Epilepsy, Myasthenia Gravis, and Effect of Plasmapheresis on Antiepileptic Drug Concentrations Chi-Wan Lai, MD; Ilo
E.
Leppik, MD;
David C. Jenkins, MD; Paul Sood, PhD
28-year-old woman developed complex partial seizures at the age of 17 years and was treated with phenytoin sodium. Five years later she developed myasthenia gravis, and phenytoin was replaced by valproic acid and phenobarbital. She required plasmapheresis (PP). During one course of PP, total and unbound concentrations of valproic acid and phenobarbital were measured in serum sampled before, during, and after PP and in plasma removed by PP. It was determined that the magnitude of loss of valproic acid or phenobarbital by PP was small, and the changes of unbound/total ratio did not reach clinical importance. (Arch Neurol. 1990;47:66-68) \s=b\ A
"WThile removal of phénobarbital or valproic acid by peritoneal dialysis1·2 and hemodialysis3·4 has been established, extraction of these drugs
in patients undergoing plasmaphere¬ sis (PP) has not been studied in detail. Plasmapheresis has become an ac¬ cepted effective treatment for myas¬ thenia gravis,5·6 which has been re¬ ported in association with epilepsy7·8 and epileptiform patterns in electroencephalography.9 It has been shown to significantly decrease pheny¬ toin concentration and result in
Accepted for publication July 11, 1989. From the Departments of Neurology (Dr Lai) and Clinical Pharmacology (Dr Sood), University of Kansas Medical Center, Kansas City; the Department of Neurology, University of Minnesota, Minneapolis (Dr Leppik); and the Community Blood Center of Greater Kansas City (Mo) (Dr Jenkins). Reprint requests to Department of Neurology, University of Kansas Medical Center, 39th and Rainbow Boulevard, Kansas City, KS 66103 (Dr Lai).
our knowledge, its effect on valproic acid and phénobar¬ bital concentrations has not been stud¬ ied. We evaluated the effect of PP on valproic acid and phénobarbital levels in a patient with epilepsy and myas¬ thenia gravis.
seizures,10 but,
to
REPORT OF A CASE
A 28-year-old woman was diagnosed as having complex partial seizures at the age of 17 years. Electroencephalography re¬ vealed an epileptogenic focus in the left an¬ terior temporal lobe. The patient was ini¬ tially treated with phenytoin sodium (Di¬ lantin) (500 mg/d), with good control. At the age of 22 years, she developed weakness of legs, diplopia, and difficulty with swal¬ lowing and respiration. Electromyography and an edrophonium test confirmed the di¬ agnosis of myasthenia gravis. Pyridostigmine bromide therapy (60 mg four times daily) initially improved the condition, and the patient subsequently underwent a thymectomy and was treated with pred¬
nisone. One year later, she noticed a tem¬ poral relationship between taking pheny¬ toin and increasing weakness; phenytoin therapy was discontinued and, since then, she has been taking valproic acid (Depakene) and phénobarbital. The patient first underwent PP 5 years before this study began. She was also treated with azathioprine (25 mg four times daily) and prednisone (50 mg every other day). Her muscle strength improved dra¬ matically, and the titer of acetylcholine re¬ ceptor antibody decreased from 10 U to less than 1 TJ. Over a 4-year period she under¬ went a total of 13 courses of PP. Each course consisted of 2 to 3 consecutive days of plasma exchange, and each procedure lasted approximately 2 hours. Thus, during this 4-year period the patient had 35 plasma exchanges. She continued to have seizures about once every 1 to 4 months, and she ex¬ perienced seizures during two different PP courses.
Downloaded From: http://archneur.jamanetwork.com/ by a University of Manitoba User on 06/13/2015
METHODS PP Procedures
The plasma exchange was performed by continuous-flow cell separator at the Com¬ munity Blood Center of Kansas City (Mo) on three consecutive mornings. A total of 2500 mL of plasma was removed (1.4 times the estimated plasma volume) and ex¬ changed with 1800 mL of saline and 1250 mL of 5% normal serum albumin. During the exchange, 450 mL of adenine citrate dextrose A solution was introduced into the system. The exchange lasted for 2 hours on days 1 and 2, without technical difficulty, but an occluded arteriovenous fistula was encountered on day 3. Therefore, data from day 3 were not included in this study. Medication Schedule
The patient had been on the following drug regimen for 4 months before this PP: pyridostigmine bromide (120 mg three times a day before meals and a 180-mg time-span capsule at bedtime), valproic acid (250 mg at 8 am, 1 pm, and 5 pm and 500 mg at 9 pm, for a total daily dose of 1250 mg), and phénobarbital (115 mg at 9 pm). To as¬ sure that this study was carried out during steady state conditions, the patient was
asked to maintain her usual schedule dur¬
ing all 3 days of PP. Specimen Collection Five serum samples were drawn as fol¬ lows: sample 1, at 8 am prior to the first dose (trough level); sample 2, immediately be¬ fore PP at 9 am; samples 3 and 4, during PP (after the exchange of bag 1 of 1000 mL and after bag 2 of another 1000 mL); and sam¬ ple 5, immediately after the completion of PP (after the collection of bag 3 of 500 mL). Three additional samples were collected from the aliquot of the removed plasma from bags 1, 2, and 3.
