During the completion of a CPET, the symptoms of dyspnea, leg fatigue, and rate of perceived exertion (RPE) were reported in all subjects. The complementary use of the NIRS devices did not add any discomfort to the subjects' sensation assessment. Also, we did not stop the CPET assessments by any risk event associated with excessive physiological stress.
We studied two competitive male cyclists recruited from a national cycling club. The inclusion criteria for this study were physically active participants (≥150 min of moderate or ≥75 min of vigorous physical activity per week) with a normal body mass index (20–25 kg·m-2). Exclusion criteria for this study were a history of respiratory, cardiovascular, metabolic, musculoskeletal, or neoplastic disease or an infectious or inflammatory process at least 2 weeks before the study assessments.
The completion of a CPET, accompanied by non–invasive recording of hemodynamic and tissue oxygenation changes resulting from metabolic changes induced by increased exercise intensity using devices equipped with NIRS technology, allows the identification of central limiting factors associated with changes in brain areas versus peripheral limiting factors related to respiratory or musculoskeletal responses to increased exercise intensity. The testing protocol encompasses a series of steps, including the preparation of the participant, the execution of the exercise test, and the collection of relevant physiological data.
Successful execution of the CPET–NIRS protocol is demonstrated by clear and consistent data across several parameters. During the initial rest stage, measurements such as heart rate (HR), pulse oxygen saturation (SpO2), and NIRS readings are recorded to establish a baseline. The warm-up phase, characterized by low workload pedaling, prepares the participant for the incremental exercise phase, where the workload progressively increases (see Figure 1).

Figure 1: Experimental design of exercise protocol scheme. Schematic representation of the stages of the exercise protocol used in the study, highlighting key events such as rest (R), warm-up (W), exercise (E), finalize (F), and stop (S), which correspond to the flow of the cardiopulmonary exercise testing protocol. Please click here to view a larger version of this figure.
Representative results of the CPET (Figure 2) and NIRS (Figure 3 and Figure 4) data from two male athletes assessed under controlled laboratory ambient conditions (air temperature ~20 ± 2 °C; relative humidity ~40% ± 5 %) are shown: (i) Participant 1 (Figure 2A and Figure 3) (age: 33 years, weight: 80 kg, height: 178 cm, Workload max: 300 W, VO2-max: 46 mL·kg-1·min-1, VE: 177 L·min-1, HR-max (%predicted, 220-years): 100%, PetCO2: 27 mmHg); and (ii) Participant 2 (Figure 2B and Figure 4) (age: 26 years, weight: 67 kg, height: 178 cm, Workload max: 300 W, VO2-max: 51 mL·kg-1·min-1, VE: 131 L·min-1, HR-max (% predicted, 220-years): 93%, PetCO2: 33 mmHg).
In both participants, the VO2 (oxygen consumption), VCO2 (carbon dioxide production), RQ (respiratory quotient, VCO2·VO2-1), HR, VE (lung ventilation), and RR (respiratory rate) exhibit a continuous rise as the exercise intensity increases until the maximal value of VO2 is reached (see Figure 2).

Figure 2: Changes in physiological variables assessed during CPET. The progression of physiological variables measured during cardiopulmonary exercise testing, including oxygen consumption (VO2), minute ventilation (VE), end-tidal CO2 pressure (PetCO2), and power output (Watts), are shown. The transitions between ventilatory thresholds 1 and 2 (VT1 and VT2) are indicated within the exercise stages. (A) Participant 1 and (B) Participant 2. Please click here to view a larger version of this figure.
The NIRS data provided insights into the local metabolic demand during CPET. The exercise-induced changes observed in the target tissue (muscular versus brain) vary depending on the specific tissue and the intensity of the exercise analyzed. Therefore, a useful physiological framework to interpret the exercise-induced NIRS data is the triphasic model of exercise intensity proposed by Skinner and McLellan34. In this model, the authors delineate three zones defined by VTs: Phase I or aerobic, (ii) Phase II or aerobic-anerobic transition, and (ii) Phase III or "metabolic instability".
At exercise intensities below VT2 (Phases I and II), a decrease in O2-Hb and an increase in H-Hb will occur at the muscle level – if there is no significant variation in T-Hb as a parameter of local blood flow. Our CPET-NIRS protocol consists of cyclic exercises with repetitive muscle contractions/relaxations, so minimal variation in T-Hb is expected. However, exercise-induced changes vary depending on the target muscle tissue being evaluated. In locomotor muscles, as the m.Vastus Lateralis, the progressive increase in exercise intensity, induces NIRS data changes concomitant with the workload (see Figure 3A). In contrast, in accessory respiratory muscles, such as the m.Intercostales, changes are concurrent with ventilatory changes rather than workload (see Figure 3B). In PFC, an increase of O2-Hb, H-Hb, and T-Hb is observed because blood flow exceeds the local demand induced by exercise; also, a slight decrease in TSI could be seen (see Figure 3C). In all tissues evaluated, the TSI parameter decreases as exercise intensity increases, making the changes in m.Vastus Lateralis more notorious than m.Intercostales and PFC (see Figure 3D).

Figure 3: Example of central limitation (Participant 1). NIRS data during CPET protocol (Events: W = Warm-up, E = Exercise, VT1 = Ventilatory threshold 1 or aerobic ventilatory threshold, VT2 = Ventilatory threshold 2 or anaerobic ventilatory threshold, F = Finalized exercise or VO2-max). (A) m.Vastus Lateralis, (B) m.Intercostales, and (C) Prefrontal cortex (PFC). Please click here to view a larger version of this figure.
As exercise-induced metabolic demand increases, particularly at intensities above VT2 (Phase III or "metabolic instability"), interesting physiological responses to be studied occur both at the muscular (locomotor and respiratory muscles) and PFC levels. These consist of a marked decrease in O2-Hb and tHb, alongside a remarkable increase in H-Hb, aspects that support the pronounced decline in TSI.
In subjects with high ventilatory demand during high-intensity exercise, the exponential rise in VE and RR causes elevated hyperventilation because of the increase in CO2 of "metabolic origin". This hyperventilation can induce pronounced brain vasoconstriction, thereby limiting performance by central limitation, as seen in this representative subject. Theoretically, the changes observed in NIRS data result from brain vasoconstriction induced by the hypocapnia inferred from the abrupt decrease of pressure end-tidal of CO2 (PetCO2) registered in the CPET (see Figure 2). These physiological changes have been shown to have a high relationship with increased dyspnea induced by exercise, registered using the modified Borg's scale35,36.
On the other hand, subjects with high locomotor demand but not high respiratory demand do not exhibit brain vasoconstriction by hypocapnia. Consequently, NIRS data may continue to reflect changes like those observed at moderate exercise intensities. In these subjects, exercise performance is limited by peripheral rather than central limiting factors (see Figure 4). These physiological changes have been shown to have a high relationship with increased leg fatigue induced by exercise.

Figure 4: Example of peripheral limitation (Participant 2). NIRS data during CPET protocol (Events: W = Warm up, E = Exercise, VT1 = Ventilatory threshold 1 or aerobic ventilatory threshold, VT2 = Ventilatory threshold 2 or anaerobic ventilatory threshold, F = Finalized exercise or VO2–max). (A) m.Vastus Lateralis, (B) m.Intercostales, and (C) Prefrontal cortex (PFC). Please click here to view a larger version of this figure.