Content uploaded by Nathanial R Eckert
Author content
All content in this area was uploaded by Nathanial R Eckert on Jan 18, 2022
Content may be subject to copyright.
2022, Volume 2 (Issue 1): 3 OPEN ACCESS
Research Directs in Health Sciences
Test-Retest Reliability of the Microvascular
Oxygenation Recovery Response Subsequent to
Submaximal Cycling Exercise
Original Research
Trent E. Cayot1, Alicia M. Otto1, Alex Sikora2, Alexandria C. Frick1, Nathanial R. Eckert1, Stacey
L. Gaven2
1University of Indianapolis, Department of Kinesiology, Health, and Sport Sciences, Indianapolis,
Indiana/USA
2University of Indianapolis, Department of Athletic Training, Indianapolis, Indiana/USA
Abstract
Introduction: The purpose was to quantify the within-session and between-session
reliability of halftime (HT) and monoexponential curve fitting (EXP) analyses, when
assessing the microvascular tissue oxygenation (StO2) recovery response via near-
infrared spectroscopy (NIRS).
Methods: Seventeen subjects completed a submaximal cycling test and 6-minute
cool-down on three occasions. The protocol was completed twice during session 1
and once during session 2. StO2 were collected via NIRS from a randomized vastus
lateralis. StO2 response from the last minute of exercise and the entire cool-down was
analyzed using HT and EXP. Within-session and between-session reliability were
examined by mixed, absolute agreement intraclass correlation coefficients (ICC) and
standard error of the measurement (SEM).
Results: HT resulted in higher within-session reliability compared to EXP for
exercising StO2 (ICCHT=0.920, ICCEXP=0.865, SEMHT=4.9 ∆BSL, SEMEXP=6.2
∆BSL) and StO2 recovery time (ICCHT=0.772, ICCEXP=0.720, SEMHT=7 sec,
SEMEXP=9 sec). Similar between-session reliability for exercising StO2 was observed
(ICCHT=0.895, ICCEXP=0.879, SEMHT=5.2 ∆BSL, SEMEXP=5.4 ∆BSL), however HT
elicited higher between-session reliability for StO2 recovery time (ICCHT=0.583,
ICCEXP=-0.211, SEMHT=7 sec, SEMEXP=15 sec).
Conclusions: Due to the better within-session (exercising StO2, StO2 recovery time)
and between-session (StO2 recovery time) reliability, practitioners are encouraged to
use HT when assessing exercising StO2 and StO2 recovery time.
Key Words
: Near-Infrared Spectroscopy, Performance Assessment, Curve Fitting
Analysis
Corresponding author: Trent E. Cayot, cayott@uindy.edu
Introduction
Performance of high intensity exercise repeatedly in an intermittent fashion is an important physical characteristic for
many athletic events and a characteristic that practitioners might be interested in monitoring throughout a training or
rehabilitation intervention. The phosphagen system is heavily relied upon during high intensity, short duration exercise
and uses phosphocreatine (PCr) to help meet the increased adenosine triphosphate (ATP) demand. Therefore, to
Published: January 17,
2022
Copyright, 2022 by the
authors. Published by
Research Directs and
the work is licensed
under the Creative
Commons Attribution
4.0 International
License. To view a copy
of this license, visit
http://creativecommons
.org/licenses/by/4.0/
Research Directs in
Health Sciences: 2022,
Volume 2 (Issue 1): 3
ISSN: 2768-492X
Open Access
2022, Volume 2 (Issue 1): 3
Research Directs in Health Sciences
2
sustain performance during subsequent exercise bouts PCr must be resynthesized from creatine during recovery. Since
decreased PCr concentrations have been associated with increased mitochondrial respiration, an exercise bout which
decreases the PCr/creatine ratio will promote PCr resynthesis1, potentially aiding in sustained performance during
subsequent exercise2. More specifically, creatine can be rephosphorylated via a creatine shuttling mechanism within
the inner mitochondrial membrane and the resulting PCr is then transferred to the cytosol via a voltage-dependent
anion channel3,4. Since the creatine shuttling mechanism is ATP dependent, requiring oxygen uptake, many studies
have investigated the microvascular tissue oxygenation (StO2) response during (exercising StO2) and immediately
following (StO2 recovery time) an exercise bout using near-infrared spectroscopy (NIRS)2,5-7. After cardiorespiratory
training, the exercising skeletal muscle can become more economical during submaximal exercise6 and the StO2
response can recover faster immediately following exercise2. The faster StO2 recovery time has been associated with
PCr resynthesis rates7, maximal aerobic speed2, the ability to sustain high intensity efforts during multiple exercise
bouts2, and being a non-invasive indicator of skeletal muscle oxidative capacity8.
