Quenby, Siobhan MD*; Pierce, Sue J. MBBS*; Brigham, Sara MRCOG*; Wray, Susan PhD†
There is increasing public health concern regarding the rise in the cesarean delivery rate in the last 30 years. In 1985, the World Health Organization issued a consensus statement suggesting there were no additional health benefits associated with a cesarean delivery rate above 10–15%. However, rates in some Latin American countries are approaching 40%,1,2 in the United States they are 24%,3 and in England and Wales they are 21.5%.4 A substantial proportion (20%) of cesarean deliveries are for failure to progress in labor (dystocia).4 Dystocia is associated with significantly increased morbidity and mortality for the mother and her child.5,6 Inefficient uterine contractions are the most common cause of poor progress in labor.7 According to the U.K. National Audit, 80% of primigravida women who had a cesarean delivery received oxytocin before delivery4; however, oxytocin has not been found to lower the cesarean delivery rate.8–10
Much progress has been made in the understanding uterine contractility by in vitro studies of both animal and human myometrial samples. Hypoxia is known to reduce the force of smooth muscle contraction.11 During labor, the pregnant uterus has to contend with intermittent hypoxia because the intrauterine pressure during contractions is sufficient to reduce arterial blood flow in both animals and humans.12–14 Hypoxia was found to decrease agonist-induced force in the uterus.15 In vitro, acidification was found to have a profound inhibitory role on uterine contractions in human pregnant myometrium.16,17
Because acidification is associated with hypoxia, we hypothesized that
* The myometrial capillary pH in dysfunctional labor is lower than that in normally contracting uteri.
* The difference in myometrial capillary pH between normal and dysfunctional labor in vivo is sufficient to decrease the force and frequency of human pregnant myometrial contractions in vitro.
MATERIALS AND METHODS
Local ethics committee approval was obtained for all work in the study, and written, informed consent was obtained from all women. Women undergoing lower-segment cesarean delivery were asked if they would consent to the study if their cesarean delivery was planned to occur under regional anesthesia during delivery suite shifts when the authors S.Q. and S.B. worked as clinicians between November 2001 and November 2002 operations. The inclusion criteria for all groups were that they were more than 37 completed weeks of gestation, that they were in the first or second pregnancy that progressed beyond 24 weeks gestation, and that they had ability to give informed consent. Further inclusion criteria for the laboring samples were women in the first stage of labor for at least 6 hours and in whom 2 experienced clinicians (S.Q. and S.B.) were able to independently review all the clinical notes and arrive at the same diagnosis. Women were excluded if general anesthesia was used, if they were unable to give informed consent (eg, too distressed in labor, insufficient time to consent the women before the procedure because of severe fetal distress, unable to speak sufficient English), they had more than 2 previous pregnancies that progressed beyond 24 weeks gestation, if ketonuria was detected on urinalysis, or uterine hyperstimulation occurred with oxytocin, or hypertonic or tetanic type contractions occurred. The patients selected were all those receiving cesarean deliveries who were eligible for the study performed by S.Q. or S.B. within the study period, thus ensuring consistency in sampling technique. Five women declined to consent to the study. Insufficient blood was obtained for analyses in 11 cases, 6 of which had a previous lower-segment uterine incision, so that the part of the uterus that was initially incised was less vascular than usual. The following diagnostic groups were used:
Group 1, nonlaboring: Twenty-three women having elective cesarean delivery at term (38–40 weeks of gestation) before the onset of labor.
Group 2, failed induction of labor: These 4 women had 2 doses of prostaglandin 6 hours apart. The next day they had artificial rupture of the membranes and at least 6 hours of maximal dose oxytocin infusion. Despite this, no cervical dilation occurred, and they never reached 2 cm of cervical dilation or the active phase of labor.
Group 3a, normally contracting: Seven women with regular, spontaneous contractions that were clinically strong and who were dilated the cervix at a rate of greater than 1 cm an hour. These women underwent a cesarean delivery for other reasons; namely fetal distress (n = 2) and breech diagnosed late in labor (n = 5).
Group 3b, normally contracting on oxytocin: Eight women with normally contracting uteri and oxytocin administered for induction/augmentation of labor. These women had their cesarean deliveries for relative cephalopelvic disproportion (inadequate progress in the active phase of labor despite adequate uterine contractions; n = 6) or for fetal distress (n = 2). They all had good regular contractions, as defined by an intrauterine pressure catheter reading of greater than 200 Montevideo units (mu).