Drug Assay Both total and unbound (free) concentra¬ tions were measured. Unbound valproic
acid and phénobarbital in serum and re¬ moved plasma were determined using the principle of ultrafiltration and centrifugation" (Amicon MPS-1 Centrifree Micropartition System, Amicon Corp, Scientific Systems Division, Danvers, Mass). For the determination of the concentration of un¬ bound drugs, 1 mL of serum or removed plasma was poured into the sample reser¬ voir of the filtration system and capped. After being tightly attached to the filtrate cup, it was centrifuged at 2500 X g for 10 minutes. Approximately 0.5 mL of the fil¬ trate was collected. Both unbound and total valproic acid and phénobarbital levels were determined by fluorescence polarization immunoassay (TDX Analyzer System, Abbott Laborato¬
Table 1.—Each Bag of Removed Plasma Was Checked for Concentrations of Valproic Acid (VPA) and Phénobarbital (PB)* VPA
PB
9/27/83
9/28/83
9/27/83
9/28/83
Drug concentration, µ /L Bag 1 (1000 mL)_273.4_263.0_802_74.6 Bag 2 (1000 mL)_233.9_243.6_74JS_72.0 Bag 3 (500 mL)_233.9_236.0_7a4_73.7 Drug removed, mg Bag 1_304_3 9_18J5_17.3 Bag 2_33^_35J_VT3_16.7 Bag 3_1^4_17 _gj_a6_ Total Removedt
89.0
89.5
*The numbers are the mean of duplicate tExpressed as milligrams per exchange.
45.0
42.6
measures.
ries, Diagnostics Division, Irving, Tex).
Both unbound and total concentrations determined the same day; they were run in duplicate, and coefficients of varia¬ tion were 2.9% for total-form valproic acid, 6.3% for unbound valproic acid, 2.3% for total-form phénobarbital, and 3.6% for un¬ bound phénobarbital. These were well within previously reported limits of reli¬ ability of the drug assay.12 The mean of each pair of duplicate measurements was used in the analysis. were
RESULTS Magnitude of Removal of Valproic Acid and Phénobarbital by PP
From the measured volume and con¬ centration of the removed plasma in each bag, the amount of valproic acid removed was calculated to be 89.5 and 89.0 mg per exchange and that of phé¬ nobarbital was 45 and 42.6 mg per ex¬ change on 2 consecutive days (Table 1). The change in the amount of val¬ proic acid or phénobarbital in the body can be calculated from the volume of distribution and change in serum level. In adults, the volume of distribution is in the range of 0.16 L/kg for valproic acid1315 and 0.54 ± 0.03 L/kg for phénobarbital.16 For valproic acid, the serum level decreased by 100.6 ¿umol/L (14.5 mg/L) on September 27,1983, and by 110.3 µ /L (15.9 mg/L) on Sep¬ tember 28, 1983 (Table 2). Using the equation for volume of distribution and the patient's weight of 46 kg, it can be calculated that 106.7 and 117 mg, respectively, were cleared during PP. For phénobarbital, the serum level de¬ creased by 14.7 /¿mol/L (3.4 mg/L) on September 27 and by 12.1 µ /L (2.8 mg/L) on September 28 (Table 2). This is equivalent to 84.5 and 69.6 mg, re¬ spectively. Thus, PP could account for the major proportion of removal of valproic acid (day 1, 89.5/106.7 mg [83.9%]; day 2, 89/117 mg [76.1%]) and phénobarbital (day 1, 45/84.5 mg [53.3%]; day 2, 42.6/69.6 mg [61.2%]) during PP, with hepatic metabolism and renal clearance accounting for the rest.
