The body weight-walking distance product as a superior parameter in determining the VO2 on-kinetics in coronary artery disease

Background The 6-minute walk test distance is frequently used to assess functional capacity of cardiac disease population. Nevertheless, anthropometric differences can confound or misestimate performance, which highlights the need for new parameters. This study aims to investigate the potential of the body weight-walking distance product (D.W), compared to 6-minute walk test distance, to predict exercise capacity measured by VO 2 on-kinetics in coronary artery disease (CAD) patients.Methods Cross-section study in a tertiary-care reference institution. Forty-six participants with multiarterial CAD with and without left ventricular dysfunction underwent a 6-minute walk test with simultaneous use of mobile telemetric cardiopulmonary monitoring to evaluate oxygen uptake (VO 2 ) on-kinetics and other cardiorespiratory responses.Results Perceived effort Borg for lower limb fatigue was only correlated with the D.W (p=0.007). The percent-predicted and actual distance were only modestly to moderately correlated with VO 2 on-kinetics (p<0.05). All the associations of VO 2 on-kinetics parameters were improved by showing a stronger correlation to the D.W (p<0.0001), which also had a larger effect size to identify differences between coronary disease patients compared to distance ( d =1.32 vs d =0.84).Conclusion The D.W demonstrates potential as a measure superior to the distance in determining VO 2 on-kinetics in participants with CAD with and without left ventricular dysfunction. powerful to discriminate the level of disease impairment. These findings suggest that D.W may be a

rest to exercise, defined as VO 2 on-kinetics [2,3]. As with peak VO 2 , VO 2 on-kinetics has shown to be a significant prognostic marker in several chronic disease populations [4], especially in the coronary artery disease (CAD) with left ventricular systolic dysfunction (LVSD). In clinical practice, a number of field tests based on walking performance are frequently performed to assess the submaximal functional capacity in these patients [5].
The 6-minute walk test (6MWT) is easily administered and represents daily-life activities effort [6]. In a fashion similar to constant load ergometer testing, simultaneous use of mobile telemetric cardiopulmonary monitoring (MOB) while performing the 6MWT can provide the assessment of the VO 2 on-kinetics [7]. The mainly parameter obtained by the 6MWT is the distance in meters [8]. However, measurement of walking-distance has limitations in accurately predicting exercise capacity due to the presence of confounding factors (i.e. body weight, step length) [9] and also fails to provide clinical utilization for less impaired patients.
Chuang and Wasserman [10] have described a measurement of walking-work represented as the product of the body weight and distance(D.W). The D.W has been reported to better correlate with peak VO 2 and VO 2 at ventilatory threshold compared to walking-distance in patients with chronic obstructive pulmonary disease [10,11]. Nevertheless, to our knowledge, there is no current analysis that sustains whether the D.W could be more effective than the 6MWT distance in assessing more accurately submaximal exercise capacity in CAD patients, beyond a severe impairment. Thus, the aim of the present study was to investigate the potential of the D.W, compared to 6MWT distance, to reflect variations in VO 2 on-kinetics in CAD patients with and without LVSD.

Study design
This cross-sectional study was conducted at the academic Hospital from a Federal University. Participants with a confirmed diagnosis of stable CAD were included in this study, which was approved by the Institutional Ethics Committee, in accordance with ethics' code of the Declaration of Helsinki. All patients were informed about the study and signed a written consent form.
CAD diagnosis was obtained by coronary angiography. Through echocardiography, a left ventricular ejection fraction (LVEF)<45% was used as a threshold to define participants with LVSD [12]. Exclusion criteria consisted of recent myocardial infarction (<6months), presence of chronic or acute pulmonary disease, morbid obesity, neurological or orthopedic diagnoses impacting the ability to complete the study protocol, inability to comprehend and perform the tests or hemodynamic instability/severe arrhythmias during tests protocols.

6-minute walk test
Submaximal functional capacity was evaluated by the 6MWT, guided by the American Thoracic Society (ATS) recommendations [13]. Prior to the 6MWT, volunteers rested in a chair for 3-5 min to record baseline parameters. At the end of the 6MWT, participants again rested in the seated position for a recovery phase. Dyspnea and leg fatigue were rated by the perceived exertion Borg scale (PEB) before, after and every 2 min during the 6MWT. The highest distance achieved during the 6 minute test was recorded in meters.
The D.W was calculated by multiplying the distance in kilometers (km) and body weight in kilograms (kg) [10]. The prediction equation proposed by Pereira e Soares 14 was used to predict walking distance for all participants.

