Primidone

Primidone is an anticonvulsant of the pyrimidinedione class, the active metabolites of which, phenobarbital (major) and phenylethylmalonamide (PEMA) (minor), are also anticonvulsants. It is used mainly to treat complex partial, simple partial, generalized tonic-clonic seizures, myoclonic, and akinetic seizures and since the 1980s it has been a valuable alternative to propranolol in the treatment of essential tremor. Unlike other anticonvulsants such as carbamazepine and valproic acid, primidone is rarely used in the treatment of bipolar disorder or any other psychiatric problem. It is also not widely used in treatment of neuropathic pain, trigeminal neuralgia, or migraine. Primidone has been occasionally used to treat long QT syndrome, cerebral palsy, and athetosis.

Primidone was once a mainstay anticonvulsant in the treatment of partial and generalized seizures and was the treatment of choice for secondarily generalized seizures originating in the temporal lobes, especially when combined with phenytoin, but by the early 1980s, carbamazepine had surpassed it in popularity due to the latter's lower incidence of sedation. As time passed and more and more anticonvulsants came on the market, primidone was pushed further and further away from its former place of prominence and major Western pharmaceutical corporations became less and less interested in manufacturing and selling it. It has largely fallen into disuse in the developed world as more and more anticonvulsants entered the market and has been withdrawn from markets all over the world.

Unlike many other anticonvulsants, primidone works via interactions with voltage-gated sodium channels which inhibit high-frequency repetitive firing of action potentials. It is much less toxic in overdose than phenobarbital. Along with carbamazepine, phenobarbital, and phenytoin, primidone is an inducer of metabolic enzymes in the liver, which means that it accelerates the metabolism of many other pharmaceuticals. Like other anticonvulsants that entered the market before 1989, primidone has been associated with hyperhomocysteinemia, folate deficiency and its various symptoms (birth defects, depression, and megaloblastic anemia), reduced calcium absorption, and various bone diseases. It also shares with most other anticonvulsants the tendency to cause sedation.

History
Primidone's effectiveness for epilepsy was first demonstrated in 1949 by Dr. Yule Bogue. It was introduced a year later by the Imperial Chemical Industry (ICI), now known as AstraZeneca in the United Kingdom and Germany. In 1952, it was approved in the Netherlands. That same year, Drs. Handley and Stewart demonstrated its effectiveness in the treatment of patients who failed to respond to other therapies; it was noted to be more effective in people with idiopathic generalized epilepsy than in people whose epilepsy had a known cause. Dr. Whitty noted in 1953 that it benefitted patients with psychomotor epilepsy, who were often treatment-resistant. Toxic effects were reported to be mild. That same year, it was approved in France. Primidone was introduced in 1954 under the brandname Mysoline by Wyeth in the United States.

The first report associating it with megaloblastic anemia came in 1954 from Drs. Chalmers and Boheimer. Between 1954 and 1957, twenty-one cases of megaloblastic anemia associated with primidone and/or phenytoin were reported. Most cases were due to folic acid deficiency; however, there was one that only responded to B12 therapy and one that required Vitamin C. Few cases were associated with deficient diets; one patient ate mostly bread and butter, another ate bread, buns, and hard candy, and another could rarely be persuaded to eat in the hospital. The idea that folic acid deficiency could cause megaloblastic anemia was not new. What was new was the idea that drugs could cause this in well-nourished people with no intestinal abnormalities. In many cases, it was not clear which drug had caused it. It was speculated that this might be related to the structural similarity between folic acid, phenytoin, phenobarbital, and primidone. Folic acid had been found to alleviate the symptoms of megaloblastic anemia in the 1940s, not long after it was discovered, but the typical patient only made a full recovery&mdash;cessation of CNS and PNS symptoms as well as anemia&mdash;on B12 therapy. Five years earlier, folic acid deficiency was linked to birth defects in rats. Primidone was seen by some as too valuable to withhold based on the slight possibility of this rare side effect and by others as dangerous enough to be withheld unless phenobarbital or some other barbiturate failed to work for this and other reasons (i.e., reports of permanent psychosis.

