Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2094 - 1/11
`
`

`
`Seizure 1992; 1 : 89-98
`
`The occurrence, management and outcome of
`anti epileptic drug side effects in 767 patients
`
`NEIL BUCHANAN
`
`Comprehensive Epilepsy Centre and the Department of Paediatrics, Westmead Hospital, Westmead, Sydney
`2145, Australia
`
`Correspondence to Professor Neil Buchanan at the ~bove address
`
`This study reports the nature of adverse drug reactions (ADR) occurring in 767 epilepsy clinic patients (adults
`and children), the drugs most commonly involved, how they were managed and the outcome of such manage(cid:173)
`ment. One hundred and thirty four patients were found to have 155 separate ADRs. The majority appeared to
`be pharmacodynamic in nature, although 21 were clearly pharmacokinetic in origin and four due to drug
`interactions. The antiepileptic drugs (AED) perceived to be causative, in order of frequency were phenytoin,
`sodium valproate, carbamazepine, clonazepam, barbiturates, vigabatrin and clobazam. Management most
`often involved withdrawing the offending drug(s), usually replacing them with another AED. Of the 155 ADRs,
`40.6% resolved totally, 27.7% showed a marked improvement, 16.1% improved, 14.8% did not change and one
`patient deteriorated.
`This study emphasizes the need to be vigilant for ADRs and demonstrates that their management is
`essentially clinical with some 85% of patients experiencing benefit.
`
`INTRODUCTION
`
`As for all persons taking medication on a long
`term basis, people with epilepsy may be subject
`to adverse drug reactions (ADR). It is generally
`believed that antiepileptic drug ADRs are
`common1
`. The results of a questionnaire
`survey of people with epilepsy showed that
`59% of the patients felt they were suffering
`some drug side effects, although this was likely
`to have been an overestimate. In this survey,
`the antiepileptic drugs (AEDs) most commonly
`incriminated were phenytoin (PHT, 34%), car(cid:173)
`bamazepine (CBZ, 30%), and sodium valproate
`(VPA, 23%)2 .
`Since the study by the Mario Negri Institute
`Group3
`, there has been little published data on
`antiepileptic drug ADRs. That study showed
`that 31% of the patients surveyed complained
`of ADRs with 30% of patients taking PHT
`reporting ADRs, 23% with phenobarbitone
`(PB), 15% with CBZ and 12% with VPA. Most
`studies have looked at ADRs from an epidemio(cid:173)
`logical viewpoint4
`, with a recent emphasis on
`cognitive side effects of AEDs5
`7
`-
`•
`The present study looks at the problem of
`
`antiepileptic drug ADRs from a different view(cid:173)
`point without epidemiological significance.
`The study looks at ADRs in a clinical practice
`and asks the following pragmatic questions:
`
`-What are they?
`-What AEDs are most commonly involved?
`-How were they managed?
`-What was the outcome of such management?
`
`The present study represents a clinical audit of
`the common problem 1 of drug side effects in
`767 patients with epilepsy.
`
`METHODS
`
`All data were obtained from the author's adult
`and paediatric epilepsy practice over a period
`of 10 years. From the commencement of the
`practice, which deals mainly with people with
`'difficult' epilepsy and has a referral base with
`a strong psychosocial emphasis, specific atten(cid:173)
`tion has been paid tq.t}le side effects of AEDs.
`At each consultation, patients or the parents of
`children were asked about drug side effects and
`if these were felt to be clinically valid, were
`
`1059-1311/92/020089+10 $08·00/0
`
`© 1992 Bailliere Tindall
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2094 - 2/11
`
`

`
`90
`
`N. Buchanan
`
`noted and coded. Using a coding system, it was
`possible to identify, from the clinical notes,
`which patients had drug side effects at the time
`of initial presentation and who developed them
`subsequently. Adverse drug reactions were
`defined as clinical symptoms or signs recog(cid:173)
`nized from past experience and published
`reports as being related to the drug(s) in
`question. Since most patients were regular
`clinic attenders, it was almost always possible
`to follow them up and monitor progress over
`time, especially if a medication was with(cid:173)
`drawn. Patients were generally seen every 1 to
`3 months, at least initially. The majority of
`patients had been receiving treatment for some
`time and thus it was not anticipated that
`initial ADRs such as rashes would be observed.
`Information on previous ADRs was not
`recorded for study purposes since they could
`not be verified.
