Epidural Neostigmine Combined with Sufentanil Provides Balanced and Selective Analgesia in Early Labor

After approval by the Clinical Research Practices Committee and obtaining informed consent, 125 American Society of Anesthesiologists physical status I and II healthy parturients whose gestational age was greater than 36 weeks and who requested epidural analgesia during labor were enrolled in the study. Lumbar epidural puncture was performed with an 18-gauge Tuohy needle, and an epidural catheter was inserted 4 cm at the L3–L4 level in all parturients. Exclusion criteria included parturients with obstetric complications (multiple pregnancy, premature labor, or non-vertex presentation) or contraindications to regional analgesia. Accidental dural puncture when the epidural was performed also excluded patients from participation in the study. All parturients were in the established first stage of labor (cervical dilatation inferior to 5 cm) when the epidural technique was performed, and all received an oxytocin infusion during the course of labor.

Pain was assessed with a 100-mm visual analog scale (VAS; 0–100). When the VAS score reached 30, a test dose of 3 ml of lidocaine 2% with epinephrine 1:200,000 was administered via  the catheter, and the parturients were randomly allocated to one of the study groups. Patients in the two first groups received sufentanil 10 μg (S10; n = 25) and sufentanil 20 μg (S20; n = 25), respectively. In the other groups, sufentanil 10 μg was combined with neostigmine 250 μg (S10 N250; n = 25), neostigmine 500 μg (S10 N500; n = 25), or neostigmine 750 μg (S10 N750; n = 25). Neostigmine was provided from the commercial solution of neostigmine methylsulfate (Prostigmine®, 0.5 mg/ml; Roche, Somerville, NJ), for which intrathecal safety assessment has been reported.7All the parturients received the first epidural bolus in a total volume of 12 ml. The data were collected by an anesthesiologist, either a resident or an attending blinded to the patient group and to the epidural drugs.

Maternal noninvasive blood pressure, heart rate, and oxygen saturation were monitored. Fetal heart rate was also continuously recorded on a cardiotocograph. Pain score (VAS) was recorded 3, 5, 10, 15, 20, and 30 min after the epidural injection. Level of sensory block was evaluated by using the ether test and motor block in the lower limb was evaluated by using a modified Bromage scale (1 = complete motor block; 2 = almost complete motor block: the patient is able only to move the feet; 3 = partial motor block: patient is able to move the knees; 4 = detectable weakness of hip flexion: the patient is able to raise the leg but is unable to keep it raised; 5 = no detectable weakness of hip flexion: the patient is able to keep the leg raised during 10 s at least; 6 = no weakness at all) were recorded 15 and 30 min after the epidural injection as well as maternal and fetal vital parameters. The duration of analgesia was considered as the time elapsed to the patient’s first request for further analgesia (new epidural injection). From that second epidural injection until delivery, the maintenance of analgesia was provided by continuous infusion with ropivacaine 0.1% (8 to 10 ml/h) and rescue doses as needed.

The total duration of the labor from the time of the first epidural injection until delivery, the mode of delivery (operative, vaginal instrumental, or spontaneous), and cervical dilatation at the time of second epidural injection were recorded. Maternal adverse events (nausea and vomiting, pruritus, sedation) and fetal side effects (bradycardia during labor, low Apgar score) were also closely monitored.

Results were expressed as mean ± SD, 95% confidence interval, or median (interquartile range), as indicated. Demographic data and continuous variables were compared among the groups by analysis of variance and repeated measures, followed, if appropriate, by multiple comparison with the Tukey HSD test (Statistica; StatSoft, Tulsa, OK). Comparison of ordinal categorical data among groups was realized by application of the Kruskal-Wallis test followed by the Wilcoxon rank sum test for multiple comparisons. Incidence of adverse effects and installation of satisfactory analgesia were compared among the groups by chi-square analysis corrected for multiple tests. A P < 0.05 was considered significant.

