Mulroy, Michael F. MD; Greengrass, Roy MD; Ganapathy, Sugantha MD; Chan, Vincent MD; Heierson, Agneta PhD
Certain compounds, such as meperidine, have both opioid and local anesthetic properties  and can produce spinal anesthesia as well as postoperative analgesia [2,3]. Sameridine (N-ethyl-1-hexyl-N-methyl-4-phenyl-4-piperidine carboxamide hydrochloride, structurally similar to meperidine) is a new chemical compound that has both local anesthetic and analgesic properties. Animal experiments suggest a prolongation of analgesia after the resolution of motor block with sameridine. Preliminary clinical use for spinal anesthesia in human subjects undergoing knee arthroscopy demonstrated adequate surgical anesthesia . We conducted a multicenter, randomized, prospective, double-blinded, dose-ranging study of intrathecal sameridine compared with lidocaine when used for spinal anesthesia for hernia repair in healthy men. Inguinal hernia repair was chosen for its potential to demonstrate postoperative analgesia after resolution of spinal anesthesia.
The study was performed in 10 medical centers in the United States and Canada (Appendix 1 Table 6). The same protocol was used in all institutions, and institutional review board approval was obtained at each site. After informed consent, 193 male patients scheduled for elective unilateral inguinal hernia repair were enrolled in the study. Patients were ASA physical status I or II, aged 18-80 yr, with a body weight of 50-110 kg, and height >150 cm. After a baseline electrocardiogram (ECG) on the day of surgery, patients were premedicated with up to 5 mg of midazolam IV and given an IV infusion of 5 mL/kg lactated Ringer's solution. In the operating room, patients were monitored with an ECG, pulse oximeter, and a noninvasive blood pressure device, and respiratory rate was recorded every 20 min.
Anesthetics were administered by a physician blinded to the randomization. After skin anesthesia with 1% lidocaine, spinal anesthesia was performed with the patient in the lateral decubitus position (with the operative side dependent) at the L2-3 or L3-4 interspace using a midline approach with a 25-gauge Whitacre needle. After identification of cerebrospinal fluid, 4 mL of study drug was injected over 20 s with the needle orifice directed cephalad. The study solution contained either 15, 20, or 23 mg of isobaric sameridine or 100 mg of hyperbaric lidocaine. Immediately after injection, patients were returned to the supine position. Onset of sensory block was assessed bilaterally using a 27-gauge dental needle to detect loss of sensation at 5-min intervals until the start of surgery. If the dermatomal level of analgesia had not reached the T8 level within 30 min after injection or if anesthesia was inadequate for surgery, the patient received general anesthesia with propofol and nitrous oxide and was excluded from further efficacy evaluations and classified as a treatment failure-efficacy (TFE). If the duration of anesthesia was inadequate for completion of the procedure, the patient was given a similar general anesthetic; and these patients were classified as treatment failure-duration (TFD). During surgery, the upper limit of sensory block was recorded at 10-min intervals. After surgery, the sensory block was evaluated at upper and lower dermatomes at 15-min intervals (calculated from the time of study drug injection) until the return of normal sensation. Motor block was evaluated using a modified Bromage scale. Patients who attained a motor scale score of 0 were assessed every 30 min for the ability to walk without support. The time to micturation was also recorded. Patients were discharged from the postanesthesia care unit (PACU) according to institutional criteria. Postoperative care and data collection were performed by personnel blinded to the randomization.
Postoperative pain was assessed using a visual analog scale (VAS) at rest, after a vigorous cough, and on moving from a lying to a sitting position every hour (calculated from injection of study drug) for 24 h. Time until first request for postoperative supplemental analgesics was recorded. If the patient was in the PACU, IV morphine was administered. If the patient had been transferred from the PACU, the initial analgesic was generally oral hydrocodone and acetaminophen. For mild pain, patients received acetaminophen only. The total amount of both oral hydrocodone and acetaminophen and morphine administered in the first 24 h were recorded. Postoperative monitoring included noninvasive blood pressure, heart rate, and respiratory rate every hour after surgery up to 24 h after injection of the study drug. Hypotension was defined as a systolic blood pressure <80 mm Hg and was treated at the discretion of the anesthesiologist. ECGs were obtained twice before premedication; 5 min before study drug injection; 20, 40, and 60 min after study injection; and again 2, 3, 4, 5, 6, and 12 h after injection of the study drug. Mean QS width, QT interval and QTC interval were calculated for each ECG. Patients were monitored with pulse oximetry continuously during surgery, and the pulse oximetry reading was recorded again coincident with the respiratory rate evaluations after surgery. Supplemental oxygen was discontinued in patients with an oxygen saturation >94%.
