A Survey of Pregnant Women Using Isotretinoin

Julia Robertson,1* Janine E. Polifka,2 Marina Avner,3 Christina Chambers,4 George Delevan,1

Gideon Koren,3 Sharon Voyer Lavigne,5 Lynn P. Martinez,1 Richard K. Miller,6 and John C. Carey7

1Utah Department of Health, Division of Community and Family Health Services, Birth Defects and Genetics Program, PregnancyRiskLine, Salt Lake City, Utah

2CARE Northwest and TERIS, Department of Pediatrics, Center on Human Development and Disabilities, University of Washington,Seattle, Washington

3Motherisk Program, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada4University of California, San Diego, Departments of Pediatrics and Family and Preventive Medicine, La Jolla, California

5Connecticut Pregnancy Exposure Information Service, Division of Human Genetics, Department of Genetics and DevelopmentalBiology, University of Connecticut Health Center, Farmington, Connecticut

6PEDECS (National Institute of Environmental Health Sciences/New York Teratogen Information Service), Department of Obstetricsand Gynecology, University of Rochester School of Medicine and Dentistry, Rochester, New York

7Department of Pediatrics, Division of Medical Genetics, University of Utah Health Sciences Center, Salt Lake City, Utah

Received 26 April 2005; Accepted 28 July 2005

BACKGROUND: Isotretinoin is a known human teratogen, causing birth defects and/or subnormal cognitiveperformance in prenatally-exposed children. METHODS: A survey was conducted among women who calledteratology information services throughout North America. Using a structured questionnaire, women with anisotretinoin-exposed pregnancy were prospectively interviewed before the outcome of the pregnancy wasknown. RESULTS: Almost 1/4 of the women surveyed (24%; 8/34) did not recall having contraceptioncounseling before starting their medications. Once therapy was initiated, 62% (21/34) recalled using a birthcontrol method, but only 29% (6/21) recalled using 2 forms of birth control, as specified by the voluntarypregnancy prevention programs. Monthly pregnancy tests were not always conducted during treatment, asrecalled by the surveyed women (56%; 19/34). As many as 24% (8/34) of the women surveyed recalled that theywere not screened using 2 pregnancy tests before receiving a prescription, another recommendation of theprograms. Only a small number of the women (30%; 6/20) in the United States recalled being enrolled in anymanufacturers’ voluntary pregnancy prevention survey. CONCLUSIONS: Results demonstrate that essentialcomponents of voluntary pregnancy prevention programs were not consistently followed, which resulted infetal exposures. Birth Defects Research (Part A) 73:881–887, 2005. © 2005 Wiley-Liss, Inc.

Key words: isotretinoin; Accutane; pregnancy; birth defects

INTRODUCTION

Isotretinoin is a vitamin A congener approved for use inthe treatment of severe, recalcitrant nodular acne (Perryand McEvoy, 1983; Layton et al., 1993; Shahidullah et al.,1994; Hermes et al., 1998). If left untreated, the cysts orlesions can cause pain as well as emotional and physicalscarring. Unlike other anti-acne medications, isotretinoin iseffective in treating a broad spectrum of dermatologicdisorders (Goldsmith et al., 2004). As a result, its use inless-severe acne that is refractory to conventional therapieshas been increasingly endorsed (Lowenstein, 2002;O’Donnell, 2003; Honein et al., 2004). During the meetingof the Subcommittee on Oversight and Investigations ofthe House of Representatives, in 2002, it was estimated thatas many as 90% of prescriptions for isotretinoin were foroff-label uses (Honein et al., 2004).

Although isotretinoin provides tremendous benefits toacne patients, its use is not without risks, including the riskof birth defects when women take the drug during preg-nancy. Isotretinoin is recognized as a human teratogen andcauses a spectrum of congenital anomalies (referred to as“isotretinoin embryopathy”) in �35% of exposed infants.Of the affected infants, �70% will have microtia/anotia,

Grant sponsor: Centers for Disease Control and Prevention in cooperation withthe Association of American Medical Colleges and the University of Utah HealthSciences Center; Grant number: MM-0404/03/03.*Correspondence to: Julia A. Robertson, Utah Department of Health, Divisionof Community & Family Health Services, Birth Defects and Genetics Program,Pregnancy RiskLine, P.O. Box 144691, 44 North Medical Drive, Salt Lake City,UT 84114-4691. E-mail: jrobertson@utah.govPublished online 14 October 2005 in Wiley InterScience (www.interscience.wiley.com).DOI: 10.1002/bdra.20197

© 2005 Wiley-Liss, Inc. Birth Defects Research (Part A) 73:881–887 (2005)

Birth Defects Research (Part A): Clinical and Molecular Teratology 73:881–887 (2005)

