A Comparison of Aerobic Fitness Testing on a Swim Bench and Treadmill in a Recreational Surfing Cohort: A Pilot Study

e intermittent manner of sur ng accentuates the importance of both the aerobic and anaerobic energy systems. Currently, the optimal method of assessing sur ng-speci c aerobic tness is using a swim bench (SWB) ergometer; however, their limited availability presents a barrier to surfers wanting to know their maximal aerobic power (VO 2peak). As a result, the aims of this pilot study were to determine the VO 2peakof recreational surfers using a new commercial SWB ergometer and to propose and examine the feasibility of a regression model to predict SWB ergometer VO 2peak values. A total of nine recreational surfers were assessed where body measurements were conducted followed by maximal aerobic capacity testing (swim bench and treadmill) to pro le the cohort. Findings demonstrated that VO 2peakvalues were signi cantly greater (p< 0.001) on the treadmill compared to the SWB ergometer (M= 66.01 8.23 vs. 37.41 8.73 mL/kg/min). Peak heart rate was also signi cantly greater on the treadmill compared to the SWB ergometer. Multiple regression analysis was used to produce a model which predicted SWB VO 2peakvalues with an R 2 value of 0.863 and an adjusted R 2 value of 0.726. The physiological pro ling of the recreational cohort coupled with a surfer's predicted SWB VO 2peakvalue will allow for identi cation of sur ng-speci c aerobic tness levels and

SWB ergometer. Multiple regression analysis was used to produce a model which predicted SWB VO 2peakvalues with an R 2 value of 0.863 and an adjusted R 2 value of 0.726. The physiological pro ling of the recreational cohort coupled with a surfer's predicted SWB VO 2peakvalue will allow for identi cation of sur ng-speci c aerobic tness levels and evidence-based training recommendations. Keywords:aerobic tness; aerobic power; VO 2peak; VO2max; ergometer; treadmill 1. Introduction Sur ng is regarded as part of a lifestyle and culture for those living on the coastal borders of countries like Australia and the United States. Today, sur ng is enjoyed as an iconic pastime by 2.7 million Australians and 37 million individuals worldwide [1,2]. With its growth into a global multi-million-dollar industry and its recent induction into the Olympics, it is reasonable to assume that sur ng's impressive growth over the last decade will continue [3]. Using time-motion analysis, sur ng has been subdivided into intermittent periods of arm paddling, prolonged periods of rest, and wave riding [3]. Monitoring of activity requirements for a 20 min heat in surfers using global positioning system (GPS) technology has revealed that paddling encompassed 42.6–44% of the total time with the majority of these paddling bouts (60%) ranging from 1 to 20 s long [4,5]. This intermittent manner of sur ng accentuates the importance of both the Sports2018,6, 54; doi:10.3390/sports6020054

Sports2018,6, 54 2 of 11 aerobic and anaerobic energy systems [6]. The repeated high and low intensity paddling bouts within a heat may promote a high capacity of oxygen uptake, while allowing for adequate recovery between paddling spells [3]. Meir, et al. [5] reported that during one hour of recreational sur ng, the mean heart rates during arm paddling represented 80% of the laboratory peak heart rate (HR peak) achieved by the surfers during a progressive swim bench (SWB) ergometer peak oxygen uptake (VO 2peak) test. This suggests that a high level of aerobic tness may be essential in sur ng. As evident through the existing body of research surrounding aerobic tness testing in surfers, the preferred and ideal method of testing is completed with an SWB ergometer. Although dynamic leg exercise such as a cycle ergometer or treadmill are the most popular mode of exercise testing, research strongly supports the speci city of tness testing [7]. For instance, VO 2peakvalues during arm work equate to approximately 70% of the values obtained during leg exercise [8]. However, athletes with a highly trained upper-body may achieve arm-crank values approaching 90% of their cycle VO 2peak[8]. Through a systematic review of the literature speci c to aerobic tness assessments in the sur ng population, it is evident that the use of paddling ergometers to measure aerobic tness in the elite cohort is steadily growing however remains limited in the recreational cohort [3,6,9,10] when compared to mainstream sports like kayaking [11–17] (see Appendix To date, only three studies have assessed recreational surfers using SWB ergometers [5,6,10]. Results from these studies ranged from 31.25 6.31–54.20 10.2 mL/kg/min indicating variations in VO 2peakacross the recreational sur ng cohort. Discrepancies in the results may be explained by differences in ergometers used and testing protocols employed. Previous studies measuring VO 2peak in sur ng cohorts have employed a variety of equipment like the arm crank, tether board, treadmill and bicycle, which in theory could be the result of the limited availability of SWB ergometers [18–20]. Despite the treadmill and cycle ergometer being considered the optimal form of