Because StO2 training adaptations may be advantageous for performance2,6-8, practitioners may be interested in
monitoring the StO2 response throughout a training2,6 or rehabilitation9 intervention. Previous investigations use
multiple analysis techniques to quantify the StO2 variables, including halftime (HT)2,5 and monoexponential curve
fitting (EXP)7,10 analyses. Identifying the most reliable StO2 analysis method would be beneficial for practitioners
interested in monitoring potential StO2 changes across training or rehabilitation interventions. Therefore, the primary
purpose of the present investigation was to examine the within-session and between-session reliability of two StO2
variables when analyzed by HT and EXP. It was hypothesized that both HT and EXP would provide reliable StO 2
measurements.
Scientific Methods
Participants
Seventeen healthy, active subjects (age = 24 ± 3 years, height = 1.74 ± 0.08 m, weight = 72.2 ± 11.7 kg, thigh skinfold
= 11 ± 5 mm, estimated maximal oxygen uptake = 40.4 ± 5.8 ml/kg/min) participated in the investigation. Based
upon a previous NIRS investigation11, a power analysis indicated that nine subjects were needed for the present
investigation (power = 0.80, alpha = 0.05). Subjects were informed of the investigation’s purpose, procedure, and
possible risks. All subjects completed an informed consent that was approved by the university’s Institutional Review
Board for Human Subject Research and was in accordance with the Declaration of Helsinki. Any individual who self-
reported a history of metabolic, pulmonary, or cardiovascular disease, an orthopedic related injury in the past 12
months, or who were physically inactive were excluded. Physical activity was defined as participating in structured
exercise at least 30 minutes per session, three days a week for the last three months12.
Protocol
Subjects reported to the laboratory on two occasions at approximately the same time of day (±1 hour) separated by 48
hours. Participants were outfitted with a heart rate monitor (Garmin International Inc., Olathe, Kansas, USA) around
their chest and a wireless, continuous-wave NIRS sensor (Moxy Muscle Oxygen Monitor, Fortiori Design, LLC,
Hutchinson, Minnesota, USA) on a randomized thigh. Prior to the NIRS sensor placement, the thigh was shaved and
cleansed with an alcohol preparation pad. The NIRS sensor was placed midway between the anterior superior iliac
spine and the proximal patella over the vastus lateralis11 and StO2 data were sampled at 2 Hz. The NIRS sensor was
covered with a flexible plastic cover to help prevent stray visible light from affecting the NIRS signal and was secured
to the thigh with tape and elastic wraps to limit movement artefact.
After five minutes of rest, resting heart rate was recorded and maximal heart rate was estimated (Estimated HRMAX =
208 – [0.7 x age])12. The resting and estimated maximal heart rates were used to calculate 70% of the subject’s heart
rate reserve (HRR). Subjects performed the YMCA Submaximal Cycling Test12 (bout 1) on an electromagnetic cycle
ergometer (Corival, Lode, Groningen, The Netherlands). All subjects cycled at 25 watts during the first three-minute
stage (stage 1) and the work rates used for subsequent stages were dependent upon the subject’s heart rate at the end
of stage 1 (<80 bpm = 125 watts, 80-89 bpm = 100 watts, 90-100 bpm = 75 watts, >100 bpm = 50 watts). After the
initial work rate increase from stage 1 to stage 2, the work rate was increased by 25 watts for all subsequent stages.
