Neurology 1996 Krauss 1534 9

DOI 101212WNL46615341996461534-1539 Neurology

G L Krauss J Brandt M Campbell et al national survey of neurologists and obstetricians

Antiepileptic medication and oral contraceptive interactions A

This information is current as of June 1 1996

httpwwwneurologyorgcontent4661534fullhtmlthe World Wide Web at

The online version of this article along with updated information and services is located on

Inc All rights reserved Print ISSN 0028-3878 Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright Copyright 1996 by Advanstar Communications

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

Antiepileptic medication and oral contraceptive interactions

A national survey of neurologists and obstetricians GL Krauss MD J Brandt PhD M Campbell RN C Plate BS and M Summerfield

Article abstract-Hepatic enzyme-inducing antiepileptic drugs (AEDs) lower oral contraceptive (OC) sex hormone levels approximately 40 and increase the risk of unplanned pregnancies in women with epilepsy AEDs also increase the risk of birth defects in offspring of women with epilepsy We performed a national survey to determine obstetriciansrsquo and neurologistsrsquo knowledge of OC and AED interactions and the risk of birth defects for women with epilepsy taking AEDs We received responses to a mailed questionnaire from 160 of 1000 neurologists (16) and 147 of 1000 obstetricians (15) from 47 states Practice demographics and ages of responders were typical for US neurologists and obstetricians Ninety-one percent of neurologists and 75 of obstetricians said they treat women with epilepsy of child-bearing age Only 4 of the neurologists and none of the obstetricians however knew the effects of the six most common AEDs on OCs even though 27 of neurologists and 21 of obstetricians reported OC failures in their patients taking AEDs Although increasing OC doses can compensate for insufficient OC sex hormone levels due to AEDs most physicians do not increase the doses Even though the risk of birth defects for the offspring of women with epilepsy is 4 to 6 up from the background level of 296 44 of neurologists thought the risk was lower (0 to 3) and some of the respondents guessed that it was as high as 50 Many neurologists and obstetricians do not have accurate information to counsel women with epilepsy properly about their contraceptive and pregnancy choices NEUROLOGY 199646 1534-1539

Antiepilepsy drugs (AEDs) are associated with two major reproductive risks for women with epilepsy of child-bearing age (1) an increased risk of unplanned pregnancies due to hepatic enzyme-inducing AEDs lowering sex hormone levels of oral contraceptives (OCs)rsquo (2) an increased risk of birth defect^^^ We conducted a national survey of obstetricians and neurologists to determine whether they were aware that enzyme-inducing AEDs decrease the effectiveness of OCs and whether their patients taking AEDs had become accidentally pregnant while taking OCs We also determined whether physicians accurately counsel women about their increased risk of birth defects due to AEDs

During the past 2 years we saw five women at Johns Hopkins Hospital with hormonal contraceptive failures while taking AEDs These unexpected pregnancies caused major social and emotional dis- tress one woman also had a brain tumor one had mental retardation and epilepsy surgery was can- celed for a third We became concerned that neurologists and obstetricians may not be aware of this drug interaction and that women with epilepsy taking AEDs are not being warned of their risk for accidental pregnancies This is especially important because women with epilepsy are at an approximately 4 to 6 risk of birth defects (2-3X background) in their

offspring3 with even greater risks for high-risk group^^^ Physicians have been held legally responsi- ble for a ldquowrongful liferdquo when a woman was not in- formed of specific pregnancy risks associated with her AEDslsquoj

Over 80 women have been reported in the United States and United Kingdom with accidental pregnancies due to enzyme-inducing AEDs interfering with O C S ~ gt ~ - ~ The enzyme-inducing AEDs associated with OC failures are carbamazepine phenytoin phenobarbital primidone and ethosuximide All reduce OC estradiol levels approximately 40 and may also decrease free progestin levels by increasing sex hormone-binding globulins12 Enzyme-inducing AEDs increase the expected OC failure ratelo of 01 to 3 by severalfold In one small study 7 of epilepsy patients taking OCs with AEDs had OC fai1ureslrsquo The nonhepatic enzyme-inducing AEDs -valproic acid and gabapentin- do not decrease OC levels and have not been associated with accidental pregnancies

Physicians can compensate for the influence of enzyme-inducing AEDs on OC sex hormone levels by increasing OC estradiol doses from 20 to 35 pg to 50 kg especially if there is irregular or breakthrough menstrual bleeding on low-dose OCs27rsquo2-15 Because most reported OC failures have been with 50- to

From the Departments nf Neurology IDr Krauss Ms Campbell Ms Plate and Mr Summerfield and Psychiatry (Dr Brandt) Johns Hopkins University Baltimore MD Received July 11 1995 Accepted in final form October 2 1995 Address correspondence and reprint requests to Dr Gregory L Krauss Johns Hopkins Hospital Meyer 2-147 600 N Wolfe Street Baltimore MD 21287-7247

