American Journal of Medical Genetics 35:574-578 (1990)
Distal Phalangeal Hypoplasia in Children With Prenatal Phenytoin Exposure: Results of a Controlled Anthropometric Study Eija Gaily Children's Hospital, University of Helsinki, and Children$ Castle Hospital, Helsinki, Finland Phalangeal and metacarpal bone length was measured from hand radiographs in 111children of epileptic mothers and 96 control children of nonepileptic mothers. Seventy-six children of the study group had been exposed to phenytoin in the first 20 weeks of pregnancy, 21 had been exposed to other antiepileptic drugs excluding phenytoin, and 14 had not been exposed. Distal phalangeal lengths were significantly reduced in phenytoin-exposed children. The second and fifth digits were most affected. Phenytoin exposure was associated with a significantly elevated prevalence (Ilk) of radiologically defined distal phalangeal hypoplasia. "he subgroup of children exposed to phenytoin levels over 40 pmolA showed more prominent effects than did the subgroup exposed to lower or unknown concentrations. These results confirm that early fetal exposure to phenytoin decreases distal phalangeal size, as suggested by several previous studies relying on clinical examination only. Distal phalangeal hypoplasia was not accompanied by other serious abnormalities.
ent studies [Andermann et al., 1982;Rating et al., 1982; Gaily et al., 1988al. Radiological anthropometry offers a more discrete and objective method of measuring the degree of distal digital hypoplasia than does the clinical examination of phalangeal size and dermatoglyphic patterning. There is one previous uncontrolled study [Kelly, 19841, where distal phalangeal hypoplasia was diagnosed on the measurements of hand radiographs. In that study, distal phalangeal hypoplasia was observed in 32% of 47 phenytoin-exposed children and in only one (10%)of ten children not exposed to phenytoin. The aims of the present study were 1)to investigate the effect of prenatal phenytoin exposure on distal phalangeal length measured from hand radiographs, and 2) to estimate the prevalence of radiologically diagnosed distal phalangeal hypoplasia among phenytoin-exposed and nonexposed children. SUBJECTS AND METHODS From December 1975 to December 1979,140epileptic mothers were enrolled in a prospective study during 153 pregnancies at the Obstetric Clinic of the Helsinki University Hospital. The 148 surviving children from these pregnancies were included in the present study. They represent approximately 75% of the children born to epileptic mothers in the Uusimaa province in southern KEY WORDS: maternal epilepsy, phenytoin, Finland during the study period [Gaily et al., 1988bl. teratogenicity The control group comprises 105children of nonepileptic mothers born in the same area during the study period. Details of the recruitment of the study and the control INTRODUCTION children have been given in a previous report [Gaily et Clinical signs of distal phalangeal hypoplasia have al., 1988al. Radiographs of the left hand were obtained a t 5.5 been repeatedly associated with prenatal exposure to phenytoin in prospective studies of children of epileptic years of age in 111(75%)children of the study group and mothers [Andermann et al., 1982; Kelly et al., 1984; in 96 (91%) control children. The mean age of both Rating et al., 1985; Gaily et al., 1988al. However, the groups at the time of the examination was 5.5 years clinical assessment of phalangeal size is indirect and (range 5.2-5.8 yrs). Seventy-six children of the study easily affected by observer bias. Consequently, there is group had been exposed to phenytoin (44 monotherapy), considerable variation in the frequency of distal pha- 21 to other antiepileptic drugs (mostly carbamazepine), langeal hypoplasia in the control populations in differ- and 14 had not been exposed to drugs in the first 20 weeks of pregnancy. At least one determination of maternal phenytoin level from the first 20 weeks of pregReceived for publication March 17, 1989; revision received July nancy was available for 39 children; in eight of them, the 20, 1989. maximum level measured exceeded 40 pmolil (range Address reprint requests to Eija Gaily, M.D., Children's Castle Hospital, Lastenlinnantie 2, SF-00250 Helsinki, Finland. 52-86 pmol/l). 0 1990 Wiley-Liss, Inc.