Table 2.—Serum Levels of Valproic Acid (VPA) and Phénobarbital (PB) Were Determined at Early Morning (Trough Level) Before, During, and After
Plasmapheresis (PP) After the
Early
Pre-PP
am
1000 mL
Exchange of
2000 mL
2500 mL
VPA, /imol/L Day
1
(9/27/83)
Total
323.4
372.0
298.4
61.8
61.8
16.6
20.7
20.7
18.2
306.7
375.5
284.5
263.7
265.1
13.8
13.3
18.3
18.2
17.0
Unbound
Ratio, %* Day 2 (9/28/83) Total Unbound
Ratio,
274.8
49.3
45.1
PB, (imol/t Day
1
(9/27/83)
Total Unbound * Ratio, %
119.8
114.2
98.3
98.3
99.6
62.6
63.4
65.8
64.5
63.4
Total Unbound
119.0
106.0
93.1
91.4
94.0
70.3
68.5
Ratio, %
59.1
62.1 66.7
66.5
63.8
Day
2
'Ratio
(9/28/83)
of unbound to total VPA
Changes of Serum
Level
or
64.6 PB.
During
COMMENT
PP
Table 2 traces serum levels over the of PP. The serum levels of both valproic acid and phénobarbital dropped appreciably during the early phase of PP, and the magnitude of dif¬ ference was greater for valproic acid than for phénobarbital. course
Effect of PP on Unbound/Total Ratio of Drugs
In comparing the data from pre- and post-PP levels, it was found that un¬ bound/total ratios were unchanged for phénobarbital and that the magnitude of changes was not of clinical impor¬ tance for valproic acid (Table 2). Even after the exchange of significant amounts of plasma protein, drugs were able to quickly reestablish the same proportion between unbound and pro¬ tein-bound form.
Downloaded From: http://archneur.jamanetwork.com/ by a University of Manitoba User on 06/13/2015
Epilepsy coexisting with myasthe¬ gravis is not a common clinical problem; however, it has been noted by Patten8 that myasthénie patients have a higher than normal frequency of sei¬ zures. In one series of 149 patients with juvenile myasthenia, four had associ¬ ated epilepsy.7 In another series of 118 patients with myasthenia, two pa¬ nia
tients had seizures.9 Phenytoin has been associated with disorders in neuromuscular transmis¬ sion in both human experiences and rat experiments.1'18 One instance of
myasthenia developing during pheny¬
toin Our
use
has been well documented.19
patient developed myasthenia gravis after being treated with pheny¬ toin for 5 years. During her first year of myasthenia gravis, symptoms tem¬ porally related to the intake of pheny-
toin
were
observed. This further
ex¬
emplifies the potential of this antiepi¬ leptic drug to affect the myasthénie
condition.
Plasmapheresis has been noted to rapid fluctuations of antimyasthenic drugs20 and phenytoin.10 In one patient with thrombotic thrombocy¬ topenic purpura who underwent two plasma-volume exchanges lasting for 4 to 5 hours, the procedures removed a significant amount of phenytoin, and additional doses of phenytoin were re¬ quired to control the seizures.10 In our patient, enough time did not pass dur¬ ing PP to permit calculation of a valid half-life, and the patient did not have the same magnitude of exchange (only 1.4 plasma-volume exchanges in 2 hours). The magnitude of loss of val¬ proic acid and phénobarbital by PP, on 2 different days, was relatively small, amounting to 7.1% to 7.2% and 37% to 39% of the daily dose of valproic acid and phénobarbital, respectively. It is known that PP removes plasma cause
factors most efficiently during the first hour of exchange,21 and our patient showed an appreciable drop of serum level of valproic acid, and to a lesser degree of phénobarbital, in the first 1000-mL exchange. Studies of antiepi¬ leptic drug concentrations in the hu¬ man epileptogenic brain have shown a good correlation between brain and plasma ratio in phenytoin, phénobar¬ bital, and valproic acid.22 24 In contrast to phénobarbital and phenytoin, how¬ ever, valproic acid concentration in the brain is thought not to reflect selective brain receptor binding, although it is related to the free valproic acid level in plasma.24 Goldberg and Todoroff25 found that valproic acid was not bound to brain or its subfraction; and there was little, if any, binding of valproic acid to either protein or phospholipid in the brain. In our patient, the effect of PP in removing valproic acid is sub¬ stantially higher than in removing
phénobarbital. In the past,
our
patient had experi-
enced seizures during 2 of her 35 plasma exchanges. This appeared to be greater than expected from her usual frequency of seizures. These two sei¬ zures, nonetheless, occurred exclu¬ sively on the first days of two different 3-day courses of PP. The patient ad¬ mitted that she had not been compli¬ ant to antiepileptic drugs before her hospitalization on these two occasions. Therefore, the combined effects of noncompliance and the subsequent rapid decrease of antiepileptic drug levels by PP may have resulted in the occur¬ rence of these two seizures. Conse¬ quently, we believe that it is important to keep the serum level of valproic acid or phénobarbital in therapeutic range before PP, but it may not be necessary to give additional doses of valproic acid or phénobarbital during PP. This study was supported in part by grant P5016308 from the National Institutes of Health, Be¬ thesda, Md. The authors wish to thank Patricia Melching for typing the manuscript.