In Europe, up until 1963, it was not uncommon to prescribe primidone and phenobarbital in combination, often with a stimulant. They were believed to be the most useful for seizures occurring upon awakening, while phenytoin was the most useful for nocturnal seizures. Primidone and phenobarbital were prescribed in combination with phenytoin in diffuse epilepsies. They were third and fourth line-agents, respectively, in the treatment of partial seizures. By 1963, carbamazepine was marketed in most of Europe. It soon became clear that its efficacy in generalized tonic-clonic seizures was the same as phenytoin, that its ability to control partial seizures was superior, along with its tolerability. Sodium valproate was approved in France in 1967. Because the doses were so low (200–400 mg/day), it was viewed as a moderately effective anticonvulsant whose best quality was its nonsedating nature. In spite of the availability of carbamazepine and valproate, physicians practicing in Mediterranean countries still preferred phenobarbital. Other medications were seen as not usually necessary. Phenytoin was to be used as adjunctive therapy only. In spite of these advances, primidone was still considered to be a "sheet-anchor" anticonvulsant in the United Kingdom as late as 1969, along with phenobarbital and phenytoin, for the treatment of pediatric epilepsy in spite of its side effects and one of two drugs (the other being sultiame) that were tried in adult patients if a combination of phenytoin and phenobarbital failed to control seizures. A review published that same year stated that carbamazepine did not live up to the claims of those who advocated its use in epilepsy.

In 1968, a Dr. Meadow encountered six babies with cleft lip and palate in addition to other congenital abnormalities whose mothers had been taking anticonvulsants. Meadow wrote a letter in The Lancet asking for cases of cleft lip and palate in babies whose mothers had taken anticonvulsants. That same year, Milunsky, Graef, and Gaynor reported cases of cleft lip and palate associated with attempted abortion with methotrexate and aminopterin, which are folic acid antagonists. While it was widely accepted by 1969 that these drugs could interfere with folic acid and that folic acid supplementation might have beneficial somatic effects, it was believed that folic acid supplementation could exacerbate seizures; this included the routine supplements given to pregnant women to prevent megaloblastic anemia. By 1970, the doctor had collected thirty-two cases. Sixteen of them were born to women taking primidone. 25% of the thirty-two cases had congenital heart defects; the reported rate in cleft lip and palate was 3-5%. Meadow emphasized that there was no proof of an association, the immense value of anticonvulsants and the probably small odds of any one epileptic woman having a child with a congenital abnormality. Roman and Caratzali reported in 1971 that the offspring of mice treated with primidone had more abnormal bone metaphases than controls; this was due to chromosomal lesions. By the mid-1970s, it was obvious that this antagonistic effect of primidone was not due to the inhibition of dihydrofolate reductase, the enzyme responsible for the reduction of dihydrofolic acid to tetrahydrofolic acid, but rather to defective folate metabolism.

Carbamazepine was approved in the United States for the treatment of adult epilepsy in 1974. Its lack of sedating properties relative to phenobarbital and lack of somatic effects relative to phenytoin generated much interest. Within two years, primidone was no longer seen as the drug of choice for psychomotor epilepsy in the United States. This was because while carbamazepine and primidone are of roughly equal effectiveness, the former is less likely to cause sedation and cognitive impairment. Also, primidone has a greater tendency to cause undesirable psychiatric side effects compared with carbamazepine, which was noted to lessen pre-existing depressive symptoms. By 1978, it looked as if its superior side effect profile would increase its use in epilepsy in the United States. That same year, a review was published stating that primidone and phenobarbital were used less often in the United Kingdom due to their greater tendency to cause behavior disorders and interfere with learning in young children. Valproate was heralded by some as "the greatest thing since Greta Garbo" and carbamazepine had also risen in popularity.

By 1980, primidone was seen as not worth mentioning as an option for childhood temporal lobe epilepsy by doctors in the United Kingdom. In January 1981, Dr. O'Brien and colleagues reported that primidone had a positive effect on the essential tremor of one of their patients. This led them to initiate a twenty-person prospective study. Twelve of the participants responded well. By 1984, valproic acid was the drug of choice for juvenile myoclonic epilepsy and not the equally effective primidone. However, as late as 1985, primidone was still one of the most widely used anticonvulsants. At the close of the 1980s, primidone was still the preferred anticonvulsant for complex partial seizures in Germany. In 1989, Dainippon Pharmaceutical launched Exegran (zonisamide), the first new, chemically unique, non-benzodiazepine anticonvulsant in decades.

