7/27/2019 02-077

1/5

624

Maternal Drug Use and Infant Cleft Lip/Palate With Special Reference

to Corticoids

BENGT KALLEN, M.D., PH.D.

Objective:To study the association between maternal drug use in early preg-

nancy and orofacial cleft in the infant.Design: Register analysis based on prospectively collected information.

Patients:All delivered women in Sweden July 1, 1995, through December 31,

2001.

Main outcome measure: Presence of orofacial cleft in infant.

Results: Prospective information on maternal drug use during the first tri-

mester, as reported in early pregnancy, was studied in 1142 infants with oro-

facial clefts, isolated or with other malformations, excluding chromosome

anomalies. Any drug use was not associated with clefts (odds ratio [OR]

0.98, 95% confidence interval [95% CI] 0.85 to 1.13), with isolated clefts (OR

0.92) with isolated median cleft palate (OR 1.03, 95% CI 0.79 to 1.36) or

with isolated cleft lip with or without cleft palate (OR 0.86, 95% CI 0.71 to

1.05). Reported use of multivitamins, folic acid, or B12 was not associated with

a decrease in orofacial cleft risk (OR 1.00, 95% CI 0.63 to 1.52). ORs above

2 were seen for some drugs: sulfasalazine, naproxen, and anticonvulsants, butonly a few exposed cases occurred. An association between glucocorticoid

use and infant cleft was indicated and seemed to be strongest for median cleft

palate.

Conclusion:Maternal drug use seems to play only a small role for the origin

of orofacial clefts, at least in Sweden.

KEY WORDS: cleft lip, cleft palate, drugs, glucocorticoid, naproxen, sulfasala-

zine, vitamins

Orofacial clefts are relatively common congenital malfor-

mations. In Sweden the rate is about 1:500 births, which is ahigh rate, compared with that in many other populations. Rob-

ert et al. (1996) reported rates of 19.2 per 10,000 in Sweden,

9.9 in France, and 16.1 among whites in California. Some

orofacial clefts appear in association with other congenital

malformations, but the majority is isolated, nonsyndromic. Nu-

merous studies on the etiology of the clefts have been per-

formed. A clear-cut genetic factor is involved, but nongenetic

factors are also of importance. Wyszynski and Beaty (1996)

proposed a number of environmental factors that may be of

importance for the origin of clefts, including cigarette smok-

ing, multivitamins, anticonvulsants, organic solvents, agricul-

tural chemicals, common cold, and alcohol consumption.

Many articles have been published trying to identify variousdrugs as a cause of orofacial clefts, and it has also been sug-

gested that vitamins and notably folic acid may have a protec-

tive effect.

The purpose of the present study was to investigate the im-

pact of maternal drug use in the origin of orofacial clefts based

Dr. Kallen is a Professor, Tornblad Institute, University of Lund, Sweden.

Submitted June 2002; Accepted March 2003.

Address correspondence to: Professor Bengt Kallen, Tornblad Institute, Bis-

kopsgatan 7, SE-223 62 Lund, Sweden. E-mail embryol@embryol.lu.se.

on prospectively collected data to eliminate recall bias. The

primary purpose was to see whether maternal drug use wouldbe a serious confounder in studies of other exposures but also

to search for associations between specific drugs and orofacial

clefts.

MATERIAL AND METHODS

The Swedish Medical Birth Registry (Cnattingius et al.,

1990) is based on copies of medical documents concerning

antenatal care, delivery care, and the pediatric examination of

the newborn. It covers all of Sweden, although 1% to 2% of

deliveries are missing in the register. Data from the antenatal

care contain information obtained by the attending midwife atthe first antenatal care visit (usually week 10 to 12). Among

other things, maternal smoking habits are recorded (0, 10,

and 10 cigarettes/day) and since July 1994, names of drugs

that the woman states she has used during pregnancy before

the antenatal visit. By and large, this represents first-trimester

exposures. The drug names are then transformed into Anatom-

ical Therapeutic Chemical Classification (ATC) codes and

stored in the register.

