Analgesic and Hemodynamic Effects of Intrathecal Clonidine as the Sole Analgesic Agent during First Stage of Labor : A Dose-Response Study

After institutional review board approval and written informed consent, 36 women at term pregnancy (American Society of Anesthesiologists physical status I or II; gravida 1 or 2) were recruited for this double-blind prospective study. Only patients with singleton pregnancies and vertex presentation who were in established labor and requesting analgesic medication, with a cervical dilatation < 6 cm and normal FHR tracings were enrolled. Each patient was randomized when she requested labor analgesia to receive one of the three intrathecal study solutions: 50, 100, or 200 [micro sign]g clonidine (Boehringer Ingelheim, Ingelheim, Germany) in a total volume of 4 ml. The solutions were prepared using 0.9% saline by the hospital pharmacy and coded. Solution of clonidine in saline is isobaric at room temperature but slightly hypobaric compared with cerebrospinal fluid (CSF) at body temperature. [16] The anesthesiologist performing the block and the investigator were blinded to the clonidine dose administered.

Each patient received an intravenous bolus of at least 10 ml/kg Ringer's solution before initiation of analgesia, which was performed using the combined spinal epidural technique. With the patient in the sitting position, the epidural space was located at the L2-L3 or L3-L4 interspace with an 18-gauge epidural needle using the loss of resistance to saline technique. A 26-gauge Whitacre spinal needle (Espocan, Braun, Melsungen, Germany) was inserted through the epidural needle, clear free-flowing CSF was identified, and the study solution was injected through the spinal needle. The spinal needle was removed, and a 20-gauge epidural catheter was inserted 3 cm into the epidural space. To detect accidental intravascular or intrathecal placement of the catheter, aspiration from the epidural catheter was performed, but no epidural local anesthetic test dose was administered at this time. The epidural catheter was secured, and the patient positioned supine with slight head elevation and left uterine displacement. Time elapsed from intrathecal drug administration until the parturient was positioned in bed was around 5 min in all patients.

Pain at the peak of a contraction was assessed with a 10-cm visual analog pain score (VAPS) immediately before combined spinal epidural placement, every 5 min for 30 min and then every 15 min until request for further analgesic drug. Maternal blood pressure and heart rate were measured at the same intervals. Hypotension was defined as a systolic blood pressure < 90 mmHg or a 20% decrease from baseline and was treated immediately with a fluid bolus or intravenous ephedrine as required. At the same times, degree of somnolence was assessed using a four-point ordinal scale for sedation (1 = wide awake and alert; 2 = awake but drowsy, responding to verbal stimulus; 3 = arousable, responding to physical stimulus; and 4 = not arousable, not responding to physical stimulus). Other side effects such as pruritus, nausea, vomiting, or any other discomfort were also recorded at these times. When additional analgesia was requested, patients received epidural bupivacaine as per usual clinical routine for the remainder of their labor and the study protocol and data collection during labor were terminated. We defined duration of analgesia as time from injection of the study solution until the patient requested additional analgesia.

All patients had continuous external electronic FHR and tocodynamometric monitoring throughout the study period. FHR and uterine activity tracings from 20 min before until initiation of epidural analgesia were analyzed by a perinatologist blinded to the dose of the study solution. Tracing analysis included baseline FHR, number of accelerations per 20 min, long-term variability, decelerations (early, late, and variable), and number of uterine contractions per 20 min. The requirement for oxytocin augmentation, an instrumented (forceps) or cesarean delivery, and neonatal Apgar scores were also recorded. Patients were seen daily for 5 days to inquire about any postpartum problems including headache.

Data are expressed as mean +/- SD or median and range, as indicated. Two way analysis of variance (ANOVA) for repeated measurements was used to assess changes over time within as well as between groups. For comparison of raw data between groups as well as duration of analgesia, one-way ANOVA was performed with post hoc analysis (Scheffe) to allow for multiple comparisons. Paired Student t test was performed to make single comparisons of pre- with posttreatment data from the FHR tracings. For evaluation of sedation, time from intrathecal drug administration until delivery and Apgar scores, the Kruskal-Wallis test with Bonferroni correction allowing for multiple comparisons was performed as indicated. Proportional data were evaluated using the chi-square test. A P value < 0.05 was considered significant.

