Volume 88, Issue 8 p. 948-952
Free Access

Association between aerobic fitness in late pregnancy and duration of labor in nulliparous women


Corresponding Author


Department of Obstetrics and Gynecology, Oslo University Hospital – Rikshospitalet, Oslo, Norway

Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway

: Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, P.O. Box 1046, Blindern, 0316, Oslo, Norway E-mail: [email protected]Search for more papers by this author


Department of Respiratory Medicine, Oslo University Hospital – Rikshospitalet, Oslo, Norway

Search for more papers by this author


Department of Obstetrics and Gynecology, Oslo University Hospital – Rikshospitalet, Oslo, Norway

Search for more papers by this author


Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway

Department of Hematology, Oslo University Hospital – Ullevaal, Oslo, Norway

Search for more papers by this author


Department of Obstetrics and Gynecology, Oslo University Hospital – Rikshospitalet, Oslo, Norway

Search for more papers by this author
First published: 31 December 2010
Citations: 36


It is established that the level of physical activity in the population generally is too low, and data indicate that pregnant women are no exception. Studies of the effects of aerobic fitness on delivery outcomes are limited. In this observational study, we investigated the effect of aerobic fitness on duration of labor in nulliparous women who started labor spontaneously. Maximal oxygen uptake was measured in 40 nulliparous women at 35–37 weeks of gestation. Duration of labor was defined as the time between 3 cm cervical dilation with regular uterine contractions and delivery. The mean (SD) maximal oxygen uptake was 2.1 (0.3) L/min and duration of labor 583 (317) minutes. Duration of labor was inversely associated with maximal oxygen uptake after adjusting for birthweight (p = 0.034). We conclude that measurement of maximal oxygen uptake is safe in pregnancy at 35–37 weeks and that increased aerobic fitness was associated with shorter labor in nulliparous women who started labor spontaneously.


  • BMI
  • body mass index (kg/m2)
  • CO2
  • carbon dioxide
  • O2
  • oxygen
  • CTG
  • cardiotocographic registration
  • RPE
  • ratings of perceived exertion
  • VO2
  • oxygen uptake
  • VO2max
  • maximal oxygen uptake
  • Introduction

    Physical activity constitutes a substantial part of the energy balance, and hence is an important determinant of body weight. A sedentary lifestyle and being overweight increases the risk of several obstetrical complications, stillbirth, and neonatal deaths. Reportedly, physical activity during pregnancy in fit women has no adverse effects on either the mother or the fetus (1, 2). In fact, some studies found reduced risks of pre-eclampsia, gestational diabetes and gestational hypertension (3), and maintenance or improvement of aerobic fitness (4).

    Much less is known about how aerobic fitness affects the course of labor. Prolonged labor is associated with increased risk of fetal asphyxia (5), maternal complications including infections (6), and obstetrical fistulas (7). Physical fitness could influence the course of labor because exercise induces a number of metabolic and hormonal changes that may affect uterine contractility and endurance. Furthermore, during pushing in the second stage of labor, several groups of skeletal muscles are actively involved.

    Previous studies of the relation between duration of labor and aerobic fitness have given inconsistent results. The first stage of labor was significantly longer for a group of heavy-exercising pregnant women compared to a light-exercise group (8). Likewise, we found a tendency toward longer labor in a high-volume compared to a medium-volume exercise group (2). At somewhat variance with this, Bungum et al. (9) reported that the odds ratio for cesarean section delivery in sedentary nulliparous women was four-fold compared to women who had been physically active during the first two trimesters. Rice and Fort (10) found that inactive pregnant women reported higher perceived physical exertion during labor, and had longer second stage of labor than physically active women. Compared with women who stopped regular physical activity during the first trimester, well-trained women who continued their exercise during pregnancy, had (i) fewer complications during labor, (ii) shorter duration of labor, (iii) less frequently needed epidural anesthesia, and (iv) more vaginal deliveries (1). Similar findings were reported by Beckmann et al. (11). One possible explanation of the conflicting results may be that parity was not taken into consideration. Duration of labor is longer in nulliparous women (12).

    Measurement of maximal oxygen uptake is among the most valid estimate of aerobic fitness. This method has also been applied late in pregnancy in a few studies (13, 14). None of these studies had labor outcomes as endpoints. The main aim of the present study was to examine the effect of aerobic fitness in late pregnancy on duration of labor in nulliparous women. To this end, we determined maximal oxygen uptake at gestational week 36 in healthy nulliparous women subjected to a standardized ergometer bicycle test.