Distal Phalangeal Hypoplasia and Phenytoin Exposure The length of all phalanges and metacarpal bones was measured with a direct reading caliper taking the maximum length of each bone along its longitudinal axis [Poznanski, 19841, including the epiphysis, and recorded with an accuracy of 0.1 mm. Because of technical reasons, the first distal phalanx of one child (phenytoin exposure with low levels), and the third distal and middle phalanges of another child (carbamazepine monotherapy), could not be reliably measured and were recorded as missing values. All measurements were done by one person (E.G.) without knowledge of group assignment. Metacarpophalangeal profile patterns with bone lengths expressed as standard deviation units (using the control group of this study as the standard) were constructed as described by Poznanski [19841. Kelly [19841 diagnosed distal phalangeal hypoplasia to be present in a digit if two of the three following length ratios-l) dista1:middle phalanx, 2) dista1:proximal phalanx, and 3) distal pha1anx:metacarpal-were below - 2 SD from the mean for age, and in a child, if at least two of four digits (the thumb was excluded) showed distal hypoplasia by the first criterion. In the present study, these criteria were slightly modified. The thumb was included, and the first distal phalanx was considered hypoplastic if both length ratios (dista1:proximal phalanx and distal pha1anx:metacarpal bone) were below - 2 SD. For digits two to five, similar criteria as in Kelly’s [19841study were used. Cutoff points at - 2 SD were defined separately for boys and girls using the control data of this study. Distal phalangeal hypoplasia was deemed to be present in a child if at least two of five digits showed distal hypoplasia as defined above. Other examinations done to the study and control children at 5.5 years included psychometric testing by the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) [Wechsler, 19671, the Leiter International Performance Scale (LIPS) [Levine, 19821, and the measurement of height [Gaily and Granstrom, 19891. Total hand length (middle finger plus palm length, with an accuracy of 1mm) of the parents was measured by the author a t a clinical examination as shown by Smith [19821. Both parents of 76 (68%) study and 51 (53%)control children were available for the measurement. Midparent hand length was defined as the mean of maternal and paternal hand length. One- and two-way analysis of variance was used in the comparison of the bone lengths and the length ratios with distal phalanx as numerator and middleiproximal phalanges and metacarpal bones as denominators. The numbers of children in different categories were compared with the x2 test. The computations were performed with the BMDP statistical package [Dixon et al., 19851. Informed consent was obtained from the parents of the participating children. The study was accepted by the ethical committee of the Pediatric Clinic of the Helsinki University Hospital.
RESULTS Distal phalanges 2-5 were significantly shorter in the study group compared with the controls (Table I). Lengths of other phalanges or metacarpals did not differ
575
significantly between the groups. Males and females differed significantly in three bone lengths: boys had longer fifth middle and proximal phalanges but shorter first metacarpal bone than do girls. One child exposed t o high phenytoin levels (maximum 86 FmoL’l measured at 18 weeks) was observed to have complete aplasia of the distal phalanges of the second and fifth digits, and severe hypoplasia of the third and fourth distal phalanges. His phalangeal and metacarpal lengths are presented separately from the other phenytoin-exposed children in Figure 1. Relative shortness of distal phalanges compared to the other bones is seen also in the other seven children exposed to phenytoin levels over 40 kmolil (Fig. 1).Both in the severely affected child and in the other seven children exposed to phenytoin over 40 kmol/l, the distal phalanges of digits 2 and 5 were the shortest and the middle phalanges were relatively long. The children exposed to phenytoin levels below 40 kmolil or unknown levels had metacarpal and proximal phalangeal lengths that were very close to the control group and somewhat longer middle and shorter