Section Editor: David J. Birnbach.
The pain and stress of labor result in large increases in circulating catecholamines in the parturient [1-3]. Relief of pain, particularly with epidural analgesia, modifies this pattern, causing a decrease in epinephrine but no change in norepinephrine [4-6].
It has been suggested for decades that the increase in maternal catecholamines could have an effect on uterine activity because the smooth muscle of the uterus is endowed with alpha- and beta-adrenergic receptors [7,8]. Many previous investigations have assessed the effects of epinephrine or norepinephrine on uterine activity, and this work has been extensively reviewed [9-10]. In general, norepinephrine has been found to exhibit uterine-stimulating effects mediated by alpha-adrenergic receptors, whereas epinephrine has been shown to have widely varying effects depending on the concentrations tested. None of these studies, however, has closely modeled the clinically encountered situation. Methodological difficulties with previous work includes the use of very large doses of catecholamines; studying both pregnant and nonpregnant animals in a variety of estrous states; and testing epinephrine or norepinephrine alone, rather than in combination, as is encountered clinically. Moreover, although it is generally agreed that alpha-adrenergic agonists increase and beta-adrenergic agonists decrease uterine activity, epinephrine and norepinephrine are mixed agonists and can stimulate both receptor types at different concentrations .
The purpose of this study was to assess the effect of epinephrine and norepinephrine on uterine activity at physiologic concentrations and in combination. We further sought to model the effects of oxytocin augmentation and of pain relief (decreased epinephrine) on catecholamine-induced changes in uterine activity.
This study protocol was approved by the Harvard Medical Area Standing Committee on Animals. Pregnant Sprague-Dawley rats at 20-22 days' gestation (term = 22 days) were killed with intraperitoneal pentobarbital, and the uterine horns were removed. Fetuses and placentas were gently removed. Cross-sectional rings 3 mm wide were cut from the uterus and mounted in water-jacketed (37[degree sign]C) and continuously gassed (95% O2/5% CO2) tissue baths in pH 7.4 Krebs-Henseleit buffer (KHB; composition (in mM): Na+ 143, K+ 5.9, Ca2+ 2.6, Mg2+ 1.2, Cl- 128, H2 PO4- 2.2, HCO3- 24.9, SO4 (2-) 1.2). Uterine rings were positioned between a rigid stainless steel wire support and a similar wire connected to an isometric force transducer and pretensioned with 0.5 g. Spontaneous contractions were allowed to stabilize over 1 h, and the bathing buffer was changed every 15 min. Preliminary time control experiments with no further drug additions showed the model to exhibit stable uterine activity for at least 4 h after preparation in this manner.
Cumulative concentration-response relationships for epinephrine, norepinephrine, and their combination on uterine activity were performed over the range of 10-12 to 10-6 M. Drugs were added directly to the tissue baths at 10-min intervals. Rings were not sequentially studied under different conditions but were exposed to only one drug or drug combination. The preparation was not washed between doses, and uterine activity was not allowed to return to baseline between additions. Tissue viability was verified at the end of each concentration-response analysis by observing a return to baseline activity after washing the preparation with KHB.
Concentration-response curves for the individual catecholamines were repeated in the absence (control) or presence of the alpha-adrenergic antagonist, phentolamine (10-4 M) or the beta-adrenergic antagonist, propranolol (10-4 M). The ability of oxytocin to antagonize catecholamine-induced tocolysis was tested by incubating contracting uterine rings in KHB alone or in KHB with oxytocin 1 mU/mL for 10 min, then treating them with 10-9 M each epinephrine and norepinephrine. Finally, the effect of analgesia-mediated changes in catecholamines was simulated in washout experiments. After measuring baseline activity, 10-9 M norepinephrine and epinephrine were added, and activity was measured 15 min later. The preparations were then washed with plain KHB or buffer containing norepinephrine 10-9 M. Activity was recorded again 15 min later.
Uterine contractions were recorded by using a custom-designed digital data acquisition system (LabView Software 2.0; National Instruments, Austin, TX), integrated over time to yield total uterine activity (area under the force-time curve), and analyzed by comparing activity with a baseline before any interventions. Five-minute intervals were integrated after each drug addition after a stabilization period of 5 min. Effects of catecholamines on uterine activity and of antagonists on catecholamine responses were tested by using repeated-measures analysis of variance (ANOVA). Uterine activity was treated as the outcome measure, and drug and dose were used as the main effects in the model. Change in activity in oxytocin and washout experiments was tested by paired t-test. Statistical significance was assumed when P was <0.05.