by differences in ergometers used and testing protocols employed. Previous studies measuring VO 2peak in sur ng cohorts have employed a variety of equipment like the arm crank, tether board, treadmill and bicycle, which in theory could be the result of the limited availability of SWB ergometers [18–20]. Despite the treadmill and cycle ergometer being considered the optimal form of equipment for VO 2peaktesting in both children and adults, the aforementioned differences between leg and arm work during maximal and submaximal exercise supports the use of a sur ng-speci c SWB ergometer [3,7]. However, the limited availability of SWB ergometers outside of leading sports performance institutes makes them dif cult to access by recreational athletes. Given these concerns the aims of this pilot study were; to determine the VO 2peakof recreational surfers using a new commercial SWB ergometer (Figure) and to propose and examine the feasibility of a regression model to predict SWB ergometer VO 2peakvalues.Sports 2018, 6, x FOR PEER REVIEW 2 of 11 within a heat may promote a high capacity of oxygen uptake, while allowing for adequate recovery between paddling spells [3]. Meir, et al. [5] reported that during one hour of recreational surfing, the mean heart rates during arm paddling represented 80% of the laboratory peak heart rate (HRpeak) achieved by the surfers during a progressive swim bench (SWB) ergometer peak oxygen uptake (VO2peak) test. This suggests that a high level of aerobic fitness may be essential in surfing. As evident through the existing body of research surrounding aerobic fitness testing in surfers, the preferred and ideal method of testing is completed with an SWB ergometer. Although dynamic leg exercise such as a cycle ergometer or treadmill are the most popular mode of exercise testing, research strongly supports the specificity of fitness testing [7]. For instance, VO2peak values during arm work equate to approximately 70% of the values obtained during leg exercise [8]. However, athletes with a highly trained upper-body may achieve arm-crank values approaching 90% of their cycle VO2peak [8]. Through a systematic review of the literature specific to aerobic fitness assessments in the surfing

strongly supports the specificity of fitness testing [7]. For instance, VO2peak values during arm work equate to approximately 70% of the values obtained during leg exercise [8]. However, athletes with a highly trained upper-body may achieve arm-crank values approaching 90% of their cycle VO2peak [8]. Through a systematic review of the literature specific to aerobic fitness assessments in the surfing population, it is evident that the use of paddling ergometers to measure aerobic fitness in the elite cohort is steadily growing however remains limited in the recreational cohort [3,6,9,10] when compared to mainstream sports like kayaking [11–17] (see Appendix A for search strategy). To date, only three studies have assessed recreational surfers using SWB ergometers [5,6,10]. Results from these studies ranged from 31.25 ± 6.31–54.20 ± 10.2 mL/kg/min indicating variations in VO2peak across the recreational surfing cohort. Discrepancies in the results may be explained by differences in ergometers used and testing protocols employed. Previous studies measuring VO2peak in surfing cohorts have employed a variety of equipment like the arm crank, tether board, treadmill and bicycle, which in theory could be the result of the limited availability of SWB ergometers [18– 20]. Despite the treadmill and cycle ergometer being considered the optimal form of equipment for VO2peak testing in both children and adults, the aforementioned differences between leg and arm work during maximal and submaximal exercise supports the use of a surfing-specific SWB ergometer [3,7]. However, the limited availability of SWB ergometers outside of leading sports performance institutes makes them difficult to access by recreational athletes. Given these concerns the aims of this pilot study were; to determine the VO2peak of recreational surfers using a new commercial SWB ergometer (Figure 1) and to propose and examine the feasibility of a regression model to predict SWB ergometer VO2peak values. (a) (b) Figure 1. (a) Subject completing the catch phase of the paddling stroke on the swim bench ergometer setup in the laboratory; (b) Subject completing the pull phase of the paddling stroke on the swim bench ergometer setup in the laboratory. Figure 1. (a) Subject completing the catch phase of the paddling

