Effect of bicycle racing saddle design on transcutaneous penile oxygen pressure

Effect of bicycle racing saddle design on transcutaneous penile oxygen pressure
J.D. Cohen, M.T. Gross

Aim. To dietermine the reliability of monitoring penile transcutaneous oxygen (tpO2) during cycling, and to assess the influence of seat design and cycling position on tpO2.

Methods. Expermintal design: repeated measures analysis of the effects of seat design and riding position on tpO2 values. Participants: 31 male cyclists between the ages of 20 and 50 years participated . Subject inclusion criteria were: averaged≥80 miles of road bicycling per week during the 2 months prior to enrollment in this study; no history of vascular disease, diabetes, or sexual dysfunction; and had an erection within 15 days prior to study. Measures: mean tpO2 values were calculated for seated and standing positions using 3 current bicycle seat designs. Results. Test-retest reliability for seated cycling tpO2 had an ICC (3,1) of 0.88 and mean absolute difference of 7.23 mmHg. No interaction effect occurred between seat design and position. Seat design had no significant effect on tpO2 values. Seated cycling significantly reduced tpO2 levels compared with standing cycling (P<0.05). Mean percent decreases in tpO2 from standing to seated cycling were; Vetta® 76%, Terry® 73%, and Specialized® 62%.

Conclusion. The data suggest that penile tpO2 monitoring is reliable for use during cycling studies. None of the seats exhibited any significant ability to spare penile tpO2. The implications of decreased penile tpO2 over different time intervals on penile physiology remain to be investigated.
KEY WORDS: Blood gas monitoring – Penis – Bicycling – Impotence.

Modern day bicycle seats have been reported to cause various clinical syndromes including saddle sores, subcutaneous nodules, pudendal neuropathies, impotence, testicular torsion, and prostatitis.¹ Clinical experience suggests that many individuals discontinue cycling or do not increase bicycle use because of various discomforts related to the bicycle saddle.

The primary focus of recent ergonomic seat designs has been an attempt to relieve pressure through the perineal region by removing large amounts of material from the center portion of the seat. The companies that manufacture these seats claim that the removal of the center portion of material results in lower contact pressures imposed on the perineum. This approach tends to decrease the total seat/buttocks contact area and, therefore, increases the contact pressure through the ischeal tuberosities. This increased ischeal contact pressure has been associated with seat-related injuries such as saddle sores and skin callosities.²

Regardless of the particular seat design, many bicycle seat-related traumas are caused, in part, by the rider’s position on the seat. Cyclists often attempt to minimize air drag by lowering the torso towards a horizontal position and tiliting the pelvis anteriorly towards a horizontal position and tilting the pelvis anteriorly to maintain lumbar flexion within normal limits. Many bicycles have been intentionally designed with drop style handlebars and frame geometry that automatically position the rider in this aerodynamic posture. The anterior tilting of the pelvis causes the cyclist to unload the ischeal tuberosities and transfer pressure towards the nose of the saddle. This will tend to increase pressure through the medial border of the pubic rami and compress the perineal vascular and neural structures. Several authors have suggested this compromise of neurovascular in this anatomic region may explain some of the perineal clinical syndromes experienced by cyclists.
Nayal et al. have used transcutaneous oxygen monitors to record transcutaneous penile oxygen (tpO2) levels for subjects during cycling.³ Their results suggest that compression of the internal pudendal artery, which supplies the dorsal penile artery, can cause significant decreases in transcutaneous penile oxygen levels. Nayal et al. observed a 68% decrease in penile oxygen levels after subjects had cycled for 3 mins, compared with oxygen levels observed when subjects stood prior to cycling. Sommer et al. have reported very similar results. ⁴’⁵ Direct measurement of neural compression during cycling has not been studied due to the invasive nature of current measurement techniques. The results of studies by Nayal et al.³ and Sommer et al., ⁴’⁵ however, suggest that use of some bicycle seat designs compromise vascular structures in Alcock’s canal, an anatomic region which also houses neural structures that supply the penile musculature. These studies also suggest that transcutaneous penile oxygen instrumentation may be a useful method for quantifying the ability of bicycle seat designs to relieve pressure from sensitive neurovascular bundles within the perineum. The reliability of this instrumentation, however, has not been reported. The purpose of this study, therefore, is to determine the reliability of monitoring transcutaneous penile oxygen levels during dynamic bicycling conditions, and to determine if current bicycle seat designs significantly influence transcutaneous penile oxygen levels. The specific research questions addressed were:
1) What is the reliability of transcutaneous penile oxygen measurements?
2) Does seat design have a significant effect on penile transcutaneous oxygen pressures?
3) Does cycling position (standing versus sitting) have a significant effect on penile transcutaneous oxygen pressures?
4) Do cycling position (standing versus sitting) and seat design interact significantly to affect penile transcutaneous oxygen pressures?