In 1990, primidone, along with phenobarbital, was a second-line agent in partial epilepsy with or without secondarily generalized tonic-clonic seizures and was one of four agents (the others being carbamazepine, phenytoin and phenobarbital) that was used along with ethosuximide or a benzodiazepine for any absence or myoclonic seizures when valproate failed to control tonic-clonics (at least in the United States). After zonisamide, other new anticonvulsants came onto the market: felbamate, gabapentin, lamotrigine, and vigabatrin. All four were structurally distinct both from other anticonvulsants already on the market. They all had larger protective indexes than conventional agents and unlike these agents, the new ones did not cause birth defects in laboratory animals or antagonize folic acid. They seemed to be relatively mild in terms of side effects. Out of all of them lamotrigine was the most similar to phenytoin in its pattern of efficacy. Felbamate was the most effective for Lennox-Gastaut syndrome and was seen as a second-line agent in juvenile myoclonic epilepsy after valproate. These new agents were aggressively marketed. In 1994, felbamate became the anticonvulsant of last resort after ten people out of 100,000 came down with aplastic anemia.

By 1994, primidone was no longer one of the most widely used anticonvulsants. Phenytoin was still regarded as the drug of choice for partial seizures due to its long half-life and low cost; but for children, carbamazepine was seen as the best one due to phenytoin's effects on physical appearance. Topiramate was approved two years later. It, along with the others, was mainly used in patients refractory to carbamazepine and valproate. These new agents were often described as having "innovative" and "selective" mechanisms of action; in reality, most of them also worked similarly to older agents. On February 28, 1998, Élan Corporation, plc, bought the trademark and exclusive product distribution rights for Mysoline from Wyeth in Canada and the United States at a cost of $46 million and a royalty on future sales. The actual manufacture and distribution was done by Athena Neurosciences; their name appeared on a Mysoline package information sheet dated June 1998. On November 30, 1999, levetiracetam was approved for the adjunctive treatment of partial epilepsy in adults in the United States.

By the year 2000, primidone was something that was prescribed in the event that the patient had tried all other anticonvulsants and was not a candidate for surgery in the United States. In April 2001, Élan decided to concentrate its efforts towards Zanaflex, Zonegran, Skelaxin, Abelcet, Azactam, Maxipime, Myobloc, and Cutivate. Mysoline was rationalized along with many other products that did not meet "certain commercial criteria." Yamanouchi Pharma Technologies, a Palo Alto-based subsidiary of Yamanouchi Pharmaceutical Co., Ltd, manufactured the actual drug. By 2003, most of the people taking primidone for epilepsy were elderly people who had been taking the drug for many years. In July 2004, Acorus Therapeutics Ltd. took over the manufacture and distribution of Mysoline from AstraZeneca February 3, 2005: almost four years after acquiring it from Elan, Xcel was acquired by Valeant Pharmaceuticals International. On April 1, 2005, Yamanouchi merged with Fujisawa Pharmaceutical Co., Ltd to form Astellas Pharma. As of 2005, it is widely used in the treatment of many forms of epilepsy in developing countries.

Epilepsy
Licensed for generalized tonic-clonic and complex partial seizures in the United Kingdom. In the United States, primidone is approved for adjunctive (in combination with other drugs) and monotherapy (by itself) use in generalized tonic-clonic seizures, simple partial seizures, and complex partimple partial seizures, and myoclonic seizures. In juvenile myoclonic epilepsy (JME), it is a second-line therapy, reserved for when the valproates and/or lamotrigine do not work and when other second-line therapies&mdash;acetazolamid work either.

Open-label case series have suggested that primidone is effective in the treatment of epilepsy. Primidone has been compared to carbamazepine, phenytoin,  phenobarbital,  mephobarbital, ethotoin, metharbital, and mephenytoin. Compared to carbamazepine, primidone has been found to be equally effective, less effective at controlling partial seizures but just as effective at controlling generalized tonic-clonics, less likely to cause side effects but more likely to cause side effects requiring withdrawal of the drug, half as likely to reduce seizures in patients being considered for surgery by at least 80%, more likely to cause depression, significantly more likely cause intolerable side effects, more likely to cause impotence and decreased libido, and cause more adverse effects on motor performance and attention/concentration tests. In adult comparison trials, primidone higher a incidence of intolerable side effects than phenytoin, a higher incidence of decreased libido and impotence, similar control of tonic-clonic seizures, more likely to cause nausea, vomiting, dizziness, and sedation; twice as likely to be effective in controlling seizures in epilepsy surgery candidates, more acute effects such as nausea, vomiting, dizziness, and sedation, and to be just as effective.

