A cross-sectional and experimental study was conducted in Brazzaville, capital of the Republic of Congo, from 10 to 21 August 2018. A total of 76 men, all amateurs, participated in the study: 39 amateur runners, not affiliated to the Congolese Athletics Federation; 37 amateur runners, specialists in endurance race at sports clubs affiliated to the Congolese Athletics Federation. They were selected among the 3105 regular participants in the half marathon event on August 14, 2018; the selection process was the probabilistic method and the technique of two random draws with two fractions of 1/5, that is to say 124 subjects. At the end of this process, 124 subjects were selected. Those were to obey criteria for inclusion following: Congolese of ethnic origin and ethnicity; reside on Congolese territory for at least one year (specific dietary habits); be unharmed one month before diseases (pulmonary infections, arterial hypertension, diabetes, malaria, viral hepatitis); a hemoglobin level more than 11.5 g/dL; have participated in at least one Brazzaville Half Marathon competition in the last five years; be 18 years of age and under 40 years of age; informed and written consent to participate in the study. Since the number of women in the top 100 ranks was small (n = 9), they were excluded from the sample. The abandonment of the race during the competition was also an exclusion criterion. Into definitive, all the 76 participants filled the criteria of selection referred to above. The experimental protocol was accepted by the National Committee of Ethics for Research in Health Sciences of the Congolese Ministry of Scientific and Technical Research.

An individual information sheet was used to collect information from each subject on age, history of pathology, number of years of endurance training and weekly frequency of training sessions of running practice (see appendix).

Anthropometric measurements were performed, including determination of body mass using a digital beam balance (Seca; accuracy, 0.13 kg by default) and height measurement using a wall scale; then the body mass index (BMI) was calculated.

Tubes containing EDTA (1.5 mg/ml as anticoagulant) collected 10 ml of venous blood. Twelve peripheral blood parameters were subsequently determined using a Coulter counter analyzer (Coulter AcT diff, Beckman Coulter, Miami, FL, USA). Hemoglobin (Hb) concentration was determined by HemoCue Hb 201 + (USA/Canada, Friwo, Mod nR FE15 1060 D035). Total leukocyte counts were determined by haemocymeter methods. Serum haptoglobin (Hp) was measured by radial immunodiffusion (Behring Diagnostics, Marburg, Germany). A blood smear was performed for manual white blood cell, differentiation and determination of percentages and absolute values of lymphocytes, neutrophils, eosinophils, basophils and monocytes for each subject. All hematological analyses were performed at the Hematology Laboratory of the Pasteur Institute of Brazzaville. The storage of blood samples at room temperature lasted four hours (first samples) or two hours (second samples).

On the event of competition, all racers of Congolese nationality and black ethnicity filled in the individual information sheets and the informed consent cards. Then, pre-screening tests included a physical examination and rapid haemoglobin. On the day of the race and three hours before, the first blood samples were taken. After the test the second samples were taken after arrival after 24 hours of recovery (3 h, 6 h and 24 h).

All samples were taken at the cubital fold of the left elbow and the single-blind assays, i.e., the laboratory technician did not know the source of the blood sample he was analyzing. The manufacturers’ instructions for the devices and reagents have been strictly adhered in order to reduce the error in measurement accuracy. At the end of the race and samples, subjects were allowed to drink plain or mineral water at their convenience.

The 21.1 km half-marathon race took place between 8:30 am and 12:00, along the avenues of Brazzaville. The elevations varied between 5% and 37% of slope. The climatic conditions indicated an ambient temperature of 28.5˚C and a relative humidity of 90.3%, as recorded in the morning for the city of Brazzaville, by the National Meteorological Department (ANAC) of the Congo.

