The Effect of Naloxone on Ketamine-induced Effects on Hyperalgesia and Ketamine-induced Side Effects in Humans

Twenty-five healthy male volunteers were included in the study. None received any medication for at least 48 h before the study periods. Each participant in the study had been familiarized with the experiment protocol and the extent of the burn injury on a separate day. Informed consent was obtained, as was approval by the local ethics committee and the Danish National Board of Health.

Two subjects were excluded during the study, one because of a psychotomimetic reaction during infusion of study drug, and one because of recurring second-degree burn injuries to the test area.

The study was three-way cross-over, double-blind, randomized, and placebo-controlled. According to a computer-generated randomization list, the study drugs were prepared by a certified registered nurse with no other connection to the study. The three treatment regimens were:

1. Injection of naloxone, 0.8 mg/15 min, followed by 0.4 mg/h; 30 min after injection of naloxone, injection of intravenous ketamine 0.3 mg [middle dot] kg^-1[middle dot] 15 min^-1followed by 0.3 mg [middle dot] kg^-1[middle dot] h^-1

2. Injection of normal saline solution followed by infusion of normal saline; 30 min after injection of saline, injection of intravenous ketamine 0.3 mg [middle dot] kg^-1[middle dot] 15 min^-1followed by 0.3 mg [middle dot] kg^-1[middle dot] h^-1

3. Injection of normal saline solution followed by infusion of normal saline; after 30 min, injection of normal saline followed by infusion of saline

At 0 min, heat-pain detection threshold (HPDT) values were obtained, followed by induction of hyperalgesia.

At 60 min, measurements of HPDT values, areas of secondary hyperalgesia, and sedation were made.

At 75 min, initiation of naloxone-placebo-placebo infusion took place.

At 90 min, measurements of HPDT values, areas of secondary hyperalgesia, and sedation were made.

At 105 min, initiation of ketamine-ketamine-placebo infusion took place.

At 135 min, measurements of HPDT values, areas of secondary hyperalgesia, and sedation were made.

Heat-pain detection threshold was defined as the lowest temperature perceived as painful. A thermode identical to the thermode used for induction of the burn injury was applied to the skin.

Heat-pain detection thresholds were determined by increasing temperature of the thermode from 35 [degree sign]C with a rate of change of 1 [degree sign]C/s, having instructed the subjects to press a button on a device connected to the thermode indicating that the pertinent temperature had been reached. This temperature was registered, and the thermode automatically returned to baseline. A cut-off temperature limit of 52 [degree sign]C was defined, above which the thermode was to return to baseline and a threshold of 52 [degree sign]C would be registered. In no instances was this cut-off limit reached. Each threshold was calculated as the average of three measurements with randomized intervals (6-10 s) between each stimulation. Thermal thresholds were determined at the site of injury and at the corresponding site of the contralateral, unburned calf.

A model that has previously proven reliable was used to produce hyperalgesia. A first-degree burn injury was induced in the subjects by heating the medial surface of the right calf (L3-L4 dermatome) with a rectangular thermode measuring 25 x 50 mm (Thermotest, Somedic A/B, Sollentuna, Sweden) strapped to the skin. The borders of the thermode were carefully marked on the skin for subsequent accurate placement of the thermode at the same site. The temperature was 47 [degree sign]C, and heating time was 7 min. One hour after heating the area, primary hyperalgesia in the burned area and a zone of persistent secondary hyperalgesia around the burn injury could be demonstrated, as had previously been shown [24,25]and had recently been reproduced. [26] 

The temperature of the injured site on the right calf was stabilized at 35 [degree sign]C with the thermode, from 1 min before assessment, and throughout assessment of secondary hyperalgesia. The border area of hyperalgesia was determined by stroking with a brush or by pinpricking the calf along four linear paths arranged radially around the thermal injury. Pinprick was performed with the von Frey technique using a nylon filament with a diameter of 1 mm and a bending force of 1.15 N. Stroke hyperalgesia was induced by gently stroking the calf with a brush made of foam. The stimulation was initiated in an area well outside the area of hyperalgesia and gradually continued toward the area of the burn injury in steps of 5 mm at intervals of 1 s. Immediately after the determinations of hyperalgesia, the thermode was removed.

Sedation was assessed on an 11-point numeric scale (0 = completely awake; 10 = almost asleep). Discomfort was assessed on an 11-point numeric scale (0 = no discomfort; 10 = maximal discomfort). Nausea was registered on a four-grade verbal rating scale (none, light, moderate, or severe). Finally, subjects were asked if they experienced hallucinations or any other sensations.

Data are presented as medians (quartiles). Friedman's analysis of variance has been applied. All significant P values have been corrected, using Bonferroni correction for repeated measurements. Statistically significance was considered to be at P < 0.05.

Before entering statistical analyses, data regarding HPDT and areas of secondary hyperalgesia were normalized to achieve the same point of reference in subjects from all of the three study days.

In all three experimental settings, HPDT inside the burned area on the right calf decreased significantly 60 min after inducing a first-degree burn injury and remained decreased throughout the study (P < 0.05).

Heat-pain detection thresholds did not differ in the subjects regardless of infusion of naloxone or placebo (P = 0.96), nor did the combined effects of placebo followed by placebo, placebo followed by ketamine, or naloxone followed by ketamine induce any changes in HPDT (P = 0.07; Table 1). Thermal thresholds in the contralateral unburned calf did not differ from a baseline value, regardless of treatment (Table 1).