Essential tremor
Primidone is considered to be a first-line therapy for essential tremor along with propranolol. In terms of tremor amplitude reduction, it is just as effective as propranolol, reducing it by 50%. Both drugs are well studied for this condition, unlike other therapies, and are recommended for initial treatment. 250 mg/day (low-dose therapy) is just as good as 750 mg/day (high-dose therapy).

Primidone is not the only anticonvulsant used for essential tremor; the others include topiramate, clonazepam, and gabapentin. Other pharmacological agents include alprazolam (Xanax), atenolol, sotalol, nadolol, clozapine, nimodipine, and botilinum toxin A. Many of these drugs were less effective (according to Table 1), but a few were not. Only propranolol has been compared to primidone in a clinical trial.

Long QT syndrome
The first case report in which the shortening of the QT interval by primidone was documented, published in the July 1980 issue of Annals of Internal Medicine, involved three patients, a 31-year-old woman, her 15-year-old nephew, and his 16-year-old sister. The woman still had ventricular fibrillation, syncope, and seizures even after the removal of her left stellate ganglion and a thoracic chain dissection. The woman had previously been tried on a combination of phenobarbital and phenytoin when she was thought to have only seizures, followed by phenobarbital combined phenytoin and procainamide. The procainamide was replaced with propanolol, which in turn replaced the propanolol when the latter brought the tachycardia back almost instantaneously upon the first dose. The phenobarbital and phenytoin were then combined with atropine and acetyl strophanthidin. When this failed, the atropine and acetyl strophanthidin were replaced with lidocaine prior to the surgery. After the surgery, the QT-prolongation returned, so the phenytoin was doubled to 200 mg four times daily. A month later, she was admitted to the hospital for phenytoin toxicity, where it was found that she had slow spike and abortive wave activity in her left temporal lobe. It was after this that primidone was substituted for phenytoin. The primidone suppressed the fibrillations and lengthened the QT interval for two years and eight months in the patient. Her 16-year-old niece was started on primidone after an unsuccessful trial of phenytoin. Following this, her 15-year-old nephew, the niece's brother, was started because of family history.

The second case report, published in the December 1986 issue of Zhonghua Xin Xue Guan Bing Za Zhi, describes four cases, men, women, adults and adolescents, who were put on primidone for LQTS. The 2002 case focuses on hypocalcemia stemming from such treatment in an adolescent male. The primidone he had been taking suppressed his hypocalcemia-induced QT prolongation.

Psychiatric disorders
In 1965, Monroe and Wise reported using primidone along with a phenothiazine derivative antipsychotic and chlordiazepoxide in treatment-resistant psychosis. What is known is that ten years later, Monroe went on to publish the results of a meta-analysis of two controlled clinical trials on people displaying out-of-character and situationally inappropriate aggression, who had abnormal EEG readings, and who responded poorly to antipsychotics; one of the studies was specifically mentioned as involving psychosis patients. When they were given various anticonvulsants they were administered not only did their EEGs improve, but so did the aggression.

In March 1993, S.G. Hayes of the University of Southern California School of Medicine reported that nine out of twenty-seven people (33%) with either treatment-resistant depression or treatment-resistant bipolar disorder had a permanent positive response to primidone. It should be noted that a plurality of subjects were also given methylphenobarbital in addition to or instead of primidone.

Five months later, Brown, Stone, and Rathbone published a case report titled, "Primidone and rapid cycling affective disorders" describing a 62-year-old woman who had rapid-cycling bipolar disorder starting in 1978. Lithium treatment was started two years later, but it only eliminated the manic swings, leaving her depressions unaffected. Between 1980 and 1989, the patient had six to ten episodes a year, each lasting between eleven and twenty-four days that left her with little energy, made tasks seem more arduous than they actually were, a smaller appetite, a tendency to sleep too much, anxious, and weepy. No antidepressant eliminated all her symptoms. At the age of fifty-eight, she was started on 125 mg/day of primidone for hand tremor. Her depression, which was resistant to all the antidepressants she tried, gradually remitted during primidone therapy; the expected depressive episode lasted three days instead of 13-17. Eight weeks later, she had an episode lasting twenty-eight days. Her last two episodes were nineteen weeks apart, lasting twenty-five and twenty-seven days, respectively. It was during the final episode, in the middle of 1990, that her dose of primidone was stabilized at 500 mg/day. She had been free of depression for two and a half years before the case report was written up.