`Information was obtained about age, sex,
`seizure type(s), the number of AEDs being
`taken at the time, which AEDs these were and
`which was likely to be causing the side effects.
`Additional information was obtained with
`respect to the side effects themselves, whether
`there was any specific pharmacological expla(cid:173)
`nation for the side effects, the value of blood
`level assessments in the diagnosis of drug side
`effects, action taken to alleviate the problem
`and the result thereof.
`Serum AED concentrations were only
`measured when it was felt to be clinically
`indicated.
`The outcome was graded as: (i) complete re(cid:173)
`covery, when the symptoms/signs regressed
`totally; (ii) marked improvement, consider(cid:173)
`able, but not total, regression; (iii) Improve(cid:173)
`ment, regression which was obviously incom(cid:173)
`plete, but which the patient found beneficial;
`(iv) no change and (v) deterioration, a deterio(cid:173)
`ration in symptoms/signs as a result of altering
`therapy to try and modify the side effect(s).
`Outcome was assessed by discussion between
`the patient and the author.
`The notes of 767 patients seen during the 10
`year period were reviewed. Twelve patients,
`coded as having side effects, were excluded
`from the study because they had been seen
`only once or twice and thus follow up was inad(cid:173)
`equate to assess the effect of suggested thera(cid:173)
`peutic changes to modify the side effects. In the
`patients reported, follow up was from 6 months
`to 10 years. There were 77 patients who
`initially presented to the clinic with side
`effects, 40 other patients developed side effects
`
`whilst attending the clinic and finally 17
`patients had side effects at the time of presen(cid:173)
`tation and developed additional/different side
`effects whilst attending the clinic.
`The present study was designed specifically
`as an audit of a clinical practice solely dealing
`with people with epilepsy. It is of no epidemio(cid:173)
`logical value because of the nature of the clini(cid:173)
`cal practice.
`
`RESULTS
`
`1. Patients with ADRs at initial presentation
`
`Of the 77 patients with ADRs at initial presen(cid:173)
`tation, 56 were adults (aged 14-59 years) and
`21 children (1-13 years) with 35 being male.
`Twenty-four of the patients had primary
`generalized epilepsy, 19 had complex partial
`seizures (CPS), nine simple partial seizures
`(SPS), eight secondarily generalized seizures,
`juvenile myoclonic epilepsy and
`the
`five
`remainder had miscellaneous seizure types.
`At the time of presentation, 28 of the 77
`patients were receiving monotherapy, 27 were
`receiving two AEDs, 20 three AEDs and two
`were taking four anticonvulsants. Thirty eight
`of the patients were receiving CBZ, 34 PHT, 29
`VPA, 19 barbiturates, 18 clonazepam (CZP),
`four diazepam (DZP), two stilthiame (SUL),
`one ethosuximide (ESM), one nitrazepam (NZ)
`and one clobazam (CLB).
`Of the 28 patients on monotherapy, 10 were
`taking PHT, six VPA, five CBZ, four barbitur(cid:173)
`ates, two CZP and one ESM. The most common
`combinations amongst the 27 patients taking
`two AEDs were PHT/CBZ (9), VPA/CBZ (4),
`CBZ/barbiturate
`(3)
`and PHT/VPA
`(3).
`Amongst the 20 patients on three AEDs the
`most common combinations were VP A/CBZ/
`CZP (4), VPA/PHT/CBZ (3), PHT/CZP/barbi(cid:173)
`turates (3) and PHT/VPA/benzodiazepine (3).
`The AEDs presumed to be causative of the
`side effects at the ti:tne of presentation are
`shown in Table 1. The majority were related to
`PHT (23), CZP (10), CBZ (9), barbiturates (9)
`and VPA (9). In three patients who were
`receiving three AEDs each, it was not possible
`to ascertain which drug(s) were causative and
`it was simply felt that these patients were
`over-medicated.
`Subsequent analysis will look in detail at
`those patients whose side effects were felt to be
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2094 - 3/11
`
`

`
`Antiepileptic drug side effects
`
`91
`
`Table 1: The AEDs presumed to be responsible for the side
`effects in 77 patients initially presenting with ADRs
`
`Presumed causative AED(s)
`
`Number of patients
`
`b. ADR with clonazepam (CZP) seen as the
`problem (n = 1 0)
`
`PHT
`CZP
`CBZ
`Barbiturates
`VPA
`Barbiturates/benzodiazepine
`PHT/CZP
`PHT/barbiturate
`VPA/PHT
`Two benzodiazepines
`PHT/DZP
`PHT/CLB
`VPA/CBZ
`PHT/CBZ/barbiturate*
`VPA/CBZ
`PB/VPA/acetazolmide*
`VP A/SUL/barbiturate/CZP*
`PB = Phenobarbitone.
