Intravenous Ketamine or Fentanyl Prolongs Postoperative Analgesia After Intrathecal Neostigmine

Lauretti, Gabriela R. MD, MSc, PhD; Azevedo, Vera M. S. MD

Anesthesia & Analgesia:
Regional Anesthesia and Pain Management

The purpose of this study was to determine whether intravenous (IV) ketamine would enhance analgesia from intrathecal (IT) neostigmine compared with combining IV fentanyl with IT neostigmine.Sixty patients undergoing vaginoplasty under spinal anesthesia were assigned to one of six groups (n = 10). Patients were premedicated with midazolam plus the IV test drug. The IT drugs were 20 mg bupivacaine plus saline or 50 micro gram neostigmine. The control group (CG) received saline IV and IT. The neostigmine control group (NCG) received saline IV and neostigmine IT. The ketamine group (KG) received ketamine 0.2 mg/kg IV and saline IT, and the ketamine neostigmine group (KNG), ketamine IV and neostigmine IT. The fentanyl group (FG) received fentanyl 1 micro gram/kg IV and saline IT, and the fentanyl neostigmine group (FNG), fentanyl IV and neostigmine IT. The time to first rescue analgesic was longer for the FNG and KNG compared with the CG, with less rescue analgesic consumption (P < 0.02 and P < 0.01, respectively). Only the FNG had significantly intraoperative nausea/vomiting (P < 0.02). In conclusion, the combination of IV ketamine and IT neostigmine results in prolonged postoperative analgesia and less intraoperative nausea and vomiting than the combination of IV fentanyl and IT neostigmine.

(Anesth Analg 1996;83:766-70)

Author Information

Discipline of Anesthesiology, Department of Surgery, Orthopedics and Traumatology, Hospital das Clinicas-Faculdade de Medicina de Ribeirao Preto-USP, Sao Paulo, Brazil.

Accepted for publication June 6, 1996.

Address correspondence and reprint requests to Gabriela R. Lauretti, MD, MSc, PhD, rua Mantiqueira, 460, Ribeirao Pretto, Sao Paulo, Brazil 14025/600.

Article Outline

Despite advances in the management of postoperative pain, many patients still suffer from postoperative discomfort, probably due to difficulties in balancing a reliable, prolonged, and effective pain regimen with acceptable side effects [1]. Recent research on the pharmacology of nociception has shown the involvement of cholinergic transmission and its safe use at the spinal level [2] to be among the numerous candidates for spinal pain modulation [3,4]. Unfortunately, dose-dependent analgesia from intrathecal (IT) neostigmine is also accompanied by a dose-dependent incidence of side effects, mainly epigastric discomfort, which can delay postoperative recovery [2,5,6]. One approach to maintaining analgesia while decreasing the incidence of untoward effects is to combine small doses of analgesics from different pharmacological classes [7-9]. The purpose of this study was to determine whether a combination of intravenous (IV) ketamine would enhance analgesia from IT neostigmine compared with the combination of IV fentanyl with IT neostigmine under spinal bupivacaine anesthesia.

Back to Top | Article Outline


The study protocol was approved by the Ethical Committee of the University of Sao Paulo's Teaching Hospital, Ribeirao Preto. After giving written, informed consent, 60 ASA physical status I and II patients undergoing anterior and posterior vaginoplasty were randomly assigned to one of six groups in a double-blind way (Table 1; n = 10) and prospectively studied using a placebo-controlled design to examine analgesia and adverse effects. The visual analog scale (VAS) scores were taken by a single interviewer, and a consistent set of instructions was used throughout the study. The concept of the VAS, which consisted of a 10-cm line with 0 equaling "no nausea or vomiting" or "no pain at all" and 10 being "worst possible nausea or vomiting" or "the worst possible pain" was introduced the day before.

Patients were premedicated with IV midazolam 0.05 mg/kg plus the IV drug test to a 4-mL final volume. Ten to 15 minutes later, the spinal anesthesia was initiated using the L3-L4 interspace with a 25-gauge spinal needle. The IT drugs were 20 mg hyperbaric bupivacaine 0.5% plus saline or 50 micro gram preservative-free neostigmine diluted with saline Table 1. The IT neostigmine dose was chosen based on previous experience of this group. The final spinal drug volume was 5 mL, which was injected at 1 mL/7 s. Patients were placed supine immediately after the spinal injection.

