Competitive swimming suits made from a proprietary fabric purported to reduce water resistance (active drag) and increase buoyancy were recently designed and are being worn by these athletes in hopes of improving performance. Although it is widely accepted that reducing active drag and increasing buoyancy will decrease the physiological cost of any given swimming speed (2–5,8–18), currently only one study has investigated the drag characteristics of these swimsuits. Toussaint et al. (17) found no difference in active drag with the use of one of these body suits (FastSkinTM by Speedo®). However, this study provided no insight into the buoyant characteristics of these suits and, more importantly, examined no physiological variables as they related to the use of these suits.
Trappe et al. (18) showed that wearing a neoprene wet suit significantly reduced both V̇O2 and V̇E during swimming at velocities ranging from 0.9 to 1.3 m·s−1. Furthermore, the magnitude of this effect was positively related to the amount of body surface covered by the suits. Because neoprene is highly buoyant, part of the explanation for the reduced physiological demands of swimming is likely the added buoyancy provided by the wet suit. However, it is also possible that simply covering the skin might also reduce physiological cost by decreasing active drag independent from changes in buoyancy. Starling et al. (14) showed that wearing torso suits made from the same fabric as standard brief-style swimsuits reduced both V̇O2 (by 4%) and blood lactate concentration (by 16%) while increasing distance per stroke (by 4%) during freestyle swimming at a standardized speed. These findings are consistent with the results of Sharp and Costill (12), who showed that shaving exposed body hair reduced the physiological cost of breaststroke swimming at a standardized speed. In this study V̇O2 was reduced 9%, postswim blood lactate concentration was reduced 20%, and distance per stroke during the swim was increased 12%. Combined with the reduced velocity decay during a push-off and glide test, these results were attributed to a reduction in active drag after shaving exposed body hair.
Collectively, the results of these studies suggest that successful attempts to either reduce active drag or increase the buoyancy of the swimmer will reduce physiological cost of swimming and increase distance per stroke. It was the purpose of the present study to examine the effects of wearing a FastSkinTM suit made from a putative drag-reducing fabric on physiological and biomechanical responses to freestyle swimming.
Ten male swimmers (age = 20.2 ± 1.5 yr, height = 183.8 ± 5.5 cm, and weight = 79.6 ± 7.4 kg) currently training and competing at the collegiate level (NCAA Division I) provided informed written consent before participation in the study. Competitive swimmers were selected as subjects to control for differences in performance attributable to technique and skill level. The Institutional Review Board for Iowa State University approved this study.
A randomized repeated measures design was used to compare passive drag, buoyancy, and submaximal physiological responses to swimming, when wearing a traditional competition brief-style suit to wearing a FastSkinTM (hereafter referred to as the FS suit). For each measurement, the subjects completed testing once wearing an FS suit and again wearing the brief-style suit. The style of the FS suit chosen for this study was the sleeveless, full-torso, ankle-length suit. Appropriate sizing of the suits was determined by the supplier based on height, weight, and waist circumference of the subjects. Once the FS suits were received, they were worn by the subjects only for the in-water testing sessions not including warm-up. The competitive briefs were made of Lycra.
Subjects reported for testing on 2 d, separated by 1 wk. Each subject performed three 183-m freestyle swims. To balance the presentation of suit conditions between data collection sessions, 5 of the 10 subjects were randomly chosen to use the body suit on the first day of collection, whereas the other 5 used the brief-style suit. These conditions were switched on the second day of testing. After checking in, the subjects’ height and weight were measured. If the swimmer was using the brief-style suit, buoyancy measurements were made before warm-up. If the swimmer was using the FS suit, he warmed-up in a brief-style suit, and then put on the FS suit so the buoyancy measurement was made in a dry body suit. Warm-up was self-determined by each swimmer and was repeated before the second data collection session. After completion of the static buoyancy measurement and warm-up, the swimmer received instructions for the swim test. The subject was asked to perform the first swim at a “moderate” pace, the second swim at a “moderately hard” pace, and the third swim at a “hard” pace. In addition, it was suggested that the subject decrease his time by roughly 5 s across consecutive swims. The swims were timed from when the feet broke contact with the starting wall and finished when the hand touched the finishing wall. Immediately upon finishing the swim, postexercise oxygen uptake was collected for 45 s. One minute into the recovery period, a blood sample was collected. Following that, the subject was presented with the Borg scale and selected a representative value for that particular swim (1). The three swims were separated by 2 min of passive recovery. Water temperature (28°C) and environmental conditions (24°C, 75% humidity) were similar for both test days. The subjects performed the same protocol 1 wk later and were asked to duplicate the times recorded for the swims from the previous week.
