SciELO - Scientific Electronic Library Online

vol.13 issue3High intensity interval training series as indices of acidosis tolerance determination in swimming anaerobic performance predictionValidity of 30 minutes test (T-30) in aerobic capacity, stroke parameters and aerobic performance determination of trained swimmers author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Medicina do Esporte

Print version ISSN 1517-8692On-line version ISSN 1806-9940

Rev Bras Med Esporte vol.13 no.3 Niterói May/June 2007 


Adaptation of the VO2000 gas analiser's mask to mensuration of cardiorespiratory parameters in swimming*



Marcelo PapotiI,II; Pedro Balikian JuniorIII; Benedito Sergio DenadaiIV; Manoel Carlos Spiguel LimaI,III; Adelino Sanchez da SilvaI; Vanessa SanthiagoI; Claudio Alexandre GobattoI

ILaboratório de Fisiologia Aplicada ao Esporte – UNESP – Rio Claro
IICentro de Estudos e Pesquisa da Atividade Física – FIB – Bauru
IIIUniversidade do Oeste Paulista – Presidente Prudente
IVLaboratório de Avaliação da Performance Humana

Correspondence to




The aim of the present study was to develop a snorkel (SNK) with low cost for cardio respiratory parameters measurement in swimming. So, a mask of gas analyzer VO2000 (MASK) was adapted to a SNQ developed by man craftwork form with dead space of 250 ml. Eight subjects were submitted to two incremental tests (IT) on cycloergometer using the MASK and the SNK. The IT(s) occurred until voluntary exhaustion and it was composed per stages of 3 min and with initial load and grade increment of 35 W. In both situations, gaseous samples were collected in intervals of 10 s for determination of the amount of oxygen (VO2), of carbonic gas (VCO2) and of ventilatory volume (VE); heart rate (HR) measurement was done with monitor frequency heart (Polar). The comparison of the cardiorespiratory parameters (VO2, VE, VCO2) and HR measured with a SNK and a MASK was performed with the Student's t test of dependent samples, while the Pearson's correlation test and graphic analyzer of Bland and Altman were performed to verify the associations and agreement among parameters. In all cases the level of significance was P < 0.05. The appropriate equations of the correction of values originating from SNK was verified by systematic error (bias), aleatory (precision) and accuracy (ac). It was not found significant differences among the values of VO2, VCO2 and HR obtained with the MASK and SNK. The values of VE measured with SNK were found significantly higher than the obtained with the MASK. However, all parameters presented high agreement and correlation (0.88 to 0.97). Besides, low values of bias (VO2 = 0.11 L/min; VE = 4.11 L/min; VCO2 = 0.54 L/min; 8.87 bpm), precision (VO2 = 0,24 L/min; VE = 11.02 L/min; VCO2 = 0,18 L/min; 7.42 bpm) and accuracy (VO2 = 0,27 L/min; VE = 11.76 L/min; VCO2 = 0.56 L/min; 11.56 bpm) were verified. Thereby, we could conclude that the SNK developed in this study allow a validate measurement of cardiorespiratory parameters in swimming.

Keywords: Swimming. Snorkel. Oxygen uptake. Ventilatory volume. Heart rate.




The measurement of cardiorespiratory parameters such as ventilatory volume, heart rate, carbon dioxide volume (VCO2) and especially oxygen uptake (VO2) has been topic of several investigations in Physiology of the Exercise, especially in swimming(1). Initially, the gases collected during the efforts were stored in a Douglas bag(2-4) for later analysis. Afterwards, the measurements of these parameters were made easier with the development of gas analyzers with electronic sensors(5-6) which enabled the acquisition of a great number of gas samples and also provide the results in a short period of time. Although the technological advance has made this kind of test performance easier, especially in swimming, it is extremely difficult to obtain a valid estimation for cardiorespiratory parameters(7). However, the importance of these data is crucial, since the technique applied during swimming can considerably modify the energy spent for the individual to move in the liquid environment(3,8). Dal monte et al.(7) reported that the power generated by the swimmer to win the drag may be affected by the use of maladjusted equipment, resulting hence in higher energy expenditure to perform swimming at a given velocity(8-9).

