Low-dose S+ ketamine in target-controlled intravenous anaesthesia with remifentanil and propofol for open gynaecological surgery: A randomised controlled trial.

BACKGROUND AND AIMS: Using remifentanil-propofol target-controlled infusion (TCI) in open gynaecological surgeries could be associated with opioid-induced hyperalgesia postoperatively. This study's aim was to investigate the effect of low-dose S-ketamine compared with control on cumulative morphine consumption 24 h postoperatively in women undergoing open abdominal hysterectomy with remifentanil-propofol TCI technique.
METHODS: Ninety female patients above 21 years old who underwent elective open abdominal hysterectomy under general anaesthesia with remifentanil-propofol TCI were recruited. They were randomised to receive either normal saline as control (n = 44) or 0.25 mg/kg intravenous boluses of S-ketamine before skin incision and after complete removal of uterus (n = 45). The primary outcome measure was cumulative morphine consumption measured over 24 h postoperatively. The secondary outcome measures were incidences of opioid-related and psychotomimetic side effects, pain and level of sedation scores.
RESULTS: The cumulative 24-h morphine consumption postoperatively (P = 0.0547) did not differ between both the groups. S-ketamine group had slower emergence from general anaesthesia (P = 0.0308) and lower pain scores (P = 0.0359) 15 min postoperatively. Sedation level, common opioid-related side effects (nausea, vomiting, pruritus), respiratory depression and psychotomimetic side effects were similar between both the study groups.
CONCLUSION: Low-dose S-ketamine did not reduce the total cumulative morphine consumption in patients undergoing major open gynaecological surgeries with remifentanil-propofol TCI.

INTRODUCTION

The use of general anaesthesia using the remifentanil–propofol target-controlled infusion (TCI) system is a viable anaesthetic option in gynaecological procedures. The remifentanil–propofol TCI regimen is associated with better haemodynamic stability intraoperatively, faster emergence from anaesthesia postoperatively and lower incidence of postoperative nausea and vomiting (PONV).[1] Gynaecological surgeries are associated with a higher risk of PONV.[2] However, the use of remifentanil may lead to secondary hyperalgesia and increased opioid requirements in the postoperative period.[3] To mitigate this, the use of low-dose ketamine could reduce the risk of opioid-induced hyperalgesia and reduce postoperative pain and opioid consumption.[4]

There is interest in the use of S-ketamine, instead of racaemic mixture, to reduce opioid-induced hyperalgesia. S-ketamine is twice as potent as the racaemic mixture,[5] and therefore lower doses of S-ketamine is needed to achieve the same clinical effects and to reduce the associated psychotomimetic side effects. There is limited knowledge of the use of low-dose S-ketamine in open gynaecological surgeries. We investigated the effect of low-dose S-ketamine on postoperative cumulative morphine consumption, opioid-induced and psychotomimetic side effects in women undergoing elective open abdominal hysterectomies with remifentanil–propofol TCI.

METHODS

A randomised, double-blinded clinical trial was conducted to compare the effects of S-ketamine in women undergoing elective open abdominal hysterectomy with remifentanil–propofol TCI. Written informed consent was obtained from all patients in this study before any study procedures. The study period was between October 2014 and January 2015. Ninety women above 21 years and below 70 years belonging to the American Society of Anesthesiologists' (ASA) status 1 or 2 were recruited. The exclusion criteria were patients with contraindications to S-ketamine listed in the product label, uncontrolled hypertension, psychiatric disorders, body mass index above 35 kg/m2, history of drug or alcohol abuse, regular use of analgesics or opioids within 12 h prior to surgery, chronic use of benzodiazepines or neuroleptics and on thyroid hormone replacement therapies. Pregnant and breastfeeding patients were also excluded from this study.

Patients were randomly assigned into two groups with a 1:1 allocation ratio. Allocation concealment was performed using random numbers pregenerated by an off-site statistician not involved in patient recruitment. Implementation was carried out via serially numbered opaque sealed envelopes, and the serial numbers were performed using a computerised random number generator with a permuted block randomisation scheme.