    Material and methods

    Fifty-nine healthy nulliparous Caucasian women (20–40 years) with single pregnancies between gestational weeks 35 and 37 were recruited between March 2004 and May 2006 from the Store barn kohorte [STORK] cohort (15) at Oslo University Hospital – Rikshospitalet, Oslo, Norway. Exclusion criteria included fetal pathology, maternal disease (i.e. cardiovascular, rheumatic, lung disease, epilepsia), or pregnancy complications (i.e. gestational diabetes, hypertension, pre-eclampsia). Body mass index (BMI) was calculated from maternal weight and height recorded during weeks 14—16, whereas the weight at weeks 35–37 was recorded on the test day. Approval was obtained from The Norwegian Medical Ethics Committee for Health Region South (no. S-03273) and Norwegian Social Science Data Service (no. 10594). All women gave written informed consent to participate.

    All pulmonary measurements were carried out on a Vmax Pulmonary Function Unit (VIASYS Respiratory Care Inc., Yorba Linda, CA). The calibration of the flow-meter was performed daily with a three liter syringe. The gas analyzer was calibrated for oxygen and carbon dioxide by two gas mixtures containing 26% O2 and 0% CO2, and 16% O2 and 4% CO2. For the exercise test, the Vmax unit was coupled to an electrically braked ergometer bicycle (Ergoline 800, Ergoline GmbH, Bitz, Germany). Maternal heart rate was measured by a three-leaded ECG unit (for Vmax, VIASYS) and peripheral oxygen saturation by a pulse oximeter (Nonin Medical Inc., MN). Fetal heart rate was monitored with a cardiotocographic (CTG) registration unit (Sonicaid TEAM Cardiotocograph System, Oxford Instruments Medical, Oxon, England).

    We used a symptom-limited breath-by-breath technique. Aerobic fitness was based on the reference values from Cooper (16), since it has been shown that pregnancy does not effect aerobic fitness (13, 14). Testing consisted of three phases: (1) a two-minute warm-up with 20 Watts work load, (2) an exercise phase varying between 8 and 12 min, and (3) a two-minute recovery phase. During the exercise phase, the work load was ramped up so that increase in work load per minute was 10% of the predicted maximal load. Before testing, the women were familiarized with Borg's scale for ratings of perceived exertion (RPE) (17), and instructed to give their RPE at two-minute intervals.

    Testing continued until limiting symptom like dyspnoea, general fatigue, or leg fatigue occurred. By this method we obtained peak VO2 calculated as the highest average of 20 seconds at the end of the exercise. During testing, we continuously recorded heart rate, work load, and oxygen uptake (VO2). Oxygen uptake per kilogram was derived from the measured variables. Prior to, and immediately after the exercise test, a 20-minute CTG was performed. Of 59 post-exercise CTG measurements, one fetus showed a transient tachycardia of 170 beats per min over 20 minutes, where after the pattern normalized. All other CTG measurements were normal before and after exercise.

    The first stage of labor was defined as the time between 3 cm cervical dilation with regular uterine contractions until complete dilation (10 cm). The second stage of labor was defined as the time between fully dilated cervix and delivery. The term ‘duration of labor’ is used for stages 1 and 2 combined.

    The data were normally distributed, and we calculated mean and SD. Potential determinants of duration of labor, identified by Pearson correlation tests, were tested in a multivariate model. Those with p-values < 0.25 were included in a multiple linear regression analysis. Backward multiple linear regressions with removal probability set to 0.10 was used for final selection of independent variables in the analysis. Final determinants were: (i) epidural analgesia during labor, (ii) stimulation during labor, (iii) birthweight, and (iv) maternal age and height. Maternal weight was not among the final determinants (p>0.25). A significance level of 5% was chosen. The criteria for performing regression analysis were fulfilled. SPSS version 14.02 (SPSS Inc. Headquarters, Chicago, IL, USA) was used for all analysis.


    Nineteen of the 59 women did not start labor spontaneously (induced or elective cesarean section) and were according to protocol excluded from the analysis. The mean (SD) maternal age, height, weight at gestational weeks 14–16 and 35–37, and BMI at gestational weeks 14–16 were 30.3 (3.9) years, 169 (5.0) cm, 67.1 (8.0) kg, 76.5 (8.6) kg, and 23.5 (2.4) kg/m2, respectively. They did not differ from the remaining ‘STORK’ participants (15) with respect to maternal age, BMI, physical fitness, educational level, marital status, parity, gestational age at delivery, or birthweight. There were no breech deliveries, but one emergency cesarean section and 12 instrumental deliveries (forceps/vacuum). Thirteen women needed epidural analgesia, whereas 23 women were stimulated with oxytocin and 15 by amniotomy during labor. Epidural analgesia, stimulation with oxytocin/amniotomy or mode of delivery (vaginal/emergency cesarean section) did not effect VO2max (p>0.05). Maternal weight was not related to duration of labor (p>0.25).

    Table I shows exercise and labor-related variables and birthweights. Univariate analysis showed no association between VO2max and duration of labor (data not shown). Table II shows the results of the multiple regression analysis in two models, one in which any stimulation of labor was excluded (Model 1) and one in which it was included (Model 2). Both models showed a significant association between VO2max and duration of labor (p = 0.034 and 0.047, respectively). This indicates that an increase in maximal oxygen uptake of 0.1 L/min (corresponding to 5% increase), would reduce the duration of labor with on average 33 min (Model 1) and 28 min (Model 2).