After induction of burn injury, no spontaneous pain or other sensations were experienced from the site of injury. Areas of secondary hyperalgesia toward pinprick or stroking with a brush around the injured area on the right calf were detected easily in all subjects from 60 min after induction of burn injury. Administration of naloxone compared with administration of placebo induced no significant changes in areas of secondary hyperalgesia toward pinprick (P = 0.20) or toward brush (P = 0.48; Figure 1and Figure 2). The ensuing infusion of ketamine, however, was found to reduce the areas of secondary hyperalgesia for both pinprick (P < 0.05) and stroking (P < 0.05) on the calf compared with placebo, regardless of preceding naloxone or placebo infusion (Figure 1and Figure 2). No significant differences in the degree of reduction of secondary hyperalgesia were observed between the two regimens in which ketamine was infused in the area of secondary hyperalgesia using either pinprick (P = 0.53) or stroke stimuli (P = 0.67).

No sedation was noted when infusing naloxone or placebo. However, when the subsequent infusion of ketamine was initiated, median sedation scores of 4 (quartiles, 2-6) for naloxone-ketamine infusion and 5 (quartiles, 3-6) for placebo-ketamine infusion were achieved. These scores did not differ (P = 0.82). The scores, however, proved significantly different from the infusion of placebo-placebo, for which the median sedation score was 0 (quartiles, 0; P < 0.05).

In six instances, reports of changed perceptions of body parts were encountered during infusion of ketamine regardless of concomitant infusion of naloxone. No nausea was reported. No other forms of discomfort were noted, and no definite hallucinations were reported.

The objective of this study was to investigate to what extent opioid receptor blockade by naloxone could inhibit a well-known clinical response to intravenous injection of ketamine. A prerequisite for the present study was the reproduction of results obtained by Ilkjaer et al., [4]showing that infusion of ketamine, 0.3 mg [middle dot] kg^-1[middle dot] h^-1, can decrease the secondary hyperalgesia occurring after a first-degree burn injury in humans. These findings have been reproduced.

In the present study, no opioid-mediated effect of ketamine was demonstrated, as the clinical effect of ketamine remained unchanged regardless of concomitant infusion of naloxone. This finding is in line with those of Maurset et al. [27]- that naloxone, 1.6 mg, prevented the analgesic effect of meperidine but had no effect on ketamine analgesia. Our findings, however, contradict animal studies, indicating that there is an effect of ketamine on the opioid receptor. [10,28] 

Regarding the size of our study population, the power of the study is 80%, given that the smallest reduction of the area of secondary hyperalgesia not to be overlooked would amount to 21%.

It could be argued that the dose of naloxone chosen was too small. A larger dose might have been required to antagonize any opioid-mediated effect of ketamine. In a study by Brennum et al., [17]which investigated the effect of opioids on the secondary hyperalgesia after a first-degree burn injury, the inhibitory effect of 4 mg epidurally injected morphine on this secondary hyperalgesia was antagonized by intravenous injection of naloxone, 0.1 mg/kg. However, Smith et al. [28]have demonstrated that the in vitro affinity of ketamine to opioid receptors is considerably weaker than other drugs showing primarily opioid receptor antagonism, and as such it should be possible to antagonize effects on the opioid receptor with a relatively smaller dose than that necessary to revert opioid receptor agonism in a normal clinical setting. It should be noted that the cumulative dose of naloxone given in the present study is the same as that used to reverse heroin-induced coma in drug addicts, [29]and as such any clinically significant effect of ketamine on the opioid receptors should be antagonized using this dose. Doses similar to or even smaller than those used in the present study have been effective in managing opioid-induced side effects. A dose of 0.8 mg (0.4 mg intravenously and 0.4 mg subcutaneously) has been shown to improve respiratory parameters after high-dose alfentanil anesthesia, [30]and considerably smaller doses of naloxone have been shown to antagonize pruritus after epidural morphine. [31] 

Apart from studying the effects of naloxone on the well-known effects of ketamine on secondary hyperalgesia, we have investigated the effects of naloxone on ketamine-induced sedation. No difference in sedation scores was found after ketamine infusion, regardless of concomitant naloxone infusion. This finding does not support findings by Stella et al., [14]who reported inhibition of the loss of consciousness after injection of ketamine, 0.4 mg/kg, by injection of naloxone, 0.006 mg/kg. Our findings rather support the findings of Hao et al., [15]who found that ketamine does not possess any opioid agonistic effect but acts by antagonizing excitatory amino acids. Based on our findings in this experimental setting, the beneficial effects of a combination of opioids and ketamine in controlling acute (postoperative) pain, as observed by Bristow and Orlikowski, [32]Javery et al., [33]and Wong et al., [34]may be explained by the different receptors involved and the possibility of acquiring an additive [35]or even a synergistic effect [36]by combining opioids and NMDA-receptor antagonists. This topic, however, is not fully elucidated yet, as Edwards et al. [37]failed to demonstrate any improvement in analgesia by adding ketamine to an infusion of morphine, and Ilkjaer et al. [38]were unable to improve postoperative pain control by adding ketamine to a combination of epidural local anesthetics and patient-controlled morphine injections.

Infusion of ketamine can decrease secondary hyperalgesia occurring after a first-degree burn injury in humans. In the present study, this effect was not significantly abolished by the concomitant administration of naloxone, although it is possible that some effect of naloxone might have been overlooked because of a type II error.

Sedation observed after administration of ketamine is not affected by naloxone. In this experimental setting, the effect of ketamine on secondary hyperalgesia was not affected by blocking opioid receptors with naloxone.