distal phalanges (Fig. 1). The metacarpal and phalangeal length profiles of the children exposed t o antiepileptic drugs other than phenytoin, and of those not exposed to any drugs during pregnancy, followed very closely the control profile. All the length ratios of distal to middle phalanx, and also some of the ratios distal to proximal phalanx and distal phalanx to metacarpal bone ratios, were significantly reduced in phenytoin-exposed children, even when the severely affected child was excluded (Table 11). The length ratios of children exposed to other anticonvulsants and of those not exposed to drugs were similar to controls. The mean (SD)midparent total hand length was 18.5 (0.6) cm in both the study and the control group. The mean values expressed as standard deviation units (control mean = 0) in the drug-exposure subgroups were high phenytoin, + 0.5 SD; low or unknown phenytoin, + 0.1 SD; other drugs, - 0.7 SD; and no drugs, + 0.9 SD. Significantly more phenytoin-exposed children (8175 = 11%) had radiologically diagnosed distal phalangeal hypoplasia than did children not exposed to phenytoin (11130 = 1%) (x2 = 8.9, P = .003). If the thumb was excluded according to the criteria used by Kelly [19841, the prevalence of distal phalangeal hypoplasia was 7175 (9%) in phenytoin-exposed and Oil30 (0%) in nonexposed children (x2 = 9.7, P = .002). Maternal serum levels of phenytoin in the first 20 weeks of pregnancy had been over 40 Fmolil in four, and unknown in the remaining four phenytoin-exposed children with distal phalangeal hypoplasia. The mean intelligence scores of seven of the eight phenytoin-exposed children showing distal phalangeal hypoplasia by radiological criteria were 104.7 (WPPSI) (range 77-125) and 103.1 (LIPS) (from 93 t o 113). The remaining child, who was not tested, was attending normal school with average achievement at age 8 years. The mean height at 5.5 years of the eight children was 112.8 (from 110.0 to 116.9)cm. These mean values were within 1 standard deviation of the mean values (re-
576
Gaily TABLE I. Phalangeal and Metacarpal Lengths of the Study and Control Children by Gender Study
Bones Phalanges Distal
1 2 3 4 5 Middle 2 3 4 5 Proximal 1 2 3 4 5 Metacarpal 1 2 3 4 5
Males (n = 57) Mean SD
Females (n = 541 Mean SD
Control Males Females (n = 43) (n = 351 Mean SD Mean SD
14.5
14.4" 10.3 11.2" 11.3 10.0 15.2 18.5 17.5 11.9 19.2 26.3 29.1a 27.0 20.9 27.9 42.6 40.4 35.9 33.0
14.7 10.5 11.3 11.7 10.4 15.5 18.4 17.7 12.3 19.6 26.6 29.3 27.5 21.4 27.5 42.8 40.4 35.8 32.8
1.1 1.6 1.2 1.3 1.6 0.9 1.1
10.0
10.9 11.2 9.8 15.2 18.4 17.6 12.1 19.3 26.1 28.9 27.0 20.9 27.2 42.6 40.2 35.5 32.9
1.1 1.2 1.1 1.4 1.6 1.6 1.2 1.5 2.4 2.3 2.1 1.9
1.2 0.9 0.9 0.9 0.9 1.5 1.6 1.8 1.6 1.6 1.8 2.0 1.9 1.5 2.2 2.8 2.7 2.4 2.4
0.8 0.8 0.8 0.8 0.7 1.0 1.0
1.1 1.2 1.3 1.7 1.7 1.6 1.2 1.7 2.6 2.3 2.0 1.7
14.5 10.5 11.4 11.7 10.3 14.9 18.1 17.2 11.5 19.0 26.0 28.9 26.8 20.7 27.9 42.8 40.5 36.0 33.1
1.o 0.8 0.9 0.8 0.8 1.1 1.1 1.1 1.3 1.2 1.4 1.4 1.4 1.1 1.n 2.1 2.1 2.1 1.7
Summary of ANOVA Group F
Gender F
1.78 5.66* 4.80* 8.37** 8.36** 0.19 1.51 0.45 0.26 0.29 0.43 0.12 0.26 0.42 0.40 0.33 0.37 0.28 0.01
0.82 1.03 1.35 0.41 0.03 3.46 0.37 2.44 6.27" 3.20 0.92 0.21 2.70 4.2n* 4.13* 0.02 0.28 0.82 0.35
All values are in mm. ANOVA =analysis of variance, SD = standard deviation. = 52.
* P < .05. ** P i.01. Proximal
+Metacarpal+
- *I+
Middle + l e Dlstal
-+
1 2 3 4 5 ' 1 2 3 4 5 ' 2 3 4 5 ' 12 3 4 5 +1
0 h
n rn
v
-1
-2
z
'0 *
P
2 >
n
ALL BONES: MEAN=O, SD=l
-5-
-6-7-8-
PHENYTOIN =+40pmol/l (N=7)
b
PHENYTOIN $40 pmolll OR UNKNOWN (N=68) A NO PHENYTOIN BUT EXPOSED TO OTHER DRUGS [N=21) 0 NO DRUGS (N=14) A CHILD WITH SEVERE DISTAL DIGITAL HYPOPLASIA AND PHENYTOIN 2 4 0 pmol/l (N=l) 0
-'-10-
-11-12-13l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
t
1 2 3 4 5 1 2 3 4 5 2 3 4 5 1 2 3 4 5 +Metacarpal+
Proximal
+ItMdidel++-
Distal
Fig. 1.Metacarpophalangealprofile patterns in study group children with different drug exposures expressed as standard deviation units with the control group as the standard.
ported earlier) of the total study group [Gaily et al., 1988b; Gaily and Granstrom, 19891. The only major malformation observed in these children was very mild hypospadias occurring in the child with severe distal phalangeal hypoplasia.