Buffer solutions were prepared fresh daily from concentrated stock solutions. Epinephrine was obtained from American Regent Laboratories (Shirley, NY) as a 1-mg/mL solution, and norepinephrine was obtained from Abbott Laboratories (Chicago, IL) as a 1-mg/mL solution. Catecholamines were diluted in distilled water to which ascorbic acid 10-4 M was added to prevent auto-oxidation. This catecholamine buffer alone had no effect on uterine activity. All other chemicals were purchased from Sigma (St. Louis, MO) or ICN (Irvine, CA) as reagent grade or purer and diluted in distilled water.
The effects of individual and mixed catecholamines on uterine activity are shown in Figure 1. Epinephrine caused dose-dependent decreases in uterine activity, with the peak effect at 10-7 M, followed by an increase in activity at higher concentrations. Norepinephrine caused a dose-dependent increase in activity in the same concentration range, followed by a decrease at larger doses. The combination of the two catecholamines produced a decrease in activity similar to the effect of epinephrine alone, which was maximal in the clinical range (10-9 to 10-8 M). For each catecholamine or combination tested, the effect of dose on uterine activity was statistically significant (ANOVA, P < 0.05). Over the entire dose range analyzed together, differences between catecholamines and their combination were statistically significant (ANOVA and Scheffe's test, P < 0.0001).
The effect of phentolamine on the catecholamine responses is shown in Figure 2. alpha-Adrenergic blockade antagonized the uterotonic effect of norepinephrine (Figure 2A) (P < 0.0001), which then produced concentration-dependent uterine relaxation. Phentolamine had no effect on epinephrine-induced tocolysis (Figure 2B) (ANOVA for 10-12 to 10-8, P = 0.10) but blocked the increase in activity seen at larger doses (ANOVA for 10-7 to 10-6, P = 0.05). beta-Adrenergic blockade with propranolol enhanced norepinephrine-induced uterine stimulation (Figure 3A) (P = 0.0051) and blocked epinephrine-induced uterine relaxation (Figure 3B) (P = 0.0058).
Oxytocin (1 mU/mL) had little effect on baseline uterine activity, but it blocked the effect of simulated maternal catecholamines in labor, 10-9 M epinephrine and norepinephrine (Figure 4).
When catecholamines were added at 10-9 M each, followed by washout with KHB alone, uterine activity increased approximately to baseline (Figure 5). If norepinephrine 10-9 M remained in the bath but epinephrine was washed out, a greater increase in activity was observed (Figure 5). Both changes in activity with washing were significant (ANOVA, P = 0.02).
The results of this investigation suggest that concentrations of catecholamines encountered in laboring women are significantly tocolytic. Endogenous catecholamines measured during parturition have yielded mean values of approximately 100-300 pg/mL for epinephrine and 100-900 pg/mL for norepinephrine, with some women having levels several times as high [1-6]. These concentrations are equivalent to approximately 1-10 nM (10-9 to 10-8 M). The present study indicates that such levels of norepinephrine and epinephrine can reduce uterine activity by at least one-third below baseline (Figure 1).
Many previous studies have investigated the complex role of adrenergic agonists in modulating uterine activity (for review see [9-10]), but this is the first to investigate the effect on gravid myometrium of physiologic concentrations of catecholamines in combination, as occurs naturally in parturients. Our results confirm that nanomolar norepinephrine is uterotonic and that the effect is mediated by alpha-adrenergic receptors because the effect was blocked by phentolamine. Norepinephrine clearly has some beta-adrenergic activity at this concentration, however, because propranolol enhanced its stimulatory effect. Epinephrine in the nanomolar range was tocolytic, and the effect was blocked by beta-adrenergic blockade. Epinephrine apparently had little alpha-adrenergic activity at the very low concentrations tested in our preparation because phentolamine did not modify its effect on uterine contraction, except at supraphysiologic (micromolar) concentrations. Most significantly, the combination of the two catecholamines had an effect similar to epinephrine alone in the physiologic range, which indicates a beta-adrenergic predominance.
A potential weakness of our model is that we studied full-thickness cross-sectional myometrial rings, rather than isolated circular or longitudinal layers. Some investigators have found differences in adrenergic sensitivity among the different uterine muscle layers of the rat, but at very high concentrations of norepinephrine and epinephrine [12,13]. The orientation of muscle fibers in our rings suggests that the contractions were primarily due to circular muscle, although the contribution of short segments of longitudinal muscle cannot be excluded. In this sense, our preparation perhaps more closely reflects the less distinct layers of the human myometrium . Moreover, we have found this preparation to closely mirror clinical experience in its response to several other uterotonic and tocolytic drugs .
