Intrathecal sufentanil provides rapid-onset and complete analgesia for the first stage of labor. The dose required to produce this effect can be associated with maternal respiratory depression, hypotension, nausea, or pruritus. Because clonidine potentiates the analgesic effects of opioids without increasing their side effects, the authors wanted to determine the efficacy of low doses of intrathecal clonidine (15 and 30 microg) combined with sufentanil.
Ninety-eight parturient requesting labor analgesia were studied. In a combined spinal-epidural technique, patients were randomly assigned to receive one of the following intrathecal solutions: either 15 microg clonidine (n = 10); 30 microg clonidine (n = 10); 2.5 microg sufentanil (n = 13); 5 microg sufentanil (n = 13); 2.5 microg sufentanil and 15 microg clonidine (n = 13); 2.5 microg sufentanil and 30 microg clonidine (n = 13); 5 microg sufentanil and 15 microg clonidine (n = 13); or 5 microg sufentanil and 30 microg clonidine (n = 13). Visual analog scores for pain, blood pressure, heart rate, sensory levels, incidence of nausea and pruritus, and motor blockade, and maternal and cord blood concentrations of clonidine were recorded.
Patients receiving 30 microg intrathecal clonidine with 2.5 or 5 microg intrathecal sufentanil had significantly longer-lasting analgesia (145 +/- 36 and 145 +/- 43 min vs. 104 +/- 35 for those receiving 5 microg intrathecal sufentanil alone). Clonidine levels were undetectable in maternal serum.
Thirty micrograms of intrathecal clonidine combined with 2.5 or 5 microg intrathecal sufentanil significantly increased the duration of analgesia during the first stage of labor without adverse maternal or fetal effects.
THE combined spinal epidural technique using the intrathecal administration of sufentanil to parturients is becoming routine. The intrathecal administration of 10 micro gram sufentanil provides rapid-onset, potent, and reliable analgesia lasting 1–3 h without motor blockade during the first stage of labor. [1–4]At this dose, however, pruritus and nausea and moderate maternal hypotension were observed. Extensive dermatomal spread suggesting that early-onset respiratory depression could occur was also noted. Further, the intrathecal administration of 10 micro gram sufentanil has been shown to depress the maternal central respiratory drive for at least 60 min.* Although decreasing the dose of intrathecal sufentanil may decrease side effects, it may also decrease the duration and reliability of analgesia. The effective dose in 95% of the patients (ED95) of intrathecal sufentanil for labor analgesia has been shown to be between 7.5 and 11.5 micro gram. An alternative would be to add an agent that acts synergistically with sufentanil and then increases the duration of analgesia without increasing side effects. Clonidine may be such an agent if large doses are avoided. Doses of 100 and 200 micro gram clonidine given intrathecally combined with 2 micro gram intrathecal sufentanil were reported by Chiari et al. ** to significantly exacerbate maternal hypotension and ephedrine requirements.
The present study was designed to identify the minimal effective dose of intrathecal clonidine that can be combined with sufentanil to alleviate labor pain.
Materials and Methods
The protocol was approved by the Clinical Research Practices Committee and informed consent was obtained from each patient. Ninety-eight parturients classified as American Society of Anesthesiologists physical status I or II who had singleton pregnancies in the vertex presentation were enrolled in the study. Uncomplicated course of pregnancy and normal fetal heart rate at the admission monitoring were mandatory inclusion criteria. All patients were either nulliparous or uniparous and in active labor with cervical dilation less than 5 cm when analgesia was requested.
Before spinal injection, patients were placed in the lateral position to achieve left uterine displacement, and an intravenous bolus of at least 250 ml lactated Ringer's solution was given. After these precautions, a 8-cm, 16-gauge Tuohy needle (Portex, England) was inserted into the epidural space at the approximate L3-L4 interspace by a loss-of-resistance technique. A 11.9-cm, 29-gauge Quincke needle (B-D) was inserted through the epidural needle into the subarachnoid space, and a 1.2-ml intrathecal injection was administered. The spinal needle was removed, and an epidural catheter inserted 3 cm into the epidural space.