�25% will have micrognathia, and �10% will have cleftpalate. Heart defects (conotruncal and aortic-arch abnor-malities) are seen in as many as 40%, with thymic defectsoccurring in �30% of exposed infants. Retinal or opticnerve abnormalities are present in almost 20% of thosewith the embryopathy. In addition, defects of the centralnervous system are seen frequently, including microceph-aly, poor cognitive performance, and learning disabilities,which occur in �80% of affected cases (Lammer et al., 1985;Adams and Lammer, 1991; Dai et al.,1992). Furthermore,problems in cognitive performance are seen even amongchildren who have no structural defects; overall, 40% ofthose who are prenatally exposed have impaired learningability (Adams and Lammer, 1991). Isotretinoin also in-creases the risk for miscarriage or stillbirth, which occursin as many as 40% of exposed pregnancies (Lammer et al.,1987; Dai et al., 1992). In addition, women who becomepregnant while taking isotretinoin have about a doublingof risk for delivering a premature baby (Adams and Lam-mer, 1991).

Because of the teratogenic potential of isotretinoin, theU.S. Food and Drug Administration (FDA) and the man-ufacturer of Accutane (Roche Pharmaceuticals, Nutley, NJ)developed a voluntary pregnancy prevention program(PPP) in 1988. This program was designed to providehealth care professionals with prescribing guidelines forwomen of reproductive age with the intention of prevent-ing conception. Because isotretinoin-exposed pregnanciescontinued to occur between 1988 and 2000 during theimplementation of this program, the PPP was replaced inthe United States in 2002 with the System to ManageAccutane Related Teratogenicity (SMART) (Mitchell et al.,1995; Atanackovic and Koren, 1999; Honein et al., 2001,2004; Brinker et al., 2002; O’Donnell, 2003; Goldsmith et al.,2004). This program was similar to the PPP but placedmore emphasis on the need for pregnancy testing (beforeand during isotretinoin therapy) and the use of 2 forms ofbirth control (see Table 1). It was hoped that by puttingmore stringent requirements in place, fewer unintentionalpregnancies would occur in women being treated withisotretinoin. Many professionals feel that the teratogenic

potential of isotretinoin can be effectively managedthrough voluntary compliance with industry-sponsoredpregnancy prevention programs (Cooper, 2003). However,the number of isotretinoin-exposed pregnancies reportedto the FDA during the years before SMART (127 pregnan-cies reported) and after implementation (120 pregnanciesreported) remains essentially the same despite a 6% dropin the total number of written prescriptions, suggestingthat SMART may not be achieving its goal (Uhl, 2004).Similar pregnancy prevention programs were developedfor the generic versions of isotretinoin that became avail-able in 2002 and 2003 (Honein et al., 2004; Koren et al.,2004).

During the 13th Annual Meeting of the Organization ofTeratology Information Services (OTIS), an abstract pre-sented by Lavigne and Rosengren (2000) indicated an in-crease in the number of isotretinoin-exposure calls to ateratogen information service in Connecticut. Thisprompted an informal inquiry of OTIS and its pregnancyrisk-counseling specialists, which showed that between2000 and 2003, a total of 136 isotretinoin-exposed pregnantwomen contacted 1 of these services—far more than inprevious years. This report and others provide furtherevidence that isotretinoin-exposed pregnancies are still oc-curring despite the implementation of SMART (Carver etal., 2001; Honein et al.; 2001, 2004; Jones et al., 2001; Brinkeret al., 2002; Robertson et al., 2002; Wysowski et al., 2002;Uhl, 2004). However, little information is available aboutwhy these pregnancies occur despite the requirements ofthe prevention programs. Therefore, the present surveywas undertaken to identify how isotretinoin is dispensedamong women who become pregnant while taking thedrug and possible reasons for failure of the PPP or SMARTprograms to prevent those pregnancies. During this sur-vey, the PPP was used in Canada and the SMART programwas used in the United States.

MATERIALS AND METHODSDesign

This survey of isotretinoin-exposed pregnancies in-cludes responses from women who voluntarily contactedan OTIS member service. A target sample size of 30 sub-jects was projected based on the number of eligible callersand expected participants. It was anticipated that partici-pation rates would be relatively low, as previous anecdotalreports from OTIS and a study by the Slone EpidemiologyUnit (Mitchell et al., 1995) showed that most women (72%)with isotretinoin-exposed pregnancies elect to discontinuethe pregnancy and, therefore, may be less likely to partic-ipate in any survey.

SettingThe present survey was conducted within the OTIS net-

work of teratology information services (TIS). OTIS is aNorth American nonprofit organization comprised of 19teratology information services and 16 professionals inter-ested in teratology. OTIS services and members providetelephone consultation to healthcare professionals andtheir patients who desire up-to-date scientific informationregarding the effects of maternal exposures to environmen-tal agents on the developing embryo or fetus (Leen-Mitch-ell et al., 2000). OTIS member services receive �70,000inquiries per year. The collaboration of OTIS services pro-vides a unique opportunity to develop evidence-based

Table 1SMART Requirements

• The approved indication for use of isotretinoin is for“treatment of severe recalcitrant nodular acne.”

• Women must have 2 negative pregnancy tests, the second ofwhich must be conducted during the first 5 days of thewoman’s menstrual period before receiving a prescription.