Once the exercising heart rate reached 70% HRR the subject finished the three-minute stage and then actively
recovered for six minutes at 25 watts. In order to account for potential effects that NIRS sensor placement across
multiple sessions might have, both within-session (two exercise bouts within the same session where the NIRS sensor
was not moved; Bout 1 and Bout 2) and between-session (one exercise bout performed during two sessions where the
NIRS sensor was removed and then reapplied; Bout 1 and Bout 3) test-retest reliability analyses were performed.
2022, Volume 2 (Issue 1): 3
Research Directs in Health Sciences
3
Therefore, subjects performed the cycling test and six-minute active recovery again 20 minutes later (bout 2) and once
more 48 hours later (bout 3).
All raw StO2 data from the last minute of exercise (-60 sec to 0 sec) and the six-minute active recovery (0 sec to 360
sec) were interpolated and smoothed using a three-second moving average2. For EXP (Figure 1), the StO2 variables
were determined by fitting the data to the following monoexponential curve (Sigma Plot 14.0, Systat Software Inc.,
Chicago, Illinois, USA), similar to previous investigations7,10:
StO2(t) = StO2(BSL) + Amp · (1-e-t/tau)
StO2(BSL), Amp, and tau represents the exercising StO2, StO2 recovery amplitude, and StO2 recovery time,
respectively. For HT, a custom macro function was written (Excel, Microsoft, Redmond, Washington, USA) to
calculate the StO2 variables (Figure 1). Specifically, the exercising StO2 was calculated as the average of the last minute
of exercise (-60 sec to 0 sec). The StO2 recovery time was determined as the time the StO2 response took to recover
50% between the exercising StO2 and peak StO25. Regardless of analysis method, all exercising StO2 values were then
normalized to the last 30 seconds of the first exercise stage and thus are presented as a change from baseline (0 ∆BSL
= 25 W baseline)11.
Figure 1. Representative Example of Monoexponential Curve Fitting Analysis and Halftime Analysis
A representative example of the exercising tissue oxygen saturation (StO2) response (-60 sec to 0 sec) and tissue oxygen
saturation (StO2) recovery response (0 sec to 360 sec) analyzed via monoexponential curve fitting analysis (red dashed
line in left figure) and halftime analysis (red dashed line in right figure).
Statistical Analysis
Relative and absolute within-session (bout 1, bout 2) and between-session (bout 1, bout 3) test-retest reliability of the
StO2 variables were assess by mixed, absolute agreement intraclass correlation coefficients (ICC) and standard error of
the measurement (SEM), respectively. Additionally, the minimal detectable change (MDC) was calculated at a 95%
level of confidence for each StO2 variable. ICC were interpreted as 0.90-1.00 = “excellent”, 0.75-0.89 = “good”, 0.50-
0.74 = “moderate”, and <0.50 = “poor”. All statistical analyses were performed using Statistical Package for Social
Sciences, version 23 (IBM Corporation, Armonk, New York, USA). Statistical significance was set at p<0.05.
Results
StO2 recovery responses had a moderate-good fit when examined with EXP during bouts 1 (r=0.635-0.997), 2
(r=0.501-0.996), and 3 (r=0.791-0.997). The within-session and between-session StO2 reliability results are described
in Table 1 and 2, respectively.
