|Home | About | Journals | Submit | Contact Us | Français|
The purpose of this study was to determine whether carbidopa (Lodosyn), an inhibitor of dopa-decarboxylase that blocks the synthesis of dopamine outside the brain, is an effective antiemetic in patients with familial dysautonomia (FD) and hyperdopaminergic nausea/retching/vomiting attacks.
We enrolled 12 patients with FD in an open-label titration and treatment study to assess the safety of carbidopa. We then conducted a randomized, double-blind, placebo-controlled, crossover study to evaluate its antiemetic efficacy.
Previous fundoplication surgery in each patient studied prevented vomiting, but all of the subjects experienced severe cyclical nausea and uncontrollable retching that was refractory to standard treatments. Carbidopa at an average daily dose of 480 mg (range 325–600 mg/day) was well tolerated. In the double-blind phase, patients experienced significantly less nausea and retching while on carbidopa than on placebo (p < 0.03 and p < 0.02, respectively). Twenty-four-hour urinary dopamine excretion was significantly lower while on carbidopa (147 ± 32 µg/gCr) than while on placebo (222 ± 41µg/gCr, p < 0.05).
Carbidopa is a safe and effective antiemetic in patients with FD, likely by reducing the formation of dopamine outside the brain.
This study provides Class II evidence that carbidopa is effective in reducing nausea/retching/vomiting in patients with FD.
Recurrent bouts of nausea, retching, and vomiting accompanied by tachycardia and hypertension following emotional or physiologic stresses are a common disabling feature of familial dysautonomia (FD) (Riley-Day syndrome or hereditary sensory and autonomic neuropathy type III).1,2 Standard antiemetics have little effectiveness during these episodes. Traditionally, patients with FD were advised to manage their episodes with benzodiazepines.3 Many quickly become tolerant and require increasing doses, which have been associated with respiratory depression. The central sympatholytic agent, clonidine, has been used with some success, but its use is limited by the unwanted side effects of sedation and hypotension.3 Metoclopramide has also been used, but it is frequently associated with tardive dyskinesia, and is therefore not suitable as a long-term therapy.4 Newer agents like pregabalin are only moderately effective.5
Because of a developmental abnormality in the afferent neurons of the baroreflex, patients with FD are unable to restrain sympathetic outflow.6 Vomiting/retching attacks occur during sympathetic activation when dopamine spills over into the circulation.2 We hypothesize, therefore, that vomiting is due to activation of dopamine receptors in the chemoreceptor trigger zone, which lies outside the blood–brain barrier.7
The purpose of this study was to determine whether carbidopa (Lodosyn, Merck & Co., Inc., Whitehouse Station, NJ), an inhibitor of dopa-decarboxylase that blocks the synthesis of dopamine outside the brain,8 is an effective antiemetic in patients with FD. We present the results of a randomized, placebo-controlled trial conducted in patients with FD.
We enrolled 12 patients homozygous for the common FD intron-20 gene mutation9 who complained of severe nausea and retching and had failed multiple treatment attempts (inclusion criteria). All were being followed as outpatients at the Dysautonomia Center in New York University School of Medicine. Patients younger than 12 years or with severe renal disease (serum creatinine >2.0 mg/dL) or significant electrocardiographic abnormalities were ineligible. Patients taking dopamine blockers (metoclopramide, domperidone, risperidone), monoamine oxidase inhibitors, tricyclic antidepressants, or neuroleptics were also excluded.
The study was approved by the US Food and Drug Administration's Office of Rare Disorders and ethical approval was obtained from the Institutional Review Board of New York University. The trial was registered on ClinicalTrials.gov (NCT01212484). Written informed consent (or assent) was obtained from all participants. Recruitment ran from February 2010 to September 2011.
Participants first monitored their symptoms of nausea and retching in a daily diary on a validated patient-reported outcome scale for 1 month.10 Twenty-four-hour urine collections were assayed to measure daily dopamine excretion the day before scheduled study visits, at baseline, at the end of the open-label treatment phase, and at each crossover visit. Baseline laboratory values, EKG, and blood pressures were obtained at scheduled office visits (table 1).
