Journal Logo

CLINICAL SCIENCES: Clinically Relevant

V̇O2peak Prediction and Exercise Prescription for Pregnant Women

MOTTOLA, MICHELLE F.1,2; DAVENPORT, MARGIE H.1; BRUN, CHANTALE R.1; INGLIS, STUART D.1; CHARLESWORTH, SARAH1; SOPPER, MAGGIE M.1

Author Information
Medicine & Science in Sports & Exercise: August 2006 - Volume 38 - Issue 8 - p 1389-1395
doi: 10.1249/01.mss.0000228940.09411.9c
  • Free

Abstract

The latest medical advice from the American College of Obstetricians and Gynecologists (ACOG) states that a woman with a low-risk pregnancy can participate in moderate exercise for ≥ 30 min·d−1 on most, if not all, days of the week (1). Although this advice promotes exercise during pregnancy in the United States, it does not give women concrete target HR guidelines for exercise.

In Canada, target HR zones and guidelines for exercise during pregnancy are found in the PARmed-X for Pregnancy document (21). This document was recently endorsed by the Society of Obstetricians and Gynecologists of Canada (SOGC) (9) and the Canadian Society for Exercise Physiology (CSEP) (10) in a SOGC/CSEP joint position paper, which is available on the CSEP Web site (www.csep.ca). The American College of Sports Medicine (ACSM) recently endorsed the SOGC/CSEP joint position statement (9,10) in which the PARmed-X for Pregnancy document is highlighted (3).

The validated target HR zones for exercise prescription (20) found in the PARmed-X for Pregnancy document (21) were modified from target HR zones suggested for nonpregnant individuals (2) for two reasons: 1) maximum HR during pregnancy is attenuated during maximal exercise testing, resulting in a significant decrease in maximum HR reserve (19); and 2) resting HR increases during the first trimester, followed by further increases as pregnancy advances to about 15-20 bpm over nonpregnant values (18). These pregnancy alterations are the result of sympathoadrenal modulation and a reduced parasympathetic/vagal response (4), respectively. Although the target HR zones found in the PARmed-X for Pregnancy represent approximately 60-80% of aerobic capacity based on age for the average pregnant woman (18), we have found that women who are on either end of the exercise continuum-either overweight, with a body mass index (BMI) ≥ 25 kg·m−2, or very fit-the current target HR zones may not be appropriate (Mottola et al., unpublished observations, 2005).

The present study was designed to 1) use a progressive treadmill test to develop a prediction equation for peak oxygen consumption (V̇O2peak), 2) validate the prediction equation, and 3) prescribe and refine the current target HR exercise guidelines for pregnancy (PARmed-X for Pregnancy) (21) that are based on age alone by also including fitness levels.

METHODS

One hundred fifty-six women between 16 and 22 wk of gestation with medical clearance from their healthcare provider (PARmed-X for Pregnancy) (21) performed a progressive treadmill test modified from the Balke protocol (5) to volitional fatigue (peak exercise test). We used a modified Balke protocol because of the 2-min stages that allow pregnant women time to adjust to each new work rate. Sixteen to 22 wk of gestation were chosen to represent the most appropriate gestational age for testing. By this time, the symptoms of pregnancy (nausea, vomiting, fatigue, etc.) are usually minimal, and women have had medical approval to exercise. All women had low-risk pregnancies with no contraindications to exercise. Written informed consent was obtained from each participant. The human research ethics board for health sciences at the University of Western Ontario approved the protocol.

To ensure adequate maternal blood glucose for the duration of the test, 1 h prior to the start of the exercise session, each participant ingested a standard meal (one pouch (38-g serving) of Carnation Instant Breakfast™ mixed with 250 mL of milk; 248 kcal, 14.2 g of protein, 3.8 g of fat, and 39.3 g of carbohydrate). Height and weight were measured and recorded to the nearest centimeter and kilogram, respectively. Room temperature was maintained at 20 ± 3°C with 55% humidity.

Prior to each exercise test, the SensorMedics (Yorba Linda, CA) V̇Omax 29c breath-by-breath gas-analysis unit was calibrated according to company instructions, using two tanks of calibration gases (tank 1 = 4% carbon dioxide and 16% oxygen; tank 2 = 26% oxygen, 0% carbon dioxide; nitrogen balanced; SensorMedics), with an accuracy of 0.100 for the oxygen and carbon dioxide analyzer. The flow sensor meter was calibrated to a 3-L calibrator syringe (SensorMedics). The acceptable range for calibration was within ± 2% variability.