From the Medical Birth Register, infants with a diagnosis

of an orofacial cleft (but without a chromosome anomaly di-

agnosis) were identified. This data set was supplemented with

7/27/2019 02-077

2/5

Kallen, MATERNAL DRUGS AND OROFACIAL CLEFTS 625

TABLE 1 Types of Orofacial Clefts and Presence of Other

Congenital Malformations, With Known Chromosome Anomalies

Excluded

Other

Malformations

Cleft

Palate

Cleft

Lip

Cleft

Lip/Palate Total

None

With holoprosencephaly

Other syndromeOne other major

Two or more other major

Other minor

274

4

3445

15

16

249

1

36

5

7

334

5

519

14

8

857

10

4270

34

31

Total 388 271 385 1044

cases reported to the Swedish Registry of Congenital Malfor-

mations (Kallen, 1987) and also (for the period 1995 to 2000)

with infants who were identified from the Hospital Discharge

Registry with an orofacial cleft diagnosis. Only infants present

in the Medical Birth Registry were included in the study. In-

fants immigrating after birth or not identifiable in the Medical

Birth Registry were not included.The study was restricted to infants born from July 1995

through December 2001.

Drug use in women who had an infant with an identified

orofacial cleft was compared with drug use in all women reg-

istered in the Medical Birth Registry. Comparisons were made

as Mantel-Haenszel odds ratio (OR) estimates, stratifying for

year of birth, maternal age (5-year class), parity (1, 2, 3, 4),

smoking habits (unknown, 0, 10, 10 cigarettes/day), and

period of involuntary childlessness (0, 1, 2, 3, 4, 5 years).

Confidence intervals (95% CI) were estimated with Miettinens

method. For individual drugs (when exposures were few) the

expected number was calculated, stratified as above, and com-

pared with the observed number as a risk ratio (RR; observed/

expected) with 95% CIs based on exact Poisson distributions.

RESULTS

One thousand forty-four infants with an orofacial cleft and

no known chromosome anomaly were identified. The total

number of births in the Medical Birth Registry for the studied

period was 576,873. The orofacial cleft rate (excluding known

chromosomal cases) was thus 18.1 per 10,000 births.

Table 1 shows the distribution of type of orofacial cleft and

the presence of other malformations. Infants with syndrome

diagnoses were not excluded from the study; however, they

were few in number (about 5%) and the quality of the clinical

syndrome diagnoses sometimes questionable. It is also possible

that a drug exposure may modify the phenotypic expression

of a syndrome.

Drugs were reported by 261 mothers of infants with an oro-

facial cleft (25.0%) and by 149,932 of all women (26.0%).

The OR (stratified for year of birth, maternal age, parity, smok-

ing in early pregnancy, and period of involuntary childless-

ness) for having used any drug in early pregnancy when the

infant had an orofacial cleft was 0.99 (95% CI 0.86 to 1.14).

The corresponding OR for isolated clefts was 0.92 (95% CI

0.79 to 1.08), isolated cleft palate was 1.05 (95% CI 0.80

to 1.39), and cleft lip/palate was 0.86 (95% CI 0.71 to 1.05).

The latter two ORs did not differ significantly (z 1.20, p

.12).

Table 2 shows the distribution of the observed numbers of

specific reported groups of drugs or specific drugs and for each

one the expected number and the RR with 95% CI. For mostdrugs or drug groups, the number of exposures was low and

confidence intervals large.

For only one of the drugs or drug groups tabulated in Table

2 (naproxen) did a formal statistical significance appear.

Among the eight patients, four reported no further drug use,

one had been exposed for systemic glucocorticoid and sulfa-

salazine, two for antibiotics, and one for an antitussive.

High RRs were found for some additional exposures: sul-

fasalazine, topical glucocorticoids, systemic glucocorticoids,

and anticonvulsants. It can be noted that the reported use of

vitamins showed no protective effect. The estimated RR was

1.00, multivitamins had an RR below, and folic acid and B12

an RR above unity, neither statistically significant.The high RR for sulfasalazine was based on only three ex-

posures because of maternal ulcerative colitis or Crohns dis-

ease. One of these infants had also been exposed to glucocor-

ticoids and naproxen.

Only five infants with clefts had been exposed to anticon-

vulsants. Two of the infants were exposed to valproic acid (one

of the mothers reported use of folic acid), two to carbamaze-

pine (one of them also to terbutaline), and one infant was

exposed to lamotrigine (the mother also reported the use of

B12

, folic acid, and vitamin B6).

Corticoid exposure could have occurred by systemic admin-

istration or topical administration as a dermatological prepa-

ration, nose drops, or an inhalation antiasthmatic. Table 3

shows the distribution of reported uses.

If all types of glucocorticoid exposure and clefts were stud-

ied together, a 44% risk increase was seen (Table 4). It seemed

to be stronger for infants with other malformations than for

infants with only a cleft and higher for median cleft palate

than for cleft lip with or without cleft palate, but the observed

differences (RR 1.94 versus 1.32 and 1.70 versus 1.26, re-

spectively) could well be random. For isolated cleft lip/palate,

there seemed to be no risk increase, but the low number of

such clefts resulted in a large confidence interval.

