Steven Lamm, M.D., DTPH, gave the following expert report in a Reglan lawsuit. The report establishes that some people who used Reglan or Metoclopramide for over 12 weeks developed movement disorders such as Tardive Dyskinesia.
"The literature on tardive dyskinesia and Metoclopramide begins with case reports and reports of case series. These reports demonstrate an association between metoclopramide and tardive dyskinesia but do not provide a quantitative estimate of risk.
Medical records that report drug use and pharmacy records have been used to describe the population characteristics of metoclopramide users, but without reports of tardive dyskinesia cases, they are unable to provide risk estimates.
The first data that do yield an initial estimate of risk come from national adverse drug reaction registries in which reported cases of tardive dyskinesia in metoclopramide patients have been compared with drug sales data.
Cross-sectional studies report the prevalence of signs of tardive dyskinesia among metoclopramide users, but without a time factor that would allow an assessment of the frequency of the development of tardive dyskinesia, they do not provide a basis for calculating the risk.
Clinical studies in which patients placed onto metoclopramide have been followed over time during which the development of tardive dyskinesia can be observed would provide a basis for calculating the risk of acquiring tardive dyskinesia. These observations are generally found within data submitted to the drug regulatory agencies and not specifically in the published literature."
This report is a little outdated, as newer evidence has found a much clearer link between movement disorders and Reglan.
Re: Tardive Dyskinesia and Metoclopramide
Expert Qualification:
I am Steven H. Lamm, MD, DTPH, a physician-epidemiologist in practice in Washington, DC. I received my medical degree (MD) from the University of Southern California, School of Medicine in 1968 and my public health degree (DTPH) from the University of London, School of Hygiene and Tropical Medicine in 1974. I am Board-certified in Pediatrics, Preventive Medicine, and Occupational Medicine and am a charter-member of the American College of Epidemiology.
I have held positions at the Centers for Disease Control and Prevention (Atlanta, GA) and the National Institutes of Health (National Center for Child Health and Human Development; Bethesda, MD). I have been a consultant to a variety of governmental organizations. I hold, or have held, positions on the faculties of Johns Hopkins University-Bloomberg School of Public Health (Department of Health Policy and Management), Georgetown University School of Medicine (Department of Pediatrics), George Washington University School of Public Health and Health Services, and the Uniformed Services University of the Health Sciences, F. Edward Hebert School of Medicine (Department of Preventive Medicine and Biometrics). I am a Visiting Professor at the Institute for Environmental Engineering at National Cheng Kung University (Tainan, Taiwan).
My research work has concentrated on assessing the risk to human health from exposures to chemicals in the workplace, ambient environment and pharmaceutically. I have over 120 contributions to the scientific literature. My curriculum vitae contains a list of my publications and is attached. I have been active in numerous professional organizations (American College of Occupational and Environmental Medicine, American Thyroid Association, Organization of Teratology Information Specialists, American Industrial Hygiene Association, Society for Epidemiological Research, and American Public Health Association). I have been a peer reviewer for nearly 20 scientific and medical journals.
I am President of Consultants in Epidemiology and Occupational Health, LLC (CEOH) and have been in the private practice of medical epidemiology for over 25 years. My rate for consulting is $450/hour and a list of my testimonies over the past four years is attached.
Report Task:
I have been asked to review the case records, and relevant literature, pertaining to Mr. Paul Burnett, who developed tardive dyskinesia while taking metoclopramide. I have reviewed the medical records of Paul Burnett and have attempted to summarize his clinical history. I have presented the epidemiological approach to examining literature related to a drug and an adverse health effect. I have then reviewed in an historical or chronological context the literature on metoclopramide as an oral pharmaceutical agent and tardive dyskinesia.
Case Background:
Mr. Paul Burnett (Date of Birth XX/XX/1942) is a patient with long-standing cardiac and gastrointestinal disease. He had a heart attack in February 1999 which led to the diagnosis of coronary artery disease and the insertion of a cardiac stent by Dr. John Bret (Interventional Cardiologist). He has been followed by Dr. Bret and has subsequently had additional cardiac stents inserted.
He had been seen for his gastrointestinal disease by Dr. Andrew R. Gottesman (Gastroenterologist) since February 1997 and has been diagnosed as having colonic polyps and reflux esophagitis (treated with Zantac, Alka-selzer). Dr. David W. Bragg (Family Medicine) has been his primary physician since March 1999.