Materials and methods

Subjects
Thiry-one male subjects betweent he ages of 20 and 50 years from local cycling clubs partidicpated. Subject inclusion criteria were:
-Averaged at least of 80 miles of road bicycling per week during the 2 months prior to enrollment in this study.
-No history of vascular disease, diabetes, or sexual dysfunction.
-Had an erection within 15 days prior to study.

Our inclusion criteria are similar to those used by Nayal et al.³ and Sommer et al. ⁴’⁵ and were meant to exclude any subjects that may have had preexisting penile vascular dysfunction that might confound the measurements.³⁻⁵ A minimum number of weekly miles was established to define the subjects’ level of fitness and ensure that all subjects would be able to complete the 40 minute testing period without limitations secondary to aerobic deconditioning. Although studies performed by Kerstein et al.⁶ and Nayal et al.³ focused primarily on subjects between 20 and 35 years of age, penile dysfunction has been reported in cyclists as old as 54. ³’⁶’⁷ Subjects ranging in age from 20 to 50 years were included in this study to reflect the majority of participants who experience penile related dysfunctions secondary to long distance cycling. Each subject signed a statement of informed consent, and the study was approved by the Committee for the Protection of the Rights of Human Subjects at the University of North Carolina at Chapel Hill.

Instrumentation
The Novametrix® TCO2M Monitor Model 860 (Novametrix®, Connecticut) was used to monitor transcutaneous oxygen pressures. This machine is able to measure tha mount of oxygen that diffuses through the skin. The amount of oxygen that diffuses through the skin at a given temperature is significantly correlated (r=0.80, P<0.001) with the trend in arterial partial oxygen levels.⁸ The Novametrix® TCO2M Monitor Model 860 has 2 independent methods to monitor skin temperature and shut down monitor operation if a temperature fault occurs. Temperature faults can be detected by both a hardware and software heater control. If the skin surface temperature beneath the sensor face exceeds 45.5° Celsius, the monitor will signal an alert light and shut down monitor operation. Monitor function can only be resumed by restarting the monitor. An alert light will also be activated if sensor temperature varies by ±1° Celsius.⁹

All bicycles were affixed to a Performance Indoor Roller Trainer with front fork stabilizer (Performance Inc., NC, USA). Traning rollers were used to approximate flat road riding requirements. Fork stabilizers were used to minimize the confounding effects of task novelty for those riders who have not previously ridden on rollers.
A Polar® A3 heart rate monitor (Polar®, NY, USA) with chest strap sensor was used to monitor each subject’s heart rate during testing to assure that all subjects remained at a consistent, aerobic work rate. All seats were preattached to seat posts to expedite seat transfers.

Three male road bicycle seats were chosen to represent many current seat designs that are purported to be effective for relieving perineal pressure for male cyclists. The three seats tested include a standard road seat used as the control (Vetta Lite®, Figure 2), the Terry Dragonfly Ti® seat (Figure 3), and the Specialized Body Geometry Comp® seat (Figure 4).

The Vetta Lite® seat (Figure 2) has minimal padding and a seamless surface. No other unique surfact features are present in this seat. This seat also does not exhibit any features specifically designed to reduce perineal compression and was chosen, therefore, as the control design.

