ORIGINAL RESEARCH published: 06 May 2020 doi: 10.3389/fphys.2020.00352 Relationship Between Bioimpedance Vector Displacement and Renal Function After a Marathon in Non-elite Runners Lexa Nescolarde 1* , Emma Roca 1 , Paco Bogónez-Franco 1 , José Hernández-Hermoso 2,3 , Antoni Bayes-Genis 4,5,6 and Jordi Ara 7 1 Department of Electronic Engineering, Universitat Politècnica de Catalunya, Barcelona, Spain, 2 Department of Orthopedic Surgery and Traumatology, Hospital Universitari Germans Trias i Pujol, Barcelona, Spain, 3 Department of Surgery, Faculty of Medicine, Universitat Autònoma de Barcelona, Campus UAB, Barcelona, Spain, 4 Department of Cardiology, Hospital Universitari Germans Trias i Pujol, Barcelona, Spain, 5 Department of Medicine, Faculty of Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain, 6 Research Program, Fundació Institut d’Investigació en Ciències de la Salut Germans Trias Edited by: i Pujol, Barcelona, Spain, 7 Department of Nephrology, Hospital Universitari Germans Trias i Pujol, Barcelona, Spain Volker Scheer, Fondation de la Science des Sports Ultra, France Purpose: This study investigates the relationship between whole-body bioimpedance Reviewed by: vector displacement, using bioelectrical impedance vector analysis (BIVA), and renal Henry Lukaski, University of North Dakota, function through serum biomarkers [creatinine, urea, sodium, C-reactive protein (CRP), United States and creatine kinase] and urine biomarkers after a marathon. James M. Kovacs, University of Colorado Colorado Methods: Bioimpedance measurements were taken among 19 non-elite runners at Springs, United States 24 h pre-race, immediately post-race, and at 48 h post-race. The bioimpedance *Correspondence: measurements were analyzed by BIVA using the Hotelling’s T2 test. The runners were Lexa Nescolarde lexa.nescolarde@upc.edu divided according to a cutoff of serum creatinine level immediately post-race in G1 (<1.2 mg/dl of serum creatinine level) and G2 (≥1.2 mg/dl of serum creatinine level). Specialty section: The increase of the serum creatinine levels in 83% of G2 runners was related to acute This article was submitted to Exercise Physiology, kidney injury (AKI) stage 1. a section of the journal Frontiers in Physiology Results: Neither G1 nor G2 showed a creatinine clearance rate (CCr) lower than Received: 16 October 2019 60 ml/min. G2 showed a significant increase in CRP values at 48 h post-race vs Accepted: 26 March 2020 baseline compared to G1 (P < 0.05), with over 5 mg/L (6.8–15.2) in 92% of the runners, Published: 06 May 2020 and in CK values with over 215 U/L (282–1,882) at 48 h post-race in 100% of the Citation: Nescolarde L, Roca E, runners. By BIVA, the 95% confidence ellipses of G2 showed shorter bioimpedance Bogónez-Franco P, vectors than G1, with a noticeable minor Xc/H (P < 0.01), indicating an expansion Hernández-Hermoso J, on extracellular water and inflammation. The runners with 48 h post-race Xc/H values Bayes-Genis A and Ara J (2020) Relationship Between Bioimpedance ≤30.5 , with a decrease from −3 to −12% with respect to the Xc/H value at 24 h Vector Displacement and Renal pre-race, indicated AKI stage 1 with 85.7% sensitivity and 91.7% specificity, with a Function After a Marathon in Non-elite Runners. Front. Physiol. 11:352. direct correlation between AKI stage 1 with greater CRP values at 48 h post-race and doi: 10.3389/fphys.2020.00352 bioimpedance vector displacement, but not with CK values at 48 h post-race. Frontiers in Physiology | www.frontiersin.org 1 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners Conclusion: Through this data collection, it was evidenced that a transient reduction in renal function is more related to inflammatory factors than muscle damage. The BIVA method along with serum biomarkers could be used to follow up the kidney function in runners. Keywords: marathon race, BIVA, serum creatinine, C-reactive protein, serum creatine kinase, AKI stage 1 specific, marker of muscle tissue damage. Exertional (exercise- INTRODUCTION induced) rhabdomyolysis (ER) is well described in athletes The usual practice of sport has a direct positive impact on and is the general term for muscle breakdown associated with physical health in general (Hawley et al., 2014). In marathon strenuous exercise. In endurance exercise, there is no consensus races (42.195 km), non-elite participants increased by 25-fold in about the CK level threshold, although healthy population the last 40 years. For example, in the 2017 Barcelona marathon, CK values >15,000 U/L are predictive of renal impairment there were 19,740 runners, whereas in the 2007 marathon (Rawson et al., 2017). In marathon races, CK can increase there were 7,430 runners (Zurich Barcelona Marathon, 2019). threefold at 24 h post-race (Kim et al., 2009) without showing However, marathon running can induce transient changes in a correlation between AKI and muscle breakdown (Mansour renal filtration function due to neurohormonal, inflammatory, et al., 2017). This is due to the fact that the marathon effort and oxidative stress factors or to hemodynamic influences on does not involve eccentric contractions, and rhabdomyolysis renal blood flow (McCullough et al., 2011; Hewing et al., 2015; is also less common (Scalco et al., 2016). In ultra-endurance Lipman et al., 2016). The glomerular filtration rate (GFR) races, CK can increase 35-fold (Kim et al., 2009). Indeed after provides the best index of overall kidney function, whereas serum a 161-km ultramarathon, a CK of greater than 100,000 U/L has creatinine is the most sensitive serum biomarker for detecting been frequently observed without apparent health consequences small changes in GFR (Dalton, 2010; Ferguson and Waikar, (Hoffman et al., 2012). Military recruits (Kenney et al., 2012) in 2012). Acute kidney injury (AKI) was defined using the AKI basic training (n = 499) reported mean CK values of 734, 1,226, Network (AKIN) criteria (Mehta et al., 2007). Stage 1 AKI was and 667 U/L on days 3, 7, and 14 post-exercise, respectively, defined as a 1.5-fold to twofold increase or 0.3 mg/dl increase and a maximum of 35,056 U/L; however, nobody developed in serum creatinine from day 0 to peak creatinine value on clinical exertional rhabdomyolysis. The authors suggested that either day 1 or day 2 (McCullough et al., 2011; Kellum et al., an increase of CK values by 50-fold above the normal limit 2013). Other authors relate an acute kidney injury after contrast can be a specific marker for exertional rhabdomyolysis in medium (CI-AKI) as an absolute increase ≥0.5 mg/dl over military recruits. baseline serum creatinine (Maioli et al., 2012). The measurement The bioelectrical impedance vector analysis (BIVA) (Piccoli of urea concentration, in blood or urine, is another indicator et al., 1994) is a method to classify soft tissue mass and hydration of kidney function similar to creatinine (Kratz et al., 2002; based on patterns of direct bioimpedance measurements. McCullough et al., 2011). Urea is the principal nitrogen waste Bioimpedance (Z) is measured in a whole-body configuration product of protein metabolism, and it