Mr. Burnett was seen by Dr. Bragg in March 1999 for upset stomach (belching, stomach rumbling, some nausea). Dr. Bret had placed him on Prilosec. His chest pain had been on and off for 14 hours. He had no shortness of breath. His hands were tingling on and off. He had run out of Xanax three days earlier. His weight was 230 pounds and his blood pressure 148/88. Dr. Bragg started him on Propulsid BID (twice a day) (March 16, 1999) which Dr. Gottesman later (May 15, 1999) said could be increased to QID (four times a day).
Over the next years he was seen by Dr. Bragg on almost a monthly basis and treated for a wide range of medical conditions with medications for his specific conditions. He has had a sequence of medications to treat his symptoms of anxiety and depression. Medications were switched or modified when signs or symptoms suggested problems with safety or efficacy. When issues had been raised at the FDA about the cardiac safety of Propulsid, Dr. Bragg switched his treatment to Reglan (February 1, 2000). His Reglan prescription dosage was increased from BID to QID on March 27, 2001 as he reported continuing gastrointestinal symptoms (belching and burping).
On August 25, 2003, he reported that his reflux was doing great, that he had been taking Reglan BID, and that he was gritting his teeth more. Dr. Bragg considered that he probably had central nervous system side effects from his Reglan and discontinued his Reglan medication. His symptoms continued and in December 2004 he was referred for a neurological consultation to Dr. P. Terrance Moore (Neurology and Sleep Medicine). Dr. Moore diagnosed him (October 18, 2005) as having (1) Tardive dyskinesia following Reglan usage, (2) Obstructive sleep apnea syndrome, and (3) Restless leg syndrome. His Tardive dyskinesia had been treated with Zyprexa and Lexapro. His sleep apnea had been treated with Ambien and Ativan. His restless leg syndrome had been treated with Klonopin. He was later (October 19, 2006) seen by Dr. Richard B. Dewey at University of Texas Southwestern Medical Center at Dallas where his Tardive dyskinesia was treated with Botox injections (from which he developed an allergic reaction). He was found to have blepharospasm and orofacialdyskinesia. He continues to be followed for his multiple medical problems by Dr. Bragg, including Tardive dyskinesia.
I shall review the literature on tardive dyskinesia in patients taking metoclopramide for gastrointestinal conditions.
Epidemiologic Approach:
Given the hypothesis that a particular disease condition is caused by a particular exposure, customarily a literature develops to assess the strength of evidence supporting, clarifying, or rejecting that hypothesis. Typically, the literature begins with specific case reports of individuals who have the disease condition, the exposure, and a temporal relationship in which the exposure preceded the disease. This is then followed by reports of case series, in which authors report a number of cases seen or collected by a specific individual or found in the records of a particular clinic. These types of information are informative in raising a causal hypothesis or hypothesis of association but they are not sufficiently strong to provide proof of the hypothesis.
Epidemiological studies are necessary to demonstrate either that persons with the disease are significantly more likely to have had the exposure than are persons without the disease (case-control studies) or that persons with the exposure are significantly more likely to subsequently develop the disease than are similar persons without the exposure (cohort studies).
Three critical elements in an epidemiologic study are the definition of exposure, the definition of disease, and the definition of the study population. The design and analyses of these studies may clarify (1) the specific conditions of exposure, (2) the specificity of diagnosis, and (3) the definition of the study population related to an association. A significant factor in these studies is the need to demonstrate that the exposure occurred during a reasonable time interval prior to the onset of the disease.
Structurally, i.e., with respect to study design, case-control studies and cohort studies are able to provide an estimate of risk. Case-control studies can demonstrate (odds ratio) whether the cases are significantly more likely to previously have had the exposure than are the non-cases (controls). Likewise, cohort studies demonstrate (risk ratio or relative risk) whether those with exposure are significantly more likely to subsequently acquire the disease than are those without the exposure.
Cross-sectional (prevalence) studies are not able to provide estimates of the risk of acquiring disease (e.g., dyskinesia) from exposure (e.g., metoclopramide) since they lack the critical factor of knowing that the exposure came before the disease.
I will review below the literature on tardive dyskinesia and metoclopramide. I have attempted to limit this to the literature on metoclopramide as an oral medication and to exclude the literature on metoclopramide as an intravenous medication. Most of this literature relates to metoclopramide as a 10 milligram tablet taken generally three or four times per day.