In 1999, Drs. Linda C. Schaffer, Charles B. Schaffer, and J. Caretto conducted a follow-up study on those earlier reports, as no one else had done so in the six years following their publication, and found it to be roughly as (permanently) effective for refractory bipolar disorder as Hayes had reported it to be (31% vs. Hayes's 33%). All subjects with a permanent and complete treatment response were either Bipolar I or Bipolar NOS. In this study, unlike the 1993 case report, primidone had an impact on manic symptoms&mdash;especially insomnia, anxiety, and racing thoughts&mdash;and only on manic symptoms.

Other
Unlike carbamazepine, there are few case reports mentioning the use of primidone in the treatment of trigeminal neuralgia. The first, published in the October 10, 1957 issue of Gazette Médicale de France, has no abstract. The second one was a case report of a woman taking primidone and then phenobarbital for trigeminal neuralgia. She developed toxic epidermal necrolysis, as well as endocarditis and gastrointestinal hemorrhage. The second recurrence, induced by phenobarbital, killed her within five weeks of onset.

Two trials of primidone for athetosis have been published. In one, small doses of primidone successfully treated eighteen out of thirty-one children. In a subsequent pilot trial in which children who experienced improvement were switched to placebo, the results were much more disappointing, with the two subjects who experienced subjective improvement also experiencing this on placebo.

Primidone has been used in cerebral palsy. In 1953, Plum and Sparup found that out of thirty-three patients, twenty had a favorable or moderate response. Thorn reported in 1962 that sixteen patients responded favorably to both primidone and chlordiazepoxide and four to only primidone. Nine patients responded only to chlordiazepoxide and four responded to neither drug. The best results were obtained in athetosis and rigidity.

Primidone has veterinary uses, including the prevention of aggressive behavior and cannibalism in gilt pigs, and treatment of nervous disorders in dogs and other animals.

Mechanism of action
The exact mechanism of primidone's anticonvulsant action is still unknown after over fifty years. It is believed to work via interactions with voltage-gated sodium channels which inhibit high-frequency repetitive firing of action potentials. The effect of primidone in essential tremor is not mediated by PEMA.

Overdose
The most common symptoms of primidone overdose are coma with loss of deep tendon reflexes and, during the recovery period, if the patient survives, disorientation, dysarthria, nystagmus, and ataxia, lethargy, somnolence, vomiting, nausea, and occasionally, focal neurological deficits which lessen over time. Complete recovery comes within five to seven days of ingestion. The symptoms of primidone poisoning have generally been attributed to its biotransformation to phenobarbital; however, primidone has toxic effects independent of its metabolites in humans. The massive crystalluria that sometimes occurs sets its symptom profile apart from that of phenobarbital. The crystals are white, needle-like, shimmering, hexagonal plates consisting mainly of primidone.

In the Netherlands alone, there were thirty-four cases of suspected primidone poisoning between 1978 and 1982. Out of these, Primidone poisoning was much less common than phenobarbital poisoning. Twenty-seven of those adult cases were reported to the Dutch National Poison Control Centre. Out of these, one person taking it with phenytoin and phenobarbital died, twelve became drowsy and four were comatose.

Treatments for primidone overdose have included hemoperfusion with forced diuresis, a combination of bemegride and amiphenazole; and a combination of bemegride, spironolactone, caffeine, pentylenetetrazol, strophanthin, penicillin and streptomycin.

In the three adults who are reported to have succeeded, the doses were 20–30 g.  However, two adult survivors ingested 30 g 25 g, and 22.5 g. One woman experienced symptoms of primidone intoxication after ingesting 750 mg of her roommate's primidone.

Adverse Effects
Primidone can cause drowsiness, listlessness, ataxia, visual disturbances, nystagmus, headache, and dizziness. These side effects are the most common, occurring in more than 1% of users. Transient nausea and vomiting are also common side effects.