DISCUSSION

Information on maternal drug use during pregnancy is usu-

ally obtained retrospectively with the possibility of recall or

interviewer bias. In the present study, drug information was

obtained from the women. Because it was obtained prospec-

tively, possible information errors were unrelated to pregnancy

outcome and could therefore not bias the results. An alternative

was the use information linked from prescription registers, but

then the actual use of the drugs during early pregnancy be-

comes uncertain and no information will be obtained on over-

the-counter drugs.

7/27/2019 02-077

3/5

626 Cleft PalateCraniofacial Journal, November 2003, Vol. 40 No. 6

TABLE 2 Observed (Obs) and Expected (Exp) Numbers of Specific Drug Exposures of Infants With Any Orofacial Cleft*

Drug (Group)

Number Exposed

Pop CP CLP

All

Obs Exp

All

RR 95% CI

Ulcus drugs

Sulfasalazine

Insulin

VitaminsMultivitamins

Folic acid

B12

1425

515

1646

130165820

7018

1208

1

1

3

102

7

3

2

2

1

126

8

1

3

3

4

228

15

4

2.3

1.0

3.0

21.99.4

12.0

2.3

1.28

3.00

1.34

1.000.85

1.25

1.75

0.263.75

0.628.77

0.373.44

0.631.520.371.68

0.702.07

0.484.47

Antihypertensives

Topical glucocorticoid

Oral contraceptive

Fertility drugs

Systemic glucocorticoids

1996

1094

1878

1506

2050

2

1

2

1

2

2

3

4

2

5

4

4

6

3

7

3.4

2.0

3.3

2.8

3.6

1.18

2.01

1.82

1.07

1.94

0.323.03

0.555.15

0.673.96

0.223.12

0.783.99

Thyroxine

Antibiotics

Penicillin V

Nonsteroidal anti-

inflammatory drugs

Naproxene

4951

18494

7923

7698

1679

7

10

3

5

3

4

19

11

9

5

11

29

14

14

8

8.8

33.2

14.3

12.9

2.9

1.24

0.87

0.98

1.09

2.72

0.622.23

0.591.26

0.531.64

0.591.82

1.175.36

Analgesics

Acetyl salicylic acid

ParacetamolAnticonvulsants

Sedatives/hypnotics

Bensodiazepine

43750

5920

366261370

3513

1009

29

5

254

1

0

41

2

371

4

1

70

7

625

5

1

77.3

10.4

64.82.3

5.7

1.7

0.91

0.67

0.962.18

0.88

0.59

0.711.14

0.271.39

0.731.230.715.10

0.282.05

0.012.59

Nose drops

With glucocorticoids

Antiasthmatics

Inhaled glucocorticoids

Antihistamines

Meclozine

Antitussives

5365

2872

16582

7404

29155

16821

2088

6

4

20

9

22

12

3

7

3

18

7

26

14

4

13

7

38

16

48

26

7

9.3

5.1

29.7

13.4

52.0

29.7

4.6

1.39

1.39

1.28

1.19

0.92

0.88

1.51

0.742.39

0.562.86

0.911.76

0.681.94

0.681.22

0.571.28

0.613.11

* Risk ratios (RR observed/expected) with exact 95% confidence intervals (95% CI) from Poisson distributions. Number of exposed infants given for population (Pop), for isolated cleft palate

(CP) and cleft lip with or without cleft palate (CLP).

TABLE 3 Distribution of Reported Use of Glucocorticoid Drugs

According to Administration Mode When Infant Had an

Orofacial Cleft

Administration Mode Number

Only systemic

Only dermatological

Only nose drop

Only inhalation

Systemic and dermatological

Nose drop and inhalation

6

3

6

15

1

1

Total 32

TABLE 4 Observed (Obs) and Expected (Exp) Numbers ofInfants With Orofacial Cleft After Exposure to Any

Glucocorticoid Preparation*

Cleft Category Obs Exp RR 95% CI

All clefts 32 22.3 1.44 0.982.03

Isolated clefts

Clefts with other malformations

All cleft palate

Isolated cleft palate

24

8

15

11

18.2

4.2

8.8

6.2

1.32

1.93

1.70

1.76

0.851.97

0.833.80

0.952.80

0.883.15

All cleft lip/palate

Isolated cleft lip/palate

17

13

13.5

11.9

1.26

1.09

0.732.02

0.581.87

* Observed/expected quotient (RR) with 95% confidence intervals (95% CI) are given, based

on exact Poisson distributions.