model to predict SWB ergometer VO2peak values. (a) (b) Figure 1. (a) Subject completing the catch phase of the paddling stroke on the swim bench ergometer setup in the laboratory; (b) Subject completing the pull phase of the paddling stroke on the swim bench ergometer setup in the laboratory. Figure 1. (a) Subject completing the catch phase of the paddling stroke on the swim bench ergometer setup in the laboratory; (b) Subject completing the pull phase of the paddling stroke on the swim bench ergometer setup in the laboratory.

Sports2018,6, 54 3 of 11 2. Materials and Methods 2.1. Subjects This study involved nine male recreational surfers aged 18–42 years. To be classi ed as a recreational surfer, participants were to have a minimum of 12 months sur ng experience, be currently sur ng and list the sport as their main form of activity, and not be competing at higher than local club level. All testing of participants was conducted at the Bond University Institute of Health and Sport. Subjects were tested following their normal routine of sleep, nutrition and hydration levels. Ethics was granted through the Bond University Human Research Ethics Committee (RO1550) prior to commencement. Participants were informed of the associated risks and bene ts of the study prior to signing an informed consent form. 2.2. Procedures Testing was conducted by two experienced exercise scientists under the supervision of a senior researcher with expertise in maximal aerobic and anaerobic testing of recreational and professional surfers. Body measurements were conducted followed by the aerobic testing. All subjects underwent both a SWB ergometer and treadmill maximal oxygen consumption test. To minimize systematic bias and reduce the effects of fatigue the order of maximal oxygen consumption testing (i.e., treadmill or SWB) was randomized. 2.3. Body Measurements Body measurements collected included height (EcoMed Seca, Hamburg, Germany), mass (Wedderburn, WM204, Sydney, Australia), with body mass index (BMI) subsequently calculated. Height was initially measured to the nearest 0.1 cm and body mass was measured to the closest 100 g with minimal clothing using a standard medical balance scale. 2.4. Aerobic VO 2peakTesting Prior to commencing the incremental exercise testing, subjects were provided with standardized instructions on the use of the SWB ergometer and treadmill. Subsequently subjects completed a 2-min warm-up on the apparatus they were to be tested on that day. Warm-up on the SWB ergometer included paddling at below 10 watts, while warm-up on the treadmill included running at a speed of 8 or 10 km/h (subject preference) at an incline of 0 percent. This served to reduce the risk of injury and familiarize the subject with the equipment. The subjects VO

