Guidelines for cancer pain management include nonsteroidal antiinflammatory drugs with opioids administered in a time-contingent manner. This study was designed to evaluate the role of oral ketamine or transdermal nitroglycerin polymer, a nitric oxide donor, as coadjuvants to oral morphine in cancer pain therapy.
After institutional approval and informed patient consent were obtained, 60 patients with cancer pain were randomized to one of four groups (n = 15) and studied prospectively to evaluate analgesia and any adverse effects. A visual analog scale that consisted of a 10-cm line with 0 representing "no pain at all" and 10 representing "the worst possible pain" was introduced. All patients were regularly taking oral amitriptyline 50 mg at bedtime. The morphine regimen was adjusted individually to a maximal oral dose of 80-90 mg/day to keep the visual analog scale score less than 4. When patients reported pain (visual analog scale of 4 or more), despite taking 80-90 mg oral morphine daily, the test drug was added as follows: the control group (CG) received an additional 20 mg oral morphine (10 mg at 12-h intervals); the nitroglycerin group (NG) received a 5-mg nitroglycerin patch daily; the ketamine group (KG) received 0.5 mg/kg oral ketamine at 12-h intervals; and the dipyrone group (DG) received 500 mg oral dipyrone at 6-h intervals. Patients were free to manipulate their daily morphine consumption when the test drug was introduced to keep their visual analog scale score less than 4.
The groups were similar with respect to demographic data and visual analog scale pain scores before treatment. The visual analog scale scores after the test drug was introduced were similar among the groups. The daily consumption of oral morphine was as follows: on day 15: CG = DG = NG (P > 0.05), CG > KG (P = 0.036); on day 20: CG > NG = KG (P < 0.02) (CG > KG, P < 0.005; CG > NG, P < 0.02), DG > KG (P < 0.05); on day 30: CG = DG > KG = NG (P < 0.05). Patients in the CG and DG groups reported somnolence, but patients in the NG and KG groups did not.
Low-dose ketamine and transdermal nitroglycerin were effective coadjuvant analgesics. In conjunction with their opioid tolerance-sparing function, joint delivery of ketamine or nitric oxide donors with opiates may be of significant benefit in cancer pain management.
Key words: nitric oxide; N-methyl-D-aspartate antagonist; opioid; tolerance.
This article is featured in "This Month in Anesthesiology." Please see this issue of Anesthesiology, page 7A.
THE World Health Organization has established fundamental guidelines for systemic analgesic therapy in an effort to alleviate pain from cancer by the year 2000, and they include the use of nonsteroidal antiinflammatory drugs and opioids. # Despite aggressive therapy, some patients with cancer continue to have severe pain because of dose-limiting side effects from opioids and because they become tolerant of or poorly responsive to opioids. Tolerance presents a challenge for pain management physicians. 
Recent insights into the mechanism underlying the development of tolerance to the analgesic effects of morphine have indicated the involvement of glutamate receptors. This finding suggests that oral N-methyl-D-aspartate (NMDA) antagonists combined with opiate analgesics may be an effective approach to simultaneously prevent opiate tolerance and enhance analgesia in rodents [2-4]and patients. In addition, there is also evidence that endogenous nitric oxide (NO) is necessary to inhibit nociceptive transmission. [6-8]Systemic morphine increases spinal cord NO metabolite concentrations, and behavioral analgesia in normal animals from systemic morphine is blocked by NO synthase inhibitors. Based on these data, we may be able to alter the time course of tolerance development or enhance analgesia with daily administration of ketamine or an NO donor. However, the usefulness of the administration of ketamine or NO donors to opioids in the early phase of cancer pain therapy has not been evaluated.
The current study evaluated the potential role of oral ketamine, an NMDA antagonist, or transdermal nitroglycerin, an NO donor, as coadjuvants to oral morphine in cancer pain therapy, compared with oral morphine alone or with the combination of a nonsteroidal antiinflammatory drug (dipyrone) and oral morphine.