is eliminated from the (standard bioimpedance measurement) (Foster and Lukaski, body almost exclusively by the kidneys in urine, although serum 1996) at 50 kHz using phase-sensitive devices. The components of creatinine is the main biomarker to define acute kidney injury the bioimpedance vector—resistance (R) and reactance (Xc)—are (Kellum et al., 2013). normalized by the height (H) of the subjects (R/H and Xc/H) and C-reactive protein (CRP) is one of the proteins involved in are represented in the resistance–reactance graph (RXc-graph). physical stress reactions, implicated in renal function acting as a The vector measured on an individual is compared against risk factor for AKI and increasing in AKI (Tang et al., 2018). After the normal intervals of the reference population, expressed in a marathon, the CRP serum concentrations start to rise within percentiles of 50, 75, and 95% of normal (Gaussian) distribution 8 h after exercise and peak at 24 h after due to the inflammatory in a bivariate, probabilistic plot (Morrison, 1967). The RXc- response caused by tissue injury. In non-elite marathon runners, graph method allows (1) the evaluation of an individual vector: CRP increases threefold at 24 h post-race, with this increase the displacements parallel to the major axis of tolerance ellipses being even greater (40-fold) in ultraendurance races (Kim et al., indicating progressive changes of fluids, whereas displacements 2009). CRP is considered as a biomarker of inflammation in AKI, parallel to the minor axis are associated to variations of impairing tubular regeneration and promoting fibrosis of injured cell population density contained in soft tissues and different renal tissue (Tang et al., 2018). trajectories indicate combined changes in both fluid and soft A strenuous, untrained exercise can damage muscle fibers, tissue mass; (2) the evaluation of bioimpedance follow-up, by increasing membrane permeability and sarcomere degeneration. the trajectory of successive measurements of the impedance As a consequence, at 24 h post-race, large amounts of muscle vector; and (3) the evaluation of groups of subjects using the proteins will be released into the bloodstream, like creatinine bivariate 95% confidence ellipses of the mean vectors. Two kinase (CK) (Brancaccio et al., 2007, 2010; Shin et al., 2016), mean vectors, from two independent groups of subjects, can be which is easily assessed in the serum and is a sensitive, but not compared with the two-sample Hotelling’s T2 test (Hotelling, Frontiers in Physiology | www.frontiersin.org 2 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners 1931). A mean vector displacement in one group of subjects Core temperature was measured using a tympanic thermometer can be evaluated with the paired one-sample Hotelling’s T2 (Braun ThermoScan 7—IRT6520). test (Piccoli and Pastori, 2002). The main advantage of BIVA The Barcelona Marathon 2017 (42.195 km on asphalt at sea is that it does not require any assumption of soft tissue level) began at 8:30 h at a temperature of 11–14◦ C and 68– hydration (normal or abnormal) and body composition and is 79% humidity. During the race, the runners were provided independent of body weight. The assessment of soft tissue is done with guidelines to maintain adequate levels of hydration by through direct phase-sensitive bioimpedance measurement— eating and drinking at aid stations every 5 km, with such without requiring the use of prediction equations, avoiding sustenance as mineral water, sports drinks, fruits, and nuts. The imprecision of individual estimations (> 10% variability) first liquid intake was programmed at 60 min into the race: associated with regression equations (Piccoli et al., 1994; 400 ml for lighter/slower runners and 800 ml for heavier/faster Lukaski et al., 2019). In addition, severe dehydration can runners, drinking 100–150 ml every 15–20 min. Commercialized occur before the appearance of clinical signs; however, these beverages were provided to runners with an average of 480 mg/l can be assessed by combining measurements of phase angle for Na+ , 85 mg/l for K+ , and 45 mg/l for Mg2+ . The race (PA) and Z vector length from phase-sensitive bioimpedance time recorded of each participant was the official time from the devices operating at 50 kHz and plotted on the RXc-graph race organization. (Lukaski et al., 2019). Ethics approval was obtained from the Hospital Universitari The major clinical applications of BIVA are focused on kidney Germans Trias i Pujol (ICOR-2017-04, REF.CEI, PI-17-037) failure to adjust the ultrafiltration sessions (Piccoli, 1998, 2004; according to the principles of the Helsinki Declaration for Pillon et al., 2004), to relate edema with mortality (Piccoli experiments with human beings. All the participants gave signed et al., 1996; Nescolarde et al., 2004; Piccoli, 2004; Bouchard informed consent. et al., 2009), and more recently (Maioli et al., 2018) as a valid predictor of CI-AKI. The applications of BIVA method Clinical Data on endurance events are focused on relating fluid loss after a Data on serum and urine biomarkers of kidney function bout of exercise-induced dehydration with a lengthening of the (serum creatinine, urea, serum sodium, and urine test strip), impedance vector along the major axis of the tolerance ellipse inflammation (C-reactive protein), and muscle damage (creatine compared to pre-trial measurements (Gatterer et al., 2014) and, kinase) were collected 24 h pre-race, immediately post-race, and with body mass changes concomitant with training and intensity at 48 h post-race. of the exercise, finding a significant decrease in body mass and Fasting blood samples (3 × 10 ml) were obtained from an increase in bioimpedance variables (Carrasco-Marginet et al., the antecubital vein in EDTA vacutainers at 24 h pre-race, 2017; Giorgi et al., 2018). immediately post-race, and at 48 h post-race. The blood samples Although previous studies have examined the effect of were centrifuged at 3,000 rpm at 4◦ C for 10 min in a bench- endurance exercise in the displacement of the bioimpedance top centrifuge. The supernatant serum was aliquoted and stored vector, none of them has examined the relationship between on dry ice until all samples were frozen at −80◦ C in sealed the serum biomarkers of kidney function with bioimpedance Eppendorf tubes to avoid evaporation. Serum urea, creatinine, measurements using BIVA. This study investigates, for the CRP, CK, and Na+ were determined using an AU-5800 Chemistry first time, the relationship between whole-body bioimpedance Analyzer (Beckman Coulter Inc., CA, United States). Serum vector displacement by BIVA, using a 50-kHz phase-sensitive osmolality was determined with an Advanced Micro-Osmometer, bioimpedance analyzer, with renal function through serum Model 3300 (Advanced Instruments Inc., MA, United States) and urine biomarkers in non-elite marathon runners using freezing point depression. Urine tests were measured with after a marathon. AUTION MAX AX-4030 (ARKRAY Inc., MN, United States) analyzer with Uriflet S 9UB strips. MATERIALS AND METHODS Whole-Body Bioimpedance