References 1. Chow-Tung E, Lau AH, Vidyasagar D, John EG. Clearance of phenobarbital by peritoneal dialysis in a neonate. Clin Pharm. 1982;1:268-271. 2. Orr JM, Farrell K, Abbott FS, Ferguson S, Godolphin WJ. The effects of peritoneal dialysis on the single dose and steady state pharmacokinetics of valproic acid in a uremic epileptic child. Eur J Clin Pharmacol. 1983;24:387-390. 3. Zawada ET, Nappi J, Done G, Rollins D. Advances in the hemodialysis management of phenobarbital overdose. South Med J. 1983;76:6-8. 4. Mortenson PB, Hansen HE, Pedersen B, Hartmann-Andersen F, Husted SE. Acute valproate intoxication: biochemical investigations and hemodialysis treatment. Int J Clin Pharmacol Ther Toxicol. 1983;21:64-68. 5. Seybold ME. Plasmapheresis in myasthenia gravis. Ann NY Acad Sci. 1987;505:584-587. 6. Cornelio F, Peluchetti D, Mantegazza R, Sghirlanzoni A, Collarile C. The course of myasthenia gravis in patients treated with corticosteroids, azathioprine, and plasmapheresis. Ann NY Acad Sci. 1987;505:517-525. 7. Rodriguez M, Gomez MR, Howard FM Jr, Taylor WF. Myasthenia gravis in children: long\x=req-\ term follow-up. Ann Neurol. 1983;13:504-510. 8. Patten BM. Myasthenia gravis: review of diagnosis and management. Muscle Nerve. 1978; 1:190-205. 9. Tartara A, Mola M, Manni R, et al. EEG findings in 118 cases of myasthenia gravis. Rev
Electroencephalogr Neurophysiol Clin. 1982;
12:275-279. 10. Liu E, Rubenstein M. Phenytoin removal by plasmapheresis in thrombotic thrombocytopenic purpura. Clin Pharmacol Ther. 1982;31:762-765. 11. Fischer JH, Cloyd JC, Ehresman DJ, Lakatua DJ. Comparison of two ultrafiltration methods with equilibrium dialysis for the determination of free valproic acid plasma concentration. Epilepsia. 1983;24:112-113. 12. Gerson B. Quality control in therapeutic drug monitoring: intralaboratory precision and medical requirements. Ther Drug Monit. 1980; 2:225-232. 13. Gugler R, von Unruh GE. Clinical pharmacokinetics of valproic acid. Clin Pharmacokinet.
1980;5:67-83.
14. Bryson SM, Verma N, Scott PJ, Rubin PC. Pharmacokinetics of valproic acid in young and elderly subjects. Br J Clin Pharmacol.
1983;16:104-105. 15. Eadie MJ, Heazlewood V, McKauge L, Tyrer JH. Steady-state valproate pharmacokinetics
during long-term therapy. Clin Exp Neurol.
1983;19:183-191. 16. Wilensky AJ, Friel PN, Levy RH, Comfort CP, Kaluzny SP. Kinetics of phenobarbital in
normal subjects and epileptic patients. Eur J Clin Pharmacol. 1982;23:87-92. 17. Norris FH Jr, Colella J, McFarlin D. Effect of diphenylhydantoin on neuromuscular synapse.
Downloaded From: http://archneur.jamanetwork.com/ by a University of Manitoba User on 06/13/2015
Neurology. 1964;14:869-877. 18. Kaeser HE. Drug-induced myasthenic syndromes. Acta Neurol Scand Suppl. 1984;100:39-47. 19. Milonas J, Kountouris D, Scheer E. Myasthenic syndrome following long-term diphenylhydantoin therapy. Nervenarzt. 1983;54:437-438. 20. Shumak KH, Rock GA. Therapeutic plasma exchange. N Engl J Med. 1984;310:762-771. 21. Orkin JB, Berkman EM. Partial plasma exchange using albumin replacement. Blood. 1980;56:1055-1059. 22. Sherwin AL, Harvey CD, Leppik IE. Quantitation of antiepileptic drugs in human brain. In: Kellaway P, Petersen I, eds. Quantitative Analytic Studies in Epilepsy. New York, NY: Raven Press;
1976:171-182. 23. Vajda F, Williams FH, Davidson S, Falconer MA, Breckenridge A. Human brain, cerebrospinal fluid, and plasma concentrations of diphenylhydantoin and phenobarbital. Clin Pharmacol Ther. 1974;15:597-603. 24. Vajda FJE, Donnan GA, Phillips J, Bladin PF. Human brain, plasma, and cerebrospinal fluid concentration of sodium valproate after 72 hours of therapy. Neurology (NY). 1981;31:486\x=req-\ 487. 25. Goldberg MA, Todoroff T. Carbamazepine and valproic acid binding to brain. Neurology
(NY). 1979;29:577.