Materials and Methods
The Ethics Committee at the University of Sao Paulo's Teaching Hospital, Ribeirao Preto, approved the study protocol. After giving written informed consent, 60 patients with cancer pain for whom tramadol or nonsteroidal drugs were ineffective were randomized to one of four groups (n = 15) and studied prospectively using a placebo-controlled design to evaluate analgesia and adverse effects. The number of patients enrolled was determined by a power analysis, based on preliminary data in the visual analog scores for pain, in which the [Greek small letter alpha] level was 5% and the [Greek small letter beta] level was 0.1. The concept of a visual analog scale (VAS), which consisted of a 10-cm line with 0 representing "no pain at all" and 10 representing "the worst possible pain" was introduced. The study investigators were not blinded because the test drugs were distinct (oral or transdermal, administered at fixed intervals based on the drug half-life). Nevertheless, this prospective study was intended to be a pilot work in preparation for a more definitive blinded clinical trial in the future. Patients brought their VAS scores recorded each day to their weekly appointments.
All patients were regularly taking 50 mg oral amitriptyline at bedtime. The morphine regimen was adjusted individually to a maximal oral dose of 80-90 mg/day to keep the VAS pain score less than 4. All patients in all groups had free access to as much morphine as they needed to a maximum dose of 80-90 mg/day. At that point, when patients reported pain (VAS >or= to 4), despite taking 80-90 mg oral morphine daily, the test drug was added as follows: the control group (CG) received 20 mg additional oral morphine (10 mg at 12-h intervals); the dipyrone group (DG) received 500 mg oral dipyrone at 6-h intervals; the ketamine group (KG) received 0.5 mg/kg oral ketamine at 12-h intervals; and the nitroglycerin group (NG) received a patch of 5 mg nitroglycerin daily (available commercially in Brazil as a 5-mg transdermal patch). After the test drug was introduced, patients were also free to manipulate their daily morphine consumption by adding more morphine to the 80- to 90-mg dose, to keep pain VAS less than 4. The daily consumption of morphine and the VAS scores for pain were noted on days 1, 5, 10, 15, 20, and 30 after the test drug was introduced. Any adverse effects were noted and treated, if necessary.
The normality of the distributions was assessed using the Shapiro-Wilk test. Groups were compared for demographic data (age, weight, and height) using one-way analysis of variance. The incidence of adverse events, patient gender, and the site of primary disease were compared among groups using chi-square analysis corrected for multiple comparisons. P < 0.0125 was considered significant. The VAS scores and the daily morphine consumption on days 1, 5, 10, 15, 20, and 30 were compared among groups using two-way analysis of variance for repeated measures. The statistical analysis compared the four groups on each established day of the study. Tukey honest analysis was applied to correct P values for multiple group comparisons. P < 0.05 was considered significant. Data are expressed as the mean +/- SD.
The groups showed no differences regarding gender, weight, age, height (Table 1), or distribution of the primary cancer site (Table 2). The VAS scores for pain before the oral morphine treatment were similar among groups: CG (7.6 +/- 1.9) = DG (7.6 +/- 1.7) = KG (7.4 +/- 1.5) = NG (7.9 +/- 1.6) (P = 0.774).
The period from the first dose of oral morphine administration until the time of the test drug administration was similar among groups (28-74 days, P = 0.768). The VAS pain scores after the test drug was introduced were similar among groups: CG (3.9 +/- 2.6) = DG (4 +/- 1.6) = KG (3.3 +/- 1.6) = NG (3.5 +/- 1.7) (P = 0.967). Figure 1shows the daily consumption dose of oral morphine on days 1, 5, 10, 15, 20, and 30 after the test drug was introduced. The daily consumption of oral morphine on day 15 was: CG = DG = NG (P > 0.05); DG = NG = KG (P > 0.05); and CG > KG (P = 0.036). On day 15, the KG was only statistically different from the CG. On day 20, the morphine consumption was CG = DG (P = 0.873); CG > NG = KG (P < 0.02) (CG > KG, P = 0.004; CG > NG, P = 0.0183); DG = NG (P = 0.10); and DG > KG (P = 0.04). On day 30, the oral morphine consumption was CG = DG (P = 0.815); CG > KG (P = 0.003); CG > NG (P = 0.012); DG > NG (P = 0.021); DG > KG (P = 0.001); KG = NG (P = 0.964).