• Women must receive a pregnancy test each month of therapybefore refilling their prescription.

• Women must use 2 reliable forms of birth controlsimultaneously, starting 1 month before, during, and 1 monthafter isotretinoin therapy.

• Pharmacists must only fill prescriptions that bear amanufacturer-issued yellow qualification sticker.

• Pharmacists must fill prescriptions within 7 days of theirbeing written and to dispense no more than a 30-day supplyat 1 time.

• Women must sign an informed consent form prior to startingisotretinoin therapy.

• Women must receive verbal and written warnings aboutpregnancy avoidance during isotretinoin therapy and for 2months following discontinuation of treatment.

882 ROBERTSON ET AL.

Birth Defects Research (Part A) 73:881–887 (2005)

information on the safety of certain environmental agentsduring pregnancy. All OTIS member services were askedto participate in the survey and 16 agreed. Each participat-ing TIS recruited pregnant women who had been exposedto isotretinoin and who had voluntarily contacted their TISbetween April 2002 and September 2004.

Various marketing strategies to increase awareness ofthe project were used, including newspaper and newsletterarticles, website links, letters to dermatologists, presenta-tions at conferences, and advertising in journals.

Pregnant women who had been exposed to isotretinoinand who called a TIS during the survey period receivedcounseling by a specialist trained in teratogen risk assess-ment and counseling. After the counseling was completed,the caller was asked to participate in the survey. If sheagreed, the counselor read the questionnaire cover letterfor consent to participate. Once consent was given, contactinformation was obtained and forwarded to the projectcoordinator who arranged for a telephone interview con-ducted by 1 of 2 specially-trained project interviewers.Most women were surveyed within 3 months of a con-firmed pregnancy. The questionnaire consisted of 3 parts:1) the intake interview; 2) the interim interview; and 3) theoutcome interview. The intake interview took �30 minwhereas the other interviews took no more than 10 mineach to complete. The study protocol was approved by theOTIS Research Committee; the University of California,San Diego Human Research Protections Program; the In-stitutional Review Board Health Sciences Section at Uni-versity of Missouri-Columbia; and the Utah Department ofHealth Institutional Review Board; as well as the Univer-sity of Utah Health Sciences Center Institutional ReviewBoard and The Hospital for Sick Children Research EthicsBoard, Toronto, Canada.

Survey QuestionnaireSurvey questions were designed and reviewed by sev-

eral experts in the field of clinical teratology. Among themany questions asked of callers were queries regardingtheir use of contraception, frequency and results of preg-nancy testing, and participation in manufacturers’ preg-nancy prevention programs (see Tables 2 and 3).

The responses were examined separately for the U.S. (20)and Canadian (14) women to determine differences be-tween versions of pregnancy prevention programs.

Table 2Percentage of Women Answering “Yes” to Survey Questions Pertaining to Contraception and Pregnancy Tests

%

Canada United States Total

While taking isotretinoin did you ever telephone the Confidential ContraceptionCounseling Line? 0% (0/14) 0% (0/20) 0% (0/34)

While taking isotretinoin did you ever receive contraception counseling? 36% (5/14) 15% (3/20) 24% (8/34)While taking isotretinoin did you ever take the morning-after pill? 0% (0/14) 15% (3/20) 9% (3/34)Before starting isotretinoin, were you practicing birth control? 64% (9/14) 70% (14/20) 68% (23/34)During the time that you took isotretinoin while pregnant, were you using birth

control? 57% (8/14) 65% (13/20) 62% (21/34)Did you change your birth control method during any time that you were taking

isotretinoin? 7% (1/14) 35% (7/20) 24% (8/34)During the time that you took isotretinoin, while pregnant, were you using two

forms of birth control? 13% (1/8)a 38% (5/13)a 29% (6/21)a

During the time that you took isotretinoin, did you have a negative pregnancy testbefore your health care provider prescribed isotretinoin? 64% (9/14)b 68% (13/19)b 67% (22/33)b

During the time that you were taking isotretinoin, were monthly pregnancy testsdone? 86% (12/14) 50% (10/20) 65% (22/34)

During the time that you took isotretinoin, did you have a second pregnancy testduring your menstrual period? 21% (3/14)b 26% (5/19)b 24% (8/33)b

During the time that you took isotretinoin, while pregnant, were monthlypregnancy tests done? 64% (9/14) 55% (11/20) 56% (19/34)

During the time that you took isotretinoin while pregnant, did pregnancy testsindicate you were not pregnant, when in fact you were pregnant? 29% (4/14) 50% (10/20) 41% (14/34)aDenominator represents the 8 Canadian and 13 U.S. women who were using any birth control method.bDenominator represents the 14 Canadian and 19 U.S. women who received medication in Canada or in the United States.