The present study was basically a case-control study. Drug

exposure was compared between infants with orofacial cleftsand all other infants. As such, it was a relatively large study

of the association between maternal drug use and orofacial

clefts, comprising 1114 patients and a very large number of

controls. Because the total number of exposures for each drug/

drug group is given in Table 2, risk estimates as from a cohort

study can be made. Most of the tabulated drugs or drug groups

contained more than 1000 exposures, the only exceptions be-

ing sulfasalazine and B12

.

The outcome of this study is basically negative. No asso-

ciation was found between general drug use and the occurrence

of orofacial clefts. The OR was close to 1 and the upper con-

fidence limit was 1.08 when year of birth, maternal age, parity,

smoking, and period of involuntary childlessness are consid-ered. It seems that drugs are not a major contributor to oro-

facial clefts in Sweden. This contradicts other studies, based

on retrospective exposure information in case-control designs

in which associations have been found with many drug groups.

A weakness in the present study is that the number of pa-

tients exposed to specific drugs was often low. To some extent

this was due to an underreporting of drug use in early preg-

nancy. The amount of underreporting is difficult to estimate

and may well vary among different drugs. So, for instance,

only 2.2 women per 1000 reported the use of anticonvulsants

7/27/2019 02-077

4/5

Kallen, MATERNAL DRUGS AND OROFACIAL CLEFTS 627

and probably 3/1000 have epilepsy in early pregnancy. For

drugs like multivitamins, underreporting is certainly larger.

The loss of data was, however, unbiased by the outcome be-

cause the information was obtained in early pregnancy, long

before the presence of a facial cleft was known. For drugs that

are relatively seldom used, data loss would hardly affect risk

estimates, but for very frequently used drugs (e.g., multivita-mins and over-the-counter analgesics), loss of data could bias

the risk ratio estimates toward 1.0. In the present study, only

about 2% of the women reported vitamin use.

In a questionnaire study performed during the same period

(Ericson et al., 2001c), 18% of the women stated the use of

vitamins in early pregnancy, indicating that only about 10%

of the women reported (or the midwives recorded) their vita-

min use. If the RR for orofacial cleft was 0.50 after full as-

certainment of vitamin use (18%), it would increase to 0.55 if

only 2% exposure is ascertained. As the estimated lower 95%

CI of the RR after the use of vitamins was 0.61 and after the

use of folic acid 0.70, it is unlikely that this data loss could

have hidden a marked protective effect of vitamins includingfolic acid. Such an effect was identified in retrospective case-

control studies (e.g., Shaw et al., 1995; Itikala et al., 2001) but

not in others (Hayes et al., 1996). In a study from Hungary

(Czeizel et al., 1999), a protective effect was found only for

high doses of folic acid (6 mg/day) but not doses of1 mg/

day, which will correspond to the content of multivitamin tab-

lets and the 0.4 mg folic acid tablets used by pregnant women.

Folic acid is available in 5-mg doses on prescription in Swe-

den, used periconceptionally under special conditions such as

previous neural tube defect fetus, together with anticonvul-

sants, and sometimes after in vitro fertilization. The vast ma-

jority of exposures concern 400-g tablets, recommended for

periconceptional use to reduce the risk for a neural tube defect.

Even though no general association could be seen between

drug use and the occurrence of orofacial clefts, a specific drug

may well cause such malformations, notably if such a drug is

rarely used or has a low teratogenic potential. Maternal use of

anticonvulsants and notably phenytoin or phenobarbital (Bossi,

1983; Kallen et al., 1989; Arpino et al., 2000) has been as-

sociated with a markedly increased risk for orofacial clefts. In

the present study, only five infants with orofacial clefts had

been exposed to anticonvulsants, and the OR of 2.2 was not

statistically significant. Phenobarbital as a sole anticonvulsant

is seldom used in Sweden. Only nine women reported the use

of phenobarbital or primidone and 90 the use of phenytoin in

monotherapy. Even an eightfold increased rate of orofacial

clefts (as has been suggested) would mean that only one to

two infants with clefts were born after exposure to one of these

drugs, and none was found.

Sedatives/hypnotics and notably bensodiazepines have been

associated with orofacial clefts, but no such effect was found

in the present study, even though only few exposures occurred

and the upper confidence limit for any sedative was 1.87. The

meta-analysis of Dolovich et al. (1998) found an association

in retrospectively performed studies (OR 1.79, 95% CI

1.13 to 2.82) but not in studies in which exposure data were

collected prospectively (OR 1.19, 95% CI 0.34 to 4.15).

