical activity from sitting and standing to slow walking. This kind of high training load could be a risk factor for OR for subjects with a low fitness level. Surprisingly, OR and non-OR subjects did not differ from each other according to V˙O2max or body composition at the beginning of BT. This suggests that fitness and fatness in this study are not strong underlying factors for overtraining. On the other hand, OR were more physically active both during the day- and nighttime than non-OR. These results support the hypothesis that the consequences of physical training are influenced not only by the intensity and duration of training but also by the duration of recovery time. Among the OR subjects, high physical activity during the day- and nighttime increased the accumulated training load at the end of BT, which may have contributed to the symptoms and indicators of OR. However, the OR and non-OR subjects did not differ regarding to tasks of the military service. Thus, higher physical activity might more likely be an individual property rather than an external factor.

Classification of OR/OTS is challenging, and it is rarely classified by a single criterion. In this study, OR subjects had to fulfill three of five criteria, including a decrease in aerobic physical performance. Absence from V˙O2max tests or submaximal exercise was set as an additional criterion for a decreased performance. In 71% of the cases, the main reason for sick leave during BT was an upper respiratory track infection. Sick leave itself could affect training responses, even resulting in a detraining effect. However, OR and non-OR did not differ according to sick leave during the first 7 wk of BT. Thus, the decrease in performance probably was mainly caused by an excessive training load. Furthermore, we have recently reported that these OR subjects had higher basal sex-hormone-binding globulin than non-OR subjects at baseline and also after 4 and 7 wk of training (30). In addition, OR subjects had higher basal serum cortisol than non-OR subjects after 7 wk of training (30). Moreover, OR subjects had a decrease in the testosterone/cortisol ratio at rest from weeks 4 to 7 and in the maximal blood lactate/RPE ratio in the V˙O2max test from baseline to the end of BT, whereas non-OR subjects did not (30).

In this study, day activity time increased and rest time decreased during the first 4 wk of BT, with a concomitant increase in aerobic performance, whereas BM remained the same. Among all 35 subjects, decreased oxidative stress at rest, evident as lower oxidized oxidation (GSSG), was observed after the first 4 wk of training. The decreased oxidative damage can be explained either by attenuated generation of ROS (caloric restriction) or by enhancement of tissue protection and antioxidant systems because of adaptation to regular exposure to a small amount of ROS (such as from exercise) (20). Aerobic training itself has been found to decrease oxidative stress at rest also without changes in BM (5). Thus, the decreased oxidative stress observed in this study may be a consequence of an enhanced antioxidant defense system in response to a tolerable training load (10,14), also among OR subjects. The high capacity to consume oxygen (enhanced V˙O2max) and, consequently, to produce a larger amount of ROS is accompanied by high erythrocyte glutathione peroxidase activity and high glutathione concentration, which both serve to protect the organism from lipid peroxidation and cell membrane damage (14).

In contrast to the decreased oxidative stress at rest, 4 wk of BT increased the susceptibility to acute exercise-induced oxidative stress evident as higher glutathione oxidation and lower antioxidant capacity after a submaximal exercise. At baseline, the subjects were not familiar with an acute physical load that may lead to increased oxidative stress. In contrast, we observed a higher exercise-induced oxidative stress at week 4 than at baseline in the trained subjects. In previous studies, both aerobic (8) and strength (17) training has been found to reduce postexercise oxidative stress at the same physical task. However, Vollaard et al. (35) found that neither tapering nor overload endurance training affected the exercise-induced increase in GSSG among endurance-trained athletes. Our results may indicate OR due to the intensive BT period. In the entire group of 35 subjects and in the OR subjects, a U-shaped curve of oxidative stress at rest was observed during BT; GSSG decreased at rest from baseline to week 4, increasing back to baseline values at week 7. Interestingly, submaximal exercise decreased ORAC at week 4. This might be explained by the increased consumption of the body's endogenous antioxidant resources. While coping with oxidative stress, plasma antioxidants are mobilized into the tissue (15).

During the second half of the 8-wk training period, although day activity time and rest time remained the same, GSSG and the GSSG/TGSH ratio at rest increased, and submaximal exercise resulted in a higher induction of GSSG/TGSH ratio. Also, stagnation in V˙O2max was observed and all the subjects had a decrease in FM. All these findings indicate that that training load was too strenuous during the second half of BT causing oxidative stress (10,15). However, ORAC did not respond to submaximal exercise at the end of BT, as was the case at week 4. This could be a marker of OR because diminished antioxidant protection has been observed among overtrained athletes (28).