2022, Volume 2 (Issue 1): 3
Research Directs in Health Sciences
4
Table 1. Within-Session Test-Retest Reliability
Variable
Bout 1
Bout 2
ICC
SEM
MDC
Halftime (HT) Analysis
Exercise StO2
-13.1 ± 17.3
∆BSL
-16.0 ± 17.7
∆BSL
0.920*
0.783-0.971
“Excellent”
4.9 ∆BSL
13.6 ∆BSL
StO2 Recovery Time
25 ± 13 sec
27 ± 17 sec
0.772*
0.479-0.911
“Good”
7 sec
20 sec
Monoexponential Curve Fitting (EXP) Analysis
Exercise StO2
-12.4 ± 16.2
∆BSL
-15.6 ± 17.6
∆BSL
0.865*
0.669-0.949
“Good”
6.2 ∆BSL
17.1 ∆BSL
StO2 Recovery Time
24 ± 18 sec
24 ± 16 sec
0.720*
0.373-0.889
“Moderate”
9 sec
24 sec
Group average ± standard deviation is reported for each variable during Bout 1 and Bout 2. Intraclass correlations
(ICC; single measure results with 95% confidence intervals and interpretations), standard error of the measurement
(SEM), and minimal detectable change (MDC) are provided for each variable. *Significant Intraclass Correlations
(p<0.05).
Table 2. Between-Session Test-Retest Reliability
Variable
Bout 1
Bout 3
ICC
SEM
MDC
Halftime (HT) Analysis
Exercise StO2
-13.1 ± 17.3
∆BSL
-17.8 ± 14.7
∆BSL
0.895*
0.687-0.963
“Good”
5.2 ∆BSL
14.4 ∆BSL
StO2 Recovery
Time
25 ± 13 sec
21 ± 9 sec
0.583*
-0.101-0.846
“Moderate”
7 sec
21 sec
Monoexponential Curve Fitting (EXP) Analysis
Exercise StO2
-12.4 ± 16.2
∆BSL
-17.8 ± 14.7
∆BSL
0.879*
0.614-0.958
“Good”
5.4 ∆BSL
14.9 ∆BSL
StO2 Recovery
Time
24 ± 18 sec
17 ± 7 sec
-0.211
-1.974-0.541
“Poor”
15 sec
42 sec
Group average ± standard deviation is reported for each variable during Bout 1 and Bout 3. Intraclass correlations
(ICC; average measure results with 95% confidence intervals and interpretations), standard error of the measurement
(SEM), and minimal detectable change (MDC) are provided for each variable. *Significant Intraclass Correlations
(p<0.05).
Discussion
HT demonstrated better relative (higher ICC) and absolute (lower SEM) within-session reliability for both exercising
StO2 and StO2 recovery time compared to EXP (Table 1). Practitioners may be interested that when using HT, they
can be confident that a real change occurred if the exercising StO2 or StO2 recovery time changes within a single
2022, Volume 2 (Issue 1): 3
Research Directs in Health Sciences
5
session by 13.6 ∆BSL or 20 sec, respectively. These minimal detectable changes could be used as reference values when
interpreting the effects of acute interventions on the StO2 response.
Both HT and EXP provided “good” between-session reliability for exercising StO2 (Table 2). The present exercising
StO2 between-session reliability results are similar to previous findings which demonstrated reliable exercising StO2
responses at moderate (50% peak work rate, ICC = 0.841) and peak (100% peak work rate, ICC = 0.873) cycling
intensities11. The “good” between-session reliability for exercising StO2 may allow practitioners to further investigate
the efficacy that training6 and rehabilitation9 interventions have on skeletal muscle exercise economy. According to our
results, practitioners using HT can be confident that a change occurred between sessions when the exercising StO2
response changes by 14.4 ∆BSL.
HT also resulted in slightly better and statistically significant between-session reliability for StO2 recovery time
compared to EXP (Table 2). The lower between-session reliability for StO2 recovery time could have occurred due to
differences in NIRS sensor placement between sessions, inter-subject self-selected cycling cadences used, and/or
because only one recovery trial was analyzed instead of multiple recovery trials being ensemble averaged and then
analyzed10. In order to mitigate some potential sources of measurement error, investigators used prominent anatomical
landmarks to standardize the NIRS sensor placement11. Additionally, an electromagnetic cycle ergometer, which
automatically adjusted the resistive torque to maintain constant power output dependent on the cycling cadence, was
used to decrease the potential effects that inter-subject, self-selected cycling cadences may have had on the present
reliability results. It is important to mention that no differences in exercising StO2 were identified when cycling was
performed at different cadences while using the same electromagnetic cycle ergometer13. It was a priority of the present
investigation to assess testing methods that could easily be implemented in the field or clinical setting. Therefore,
instead of having subjects complete multiple exercise and active recovery bouts10, only one exercise and active recovery
bout was performed. This methodological decision might have led to decreased between-session reliability (Table 2)
compared to if multiple exercise and recovery bouts were completed, ensembled averaged, and then analyzed.