Continuous electrocardiogram, noninvasive blood pressure, pulse oximeter, capnogram and bispectral index monitoring were placed for all patients. The study patients received midazolam 0.03 mg/kg intravenously before induction of anaesthesia. In the control group, intravenous normal saline 0.25 mg/kg bolus was administered before skin incision as preemptive analgesia. A second dose of normal saline 0.25 mg/kg bolus was given upon complete surgical removal of uterus. In the treatment (S-ketamine) group, intravenous S-ketamine (Ketanest® S; Pfizer, Berlin, Germany) of similar volume was given in place of normal saline at the same time points of the surgery.

After adequate preoxygenation with a face mask, anaesthesia was induced and maintained with 1% propofol and remifentanil 20 μg/mL titrated doses through the TCI pump (Alaris® Asena PK; Cardinal Health, Basingstoke, UK). Muscle relaxation was achieved with weight-adjusted atracurium dosing by the attending anaesthetist. Intubation was performed, using direct or video laryngoscopy, using appropriate sized endotracheal tube by the attending anaesthetist and ventilated with 50% oxygen in air mixture. Maintenance of general anaesthesia was achieved using total intravenous anaesthesia with remifentanil infusion according to Minto pharmacokinetic model and propofol infusion according to Schneider pharmacokinetic model through the TCI pump.[67] Intravenous morphine 0.05 mg/kg was administered after the complete surgical removal of uterus. Vital signs (heart rate, blood pressure, oxygen saturation, capnography, bispectral index) were monitored intraoperatively. After the completion of surgery, neuromuscular blockade was reversed with standard doses of neostigmine and atropine by the attending anaesthetist.

Upon recovering consciousness from general anaesthesia postoperatively, patients were provided with patient-controlled analgesia (PCA) pump using intravenous morphine 1 mg bolus with a 5-min lockout interval and no background infusion. The maximum hourly limit was 8 mg morphine. Preoperative education on the use of pain-controlled analgesia pump and assessment of pain experienced at rest using the visual analogue scale, a graded scale with numbers from 0 to 10 drawn for patients to indicate their pain score, was performed. PCA morphine was maintained for at least the next 24 h according to hospital and acute pain service guidelines. Nonopioid analgesia, such as paracetamol, and nonsteroidal anti-inflammatory drugs (mefenamic acid, naproxen, etoricoxib) were provided to the patients as needed for pain relief as reviewed and decided by the attending anaesthetist, and the respective dosages were recorded accordingly. Should patients verbalise the need for additional pain relief, the attending anaesthetist would review and provide intravenous boluses of morphine (rescue morphine). Analgesia dose frequencies were given according to standard hospital and acute pain service guidelines. Similarly, rescue ondansetron could also be administered for PONV.

The anaesthetists who were on acute pain service, blinded to the randomisation procedure, were involved in the postoperative data collection. Any opioid-related and psychotomimetic side effects would be treated and monitored till discharge using hospital and acute pain service guidelines. Patients were assessed for pain at 15 min, 30 min, 45 min, 1 h, 2 h, 4 h, 6 h, 12 h, and 24 h, postoperatively. We assessed for opioid-related side effects (nausea, vomiting, pruritus, respiratory depression) in the first 24 h postoperatively. Respiratory depression was defined as having a respiratory rate of 8 or less breaths per minute. The level of alertness/sedation was assessed using the responsiveness component of the Observer's Assessment of Alertness/Sedation (OAA/S) scale at 15 min, 30 min, 45 min, 1 h, 2 h, and 6 h, postoperatively.[8] Psychotomimetic side effects were evaluated 24 h postoperatively using Profile of Mood Scale (POMS) to assess the patients' mood states[9] and a questionnaire to look for any abnormal sensory perception. Patients were also asked to rate their satisfaction with their overall experience using a 3-point verbal rating scale (1 = very satisfied, 2 = satisfied, and 3 = dissatisfied).

The primary outcome is the cumulative morphine consumption in the first 24 h postoperatively. The required sample size for the study was estimated to be 88 patients (44 patients per group). Sample size calculation was based on the following parameters using a pilot study at our centre on morphine consumption in open abdominal hysterectomy: the mean ± standard deviation (SD) of total cumulative morphine consumption at 20 ± 15 mg, expected reduction in morphine consumption of 45% in S-ketamine group to 11 ± 15 mg, power of 80%, a type 1 error (alpha) of 5%[10] and an allocation ratio of 1:1. Using unequal variance two-sample t-test for comparing means of two independent groups, 44 patients would be required per group. Given that the patients would remain in the hospital after surgery for longer than the time required for this study, a low drop-out rate is anticipated in this study.