During a third testing session, subjects were tested in groups of five for the passive drag measurement. It was easier for the subjects to put the FS suit on when dry. Therefore, 9 of the 10 subjects were tested wearing the FS suit first to facilitate changing suits. The remaining subject wore the regular competition suit first in order to share an FS suit with one of the other swimmers. Subjects completed three trials each at a slow and fast speed. Subjects were removed from the drag measuring apparatus between trials such that each subject within a group of five completed one trial before a second trial of any subject was collected. Once all subjects within a group had completed three trials at a given speed, flow speed was adjusted and the protocol was repeated. After all trials were completed for a given condition, subjects changed suits and repeated the protocol using the other suit.
Postswim oxygen uptake collection began immediately after completion of the 183-m swim. A mouthpiece was inserted, and the subject pinched his nose shut as he breathed through the mouthpiece for 45 s (Physio-Dyne Max-1 Metabolic System, Quogue, NY). Expired air was collected in a 3-L mixing chamber and V̇O2, V̇CO2, and RER were measured on a breath-by-breath basis and recorded every 5 s. Each subject’s postswim oxygen uptake was corrected to reflect peak exercise oxygen uptake using a regression equation fitted to each 5-s integrated V̇O2 and extrapolated back to time zero (6). To compare the postswim physiological responses (V̇O2, blood lactate, and rating of perceived exertion) and between testing sessions, linear regression of physiological response versus velocity cubed was used for each subject × suit condition. The correlation coefficients for these regressions ranged from 0.86 to 1.00. The individual regression equations were then used to predict physiological responses at standardized speeds of 1.40 and 1.60 m·s−1. Standardized postswim oxygen uptake, blood lactate, and rating of perceived exertion were then used in ANOVA to compare responses between suits at the two swimming speeds.
Blood samples collected 1-min postswim were analyzed for blood lactate concentrations (12). A 20-μL fingertip blood sample was collected and immediately deproteinized in 2-N perchloric acid. The blood samples were then centrifuged and stored at 4°C until later analysis of lactic acid concentration, using an enzymatic spectrophotometric assay. The blood lactate values were then transformed to common logarithm (13) and standardized to velocities of 1.4 and 1.6 m·s−1 by using linear regression as described above. The correlation coefficients for these trials ranged from 0.85 to 1.00.
Rating of perceived exertion was measured using the Borg scale to indicate their perception of effort for that particular trial (1). The scale was presented to each subject approximately 1.5 min into recovery.
Stroke characteristic measurement.
Average speed over the entire 183-m swim, hereafter referred to as average speed, was computed by dividing total swim distance (183 m) by total swim time. In addition, stroke characteristics were determined for a swimming-only portion (mid-pool) of the 183-m distance in the following manner. Each swim trial was videotaped. Breakout distance, stroke rate (SR), and stroke length (SL) were determined from an analysis of the videotape and averaged for all eight lengths of the pool. Breakout distance was defined as the distance from the wall to the point the head broke the surface of the water. The total number of complete stroke cycles were counted from the first visible hand entry during a given length of the swim to the instant when the head of the swimmer reached a point 4.6 m from the wall. Stroking distance was then computed by determining the location of the swimmer’s head where the stroke count began and the point where the stroke count ceased. SL was then calculated by dividing the stroking distance by the number of arm strokes (left-hand entry to left-hand entry).
Stroking time was measured using a manual stopwatch to measure the time needed to complete the stroking distance. SR was then calculated by dividing the number of arm strokes by time. Mid-pool swimming speed was then determined as the product of SR and SL. Average SL, SR, mid-pool swimming speed, and breakout distance for all eight measurements during each swim was used in statistical analysis to compare responses between suits.
Passive drag measurements.