DISCUSSION The phalangeal and metacarpal lengths, length ratios, and standard deviations obtained in our control group are very close to the previously published standards [Poznanski, 19841. Because no consistent differences were observed between boys and girls in our study, no adjustment of the mean values for gender was made. The relatively long metacarpal and proximal phalangeal lengths in the subgroup exposed t o phenytoin levels over 40 pmol/l, and the relative shortness of all measured bones in the subgroup exposed t o other antiepileptic drugs, are probably explained by genetic determination of hand size, as there was close agreement with the mean total hand lengths observed in their respective parents. The results of this study confirm that maternal phenytoin treatment during pregnancy is associated with a reduction of distal phalangeal length in the offspring,as was suggested by the observations from previous clinical [Andermann et al., 1982;Kelly et al., 1984;Rating et al., 1985; Gaily et al., 1988al and an uncontrolled anthropometric study [Kelly, 19841. The effect seems to be dose-related, as the smallest values of distal phalangeal length and the distal to proximal bone length ratios, and the highest prevalence of radiologicallydiagnosed distal phalangeal hypoplasia were seen in the group exposed to the highest phenytoin levels. A significant reduction in distal phalangeal length
Distal Phalangeal Hypoplasia and Phenytoin Exposure
577
TABLE 11. Ratios of Distal Phalanees to the More Proximal Phalanges and Metacamal Bones in the Left Hand High PHT Low PHT Other AED No drugs Control Summary of (n = 7)” (n = 68) (n = 21) (n = 14) (n = 96) ANOVA Ratio Mean SD Mean SD Mean SD Mean SD Mean SD F P Distal/ Middle
2 3 4 5 Proximal 1 2 3 4 5 Metacarpal 1 2 3 4 5
.59*** .57** .59*** .70*** .73 .36*** .38 .40* .43*** 52 .22** .27 .30* .27***
.03 .02 .02 .05 .02 .02 .03 .02 .03 .03 .01 .01 .02 .03
.67” .60*** .64””* .83** .76b .39 .38 .42** .48* .53b .24 27 .32* .30*
.05 .04 .04 .08 .05 .03 .03 .03 .04 .04 .02 .02 .02 .02
.70 .63” .68 .92 .75 .40 .39‘ .42 .49 .51 .25 .28‘ .32 .31
.05 .04 .05 .10 .04
.02 .02
.02 .03 .03 .01 .02 .02 .02
.68 .61 .67 .86 .76 .40 .39 .43 .49 .53 .25 .28 .33 .31
.04
.04 .04 .08 .04 .03 .03 .03 .03 .03 .01
.02 .02 .02
.70 .63 .67 .88
.76 .40 .39 .43 .49 53 .25 .28 .33 .31
.05 .04 .04 .10 .04 .03 .02 .02 .03 .03 .02 .02 .02 .02
11.7 8.7 13.2 11.4 0.9 5.3 2.1 5.4 7.6 1.0 4.5 1.9 4.0 5.6
.OOOO .OOOO .OOOO .OOOO ,4463 ,0004 ,0786 .0003 .OOOO ,4238 ,0017 .I075 ,0037 .OOOO
High vs. low PHT=maximum measured maternal phenytoin level over 40 p,mol/l vs. 540 p,mol/l or unknown in 20 first weeks of pregnancy. AED =antiepileptic drug, ANOVA = analysis of variance, SD = standard deviation. Child with aplasiaisevere hypoplasia of the distal phalanges is excluded. ‘n=67. ***P< ,001, **P< .01,“P < .05 against controls (pairwise t-tests with Bonferroni correction). ~n=20.
was seen in the children exposed to phenytoin levels over 40 pmol/l even after the exclusion of the most severely affected child. The small variance observed in this subgroup suggests that all or nearly all of the exposed children were affected to some degree. As in Kelly’s [1984] study, the most prominent effects were noted in digits 2 and 5. The length ratios between phalangeal and metacarpal bones have been shown to change minimally with age [Poznanski, 19841, and “adult-type” length proportions are attained already at the 13th week of gestation [Garn et al., 19751. This suggests that phenytoin must affect distal phalangeal length very early in pregnancy. Perhaps the vulnerable period is by the 8th-9th gestational week, when the distal phalanges show rapid disproportionate elongation [Garn et al., 19751. The factors determining the degree of distal hypoplasia in individual children are not obvious. It is possible that severe distal phalangeal hypoplasia or aplasia occurs only when phenytoin levels at the 8th-9th week are in the toxic range. In most mothers in this study, including the mother of the severely affected child, the first drug levels during pregnancy were obtained only after the 12th week. Furthermore, genetic differencesin susceptibility to phenytoin teratogenic effect might play a role [Aase, 19741. The prevalence of distal phalangeal hypoplasia among phenytoin-exposed children in this study was lower than that observed by Kelly [19841. As no maternal drug levels were reported in Kelly’s study, it is possible that there were more children with early exposure to high phenytoin concentrations among his subjects than in the present study. Our methods also differed, as epiphyses could not be included in the measurements obtained at approximately 1year of age in Kelly’s study. Significant catch-up growth, however, should not occur,
as the effect of phenytoin on the developing distal phalanges is presumed to be teratogenic. Although irrelevant for the well-being of the child as such, mild distal phalangeal hypoplasia is a manifestation of a prenatal effect of phenytoin, and one may ask if the affected children show other consequences of this exposure. The present results support the findings of Kelly et al. [19841 that mild distal phalangeal hypoplasia in phenytoin-exposed children is not associated with a high risk of serious developmental abnormalities.
