Skip to main content
NCBI home page
European Journal of Sport Science logoLink to European Journal of Sport Science
. 2024 Mar 5;24(4):431–439. doi: 10.1002/ejsc.12073

Are pre‐race serum blood biomarkers associated with the 24‐h ultramarathon race performance?

Stefano Amatori 1, Davide Sisti 1, Alexander Bertuccioli 1, Marco BL Rocchi 1, Francesca Luchetti 1, Maria Gemma Nasoni 1, Stefano Papa 1, Roberto Citarella 2, Fabrizio Perroni 1,✉, Serena Benedetti 1
PMCID: PMC11235809

Abstract

The 24‐h ultramarathon (UM) race is one of the most demanding competitive sports in terms of muscular and physiological exertion. In this context, predictors of UM athletes' physical performance are in high demand; however, data on the predictive capabilities of hematological variables are still sparse. In the present paper, we retrospectively took into consideration the pre‐race blood biomarker levels (including basic blood count, leukocyte subpopulations, markers of inflammation and organ function, metabolic profile, and electrolytes) of 50 UM athletes (M = 33, F = 17) who completed a 24‐h competition in order to identify a combination of analytes capable of predicting the athletic performance in terms of distance covered during the 24‐h run. The multiple regression analysis produced a model that explained a significant portion of the variance in the dependent variable, with an adjusted R‐squared value of 0.783 (F(13, 36) = 14.58, p < 0.001). A greater race distance was correlated with higher pre‐race values of hematocrit, lactate dehydrogenase (LDH), total cholesterol, HDL/LDL ratio, and triglycerides and lower levels of monocytes, eosinophils, alanine aminotransferase (ALT), gamma‐glutamyl transferase (GGT), total proteins, and sodium. This study represents the first of its kind conducted on 24‐h UM athletes that investigated the association between blood markers and endurance performance. Our model, given its promising predictive power, would serve as a starting point that will require refinement and integration with other traditional performance prediction measures, in order to support athletes and coaches in better managing the training loads during the race‐approaching phases.

Keywords: hematological variables, performance, predictive model, running, ultra‐endurance

Highlights

  • A combination of pre‐race hematological parameters was found to predict the 24‐h ultramarathon race distance in ultra‐endurance athletes.

  • A greater race distance was correlated with higher pre‐race values of hematocrit, lactate dehydrogenase, total cholesterol, HDL/LDL ratio, and triglycerides, and lower levels of monocytes, eosinophils, alanine aminotransferase, gamma‐glutamyl transferase, total proteins, and sodium.

  • The predictive model generated by the multiple regression analysis explained a significant portion of the variance in the dependent variable (race distance), with an adjusted R‐squared value of 0.783.

1. INTRODUCTION

Ultramarathons (UMs) are defined as races longer than the traditional marathon length (42.195 km) (Spittler & Oberle, 2019). UMs may be performed as distance‐limited runs (50‐km, 100‐km, 50‐miles, and 100‐miles UMs), or as time‐limited events (6‐, 12‐, 24‐, 48‐, 72‐h, or 6‐ and 10‐day UMs). As with other types of running competitions, UMs are rapidly gaining popularity. Among them, the 24‐h UM is an extremely demanding race in terms of muscular and physiological exertion, characterized by acute cardiorespiratory and metabolic responses to guarantee sufficient oxygen supply to cells and tissues during the competition (Bizjak et al., 2022; Hoffman, 2016; Knechtle & Nikolaidis, 2018). As a consequence of the high level of physical demand, injuries to tissues and organs may occur, leading to the migration of leukocytes (in particular, neutrophils and monocytes) to the damaged areas and the induction of an acute phase inflammatory response (Kim et al., 2007; Waskiewicz et al., 2012; Wu et al., 2004).

Accordingly, in our previous investigations involving 24‐h UM athletes subjected to pre‐ and post‐race blood sample collection (Benedetti et al., 2018, 2021), we observed that the serum biomarkers of skeletal muscle damage [such as creatine phosphokinase (CPK), lactate dehydrogenase (LDH), and myoglobin] significantly increased at the end of the competition in comparison with the corresponding baseline levels, as a consequence of the prolonged muscular exertion. Similarly, the transitory myocardial overstimulation following the strenuous contraction of the heart muscle led to a post‐race increment of the markers of cardiac damage, namely CPK‐MB, troponin‐I, and NT‐proBNP. Tissue injury was associated with immune system activation and increased levels of C‐reactive protein (CRP) as a marker of inflammation. The 24‐h race also had a major impact on fat metabolism; indeed, triglyceride levels significantly decreased at the end of the event, reflecting the use of fatty acids as fuels for energy production and muscle contraction. A significant decline in total cholesterol and LDL levels and a significant rise in HDL were also evidenced. Noteworthy, the length of the 24‐h UM distance directly influenced the post‐race changes of the investigated parameters, being positively correlated with the markers of inflammation and organ damage.

In this context, predictors of UM athletes' physical capacities are in high demand. Most studies aiming to predict the physical performance of endurance athletes primarily focus on physiological and anthropometric parameters (Knechtle et al., 2010; Knechtle, Knechtle, Rosemann, & Senn, 2011). In particular, key predictors of a successful 24‐h UM race included fast personal best running times, extensive previous race experience, and a high running speed and a high running volume during training (Knechtle, Knechtle, Rosemann, & Lepers, 2011).