We verified the normality of data distribution and considered the absolute difference in means between S-ketamine and control groups by taking differences between log-transformed total cumulative morphine consumption in the first 24 h postoperatively. These were expressed as absolute differences with its corresponding 95% confidence interval (CI). The differences were tested using unequal variances two-sample t-test for statistical significance. All covariates were summarised in terms of two intervention groups. Continuous variables were summarised as mean ± SD or median (interquartile range), whichever appropriate, and categorical variables were summarised as frequency (proportion). Associations between intervention groups and other covariates were tested using Chi-square test and two-sample t-test/Mann–Whitney U test for categorical and continuous data, respectively. Differences between S-ketamine and control groups at each time point were estimated using repeated measures mixed-models (repeated analysis of variance) regression analyses, which accounted for the dependence among repeated measurements of the same patient. Mixed-models for continuous data were used with time and intervention group interaction as fixed-effects, time as repeated measure and variance components as covariance matrix. The estimation method was based on a residual (restricted) maximum likelihood technique and the variance–covariance matrix of the parameter estimates computed using a sandwich (empirical) estimator. Differences between the two groups at different time points were reported as mean difference with 95% CI. Significance level was set at 0.05, and all statistical tests were two-sided. All statistical analyses were performed using SAS version 9.4 software (SAS Institute, Cary, NC, USA).

RESULTS

A total of 90 patients were recruited into this study. One patient was excluded due to persistent preoperative hypertension necessitating cancellation of elective surgery (control group n = 44, S-ketamine group n = 45), and the CONSORT diagram is shown [Figure 1]. There were no significant differences between the two groups in terms of patient characteristics and anaesthesia care [Table 1]. Patients in the control group (mean ± SD; 11.6 ± 6.37 s) had faster emergence from general anaesthesia when compared with patients in the S-ketamine group (14.6 ± 7.55 s; P = 0.0308).

Figure 1

CONSORT diagram for study recruitment

Table 1

Patient demographics and anaesthetic data

	Control (n=44)	S-ketamine (n=45)	P
Age (years)	48.1±4.86	48.1±3.55	0.9049
Weight (kg)	62.2±8.35	62.7±11.09	0.7834
Body mass index (kg/m2)	25.0±3.20	25.1±4.17	0.9705
Duration of anaesthesia (min)	132.5±37.65	134.0±43.13	0.9935
Intraoperative remifentanil usage (µg)	1019.7±351.70	1020.8±351.76	0.9935
Intraoperative propofol usage (mg)	1246.6±412.93	1274.8±342.93	0.6340
BIS at tracheal extubation	79.5±6.54	79.2±4.69	0.4151
Time to eye opening after stopping TCI (s)	11.6±6.37	14.6±7.55	0.0308
Duration of stay in recovery unit (min)	98.0±31.97	99.3±33.73	0.8922

TCI – Target-controlled infusion; SD – Standard deviation. Values are represented as mean±SD

The primary outcome measure of total cumulative morphine consumption (both in recovery and 24 h postoperatively) was not significantly different between both the groups [Table 2]. The corresponding logarithm of total cumulative morphine consumption at 24 h postoperatively found that the S-ketamine group had significantly lesser cumulative morphine consumption than the control group (P = 0.0263), with the estimated absolute difference (95% CI) between both the groups being 0.352 (0.043, 0.661). Both groups also had similar demands for morphine boluses from PCA during their time in the recovery unit. The amount of nonopioid analgesic consumption over 24 h postoperatively by the patients in the S-ketamine group did not differ from those in the control group.