Table 3Percentage of U.S. Women Answering “Yes” to

Questions Regarding Enrollment in theManufacturer’s Survey

%

During the time that you took isotretinoin,did you enroll in any of themanufacturer’s surveys? 30% (6/20)

Accutane surveyDid you receive a check for $20? 34% (2/6)a

Isotretinoin surveyDid you receive a check for $20? 0%

Both SurveysDid you receive a check for $20? 0%

Did you know about the survey andchoose not to enroll? 15% (3/20)b

aDenominator consists of U.S. women who recalled enrolling inthe manufacturer’s survey.

bDenominator refers to 20 U.S. women who recalled hearingabout the manufacturer’s survey.

883PREGNANCY SURVEY: ISOTRETINOIN USE

Birth Defects Research (Part A) 73:881–887 (2005)

RESULTS

Of the 109 women who called 1 of 16 participating OTISmember services during the survey period because theyhad become pregnant while taking isotretinoin, 34 agreedto participate in the survey and all had inadvertently ex-posed their fetus during the first trimester. Characteristicsof the survey participants in Canada and the United States,and the combined participants are presented in Table 4.

Of the women surveyed, 14 (44%) of the 32 had beenpregnant before. (Note: 2 women from Canada did notrespond to this question.) In addition, 1 of these womenreported a previous pregnancy with exposure to isotreti-noin. Of the various brand names for isotretinoin, Accu-tane (Roche, Nutley, NJ) was taken by 73% (25/34); Am-nesteem (Genpharm Inc., Quebec, Canada, an affiliate ofMerck KGaA, Darnstadt, Germany) by 12% (4/34); Clara-vis (Barr, Woodcliff Lake, NJ) by 3% (1/34); Oratane(Douglas Pharmaceuticals, Auckland, New Zealand) by3% (1/34), and Isotretinoin (Ranbaxy, Princeton, NJ) by 3%(1/34) . One woman had taken both Accutane and Am-nesteem. Roche’s Soriatane (acitretin), a retinoid for use inpsoriasis, was taken by 1 of the respondents. We includedher responses in this isotretinoin survey because she tookthe drug for severe acne and thought she had taken Accu-tane.

For 25/32 (78%) of the surveyed women, this was thefirst time they had taken isotretinoin. (Note: 2 women fromCanada did not respond to this question.) Concomitantmedical treatments for acne, such as oral antibiotics andtopical agents, were used in 82% (28/34) of the womensurveyed.

In both Canada and the United States, 76% (26/34) of thewomen surveyed learned about the benefits of takingisotretinoin from a health care provider such as a derma-tologist (42%; 11/26), other doctor (12%; 3/26), or pharma-

cist (4%; 1/26). Family members (6%; 2/34) were also asource for information, as well as friends (12%; 4/34). Nowomen reported getting information from magazines,newspapers or television advertisements, or the Internet.Two women (6%) could not remember where they learnedabout the benefits of taking isotretinoin.

More than 90% of the women received the medication ora prescription for the medication from a health care pro-vider (91%; 31/34); in most cases, that was a dermatologist(90%; 28/31). Two of the women received the medicationin the form of samples from their dermatologist. Only 1woman received her prescription in another country (Mex-ico). None of the women reported ordering isotretinoinfrom the Internet or getting it from a friend or relative.

Of the women surveyed, 82% (28/34) recalled receivinginformation about the risk for birth defects if they becamepregnant while taking isotretinoin. The majority of thesewomen (89%; 25/28) recalled obtaining this informationfrom the manufacturer in the form of a booklet or pam-phlet. Although 79% (27/34) of the women surveyed tookthe prescription to a pharmacy to be filled, only 30% (8/27)recalled their pharmacists’ providing information aboutthe potential teratogenic risk of isotretinoin. Of the womensurveyed, 97% (33/34) denied knowing about or attempt-ing to call the Accutane InfoLine listed in the manufactur-er’s product materials.

As shown in Table 2, the percentage of women whoreported accessing or receiving appropriate contraceptioncounseling was low in both Canada and the United States.None of the women surveyed called the Confidential Con-traception Counseling Line mentioned in the manufactur-er’s package insert, and only 24% (8/34) recalled receivingcontraception counseling before starting or at any timeduring isotretinoin therapy. Of the 8 women who receivedcontraception counseling, dermatologists (6; 75%) and

Table 4Characteristics of Survey Participants in Canada and the United States

Canada(n � 14)

United States(n � 20)

Canada and United States(n � 34)

Median maternal age, years (range) 30 (20–43) 25 (18–40) 26 (18–43)Maternal race/ethnicity

White, non-Hispanic 6/14 (43%) 17/20 (85%) 23/34 (68%)Hispanic 2/14 (14%) 2/20 (10%) 4/34 (12%)Asian or Pacific Islander 1/14 (7%) 0/20 (0%) 1/34 (3%)Black 0/14 (0%) 1/20 (5%) 1/34 (3%)East Indian 1/14 (7%) 0/20 (0%) 1/34 (3%)Didn’t know or declined to answer 4/14 (29%) 0/20 (0%) 4/34 (12%)