Data on the predictive capabilities of blood test results regarding physical performance are still sparse, and the few relevant studies mainly deal with the half‐marathon race. For example, Lippi et al. (2014) monitored a set of standard laboratory analytes immediately before a half‐marathon run and found a significant inverse association (r = −0.450; p = 0.042) between pre‐race mean platelet volume (MPV) and running performance. In another report, the baseline serum value of α‐amylase was a significant predictor (r = −0.598; p = 0.021) of half‐marathon running performance (Lippi et al., 2015). More recently, a combination of routine blood analytes has been shown to predict the fitness decrease in elderly endurance athletes (Haslacher et al., 2017) and to predict physical performances in a group of youth soccer players (Perroni et al., 2020).

With this in mind, in the present study we hypothesized that pre‐race serum biomarkers might provide a valuable contribution in predicting the 24‐h UM performance. To this aim, we retrospectively took into consideration a large panel of pre‐race blood biomarkers (including basic blood count, leukocyte subpopulations, markers of inflammation and organ function, metabolic profile, and electrolytes) of UM athletes who completed a 24‐h competition in order to explore and identify possible analytes capable of predicting athletic performance in terms of distance covered during the 24 h.

2. MATERIALS AND METHODS

2.1. Study design

The study was designed in collaboration with the University of Urbino Carlo Bo (Italy) and the Centre of Rehabilitation Therapy of Reggio Emilia (Italy). As previously reported (Benedetti et al., 2018, 2021), the ultra‐endurance athletes were enrolled from the Italian Ultramarathon and Trail Association (IUTA). All the runners were completely familiarized with the UM races (on average, 10 years of long run experience and at least one 24‐h race per year) and were eligible for competitive sports, having passed all the foreseen checks (information on the athlete's state of health and their family pathologies, work quality and sporting activities, smoking and alcohol habits, and supplement intake). Anthropometric data (weight and height) were also collected, and a clinical visit, including blood pressure measurement and electrocardiogram at rest and under stress, was performed. Athletes were informed about the experimental procedures and gave their written consent to participate in the study. The Institutional Review Board of IUTA approved the study protocol in accordance with the Declaration of Helsinki, as previously indicated (Benedetti et al., 2018, 2021).

2.2. Ultramarathon competitions

Three official 24‐h UM competitions (certified by the Italian Athletics Federation—FIDAL) were taken into consideration and retrospectively scrutinized. The first 24‐h running was organized in Reggio Emilia (Italy) on March 12, 2016 (Benedetti et al., 2018). The race was performed on a flat 1000 m certified trail (Pista Cimurri, 58 m above sea level). The temperature at the competition site varied from 7°C to 15°C while humidity ranged from 51% to 93%. Similarly, the second 24‐h race was held in Reggio Emilia, Italy (November 11, 2017) on the same certified trail described above (Benedetti et al., 2021). Temperature varied from 5°C to 12°C and humidity ranged from 76% to 100%. The third 24‐h event was performed in Verona, Italy (September 18, 2021) on a flat 1500‐m certified trail (Pista Consolini, 59 m above the sea level). Temperature ranged from 18°C to 26°C and humidity from 54% to 94%. All the 24‐h competitions started at 10.00 a.m. and ended at 10.00 a.m. the following day. Athletes could rest and ingest food and liquids without restrictions during the race. The total distance covered in the 24‐h was used as a performance indicator. Times and distances were recorded every lap by an electronic chip timing system attached to the runner's shoelace. Distance records may be assessed online at DUV Ultra Marathon Statistics (https://statistik.d‐u‐v.org/).

2.3. Participants

The study population comprised 50 UM runners (M = 33, F = 17, and age 26–76 years), of which 8 (M = 5, F = 3, and age 30–58 years) completed the 24‐h UM race organized in Reggio Emilia (Italy) on March 12, 2016 (https://statistik.d‐u‐v.org/getresultevent.php?event=32723), 22 (M = 12, F = 10, and age 26–71 years) completed the 24‐h competition held in Reggio Emilia (Italy) on November 11, 2017 (https://statistik.d‐u‐v.org/getresultevent.php?event=44540), and 20 (M = 16, F = 4, and age 38–76 years) completed the 24‐h race held in Verona (Italy) on September 18, 2021 (https://statistik.d‐u‐v.org/getresultevent.php?event=75735).

On the basis of training volume and performance results, participants could be classified as Highly Trained/National Level (Tier 3) or Elite/International Level (Tier 4) (McKay et al., 2022). Males reported running a median of 300 km/month [Q1: 200; Q3: 400], and females a median of 200 km/month [Q1: 180; Q3: 250].

2.4. Procedures

Venous blood samples were collected from each participant 3 h before the race after overnight fasting by healthcare personnel according to standard procedures. Both sterile Vacutainer EDTA‐containing tubes and serum separator tubes were used. Samples were immediately transported at a controlled temperature to a clinical laboratory located in Reggio Emilia (Centro Analisi Reggio Emilia) for routine blood analyses.

2.5. Blood parameters and analytical instruments

Due to the exploratory nature of the study, a large panel of the most common blood biomarkers was taken into consideration. In detail, basic blood count included white blood cells (WBC), red blood cells (RBC), hematocrit (HCT), hemoglobin (HGB), and platelets; leukocyte subpopulations were also assessed. High sensitivity C‐reactive protein (hs‐CRP) was evaluated as a biomarker of inflammation; creatine phosphokinase (CPK), lactate dehydrogenase (LDH), and myoglobin as markers of muscle damage; cardiac creatine phosphokinase (CPK‐MB), troponin‐I, and N‐terminal pro‐brain natriuretic peptide (NT‐proBNP) as markers of cardiac damage; uric acid, blood urea nitrogen (BUN), creatinine, and glomerular filtration rate (MDRD) as markers of kidney function; and alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma‐glutamyl transferase (GGT) as markers of liver injury. As regards the metabolic profile, total proteins, glucose, triglycerides (TG), total cholesterol (TC), low‐density lipoproteins (LDL), and high‐density lipoproteins (HDL) were determined. Other serum parameters included iron, ferritin, homocysteine (Hcy), and electrolytes (calcium, chlorine, sodium, potassium, and magnesium). Routine analyses were conducted using automated analyzers (Access 2, Beckman Coulter, Olympus AU480, Beckman Coulter, Sysmex XT1800i, Dasit).