    Table I. Exercise and labor-related variables and birthweights.
    n Mean SD
    Maximal VO2 (L/min) 40 2.1 0.3
    Maximal VO2 (mL/kg·min) 40 27.8 4.4
    Maximal work load (Watt) 40 171.3 29.3
    Maximal heart rate (beats/min) 38 174 8.5
    RPE (Borg's scale) at the end of the test 40 8.6 1.2
    Gestational age at delivery (weeks) 39 40.3 1.0
    Duration of labor (minutes) 39 583 317
    Stage 1 of labor (minutes) 34 569 309
    Stage 2 of labor (minutes) 34 45 29
    Birthweight (g) 40 3,516 437
    • Note: Since complete datasets were not available, n for individual variables is in some cases less than 40. RPE: Borg's scale for ratings of perceived exertion.
    Table II. Factors affecting duration of labor.
    Linear regression models R2 B 95% CI for B n p
    Duration of labor
     Model 1: maximal VO2 (L/min) 0.26 −326.8 −627.2 to −26.4 39 0.034
     Birthweight (g) 0.3 0.1 − 0.6 39 0.003
     Model 2: maximal VO2 (L/min) 0.43 −283.0 −542.8 to −3.3 39 0.047
     Birthweight (g) 0.3 0.1 − 0.5 39 0.006
     Any stimulation during labor 274.8 100.0 − 449.6 39 0.003
    Duration of first stage
     Model 1: birthweight (g) 0.15 0.3 0.0 − 0.5 34 0.026
     Model 2: birthweight (g) 0.41 0.2 0.0 − 0.4 34 0.050
     Any stimulation during labor 322.9 141.9 − 503.9 34 0.001
    Duration of second stage
     Birthweight (g) 0.27 0.03 0.01 − 0.05 34 0.010
     Maternal height (cm) −1.9 −3.6 − 0.1 34 0.039
    • Note: Association between aerobic fitness (VO2max) and duration of labor, first, and second stage estimated from multiple linear regression models adjusted for birthweight, stimulation during labor and maternal height. R2: amount of variation in the data that is explained by the regression model. B: regression coefficient (min).

    The duration of the first stage of labor was positively associated with birthweight (p = 0.026 and 0.050, respectively, for Models 1 and 2). This indicates that an increase in birthweight by 100 g would prolong the first stage of labor by on average 30 and 20 minutes, respectively. Furthermore, the duration of the second stage of labor was positively associated with birthweight (p = 0.010), but was inversely related to maternal height (p = 0.039). In other words, a 1-cm difference in maternal height will give a 1.9-minute difference in duration of the second stage of labor.


    The present study indicates that nulliparous women with higher VO2max have shorter duration of labor than nulliparous women with lower aerobic fitness. Our VO2max values were similar to previously reported maximal oxygen uptake values (13, 14). Interestingly, maternal weight was not a confounder in the present study.

    The main strength of this study is that a standardized measure of maximal oxygen uptake was obtained close to labor in a well-defined cohort of healthy, nulliparous women. Moreover, such an exhausting test did not have any adverse effect either on the mother, the fetus, or labor outcomes. Duration of labor differs considerably between nulliparous and multiparous women (12). The present study was limited to nulliparous women.

    Earlier studies have not taken parity into consideration (1, 10). Nevertheless, Clapp (1) found that his well-trained group had a significantly shorter first stage of labor, but also gave birth to significantly lighter babies (difference of about 400 g). Beckmann and Beckmann (11) found that nulliparous women who exercised regularly had significantly shorter first and second stages of labor than did non-exercising nulliparous.

    We found effect of aerobic fitness on total duration of labor, but not on first and second stage separately. This implies an effect of aerobic fitness on first stage of labor, suggesting that physical fitness in some way may affect the work of the uterus muscles. Although this seems apparently inconceivable, there are studies indicating that metabolic changes associated with overweight may affect the contractile activity of the uterine muscles (18).

    Intensity, duration, frequency, and type of physical activity may cause different effects on pregnant women. Controlled exercise in regard to these four qualities should therefore be assessed in a larger population of nulliparous and multiparous women to determine the general validity of the aerobic fitness in relation to duration of labor and labor outcomes.

    Physical activity during pregnancy appears to be beneficial for maternal health (1). Our study confirms that measurement of aerobic fitness by maximal oxygen uptake is safe, and aerobic fitness in healthy nulliparous women has a positive effect on duration of labor since shorter labor is generally associated with fewer complications.


    The study was funded partly by The Throne Holst Foundation, The Medical Faculty of the University of Oslo, The Norwegian Women's Public Health Association, and The South-East Norway Regional Health Authority.

    Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.