All measurements of passive drag were performed in a swimming flume in which water was circulated via two BaduStream jets (Speck Pump, 4 HP, 400 GPM, Jacksonville, FL) capable of producing flow velocity ranging from 1.2 to 3.0 m·s−1. Flow velocity was measured to the nearest 0.1 m·s−1 by using an FP6–201 Flow Probe (Omega, Stamford, CT) by placing the flow meter approximately 30 cm from the flume jets. Flow velocities of 2.0 and 2.5 m·s−1 were chosen for the two test conditions to provide race-pace and faster-than-race-pace speeds. For each measurement of passive drag, the subjects were instructed to submerge themselves in a prone position, grasp a nylon handle attached to a tethering cable, and assume a streamline position similar to one used during an underwater glide (Fig. 1). The subject was maintained in a horizontal position approximately 40 cm below the water surface in the flume’s flow. To achieve this position, it was necessary for some subjects to use a floatation device between their legs to prevent them from sinking. Drag force was measured as the tension developed in the cable holding the swimmer in a stationary position. This tension was measured using a SM-250 strain-gauge force transducer (Interface Inc., Scottsdale, AZ) calibrated to the nearest 0.25 N. Transducer output was sampled at a frequency of 50 Hz and low-pass filtered at 1 Hz by using an MP-100 A/D board and Acqknowledge 3.6 software (Biopac. Inc., Goleta, CA). Average drag force was computed for the final 8 s of a 10-s trial.
The buoyant force was measured using a device that allowed the measurement to be made with the swimmer in a horizontal prone body position similar to one used when the swimmers glided under water (Fig. 2) (11). The body was supported using a cranial tether attached to the trunk around the chest and a caudal tether attached to the ankles around the lateral malleoli. The position between the two tethers was set at 1.2 m. This caused the placement of the cranial tether to vary slightly depending on the height of the subject. For all trials, each subject was at full inhalation and was completely submerged to a position 30 cm below the surface of the water in a prone position. Mass was added to the cranial tether (5.5 kg) to maintain the swimmer in a horizontal position underwater. Supporting force measurements were measured using SM-250 strain-gauge force transducers (Interface Inc., Scottsdale, AZ) calibrated to the nearest 0.25 N. Transducer output was sampled at a frequency of 25 Hz and low-pass filtered at 1 Hz. The measured resultant cranial and caudal supporting forces were adjusted by subtracting the forces measured by the force transducers when supporting only the tethers and the stabilizing mass. Buoyant force was then calculated as MATHwhere W was the subject’s weight, and Scranial and Scaudal were the adjusted supporting tether forces, respectively.
Absolute data for average swim speed, postswim blood lactic acid concentration, oxygen uptake, ratings of perceived exertion, SL, SR and breakout distance were analyzed using separate 3 × 2 (intensity × suit) repeated measures ANOVAs. The standardized postswim oxygen uptake, blood lactate concentration, and ratings of perceived exertion were analyzed using separate 2 × 2 (intensity × suit) repeated measures ANOVA. A 3 × 2 × 2 (trial number × suit × speed) repeated measures ANOVA was used to analyze the passive drag measurements. Buoyancy was evaluated using one-way repeated measures ANOVA. When a significant interaction effect was noted (P < 0.05), a Newman-Keuls multiple range test was used to locate the significant mean differences.
Although the subjects were instructed to swim the three 183-m swims with the same effort in both trials, average swim speed for the 183-m freestyle swims was faster (P < 0.01) when wearing the FS suit than the brief-style suit (Table 1). This effect was present at all three of the submaximal speeds and occurred in 9 of the 10 subjects. Commensurate with the faster velocities chosen by the swimmers while wearing the FS suit, the postswim blood lactate concentration was significantly elevated (main effect for suit, P = 0.02) (Table 1). Postswim oxygen uptake was elevated in the FS suit trial, but this difference was significant (interaction effect, P = 0.04) only in the two fastest swims (Table 1).
There was no difference in mid-pool swimming speed (SL × SR) between suit conditions although the interaction between suit condition and swimming effort was significant (P < 0.05). Post hoc analysis revealed that there was a significant difference in mid-pool speed when wearing the FS suit at the lowest effort (P < 0.05) but there was no significant difference in mid-pool swimming speed between suit conditions at the moderate and fast efforts.