Of the 36 patients enrolled in the study, 12 were assigned to each group. Participants in this study did not differ in most demographic variables, gestational age, parity, percentage receiving oxytocin, baseline VAPS, or cervical dilatation at the time when the intrathecal study solution was administered (Table 1). Patients in the 100-[micro sign]g group were found to be significantly taller (6 cm) than patients in the 50-[micro sign]g group, although this difference was irrelevant.

Intrathecal clonidine provided a significant reduction in VAPS for parturients in all three groups (Figure 1; P < 0.001). Onset of analgesia was similar among groups. Pain scores were significantly smaller 25 and 30 min after intrathecal clonidine administration in the 200-[micro sign]g group compared with those in the 50-[micro sign]g group (Figure 1). A reduction in pain score of 70% from baseline was reported by 5 of 12 patients in the 50-[micro sign]g group, 8 of 12 patients in the 100-[micro sign]g group, and 11 of 12 patients in the 200-[micro sign]g groups. This difference between groups was statistically significant (P < 0.05). Analgesic duration in the 50-[micro sign]g group was shorter than in the 100-[micro sign]g and 200-[micro sign]g groups (Figure 2). Blood pressure decreased significantly following clonidine injection in all groups (Figure 3A). Although onset of hypotension compared with baseline occurred earlier after higher clonidine doses, blood pressure was not significantly different among groups over the time. Maximum reduction of mean arterial blood pressure in individual patients occurred within 1 h after intrathecal clonidine administration in all cases (Table 2), and all episodes of decreased blood pressure were easily reversed by supplemental fluid and ephedrine administration. As indicated, maximum decrease in mean arterial blood pressure was significantly greater in the 200-[micro sign]g group than in the 50-[micro sign]g group (P < 0.05), and ephedrine administration was required significantly more often in the 200-[micro sign]g group than in the lower dose groups (P < 0.05;Table 2). Similarly, maternal heart rate decreased significantly from baseline in the 100-[micro sign]g and 200-[micro sign]g groups, but not in the 50-[micro sign]g group (Figure 3B). Again, there was no difference in heart rate over the time among groups. Maximum reduction in heart rate occurred up to 90 min after intrathecal clonidine (data not shown), but all episodes resolved spontaneously, and none of these parturients required atropine treatment.

Intrathecal clonidine produced sedation at all doses studied. Starting at 15 min after intrathecal clonidine injection, sedation scores were increased in a dose-dependent manner (Figure 4; P < 0.05). Although we did not actively assess motor function in this study, all patients could move their legs easily and reported normal sensation without numbness. Patients in all groups could micturate spontaneously during the period of clonidine analgesia. No patient was allowed to walk during the study period due to the need for continuous maternal and fetal monitoring for safety reasons. None of the parturients reported any nausea, vomiting, or itching, but dryness of the mouth was observed by 1 of 12, 2 of 12, and 3 of 12 patients in the 50-[micro sign]g, 100-[micro sign]g, and 200-[micro sign]g groups, respectively. Daily questioning for 5 days postpartum revealed no case of postdural puncture headache.

There was no significant difference among groups for number of participants whose labors were induced or required oxytocin augmentation (2–20 mU/min). There was no difference in fetal variables between observation periods before and after intrathecal clonidine administration (Table 3). All FHR decelerations prior to and after intrathecal clonidine administration were transient and no fetus experienced severe bradycardia. None of the FHR changes seen after intrathecal clonidine administration were considered clinically significant, and none resulted in intervention for fetal compromise. No episodes of uterine hyperstimulation or decreased uterine activity were noted after intrathecal clonidine. The progress of labor as indicated by the time of intrathecal injection to delivery (Table 4), was similar among groups. The number of operative deliveries was similar for the three groups (Table 4). Indications for cesarean section were cephalopelvic disproportion, slow progress of labor, abruptio placentae, and twisted umbilical cord. All operative and instrumental deliveries were performed after the analgesic effect of the clonidine dose had disappeared and epidural analgesia had been established. Apgar scores were within normal range in all neonates.