The Terry Dragonfly Ti® seat (Figure 3) has a complete hole in the center of the seat in the form of an ellipse. The long axis of the cutout extends from the rear to the front of the seat. The borders of the cutout and the remainder of the seat surface are padded with dense, thin foam. The padding in the Terry Dragonfly Ti® seat is thinner and firmer than the padding in the Specialized Body Geometry Comp® seat (Figure 4). The surface of the Terry® seat (Figure 3) exhibits a seam extending from the front portion of the cutout to the tip of the front end of the seat. A recession also bisects the posterior half of the seat into left and right portions.
The Specialized Body Geometry Comp® seat (Figure 4) features a central depression extending from the back to the front of the seat. The rear portion of the seat is split to form a V shaped notch with the apex of the V pointing anteriorly. The seat is moderately padded with a foam material. The surface of the seat is seamless.

Testing procedures
Each subject’s bicycle was affixed to stationary training rollers. The cyclists rode their own road-racing style bicycles fitted with their own pedal-shoe attachment systems. The front wheels were removed and the bicycles were attached via the front fork stabilizing post. The subject’s current seat heights were measured. Seat height was defined as the distance between the pedal axle and the apex of the seat as measured along the longitudinal axis of the seat tube when the crank was in the down postion in alignment with the seat tube. Descriptive data were then collected for each subject including age, height, weight, years cycling, and average number of miles ridden per week during the two months preceding enrollment in the study. Leg length was calculated with each subject standing in their cycling shoes, from the floor to the approximate center of the greater trochanter. The current seats were preaffixed to seat posts. All seat posts were marked with tape so that the allowable amount of seat post insertion into the seat tube assured a seat height equal to 100% of trochanteric height to expedite seat transfers during testing. This seat height has been reported to minimize VO2 requirements during cycling.¹⁰ Anterior-posterior seat position was also adjusted until the anterior aspect of the knee was directly above the pedal axle when the crank was in the forward horizontal position.²

Before testing each subject, the transcutaneous oxygen monitor was calibrated using the one point calibration procedure detailed in the Novametrix® TCO2M Model 860 user’s manual. The one point calibration procedure requires that a drop of solution be placed on the sensor face to create an anoxic environment and set the lower limit for the monitor. The sensor was then allowed to stabilize while exposed to room air for 15 min. The sensor temperature was set to 44° Celsius to create a local hyperemic reaction that improves transcutaneous oxygen diffusion across the skin. This temperature is also slightly above normal body temperature so that local fluctuations of surface temperate in nearby regions did not affect the sensor temperature.

The subject was then instructed in proper application of the sensor of the glans penis. A picture of the penile anatomy was used to describe the procedure with particular attention paid to the importance of creating a secure seal between the adhesive ring and the glans penis (Figure 1). The subject then cleaned the glans with an alcohol guaze pad to remove any surface residues. The Novadisk® sensor assembly with adhesive ring was fully prepared by the principal investigator prior to allowing the subject to apply the sensor to the glans penis. Average transcutaneous penile oxygen level readings (less than 75 torr) are significantly lower than room air oxygen levels (greater than 120 torr). If the seal around the sensor face was not complete, tpO2 reading would read above 120 torr. The subject ¹⁻was then instructed to press firmly on all portions of the adhesive ring to complete the seal until tpO2 readings were withing normal limits below 75 torr. Sommer et al. used similar methods for penile tpO2 monitoring and reported no patient complaints while maintaining the recommended 44° Celsius temperature for the tpO2 sensor.⁴

The Novadisk® assembly was replaced whenever: a) there was more than 50% air in the annulus, b) the membrane had been damaged or the Novadisk® had loosened from the sensor, or c) after 7 days of use or extended periods of non-use. Novadisk® assembly replacement was completed following the detailed instructions in the Novametrix® users’ manual.