Maternal education Bachelor’s degree or higher 3/14 (21%) 10/20 (50%) 13/34 (38%)Annual household income $40,000 or greater 5/14 (36%) 13/20 (65%) 18/34 (53%)Gravidity �1 3/12 (25%)a 11/20 (55%) 14/32 (44%)a

Drug takenAccutane 14/14 (100%) 11/20 (55%) 25/34 (74%)Amnesteem – 4/20 (20%) 4/34 (12%)Accutane and Amnesteem – 1/20 (5%) 1/34 (3%)Claravis – 1/20 (5%) 1/34 (3%)Oratane – 1/20 (5%) 1/34 (3%)Soriatane – 1/20 (5%) 1/34 (3%)Isotretinoin – 1/20 (5%) 1/34 (3%)

First time taking isotretinoin 12/12 (100%)a 13/20 (65%) 25/32 (78%)a

Concomitant medical treatment (e.g., antibioticsand topical agents) 9/14 (64%) 19/20 (95%) 28/34 (82%)a2 women did not respond to the question.

884 ROBERTSON ET AL.

Birth Defects Research (Part A) 73:881–887 (2005)

family practitioners (2; 25%) were the only health careproviders who educated women about contraception be-fore they started or at any time during their isotretinointherapy.

Only 30% (6/20) of the U.S. women enrolled in themanufacturers’ voluntary survey, called “the AccutaneSurvey” (see Table 3). None of the women knew if theirhealth care provider had completed a MedWatch adversereport form to notify the FDA of an isotretinoin-exposedpregnancy.

Further results from the survey indicate that healthcareprofessionals and their patients failed to comply with anumber of key SMART requirements (see Table 1). Theserequirements and corresponding observations from theOTIS survey are as follows:

SMART Requirement 1. The approved indication foruse of isotretinoin is for “treatment of severe recalcitrantnodular acne.”

Observations from the OTIS Survey. Of the 34 womensurveyed, 91% (31) of the women surveyed reported re-ceiving a diagnosis of a skin problem by a health careprovider before starting treatment with isotretinoin. How-ever, few women (32%; 11/34) described their skin prob-lem as cystic or nodular acne, and only 39% (12/31) re-called a doctor’s diagnosis of cystic or nodular acne.

SMART Requirement 2. According to the PPP and theSMART program, women must have 2 negative pregnancytests before receiving a prescription.

Observations from the OTIS Survey. Twenty-five (76%) ofthe 33 women who enrolled in the survey said they did nothave a second pregnancy test during their menstrual pe-riod before beginning isotretinoin (see Table 2). (Note: Wedid not include the woman who received her medicationfrom Mexico.)

SMART Requirement 3. According to the SMART pro-gram, women must receive a pregnancy test each monthbefore refilling a prescription.

Observations from the OTIS Survey. Twelve (35%) of the34 women who enrolled in the survey said that they didnot have monthly pregnancy tests during the course oftherapy (see Table 2).

SMART Requirement 4. According to the SMART pro-gram, women must use 2 forms of birth control simulta-neously, starting 1 month before receiving a prescription.

Observations from the OTIS Survey. Of the 62% (21/34)who were using a birth control method while receivingisotretinoin therapy, only 29% (6/21) were using 2 forms ofbirth control. One survey participant reported her onlybirth control was the “rhythm method” (see Table 2).

SMART Requirement 5. According to the SMART pro-gram, pharmacists must only fill prescriptions that bear ayellow sticker. Note: The yellow sticker indicates that preg-nancy testing for the patient was negative and that 2 formsof birth control were in use. The intent of this programfeature was that the pharmacist would only dispenseisotretinoin prescriptions that had a yellow sticker.

Observations from the OTIS Survey. The following dataare presented for the 20 women surveyed in the UnitedStates where the yellow sticker requirement had been im-plemented. Of these U.S. women, 31% (5/16) recalled thatthey had received a prescription but did not recall seeing ayellow qualification sticker on the prescription. (Note: 2women were given samples, 1 went to a pharmacy inMexico and did not have a prescription, and 1 woman’s

mother took the prescription to the pharmacy.) Evenamong women who did recall a yellow sticker on theirprescription, not all recalled having had pregnancy testingor recalled birth control use prior to receiving the medica-tion from their pharmacists.

SMART Requirement 6. According to the SMART pro-gram, pharmacists must fill prescriptions within 7 days ofbeing written and dispense no more than a 30-day supplyat 1 time.

Observations from the OTIS Survey. The questionnaire didnot gather prescription dates or refill information.

SMART Requirement 7. According to the SMART pro-gram, women must sign an informed consent form prior tostarting isotretinoin therapy.

Observations from the OTIS Survey. Eighteen women(53%; 18/34) recalled signing a consent form prior to start-ing isotretinoin. Thirteen (72%; 13/18) were able to give anestimated date on which they signed the consent form.

SMART Requirement 8. According to the SMART pro-gram, women must receive verbal and written warningsabout pregnancy avoidance during isotretinoin therapyand for 2 months following discontinuation of treatment.