Dupuytren's contracture, a disease of the fasciae in the palm and fingers that permanently bends the fingers (usually the little and ring fingers) toward the palm, was first noted to be highly prevalent in epileptic people in 1941 by a Dr. Lund, fourteen years before primidone was on the market. Lund also noted that it was equally prevalent in individuals with idiopathic and symptomatic epilepsy and that the severity of the epilepsy did not matter. However, only one quarter of the women were affected vs. half of the men. Thirty-five years later, Critcheley et al. reported a correlation between how long a patient had had epilepsy and his or her chance of getting Dupuytren's contracture. They suspected that this was due to phenobarbital therapy, and that the phenobarbital was stimulating peripheral tissue growth factors. Dupuytren's contracture is almost exclusively found in Caucasians, especially those of Viking descent, and highest rates are reported in Northern Scotland, Norway, Iceland, and Australia. It has also been associated with alcoholism, heavy smoking, diabetes mellitus, physical trauma (either penetrating in nature or due to manual labor), tuberculosis, and HIV. People with rheumatoid arthritis are less likely to get this, and Drs. Hart and Hooper speculate that this is also true of gout due to the use of allopurinol This is the only susceptibility factor that is generally agreed upon. Anticonvulsants do not seem to increase the incidence of Dupuytren's contracture in non-whites.

Primidone has other cardiovascular effects in beyond shortening the QT interval. Both it and phenobarbital are associated with elevated serum levels (both fasting and six hours after methionine loading) of homocysteine, an amino acid derived from methionine. This is almost certainly related to the low folate levels reported in primidone users. Elevated levels of homocysteine have been linked to coronary heart disease. In 1985, both drugs were also reported to increase serum levels of high density lipoprotein (HDL) cholesterol, total cholesterol, and apolipoproteins A and B.

It was first reported to exacerbate hepatic porphyria in 1975. In 1981, it was shown that phenobarbital, one of primidone's metabolites, only induced a significant porphyrin at high concentrations in vitro. It can also cause elevations in hepatic enzymes such as gamma-glutamyl transferase and alkaline phosphatase.

Less than 1% of primidone users will experience a rash. Compared to carbamazepine, lamotrigine, and phenytoin, this is very low. The rate is comparable to that of felbamate, vigabatrin, and topiramate. Primidone also causes exfoliative dermatitis, Stevens-Johnson syndrome, and toxic epidermal necrolysis.

Primidone, along with phenytoin and phenobarbital, is one of the anticonvulsants most heavily associated with bone diseases such as osteoporosis, osteopenia (which can precede osteoporosis), osteomalacia and fractures. The populations usually said to be most at risk are institutionalized people, postmenopausal women, older men, people taking more than one anticonvulsant, and children, who are also at risk of rickets. However, it has been suggested that bone demineralization in most pronounced young people (25–44 years of age) and one 1987 study of institutionalized people found that the rate of osteomalacia in the ones taking anticonvulsants&mdash;one out of nineteen individuals taking an anticonvulsant (vs. none among the thirty-seven people taking none) &mdash;was similar to that expected in elderly people. The authors speculated that this was due to improvements in diet, sun exposure and exercise in response to earlier findings, and/or that this was because it was sunnier in London than in the Northern European countries which had earlier reported this effect. In any case, the use of more than one anticonvulsant has been associated with an increased prevalence of bone disease in institutionalized epilepsy patients versus institutionalized people who did not have epilepsy. Likewise, postmenopausal women taking anticonvulsants have a greater risk of fracture than their drug-naive counterparts.

Anticonvulsants affect the bones in many ways. They cause hypophosphatemia, hypocalcemia, low Vitamin D levels, and increased parathyroid hormone. Anticonvulsants also contribute to the increased rate of fractures by causing somnolence, ataxia, and tremor which would cause gait disturbance, further increasing the risk of fractures on top of the increase due to seizures and the restrictions on activity placed on epileptic people.Increased fracture rate has also been reported for carbamazepine, valproate and clonazepam. The risk of fractures is higher for people taking enzyme-inducing anticonvulsants than for people taking non-enzyme-inducing anticonvulsants. In addition to all of the above, primidone can cause arthralgia.

Granulocytopenia, agranulocytosis, and red-cell hypoplasia and aplasia, and megaloblastic anemia are rarely associated with the use of primidone. Megaloblastic anemia is actually a group of related disorders with different causes that share morphological characteristics&mdash;enlarged red blood cells with abnormally high nuclear-cytoplasmic ratios resulting from delayed maturation of nuclei combined with normal maturation of cytoplasm, in to abnormal megakaryocytes and sometimes hypersegmented neutrophils; regardless of etiology, all of the megaloblastic anemias involve impaired DNA synthesis. The anticonvulsant users who get this also tend to eat monotonous diets devoid of fruits and vegetables.

