In recent years, maternal mortality from acid pulmonary aspiration (Mendelson’s syndrome) (1) has dramatically declined. In the Report on the Confidential Enquiries into Maternal Deaths in England and Wales (1991–1996), only one mother died from aspiration (2). There are several factors that may be associated with this audited improvement. These include the increased use of regional anesthesia for cesarean delivery, improved training of anesthesiologists, and, possibly, the introduction of nonparticulate antacids and H2-receptor antagonists. The role of nothing by mouth during labor, as recommended in the first Report on the Confidential Enquiries into Maternal Deaths (1952–1954), is less clear (2).
Women in labor exhibit a state of “accelerated starvation,” with rapid increases in the blood levels of β-hydroxybutyrate, acetoacetic acid, and the nonesterified fatty acids (NEFAs) from which they are derived and with a concomitant decrease in blood glucose (3). It has been suggested, although never scientifically proven, that these changes may have detrimental effects on uterine activity and the progress of labor (4).
A previous study demonstrated that allowing laboring women to eat a light diet prevented the increase of plasma ketones and NEFAs (5). However, not surprisingly, feeding resulted in a significant increase in residual gastric volume, which could predispose to pulmonary aspiration should a complication of neuroaxial anesthesia occur or should general anesthesia be required unexpectedly. Isotonic drinks are rapidly emptied from the stomach and absorbed by the gastrointestinal tract (6,7) and therefore may theoretically provide a safer alternative to solid food. The aim of this study was to evaluate whether isotonic drinks would prevent ketosis without increasing the risk of potential aspiration.
St. Thomas’ Hospital Ethics Committee granted approval for this project. After informed written consent, 60 women presenting in early labor (cervical dilation <5 cm) were randomized to receive either oral isotonic “sport drinks” (28 kcal/dL) or water only (Control group) throughout the course of labor. Randomization was achieved with computer random number generation. Sealed envelopes were opened after recruitment to the study. All women were at 37-wk gestation or longer and had a singleton fetus with cephalic presentation. Any mother with a known obstetric or medical complication that could have increased the likelihood of instrumental delivery or cesarean delivery was excluded. Mothers requesting IM meperidine for analgesia were excluded because of its well documented effects of delaying gastric emptying (8). Epidural lumbar analgesia, when requested, was provided with a small-dose infusion of bupivacaine 0.1% plus fentanyl 2 μg/mL. The oral isotonic fluid used was Lucozade Sport® (still), with the choice of either orange or lemon flavor. Lucozade Sport (still) contains a mixed carbohydrate profile (dextrose, maltodextrin, and glucose) of 64 g/L, a sodium of 24 mmol/L, potassium of 2.6 mmol/L, and calcium of 1.2 mmol/L and has a tonicity of 300 mOsm/kg.
Women in the Sport Drinks group were encouraged to consume up to 500 mL (one bottle) in the first hour and then a further 500 mL every 3 to 4 h. Additionally, they were allowed to take small quantities of water as desired. Women randomized to the Water-Only group could consume as little or as much water as they wanted.
For metabolic assessment, plasma β-hydroxybutyrate, NEFAs, and glucose were measured in early labor and again at the end of the first stage by using blood samples. Real-time ultrasonography was used to compare residual gastric volumes between the two groups (9,10). Examinations were performed with a high-resolution scanner (SSD 620; Aloka, Tokyo, Japan), by use of a 5-MHz curvilinear transducer in the midline of the epigastrium with the mother sitting up at 45°. All the ultrasound scans were performed within 45 min after delivery and by one investigator (MK). The incidence and volume of vomiting that occurred during labor and within the hour of delivery were recorded.
Duration of the stages of labor, oxytocin requirements, mode of delivery, Apgar scores, and umbilical artery and venous blood gases were recorded.
