Safety, Tolerability, and Pharmacokinetics of the N-Methyl-d-Aspartate Antagonist Dextrorphan in Patients With Acute Stroke
Background and Purpose Dextrorphan hydrochloride is a noncompetitive N-methyl-d-aspartate antagonist that is neuroprotective in experimental models of focal brain ischemia. The purpose of this study was to determine the maximum loading dose and maintenance infusion of dextrorphan hydrochloride that are well tolerated in patients with an acute stroke.
Methods An intravenous infusion of dextrorphan or placebo was begun within 48 hours of onset of a mild-to-moderate hemispheric stroke. Initially, patients were treated with either placebo (n=15) or dextrorphan (n=22) using a 1-hour loading dose (60 to 150 mg) followed by a 23-hour ascending-dose maintenance infusion (maximum total dose, 3310 mg). Subsequently, 29 patients were treated with dextrorphan in an open trial using a 1-hour loading dose (145 to 260 mg) followed by an 11-hour constant rate (30 to 70 mg/h) infusion.
Results Transient and reversible adverse effects, including nystagmus, nausea, vomiting, somnolence, hallucinations, and agitation, commonly occurred in dextrorphan-treated patients. Loading-dose escalation was stopped because of rapid-onset, reversible, symptomatic hypotension in 7 of 21 patients treated with doses of 200 to 260 mg/h. At the highest rates of maintenance infusion (>90 mg/h), 3 patients developed deep stupor or apnea. The maximum tolerated loading dose was 180 mg/h, and the maximum tolerated maintenance infusion was 70 mg/h. Maximum plasma levels of 750 to 1000 ng/mL were obtained in 9 patients. There was no difference in neurological outcome at 48 hours between the dextrorphan-treated and placebo-treated patients.
Conclusions The highest doses of dextrorphan administered were associated with serious adverse experiences in some patients. Lower doses (loading doses of 145 to 180 mg, maintenance infusions of 50 to 70 mg/h) were better tolerated and rapidly produced potentially neuroprotective plasma concentrations of dextrorphan. These doses were associated with well-defined pharmacological effects compatible with N-methyl-d-aspartate receptor antagonism.
Experimental studies indicate that extracellular levels of endogenous excitatory amino acid (EAA) neurotransmitters such as glutamate and aspartate are elevated in acute cerebral ischemia and cause excessive activation of neuronal EAA receptors.1 One subclass of EAA receptors, the N-methyl-d-aspartate (NMDA) receptor, has been implicated as a key mediator of ischemic neuronal injury. The NMDA receptor activates an ion channel that has a high sodium and calcium conductance. Excessive opening of this channel can lead to rapid neuronal swelling, excessive intracellular calcium accumulation, and eventually neuronal death.2
Considerable experimental evidence indicates that pharmacological antagonism of the NMDA receptor can substantially attenuate hypoxic neuronal injury in cell culture and reduce infarct size in animal models of focal ischemia.3 4 Because of these encouraging experimental results, several NMDA antagonists are undergoing clinical evaluation in human stroke patients. The evaluation of these agents is proceeding cautiously, however, because of the potential for adverse pharmacological effects such as central nervous system depression, depressed ventilation,5 hypotension,6 psychotomimetic properties,7 and morphological neuronal changes8 with this class of compounds.
Dextrorphan, the O-demethylated metabolite of the commonly used antitussive dextromethorphan, is an NMDA antagonist that has been shown to attenuate hypoxic neuronal injury in culture9 10 and to significantly reduce ischemic neuronal injury in animal stroke models.11 12 13 14 The neuroprotective effect of dextrorphan is dependent on plasma and brain levels, with greater protection obtained at higher concentrations.13 To assess the potential for clinical use of dextrorphan in patients with acute ischemic stroke, we conducted a safety and tolerability study to define the safety profile of dextrorphan, determine the maximum safe and tolerated loading dose and maintenance infusion, and evaluate pharmacokinetic parameters in patients with acute stroke.
Subjects and Methods
This trial was a seven-center, two-part dose escalation study. Patients between the ages of 18 and 80 years who presented with an acute supratentorial ischemic stroke were eligible for enrollment. They were required to be alert and have a stable neurological deficit. Patients were excluded if their symptoms were referable to the vertebrobasilar arterial system or if they could not be enrolled within 48 hours after the onset of ischemic symptoms. Additional exclusion criteria included a significant psychiatric disorder (eg, psychosis or history of suicide attempt), unstable cardiovascular condition (eg, recent myocardial infarction or congestive heart failure), alcohol or drug dependence, febrile illness within 1 week before study entry, blood transfusion within 2 months before study entry, or treatment with an investigational agent within 1 month before study entry. Patients taking specific concomitant medications, including digoxin, warfarin, certain antiarrhythmic drugs, and dextromethorphan, were also excluded. No minimum neurological deficit was required.