Patients received, as determined by a table of random numbers, one of the following eight intrathecal solutions: 15 micro gram clonidine alone (C 15); 30 micro gram clonidine alone (C 30); 2.5 micro gram sufentanil alone (S 2.5); 5 micro gram sufentanil (S 5); 2.5 micro gram sufentanil combined with 15 micro gram clonidine (S2.5C15); 2.5 micro gram sufentanil and 30 micro gram clonidine (S2.5C30); 5 micro gram sufentanil and 15 micro gram clonidine (S5C15); or 5 micro gram sufentanil and 30 micro gram clonidine (S5C30). The study solutions were prepared by an anesthesiologist not involved in the patients' care, and the patient and the anesthesiologist who delivered spinal analgesia were blinded to the study solutions.
Pain was assessed with a 10-cm linear visual analog score immediately before spinal-epidural placement and 5, 10, 20, 30, 40, 60, 120, 180 min after intrathecal injections. Additional visual analog scores were obtained when additional medication was requested. When the patient requested additional analgesia, 10 ml of a solution containing 12.5 mg bupivacaine, 12.5 micro gram epinephrine, and 5 micro gram sufentanil were given by the anesthesiologist in charge through the epidural catheter. Subsequent administrations were done by the nurses using a patient-controlled epidural analgesia device (PMP; Abbott Laboratories, North Chicago, IL). The duration of the analgesia provided by the intrathecal solutions was taken as the time elapsed between the intrathecal injection and the first additional analgesic request.
At the time of each visual analog score assessment, patients subjectively rated nausea, sedation, and pruritus. Motor blockade of the abdominal musculature was assessed objectively by the investigators using a RAM test (the parturient is asked to come slowly from a supine to a sitting position). Her score is 100 if she succeeds with her hands clasped behind the head; 80 if she succeeds by extending the arms; 60 if she only lift the scapulae from the bed; 40 if she can only lift the shoulder; 20 if the observer can only feel an increase in tension in the abdomen. This test was performed before and 15 min after intrathecal injection. Cephalad level of anesthesia by sensory changes to pinprick were ascertained 15 and 30 min after intrathecal injection.
Blood pressure and heart rate were monitored with a noninvasive monitor (Cardiocap Datex, Helsinki, Finland) every 5 min (at least three times) before intrathecal injection and at 5, 10, 15, 20, 30, 45, 60, 90, 120, 180 min afterward. Baseline values for blood pressure and heart rate were the mean of the three measurements performed before intrathecal injection. Respiratory rate was recorded every 15 min throughout the study, and oxyhemoglobin saturation was monitored by pulse oximetry in all patients. Fetal heart rate and uterine activity were monitored continuously throughout labor. The duration of labor, mode of delivery, and neonatal Apgar scores were recorded. All patients were seen 5 days after delivery and questioned for the occurrence of a postdural puncture headache.
Maternal blood samples for clonidine assessment were obtained in 10 parturients 15, 30, and 60 min after the intrathecal administration of 30 micro gram clonidine alone or combined with sufentanil (two in group C30, four in group C30 S5, and four in group C30 S2.5). Cord blood samples for clonidine assessment were obtained in 10 corresponding neonates. Clonidine concentrations were determined using a radioimmunoassay method. The sensitivity of this assay is 0.1 ng/ml with a coefficient of variation < 15%. 
Data are presented as means +/- SD. The statistical analysis of the duration of analgesia assessed in the different groups was performed as follows: The normality of the distributions in the eight groups was assessed using the Shapiro-Wilk's test in the PROC UNIVARIATE procedure in SAS (Windows version, 6.12 Orlando II; SAS Institute, Cary, NC). The distribution was normal in seven of the eight groups. The fact that for one group (C15) the null hypothesis of normality was rejected does not exclude that in general the distribution of the input data values from the primary variable was normal. Under the assumption of normal distributions, the use of parametric one-way analysis of variance technique (ANOVA) could be justified. After a significant overall probability value after a one-way ANOVA test, multiple comparisons were applied using the Student-Newman-Keuls procedure. For completeness, a nonparametric alternative for one-way ANOVA, the Kruskall-Wallis test, was applied along with a nonparametric multiple comparison procedure, the Wilcoxon rank sum test. The nonparametric tests, however, confirmed the findings of the parametric approach. Categorical variables were analyzed by chi-square test or the Fisher's exact test, as appropriate. A P value <or= to 0.05 was considered significant.