`* In these 3 patients, it was not possible to define which
`drug might be causative.
`
`23
`10
`9
`8
`8
`4
`3
`3
`2
`1
`1
`1
`, 1
`1
`1
`1
`1
`
`due to PHT, CZP, CBZ, barbiturates, VPA and
`the remaining miscellaneous combinations.
`
`Signs and symptoms included drowsiness (7),
`aggression (4), irritability (3), feeling 'doped
`up' (3), dribbling/drooling in two children and
`hyperactivity in one child. No serum clonaze(cid:173)
`pam concentrations were measured.
`Clonazepam was withdrawn in nine of the 10
`patients, being replaced with VP A or CBZ in
`two patients each. Of these nine patients, one
`with profound drowsiness went into status epi(cid:173)
`lepticus requiring ICU management, but came
`off CZP and is now well controlled with fewer
`side effects. A second patient was fine for 3
`years after coming off CZP, but then had recur(cid:173)
`rent seizures and was represcribed CZP with a
`return of his side effects. One patient had a
`single seizure 3 months after beginning slowly
`to withdraw CZP and chose to go back onto the
`drug: her side effects returned. One patient re(cid:173)
`covered fully from CZP side effects, four
`showed
`a marked
`improvement,
`three
`improved and two showed no change.
`
`a. ADR with phenytoin (PHT) seen as the
`problem (n = 23)
`
`c. ADR with carbamazepine (CBZ) seen as the
`9)
`problem (n
`
`Signs and symptoms included drowsiness and
`being 'slowed down' (19), cosmetic problems
`such as gum swelling, hirsutism and acne (10),
`slow, slurred speech (9), a deteriorating
`memory (8), acute intoxication (3), recurrent
`episodes of acute
`intoxication (2), weight
`increase (1) and break through bleeding with
`the oral contraceptive pill (1).
`Serum phenytoin levels were only of value
`on six of the 23 occasions on which they were
`measured, with levels being above the rec(cid:173)
`ommended therapeutic range (40-80 1-1mol/l).
`In five of these six patients, the diagnosis of
`phenytoin intoxication was clinically evident
`and blood levels were solely confirmatory.
`In 22 of 23 patients, PHT was withdrawn
`and was usually replaced with CBZ or VP A
`with no difficulties. In one patient, as the PHT
`was reduced there was an increase in seizure
`frequency and it was re-instated. With respect
`to the response to changing therapy, nine
`patients recovered completely from their ADR,
`nine showed a marked improvement, four some
`improvement and one patient, who had with(cid:173)
`drawal fits, deteriorated. The smallest change
`was seen in those patients whose predominant
`complaint was of a deteriorating memory.
`
`symptoms of carbamazepine
`Signs and
`included drowsiness (5), aggression (2), dip(cid:173)
`lopia (2), nystagmus (2), irritability (1), person(cid:173)
`ality change (1) and there were two acutely
`intoxicated patients (one therapeutically and
`the other, a suicidal gesture). -
`Serum CBZ concentrations, using a thera(cid:173)
`peutic range of 15-401-Lmol/1, were measured in
`seven of the nine patients receiving CBZ. They
`were useful on six of these occasions. In four of
`the nine patients, CBZ was withdrawn, in
`three the dose was reduced and in one patient
`no action was taken. In the
`intoxicated
`patients, CBZ was stopped for 48 h and then
`recommenced. In five patients there was com(cid:173)
`plete recovery from their ADRs, a marked im(cid:173)
`provement was seen in three cases and in one
`person where no action was taken since they
`had complete seizure control, their side effects,
`which were quite mild, persisted.
`
`d. ADR with barbiturates seen as the problem
`(n= 8)
`
`Marked drowsiness was the predominant ADR
`seen in four patients. Hyperactivity/irritability
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2094 - 4/11
`
`

`
`92
`
`(2), tantrums (1), aggression (1) and feeling
`'slowed down' (1) were also noted. Serum phe(cid:173)
`nobarbitone concentrations were not measured
`in any of these patients since it was felt, clini(cid:173)
`cally, that this would not assist in manage(cid:173)
`ment.