The number of patients assigned to the different groups was based on preliminary data based on the duration of analgesia resulting from a first epidural bolus of sufentanil 20 μg. A prospective sample size calculation (SigmaStat; SPSS Science Software, Surrey, UK) indicated that 20 subjects were required in each group to have 80% power to detect a 25% difference at α-level of 0.05 in the duration of analgesic effect.

One hundred and twenty-five patients were enrolled in the study. Variables measured included age, weight and parity were similar among the groups, as well as cervical dilatation at the first injection, and when a second epidural injection was required (table 1). Maternal hemodynamic variables did not differ between groups and did not change into the group after epidural injection. Epidural neostigmine, 250 to 750 μg, was devoid of significant effect on maternal blood pressure and heart rate during the first 30 min postinjection (table 2) as well as during the rest of labor (unshown data).

All epidural solutions provided efficient analgesia and significantly reduced the initial VAS. From a similar initial pain score, epidural sufentanil 10 and 20 μg provided a significant VAS reduction after 20 min, whereas such a significant effect was recorded after 15 min in the S10 N250 and S10 N 750 groups and after 10 min in the S10 N500 group (fig. 1). At 30 min postinjection, only the S10 group presented a pain score statistically higher than all the other groups that did not differ from the initial VAS within the group. To better compare the effectiveness of the various epidural solutions, we recorded the percentage of parturients presenting a satisfactory analgesia (defined as a VAS strictly <30) at different time scores postinjection (fig. 2). At first sight, the S20 and S10 N500 groups displayed a very similar time course of effectiveness. The duration of analgesia resulting from this first top-up dose was significantly longer in the S20, S10 N500, and S10 N750 groups than in the S10 and S10 N250 groups (table 3). Epidural neostigmine 250 μg did not prolong sufentanil duration of action (S 10–S10 N250, P = 0.15). Further, neostigmine 750 μg was no more effective than neostigmine 500 μg (S10 N500–S10 N750, P = 0.09). The level of sensory block assessed at 30 min after epidural injection was similar between all the groups (median value, T10; T9– 12), as was the degree of motor block. No significant motor impairment was observed (table 3).

Epidural administration of neostigmine neither significantly modified the total duration of labor nor interfered with the mode of delivery (table 4). No harmful fetal side effects, e.g. , bradycardia, were noted during labor and Apgar scores did not differ between the groups. Cesarean delivery was required for two parturients in the S10 group, two parturients in the S10 N500 group, and one parturient in the S10 N750 group.

Our results provide evidence that epidural neostigmine 500 μg combined with sufentanil 10 μg produces effective analgesia in early labor (VAS <30 within 10 min in 72% of parturients and within 15 min in 85% parturients, average duration of 119 min, confidence interval, 96–142 min) without motor block or other side effect in mother and fetus. These results match the current definition of selective analgesia during the first stage of labor, also called “mobile epidural.” The term, which is becoming popular, does not necessarily imply maternal ambulation but rather refers to a regional analgesic technique that interferes least with the normal mechanism of labor.1 

To meet such a goal, potent opioids such as fentanyl and sufentanil are mainly used by either the intrathecal or the epidural route. Most studies refer to the use of the combined spinal-epidural technique with the injection of intrathecal sufentanil,2which results in quick and almost complete pain relief (within 5 to 10 min) lasting approximately 90 min (SD, 40 min).10,11However, many anesthesiologists still prefer the traditional epidural technique, which is less expensive than the combined spinal-epidural technique and which does not involve an intentional dural puncture.3,12The minimum analgesic dose for epidural sufentanil (21 μg, confidence interval, 20.2–21.9 μg) provides an effective analgesia, similar in duration to that of intrathecal sufentanil in the first stage of labor (90 min; SD, 38 min).9When preceded by a test dose of lidocaine with epinephrine, the duration of analgesia increases to 138 min (SD, 50 min).3Among the differences between both routes, the onset of analgesia is slower with the epidural administration of opioids (between 10–20 min), but such concern is probably of little relevance in early labor, in contrast with advanced labor. Furthermore, the risk of harmful side effects, as well as nausea and pruritus, is lower after epidural than after intrathecal injection.12 