Participants Measurements Two months before the 2017 Barcelona Marathon, 23 Caucasian Tetra-polar bioimpedance measurements were taken at 24 h pre- non-elite runners were voluntarily recruited through the event’s race, immediately post-race, and at 48 h post-race according to organizing committee; four of them did not finish the marathon the protocol of the standard distal BIA configuration, known and thus did not complete the study. The participants who as whole-body (Foster and Lukaski, 1996), with the participant were under 18 years or over 50 years, along with those of in supine decubitus. Four adhesive contact electrodes Ag/AgCl known chronic pathology, were excluded. The sample size of (COVIDIEN Ref. 31050522, COVIDIEN LLC, Mansfield, IL, 19 males (41 ± 4 years old; 77.4 ± 7.1 kg; 1.80 ± 0.1 m; United States)—two for injection of current (I) and two for 24.0 ± 2.1 kg/m2 ) ran 2.7 ± 1.6 h/training per week and had detecting voltage (V)—were dorsally placed on the right hand in a running history of 8.2 ± 5.1 years, on average, prior to the the third metacarpo-phalangeal articulation and in the carpus, race. All of them were weighed, only with running pants on, 2 h respectively, 5 cm apart. The pair on the foot was located in before the race, immediately after completing the race, and at 48 h the third metatarso-phalangeal and in the articulation, 6 cm after the race using the same scale (Jata 565, Barcelona, Spain). apart (Figure 1). Frontiers in Physiology | www.frontiersin.org 3 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners version 24.0 (Armonk, NY: IBM Corp., United States) was used for data analysis. Bioelectrical Impedance Vector Analysis In this study, the individual bioimpedance vectors of runners [normalized by the runner’s height (H) R/H, Xc/H, and PA] measured at 24 h pre-race, immediately post-race, and at 48 h post-race were plotted on the tolerance ellipses (RXc-graph) of the Caucasian ethnicity adult population (Piccoli et al., 1995) by using the BIVA software (Piccoli and Pastori, 2002). The differences in the mean bioimpedance vectors obtained FIGURE 1 | The standard distal BIA configuration electrode placement. at 24 h pre-race and immediately post-race were plotted with their 95% confidence on the dRXc-graph. The paired one- sample Hotelling’s T2 test (Hotelling, 1931), a multivariate The measurements were obtained at 50 kHz with a phase- extension of Student’s t-test for paired data, was used to compare sensitive bioimpedance analyzer (BIA-101 Anniversary; AKERN- the differences of the mean bioimpedance vectors. If a 95% Srl, Florence, Italy), which injected a constant, sinusoidal confidence ellipse of a vector displacement covers the null alternating current of 245 µARMS. The measurement or vector (i.e., abscissa 0, ordinate 0), the vector displacement is technical errors of the system, determined with a parallel circuit not statistically significant (P < 0.05) at a paired one-sample of precision resistor and capacitor, were <1  for R and <2% Hotelling’s T2 test. for capacitance. The two-sample Hotelling’s T2 test, a multivariate extension The supporting staff used a towel to remove dirt and sweat of Student’s t-test for unpaired data, was used to compare before the electrode application. The measurement was done the mean vectors from two groups (G1 and G2) according to 15 min immediately post-race, with a core temperature at around the cutoff of ≥1.2 mg/dl serum creatinine value immediately 36◦ C. Inside the sanitary facility, the temperature was 21–24.5◦ C post-race. If the 95% confidence intervals of group mean and the humidity was 36–38%. do not overlap, the group means are significantly different (P < 0.05), but the reverse is not necessarily true. Hotelling’s T2 test is more sensitive than Student’s t-test performed on each Statistical Analysis variable and entails a smaller risk of erroneously rejecting the Data were collected at 24 h pre-race, immediately post-race, and null hypothesis. at 48 h post-race, considering the values taken at 24 h pre- The sample size was estimated for independent groups race as baseline. The normality of distribution of the variables according to R/H (n = 6 and 3 by group) and Xc/H (n = 14 was checked by the Shapiro–Wilk test and the homogeneity of and 7 by group) value to have a power (β = 0.80) to detect variances by Levene’s test. an effect size of 0.996, with a 0.05 significance level using Repeated-measures ANOVA test was used to determine the MedCalc statistical software version 19.0.3 (MedCalc Software effect of the marathon on variables measured at 24 h pre-race, bvba, Ostend, Belgium). immediately post-race, and at 48 h post-race using the multiple- In addition, the area under the receiver operating comparison Bonferroni tests. Friedman test was used for repeated characteristic (ROC) curve was used to measure the measure of data non-normally distributed with the Wilcoxon test. discriminative capacity of R/H, Xc/H, and PA immediately In addition, the samples were separated in two groups post-race and at 48 h post-race in a clinical situation of according to the cutoff of ≥1.2 mg/dl of serum creatinine AKI stage 1 according to the cutoff of ≥1.2 mg/dl of serum post-race (G1, serum creatinine post-race <1.2 mg/dl and G2, creatinine post-race. serum creatinine post-race ≥1.2 mg/dl) considering normal values of 0.6–1.2 mg/dl by the Clinical Laboratory of the Hospital Universitari Germans Trias i Pujol (AU-5800 Chemistry RESULTS Analyzer (Beckman Coulter Inc; California, United States). Creatinine clearance rate (CCr or CrCl) is a useful measure Table 1 shows the mean ± SD values of body weight, body mass for approximating the GFR (Stevens and Levey, 2009). It index (BMI), race time, and percentage change of body weight was calculated according to the Cockcroft–Gault equation as well as the result of the statistical analysis. A decrease of (Cockcroft and Gault, 1976) and taking into account that CCr body weight is related to a negative percentage change of body <60 ml/min/1.73 m2 precedes the onset of kidney failure (Kdigo., weight, while an increment of body weight is related to a positive 2013). Mann–Whitney’s U test and Student’s t-test were applied percentage change with respect to the baseline values. to analyze non-parametric and parametric variables, respectively. The repeated-measures ANOVA test showed statistical The variables normally distributed are shown as mean ± SD, significance (P < 0.01) for both body weight and BMI. while the non-normally distributed data are shown as median The multiple-comparison Bonferroni tests showed significant and interquartile range (IQR). The level of statistical significance differences (P < 0.01) immediately post-race compared to was set at P < 0.05. The statistical software IBM SPSS R R 24 h pre-race and immediately post-race compared to 48 h Frontiers in Physiology | www.frontiersin.org 4 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners TABLE 1 | Anthropometric and training data of 19 non-elite marathon runners. Parameters 24 h pre-race (1) Immediately 48 h F P valuea SE P SE P SE P post-race (2) post-race (3) value1-2b value1-3b value2-3b Weight (kg) 77.4 ± 7.1 75.0 ± 7.0 78.1 ± 7.1 79.691 0.285 0.233 0.219 0.000 0.000 0.492 0.000 BMI (kg m−2 ) 24.0 ± 2.1 23.2 ± 2.0 24.2 ± 2.1 75.945 0.092 0.073 0.070 0.000 0.000 0.0561 0.000 Difference Weight (%)2−1 −3.2 ± 1.5 – – – – – Difference Weight (%)3−1 – 0.5 ± 1.3 – – – – Race time (hh:mm:ss) 3:33:00 ± 0:24:16 a Repeated measures ANOVA test; b Bonferroni test. BMI, body mass index. post-race for body weight and BMI values. However, no statistical The serum urea post-race values increased significantly significance (P > 0.05) was found between values in body (P < 0.01) compared to baseline, and 10 of 19 runners (53%) weight and BMI at 24 h pre-race and at 48 h post-race. Among were above 43 mg/dl. Serum creatinine increased significantly the runners, 63.2% recovered their body weight baseline values post-race compared to baseline (P < 0.01), and 12 of 19 runners at 48 h post-race. (63%) had values over the cutoff of ≥1.2 mg/dl. The CRP at 48 h post-race increased significantly (P < 0.01) compared to baseline, and 13 of 19 runners (68%) showed values above 5.0 mg/L (5.2– Clinical Data and BIVA at 24 h Pre-race, 15.5 mg/L), with an increase of more than 800% of their baseline Immediately Post-race, and at 48 h values. The CK at 48 h post-race increased significantly (P < 0.01) Post-race compared to baseline, and 18 of 19 runners (95%) showed values Table 2 shows the clinical data measured at 24 h pre-race, above 215 U/L (256–1,882 U/L), with an increase from 70 to immediately post-race, and at 48 h post-race as mean ± SD 1,500% of their baseline values. Any CK values at 48 h post-race for normally distributed variables and as median (IQR) for were greater than 2,000 U/L. non-normally distributed data. In addition, also presented are The sodium serum levels also increased significantly the results of repeated-measures ANOVA test using multiple- immediately post-race with respect to baseline (P < 0.01), but comparison Bonferroni tests and of Friedman test with Wilcoxon within normal reference values; however, only two of 19 runners test for data which were non-normally distributed. (10%) had sodium serum levels above 145 mmol/L post-race. A repeated test showed statistical significance (P < 0.01) in The osmolality serum levels also increased significantly serum creatinine, serum urea, serum sodium, C-reactive protein, immediately post-race with respect to baseline (P < 0.01), and creatine kinase. but within normal reference values (285–295 mmol/kg), TABLE 2 | Serum and urine biomarkers of 19 non-elite marathon runners. Parameters (normal reference values) 24 h Immediately 48 h F P valuea SE P SE P SE P pre-race (1) post-race (2) post-race (3) value1-2b value1-3b value2-3b Serum Creatinine (mg/dL) (0.6–1.2 mg/dL) 0.96 ± 0.09 1.28 ± 0.20 0.91 ± 0.09 46.491 0.039 0.013 0.038 0.000 0.000 0.014 0.000 Serum Urea (mg/dL) (17–43 mg/dL) 35.2 ± 8.9 43.2 ± 9.4 35.2 ± 6.9 10.326 1.919 1.496 1.851 0.001 0.002 1.000 0.001 Serum Sodium (mmol/L) (135–145 mmol/L) 139.3 ± 1.1 141.6 ± 3.0 139.2 ± 1.5 6.214 0.720 0.449 0.716 0.009 0.010 1.000 0.007 Serum Osmolality (mmol/kg) (285–295 mmol/kg) 287.0 ± 3.9 292.7 ± 6.6 285.2 ± 2.2 14.109 1.488 1.488 1.488 0.000 0.001 0.663 0.000 Parameters (normal reference values) 24 h Immediately 48 h Chi- ZP ZP ZP pre-race (1) post-race (2) post-race (3) square P value2-1d value3-1d value3-2d valuec Serum CRP (mg/L) (0–5.0 mg/L) 0.70 0.50 7.50 30.778 −1.880 −3.824 −3.825 (0.20–5.30) (0.20–3.00) (3.20–15.50) 0.000 0.050 0.000 0.000 Serum CK (U/L) (20–215 U/L) 140 (76–400) 376 (260–946) 685 (194–1882) 31.684 −3.823 −3.823 −3.099 0.000 0.000 0.000 0.000 Hematuria 0 1 case 0 – – – – Proteinuria 0 5 cases 0 – – – – a Repeated measures ANOVA test; b Bonferroni test; c Friedman test with Chi-square; d Wilcoxon test. CRP, C-reactive protein; CK, creatine kinase. Frontiers in Physiology | www.frontiersin.org 5 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners TABLE 3 | Whole-body bioimpedance measurements of 19 non-elite marathon runners. Parameters 24 h pre-race Immediately 48 h F P valuea SE P SE P SE P (1) post-race (2) post-race (3) value1-2b value1-3b value2-3b R/H (/m) 271.7 ± 15.3 282.5 ± 19.4 264.6 ± 18.0 23.977 3.595 2.740 2.608 0.000 0.023 0.055 0.000 Xc/H (/m) 31.2 ± 2.8 34.0 ± 3.6 29.6 ± 2.8 41.134 0.665 0.368 0.547 0.000 0.000 0.000 0.001 PA (◦ ) 6.5 ± 0.4 6.9 ± 0.5 6.4 ± 0.5 11.371 0.121 0.086 0.102 0.001 0.048 0.218 0.000 r 0.69 0.79 0.62 – – – – a Repeated measures ANOVA test; b Bonferroni test. R/H, resistance/height; Xc/H, reactance/height; PA, phase angle. except in four of 19 runners (21%) who had levels above The 95% confidence ellipse did not cover the null vector 295 mmol/kg post-race. (i.e., dZ/H = 0.0), indicating a statistically significant vector Regarding the urine test, there were some cases of hematuria displacement (P < 0.05) at a paired one-sample Hotelling’s T2 (one of 19 runners, 5%) and proteinuria (five of 19 runners, 25%) test (Hotelling, 1931). found only after the race. Table 3 shows the values of R/H, Xc/H, and PA described as mean ± SD, the Pearson’s correlation coefficient, r, between R/H Clinical Data and BIVA According to the and Xc/H, and the result of repeated-measures ANOVA test using Cutoff of ≥ 1.2 mg/dl of Serum multiple-comparison Bonferroni tests. Creatinine Immediately Post-race A repeated ANOVA test showed high statistical significance Table 4 shows the results of anthropometric and clinical data (P < 0.01) in R/H and Xc/H as well as PA. Using multiple- according to a cutoff of ≥ 1.2 mg/dl of serum creatinine comparison Bonferroni tests, Xc/H showed significant differences level immediately post-race for G1 (< 1.2 mg/dl of serum (P < 0.01) immediately post-race compared to values at 24 h pre- creatinine level) and G2 (≥1.2 mg/dl of serum creatinine race and at 48 h post-race compared to values at 24 h pre-race. level) as mean ± SD for normally distributed variables and as In addition, Xc/H showed significant differences (P < 0.01) at median (IQR) for non-normally distributed data. In addition, 48 h post-race compared to immediately post-race. However, R/H also presented are the statistical results of Mann–Whitney’s and PA only showed significant differences immediately post-race U test or Student’s t-test for non-parametric and parametric compared to values at 24 h pre-race (P < 0.05) and at 48 h post- variables, respectively. race compared to the values immediately post-race (P < 0.01). According to the cutoff of ≥ 1.2 mg/dl serum creatinine, Any significant differences were found in R/H and PA at 48 h 63% of the runners (12 of 19) corresponding to G2 showed an post-race compared to values at 24 h pre-race (P > 0.05). increment of serum creatinine post-race and 58% of them (seven Using the BIVA software (Piccoli and Pastori, 2002), Figure 2 of 12) showed an increment in serum urea post-race of above shows the individual bioimpedance vectors at time points 24 h 43 mg/dl (cutoff value of serum urea). Serum creatinine reached pre-race, immediately post-race, and 48 h post-race for 19 non- values of AKI stage 1, showing