Observations from the OTIS Survey. Of the women sur-veyed, 82% (28/34) recalled receiving information aboutthe concerns of using isotretinoin while pregnant. Eighty-nine percent (25/28) of the women who did recall receiv-ing information stated that the information was in the formof a “booklet” from the manufacturer. Other than the pre-scriber or the pharmacist, 24% (8/34) recalled receivinginformation (written or verbal) from other sources (friends,other health care providers, or the Internet).

Data on Pregnancy OutcomesOf the 34 pregnancies followed, 41% (14/34) ended in

elective termination, 9% (3/34) resulted in miscarriage, 3%(1/34) had an ectopic pregnancy, and 35% (12/34) deliv-ered live-born babies. No stillbirths have been reportedthus far. Outcomes are pending for 12% (4/34) of thepregnancies. Of the women who electively terminatedtheir pregnancies, none knew if birth defects were present.Isotretinoin embryopathy was described in 1 of the 12live-born babies whose mother took isotretinoin at doses of40 mg/day for 1 month and 60 mg/day for 2.5 monthsduring the first and second trimesters.

DISCUSSION

The results of the OTIS survey show that the majority ofwomen who participated were treated for less-severe dis-ease than is recommended in the SMART program. Thisfinding is consistent with other studies that found off-labeluse of isotretinoin is common for less-severe disorders(Doering et al., 1992; Wysowski et al., 2002). Some derma-tologists advocate using isotretinoin to treat even mildcases of acne that are unresponsive to standard therapies,not just cystic acne (Layton et al., 1993; Honein et al., 2001).Others believe that treatment with isotretinoin should bestarted in patients with severe acne before scarring occurs.

Seventy-five percent (18/24) of women surveyed indi-cated they were not following the SMART recommenda-tion to use 2 forms of birth control. Furthermore, the ad-visability of using the rhythm method for birth controlwhile taking isotretinoin is questionable. This is particu-larly disconcerting in light of other findings in this surveythat the majority of women surveyed reported they were

885PREGNANCY SURVEY: ISOTRETINOIN USE

Birth Defects Research (Part A) 73:881–887 (2005)

not screened monthly for pregnancy as suggested bySMART, and some women did not recall any testing.

Furthermore, in 31% (5/16) of the women surveyed inthe United States, the yellow sticker was not apparently inplace when a pharmacist filled the prescription.

Use of effective contraceptives and screening for preg-nancy prior to the start of therapy with isotretinoin arecritical elements of any pregnancy prevention program.Despite the widespread availability and use of contracep-tive methods, Henshaw (1998) estimated that 49% of allpregnancies in the United States are unintended. Unin-tended pregnancies are generally the result of failed orimproperly or inconsistently used contraceptives, or no useof contraception at all. Reasons for some sexually activewomen not using contraception are varied, but includelack of access to contraceptives, health concerns aboutcontraceptives, and/or religious or cultural beliefs. Mis-perception of pregnancy risk can lead some women to notuse contraception or choose less-effective contraceptives.

As case report 1 (Fig. 1) illustrates, misinterpretation ofinformation provided by the physician resulted in the pa-tient’s misunderstanding of her risk to become pregnantand, subsequently, exposure of her fetus to isotretinoin.This case highlights the importance of fully informing apatient about her pregnancy risk at a level that she canunderstand. It also illustrates the important role that phy-sicians have in influencing their patient’s perception of riskand the future decisions they make regarding using (or notusing) contraception.

In case report 2 (Fig. 2), failure of the physician to verifya positive pregnancy test when his patient denied beingsexually active resulted in an exposed pregnancy. Bothcases demonstrate the importance of requiring the use of 2forms of contraception and screening for pregnancy beforeisotretinoin is prescribed.

The strengths of this survey include the use of OTIScounselors with their extensive experience in communicat-ing with women about their reproductive concerns. Thesurvey used a detailed, structured interview questionnaireand most intake interviews were completed within 3months of exposure, before fetal outcome was known.

Survey limitations include the small number of womensurveyed, which limits interpretation. Women who call aTIS and agree to participate in a survey may not be repre-sentative of the general public. However, OTIS callers pro-vide a unique and otherwise unavailable resource for iden-tifying women who had both failed the pregnancyprevention program and were not captured as part of themanufacturer’s survey.

The survey responses are based on patient self-reportingand may be subject to recall error. Nonetheless, womenwere surveyed prior to the outcome of the pregnancy, sowe believe the responses provided in this report can beuseful in elucidating important factors that hinder thesuccess of pregnancy prevention programs.

The FDA Joint Drug Safety and Risk Management andDermatologic and Ophthalmic Drugs Advisory Commit-tees met in February 2004 to address concerns that volun-tary safeguards in current programs had failed to reducethe number of isotretinoin-exposed pregnancies (Honein etal., 2004). As result, the Advisory Committee recom-mended replacement of the current SMART program witha stricter plan for all brands of isotretinoin. In November2004, it was announced that 4 manufacturers of isotretinoinhave contracted with Covance Inc. (Princeton, NJ) to de-sign, build and manage a new pregnancy risk programhttp://www.rocheusa.com/newsroom/current/2004/pr2004112301.html.