Previous metabolic studies in pregnancy suggested that a change in β-hydroxybutyrate of 0.15 mmol/L (mean, 0.24 mmol/L; sd, 0.20 mmol/L), in NEFAs of 0.2 mmol/L (mean, 0.85 mmol/L; sd, 0.2 mmol/L), and in glucose of 0.7 mmol/L (mean, 5.2 mmol/L; sd, 0.7 mmol/L) would be both plausible and medically important (11). Power calculations indicated that 30 women in each group would be needed to detect such a difference in β-hydroxybutyrate with 80% power and at a 5% level of significance. Much smaller numbers would be needed for the other measures. Metabolic, gastric volume, and vomiting data were analyzed by parametric methods by using linear regression analysis, adjusting for baseline as appropriate (12). The standard errors were corrected for nonnormality and unequal variances by use of the Huber-White sandwich estimator (13). Demographic data and maternal and fetal outcomes were analyzed with Student’s t-test and χ2 analysis as appropriate. All data were analyzed by the intent-to-treat principle by using the commercially available statistics software package Stata® version 6.0 (Stata Corp., College Station, TX).
Sixty women were recruited to the study: 30 to the Isotonic Sport Drink group and 30 to the Water-Only (Control) group. Ninety-one women were asked to join the study; 31 declined consent. No woman withdrew from the study. Women in the two groups were similar with respect to age, parity, induction, and cervical dilation at the time of randomization (Table 1). There was no difference between the two groups in the duration of labor, use of oxytocin, mode of delivery, or use of epidural analgesia (Table 2). The babies had similar Apgar scores and umbilical artery and venous gases (Table 2).
The isotonic sport drinks were acceptable to most mothers in the Sport Drink group, with only one woman refusing to consume more than 200 mL for the study. There was a progressive decrease in the desire to drink the sport drink toward the end of the labor and after consuming 750–1000 mL. The two groups were similar in all of the baseline metabolic indices measured on entry to the study in early labor (Table 3). However, by the end of labor, plasma β-hydroxybutyrate and NEFAs were significantly increased and plasma glucose significantly decreased in the Water-Only group (Table 3). In one patient in the Isotonic Sport Drink group, a result was not obtained from an entry sample for β-hydroxybutyrate, and this result was therefore excluded from the metabolic analysis. Gastric antral cross-sectional area measured within 45 min of delivery was not significantly different between the groups (Table 4). There was no difference between the groups in the volume vomited or number of vomiting episodes within 1 h of delivery or throughout labor (Table 4). The total quantity of liquid consumed was significantly more (95% confidence interval, 193–701;P = 0.001) in the Sport Drink group (mean, 925 mL; sd, 384) compared with the Water-Only group (mean, 478 mL; sd, 579). The mean calorific intake in the Sport Drink group was 47 (sd, 16) kcal/h and was none in the Water-Only group.
It has previously been shown that consuming a light diet during labor prevents the increase in ketone production, but this happens at the expense of increasing the residual gastric volume (5). In this study, isotonic sport drinks significantly reduced the increase in ketone production (Table 3) but did not increase the residual gastric volume when compared with the Water-Only group. This suggests that isotonic sport drinks may be a safer alternative to solids in labor. Ketosis occurs when the body metabolizes fats because carbohydrates are not available in sufficient quantity for use as a source. In labor, β-hydroxybutyrate is the principle ketone produced as a result of starvation (14), and in our study the sport drinks significantly reduced the production of this metabolite (Table 3). Plasma NEFAs, from which β-hydroxybutyrate is derived, are mobilized from the liver in response to starvation (14), and, once again, the increase was significantly reduced in the Isotonic Sport Drink group (Table 3). Indeed, the mean β-hydroxybutyrate was decreased and the mean plasma NEFA level remained unchanged throughout the first stage of labor in the Isotonic Sport Drink group. Mean plasma glucose remained unchanged in the Isotonic Sport Drink group, demonstrating that sufficient calories were being absorbed and used compared with the Water-Only group, in which the level decreased significantly (Table 3).