Before study drug administration, all patients underwent a complete physical and neurological examination including the National Institutes of Health (NIH) Stroke Scale,15 computed tomographic scan of the head, 12-lead electrocardiogram, and screening laboratory tests (chemistry panel, complete blood count, platelet count, and coagulation profile). Cardiac telemetry (2 to 4 hours) was also performed before study drug infusion. Respiratory rate was closely monitored during drug administration. Dextrorphan loading doses of 60 to 260 mg were administered intravenously over 1 hour.
Two different maintenance-infusion regimens were evaluated. Initially, a 23-hour escalating-dose maintenance infusion (15 to 135 mg/h) was used. After the 1-hour loading dose infusion, patients received a maintenance infusion divided into a 5-hour infusion (hourly rate, 25% of the loading dose) followed by three 6-hour infusions at hourly rates of 50% of the loading dose, then 75% of the loading dose, and finally 100% of the loading dose. This phase of the study was placebo controlled and double blinded. In the second phase of the study, which was open-label, an 11-hour constant-rate infusion of dextrorphan (30 to 70 mg/h) was given after the 1-hour loading dose (145 to 260 mg).
Throughout the infusion of study drug, patients remained on bed rest; other medical treatments, including antiplatelet agents, anticoagulants, and supplemental oxygen, were allowed if indicated. Neurological examinations, including the NIH Stroke Scale, were performed at baseline, 12 hours, 24 hours, and 48 hours after the start of the infusion. Blood and urine samples were obtained for determination of dextrorphan plasma levels as well as for routine laboratory studies at specified time points during and after the infusion. Patients were seen for a follow-up visit at day 30 to assess adverse events. All adverse events, whether related to study drug infusion or not, were recorded by the investigators during the study period.
Sixty-seven patients were enrolled in the study (51 received dextrorphan and 16 received placebo). The baseline characteristics of the study population, grouped by loading dose, are summarized in Table 1⇓. An equal number of men and women were enrolled. The mean time from symptom onset to treatment was slightly longer in the placebo group (37 hours compared with 31 hours in the dextrorphan-treated patients). Patients in both the placebo- and dextrorphan-treated groups had comparable demographic characteristics. Most patients had relatively mild stroke symptomatology with a median NIH Stroke Scale score of 5 in the placebo group and 6 to 8 in the dextrorphan groups.
Patients treated with dextrorphan experienced a wide variety of pharmacological effects that differed from those commonly reported by placebo-treated patients (Table 2⇓). At least one adverse event was reported for 100% of dextrorphan-treated patients compared with 81% of the patients who received placebo. Nystagmus, somnolence, agitation, hallucinations, and confusion occurred in more than half of the dextrorphan-treated patients. Other common effects included hypertension, nausea, dizziness, stupor, and vomiting. Most of the adverse effects occurred in a relatively predictable manner, with certain reactions occurring more frequently during the loading dose and others during the maintenance infusion.
Pharmacological effects of dextrorphan that occurred commonly during the loading dose are shown in Fig 1⇓. Patients who were receiving dextrorphan typically developed nystagmus during the first 15 to 20 minutes of the infusion. Subsequently, approximately one third developed nausea or vomiting that usually responded rapidly to small doses of parenteral prochlorperazine (5 to 15 mg) or droperidol (0.6 to 1.2 mg). Vomiting occurred approximately twice as frequently in patients who received higher loading doses (200 to 260 mg) compared with lower doses (145 to 180 mg) (Fig 2⇓). Patients usually became drowsy toward the end of the loading dose infusion; about half developed somnolence irrespective of the loading dose administered (Fig 2⇓).
At loading doses of 200 mg/h or more, rapid, reversible, symptomatic (eg, pallor, diaphoresis, nausea) drops in blood pressure occurred in 7 of 21 patients. Six of these patients had systolic blood pressure drops of 50 mm Hg or more. These hypotensive events always occurred within 90 minutes of the start of the loading dose and responded rapidly to treatment with intravenous fluids and placement of the patient in a supine position. In addition, brief infusions of low-dose dopamine were administered to 2 patients. The single most important factor in the production of hypotension appeared to be the rate at which the loading dose was administered. No other predisposing factors (such as concomitant medications, lean body mass, or dextrorphan plasma concentrations) could be determined. Because of the occurrence of these hypotensive episodes, loading-dose escalation was stopped at 260 mg/h. None of the patients who experienced hypotension had any adverse neurological sequelae.