E.
Leppik, MD;
David C. Jenkins, MD; Paul Sood, PhD
28-year-old woman developed complex partial seizures at the age of 17 years and was treated with phenytoin sodium. Five years later she developed myasthenia gravis, and phenytoin was replaced by valproic acid and phenobarbital. She required plasmapheresis (PP). During one course of PP, total and unbound concentrations of valproic acid and phenobarbital were measured in serum sampled before, during, and after PP and in plasma removed by PP. It was determined that the magnitude of loss of valproic acid or phenobarbital by PP was small, and the changes of unbound/total ratio did not reach clinical importance. (Arch Neurol. 1990;47:66-68) \s=b\ A
"WThile removal of phénobarbital or valproic acid by peritoneal dialysis1·2 and hemodialysis3·4 has been established, extraction of these drugs
in patients undergoing plasmaphere¬ sis (PP) has not been studied in detail. Plasmapheresis has become an ac¬ cepted effective treatment for myas¬ thenia gravis,5·6 which has been re¬ ported in association with epilepsy7·8 and epileptiform patterns in electroencephalography.9 It has been shown to significantly decrease pheny¬ toin concentration and result in
Accepted for publication July 11, 1989. From the Departments of Neurology (Dr Lai) and Clinical Pharmacology (Dr Sood), University of Kansas Medical Center, Kansas City; the Department of Neurology, University of Minnesota, Minneapolis (Dr Leppik); and the Community Blood Center of Greater Kansas City (Mo) (Dr Jenkins). Reprint requests to Department of Neurology, University of Kansas Medical Center, 39th and Rainbow Boulevard, Kansas City, KS 66103 (Dr Lai).
our knowledge, its effect on valproic acid and phénobar¬ bital concentrations has not been stud¬ ied. We evaluated the effect of PP on valproic acid and phénobarbital levels in a patient with epilepsy and myas¬ thenia gravis.
seizures,10 but,
to
REPORT OF A CASE
A 28-year-old woman was diagnosed as having complex partial seizures at the age of 17 years. Electroencephalography re¬ vealed an epileptogenic focus in the left an¬ terior temporal lobe. The patient was ini¬ tially treated with phenytoin sodium (Di¬ lantin) (500 mg/d), with good control. At the age of 22 years, she developed weakness of legs, diplopia, and difficulty with swal¬ lowing and respiration. Electromyography and an edrophonium test confirmed the di¬ agnosis of myasthenia gravis. Pyridostigmine bromide therapy (60 mg four times daily) initially improved the condition, and the patient subsequently underwent a thymectomy and was treated with pred¬
nisone. One year later, she noticed a tem¬ poral relationship between taking pheny¬ toin and increasing weakness; phenytoin therapy was discontinued and, since then, she has been taking valproic acid (Depakene) and phénobarbital. The patient first underwent PP 5 years before this study began. She was also treated with azathioprine (25 mg four times daily) and prednisone (50 mg every other day). Her muscle strength improved dra¬ matically, and the titer of acetylcholine re¬ ceptor antibody decreased from 10 U to less than 1 TJ. Over a 4-year period she under¬ went a total of 13 courses of PP. Each course consisted of 2 to 3 consecutive days of plasma exchange, and each procedure lasted approximately 2 hours. Thus, during this 4-year period the patient had 35 plasma exchanges. She continued to have seizures about once every 1 to 4 months, and she ex¬ perienced seizures during two different PP courses.
Downloaded From: http://archneur.jamanetwork.com/ by a University of Manitoba User on 06/13/2015
METHODS PP Procedures
The plasma exchange was performed by continuous-flow cell separator at the Com¬ munity Blood Center of Kansas City (Mo) on three consecutive mornings. A total of 2500 mL of plasma was removed (1.4 times the estimated plasma volume) and ex¬ changed with 1800 mL of saline and 1250 mL of 5% normal serum albumin. During the exchange, 450 mL of adenine citrate dextrose A solution was introduced into the system. The exchange lasted for 2 hours on days 1 and 2, without technical difficulty, but an occluded arteriovenous fistula was encountered on day 3. Therefore, data from day 3 were not included in this study. Medication Schedule
The patient had been on the following drug regimen for 4 months before this PP: pyridostigmine bromide (120 mg three times a day before meals and a 180-mg time-span capsule at bedtime), valproic acid (250 mg at 8 am, 1 pm, and 5 pm and 500 mg at 9 pm, for a total daily dose of 1250 mg), and phénobarbital (115 mg at 9 pm). To as¬ sure that this study was carried out during steady state conditions, the patient was
asked to maintain her usual schedule dur¬
ing all 3 days of PP. Specimen Collection Five serum samples were drawn as fol¬ lows: sample 1, at 8 am prior to the first dose (trough level); sample 2, immediately be¬ fore PP at 9 am; samples 3 and 4, during PP (after the exchange of bag 1 of 1000 mL and after bag 2 of another 1000 mL); and sam¬ ple 5, immediately after the completion of PP (after the collection of bag 3 of 500 mL). Three additional samples were collected from the aliquot of the removed plasma from bags 1, 2, and 3.
