Oxygen Consumption and Ventilation during Normal Labor* LTC Arn H. Eliasson, M.D.,F.C.C.P.,; LTC Yancy Y Phillips, M.D., F.C.C.P.; MAl Karl C. Stajduhar, M.D.; MAJ Michael A. Carome, M.D.; and LTC John D. Cowsar, fr., D.Q

Oxygen consumption (V02 ) and minute ventilation (VE) were measured breath-by-breath for 10-min periods in the third trimester of pregnancy in 16 healthy women. These measurements were repeated during the first stage of labor in eight of the women. The 10-min mean VOl was 3.56 mlJkgl min (±0.82 SD) at term and 4.28 ml/kglmin (±0.93) during labor, for an average increase of 23 percent ( ± 28 percent, p=O.04) from third trimester to labor. The mean VE was 0.15 Ukglmin (±0.03) at term and increased significantly (p=0.05) to 0.24 Ukglmin (±0.11) during labor for an average increase in VE of 65 percent (± 78 percent). Peak V02 and VE occurred during contractions with five-breath

obstetric patients have significant acute or M anychronic illnesses that dramatically reduce their

cardiopulmonary reserve. This may result from the illness itselp·2 or from deconditioning due to a limitation on physical activity by the illness. In this clinical setting, the most important question is whether the mother can meet the respiratory requirements oflabor and delivery. Delivery by cesarean section may be a prudent alternative in cases where the work imposed by labor and delivery is more than the patient can be expected to perform. A knowledge of the energy and ventilatory requirements of normal labor and delivery would be useful in the evaluation of high-risk pregnant patients. Measurements of the work performed by normal subjects could be used to estimate the ventilatory and metabolic capacity required of the high-risk patient in labor. Ideally these data would provide the clinician with objective information to help in the decision about whether to attempt a trial of labor in patients with significant cardiopulmonary limitations. There are few reliable data available that characterize the energy and ventilatory requirements of labor and delivery in healthy or compromised patients. *From the Departments of Medicine and Clinical Investi~ation, Walter Reed Army Medical Center, \Vashin~ton, DC; the Department of Family Practice, De"ritt Army Community Hospital, Fort Belvoir, Va; and the Department of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Md. The opinions or assertions contained herein are the private views of the authors and are not to be constnled as official or as reflecting the views of the Department of the Army or the Department of Defense. Manuscript received August 12; revision accepted November 19. Reprint requests: Dr. Eliasson, Walter Reed Anny ltledical Center, Washington, DC 20307-5001

average peak VOl being 86 percent (± 53%) above the 10min mean value at term and VE increasing 167 percent ( ± 154 percent) from third trimester to peak values during labor. These data may be useful in identifying patients at risk for developing respiratory insufficiency during labor. We propose an algorithm for approaching the obstetric patient with respiratory disease. (Chest 1992; 102:467-71)

FEV1 = forced expired volume in 1 s; FVC ~ forced vital capacity; MVV = maximal voluntary ventilation; VE = minute ventilation; VOl = oxygen consumption

Several studies have shown an increase in oxygen consumption (V02) and minute ventilation (VE) which is progressive during pregnancy.3-li Other studies have examined arterial oxygen saturation and carbon dioxide content during labor.9-11 However, this research has focused on either the standardization of transcutaneous measurement devices9 or the effect of pain on oxygen saturation and lactate production. 10.11 Two studies have measured V02 during brief portions of labor. Hagerdal and coworkers l2 reported a significant increase in V02 and VE with contractions as compared with resting between contractions. They also demonstrated that effective pain relief abolished this difference, reducing V02 by 30 percent during contractions. Sangoul and associates l3 reported similar findings and documented a 14 percent decrease in V02 with effective analgesia. However, these studies did not compare the values obtained during labor with measurements on the same pregnant nonlaboring women at term; therefore, they do not directly measure the increase in V02 imposed by the transition to labor. Furthermore, neither of these studies measured breath-by-breath values ofV02 and VEe The data from Sangoul et al l3 and Hagerdal et al 12 are derived from measurements of pooled expired gases-a technique that necessarily yields averages over the intervals of timed collection. Breath-by-breath analysis allows for a more precise measurement of ma.ximal, transient V02 and VE during the contractions of labor. The objective of our study was to measure the V02 and VE in healthy, uncomplicated pregnancies during the third trimester and during the first stage of labor using breath-by-breath analysis. CHEST / 102 / 2 / AUGUST; 1992