Of the 98 patients enrolled in the study, 97 completed the study protocol and were included in the data analysis. One patient receiving 5 micro gram sufentanil gave birth before requesting additional analgesia and was excluded. Four parturients underwent cesarean section for nonprogression of labor, but this occurred after the first epidural injection and their data were considered.
Treatments groups did not differ in demographic variables, visual analog pain scores, parity, percentage receiving oxytocin, or cervical dilatation at the time of entry into the study. The duration of labor after intrathecal injections and the method of delivery were similar among the different groups (Table 1).
Intrathecal clonidine used as the sole analgesic agent at the dose of 15 micro gram provided adequate analgesia (100 and 84 min, respectively) in only 2 of the 10 patients considered. Clonidine at 30 micro gram was sufficient to alleviate labor pain for more than 30 min in 80% of the patients considered. These results are comparable to those obtained with 2.5 micro gram sufentanil (Table 2, Figure 1). The combination of 30 micro gram clonidine with 2.5 micro gram or 5 micro gram sufentanil significantly improved the duration of analgesia when compared with 5 micro gram sufentanil alone (Table 2, Figure 1; P <0.05). The association of 15 micro gram clonidine with 2.5 micro gram or 5 micro gram sufentanil produced analgesia comparable to 5 micro gram sufentanil alone (Table 2, Figure 1). A similar median level of sensory loss to pinprick in the midclavicular line was observed in all patients in the study group (T5). The highest level recorded (T2) occurred in one patient of the group S2.5 C30.
Intrathecal sufentanil produced pruritus in most of the patients treated. Parturients who received clonidine alone did not suffer this side effect (Table 3). Pain relief was associated with light sedation in all the patients. Pain relief was also associated with a moderate decrease in maternal baseline arterial blood pressure and heart rate. There was, however, no differences among the treatment groups. All the episodes of decreased blood pressure were short and easily reversed by supplemental fluid administration. No patients required the administration of ephedrine. None of these episodes were associated with an abnormal fetal heart rate pattern.
None of the patients had nausea or detectable motor blockade. The groups did not differ in mean respiratory rate and in no parturient did respiratory rate decrease to < 12 breaths/min. Oxyhemoglobin saturation was > 95% in all patients at all times. Neonatal Apgar scores did not differ among groups, and no postdural puncture headaches were reported. The subsequent course of the parturients who required early epidural analgesics was uneventful.
No clonidine was detected in maternal serum 10, 30, and 60 min after intrathecal injections. No clonidine was detected in the serum of the cord of the neonates immediately after delivery.
Our results show that the addition of 30 micro gram intrathecal clonidine significantly improves the analgesic efficacy of low doses of intrathecal sufentanil (2.5 and 5 micro gram) for the first stage of labor.
Because poor results were expected in patients receiving clonidine alone, only 10 patients were enrolled in each of these two groups. The efficacy of 30 micro gram intrathecal clonidine to alleviate labor pain reported in this study is somewhat surprising. Intrathecal and epidural clonidine, used as the sole analgesic agent, have already been demonstrated to alleviate postoperative pain after cesarean sections or abdominal surgery. [10,11]In these circumstances, however, the doses used were far higher than these used in the present study. Because labor pain may be as severe as postoperative pain, an explanation for this apparent discrepancy can be found in the pregnancy-induced physiologic changes. The end of pregnancy is associated with high levels of endorphins in humans, and naloxone sensitive autoanalgesia has been demonstrated in pregnant animals. A potentiation of endorphin-mediated autoanalgesia by the low doses of intrathecal clonidine may account for the results obtained in this study. Potentiation of “naloxone sensitive stress induced autoanalgesia” has recently been observed in animals after the administration of alpha2-adrenergic agonists. 