The results demonstrated no significant difference in either maternal or fetal outcomes despite these metabolic improvements. To detect a 10% change in duration of labor or a 6.6% increase in the spontaneous delivery rate (power at 90%, 5% significance level), it is estimated that 2400 primiparous women would need to be studied. Production of ketones from the mobilization and breakdown of NEFAs is the physiological consequence of starvation, and in normal pregnancy women are prone to ketosis because of a change in the maternal hormonal milieu (14). Ketone bodies are not waste products of metabolism but are readily available alternate energy sources for the contracting uterus and the brain of both mother and fetus. Although ketones may increase with the duration of labor in a mother deprived of a calorific source, no causal relationship with outcome has ever been proven (4). Indeed, this increase in ketones may prove to be inconsequential to the progress and outcome of labor. A study evaluating the effect of feeding on labor outcome is now in progress.
Despite the greater quantity of liquid taken during labor in the Isotonic Sport Drink group, residual gastric volumes were similar in the two groups, indicating rapid gastric emptying of the isotonic drinks. Lucozade Sport (still), the isotonic sport drink used in the study, contains a mixed carbohydrate profile (dextrose, maltodextrin, and glucose) of 64 g/L and has a tonicity of 300 mOsm/kg. Mixed carbohydrate solutions empty faster than glucose-alone solutions (6), as do isotonic carbohydrate solutions, compared with hypertonic carbohydrate-containing solutions (7). Large concentrations of carbohydrates in solutions have an increased osmolality and are associated with a slower rate of gastric emptying (7). The source of carbohydrate will influence fluid osmolality, and to avoid very high osmolalities, the quantity of monosaccharides must be smaller than that of disaccharides or polysaccharides. Indeed, in a study in exercising athletes (15), an ideal solution to achieve maximal carbohydrate availability without impairing fluid homeostasis was found in beverages with 60–80 g/L of carbohydrate and isoosmolar 300 mOsm/kg. The sport drink used in the study has advantages of being isotonic with faster gastric emptying, but having a smaller carbohydrate content; therefore, fewer calories are available to the parturient.
The mean values for the gastric antral cross-sectional area in both the Isotonic and Water groups were similar to those of a group who drank water only in a previous study (5). The use of the ultrasound as a two-dimensional modality to provide an estimate of gastric volume, a three-dimensional measure, has been validated previously (9,10) and was used in a similar study evaluating the effect of eating solid food during labor (5). Three of the participants were excluded from the gastric analysis because of inability to visualize the stomach: two in the Sport Drink group and one in the Water group. Volumes vomited within an hour of delivery and throughout labor were also not significantly different, again showing that isotonic drinks were being emptied from the stomach as fast as water. Evidence to date indicates that pregnancy may not significantly alter gastric emptying of liquids or solids but that established labor may cause an unpredictable delay in gastric emptying (16). Opioid drugs given parenterally for pain relief cause a marked delay in gastric emptying (8,17). Studies of the effect of small-dose epidural infusions containing fentanyl 2 μg/mL have not demonstrated a profound delay in gastric emptying (18,19). However, in one study, a slight delay (not statistically significant) was noted at 4.5 hours after 125 μg of fentanyl, leading the investigators to suggest that delay in gastric emptying may occur in a dose-dependent manner (18). The evidence is conflicting (20,21). The mean total doses of epidural fentanyl in the Isotonic Sport Drink and Water groups during labor were 181 and 205 μg, respectively, which equates to 27 and 24 μg/h, of fentanyl. It is possible that epidural fentanyl in small-dose infusions may have an effect, but it is likely to be small compared with the delay in gastric emptying produced by parenteral opioids. The women who required cesarean delivery in this study had their lumbar epidural analgesia converted to anesthesia for surgery.
This study demonstrates that only a relatively small calorific intake of 47 kcal/h prevents the development of ketosis and that isotonic drinks are rapidly emptied from the stomach and absorbed by the gastrointestinal tract in the laboring mother. Isotonic drinks provide an alternative source of nutrition to food. To be isotonic, the calorific load of such drinks is limited to approximately 30 kcal/dL.
It is not known whether the degree of ketosis that occurs in some women during labor is a harmless physiological state or a pathologic condition that interferes with uterine activity, but further research in this field is needed. Our study supports the policy of allowing mothers who have not received parenteral opioids to consume isotonic drinks once they are in established labor.