Drug Assay Both total and unbound (free) concentra¬ tions were measured. Unbound valproic
acid and phénobarbital in serum and re¬ moved plasma were determined using the principle of ultrafiltration and centrifugation" (Amicon MPS-1 Centrifree Micropartition System, Amicon Corp, Scientific Systems Division, Danvers, Mass). For the determination of the concentration of un¬ bound drugs, 1 mL of serum or removed plasma was poured into the sample reser¬ voir of the filtration system and capped. After being tightly attached to the filtrate cup, it was centrifuged at 2500 X g for 10 minutes. Approximately 0.5 mL of the fil¬ trate was collected. Both unbound and total valproic acid and phénobarbital levels were determined by fluorescence polarization immunoassay (TDX Analyzer System, Abbott Laborato¬
Table 1.—Each Bag of Removed Plasma Was Checked for Concentrations of Valproic Acid (VPA) and Phénobarbital (PB)* VPA
PB
9/27/83
9/28/83
9/27/83
9/28/83
Drug concentration, µ /L Bag 1 (1000 mL)_273.4_263.0_802_74.6 Bag 2 (1000 mL)_233.9_243.6_74JS_72.0 Bag 3 (500 mL)_233.9_236.0_7a4_73.7 Drug removed, mg Bag 1_304_3 9_18J5_17.3 Bag 2_33^_35J_VT3_16.7 Bag 3_1^4_17 _gj_a6_ Total Removedt
89.0
89.5
*The numbers are the mean of duplicate tExpressed as milligrams per exchange.
45.0
42.6
measures.
ries, Diagnostics Division, Irving, Tex).
Both unbound and total concentrations determined the same day; they were run in duplicate, and coefficients of varia¬ tion were 2.9% for total-form valproic acid, 6.3% for unbound valproic acid, 2.3% for total-form phénobarbital, and 3.6% for un¬ bound phénobarbital. These were well within previously reported limits of reli¬ ability of the drug assay.12 The mean of each pair of duplicate measurements was used in the analysis. were
RESULTS Magnitude of Removal of Valproic Acid and Phénobarbital by PP
From the measured volume and con¬ centration of the removed plasma in each bag, the amount of valproic acid removed was calculated to be 89.5 and 89.0 mg per exchange and that of phé¬ nobarbital was 45 and 42.6 mg per ex¬ change on 2 consecutive days (Table 1). The change in the amount of val¬ proic acid or phénobarbital in the body can be calculated from the volume of distribution and change in serum level. In adults, the volume of distribution is in the range of 0.16 L/kg for valproic acid1315 and 0.54 ± 0.03 L/kg for phénobarbital.16 For valproic acid, the serum level decreased by 100.6 ¿umol/L (14.5 mg/L) on September 27,1983, and by 110.3 µ /L (15.9 mg/L) on Sep¬ tember 28, 1983 (Table 2). Using the equation for volume of distribution and the patient's weight of 46 kg, it can be calculated that 106.7 and 117 mg, respectively, were cleared during PP. For phénobarbital, the serum level de¬ creased by 14.7 /¿mol/L (3.4 mg/L) on September 27 and by 12.1 µ /L (2.8 mg/L) on September 28 (Table 2). This is equivalent to 84.5 and 69.6 mg, re¬ spectively. Thus, PP could account for the major proportion of removal of valproic acid (day 1, 89.5/106.7 mg [83.9%]; day 2, 89/117 mg [76.1%]) and phénobarbital (day 1, 45/84.5 mg [53.3%]; day 2, 42.6/69.6 mg [61.2%]) during PP, with hepatic metabolism and renal clearance accounting for the rest.