467

r - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 100

50

~

12 TIME IN MINUTeS

FIG~lRE 1. Plot of V0 2 (solid line), Vc0 2 (dashed line), and VE (dotted line) as displayed bv metabolic cart

dunn~ breath-b.y-breath analysis. Uterine c0!1tractiol1s occurred approximately 2 to 3' min apart and

corresponded \\'lth the measured increases in V0 2 and VE. METHODS

Healthy \\'omen a~ed 16 throu~h 36 years old, in the 36th to 41st week of ~estation, \\'ere recruited durin~ their routine obstetric visits. A careful review of the patient's medical re(.'()fd, history, and physical examination was made to exclude patients with complications such as preeclampsia, diabetes mellitus, si~nificant cardiopulmonary disease, planned cesarean-section delivery (eg, breech presentation), prior cesarean-section delivery, or kno\\'n metabolic disease (eg, hyperthyroidism). Patients were enrolled after ~iving informed t'Onsent under a protocol approved by the institution's Department of Clinical Investi~ation. While study subjects were in a restin~, seated position, measurements were made of forced vital capacity and rnaximal voluntarv ventilation using a metabolic cart (Medical Graphics Corp()ratio~ CADINET System 2(01). Pulmonary function tests (PITs) were performed in accordance with standards established by the American Thoracic Society. I. The maximal voluntary ventilation (MVV) was the volume of air exhaled in a 12-s interval durin~ repetitive maximal respiratory efforts. Restin~ V0 2 and VE \\'ere measured breath-by-breath on the same metabolic cart over a lQ-min period while the subjects were sittin~ quietly, usin~ nose dips and mouthpiece with a two-way nonrebreathin~ valve (Hans Rudolph 2700 series). On admission to the labor and delivery suite, patients were restudied to measure V0 2 and VE over a lQ-min interval during active labor. All patients were studied in a semirecumbent position, sitting back at an angle of 450 from vertical. Patients were studied

during the first sta~e of labor, which was defined as that phase of active interval uterine contractions with cervix not yet fully dilated or effaced. The use of Lamaze techniques or analgesia was recorded and patients \\'ere excluded if anesthesia was given. Patients were not restudied if transferred directly to the delivery room for rapid pro~ression of labor, if medical or obstetric t'()mplications were detected, or if the patient declined further participation The 10-min periods of data collection were examined for episodes of uterine contraction (Fig 1). Data points from the five consecutive breaths ofhi~~est V0 2 and V~ during contractions were averaged to ~ive a peak V0 2 and peak VE. If more than one clear peak was identified durin~ the collection period, the peak V02 and peak VE \\'ere taken as the mean of the individual peaks. Group data are presented as mean ± standard deviation. Student's paired t test was used for comparison of values before and during labor with significance set at p = 0.05. V0 2 and VE data were normalized by dividin~ by term weight in kilograms. RESULTS

Twenty subjects were enrolled in the study. Four patients were excluded from the study after enrollment because of technical difficulties in data collection and calibration problems at the time of study. Of the 16 remaining study subjects, data collection for eight patients was limited to PFrs and baseline measurements of \10 2 and \1E during the final weeks

Table I-Oxygen Consumption· Subject 1 2 3 4 5 6 7 8

Mean V0 2 at Term

Mean V0 2 in Labor

3.3 3.1 3.3 2.7 5.3 4.0 3.8 3.0 3.56 (±0.82)

3.8 5.1 3.8 4.5 6.1 3.9 3.8 3.2 4.28t (±0.93)

Percent Increase from Term 15

65

15 67 15 -3 0 7 23 (±28)

PeakV0 2 in Labor 4.9 8.7 5.2 6.0 10.7 5.0 6.2

6.67:1: (±2.20)

Percent Increase from Term 48

181 58

122 102 25

63 86

(±53)

*Oxyge~ consumption in ml/k~min for the eight subjects who completed the stud\'. tMean Y02 in labor> mean"":02 at term, p=O.04. '

tPeak V0 2 in labor> mean V0 2 at term, p=O.OO5. 468

Oxygen Consumption and Ventilation during Normal Labor (Eliasson et al)