MOB device during 6MWT
The 6MWT was performed with simultaneous MOB device (Oxycon Mobile-Viasys Healthcare, USA) to measure real time breath-by-breath cardiopulmonary responses [7].
The unit was harnessed to participants in a way that it had no effect on walking performance. Heart rate was recorded by a 12-lead electrocardiogram (Oxycon ECG module, Viasys Healthcare, USA). A facemask (with a dead space <70ml) linked to a turbine volume transducer was used to continuously sample gas exchange, tidal volumes and breathing frequency. Breath-by-breath calculations of VO 2 and carbon dioxide output (VCO 2 ) were digitized. Before each 6MWT, spirometry was performed with the MOB device according to ATS guidelines [15]. Steady-state variables were calculated as an average of the last 2 minutes of exercise. The transition of VO 2 through exercise was registered to obtain curve fitting of VO 2 on-kinetics.

Curve fitting of VO 2 kinetics
Raw breath-by-breath data obtained by the MOB device was preprocessed by the average of consecutive periods of 15 seconds. The fit of VO 2 on-kinetics was performed by a monoexponential regression model [2,16], as follows: where the f(t) represents VO 2 at certain time (t); y0 indicates the lower limit at t=0, i.e. the VO 2 at rest (the mean value of the last minute VO 2 prior to the test); y1 represents the upper limit, indicating theVO 2 steady-state (VO 2SS ); and τ indicates the time constant, i.e., the time needed to reach 63% of the VO 2SS . Since time delay was found to be undistinguishable from the second exponential phase [17], phase I was not modelled in this study. Whereas the time delay was not taken into account, the time needed for a 63% increase in VO 2SS , i.e. time constant (τ), also corresponds to the Mean Response Time (MRT). The MRT was corrected for the work rate (wMRT) during the 6MWT to avoid possible differences in intensity amongst participants. The work rate was obtained using the difference between VO 2 at rest and during effort (VO 2SS −VO 2rest ) [18,19]. The quality of fit was assessed visually by two independent investigators to avoid data with obvious lack-offit, leading to 6 exclusions.

Statistical analysis
Categorical data were expressed in absolute and relative frequency and continuous variables were expressed as mean and standard deviation. The Shapiro-Wilk test was used to investigate normality distribution of data. Pearson correlation test was performed to investigate the relationship between VO 2 on-kinetics and 6MWT parameters. Comparison of 6MWT performance in CAD participants with and without LVSD was assessed by unpaired Student-t test or Mann-U-Whitney, according to data distribution. Cohen's d test was used to investigate the effect size of 6MWT parameters. A post hoc sample size analysis was performed to comprehend whether the effect size of 6MWT parameters would evolve with satisfying power of analysis. A p-value<0.05 was considered statistically significant for all tests.
The VCO 2 , METS, VE, the amount of change in VO 2 and PEB for dyspnea, were all significantly and positively-correlated to the 6MWT parameters ( Table 2). However, all observed associations demonstrated a stronger relationship when the D.W was used.
Notably, the PEB for lower limb fatigue was only found to be correlated with the D.W ( Table 2).
There was a stronger positive correlation between D.W and LVEF compared to the walking distance or to the % predicted distance (r= 0.72, p<0.001 vs r= 0.55, p=0.001 or r= 0.43, p=0.005, respectively). When participants were dichotomized according to LVEF, those with LVSD (LVEF<45%) had worse submaximal functional capacity measures compared to participants without LVSD (LVEF >45%). Participants with LVSD achieved a lower 6MWT distance and lower D.W( Table 3). The effect size Cohen d' test demonstrated that D.W was superior in identifying the difference between the groups compared to 6MWT distance and percent-predicted distance (d=1.32 vs d=0.84 and d=0.02, respectively - Table 3).
Also, because of a large effect size was founded when D.W was used, the analysis of group comparisons achieved a power of 95% (Table 3).