Adverse effects did not limit escalation of the maintenance dose. Common adverse effects associated with the maintenance infusion are shown in Fig 3⇓. Patients typically became confused and frequently had vivid auditory and visual hallucinations, often of religious or death-related content. Agitation was also seen frequently but was usually easily controlled with verbal reassurance; agitated patients frequently became hypertensive and required treatment with antihypertensive medications. Patients treated with higher maintenance infusions (71 to 135 mg/h) experienced hallucinations and hypertension more than twice as frequently as those who received lower maintenance doses (50 to 70 mg/h) (Fig 4⇓). Confusion and agitation occurred frequently even in the patients who received lower maintenance doses (Fig 4⇓). In 3 patients who were receiving high maintenance doses (>90 mg/h), the infusion was discontinued because 2 of the patients became stuporous to the point of unresponsiveness and 1 patient developed apnea that required intubation. The latter patient was subsequently found to have an occult carcinoma metastatic to the lung. All adverse effects resolved rapidly (usually over 8 to 12 hours) after discontinuation of the drug infusion. The maximum total dose of dextrorphan administered was 3310 mg over 24 hours.
A comparison between the baseline NIH scores and the 48-hour scores is shown in Table 3⇓. The mean change in these scores was virtually identical for both dextrorphan- and placebo-treated patients; in general, most patients showed approximately a 1-point improvement over 48 hours. The mean change in NIH scores was no different in patients who received low, moderate, or high doses of dextrorphan (Table 3⇓). In addition, the 48-hour outcome (as measured by NIH scores) was no different in patients who had symptomatic transient hypotension during the loading infusion.
There was no evidence of systemic toxicity, and means of routine laboratory tests (complete blood count, chemistry panel, coagulation studies) were within the normal ranges at 24 and 48 hours after infusion. In addition, no significant electrocardiographic changes were noted on routine monitoring. There was one death 30 days after drug infusion (the patient with the occult carcinoma) due to respiratory failure secondary to the carcinoma.
Pharmacokinetic studies indicated that intravenous dextrorphan has a mean half-life of 1.7 to 5.4 hours, a large volume of distribution (approximately 300 to 650 L), and a high clearance (approximately 70 to 220 L/h). Plasma concentrations obtained at the end of the loading dose are summarized in Fig 5⇓. Serum levels obtained during and after discontinuation of the infusion are shown in Fig 6⇓ for two groups of patients. Nine patients achieved maximum plasma concentrations of >750 ng/mL. There was no difference between the levels obtained in men and women, and there was no significant correlation between plasma concentration and patient weight.
The results of this study indicate that intravenous dextrorphan can be administered to acute stroke patients to rapidly achieve and maintain plasma concentrations between 400 and 1000 ng/mL throughout a 12- to 24-hour infusion. Because experimental ischemia models suggest that the neuroprotective effects of dextrorphan are dose dependent and greatest at higher plasma concentrations, the goal of this study was to achieve the highest tolerated plasma concentrations by escalating the dextrorphan dose until significant adverse events were encountered. The highest doses administered were associated with a large number of adverse experiences including hypotension, stupor, and hypoventilation. These serious side effects could potentially have a deleterious effect on the clinical outcome of acute stroke patients.
Loading-dose escalation was limited by abrupt symptomatic drops in blood pressure that occurred at doses of 200 mg/h or greater, but clinically significant hypotension was not seen at lower loading doses. The rate of loading-dose administration appeared to be the determining factor in the production of hypotension. No consistent effect of weight, age, previous or concomitant medication, or coexisting medical illness was found. Other potentially serious adverse effects, such as unresponsiveness and hypoventilation, occurred in a few patients who were treated with very high-dosage maintenance infusions. These adverse events suggest that significant antagonism of excitatory neurotransmission was obtained at these doses.16
Dextrorphan loading doses of 145 to 180 mg and maintenance infusions of 50 to 70 mg/h had a better safety profile than higher doses. Patients treated with these doses did experience a wide variety of adverse effects, including a high incidence of somnolence, agitation, and confusion. In general, these experiences were rated as mild to moderate in severity by the investigators. However, these effects can interfere with monitoring of the neurological examination and could contribute to other complications such as aspiration pneumonia. Most of the patients treated in the lower dose range maintained plasma levels between 400 and 600 ng/mL; these levels are comparable to those required to obtain neuroprotective effects in both cell culture10 and animal stroke models.13
The goal of this study was not to evaluate the efficacy of dextrorphan, and patients probably were entered into the trial too long after stroke onset to demonstrate any significant neuroprotective effects. Dextrorphan-treated patients did not appear to have any difference in neurological outcome at 48 hours. Patients who experienced brief hypotensive episodes did not suffer acute clinical deterioration; however, an effect on the long-term outcome of these patients could not be assessed by this study. In patients who did experience significant adverse effects, the effects resolved rapidly after discontinuation of the infusion in accordance with the short half-life of the drug. In addition, at the doses evaluated in this study, dextrorphan did not exhibit any cardiac or other systemic toxicity.