Literature Analysis:
Literature Review on Tardive Dyskinesia and Metoclopramide
The literature on tardive dyskinesia and Metoclopramide begins with case reports and reports of case series. These reports demonstrate an association between metoclopramide and tardive dyskinesia but do not provide a quantitative estimate of risk.
Medical records that report drug use and pharmacy records have been used to describe the population characteristics of metoclopramide users, but without reports of tardive dyskinesia cases, they are unable to provide risk estimates.
The first data that do yield an initial estimate of risk come from national adverse drug reaction registries in which reported cases of tardive dyskinesia in metoclopramide patients have been compared with drug sales data.
Cross-sectional studies report the prevalence of signs of tardive dyskinesia among metoclopramide users, but without a time factor that would allow an assessment of the frequency of the development of tardive dyskinesia, they do not provide a basis for calculating the risk.
Clinical studies in which patients placed onto metoclopramide have been followed over time during which the development of tardive dyskinesia can be observed would provide a basis for calculating the risk of acquiring tardive dyskinesia. These observations are generally found within data submitted to the drug regulatory agencies and not specifically in the published literature.
Thus, there is little literature upon which to quantify the risk.
1978-1985:
Case Reports and Case Series:
Tardive dyskinesia was first reported in association with Metoclopramide therapy in a case report from Israel by Lavy et al. published in the British Medical Journal in 1978. This 48 y/o male had been taking Metoclopramide at a daily dosage of 20-40 mg for six years and then increased gradually to 80 mg/day for four years. Ten days after he ran out of medications, he developed facial dyskinesia that was diagnosed as tardive dyskinesia. Re-administration of Metoclopramide at 30 mg/day resulted in the almost complete disappearance of the involuntary movements. The dyskinesia returned after another attempt at withdrawal and subsided when Metoclopramide was given again.
This case report was then followed by a report by Kataria et al. (1978) London, UK in the journal Lancet of three cases of parkinsonism with chronic tardive type dyskinesia in older (76, 84, and 66 y/o) women taking Metoclopramide for years, greater than 5, 1, and 2 years, respectively. Their symptoms generally resolved with or without treatment within 3 – 9 months after stopping Metoclopramide.
Grimes et al. reported first in the New England Journal of Medicine (1981) and then in the Canadian Medical Association Journal (1982), and the Lancet (1982) a series of tardive dyskinesia cases associated with Metoclopramide. Their first report was of four men and three women for a total of seven cases; their third report was of 12 cases. The mean age of these patients was 72, and their duration of metoclopramide use was 8 to 60 months (mean 26 months). Their patients had developed facial dyskinesia either during therapy or after Metoclopramide withdrawal. These cases, most of whom had parkinsonism, had been seen in the Parkinson's disease Clinic in Ottawa, Canada.
Haggstrom (1981) reported in the Swedish journal Lakartidningen two cases of metoclopramide-induced tardive dyskinesia. One was a 72 year old female with symptoms beginning after 1 3/4 years of metoclopramide treatment, which was discontinued nine months later. Symptoms became disabling six months after withdrawal of metoclopramide. The other was a 74 year old female whose symptoms began six months into two years of metoclopramide therapy. This was the first case report with symptoms beginning after less than one year of treatment. Symptoms became accentuated after the drug was withdrawn. These cases were included in the subsequent report from the Swedish Adverse Drag Reactions Advisory Committee report (Wiholm et al., 1984) [See below].
Indo and Matsuoka (1984) reported from Nagoya, Japan in the Japanese Journal of Medicine a case report of tardive dyskinesia in a 64 year-old male which developed after 260 days of metoclopramide treatment.
Thus, case reports of tardive dyskinesia in metoclopramide patients have been reported from Israel, from England, from Canada, from Sweden, and from Japan. It is a problem that has been recognized world-wide. Cases with onset within one year of exposure had now been reported from Canada, Sweden, and Japan.
Adverse Drug Effects Registries:
The publications of the cases in the British and Swedish medical literature led to the examination of data within their national adverse drug reaction registers on the frequency of reported cases of tardive dyskinesia among metoclopramide users.