2.6. Statistical analyses

Descriptive statistics were used to describe participants' characteristics, race distance, and blood parameters; medians [first and third quartile] were reported for continuous variables. Differences in anthropometric variables and ultramarathon performance, considering sex as a predictor, were checked using (with a conservative approach) a Mann–Whitney U test for independent samples. To handle the missing data, Bayesian Stochastic regression imputation was used to impute missing values for blood parameters. This method was chosen as it considers the uncertainty when estimating the regression coefficients of the imputation model. Missing data were about 2% of the total, and were related to unreliable results derived from blood analyses. Multicollinearity among blood parameters was also checked, and multicollinear variables were removed if the correlation (r) was greater than 0.8. This was done to ensure that the independent variables were not highly correlated, which could lead to unreliable estimates of the regression coefficients. A multiple regression analysis with backward stepwise elimination (p to enter 0.05; p to removal 0.10) was then performed with race distance as the dependent variable and blood parameters, age, BMI, and sex as predictors. Cohen's f were calculated as effect size measures and were interpreted as follows: f = 0.02, small effect; f = 0.15, medium effect; and f = 0.35, large effect (Cohen, 1992). The analyses were conducted using SPSS v.26 (IBM, Armonk, NY, USA) and RStudio (Posit Software, Boston, MA, USA). An alpha value of 0.05 was considered for all statistical tests.

3. RESULTS

Descriptive statistics of the participants' characteristics are reported in Table 1. Differences between males and females were related to height, weight, and BMI, while age and ultramarathon records were comparable. The participants' characteristics grouped for the three races are reported in Supplementary Material (S1).

TABLE 1.

Baseline characteristics and ultramarathon records of the runners who completed the three 24‐h races (n = 50), reported as median [first and third quartile] and Mann–Whitney test significance value (p(U)).

Males (n = 33) Females (n = 17) p(U)
Anthropometrics
Age (years) 48 [38–58] 44 [40–50] 0.277
Height (cm) 172 [170–177] 160 [156–164] < 0.001
Weight (kg) 69.0 [65.8–74.0] 55.4 [54.0–57.5] < 0.001
BMI (kg/m2) 23.1 [21.8–23.9] 21.4 [20.0–22.2] < 0.001
Ultramarathon results
Distance completed (km) 145 [128–176] 135 [110–185] 0.319
Average speed (km/h) 6.1 [5.4–7.3] 5.6 [4.6–7.7] 0.319
Open in a new tab

Baseline blood parameters are reported in Table 2. All measures fell within the reference ranges except for total cholesterol (TC) and LDL, which were slightly higher than recommended. However, it should be noted that even HDL values were above the cutoff value. Correlation matrix of the blood parameters used is reported as Supplementary Material (S2).

TABLE 2.

Baseline blood parameters presented as median (first and third quartile).

Blood parameters Pre‐race measure (n = 50) Reference range
Basic blood count and leukocyte subpopulations
WBC (x103/μl) 5.80 (4.80–6.70) 3–10.8
RBC (x106/μl) 4.92 (4.70–5.18) 4–5.9
HGB (g/dl) 14.90 (14.30–15.80) 13–17.5
HCT (%) 43.90 (42.15–45.65) 39–55
MCV (fL) 88.50 (86.40–91.40) 80–99
MCH (pg) 29.90 (29.43–30.80) 25–35
MCHC (g/dl) 33.75 (33.23–34.40) 28–36
RDW (%) 13.30 (12.90–13.90) 10–16
Platelets (x103/μl) 244.00 (198.25–282.75) 140–450
MPV (fL) 10.80 (10.23–11.35) 6.5–12.5
Neutrophils (x103/μl) 3.20 (2.53–4.08) 1.8–7
Eosinophils (x103/μl) 0.10 (0.10–0.20) 0–0.7
Basophils (x103/μl) 0.00 (0.00–0.00) 0–0.2
Monocytes (x103/μl) 0.50 (0.40–0.60) 0–1.1
Lymphocytes (x103/μl) 1.80 (1.53–2.00) 1–4.8
Markers of organ function and inflammation
Uric acid (mg/dl) 4.60 (4.00–5.28) 3.5–7.2
BUN (mg/dl) 35.00 (29.00–41.00) 10–50
Creatinine (mg/dl) 0.83 (0.70–0.92) 0.7–1.3
MDRD (ml/min/1,73mq) 95.00 (84.00–106.00) >60
AST (U/l) 22.50 (19.00–28.00) 7–45
ALT (U/l) 18.50 (15.25–26.75) 7–45
GGT (U/l) 19.50 (15.00–30.75) <55
CPK (U/l) 115.00 (93.00–169.75) <171
CPK‐MB (ng/ml) 4.20 (2.60–6.80) 0.6–6.3
Troponin‐I (ng/ml) 0.02 (0.00–1.50) <0.04
Myoglobin (μg/L) 27.70 (19.10–45.63) <70
LDH (U/l) 174.00 (158.25–196.30) <248
NT‐proBNP (pg/ml) 50.00 (29.65–134.53) <125
hs‐CRP (mg/L) 0.05 (0.03–0.11) <1
Metabolic profile and electrolytes
Glucose (mg/dl) 95.00 (89.00–103.75) <100
TC (mg/dl) 221.50 (191.00–243.75) <200
HDL (mg/dl) 69.00 (64.00–85.00) >45
LDL (mg/dl) 128.00 (112.00–157.00) <115
HDL/LDL ratio 0.55 (0.45–0.71) ‐
TG (mg/dl) 80.50 (63.00–95.25) <150
Total proteins (g/dl) 7.50 (7.25–7.80) 6–8.3
Iron (μg/dl) 84.00 (62.35–125.50) 60–180
Ferritin (ng/ml) 46.00 (24.00–77.75) 24–336
Hcy (μmol/L) 12.87 (10.75–14.82) 5–15
Calcium (mg/dl) 9.60 (9.40–9.88) 8.8–10.6
Chlorine (mEq/L) 104.00 (102.00–105.00) 96–110
Sodium (mEq/L) 141.50 (140.00–142.00) 135–150
Potassium (mEq/L) 4.10 (3.83–4.30) 3.5–5.3
Magnesium (mg/dl) 2.00 (2.00–2.10) 1.6–2.6
Open in a new tab