The current investigation is not without limitations. NIRS measurements were only recorded from one muscle site
and heterogeneous exercising StO2 responses have been reported within the same muscle14 and after injury9.
Additionally, since the StO2 response reflects the balance between oxygen delivery and oxygen utilization within the
microvasculature, practitioners would not be able to identify the mechanism responsible for any potential change in
the StO2 recovery time using the current methodological approach. Suprasystolic occlusion protocols that temporarily
occlude local blood flow during the recovery after exercise have been validated in non-invasively assessing
mitochondrial function15. However, since StO2 recovery time has been shown to be an indicator of muscle oxidative
capacity8 the current study did not include the suprasystolic occlusion protocol during recovery as practitioners in the
field or clinical setting may not have access to the required equipment (rapid cuff inflation system) or the testing
population may have contraindications to perform an occlusion protocol. Lastly, cycling exercise was the only exercise
modality assessed during the present investigation. The investigators chose to examine cycling exercise due to the low
weight bearing properties and increased potential use throughout a clinical/rehabilitation intervention. Therefore, it
would be prudent to assess the reliability of StO2 variables using other exercise modalities, such as isometric exercise9.
Conclusions
Based upon the present findings, it is recommended that practitioners interested in non-invasively assessing oxidative
capacity after a single exercise bout and active recovery on a cycle ergometer use HT to analyze the exercising StO2
and StO2 recovery time due to the slightly higher within-session and between-session reliability. Practitioners interested
in observing if an exercise modality or training/rehabilitation intervention had a real effect on the exercising StO2
response or StO2 recovery time could be confident a change occurred when using the reference values of ~14 ∆BSL
and ~20 sec, respectively. Lastly, HT can be analyzed using a commonly available software compared to the more
advanced statistical software needed for EXP, possibly making HT accessible to more practitioners within the field
and clinical settings.
Acknowledgements
The authors would like to sincerely thank the participants for volunteering their time and efforts during the research
investigation. This study was funded by the University of Indianapolis InQuery Collaborative Research Grant. The
authors declare no conflicts of interest.
2022, Volume 2 (Issue 1): 3
Research Directs in Health Sciences
6
References
1. Walsh B, Tonkonogi M, Soderlund K, Hultman E, Saks V, Sahlin K. The role of phosphorylcreatine and
creatine in the regulation of mitochondrial respiration in human skeletal muscle. J Physiol. 2001;537:971-978.
2. Buchheit M, Cormie P, Abbiss C, Ahmaidi S, Nosaka K, Laursen P. Muscle deoxygenation during repeated
sprint running: effect of active vs. passive recovery. Int J Sports Med. 2009;30:418-425.
3. Guzun R, Gonzalez-Granillo M, Karu-Varikmaa M, et al. Regulation of respiration in muscle cells in vivo by
VDAC through interaction with the cytoskeleton and MtCK within mitochondrial interactosome. Biochim
Biophys Acta. 2012;1818:1545-1554.
4. Perry C, Kane D, Herbst E, et al. Mitochondrial creatine kinase activity and phosphate shuttling are acutely
regulated by exercise in human skeletal muscle. J Physiol. 2012;590:5475-5486.
5. Chance B, Dait M, Zhang C, Hamaoka T, Hagerman F. Recovery from exercise-induced desaturation in the
quadriceps muscles of elite competitive rowers. Am J Physiol. 1992;262:C766-C775.