Group 4, dysfunctional labor: Ten women undergoing cesarean for dysfunctional labor, as diagnosed by being in active labor but with poor contractile pattern and activity (less than 200 mu). These women were managed in accordance with the hospital protocol. Thus, failure to progress in labor associated with clinically poor uterine contraction was initially treated with standard oxytocin augmentation (start at 2 mu/min and double every 30 minutes until contractions occur at a rate of 3–4 in 10 minutes and last 45–60 seconds or until 16 mu/min in multiparous women or at 32 mu/min in primigravida women is reached); if this fails, an intrauterine pressure catheter is inserted. If the catheter readings were less than 200 mu, then high-dose (64 mu/min in primigravida women and 32 mu/min in multigravida women) oxytocin was used. If this higher dose also failed to progress the labor (n = 10), the women had cesarean delivery for dysfunctional labor.
All women were fed in labor per hospital protocol. Most women who were in this study were in labor for many hours. In the initial part of the labor, vaginal delivery was anticipated and therefore feeding was appropriate.
Per hospital protocol, all laboring women had 8 hourly dipstick urinalyses for protein, glucose, and ketones analysis on an automated analyzer. All women in this study had a urinalysis test within 1 hour of the cesarean procedure. All babies were delivered in good condition; none needed admission to a special-care baby unit or showed signs of significant hypoxia.
When taking the blood sample, all visible blood vessels on the surface of the uterus were avoided. The first bead (1 mL) of blood that oozed out of the first small (1 cm), superficial, lower-segment uterine incision was aspirated into a preheparinized syringe. If insufficient blood was obtained at this incision, the sample was excluded from the analysis. Therefore, this sample was as close to myometrial capillary blood as possible. However, we cannot entirely discount the possibility that myometrial sinusoids were contributing to the sample. If this was the case, then the blood samples represent mixed venous and arterial blood. Great care was taken to ensure that only blood from the uterine incision, and not from the skin (or any other area), entered the syringe. The blood was collected before the uterine cavity was opened so that it would not be contaminated by liquor. This blood sample was then immediately analyzed in the blood gas analyzer (Bayer) on delivery suite at Liverpool Women's Hospital. This machine undergoes daily quality control. The blood was not collected anaerobically. However, it was collected in a similar way to that established, validated technique used for fetal scalp blood sampling in labor. The theoretical contamination by air is the same for both the sampling technique described here and the established fetal blood sampling technique. Air contamination has not proved to be a problem in the internationally accepted fetal blood sampling technique and in any case would give a systematic error in all samples. Peripheral oxygen saturation was monitored with pulse oximetry throughout the operation.
The results were analyzed on Arcus software (Research Solutions, Cambridge, UK) for personal computers. The Shapiro–Wilks test found the data not to be normally distributed. The Kruskal–Wallis test was used to detect significant variance among the groups. A significance level of P < .05 was used, and interquartile ranges are given.
Human myometrial tissue was obtained from 5 women undergoing elective lower-segment cesarean delivery at term, before the onset of labor, that is, group 1 above. Full-thickness tissue biopsy specimens were obtained from the upper lip of the uterine incision. The tissue was stored in cold physiological saline for a maximum of 3 hours, until dissection was performed.
Strips of longitudinal fibers (1 × 5 mm) were dissected and placed in a small organ bath and attached to a force transducer. The tissues were continuously superfused with physiological saline at 37°C, pH 7.5, buffered with 10.9 mmol/L N-[2hydroxyethyl]piperazine-N′-[2enthanesulfonic acid]. Changes in extracellular pH were made by the addition of HCl to the perfusing solution. Previous work has shown that identical data are obtained irrespective of whether the pH buffer is N-[2hydroxyethyl]piperazine-N′-[2enthanesulfonic acid] or bicarbonate/CO2.18,19 The mean amplitude of contractions at pH 7.5 was compared with those at pH 7.3 by using a Student t test.
The groups of laboring patients in this study were clinically similar in all respects other than those patients in groups 3a and 3b, who were contracting normally, and those in group 4, who were contracting poorly despite standard intervention with oxytocin to improve contractions. The duration of active labor was similar in these groups (Table 1). All groups of women had normal peripheral O2 saturation (98/100%), and no ketones were present in the urinalysis tests performed during labor.