Note: Reference ranges for healthy subjects are also reported.

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CPK, creatinine phosphokinase; CPK‐MB, creatine phosphokinase‐MB; GGT, gamma‐glutamyl transferase; HCT, hematocrit; Hcy, homocysteine; HDL, high‐density lipoproteins; HGB, hemoglobin; hs‐CRP, high sensibility C‐reactive protein; LDH, lactate dehydrogenase; LDL, low‐density lipoproteins; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; MCV, mean corpuscular volume; MDRD, glomerular filtration rate; MPV, mean platelet volume; NT‐proBNP, B‐type natriuretic peptide; RBC, red blood cells; RDW, red cells distribution width; TC, total cholesterol; TG, triglycerides; WBC, white blood cells.

The multiple regression analysis produced a model that explained a significant portion of the variance in the dependent variable, with an adjusted R‐squared value of 0.783 (F(13,36) = 14.58, p < 0.001). In Supplementary Material (S3) the mean distance predicted values (95% CI) were reported for each observation. The results showed that 13 variables were significant predictors of the dependent variable. The regression coefficients for each of the predictors in the model, along with their confidence intervals and p‐values, are reported in Table 3. A greater race distance was correlated with higher pre‐race values of HCT, LDH, TC, HDL/LDL ratio, and TG, while the remaining variables showed negative correlations.

TABLE 3.

Regression coefficients of the model, with 95% confidence intervals (only the variables entered into the model are reported).

Variable Unstandardized beta [95% CIs] p‐value Cohen's f
Age −1.32 [−2.34; −0.29] 0.013 0.12
HCT 8.94 [3.58; 14.30] 0.002 0.27
Monocytes −85.48 [−160.81; −10.14] 0.027 0.07
Eosinophils −69.86 [−117.21; −22.50] 0.005 0.01
ALT −4.15 [−6.69; −1.62] 0.002 0.02
GGT −1.25 [−1.93; −0.56] 0.001 0.17
LDH 1.03 [0.57; 1.49] <0.001 0.06
TC 0.61 [0.16; 1.05] 0.009 0.12
HDL/LDL ratio 163.95 [89.77; 238.14] <0.001 0.09
TG 0.307 [0.082; 0.532] 0.009 0.01
Total proteins −32.85 [−55.29; −10.41] 0.005 0.37
Hcy −5.18 [−8.38; −1.99] 0.002 0.05
Sodium −8.15 [−15.32; −0.98] 0.027 0.01
Open in a new tab

Abbreviations: ALT, alanine aminotransferase; GGT, gamma‐glutamyl transferase; HCT, hematocrit; Hcy, homocysteine; HDL, high‐density lipoproteins; LDH, lactate dehydrogenase; LDL, low‐density lipoproteins; TC, total cholesterol; TG, triglycerides.

Finally, the equation of the predictive model was:

Race distance (km) = 1019.27–1.32 × age + 8.94 × HCT ‐ 85.48 × monocytes ‐ 69.86 × eosinophils ‐ 4.15 × ALT ‐ 1.25 × GGT + 1.03 × LDH + 0.61 × TC + 163.95 × HDL/LDL + 0.31 × TG ‐ 32.85 × protein ‐ 5.18 × Hcy ‐ 8.15 × sodium

The observed and predicted race distance values are presented in Figure 1.

FIGURE 1.

FIGURE 1

Open in a new tab

Predicted (x‐axis) versus observed (y‐axis) race distances. The identity line (dotted line) and the prediction intervals (gray lines) are reported.

4. DISCUSSION

Ultramarathon running performance may depend on several physiological factors, including anthropometric variables, functional and training characteristics (Knechtle, Knechtle, Rosemann, & Senn, 2011). In addition, age also has a major impact on UM performance. In agreement with literature describing an age‐related decline in the running performance (Knechtle et al., 2012, 2014; Romer et al., 2014), this study confirmed a negative association between age and UM distance. Referring to blood parameters, the association between hematological variables and race performance has not yet been investigated in the context of 24‐h UM running to the best of our knowledge. The present study aimed to explore retrospectively whether a combination of pre‐race blood parameters might predict the 24‐h UM race distance in highly‐trained/elite ultra‐endurance athletes. As a result, the regression model explained a significant portion of the variance in the dependent variable (24‐h distance), with an adjusted R‐squared value of 0.783 (Figure 1). In particular, a greater race distance was significantly correlated with higher pre‐race values of HCT, LDH, TC, HDL/LDL ratio, and TG and lower monocytes, eosinophils, ALT, GGT, total proteins, and sodium levels (Table 3).

Despite the limited information on the 24‐h UM, some reliable explanations can be brought to support the associations observed in our study. As expected, pre‐race hematocrit (HCT) values positively correlated with running performance in the multivariate analysis (p = 0.002; f = 0.27). HCT is a reference point for blood's ability to deliver oxygen to working muscles during prolonged aerobic exercise (Mairbäurl, 2013); consequently, higher HCT levels may confer a performance advantage to the UM footrace, as previously suggested in the case of multi‐stage UM competitions (Rama et al., 2016).