(Table 3) shows the incidence of adverse effects. Patients from the CG and the DG reported more somnolence compared with the KG and the NG (P < 0.013). One patient from the NG was withdrawn from the study because of intense headache and was replaced by another patient to keep 15 patients in each group. One patient from the KG reported frequent hallucinations, and the oral dose was changed from 0.5 mg/kg to 0.25 mg/kg twice daily. This patient was included in the data analysis and compilation of adverse effects, even after the hallucinations resolved with manipulation of the ketamine dose.
In our clinic, we routinely combine the nonsteroidal antiinflammatory analgesic dipyrone (500 mg at 6-h intervals in adults) with morphine to treat cancer pain. Dipyrone has been shown to have peripheral and central actions on the synthesis of prostaglandins and indirect action on central opioid receptors, after enkephalin and endorphins are liberated. Because animal data suggest a possible role for NMDA antagonists [2-4]and NO [6-9]in pain control, we decided to evaluate the analgesic action of oral ketamine and transdermal nitroglycerin, compared with morphine alone or combined with dipyrone. Nevertheless, patients who received oral dipyrone at 6-h intervals did not have a particularly impressive outcome compared with the CG during the observation period.
In the current study, the analgesic effect was maintained during the 30-day observation period with minimal adverse effects and without the need to increase the morphine dose in patients with chronic cancer pain treated with low doses of oral ketamine and oral morphine. An analgesic effect of ketamine or its effect on the modulation of morphine's tolerance was evident after day 15, compared with the CG. Similarly, patients who received daily transdermal 5-mg nitroglycerin patch therapy also displayed an analgesic effect that was evident after day 20 compared with the CG. After day 30, both the KG and the NG consumed less morphine compared with the DG or the CG. Patients from the KG and the NG reported significantly less somnolence and apparently less constipation and nausea or vomiting, and had a better outcome and delayed need for invasive procedures, such as spinal pain control.
In this study, we fixed the maximum daily morphine dose to 80-90 mg/day. The protocol for cancer pain management in the Center for Pain in our institution includes a maximum oral morphine dose of 80-90 mg/day combined with the nonsteroidal antiinflammatory drug dipyrone on a time-contingent schedule. Although the World Health Organization recommends titration of morphine to effect and higher doses are used in the literature, the limited dose of morphine (80-90 mg) is based on (1) the observation of fewer adverse effects, such as constipation and nausea, and consequently less need for intervention, such as with bowel obstruction or antiemesis; (2) the finding that the hospital facility gave 60 oral tablets of 10-mg morphine weekly/patient, free of charge, when this study was conducted, which is sufficient for 7 days of treatment, and normally the patients cannot afford to buy extra morphine tablets; and finally (3) the finding that the NMDA receptor antagonist MK801 administered to morphine-tolerant rats was effective in restoring their sensitivity to morphine. Nevertheless, previous long-term blockade of NMDA receptors with ketamine (40 mg/kg, 7 times daily for 5 days) intensified morphine dependence in rats exposed to morphine implanted subcutaneously. In addition, our policy generally includes epidural analgesia after systemic treatment using a multimodal approach.