Toussaint et al.(10), developed a specific snorkel for swimming evaluation with the advantage to generate a low drag in relation to conventional snorkels. Dal Monte et al.(7) presented a snorkel similar to the one proposed by Toussaint et al.(10), but with lower dead space. These snorkels were designed however for Douglas bag. Keskinen et al.(6) validated in cycle ergometer a snorkel and a valve system for gas analysis breath to breath in swimming using the K4 gas analyzer.

Considering the lack of specific equipment for measurement of cardiorespiratory parameters in swimming, the present study had the aim to adapt the mask of the VO2000 gas analyzer to measure the VE, VO2 and VCO2 of swimmers.



Eight active individuals, age range of 27 ± 4.1 years and body mass of 75 ± 5.8 Kg participated in the present study.

After having agreed and signed a written consent form approved by the Ethics in Research Committee of the Institute of Biosciences of the State University of the São Paulo Rio Claro Campus, the participants were submitted to two progressive tests in cycle ergometer separated by a minimum rest period of 24 hours. The efforts were performed using a VO2000 mask (MASK) and a swimming snorkel (SNK).

Prior to the tests, the participants remained seated in the cycle ergometer during 5 minutes (determination of the values concerned with the baseline) and later performed the incremental test which consisted of progressive efforts in cycle ergometer of mechanical breakage (Monark) with initial load of 35 w (warm-up) and increment of 35 w until voluntary exhaustion, which was assumed as the participant's inability to remain in the pre-set load for a period of 10 seconds. After exhaustion, the participants continued to exercise for approximately 5 minutes with load of approximately 35 w (back to calmness). The cardiorespiratory variables were measured at ten-second intervals with the use of a gas analyzer (VO2000). The equipment was automatically calibrated prior to each evaluation according to the manufacturer's specifications. Heart rate was checked at the end of each stage through a frequency meter (Polar, Kempela, Finland).

Snorkel development

The snorkel consisted of one-inch commercial tubes and connections (PVC), connected to a rubber ending (1) joined to the 'elbow' (2) and the saliva container (3) which originally make the VO2000 mask (figure 1). The SNK was based so that the air tube was projected to the front of the participant's face and not to his side, as it is observed in conventional snorkels. This sketch was based on the snorkels elaborated by Toussaint et al.(10) and Dal Monte et al.(7) since they present reduced drag coefficients. The whole system weights 360 grams and has a dead space of 250 ml. The medium flow valve (4) was connected to the upper extremity of the snorkel (figure 1).



Statistical treatment

In order to analyze the data obtained during the incremental tests using the MASK or SNK, an algorithm was developed in Matlab 5.3® environment, specific to this purpose. The data concerning time(s) were interpolated allowing the determination of the ventilatory (VE), oxygen (VO2) and carbonic gas (VCO2) volumes for each stage. These data were assumed as the mean of the values obtained in the final minute of each stage.

The values of the cardiorespiratory parameters (VO2, VE, VCO2 and heart rate) obtained in the incremental test with the MASK and SNK were compared with the t-Student test for dependent samples and the possible associations of these parameters obtained with the MASK and SNK were used. The Pearson correlation test was also used. In all cases the significance level was preset for P < 0.05.

The points obtained of the relationship between the cardiorespiratory parameters measured with the MASK and the SNK were submitted to the linear regression procedure with the purpose to elaborate correction equations for the values derived from the SNK.

The coefficients of determination and residue (res) were analyzed in order to verify the validity of the correction equations. Moreover, the systematic (bias), random (precision) and accuracy (ac) errors were determined.

The res was assumed as the difference between the cardiorespiratory parameters values of the MASK theoretically estimated through regression equations from values derived from the SNK.