We have found that the myometrial capillary blood taken from women having a dysfunctional labor (Table 1) was at a significantly, lower pH (7.35) than that from women having elective cesarean delivery (7.49), cesarean with normal contractions, with (7.47) or without (7.48) oxytocin, or cesarean for failed induction of labor (7.46). The myometrial capillary percent oxygen saturation was also lower in women in dysfunctional labor compared with normally contracting women and women with failed induction of labor (Table 1). The myometrial capillary lactate concentration was significantly higher in dysfunctional labor than normal labor or nonlaboring women (Table 1). Chloride, calcium, and base excess were similar in all groups of patients (Table 1).
To determine the in vitro effect on contractility of acidifying human myometrium in the range found above, the pH was lowered from 7.5 to 7.3. At pH 7.3, the uterine contractions were less regular and of reduced amplitude compared with control activity before relative acidification (n = 5; Figure 1). The mean decrease in amplitude was 41% ± 5% and the decrease in contractile activity (ie, the integral force in unit time) was 44% ± 4% compared with the preceding control activity (P < .001). When the pH was returned to 7.5, regular contractions were restored. In control strips not subjected to acidification, the contractile pattern was unchanged (data not shown).
Our data support our hypothesis that the myometrial capillary pH in women undergoing cesarean delivery for dysfunctional labor was significantly lower (7.35) than that in women having this procedure when they are contracting effectively (7.48). Furthermore, lowering the pH from 7.5 to 7.3 was sufficient to decrease the force and frequency of human pregnant myometrial contraction in vitro. Our unique data show that myometrial acid-base balance is associated with dysfunctional labors.
Before considering the mechanisms underlying our findings, it is necessary to discuss the definition of dysfunctional labor. Despite its being a common clinical problem, it is difficult to find precise, agreed definitions. We have therefore used the hospital protocol as the criteria for a dysfunctional labor and how it was distinguished from oxytocin-augmented normal labor. In addition, the diagnoses had to be arrived at independently by 2 experienced clinicians. Thus, the reason for cesarean delivery in the dysfunctional labor group would be failure to progress because of insufficient uterine activity, whereas in the normally laboring uterine group, it was because of relative cephalopelvic disproportion or fetal distress.
The relatively high pH of the elective cesarean group (7.49) is consistent with the respiratory alkalosis of pregnancy,18,20 and women in normal labor had a similar myometrial pH. However, the women in dysfunctional labor of the same duration had a dramatic decrease in the pH of 0.14 pH units. This was associated with greater lactate levels and reduced oxygen saturation. The question then becomes what mechanism can account for these changes found only in the dysfunctional labor group?
The changes were not caused by oxytocin administration because the other groups of women who received oxytocin (groups 2 and 3b) had similar myometrial pH levels to the elective cesarean group, as well as similar oxygen saturation and lactate. It also has been suggested that the laboring woman is in a state of “accelerated starvation,” whereby she is starved, and the metabolic requirements of her uterus are met by gluconeogenesis, lipolysis, and glycogenolysis,21 conditions that might change pH. Such an explanation is not the case in any of our women because they were fed in labor and no ketones were detected in their urine.
Venous, maternal capillary lactate levels have been found to increase in the second stage of labor.22 This was thought to be the result of maternal skeletal muscle activity while the woman was “pushing” in order to deliver the fetus.23 However, our data, where myometrial capillary lactate levels were measured, suggest that the uterine muscle itself is a source of lactate in the first stage of labor (when no “pushing” occurs). Our women were semirecumbent and, therefore, the only tissue that was likely to have a more active metabolism when compared with women having an elective cesarean delivery was the myometrium. It was not, however, the length of labor that determined the lactate accumulation and fall in pH because both the normally contracting women and the dysfunctional group had similar amounts of time in the first stage before their cesarean.