increases of ≥ 0.3 mg/dl in 83% of elite marathon runners compared with the tolerance ellipses for the runners (10 of 12) and increases of ≥ 0.5 mg/dl in 42% of the healthy Caucasian adults (Piccoli et al., 1995) on RXc-graph. Two runners (five of 12). runners (runner_11 and runner_17) had a vector displacement The CCr post-race showed a significant decrease (P < 0.01) immediately post-race outside of the 50% tolerance ellipse for upon comparing G1 and G2, although none of the runners (G1 healthy Caucasian adults; however, this reverted to 50% tolerance and G2) showed CCr values below 60 ml/min/1.73m2 . The CK ellipse at 48 h post-race. Thus, all vector positions were within at 48 h post-race showed a significant increase (P < 0.05) upon the range of normal values. Two other runners (runner_14 and comparing G1 and G2. All G2 runners had CK values over runner_15) had a vector displacement at 48 h post-race outside 215 U/L (282–1,882 U/L). The C-reactive protein at 48 h post- of the inferior pole of the 50% tolerance ellipse for healthy race showed a significant increase (P < 0.01) upon comparing Caucasian adults, indicating an expansion on extracellular water G1 and G2. It should be noted that in G2 there are the only and inflammation. cases of hematuria (8%) and proteinuria (42%) among all the Figure 3A shows the mean bioimpedance vector. The runners analyzed. Regarding serum urea, serum sodium, and vector displacement and lengthening along the major axis was serum osmolality, no significant differences were seen between significant compared to baseline but did not cross through the groups (P > 0.05). the 75% area of tolerance ellipses of the reference population. According to the cutoff of ≥ 1.2 mg/dl serum creatinine, no The paired graph (Figure 3B) shows the 95% confidence significant differences were observed in weight loss immediately ellipse corresponding to the mean of the difference obtained post-race (P > 0.05). G1 showed a weight loss immediately from immediately post-race compared to the vector (baseline) at 1.6 to 4.1% and likewise for G2 from 1.9 to 5.0%. However, 24 h pre-race and at 48 h post-race compared to the baseline significant differences were observed in both body weight and vector using the BIVA software (Piccoli and Pastori, 2002). BMI (P < 0.05). Frontiers in Physiology | www.frontiersin.org 6 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners FIGURE 2 | Continued Table 5 shows the values of R/H, Xc/H, and PA immediately Figure 4A shows an overlapping of 95% confidence ellipses post-race and at 48 h post-race, described as mean ± SD of G1 between G1 and G2 immediately post-race according to the cutoff and G2. In addition, also presented are the Pearson’s correlation of ≥ 1.2 mg/dl serum creatinine level immediately post-race. coefficient, r (R/H, Xc/H), and the statistical significance of BIVA method with Hotelling’s T2 test shows no significant Student’s t-test for two independent samples. differences (T2 = 3.87, F = 1.82, P > 0.05). However, Figure 4B No statistical significance was found in whole-body shows separated 95% confidence ellipses between G1 and G2 bioimpedance measurements immediately post-race upon at 48 h post-race according to the cutoff of ≥ 1.2 mg/dl comparing G1 vs. G2. However, a high statistical significance serum creatinine level immediately post-race. Through BIVA (P < 0.01) was found in Xc/H at 48 h post-race, with a tendency method with Hotelling’s T2 test, the mean groups are significantly to be statistically significant in R/H and PA. different (T2 = 18.54, F = 8.73, P < 0.05). Frontiers in Physiology | www.frontiersin.org 7 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners FIGURE 2 | Individual vectors on RXc-graph for 19 non-elite marathon runners 24 h pre-race (1), immediately post-race (2), and 48 h post-race (3); compared to the tolerance ellipses (50%, 75% and 95% percentiles) of the reference population. Frontiers in Physiology | www.frontiersin.org 8 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners FIGURE 3 | Mean bioimpedance vectors (A) 24 h pre-race, immediately post-race, and 48 h post-race on RXc-graph and paired graph (B) for 19 non-elite marathon runners. The paired graph shows the 95% confidence ellipse corresponding to the mean bioimpedance vector difference obtained immediately post-race and 48 h post-race compared to baseline (24 h pre-race). According to Figure 2, the runners that belong to the group According to AKI stage 1, the Xc/H at 48 h post-race variable G2 are runner_1, runner_2, runner_5, runner_6, runner_7, shows a high significant difference (P = 0.003), with area below runner_10, runner_11, runner_14, runner_15, runner_16, the ROC curve of 0.917 and a standard error of 0.067. runner_18, and runner_19. Figure 5 shows the area under the ROC curve of R/H, Xc/H, and PA at 48 h post-race in the clinical situation of AKI stage 1 DISCUSSION according to the cutoff of ≥ 1.2 mg/dl serum creatinine post-race. This study analyzes the relationship between whole-body bioimpedance vector displacement, by BIVA, and renal function TABLE 4 | Anthropometric and clinical data of 19 non-elite marathon runners through serum and urine biomarkers, after a marathon, in non- according to cut-off of ≥1.2 mg/dL of serum creatinine level immediately elite marathon runners. post-race. Transient significant increases of urea and creatinine, Parameters Immediately G1, n = 7 G2, n = 12 P value indicating renal dysfunction, have been described in runners after post-race (<1.2 mg/dL) (≥1.2 mg/dL) completion of the marathon (Kratz et al., 2002; McCullough et al., 2011). A repeated-measures test showed high statistical Weight (kg) 70.6 ± 5.7 77.5 ± 6.5 0.035 significance (P < 0.01) in serum creatinine, urea, serum sodium, Weight-lost (%) −2.83 ± 1.28 −3.44 ± 1.55 0.397 CRP, and CK. Using multiple-comparison tests, serum creatinine BMI (kgm−2 ) 22.0 ± 1.7 23.9 ± 1.9 0.047 showed statistical significance (P < 0.01) immediately post- Serum Urea (mg/dL) 40.9 ± 9.2 44.4 ± 9.2 0.429 Serum Creatinine (mg/dL) 1.07 ± 0.07 1.39 ± 0.15 0.000 race and at 48 h post-race with respect to the values at 24 h Increment Serum Creatinine 0.17 ± 0.04 0.40 ± 0.16 0.000 pre-race and at 48 h post-race with respect to the values (mg/dL) immediately post-race. The inflammation biomarker CRP and CCr (mL/min) 93.2 ± 10.1 76.4 ± 9.6 0.002 the muscle damage biomarker CK showed significantly different Serum CRP§ (mg/L) 4.40 7.50 0.001 values (P < 0.01) at 48 h post-race compared to baseline values (2.80–5.50) (3.90–8.70) according to its rate of production, with concentration peaks Serum CK§ (U/L) 389 871 0.035 at 48 h. The mean bioimpedance vector showed a lengthening (194–890) (296–1882) along the major axis, with a migration above (left) the minor Serum Sodium (mmol/L) 142.0 ± 2.2 141.5 ± 3.4 0.730 axis of the tolerance ellipse, immediately post-race compared to Osmolality (mmol/kg) 290.1 ± 5.0 294.3 ± 7.1 0.286 baseline, relating to the fluid loss and muscle activation during Hematuria 0 cases 1 case – a marathon. A 50-kHz, phase-sensitive BIA analyzer yields Proteinuria 0 cases 5 cases – passive bioelectrical measurements of bioimpedance module (Z) § 48hpost-race values. BMI, body mass index; CCr, creatinine clearance rate; CRP, and its components R (which is the opposition to the flow of C-reactive protein; CK, creatine kinase. an alternating current through intra- and extra-cellular ionic Frontiers in Physiology | www.frontiersin.org 9 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners TABLE 5 | Whole-body bioimpedance measurements (R/H, Xc/H, and PA) immediately post-race, 48 h post-race and statistical results between G1 and G2, according to cut-off of ≥1.2 mg/dL serum creatinine level immediately post-race. G1, n = 7 (<1.2 mg/dL) G2, n = 12 (≥1.2 mg/dL) P value2 P value3 (G1 vs. G2) (G1 vs. G2) Immediately 48 h Immediately 48 h post-race (2) post-race (3) post-race (2) post-race (3) R/H (/m) 292.9 ± 14.4 274.1 ± 6.7 276.4 ± 19.8 259.0 ± 8.3 0.072 0.077 Xc/H (/m) 35.6 ± 2.6 31.9 ± 1.9 33.1 ± 3.9 28.2 ± 2.4 0.151 0.003 PA (◦ ) 6.94 ± 0.41 6.64 ± 0.40 6.82 ± 0.49 6.23 ± 0.44 0.609 0.065 r 0.60 0.52 0.65 0.59 – – 2 Immediately post-race G1 vs. G2; 3 48 h post-race G1 vs. G2. R/H, resistance/height; Xc/H, reactance/height; PA, phase angle. FIGURE 4 | 95% confidence ellipse for both groups immediately post-race (A) and 48 h post-race (B) separated according to cut-off of ≥1.2 mg/dL of serum creatinine level immediately post-race. solutions) and Xc (which indicates the capacitive component developed AKI stage 1 (G2) had more CK and more CRP at 48 h of tissue interfaces and cell membranes). The parameter Xc is post-race than the group which did not develop AKI stage 1. G2 associated with the integrity of the soft tissue structures and R showed an increase of CK at 48 h post-race by threefold to 16- with the hydration state of the soft tissue. The phase angle (PA), fold above the reference values measured at 24 h pre-race, while calculated as the arc tangent of Xc/R, allows the non-invasive G1 showed an increase of one- to sixfold above the reference assessment of intra- to extracellular fluid (Lukaski et al., 2019). values. Although no runners presented CK over 1,890 U/L at Changes in bioimpedance variables are indicators of fluid (R) 48 h post-race (ninefold above the normal reference values), not and cell structure integrity (Xc and PA). Muscle contraction can close to exertional (exercise-induced) rhabdomyolysis (Kenney be described based on two variables, length and tension, which et al., 2012), we believe that the inflammation situation that imply a major delay between I and V and, as consequence, major occurs, together with other factors that may be intra-renal or Xc and PA post-race. thermal hemodynamic, may be the cause of the kidney damage The paired graph (mean of the difference post-race compared (Mansour et al., 2017). Although it is not known which renal to the baseline vector) did not cover the null vector (i.e., structure is damaged, there are data that suggest that the tubule dZ/H = 0.0), indicating a statistically significant vector may be damaged (Mansour et al., 2017). The alterations in the displacement (P < 0.05) at a paired one-sample Hotelling’s urinary sediment and the urinalysis observed (hematuria and T2 test. Although the mean differences at 48 h post-race proteinuria) are less described and also certify the involvement compared to the baseline vector did not cover the null vector, of renal parenchyma. To date, an increased risk of chronic renal there is a tendency to cover it (recovering the pre-race values). failure in marathoners has not been described, but we cannot Several factors induce transient changes on renal filtration deny that repeated kidney damage may have long-term negative function, such as hemodynamic influences on renal blood effects. AKI stage 1 showed a significant Spearman correlation flow, neurohormonal, inflammatory, or oxidative stress factors with CRP values at 48 h post-race (r = 0.767, P = 0.000) (McCullough et al., 2011). In the current study, the subjects who and with CCr immediately post-race (r = −0.869, P = 0.000). Frontiers in Physiology | www.frontiersin.org 10 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners According to the cutoff of ≥ 1.2 mg/dl serum creatinine post- race, neither has there been significant differences (P > 0.05) between G1 and G2 in terms of serum osmolality post-race values used to evaluate the hemoconcentration influences on renal blood flow. By BIVA, the 95% of confidence ellipses at 48 h post- race showed significant differences (T2 = 18.54, F = 8.73, P < 0.05) between runners with serum creatinine levels post-race of ≥ 1.2 mg/dl (G2) compared to those with serum creatinine levels < 1.2 mg/dl (G1). This significant difference is mainly due to the parameter Xc/H (P < 0.01), which showed a decrease in G2 at 48 h post-race with respect to values at 24 h pre-race from -3.3 to -12.4% and in G1 from 0 to -3.1%. According to the BIVA method (Piccoli et al., 1994), this decrease is due to the expansion of extracellular water (fluid overload) and inflammation. The Xc/H at 48 h post-race showed a significant Spearman rho correlation with CRP at 48 h post-race (r = −0.467, P = 0.044). AKI stage 1 showed a significant Pearson correlation with Xc/H at 48 h post-race (r = − 0.710, P = 0.001) and phase-angle (PA) at 48 h post-race (r = − 0.586, P = 0.008). In addition, among FIGURE 5 | ROC Curve of R/H, Xc/H, and PA 48 h post-race according to R/H, Xc/H, and PA at 48 h post-race, the variable with major cut-off of ≥1.2 mg/dL of serum creatinine post-race. area under ROC curve was Xc/H 48h post-race (AUC = 0.917). The runners with serum creatinine levels post-race ≥ 1.2 mg/dl (G2) showed Xc/H at 48 h post-race values ≤ 30.5 , with 85.7% However, AKI stage 1 shows no correlation with CK at 48 h sensitivity and 91.7% specificity. post-race (r = 0.376, P = 0.113). Exertional rhabdomyolysis The major limitation of this study is the small sample size can be a complication presenting as asymptomatic physiologic and the fact that only the male gender was analyzed. When elevations of creatine kinase after exercise, leading to acute performing the BIVA method, the male and the female samples kidney injury (Szczepanik et al., 2014), but none of the non- cannot be processed together because of the reference values elite runners analyzed presented with rhabdomyolysis. This (R and Xc) of the healthy population which change according could indicate that due to the fact that marathon does not to gender (male and female) and age range (childhood and involve eccentric contractions, rhabdomyolysis is less common adults). For this reason, we found the tolerance ellipses (50, 75, (Clarkson, 2007). and 95%) of the reference population to beseparated by gender Previous studies have related the effects of endurance exercise (Piccoli et al., 1995). However, the strength of this study is the to serum C-reactive protein and inflammatory markers (Kasapis use of a non-invasive method (BIVA) to assess renal function and Thompson, 2005). All runners analyzed had increases in and muscle damage. CRP of 190–7,200% at 48 h after the race compared to the reference