they were to be tested on that day. Warm-up on the SWB ergometer included paddling at below 10 watts, while warm-up on the treadmill included running at a speed of 8 or 10 km/h (subject preference) at an incline of 0 percent. This served to reduce the risk of injury and familiarize the subject with the equipment. The subjects VO 2peakwas obtained using incremental exercise testing on a SWB ergometer and treadmill. Peak oxygen consumption is considered the gold-standard for quantifying aerobic tness [21]. SWB ergometry has previously been shown to be both valid and reliable to test peak aerobic levels in both recreational and competitive surfers [6,10]. Oxygen consumption was analyzed using an automated gas analysis system (Parvo Medics, TrueOne ® , 2400, Sandy, UT, USA) (O2analyzer, CO2 analyzer, pneumotach) that was calibrated with laboratory grade standard gasses (O2and CO2) prior to each test. The SWB ergometer (KayakPro SwimFast, Miami, FL, USA) incremental test commenced at 10 watts, with increments of 10 watts every minute. This incremental test was adapted from a previously validated protocol by Furness, et al. [6]. Treadmill testing commenced at a speed of 8 or 10 km/h based upon the subject's preference with an incline of 2% for the rst minute [22]. At the start of the second minute the incline was increased to 4% then increased by 1% every subsequent minute [22]. The incremental treadmill test was developed in accordance with research by Sperlich, et al. [22] whose ndings suggested that VO 2peakscores attained by employing individually designed treadmill exercise protocols allow the athlete to pace himself or herself according to their present biological state thus achieving higher VO 2peakvalues when compared with standardized protocols. The testing termination criterion was based upon the ACSM guidelines for exercise testing and prescription [23]. Testing was terminated if age-predicted maximal heart rate was exceeded, respiratory

Sports2018,6, 54 4 of 11 exchange ratio (RER) reached greater than 1.5, oxygen consumption did not increase concurrently with power output, required power output was not maintained for greater than 10 s, volitional exhaustion was achieved, or any symptoms of chest pain were expressed by the participant. Heart rates (HR) were monitored throughout testing via telemetry using a Polar Team Pro HR sensor (Polar H7 Bluetooth HR Sensor) connected to Polar Team Pro software, which was interfaced with the Parvo Medics system. 2.5. Statistical Analysis Statistical analyses were performed using SPSS (Version 25.0; IBM Corp, Armonk, NY, USA). Normal distribution of the data was con rmed through a Shapiro-Wilks test and visual inspection of box plots, normal Q-Q plots and frequency histograms. Descriptive statistics including means (M), standard deviations (SD) and ranges were calculated for key performance variables (VO 2peak, HR peak, percent of age predicted HRmaxand peak aerobic power). A paired samplet-test was used to determine whether there was a statistical mean difference between participants VO 2peakscores on the SWB ergometer and treadmill. A multiple regression analysis was also conducted to produce a model to predict SWB ergometer VO 2peakscores. Prior to analyzing the data, the eight assumptions of multiple regression were considered and satis ed to ensure accuracy of the predictive model [24]. These assumptions are addressed in the results section. To assess the validity of the multiple regression model, a one-samplet-test was used to determine whether the mean difference between the two measures (SWB ergometer VO 2peakvalues versus predicted SWB ergometer VO 2peakvalues) were statistically different from zero. The level of agreement between the two measures was then represented through a Bland Altman plot with the associated 90% limits of agreement. 3. Results 3.1. Recreational Surfer Body Measurements and Aerobic Pro le All nine subjects successfully completed all body measurements and maximal oxygen consumption tests (treadmill and SWB) without incident. The body measurements and sur ng experience of the surfers are listed in Table. Surfers had a mean BMI in the normal range (BMI 20.0 to 24.99 kg/m 2 ) and a mean sur ng experience of

Measurements and Aerobic Pro le All nine subjects successfully completed all body measurements and maximal oxygen consumption tests (treadmill and SWB) without incident. The body measurements and sur ng experience of the surfers are listed in Table. Surfers had a mean BMI in the normal range (BMI 20.0 to 24.99 kg/m 2 ) and a mean sur ng experience of greater than 12 years. Subjects were sur ng 5.22 h each week. Descriptive statistics of physical attributes and experience as well as key performance variables are presented in Tables, respectively. Results of the paired samples t-tests in Table p< 0.05) between the means of treadmill and SWB ergometer key performance variables. The mean relative peak aerobic capacity for the recreational surfers on the treadmill was 66.01 mL/kg/min (range 49.0–70.1 mL/kg/min), which is signi cantly (p< 0.001) greater (+76.4%) than the SWB (M 37.41 mL/kg/min, range 24.0–43.6 mL/kg/min). Participants also had a signi cantly greater (+11.8%,p< 0.003) age-predicted HR peakon the treadmill as compared to the SWB. Table 1.Physical Attributes and Experience of Recreational Surfers M SD (n= 9). Measure Value Height (cm) 176.90 3.97 Mass (kg) 78.37 10.16 BMI (kg/m 2 ) 24.98 2.28 Sur ng Experience (years) 12.89 7.82 Sur ng Frequency (hours per week) 5.22 2.43

