The measurements of energy expenditure in relation with an increased oxygen uptake are commonly effected during incremental cycling or running exercises performed in dry air condition. They give essential information on the performance and/or the limitations of subjects. Sport and occupational activities during immersion concern both civil and military swimmers and divers and represent an actual health problem.

We felt the necessity to conceive an innovative method to improve the accuracy in measurements and also to do not impede the swimming motion. Indeed, the literature brings numerous data on measurements of the propulsive force during free or tethered (stationary) swimming but the methods often required the use of an external mechanical load which might impede the swimming motion. During free swimming, the propulsive force was already estimated using either 1) a threedimensional kinematic analysis [1,2], 2) a numerical technique of computational fluid dynamics to calculation of the steady flow around the swimmer’s hand and arm [3-5], or 3) an evaluation of the thrust force by performing unsteady flow measurements using the method of particle image velocimetry in several horizontal planes for subjects swimming in a flume [6]. During tethered swimming, some authors have already measured either the mean drag force in subjects pushing off against handle grips [7,8] or against a rigid arm fixed at the shoulder but the methods required the use of external mechanical loads impeding the swimming motion [4]. Only Pilmanis et al. [9] had conceived a light-weight ergometer to be used used in open-ocean diving physiology research.Few purely ergonomic studies concerned the use of fins for swimming [9-12].

A lot of previous studies have measured the oxygen uptake (VO₂) during fin swimming [13-17]. However, these studies did not simultaneously measure VO₂ and the propulsive force. We developed a new method to simultaneously measure the propulsive force and VO₂ in fin swimming with no forward displacement. The mechanical device built to measure the force did not counteract the swimming motion with fins. The force measurement was coupled with breath-by-breath VO₂ measurement during an incremental exercise produced by increasing step-by-step the kick frequency. The device has been briefly cited in a previous study comparing the changes in respiratory gas exchange, breathing pattern, and heart rate response during incremental fin swimming and dry land cycling protocols [18].The major focus of the present study is the technical manner in which the measurement of propulsive force and VO₂ are possible.

Ethical approval and subjects

This research adheres to the principles of the Declaration of Helsinki. The procedures were carried out with the adequate understanding and written informed consent of the subjects and the protocol was approved by our Institutional Ethics Committee. Fourteen subjects participated to this study (mean age: 39 ± 4 year; body mass: 70 ± 4 Kg; height: 178 ± 3 cm). Seven practised or had practised fin swimming during diving using self contained underwater breathing apparatus (SCUBA), the seven others had no experience of fin swimming. The subjects only wore swimming trunks. The trials were conducted in summer in an open swimming pool and the water temperature was 28.0 ± 0.3°C.

Measurements of physiological variables

During immersion, the subjects breathed through a mouthpiece connected to a snorkel (internal diameter: 2.5 cm; length: 24 cm). The added dead space constituted by the snorkel in fin swimming was 105 ml. Minute ventilation was measured with a volumetric rotor transducer (Triple V digital volume transducer, Jaeger, Germany), connected to the snorkel, and a side pore was connected to fast-response (90 % response time in 100 ms) paramagnetic O₂ and infrared CO₂ analyzers (Oxycon beta, Sebac MSR, Gennevilliers, France). A plastic cone connected to the snorkel protected the volumetric rotor transducer against the water projections (Figure 1). The software computed breath-by-breath VO₂ data of from the simultaneous measurements of ventilation and O₂ partial pressure in the expired gas. VO₂ was expressed in ml STPD. min^-1.s^-1.

Figure 1: Illustrations showing the placement in the swimming pool of the original device created to measure the propulsive force and the position of the subjects during stationary fin swimming. We also showed the plastic cone connected to the snorkel to protect the volumetric rotor transducer against water projections.

Measurement of propulsive force

Figure 2 shows the front and lateral views of the device built to measure the propulsive force during fin swimming. A rigid lever arm supporting the handles and the apparatus stand pivoted out of the water around an axis with ball bearing. It was connected by an iron chain to a load cell (SCAIME model ZF 100, AS Technologies, Langlade, France), linear from 0 to 300 N, and its sensitivity was 3 mV/N. The signal output of the load cell was addressed to a computer and simultaneously recorded with ventilation and VO₂. The load cell data acquisition frequency was 20 Hz.

Figure 2: Front and lateral views of the device created to measure the propulsive force during stationary fin swimming in a pool. The length of immersed rigid arm can be adjusted to increase the gain of the load applied to the force transducer. The distance between the vertical handles may be adapted to each subject. Adjustable rubber stops allow avoiding any transversal displacement of the device during swimming.

The whole apparatus was calibrated with weights in the range of values measured at the maximal power during fin swimming, i.e., up to 150 N (Figure 3). During calibration of the device, the length of the immersed lever arm was adjusted to give twice the true propulsive force to take advantage of the load cell range for sensitivity optimization. Thus, the computer divided by two the force data given by the load cell to measure the true propulsive force. The computer only considered the peak force developed at each kick. The values of the peak forces and VO₂ were averaged for consecutive 30-s periods, i.e. four times for each 2-min step increase in kick frequency. Adjustable rubber stops allowed avoiding any transversal displacement of the device against the front wall of the pool during swimming. During fin swimming, the subjects had no forward displacement. They stayed in horizontal position, and pushed against vertical handles with the arms stretched out (Figure 1). The subject was given a visual feedback from the load cell display. To produce an incremental fin swimming exercise, the kick frequency was increased by steps (10 beats every 2 min) from an initial step at 20 beats/min until VO₂ had reached a plateau level (VO₂ max). An electronic metronome connected to a loudspeaker gave the rate of kicking. An immersed observer confirmed that the swimmer really obeyed to the acoustic pace maker and also estimated the amplitude of leg motion. The knee angle was estimated between 30 and 45 degrees.