levels, showing that global inflammation may be related to the inflammation of the nephron (correlated with a significant CONCLUSION increase of creatinine levels). On the other hand, G2 showed a significant increase (P < 0.05) The transient reduction in renal filtration function is more related in CRP values at 48 h post-race (290–7000%) compared to to the inflammatory factors rather than to muscle damage. In baseline (24 h pre-race). Among G2 runners, 92% showed very all the runners observed, serum creatinine and urea values at high levels of CRP values over the cutoff (> 5 mg/L), which is 48 h post-race compared to baseline values were recovered. associated with an independent predictor of AKI (Tang et al., There is a direct relationship between a marathon race in non- 2018), and 83% of them showed high increase levels of creatinine elite runners and the serum biomarkers of kidney function and (≥ 0.3 mg/dl), which is associated with AKI stage 1 (McCullough inflammation, with a high percentage of AKI stage 1. The runners et al., 2011). Thus, the transient reduction in renal filtration with AKI stage 1 presented greater CRP values at 48 h post- function is more related to inflammatory factors (CRP, P < 0.01) race, which is associated with systemic inflammation. Phase- than to muscle damage (CK, P < 0.05). sensitive measurements at 50 kHz with BIVA could be a non- According to the cutoff of ≥ 1.2 mg/dl serum creatinine, no invasive method to assess kidney function in runners through significant differences were observed in weight loss immediately the Xc/H value at 48 h post-race, with a decrease of > 3% after the race (P > 0.05) between G1 and G2. After a marathon, with respect to the baseline value. It would be necessary to body weight loss is at around 2–3% (Sawka et al., 2007; Thomas study several consecutive marathon races, adding 24 h post-race et al., 2016) and at 5–6% after ultra-endurance events without measurements, on runners who frequently suffer transient AKI clinical implications (Tam and Noakes, 2013). G1 showed a stage 1, including female runners, to determine the long-term weight loss of 2–4% and G2 of 2–5% without renal implications. negative effects on renal function. Frontiers in Physiology | www.frontiersin.org 11 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners DATA AVAILABILITY STATEMENT version of the manuscript. LN, processed the data and prepared the tables and figures. LN and ER, drafted the manuscript. The datasets generated for this study are available on request to the corresponding author. FUNDING ETHICS STATEMENT This study was supported by the Electronic and Biomedical Instrumentation Research Group of the Electronic Engineering Ethics approval was obtained from the Hospital Universitari Department, Universitat Politècnica de Catalunya and Hospital Germans Trias i Pujol (ICOR-2017-04, REF.CEI, PI-17-037) Universitari Germans Trias i Pujol, Barcelona, Spain. according to principles of the Helsinki Declaration for experiments with human beings. The patients/participants provided their written informed consent to participate in this study. ACKNOWLEDGMENTS We would like to especially thank all the runners, without whom AUTHOR CONTRIBUTIONS this study would not have been possible. In addition, we would like to thank Jaume Barallat from the Clinic Laboratory of LN, ER, PB-F, JH-H, AB-G, and JA designed the experiments, Hospital Universitari Germans Trias i Pujol (Barcelona, Spain) analyzed the data, revised the manuscript, and approved the final for his invaluable collaboration in this study. REFERENCES endurance run. Wilderness Environ. Med. 23, 56–60. doi: 10.1016/j.wem.2011. 11.001 Bouchard, J., Soroko, S. B., Chertow, G. M., Himmelfarb, J., Ikizler, T. A., Paganini, Hotelling, H. (1931). The generalization of Student’s ratio. Ann. Math. Statist. 2, E. P., et al. (2009). Fluid accumulation, survival and recovery of kidney function 360–378. in critically ill patients with acute kidney injury. Kidney Int. 76, 422–427. doi: Kasapis, C., and Thompson, P. D. (2005). The effects of physical activity on serum 10.1038/ki.2009.159 C-reactive protein and inflammatory markers: a systematic review. J. Am. Coll. Brancaccio, P., Lippi, G., and Maffulli, N. (2010). Biochemical markers of muscular Cardiol. 45, 1563–1569. doi: 10.1016/j.jacc.2004.12.077 damage. Clin. Chem. Lab. Med. 48, 757–767. doi: 10.1515/CCLM.2010.179 Kdigo. (2013). Chapter 1 Definition and classification of CKD. Kidney Int. Suppl. Brancaccio, P., Maffulli, N., and Limongelli, F. M. (2007). Creatine kinase 3, 19–62. doi: 10.1038/kisup.2012.64 monitoring in sport medicine. Br. Med. Bull. 8, 209–230. doi: 10.1093/bmb/ Kellum, J. A., Lameire, N., Aspelin, P., Barsoum, R. S., Burdmann, E. A., Goldstein, ldm014 S. L., et al. (2013). Diagnosis, evaluation, and management of acute kidney Carrasco-Marginet, M., Castizo-Olier, J., Rodríguez-Zamora, L., Iglesias, X., injury: a KDIGO summary (Part 1). Crit. Care 17, 1–15. doi: 10.1186/cc11454 Rodríguez, F. A., Chaverri, D., et al. (2017). Bioelectrical impedance vector Kenney, K., Landau, M. E., Gonzalez, R. S., Hundertmark, J., O’Brien, K., and analysis (BIVA) for measuring the hydration status in young elite synchronized Campbell, W. W. (2012). Serum creatine kinase after exercise: drawing the line swimmers. PLoS One 12:e178819. doi: 10.1371/journal.pone.0178819 between physiological response and exertional rhabdomyolysis. Muscle Nerve Clarkson, P. M. (2007). Exertional rhabdomyolysis and acute renal failure in 45, 356–362. doi: 10.1002/mus.22317 marathon runners. Sport Med. 37, 361–363. doi: 10.2165/00007256-200737040- Kim, H. J., Lee, Y. H., and Kim, C. K. (2009). Changes in serum cartilage oligomeric 00022 matrix protein (COMP), plasma CPK and plasma hs-CRP in relation to running Cockcroft, D. W., and Gault, M. H. (1976). Prediction of creatinine clearance from distance in a marathon (42.195 km) and an ultra-marathon (200 km) race. Eur. serum creatinine. Nephron 16, 31–41. doi: 10.1159/000180580 J. Appl. Physiol. 105, 765–770. doi: 10.1007/s00421-008-0961-x Dalton, R. N. (2010). Serum creatinine and glomerular filtration rate: perception Kratz, A., Lewandrowski, K. B., Siegel, A. J., Chun, K. Y., Flood, J. G., Van Cott, and reality. Clin. Chem. 56, 687–689. doi: 10.1373/clinchem.2010.144261 E. M., et al. (2002). Effect of marathon running on hematologic and biochemical Ferguson, M. A., and Waikar, S. S. (2012). Established and emerging markers laboratory parameters, including cardiac markers. Am. J. Clin. Pathol. 118, of kidney function. Clin. Chem. 58, 680–689. doi: 10.1373/clinchem.2011. 