Share this article on:

Hypercarbic and Hypoxic Ventilatory Responses After Intrathecal Administration of Bupivacaine and Sameridine

Modalen, Åsa Österlund MD; Westman, Lars MD, PhD; Arlander, Eva MScPharm; Eriksson, Lars I. MD, PhD; Lindahl, Sten G. E. MD, PhD

doi: 10.1213/00000539-200302000-00049

Sameridine is a new compound with both local anesthetic and opioid properties (partial μ-opioid receptor agonist). It was intended for intrathecal administration to provide anesthesia for surgery and extended postoperative analgesia. In this double-blinded pharmacodynamic study with a two-parallel-group design, we investigated, during a 24-h period, the effects of intrathecal sameridine and bupivacaine on ventilation at rest and at ventilatory challenges during hypercarbia and hypoxia. Twenty-four healthy volunteers received either 25 mg of sameridine or 15 mg of bupivacaine intrathecally. Ventilation was measured by pneumotachography and in-line capnography. Sedation was rated by a visual analog scale. Segmental spread and development of motor and sensory block were similar in both groups. There was a decrease in tidal volume 2.5 to 6 h after injection in the bupivacaine group. This was seen only at 4 h in the sameridine group. There were no other major ventilatory differences between sameridine and bupivacaine during resting ventilation. Hypercarbic (tidal volume, mean inspiratory flow) and hypoxic (mean inspiratory flow) ventilatory responses were slightly decreased in the sameridine group, but not in the bupivacaine group. We conclude that intrathecal administration of sameridine or bupivacaine in healthy volunteers produces similar, minor effects on ventilatory responses over a 24-h observation period.

Department of Anesthesiology and Intensive Care, Karolinska Hospital and Institute, Stockholm, Sweden

October 9, 2002.

Address correspondence and reprint requests to Åsa Österlund Modalen, MD, Karolinska Hospital and Institute, S-171 76 Stockholm, Sweden. Address e-mail to

The combined use of local anesthetics and opioids has been increasingly used for spinal anesthesia (1,2). This has improved analgesia postoperatively (3). Sameridine is a new compound with both local anesthetic properties and partial μ-receptor agonistic effects. Experimental and clinical studies have previously demonstrated that sameridine provides good analgesia after both IV and intrathecal (IT) administration. In an investigation of knee arthroscopy, the authors concluded that IT sameridine doses up to 25 mg resulted in a good sensory block that lasted 1–2 h longer than the motor block of 4–5 h (4). Mulroy et al. (5) compared IT sameridine with lidocaine in humans and concluded that the sensory block was similar during surgery. The use of postoperative systemic analgesics was, however, less when sameridine was used.

To evaluate the effects on resting ventilation after different IV sameridine doses (0.15 and 0.73 mg/kg), a comparative study was performed with morphine (0.10 mg/kg). The larger sameridine dose reduced minute ventilation (V̇e) to a greater extent than 0.10 mg/kg of morphine. However, the smaller dose of sameridine (0.15 mg/kg), corresponding to an IT dose, had no effect on resting ventilation (6). Furthermore, the larger (IV) sameridine dose (0.73 mg/kg) depressed the ventilatory response to hypercapnia, but this was not the case with the smaller dose (7). Because the goal for clinical use of sameridine is IT administration for surgery, it is of interest to evaluate its ventilatory effects.

The aim of this double-blinded pharmacodynamic study with a two-parallel-group design was to compare the ventilatory effects of sameridine and bupivacaine. Because of the μ-agonist effects of sameridine, it was hypothesized that IT sameridine impaired not only resting, but also ventilatory, responses to hypercarbia and hypoxia to a greater extent than bupivacaine.

Back to Top | Article Outline


The Ethical Committee on Human Research at the Karolinska Hospital and Institute, Stockholm, Sweden, approved the study protocol, and all volunteers gave their written, informed consent before inclusion. Twenty-four healthy volunteers (13 men and 11 women), 23 to 44 yr old, height between 160 and 190 cm and body weight between 55 and 93 kg, participated in study sessions that started in the morning. They were fasted for 8 h before drug administration. All subjects had to refrain from intake of alcohol within 24 h before the study start and from coffee and tea from 10:00 pm the evening before the study. They were advised to sleep at least 6 h the night before drug administration. All volunteers were nonsmokers and were screened by medical history, physical examination, and electrocardiogram (ECG). They were also informed about the breathing equipment and allowed to test it. The study was double-blinded and randomized. There were 2 parallel groups with 12 volunteers in each.