All subjects wore their own padded cycling shorts, as is the usual custom during recreational cycling. Each subject was instructed to find a comfortable position as soon as possible on each seat and maintain that position for the duration of the 5-minute cycling interval. After 2.5 min of cycling, the subject was asked if he had achieved a comfortable position and was then reminded to hold that position for the remainder of the testing interval. Each subject was also instructed to remain with both hands positioned in the lowest portion of the handlebars (drops) throughout the testing session.

Testing sessions began by having subjects stand for 10 min next to the bicycle to establish a baseling reading for the tpO2 levels. This interval of time was sufficient to produce stable tpO2 readings during pilot testing. The subject was then instructed to begin pedaling the bicycle in a seated position using the first test seat, until hios heart rate had reached and remained at 65% of heart rate reserve for 5 min. To maintain the assumption that any decrease in tpO2 during cycling is due to arterial compression, a constant physiological work rate was maintained. By maintaining a constant physiological work rate, changes due to shunting of blood and shifts in metabolic activity were minimized. For aerobically conditioned males between 20 and 50 years old, preventing subjects from exceeding 65% heart rate reserve could reasonably prevent a physiological shift towards anaerobic metabolism during the testing, and therefore produce a reasonably constant metabolic work rate. Percentage of heart rate reserve was calculated using the Karvonen Formula as follows:¹¹⁻¹³
Heart rate reserve= [(0.65x(200-70))+70]=154.5 bpm

Each subject was then instructed to retain a heart rate within 5 beats of 65% of his heart rate reserve throughout the testing. After this seated warm-up period, all subjects completed the following intervals. Subjects dismounted briefly for seat changes. In pilot testing, the time period for seat changes averaged approximately 45 s to 1 min. Seat order was randomized among subjects using random card drawing without replacement:
1) 10 min subject – monitor calibration in relaxed standing;
2)warm-up @ 65% HRR for 5 min with seat #1;
3) 3 min of standing cycling;
4) 5 min seated cycling with seat #1;
5) 3 min of standing cycling;
6) 5 min seated cycling with seat #2;
7) 3 min of standing cycling;
8) 5 min seated cycling with seat #3;
9) 3 min standing cycling;
10) 5 min seated cycling with seat #1;
11) 3 min standing cycling.
Total cycling time: warm-up + 35 min.
The length of 5-minute intervals per testing condition was based on pilot tests and the finding my Sommer et al. that penile tpO2 displayed the largest reduction within the first 3 min of seated cycling.¹⁴ This also ensured that the total amount of time in any one position would not be sufficient to cause the subject any lasting discomfort or injury.¹⁵ After the third seat had been tested, the subject continued the testing intervals with seat #1 to determine the repeatability of transcutaneous penile oxygen values for the same seat.
Means and standard deviations of tpO2 measurements were calculated for the final minute and 4 s of each testing interval. The tpO2 monitor automatically samples tpO2 readings at a frequency of one sample every 8 s. Due to this 8-second interval between sampling of tpO¬2 levels, 1 min and 4 s intervals of data were used to approximate the final minute of testing for each interval. The final minute of each interval was chosen to allow sufficient time for the oxygen level to adjust to the new testing condition, as based on the findings Sommer et al. ⁴’⁵
Subjects were given the opportunity to bicycle on their own saddle while remaining connected to the tpO¬2 monitor after they had completed all study procedures. This was offered to subjects as an inducement for participation in the study so that they could observe the relative amount of perineal compression caused by their saddle compared to other saddles in the study. Participants were instructed that these readings may not be as valid as previous readings taken in the context of formal scientific methodology.