To achieve better pain control associated with a reduction of side effects, the principle of balanced analgesia has been applied in obstetric analgesia using spinal drugs combinations which target different mechanisms.4Neostigmine inhibits the breakdown of endogenous acetylcholine, and its intrathecal injection induces selective analgesia devoid of sympathetic and motor block. However, despite a positive interaction with spinal opioids during labor analgesia, the incidence of major gastrointestinal side effects (nausea and vomiting, diarrhea) consecutive to the intrathecal administration, even at doses as low as 10 μg, has precluded any routine use of the drug.11 

Recently, several studies in postoperative patients have reported an analgesic effect of epidural neostigmine that, interestingly, was devoid of unacceptable side effects,6and we made the first report about the use of epidural neostigmine in labor.8However, at doses up to 4 μg/kg, that provided analgesia and opioid-sparing effects in postoperative conditions involving somatic pain, we did not observe a significant benefit nor did we note relevant side effects in our parturients. One explanation might be that the dose we used was too low, especially in laboring parturients with potent painful stimuli from a visceral origin. In a previous study, Lauretti and Lima observed that intrathecal neostigmine was more effective to relieve somatic than visceral pain.13Effectively, in contrast with somatic fibers, the central terminal of visceral afferents display an extensive and deeply located intraspinal arborization,14and neostigmine is a quaternary ammonium compound, ionized at physiologic pH and with a low lipid solubility that crosses biologic membranes to a limited extent.15It is thus conceivable that the dose of neostigmine that really accessed the spinal cord was ineffective. Therefore, in this study, we decided to test higher doses of epidural neostigmine: 500 μg (equivalent to 6–7 μg/kg) and up to 750 μg (10 μg/kg). Besides, others have previously used such doses: Nakayama et al . found that a dose of 10 μg/kg of epidural neostigmine was necessary to prolong the duration of postoperative analgesia after abdominal hysterectomy; 5 μg/kg was ineffective.16 

In our parturients, the best effect was obtained with a dose of 500 μg combined with sufentanil 10 μg. A dose of 750 μg was no more efficient and even seemed to produce a plateau effect. However, such an observation might also result from a bias caused by the small number of parturients included.

At the effective dose, we have observed that the epidural combination of neostigmine with sufentanil decreases the previously established sufentanil ED⁵⁰by 50%. A positive interaction between cholinesterase inhibitors and opioid derivatives has been described in several studies involving visceral pain, particularly after intrathecal administration, both in gynaecologic surgery17,18and during labor,11although nausea limited clinical usefulness. Nelson et al . have found a 25% reduction of intrathecal sufentanil ED⁵⁰with 10 μg neostigmine, the combination of both resulting in approximately 100 min (SD, 40 min) of analgesia.11 

In agreement with our previous study and other published reports, we did not observe undesirable side effects after epidural neostigmine, even when increasing the doses. Particularly the lack of nausea and vomiting strongly contrasts with the reports about intrathecal administration in which side effects are correlated to the dose.6The role played by the meninges, which express both acetylcholinesterases and butyrylcholinesterases, probably prevents the diffusion of high doses of neostigmine into the cerebrospinal fluid as well as its spread to supraspinal levels.19The risk of maternal hypotension is not a concern with neostigmine. Further, after epidural administration the chemical characteristics of the drug certainly limit its transfer to the maternal circulation and hence to the fetus, decreasing the risk for direct and indirect fetal bradycardia, which was observed after intravenous administration of the drug to reverse muscle relaxants.15Finally, because muscarinic receptors mediate uterine tone,20although we did not observe bothersome side effects in parturients, a potential effect of epidural neostigmine on basal tone, intensity, and frequency of uterine contractions can not be excluded.

In conclusion, in the context of “mobile epidural” involving a selective block of pain by spinal opioids in the absence of sympathetic and motor block, epidural neostigmine represents a useful tool. Combination with epidural sufentanil provides very effective pain relief, similar in duration to what is usually reported with intrathecal analgesics. Results from this study show that a dose of neostigmine 6 to 7 μg/kg (around 500 μg) is mandatory to be effective in the first stage of labor.