The ac was determined by the sum of the bias and precision. In this case the precision was assumed as being the standard deviation of the res while the bias corresponded to the linear coefficients (intercept-y) of each of the regressions (ac2 = precision2 + bias2).

The agreement of the cardiorespiratory parameters values obtained with the MASK and SNK were quantified through the differences between the two observations using Bland and Altman graphic analysis(11). An agreement limit (AL) of 95% was used by the following equation: AL = (1.96 x sd) ± Mdif; where sd = standard deviation and Mdif = mean of differences.



The parameters obtained with the SNK presented high correlations with the ones obtained through the MASK (table 1). However, the VE values determined in the SNK were significantly higher than the ones obtained with the MASK. Except for the VE, the behavior of the remaining respiratory parameters, especially the VO2 measured with the SNK and MASK presented similar behavior (figure 2).





The mean values ± standard deviation of the coefficients of linear and angular determinations of the individual regression equations of VE, VO2, VCO2 and HR are presented in table 2.



When the cardiorespiratory responses relationships are analyzed as a group, one may observe high correlation values (0.88 to 0.97) and determination coefficient (0.77 to 0.94). These results enabled the correction equations for all parameters (VE, VO2, VCO2 and HR) when measured with the SNK (figures 3abcd).

When the distribution of the difference of the values obtained observed. This data behavior shows high agreement between the with the MASK and SNK is analyzed, a random distribution around values from the MASK and SNK (figures 4abcd). zero and majority of the cases within the high and low limits is observed. This data behavior shows high agreement between the values from the MASK and SNK (figures 4abcd).

The values of VE, VO2 and VCO2 for the systematic, random and accuracy errors were similar to the rest values. Moreover, the HR errors were lower than the ones verified at rest.

Specifically analyzing the accuracy values, which is composed by the sum of the systematic and random errors, it was observed that they approximately corresponded to 12, 8, 15 and 6% to the maximal values of VE, VO2, VCO2 and HR respectively, obtained during the incremental tests (table 3).




The main findings of the present study were the high correlations and agreements among the VO2, VCO2, VE and HR values measured with the MASK and SNK during incremental test in cycle ergometer.

Once one believes in the importance of the measurement of cardiorespiratory variables, especially the oxygen consumption in swimming(3,8,12); the validity of the results may be compromised during swimming due to an additional drag, depending on the respiratory valve used(10) and the lack of a specific swimming snorkel for the VO2000 gas analyzer. The present study manually adapted the MASK of the VO2000 to a swimming SNK, using some measurement and analysis procedures similar to the ones used by Keskinen et al.(6). Although the drag coefficient had not been measured, the air tube of the SNK developed in the present investigation was designed in front of the swimmer's face and not on his side, as it is observed in conventional snorkels. Such design was based on the snorkels elaborated by Toussaint et al.(10) and Dal Monte et al.(7) since they present reduced drag values.

The results of all parameters measured were highly correlated (0.88 to 0.99) and except for the VE, they did not present significant differences. Investigations have reported that the VE measured in swimming is usually lower than the ones seen in running(2,4).

In the present study, the VE values determined with the SNK were higher than the ones determined with the MASK, although both tests have been performed in a cycle ergometer. Keskinen et al.(6) when validated a specific snorkel to swimming for the K4 gas analyzer (Cosmed, Italy) also verified that the VE values derived from the SNK were significantly higher than the ones obtained with the MASK. These differences were attributed by the researchers to several factors such as the difference in the respiratory mechanisms due to the use of the SNK, increase in the dead space, and distance of the ending until the turbine as well as differences in resistance in the air flow.

Gyton and Hall(13) reported that the air keep in the dead space (anatomic and physiological) is expired before the alveolar air. Therefore, the additional dead space (250 ml) may have been the responsible for the differences in the VE values since they have made the gas exchanges difficult, which resulted in increases of PCO2 in the blood and consequently stimulated the increases of VE. Toklu et al.(14), observed significant increases in VE, VO2, VCO2, heart rate and tidal volume with an addition of a snorkel with 165 ml of dead space and proposed a new SNK with a low resistance and with a 'two-way' valve (Two-Way). These researchers believe that with this SNK the expired air (containing high values of CO2) is not re-inspired and therefore would not influence in the cardiorespiratory parameters values.