`The barbiturates were withdrawn in all
`patients and were replaced by CBZ or VP A in
`five cases. Three patients recovered fully from
`their ADR, four showed a marked improve(cid:173)
`ment and the other patient improved a little.
`
`e. ADR with sodium valproate .(VPA) as the
`problem (n = 8)
`
`The predominant problems were weight gain
`(3), tremor (3), hyperactivity (2), drowsiness (1)
`and hair loss (1). The magnitude of the weight
`gain was such that over 1 to 2 years these three
`patients reported gains from 60 to 74kg, 51 to
`63 kg and 58 to 67 kg, respectively. Serum val(cid:173)
`proate concentrations were not measured.
`In two patients VP A was withdrawn, in
`three patients with juvenile myoclonic epilepsy
`and absolute seizure control no action was
`taken, in two VP A dosage was reduced and
`strict dietary advice was provided for the final
`patient.
`In terms of response to these actions, a com(cid:173)
`plete recovery was seen in one patient, a
`marked improvement in two, mild improve(cid:173)
`ment in one and no change in four patients.
`The latter included the three people with
`weight gain. Tremor improved greatly with
`VP A dose reduction.
`
`f. ADR in the remaining 19 patients
`
`In this group of 19 patients, it was less easy to
`be certain of the drugs causing the ADRs. On
`clinical grounds, it was felt that the following
`combinations were responsible: barbiturate/
`benzodiazepine (4 patients), PHT/CZP (3),
`PHT/barbiturate (3), VPA/PHT (2), two benzo(cid:173)
`diazepines (1), PHT/DZP (1), PHT/CLB (1),
`VPA/CBZ (1) with three patients falling into
`the )ust too many drugs' category PHT/CZP/
`barbiturate, barbiturate/VP A/acetazolamide,
`VP A/SUL/CZP/barbiturate.
`The predominant side effects were drowsi(cid:173)
`ness (17 patients), slurred speech (6), falling
`asleep at work (6), feeling 'slowed down' (5),
`ataxia (4), deteriorating memory (4), nystag(cid:173)
`mus (2) and one patient each complained of
`
`N.Buchanan
`
`learning problems,
`depression,
`diplopia,
`weight increase, cosmetic problems, drooling
`and aggression.
`Serum AED concentrations were measured
`in 10 of 19 patients, but were only useful twice.
`On both occasions, the patients were felt to
`have an ADR owing to a VPA/PHT combi(cid:173)
`nation and in both cases the free PHT fraction
`was increased accounting for the symptomato(cid:173)
`logy of PHT intoxication. One patient had a
`total PHT of 60 1-Lmol/l and a free concentration
`of 16% and the other a total PHT of 80 j..Lmol/1
`and a free concentration of 17%.
`The drugs felt to be causing the ADR were
`withdrawn in all 19 patients and in six
`patients, CBZ or VP A were added. Two
`patients had withdrawal seizures, but were
`successfully withdrawn from the drugs in
`question. In addition, three patients had the
`doses of their other medications reduced. This
`led to be a complete recovery from ADR symp(cid:173)
`tomatology in three patients, a marked im(cid:173)
`provement in 11, improvement in four and no
`change in one patient.
`
`2. Patients developing ADRs whilst attending
`the clinic
`
`Forty patients developed ADRs whilst attend(cid:173)
`ing the clinic, of whom 24 were adults (14-57
`years old) and 16 children (2-13 years) with 23
`being female. Twelve had primary generalized
`epilepsy, 11 CPS, six tonic seizures, three
`secondarily generalized seizures, three myoclo(cid:173)
`nic seizures, two each had SPS, benign focal
`epilepsy (BFE) of childhood and juvenile myo(cid:173)
`clonic epilepsy and one patient had the Len(cid:173)
`nox-Gastaut syndrome.
`At the time that the ADRs occurred, 13
`patients were on monotherapy, 16 taking two
`AEDs, seven taking three AEDs and four
`patients were receiving four AEDs. Twenty-six
`of the patients were taking VPA, 21 CBZ, 14
`PHT, seven vigabatrin (GVG), six CLB, three
`barbiturates, two ESM and one each CZP, DZP
`and NZ.