Patients who reported severe, frequent (>3/week) attacks of nausea and retching in their diaries were enrolled in an open-label dose titration and active treatment phase. During the trial, the patients continued to take the same doses of standing medications (diazepam, clonidine, clonazepam, pregabalin) that they had taken during the previous 4 weeks. Doses of carbidopa were increased in a stepwise fashion over an average of 38 days. Doses began at 12 mg and were increased in 25 mg/day increments every other day. The duration of the titration phase depended on the dosage, but on average lasted 38 days. Adult patients were titrated up to 600 mg of carbidopa per day (divided into 3 200-mg doses every 6 hours by mouth or via gastrostomy tube, if preferred). Doses were adjusted accordingly in pediatric patients who weighed less than 50 kg using the Lack and Stuart-Taylor formula.11 Patients remained on their maximum dose of carbidopa for 4 weeks to evaluate safety and tolerability.
To evaluate the efficacy of carbidopa, we used a randomized, double-blind, placebo-controlled crossover design in which each patient served as his or her own control. The order in which carbidopa or placebo was administered was determined randomly (computer generated) and none of the treating physicians, patients, or caregivers knew the identity of the drugs given. To maintain allocation concealment, we prepared identical orange capsules containing either carbidopa powder (active agent) or an identical-looking lactose powder (placebo). Patients were instructed to take equal numbers of either placebo or carbidopa capsules every 6 hours. The 2 crossover periods ran consecutively and each lasted 4 weeks.
Table 1 provides a schedule of the main assessments during the trial. Safety measures included weight and temperature, supine/seated/standing blood pressures, complete blood counts, metabolic panels, 12-lead EKGs, and full neurologic examinations (table 1). Side effects were monitored throughout.
Our primary research question was to determine whether carbidopa (Lodosyn) is an effective treatment for nausea/retching/vomiting attacks in patients with FD. To assess efficacy, patients were given a printed diary and asked to grade their symptoms each night for 1 month. On each diary page, nausea and retching were defined in detail and assessed individually. Daily scores were reported on a modified version of the Rhodes Index of Nausea, Vomiting and Retching,12 which included all 5 items relating to nausea and retching. Items addressing vomiting/throwing up were omitted, as all participants had antireflux surgery that prevented vomiting (Nissen fundoplication). Retching distress, nausea distress, number of nausea episodes per day, number of retching episodes per day, and the amount of time spent feeling nauseous were graded on a 5-point scale. Patients were also asked to note whether they took additional drugs to control their nausea that day.
Patients were asked to make global clinical assessment of symptom severity, using a 7-point scale, ranging from 1 (no nausea) to 7 (extremely ill with nausea), at scheduled office visits. Patients were also asked to rate their global impressions of improvement on a 7-point scale. Scores ranged from 1 (very much improved) to 7 (very much worse) and were made in relation to baseline (screening prior to dose titration). Twenty-four-hour urine was collected in standardized darkened collection bottles containing hydrochloric acid as a preservative. Samples were refrigerated before being assayed using quantitative high-performance liquid chromatography to assess free dopamine and norepinephrine levels (table 1).
An average daily score was calculated from the nausea diaries at baseline, during open-label active treatment, and in both crossover phases. Analysis was carried out on the individual item scores and the composite score generated from the sum of all 5 domains.12 Datasets were first tested for normality using the Shapiro-Wilk and Kolmogorov-Smirnov tests. Depending on the outcome, mean unadjusted scores as well as change from baseline were compared during placebo and carbidopa using paired t tests, 2-way analysis of variance, or Wilcoxon signed-rank tests. Parametric and nonparametric evaluations were used as appropriate. Data were analyzed using Prism version 5.0 (GraphPad Software, Inc., 2011). All p values were 2-tailed and p < 0.05 was considered statistically significant.