For clinical efficacy, groups were described and compared with respect to their VAS scores at rest and on mobilization and coughing, time to first analgesic, total dose of analgesics, onset of sensory block, duration of sensory block, maximal spread of sensory block, onset of motor block, duration of motor block, time to ambulation, and time to micturation. Formal pairwise comparisons of groups were made using a center-adjusted Wilcoxon rank sum test (with an associated estimate and 95% confidence interval for each difference for relevant variables) or a Mantel-Haenszel test. Differences referred to as statistically significant correspond to a two-sided P value <or=to 0.05, unadjusted for multiple comparisons. For safety evaluation, the groups were described and informally compared using graphs and descriptive statistics with respect to blood pressure, heart rate, ECG, respiratory rate, arterial oxygen saturation, and hematological and biochemical variables.
The intended sample size of 50 patients in each group was based on considerations of the length of the 95% confidence intervals for the primary efficacy variable, VAS scores at active flexion (a within-subject mean over the first 24 h) from a previous study of patients undergoing knee arthroscopy. The length of such an interval was predicted to be <or=to12 mm with power 80%. For ethical reasons, it was decided that the recruitment of patients to a treatment group should be terminated if the number of TFEs attained eight in the group (>15% projected failure rate). The frequency of TFEs occurring in each group was monitored by a separate unblinded member of the Department of Pharmacy at Astra Pain Control, Sweden.
One hundred ninety-three patients were enrolled in the study. Of these, 189 received the study drug and were included for evaluation for safety purposes. One hundred forty patients completed the study according to the protocol and were evaluated for efficacy (Table 1). The sameridine 15 mg dose group was excluded from further accrual after 35 subjects were enrolled because of the occurrence of eight failures. There were no technical failures of spinal anesthesia during the protocol. The groups were similar with respect to demographic variables and duration of surgery (Table 2).
The sameridine groups had similar onset of sensory block to the T8 level, time to maximal sensory level, maximal block height, and duration of sensory and motor block (Table 3). A dose response was not demonstrated for sameridine. Compared with sameridine, lidocaine had a slightly faster onset and a shorter duration of sensory anesthesia (Figure 1). In one group (sameridine 20), the maximal block height was lower than that in the lidocaine group. Motor blockade was less intense but had a longer duration with the sameridine patients. Time to void and ambulate was similar among the groups. Although enrollment in the 15 mg sameridine group was discontinued because of the predetermined failure limit, there was a similar failure rate in the other three groups (Table 1).
The sameridine groups had lower VAS scores at rest, with coughing, and with mobilization compared with the lidocaine group in the first 4 h, but the four groups were similar thereafter. More subjects receiving lidocaine spinal anesthesia (26 of 32, 81%) requested morphine than those receiving sameridine (39 of 108, 36%), and the first request for analgesic medication was earlier in the lidocaine group (Table 4). Total morphine consumption in 24 h was greater in the lidocaine group compared with that in the sameridine groups. The average dose of morphine for sameridine patients who requested morphine was similar to the average dose for lidocaine patients. The total oral analgesic consumption over 24 h was similar among the four groups. A dose response was not identified for postoperative analgesia after the administration of sameridine.
Serious adverse events included a wound infection in the 15 mg sameridine group, a transient paresthesia in the 20 mg sameridine group, and a wound hematoma and two incidences of bradycardia in the lidocaine group. Other adverse events (Table 5) were similar among the groups. The groups were also similar in the blood pressure changes and heart rate changes during the course of the anesthetic and during the recovery period, although hypotension was more frequent in the lidocaine group and pruritus more common with sameridine. There was a slight increase in the QTc interval among the patients receiving sameridine. Although the prolongation was somewhat greater in the sameridine group than in the lidocaine group (9%-12% with sameridine, 8% with lidocaine), the QTc interval was similar to the baseline in both groups. There were no dysrrhythmias associated with the QTc changes. The number of patients whose respiratory rates were <8 breaths/min (Table 5), as well as the average pulse oximetry readings, was similar in the groups. No patients had clinical evidence of respiratory depression that required naloxone.