`Of the 13 patients on monotherapy, six were
`taking VP A, five CBZ, one PHT and one was
`taking ESM. Amongst the 16 people taking
`two AEDs the combinations were-CBZ/VP A
`(5), PHT/VPA (4), VPA/CLB (2), VPA/GVG (1)
`and VPA/NZ .. (1). Among the seven patients
`taking three AEDs there were seven different
`combinations: PHT/VP A/CLB, PHT/VP AICBZ,
`VPA/PHT/barbiturate, PHT/VPA/GVG, PHT/
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2094 - 5/11
`
`

`
`Antiepileptic drug side effects
`
`93
`
`CLB/GVG, CBZ/GVG/barbiturate and PHT/
`CLB/GVG. Finally, the four patients received
`four AEDs in the following three combinations:
`VPA/CBZ/PHT/GVG (2), VPA/CBZ/ESM/bar(cid:173)
`biturate (1) and PHT/CBZ/CLB/DZP (1).
`The AEDs presumed to be causative of the
`ADRs are shown in Table 2. The majority were
`related to VPA (14), CBZ (7), vigabatrin (7)
`and PHT (5). One patient, a 3 year old child
`who had Lennox-Gastaut syndrome and pre(cid:173)
`sented in status epilepticus, required a combi(cid:173)
`nation of these four drugs to control her
`seizures but, not surprisingly, developed side
`inco-ordination and
`effects of drowsiness,
`marked ataxia.
`
`Table 2: The AEDs presumed to be responsible tor the side
`effects which developed in the 40 regular clinic attenders
`
`Presumed causative AEDs
`
`Number of patients
`
`VPA
`CBZ
`GVG
`PHT
`CLB
`NZ
`ESM
`CZP
`VPA/PHT
`PHT/CBZ/CLB/DZP
`
`14
`7
`7
`5
`2
`1
`1
`1
`1
`1
`
`a. ADR with sodium valproate (VPA) as the
`14)
`problem (n
`
`Signs and symptoms included weight gain in
`seven patients (6 female). Weight gains in each
`patient, assessed over 9-12 months, were 56-
`66kg, 51-60kg, 63-77kg, 68-79kg, 36-
`48 kg, 43-52 kg and 54-62 kg. Other side
`effects
`included significant hair
`loss
`(3),
`increased seizure frequency (1), tremor (1),
`drooling (1), rash (1) and oedema (1).
`VP A blood levels were not measured. The
`response to these ADRs was such that nothing
`was done in five patients since they had com(cid:173)
`plete seizure control, VPA was withdrawn in
`three patients and VP A dose was reduced in
`three patients. In the remaining patients, in
`one case all other drugs were withdrawn, in
`another patient VP A was withdrawn, with a
`resultant marked increase in seizure frequency
`and so was reinstituted, and in the final
`patient, a very low calorie diet was commenced
`with resultant weight loss (77 to 70 kg).
`
`In only two patients was there a complete
`recovery from the ADR; the patient with
`tremor. recovered with a dose reduction from
`2500 mg to 1500 mg/day and the patient who
`had developed a bullous rash 3 days after VP A
`was commenced, recovered when the drug was
`withdrawn. Five patients showed improve(cid:173)
`ment and seven no change. This latter group
`contained six of the seven patients who had
`gained weight.
`
`b. ADR with carbamazepine (CBZ) seen as the
`problem (n = 7)
`
`Signs and symptoms were those of acute CBZ
`intoxication in five patients; two as a result of
`interactions with danazol/erythromycin, two
`patients accidentally overdosed themselves
`and the final patient was therapeutically over(cid:173)
`dosed. Two other patients complained of pro(cid:173)
`found fatigue.
`Serum CBZ levels were useful in all cases
`where measured as follows: the patient with
`the danazol interaction (CBZ 711-Lmol/l), thera(cid:173)
`peutic overdose (39 j.Lmolll), accidental over(cid:173)
`doses (45 and 59!-Lmol/l), erythromycin interac(cid:173)
`tion (661-1mol!l) and in the two patients with
`fatigue ( 48 and 52 1-1mol/l).
`In response to these ADRs, the CBZ dose was
`reduced in four cases, the interacting drug was
`ceased in the two affected patients and no
`action was taken in one case since there was
`complete seizure control. Six patients re(cid:173)
`covered fully and in the patient in whom no
`action was taken, there was no change.