As well as a decreased pH, the dysfunctional labor group had a significantly lower myometrial O2% saturation compared with the normally laboring group. This in turn provides an explanation for the raised lactate and lowered pH, as anaerobic metabolism will be stimulated. During labor, contractions can be of such intensity that blood vessels supplying and traveling through the myometrium are occluded.24,25 This occlusion will cause lowering of myometrial oxygen levels and accumulation of lactic acid, as detected in our study, and suggests that it is more pronounced in dysfunctionally laboring women. It is not possible on the basis of this cross-sectional study to determine whether the lactate acidosis or the poor contractility occurred first. However, a poorly contracting uterus should be associated with less myometrial blood vessel constriction than a normally contracting one. Therefore, interindividual response to intermittent myometrial hypoxia may be a determining factor in the efficacy of myometrial contractility in labor. It may be that there is variation in the myometrial response to intermittent hypoxia, such that in normally laboring women lactate accumulation and lowering of the pH are avoided more effectively than in those women who endure dysfunctional labor, for example, if the myometrial blood vessels do not recover from the episodes of occlusion as rapidly or fully in some women. It has also been shown in animal studies that there is a shift in the isoenzyme form of lactate dehydrogenase at term and a general metabolic preparation for hypoxic conditions.26 Any variation in these changes could also contribute to the differences observed in the dysfunctional labor group. This is consistent with the report that a deficiency in lactate dehydrogenase leads to inadequate and arrested labor and cesarean delivery.27 It may also be that the contractile pattern in dysfunctional labors leads to incomplete relaxation of the uterus between contractions and, thus, despite the inefficient contractions, there is inadequate reoxygenation of the uterus.
Irrespective of the mechanism involved, there will be functional implications of the acidification. There is substantial evidence that a fall in pH levels leads to inhibition of contractions in both rat14,19,28 and human17,18 uterus. When the pH in vitro was specifically altered over the range found in vivo, a significant decrease in the force of myometrial contractions occurred (Figure 1). Thus, in vivo acidification within the uterus would be expected to impair contractility and labor.
Although the numbers are not large, it appears that myometrial lactic acidosis is not the explanation for failure of induction of labor; this group of women had similar myometrial pH and oxygen saturation levels to the nonlaboring and normal contracting women (Table 1).
Our findings are of clear clinical significance. We have found a mechanism that can account for some dysfunctional labors, that is, hypoxic episodes reducing pH and the effectiveness of uterine contractions. Given the significant increase in morbidity associated with dystocia, its frequency in labors and its impact on cesarean rates, we suggest that there is an urgent need to extend this study. We would also suggest that our data shed light on why the current standard management for dysfunctional labor, that is, oxytocin infusion, can be ineffective in many labors and has not reduced the cesarean delivery rate. If oxytocin initially stimulates contractions, this may further worsen the myometrial blood supply, thereby lowering the local oxygen supply and increasing lactic acid accumulation, lowering the capillary pH, and ultimately reducing the force of the contractions. We have previously found that when oxytocin-stimulated myometrium was acidified, force fell markedly, consistent with this suggestion.17 Dysfunctional labor is a common cause of emergency cesarean delivery.4 The effect of dysfunctional labor on the cesarean delivery rate is compounded by patient dissatisfaction with the available management for dysfunctional labor, leading to an increasing demand for elective cesarean delivery in subsequent pregnancies.1–3 This seems likely to continue until we have a better understanding of the mechanisms producing dysfunctional labor and new therapeutic strategies. There is an urgent need to further investigate the role of myometrial pH in labor.
1.Behague DP, Victora CG, Barros FC. Consumer demand for caesarean sections in Brazil: informed decision making, patient choice, or social inequality? A population based birth cohort study linking ethnographic and epidemiological methods. BMJ 2002;324:942–5.
2.Belizan JM, Althabe F, Barros FC, Alexanser S. Rates and implications of caesarean sections in Latin America: ecological study. BMJ 1999;319:1397–402.
3.Langer A, Villar J. Promoting evidence based practice in maternal care. BMJ 2002;324:928–9.
4.Thomas J, Paranjothy S. R. C. O. G. Clinical Effectiveness Support Unit. National Sentinel Caesarean Section Audit Report. 1-1-2001. London (UK): RCOG Press; 2001.
5.Saunders NS, Paterson CM, Wadsworth J. Neonatal and maternal morbidity in relation to the length of the second stage of labour. Br J Obstet Gynaecol 1992;99:381–5.
6.Chelmow D, Kilpatrick SJ, Laros RK. Maternal and neonatal outcomes after prolonged latent phase. Obstet Gynecol 1993;81:486–91.