Before and after the run of the half-marathon, the variables studied were: Red Blood Cells (RBC), hemoglobin (Hb), hematocrit (Hct), Mean Cell Volume (MCV), Plasma Volume Variation (ΔVP), Mean Cell Volume (MCV), Mean Cell Hemoglobin (MCH), Mean Cell Hemoglobin Concentration (MCHC), Red Cell Distribution Width (RDW), White Blood Cell (WBC), neutrophil count, lymphocyte count, neutrophil count, monocyte count, eosinophil count, basophil count, platelet count (PLT), Mean Platelet Volume (MPV) and serum haptoglobin. The motor performances achieved (run times, in hours and minutes) by the riders were recorded. The variations of these parameters of interest were calculated, as well as that of the plasma volume variation (VP), between the rest and the end of the race determined using formula of Van Beaumont [10] following:

Δ V P ( % ) = 100 × ( H b 1 / H b 2 ) × { [ 1 − ( H c t 2 / 100 ) ] / [ 1 − ( H c t 2 / 100 ) ] } − 100

In this equation, Hb1 and Hct1 represent respectively the rates of Hb and hematocrite of rest, then Hb2 and Hct2 those measured in race end of half-marathon.

The recorded data were treated by software Statistica (Stat Soft Inc.), version 5.5. Descriptive statistics were used by calculation percentages, averages and standard deviations. The results are expressed in percent (%) or average values (x) accompanied by standard deviation (SD). After having checked the normality of the distribution of the variable (Kolgomorov-Smirnov test’s), the comparisons between two averages carried out using the Student’s t test. In the event of non-normality, the significant character of the possible differences was judged starting from the nonparametric test of Freedman. For two percentages, it was judged by the Sokal’s S test [11] . The relationship between two variables was analyzed by calculating the r correlation spearman coefficients. During the 24 hours following the competition, data were analyzed by two factors and one way Anova method, completed by post hoc Scheffe’s test. Level of significance of all tests considered at p < 0.05.

The mean anthropometrical data are shown in Table 1. One can notice on this table that the two groups (specialists in the endurance races vs no specialists) were comparable for age, size and weight. The subjects presented a normal stoutness, with an average BMI of 22.6 ± 1.4 kg/m²; however, the values of BMI at the specialists subjects of endurance race were statistically lower (p < 0.01) than those of no specialists.

The performances carried out by the specialist runners were better than those of the no specialists: 121.1 ± 10.5 min against 137.5 ± 11.4 min (p < 0.05).

Of the 76 runners studied, 12 (15.8%) were found to have a mild anemia (Hb < 14.0 g/dL). All essentially of normochromic, normocytic type as evaluated by MCV, MCH, and microscopic appearances. In 34 subjects (or 44.7%), the Hb level was only slightly reduced (13 - 14 g/dL), whereas the remaining 30 (28.5%) had Hb levels comprised with Hb Hb13.0 g/Dl. Microscopy of the peripheral blood film showed the characteristic morphological features of asplenism or hyposplenism.

The mean values (±SD) and the statistical significance of the red cell, leukocyte, and platelet parameters along with the serum haptoglobin levels before and after the half marathon are presented in Table 2.

BMI: Body mass index.

**, p < 0.01; *, p < 0.05.

Of the 76 male runners studied, 22 were found to have a mild anemia (Hb < 14.0 g/dl), all essentially of normochromic, normocytic type as evaluated by MCV, MCH, and microscopic appearances.

In 19 subjects, the Hb level was only slightly reduced (13 - 14 g/dl), whereas the remaining 3 had significantly lower levels of 12.6, 12.6, and 12.5, respectively. In one subject with Hb 13.0 µg/dL, microscopy of the peripheral blood film showed the characteristic morphological features of asplenism or hyposplenism.

The most significant red cell change following the marathon was the increase in ΔMCV (+14.7%) accompanied by a corresponding increase in red cell count (+8.3%) and Hb level (+2.0%). These changes indicate a post-race decrease in plasma volume of 6.5% ± 8.8% (p < 0.0001). It should also be noted that in all three studies there is a significant increase in the immediate post-race MCV.