Table 2

Analgesic and antiemetic agent consumption in recovery unit and 24 h postoperatively

	Control (n=44)	S-ketamine (n=45)	P
In recovery
Number of times total demands for PCA	12.2±13.35	10.4±9.43	0.5706
Number of successful demands from PCA	7.3±4.79	6.0±4.26	0.2119
Rescue morphine consumption (mg)	1.7±2.62	1.2±2.11	0.3311
Rescue ondansetron consumption (mg)	0.4±1.16	0.4±1.27	0.7533
Total morphine consumption (mg)	8.9±6.45	7.2±5.21	0.2184
24 h postoperatively
Total morphine consumption (mg)	21.0±11.4	16.4±10.52	0.0547
Logarithm of total morphine consumption (mg)	2.9±0.61	2.5±0.84	0.0263
Total nonopioid analgesia consumption
Oral paracetamol (g)	1.5±1.32	1.5±1.14	0.9797
Per-rectal paracetamol (mg)	265.9±643	216.7±460	0.9235
Oral mefenamic acid (mg)	318.2±471.16	466.7±578.01	0.2363
Oral etoricoxib (mg)	7.5±24.3	1.3±8.94	0.1529
Oral naproxen (mg)	37.5±184	6.1±41.0	0.5288
Total antiemetic consumption
Intravenous ondansetron (mg)	9.7±4.83	9.3±4.90	0.6506
Intravenous metoclopramide (mg)	1.4±6.32	2.4±6.45	0.1978

PCA – Patient-controlled analgesia; SD – Standard deviation. Values are represented as mean±SD

The incidence of nausea was 40.0% and 31.8% and the incidence of vomiting was 28.9% and 13.6% in the S-ketamine and control groups, respectively [Table 3]. There were no significant differences in nausea and vomiting between groups. The incidence of pruritus was 11.1% and 13.6% in the S-ketamine and control groups, respectively (P = 0.7578). None of the patients experienced respiratory depression. With regard to the POMS results, there was no overall significant difference in patient mood states between groups. In terms of psychotomimetic side effects, there were no significant differences between the two study groups. The most commonly experienced side effects were feelings of altered physical strength, dizziness and abnormal movements. We found that 51.7% of patients reported 'feelings of altered physical strength', defined as patients feeling weaker or stronger in physical strength than before intervention. Furthermore, 39.3% of patients reported dizziness and 12.4% of patients experienced 'abnormal movements', defined as occurrence of involuntary sudden twitches, in the first 24 h postoperatively. Three (3.4%) patients experienced visual and/or auditory hallucinations in the postoperative setting. Two of them required clinical assessment and verbal reassurance from the medical team, while one patient did not alert the team until at the conclusion of the study. Nevertheless, all three patients continued with the study as their hallucinations resolved spontaneously over time. Two (2.2%) patients experienced reduced visual acuity: one patient experienced initial blurring of vision for 2 h before spontaneous resolution, while the other experienced transient diplopia. One patient noted that she had reduced hearing because she heard voices in her ear while she was sleeping; this episode resolved spontaneously.

Table 3

Patient-reported functional status experienced 24 h postoperatively

	Control (n=44)	S-ketamine (n=45)	P
Patient satisfaction			0.6377
Highly satisfied	19 (43.2%)	23 (51.1%)
Satisfied	23 (52.3%)	21 (46.7%)
Unsatisfied	2 (4.5%)	1 (2.2%)
Opioid-related side effects
Nausea	14 (31.8%)	18 (40.0%)	0.5092
Vomiting	6 (13.6%)	13 (28.9%)	0.1198
Pruritus	6 (13.6%)	5 (11.1%)	0.7578
Profile of Mood State
Anger	0.0±0.00	0.1±0.36	0.0832
Depression	0.0±0.00	0.0±0.15	0.3227
Confusion	0.0±0.15	0.0±0.15	0.9872
Fatigue	0.5±0.63	0.5±0.55	0.9293
Anxiety	0.1±0.36	0.1±0.25	0.9548
Hyperactivity	0.6±0.62	0.5±0.59	0.3267
Total score	1.3±1.08	1.2±1.07	0.5823
Psychotomimetic side effects
Feeling of altered physical strength	21 (47.7%)	25 (55.6%)	0.5273
Dizziness	21 (47.7%)	14 (31.1%)	0.1315
Abnormal movements	6 (13.6%)	5 (11.1%)	0.7578
Anxiety	3 (6.8%)	4 (8.9%)	1.0000
Visual and/or auditory hallucinations	1 (2.3%)	2 (4.4%)	1.0000
Increased salivation	2 (4.5%)	2 (4.4%)	1.0000
Reduced visual acuity	2 (4.5%)	0	0.2416
Reduced hearing	0	1 (2.2%)	0.1198
Aggression	0	0	0.5092