Conversely, monocytes, eosinophils, ALT, and GGT were inversely correlated to the 24‐h UM distance (f ranging from 0.01 to 0.17). This means that athletes who covered a lower distance showed higher pre‐race levels of these serum biomarkers, possibly suggesting the presence of immune system activation and/or organ damage before the race in less‐performing UM runners. Immune system activation might be due to microtrauma to the musculoskeletal system or upper respiratory tract infections occurring during practice for the UM competition (Partyka & Waśkiewicz, 2021; Peters et al., 2010; Shin & Lee, 2013). Similarly, higher pre‐run ALT and GGT levels could be linked to transient hepatocellular injury after overly UM training (Shin et al., 2016; Tirabassi et al., 2018). Unlike ALT and GGT, pre‐race LDH levels showed a positive correlation with the 24‐h UM distance (p < 0.001; f = 0.06). Higher pre‐run LDH might reflect the occurrence of a higher running volume during training, which is one of the most important predictor variables for a successful UM performance (Knechtle, Knechtle, Rosemann, & Senn, 2011). Previous observations on blood predictors of performance evidenced no significant associations between running performance and pre‐race monocyte, eosinophil, ALT, or GGT levels, while LDH displayed a negative correlation (Lippi et al., 2014, 2015). However, these findings were related to athletes who completed a 21.1 km half‐marathon run and not a 24‐h UM race, highlighting a different correlation based on the type of competition.

As regards the predictors related to the lipid profile, previous findings by Lippi et al. (2015) reported a positive association between pre‐run HDL levels and half‐marathon time in univariate analysis, while total cholesterol (TC) and triglycerides (TG) showed no significant correlations. In our study, pre‐race TC and TG levels were positively associated with the 24‐h running distance (p = 0.009 for both, f = 0.12, and f = 0.01 for TC and TG, respectively). The homeostatic control of energy balance is a major concern in UM due to its large energetic demands, and TG stored in adipose tissue and contained in lipoproteins widely contribute as energy substrates for muscle contraction (Benedetti et al., 2018, 2021; Emed et al., 2016; Gorecka et al., 2020; Waskiewicz et al., 2012; Wu et al., 2004), possibly conferring a performance advantage. It is to be noted that the pre‐race HDL/LDL ratio was also positively correlated with the UM distance (p < 0.001; f = 0.09). Higher HDL levels favorably lead to an increase in serum concentrations of paraoxonase‐1 (PON1), an HDL‐associated enzyme with antioxidant and anti‐atherosclerotic properties that may protect against lipid oxidation and endothelial dysfunction (Benedetti et al., 2018; Chistiakov et al., 2017). Indeed, the 24‐h UM race is characterized by high oxygen consumption and high production of reactive oxygen species (ROS) during muscle contraction that may negatively affect athletes' cardiovascular health (Guerrero et al., 2021; Jee & Jin, 2012; Turner et al., 2014; Vezzoli et al., 2016). Accordingly, homocysteine (Hcy), a well‐recognized independent risk factor of atherosclerosis, contributing to plaque formation through mechanisms involving ROS generation and lipoprotein oxidation (Ganguly & Alam, 2015; Jakubowski, 2019), resulted inversely correlated with the 24‐h UM distance (p = 0.002; f = 0.05).

The prediction model revealed a negative association also for baseline total protein levels and the 24‐h UM distance (p = 0.005; f = 0.37). Very few studies have investigated the effects of protein supplementation on endurance performance, and to date, the role of protein intake in promoting athletes' yield is still doubtful (Jager et al., 2017). Dietary protein is not the preferred energy source during an endurance event; in fact, UM racing heavily depends on oxidative metabolism to utilize glycogen and fat stores efficiently (Tiller et al., 2019). In the present study, athletes with lower baseline protein levels could have adopted a higher carbohydrate‐rich diet in the pre‐competition period (at the expense of protein intake) so as to increase muscle glycogen stores, which would benefit their UM performance (Burke et al., 2011; Costa et al., 2019).

Finally, we found a negative association between baseline sodium levels and the 24‐h UM running distance (p = 0.027; f = 0.01), while the remaining electrolytes (calcium, chlorine, magnesium, and potassium) showed no significant correlations. Previous evidence on performance predictors revealed no associations between pre‐run electrolyte levels and running performance in half‐marathon athletes (Lippi et al., 2015).

The limitations of this study are primarily associated with the sample size that, despite being large enough according to the power analysis, is limited, and must be acknowledged when interpreting the findings. However, it should be noted that 24‐h UM competitions typically involve a relatively small number of athletes compared to other endurance events such as marathon running. Moreover, fluid and nutrition intake during the race was not measured in this study, and these factors could have significantly impacted the race performance. In fact, energy deficits or disorders in fluid or electrolyte metabolism have been shown to limit the ultra‐endurance performance (Costa et al., 2014). Finally, other aspects not taken into consideration that could have influenced the 24‐h performance were psychological variables (e.g., motivation and stress) (Roebuck et al., 2018) and environmental conditions (e.g., air temperature, wind speed, relative humidity, and barometric pressure) (El Helou et al., 2012).