`
`c. ADR with vigabatrin seen as the problem (n
`=7)
`
`The predominant problem was the develop(cid:173)
`ment of a marked deterioration of behaviour or
`an acute psychosis in four patients, all of whom
`had past histories of behaviour disturbance/
`mental illness. One patient became very
`drowsy, another showed an increase in seizure
`frequency and the final patient had an increase
`in weight from 48-54 kg over 3 months. Blood
`levels of GVG were not measured.
`Vigabatrin W~:§. withdrawn in all cases with
`complete recovery in all patients except the
`person whose weight increased, in whom there
`was no change.
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2094 - 6/11
`
`

`
`94
`
`d. ADR with phenytoin (PHT) seen as the
`problem (n = 5)
`
`Side effects included marked drowsiness/fati(cid:173)
`gue (3), dizziness (1), diplopia (1) and gross
`gum hypertrophy (1).
`Serum PHT measurement was useful in four
`instances: in one patient who complained of
`fatigue, the dose had been increased from 360
`to 400 mg/day with an increase in the serum
`concentration from 46 to 132 1-1mol/l. In a
`further patient, in association with a severe
`viral
`infection,
`the serum concentration
`increased from 65 to 204 j-Lrnol/1. The patient
`with dizziness and diplopia was found to have a
`PHT concentration of 1081-1mol/l for no obvious
`reason and, finally, one patient had a PHT/
`cimetidine interaction with a serum PHT of
`1461-Lmol/1.
`PHT was withdrawn in three patients,
`dosage was reduced in one and cimetidine was
`changed to ranitidine in the final case. All
`patients recovered completely.
`
`e. ADRs in the remaining seven patients
`
`The causative agents were believed to be CLB
`(2), NZ (1), ESM (1), CZP (1), VPA/PHT (1) and
`PHT/CBZ/CLB/DZP (1). Side effects included
`in
`two patients, ataxia
`(2),
`drowsiness
`increased seizure frequency (2), and one each of
`hyperactivity, immobility, urinary frequency
`and pedal oedema.
`Blood levels were only measured in the VP A/
`PHT patient in whom the serum PHT concen(cid:173)
`tration was 138 1-1mol/l with a free PHT concen(cid:173)
`tration of 16% and serum VPA 280 j-Lmol/1. The
`PHT dose was reduced with complete recovery
`from the drowsiness.
`The offending drug was withdrawn in five
`cases, PHT dose was reduced in one and in the
`patient on four drugs, they were all withdrawn
`and replaced with VP A. Complete recovery
`was seen in six cases and a marked improve(cid:173)
`ment in the final patient.
`
`3. Patients initially presenting with an ADR and
`subsequently developing further ADRs
`
`There were 17 patients in this group: two chil(cid:173)
`dren and 15 adults of whom 10 were female.
`Five patients had primary generalized
`seizures, four had each CPS, secondarily gener(cid:173)
`alized seizures and juvenile myoclonic epi-
`
`N.Buchanan
`
`lepsy, two with myoclonus and one with tonic
`seizures. Of these 17 patients who presented
`with an ADR, 14 developed an ADR on a
`further occasion, 2 on two further occasions
`and one patient developed three subsequent
`ADRs.
`Because of the complexity of these data it is
`presented in tabular form (Table 3). In many
`respects the data are similar to those of the
`previous two groups, but demonstrate the pro(cid:173)
`pensity of some individuals to develop side
`effects. Again PHT was the most common AED
`involved (14 patients) followed by VPA and
`CBZ in seven cases each. Serum AED measure(cid:173)
`ments were useful in 15 of the 18 instances in
`which they were carried out. Management of
`the ADRs is detailed in Table 3 and in terms of
`outcome of the 38 ADRs observed in these
`patients, complete recovery occurred in 15
`cases, marked improvement in nine, improve(cid:173)
`ment in eight and no change in six patients. Of
`interest is the finding that three of the patients
`who showed an improvement in their ADRs
`unfortunately had more seizures and were thus
`no better off. Again, the no change group rep(cid:173)
`resented those patients who had gained weight
`whilst taking VP A.