1534 Copyright S 1996 by the American Academy of Neurology

Table 1 Physician respondents

Survey neurologists National neurologists Survey obstetricians National obstetricians (1994) (1991-1992) (1994) (1992-1993)

-

Total responders

Mean age (range) (yr)

Work settings ()

Solo

Group Community Hospital

Med school

Research other

Board-Certified ()

Year of MD

160 (16)

453 (30-75) 445

316 232

513 373

33 154

112 203

26 38

925 776

1943-1990

80-kg doses these steps will not prevent all OC failures but are likely to significantly reduce their frequency

Physicians caring for women with epilepsy may not be aware of these issues I t is particularly important to determine whether women with epilepsy are receiving accurate information about contraception because they are at increased risk for pregnancy complications and birth defects in their ~ffspr ing~

Methods Suruey We mailed surveys to a sample of 1000 neurologists and 1000 obstetriciadgynecologists randomly selected from a geographically stratified Ameri- can Medical Association ( M A ) master list of 10034 licensed neurologists and 33208 licensed obstetricians in the United States This is a comprehensive list of nearly all active specialists The random selection of physicians was performed by sequentially selecting physicians from the AMA database listed by zip codes This resulted in a sample that included all states but also represented the relative population of physicians in various regions We asked the physicians to respond to a brief anonymous knowledge questionnaire without consulting texts or other sources Twenty questions probed physiciansrsquo practice demographics familiarity with AED and OC interactions knowledge of interactions between specific AEDs and OCs and whether physicians adjusted OC doses for patients taking AEDs To assess physiciansrsquo general awareness of reproductive health issues in women with epilepsy we also asked them to estimate the risk of birth defects in pregnancies of women with epilepsy and whether they counsel women with epilepsy against pregnancy

The following questions were asked (1) Are you currently treating any women of child-bearing age with epilepsy (2) In counseling a pregnant woman with good seizure control on a single anticonvulsant what would you tell her the risk is of her having a baby with a birth defect (3) Do you generally recommend that women taking a single anticonvulsant not become pregnant because of this risk (4) How familiar are you with the interactions between oral contraceptives and antiepileptic medications (5-12) To the best of your knowledge do the following antiepileptic medications interfere with levels of oral contraceptives phenytoin (Dilantin) carbamazepine (Tegretol) valproic acid (Depakote) phenobarbital primi-

Survey questions

147 (147)

424 (28-74) 420

341

507

58

58

876

66

12

36 33

925 71 1943-1992

done (Mysoline) ethosuximide (Zarontin) gabapentin (Neurontin) felbamate (Felbatol) (13) Do you have your patients adjust their oral contraceptive dose if they are taking antiepileptic medication (14) If ldquoYesrdquo what adjust- ment (15) Have any of your patients on antiepileptic medication accidentally become pregnant while taking oral contraceptives

Demographic questions were (16) age (17) year of MD (18) board certification and (19) primary work setting a general comments section (20) was provided

Questionnaires returned within 6 months of the mailing were included Responses were coded and analyzed using the SPSS for Windows statistical program Data for categorical responses were tabulated and continuous data are summarized by means ranges and variance One post-hoc response to the questionnaire was added Some neurologists did not adjust OC doses if their patients were taking antiepileptic medications (ldquonordquo to question 131 but they did refer patients to obstetricians for OC adjustments (comment under 14) Therefore ldquoyesrdquo ldquonordquo and ldquoreferrdquo were possible responses for neurologists to question 13

We tested six specific hypotheses separately for neurologists and obstetricians using nonparametric (chi-square) and parametric analysis (independent t test and bivariate correlation) We did not correct for multiple comparisons to avoid type I1 statistical error The hypotheses were that obstetricians and neurologists are more likely to adjust OC doses for patients taking AEDs if (1) they knew that enzyme-inducing AEDs interfere with OC levels (2) they were familiar with interactions between AEDs and OCs but did not necessarily recognize specific drug interactions (3) they had previously had patients on AEDs become accidentally pregnant while taking OCs (4) physicians with recent medical training would be most knowledgeable about OC-AED interactions (5) physiciansrsquo estimates of the risks of birth defects would influence whether they recommend against pregnancy and (6) physiciansrsquo knowledgeable about OC-AED interactions would provide more accurate estimates of the risks of birth defects than would physicians who were not aware of OC-AED interactions We also compared obstetriciansrsquo and neurologistsrsquo familiarity with and knowledge of OC-AED interactions and their estimates of the risk of malformations in offspring of women with epilepsy taking AEDs

Statisticat analysis

June 1996 NEUROLOGY 46 1535

Figure 1 Physiciansrsquo knowledge of OC interactions with six common AEDs (new AEDs excluded) AEDs carbamazepine phenytoin valproic acid phenobarbital primidone ethosuximide