This antagonistic effect is not due to the inhibition of dihydrofolate reductase, the enzyme responsible for the reduction of dihydrofolic acid to tetrahydrofolic acid, but rather to defective folate metabolism.

In addition to increasing the risk of megaloblastic anemia, primidone, like other older anticonvulsants also increases the risk of neural tube defects, and like other enzyme-inducing anticonvulsants, it increases the likelihood of cardiovascular defects, and cleft lip without cleft palate. Epileptic women are generally advised to take folic acid, but there is conflicting evidence regarding the effectiveness of vitamin supplementation in the prevention of such defects.

Additionally, a coagulation defect resembling Vitamin K deficiency has been observed in newborns of mothers taking primidone. Because of this, primidone is a Category D medication.

Primidone, like phenobarbital and the benzodiazepines, can also cause sedation in the newborn and also withdrawal within the first few days of life; phenobarbital is the most likely out of all of them to do that.

In May 2005, Dr. M. Lopez-Gomez's team reported an association between the use of primidone and depression in epilepsy patients; this same study reported that inadequate seizure control, posttraumatic epilepsy, and polytherapy were also risk factors. Polytherapy was also associated with poor seizure control. Out of all of the risk factors, usage of primidone and inadequate seizure control were the greatest; with ORs of 4.089 and 3.084, respectively. They had been looking for factors associated with depression in epilepsy patients. Schaffer et al. 1999 reported that one of their treatment failures, a 45-year-old woman taking 50 mg a day along with lithium 600 mg/day, clozapine 12.5 mg/day, trazadone 50 mg/day, and alprazolam 4 mg/day for three and a half months experienced auditory hallucinations that led to discontinuation of primidone. It can also cause hyperactivity in children; this most commonly occurs at low serum levels. There is one case of an individual developing catatonic schizophrenia when her serum concentration of primidone went above normal.

Primidone is one of the anticonvulsants associated with anticonvulsant hypersensitivity syndrome, others being carbamazepine, phenytoin, and phenobarbital. This syndrome consists of fever, rash, peripheral leukocytosis, lymphadenopathy, and occasionally hepatic necrosis.

Hyperammonemic encephalopathy was reported by Katano Hiroyuki of the Nagoya City Higashi General Hospital in early 2002 in a patient who had been stable on primidone monotherapy for five years before undergoing surgery for astrocytoma, a type of brain tumor. Additionally, her phenobarbital levels were inexplicably elevated post-surgery. This is much more common with the valproates than with any of the barbiturates. A randomized controlled trial whose results were published in the July 1985 issue of The New England Journal of Medicine found that primidone was more likely to cause impotence than phenytoin, carbamazepine, or phenobarbital. Like phenytoin, primidone is rarely associated with lymphadenopathy. Primidone can also cause vomiting; this happens in 1.0–0.1% of users.

Interactions
Taking primidone with monoamine oxidase inhibitors (MAOIs) such as isocarboxazid (Marplan), phenelzine (Nardil), procarbazine (Matulane), selegiline (Eldepryl), tranylcypromine (Parnate) or within two weeks of stopping any one of them may potentiate the effects of primidone or change one's seizure patterns. Isoniazid, an antitubercular agent with MAOI properties, has been known to strongly inhibit the metabolism of primidone.

Like many anticonvulsants, primidone interacts with other anticonvulsants. Clobazam decreases clearance of primidone, Mesuximide increases plasma levels of phenobarbital in primidone users, both primidone and phenobarbital accelerate the metabolism of carbamazepine via CYP3A4, and lamotrigine's apparent clearance is increased by primidone. In addition to being an inducer of CYP3A4, it is also an inducer of CYP1A2, which causes it to interact with substrates such as fluvoxamine, clozapine, olanzapine, and tricyclic antidepressants. It also interacts with CYP2B6 substrates such as bupropion, efavirenz, promethazine, selegiline, and sertraline; CYP2C8 substrates such as amiodarone, paclitaxel, pioglitazone, repaglinide, and rosiglitazone; and CYP2C9 substrates such as bosentan, celecoxib, dapsone, fluoxetine, glimepiride, glipizide, losartan, montelukast, nateglinide, paclitaxel, phenytoin, sulfonamides, trimethoprim, warfarin, and zafirlukast. It also interacts with estrogens.