`
`OVERALL OBSERVATIONS
`
`The 134 patients reported here suffered 155
`separate ADRs. In 50 cases the patient was
`receiving monotherapy, 62 were taking two
`AEDs, 35 three AEDs and 8 four AEDs. The
`AEDs perceived as causative in order of fre(cid:173)
`quency were PHT, VPA, CBZ, CZP, barbitur(cid:173)
`ates, GVG and CLB. In 21 .instances, the ADR
`was pharmacokinetic in origin and was due to
`overdosage (accidental or therapeutic) or an
`associated
`intercurrent
`illness.
`In
`five
`instances there was PHT intoxication when
`used in association with VP A leading to ele(cid:173)
`vated free PHT concentrations. There were
`four further drug interactions noted: PHT/oral
`contraceptive, CBZ/danazol, PHT/cimetidine
`and CBZ/erythromycin. The remaining 125
`ADRs were felt to be predominantly pharmaco(cid:173)
`dynamic in nature.
`In terms of management, in 105 instances
`the presumed offending drug was withdrawn,
`most often being replaced by VP A or CBZ. In
`24 instances,dosage was reduced, there was no
`change made on 13 occasions and dietary
`advice was given on six occasions. In three
`patients who were acutely intoxicated, the
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2094 - 7/11
`
`

`
`Antiepileptic drug side effects
`
`drug was stopped for 48-72 h and then recom(cid:173)
`menced, whilst in the four patients who suf(cid:173)
`fered drug interactions, the offending drug was
`withdrawn.
`Finally, following withdrawal of the drugs
`causing the 155 ADRs, improvement was com(cid:173)
`plete in 63 cases (40.6%), marked in 43 (27.7%),
`slight in 25 (16.1 %) and unchanged in 23
`(14.8%) (of which 15 represented weight gain
`associated with sodium valproate) and in only
`one instance, the patient deteriorated.
`
`DISCUSSION
`
`The present clinical audit set out to investigate
`four questions about ADRs in an epilepsy clinic
`population. Firstly, in 80.6% of instances it
`was felt that the ADRs were pharmacodynamic
`in nature. In other words, the minority were
`pharmacokinetic, although there was value in
`measuring serum AED concentrations on a
`limited number of occasions, especially with
`CBZ and PHT. Of note are the five patients in
`whom the combination ofPHT/VPA led to PHT
`intoxication associated with protein binding
`displacement leading to elevated free concen(cid:173)
`10
`trations of PHT8
`. Whilst the frequency of
`-
`clinically significant changes in PHT -protein
`binding remains uncertain, the finding in
`these five patients suggests that this problem
`should be borne in mind and free PHT concen(cid:173)
`trations measured if clinically indicated. Past
`experience suggests VP A to be the commonest
`drug involved in this interaction, but salicy(cid:173)
`lates and diazepam may produce the same
`effect11
`. Secondly, the AEDs presumed to be
`involved in the ADRs, in order of frequency in
`this particular population were PHT, VP A,
`CBZ, barbiturates, GVG and CLB. This gener(cid:173)
`ally coincides with the questionnaire responses
`mentioned earlier2
`, but differs from the Mario
`Negri observations3
`. This difference probably
`reflects changed prescribing habits in the
`decade since those data were collected.
`Thirdly, as far as management of ADRs are
`concerned, in 67.7% of instances the presumed
`offending drug(s) were withdrawn, frequently
`being replaced by an alternative AED, usually
`CBZ or VP A. Whilst this led to an increase in
`seizure frequency in six patients, in five
`instances this settled and was acceptable to the
`patient. In the final individual, it proved very
`difficult
`to
`re-establish
`seizure
`control,
`although the original ADR improved. Overall,
`however, this patient was worse off.
`
`95
`
`This leads to the fourth question, namely the
`outcome of ADR management. To the author's
`knowledge, at least in the recent past, this has
`not been formally studied. It is encouraging
`that in the patient's view, complete recovery
`occurred in 40.6% of ADRs, a marked improve(cid:173)
`ment occurred in 27. 7%, some improvement
`was seen in 16.1% of ADRs and, as mentioned
`above, one patient deteriorated. In those
`patients in whom no change occurred, it might
`be assumed that the drug(s) felt to be causing
`the problem, may not, in fact, have contributed
`to the reported symptoms.
`Thus there was complete recovery or a
`marked improvement in 68.3% of instances.
`This appears encouraging and reinforces the
`need to be vigilant for AED side effects and
`suggests the majority can be successfully
`managed with benefit to the patient.