Physicians were asked which of eight AEDs interfered with OCs Their knowledge of which AEDs interfere with OCs was estimated however using only six of eight AEDs (phenytoin carbamazepine valproic acid phenobarbital primidone and ethosuximide ) Gabapentin and felbamate were not included because they had been recently released at the time of the survey and were not recognized by many neurologists and most obstetricians Correct responses were tha t phenytoin carbamazepine phenobarbital primidone and ethosuximide interfere with OCs whereas valproic acid does notlsquo27-9

Results Characteristics of survey respondents A total of 160 neurologists (16) and 147 obstetricians (147) responded to the questionnaire (table 1) The ages work setting and geographic distribution for responders were representative of neurologists and obstetricians national- l ~ rsquo ~ rdquo The responders were from 47 states The largest numbers were in office-based practices with medical school graduation in the 1960s through 1980s

Most neurologists (91) and obstetricians (7596) said they treat women with epilepsy of child-bearing age Most neurologists (75) and most obstetricians (59) believed they were moderately or very familiar with the interactions between AEDs and OCs Physiciansrsquo actual knowledge of OC- AED interactions however was extremely poor When physicians were scored on their knowledge of whether six common AEDs interfered with OCs (phenytoin carbamazepine valproic acid phenobarbital primidone and ethosuximide) only 4 of the neurologists and none of the obstetricians were correct for all six AEDs (figure 1) The two new AEDs gabapentin and felbamate were excluded because they were not recognized by most physicians The average score for neurologistsrsquo knowledge of OC interactions with the six AEDs was 61 5 22 and for obstetricians was 378 f 19 Neurologists were both more familiar with ( t = 350 d f 297 p lt 0001) and more knowledgeable about ( t = 782 d f 292 p lt 0001) AED-OC interactions than obstetricians

Most neurologists (77 to 84) knew that phenytoin carbamazepine and phenobarbital interfered with OCs but only 38 knew that valproic acid does not interfere with

Specialistsrsquo knowledge of OC-AED interactions

1536 NEUROLOGY 46 June 1996

Figure 2 Estimate of birth defect risk associated with AEDs

OCs Most obstetricians also knew that phenytoin interfered with OCs (77) but fewer knew that phenobarbital (54) carbamazepine (44) and primidone (17) lowered OC levels and only 29 know that valproic acid did not interfere with OCs Few neurologists and obstetricians knew whether ethosuximide or the two new AEDs gabapentin and felbamate interfere with OCs

Younger physicians were slightly more knowledgeable than older physicians about common AED-OC interactions although this was significant only for obstetricians (neurologists r = 014 p = 009 obstetricians r = 030 p lt 0001)

Fewer than half of neurologists (41) and obstetricians (43) have patients adjust their OC doses if they are taking AEDs Moreover a number of physicians noted that they ldquoadjusted OC doses by lowering estrogen doses or using relatively low-dose 35-kg estrogen pills An additional 8 of neurologists refer their patients to obstetricians to have OC doses adjusted

There was a strong association between neurologistsrsquo and obstetriciansrsquo knowledge of which AEDs interfere with OCs and whether they adjust OC doses to compensate for AED effects (neurologists t = 277 d f 143 p = 0006 obstetricians t = 421 df 128 p lt 0001) Even physicians who were ldquofamiliarrdquo with this drug interaction but did not have accurate knowledge of which AEDs were involved usually adjusted their patientsrsquo OC doses for patients on AEDs 89 of neurologists and 76 of obstetricians familiar with OC-AED interactions adjust OC dosages for patients on AEDs whereas only 19 of neurologists and 24 of obstetricians not familiar with OC-AED interactions adjust OC dosages (neurologists chi-square 1043 p = 0001 obstetricians chi-square 1464 p = 00001)

Twenty-seven percent of neurologists and 21 of obstetricians reported OC failure in their patients taking AEDs Neurologists whose patients had accidental pregnancies on OCs and AEDs were more likely to adjust OC doses (58 versus 35) than neurologists whose patients had not had such pregnancies (chi-square 633 p = 0011) Obstetri- cians whose patients had accidental pregnancies on OC and AEDs were more likely to adjust OCs than obstetricians who had not had patients with accidental pregnancies (57 versus 39) although this did not reach signifi- cance (chi-square 296 p = 0086)

Pregnancy risks and counseling Women with epilepsy have significant risks of approximately 4 to 6 for congen-