Several compounds have both opioid and local anesthetic properties . Meperidine is the only one that has been assessed in clinical practice. Norris et al.  have shown that meperidine is similar to lidocaine in providing effective anesthesia for postpartum tubal ligation, with a longer delay in requesting postoperative analgesia (448 +/- 184 vs 83 +/- 33 min). Patel et al.  have also shown similar efficacy between meperidine and lidocaine when used for urologic procedures, again with postoperative analgesia persisting longer in the meperidine group. Meperidine is effective for cesarean section  and lower extremity surgery . However, in each of these studies, meperidine produced a higher incidence of side effects (nausea, drowsiness, pruritus).
Sameridine is a new compound that also seems to have both local anesthetic and analgesic properties. Our data confirm that sameridine has an anesthetic effect similar to that of lidocaine in humans, as well as residual analgesic efficacy, which delays the need for supplemental opioids compared with lidocaine. In contrast to meperidine, the side effect profile seems to be similar to that of lidocaine.
As a local anesthetic, the duration of sameridine anesthesia in isobaric doses of 15, 20, or 23 mg was equivalent to that of 100 mg of hyperbaric lidocaine. There was no apparent dose response over the range of doses in this study. All three produced equivalent extent and duration of anesthesia. There was a suggestion that the smaller dose (15 mg) is not as effective, based on the higher percentage of treatment failures in this group. In those patients who did achieve spinal anesthesia, however, the extent and duration of the blockade were similar to those of lidocaine in all three sameridine groups. The major difference between sameridine and lidocaine was the time to reach maximal sensory block height. Sameridine is equally rapid in the initial onset of anesthesia but is slower to attain the maximal block height. This may be due to the isobaric formulation of sameridine, compared with the hyperbaric lidocaine solution. Like meperidine, sameridine also produces less intense motor blockade than lidocaine in these doses. Time to complete resolution of sameridine anesthesia and motor blockade was longer. However, time to micturation and ambulation were similar, which suggests that discharge times would be similar in the outpatient setting.
In analgesic efficacy, there were two differences compared with lidocaine. First, time to request analgesic medication was significantly longer after sameridine anesthesia (5-5.5 h vs 2.1 h), despite resolution of sensory anesthesia at 3.1-3.6 h and 2.0 h, respectively. Another finding was that the severity of pain at the time of this first request was less in the sameridine group, as evidenced by the VAS scores at time for first analgesia and by the percentage of patients in the lidocaine group who requested morphine as their first analgesic versus those in the sameridine group who requested the oral oxycodone or acetaminophen as their first analgesic. This difference in pain intensity persisted for the first 4 h after the subarachnoid injection of sameridine. After 4 h, there was no difference between the lidocaine and sameridine groups in VAS scores or total consumption of analgesics. This finding is similar to the results of a parallel study of sameridine compared with bupivacaine for spinal anesthesia for hip surgery, in which the duration of anesthesia and analgesia was similar in both groups and there was no difference in analgesic consumption after 4 h in the first 24 h of treatment .
Sameridine seems to have acceptable side effects in humans based on the findings of our study. Frequency of serious adverse events was rare (5 of 189). One event (bradycardia in a lidocaine group patient) was reported as "probably related to the study drug." Respiratory depression is a major concern for any compound with opioid properties. In a study comparing sameridine 25 mg with bupivacaine 15 mg in healthy human volunteers, there were similar minor effects on the ventilatory response to hypercarbia and hypoxia . In our subjects, there were no significant episodes of respiratory depression that required reversal with opioid antagonists, and the frequency of bradypnea (respiratory rates <8 breaths/min) was evenly distributed between the sameridine and the lidocaine groups. Mild opioid side effects such as pruritus were low in frequency. There were no complaints in any group of radiating back pain similar to the transient radicular irritation syndrome described with lidocaine . Animal studies suggest a change in cardiac conduction with IV sameridine manifesting as a prolongation of the QTc interval; our patients showed some mild prolongation in QTc interval, which also occurred with the lidocaine treatment group. Overall, there were no significant adverse events that seem to contraindicate the use of this drug in humans.