5. CONCLUSION

Ultramarathon running performance may depend on several factors, but hematological variables have been poorly considered in previous studies. This research demonstrates that a combination of pre‐race blood parameters, including hematocrit, leukocytes, markers of organ function, and markers of metabolic profile, can predict the 24‐h ultramarathon race distance in highly‐trained/elite ultra‐endurance athletes. Overall, understanding which hematological variables can be the predictors of UM performance in a laboratory setting should contribute to a better sport and health management of the ultra‐endurance athlete both in terms of training and nutritional protocols, while taking into consideration the physiological and anthropometric parameters. This study represents the first of its kind conducted on ultra‐endurance athletes and further research is necessary to confirm and strengthen these findings in order to deepen our understanding of the association between blood markers and endurance performance. Our model, despite exhibiting promising predictive power, represents a starting point for future prediction analysis, in order to integrate traditional performance prediction models and enhance predictive accuracy.

CONFLICT OF INTEREST STATEMENT

The authors report there are no competing interests to declare.

Supporting information

Supporting Information S1

EJSC-24-431-s003.docx (16.7KB, docx)

Supporting Information S2

EJSC-24-431-s001.docx (111.5KB, docx)

Supporting Information S3

EJSC-24-431-s002.docx (23.3KB, docx)

ACKNOWLEDGMENTS

Special thanks to the UM runners that agreed to participate in the study and for the organization of the UM races and athletes' enrollment. Thanks for the support in blood sample collection.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

