I'm underwhelmed by the review. It seems Jammes et al. have little new insight beyond their previous studies (which were good!).
In particular, a lack of discussion of the work of Light and colleagues on the role of sensory chemoreceptors on sensation of fatigue associated pain, nor discussion of why the power at the gas exchange threshold (VT1) on the 2nd CPET is lower than the first day.
This research group has previously shown particular interest in trying to understand the relationship between EMG signals, fatigue, sensory pathways and this ventilatory threshold. See:
https://www.ncbi.nlm.nih.gov/pubmed/12914560 (though I don't agree with this conclusion.
I've just found another paper by this group which I had not read before "EMG Changes in Thigh and Calf Muscles in Fin Swimming Exercise":
https://www.thieme-connect.com/products/ejournals/pdf/10.1055/s-0030-1251993.pdf]
Notably:
Jammes 2010 said:
Based on the observations which indicated a reduced anaerobic muscle metabolism in water exercise associated with a delayed or absent VTh, we hypothesized that the changes in EMG power spectrum, triggered by the activation of the muscle metaboreceptors, are attenuated or absent in fin swimming
(I wouldn't have suggested this hypothesis at all...)
During incremental fin swimming, the MF declined in the three studied muscles, the MF changes began at higher VO2 values in the tibialis anterior and vastus medialis which do not participate to propulsion in fin swimming.
In the present experiment the water temperature ranged between 28 ° C and 31 ° C and we confirmed the absence of M-wave changes.
MF decrease might solely result from an enhanced recruitment of motoneurones which drive the slow-firing motor units.
In dry land cycling exercise, the MF decrease in the vastus lateralis coincided with the determination of the ventilatory threshold when EqO2 increased, i.e., when the ventilatory response was accentuated [17] . Moreover, during a static handgrip executed in dry land conditions, the MF decline in the flexor digitorum superficialis was concomitant with VTh determination, and this also occurred in the absence of any release of metabolites in the circulation produced by an arterial and venous blood flow interruption [18] . These observations lead us to postulate that the metaboreflex, represented by neural pathways arising from contracting muscles, might be responsible for the concomitant changes in the motor drive to muscle and the adaptive enhanced ventilatory response. Based on this interpretation of data, the VTh determination could be used to assess the occurrence of the metaboreflex. Our present data in fin swimming exercise do not support this hypothesis. Indeed, they clearly show a dissociation between the two components of the metaboreflex, i.e., the adjustment of the motor drive to contracting muscles (MF decline) and the enhanced ventilatory response (EqO2 increase). Based on observations that immersion reduces the lactate production in resting and exercising muscles [12, 16, 35], we suppose that the increased lactate level in exercising muscle does not play a key role in the motor component of the metaboreflex.
Because the ROS production is directly proportional to the importance of aerobic pathways [31] which seem to be accentuated in water exercise, fin swimming might constitute a situation of prevailing influence of the oxidative stress. This hypothesis deserves to be verified in further immersion studies in humans.
The fundamental difference between a ramped power CPET and the fin swimming test is that the power output during the ramped CPET is increased in a mostly linear manner, and since motor recruitment is mainly dependent on the force output required, this means that the neural drive must increase to match. (it is also noteworthy that perception of effort is based on the signal generated by the brain and is not altered by peripheral afferents)
Whereas during fin swimming, which doesn't have a feedback mechanism requiring constantly increasing power, a more adaptive motor response is possible namely the increase in frequency doesn't lead to a linear increase in power, if there is a reduction in force. It is notable that the VO2Peak during Fin swimming is achieved at a fairly low heart rate (145 BPM, yet the oxygen consumption at this heart rate is much higher than the equivalent during cycling), and the lack of evidence of significant anaerobic metabolism suggests the output is not limited by peripheral ox. phos. capacity, nor cardiopulmonary limits, instead the number of motor units recruited are far below maximum (reflects lower neural drive in comparison to cycling), presumably due to mechanical limitations of the activity that limits the amount of force per stroke, even at high stroke frequency. It is hard to directly compare, but the ΔRMS (%) was substantially lower during fin swimming compared to cycling.
I would argue, on balance (and I am not the first to suggest this), that the metabolic balance is a simply result of the pattern of motor units that are recruited, their firing rates and their resulting metabolic balance (based on the structure, specifically the size of the motor units, the location of the muscle fibres, the firing rate and their adapted metabolic balance). So the ventilatory threshold is ultimately an artefact of significantly increasing the amount of motor units being recruited during a ramped power exercise test. A decline in mean frequency (the optimal frequency depends on the motor unit itself, and while fatigue can alter this frequency, it can be higher for some units and lower for others) when associated with lower force output, in itself does not lead to more motor units being driven - unless there is feedback requiring power to be maintained or increased, such as during a ramped CPET.
I would have liked to see discussion of the above in the aforementioned review, but alas...