Plasma nitrotyrosine and protein carbonyls, markers of oxidative protein modification, remained the same during the entire 8-wk training period, both at rest and after exercise. There were also no differences between the OR and non-OR subjects. Plasma nitrotyrosine levels had a wide variation, which may partly explain the observed findings. The present results also support the previous findings that plasma nitrotyrosine level may not be a sufficiently sensitive marker to assess training or acute exercise-induced changes in oxidative stress (22,28). Proteins are an important target for oxidative challenge. ROS modify amino acid side chains of proteins to form protein carbonyls (1). Serum protein carbonyls have been found to increase after an intensive 12-wk resistance training period (15), to decrease after exhaustive marches (50 and 80 kg) in well-trained soldiers (4), and to increase after an exercise test to exhaustion in athletes (28). The turnover time for protein carbonyls varies from many hours to days (4). The 8-wk training period might have been too short, and submaximal exercise load in the present study may not be strenuous enough to modify proteins. Exercise may also have activated a mechanism that removes the oxidatively modified proteins from the circulation or, alternatively, activated antioxidant mechanisms that remove the ROS (4). In addition, differences among studies in oxidative stress responses to training and exercise may be explained by the differences in energy availability, training mode and length of the training period, fitness levels, and genetic background of the subjects. The subjects in the present study were low to moderately fit conscripts in the compulsory army, and training consisted of both strength and endurance training.

In conclusion, the present results suggest that increased oxidative stress may be associated with OR. Although the training had favorable effects on oxidative stress markers during the first 4 wk, sustained heavy training blunted this effect at the latter part of BT. At the end of the study, military training increased oxidative stress at rest back to pretraining levels and beyond the pretraining levels in response to submaximal exercise. Therefore, we may speculate that oxidative stress may also be a marker of insufficient recovery resulting in OR. Because one-third of the subjects were classified as overreached, the variability of the BT program has to be taken into careful consideration. Especially, the program should involve more recovery phases, both daily and periodically.

The authors thank the conscripts who participated in this study and the staff of the Brigade of Kajaani.

The authors also thank Ms. Taija Vaarala for her technical assistance in the blood analysis, Ms. Elina Kokkonen for her assistance in the statistical analysis, and Dr. David Laaksonen for editing the language.

The authors have no conflict of interest to disclose.

The study was granted by the Finnish Ministry of Education, Finnish Cultural Foundation, Polar Electro Oy, and the Scientific Advisory Board for Defence.

The results of this study do not constitute endorsement by the American College of Sports Medicine.