A second possible criticism of our methods is the use of near-term but nonlaboring rats. There is evidence for down-regulation of beta-adrenergic receptor number during the progesterone withdrawal of parturition, as well as a decreased adenylate cyclase activity [15,16]. We opted to study rats just before delivery because of the methodological difficulties in killing the animals at the exact time of spontaneous labor. Furthermore, we did allow the myometrial preparations to contract for >1 h in the absence of progesterone before studying the catecholamine responses, in an attempt to model spontaneous labor. Nonetheless, our preparation may overestimate the tocolytic effect of epinephrine.
Taken together, these limitations may explain some of the variability of the responses we observed, especially in the uterine-stimulating effect of norepinephrine (Figure 1, Figure 2, and Figure 3). The use of traditional cumulative dose-response curves, in which the preparation does not return to baseline before adding the next concentration of drug, could have confounded the results in some cases. For example, the response to one concentration might not be independent of the previous response, or there might be fatigue or other time-dependent variation in the responses.
Extreme caution should be exercised when attempting to infer clinical relevance from laboratory investigations such as ours. However, our results do mirror the clinical observations that oxytocin can antagonize catecholamine-induced tocolysis (Figure 4) and that maneuvers that reduce epinephrine, such as regional analgesia, can significantly increase uterine activity (Figure 5). The effectiveness of oxytocin has been appreciated for decades, but the effect of regional analgesia on uterine activity has been more controversial. Previous work has demonstrated that administration of lumbar epidural analgesia during labor reduces plasma epinephrine but has little effect on norepinephrine [4-6]. Our results indicate these catecholamine changes should result in increased uterine activity. Certainly, some evidence supports a positive role for epidural analgesics in managing the progress of painful, exhaustive labors . Other work has suggested that analgesia after intrathecal opioids may significantly enhance uterine contraction and accelerate cervical dilation . There is little controversy, however, that substantially increased maternal catecholamines correlate with slower progress of labor and nonreassuring fetal status [2,3,5,19]. Furthermore, several investigations have demonstrated the effectiveness of propranolol in increasing uterine activity in normal parturients  and in the management of dysfunctional labor [21,22].
Catecholamines are not routinely measured in laboring women, and the large variation in reported levels [1-6] makes it difficult to predict which parturients might be suffering from catecholamine-induced tocolysis. Lederman et al. [2,3] suggested that maternal anxiety correlates with higher circulating catecholamine levels and slower progress of labor, and others have clearly related pain to increased levels of endogenous epinephrine [4-6]. The results of the present investigation, as well as some clinical data , suggest that techniques that either reduce (regional analgesia) or antagonize (propranolol, oxytocin) circulating epinephrine may be beneficial in patients whose labors are progressing slowly.
Our results may also help to explain the clinical observation that uterine hypertonus and fetal bradycardia may occasionally complicate the onset of regional analgesia [23-25], even in the absence of maternal hypotension. Although controversial, this complication may occur more often when the combined spinal-epidural technique of analgesia is used. The very rapid onset of pain relief that follows the injection of intrathecal opioids in this technique has been implicated in the pathogenesis of several cases of "rock hard" uterine contractions and fetal bradycardia . Interestingly, Cascio et al.  reported a faster decrease in plasma epinephrine in parturients who received intrathecal opioids versus those who received epidural bupivacaine.
In summary, our results suggest that epinephrine and norepinephrine exert competing effects on uterine contractions and that epinephrine's effect predominates in clinically encountered concentrations. A reduction in circulating epinephrine, such as accompanies regional analgesia, may result in a substantial increase in uterine activity.
1. Lederman RP, McCann DS, Work B, Huber MJ. Endogenous plasma epinephrine and norepinephrine in last-trimester pregnancy and labor. Am J Obstet Gynecol 1977;129:5-8.
2. Lederman RP, Lederman E, Work BA, McCann DS. The relationship of maternal anxiety, plasma catecholamines, and plasma cortisol to progress in labor. Am J Obstet Gynecol 1978;132:495-500.
3. Lederman RP, Lederman E, Work BA, McCann DS. Anxiety and epinephrine in multiparous women in labor: relationship to duration of labor and fetal heart rate pattern. Am J Obstet Gynecol 1985;153:870-7.