Mean SL was 3–5% longer in the FS suit trials than in the brief-style suit trials (main effect for suit, P = 0.03) (Table 1), but there was no difference in SR (Table 1). There also was no difference in breakout distance when wearing the FS suit compared with the brief-style suit (Table 1).
Because the swimmers did not duplicate their 183-m swim velocities in the two swim trials, physiological responses at given submaximal speeds of 1.4 and 1.6 m·s−1 were predicted from linear regression (V̇O2 vs velocity (3), lactate vs velocity (3)). At standardized speeds of 1.4 and 1.6 m·s−1, predicted values for postswim blood lactate concentration (P = 0.91), postswim oxygen uptake (P = 0.91), and rating of perceived exertion (P = 0.70) were not different between suit conditions (Table 2).
The mid-pool measurements of stroke characteristics for each condition were standardized to 1.4 and 1.6 m·s−1 speeds by fitting the data (SR vs average speed, SL vs average speed) to quadratic curves by using a least-squares method. A quadratic curve was used based on the observation that individual data did not appear to be linear and on personal experience with additional data that these relationships were not linear. At the standardized speeds of 1.4 and 1.6 m·s−1, predicted values for SR and SL were not different between suit conditions (Table 2).
Passive drag and buoyancy measures.
Passive drag measured in the flume was not affected at flow rates of either 2.0 m·s−1 (brief: 77.0 ± 4.9 N; FS: 73.9 ± 4.0 N) or at 2.5 m·s−1 (brief: 101.9 ± 3.1 N; FS: 103.2 ± 3.1 N) when wearing the FS suit compared with the brief-style suit. The reliability coefficient for the three replicate measures of passive drag at each flow rate was r = 0.98.
Buoyant forces measured while wearing the brief suit (801 ± 22 N) were slightly but significantly (P < 0.05) higher than the FS suit (794 ± 21 N), but the subjects were significantly (P < 0.05) heavier (785 ± 74N) when wearing the brief-style suit than when wearing the FS suit (779 ± 69). The overall body density (uncorrected for lung volume) when wearing the brief-style suit (0.98 ± 0.01 g·cm−3) was not different (P = 0.33) than when wearing the FS suit (0.98 ± 0.01 g·cm−3).
The major finding of this study was that use of the FS suit did not affect the metabolic cost of swimming at given submaximal speeds. The 1.4 and 1.6 m·s−1 standardized velocities were selected to represent a slower and a faster swim speed. They were also selected because these speeds were close to falling within the range of speeds chosen by the swimmers in their three 183-m swim trials. The standardized physiological results revealed that wearing a FS suit did not reduce the physiologic cost of swimming compared with the brief-style suit at the speeds of 1.4 and 1.6 m·s−1. At standardized speeds of 1.08 and 1.30 m·s−1, Sharp et al. (13) found that after shaving body hair, blood lactate concentrations were reduced 28% and 23% for the two speeds, respectively. In a later shaving study, Sharp and Costill (12) controlled for speed by using pacer lights and found that V̇O2 and blood lactate accumulation post–364-m breaststroke swim were significantly lower by 9% and 20% from pre- to postshave, respectively.
A study involving a torso swimsuit made of a conventional material (80% polyester, 20% polyurethane) found a 4% decrease in V̇O2 and a 16% decrease in blood lactate accumulation after swimming a 366-m freestyle swim at a fixed speed (14). Presumably a positive effect of the torso suit on the drag characteristics of these swimmers was responsible for the improved performance. Although the torso suits may have affected surface drag due to a large portion of the body being covered by swimsuit material, it is possible that any reduction in drag was a reduction in form drag. In this earlier study, older swimmers were used as subjects (mean ± SD age = 30.4 ± 14.2 yr), which may account for a different finding from the present study. It is possible that the skin of older swimmers is more flaccid than in younger swimmers. Consequently, there can be considerable deformation of the skin as the body moves through the water and encounters turbulence. Such changes in the form characteristic of the skin might therefore increase form drag. Wearing a tight-fitting torso suit may therefore reduce surface compliance in older swimmers, resulting in less drag and less metabolic cost of swimming a given speed. Both shaving studies and the torso suit study suggest that a reduction in active drag was responsible for the reduced metabolic cost. In the present study, the finding of no effect of the FS suit on metabolic responses to standardized swimming speeds suggests that the FS does not have a measurable effect on active drag.