The first study is that of Wiholm et al. (1984) in the British Medical Journal which reported from the Swedish Adverse Drug Reactions Advisory Committee eleven cases of tardive dyskinesia associated with Metoclopramide in the five year period 1977-1981, ten of which occurred during 1979-1981. All cases were in women with a mean age of 76 years (range 69-86) who had been treated for a median of 14 months (range 4-44) prior to onset. Four of the 11 cases began within 4-6 months of metoclopramide treatment. The metoclopramide sales during 1977-1981 was 11 million treatment days (prescribed daily doses). The case rate was thus about 1 per million treatment days or 1 per 2,800 treatment years. A total of 196,000 prescriptions for metoclopramide were written. With a total of 11 reported cases, that means one tardive dyskinesia case per 17,800 prescriptions. They calculated that 0.3% of the male and 0.8% of female population aged 70 years or more could have taken metoclopramide for at least six months. Further, they calculated that for elderly patients on treatment for at least six months, the incidence rate was greater than one case in 1,000 patients. Correction of a few technical errors in this calculation would yield an incidence rate of less than one in 1,000 patients. Two of the cases had onset prior to six months treatment, and the boundary conditions of the calculation were set only after the characteristics of the cases were known. Orme and Tallis (1984) provided a commentary on the Wiholm et al. (1984) report in a subsequent issue of the British Medical Journal.
The second study is that of Bateman et al. (1985) in the British Medical Journal which reported from the Adverse Reactions Register of the Committee on the Safety of Medicines (UK) four cases of Metoclopramide-associated tardive dyskinesia during 1967-1982, a period during which there were 15.9 million prescriptions written for Metoclopramide. This would be a rate of about one case per 4 million prescriptions or 0.25 reported cases per million metoclopramide prescriptions.
The demographic and treatment characteristics of the tardive dyskinesia cases in Bateman et al. (2004) were not described. The tardive dyskinesia reports were part of a broader study of reports of extrapyramidal reactions with metoclopramide. During the same time period, the committee found a total of 479 reports of extrapyramidal reactions for a rate of one case per 33,000 prescriptions. Most of these (455) were for acute dystonia-dyskinesia and among young adults. The highest risk (0.02%) was among 12-19 year old females. Tardive dyskinesia represented less than one percent of the reported extrapyramidal reactions; 95 % were acute dystonic type reactions.
Thus, the two studies above produce overall point estimates of the risk of tardive dyskinesia with metoclopramide use as 1 per about 20,000 prescriptions and 1 per about 4 million prescriptions. Clearly these two estimates are considerably different and a number of technical issues can be raised to explain that difference, but neither approaches the range of 1 per 500 prescriptions, the rate the label indicates for extrapyramidal symptoms, primarily acute dystonic reactions.
This is the status of the literature as of 1985.
1986-2006:
Case Reports and Case Series:
The first case report published from the USA was that of Lazzara et al. (1986) in General Hospital Psychiatry which reported from Emory University a case of an 83 year-old woman who developed tardive dyskinesia after being on metoclopramide for eight months. She had mild renal insufficiency and a daily dosage of 30 mg/day.
Beauclair and Fontaine (1986), in the Canadian Medical Association Journal, reported a case of a 74 year-old women who developed severe tardive dyskinesia two weeks after having raised her dosage from about one tablet every other day for two years to three to four tablets per day.
The second case report published from the USA was that of Pate] and Louis (1986) which reported in the New York State Journal of Medicine a case of an 84 y/o woman treated with metoclopramide for eight moths. She developed parkinsonism-type symptoms during treatment and dyskinesia-type symptoms two weeks after ceasing treatment.
The third case report published in the USA was that of Sarnie et al. (1987) which reported from Buffalo NY in the journal Movement Disorders the case of a 66 year-old man who developed parkinsonism after 10 months on 30-40 mg/day metoclopramide and tardive dyskinesia two months later. This patient proceeded to have life-threatening swallowing and breathing difficulties for which intervention was needed.
In response to a case report in the New England Journal of Medicine (Breitbart, 1986) on tardive dyskinesia associated with high-dose intravenous metoclopramide, the AH Robins Company reported (Broad, 1986) that since first marketing Reglan in the United States in 1979 and through April 1986 it had received notice of 26 cases of involuntary movement disorders consonant with tardive dyskinesia. Onset occurred after four to 37 months of therapy (average 13 months), primarily in women, and with a mean age of 67 years. Most patients had either complete recovery or considerable improvement.