The SNK developed in the present study was manually designed and is simpler than the SNK presented by Toklu et al.(14). However, the correction equations of the respiratory variables (VE, VO2, VCO2 and HR) obtained from the relationship between the MASK and SNK presented reduced values of systematic, random and accuracy errors. Besides the high correlations observed among these results, the Bland and Altman graphic analysis(11) demonstrated high agreement. In addition to these responses, the participants did not report additional discomfort while using the SNK in relation to the MASK.

Although further research is needed in order to quantify the PCO2 and PO2 as well as the existence of additional drag with the use of SNK, one may conclude that the SNK manually developed in the present investigation is an accessory that may be used joined with the VO2000 gas analyzer for measurement of cardiorespiratory parameters valid in swimming.



1. Lavoie JM, Montpetit RR. Applied physiology of swimming. Sports Med. 1986; 3:166-89.        [ Links ]

2. Magel JR, Faulkner JA. Maximum oxygen uptakes of college swimmers. J Appl Physiol. 1967;22:924-8.        [ Links ]

3. Holmér I. Oxygen uptake during swimming in man. J Appl Physiol. 1972;33:502-9.         [ Links ]

4. Holmér I, Lundin A, Eriksson BO. Maximum oxygen uptake during swimming and running by elite swimmers. J Appl Physiol. 1974;36:711-4.         [ Links ]

5. Wakayoshi K, D'Aquisto LJ, Cappaert JM, Troup JP. Relationship between oxygen uptake, stroke rate and swimming velocity in competitive swimming. Int J Sports Med. 1995;16:19-23.        [ Links ]

6. Keskinen KL, Rodríguez FA, Keskinen OP. Respiratory snorkel and valve system for breath-by-breath gas analysis in swimming. Scand J Med Sci Sports. 2003; 13:322-9.        [ Links ]

7. Dal Monte A, Sardella F, Alippi B, Faina M, Mantea A. A new respiratory valve system for measuring oxygen uptake during swimming. Eur J Appl Physiol. 1994; 69:159-62.        [ Links ]

8. Di Prampero PE, Pendergast DR, Wilson DW, Rennie DW. Energetics of swimming in man. J Appl Physiol. 1974;37:1-5.        [ Links ]

9. Toussaint HM, Knops W, Groot G, Hollander AP. The mechanical efficiency of front crawl swimming. Med Sci Sports Exerc. 1990;22:402-8.        [ Links ]

10. Toussaint HM, Meulemans A, Groot G, Hollander AP, Schreurs AW, Vervoorn K. Respiratory valve for oxygen uptake measurements during swimming. Eur J Appl Physiol. 1987;56:363-6.        [ Links ]

11. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307-10.        [ Links ]

12. Pendergast DR, Di Prampero PE, Craig AB, Wilson DR, Rennie DW. Quantitative analysis of the front crawl in men and women. J Appl Physiol. 1977;43:475-9.        [ Links ]

13. Guyton AC, Hall JE. Tratado de fisiologia médica. 9ª ed. Rio de Janeiro: Guanabara Koogan; 1997.        [ Links ]

14. Toklu AS, Kayeserilioglu A, Unal M, Ozer S, Aktas S. Ventilatory and metabolic responses to rebreathing the expired air in the snorkel. Int J Sports Med. 2003; 24:162-5.        [ Links ]



Correspondence to:
Claudio Alexandre Gobatto
Departamento de Educação Física, IB-UNESP
Rio Claro, Avenida 24-A, 1.515
Bela Vista – 13506-900 – Rio Claro, SP

Approved in 29/11/06



All the authors declared there is not any potential conflict of interests regarding this article.
* Apoio Financeiro: CAPES e CNPq (proc. nº 301601/ 2006-2).

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License