Intrathecal injection of sufentanil has been reported to provide excellent and rapid-onset analgesia without evidence of motor blockade for the first stage of labor. [1–4]Usually a dose of 10 micro gram is advocated because it produces complete analgesia lasting 1–3 h. In the present study, however, the dose of 5 micro gram intrathecal sufentanil was chosen because this is the standard dose at our institution and evidence suggests that 10 micro gram sufentanil may be too high. Courant et al.,*** in a dose-response study, showed that intrathecal sufentanil at doses of 1.5, 3, 5, or 10 micro gram provides profound analgesia without significant differences among the four doses. The results obtained in this study confirm the lack of major differences in analgesic efficacy between 2.5 and 5 micro gram intrathecal sufentanil.
The addition of 30 micro gram clonidine improves the analgesic efficacy of the low doses of intrathecal sufentanil. This observation is in accordance with the many experimental and clinical reports that show a positive interaction that is either synergistic or simply additive for opioids and alpha2-adrenergic agonists. [16,17]Interestingly, 30 micro gram clonidine combined with 2.5 micro gram sufentanil produces labor analgesia of similar duration to 200 or 100 micro gram clonidine combined with 2 micro gram sufentanil. * It indicates that 30 micro gram intrathecal clonidine is as effective as 100 or 200 micro gram clonidine in potentiating sufentanil analgesia. Before asserting the clinical utility of this combination, the exact incidence of the induced side effects should be established in studies including large populations. Despite the limited number of parturients included in our study, some information can be found concerning the hemodynamic side effects induced by the studied drug combinations. In contrast to the results of Chiari et al., * the intrathecal administration of 30 micro gram clonidine and 2.5 or 5 micro gram sufentanil did not exacerbate the decrease in maternal blood pressure associated with pain relief. Clonidine decreases arterial blood pressure mainly by a central sympatholytic effect. In our study, this action is unlikely to occur because no clonidine could be detected in the serum of the mothers. Nevertheless, because of the limited number of patients included in the different groups, a type II statistical error cannot be precluded. Consequently, the results obtained in this study do not unequivocally affirm that the low doses of clonidine and sufentanil used do not cause maternal hypotension.
Intrathecal clonidine is not the sole agent that can be used to potentiate the analgesic effects of sufentanil. Campbell et al. reported that 2.5 mg intrathecal bupivacaine significantly prolonged the duration of analgesia of 10 micro gram intrathecal sufentanil (114 +/- 26 min for sufentanil alone vs. 148 +/- 27 min for the combination). This was obtained without adverse maternal or fetal effects in 15 consecutive parturients. The results from our study, however, show that 30 micro gram clonidine affords a greater reduction in the dose of sufentanil required compared with 2.5 mg bupivacaine because a similar duration of analgesia is obtained when combining 30 micro gram clonidine with 2.5 or 5 micro gram sufentanil compared with combining 2.5 mg bupivacaine with 10 micro gram sufentanil (145 +/- 36 min and 145 +/- 43 vs. 148 +/- 27 min).
In conclusion, the administration of 30 micro gram intrathecal clonidine combined with 2.5 or 5 micro gram intrathecal sufentanil significantly increased the duration of analgesia during the early first stage of labor. Our results, however, do not determine the efficacy of this combination in advanced labor. Although hypotension may be a concern with the administration of intrathecal clonidine, our results, obtained in a limited number of patients, did not reveal any adverse maternal or fetal effects.
The authors thank G. Byttebier, M.Sc., for helpful suggestions regarding statistical analysis, C. Jones, M.B.B.S., for critical review of the manuscript, and M. Deboschere for technical assistance.
*Arkoosh V, Torjman MC, Montgomery OC, Leighton BL: Does intrathecal sufentanil depress the ventilatory response to CO2 in the parturient [Abstract]? Anesthesiology 1994; 81:A1137.
**Chiari A, Berger M, Lorber C, Gosch M, Klimscha W: Intrathecal sufentanil and clonidine for obstetric analgesia [Abstract]. Anesthesiology 1994; 81:A1141.
***Courant N, Raiga J, Monteillard C, Curt 1, Schoeffler P: intrathecal sufentanil for labor analgesia. A dose response study [Abstract]. Anesthesiology 1994; 81:A1142.