Primidone and the other enzyme-inducing anticonvulsants can cut the half-life of antipyrine roughly in half (6.2 ± 1.9 h vs. 11.2 ± 4.2 h), and increases the clearance rate by almost 70%. Phenobarbital reduces the half-life to 4.8 ± 1.3 and increases the clearance by almost 109%. It also interferes with the metabolism of dexamethasone, a synthetic steroid hormone, to the point where its withdrawal from the regimen of a 14-year-old living in the United Kingdom made her hypercortisolemic. Tempelhoff and colleagues at the Washington University School of Medicine's Department of Anesthesiology reported in 1990 that primidone and other anticonvulsant drugs increase the amount of fentanyl needed during craniotomy based on the patient's heart rate.

Pharmacokinetics
Primidone converts to phenobarbital and PEMA; it is still unknown which exact cytochrome P450 enzymes are responsible. The phenobarbital, in turn, is metabolized to p-hydroxyphenobarbital. The rate of primidone metabolism was greatly accelerated by phenobarbital pretreatment, moderately accelerated by primidone pretreatment, and reduced by PEMA pretreatment. In 1983, a new minor metabolite, p-hydroxyprimidone, was discovered.

Primidone, carbamazepine, phenobarbital and phenytoin are among the most potent hepatic enzyme inducing drugs in existence. This enzyme induction occurs at therapeutic doses. In fact, people taking these drugs have displayed the highest degree of hepatic enzyme induction on record. In addition to being an inducer of CYP3A4, it is also an inducer of CYP1A2, which causes it to interact with substrates such as fluvoxamine, clozapine, olanzapine, and tricyclic antidepressants. Its metabolite, phenobarbital, is a substrate of CYP2C9, CYP2B6, CYP2C8, CYP2C19, CYP2A6, CYP3A5, CYP1E1, and the CYP2E subfamily. The gene expression of these isoenzymes is regulated by human pregnane receptor X (PXR) and constitutive androstane receptor (CAR). Phenobarbital induction of CYP2B6 is mediated by both. Primidone does not activate PXR.

The rate of metabolism of primidone into phenobarbital was inversely related to age; the highest rates were in the oldest patients (the maximum age being 55). People aged 70–81, relative to people aged 18–26, have decreased renal clearance of primidone, phenobarbital, and PEMA, in ascending order of significance, and that there was a greater proportion of PEMA in the urine. The clinical significance is unknown.

The percentage of primidone converted to phenobarbital has been estimated to be 5% in dogs and 15% in humans. Work done twelve years later found that the serum phenobarbital 0.111 mg/100 mL for every mg/kg of primidone ingested. Authors publishing a year earlier estimated that 24.5% of primidone was metabolized to phenobarbital. However, the patient reported by Kappy and Buckley would have had a serum level of 44.4 mg/100 mL instead of 8.5 mg/100 mL if this were true for individuals who have ingested large dose. The patient reported by Morley and Wynne would have had serum barbiturate levels of 50 mg/100 mL, which would have been fatal.

Available forms
Primidone is available as a 250 mg/5mL suspension, and in the form of 50 mg, 125 mg, and 250 tablets. It is also available in a chewable tablet formulation in Canada.

It is marketed as several different brands including Mysoline (Canada, Ireland, Japan, the United Kingdom, and the United States ), Prysoline (Israel, Rekah Pharmaceutical Products, Ltd.), Apo-Primidone, Liskantin (Germany, Desitin), Resimatil (Germany, Sanofi-Synthélabo GmbH), Mylepsinum (Germany, AWD.pharma GmbH & Co., KG)., and Sertan (Hungary, 250 mg tablets, ICN Pharmaceuticals Inc.)

Primidone has been available in the United States as a generic drug from Lannett since 1978.

Chemistry
Primidone, 5-ethyl-5-phenylhexahydropyrimidine-4,6-dione is synthesized by reacting ethylphenylmalonic acid diamide with formamide. An alternative method is the electrolytic reduction of phenobarbital or the catalytic reduction of the appropriate 2-thiobarbituric acid.
 * W.R. Boon, H.C. Carrington, C.H. Vasey, (1951).
 * W.R. Boon, N. Greenhalgh, E. Lenden, Ch.H. Vasey, (1952).
 * Ch.H. Vasey, W. Robert, (1952).