Cumulative pre-cooling methods do not enhance cycling performance in tropical climate

menthol (L-menthol) beverage was a 0.025% natural menthol aroma (% vol: 86.0±1.0%; dos- age: 0.5 g.L-1) (Robertet, Grasse, France; see [20] for procedure) whose concentration has pre- viously been shown to be beneficial in tropical climate during efforts between 20 min and 1h30 [19, 20]. The cooling vest was kept cool for the duration of the warm-up by soaking it every 10 min (T0, T10 and T20) in ice water (3˚C) for the ICE-VEST condition or ice water + menthol for the MEN-VEST condition (4˚C; dosage: 1%); the vest was adjusted to the body with belts to optimize contact surface with the body. The temperature of each beverage and the ice water for the soaked vest was measured with a digital thermometer (YSI 409B, Yellow Springs Instruments, OH, USA). After the warm-up with the pre-cooling strategies, the cyclists covered the greatest distance possible in 20 min on an open-air velodrome track. Assessment The cycling performance on the velodrome track was measured as the total distance traveled (m) over 20 min, and the distance traveled (m) per 2-min block was also recorded. The velo- drome measured 333m and was marked every 5m. We measured the final distance traveled between marks with a tape meter, thus enabling a precise performance report (i.e., near 50 cm). Heart rate (HR) was continuously monitored during the entire experimental session using a portable telemetry unit (Suunto Memory Belt, Suunto, Vantaa, Finland), and averaged by 1-min blocks for the warm-up and 2-min blocks for the cycling performance. Core tempera-

ture (i.e., gastrointestinal temperature-Tgi) was recorded every 5 min during warm-up and at the beginning and end of the cycling performance using ingestible temperature measurement pills (CorTemp, HQ, Inc., Palmetto, FL, USA). To ensure the pill was out of the stomach and thereby avoid Tgi variability due to pill movement or fluid/food consumption, the athletes were invited to ingest these pills 6 hr before all experimental sessions. The cooling rate which corresponds to a temperature difference (ΔTgi) was calculated as the difference in Tgi from pre- to post-warm-up and from pre- to post-cycling performance. Perception was measured every 5 min during warm-up: from T0 to T30 for thermal com- fort (TC) and thermal sensation (TS), and from T5 to T30 for RPE. For the cycling perfor- mance, TC and TS were recorded at the beginning and the end of the cycling performance and RPE at the end. The perception of exertion was evaluated with the 15-point Borg’s RPE scale [36] that ranges from 6 (no exertion) to 20 (maximal exertion). Regarding the perception of the thermal environment, participants were asked to rate their perceived TC and TS on two continuous rating scales ranging from, respectively, 1 (“not uncomfortable”) to 4 (“very uncomfortable”), and from 1 (“slightly cool”) to 7 (“extremely hot”) [37]. Statistical analyses Kolmogorov–Smirnov and Levene tests were used to check the normality of distribution and the homogeneity of variance, respectively. When the assumption of normality of distribution

and/or homogeneity of variance was violated, non-parametric analyses were conducted to determine whether differences existed between periods and conditions. Accordingly, a Fried- man ANOVA and a Wilcoxon test were used for the TS, the TC and the RPE scores collected during the warm-up to assess whether differences existed between T0, T5, T10, T15, T20, T25 and T30 (T5, T10, T15, T20, T25 and T30 for RPE scores) for all conditions (i.e., ICE-VEST, MEN-VEST, NO-VEST) and between the conditions for all 5-min blocks. Similar statistical analyses were conducted for these scales during the cycling performance to assess (1) condi- tion differences before the cycling performance for TS and TC and after the cycling PLOS ONE Tropical cycling performance and cumulative pre-cooling PLOS ONE | https://doi.org/10.1371/journal.pone.0291951 October 12, 2023 4 / 15