Critical to a new program is mandatory registration ofprescribers, patients, and pharmacists in a single central-ized registry for purposes of follow-up and evaluation, asis done in the thalidomide program called The System forThalidomide Education and Prescribing Safety (STEPS).Although it is important to note differences in the womenwho take isotretinoin and those who take thalidomide,only 1 pregnancy has been reported, thus far, with theSTEPS program (Honein, et al., 2004). It is hoped that therequirements contained in a stricter program will be suffi-cient to reduce (or possibly prevent) fetal exposures.

ACKNOWLEDGMENTSThe authors express their appreciation to Edward Lam-

mer, Children’s Hospital Research Institute, Oakland, CA;Nancy Rose, Intermountain Health Care, Salt Lake City,UT; and the Teratogen Information Services and GeneticCounselors who sent referrals to our survey. Thanks toJane Adams, University of Massachusetts, Boston, MA;Margaret Honein, Centers for Disease Control and Preven-tion (CDC), Atlanta, GA; James Mills, the National Insti-tute of Child Health and Human Development, the Na-tional Institutes of Health , US Department of Health andHuman Services, Bethesda, MD; Allan A. Mitchell, SloneEpidemiology Center, Boston University, Boston, MA;Cynthia Moore, CDC, Atlanta, GA; Carla Van Bennekom,Slone Epidemiology Center, Boston University, Boston,MA, Martha Werler, Slone Epidemiology Center, BostonUniversity, Boston, MA and the OTIS Research Committeefor their help in creation of the survey questionnaire:Chair: Gerald G. Briggs, Long Beach Memorial Medical

Figure 1. Case Report: Mrs. L., a pregnant woman, calledthe TIS and reported that during the removal of her intra-uterine device (IUD), she heard her doctor say that for ayear or so after removal of the IUD, it may be “harder” forher to become pregnant. She interpreted this statement tomean that she could not become pregnant and could takeisotretinoin without using any form of birth control. Shebecame pregnant within 3 months after removal of the IUDand 1 month of starting isotretinoin therapy. Mrs. L. de-cided to terminate the pregnancy.

Figure 2. Case Report: Ms. Z., a 19-year-old, called a TIS tolearn about potential problems that her baby might havebecause of exposure to isotretinoin. She reported that shereceived the news of her pregnancy during a telephone callfrom her dermatologist. She had never taken isotretinoinbefore but wanted to clear up her severe acne. Because hermother was standing in the same room while she was onthe telephone, she reported to her dermatologist that shewas not sexually active and the test MUST be incorrect (shenow realizes that she was in denial). She recalls that herdermatologist responded with “Oh, I guess I’ve misreadthe test results.” She was then given a prescription forisotretinoin, which she began taking the next day. Thepregnancy test was correct; she was pregnant. Ms. Z. de-cided to terminate the pregnancy.

886 ROBERTSON ET AL.

Birth Defects Research (Part A) 73:881–887 (2005)

Center, Long Beach, CA; Adrienne Einarson, University ofToronto, Toronto, Ontario, Canada; Anick Berard, Ste. Jus-tine Hospital, Montreal, Quebec, Canada; Offie P. Soldin,Georgetown University, Washington, D.C.; StephanieWan, Long Beach Memorial Medical Center, Long Beach,CA; Steve Lamm, Center for Environmental and Occupa-tional Health, Washington, D.C.; Coordinators: BethConover, University of Nebraska, Omaha, NE; Carrie L.Chou, University of Missouri-Columbia, Columbia, MO;Kelly Kao, University of California, San Diego MedicalCenter, San Diego, CA; Leslie Evans, Fullerton GeneticsCenter, Asheville, NC; Lori Wolfe, University of NorthTexas, Denton, TX; Mara Gaudette, Illinois Teratogen In-formation, Chicago, IL; Mary Riske, University of NorthDakota, Grand Forks, ND; Myla Moretti, University ofToronto, Toronto, Ontario, Canada; and Shanon Kolb,Pregnancy Riskline Network, Binghamton, NY. Dysmor-phologists: Stephen Braddock, University of Missouri-Co-lumbia, Columbia, MO; Kenneth L. Jones, University ofCalifornia, San Diego Medical Center, San Diego, CA;Luther Robinson, University of Buffalo, Buffalo, NY. Weare indebted to Elizabeth Balken, Marsha Leen-Mitchell,Mary Ness and Jenny Starks with the Utah PregnancyRiskLine for their never-ending support of this project.

REFERENCESAdams J, Lammer EJ. 1991. Relationship between dysmorphology and

neuropsychological function in children exposed to isotretinoin “inutero”. In: Fujii T, Boer GJ, editors. Functional neuroteratology of shortterm exposure to drugs. Tokyo: Teikyo University Press. pp. 159–170.