In the pre-race leukocyte counts, seven subjects (or 9.2%) were found to have a marginal thrombocytopenia with respective events of 2300 - 2400/µL (3 cases) and 2100 - 2500/µL (4 cases). The striking change in the post-race sample was the dramatic increase (7 cases) in the total counts due almost exclusively to an increase in neutrophils but involving lymphocytes and monocytes also. It is of interest that the four runners with low initial counts had an even greater rise (fivefold) in their post-race samples. The runners with high initial counts had a neutrophil/lymphocyte ratio of 2.09 ± 1.34 and those with low counts a ratio of 1.42 ± 0.65. There was no correlation of time taken to complete the race, with either the post-race leukocyte count (r = 0.055) or the increment incurred during the race (r = +0.008).

In the pre-race platelet measurements, nine subjects (or 13.2%) were found to have a marginal thrombocytopenia (<150 × 10⁶/µL). the changes in platelet number following the half-marathon generally mirrored those of the leukocytes (r = 0.387), the being a substantial increase from the pre-race level. The pre-race platelet counts also correlated positively with the pre-race leukocyte counts (r = 0.426). The MPV did not change significantly after the race (Table 2).

In the pre-race Hp measurements, 5 runners (or 6.6%) were found to have subnormal levels (<0.3 g/l). In the post-race samples, there was a significant fall (p < 0.01) in the mean Hp level on 72 runners. No correlation was established with race time, age, height, pre-race body weight, Hb, and RBC or post- and post-race leukocyte counts. However, in the 5 runners with pre-race hypothaptoglobinemia there was a strong correlation with race time (mean 116 min; range 130 - 175 min). In the post-race samples overall, 92% (n = 70) of subjects exhibited a fall in Hp 3.4% (n = 3) had no change, and 3 showed a slight rise.

With regard to the effects of endurance training on the hematological profile (Table 3), no significant difference was observed before the race between specialists and non-specialists at the level of training for. Platelet Counts, MCV, MCH, WBC, neutrophil Count, lymphocyte count and basophil count. Higher values (p < 0.05) were noted among specialists at RBC (Δ = +9.0%), Hbc (Δ = +10%), RDW (Δ = +4.6%), monocytes (Δ = +5.1%), eosinophils (Δ = +8.7%) and haptoglobin (Δ = +17.6%); they were found in no specialists for Hb, Hct, MCHC and MPV. Immediately after the race the non-significant differences remained for MCV, MCH, WBC, Neutrophil count and Lymphocyte count. However, the significantly higher values were noted in specialists runners. They concerned Hb (Δ = +2.6%), ΔPV (Δ = +2.6%), RDW (Δ = +5.3%), basophil counts (Δ = +24.0%), MPV (Δ = +17.6%) and haptoglobin. However, the higher values noted in nonthe specialists were noted for RBC and PLT.

During the recovery period (Table 4), the two-factor Anova shows the evolution of haematological parameters: a significant “group” effect [F (1.74) = 15.73; p = 0.001]; as well as a significant “group × recovery” interaction [F (3.72) = 8.31; p < 0.01]. One-way analysis of variance (recovery) in the non-specialist endurance race group shows a significant effect [F (4.33) = 4.8; p < 0.01], the same for the group of specialists [F (4.35) = 3.17; p < 0.05]. The post hoc Schoffe’s test shows that RBC, PLT, RDW, neutrophil counts, lymphocyte counts and MPV decreased without reaching basal values throughout the sample The rates of decrease between the end of the effort and the 24 hours of recovery were respectively: −8.3% (p < 0.05), −20.9% (p < 0.001), −3.9% (p > 0,05), −40% (p < 0.001), −4.3% (p > 0.05) and −4% (p > 0,05). However, an increase in values was noted for MCH and MCHC: +5.1% (p < 0.05) and +4.3% (p > 0.05). The decline observed for Hb was not significant.

**: p < 0.01; *: p < 0.05;