Factors influencing OC dosing

Table 2 Effects of AEDs on OC hormone levels

Enzyme-inducing AEDs

Phenobarbital

Phenytoin

Carbamazepine

Oxacarbamazepine Nonenzyme-inducing AEDs

Valproic acid

Gabapentin

-381

-490 -422

-470

110

60

Estradiol () -

~

Changes in 24-hour serum levels from baseline or control values No significant change References 1 2 7 20 23-25

ital malformations in their offspring compared with a background risk of approximately 23 This 2-3X increased risk is increased further for women with uncon- trolled seizures taking multiple AEDs5 or who belong to high-risk socioeconomic group^^ Despite this there was a wide variation in physiciansrsquo estimates of the risk of malformations with many physicians particularly neurologists underestimating the risk (figure 2) When counseling a pregnant woman with good seizure control on a single antiepileptic medication obstetricians said they would estimate a mean risk for birth defects of 689 (range 0 to 50) Most obstetricians (95) estimated a birth defect risk due to a single AED between 1 and 10 Over one- fifth (23) of obstetricians underestimated the risk as 0 to 3 In counseling a woman with epilepsy neurologists estimated a lower mean risk for birth defects than did obstetricians 475 (range 0 to 50) (neurologistsrsquo versus obstetriciansrsquo estimates t = 351 df 286 p = 001) Most neurologists (95) estimated a risk of birth defects between 0 and 9 A large proportion of neurologists (44) underestimated the risk of birth defects as 0 to 3

Only 3 of neurologists and 5 of obstetricians recommend that their patients taking AEDs not become pregnant Physiciansrsquo estimates of the risk of birth defects associated with a single AED did not influence whether they would recommend that women taking single anticonvulsants should avoid pregnancy (neurologists t = 139 df 147 p = 0165 obstetricians t = 078 df 132 p = 044) There was also no correlation between physiciansrsquo estimates of the risk of birth defects and their knowledge of OC-AED interactions (neurologists r = 005 p = 056 obstetricians r = 0 0 1 ~ = 081)

Table 3 Reported OC failures with AEDs

AED Accidental pregnancies

Phenobarbital

Phenytoin

Carbamazepine

Valproic acid

Ethosuximide

51

42

19

6 (0 monotherapy)

4

References 1 7 11 28

Norgestrel or norethindrone ()

- 187

-420

-397

-360

oo 30rsquo

Discussion Over one-fifth of neurologists and obstetricians reported accidental pregnancies in their patients taking AEDs with OCs Most physicians however did not know which common AEDs interfere with OCs do not adjust OC doses to overcome the effects of enzyme-inducing AEDs on sex hormone levels and did not know that valproic acid and gabapentin are nonenzyme-inducing alternatives to enzyme-inducing AEDs Although physicians who were at least moderately familiar with OC-AED interactions tended to adjust OC doses for their patients it is unclear how they would do this without recogniz- ing specific AED-OC interactions These results occur despite warnings of AED-OC interactions in pharmaceutical product inserts the Physiciansrsquo Desk Reference18 and in several medical reviews and edi- t o r i a l ~ ~ ~ ~ J ~ - ~ ~ Our survey suggests that a large number of women in the United States with epilepsy are at risk for unplanned pregnancies while taking OCs

There are several limitations in our survey A response bias to the questionnaire is possible It is typical however for only a few mailed questionnaires to be returned19 The geographic distribution practice types and ages of respondents were typical for American neurologists and obstetricians (see table l)16J7 Nonetheless the survey may have overes- timated physiciansrsquo lack of knowledge about AED complications for women with epilepsy particularly if nonrespondents were physicians who already knew about the reproductive risks of AEDs More likely however is that those who believed these issues are important would respond rather than those who do not recognize the reproductive complications associated with AEDs This suggests in fact that any skew in the study would be toward overestimating physiciansrsquo knowledge of these issues Evidence that less-knowledgeable physicians were not over- represented in the survey is that a larger proportion of responders were board-certified specialists than the national average (see table 1) Another limitation is that several questions could not be asked directly without cuing other responses We could not ask for example questions to determine precisely how phy-


Table 4 Suggestions for combining OCs and AEDs for women with epilepsy

1 Discuss contraceptive choices and the effects of enzyme- inducing AEDs (carbamazepine phenytoin phenobarbital primidone and ethosuximide) on OCs with all women of child-bearing years who have epilepsy This can be part of a general discussion of the risks of pregnancy and epilepsy

When treating women with epilepsy taking OCs continue to select AEDs according to recommended indications for epilepsy33

AED (valproic acid gabapentin vigabatrin) for women taking OCs

equivalent) for women taking enzyme-inducing AEDs who desire effective contraception Obstetricians and family physicians usually prescribe the OCs Counsel women that increased OC doses do not always protect from contraceptive failures due to AEDs but that OC failure rates are usually lower than with barrier methods alone

irregular menstrual bleeding while taking OCs Increase OC doses or switch contraceptives if irregular bleeding occurs