Twelve patients entered the trial (figure e-1 on the Neurology® Web site at www.neurology.org). The average age of the patients was 18 years (range 12–36 years). Eight were male. All had histories of multiple hospitalizations for uncontrollable retching attacks. Despite chronic treatment with benzodiazepines (all 12) and clonidine (11 out of 12), all enrolled patients experienced daily episodes of nausea/retching (table 2). Seven of the 12 had failed a trial of pregabalin. On average, patients had 3 attacks per day, rated them as moderately distressing, and spent 1.7 ± 1.5 hours each day feeling nauseated. Average dopamine excretion in urine was 271 µg/gCr (range 782–552 µg/gCr).
All 12 patients entered the dose-escalation phase. Despite symptomatic improvement, one patient withdrew because of constipation. His medical records revealed similar past episodes.
The remaining 11 patients were titrated to their maximum dose (on average 480 mg, range 325–600 mg/day). There were no changes in electrocardiographic parameters and there were no signs of renal or hepatotoxicity. One patient was hospitalized for an unrelated respiratory complication. He opted to continue in the trial.
During the 4 weeks of open-label treatment, all but 2 patients showed an improvement (composite nausea score; baseline 14 ± 5 vs active treatment 9 ± 2 units, p < 0.01). On average, there was a 50% improvement in overall symptom scores. Two patients were asymptomatic. Urinary dopamine output decreased significantly compared to baseline (figure 1). Urinary dopamine levels were positively correlated with average daily symptom scores (figure 1).
Of the 11 patients who completed the open-label treatment phase, one withdrew because his nausea had improved and he was unwilling to take placebo. A second patient was withdrawn in the first arm of the randomization because of severe anemia, a common complication in FD. Drug accountability logs showed that he was on active carbidopa, but did not comply with the dosing regimen. A third patient withdrew at the end of the first randomization. He thought, as his symptoms were improved, he was taking active agent and did not want to be randomized to placebo. He was correct.
Eight patients completed both arms of the double-blind, randomized, placebo-controlled phase. Three were randomized first to carbidopa and 5 were randomized first to placebo. There were no significant baseline differences in clinical characteristics, nausea scores, or dopamine levels between the patients who received carbidopa first and the patients who received placebo first (Class II level of evidence).
There was less dopamine in the urine of patients when they were taking carbidopa than when they were taking placebo (p < 0.05, table 3). Figure 2 shows change from baseline in nausea/retching subscale and composite scores during the randomized period. The composite and subscale scores fell more on carbidopa than on placebo, indicating a greater improvement in symptoms while on active study drug (table 3 and figure 2). When randomized to carbidopa, patients rated their nausea and retching as less distressing (p < 0.01 and p < 0.05, respectively), they reported fewer episodes of nausea (p < 0.02) and retching (p < 0.02), and they spent less time feeling unwell (p < 0.05). On the clinical global impressions scale, patients rated their nausea as less severe (p < 0.02) on carbidopa compared with placebo. Compared to baseline, they reported no change while on placebo, while feeling much improved on carbidopa (p < 0.002, see table 3 for details). According to the daily nausea diaries, patients required additional doses of benzodiazepines and clonidine less often when randomized to carbidopa (p < 0.01, table 3). Interestingly, 2 patients reported missing days of school while taking placebo; none missed school while randomized to carbidopa.
Supine and standing blood pressures were similar at baseline and the end of active open-label treatment. During the double-blind phase, standing systolic blood pressure was lower on carbidopa compared with placebo (p < 0.02, table 3). Diastolic and mean pressures were similar. Urinary norepinephrine levels tended to be lower on carbidopa than at baseline (table 3).
Generalized weakness, a common complaint in patients with FD, was reported as an adverse event by one patient while randomized to carbidopa and while randomized to placebo.
This is the first double-blind, placebo-controlled clinical trial of a specific symptomatic therapy in patients with FD. The results showing that blocking dopamine synthesis outside the brain significantly reduces the severity and frequency of the nausea/retching crises strongly support the hypothesis that excessive dopamine is the likely cause of retching and nausea in these patients. Except in patients with Parkinson disease who are taking levodopa, carbidopa has never before been used as an antiemetic.