Preexercise respiratory gases were collected for 5 min while the subject stood quietly on the treadmill with continuous monitoring of oxygen consumption. HR was recorded via four ECG leads (SensorMedics, V̇Omax 29c). The test began with a 5-min warm-up at 3 mph (4.8 km·h−1), 0% grade. This is generally considered a normal walking pace (11). During the test, the treadmill speed was held constant at this pace (3 mph), with the incline increased every 2 min by 2% until volitional fatigue. If fatigue was not reached by 12%, then the speed was increased slightly (by 0.2 mph (0.3 km·h−1) at each stage) until volitional fatigue was reached. At the start of each stage (within 30 s), the subject rated her perceived exertion using the Borg scale (7). Once volitional fatigue was reached (Borg rating = 9 or 10; maximal on a 10-point scale; or 19-20 on the 20-point scale) (7), a 5-min cool-down at 3 mph and 0% grade was immediately initiated. Following the cool-down, the subject stood on the treadmill while recovery respiratory gases were collected for five additional minutes. During the test, subjects were instructed to lightly grip the hand rail at all times (6). Peak V̇O2 was determined from the average of the last 30 s of the breath-by-breath analyses recorded by the computer software, once volitional fatigue was reached (14).

The 156 women were separated into two groups: one to develop the equation (N = 117) and a second to cross-validate the equation (N = 39). Every fourth subject was removed from the subject pool to form a cross-validation group (16).

The 156 women were separated into two age groups: 20-29 (N = 60) and 30-39 (N = 96) yr of age. Each age group was further separated into fit, active, and unfit. Fit women were defined as having a V̇O2peak ≥ 75th percentile for their age group (13). Unfit women were defined as having a V̇O2peak ≤ 25th percentile for their age group, and active women were between these two ranges. A linear regression was performed between peak HR and V̇O2peak (15) for each of the six groups. Because the Canadian guidelines (21) suggest that pregnant women should exercise between 60 and 80% of their aerobic capacity, these V̇O2 values were determined for the two age groups (20-29 yr and 30-39 yr) within each of the three fitness levels (fit, active, unfit) and used in the regression equation to predict target HR ranges at 60 and 80% V̇O2peak.

Statistical analysis.

Statistical analysis included subject characteristics (mean ± SD) and Pearson product-moment correlations to measure the relationship between variables. Correlations (R2) were adjusted for the degrees of freedom in the model. A nonparametric Mann-Whitney U-test and an independent samples t-test were used to determine significant differences between predicted and measured V̇O2peak. A multivariate linear regression was used to develop the V̇O2peak prediction equation and the HR equations. Significance was accepted at P ≤ 0.05. All analyses were performed using SPSS software (version 13).

Correlations between peak V̇O2 and weight, height, BMI (at time of test), age, gestational age, distance at peak, time to peak, speed at peak, peak incline, and HR were assessed for the equation group (N = 117). Correlations were also performed between weight, height, BMI, age, gestational age, distance at peak (distance covered during test to volitional fatigue), time to peak, speed at peak, peak incline, and HR to determine whether any of the variables were linearly related. Multivariate linear regression was performed using variables that were significantly correlated with peak V̇O2 only. The derived equation was used in the validation group to estimate peak V̇O2.