This is the first study showing that intrathecal clonidine by itself produces dose-dependent analgesia for pain during first stage of labor. The current study determined that a dose of 50 [micro sign]g intrathecal clonidine provides labor analgesia by itself, although degree and duration of analgesia were significantly greater with the larger doses, but also accompanied by hypotension.

Intrathecal and epidural administration of [Greek small letter alpha]^2-adrenergicagonists prolong and intensify labor analgesia from local anesthetics [17,18] and opioids. [12,13,19] However, with the exception of an early report describing epinephrine as a sole analgesic for labor, [20] no information has been available regarding the dose-response relationship of intrathecal [Greek small letter alpha]-adrenergic agonists in this clinical setting.

This is the first study to demonstrate that intrathecal clonidine alone produces dose-dependent analgesia for pain during the first stage of labor. We determined that a dose of 50 [micro sign]g intrathecal clonidine provides moderate analgesia to labor pain, although degree and duration of analgesia were significantly more satisfactory with 100 [micro sign]g and 200 [micro sign]g. The 200-[micro sign]g dose, however, did not offer advantages over 100 [micro sign]g with respect to duration or intensity of analgesia. We found a shorter duration of pain relief than in a study by Gautier et al., [12] wherein 30 [micro sign]g intrathecal clonidine alleviated labor pain for about 1 h. Different methodologies in pain assessment and labor management may account for these discrepancies. First, the degree of reduction in VAPS indicating sufficient pain relief is variable among different investigators. Gautier et al. [12] did not provide VAPS during the period of "adequate analgesia" of their patients, making a direct comparison difficult. Whereas a 50% reduction in VAPS has been used to indicate successful analgesia during labor, [10,18] other authors define criteria as a VAPS <or= to 25 mm. [13,15] Second, we measured time from intrathecal drug injection until additional analgesia was requested by the parturient rather than when it was administered. The continuous presence of the anesthesiologist with the patient in our study may have encouraged earlier reporting of pain. Third, oxytocin-augmented labor is known to be associated with more painful contractions, depending on the administered oxytocin dose. Unfortunately, Gautier et al. [12] do not provide these doses and therefore the possibility of different oxytocin regimens, which may account for different pain levels, cannot be excluded.

Intrathecal clonidine, used as the sole analgesic agent after cesarean section, [4] provided dose-dependent analgesia with 150 [micro sign] relieving postoperative pain for 6 h, compared with only 2.5 h with 200 [micro sign]g in our study for labor pain. It is possible that different qualities of pain may account for that difference, since labor pain increases with progression of labor, but postoperative pain has a tendency to decrease over time. As labor progresses and the nature of labor pain turns from visceral into somatic, the analgesic effect of opioids without local anesthetic effects decreases. [21] Although clonidine itself has weak local anesthetic properties in extremely high concentrations in vitro, [22] a direct inhibitory effect of clonidine on nerve conduction in clinical relevant doses used for spinal analgesia seems unlikely. Intrathecal clonidine by itself might, even at higher doses than we administered, therefore not be effective to treat more advanced labor pain since it has also failed to provide surgical anesthesia, despite doses up to 450 [micro sign]g. [23]

As expected, blood pressure decreased significantly over time in each group (Figure 3A). Although analysis of variance showed no difference among groups, maximum blood pressure decrease from baseline was significantly more profound with the highest dose and consequently, ephedrine treatment was required significantly more often to restore adequate blood pressure in this group (Table 2). Our data confirm results obtained by Gautier et al. [12] and Filos et al., [4] who found maximum decreases of blood pressure by 15% after 30 [micro sign]g intrathecal clonidine and 21% after 150 [micro sign]g, respectively. Neuraxial clonidine decreases blood pressure by inhibiting preganglionic sympathetic neural activity in the spinal cord. [24] The degree of hypotension after neuraxial clonidine seems to be associated with the level of injection, with more profound hypotension occurring after injection at thoracic levels, [25,26] close to the preganglionic sympathetic neurons. In addition, activation of postsynaptic [Greek small letter alpha]²receptors in the brainstem and peripheral presynaptic [Greek small letter alpha]²receptors reduce sympathetic drive.