Figure 3: The process used to calibrate with added weights the device measuring the propulsive force. The response of the load cell was linear in the range of the force measured during our protocol of maximal incremental fin swimming exercise. The length of the immersed arm of the device was adjusted to give twice the value of force.

In the most efficient subjects who reached the kick frequency of 80 min^-1, the maximal duration of immersion, including the 20-min period of measurements at rest and the incremental fin swimming exercise maximally lasted 40 min. Each subject performed only one trial.

Statistical analysis

Data are presented as the mean ± SEM. Statistical inferences were made by a one-way analysis of variance (ANOVA) with repeated measures at predetermined epochs (rest and each step of the fin swimming bout).Correlations between variables (Propulsive force, kick frequency, and VO₂) were evaluated with the Spearman’s test and significant differences assessed by the Rank Sum test. Significance was accepted if p<0.05. A Student’s t test was used to compare the slopes and ordinate intercepts with their SEM of the regression lines obtained between the propulsive force and the kick frequency or VO₂.

Results

For each participant, the mean peak value of the propulsive force was proportional to the kick frequency and VO₂ during the incremental fin swimming trial (Figure 4). The slopes of regression lines between the mean peak propulsive force and the kick frequency did not significantly differed between subjects having an experience or not of fin swimming, but the ordinate intercept was significantly higher in experienced subjects. Thus, they developed higher peak propulsive force for the same kick frequency than subjects having no experience. On the other hand, the slope of the peak propulsive force vs. VO₂ relationship was significantly (p<0.01) elevated in experienced subjects (Figure 4). The measured VO₂ max was in the range of that reported in previous swimming studies [18,19].

Figure 4: Regression lines with 95% confidence intervals obtained between the propulsive force, the kick frequency and oxygen uptake (VO₂) measured during the maximal incremental fin swimming protocol in subjects having or not an experience of swimming with fins.

Data summary

We conceived a new device helped to measure the propulsive force during fin swimming which did not impede the motions of the swimmer. We showed that the propulsive force was linearly proportional to the kick frequency and VO₂. This device allowed to demonstrate that experienced subjects produced higher propulsive force for the same kick frequency and VO₂ than subjects having no experience of fin swimming,

Limitations of the study

The swimmers breathed through a snorkel. This will result in a negative breathing pressure that may affect ventilation and work output. However, the hydrostatic pressure difference between the lung centroid pressure and the ambient air pressure must be the same than that encountered by fin swimmers during their conventional activities. Indeed, the subjects were immersed under 10 to 15 cm of water and the length of the snorkel was the same than that used during the conventional fin swimming.

Since the expiratory gases were sampled and analyzed at the outlet of the snorkel, the measurements of end tidal values of oxygen and carbon dioxide might be altered, resulting in a wrong calculation of VO₂. In our previous fin swimming study [18], the participants repeated the same dry-land cycling protocol with the device used to measure ventilation and respiratory gas exchange in fin swimming, which consisted of a mouth piece and a 24-cm snorkel connected to the volumeter. We showed that the mean slopes of the ventilation vs. VO₂ regressions did not significantly differ from those measured when the subjects breathed through the face mask when swimming. Also, the baseline end-tidal values of oxygen and carbon dioxide in expired gas did not significantly differ in resting participants while breathing through the face mask or the snorkel.

During our fin swimming trials, the subjects stayed in horizontal position, and pushed against the vertical handles with the arms stretched out. This position was rarely reproduced during fin swimming which often obliges to actively move the arms. Because our device obliged the subject to thrust the handles and thus to do not exercise with the arms, one may suppose that both the propulsive force and the oxygen uptake should be higher in the actual fin swimming activity.

Ergonomic data

A lot of previous studies have measured during swimming the oxygen uptake using different processes often based on the collection of expiratory gas to measure VO₂ and CO₂ but the breath-by-breath measurements of VO₂ coupled to the concomitant measurement of propulsive force were never performed [13-17,20]. Moreover, as said above, our device did not impede swimming movements, as done in the protocols using active or passive drags to measure the propulsive force. It merits being underlined that our mechanical arm gave values of propulsive force in the same range than that reported in other studies measuring either the active drag or the passive drag on a towed swimmer who was not moving [8,21]. In our protocol, the drag force was null because the swimmer stayed strictly stationary in the pool. The peak propulsive force measured in stationary fin swimming does not represent the total mechanical power because there was no effective measurement of the kick depth which must markedly influence the energetic cost of fin swimming at the same kick frequency [11]. The observer in the swimming pool roughly estimated that the knee angle varied between 30 to 45 degrees among individuals. Thus, we are not authorized to compute the total mechanical work load developed during our protocol of incremental fin swimming exercise. However, the propulsive force was proportional to VO₂ that represented the total energy expenditure during swimming at a given kick frequency and closely corresponded to the total mechanical work This suggests that the thrust force measured when the swimmers pushed with the arms against the cell load is representative of the total work developed during swimming. Our data also clearly indicated the higher performances of subjects having an experience of fin swimming who developed higher propulsive force for a given kick frequency and VO₂.

The combination of measurements of the propulsive force and oxygen uptake during stationary fin swimming might constitute a useful tool to assess the individual capacities to swim with fins and to evaluate the benefits of training. The dimensions of our device conceived to measure the propulsive force and the possibility to easily fit watertight load cells authorize to use it in underwater experiments. Obviously, the apparatus measuring ventilation and VO₂ cannot be used in these conditions.

All authors disclose any financial and personal relationships with other people or organisations that could inappropriately influence their work.