The subjects in this study swam at a faster average speed when wearing the FS suit for the three swims. Concomitant with the faster speeds, there was a significant 4–6% increase in postswim V̇O2 and a significant 5–15% increase in postswim blood lactate concentration. If the FS suit had allowed the faster swimming speeds by reducing drag, V̇O2 and blood lactate concentration responses would be lower or unchanged, indicating improved economy. This implied the swimmers chose to swim at a harder intensity during the FS suit trials, despite the ratings of perceived exertion not being significantly different between the two suits. This suggested the swimmers felt like they were exerting the same effort in both conditions but in reality were swimming at a higher physiological intensity with the FS suit.
The faster average speed while wearing the FS suit was due to a longer SL and unchanged SR. This is similar to the findings of the shaving studies where the longer SL led to an improved swim speed that the authors speculated was due to a decrease in active drag (12,13). Although the subjects swam faster with a longer SL while wearing the FS suit, it was done so at an elevated metabolic cost. A possible explanation for the results could be a psychological advantage of wearing an FS suit. It is proposed that subjects expected to swim faster with the FS suit and thus performed accordingly.
Although the FS suits had no effect on physiological responses to submaximal swimming, some of the swimmers reported that they perceived an improved turn performance, particularly during the glide phase after push-off. If the subjects’ perception of improved turn performance during the glide phase was valid, the measure of breakout distance should reflect this. There was, however, no effect of the FS suit on breakout distance. Moreover, there was also no effect of the FS suit on the passive surface drag measurement used in this study. Because the body position of the glide phase after wall push-off and the body position of our passive drag measurement were similar, neither the decrease in passive surface drag nor the improved turn performance can explain the faster swim time. Again, these findings suggest that faster chosen speeds and perception of improved turn performance were more likely a result of subjects’a priori expectations than the effect of the swimsuit design.
Consistent with previous work (17), no differences were found in drag measurements. The use of completely submerged measurements and one speed faster than swimming speeds (i.e., 2.5 m·s−1) permitted a better environment to examine the drag characteristics of both suit conditions. Lyttle et al. (7) found that passive towing when submerged resulted in lower drag values. This presumably was due to the lack of wave drag. Given that an FS suit would be expected to affect primarily surface (frictional) drag, use of a completely submerged condition would provide an environment not influenced by wave drag. Furthermore, drag is proportional to the square of velocity. Given the limitations in measurement of drag, it may not be possible to observe differences in drag between conditions at slower speeds because the drag forces would be relatively small. However, at higher speeds, the differences would be increased quadratically, thus any differences would be easier to discern. The magnitude of passive drag measurements was similar to previously measured (7,17). However, no suit effect was observed.
Previous research has found that the use of neoprene wet suits can alter buoyancy characteristics, which in turn can lead to improved performance through decreased drag and decreased metabolic cost (4,5,12,15,16,19). Results of the present study indicate that the subjects were slightly more buoyant while wearing the brief-style suit than the FS suit. However, the buoyant force was calculated as the difference between the supporting forces (through the load cells) and the body weight. Therefore, the greater buoyant force when wearing the brief-style suit was a result of the subjects being slightly heavier on the testing day when they wore the brief-style suit. When this difference in weight was taken into account through the calculation of overall body density, there was no statistical difference between the two suit conditions, suggesting that the FS suit did not affect the buoyancy characteristics of the swimmers.
In summary, no evidence was found to indicate physiological or biomechanical benefits of wearing an FS suit as compared with a brief-style suit. The possible psychological effects to improve performance cannot be discounted, however, and may account for the observation of a longer SL and faster chosen speeds on the three 183-m freestyle swims in this study. Therefore, it is concluded that the combined effect of increased body surface area covered and design of the suit material had no measurable effect on submaximal responses to swimming. Because no attempt was made to evaluate effects on maximal performance this question remains for future investigations.