Miller and Jankovic (1989), in the Archives of Internal Medicine, added 10 cases seen at the movement disorder clinic at Baylor College of Medicine to the published literature. Their patients ranged between 53-85 years of age. Four of the cases had 2.5 to 7 months exposure before symptom onset; four had 24 to 48 months exposure before symptom onset. Two cases lacked information. Miller and Jankovic (1989) also included a review of the literature on metoclopramide-induced movement disorders, including most of the tardive dyskinesia cases reported above. This was the first major literature review on metoclopramide and movement disorders, including tardive dyskinesia.
Lang (1990), in the Canadian Journal of Neurological Sciences, reported a case of metoclopramide-induced tardive dyskinesia in a 74 year-old diabetic woman who developed pelvic thrusting while on metoclopramide and mild orofacial dyskinesias three months after withdrawal. He reported nine cases of metoclopramide tardive dyskinesia from the Toronto Western Hospital and found them to have more pelvic thrusting and respiratory dyskinesia than his 33 anti-psychotic related tardive dyskinesia cases.
Stewart et al. (1992) reported in Annals of Pharmacotherapy having seen two elderly patients with possible metoclopramide-induced tardive dyskinesia at the University of Florida Shands Hospital after eight or more years of treatment. One was a 57 year-old registered nurse; the other was not described. Having seen these cases in elderly adults, they chose to review the health surveillance records at the Florida Geriatric Research Program to estimate the use of metoclopramide in the elderly. They found that 34 of the 4,515 subjects (0.75%) had recorded metoclopramide in their medication history at least once. One-third of them (11) reported having taken metoclopramide for a year or more, or 0.25% of the elderly subjects. This use frequency can be compared with that from Jagtland, Sweden, showing that 0.3% of the males and 0.8% of the females aged 70 or more could have taken metoclopramide for at least six months.
A second review of the literature on mctoclopramide-associated dyskinesia was published by Sewell and Jeste (1992) based on a total of 67 published cases from 21 articles. Their case listing was comprised of 52 cases from 16 papers, 35 of which came from Grimes et al. (1982), Wiholm et al. (1984), and Miller and Jankovic (1989). Additional cases listed included Parkinsonism cases and intravenous exposures. The mean ages ranged between 46 and 84. Excluding the large case series, ten of the case were identified as male and nine as female. Eight of the 15 papers with reported mean length of exposure prior to onset of tardive dyskinesia had means of less than 12 months. The authors report that the mean length of metoclopramide treatment prior to the onset of dyskinesia was 20 months. In 15 of 21 patients on whom long-term follow-up was provided, the symptoms persisted 6 or more months after discontinuation of Metoclopramide. They report that Metoclopramide has been in clinical usage in the UK since 1967 and in the USA since 1979, with US sales in the late 1980s of about 1.5 million units annually.
The above case reports and series of cases demonstrate that tardive dyskinesia has been observed to occur in metoclopramide patients in multiple countries. These studies, lacking denominator information (e.g., number of people prescribed metoclopramide), indicate a tardive dyskinesia problem among metoclopramide patients but are inadequate to indicate the rate or risk of occurrence. Analyses of cases and case series cannot derive a risk estimate, since they include no information on the number of people in the exposed population from whom the cases came. Case series can provide an estimate of the proportion of cases seen that had a certain exposure, but these numbers are greatly influenced by both frequency of exposure and referral patterns in the community.
Adverse Drug Effects Registries:
Shaffer et al. (2004), in the Journal of the American Pharmacists Association, presented an analysis of the eighty-seven cases of tardive dyskinesia listed in the US Food and Drug Administration's (FDA) Adverse Event Reporting System (AERS) as associated with Metoclopramide use. This covered the period of 1968 through June 2003. The mean age was 60 years old (range: 11 weeks to 95 years). The median duration of use was about one year (360 days), and the mean duration was about two years (753 days). Two-thirds of the cases were female, and slightly less than half (46%) met the case definition criteria. Neither the years of diagnosis nor locations (US or Foreign) of the cases were indicated. Sales data are presented that indicate that there have been an estimated over 50 million prescriptions of Metoclopramide in the United States since 1991. A separate FDA analysis [Kaplan et al., 2007] in Pharmacoepidemiology and Drug Safety showed that 13% of metoclopramide patients used metoclopramide for longer than 90 days. This analysis was over 200,000 metoclopramide users in the 2002-2004 data base for the pharmacy benefit database, Caremark. Ages of the patients are not given.
