Introduction
A self-paced maximal oxygen uptake (
\(\dot{V}{\text{O}}_{2\hbox{max} }\)) test (SPV) allows participants to regulate their own exercise intensity according to specific Ratings of Perceived Exertion (RPE), which follows an incremental format over 5 × 2 min stages (11–20 RPE). The SPV has been shown to allow young participants to achieve a higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) when compared to a standard incremental exercise test in both cycling and running exercise modes (Astorino et al.
2015; Mauger and Sculthorpe
2012; Mauger et al.
2013), although a mechanistic explanation is yet to be identified. However, not all studies have found differences in
\(\dot{V}{\text{O}}_{2\hbox{max} }\) when comparing the SPV vs. standard incremental exercise protocols (Chidnok et al.
2013; Faulkner et al.
2015; Straub et al.
2014).
Maximal oxygen consumption is suggested to be limited by factors such as: pulmonary diffusion capacity, maximal cardiac output, oxygen carrying capacity of the blood, or skeletal muscle characteristics (Basset and Howley
2000). Mauger et al. (
2013) suggested that a higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) may be achieved during an SPV protocol due to changes in oxygen extraction at the working muscles, rather than any increase in oxygen delivery. The authors based this hypothesis on the observation that participants achieved a lower peak heart rate (HR) in the SPV without differences in peak minute ventilation (VE); they suggested that if the higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) were a result of an increase in the oxygen delivery it would be expected that both HR and VE would be increased (Mauger and Sculthorpe
2012; Mauger et al.
2013). However, it is possible that increases in stroke volume (SV), which was not measured by Mauger and Sculthorpe (
2012) and Mauger et al. (
2013), may have compensated for the lower HR achieved resulting in an increase in oxygen delivery, and thus
\(\dot{V}{\text{O}}_{2\hbox{max} }\) (Basset and Howley
2000). Nevertheless an increased oxygen extraction in the SPV may still arise from the participant being able to adjust work rate, potentially creating optimal conditions that allow enhanced rates of muscle oxygen extraction and higher work rates to be achieved. Indeed, previous research demonstrates that muscle blood flow, and thus oxygen extraction, is reduced when both muscle force and duration of contractions increase (Bjorklund et al.
2010; Hoelting et al.
2001). Research has also suggested that decreased blood transit time is associated with reduced rates of oxygen extraction (Basset and Howley
2000; Kalliokoski et al.
2001). In traditional incremental tests, the continuous increase in force requirement and consistent duration of each contraction cannot be adjusted. As a consequence this may result in a constriction of the amount of blood flow available to the muscle, and reduced time for extraction to take place. Both force and duration of muscle contractions are free for the participant to vary during the SPV, potentially optimising working muscle blood flow. In addition, it has been suggested that the self-regulation of work rate during the SPV may improve the efficiency of muscle recruitment, i.e. affording greater reliance on more oxygen efficient muscle fibres (Type I), particularly in the earlier stages of the test (Mauger and Sculthorpe
2012; Mauger et al.
2013). This self-regulation may help to conserve Type II fibres for the “all-out” final stage of the SPV protocol (RPE 20) which ultimately allows the high work rates to be achieved.
A recent study has demonstrated that the SPV may elicit a greater cardiac output (
Q) and peak HR when compared to a standard incremental ramp protocol (Astorino et al.
2015). The study by Astorino et al. (
2015) was the first to assess
Q across the SPV, and suggests that the greater
\(\dot{V}{\text{O}}_{2\hbox{max} }\) may be explained by an increase in oxygen delivery (Mortensen et al.
2005), rather than an increase in the oxygen extraction as previously suggested. Astorino and colleagues (
2015) suggested the increase in
Q during the SPV protocol is likely the result of participants adequately pacing their effort to minimise fatigue in the early stages of the test. Ultimately, this better pacing led to a greater work rate being achieved in the final stage of the SPV, potentially because participants preserved the use of type II fibres in the earlier stages, resulting in a greater HR,
Q and
\(\dot{V}{\text{O}}_{2\hbox{max} }\).