We would like to thank Stuart Isaac (Speedo International, Inc.) for donation of the FastSkinTM swimsuits. We would also like to thank Ron Leibold for assistance in building the passive drag and buoyancy equipment. Finally, we would like to thank Josh Thomas and Evan deSzoeke for their assistance in data collection.
1. Borg, G. Perceived exertion: a note on “history” and methods. Med. Sci. Sports Exerc. 5: 90–93, 1973.
2. Capelli, C., P. Zamparo, A. Cigalotto, et al. Bioenergetics and biomechanics of front crawl swimming. J. Appl. Physiol. 78: 674–679, 1995.
3. Chatard, J.C., J. M. Lavoie, and J. R. Lacour. Analysis of determinants of swimming economy
in front crawl. Eur. J. Appl. Physiol. 61: 88–92, 1990.
4. Chatard, J.-C., X. Senegas, M. Selles, P. Dreanot, and A. Geyssant. Wet suit effect: a comparison between competitive swimmers and triathletes. Med. Sci. Sports Exerc. 27: 580–586, 1995.
5. Cordain, L., and R. Kopriva. Wetsuits, body density and swimming performance. Br. J. Sports Med. 25: 31–33, 1991.
6. Costill, D. L., D. Kovaleski, D. Porter, J. Kirwan, R. Fielding, and D. King. Energy expenditure during front crawl swimming: predicting success in middle distance events. Int. J. Sports Med. 6: 266–270, 1985.
7. Lyttle, A. D., B. A. Blanksby, B. C. Elliott, and D. G. Lloyd. The role of drag in the streamlined glide. J. Swimming Res. 13: 15–22, 1999.
8. Mclean, S. P., and R. N. Hinrichs. Gender, buoyancy
and swimming performance. J. Appl. Biomech. 16: 248–263, 2000.
9. Mclean, S. P., and R. N. Hinrichs. Sex differences in the centre of buoyancy
location of competitive swimmers. J. Sport Sci. 16: 373–383, 1998.
10. Mclean, S. P., and R. N. Hinrichs. The influence of arm position and lung volume on the center of buoyancy
of competitive swimmers. Res. Q. Exerc. Sport 71: 182–189, 2000.
11. Parsons, L., and S. J. Day. Do wet suits affect swimming speed? Br. J. Sports Med. 20: 129–131, 1986.
12. Sharp, R. L., and D. L. Costill. Influence of body hair removal on physiological responses during breaststroke swimming. Med. Sci. Sports Exerc. 21: 576–580, 1989.
13. Sharp, R. L., A. C. Hackney, S. M. Cain, and R. J. Ness. The effect of shaving body hair on the physiological cost of freestyle swimming. J. Swimming Res. 4: 9–13, 1988.
14. Starling, R. D., D. L. Costill, T. A. Trappe, A. C. Jozsi, S. W. Trappe, and B. H. Goodpaster. Effect of swimming suit design on the energy demands of swimming. Med. Sci. Sports Exerc. 27: 1086–1089, 1995.
15. Toussaint, H. M., L. Bruinink, R. Coster, et al. Effect of a triathlon wet suit on drag during swimming. Med. Sci. Sports Exerc. 21: 325–328, 1989.
16. Toussaint, H. M., G. de Groot, H. H. C. M. Savelberg, et al. Active drag
related to velocity in male and female swimmers. J. Biomech. 21: 435–438, 1988.
17. Toussaint, H. M., M. Truijens, M. Elzinga, et al. Effect of a Fast-Skin “Body” Suit on drag during front crawl swimming. Sport Biomech. 1: 1–10, 2002.
18. Trappe, T. A., D. L. Pease, S. W. Trappe, J. P. Troup, and E. R. Burke. Physiological responses to swimming while wearing a wet suit. Int. J. Sports Med. 17: 111–114, 1996.
19. Zamparo, P., C. Capelli, B. Termin, D. R. Pendergast, and P. E. di Prampero. Effect of the underwater torque on the energy cost, drag, and efficiency of front crawl swimming. Eur. J. Appl. Physiol. 73: 195–201, 1996.
Keywords:©2003The American College of Sports Medicine
ACTIVE DRAG; BUOYANCY; OXYGEN UPTAKE; SWIMMING ECONOMY