Table 2.—Serum Levels of Valproic Acid (VPA) and Phénobarbital (PB) Were Determined at Early Morning (Trough Level) Before, During, and After
Plasmapheresis (PP) After the
Early
Pre-PP
am
1000 mL
Exchange of
2000 mL
2500 mL
VPA, /imol/L Day
1
(9/27/83)
Total
323.4
372.0
298.4
61.8
61.8
16.6
20.7
20.7
18.2
306.7
375.5
284.5
263.7
265.1
13.8
13.3
18.3
18.2
17.0
Unbound
Ratio, %* Day 2 (9/28/83) Total Unbound
Ratio,
274.8
49.3
45.1
PB, (imol/t Day
1
(9/27/83)
Total Unbound * Ratio, %
119.8
114.2
98.3
98.3
99.6
62.6
63.4
65.8
64.5
63.4
Total Unbound
119.0
106.0
93.1
91.4
94.0
70.3
68.5
Ratio, %
59.1
62.1 66.7
66.5
63.8
Day
2
'Ratio
(9/28/83)
of unbound to total VPA
Changes of Serum
Level
or
64.6 PB.
During
COMMENT
PP
Table 2 traces serum levels over the of PP. The serum levels of both valproic acid and phénobarbital dropped appreciably during the early phase of PP, and the magnitude of dif¬ ference was greater for valproic acid than for phénobarbital. course
Effect of PP on Unbound/Total Ratio of Drugs
In comparing the data from pre- and post-PP levels, it was found that un¬ bound/total ratios were unchanged for phénobarbital and that the magnitude of changes was not of clinical impor¬ tance for valproic acid (Table 2). Even after the exchange of significant amounts of plasma protein, drugs were able to quickly reestablish the same proportion between unbound and pro¬ tein-bound form.
Downloaded From: http://archneur.jamanetwork.com/ by a University of Manitoba User on 06/13/2015
Epilepsy coexisting with myasthe¬ gravis is not a common clinical problem; however, it has been noted by Patten8 that myasthénie patients have a higher than normal frequency of sei¬ zures. In one series of 149 patients with juvenile myasthenia, four had associ¬ ated epilepsy.7 In another series of 118 patients with myasthenia, two pa¬ nia
tients had seizures.9 Phenytoin has been associated with disorders in neuromuscular transmis¬ sion in both human experiences and rat experiments.1'18 One instance of
myasthenia developing during pheny¬
toin Our
use
has been well documented.19
patient developed myasthenia gravis after being treated with pheny¬ toin for 5 years. During her first year of myasthenia gravis, symptoms tem¬ porally related to the intake of pheny-
toin
were
observed. This further
ex¬
emplifies the potential of this antiepi¬ leptic drug to affect the myasthénie
condition.
Plasmapheresis has been noted to rapid fluctuations of antimyasthenic drugs20 and phenytoin.10 In one patient with thrombotic thrombocy¬ topenic purpura who underwent two plasma-volume exchanges lasting for 4 to 5 hours, the procedures removed a significant amount of phenytoin, and additional doses of phenytoin were re¬ quired to control the seizures.10 In our patient, enough time did not pass dur¬ ing PP to permit calculation of a valid half-life, and the patient did not have the same magnitude of exchange (only 1.4 plasma-volume exchanges in 2 hours). The magnitude of loss of val¬ proic acid and phénobarbital by PP, on 2 different days, was relatively small, amounting to 7.1% to 7.2% and 37% to 39% of the daily dose of valproic acid and phénobarbital, respectively. It is known that PP removes plasma cause
factors most efficiently during the first hour of exchange,21 and our patient showed an appreciable drop of serum level of valproic acid, and to a lesser degree of phénobarbital, in the first 1000-mL exchange. Studies of antiepi¬ leptic drug concentrations in the hu¬ man epileptogenic brain have shown a good correlation between brain and plasma ratio in phenytoin, phénobar¬ bital, and valproic acid.22 24 In contrast to phénobarbital and phenytoin, how¬ ever, valproic acid concentration in the brain is thought not to reflect selective brain receptor binding, although it is related to the free valproic acid level in plasma.24 Goldberg and Todoroff25 found that valproic acid was not bound to brain or its subfraction; and there was little, if any, binding of valproic acid to either protein or phospholipid in the brain. In our patient, the effect of PP in removing valproic acid is sub¬ stantially higher than in removing
phénobarbital. In the past,
our
patient had experi-
enced seizures during 2 of her 35 plasma exchanges. This appeared to be greater than expected from her usual frequency of seizures. These two sei¬ zures, nonetheless, occurred exclu¬ sively on the first days of two different 3-day courses of PP. The patient ad¬ mitted that she had not been compli¬ ant to antiepileptic drugs before her hospitalization on these two occasions. Therefore, the combined effects of noncompliance and the subsequent rapid decrease of antiepileptic drug levels by PP may have resulted in the occur¬ rence of these two seizures. Conse¬ quently, we believe that it is important to keep the serum level of valproic acid or phénobarbital in therapeutic range before PP, but it may not be necessary to give additional doses of valproic acid or phénobarbital during PP. This study was supported in part by grant P5016308 from the National Institutes of Health, Be¬ thesda, Md. The authors wish to thank Patricia Melching for typing the manuscript.