Table 2-Minute Ventilation· Subject 1 2 3 4 5 6 7 8

Mean VE at Tenn

Mean VE in Labor

Percent Increase from Tenn

Peak VE in Labor

0.13 0.10 0.17 0.15 0.20 0.18 0.14 0.15 0.15 (±0.03)

0.14 0.30 0.20 0.37 0.43 0.15 0.19 0.17 O.24t (±0.11)

8 200 18 147 115 -17 36 13

0.19 0.55 0.28 0.55 0.60 0.20 0.32

Percent Increase from Term 46 450 65

267 200 11 129 167 (± 154)

0.38+ (±0.18)

65

(±79)

*Minute ventilation in Ukglmin for the eight subjects who completed the study. tMean VE in labor> mean VE at tenn, p=O.05. +Peak VE in labor> mean VE at tenn, p=O.01.

of pregnancy. Measurements during labor were not performed in these eight patients either because labor progressed to delivery too quickly or because the subjects refused to participate further in the study. The eight subjects who completed the protocol ranged in age from 19 to 34 years, with an average of 27 years. Two patients were primiparous and six were multiparous (four were gravida 2, two were gravida 3). All patients labored successfully and had uncomplicated vaginal deliveries. The PFT results were normal in all cases.·5 The mean of the group's forced vital capacities (FVCs) was 3.67 L (93 percent ofpredicted), the mean of the forced expired volumes in 1 s (FEV.) was 3.12 L (99 percent of predicted), and the mean of the MVV was 112 Umin (94 percent of predicted). Resting V02 measured near term (36 to 41 weeks and labor not yet in progress) in these eight subjects ranged from 213 to 420 mVmin with a mean of296 ± 77 mVmin. Including measurements ofVo2 at term from all 16 nonlaboring subjects showed the same range of C

E

~ ry .:::L

~

0

-+-""

0.7

9

E ""-

(J"I

6

::J

4

C 0

U

3

0.5

C 0

0.4

~

I 1

I 1

0.3

~

0.2

~

-+-""

0

-+-""

c

Q)

>

Q)

:)

C

C

Q)

ry

:2

>-.. x

0

0.6

~

""--..J S

(f)

In the past, the prevalence of respiratory impairment from pulmonary tuberculosis in young women led to frequent cases of pulmonary insufficiency as a complicating factor in labor and delivery. It might be assumed that with the decline in tuberculosis, the obstetrician today is rarely confronted with this clinical situation. However, with advances in the care of other pulmonary conditions, notably cystic fibrosis, the problem of pulmonary insufficiency in pregnant patients continues to be ofconcern in the practice of obstetrics. c

8

Q.

E

DISCUSSION

10

E c

V02 and a mean of 295 ± 60 mVmin. These data were normalized by dividing by body weight in kilograms and are presented in Table 1. Measurements of V02 during labor (mean V0 2 in labor and peak V02 in labor) are also summarized in Table 2. Increases in the V02 are depicted graphically in Figure 2 and increases in the VE are shown in Figure 3.

0

0.1

T 1

0.0

term mean

labor mean

labor max

FIGURE 2. Oxygen consumption (displayed with standard deviations in brackets) increases from the third trimester (tenn mean) to active la~r (labor mean), with p = 0.04. There is a further marked increase in V0 2 measured during contractions (labor max) as compared with term baseline with p=O.OO5.

term mean

labor

labor

mean

max

FIGURE 3. Minute ventilation (shown with standard deviations in brackets) increases from the third trimester (term mean) to active labor (labor mean), with p=O.05. A marked increase in VE is found during contractions (labor max) as compared with tenn baseline with p = 0.01.