A venous cannula was placed in the left cubital vein for infusion of crystalloid solution (5 mL/kg) before injection of the study drug. The subjects rested in a comfortable bed with a 15° head-up tilt and were not allowed to leave the bed until they were able to walk again. The IT test drug was either isobaric sameridine 25 mg (6.25 mg/mL) or isobaric bupivacaine 15 mg (3.75 mg/mL). Both drugs were administered in a fixed volume of 4 mL. With the subject in sitting position, the test substance was injected via a 25-gauge Whitacre pencil-point needle placed in the lumbar region (L2-4). The injection was given by an anesthesiologist who was blinded to the drug used. At regular time intervals, 12-lead ECGs (Siemens Sicard 460; Siemens Elema, Germany) were recorded. Systolic and diastolic blood pressures were measured noninvasively.

The sensory block was determined by using ice cubes, and the motor block was determined according to a modified Bromage scale, where 0 = no motor paralysis, 1 = inability to raise extended legs, 2 = inability to flex knees, and 3 = inability to flex ankle joints. Ventilation was recorded with the subjects breathing hyperoxic air (fraction of inspired oxygen, 40%) through a transparent face mask (Gibeck; Gibeck-Dryden, Indianapolis, IN) positioned over their nose and mouth. The subjects were allowed an initial resting period of approximately 20–30 min to adjust to the apparatus. An in-line infrared capnometer (14360 A; Hewlett-Packard, Palo Alto, CA) and a pneumotachograph (Fleisch No. 0.2; linear for flows up to 2.4 L/s) were placed in the circuit. The dead space of the system was 70 mL as measured by water displacement. The inspiratory and expiratory resistances of the measuring apparatus were 2 cm H2O · L−1 · s−1. V̇e and respiratory rate (RR) were measured by integration of the flow signal from the pneumotachograph. Flow, tidal volume (Vt), and in-line end-tidal CO2 fractions (ETco2) were recorded on an ink-jet recorder (Mingograf 800; Siemens Elma) and stored in a computer. Inspired and expired concentrations of oxygen were continuously recorded by a Datex Ultima S device (Datex, Helsinki, Finland), which also measured arterial oxygen saturation by finger pulse oximetry. Ventilation volumes were presented at body temperature and pressure-saturated. The Hewlett-Packard capnometer was calibrated with certified gases before each experiment. Calibration of the pneumotachograph was performed before and after each experiment by using a high-precision syringe (Model 5530; Hans Rudolph Inc., Kansas City, MO). ETco2 values were not affected by the level of inspired CO2 and therefore were regarded to approximate arterial CO2 values. At each study occasion, the ventilatory variables recorded were V̇e (L/min), Vt (mL), RR (breaths/min), the respiratory duty cycle (Ti; s) and for expiration (s), mean inspiratory flow (Vt/Ti; mL/s), and respiratory timing (Ti/Ttot, s). These ventilatory variables were measured at resting ventilation, challenges of hypercarbia and hypoxia 30–60 min before study drug injection, and 1, 2.5, 4, 6, 9, 12, and 24 h after injection of the study drug. Prestudy drug injection measurements were used as baseline values.

Hypercarbic ventilatory responses (HCVR) were measured after adding 4% CO2 to inspired air during 5- to 6-min periods. The steady-state technique was chosen because it is routinely used in our laboratory. A steady-state for CO2 challenge was reached after 3 min. Ventilatory measurements were then performed over 2 min at steady-state breathing. HCVR was defined according to the expressionMATH

Hypoxic ventilatory responses (HVR) were measured after adding 5% oxygen/nitrogen gas to the inspired air to obtain a stable level of 80% arterial oxygen saturation. Steady-state (the technique routinely used in the laboratory) was reached after 5 min. After another 1–2 min with stable hypoxia, ventilatory mea-surements were performed over 2 min. Hence, acute HVR were measured. The isocapnic test condition was achieved by adding CO2 to inspired air to keep ETco2 constant during hyperventilation. HVR was defined according to the expressionMATH

HCVR and HVR were tested in random order. The degree of sedation was assessed by using a 10-cm visual analog scale (VAS), where the subject scaled the VAS ruler from 0 (awake, not tired) to 10 (asleep).

The effects of study drugs were estimated as the median of the absolute change from baseline. This median estimate and its 95% nonparametric confidence interval were presented for each of the variables. Comparisons between the groups were made by Wilcoxon’s ranked sum test with associated point estimates and 95% confidence intervals. The outcome of each statistical test was presented as two-sided P values and confidence intervals.