References 1. Chow-Tung E, Lau AH, Vidyasagar D, John EG. Clearance of phenobarbital by peritoneal dialysis in a neonate. Clin Pharm. 1982;1:268-271. 2. Orr JM, Farrell K, Abbott FS, Ferguson S, Godolphin WJ. The effects of peritoneal dialysis on the single dose and steady state pharmacokinetics of valproic acid in a uremic epileptic child. Eur J Clin Pharmacol. 1983;24:387-390. 3. Zawada ET, Nappi J, Done G, Rollins D. Advances in the hemodialysis management of phenobarbital overdose. South Med J. 1983;76:6-8. 4. Mortenson PB, Hansen HE, Pedersen B, Hartmann-Andersen F, Husted SE. Acute valproate intoxication: biochemical investigations and hemodialysis treatment. Int J Clin Pharmacol Ther Toxicol. 1983;21:64-68. 5. Seybold ME. Plasmapheresis in myasthenia gravis. Ann NY Acad Sci. 1987;505:584-587. 6. Cornelio F, Peluchetti D, Mantegazza R, Sghirlanzoni A, Collarile C. The course of myasthenia gravis in patients treated with corticosteroids, azathioprine, and plasmapheresis. Ann NY Acad Sci. 1987;505:517-525. 7. Rodriguez M, Gomez MR, Howard FM Jr, Taylor WF. Myasthenia gravis in children: long\x=req-\ term follow-up. Ann Neurol. 1983;13:504-510. 8. Patten BM. Myasthenia gravis: review of diagnosis and management. Muscle Nerve. 1978; 1:190-205. 9. Tartara A, Mola M, Manni R, et al. EEG findings in 118 cases of myasthenia gravis. Rev
Electroencephalogr Neurophysiol Clin. 1982;
12:275-279. 10. Liu E, Rubenstein M. Phenytoin removal by plasmapheresis in thrombotic thrombocytopenic purpura. Clin Pharmacol Ther. 1982;31:762-765. 11. Fischer JH, Cloyd JC, Ehresman DJ, Lakatua DJ. Comparison of two ultrafiltration methods with equilibrium dialysis for the determination of free valproic acid plasma concentration. Epilepsia. 1983;24:112-113. 12. Gerson B. Quality control in therapeutic drug monitoring: intralaboratory precision and medical requirements. Ther Drug Monit. 1980; 2:225-232. 13. Gugler R, von Unruh GE. Clinical pharmacokinetics of valproic acid. Clin Pharmacokinet.
1980;5:67-83.
14. Bryson SM, Verma N, Scott PJ, Rubin PC. Pharmacokinetics of valproic acid in young and elderly subjects. Br J Clin Pharmacol.
1983;16:104-105. 15. Eadie MJ, Heazlewood V, McKauge L, Tyrer JH. Steady-state valproate pharmacokinetics
during long-term therapy. Clin Exp Neurol.
1983;19:183-191. 16. Wilensky AJ, Friel PN, Levy RH, Comfort CP, Kaluzny SP. Kinetics of phenobarbital in
normal subjects and epileptic patients. Eur J Clin Pharmacol. 1982;23:87-92. 17. Norris FH Jr, Colella J, McFarlin D. Effect of diphenylhydantoin on neuromuscular synapse.
Downloaded From: http://archneur.jamanetwork.com/ by a University of Manitoba User on 06/13/2015
Neurology. 1964;14:869-877. 18. Kaeser HE. Drug-induced myasthenic syndromes. Acta Neurol Scand Suppl. 1984;100:39-47. 19. Milonas J, Kountouris D, Scheer E. Myasthenic syndrome following long-term diphenylhydantoin therapy. Nervenarzt. 1983;54:437-438. 20. Shumak KH, Rock GA. Therapeutic plasma exchange. N Engl J Med. 1984;310:762-771. 21. Orkin JB, Berkman EM. Partial plasma exchange using albumin replacement. Blood. 1980;56:1055-1059. 22. Sherwin AL, Harvey CD, Leppik IE. Quantitation of antiepileptic drugs in human brain. In: Kellaway P, Petersen I, eds. Quantitative Analytic Studies in Epilepsy. New York, NY: Raven Press;
1976:171-182. 23. Vajda F, Williams FH, Davidson S, Falconer MA, Breckenridge A. Human brain, cerebrospinal fluid, and plasma concentrations of diphenylhydantoin and phenobarbital. Clin Pharmacol Ther. 1974;15:597-603. 24. Vajda FJE, Donnan GA, Phillips J, Bladin PF. Human brain, plasma, and cerebrospinal fluid concentration of sodium valproate after 72 hours of therapy. Neurology (NY). 1981;31:486\x=req-\ 487. 25. Goldberg MA, Todoroff T. Carbamazepine and valproic acid binding to brain. Neurology
(NY). 1979;29:577.