A later analysis by the FDA published on their website (Southworth, 2005) reported that the mean duration was 635 days, that the median duration of use was only six months (180 days), and that the interquartile range was from 29.5 to 821.25 days. Thus, a quarter of the cases were reported as having had fewer than 30 days exposure. This analysis reported 66 cases in the United States and two foreign cases to January 2005. It also reported that metoclopramide sales in the US have been 45 million prescriptions for the 10 year period (1995-2004). If one assumed that all the US cases had occurred within that 10 year period, then the incidence rate for reported cases would have been 66 cases for 45 million prescriptions, or 1 1/2 cases of TD reported per one million metoclopramide prescriptions. This rate for the US data (1.5 per million) is similar to the rates for the UK (0.25 per million) reported twenty years earlier by Bateman et al. (1985). The initial rate of one case per 17,800 prescriptions as reported from the Swedish registry (Wiholm et al., 1984) has not been confirmed.
The Southworth case report count differs from that of Shaffer as Southworth, while having 1 1/2 year more data, excluded cases with onset after exposure withdrawal. The similar calculation based on Shaffer et al. (2004) would have yielded a rate of 1.8 reported tardive dyskinesia cases per million metoclopramide prescriptions.
Thus, grossly, this would indicate that tardive dyskinesia associated with Metoclopramide is a rare event, even recognizing both under-reporting and under-ascertainment. Further, this indicates that the information in 2005 is no different than it was in 1985.
Summary:
I have tried to present the published literature on the Incidents (case reports) and Incidence Rates (reports to Adverse Drug Effect Registries) regarding clinical cases of tardive dyskinesia in patients using metoclopramide. I have identified about 70 case reports (or cases in clinical case series) in the past thirty years (since 1978). It is possible that there are some additional cases that I may have missed.
I have also identified about 100 clinical cases reported to three adverse drug effect registries. As each of the studies provided metoclopramide prescription sales data, incidence rates (reported cases per million prescriptions) could be calculated. The incidence rates calculated are all in the range of 0.2-2.0 cases reported of tardive dyskinesia per million metoclopramide prescriptions.
Prevalence Studies:
A third type of study exists in the medical literature in which patient who have been on metoclopramide and comparison patients who have not been on metoclopramide are examined for fine signs of dyskinesia. Patients that meet the research definition of dyskinesia (a specific minimum level of involuntary muscle movements by examination) are said to have tardive dyskinesia (research definition) if they have had drug exposure for a certain minimum duration. In the absence of drug exposure (controls or comparison group), they are said to have spontaneous dyskinesia. These studies are called prevalence studies because they took patients based on current or recent metoclopramide use and examined them for their signs of dyskinesia without considering when those signs may have first developed with respect to their exposure history.
There are three published prevalence studies of tardive dyskinesia (research definition) among metoclopramide users. Two (Ganzini et al., 1993 and Sewell et al., 1994) were done using veterans at Veterans Hospitals in the western USA; the third (Mattson et al., 2002) was done using mentally retarded adults resident at a developmental center in Louisiana.
The Ganzini study was conducted among patients at the Veterans hospital in Portland, Oregon; the Sewell study was conducted among patients at the Veterans hospital in San Diego, California. The patients were primarily male and had an average age of 60 years or more. Pharmacy records were used to identify or confirm patient's metoclopramide use. The Ganzini study had 51 metoclopramide-exposed subjects and 51 matched controls; the Sewell study had 51 metoclopramide-exposed subjects and 34 controls. The Ganzini study required that they had been exposed for at least three months; the Sewell study required at least 30 days. The Ganzini study matched patients on diabetes history; the Sewell study did not. The Ganzini study used a matched-pair analysis; the Sewell study did not.
Both studies were designed to test the hypothesis that veterans using metoclopramide were significantly more likely to have dyskinesia (research definition) than were veterans without metoclopramide use. The Ganzini study found a dyskinesia prevalence of 29.4% in their exposed subjects and of 17.7% in their unexposed subjects. The relative risk of having dyskinesia was 1.67 (95% CI, 0.93-2.97), but this risk was not statistically significant (p = 0.08). The Sewell study found a dyskinesia prevalence of 27% in their exposed subjects and of 12% in their unexposed subjects. The relative risk of having dyskinesia was 2.33 (95% CI, 0.84-2.97), but this risk was also not statistically significant (p = 0.08). For statistical significance, the p-value must be less than 0.05. The Ganzini study permitted examining for the effect of diabetes on the dyskinesia prevalence, but the Sewell had insufficient information presented to do so. Thus, the separate effect of metoclopramide exposure across diabetes status on prevalence of dyskinesia (research definition) could not be assessed.