1. Atalay M, Laaksonen DE. Diabetes, oxidative stress and physical exercise. J Sports Sci Med. 2002;1:1-14.
2. Bloomer RJ, Goldfarb AH. Anaerobic exercise and oxidative stress: a review. Can J Appl Physiol. 2004;29(3):245-63.
3. Brugniaux JV, Niva A, Pulkkinen I, Laukkanen RM, Richalet JP, Pichon AP. Polar Activity Watch 200: a new device to accurately assess energy expenditure. Br J Sports Med. 2008;44(4):245-9.
4. Chevion S, Moran DS, Heled Y, et al. Plasma antioxidant status and cell injury after severe physical exercise. Proc Natl Acad Sci U S A. 2003;100(9):5119-23.
5. Devries MC, Hamadeh MJ, Glover AW, Raha S, Samjoo IA, Tarnopolsky MA. Endurance training without weight loss lowers systemic, but not muscle, oxidative stress with no effect on inflammation in lean and obese women. Free Radic Biol Med. 2008;45(4):503-11.
6. Dill DB, Costill DL. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J Appl Physiol. 1974;37(2):247-8.
7. Fatouros IG, Destouni A, Margonis K, et al. Cell-free plasma DNA as a novel marker of aseptic inflammation severity related to exercise overtraining. Clin Chem. 2006;52(9):1820-4.
8. Finaud J, Lac G, Filaire E. Oxidative stress: relationship with exercise and training. Sports Med. 2006;36(4):327-58.
9. Gérard-Monnier D, Erdelmeier I, Régnard K, Moze-Henry N, Yadan JC, Chaudière J. Reactions of 1-methyl-2-phenylindole with malondialdehyde and 4-hydroxyalkenals: analytical applications to a colorimetric assay of lipid peroxidation. Chem Res Toxicol. 1998;11(10):1176-83.
10. Gomez-Cabrera MC, Domenech E, Viña J. Moderate exercise is an antioxidant: upregulation of antioxidant genes by training. Free Radic Biol Med. 2008;44(2):126-31.
11. Halson SL, Bridge MW, Meeusen R, et al. Time course of performance changes and fatigue markers during intensified training in trained cyclists. J Appl Physiol. 2002;93(3):947-56.
12. Kinnunen H, Niva A, Nauha L, Miettinen J. Estimation of energy expenditure from wrist attached accelerometer signal using movement counting and filtering. Med Sci Sports Exerc. 2009;41(5 suppl):S128.
13. Kinnunen S, Hyyppa S, Lehmuskero A, et al. Oxygen radical absorbance capacity (ORAC) and exercise-induced oxidative stress in trotters. Eur J Appl Physiol. 2005;95(5-6):550-6.
14. Margaritis I, Tessier F, Richard MJ, Marconnet P. No evidence of oxidative stress after a triathlon race in highly trained competitors. Int J Sports Med. 1997;18(3):186-90.
15. Margonis K, Fatouros IG, Jamurtas AZ, et al. Oxidative stress biomarkers responses to physical overtraining: implications for diagnosis. Free Radic Biol Med. 2007;43(6):901-10.
16. Meeusen R, Duclos M, Gleeson M, Rietjens G, Steinacker J, Urhausen A. Prevention, diagnosis and treatment of the overtraining syndrome. Eur J Sport Sci. 2006;6(1):1-14.
17. Nikolaidis MG, Paschalis V, Giakas G, et al. Decreased blood oxidative stress after repeated muscle-damaging exercise. Med Sci Sports Exerc. 2007;39(7):1080-9.
18. Ogonovszky H, Sasvári M, Dosek A, et al. The effects of moderate, strenuous, and overtraining on oxidative stress markers and DNA repair in rat liver. Can J Appl Physiol. 2005;30(2):186-95.
19. Oksala NK, Paimela H, Alhava E, Atalay M. Heat shock preconditioning induces protein carbonylation and alters antioxidant protection in superficially injured guinea pig gastric mucosa in vitro. Dig Dis Sci. 2007;52(8):1897-905.
20. Radak Z, Chung HY, Goto S. Exercise and hormesis: oxidative stress-related adaptation for successful aging. Biogerontology. 2005;6:71-5.
21. Radak Z, Taylor AW, Ohno H, Goto S. Adaptation to exercise-induced oxidative stress: from muscle to brain. Exerc Immunol Rev. 2001;7:90-107.
22. Rush JW, Turk JR, Laughlin MH. Exercise training regulates SOD-1 and oxidative stress in porcine aortic endothelium. Am J Physiol Heart Circ Physiol. 2003;284(4):H1378-87.
23. Santtila M, Häkkinen K, Kraemer WJ, Kyröläinen H. Effects of basic training on acute physiological responses to a combat loaded run test. Mil Med. 2010;175(4):273-9.
24. Sen CK, Marin E, Kretzschmar M, Hänninen O. Skeletal muscle and liver glutathione homeostasis in response to training, exercise, and immobilization. J Appl Physiol. 1992;73(4):1265-72.
25. Smith LL. Cytokine hypothesis of overtraining: a physiological adaptation to excessive stress? Med Sci Sports Exerc. 2000;32(2):317-31.
26. Snyder AC, Kuipers H, Cheng B, Servais R, Fransen E. Overtraining following intensified training with normal muscle glycogen. Med Sci Sports Exerc. 1995;27(7):1063-70.
27. Spence L, Brown WJ, Pyne DB, et al. Incidence, etiology, and symptomatology of upper respiratory illness in elite athletes. Med Sci Sports Exerc. 2007;39(4):577-86.
28. Tanskanen M, Atalay M, Uusitalo A. Altered oxidative stress in overtrained athletes. J Sports Sci. 2010;28(3):309-17.
29. Tanskanen M, Uusitalo AL, Häkkinen K, et al. Aerobic fitness, energy balance, and body mass index are associated with training load assessed by activity energy expenditure. Scand J Med Sci Sports. 2009;19(6):871-8.
30. Tanskanen MM, Kyröläinen H, Uusitalo AL, et al. Serum sex hormone-binding globulin and cortisol concentrations are associated with overreaching during strenuous military training. J Strength Cond Res. 2011;25:787-97.
31. Training Division, Defence Command. Standard Direction of Conscripts' Physical Training [in Finnish]. Helsinki, Finland, Finnish Defence Forces; 2004.
32. Uusitalo AL. Overtraining: making a difficult diagnosis and implementing targeted treatment. Phys Sportsmed. 2001;29(5):35-50.
33. Uusitalo AL, Vanninen E, Valkonen-Korhonen M, Kuikka JT. Brain serotonin reuptake did not change during one year in overtrained athletes. Int J Sports Med. 2006;27(9):702-8.
34. Verde T, Thomas S, Shephard RJ. Potential markers of heavy training in highly trained distance runners. Br J Sports Med. 1992;26(3):167-75.
35. Vollaard NB, Cooper CE, Shearman JP. Exercise-induced oxidative stress in overload training and tapering. Med Sci Sports Exerc. 2006;38(7):1335-41.
36. Zoppi CC, Macedo DV. Overreaching-induced oxidative stress, enhanced HSP72 expression, antioxidant and oxidative enzymes downregulation. Scand J Med Sci Sports. 2008;18(1):67-76.


©2011The American College of Sports Medicine