RESULTS

Subject characteristics of the equation-generated and the cross-validation groups are summarized in Table 1. No significant differences were found in the variables measured. The exercise parameters (distance at peak, time to peak, speed at peak, peak incline, HR, and V̇O2peak) were all positively correlated (P < 0.01; Table 2). BMI and body weight were positively correlated with each other (P < 0.01) and negatively correlated with the exercise parameters measured (Table 2). Based on these results, multivariate linear regression was found to have the best results in predicting V̇O2peak with BMI, speed at peak, peak incline, and HR. Thus, the following equation was developed to predict V̇O2peak: V̇O2peak (predicted) = (0.055 × peak HR) + (0.381 × incline) + (5.541 × speed) (mph) + (−0.090 × BMI) − 6.846, where peak HR is in bpm, incline is the percent, and BMI (kg·m−2) is calculated at the time of the test. Analysis of this equation found that R2 = 0.72, R2adjusted = 0.71, and SEE = 2.7. When this equation was used to predict V̇O2peak in the cross-validation group (N = 39), the P value was 0.78, actual value was 23.54 ± 5.9, and predicted value was 23.38 ± 4.03 mL·kg−1·min−1. Women were considered fit if they had a V̇O2peak ≥ 27.2 mL·kg−1·min−1 and ≥ 26.1 mL·kg−1·min−1 for ages 20-29 and 30-39 yr, respectively, representing the 75th percentile. Unfit women had a V̇O2peak of ≤ 21.0 mL·kg−1·min−1 and ≤ 19.6 mL·kg−1·min−1, respectively, representing the bottom 25th percentile. Table 3 compares the characteristics of women aged 20-29 yr who were fit, active, and unfit. Unfit women in this age range had a higher BMI (29.8 ± 1.2 kg·m−2) than fit women (22.9 ± 0.7 kg·m−2) but not different than the active women (26.3 ± 1.1 kg·m−2). The BMI of the fit women was not different from that of the active women. The unfit women in this age group had lower average peak HR (159.7 ± 3.9 bpm) and V̇O2 (18.3 ± 0.6 mL·kg−1·min−1) values than both the active (171.9 ± 1.9 bpm; 23.8 ± 0.3 mL·kg−1·min−1) and fit women (176.3 ± 2.1 bpm; 31.3 ± 1.0 mL·kg−1·min−1; P < 0.05). Although the peak HR in the active women was not different from that of the fit women, peak V̇O2 in the fit women was higher (P < 0.05). Figure 1 shows the regression lines and target HR zones for the fit and unfit women in this age group (20-29 yr). Table 4 compares the characteristics of women aged 30-39 yr. In this age range, the unfit (BMI = 31.6 ± 1.0 kg·m−2) and active (BMI = 29.2 ± 0.8 kg·m−2) women had higher BMI values than the fit women (BMI = 24.8 ± 0.7 kg·m−2; P < 0.05), with no difference between the unfit and active women (P >0.05). Average peak V̇O2 values for the fit (30.7 ± 0.9 mL·kg−1·min−1) and active (22.4 ± 0.3 mL·kg−1·min−1) women were higher than those for the unfit women (18.1 ± 0.2 mL·kg−1·min−1; P < 0.05). In addition, the fit women had higher peak V̇O2 values than the active women (P < 0.05), although peak HR differed only between the unfit (160.3 ± 2.3 bpm) and the fit women (175.1 ± 1.6 bpm; P < 0.05). Figure 2 shows the regression lines and target HR zones for the fit and unfit women in this age group. No differences were found between age groups within fitness levels (unfit, active, and fit) for BMI, peak HR, and V̇O2 (P >0.05). These similarities between age groups are reflected in the target HR zones for the unfit women; however, in the active and fit women, the target HR zones are more distinct between age groups.

T1-4
TABLE 1:
Comparison of subject characteristics in the equation and cross-validation groups.
T2-4
TABLE 2:
Correlations between variables measured for 156 pregnant women.
T3-4
TABLE 3:
Characteristics of fit, active, and unfit women aged 20-29 yr (N = 60; mean ± SEM)
F1-4
FIGURE 1:
Linear regression lines of HR and V̇O2peak for fit and unfit pregnant women aged 20-29 yr with target HR zones between 60 and 80% V̇O2peak. For fit women, HR = 2.372 × V̇O2 + 100.169 (HR range = 145-160 bpm). For active women, HR = 4.146 × V̇O2 + 73.152 (HR range = 132-152 bpm, not shown on graph). For unfit women, HR = 3.933 × V̇O2 + 86.088 (HR range = 129-144 bpm).
T4-4
TABLE 4:
Characteristics of fit, active, and unfit women aged 30-39 (N = 96; mean ± SEM).
F2-4
FIGURE 2:
Linear regression lines of HR and V̇O2peak for fit and unfit pregnant women aged 30-39 yr with target HR zones between 60 and 80% V̇O2peak. For fit women, HR = 2.557 × V̇O2 + 93.194 (HR range = 140-156 bpm). For active women, HR = 4.078 × V̇O2 + 74.428 (HR range = 129-148 bpm, not shown on graph), and for unfit women, HR = 4.452 × V̇O2 + 79.310 (HR range = 128-144 bpm).