Carbidopa was well-tolerated in our current trial. All adverse events (anemia, constipation, weakness, respiratory complications) occur frequently in patients with FD and a relationship with carbidopa was unlikely.
Carbidopa was first successfully used to prevent the side effect of nausea and vomiting in patients with Parkinson disease treated with levodopa in 1969.13 It remains an adjunct therapy to levodopa. This is the first time it has been used in patients with FD.
Previous trials to assess the antiemetic efficacy of pregabalin,5 tocotrienolols,14 benzodiazepines,15 and bethanechol16 in patients with FD were all uncontrolled and without validated endpoints. Unlike clonidine and benzodiazepines, agents traditionally used to manage the hyperdopaminergic vomiting attacks in FD, treatment with carbidopa was not associated with sedative effects. Improving symptoms without depressing respiration has obvious benefits for this population, in whom respiratory drive from chemoreceptors is severely impaired.17
The present study has several limitations. The best dose could not be determined from the trial design. While we saw no evidence of central effects, it is possible that at high doses, carbidopa might cross the blood–brain barrier, and we do not know whether lower doses than those studied are as effective in FD. Free dopamine in urine does not take into account the amount that is conjugated or metabolized and therefore likely underestimates its synthesis. Nevertheless, it is believed to be clinically useful.18 In this study, samples were collected in standardized bottles with preservative, protected from ultraviolet light, and refrigerated to help stabilize the dopamine compound, and we did see a significant reduction in dopamine excretion with carbidopa. We could not measure the concentration of dopamine in the area postrema, a known vomiting center,7 but we did find a significant correlation between free dopamine in the urine and symptom severity (figure 1). Finally, because FD is a very rare disorder, our study includes a small number of patients.
As carbidopa reduces the bioavailability of dopamine,8 it is conceivable that it could also inhibit the synthesis of norepinephrine, which is synthesized from dopamine.19 Carbidopa's effects on blood pressure warrant further investigation. It could potentially be useful as an antihypertensive to lessen surges in sympathetic outflow and may mitigate hypertensive damage to the end organs.20
Although the severity and duration of the nausea and retching were improved, some patients continued to experience (milder) symptoms while taking carbidopa. In those patients, despite high doses of carbidopa, plasma dopamine levels remained within the normal range. Perhaps this is due to phenotypic variability in the expression of the enzymes involved in catecholamine synthesis and metabolism in FD.21,22 Combining carbidopa with domperidone, a dopamine receptor blocker, might prove useful in patients who continue to be symptomatic. Alternative treatment approaches with cognitive behavioral therapy or psychotherapy have been useful in some cases.23
The authors thank the research nurses of the Dysautonomia Center for their help with this study.
Supplemental data at www.neurology.org
Dr. Norcliffe-Kaufmann contributed to the study concept and design, analysis and interpretation, and critical revision of the manuscript for important intellectual content. Mr. Jose Martinez contributed to the study procedures and assigned patients to the interventions. Dr. Axelrod contributed to the acquisition of data. Dr. Kaufmann was involved in the study concept and design, analysis and interpretation, critical revision of the manuscript for important intellectual content, and was the study supervisor.
Supported by a grant from the Food and Drug Administration's Office of Rare Disorders (R01- FD003731-02). Bristol-Myers-Squibb donated carbidopa.
L. Norcliffe-Kaufmann receives research support from the NIH (R01HL103988 & U54NS065736-01). J. Martinez receives research support from the NIH (U54NS065736-01) and the Dysautonomia Foundation, Inc. F. Axelrod receives support from the Dysautonomia Foundation, Inc. H. Kaufmann receives research support from the NIH (R01HL103988, U01 NS078025-01, U54NS065736-01) and the Dysautonomia Foundation, Inc. Go to Neurology.org for full disclosures.