Intraoperative assessment included determination of the pinprick level at 5 and 10 min, arterial blood pressure, oxyhemoglobin saturation, heart rate, and adverse effects. Blood pressure was monitored noninvasively every 5 min throughout surgery. A decrease in mean arterial pressure more than 15% below preanesthetic baseline was treated by incremental ephedrine, 4 mg IV. Decrease in heart rate to less than 50 bpm was treated with incremental atropine, 0.25 mg IV. Intraoperative nausea and vomiting were scored by the patient on a 10-cm visual analog scale (VAS). Nausea (VAS > 2) at any time during the study was treated initially with metoclopramide 20 mg IV and droperidol 0.5 mg IV, if necessary.

The postoperative assessment included pain, adverse effects, and the anesthetic time, measured as time to reach Bromage 2 score [10]. Pain was assessed by the patient upon coughing and moving on the bed and scored using the 10-cm VAS at fixed intervals. Duration of effective analgesia was measured as time from the IT drug administration to the first need for analgesic administration in the recovery room, recorded in minutes. The VAS at the time for first rescue analgesic was measured using the 10-cm VAS. Diclofenac 75 mg intramuscularly was given if necessary, and patients were free to request it. The overall 24-h pain VAS quantified the overall pain impression by the patient for 24 h postoperatively. Postoperative nausea and vomiting were also determined 24 h postoperatively using the 10-cm VAS.

Groups were compared for demographic data, duration of surgery, and time to first rescue analgesics by one-way analysis of variance. Incidence of adverse events and adjuvant drug use were compared among groups by chi squared test. Categorical scale data were compared among groups by Kruskal-Wallis test. Arterial blood pressure, heart rate, and VAS scores were compared among groups by two-way analysis of variance for repeated measures. Post-hoc tests (Bonferroni and Tukey) were used to correct P values. P < 0.05 was considered significant.

Back to Top | Article Outline


The six groups were demographically the same in regard to ASA physical status, age, weight, height, and surgical and anesthetic times, the latter being defined as time to score Bromage 2 Table 2. The level to pinprick at 5 and 10 min Table 3, intraoperative mean blood pressure and heart rate at 5, 10, 15, 20, 30, and 40 min after the spinal anesthesia, and ephedrine consumption were also similar in all groups (P > 0.05; data not shown).

The time to first rescue analgesic, VAS scores at first rescue analgesic, number of im intramuscular injections of 75 mg diclofenac in 24 h, and the overall 24-h VAS pain score are presented in Table 4. The VAS pain scores at fixed intervals are shown in Figure 1. The administration of IV fentanyl and IT neostigmine, but not IV ketamine, increased the time for first need for analgesic administration in recovery room compared with the neostigmine control group (NCG) (P = 0.034639 and P = 0.367265, respectively). However, the time to first rescue analgesic was statistically longer for both the fentanyl neostigmine group (FNG) and ketamine neostigmine group (KNG) compared with the control group (CG) (P < 0.02), and all groups had similar VAS scores at the time to first rescue analgesic (P > 0.05).

Neither IT neostigmine alone nor IV ketamine or fentanyl alone increased the time for first request of rescue analgesic medication (P = 0.178406, P = 0.799933, and P = 1, respectively). The ketamine group (KG), FNG, and KNG requested fewer rescue analgesic medications during the first 24 h of observation compared with the CG (P < 0.01). However, only the KNG reported statistically lower 24-h VAS scores compared with the CG (P = 0.021848).

The number of patients having intraoperative nausea and vomiting, the overall intraoperative nausea and vomiting scores, and the overall 24-h VAS nausea and vomiting scores are shown in Table 3. Fifty percent of the patients from the FNG had intraoperative nausea and vomiting compared with 0% from the CG and KG (P = 0.026718); these patients complained of an overall intraoperative nausea and vomiting VAS assessment score of 2.4 +/- 2.6 compared with the CG (VAS = 0) and KG (VAS = 0) (P = 0.011238).

Other intraoperative adverse effects included bradycardia, hypotension, cardiac arrhythmia, and fecal incontinence. One patient from the CG and two from the KG had hypotension 30, 18, and 25 min after the spinal anesthesia, respectively, and were treated with IV ephedrine. Two patients from the NCG had bradycardia 65 and 80 min after the spinal anesthesia and were successfully treated with atropine. One patient from the KNG had bradycardia at 21 min and other cardiac arrhythmias (ventricular extrasystoles) 20 min after the spinal anesthesia and was successfully treated with 160 mg IV lidocaine divided in two equal bolus doses. This last ASA physical status II patient had a past history of occasional cardiac arrhythmia. Three patients from the FNG had bradycardia at 15, 40, and 45 min after the spinal anesthesia. Another patient from the FNG opened the bowel and urinated 35 min after the spinal anesthesia; this particular patient vomited frequently during the procedure.