DISCUSSION

This is the first study to provide a validated prediction equation of V̇O2peak for pregnant women between 16 and 22 wk of gestation. Fitness professionals who do not have access to a metabolic cart can use this prediction equation to estimate peak aerobic capacity in this population of healthy pregnant women who have been medically prescreened (21), including BMI values at 16-22 wk of pregnancy. The defined target HR zones based on age and fitness levels can be used for exercise prescription as we have further refined the target HR zones from the PARmed-X for Pregnancy document (21).

The fit women in our cohort have aerobic capacities similar to those described by Lotgering et al. (14), who reported an average value of 36 mL·kg−1·min−1 V̇O2max (calculated) at 16 wk of gestation and an average maximum HR of 180 ± 2 bpm using a treadmill test. These authors defined maximum aerobic power as the presence of two of the three following criteria: 1) oxygen consumption increase < 5% in response to an exercise intensity increase, 2) HR increase < 5% in response to an exercise intensity increase, and 3) RER > 1 (14). The slightly lower values for our women may be due to our peak versus their maximal testing protocol and the larger range in gestational age for our women (16-22 wk). When fitness levels, previous activity of the subjects, or gestational ages are not controlled, V̇O2max values can range from 20.2 to 39.1 mL·kg−1·min−1, and maximum HR can range from 167 to 197 bpm during cycle ergometry, between 20 and 34 wk of gestation (16). Maximal oxygen consumption in the cycle ergometer study was defined as the highest V̇O2 reached at volitional fatigue (16), which we have defined as peak oxygen consumption in our study.

Top-level athletes of national and international caliber at 15-19 wk of gestation had a V̇O2max range of 27.4-48.8 mL·kg−1·min−1 in a medium-volume exercise group (12) and 38.5-52.6 mL·kg−1·min−1 in a high-volume exercise group tested on a cycle ergometer (12). The maximum HR ranged from 179 ± 9 to 181 ± 9 bpm, respectively (12). The fit women in the present study were in the top 25th percentile of our cohort, but none were elite athletes. The women above our cutoff points of ≥ 27.2 mL·kg−1·min−1 and ≥ 26.1 mL·kg−1·min−1 for our younger and older age groups, respectively, are within the ranges reported for fit women (12,14,16), although the mode of assessing aerobic capacity differed between studies.

The target HR range based on age from the PARmed-X for Pregnancy document (21) suggests that between the ages of 20 and 29 yr, low-risk, medically prescreened women can safely exercise at 135-150 bpm, reflecting 60-80% of aerobic capacity (18). Data from our cohort of active pregnant women (132-152 bpm) agree with the Canadian guidelines for this age group. For fit pregnant women, these target HR guidelines may not reflect 60-80% of maximum aerobic capacity. This is apparent from the target HR zones from our women in the top 25th percentile, who present a target HR zone of 145-160 bpm, which represents 60-80% of peak aerobic capacity for this cohort. To confirm appropriate intensity, these exercise prescription HR should be coupled with the "talk test" (enabling a pregnant women to carry on a conversation without being out of breath) and the RPE scale suggested in the PARmed-X for Pregnancy document for monitoring intensity, "somewhat hard" (12-14 on the 20-point scale, or 3-4 on the 10-point scale) (21).

On the other end of the continuum, which includes those women who are in the bottom 25th percentile of our cohort, the target HR zones suggested from the Canadian guidelines (PARmed-X for Pregnancy) (21) may be inappropriate. These women also have higher BMI values and a lower aerobic capacity for exercise. The target HR zone for the unfit women in our cohort at 20-29 yr of age (129-144 bpm) started below the zone suggested in the PARmed-X for Pregnancy document (135-150 bpm) (21). If the suggested target HR zone from the present study was used in conjunction with the "talk test" and the RPE (12-14, somewhat hard on the 20-point Borg scale, or 3-4 on the 10-point scale) as further guides for intensity, the target HR zone (129-144 bpm) generated from the present study may be better suited for this group of unfit women. Again, by using these tools, the intensity of exercise is individualized to meet the needs of this special group of women.