Back to Top | Article Outline


The development and regression of sensory and motor block levels were similar in the two groups (Fig. 1, Table 1). There were no differences in hemodynamic effects. There were no signs of desaturation or late (up to 24 h) respiratory depression in either group. Side effects were somewhat more pronounced in the sameridine group (pruritus, n = 6; headache, n = 3; and nausea, n = 3). One volunteer in the bupivacaine group developed a “postdural puncture headache,” but no other side effects were noted in this group. Sedation was similar in both groups, with an increase in median VAS scores at 1–4 h (20 in the sameridine group and 10 in the bupivacaine group).

Figure 1

Figure 1

Table 1

Table 1

ETco2 was increased 2.5 h after initiation of the sameridine block (Table 2, Fig. 2). In the bupivacaine group, Vt was reduced at 2.5, 4, and 6 h (P < 0.05), which was compensated for by an increased RR, resulting in an unchanged ETco2 (Fig. 2). In the sameridine group, Vt was also reduced at one stage 4 h after the administration of the drug (Table 2).

Table 2

Table 2

Figure 2

Figure 2

In the control situation, V̇e increased by 77% (from 6.9 to 12.2 L/min) in the sameridine group and by 64% (from 7.2 to 11.8 L/min) in the bupivacaine group (Table 2). Corresponding ETco2 values increased on average 11.6 and 7.1 mm Hg, respectively (Table 2).

HCVR was reduced 1 h after the sameridine block and returned to control values at 6 h (P < 0.05;Fig. 3). There were no changes in HCVR during the bupivacaine block (Fig. 3). At 1 and 2.5 h, Vt values were higher in the bupivacaine group than in the sameridine group (P < 0.05;Fig. 4). This difference disappeared at 4 and 6 h. Acute HVRs, as reflected by V̇e, RR, Vt, and ETco2, were unaffected by sameridine and bupivacaine spinal blockade (Table 2).

Figure 3

Figure 3

Figure 4

Figure 4

Back to Top | Article Outline


In this study, bupivacaine reduced Vt (2.5–6 hours) during resting conditions, although ETco2 remained unchanged because of increased RR. This was seen only at a shorter time (four hours) with sameridine. Sameridine depressed the HCVR, whereas bupivacaine did not.

We investigated the respiratory effects after IT administration of a molecule with combined local anesthetic and opioid properties and bupivacaine, a well-known local anesthetic. On the basis of previous dose-escalating studies of sameridine (4), a dose of 25 mg was chosen in this study. Sameridine solution is isobaric and has anesthetic properties similar to those of bupivacaine, according to animal studies. 4 To be able to make a comparison between the two substances, we chose a clinically used dose of 15 mg of isobaric bupivacaine, which has a similar clinical profile with respect to sensory and motor block to 25 mg of sameridine (Fig. 1, Table 1).

Respiratory effects of local anesthetics have been studied before. In a classic study from 1968 (8), no effects were found on resting ventilation when IV doses of bupivacaine, mepivacaine, and lidocaine were administered. These findings were confirmed in 1991 by Johnson and Löfström (9), who also found that ventilation was unaffected by these three drugs during hypercarbia, but a slight stimulatory effect was noted during hypoxia. In another study, it was concluded that epidural lidocaine induces an increased ventilatory response to CO2, probably because of systemic effects (10). Gross et al. (11,12) studied the effects of lidocaine infusion on the ventilatory response to hypoxia and CO2. They found that lidocaine depressed the hypoxic drive. The ventilatory response to hypercarbia was depressed by a bolus injection, whereas an infusion of lidocaine showed an increased response.

In this study, the IT administration of bupivacaine reduced Vt at resting ventilation to a greater extent than sameridine (P < 0.05). A possible explanation could be that bupivacaine causes a more extensive motor block of respiratory muscles than sameridine. Assessments according to the Bromage scale, as well as the duration of sensory and motor blocks, do not, however, support such an explanation.

Another important factor for the differences in Vt is the doses used. The sameridine dose of 25 mg was chosen on the basis of previous dose-escalating studies in humans (4). The authors concluded that a 25-mg IT dose was clinically acceptable. The 15-mg bupivacaine dose is the ordinary clinical dose. Evaluation of sensory and motor blocks demonstrated similar development at least up to four hours after administration. At six hours, the sensory block level was somewhat higher in the bupivacaine group (Fig. 1). In addition, there were five subjects with Grade 1 motor block at six hours in the bupivacaine group and only one in the sameridine group (Table 1). This difference at six hours indicates a somewhat shorter duration of action for sameridine. It does not explain the differences in Vt between bupivacaine and sameridine at 2.5 to 6 hours, when sensory and motor blockades were the same. Still, the most likely explanation for the Vt difference is the more pronounced motor block on respiratory muscles caused by bupivacaine. Respiratory muscle action is not evaluated by the Bromage scale.