The greater hypotension with the largest clonidine dose is possibly due to rostral spread in CSF to thoracic levels and to the brainstem. Since clonidine becomes slightly hypobaric at body temperature, [16] rostral spread might have occurred with the parturient in the sitting position for a few minutes after intrathecal injection. Although clonidine's lipophilicity can be held as an argument against rapid rostral spread in CSF, [7] the lipophilic opioid sufentanil ascends rapidly from lumbar to cervical dermatomes in dogs. [27] However, the hemodynamic depression after all doses of intrathecal clonidine during labor requires careful monitoring for the entire period of analgesia.

Since clonidine is a mixed [Greek small letter alpha]^1-[Greek small letter alpha]^2-adrenergicagonist, high clonidine doses cause peripheral vasoconstriction, which results in a U-shaped hemodynamic dose-response curve. [4,26] Although Filos et al. [4] reported hemodynamic stability with 300 [micro sign]g and 450 [micro sign]g intrathecal clonidine, we chose to use 200 [micro sign]g as our maximum dose, since intrathecal injected clonidine is rapidly absorbed into plasma, [28] which could yield plasma concentrations adequate to produce uterine artery vasoconstriction and decreased uterine blood flow. [29]

Heart rate also decreased significantly from baseline in the 100 [micro sign]g and 200 [micro sign]g groups (Figure 3B). Again, no differences between all groups were detected over the time. When administered after cesarean section, intrathecal clonidine minimally reduced heart rate in doses from 150 to 450 [micro sign]g. [4] Clonidine decreases heart rate by a presynaptic mediated inhibition of norepinephrine release and by direct depression of atrioventricular nodal conduction, [30] and this decrease correlates with plasma clonidine concentrations. [28] However, no parturient in our study experienced sustained bradycardia and all episodes resolved spontaneously without treatment.

Clonidine will most likely be administered intrathecally in combination with either opioids or local anesthetics to provide labor analgesia, due to a positive analgesic interaction of these drugs in animals and humans. [12,13,31] Although clonidine does not produce an additional hypotensive effect when combined with local anesthetics, [7] there is a potential for exacerbating hemodynamic depression from the combination of intrathecal clonidine with opioids. [13,32] Hemodynamic side effects of this combination therapy warrant further investigation.

We found dose-dependent sedation in parturients from intrathecal clonidine doses of 50–200 [micro sign]g, in agreement with studies by Filos et al. [4] This represents an [greek small letter alpha]^2-adrenergiceffect, since sedation from epidural clonidine can be reversed by the specific antagonist yohimbine in postoperative patients. [33] Experimental data show that the sedative-hypnotic effect of [greek small letter alpha]^2-adrenergicagonists is caused by actions on the locus ceruleus. [34] We did not measure oxygen saturation or arterial blood gases in our study because of clonidine does not cause respiratory depression by itself or in combination with opioids. [35,36]

In our study, doses up to 200 [micro sign]g intrathecal clonidine for labor analgesia in healthy parturients were not associated with any FHR abnormalities, fetal distress, increased rate of operative delivery, or adverse neonatal outcome. Uterine activity and progress of labor were unaffected by 50–200 [micro sign]g intrathecal clonidine, as times from intrathecal drug administration to delivery did not differ between groups and were similar to intrathecal sufentanil in other studies. [12,13] Nevertheless, the relatively small number of patients included in our study precludes the conclusion of safety for mother and neonate of intrathecal clonidine in obstetric analgesia.

In conclusion, 50–200 [micro sign]g intrathecal clonidine produces a dose-dependent reduction in VAPS during first stage of labor. Duration and quality of analgesia were more pronounced with 100 [micro sign]g and 200 [micro sign]g than with 50 [micro sign]g, but 200 [micro sign]g doses were also associated with a high incidence of hypotension. Further studies are warranted to evaluate the safety for the possible clinical use of intrathecal clonidine as the sole analgesic during labor.

The authors thank midwives Lubica Parancin and Stanja Velinovska and the whole midwife staff of Ulrike Schuster, Department of Obstetrics, University of Vienna, Australia, for their support and excellent technical assistance.