There were several limitations to our study. First, as a multicenter study, there were variations in the treatment. Nevertheless, there was close adherence to the protocol, and the final analysis of data was only performed on patients whose care was conducted per protocol. A more significant limitation is that the choice of analgesia in the postoperative period was dictated, at least in part, by patient location. While in the PACU, patients could request morphine; after transfer to the overnight observation unit, they were first offered oral analgesics. Thus, the higher frequency of morphine use in the lidocaine group may have been related to the early onset of pain while patients were generally still in the PACU. Thus, morphine consumption may not be a reliable indicator of sameridine's analgesic efficacy. Nevertheless, fewer patients in the sameridine groups requested and received morphine, in equivalent doses (Table 5). This observation indicates that most of the sameridine patients were able to achieve relief with oral analgesics, which is supported by the lower VAS scores in these groups. Finally, our protocol excluded women. Surgical hernia repair in women often involves a less painful procedure (ring ligation), and substantially fewer repairs are performed in women. It was felt that inclusion of this group might produce more heterogeneity of data; thus, the protocol was restricted to men.
Another factor affecting the results is the use of an isobaric solution of sameridine. Although hyperbaric solutions (such as the lidocaine used in this study) consistently produce sensory block in the upper thoracic levels, isobaric solutions are less consistent in their spread. Total milligram dose also affects level, and one intent of the study was to demonstrate such a dose-response relationship with sameridine by using the isobaric formulation. The variability of spread of the isobaric solution may have contributed to the lack of a dose-response with sameridine in this study. It may also account, in part, for the slower time to reach maximal block height in this study. Further study with a hyperbaric solution of sameridine would be required to clarify the role of this factor.
The failure rate in all four groups was high (11%-22%). The high failure rate in the small-dose sameridine may have reflected the lower end of a dose response-curve. The failure of adequate duration in the lidocaine group probably reflects the fact that many of these procedures were performed in teaching institutions, where the duration of hernia repair exceeded expectations. The treatment failures in the other groups may be attributed to a high level of analgesia required for hernia repair in the absence of any supplemental analgesic drugs intraoperatively. Because there were no technical failures to attain some degree of anesthesia, the treatment failures seem to indicate that the doses we used were not sufficient to provide the necessary anesthesia alone.
In summary, sameridine is a new compound with both local anesthetic and analgesic properties. It has anesthetic properties similar to those of lidocaine but residual analgesic effects similar to those of meperidine, which persist for approximately 4 h. This window of analgesia may be useful in providing pain-free early discharge in an ambulatory surgery unit. The time required to void and ambulate may offset this brief window of analgesia. Nevertheless, the concept of local anesthetics with a combination of opioid properties (and fewer opioid side effects) is worthy of further research.
1. Power I, Brown DT, Wildsmith JAW. The effect of fentanyl, meperidine, and diamorphine on nerve conduction in vitro. Reg Anesth 1991;16:204-8.
2. Norris MC, Honet JE, Leighton BL, Arkoosh VA. A comparison of meperidine and lidocaine for spinal anesthesia for postpartum tubal ligation. Reg Anesth 1996;21:84-8.
3. Patel D, Janardhan Y, Merai B, et al. Comparison of intrathecal meperidine and lidocaine in endoscopic urological procedures. Can J Anaesth 1990;37:567-70.
4. Westman L, Valentin A, Eriksson E, Ekblom A. Intrathecal administration of sameridine to patients subjected to arthroscopic knee joint surgery. Acta Anaesthesiol Scand 1998;42:691-7.
5. Thi TVN, Orliaguet G, Ngu TH, Bonnet F. Spinal anesthesia with meperidine as the sole agent for cesarean delivery. Reg Anesth 1994;19:386-9.
6. Sangarlangkarn S, Klaewtanong V, Jonglerttrakool P, Khankaew V. Meperidine as a spinal anesthetic agent: a comparison with lidocaine-glucose. Anesth Analg 1987;66:235-40.
7. Muldoon T, Personne M, Belfrage M, et al. Anaesthesia for total hip replacement: the use of intrathecal sameridine. Presented at European Society of Anesthesiologists, May 1997.
8. Osterlund A, Arlander E, Westman L, et al. Ventilatory response to hypercarbia and hypoxia during 24 h after intrathecal sameridine or bupivacaine [abstract]. Anesthesiology 1997;87: A807.
9. Pollock JE, Neal JM, Stephenson CA, Wiley C. Prospective study of the incidence of transient radicular irritation in patients undergoing spinal anesthesia. Anesthesiology 1996;84:1361-7.