Data analysis
Data from one subject were eliminated from the final data set due to poor sensor adherence during testing. Poor sensor adherence was noted by gradually rising tpO2 levels that exceeded normal penile tpO¬2 values and confirmation by subject observation of an opening in the seal surrounding the sensor. No other sensor adherence problems were encountered with the 30 participants included in this study. Data for the remaining 30 subjects were analyzed using SAS Statistical Software (Cary, NC, USA). Means and standard deviations were calculated for age, height, weight, years cycling, and average miles per week ridden during the 2 months preceding enrollment in the study. Intraclass correlation coefficients and mean absolute difference values were calculated to assess reliability of repeat testing. Significance of differences was determined at the P≤0.05 level.
Mean tpO2 values were analyzed using a repeated measures 2-way analysis of variance test to determine if any significant differences existed among seats or cycling positions (seated versus standing). Data were also evaluated for interaction effects between seats and cycling position. Any significant interaction effect was followed by a test of simple main effects. Any multiple comparison of means was accomplished using Tukey’s HSD test. Percent decrease in tpO2 was also calculated for each seat using the following equation:
(tpO2 value* Standing cycling-
tpO2 value for seat ** Seated cycling)
--------------------------------------x100%
tpO2 value* Standing cycling
*) Standing cycling tpO2 value for final minute of testing interval immediately prior to seated cycling.
**) Seated cycling tpO2 value for final minute of testing interval.

Results
Descriptive statistics for the 30 subjects appear in Table I. The 30 male subjects in this study had a mean age of 35.5 years, were fairly experienced cyclists, and averaged 194.2 km/week of cycling during the 2 months preceding enrollment in the study.

The reliability of transcutaneious oxygen measurement during cycling in a standing position was calculated by comparing mean tpO2 values for standing cycling during the first interval with values for the final interval of the testing procedure. OFr this compareison, an intraclass correlation coefficient [JCC (3,1)] of 0.88 with a mean absolute difference of 7.23±0.94 mmHg was calculated. The reliability of transcutaneous oxygen measurement during seated cycling was calculated by comparing mean tpO2 values for seated cycling on the first seat tested to those values recorded on the same seat repeated at the end of the testing procedure. For this comparison, an intraclass correlation coefficient [ICC (3,1)] of 0.76 with a mean absolute difference of 5.1±1.8 mmHg was calculated.

Seat design and cycling position did not interact significantly to affect tpO¬2 levels (F=2.2; df=2,58). Seat design also did not have any significant effect on penile transcutaneous oxygen measurements (F=2; df=2,58). Mean (±SD) tpO2 values for seated cycling on the Vetta, Terry®, and Specialized® xeats were:9.1±13.8 mmHg, 10.2±13.6 mmHg, and 14.7±18.7 mmHg, respectively (Figure 5). Cycling position, however, had a significant effect on transcutaneous penile oxygen levels (F=95.4; df=1,29). Seated tpO2 values (11.4±15.5 mmHg) were significantly less than standing values (38±16.9 mmHg).

Mean (±SD) tpO2 values for standing cycling immediately prior to sitting on the Vetta, Terry®, and Specialized® seats were: 37.8±17.4 mmHg, 37.4±17.3 mmHg, and 38.7±16.4 mmHg, respectively (Figure 5). Mean (±SD) percent decreases in tpO2 from standing to seated cycling positions on the Vetta, Terry®, and Specialized® seats were 78.2% (±26.7), 75.3% (±27.5), 68.1% (±32.5), respectively (Figure 6).

Discussion and conclusions
All subjects reported averaging at least 80 miles ofbicycling per week during the 2 months prior to enrollment in this study and no history of vascular disease, diabetes, or sexual dysfunction. Subjects also stated that they had an erection within 15 days prior to study, signifying no significant penile vascular dysfunction. Participants represented a wide range of ages from the minimum of 20 years to the maximum of 50 years. Years of cycling experience also exhibited a large range from 1 year to 30 years. Riders tended to display a more narrow range of mass due to the degree of aerobic fitness needed to meet the 80 miles per week of road riding inclusion criterion.

Due to relatively short testing sessions, true resting heart rates could not be calculated for the subjects. True resting heart rate is measured immediately after an individual wakes after sleep. Relaxed heart rates were calculated after 10 min of quiet standing. These relaxed heart rate values were used in the Karvonen equation to calculate the target heart rate during testing. During pilot testing, heart rates after 10 min of quiet standing were approximately 20 beats per minute (bpm) higher than resting heart rates taken immediately after waking from sleep. When calculating 65% of heart rate reserve this would result in a mean target heart rate of 7 bpm greater when using a heart rate after 10 min of relaxed standing as compared to using a true resting heart rate. All participants were instructed to remain within ±5 bpm of their calculated target heart rate. Participants, therefore, remained as little as 2 bpm away from 65% of their heart rate reserve calculated with their true resting heart rate.