Apart from postoperative nausea and vomiting, which was similar among groups (P = 0.8437), one ASA physical status II patient (past history of hypertension) from the CG had a grand mal seizure in the recovery room 195 min after the spinal anesthesia and was successfully treated with 5 mg IV midazolam. She was the third patient from the study, and at that time we had opened her personal code to discover which group she was in. Because she was from the CG, we were confident in performing this double-blind study.

Back to Top | Article Outline


We have shown that the administration of a small dose of IV ketamine with IT neostigmine results in prolonged postoperative analgesia (mean > 7 hours) similar to that with IV fentanyl administered with IT neostigmine. However, IV ketamine results in less nausea and vomiting than does the combination of a small dose of IV 1 micro gram/kg fentanyl and 50 micro gram IT neostigmine.

While neither IV ketamine alone nor IV fentanyl alone increased the time for first request for analgesic medications, their administration with IT neostigmine showed a clear analgesic effect. Morphine given intrathecally is thought to improve IT neostigmine analgesia. Isobolographic analysis revealed a simple additivity between IT neostigmine and morphine in rats [7]. Other authors have demonstrated that the combination of IT morphine and neostigmine produces moderate analgesia in a model of noxious thermal stimuli in rats [11]. In another study, in patients undergoing vaginoplasty, IT morphine and neostigmine also produced a moderate synergistic analgesic effect [12].

This is the first time that IV ketamine has been shown to improve IT neostigmine analgesia in patients. Although systemically administered ketamine has a local anesthetic action, as well as spinal effects [13], the IV dose used in this study (0.2 mg/kg) was not large enough to have any local anesthetic action, as demonstrated by similar pinprick levels at 5 and 10 minutes and similar anesthetic times, as measured by the Bromage scale.

Although ketamine's primary antinociceptive site of action appears to be the phencyclidine receptor on the N-methyl-D-aspartate receptor complex [14], additional activity on opiate [15] and quisqualate receptors [16] is suggested. It is interesting that ketamine appears to exert its analgesic effect after the channel coupled to the N-methyl-D-aspartate receptor complex has been opened [17], but small doses of IV ketamine were successfully used preemptively, combined with general anesthesia, to provide postoperative analgesia [18]. Ketamine also inhibits central temporal nociceptive somation and has a marked hypoalgesic effect on high-intensity nociceptive electrical and mechanical stimuli [19]. Nevertheless, ketamine also profoundly inhibited both the central and peripheral muscarinic signaling within the clinically relevant concentration range [20]. It could be argued that the stimulation of spinal cord muscarinic receptors results in antinociceptive effects [2,7]. However, the ketamine analgesic effect observed could be due to a direct antispasmodic effect on the viscera through muscarinic receptors. If so, this is another example of an association between central cholinergic effects and peripheral anticholinergic effects in the treatment of postoperative pain [21].

Among the adverse effects observed, nausea and vomiting were the most troublesome in the FNG. One patient from the FNG also opened the bowels and urinated 35 minutes after the spinal anesthesia. This could be an effect of spinal neostigmine, although the etiology of bowel incontinence is uncertain, and there is no evidence that spinal cholinergic stimulation causes this effect [2]. We have previously shown that spinal bupivacaine combined with IT fentanyl and IT neostigmine were effective as anesthetics in patients undergoing abdominal hysterectomy [22]. Interestingly, while the association of IT fentanyl with IT neostigmine resulted in minimal adverse effects [22], the association of IV fentanyl with a similar spinal neostigmine dose resulted in an unacceptable incidence of nausea and vomiting. It is been suggested that the emetic effect of opioids is expressed at the chemoreceptor trigger zone through delta-opioid receptor subtype binding, while binding to the mu-receptor subtype at the vomiting center expresses an antiemetic effect [23,24]. One explanation could be that IV fentanyl expresses emesis, through a predominant binding at the chemoreceptor trigger zone, whereas IT fentanyl would be acting mainly at the vomiting center expressing an antiemetic effect. Nevertheless, the doses used in both studies were different (1 micro gram/kg IV and 25 micro gram IT). Whether a smaller IV dose of fentanyl would retain the analgesic potentiation of spinal neostigmine with an acceptable incidence of adverse effects is already under investigation.

In our study, one of the patients had a clinical seizure 195 minutes after the spinal anesthesia, which was promptly and successfully treated with IV midazolam. Fortunately, she was from the CG; otherwise, this study and many others would have been compromised, and spinal antinociceptive neostigmine could have been partially discredited.

In conclusion, the administration of a small dose of IV ketamine enhances the spinal neostigmine antinociceptive effect without increasing the incidence of side effects, showing it to be an efficient alternative postoperative pain therapy for vaginoplasty surgery.