It is well accepted that there is a decline in
\(\dot{V}{\text{O}}_{2\hbox{max} }\) with aging, which is predominantly the result of reductions in maximal
Q and muscle blood flow (Betik and Hepple
2008). Cardiac function (Lakatta and Levy
2003), lung performance (Chaunchaiyakul et al.
2004; Janssens et al.
1999) and muscle oxidative capacity (Betik and Hepple
2008; Russ and Kent-Braun
2004) have all been shown to reduce with age, which also contribute to the deterioration in
\(\dot{V}{\text{O}}_{2\hbox{max} }\). Therefore, the factors limiting factors
\(\dot{V}{\text{O}}_{2\hbox{max} }\) may differ between young and old populations. Consequently, selecting a
\(\dot{V}{\text{O}}_{2\hbox{max} }\) test protocol which adequately stresses the body’s physiological systems is an important decision when testing young and older populations. With the study of Astorino et al. (
2015) demonstrating that the SPV protocol produces higher
Q and
\(\dot{V}{\text{O}}_{2\hbox{max} }\) values compared to traditional methods, it could be speculated that a self-paced exercise test provides the best method to maximally stress the cardiorespiratory system. Furthermore, examining cardiopulmonary and muscular responses of young and old populations to both SPV and traditional test protocols may help explain why higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) values are often seen from the SPV. Therefore, the aim of the current study was to assess physiological responses to both SPV and standard incremental ramp test (RAMP) protocols in healthy younger (18–30) and older (50–75) populations, and to objectively test whether responses differ between the two groups.
Discussion
This is the first study to assess both the cardiovascular and muscular response to an incremental self-paced exercise test compared to a standard RAMP protocol, and do so in both young and older populations. In support of previous literature (Astorino et al.
2015; Mauger and Sculthorpe
2012; Mauger et al.
2013), the results of the current study demonstrate that younger participants were able to achieve a significantly higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) in the SPV compared to the RAMP protocol. However, this was not evident in the older population where there were no differences in
\(\dot{V}{\text{O}}_{2\hbox{max} }\) between the two protocols. The results also demonstrate that the SPV produced a higher peak
Q and peak SV in the young group, which was statistically different between the SPV and RAMP at the latter stages of the test. No differences in these parameters were evident between tests in the older group. However, interestingly higher physiological work rates were achieved by participants from both age groups (as evidenced by the significantly greater peak PO and RER values) in the SPV protocol, but this only lead to a higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) in the young group. In turn this suggests that the limiting factor to
\(\dot{V}{\text{O}}_{2\hbox{max} }\) and the mechanisms behind the SPV may differ between young and old populations. Results from the current study also demonstrate a lack of protocol specific patterns of deoxyHb and muscle recruitment of the VL, suggesting that oxygen extraction was not enhanced by the SPV.
In support of findings of the current study, Astorino et al. (
2015) demonstrate a significantly higher
Q and
\(\dot{V}{\text{O}}_{2\hbox{max} }\) in the SPV when compared against a standard exercise test protocol. The authors suggested that the higher
Q during the SPV is responsible for the higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) observed due to an increased oxygen delivery to the working muscles. It is well accepted that there is a strong linear relationship between
Q and
\(\dot{V}{\text{O}}_{2}\), with
Q being a principal limiting factor for
\(\dot{V}{\text{O}}_{2\hbox{max} }\) during whole-body exercise (Basset and Howley
2000). Therefore, it would be expected that a higher
Q would result in a greater
\(\dot{V}{\text{O}}_{2\hbox{max} }\) being achieved. Results from the current study demonstrate a significantly higher
Q achieved in the SPV vs. the RAMP in the young, but not the old population. Interestingly, with
\(\dot{V}{\text{O}}_{2\hbox{max} }\) only being significantly higher in SPV in the young group, it suggests that
Q might be the primary limiter. However, it must be acknowledged that peak values for
Q and
\(\dot{V}{\text{O}}_{2\hbox{max} }\) may not have necessarily occurred at the same time in the SPV, and so caution must be applied when identifying an increased
Q as the sole explanation for the higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) on the basis of these data. Nevertheless, the same maximal heart rate achieved in both protocols suggests that the enhanced
Q seen in the young group is predominantly the result of the higher SV achieved in the SPV vs. RAMP. However, as stated above, this cannot be confirmed as peak SV and peak HR may not have necessarily occurred at the same time point. Indeed, previous literature has suggested that differences in
\(\dot{V}{\text{O}}_{2\hbox{max} }\) between individuals are primarily a result of the differences in maximal SV, as less inter-individual variation is seen in maximal HR (Basset and Howley
2000). In traditional
\(\dot{V}{\text{O}}_{2\hbox{max} }\) tests it is known that SV begins to plateau/fall prior to
\(\dot{V}{\text{O}}_{2\hbox{max} }\) being reached, whilst HR continues to increase to a maximal level, thus causing a plateau in
Q near the end of the exercise test (Mortensen et al.