Atanackovic G, Koren G. 1999. Fetal exposure to isotretinoin: failure tocomply with the Pregnancy Prevention Program. CMAJ 160:1719–1720.

Brinker A, Trontell A, Beitz J. 2002. Pregnancy and pregnancy rates inassociation with isotretinoin (Accutane). J Am Acad Dermatol 47:798–799.

Carver V, Coyle B, Koren G, et al. 2001. A call for action: prevention of fetalexposure to isotretinoin. A position paper by the Organization ofTeratology Information Services public affairs committee. Reprod Toxi-col 15:729.

Cooper AJ. 2003. Australian Roaccutane Advisory Board. Treatment of acnewith isotretinoin: recommendations based on Australian experience.Aust J Dermatol 44:97–105.

Dai WS, LaBraico JM, Stern RS. 1992. Epidemiology of isotretinoin exposureduring pregnancy. J Am Acad Dermatol 26:599–606.

Doering PL, Araujo OE, Frohnaple DJ, et al. 1992. Patterns of prescribingisotretinoin: focus on women of childbearing potential. Ann Pharma-cother 26:155–161.

Goldsmith LA, Bolognia JL, Callen JP, et al. 2004. American Academy ofDermatology. American Academy of Dermatology Consensus Confer-

ence on the safe and optimal use of isotretinoin: summary and recom-mendations. J Am Acad Dermatol 50:900–906.

Henshaw SK. 1998. Unintended pregnancy in the United States. Fam PlannPerspect 30:24–29.

Hermes B, Praetel C, Henz BM. 1998. Medium dose isotretinoin for thetreatment of acne. J Eur Acad Dermatol Venereol 11:117–121.

Honein MA, Paulozii LJ, Erickson JD. 2001. Continued occurrence of Accu-tane-exposed pregnancies. Teratology 64:142–147.

Honein MA, Moore CA, Erickson JD. 2004. Can we ensure the safe use ofknown teratogens? Drug Saf 27:1069–1080.

Jones KL, Adams J, Chambers CD, et al. 2001. Isotretinoin and pregnancy[Letter]. JAMA 285:2079–2080.

Koren G, Avner A, Shear N. 2004. Generic isotretinoin: a new risk forunborn children. CMAJ 170:1567–1568.

Lammer EJ, Chen DT, Hoar RM, et al. 1985. Retinoic acid embryopathy.N Engl J Med 313:837–841.

Lammer EJ, Hayes EM, Schunior A, Holmes LB. 1987. Risk for majormalformation among human fetuses exposed to isotretinoin (13-cis-retinoic acid). Teratology 35:68A.

Lavigne SV, Rosengren S. 2000. Prenatal accutane use: the ongoing chal-lenge to prevent fetal exposure. Teratology 6:520–521.

Layton AM, Knaggs H, Taylor J, Cunliffe WJ. 1993. Isotretinoin for acnevulgaris—10 years later: a safe and successful treatment. Br J Dermatol129:292–296.

Leen-Mitchell M, Martinez L, Gallegos S, et al. 2000. Mini-review: history oforganized teratology information services in North America. Teratol-ogy 61:314–317.

Lowenstein EJ. 2002. Isotretinoin made S.M.A.R.T. and simple. Cutis 70:115–120.

Mitchell AA, Van Bennekom CM, Louik C. 1995. A pregnancy-preventionprogram in women of childbearing age receiving isotretinoin. N EnglJ Med 333:101–106.

O’Donnell J. 2003. Overview of existing research and information linkingisotretinoin (Accutane), depression, psychosis, and suicide. Am J Ther10:148–159.

Perry MD, McEvoy GK. 1983. Isotretinoin: new therapy for severe acne. ClinPharm 2:12–19.

Robertson JA, Martinez LP, Gallegos S, et al. 2002. Accutane cases: ateratogen information service’s approach [Letter]. Teratology 66:1–2.

Shahidullah M, Tham SN, Goh CL. 1994. Isotretinoin therapy in acnevulgaris: a 10-year retrospective study in Singapore. Int J Dermatol33:60–63.

Uhl K. 2004. Department of Health and Human Services, Food and DrugAdministration, Center for Biologics Evaluation and Research, DrugSafety and Risk Management Advisory Committee, in Joint Sessionwith the Dermatologic and Ophthalmic Drugs Advisory Committee.Proceedings of the FDA Joint Drug Safety and Risk Management andDermatologic and Opthalmic Drugs Advisory Committee Meeting,Gaithersburg, MD, February 27, 2004. Available at: http://www.fda.gov/ohrms/dockets/ac/04/transcripts/4017T2.htm. Accessed: Sep-tember 3, 2005.

Wysowski DK, Swann J, Vega A. 2002. Use of isotretinoin (Accutane) in theUnited States: Rapid increase from 1992 through 2000. J Am AcadDermatol 46:505–509.

887PREGNANCY SURVEY: ISOTRETINOIN USE

Birth Defects Research (Part A) 73:881–887 (2005)