Discussion
The current study demonstrates that the D.W had a better relationship with VO 2 onkinetics compared to 6MWT-distance and percent-predicted distance in patients with CAD.
Moreover, the D.W had a larger effect size than the 6MWT distance in CAD patients with and without LVSD. These findings support the premise that the D.W is an easily applying and potential meaningful measure, in the context of quantifying submaximal exercise performance and physiologic health during exertion, even in patients with less severe impairment.
Field tests have been used to assess exercise capacity in chronic disease populations for a number of years. Submaximal exertion tests enable a greater patient toleration and perhaps had a stronger indication of the ability to perform daily activities [20]. In this context, the 6MWT has been used as a prognostic marker for several diseases, as in cardiac patients, especially in those with LVSD [6,21,22]. Nevertheless, the 6MWT-distance alone is only able to detect the presence of a marked exercise limitation, i.e. in patients unable to walk more than 300 meters. In order to acquire physiological insight, several research groups around the world have simultaneously employed a MOB device during the 6MWT [18,19,23]. The breath-by-breath analysis of cardiopulmonary responses more accurately quantifies physiologic health and of the degree of an exercise limitation.
Furthermore, the analysis of VO 2 during the initial phase of the 6MWT allows for the assessment of VO 2 on-kinetics, as in our study, which is linked to the risk for future adverse events [19,24].
The walking distance in meters had historically been recognized as the primary variable obtained with the 6MWT. Several studies report threshold 6MWT-distance values to predict increased risk of adverse events such as myocardial infarction, stroke, re-hospitalization and mortality [25]. Despite its greater utilization, studies reported only a moderate association between the 6MWT-distance and peak VO 2 and other cardiopulmonary exercise testing parameters [8]. In fact, we also found a moderate relationship between 6MWTdistance with VO 2SS and wMRT in our CAD patients, explaining the discrepancy from the previously known relation between METS and walking-speed. This finding could suggest that the use of the 6MWT-distance is only able to predict an obvious impairment of functional capacity and then an already very predictable risk of morbidity and mortality.
Additionally, predicted values based on age, gender and body mass index have been studied for populations around the world to obtain normative values of performance. Since reference values are acquired in each country separately, it can be difficult to apply this as a worldwide parameter. Moreover, in the current study, only a modest association was found with VO 2 on-kinetics when the percent-predicted values of the 6MWT were used. The lack of a stronger relation could indicate that the functional contribution of participants' body processes during exercise it not well represented by current prediction models.
The current limitations in the physiological information derived from the 6MWT encourage the search of new parameters that more accurately reflect functional performance [9] and an overall health status. Previous reports have already indicated that the work of walking during the 6MWT can be correlated with the horizontal work on a treadmill (W HO ). Since the 6MWT is performed in a horizontal plane and at constant velocity, the work of walking in the 6MWT can be calculated as a product of distance and weight (i.e., D.W) [10]. In our study, we found that all associations with cardiorespiratory responses were greatly improved when the D.W was used rather than walking and percent-predicted distance values. We observed a moderate to modest correlation between both VO 2SS and wMRT with walking and percent-predicted distance. However, when the D.W was applied, a strong correlation was observed with VO 2SS (r=0.82, p<0.001) and wMRT (r=-0.70, p<0.001).
These results corroborate with other findings in the literature; Chuang, Lin and Wasserman [10] found a modest correlation between 6MWT-distance and peak VO 2 (r=0.40, p<0.05) versus a stronger correlation between D.W and peak VO 2 (r=0.67, p<0.05). Similarly, Poersch et al [11] found that peak VO 2 was modestly correlated with distance (r=0.32, p=0.084) and percent-predicted distance using Soares and Pereira [14] equation (r=0.35, p=0.058), while strongly correlated with D.W (r=0.76, p<0.01). Both aforementioned studies were performed in chronic obstructive pulmonary disease cohorts that underwent cardiopulmonary exercise testing. To the best of our knowledge, this is the first study to evaluate the association between D.W of the 6MWT with VO 2 on-kinetics in CAD patients.
It is already well established that CAD exposes the myocardial to a mismatch in oxygen supply and demand. As the time associated with this imbalance elongates, CAD evolves with greater manifestations, leading to a greater impairment in ventricular function, as in LVSD. CAD is the major cause of chronic heart failure leading to progressive degrees of impaired exercise capacity [26]. A limitation of physical activity in this population can be detected by a lower 6MWT performance [6]. In the current study, the LVEF was moderately to strongly correlated to all three 6MWT parameters evaluated in our CAD cohort. As expected based on previous literature [22], our results also revealed that when participants were separated according to LVEF, those with LVSD achieved lower distance and lower D.W during the 6MWT.
In the present study, despite the 6MWT distance already had a large effect size (d=0.84), the D.W effect size was much larger (d=1.32), indicating this as a powerful parameter to discriminate the level of disease impairment. These findings suggest that D.W may be a preferred measure in quantifying submaximal performance. Moreover, the stronger correlation observed to VO 2 on-kinetics support the widespread use of D.W, particularly if ventilatory expired gas analysis is not available.
A limitation of this study was that most of the participants were males (80.4%). Women exhibit a lower exercise capacity and as such a gender bias might impact results, especially due to the utilization of a walking test [27,28]. This was a cross-sectional study with a modest number of patients. Future research is necessary to established specific D.W values that could predict clinical outcomes in cardiac patients and reinforced this as superior to walking distance.

Conclusions
The D.W demonstrates potential as a measure superior to the 6MWT distance in determining VO 2 on-kinetics in participants with CAD. The D.W seems to reflect the work of walking and may be a stronger parameter for evaluation of submaximal exercise capacity and performance, especially if a cardiopulmonary testing device is not available.     Figure 1 A flow-chart of study evaluation protocol.

Figure 2
Pearson's correlation test to investigate association of VO2SS and wMRT in relation to 6MWT parameters. 6MWT, 6-minute walk test; D.W, body weightwalking distance product; VO2SS, oxygen uptake at steady-state.

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download.