2

3 If AEDs are otherwise equivalent use a nonenzyme-inducing

4 Recommend OC doses of 50 pg estradiol (or mestranol

5 Have women with epilepsy monitor for breakthrough or

6 Advise women seeking very high levels of contraception protection that they may add spermicide or barrier contraception to their OCs

7 Alternative contraceptives that can be used for some women are medroxyprogesterone (Depo-Provera) and intrauterine devices Many pregnancies have occurred with levonorgestrol implant (ldquoNorplantrdquo) and enzyme-inducing AEDs

8 Oxcarbazepine lowers OC hormone levels whereas gabapentin and vigabatrin are nonenzyme-inducers that are unlikely to interfere with OCs The effects of lamotrigine and felbamate on OCs are unclear

Modified from references 7 13 15

sicians adjust OC doses whether physicians monitor patients for irregular menstrual bleeding as a sign of AED effects on OCs and how many total patients had accidental pregnancies due to AEDs

Enzyme-inducing AEDs including carbamazepine phenytoin phenobarbital primidone and ethosuximide decrease the effectiveness of OCs by lowering synthetic sex hormone levels Enzyme- inducing AEDs increase cytochrome P450 en- zymes12021 resulting in increased metabolism of synthetic estrogens (ethinyl estradiol and mestranol) and lowered serum levels of approximately 40 (table 2) 1 2 T Z2~25 Free progestin levels are also decreased probably due to increased synthesis of sex hormone-binding globulins220 Valproic and two new AEDs- gabapentin and vigabatrin- do not induce hepatic P450 metabolism and are unlikely to interfere with OCs Gabapentin does not decrease estradiol levels24 and vigabatrin has not been stud- ied Oxcarbazepine lowers OC sex hormone levels2rdquo and is likely to interfere with OCs Effects of lamotrigine and felbamate on OCs have not been stud- 1538 NEUROLOGY 46 June 1996

ied Lamotrigine is metabolized by hepatic glucu- ronidation however rather than by the microsomal P450 system and possibly may not stimulate sex hormone metabolism

Over 150 accidental pregnancies have been reported in patients taking enzyme-inducing AEDs primarily women from the United K i n g d ~ m ~ ~ ~ and including the minimum of 75 American patients reported in this survey17~s1126-28 Valproic acid a nonenzyme-inducing AED has not been associated with accidental pregnancies when used in monotherapy (table 3)

There are recommendations that women taking enzyme-inducing mDs increase their OC dose from 20 to 35 pg estradiol to 50 pg particularly if breakthrough menstrual bleeding O C C U ~ S ~ ~ J ~ - ~ ~ OC failures may occasionally occur however at 50- to 80-pg estradiol doses and with no warning of irregular menstrual bleeding so women taking these doses of OCs with enzyme-inducing AEDs need counseling that they are not completely protected against pregnancy Most women who were previously reported with OC failures were taking 50- to 80-pg estradiol dose^^^^ Two of our four recent patients had OC failures despite taking OCs containing 50 pg estradiol (a fifth patient became pregnant while using levonorgestrol implant)

Physicians may use irregular or breakthrough menstrual bleeding as a clinical sign of low OC lev- els7J2 and the need to increase OC doses or add contraceptives Enzyme-inducing AEDs produce breakthrough bleeding in 71 to 76 of women taking OCs whereas only 6 of those taking valproic acid had breakthrough bleeding7 Women taking 30 pg estradiol had higher rates of bleeding than those taking 50 pg and when OC doses were increased from 50 to 75 pg estradiol breakthrough bleeding stopped in 10 of 12 women initially and in 8 of 12 chronically

This supports using ldquohigher doserdquo OCs with 50 pg estradiol for most women taking enzyme-inducing AEDs rather than the ldquolow doserdquo 20- to 35-pg estradiol pills High-dose OCs with 80 pg estradiol are no longer available Women with breakthrough bleeding on 50 pg estradiol may supplement their dose by 25 pg estradiol using an extra half pill or use other contraceptive methods A small proportion of women do not have decreased OC levels with MDsZ0 and will be exposed to slightly increased steroid levels if OC doses are increased Those a t most risk for thromboembolic complications associated with high- dose OCs are smokers and women over 35 years

There are currently no routine clinical labora- tory measures to determine which women have suff- cient OC sex hormone levels to prevent conception

Despite the effects of AEDs on sex hormone levels OCs are still usually more effective than barrier contraceptives The typical failure rate with barrier contraceptives is 12 to 211deg Seven percent of women taking AEDs with OCs had accidental pregnancies in one small seriesll Women taking OCs who seek high levels of contraceptive protection can in addition to

increasing their OC dose add a spermicidal gel or foam or other barrier methods An effective alternative hormonal therapy to OCs is the depo form of medroxyprogesterone (Depo Provera) a strong progestin contraceptive that has not been associated with contraceptive failures due to AEDs~O Levonor- gestrol implant (ldquoNorplantrdquo) is not an effective alternative for women with epilepsy Over 30 accidental pregnancies have occurred for women using levonorgestrol implant with AEDs31 this is probably due to increased sex hormone-binding globulin and decreased free norgestrol levels32 Certain women may also consider intrauterine devices instead of OCs for contraception