`Of some interest are the four patients who
`suffered ADRs as a result of drug interactions,
`reinforcing the need to provide patients taking
`medication with information on drugs to avoid,
`as well as ongoing medical education on this
`problem. There were two PHT interactions,
`both well documented12
`, one with cimetidine,
`leading to PHT intoxication, and the other in
`which PHT enhanced the biotransformation of
`an oral contraceptive leading to breakthrough
`bleeding. The CBZ interactions included the
`well-known interaction with erythromycin13
`and a less common interaction with danazol,
`which had been prescribed for endometriosis14
`.
`Also worth mentioning is the matter of
`weight-gain associated with VP A. The fre(cid:173)
`quency of this ADR is unclear, with reports
`varying from 17%15 to 40%16 or 50%17
`. The
`mechanism of weight gain with VP A remains
`unclear. In the present study, there were 19
`patients who exhibited significant weight gain
`of the magnitude outlined previously. Sixteen
`of these were associated with VPA and one
`each with PHT, CLB and GVG. In only two of
`the patients was weight loss achieved by intro(cid:173)
`ducing a very low calorie diet (Modifast(cid:173)
`Wander) for 6 weeks. Both patients have main(cid:173)
`tained their weight loss.
`The development of psychosis or a deterio(cid:173)
`ration in behaviour in persons with a past
`history of mental illness or behaviour disturb(cid:173)
`ances is also of interest. This confirms previous
`observations 18
`19 and suggests
`that GVG
`'
`should be used with care, or perhaps avoided,
`in such patie;{t"s. It does not preclude the use of
`GVG in mentally retarded patients20
`, although
`aggression may be observed.
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2094 - 8/11
`
`

`
`96
`
`N.Buchanan
`
`Table 3: Data on the 17 patients who intially presented with drug side effects who, after those had been managed, developed
`other side effects. The initial entry for each patient is the ADA with which they first presented.
`
`Patient AEDs being taken Presumed
`number at the time
`causative AED
`
`Side effects
`
`Serum AED
`concentration
`
`Management of ADR Outcome
`
`1.
`
`Barbiturate VP A
`
`Barbiturate
`
`Drowsiness
`
`VPA,PHT
`
`PHT
`
`Drowsy, ataxia,
`nystagmus
`
`PHT 108
`
`VPA, PHT, CBZ
`
`PHT
`
`Drowsiness
`
`Total PHT 60,
`but free = 20%
`
`Barbiturate
`withdrawn, VPA dose
`increased
`As serum
`concentration had
`increased from 23 to
`108 with a dose
`increase of PHT from
`150 to 200 mg/day, the
`dose was reduced
`PHT withdrawn and
`CLB added
`
`VPA,PHT
`
`PHT
`
`2.
`
`PHT, CBZ, VPA,
`Barbiturate
`CBZ, VPA
`CBZ, VPA, CLB
`
`3.
`
`PHT, CBZ, CZP
`
`Uncertain
`
`CBZ
`CLB
`
`CZP
`
`PHT, CBZ, CLB
`
`PHT
`
`CBZ,CLB
`
`CBZ
`
`4.
`
`CBZ, VPA, PHT
`
`PHT
`
`CBZ,PHT
`
`PHT
`
`Drowsy and
`dribbling
`
`Drowsy and very
`slow
`Dizziness, diplopia
`Depression
`
`132
`
`PHT dose reduced
`
`PHT and barbiturate
`withdrawn
`CBZ dose reduced
`CLB withdrawn
`
`Slow, slurred speech PHT 63, CBZ 33 CZP replaced with
`CLB
`PHT dose reduced
`
`PHT 108
`
`Classic PHT
`intoxication with a
`viral illness
`Ataxia, nystagmus CBZ 114
`
`CBZ ceased for 2 days R
`and recommenced at a
`lower dose
`
`PHT withdrawn
`
`PHT87
`
`PHT dose decreased
`
`but seizures
`increased
`R
`
`Gum hypertrophy
`and hirsutism
`
`Viral infection,
`developed ataxia
`and nystagmus
`
`R
`
`but seizures
`increased
`I
`but seizures
`increased
`
`R
`R
`
`I
`R
`
`5.
`
`CBZ, SUL, CZP
`
`CZP
`
`Very drowsy
`
`CBZ33
`
`CBZ, VPA
`
`6.
`
`PHT
`
`VPA
`
`PHT
`
`CBZ
`
`CBZ
`
`7.
`
`PHT, VPA, CBZ