2005). This demonstrates an impairment of the circulatory system to continue supplying a linear increase in oxygen delivery at higher exercise intensities (Mortensen et al.
2005). The main cause for the plateau/fall in SV is suggested to be predominately attributed to a decrease in diastolic filling time as a result of the increasing heart rate (Higginbotham et al.
1986; Vella and Robergs
2005). From the data presented in Fig.
4 it is evident that SV in the young group increases to a greater extent in the SPV than in the RAMP. A plateau in SV is also evident where the rate of increase from the early stages of the test appears to “level-off” sooner in the RAMP than in the SPV. As a result, the increase in
Q during the final stage of the SPV test is greater than in the RAMP protocol. In the older group a plateau in SV is evident in both tests, although there is a trend for SV to be higher in the latter stages of the SPV compared to the RAMP. This, combined with the significantly higher HR in minute 7 and 8 during the SPV, may partly explain the observed higher
Q in the 6th, 7th and 8th minute.
As participants are free to make adjustments in work rate, it may be that the self-paced nature of the SPV (particularly in the younger populations), contributes to the prevention of an early plateau in SV by potentially creating more optimal physiological conditions to maintain adequate oxygen delivery. However, further research is required to test this hypothesis and examine the underpinning mechanisms behind the SV response. Despite the young group’s higher peak Q, their peak a-vO2diff was lower in the SPV, which would have potentially negatively affected the magnitude of the observed \(\dot{V}{\text{O}}_{2\hbox{max} }\). Unfortunately, data from the current study cannot identify reasons for the observed lower a-vO2diff, but an oxygen diffusion limitation (caused by the higher Q), or reductions in muscle blood flow may be likely candidates. Future studies using invasive, direct measures of a-vO2diff are necessary to investigate these proposed mechanisms.
Older participants demonstrated no significant differences in
\(\dot{V}{\text{O}}_{2\hbox{max} }\) between the SPV and RAMP protocols. It is possible that the effect of the protocol is reduced in the older group due to their inability to further increase
Q in response to the increased work rates that the SPV protocol allows (as seen in the young group). Indeed, previous research has suggested that
Q plateaus at around ~80% of peak PO (Mortensen et al.
2005), and even though in the young group there were no significant differences in the increase of
Q between the final time points (6–8 min) of the SPV (
p > 0.05), the pattern of response is not the same as in the RAMP (see Fig.
4). Moreover, even though there were no differences in peak
Q in the old group between the two tests, they still achieved a higher PO in the SPV vs. the RAMP. It is not clear why this differential response in
Q has been observed between age groups across the test protocols, but it has been suggested that there are age-related changes which occur in relation to cardiac function in healthy individuals. In particular these changes in cardiac function include left ventricular wall thickness, reductions in diastolic filling, impaired left ventricular ejection fraction, and reductions in HR (Lakatta and Levy
2003). All of these changes are known to influence cardiac function (Lakatta and Levy
2003). In particular, reductions in diastolic filling time is suggested to be the primary cause of the plateau that occurs in SV above a certain exercise level in older individuals (Higginbotham et al.