After our experience with accidental pregnancies in our patients we counsel women of child-bearing years with epilepsy about their contraceptive choices using guidelines based on previous recommenda- t i o n ~ ~ ~ rsquo rdquo ~ ~ and incorporating the new data about AEDs and contraceptives reviewed above (table 4) Women taking enzyme-inducing AEDs for other neu- rologic and psychiatric conditions will also benefit from contraceptive counseling

In addition to often not being aware of contraceptive problems many physicians were not aware of the teratogenicity of AEDs Nearly one-half of neurologists (44) estimated the risk of birth defects in offspring of women treated for epilepsy as 0 to 3 This estimate is similar to the background rate of 2 for women not taking AEDs and underestimates the true risk of approximately 4 to 63 Obstetriciansrsquo estimates of the risk for malformations were more accurate than neurologists although 23 of them also underestimated the risk of birth defects

These results suggest that many neurologists and obstetricians do not have accurate knowledge to counsel women with epilepsy regarding their contraceptive and pregnancy choices Women with epilepsy might often benefit from discussing reproductive issues with both specialists because neurologists are generally more familiar with the effects of AEDs on OCs and obstetricians provide more accurate counseling about birth defect risks

References 1 Back DJ Orme ML Pharmacokinetic drug interactions with

oral contraceptives Clin Pharmacokinet 199018472-484 2 Crawford P Chadwick DJ Martin C Tjia J Back DJ Orme

M The interaction of phenytoin and carbamazepine with com- bined oral contraceptive steroids Br J Clin Pharmacol 1990

3 Yerby MS Treatment of epilepsy during pregnancy In Wyllie E ed The treatment of epilepsy principles and practices Philadelphia Lea amp Febiger 1993844-857

4 Waters CH Belai Y Gott PS Shen P De Giorgio CM Out- comes of pregnancy associated with antiepileptic drugs Arch Neurol 19945 1250 -253

5 Hauser WA HesdorfTer DC Pregnancy and teratogenesis In Epilepsy frequency causes and consequences New York Demos Publications 1990147-196

6 Beresford HR Legal issues roles of physicians in preventing fetal harm In Devinsky 0 Feldmann E Hainline B eds

30892-896

Neurological complications of pregnancy New York Raven Press 1994

7 Sonnen AE Sodium valproate and the contraceptive pill Br J Clin Prac Symp Suppl 19832731-36

8 Back DJ Evaluation of Committee on Safety of Medicines yellow card reports on oral contraceptive drug interactions with anticonvulsants and antibiotics Br J Clin Pharmacol

9 Sparrow MF Pill method failures NZ Med J 1987100102- 103

10 Trussell J Kost K A guide to interpreting contraceptive efficacy studies Obstet Gynecol 199076558-567

11 Coulam CB Do anticonvulsants reduce the efficacy of oral contraceptives Epilepsia 197920519-525

12 Diamond MP Greene JW Thompson JM VanHooydonk JE Interactions of anticonvulsants and oral contraceptives in epileptic adolescents Contraception 198531623-632

13 Mattson RH Cramer JA Darney PD Naftolin F Use of oral contraceptives by women with epilepsy JAMA 1986256238- 240

14 Cramer JA Mattson RH Hormones and epilepsy In Wyllie E ed The treatment of epilepsy principles and practices Philadelphia Lea amp Febiger 1993686-691

15 Editorial Drug interaction with oral contraceptive steroids

16 American Academy of Neurology Neurologists 1991-1992 Minneapolis AAN 1993

17 Roback G Randolph L Seldman B Pasko T Physician characteristics and distribution in the US Chicago AMA Depart- ment of Data Services 1994

18 Physiciansrsquo Desk Reference 48th ed Montvale Medical Eco- nomics Data Production Company 1994

19 Kerlinger F Foundations of behavioral research 2nd ed New York Holt Rinehart 1973414

20 Back DJ Breckenridge AM Crawford FE Orme ML Rowe PH Phenobarbitone interaction with oral contraceptive steroids in the rabbit and rat Br J Pharmacol 198069441-452

21 Levin W Welch RM Conney AH Effect of chronic phenobarbital treatment on the liver microsomal metabolism and uterotropic action of 17s-estradiol Endocrinology 196780

22 Back DJ Bates M Bowden M Breckenridge AM Hall MJ The interaction of phenobarbital and other anticonvulsants with oral contraceptive steroid therapy Contraception 1980