It is possible that the association found in retrospective case-

control studies is due to recall or interviewer bias.

We found no association between orofacial clefts and fertil-

ity drugs, but the numbers were low; only three women whose

infants had a cleft reported the use of such drugs. In a previous

study of delivery outcome after in vitro fertilization, Ericsonand Kallen (2001a) found 22 infants with cleft, compared with

the expected number of 16.5 (RR 1.3, 95% CI 0.8 to 1.9),

and in a study of infants born after ovulation stimulation but

without in vitro fertilization, 11 infants with orofacial clefts

were found, compared with the expected number of 9 (Kallen

et al., 2002). In the article by Long et al. (1992), 2.4% of

control women and 4.9% of women with an infant with an

orofacial cleft reported an induced pregnancy. Our studies do

not support the observation by Long et al. but cannot exclude

an association between fertility drugs and orofacial clefts.

The association between naproxen use and orofacial clefts

is the only one showing formal statistical significance. It may

still be a random finding because so many exposures werestudied. Ericson and Kallen (2001b) noted previously that

among 918 infants exposed to naproxen, five had an orofacial

cleft. These patients were included in the present material that

now comprises 1679 exposures and 8 infants with clefts. Thus,

among 761 added exposed infants, three had a cleft, whereas

1.3 would be expected if no relationship existed, and 4.1 would

be expected if the association found in the first sample was

true. Further observation is needed to determine whether this

association is random.

Systemic glucocorticoid treatment has been related to oro-

facial clefts. This association was suspected based on animal

data in which glucocorticoids have caused cleft palate in var-

ious species since the pioneering work of Baxter and Fraser

(1950). A Spanish case-control study (Rodriguez-Pinilla and

Martnez-Fras, 1998) found an association between maternal

systemic use of glucocorticoids and the birth of an infant with

a cleft lip/palate (OR 6.55, 95% CI 1.44 to 27.9), based

on five exposed patients, one of which had multiple malfor-

mations and may have been a trisomy 13. Similar results were

found in a California case-control study (Carmichael and

Shaw, 1999) for cleft lip/palate (OR 4.3, 95% CI 1.1 to

17.2) and isolated cleft palate (OR 5.3, 95% CI 1.1 to

26.5). The wide confidence intervals indicate that these esti-

mates are based on small numbers of exposed patients, six cleft

lip/palate and three median cleft palate. A meta-analysis (Park-

wyllie et al., 2000) estimated a common OR of 3.35 (95% CI

1.97 to 5.69) for case-control studies examining oral clefts

and maternal use of glucocorticoids. In a small prospective

study of data on 184 women exposed to prednisone in preg-

nancy, no increase in major malformations was noted, the sam-

ple size was too small to study the possible effect on orofacial

clefts.

Our data on corticoid exposure and orofacial clefts are

somewhat difficult to interpret. If all types of exposures are

analyzed together, a nearly significant risk increase with an OR

of 1.43 was found, perhaps more pronounced for median cleft

7/27/2019 02-077

5/5

628 Cleft PalateCraniofacial Journal, November 2003, Vol. 40 No. 6

palate than cleft lip/palate. These estimates are based on 15

and 17 patients, respectively. For isolated cleft/palate, no effect

was seen. When different modes of administration were com-

pared (see Table 2), high ORs were found for systemic ad-

ministration, dermatologic use, and administration as nose

drops or via inhalation. The last-mentioned mode of adminis-

tration had the lowest OR, only 1.19. All ORs are based onlow numbers and they may all estimate the same OR of 1.43.

Czeisel and Rockenbauer (1997) also found an association be-

tween nonsystemic administration of glucocorticoids and oro-

facial clefts. It seems reasonable, however, that the mode of

administration should be of importance for the possible effect

because only small amounts are absorbed when the drug is not

given systematically. Such a dilution of the present material

would result in a reduction of the OR estimate.

In conclusion, at least in Sweden, maternal drug use in early

pregnancy is not a major contributor to orofacial clefting and

will not act as a serious confounder in studies of other expo-

sures. This does not exclude, however, that some specific drugs

may increase the cleft risk, and systemic glucocorticoids seemto be the most likely ones among these.

Acknowledgments. Access to the Swedish health registers was given by the

National Board of Health and Welfare, Stockholm. The study was supported

by a grant from KA Wallenberg Foundation.