CHEST I 102 I 2 I AUGUSt 1992

469

To any woman in childbirth, there is little wonder about why the experience is called labor. The process of giving birth involves work by the mother that imposes a metabolic demand on the cardiopulmonary system. Physiologically, work is related to the amount of oxygen consumed and therefore, V02 is a measure of the amount of work that is done during labor. An understanding of how much work may be expected during labor in normal deliveries would be useful for predicting the ability of obstetric patients with cardiopulmonary limitations to meet the metabolic and ventilatory requirements of labor. Previous studies that measured V02 during labor confined themselves to evaluating the effects of pain control. HagerdaI et al 12 collected expired gas over 30s intervals during labor before and after lumbar epidural anesthesia. Their data showed a decrease in V02 of 30 percent (from 4.4 to 3.1 mllkglmin) during anesthesia. Sangoul et al 13 used expired gas collections of 7-min intervals before and after either epidural or paracervical block. Their measurements documented a decrease in V02 of 14 percent (from 317 to 272 mV min) as a consequence of pain control. This information adds to our ability to treat obstetric patients with pulmonary disease, but it does not offer data to allo\v for a prospective estimate of a patient's ability to labor. The number of subjects in our study was too small to allow for separation into categories such as the number of previous pregnancies. It is of anecdotal interest that the subject who increased her V02 the most from third trimester to labor (subject 4) \\'as gravida 3 with two previous normal deliveries. None of the other mothers in our study was more experienced at childbirth. None of the eight patients received local or systemic anesthesia or analgesia prior to our measurements. The data from our study show that there is an increase in mean V0 2 of 23 percent from sedentary third-trimester pregnancy to active stage 1 labor (from 3.6 to 4.3 mllkg/min). These values are in general agreement with the measurements of the two studies cited above. Furthermore, comparing the peak V02 during contractions with the nonlaboring state sho\\'s an 86 percent increase in work performed. This implies that women who are unable to sustain a prolonged increase in their resting term V0 2 by 23 percent would not be able to meet the metabolic requirements of labor. In a severely compromised individual, the average short-term increase in \vork during contractions might be performed by brief periods of anaerobic energy supply but with unknown risk to the fetus. Some patients will require a greater aerobic capacity for a safe labor and delivery. Therefore, the range of increases in V02 must be considered. Using this approach, it would be best if an obstetric patient could 470

sustain an increase in V02 of79 percent (mean increase plus 2 SDs) over her resting needs during the third trimester. To measure the ability of a patient to increase her V0 2 , a cardiopulmonary exercise test (eg, bicycle ergometry) may be safely performed during pregnancy.3 Data collected during the exercise test could be used to estimate the patient's cardiopulmonary reserve and whether a trial of labor is advisable. The degree of arterial desaturation with pulse oximetry during exercise may also affect management decisions. Cardiopulmonary testing may not be readily available. A simpler and less stressful method of evaluating patients, such as PFrs, would have clear advantages. However, applying the same logic to VE requirements during labor is more complicated. Our data show a mean increase in VE of65 percent from third trimester to labor with a great deal of individual variabilitychanges in VE values ranged from - 17 percent to 200 percent. During contractions, peak VE increased as much as 450 percent in the case of one individual. In normal exercise, VE follows changes in V0 2 in a linear fashion. Thus, the enormous changes in VE are not due to a metabolic requirement imposed by labor but rather must be explained by the effects ofpain, coached breathing patterns, and the birthing process. With a better understanding of the metabolic and ventilatory load imposed by labor, we can outline an approach to a patient with cardiopulmonary disease, to determine whether the patient's impairment would preclude a trial of labor. The basic question is, can the patient meet a sustained increase in V0 2 and VE over resting values at term for the indefinite period of labor? Relating the estimated ventilatory requirement to a measurable PFr parameter is best done using the MV\Z A nUlnber of studies have examined the fraction of MVV that is sustainable. 16-1H A normal subject can sustain MVV for only 20 to 30 s. The endurance time increases as the ventilatory requirement is decreased. At 76 percent of M~ for example, endurance times of 4 to 15 min have been reported. IH Breathing at or below 40 percent ofMVV can be sustained indefinitely. A term obstetric patient can readily be evaluated for resting VE and MV\Z Sustainable VE would approximate 40 percent of her measured M~ and the estimated VE to support labor should be at least 65 percent greater than her measured resting VE. Mathematically this would be as follows: 0.4xMVV=VE sustainable ~1.65xVE resting or MVV ~4 .13 VE resting. From a conservative point of view, a greater ventilatory reserve might be necessary to account for outliers who vary markedly from the expected mean increase in VE. The mean increase in VE plus 2 SDs constitutes a 222 percent increase from the resting term VE. Using these figures: Oxygen Consumption and Ventilation during Normal Labor (Eliasson et al)

0.4 x MVV = VE sustainable ~3.2 x VE restin~ or MVV ~8.0 VE resting. The dramatic but brief increases in VE measured during contractions are due in part to a response to pain rather than from a metabolic requirement imposed by labor. Adequate pain control may playa role in limiting both V0 2 and