4. Neumark J, Hammerle AF, Bieflemayer CH. Effects of epidural analgesia on plasma catecholamines and cortisol in parturition. Acta Anaesthesiol Scand 1985;29:555-9.
5. Shnider SM, Abboud TK, Artal R, et al. Maternal catecholamines decrease during labor after lumbar epidural anesthesia. Am J Obstet Gynecol 1983;147:13-5.
6. Abboud T, Artal R, Sarkis F, et al. Sympathoadrenal activity, maternal, fetal, and neonatal responses after epidural anesthesia in the preeclamptic patient. Am J Obstet Gynecol 1982;144:915-8.
7. Zuspan FP, Cibils LA, Pose SV. Myometrial and cardiovascular responses to alterations in plasma epinephrine and norepinephrine. Am J Obstet Gynecol 1962;84:841-51.
8. Kaiser IH, Harris JS. The effect of adrenaline on the pregnant human uterus. Am J Obstet Gynecol 1950;59:775-84.
9. Marshall JM. Effects of catecholamines on the smooth muscle of the female reproductive tract. Annu Rev Pharmacol 1973;13:19-32.
10. Wray S. Uterine contraction and physiological mechanisms of modulation. Am J Physiol 1993;264:C1-18.
11. Weiner N. Norepinephrine, epinephrine, and the sympathomimetic amines. In: Gilman AG, Goodman LS, Rall TW, Murad F, eds. The pharmacological basis of therapeutics. 7th ed. New York: Macmillan, 1985:145-80.
12. Chow EHM, Marshall JM. Effects of catecholamines on circular and longitudinal muscle of the rat. Eur J Pharmacol 1981;76:157-65.
13. Chernaeva L. Adrenergic sensitivity of circular and longitudinal muscles of pregnant rat myometrium. Methods Find Exp Clin Pharmacol 1984;6:73-8.
14. Segal S, Csavoy AN, Datta S. Placental tissue enhances uterine relaxation by nitroglycerin. Anesth Analg 1998;86:304-9.
15. Maltier JP, Benghan-Eyene Y, Legrand C. Regulation of myometrial beta2-adrenergic
receptors by progesterone and estradiol-17 beta in late pregnant rats. Biol Reprod 1989;40:531-40.
16. Cohen-Tannoudji J, Vivat V, Heilmann J, et al. Regulation by progesterone of the high-affinity state of myometrial beta-adrenergic receptor and of adenylate cyclase activity in the pregnant rat. J Mol Endocrinol 1991;6:137-45.
17. Maltau JM, Andersen HT. Epidural anaesthesia as an alternative to caesarean section in the treatment of prolonged, exhaustive labour. Acta Anaesthesiol Scand 1975;19:349-54.
18. Tsen L, Segal S, Datta S. Combined spinal/epidural vs. epidural analgesia: effects on progression and outcome of labor [abstract]. Anesthesiology 1996;85:A856.
19. Hollmen AI, Jouppila R, Jouppila P, et al. Effect of extradural analgesia using bupivacaine and 2-chloroprocaine on intervillous blood flow during normal labour. Br J Anaesth 1982;54:837-42.
20. Barden TP, Stander RW. The effects of propranolol upon uterine activity in term human pregnancy. J Reprod Med 1969;2:188-95.
21. Mitrani A, Oettinger M, Abinader EG, Klein A. Use of propranolol in dysfunctional labor. Br J Obstet Gynaecol 1975;82:651-5.
22. Sanchez-Ramos L, Quillen MJ, Kaunitz AM. Randomized trial of oxytocin alone and with propranolol in the management of dysfunctional labor. Obstet Gynecol 1996;88:517-20.
23. Clarke VT, Smiley RM, Finster M. Uterine hyperactivity after intrathecal injection of fentanyl for analgesia during labor: a cause of fetal bradycardia? [letter]. Anesthesiology 1994;81:1083.
24. Nielsen PE, Erickson JR, Abouleish EI, et al. Fetal heart rate changes after intrathecal sufentanil or epidural bupivacaine for labor analgesia: incidence and clinical significance. Anesth Analg 1996;83:742-6.
25. Albright GA, Forster RM. Does combined spinal-epidural analgesia with subarachnoid sufentanil increase the incidence of emergency cesarean delivery? Reg Anesth 1997;22:400-5.
26. Cascio M, Pygon B, Bernett C, Ramanathan S. Labour analgesia with intrathecal fentanyl decreases maternal stress. Can J Anaesth 1997;44:605-9.