Subject fatigue was subjectively monitored throughout testing by asking each subject if he was working at a work rate that could be comfortably sustained throughout the remainder of the testing. All subjects reported working at a comfortable, sustainable work rate throughout testing. All subjects were also able to sustain conversation throughout testing without shortness of breath. Finally, seat order was randomized with all possible scat orders represented to negate any possible effects of fatigue. The use of a resting heart rate after 10 min of quite standing, therefore, may not have significantly altered the subjects' metabolic work rates from target heart.

Previous studies performed by Sommer et al. compared transcutaneous penile oxygen levels in relaxed standing to levels recorded during seated cycling. During preliminary testing for this study, results suggested that a decrease in penile oxygen perfusion occurs during the initiation of aerobic activity, possibly attributable to the shunting of blood to the active musculature and away from the penis. For this reason, seated cycling tp02
values were compared to tp02 values for cycling out of the seat.

Reliability of transcutaneous oxygen measurement during dynamic cycling was evaluated both for repeated intervals while cycling in a standing position and repeated intervals while cycling in a seated position. The intraclass correlation coefficient [ICC(3,1)] of 0.88 calculated for repeated standing cycling intervals with a mean absolute difference of 7.23 mmHg supports the reliable use of the tp02 monitor for evaluating penile tp02 levels during dynamic cycling in a standing position. A lower ICC (3,1) value (0.76) with a mean absolute difference of 5.1 mmHg was calculated for repeated intervals of cycling in a seated position.

Repeated cycling intervals in a seated position exhibited a lower ICC (3,1) value than repeated cycling in a standing position. The mean absolute difference between repeated measurements for seated cycling, however, was less than the mean absolute difference between repeated measurements for standing cycling. An inspection of the variability of measurements for all reliability data for standing and seated cycling indifore, could produce a lesser magnitude ICC (3,1) test statistic even though the mean absolute difference for seated reliability testing was less than the mean absolute difference value for standing cycling. ICC(3,1) values for both standing and seated cycling, nevertheless, suggest acceptable reliability for tp02 measurements. The ICC (3,1) value for seated cycling intervals was also reduced by the unusually different response to the same seat during repeated intervals by one subject (time 1: 55 mmHg; time 2: 4 mmHg). Statistical measures of reliability have not been reported in any other studies of penile tp02 for dynamic cycling.

Cycling position had a significant effect on penile tp02 values (Figure 5). Large mean percent decreases in penile tp02 indicate that significant pudendal vascular compression occurred for cycling in a seated position on the 3 test seats. All subjects were instructed to remain cycling with their ischeal tuberosities over the widest portion of the seat. Anterior rotation of the pelvis during cycling, however, with hands positioned on the drops of the handlebar may have promoted increased weight bearing through the medial borders of the pubic rami, which would promote pudendal artery compression.
The dramatic decreases in tp02 values during seated cycling may be attributable to the current design of male ergonomic seats. As the male cyclist anteriorly rotates the pelvis to facilitate riding with his hands in the lower portion of the handlebars, the ischeal tuberosities may no longer be the primary weight bearing structures of the pelvis. Contact pressure may be transferred anteriorly onto the pubic rani and pubic symphysis. This may produce increased pressure through the Alcock's canal arid promote compression of the pudendal arteries, veins, and nerves. The removal of the central portion of the seat, as seen in both the Specialized® (Figure 4) and Terry® (Figure 3) designs, decreases the overall contact area of the pelvis on the seat. This type of scat design, therefore, may increase the contact pressure thnaugh the lateral aspects of the seat, a region of the seat that then makes contact with Alcock's canal. In addition, the narrow design of current racing style seats increases perineal loading, thereby increasing pressure throughout the perineal structures.

