During training and competition, competitive cyclists tend to select cadences that are higher (90 rpm) than cadences (50-80 rpm) that optimize efficiency, economy and ratings of perceived exertion (Ferguson et al., 2001; Foss and Hallen, 2005; Gaesser and Brooks, 1975; Hansen et al., 2002; Lucia et al, 2001b; 2004; Marsh and Martin, 1993; 1997; Marsh et al., 2000; Moseley et al., 2004; Nickleberry and Brooks, 1996; Nielsen et al., 2004). Moreover, during submaximal cycling, trained cyclists tend to select cadences that are higher than those that are energetically optimal, resulting in an excess energy expenditure of approximately 5% (Hansen et al., 2006). While many factors can affect which cycling cadence is most energetically optimal (including exercise duration and intensity, fitness level and muscle fiber recruitment patterns) (Lucia et al., 2001a; 2001b; Marsh and Martin, 1997; Vercruyssen et al., 2001, 2010; Whitty et al., 2009), it has been hypothesized that freely-chosen high cadences represent an innate compromise between stresses on the cardiovascular system and those on contracting skeletal muscle (Lucia et al., 2004). However, in addition to greater energy expenditure, higher cadences also increase oxygen demand, which requires greater oxygen delivery and thus greater cardiac output (Moore et al., 2008).
These potential negative impacts of choosing a higher cadence may be even more evident during long competitive road races, where ~70% of the race consists of sub-maximal cycling at 60-70% of maximal oxygen consumption (V[O.sub.2]max) (Broker, 2003; Lucia et al, 1999). Although previous studies have shown that a reduction in freely chosen cadence towards a more energetically optimal cadence (~80 rpm) occurs during prolonged periods of cycling in the laboratory (i.e., 2 h) (Argentin et al., 2006; Lepers et al., 2000; Vercruyssen et al., 2001), little is known about effects of cadence on performance in these conditions. Interestingly, most studies that have examined effects of cadence on performance utilized constant-power tests that do not accurately simulate an actual cycling race where there are multiple changes in intensity due to changes in terrain, environmental factors and race strategies. Thus, data from these studies may not translate directly to competitive cycling road races. Based on these observations, we tested the hypothesis that the use of high cadences during a prolonged cycling protocol with varying intensities (a protocol that more closely simulates competition) would decrease performance compared to cycling at a lower, more energetically optimal, cadence.
Eight healthy, competitive male road cyclists (35 [+ or -] 2 yrs) were studied. Participants had been competing in their sport for two yrs and trained 10-15 h x [week.sup.-1] (mean = 13 [+ or -] 1 h). Research was conducted in accordance with the Helsinki Declaration, and was approved by the University of California Institutional Review Board. Informed written consent was obtained from each participant.
Participants were asked to rest and follow the same hydration and eating patterns 24 h prior to all testing. Similar fluid and energy intake (600 ml x [h.sup.-1] of fluids and 40 g x [h.sup.-1] of carbohydrate) were maintained during all tests (ACSM Position Stand, 2009). Experiments were conducted on 3 separate days, at least 48 h apart. During visit 1, exercise capacity (V[O.sub.2]max) was assessed. During visits 2 and 3, participants cycled for three h at either 80 or 100 rpm with varying intensities at 50, 65 and 80% of V[O.sub.2]max, simulating a cycling road race (Lucia et al., 1999; Palmer et al., 1994).
Visit 1: Participants arrived at the laboratory 3 h after eating. Body mass, height and body compositions were measured (Jackson and Pollock, 1978). After 15 min of warm-up, each subject completed, at his preferred cadence (101 [+ or -] 11 rpm), a graded exercise test to exhaustion on his own bike, which was fitted with a power measuring device (Graber Products, Madison, WI, USA) and mounted to a compu-trainer (Racermate Inc., Seattle, WA, USA), which allows for a fixed power output to be maintained despite changes in cadence by adjusting resistance levels. The exercise test began at an intensity of 100 watts (W) and increased by 40 W every 3 min. During exercise, measurements of heart rate (HR) (Polar Heart Rate Monitor, Woodbury, NY, USA) and oxygen consumption, V[O.sub.2] ventilation (VE) and carbon dioxide production (VC[O.sub.2]) (ParvoMedics Metabolic Cart, Sandy, UT, USA) were made continuously. The metabolic cart pneumotachometer and gas analyzers were calibrated pretest and verified post-test using a 3 L syringe (Hans Rudolf Inc. Kansas City, MO, USA) at varying flow rates and a calibration gas mixture of 4% C[O.sub.2] and 16% [O.sub.2].
Rating of perceived exertion (RPE) was assessed using a 10 point scale (Noble et al, 1983) and assessed every 3 min. Ventilatory threshold (Vt) was determined using the criteria of an increase in the ventilatory equivalent for oxygen (VE V[O.sub.2.sup.-1]) without an increase in the ventilatory equivalent for carbon dioxide (VEVC[O.sub.2.sup.-1]) (Lucia et al., 2001b).
Visit 2: After a recovery period of at least 48 h, and 3 h after eating, participants performed the first experimental protocol, which began with a 10-15 min warm-up. A 190 min continuous, varied intensity cycling test was then conducted at a constant cadence of either 80 or 100 rpm (randomized) at power outputs corresponding to 50%, 65% and 80% of each subject's predetermined V[O.sub.2]max from the maximal exercise test performed during visit one. These power outputs were maintained for both varied intensity trials and were not adjusted for changes in V[O.sub.2]. Initially, participants exercised at a power that elicited 65% of V[O.sub.2]max for 30 min. Subsequently, 4 identical, 40 min varied intensity cycling bouts were performed consisting of 12 min at 80% of V[O.sub.2]max, 8 min at 65% of V[O.sub.2]max, 10 min at 50% of V[O.sub.2]max and 10 min at 65% of V[O.sub.2]max. Respiratory gases ([O.sub.2], C[O.sub.2]) were collected during the 12 min at 80% of V[O.sub.2]max and 8 min at 65% of V[O.sub.2]max for each of the 4 cycling bouts, and averaged across the whole trial. From an earlobe blood sample, blood lactate (Lactate Pro, Arkray, Inc, Kyoto, Japan) and glucose (glucose analyzer, Accu-Check, Roche, Mannheim, Germany) were determined during each 40 min cycling bout to correspond with the 65 and 80% V[O.sub.2]max intensities. Metabolic power...