We thank the mothers, midwives, and obstetricians at St. Thomas’ Hospital for their cooperation and support. We would like to acknowledge the guidance provided by Dr. L. MacDonald, Consultant Radiologist, St. Thomas’ Hospital; G. A. Metcalfe (Dietitian), St. Thomas’ Hospital; and Beechams (Ltd.) for the supply of the Lucozade isotonic sport drinks. We thank the Obstetric Anesthetists Association for funding the salary of MK and costs of the laboratory tests.
1. Mendelson CL. The aspiration of stomach contents into the lungs during obstetric anesthesia. Am J Obstet Gynecol 1946; 52: 191–5.
2. Report on the Confidential Enquiries into Maternal Deaths in England and Wales (1991–93), (1994–96). London: HMSO, 1957–96 (series of 14 triennial reports).
3. Metzger BE, Ravnikar V, Vileisis RA, Freinkel N. “Accelerated starvation” and the skipped breakfast in late normal pregnancy. Lancet 1982; 1: 588–592.
4. Dumoulin JG, Foulkes JEB. Ketonuria during labour. Br J Obstet Gynaecol 1984; 91: 97–8.
5. Scrutton MJ, Metcalfe GA, Lowy C, et al. Eating In labour: a randomised controlled trial assessing the risks and benefits. Anaesthesia 1999; 54: 329–34.
6. Sole CC, Noakes TD. Faster gastric emptying for glucose-polymer and fructose solutions than for glucose in humans. Eur J Appl Physiol Occup Physiol 1989; 58: 605–12.
7. Vist GE, Maughn RJ. Gastric emptying of digested solutions in man: effect of beverage glucose concentration. Med Sci Sports Exerc 1994; 26: 1269–73.
8. Holdsworth JD. Relationship between stomach contents and analgesia in labour. Br J Anaesth 1978; 50: 1145–8.
9. Bolondi L, Bortolotti M, Santi V, et al. Measurement of gastric emptying by real time ultrasonography. Gastroenterology 1985; 89: 752–9.
10. Ricci R, Bontempo I, Corazziari E, et al. Real time ultrasonography of the gastric antrum. Gut 1993; 34: 173–6.
11. Saleh AK, Al-Muhtaseh N, Gumaa KA, et al. Maternal amniotic fluid and cord blood metabolic profile in normal pregnant and gestational diabetics during recurrent withholding of food. Horm Metab Res 1989; 21: 507–13.
12. Frison L, Pocock SJ. Repeated measures in clinical trials: analysis using mean summary statistics and its implications for design. Stat Med 1992; 11: 1685–704.
13. Huber PJ. The behavior of maximum likelihood estimates under non-standard conditions. In: Proceedings of the Fifth Berkley Symposium in Mathematical Statistics and Probability. Berkley: University of California Press, 1967: 221–33.
14. Posner NA, Silverstone FA. Carbohydrate metabolism in pregnancy. Obstet Gynecol Annu 1977; 6: 67–125.
15. Brouns F. Heat-sweat-dehydration-rehydration: a praxis oriented approach. J Sports Sci 1991; 9: 143–52.
16. O’Sullivan G. Gastric emptying during pregnancy and the puerperium. Int J Obstet Anesth 1993; 2: 216–24.
17. Nimmo WS, Wilson J, Prescott LF. Narcotic analgesics and delayed gastric emptying during labour. Lancet 1975; 1: 890–3.
18. Porter JS, Bonello E, Reynolds F. The influence of epidural administration of fentanyl infusion on gastric emptying in labour. Anaesthesia 1997; 52: 1151–6.
19. Zimmerman DL, Breen TW, Fick G. Adding fentanyl 0.0002% to epidural bupivacaine 0.125% does not delay gastric emptying in laboring parturients. Anesth Analg 1996; 82: 612–6.
20. Wright PM, Allen RW, Moore J, Donnelly JP. Gastric emptying during lumbar extradural analgesia in labour: effect of fentanyl supplementation. Br J Anaesth 1992; 68: 248–51.
21. Ewah B, Yau K, King M, et al. Effect of epidural opioids on gastric emptying in labour. Int J Obstet Anesth 1993; 2: 125–8.