Penile tp02 measurements were not significantly different among the 3 seat designs tested. Sommer et al.14
also has reported that the Specialized Body Geometry Comp® seat did not differ from two other traditional seats with noses in terms of penile tp02 values. Agreement between our study and previous studies suggests that the current modifications of seat design tested in these studies may be inadequate at providing effective pressure relief to the pudendal blood vessels.

No significant differences were present among tp02 values for standing cycling intervals (Figure 5). Vascular supply to the penis, therefore, may return to uncompromised values after short bouts of seated cycling. Cyclists, therefore, are encouraged to bicycle frequently in standing during long periods of seated cycling to improve penile vascular circulation and minimize the duration of large drops in penile tp02 values.

The primary cause of cycling related erectile dysfunction is a source of continued debate. No current literature is available to determine if cycling related erectile dysfunction is the result of vascular compromise, neural compromise, or both. This study has demonstrated that a significant decrease in penile oxygen level occurs as a direct result of the mechanical compression of the pudendal arteries while cycling in a seated position on any of the 3 bicycle seats tested. Due to the close proximity of the pudendal artery to the pudendal nerve within Alcock's canal, mechanical compression of a magnitude large enough to produce significant compression of the pudendal artery may also produce significant compression of the adjacent pudendal nerve. The use of tp02 monitoring during dynamic seated cycling, therefore, is a reasonable method for determining the relative magnitude of compression occurring to the pudendal artery and the adjacent pudendal nerve during dynamic cycling conditions.

Further study is needed to determine the extent to which pelvic positioning impacts transcutaneous penile oxygen levels. Although this study has shown that a reasonably constant seated position compromises transcutaneous penile oxygen levels, it is not clear if cycling in an unrestricted environment would significantly improve average tp02 values due to changes in rider positioning. Vascular studies would also be helpful to determine how various durations of pressure on arteries affect the vessels' ability to remain patent after the pressure is removed. Although previous authors have reported a connection between chronic decreased penile oxygen perfusion and risk of impotence, the impact of decreased penile oxygen perfusion in the context of intermittent recreational cycling has not been clearly defined.(16) Bicycle seat manufacturers might consider using measures
of penile oxygen perfusion, such as the commercially available device used in this study, to develop and refine future male ergonomic bicycle seat designs.
Bicycle seat traumas do occur in both men and women. Injuries such as vulvar trauma and superficial skin ulcerations have been reported for female cyclists.(1) In recent years, many bicycle companies have also produced bicycle seat product lines designed to decrease perineal and vulvar compression for female cyclists. Anatomical differences in the female have made any current form of vascular testing for dynamic cycling impractical. Future scientific methods may enable ergonomic seat improvements for women cyclists.

This study was limited by the inability to determine the exact pelvic positioning and contact pressures that occur during dynamic cycling. This information would be useful in deducing the appropriate regions through which pressure would or would not be injurious. Fiscal limitations prohibited the use of sophisticated pressure sensing devices that could be used to map dynamic contact pressures. In addition to cost, many other factors threaten the reliability and validity of using pressure mapping devices during dynamic cycling. The use of this technology to map pressure distribution patterns for dynamic bicycle seat testing may be limited by shearing forces that may confound the data, and the inability to locate anatomical landmarks reliably on pressure data output. Time constraints also limited the number of seat designs tested in this study. The analysis methods used in this study may be useful to manufacturers as they test seat designs prior to mass production.

Clinical implications

The results of this study indicate that transcutaneous penile oxygen monitoring is a reliable method for measuring transcutaneous penile oxygen levels during dynamic cycling. The results also indicate that a significant decrease in penile oxygen perfusion occurs when cycling in a seated position as compared to a standing cycling position. The 3 racing seat designs tested were not significantly different in the degree to which they compromised penile oxygen perfusion during seated cycling. Periods of cycling in a standing position must occur for at least 3 min to produce relatively stable increases in penile tp02 values. Periods of seated cycling in excess of 5 min should be accompanied by frequent intervals of pressure relief (e.g., standing) to minimize the duration of pudendal artery compression and the resultant decreases in penile tp02 values that occur. Although recent studies have presented evidence for possible links between decreased penile tp02 values and impotence, no studies currently exist that support a direct correlation between low penile oxygen levels during cycling and an increased incidence of erectile dysfunction.(17) To minimize the potential risks associated with decreased penile oxygen levels, current male ergonomic seat designs should be modified to produce decreased compression of the pudendal arteries through the perineum and Alcock's canal.