Santos et al. (17) examined aerobic exercise and submaximal functional capacity in overweight pregnant women. They determined that oxygen uptake at the anaerobic threshold (AT) for women aged 27 yr with a BMI of approximately 28 kg·m−2 at 18 wk of gestation was, on average, 16 mL·kg−1·min−1, with a HR at AT of 144 bpm. These data are similar to those for our unfit women in the 20- to 29-yr cohort who are in the bottom 25th percentile (AT data not shown).

Similarly, in our older group of women, division by fitness levels may give more appropriate target HR zones for the women who have a lower aerobic capacity (the unfit women) with a higher BMI. On the other end of the continuum, the women with the higher aerobic capacity who are more fit may also benefit from the adjusted target HR. The target HR zone suggested for women aged 30-39 yr is 130-145 bpm (PARmed-X for Pregnancy) (21). Our cohort of active women from this age group produced a HR target zone of 129-148 bpm, which again is similar to the Canadian guidelines (21). The target HR zone for the unfit women in our cohort was 128-144 bpm, representing HR at 60-80% of aerobic capacity, which also fits within the Canadian guidelines. This may be because the BMI in the unfit group was not different from the BMI of the active women in this age group. Conversely, data from the fit women in this age group produced a target HR zone of 140-156 bpm, which places the range above the Canadian guidelines and may be more appropriate for women with higher aerobic capacities. Again, the "talk test" and RPE scale should be used as additional guides for intensity prescription for all the women in this age group in order to individualize exercise prescription and confirm intensity.

In the PARmed-X for Pregnancy document, guidelines for aerobic activity include advice on frequency, intensity (already discussed), time, and type of activity (21). It is suggested that the new target HR zones reported in the present study for fit and unfit pregnant women in the 20- to 29-yr and 30- to 39-yr age groups be used in conjunction with the other guidelines suggested by the PARmed-X for Pregnancy (21). This would include using the "talk test" with the RPE scale to individualize the exercise prescription and confirm the exercise intensity.

Regarding structured exercise frequency, Campbell and Mottola (8) suggested that women who engaged in structured exercise ≥ 5× wk−1 in the third trimester were 4.6 times more likely to give birth to a low-birth weight baby. In addition, they found that those women who engaged in structured exercise ≤ 2× wk−1 in late pregnancy were 2.7 times more likely to give birth to a low-birth weight baby (8). In this case-control design study of 529 women, frequency of structured exercise during late pregnancy was found to be more important as a determinant of birth weight than intensity, and thus women are cautioned about engaging consistently in structured exercise ≥ 5 or ≤ 2× wk−1 during the third trimester (8). Although the ACOG (1) suggest that pregnant women should exercise on all if not most days of the week, we recommend that those women who are more likely to engage in structured exercise ≥ 5× wk−1 decrease the frequency of activity to 3-4× wk−1, especially in the last trimester. In addition, it is important for all medically prescreened pregnant women to consistently exercise at least 3× wk−1 for the greatest health benefits.

In conclusion, this is the first study to provide a validated prediction equation of V̇O2peak for pregnant women between 16 and 22 wk of gestation using a progressive treadmill exercise test. The defined target HR zones based on age and fitness level can be used for exercise prescription in healthy pregnant women who have been medically prescreened. We suggest that fit pregnant women between the ages of 20 and 29 yr who wish to exercise at 60-80% of aerobic capacity should work at a target HR of 145-160 bpm, and in the 30-39 yr age group, target HR should be between 140 and 156 bpm. We also suggest that healthy women with lower fitness levels who are medically prescreened can exercise at target HR of 129-144 bpm if they are between the ages of 20 and 29 yr, and 128-144 bpm if they are 30-39 yr old. Target HR zones for healthy active pregnant women are confirmed in the PARmed-X for Pregnancy document (21). It is also recommended that the PARmed-X for Pregnancy be used for medical prescreening in conjunction with the new target HR zones suggested in the present study for aerobic exercise guidelines of frequency, intensity, time, and type of activity.

The authors thank the following funding sources: Molly Towell Perinatal Research Foundation, Canadian Forces Personnel Support Agency, and the Canadian Institute of Health Research-Institute of Aboriginal People's Health.