Based on the results of this study, it appears that further evaluation of dextrorphan for treatment of acute stroke should be pursued using doses that were not associated with serious adverse events in this study. In addition, the drug should be administered in a special-care unit with intensive and continuous patient monitoring to minimize the risk of serious adverse events.
Dextrorphan Study Group Participating Centers
Stanford Stroke Center, Palo Alto, Calif: Gregory W. Albers, MD, Principal Investigator; Nanette H. Hock, RN, Study Coordinator; Joni M. Clark, MD, Mark P. Goldberg, MD, David Tong, MD, Gary K. Steinberg, MD, PhD, Midori Yenari, MD, Coinvestigators.
Mercy General Hospital, Sacramento, Calif: Richard P. Atkinson, MD, Principal Investigator; Ruth Cummings, MS, RN, Deidre A. Wentworth, MSN, RN, Study Coordinators.
University of Miami (Fla) School of Medicine: Roger E. Kelley, MD, Principal Investigator; Vilma Alfonso, Clinical Coordinator; Yolanda Reyes-Iglesias, MD, Shuichi Suzuki, MD, Collaborators.
Montefiore Medical Center, Bronx, NY: Daniel Rosenbaum, MD, Principal Investigator; Emelia Klonowski, Study Coordinator; Paul Katz, MD, Coinvestigator.
University of California at San Diego Medical Center, San Diego, Calif: John Rothrock, MD, Principal Investigator; Traci Babcock, RN, Study Coordinator; Patrick Lyden, MD, Justin Zivin, MD, Coinvestigators.
Medical College of Georgia, Augusta: Robert Adams, MD, Principal Investigator; Elizabeth Carl, RN, Carol Buki, Study Coordinators; Fenwick T. Nichols, MD, David C. Hess, MD, Mark Rovick, MD, Coinvestigators.
Cooper Hospital–University Medical Center, Camden, NJ: Thomas Mirsen, MD, Principal Investigator; Carla Bruegel, RN, MSN, Study Coordinator.
Hoffmann-La Roche group: K.L. Paul, MD; L. Lesko, MD, PhD; J. Rae, MS; V. Pitman, PharmD; M. Modi, PhD; K. Yoo, PhD; L. Lehr, AAS, Nutley, NJ; G. Magni, MD, Basel, Switzerland.
This study was funded by Hoffmann-La Roche. The authors thank Phyllis Grant for preparation of the manuscript.
Reprint requests to Dr Gregory W. Albers, Stanford Stroke Center, 701 Welch Rd, Suite 325, Palo Alto, CA 94304.
- Received September 6, 1994.
- Revision received October 31, 1994.
- Accepted November 3, 1994.
- Copyright © 1995 by American Heart Association
Choi DW, Koh J, Peters S. Pharmacology of glutamate neurotoxicity in cortical cell culture: attenuation by NMDA antagonists. J Neurosci. 1988;8:185-196.
Olney JW, Labruyere J, Price MT. Pathological changes induced in cerebrocortical neurons by phencyclidine and related drugs. Science. 1989;244:1360-1362.
Choi DW, Peters S, Viseskul V. Dextrorphan and levorphanol selectively block N-methyl-d-aspartate receptor-mediated neurotoxicity on cortical neurons. J Pharmacol Exp Ther. 1987;242:713-720.
Brott T, Adams HP, Olinger CP, Marler JR, Barsan WG, Biller J, Spilker J, Holleran R, Eberle R, Hertzberg V, Rorick M, Moomaw CJ, Walker M. Measurements of acute cerebral infarction: a clinical examination scale. Stroke. 1989;20:864-870.