23 Crawford P Chadwick D Cleland P e t al The lack of effect of sodium valproate on the pharmacokinetics of oral contraceptive steroids Contraception 19853323-29

24 Eldon MA Underwood BA Randinitis EJ Posvar EL Sed- man AJ Lack of effect of gabapentin on the pharmacokinetics of a norethindrone acetateethinyl estradiol-containing oral contraceptive [abstract] Neurology 199343A307-A308

25 Jensen PK Saano V Haring P Svenstrup B Menge GP Possible interaction between oxcarbazepine and an oral contraceptive Epilepsia 199233 1149 -1 152

26 Janz D Schmidt D Antiepileptic drugs and failure of oral contraceptives Lancet 197411113

27 Kenyon TW Unplanned pregnancy in an epileptic BMJ 1972

28 Krauss GL Brandt J Campbell M Plate C Summerfield M Anti-epileptic medicationioral contraceptive interactions a national survey of neurologists and obstetricians Neurology 1996461534-1539

29 Stadel BV Oral contraceptives and cardiovascular disease part 1 N Engl J Med 1981305612-618

30 Upjohn Laboratories Data on File 1995 31 Wyeth-Ayerst Laboratories Norplant System Data on file

1995 32 Haukkamaa M Contraception by Norplant subdermal cap-

sules is not reliable in epileptic patients on anticonvulsant treatment Contraception 198633559-565

33 Mattson RH Selection of antiepileptic drug therapy In Levy RH Mattson RH Meldrum BS eds Antiepileptic drugs 4th ed New York Raven Press 1995123-135

198825527-532

BMJ 198028193-94

135-140

22495-503

1686-687



G L Krauss J Brandt M Campbell et al neurologists and obstetricians

Antiepileptic medication and oral contraceptive interactions A national survey of


ServicesUpdated Information amp

httpwwwneurologyorgcontent4661534fullhtmlincluding high resolution figures can be found at

References httpwwwneurologyorgcontent4661534fullhtmlref-list-1

This article cites 22 articles 2 of which you can access for free at

Citations httpwwwneurologyorgcontent4661534fullhtmlotherarticles

This article has been cited by 7 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsin its entirety can be found online atInformation about reproducing this article in parts (figurestables) or

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Antiepileptic medication and oral contraceptive interactions

A national survey of neurologists and obstetricians GL Krauss MD J Brandt PhD M Campbell RN C Plate BS and M Summerfield

Article abstract-Hepatic enzyme-inducing antiepileptic drugs (AEDs) lower oral contraceptive (OC) sex hormone levels approximately 40 and increase the risk of unplanned pregnancies in women with epilepsy AEDs also increase the risk of birth defects in offspring of women with epilepsy We performed a national survey to determine obstetriciansrsquo and neurologistsrsquo knowledge of OC and AED interactions and the risk of birth defects for women with epilepsy taking AEDs We received responses to a mailed questionnaire from 160 of 1000 neurologists (16) and 147 of 1000 obstetricians (15) from 47 states Practice demographics and ages of responders were typical for US neurologists and obstetricians Ninety-one percent of neurologists and 75 of obstetricians said they treat women with epilepsy of child-bearing age Only 4 of the neurologists and none of the obstetricians however knew the effects of the six most common AEDs on OCs even though 27 of neurologists and 21 of obstetricians reported OC failures in their patients taking AEDs Although increasing OC doses can compensate for insufficient OC sex hormone levels due to AEDs most physicians do not increase the doses Even though the risk of birth defects for the offspring of women with epilepsy is 4 to 6 up from the background level of 296 44 of neurologists thought the risk was lower (0 to 3) and some of the respondents guessed that it was as high as 50 Many neurologists and obstetricians do not have accurate information to counsel women with epilepsy properly about their contraceptive and pregnancy choices NEUROLOGY 199646 1534-1539

Antiepilepsy drugs (AEDs) are associated with two major reproductive risks for women with epilepsy of child-bearing age (1) an increased risk of unplanned pregnancies due to hepatic enzyme-inducing AEDs lowering sex hormone levels of oral contraceptives (OCs)rsquo (2) an increased risk of birth defect^^^ We conducted a national survey of obstetricians and neurologists to determine whether they were aware that enzyme-inducing AEDs decrease the effectiveness of OCs and whether their patients taking AEDs had become accidentally pregnant while taking OCs We also determined whether physicians accurately counsel women about their increased risk of birth defects due to AEDs