REFERENCES

  1. Benedetti, Serena , Catalani Simona, Peda Federica, Luchetti Francesca, Citarella Roberto, and Battistelli Serafina. 2018. “Impact of the 24‐h Ultramarathon Race on Homocysteine, Oxidized Low‐Density Lipoprotein, and Paraoxonase 1 Levels in Professional Runners.” PLoS One 13(2): e0192392. 10.1371/journal.pone.0192392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Benedetti, Serena , Gemma Nasoni Maria, Palma Francesco, Citarella Roberto, and Luchetti Francesca. 2021. “Serum Changes in sTWEAK and its Scavenger Receptor sCD163 in Ultramarathon Athletes Running the 24‐h Race.” Cytokine 137: 155315. 10.1016/j.cyto.2020.155315. [DOI] [PubMed] [Google Scholar]
  3. Bizjak, Daniel A. , Schulz Sebastian V. W., John Lucas, Schellenberg Jana, Bizjak Roman, Witzel Jens, Valder Sarah, et al. 2022. “Running for Your Life: Metabolic Effects of a 160.9/230 Km Non‐stop Ultramarathon Race on Body Composition, Inflammation, Heart Function, and Nutritional Parameters.” Metabolites 12(11): 1138. 10.3390/metabo12111138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Burke, Louise M. , Hawley John A., Wong Stephen H. S., and Jeukendrup Asker E.. 2011. “Carbohydrates for Training and Competition.” Journal of Sports Sciences 29(Suppl 1): S17–S27. 10.1080/02640414.2011.585473. [DOI] [PubMed] [Google Scholar]
  5. Chistiakov, Dimitry A. , Melnichenko Alexandra A., Orekhov Alexander N., and Bobryshev Yuri V.. 2017. “Paraoxonase and Atherosclerosis‐Related Cardiovascular Diseases.” Biochimie 132: 19–27. 10.1016/j.biochi.2016.10.010. [DOI] [PubMed] [Google Scholar]
  6. Cohen, Jacob . 1992. “A Power Primer.” Psychological Bulletin 112(1): 155–159. 10.1037//0033-2909.112.1.155. [DOI] [PubMed] [Google Scholar]
  7. Costa, Ricardo J. S. , Hoffman Martin D., and Stellingwerff Trent. 2019. “Considerations for Ultra‐endurance Activities: Part 1‐ Nutrition.” Research in Sports Medicine 27(2): 166–181. 10.1080/15438627.2018.1502188. [DOI] [PubMed] [Google Scholar]
  8. Costa, Ricardo J. S. , Gill Samantha K., Hankey Joanne, Wright Alice, and Marczak Slawomir. 2014. “Perturbed Energy Balance and Hydration Status in Ultra‐endurance Runners during a 24 H Ultra‐marathon.” British Journal of Nutrition 112(3): 428–437. 10.1017/S0007114514000907. [DOI] [PubMed] [Google Scholar]
  9. El Helou, Nour , Tafflet Muriel, Berthelot Geoffroy, Tolaini Julien, Marc Andy, Guillaume Marion, Hausswirth Christophe, and Toussaint J.‐François. 2012. “Impact of Environmental Parameters on Marathon Running Performance.” PLoS One 7(5): e37407. 10.1371/journal.pone.0037407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Emed, Luiz Gustavo Marin , Passaglia Daniela Gunther, Guerios Surya T., João Paula G. D., Moser André I. S., Abdalla Dulcinéia S. P., Guarita‐Souza Luis Cesar, et al. 2016. “Acute Modification in Plasma Lipid Levels in Ultramarathon Runners.” Journal of Sports Sciences 34(17): 1657–1661. 10.1080/02640414.2015.1130237. [DOI] [PubMed] [Google Scholar]
  11. Ganguly, Paul , and Alam Sreyoshi Fatima. 2015. “Role of Homocysteine in the Development of Cardiovascular Disease.” Nutrition Journal 14(1): 6. 10.1186/1475-2891-14-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Górecka, Monika , Krzemiński Krzysztof, Buraczewska Monika, Kozacz Agnieszka, Dąbrowski Jan, and Ziemba Andrzej Wojciech. 2020. “Effect of Mountain Ultra‐marathon Running on Plasma Angiopoietin‐like Protein 4 and Lipid Profile in Healthy Trained Men.” European Journal of Applied Physiology 120(1): 117–125. 10.1007/s00421-019-04256-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Guerrero, Carlos , Collado‐Boira Eladio, Martinez‐Navarro Ignacio, Hernando Barbara, Hernando Carlos, Balino Pablo, and Muriach María. 2021. “Impact of Plasma Oxidative Stress Markers on Post‐race Recovery in Ultramarathon Runners: A Sex and Age Perspective Overview.” Antioxidants 10(3): 355. 10.3390/antiox10030355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Haslacher, Helmuth , Ratzinger Franz, Perkmann Thomas, Batmyagmar Delgerdalai, Nistler Sonja, Scherzer Thomas M., Ponocny‐Seliger Elisabeth, et al. 2017. “A Combination of Routine Blood Analytes Predicts Fitness Decrement in Elderly Endurance Athletes.” PLoS One 12(5): e0177174. 10.1371/journal.pone.0177174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hoffman, Martin D. 2016. “Injuries and Health Considerations in Ultramarathon Runners.” Physical Medicine and Rehabilitation Clinics of North America 27(1): 203–216. 10.1016/j.pmr.2015.08.004. [DOI] [PubMed] [Google Scholar]
  16. Jäger, Ralf , Kerksick Chad M., Campbell Bill I., Cribb Paul J., Wells Shawn D., Skwiat Tim M., Purpura Martin, et al. 2017. “International Society of Sports Nutrition Position Stand: Protein and Exercise.” J Int Soc Sports Nutr 14(1): 20. 10.1186/s12970-017-0177-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jakubowski, Hieronim . 2019. “Homocysteine Modification in Protein Structure/Function and Human Disease.” Physiological Reviews 99(1): 555–604. 10.1152/physrev.00003.2018. [DOI] [PubMed] [Google Scholar]
  18. Jee, H. , and Jin Y.. 2012. “Effects of Prolonged Endurance Exercise on Vascular Endothelial and Inflammation Markers.” Journal of Sports Science and Medicine 11(4): 719–726. https://www.ncbi.nlm.nih.gov/pubmed/24150084. [PMC free article] [PubMed] [Google Scholar]
  19. Kim, Hyo Jeong , Lee Yoon Hee, and Kim Chang Keun. 2007. “Biomarkers of Muscle and Cartilage Damage and Inflammation during a 200 Km Run.” European Journal of Applied Physiology 99(4): 443–447. 10.1007/s00421-006-0362-y. [DOI] [PubMed] [Google Scholar]
  20. Knechtle, Beat , Knechtle Patrizia, Rosemann Thomas, and Senn Oliver. 2011. “What Is Associated with Race Performance in Male 100‐km Ultra‐marathoners‐‐anthropometry, Training or Marathon Best Time?” Journal of Sports Sciences 29(6): 571–577. 10.1080/02640414.2010.541272. [DOI] [PubMed] [Google Scholar]
  21. Knechtle, Beat , Knechtle Patrizia, Rosemann Thomas, and Lepers Romuald. 2011. “Personal Best Marathon Time and Longest Training Run, Not Anthropometry, Predict Performance in Recreational 24‐hour Ultrarunners.” The Journal of Strength & Conditioning Research 25(8): 2212–2218. 10.1519/JSC.0b013e3181f6b0c7. [DOI] [PubMed] [Google Scholar]
  22. Knechtle, Beat , and Nikolaidis Pantelis T.. 2018. “Physiology and Pathophysiology in Ultra‐Marathon Running.” Frontiers in Physiology 9: 634. 10.3389/fphys.2018.00634. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Knechtle, Beat , Rosemann Thomas, Knechtle Patrizia, and Lepers Romuald. 2010. “Predictor Variables for a 100‐km Race Time in Male Ultra‐marathoners.” Perceptual and Motor Skills 111(3): 681–693. 10.2466/05.25.PMS.111.6.681-693. [DOI] [PubMed] [Google Scholar]