SD – Standard deviation. Values are represented as mean±SD or frequency (%) as appropriate

The average pain scores on exertion between two groups were not significantly different (P = 0.9003). Based on the linear mixed effect model, the pain scores at rest recorded at 15 min postoperatively were significantly lower in the S-ketamine group when compared with the control group [mean difference (95% CI): −1.016 (−1.965, −0.067), P = 0.0359] [Figure 2]. At 24 h postoperatively, the mean (SD) pain scores at rest were 3.4 (1.83) and 3.0 (1.87) for S-ketamine and control groups, respectively [mean difference (95% CI): 0.401 (−0.549, 1.35), P = 0.4077]. In addition, levels of alertness/sedation score recorded at several time points postoperatively were higher in the S-ketamine group when compared with the control group. However, differences in alertness were not significant at any time point [Figure 3].

Figure 2

(a) Visual analogue scale (VAS) pain scores at rest over time until 24 h postoperatively. (b) VAS pain scores at rest over time until first 4 h postoperatively. Values are estimated mean (95% CI). *P < 0.05

Figure 3

Observer's Assessment of Alertness/Sedation (OAA/S) scale assessing level of sedation over time until 6 h postoperatively. Values are estimated mean (95% CI)

Overall, 97.8% of patients from the S-ketamine group and 95.5% of patients from the control group were satisfied of their perioperative care. There were no reports of adverse events during the duration of this study.

DISCUSSION

Low-dose S-ketamine did not reduce morphine consumption at 24 h postoperatively compared with control with remifentanil–propofol TCI-based anaesthesia for abdominal open hysterectomy. Similarly, the use of low-dose S-ketamine did not increase sedation, opioid-related and psychotomimetic side effects. We observed lower patient-reported pain scores at rest 15 min postoperatively. There was slower emergence from general anaesthesia in those who received S-ketamine.

The use of opioids may activate N-methyl-D-aspartate (NMDA) receptors, leading to acute opioid tolerance and hyperalgesia.[11] Among the opioids, remifentanil is especially associated with causing acute tolerance and hyperalgesia, leading to decreased sensitivity to opioids and greater postoperative pain and possibly higher opioid consumption.[1213] Remifentanil acts directly on NMDA receptor subunits, causing inadvertent activation of NMDA receptors and the development of hyperalgesia.[14] Furthermore, the use of glycine as a drug vehicle in several formulations of remifentanil could also contribute to NMDA receptor activation.[15] The concurrent use of ketamine, a noncompetitive NMDA receptor antagonist, may prevent the development of opioid-induced hyperalgesia.[16] However, the use of low-dose S-ketamine did not reduce the morphine consumption in this study.

S-ketamine is the left-handed optical isomer of racaemic ketamine. It has a fourfold higher affinity for NMDA receptors than its stereoisomer R-ketamine[17] and has twice more analgesic potency than of racaemic ketamine.[5] Moreover, it has been shown in previous studies that patients who received S-ketamine had better mood and lesser postoperative pain.[18] It is postulated that the better mood these patients experienced may play a key influence in pain control due to the interactions between the cortico-striatal-thalamic feedback loops and the dopaminergic pathways.[18]

Low-dose racaemic ketamine is defined as an intravenous bolus dose of less than 1 mg/kg.[19] As low doses, ketamine will preferentially bind to the postsynaptic NMDA receptors in the dorsal horn of the spinal cord rather than those in the brain. Hence, analgesia may be achieved with lower risk of psychotomimetic side effects of ketamine.[202122] With regard to S-ketamine, single bolus of preincision 0.5 mg/kg S-ketamine or single bolus of preincision 0.5 mg/kg followed by repeated boluses of 0.2 mg/kg at every 20-min intervals intraoperatively has been shown to lower postoperative pain than placebo in patients undergoing major abdominal surgeries.[18] Furthermore, a systematic review on the perioperative use of ketamine at subanaesthetic concentrations has suggested preincision 0.35 mg/kg S-ketamine bolus followed by repeated boluses of 0.2 mg/kg at 20-min intervals intraoperatively for major visceral surgeries.[22] In this study, we chose a conservative regimen to reduce the risk of psychotomimetic side effects resulting in a lower dose of S-ketamine.