During the past 2 years we saw five women at Johns Hopkins Hospital with hormonal contraceptive failures while taking AEDs These unexpected pregnancies caused major social and emotional dis- tress one woman also had a brain tumor one had mental retardation and epilepsy surgery was can- celed for a third We became concerned that neurologists and obstetricians may not be aware of this drug interaction and that women with epilepsy taking AEDs are not being warned of their risk for accidental pregnancies This is especially important because women with epilepsy are at an approximately 4 to 6 risk of birth defects (2-3X background) in their

offspring3 with even greater risks for high-risk group^^^ Physicians have been held legally responsi- ble for a ldquowrongful liferdquo when a woman was not in- formed of specific pregnancy risks associated with her AEDslsquoj

Over 80 women have been reported in the United States and United Kingdom with accidental pregnancies due to enzyme-inducing AEDs interfering with O C S ~ gt ~ - ~ The enzyme-inducing AEDs associated with OC failures are carbamazepine phenytoin phenobarbital primidone and ethosuximide All reduce OC estradiol levels approximately 40 and may also decrease free progestin levels by increasing sex hormone-binding globulins12 Enzyme-inducing AEDs increase the expected OC failure ratelo of 01 to 3 by severalfold In one small study 7 of epilepsy patients taking OCs with AEDs had OC fai1ureslrsquo The nonhepatic enzyme-inducing AEDs -valproic acid and gabapentin- do not decrease OC levels and have not been associated with accidental pregnancies

Physicians can compensate for the influence of enzyme-inducing AEDs on OC sex hormone levels by increasing OC estradiol doses from 20 to 35 pg to 50 kg especially if there is irregular or breakthrough menstrual bleeding on low-dose OCs27rsquo2-15 Because most reported OC failures have been with 50- to

From the Departments nf Neurology IDr Krauss Ms Campbell Ms Plate and Mr Summerfield and Psychiatry (Dr Brandt) Johns Hopkins University Baltimore MD Received July 11 1995 Accepted in final form October 2 1995 Address correspondence and reprint requests to Dr Gregory L Krauss Johns Hopkins Hospital Meyer 2-147 600 N Wolfe Street Baltimore MD 21287-7247

1534 Copyright S 1996 by the American Academy of Neurology



-

Total responders


Work settings ()

Solo


Med school

Research other

Board-Certified ()

Year of MD

160 (16)

453 (30-75) 445

316 232

513 373

33 154

112 203

26 38

925 776

1943-1990





Survey questions

147 (147)

424 (28-74) 420

341

507

58

58

876

66

12

36 33

925 71 1943-1992
























Phenobarbital

Phenytoin

Carbamazepine


Valproic acid

Gabapentin

-381

-490 -422

-470

110

60

Estradiol () -

~






Phenobarbital

Phenytoin

Carbamazepine

Valproic acid

Ethosuximide

51

42

19

6 (0 monotherapy)

4



- 187

-420

-397

-360

oo 30rsquo










2




























30892-896





























198825527-532

BMJ 198028193-94

135-140

22495-503

1686-687














Reprints




-

Total responders


Work settings ()

Solo


Med school

Research other

Board-Certified ()

Year of MD

160 (16)

453 (30-75) 445

316 232

513 373

33 154

112 203

26 38

925 776

1943-1990





Survey questions

147 (147)

424 (28-74) 420

341

507

58

58

876

66

12

36 33

925 71 1943-1992
























Phenobarbital

Phenytoin

Carbamazepine


Valproic acid

Gabapentin

-381

-490 -422

-470

110

60

Estradiol () -

~






Phenobarbital

Phenytoin

Carbamazepine

Valproic acid

Ethosuximide

51

42

19

6 (0 monotherapy)

4



- 187

-420

-397

-360

oo 30rsquo










2




























30892-896





























198825527-532

BMJ 198028193-94

135-140

22495-503

1686-687














Reprints



















Phenobarbital

Phenytoin

Carbamazepine


Valproic acid

Gabapentin

-381

-490 -422

-470

110

60

Estradiol () -

~






Phenobarbital

Phenytoin

Carbamazepine

Valproic acid

Ethosuximide

51

42

19

6 (0 monotherapy)

4



- 187

-420

-397

-360

oo 30rsquo










2




























30892-896





























198825527-532

BMJ 198028193-94

135-140

22495-503

1686-687














Reprints




Phenobarbital

Phenytoin

Carbamazepine


Valproic acid

Gabapentin

-381

-490 -422

-470

110

60

Estradiol () -

~






Phenobarbital

Phenytoin

Carbamazepine

Valproic acid

Ethosuximide

51

42

19

6 (0 monotherapy)

4



- 187

-420

-397

-360

oo 30rsquo










2




























30892-896





























198825527-532

BMJ 198028193-94

135-140

22495-503

1686-687














Reprints








2




























30892-896





























198825527-532

BMJ 198028193-94

135-140

22495-503

1686-687














Reprints













30892-896





























198825527-532

BMJ 198028193-94

135-140

22495-503

1686-687














Reprints














Reprints


Documents

Neurology 1996 Krauss 1534 9