The Sewell study reported the level of severity of the global rating of the dyskinesia (abnormal involuntary movement) scale. None of the subjects had a severe dyskinesia and the frequency of moderate dyskinesia was similar in both the unexposed and the exposed subjects (3-4%). There is no evidence that the prevalence of moderate and severe dyskinesia is greater in patients on metoclopramide than patients not on metoclopramide (Fisher exact two-tail test p = 1.00). The prevalence of mild cases was greater in the exposed than in the unexposed group, but that was not a statistically significant finding (p = 0.14).
In both studies, the diagnoses of tardive dyskinesia were made on the basis of a research case definition rather than on clinical diagnostic criteria. Subjects were specifically examined for these studies with an assessment using the Abnormal Involuntary Movement Scale [AIMS]. This scale had been developed for use in psychopharmacology drug trials in order to document “the absence of abnormal involuntary movements prior to initiation of drug treatment.” It is interesting to note that none of the subjects labeled as having dyskinesia were diagnosed by their own physicians as having dyskinesia. That is, they had not been independently diagnosed as being a clinical case of tardive dyskinesia.
The third prevalence study was the Mattson et al. (2002) study. Long-term residents with mental retardation and on either anti-psychotics or metoclopramide were entered into the study, 25 in each group. Medication records were available for the prior four years that showed that they had each been on their medication, but for how long prior to the four years or whether there was switchover prior to the four years is not known. They were examined for signs of dyskinesia four times during the subsequent year, using the total score from the DISCUS instalment. The mean Discus scores for the two groups were similar (p = 0.88) and declined over the year (p = 0.04). A group of 25 patients not known ever to have been on either anti-psychotics or metoclopramide was added to the study at the time of the fourth examination. The mean DISCUS score for the participants taking anti-psychoties was significantly greater than that of the unexposed. The mean DISCUS score of the participants taking metoclopramide was not significantly greater than that of the unexposed. Patients in the two medication groups were more likely to have met the “probable TD” criteria during at least one of the four examinations that did the unexposed group in its single examination [p = 0.005]. It is unclear what of this difference might be related to the treatment differences and what may be related to the multiple examinations.
The three prevalence studies each seem to indicate that the prevalence of dyskinesia is greater among metoclopramide users than non-users, according to the research definition. The purpose of the research definition was to identify persons who are free of signs of dyskinesia. It has a high sensitivity in that it identifies a condition [mild (research definition) dyskinesia] that is common among the unexposed (e.g., 18% in Ganzini and 12% in Sewell) but is generally not recognized as disease by the patient's own physicians. Mild (research definition) dyskinesia is unlikely to be the condition that brings patients into the courtroom.
Conclusion:
The above literature review demonstrates that the issue of whether there is an increased risk of tardive dyskinesia from Metoclopramide use has been considered for thirty years but that the evidence is insufficient to reach a quantitative answer. Numerous case reports and case-series have accumulated in the medical literature suggesting a relationship. The initial risk suggested by the data from the Swedish registry of adverse drug reactions was not supported by the reports from the registries in the UK and USA. Three cross-sectional studies suggestively support the association, but probably based on mild cases. While many case series have been published, case-control studies have not. There are no prospective studies with patients assessed prior to long-term therapy and monitored throughout therapy. There is enough information in the literature suggesting the association, but these studies do not demonstrate the magnitude of the risk. There is no reliable data for properly designed controlled study that establish the magnitude of the risk of tardive dyskinesia. The manufacturer's figure of 1 in 500 refers to extrapyramidal symptoms in Metoclopramide users. Additional analyses of large patient data bases may with modern technology be able to quantify the risks of tardive dyskinesia with a variety of different medications.
While I have found literature indicating that both diabetes (Woerner et al., 1993) and current smoking (Nillson et al., 1997) are risk factors for dyskinesia, I have not focused in on them as risk factors for Mr. Burnett as he had stopped smoking four years before the appearance of his dyskinesia and his diagnosis of mild diabetes mellitus occurred two years after the onset of his dyskinesia.
Respectively,
Steven H. Lamm, MD, DTPH