Acknowledgements.—This work was gnerously supported by lerry® Bicycles. Trek'"' Bicycles, andSpecialii.edw Bicycles through the donation of bicycle scats used in the study. We also thank Performance Bike Inc. for providing other bicycle -related equipment. Ilte support of Novametrie was also greatly appreciated for their cooperation in supplying the transcutaneous oxygen monitor used in this study. We are also Very grateful for the time and effort given by all of the dedicated cyclists who participated in the study.

References
1. Weiss BD. Clinical syndromes associated with bicycle scats. Clin Sports Med 1994;13:175-86.
2. MeIlion MB. Common cycling injuries. Management and prevention. Sports Med 1991;11:52-70.
3. Nayal W, Schwarzer U, Klotz T, Heidenreich A. Engelmann U. Transcutaneous penile oxygen pressure during bicycling. BRI in 1999;83:623-5.
4. Sommer F,Schwarter U, Graf C. Klotz T, Engelmnnn U. [Changes in penile blood flow during cycling: how does one prevent a decreased perfusion?]. Dtsch Med Wochenschr 2001;126:939-43.
5. Sommer F. Konig D. Graft C. Schwarter U. Bertram C. Klotz T et al. Impotence and genital numbness in cyclists. lot J Sports Med 7001;22:410-3.
6. Kerstein MD, Gould SA, French -Sherry E. Pirnum C. Perineal trauma and vasculop,enic impotence. 1 Urol 1982;127:57.
7. Andersen KV, Bovim G. Impotence and nerve entrapment in long distance amateur cyclists. Acts Neurol Scand 1997;95:233-40.
8. McDowell JW, Thiede WH. Usefulness of the transcutaneous Po2 monitor during exercise testing in adults. Chest 1980;78:853-5.
9. Novarnetrix. Novametrix TCO2M Monitor Model 860 User's Manual, 2000; 2000.
10. Nonleen-Snyder KS. The effect of bicycle seat height variation upon oxygen consumption and lower limb kinematics. Med Sci Sports 1977:9:113-7
11. Miller WC, Wallace JP, Eggert KF'.. Predicting max HR and the HR. V07relationship for exercise prescription in obesity. Med Sci Sports Exert 1993;25;1077-8 l
12. Omiya K. Itnh H, ()sada N. Kato M. Koilz A, Sagara Ketal. Impaired heart rate response during incremental exercise in patients with acute myocardial infarction and after coronary artery bypass grafting: evaluation of coefficients with Karvonen's formula. Jpn Circ J 2000;64:851-5.
13. Goldberg L, flint DL, Kuehl KS. Assessment of exercise intensity formulas by use of ventilatory threshold. Chest 1988;94:95-8
14. Sommer F. Schwartz/.U. Klotz T, °Isms HP, Haupt G. Engelmann U Erectile dysfunction in cyclists. Is there any difference in penile blood flow during cycling in an upright versus a reclining position? Eur Urol 2001;39:720-3.
15. Schwarzer U. Impotence rate in long distance cyclists. J Urol 1999;161.686A.
16. Maurice G. Wang X. Lehalle B. Stoltz. JF. Modeling of elastic deformation and vascular resistance of arterial and venous vasa vasorum. J Mal Vase 1998;23:282-8.
17. Nehra A. Goldstein I, Pabby A, Nugent M, Huang Yii, de as Morenas A et al. Mechanisms of venous leakage: a prospective clinicopathological correlation of corporeal function and structure. J Urol 1996;156:1320-9.