data processing from indirect calorimetry: recommendations and guidelines robert a. robergs, ph.d.,...

Data Processing From Indirect Calorimetry: Recommendations and Guidelines

Robert A. Robergs, Ph.D., FASEP, EPC

Exercise Physiology Laboratories, Exercise Science Program, University of New Mexico

Maximal Oxygen Consumption (VO2max)

Time or Intensity

VO2

VO2max

Cardiorespiratory limitations

A.V. Hill

VE

FEO2

FECO2

• No universally recommended procedures for processing VO2 data

from breath-by-breath indirect calorimetry, or from time averaged systems.

• No standardized criteria or recommended methods for detecting either of a VO2 plateau, the maximal rate of oxygen consumption

(VO2max), or a peak VO2 in the absence of a VO2 plateau

(VO2peak).

• Increasing use of breath-by-breath indirect calorimetry in education, research and professional practice

• The lack of any objective criteria to follow when processing decreases the validity of measurement.

Background

Challenges• How do researchers in exercise physiology currently collect and process data?

• What causes the “noise” in breath-by-breath VO2 data?

• Should this “noise” be reduced?

• How should this “noise” be reduced?

• What is a VO2 plateau?

• How can a VO2 plateau be objectively determined?

• What is VO2max?

• What is VO2peak?

• How can VO2max and VO2peak be objectively determined?

How do researchers in exercise physiology currently collect and process data?

• Survey conducted over internet

• International sport science email discussion list (www.sportsci.org)

• n = 75

• Breath-by-breath = 48%

• Time averaged mixing chamber = 25%

• Either depending on purpose = 27%

• Data processing = 30 s (38%), 60 s (18%), 20 s (11%), a moving average of 5-11 breaths (10%), 15 s (8%) and the middle 5 of 7 breaths (7%), other (8%)

• Check for VO2 plateau = 93%

• VO2 Plateau criteria

• < 150 mL/min (34%)

• < 2 mL/kg/min (27%)

• subjective visual (18%)

• other (19%)

• Secondary criteria = attainment of age predicted HRmax (53%), RER > 1.10 (49%) or RER > 1.15 (27%), RPE > 17, 18 or 19 (20%)

• Why these methods used? = own beliefs (32%), what they were taught (26%), what they read in research articles (22%), tradition (13%) and the influence of their colleagues (7%)

• No distinction between VO2max or VO2peak = 76%

1.01.52.02.53.03.5

VO

2 (

L/m

in)

What causes the “noise” in breath-by-breath VO2 data?

2.17 0.3 L/min, with a range of 1.4 – 3.3 L/min

10

25

40

55

70

VE

ST

PD

(L/m

in)

0 2 4 6 8 10 12 14

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

MeasuredPredicted

Time (min)

VO

2 (

L/m

in)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.00.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0 Pred VO2 = 0.9572(Meas VO2) + 0.0898

R2 = 0.9572SEE = 0.059 L/min)

Measured VO2 (L/min)

Pre

dic

ted

VO

2 (

L/m

in)

Variability 96 % explained by a two-factor model of VE and FEO2

0 2 4 6 8 10 12 140.0

0.5

1.0

1.5

2.0

2.5

5%error

Time (min)

VO

2 (

L/m

in)

Should this “noise” be reduced?

SIM

How should this “noise” be reduced?

0 5 10 15 200.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5a=bb

Time (min)

VO

2 (

L/m

in)

0 5 10 15 200

1

2

3

4b=15 s

Time (min)V

O2 (

L/m

in)

0 5 10 15 200

1

2

3

4c=30 s

Time (min)

VO

2 (

L/m

in)

0 5 10 15 200

1

2

3

4d=60 s

Time (min)

VO

2 (

L/m

in)

0 5 10 15 200

10

20

30

40

15 s

30 s

60 s

e

Time (min)

Bre

ath

s

Time Averaging

0 5 10 15 200.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5a=bb

Time (min)

VO

2 (

L/m

in)

0 5 10 15 200

1

2

3

4

5b=5 breaths

Time (min)V

O2 (

L/m

in)

0 5 10 15 200

1

2

3

4c=11 breaths

Time (min)

VO

2 (

L/m

in)

0 5 10 15 200

1

2

3

4d=21 breaths

Time (min)

VO

2 (

L/m

in)

0 5 10 15 200.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

21 breaths

11 breaths

5 breaths

bb

Time (min)

Tim

e In

terv

al (

min

)

Breath Averaging

0

1

2

3

4a

VO

2 (

L/m

in)

0

1

2

3

4d

VO

2 (

L/m

in)

0

1

2

3

4b

VO

2 (

L/m

in)

0 50 100 150 200 2500

1

2

3

4e

Data Points

VO

2 (

L/m

in)

0 50 100 150 200 2500

1

2

3

4c

Data Points

VO

2 (

L/m

in)

muscle

cardiovascular

ventilation

Digital Filtering

0 2 4 6 8 10 12 14 16 180

500

1000

1500

2000

2500

3000

3500

4000

rest exercise

Time (min)

VO

2 (

mL

/min

)

linear regression

initial deviation from linearity

VO2 plateau at VO2max

12 13 14 15 16 17 182750

3000

3250

3500

3750

Time (min)V

O2

(mL

/min

)

What is a VO2 plateau?

What is VO2max or VO2peak?

0 2 4 6 8 10 12 14 16 180

1

2

3

4

5

Time (min)

VO

2 (

L/m

in)

Data Example

10 11 12 13 14 152000

2400

2800

3200

3600

4000treadmill running

Time (min)

VO

2 (

mL

/min

)

Custom Programming Example

Conclusions

• Clear rationale for processing breath-by-breath VO2 data to decrease “noise”.

• Processing best done by digital filtering

• Still formulating and debating criteria and methods to quantify VO2 plateau, VO2max, VO2peak

• In the absence of a VO2 plateau, what are valid criteria to use to verify a “true” VO2max?

Thank you