Back to Top | Article Outline


1. Kehlet H, Dahl JB. The value of multi-modal or balanced analgesia in postoperative pain relief. Anesth Analg 1993;77:1048-56.
2. Hood DD, Eisenach JC, Tuttle R. Phase I safety assessment of intrathecal neostigmine methylsulfate in humans. Anesthesiology 1995;82:331-43.
3. Dickenson AH. Spinal cord pharmacology of pain. Br J Anaesth 1995;75:193-200.
4. Stamford JA. Descending control of pain. Br J Anaesth 1995;75:217-27.
5. Lauretti GR, Eisenach JC, Hood DD, Pfeifer BL. Intrathecal neostigmine for postoperative analgesia: dose response [abstract]. Reg Anesth 1996;21(Suppl 2):74.
6. Lauretti GR, Kitayama AT, Reis AL. A randomized double-blind prospective study of intrathecal neostigmine in orthopedic surgery [abstract]. Reg Anesth 1996;21(Suppl 2):76.
7. Naguib M, Yaksh TL. Antinociceptive effects of spinal cholinesterase inhibition and isobolographic analysis of the interaction with mu and alpha 2 receptor systems. Anesthesiology 1994;80:1338-48.
8. Dickenson AH, Sullivan AF. Combination therapy in analgesia; seeking synergy. Curr Opin Anaesth 1993;6:861-5.
9. Yaksh TL, Malmberg AB. Interaction of spinal modulatory systems. In: Fields HL, Liebeskind JC, eds. Progress in pain research and management. Seattle: IASP Press, 1994:151-71.
10. Bromage PR. A comparison of the hydrochloride salts of lidocaine and prilocaine in epidural analgesia. Acta Anesthesiol Scan 1965;16:55-9.
11. Winnie RP, Abram SE. Intrathecal morphine and neostigmine produce synergistic analgesia to noxious thermal stimuli in rats [abstract]. Reg Anesth 1994;19(Suppl 2):6.
12. Lauretti GR, Reis MP, Prado WA, Klamt JG. Dose-response study of intrathecal morphine versus intrathecal neostigmine, their combination, or placebo for postoperative analgesia in patients undergoing anterior and posterior vaginoplasty. Anesth Analg 1996;82:1182-7.
13. Gerbhardt B. Pharmacology and clinical results with peridural and intrathecal administration of ketamine. Anaesthesist 1994;43:34-40.
14. Aanonsen LM, Lei S, Wilcox GL. Excitatory amino acid receptors and nociceptive neurotransmission in rat spinal cord. Pain 1990;41:309-21.
15. Smiths DJ, Bouchal RL, DeSanctis CA, et al. Properties of the interaction between ketamine and opiate binding sites in vivo and in vitro. Neuropharmacology 1987;26:1253-60.
16. Ashford ML, Boden P, Ramsey RL, Usherwood PN. Enhancement of desensitization of quisqualate-type glutamate receptor by the dissociative anaesthetic ketamine. J Exp Biol 1989;141:73-86.
17. Ghorpade A, Advokat C. Evidence of a role for N-methyl-D-aspartate (NMDA) receptors in the facilitation of tail with-drawal after spinal transection. Pharmacol Biochem Behav 1994;48:175-81.
18. Roytblat L, Korotkoruchko A, Katz J, et al. Postoperative pain: the effect of low dose ketamine in addition to general anesthesia. Anesth Analg 1993;77:1161-5.
19. Arendt-Nielsen L, Peterson-Felix S, Fischer M, Bak P. The effect of N-methyl-D-aspartate antagonist (ketamine) on single and repeated nocicetive stimuli: a placebo-controlled experimental human study. Anesth Analg 1995;81:63-8.
20. Duriex ME. Inhibition by ketamine of muscarinic acetylcholine receptor function. Anesth Analg 1995;81:57-62.
21. Lauretti GR, Lima ICPR. Intrathecal neostigmine effects on somatic and visceral pain: improvement by association with a peripheral anticholinergic. Anesth Analg 1996;82:617-20.
22. Lauretti GR, Reis MP, Lima ICPR. Association of intrathecal fentanyl and intrathecal neostigmine: an effective postoperative analgesic for abdominal hysterectomy [abstract]. Reg Anesth 1996;21(Suppl 2):77.
23. Costello DJ, Borison HL. Naloxone antagonises narcotic self blockade of emesis in the cat. J Pharmacol Exp Ther 1977;203:222-30.
24. Harris AL. Citotoxic-therapy-induced vomiting is mediated via enkephalin pathways (hypothesis). Lancet 1982;1:714-6.
© 1996 International Anesthesia Research Society