856–863. doi: 10.1309/14TY-2TDJ-1X0Y-1V6V 167494 Lipman, G. S., Krabak, B., Rundell, S., Shea, K., Badowski, N., and Little, Foster, K. R., and Lukaski, H. C. (1996). Whole-body impedance–what does it C. (2016). Incidence and prevalence of acute kidney injury during measure? Am. J. Clin. Nutr. 64, 388S–396S. doi: 10.1093/ajcn/64.3.388s multistage ultramarathons. Clin. J. Sport Med. 26, 314–319. doi: 10.1097/JSM. Gatterer, H., Schenk, K., Laninschegg, L., Schlemmer, P., Lukaski, H., and 0000000000000253 Burtscher, M. (2014). Bioimpedance identifies body fluid loss after exercise in Lukaski, H. C., Vega Diaz, N., Talluri, A., and Nescolarde, L. (2019). Classification the heat: a pilot study with body cooling. PLoS One 9:e109729. doi: 10.1371/ of hydration in clinical conditions: indirect and direct approaches using journal.pone.0109729 bioimpedance. Nutrients 11:809. doi: 10.3390/nu11040809 Giorgi, A., Vicini, M., Pollastri, L., Lombardi, E., Magni, E., Andreazzoli, A., Maioli, M., Toso, A., Leoncini, M., Gallopin, M., Musilli, N., and Bellandi, F. et al. (2018). Bioimpedance patterns and bioelectrical impedance vector analysis (2012). Persistent renal damage after contrast-induced acute kidney injury: (BIVA) of road cyclists. J. Sports Sci. 36, 2608–2613. doi: 10.1080/02640414. incidence, evolution, risk factors, and prognosis. Circulation 125, 3099–3107. 2018.1470597 doi: 10.1161/CIRCULATIONAHA.111.085290 Hawley, J. A., Hargreaves, M., Joyner, M. J., and Zierath, J. R. (2014). Integrative Maioli, M., Toso, A., Leoncini, M., Musilli, N., Grippo, G., Ronco, C., et al. (2018). biology of exercise. Cell 159, 738–749. doi: 10.1016/j.cell.2014.10.029 Bioimpedance-guided hydration for the prevention of contrast-induced kidney Hewing, B., Schattke, S., Spethmann, S., Sanad, W., Schroeckh, S., Schimke, I., injury: the HYDRA study. J. Am. Coll. Cardiol. 71, 2880–2889. doi: 10.1016/j. et al. (2015). Cardiac and renal function in a large cohort of amateur marathon jacc.2018.04.022 runners. Cardiovasc. Ultrasound 13, 1–8. doi: 10.1186/s12947-015-0007-6 Mansour, S. G., Verma, G., Pata, R. W., Martin, T. G., Perazella, M. A., and Parikh, Hoffman, M. D., Ingwerson, J. L., Rogers, I. R., Hew-Butler, T., and Stuempfle, K. J. C. R. (2017). Kidney injury and repair biomarkers in marathon runners. Am. J. (2012). Increasing creatine kinase concentrations at the 161-km Western States Kidney Dis. 70, 252–261. doi: 10.1053/j.ajkd.2017.01.045 Frontiers in Physiology | www.frontiersin.org 12 May 2020 | Volume 11 | Article 352 Nescolarde et al. BIVA and Renal Function in Non-elite Runners McCullough, P. A., Chinnaiyan, K. M., Gallagher, M. J., Colar, J. M., Geddes, T., Scalco, R. S., Snoeck, M., Quinlivan, R., Treves, S., Laforét, P., Jungbluth, H., et al. Gold, J. M., et al. (2011). Changes in renal markers and acute kidney injury (2016). Exertional rhabdomyolysis: physiological response or manifestation of after marathon running. Nephrology 16, 194–199. doi: 10.1111/j.1440-1797. an underlying myopathy? BMJ Open Sport Exerc. Med. 2:e000151. doi: 10.1136/ 2010.01354.x bmjsem-2016-000151 Mehta, R. L., Kellum, J. A., Shah, S. V., Molitoris, B. A., Ronco, C., Warnock, D. G., Shin, K. A., Park, K. D., Ahn, J., Park, Y., and Kim, Y. J. (2016). Comparison of et al. (2007). Acute kidney injury network: report of an initiative to improve changes in biochemical markers for skeletal muscles, hepatic metabolism, and outcomes in acute kidney injury. Crit. Care 11, 1–8. doi: 10.1186/cc5713 renal function after three types of long-distance running. Medicine 95, 1–6. Morrison, D. F. (1967). Multivariate Statistical Methods. New York, NY: McGraw- doi: 10.1097/MD.0000000000003657 Hill. Stevens, L. A., and Levey, A. S. (2009). Measured GFR as a confirmatory test Nescolarde, L., Piccoli, A., Román, A., Núñez, A., Morales, R., Tamayo, J., for estimated GFR. J. Am. Soc. Nephrol. 20, 2305–2313. doi: 10.1681/ASN. et al. (2004). Bioelectrical impedance vector analysis in haemodialysis patients. 2009020171 Relation between oedema and mortality. Physiol. Meas. 25, 1271–1280. doi: Szczepanik, M. E., Heled, Y., Capacchione, J., Campbell, W., Deuster, P., 10.1088/0967-3334/25/5/016 and O’Connor, F. G. (2014). Exertional rhabdomyolysis: identification and Piccoli, A. (1998). Identification of operational clues to dry weight prescription in evaluation of the athlete at risk for recurrence. Curr. Sports Med. Rep. 13, hemodialysis using bioimpedance vector analysis. Kidney Int. 53, 1036–1043. 113–119. doi: 10.1249/JSR.0000000000000040 doi: 10.1111/j.1523-1755.1998.00843.x Tam, N., and Noakes, T. D. (2013). The quantification of body fluid allostasis Piccoli, A. (2004). Italian CAPD-BIA study group. Bioelectric impedance vector during exercise. Sports Med. 43, 1289–1299. doi: 10.1007/s40279-013-0089-y distribution in peritoneal dialysis patients with different hydration status. Tang, Y., Mak, S. K., Xu, A. P., and Lan, H. Y. (2018). Role of C-reactive protein Kidney Int. 65, 1050–1063. doi: 10.1111/j.1523-1755.2004.00467.x in the pathogenesis of acute kidney injury. Nephrology 23, 50–52. doi: 10.1111/ Piccoli, A., Nigrelli, S., Caberlotto, A., Bottazzo, S., Rossi, B., and Pillon, L. (1995). nep.13454 Bivariate normal values of the bioelectrical impedance vector in adult and Thomas, D. T., Erdman, K. A., and Burke, L. M. (2016). American college of sports elderly populations. Am. J. Clin. Nutr. 61, 269–270. doi: 10.1093/ajcn/61.2.269 medicine joint position statement. nutrition and athletic performance. Med. Sci. Piccoli, A., and Pastori, G. (2002). BIVA Software. Padova: University of Padova. Sports Exerc. 48, 543–568. doi: 10.1249/MSS.0000000000000852 Piccoli, A., Rossi, B., and Pillon, L. (1996). Body fluid overload and bioelectrical Zurich Barcelona Marathon (2019). Available online at: https://www. impedance analysis in renal patients. Miner. Electrolyte. Metab. 22, 76–78. zurichmaratobarcelona.es/esp/ediciones_anteriores.html (accessed April Piccoli, A., Rossi, B., Pillon, L., and Bucciante, G. (1994). A new method for 22, 2019). monitoring body fluid variation by bioimpedance analysis: the RXc graph. Kidney Int. 46, 534–539. doi: 10.1038/ki.1994.305 Conflict of Interest: The authors declare that the research was conducted in the Pillon, L., Piccoli, A., Lowrie, E. G., Lazarus, J. M., and Chertow, G. M. (2004). absence of any commercial or financial relationships that could be construed as a Vector length as a proxy for the adequacy of ultrafiltration in hemodialysis. potential conflict of interest. Kidney Int. 66, 1266–1271. doi: 10.1111/j.1523-1755.2004.00881.x Rawson, E. S., Clarkson, P. M., and Tarnopolsky, M. A. (2017). Perspectives on Copyright © 2020 Nescolarde, Roca, Bogónez-Franco, Hernández-Hermoso, Bayes- Exertional Rhabdomyolysis. Sport. Med. 47, 33–49. doi: 10.1007/s40279-017- Genis and Ara. This is an open-access article distributed under the terms of 0689-z the Creative Commons Attribution License (CC BY). The use, distribution or Sawka, M. N., Burke, L. M., Eichner, E. R., Maughan, R. J., Montain, S. J., and reproduction in other forums is permitted, provided the original author(s) and the Stachenfeld, N. S. (2007). American College of Sports Medicine position stand. copyright owner(s) are credited and that the original publication in this journal Exercise and fluid replacement. Med. Sci. Sports Exerc. 39, 377–390. doi: 10. is cited, in accordance with accepted academic practice. No use, distribution or 1249/mss.0b013e31802ca597 reproduction is permitted which does not comply with these terms. Frontiers in Physiology | www.frontiersin.org 13 May 2020 | Volume 11 | Article 352