VE.

tive validation. REFERENCES

2

12 . 13

These estimates and derivations lead to the following cautious approach. Any pregnant patient with cardiopulmonary disease should receive a thorough examination with spirometry, including MVV and resting VEe If the MVV is eight times the VE, the patient will likely have the ventilatory reserve necessary for successful labor and delivery. Close observation would be the only recommendation. If the MVV is only four to eight times the VE, the patient would still likely have the capacity to perform labor safely, but expert and timely attention to pain control, monitoring ofmaternal oxygen saturation, and possibly fetal monitoring would be justified. If MVV is less than four times the VE, a cardiopulmonary exercise test would be useful to measure the patient's ability to increase her V0 2 • If she can increase and sustain her V02 by 79 percent, then she probably has the respiratory reserve to perform the work of stage 1 labor, but careful attention to pain control and monitoring would be essential. Patients unable to increase V02 by 23 percent or VE by 65 percent require very close monitoring and probably lack the pulmonary reserve to tolerate labor. Our measurements of V0 2 and VE were made in healthy volunteers. It is clear that patients with pulmonary disease use a greater share of their respiratory capacity for ventilation. 19.20 This oxygen cost of breathing has been measured to be as high as 55 percent of total V02 in a patient with COPD with a mean of 20 percent in a group of cardiopulmonary patients. 20 These observations underscore the need for a cautious approach to the pregnant patient with cardiopulmonary limitations and underlie our recommendation for a conservative application of the data from normal women to the clinical setting. Our study suggests that cardiopulmonary exercise testing or spirometry may be used to estimate the ability of a patient to increase V02 and VE during labor and may have a role in identifying women with llln~ disease who can safely and successfully undergo labor and delivery. Since our study group consisted of well women, and did not include patients with cardiopulmonary limitations, these suggestions need prospec-

3

4 5 6 7

8

9

10

II

12

13

14 15

16

17 18

19

20

Matuschak G~t, ()wens GR, Rogers RM, Tibbals Sc. Pro~es­ sive intrapartum respiratory insufficieney due to pulmonary alveolar proteinosis. Chest 1984; 86:496-99 Ratto D, Balrnes J, Boylen T, Sharma O~ Pre~naney in a woman \\'ith severe pulmonary fibrosis secondary to hard metal disease. Chest 1988; 93:663-65 South-Paul JE, Raja~opal KR, Tenholder MF. The effect of participation in a regular exercise program upon aerobie capacity durin~ pregnaney. Obstet GyneeoI1988; 71:175-79 \Veinberger SE, \Veiss ST, Cohen \VR, \Veiss J\~ Johnson TS. Pregnancy and the lung. Am Rev Respir Dis 1980; 121:559-81 Leontic EA. Respiratory disease in pregnancy. Med Clin North Am 1977; 61: 111-28 Prowse C~t, Gaensler EA. Respiratory and acid-base changes during pregnancy. AnesthesioloJ.,ry 1965; 26:381-92 (;azioglu K, Kaltreider NL, Rosen M, Yu PN. Pulmonary function during pregnancy in normal \\'Olnen and in patients with cardiopulmonary disease. Thorax 1970; 25:445-50 Ueland K, Novy MJ, Metcalfe J. Cardiorespiratory responses to pregnancy and exercise in nornlal women and patients with heart disease. Obstetrics 1973; 115:4-10 Spencer JA, \Volton RS, Rolfe ~ Johnson ~ Mass spectrometer system for continuous skin-surface and intravascular blood gas measurement of maternal-fetal respiration in labour. J Biomed Eng 1987; 9:161-68 Joupilla R, Hollmen A. The effect ofsegmental epidural anal~esia on maternal and fetal acid-hase balance, lactate, serum potassium and creatine phosphokinase during labour. Acta Anaesthiol Scand 1976; 20:259-68 Fisher A, Prys-Roherts C. Maternal pulruonary ~as exchange. Anaesthesia 196R; 23:350-56 Ilagerdal M, Morgan C\~ Sumner AE, (;utsche BB. Minute ventilation and oxygen consumption during labor with epidural analgesia. AnesthesioloJ.,ry 1983; 59:425-27 Sangoul F, Fox GS, Houle G L. Effect of regional analgesia on maternal oxygen consumption during the first stage of labor. Am J Ohstet Gyne