7.Steer PJ, Carter MC, Beard RW. The effect of oxytocin infusion on uterine activity levels in slow labour. Br J Obstet Gynecol 1985;92:1120–6.
8.Fraser W, Vendittelli F, Krauss I, Breart G. Effects of early augmentation of labour with amniotomy and oxytocin in nulliparous women: a meta-analysis. Br J Obstet Gynaecol 1998;105:189–94.
9.Sadler LC, Davison T, McCowen LM. A randomised controlled trial and meta-analysis of active management of labour. BJOG 2000;107:909–15.
10.Blanch G, Lavender T, Alfirevic Z, Walkinshaw S. Dysfunctional labour: a randomised trial. Br J Obstet Gynaecol 1998;105:117–20.
11.Taggart MJ, Wray S. Hypoxia and smooth muscle function: key regulatory events during metabolic stress [review]. J Physiol Lond 1998;509:315–25.
12.Peebles DM, Edwards AD, Wyatt JS, Bishop AP, Cope M, Delpy DT, et al. Changes in human fetal cerebral hemoglobin concentration and oxygenation during labor measured by near-infrared spectroscopy. Am J Obstet Gynecol 1992;166:1369–73.
13.Brar HS, Platt LD, DeVore GR. Qualitative assessment of maternal uterine and fetal umbilical artery blood flow and resistance in laboring patients by Doppler velocimetry. Am J Obstet Gynecol 1988;158:952–6.
14.Larcombe-McDouall JB, Buttell N, Harrison N, Wray S. In vivo pH and metabolite changes during a single contraction in rat uterine smooth muscle. J Physiol Lond 1999;518:783–90.
15.Monir-Bishty E, Pierce SJ, Kupittayanant S, Shmygol A, Wray S. The effects of metabolic inhibition on intracellular calcium and contractility of human myometrium. Br J Obstet Gynaecol 2003;110:1050–6.
16.Parratt J, Taggart M, Wray S. Abolition of contractions in the myometrium by acidification in vitro. Lancet 1994;344:717–8.
17.Pierce SJ, Kupittayanant S, Shmygol T, Wray. The effects of pH change on Ca(++) signaling and force in pregnant human myometrium. Am J Obstet Gynecol 2003;188:1031–8.
18.Parratt JR, Taggart MJ, Wray S. Functional effects of intracellular pH alteration in the human uterus: simultaneous measurements of pH and force. J Reprod Fertil 1995;105:71–5.
19.Taggart M, Wray S. Simultaneous measurement of intracellular pH and contraction in uterine smooth muscle. Pflugers Arch 1993;423:527–9.
20.Sjostedt S. Acid-base balance of arterial blood during pregnancy, at delivery, and in the puerperium. Am J Obstet Gynecol 1962;6:775–9.
21.Cerri V, Tarantini M, Zuliani G, Schena V, Redaelli C, Nicolini U. Intravenous glucose infusion in labour does not affect maternal and fetal acid-base balance. J Matern Fetal Med 2000;9:204–8.
22.Nordstrom L, Achanna S, Naka K, Arulkumaran S. Fetal and maternal lactate increase during active second stage of labour. BJOG 2001;108:263–8.
23.Katz M, Lunenfeld E, Meizner I, Bashan N, Gross J. The effect of the duration of the second stage of labour on the acid-base state of the fetus. Br J Obstet Gynaecol 1987;94:425–30.
24.Griess FCJr. Effect of labour on uterine blood flow: observations on gravid ewes. Am J Obstet Gynec 1965;93:917–23.
25.Brinkman CR. Circulation in the pregnant uterus. In: Carsten ME, Miller JD, editors. Uterine function: molecular and cellular aspects. New York (NY): Plenum Press; 1990. p. 519–37.
26.Wray S. The effects of metabolic inhibition on uterine metabolism and intracellular pH in the rat. J Physiol Lond 1990;423:411–23.
27.Anai T, Urata K, Tanaka Y, Miyakawa I. Pregnancy complicated with lactate dehydrogenase M-subunit deficiency: the first case report. J Obstet Gynaecol Res 2002;28:108–11.
28.Taggart MJ, Burdyga T, Heaton C, Wray S. Stimulus-dependent modulation of smooth muscle intracellular calcium and force by altered intracellular pH. Pflugers Arch 1996;432:803–11.