It is not likely that the sameridine depression of HCVR is caused by the local anesthetic properties of the sameridine molecule, because the results differ from those in the bupivacaine group. Hence, the influence of sameridine on HCVR is most likely mediated via effects on the μ-opioid receptor. The results also support the view that HCVR responses are reduced because of interactions with centrally located μ receptors (13). In this series, there was no indication of late (24 hours) respiratory depression. Whether the small ventilatory effects of sameridine would be exaggerated with simultaneous use of systemic sedative or anxiolytic drugs will have to be evaluated in future studies.

We conclude that IT administration of 15 mg of bupivacaine and 25 mg of sameridine did not affect resting ventilation evaluated by ETco2. Bupivacaine however, reduced Vt, which was compensated for by increased RR. IT sameridine depressed HCVR more than bupivacaine, most probably because of its central partial μ agonism.

We are most grateful for the excellent technical contributions of Anette Ebberyd and Ringvor Hägglöv, Department of Anesthesiology and Intensive Care, Karolinska Hospital. Thanks also go to Pär Karlsson, MSc, Astra Pain Control AB, for statistical evaluation.

Back to Top | Article Outline


1 LaBarre M, St.-Onge S, Ask A-L, Payza K. The local anesthetic sameridine is a selective ligand and a partial agonist at cloned human mu-receptors [abstract 479.12]. Neuroscience, New Orleans, LA, 1997.
Cited Here...

2 Ask A-L, Alari L, Torsvik C. Local anesthetic and analgesic effect of sameridine in mice and inhibition of 3H-naloxone binding in guinea-pig brain [abstract D473]. 11th World Congress of Anesthesiology, Sydney, Australia, 1996.
Cited Here...

3 Gustavsson LL, Vallin H, Sjövall J, Westerling P. A drug combining local anesthetic and opioid (sameridine) first time administration in man [abstract P523]. 11th World Congress of Anesthesiology, Sydney, Australia, 1996.
Cited Here...

4 Ask A-L, Alari L, Torsvik C. Local anesthetic and analgesic effect of sameridine in mice and inhibition of 3H-naloxone binding in guinea-pig brain [abstract D473]. 11th World Congress of Anesthesiology, Sydney, Australia, 1996.
Cited Here...

Back to Top | Article Outline


1. Naulty S. Intraspinal narcotics. Clin Anaesthesiol 1986:145–56.
2. Benedetti C. Intraspinal analgesia: an historical overview. Acta Anaesthesiol Scand 1987; 31 (Suppl 85): 17–24.
3. Jacobsson L, Chabal C, Brody MCA. Dose response study of intrathecal morphine: efficacy, duration, optimal dose, and side effects. Anesth Analg 1988; 67: 1082–8.
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. Mulroy MF, Greengrass R, Ganapathy S, et al. Sameridine is safe and effective for spinal anesthesia: a comparative dose-ranging study with lidocaine for inguinal hernia repair. Anesth Analg 1999; 88: 815–21.
6. Österlund Å, Arlander E, Eriksson LI, Lindahl SGE. Effects on resting ventilation of intravenous infusions of morphine or sameridine, a novel molecule with both local anesthetic and opioid properties. Anesth Analg 1999; 88: 160–5.
7. Österlund Å, Arlander E, Eriksson LI, Lindahl SGE. Effects on hypercarbic ventilatory response of sameridine in comparison to morphine and placebo. Anesth Analg 2001; 92: 529–34.
8. Jorfeldt L, Lövström B, Pernow B, et al. The effect of local anaesthetics on the central circulation and respiration in man and dog. Acta Anaesthesiol Scand 1968; 12: 153–69.
9. Johnson A, Löfström JB. Influence of local anaesthetics on ventilation. Reg Anesth 1991; 16: 7–12.
10. Labaille T, Clerque F, Samii K, et al. Ventilatory response to CO2 following intravenous and epidural lidocaine. Anesthesiology 1985; 63: 179–83.
11. Gross JB, Caldwell CB, Shaw LM, et al. The effects of lidocaine infusion on the ventilatory response to hypoxia. Anesthesiology 1984; 61: 662–5.
12. Gross JB, Caldwell CB, Shaw LM, et al. The effects of lidocaine infusion on the ventilatory response to carbon dioxide. Anesthesiology 1983; 59: 521–5.
13. Etches RC, Sandler AN, Daley MD. Respiratory depression and spinal opioids [review]. Can J Anaesth 1989; 36: 165–85.
© 2003 International Anesthesia Research Society