Sports2018,6, 54 5 of 11 Table 2.Key Performance Variables for Recreational Surfers M ( SD). Measure Treadmill Swim Bench pValue VO 2peak(L/min) 5.16 ( 0.86) 2.98 ( 0.89) <0.001 VO 2peak(mL/kg/min) 66.01 ( 8.23) 37.41 ( 8.73) <0.001 VCO 2 5.69 ( 1.12) 3.48 ( 0.89) <0.001 VE 134.20 ( 25.60) 92.15 ( 17.61) <0.001 RQ 1.12 ( 0.068) 1.28 ( 0.10) <0.001 HR peak(b min 1 ) 184 ( 10) 165 ( 14) 0.004 Percent of age predicted HR peak(%)98.53 ( 4.61) 88.11 ( 7.79) 0.003 Peak aerobic power (W) N/A 69.91 ( 19.33) N/A 3.2. Development of a Multiple Regression Model Prior to generating the multiple regression model, SPSS was used to analyze the data and ensure all appropriate multiple regression assumptions were satis ed. Linearity of variables was assessed and veri ed by partial regression plots and a plot of studentized residuals against the predicted values. Independence of residuals was also assessed and veri ed by a Durbin–Watson statistic of 2.466. Homoscedasticity was veri ed through the visual inspection of a plot of studentized residuals versus unstandardized predicted values. There was no evidence of multicollinearity, as assessed by a tolerance value greater than 0.1. When analyzing studentized deleted residuals values, leverage points and Cook's distances, two participants were classi ed as outliers and were therefore excluded from the multiple-regression analysis as recommended by Laerd Statistics [24]. When analyzing the remaining participants (n= 7), there was no studentized deleted residuals greater than 3 standard deviations, no leverage points greater than 0.2, and values for Cook's distance above 1. The assumption of normality was met as previously discussed in the statistics section. The regression equation produced is presented below with a summary of the multiple regression analysis presented in Table. This multiple regression model predicted SWB ergometer VO 2peakscores with an R 2 value of 0.863 and an adjusted R 2 value of 0.726. Swim Bench VO 2peak(mL/kg/min) = 33.027 +(0.480 surf experience years) + (3.968 hours surfed per week) + (0.699 treadmill VO 2peak). Table 3.Summary of Multiple Regression Analysis. Variable B SE B Intercept 33.027 18.007 Surf

Table. This multiple regression model predicted SWB ergometer VO 2peakscores with an R 2 value of 0.863 and an adjusted R 2 value of 0.726. Swim Bench VO 2peak(mL/kg/min) = 33.027 +(0.480 surf experience years) + (3.968 hours surfed per week) + (0.699 treadmill VO 2peak). Table 3.Summary of Multiple Regression Analysis. Variable B SE B Intercept 33.027 18.007 Surf experience (years) 0.480 0.260 0.428 Hours surfed (per week) 3.968 1.616 0.547 Treadmill VO 2peak (mL/kg/min) 0.699 0.239 0.676 B = unstandardized regression coef cient; SEB= standard error of the coef cient; = standardized coef cient. 3.3. Validity Assessment of SWB Ergometer Predictive Model: SWB versus Predicted SWB VO 2peakValues A one-samplet-test determined the mean difference (0.0084 mL/kg/min) between the two measures (SWB ergometer VO 2peakvalues versus predicted SWB ergometer VO 2peakvalues) was not statistically (p> 0.05) different from zero. Bland–Altman plots with the associated 90% limits of agreement (5.83 and 5.79) are presented in Figure.

This pilot study assesses VO2peak in recreational surfers using a swim bench ergometer.