1986; Vella and Robergs
2005; Lakatta and Levy
2003). This could be the key reason why the older group did not achieve a higher peak
Q and therefore
\(\dot{V}{\text{O}}_{2\hbox{max} }\) during the SPV (as seen in the young group). However, interestingly, higher absolute peak
Q and SV values were found in the older group compared to the young group in both test protocols, which suggests reduced cardiac function was not evident in this older population.
A further speculative reason for the divergent protocol effects on
\(\dot{V}{\text{O}}_{2\hbox{max} }\) between the age groups could be due to the known age-related changes that occur at the periphery. Research has suggested that there is a reduced muscle oxidative capacity in older populations (Betik and Hepple
2008; Russ and Kent-Braun
2004) due to loss in mitochondrial content and function, and a reduction of muscle volume (Conley et al.
2000). This reduced oxidative capacity is likely to affect the a-vO
2diff, which according to the Fick equation, contributes to the attainment of
\(\dot{V}{\text{O}}_{2\hbox{max} }\). In support of this speculation, data from this study demonstrate that the a-vO
2diff is lower in the old compared to the young group. A further possibility is that due to the reduced lung performance associated with normal age-related decline (Chaunchaiyakul et al.
2004; Janssens et al.
1999), there is an increase in the oxygen cost of breathing meaning that more
Q is needed to be directed to the lungs to support ventilation during exercise (Proctor et al.
1998). Indeed, a 20–30% decrease in leg blood flow during cycling has been shown in older, compared to younger subjects (Proctor et al.
1998). Thus, age-related reductions in leg muscle blood flow may have affected oxygen delivery to the working muscle and consequently limited a-vO
2diff and
\(\dot{V}{\text{O}}_{2\hbox{max} }\).
Previous literature has shown the SPV to produce higher
\(\dot{V}{\text{E}}\) values when compared to a standard
\(\dot{V}{\text{O}}_{2\hbox{max} }\) protocol in a young population (Astorino et al.
2015; Faulkner et al.
2015; Hogg et al.
2014; Mauger et al.
2013). Interestingly, studies that demonstrate no difference in
\(\dot{V}{\text{O}}_{2\hbox{max} }\) between the SPV and standard RAMP protocols also failed to find differences in
\(\dot{V}{\text{E}}\) (Chidnok et al.
2013; Straub et al.
2014). The greater
\(\dot{V}{\text{E}}\) demonstrated by younger participants in the current study, and also reported in previous studies (Astorino et al.
2015; Faulkner et al.
2015; Hogg et al.
2014; Mauger et al.
2013), is likely due to the “all-out” effort required during the final stage (RPE 20) of the test. End-test lactate values (see Table
1) suggest that this “all-out” effort results in a greater level of metabolic stress than experienced in standard RAMP testing. The greater level of acidosis and metabolic buffering would therefore increase the ventilatory response during high intensity exercise (Milani et al.
2006). However, Mauger and Sculthorpe (
2012) observed no differences in
\(\dot{V}{\text{E}}\) between the RAMP and SPV protocols even though a higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) was achieved in the SPV. Therefore, it is still questionable whether or not the higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) resulting from the SPV protocol is a result of a higher rate of
\(\dot{V}{\text{E}}\).