Despite the potential benefits of low-dose S-ketamine to mitigate opioid-induced hyperalgesia, there is still controversy with regard to its clinical efficacy. In our study, low-dose S-ketamine did not reduce morphine consumption. Our findings are consistent with those seen in a randomised, double-blinded trial, which involved 30 paediatric patients who have received alfentanil–propofol TCI for major urological surgeries, that concluded that the use of low-dose S-ketamine did not reduce total morphine consumption after the first 72 h postoperatively.[23] Similarly, one study examined the effects of 0.5 mg/kg bolus of S-ketamine before skin incision and followed by a continuous infusion of 2 μg/kg/min until 2 h after emergence in 30 patients undergoing knee arthroscopy with remifentanil–propofol TCI and concluded that S-ketamine did not reduce total morphine consumption or pain scores either at rest or on exertion during the first 5 days postoperatively.[24]

Conversely, there are studies that the use of low-dose S-ketamine significantly reduces total opioid consumption. In a double-blinded, randomised trial of 90 patients undergoing elective coronary artery bypass graft surgeries with alfentanil–propofol TCI-based anaesthesia, it established that small doses of S-ketamine significantly reduced opioid consumption during the first 48 h postoperatively.[25] Similar conclusions have also been made in another randomised, double-blinded prospective study of 28 patients undergoing radical prostatectomies.[26] Based on these above-mentioned studies, it is reasonable to conclude that there is currently no general consensus pertaining to the efficacy of low-dose S-ketamine on the reduction of postoperative opioid consumption. It is possible that the negative findings from our study could be due to the inappropriately low dosage (0.25 mg/kg) of S-ketamine adopted, which is in contrast to those used in the above-mentioned studies. It is noteworthy that the logarithm of the cumulative morphine consumption 24 h postoperatively showed that the low-dose S-ketamine group has significantly lesser cumulative morphine consumption than patients on the control group.

We did not find any significant difference in pain scores at rest, except for lower pain scores at rest 15 min postoperatively. Both patient groups showed trend to a decrease in pain scores at rest in the first 12 h postoperatively with the use of S-ketamine. Similar findings also were found for patients undergoing major cardiac surgeries and knee arthroscopic surgeries.[2425]

The use of low-dose S-ketamine led to a longer emergence time from general anaesthesia, due to the sedative effects. This finding is the same as that in one published study that noted additional sedative effects from low-dose S-ketamine in paediatric patients undergoing major urological surgeries.[23] However, while our study showed statistical significance in a greater recovery time with the use of low-dose S-ketamine, this may not be clinically significant given that the difference in recovery time was 3 s.

There were no increased psychotomimetic side effects with S-ketamine use. The use of racaemic ketamine has traditionally been associated with development of dose-dependent psychotomimetic side effects, most commonly hallucinations and sensory dissociation, in up to 30% of patients.[27] S-ketamine has been shown to enhance anaesthetic potency and produce milder side effects than racaemic ketamine.[28] Moreover, the use of subanaesthetic doses further reduces the potential occurrences of these psychotomimetic side effects.[20] Our findings are in keeping with previous studies investigating on low-dose S-ketamine.[18232426] However, one study reported that 4 patients, out of 44 patients, who received small doses of S-ketamine in the study did experience psychotomimetic adverse effects, with 2 patients withdrawn from the study.[26]

CONCLUSION

Low-dose S-ketamine did not reduce the total cumulative morphine consumption in patients undergoing major open gynaecological surgeries with remifentanil–propofol TCI.

Ethics approval

This study was approved (CIRB reference number: 2008/905/D) by the SingHealth Centralized Institutional Review Board, Singapore, and registered on Clinicaltrials.gov (NCT03231683).

Financial support and sponsorship

This work was supported by the KK Women's and Children's Hospital Endowment Fund, Singapore (KKHHEF/2013/11).

Conflicts of interest

There are no conflicts of interest.