  24. Knechtle, Beat , Rüst Christoph Alexander, Rosemann Thomas, and Lepers Romuald. 2012. “Age‐related Changes in 100‐km Ultra‐marathon Running Performance.” Age 34(4): 1033–1045. 10.1007/s11357-011-9290-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Knechtle, Beat , Valeri Fabio, Zingg Matthias Alexander, Rosemann Thomas, and Rüst Christoph Alexander. 2014. “What Is the Age for the Fastest Ultra‐marathon Performance in Time‐Limited Races from 6 H to 10 Days?” Age 36(5): 9715. 10.1007/s11357-014-9715-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lippi, Giuseppe , Salvagno Gian Luca, Danese Elisa, Skafidas Spyros, Tarperi Cantor, Guidi Gian Cesare, and Schena Federico. 2014. “Mean Platelet Volume (MPV) Predicts Middle Distance Running Performance.” PLoS One 9(11): e112892. 10.1371/journal.pone.0112892. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lippi, Giuseppe , Salvagno Gian Luca, Danese Elisa, Tarperi Cantor, La Torre Antonio, Guidi Gian Cesare, and Schena Federico. 2015. “The Baseline Serum Value of Alpha‐Amylase Is a Significant Predictor of Distance Running Performance.” Clinical Chemistry and Laboratory Medicine 53(3): 469–476. 10.1515/cclm-2014-0850. [DOI] [PubMed] [Google Scholar]
  28. Mairbäurl, Heimo . 2013. “Red Blood Cells in Sports: Effects of Exercise and Training on Oxygen Supply by Red Blood Cells.” Frontiers in Physiology 4: 332. 10.3389/fphys.2013.00332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. McKay, Alannah K. A. , Stellingwerff Trent, Smith Ella S., Martin David T., Mujika Iñigo, Goosey‐Tolfrey Vicky L., Sheppard Jeremy, and Burke Louise M.. 2022. “Defining Training and Performance Caliber: A Participant Classification Framework.” International Journal of Sports Physiology and Performance 17(2): 317–331. 10.1123/ijspp.2021-0451. [DOI] [PubMed] [Google Scholar]
  30. Partyka, Alicja , and Waśkiewicz Zbigniew. 2021. “The Consequences of Training and Competition to the Musculoskeletal System in Ultramarathon Runners: A Narrative Review.” Frontiers in Physiology 12: 738665. 10.3389/fphys.2021.738665. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Perroni, Fabrizio , Migliaccio Silvia, Borrione Paolo, Vetrano Mario, Amatori Stefano, Sisti Davide, Rocchi Marco B. L., et al. 2020. “Can Haematological and Hormonal Biomarkers Predict Fitness Parameters in Youth Soccer Players? A Pilot Study.” International Journal of Environmental Research and Public Health 17(17): 6294. 10.3390/ijerph17176294. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Peters, Edith M. , Shaik Junaid, and Kleinveldt Neil. 2010. “Upper Respiratory Tract Infection Symptoms in Ultramarathon Runners Not Related to Immunoglobulin Status.” Clinical Journal of Sport Medicine 20(1): 39–46. 10.1097/JSM.0b013e3181cb4086. [DOI] [PubMed] [Google Scholar]
  33. Rama, L. , Minuzzi L., Carvalho H., Costa R., and Teixeira A.. 2016. “Changes of Hematological Markers during a Multi‐Stage Ultra‐marathon Competition in the Heat.” Int J Sports Med 37(2): 104–111. 10.1055/s-0035-1555929. [DOI] [PubMed] [Google Scholar]
  34. Roebuck Gregory S., Fitzgerald Paul B., Urquhart Donna M., Ng S.‐Ki, Cicuttini Flavia M., & Fitzgibbon Bernadette M. (2018). The Psychology of Ultra‐marathon Runners: A Systematic Review. Psychology of Sport and Exercise, 37, 43–58. 10.1016/j.psychsport.2018.04.004 [DOI] [Google Scholar]
  35. Romer, Tobias , Rüst Christoph Alexander, Zingg Matthias Alexander, Rosemann Thomas, and Knechtle Beat. 2014. “Age and Ultra‐marathon Performance ‐ 50 to 1,000 Km Distances from 1969 ‐ 2012.” SpringerPlus 3(1): 693. 10.1186/2193-1801-3-693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Shin, K.‐A. , Park Ki Deok, Ahn Jaeki, Park Yongbum, and Kim Y.‐Joo. 2016. “Comparison of Changes in Biochemical Markers for Skeletal Muscles, Hepatic Metabolism, and Renal Function after Three Types of Long‐Distance Running: Observational Study.” Medicine (Baltimore) 95(20): e3657. 10.1097/MD.0000000000003657. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Shin, Young Oh , and Lee Jeong Beom. 2013. “Leukocyte Chemotactic Cytokine and Leukocyte Subset Responses during Ultra‐marathon Running.” Cytokine 61(2): 364–369. 10.1016/j.cyto.2012.11.019. [DOI] [PubMed] [Google Scholar]
  38. Spittler, Jack , and Oberle Lauren. 2019. “Current Trends in Ultramarathon Running.” Current Sports Medicine Reports 18(11): 387–393. 10.1249/JSR.0000000000000654. [DOI] [PubMed] [Google Scholar]
  39. Tiller, Nicholas B. , Roberts Justin D., Beasley Liam, Chapman Shaun, Pinto Jorge M., Smith Lee, Wiffin Melanie, et al. 2019. “International Society of Sports Nutrition Position Stand: Nutritional Considerations for Single‐Stage Ultra‐marathon Training and Racing.” J Int Soc Sports Nutr 16(1): 50. 10.1186/s12970-019-0312-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Tirabassi, Jill N. , Olewinski Lucianne, and Khodaee Morteza. 2018. “Variation of Traditional Biomarkers of Liver Injury after an Ultramarathon at Altitude.” Sport Health 10(4): 361–365. 10.1177/1941738118764870. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Turner, James E. , Bennett Stuart J., Bosch Jos A., Griffiths Helen R., and Aldred Sarah. 2014. “Ultra‐endurance Exercise: Unanswered Questions in Redox Biology and Immunology.” Biochemical Society Transactions 42(4): 989–995. 10.1042/BST20140120. [DOI] [PubMed] [Google Scholar]
  42. Vezzoli, Alessandra , Dellanoce Cinzia, Mrakic‐Sposta Simona, Montorsi Michela, Moretti Sarah, Tonini Annamaria, Pratali Lorenza, and Accinni Roberto. 2016. “Oxidative Stress Assessment in Response to Ultraendurance Exercise: Thiols Redox Status and ROS Production According to Duration of a Competitive Race.” Oxidative Medicine and Cellular Longevity 2016: 1–13. 10.1155/2016/6439037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Waśkiewicz, Zbigniew , Kłapcińska Barbara, Sadowska‐Krępa Ewa, Czuba Milosz, Kempa Katarzyna, Kimsa Elżbieta, and Gerasimuk Dagmara. 2012. “Acute Metabolic Responses to a 24‐h Ultra‐marathon Race in Male Amateur Runners.” European Journal of Applied Physiology 112(5): 1679–1688. 10.1007/s00421-011-2135-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wu, H.‐June , Chen K. T., Shee B. W., Chang H. C., Huang Y. J., and Yang R. S.. 2004. “Effects of 24 H Ultra‐marathon on Biochemical and Hematological Parameters.” World Journal of Gastroenterology 10(18): 2711–2714. 10.3748/wjg.v10.i18.2711. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information S1

EJSC-24-431-s003.docx (16.7KB, docx)

Supporting Information S2

EJSC-24-431-s001.docx (111.5KB, docx)

Supporting Information S3

EJSC-24-431-s002.docx (23.3KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from European Journal of Sport Science are provided here courtesy of Wiley

RESOURCES