Interestingly, in contrast to the current findings, Chidnok et al. (
2013) did not find a higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\),
\(\dot{V}{\text{E}}\) or end-test blood lactate concentration from a SPV protocol compared to a standard RAMP test. However, the differences in the outcomes between Chidnok et al. and the current study could be attributed to the SPV test protocol designs. As outlined above, the SPV protocol from the current study requires an “all-out” effort to be maintained for the duration of the final stage. This “all-out” maximal effort is very different to pacing a maximal effort (RPE 20) as the highest work rate that can be sustained for the duration of the stage, as required by the protocol of Chidnok et al. (
2013). This fundamental difference between the two test protocols is the likely reason for the disparity between the findings of the two studies. Indeed, data from the current study demonstrate a mean decrease in PO in the final stage of 120 W for the young group, and 73 W for the older group. This is in contrast to the data from Chidnok’s study which demonstrate a mean reduction in PO of just 20 W during the final stage of their SPV protocol. Thus, it could be suggested the final stage “all-out” effort might be required in order to drive the mechanisms that pertain to the higher
\(\dot{V}{\text{O}}_{2\hbox{max} }\) in SPV tests using this protocol design. Interestingly, the PO values at the point of
\(\dot{V}{\text{O}}_{2\hbox{max} }\) in the current study were significantly higher in the RAMP compared to the SPV in both the young (RAMP 263 W; SPV 219 W;
p < 0.01), and old group (RAMP 222 W; SPV 195 W;
p < 0.01). These findings suggest that in the SPV,
\(\dot{V}{\text{O}}_{2\hbox{max} }\) values seem to occur when PO is submaximal. Previous research supports this finding and has demonstrated that PO can be dissociated with
\(\dot{V}{\text{O}}_{2\hbox{max} }\); i.e. PO does not necessarily have to be maximal to achieve maximal
\(\dot{V}{\text{O}}_{2}\) values (Billat et al.
2013, Milani et al.
2006). Thus, the nature of the “all-out” final stage of the SPV test (RPE 20), and time delay in the oxygen uptake kinetic response, is likely to be the reason for the dissociation between
\(\dot{V}{\text{O}}_{2\hbox{max} }\) and the peak PO values.
Limitations
A limitation of the current study is that different cycle ergometers were used to complete the RAMP and SPV. It has previously been suggested that different ergometers might cause differences in metabolic cost and cardiovascular strain (Reiser et al.
2000). Indeed, various factors such as saddle angle (Umberger et al.
1998), body positioning (Too
1991) and saddle to pedal distance (Too
1993) have been shown to influence maximal cycling performance. However, different ergometers were necessary to be able to conduct the two protocols as the SPV required participants to freely adjust their PO, and the RAMP required accurate fixing of PO. Even though different ergometers were used, the current study demonstrated similar magnitude of differences in
\(\dot{V}{\text{O}}_{2\hbox{max} }\) between RAMP and SPV as those reported in previous studies where the same ergometer was used for both tests (Astorino et al.
2015, Mauger and Sculthorpe
2012). Chidnok et al. (
2013) also used different cycle ergometers when making comparisons between the SPV and standard RAMP protocols but found no significant differences in
\(\dot{V}{\text{O}}_{2\hbox{max} }\).
The use of non-invasive techniques to estimate
Q and SV have previously been criticised for their lack or accuracy and reliability when compared against more invasive techniques (e.g. direct Fick method), with the typical error being reported to be around 9% for peak
Q and SV (Welsman et al.
2005). There is also the possibility that movement and respiratory artefacts associated with exercise may have affected our results (Siebenmann et al.
2015). However, previous research has demonstrated that non-invasive devices such as that used in the current study are a reliable and provide clinically acceptable measures of
Q and SV in adults during exercise (Charloux et al.
2000, Richard et al.
2001). Nevertheless, the authors accept that caution must be applied when drawing conclusions from such measures.
The SPV allows participants to freely adjust both workload and cadence. Previous research by Gottshall and colleagues (Gottshall et al.
1996) has suggested that large variations in cadence could influence the muscle blood flow and
Q response to exercise. Thus any differences in cadence between SPV and RAMP protocols may have influenced the current results. However, the study by Gottshall et al. (
1996) was completed during submaximal steady state exercise, and we are unaware of any studies that have presented similar data obtained during maximal incremental exercise. Therefore, it is difficult to determine whether cadence differences were a confounding variable within our study. Nevertheless, the mean cadence from both test protocols were similar in the young (RAMP 76 rev min
−1; SPV 78 rev min
−1), and identical in the old group (77 rev min
−1). We are therefore confident that cadence is unlikely to have had a substantial influence on the results.