REFERENCES

1. American College of Obstetricians and Gynecologists (ACOG). Obstet. Gynecol. 99:171-173, 2002.
2. American College of Sports Medicine. Guidelines for Exercise Testing and Prescription, 6th ed., Philadelphia: Lippincott Williams & Wilkins, 2000, pp. 180-183.
3. American College of Sports Medicine. Sports Medicine Bulletin 39, 2004.
4. Avery, N., L. A. Wolfe, C. Amara, G. Davies, and M. J. McGrath. Effects of human pregnancy on cardiac autonomic function above and below the ventilatory threshold. J. Appl. Physiol. 90:321-328, 2001.
5. Balke, B., and R. W. Ware. An experimental study of physical fitness of Air Force personnel. US Armed Forces Med. J. 10:875-888, 1959.
6. Berry, M. J., P. Brubaker, M. O'Toole, et al. Estimation of V̇O2 in older individuals with osteoarthritis of the knee and cardiovascular disease. Med. Sci. Sports Exerc. 28:808-814, 1996.
7. Borg, G. Psychophysical bases of perceived exertion. Med. Sci. Sports Exerc. 14:377-381, 1982.
8. Campbell, M. K., and M. F. Mottola. Recreational exercise and occupational activity during pregnancy and birth weight: a case-control study. Am. J. Obstet. Gynecol. 184:403-408, 2001.
9. Davies, G., L. A. Wolfe, M. F. Mottola, and C. MacKinnon. Exercise in pregnancy and the postpartum period. J. Obstet. Gynaecol Can. 25:516-522, 2003.
10. Davies, G., L. A. Wolfe, M. F. Mottola, and C. MacKinnon. Joint SOGC/CSEP clinical practice guideline: exercise in pregnancy and the postpartum period. Can. J. Appl. Physiol. 28:329-341, 2003.
11. Hendelman, D., K. Miller, C. Baggett, E. Debold, and P. Freedson. Validity of accelerometry for the assessment of moderate intensity physical activity in the field. Med. Sci. Sports Exerc. 32(9 Suppl):S442-S449, 2000.
12. Kardel, K. Effects of intense training during and after pregnancy in top-level athletes. Scand. J. Med. Sci. Sports 15:79-86, 2005.
13. Lamonte, M., L. Durstine, F. Yanowitz, T. Lim, K. DuBose, P. Davis, and B. Ainsworth. Cardiorespiratory fitness and C-reactive protein among a tri-ethnic sample of women. Circulation 106:403-406, 2002.
14. Lotgering, F. K., M. VanDoorn, P. Struijk,.et al. Maximal aerobic exercise in pregnant women: heart rate, O2 consumption, CO2 production and ventilation. J. Appl. Physiol. 70:1016-1023, 1991.
15. Pivarnik, J. M., A. Stein, and J. Rivera. Effect of pregnancy on heart rate/oxygen consumption calibration curves. Med. Sci. Sports Exerc. 34:750-755, 2002.
16. Sady, S., M. Carpenter, M. Sady, et al. Prediction of V̇O2max during cycle exercise in pregnant women. J. Appl. Physiol. 65:657-661, 1988.
17. Santos, I., R. Stein, S. Fuchs, et al. Aerobic exercise and submaximal functional capacity in overweight pregnant women. Obstet. Gynecol. 106:243-249, 2005.
18. Wolfe, L. A., and G. Davies. Canadian guidelines for exercise in pregnancy. Clin. Obstet. Gynecol. 46:488-495, 2003.
19. Wolfe, L. A., and M. F. Mottola. Aerobic exercise in pregnancy: an update. Can. J. Appl. Physiol. 18:119-147, 1993.
20. Wolfe, L. A., and M. F. Mottola. Validation of guidelines for aerobic exercise in pregnancy. In: Decision Making and Outcomes in Sports Rehabilitation, D. Kumbhare and J. Basmajian (Eds.). New York: Churchill Livingstone, 2000, pp. 205-217.
21. Wolfe, L. A., and M. F. Mottola. PARmed-X for Pregnancy, Ottawa: Canadian Society for Exercise Physiology, pp. 1-4, 2002.
Keywords:

EXERCISE; PREGNANCY; GUIDELINES; TARGET HR

©2006The American College of Sports Medicine