REFERENCES

Arpino C, Brescianini S, Robert E, Castilla EE, Cocchi G, Cornel MC, de Vigan

C, Lancaster PA, Merlob P, Sumiyoshi Y, Zampino G, Renzi C, Rosano A,

Mastroiacovo P. Teratogenic effects of antiepileptic drugs: use of an inter-

national database on malformations and drugs exposure. Epilepsia. 2000;

41:14361443.

Baxter H, Fraser FC. Production of congenital defects in offspring of female

mice treated with cortisone. Mcgill Med J. 1950;19:245249.

Bossi L. Fetal effects of anticonvulsants. In: Morsell PL, Pippenger CE, Pentry

JK, eds. Antiepileptic Drug Therapy in Pediatrics. New York: Raven Press;

1983:3764.

Carmichael SL, Shaw GM. Maternal corticosteroid use and risk of selected

congenital anomalies. Am J Med Genet. 1999;86:242244.

Cnattingius S, Ericson A, Gunnarskog J, Kallen B. A quality study of a medical

birth registry. Scand J Soc Med. 1990;18:15431548.

Czeizel AE, Timar L, Sarkozi A. Dose-dependent effects of folic acid on the

prevention of orofacial clefts. Pediatrics. 1999;104:E661E667.

Czeizel A, Rockenbauer M. Population-based case-control study of teratogenic

potential of corticosteroids. Teratology. 1997;56:335340.

Dolovich LR, Addis A, Vaillancourt JMR, Power JDB, Koren G, Einarson TR.

Benzodiazepine use in pregnancy and major malformations or oral clefts

meta-analysis of cohort and case-control studies. Br Med J. 1998;317:839

843.

Ericson A, Kallen B. Congenital malformations in infants born after IVF: a

population-based study. Hum Reprod. 2001a;16:504509.

Ericson A, Kallen B. Non-steroidal anti-inflammatory drugs in early pregnancy.

Reprod Toxicol. 2001b;15:371375.

Ericson A, Kallen B. Use of multivitamins and folic acid in early pregnancy

and multiple births in Sweden. Twin Res. 2001c;4:6366.

Hayes C, Werler MM, Willett WC, Mitchell AA. Case-control study of peri-

conceptional colic acid supplementation and oral clefts. Am J Epidemiol.

1996;143:12291234.

Itikala PR, Watkins ML, Mulinare J, Moore CA, Liu Y. Maternal multivitamin

use and orofacial clefts in offspring. Teratology. 2001;63:7986.

Kallen B. Search for teratogenic risks with the aid of malformation registries.

Teratology. 1987;52:1529.

Kallen B, Otterblad Olausson P, Nygren KG. Neonatal outcome in pregnanciesfrom ovarian stimulation. Obstet Gynecol. 2002;100:414419.

Kallen B, Robert E, Mastroiacovo P, Martnez-Fras ML, Castilla EE, Cocchi

G. Anticonvulsant drugs and malformations. Is there a drug specificity? Eur

J Epidemiol. 1989;5:3136.

Long S, Robert E, Laumon B, Pradat E, Robert JM. E pidemiologie des fentes

labiales et palatines das la region Rhone-Alpes/Auvergne/Jura. A propos de

903 cas enregistres entre 1978 et 1987. Pediatrie. 1992;47:133140.

Parkwyllie L, Mazotta P, Pastuszak A, Moretti ME, Beique L, Hunnisett L,

Friesen HH, Jacobson S, Kasapinovic S, Chang D, Diavcitrin O, Chitayat

D, Nulman I, Einarson TR, Koren G. Birth defects after maternal exposure

to corticosteroids: prospective cohort study and meta-analysis of epidemi-

ological studies. Teratology. 2000;62:385392.

Robert E, Kallen B, Harris J. The epidemiology of orofacial clefts. 1. Some

general epidemiological characteristics. J Craniofac Genet Dev Biol. 1996;

16:234241.

Rodriquez-Pinilla E, Martnez-Fras ML. Corticosteroids during pregnancy and

oral clefts: a case-control study. Teratology. 1998;58:25.

Shaw GM, Lammer EJ, Wasserman CR, OMalley CD, Tolarova MM. Risks

of orofacial clefts in children born to women using multivitamins containing

folic acid preconceptionally. Lancet. 1995;346:393396.

Wyszynski DF, Beaty TH. Review of the role of potential teratogens in the

origin of human nonsyndromic oral clefts. Teratology. 1996;53:309317.