Animal studies have shown that morphine can block the initial response of dorsal horn nociceptive neurons to noxious stimulation but has little effect on the potentiation of the response of these cells to the repeated stimulation (wind up), whereas the reverse is true for NMDA receptor antagonists. Thus, a combination of both drugs may be advantageous in chronic pain states. Ketamine has analgesic properties that are mediated by several mechanisms. After oral administration of 0.5 mg/kg ketamine, approximately 20% is absorbed, and its analgesic action seems to be mediated at least in part by its first metabolite, norketamine, which has a half-life of 12 h. Ketamine itself binds stereospecifically to opiate receptors, but a significant contribution to its analgesic effectiveness may come from an interaction with cholinergic, adrenergic, and 5-hydroxytryptamine systems. A direct action of ketamine on the dorsal horn also has been reported. Apart from this action, ketamine can prevent action-potential conduction by an effect on the sodium and potassium channels in nerve membranes, and thus it is thought to have local anesthetic properties. Finally, ketamine can selectively block the NMDA excitation of central neurons through interaction at the receptor at two distinct sites. One site is located within the channel pore and a second site is associated with a hydrophobic domain of the protein. Although binding to the site within the channel would decrease the channel open-time, binding to the membrane-associated site does not require the channel to be in the open state and decreases the frequency of channel opening through an allosteric mechanism. 
In addition to the development of central sensitization, the phenomenon of opiate analgesic tolerance, leading to a decreased analgesic response when a fixed opioid dose is administered repetitively, also complicates long-term pain management. The subtype [Greek small letter kappa]-opioid receptor, the vasopressin receptor, and the NMDA receptor are all involved in the development of morphine tolerance. After long-term morphine treatment, [Greek small letter kappa]-opioid receptor activity may be suppressed before the activation of vasopressin or NMDA receptors during the development of morphine tolerance. In addition, many effects of NMDA receptor stimulation are mediated through the intracellular formation of NO, and a role of NO in opiate tolerance and withdrawal has been shown. Thus, the development of morphine tolerance may involve the activation of NMDA receptors and NO production, which contributes to central sensitization. In contrast to animal studies, systemic L-arginine administration produces analgesia in some patients with chronic pain, and NO interacts synergistically with morphine after intravenous or spinal administration. Data from the literature suggest that in humans high doses of transdermal nitroglycerin, such as 30 mg daily, are hyperalgesic, sup ** whereas daily doses less than 6 mg are analgesic during different circumstances. [26-28]
To explain this disparity, the literature suggests that the NO contribution to antinociception or pain may be related to preferential actions on superficial spinal cord layers, but other explanations also cannot be dismissed. Nitric oxide mechanisms of action are likely to include activation of second messengers such as cyclic guanosine monophosphate. Activation of the NO-cyclic guanosine monophosphate signal transduction system contributes to the sensitization of wide-dynamic-range neurons located in the superficial and deep dorsal horns. Sensitization of the deep dorsal horn cells by cyclic guanosine monophosphate increases the responses of deep wide-range neurons to weak and strong peripheral mechanical stimulation, and simultaneously attenuates the inhibition of the same neurons produced by stimulation in the periaqueductal gray, resulting in the transmission of painful stimuli. In contrast, wide-range neurons in the superficial dorsal horn and high-threshold cells in the superficial or deep layers have lesser responses after exposure to cyclic guanosine monophosphate and result in antinociceptive effects. Recently, another article suggested that peripherally applied NO donors have no analgesic effects themselves but enhance the analgesic effects of peripherally administered morphine during inflammation. 
In conclusion, low-dose ketamine and transdermal nitroglycerin were effective coadjuvant analgesics. In conjunction with their opioid tolerance-sparing function, joint delivery of ketamine or NO donors with opiates may have significant benefits in cancer pain management. This pilot study is an important impetus that suggests that the World Health Organization ladder perhaps should include new drugs to delay morphine tolerance and decrease the incidence of adverse effects related to high doses of opiates.
The authors thank Mark J. Lema, M.D., Ph.D., of the Roswell Park Cancer Institute for reviewing the manuscript and offering helpful comments.
# Ferrante FM: ASA Guidelines for cancer pain management. The American Society of Regional Anesthesia, 1996 Annual Meeting, pp 423-69.
** Koerig HM, Chowdhury P: A comparison of the effects of nitroglycerine ointment, EMLA cream and zinc oxide on the degree of pain associated with pin prick (abstract). Anesth Analg 1996; 82:240