6. Costes F, Prieur F, Feasson L, Geyssant A, Barthelemy J, Denis C. Influence of training on NIRS muscle
oxygen saturation during submaximal exercise. Med Sci Sports Exerc. 2001;33:1484-1489.
7. McCully K, Iotti S, Kendrick K, et al. Simultaneous in vivo measurements of HbO2 saturation and PCr
kinetics after exercise in normal humans. J Appl Physiol. 1994;77:5-10.
8. Hamaoka T, McCully K, Niwayama M, Chance B. The use of muscle near-infrared spectroscopy in sport,
health and medical sciences: recent developments. Philos Trans A Math Phys Eng Sci. 2011;369:4591-4604.
9. Jaffri A, Lynch M, Saliba S, Hart J. Quadriceps oxygenation during exercise in patients with anterior cruciate
ligament reconstruction. J Athl Train. 2021;Ahead of Print:1-7.
10. Davies R, Eston R, Poole D, et al. Effect of eccentric exercise-induced muscle damage on the dynamics of
muscle oxygenation and pulmonary oxygen uptake. J Appl Physiol. 2008;105:1413-1421.
11. Cayot T, Robinson S, Davis L, et al. Estimating the lactate threshold using wireless near-infrared spectroscopy
and threshold detection analyses. Int J Exerc Sci. 2021;14:284-294.
12. Riebe D, Ehrman J, Liguori G, Magal M, eds. ACSM's Guidelines for Exercise Testing and Prescription. 10 ed.
Philadelphia, PA: Wolters Kluwer; 2018.
13. Shastri L, Alkhalil M, Forbes C, et al. Skeletal muscle oxygenation during cycling at different power output
and cadence. Physiol Rep. 2019;7:e13963.
14. Miyamoto N, Wakahara T, Ema R, Kawakami Y. Non-uniform muscle oxygenation despite uniform
neuromuscular activity within the vastus lateralis during fatiguing heavy resistance exercise. Clin Physiol Funct
Imaging. 2013;33:463-469.
15. Ryan T, Southern W, Reynolds M, McCully K. A cross-validation of near-infrared spectroscopy
measurements of skeletal muscle oxidative capacity with phosphorus magnetic resonance spectroscopy. J
Appl Physiol. 2013;115:1757-1766.
2022, Volume 2 (Issue 1): 3
Research Directs in Health Sciences
7
Appendix
A custom macro was used in Visual Basic for Applications (VBA), which is included in Microsoft Excel (Redmond,
Washington, USA), to quantify the halftime (HT) of the near-infrared spectroscopy (NIRS) recovery response. For
this macro function the time data and the microvascular tissue oxygenation (StO2) data were placed in column B and
column C, respectively. The “NIRS_Recovery” macro (see below) was then copied and pasted into the VBA dialog
box to “run” the macro. The macro calculated the average end exercise StO2 (cell I2), the peak StO2 during recovery
(cell I5), the StO2 recovery amplitude (peak StO2 – average end exercise StO2; cell I6), and the 50% StO2 amplitude
(cell I10). The conditional formatting function on the “Home” tab in Excel was then used to identify when the StO2
response (column C) was equal to or above the 50% StO2 amplitude (cell I10) and the corresponding time value (from
column B) was recorded as the HT. It is important to mention that the first two lines of the custom written macro is
dependent upon the sampling frequency in which the NIRS data was recorded and therefore might need to be adjusted
based upon the data collection and data analysis procedures. The current macro below is written for NIRS data that is
in 3 second bins (see Protocol section).
Sub NIRS_Recovery()
Range("I2").Value = Application.WorksheetFunction.Average(Range("C2", "C22"))
Range("I5").Value = Application.WorksheetFunction.Max(Range("C22:C450"))
Range("I6").Value = Application.WorksheetFunction.Sum(Range("I5"), -Range("I2"))
Range("I7").Value = Application.WorksheetFunction.Product(Range("I6") / "2")
Range("I10").Value = Application.WorksheetFunction.Sum(Range("I2"), Range("I7"))
End Sub