ACKNOWLEDGMENTS I am grateful to Hillevi Ahonen, RN, for the practical arrangements of the study, and t o Marja-Liisa Granstrom, MD, Brian Neville, MD, and Eino Marttinen, MD, for reviewing the manuscript. Vilho Hiilesmaa, MD, and Ali Bardy, MD, examined the mothers of the study children during pregnancy and provided the background data. The Rinnekoti Research Foundation, ESPOO, Finland; the Foundation of Pediatric Research, and the Orion Foundation gave financial support. These results have been presented in part at the 18th International Epilepsy Congress in New Delhi, India, October 17-22, 1989. REFERENCES Aase JM (1974): Anticonvulsant drugs and congenital abnormalities (Letter). Am J Dis Child 127:758. Andermann E, Dansky L, Andermann F, Loughnan PM, Gibbons J (1982): Minor congenital malformations and dermatoglyphic alterations in the offspring of epileptic women: A clinical investigation of the teratogenic effects of anticonvulsant medication. In J a m D, Dam M, Richens A, et al. (eds): “Epilepsy, Pregnancy and the Child.” New York Raven Press, pp 235-249. Dixon WJ, Brown MB, Engelman L, F’rane JW, Hill MA, Jennrich RI, Toporek J D (1985): BMDP Statistical Software. Berkeley, Los Angeles, London: University of California Press.
578
Gaily
Gaily E, Granstrom M-L (1989): A transient retardation of early postnatal growth in drug exposed children of epileptic mothers. Epilepsy Res. 4:147-155. Gaily E, Granstrom M-L, Hiilesmaa V, Bardy A (1988a): Minor anomalies in offspring of epileptic mothers. J Pediatr 112:520-529. Gaily E, Kantola-Sorsa E, Granstrom M-L (198813): Intelligence of children of epileptic mothers. J Pediatr 113:677-684. Garn SM, Burdi AR, Babler WJ, Stinson C (1975): Early prenatal attainment of adult metacarpal-phalangeal rankings and proportions. Am J Phys Anthropol 43:327-332. Kelly TE (1984): Teratogenicity of anticonvulsant drugs. I11 Radiographic hand analysis of children exposed in utero to diphenylhydantoin. Am J Med Genet 19:445-450. Kelly TE, Edwards P, Rein M, Miller JQ, Dreifuss FE (1984): Teratogenicity of anticonvulsant drugs. 11: A prospective study. Am J Med Genet 19:435-443. Levine ME (1982): “Leiter International Performance Scale. A Handbook.” Los Angeles, California: Western Psychological Services.
Poznanski AK (1984):“The Hand in Radiologic Diagnosis,”Vol 1, Ed 2. Philadelphia, London, Toronto, Mexico City, Rio de Janeiro, Sydney, Tokyo: W.B. Saunders Company, pp 31-54. Rating D, Jager-Roman E, Koch S, Gopfert-Geyer I, Helge H (1982): Minor anomalies in the offspring of epileptic parents. In Janz D, Dam M, Richens A, Bossi L, Helge H, Schmidt D (eds): “Epilepsy, Pregnancy and the Child.” New York: Raven Press, pp 263-288. Rating D, Koch S, Jager-Roman E, Jacob S, Helge H (1985): The influence of antiepileptic drugs on minor anomalies in the offspring of epileptic parents. In Marois M (ed): “Prevention of Physical and Mental Congenital Defects, Part C: Basic and Medical Science, Education, and Future Strategies.”New York: Alan R. Liss, Inc., pp 57-60. Smith DW (1982): “Recognizable Patterns of Human Malformation.” Philadelphia, London, Toronto, Mexico City, Rio de Janeiro, Sydney, Tokyo: W.B. Saunders Company, p 603. Wechsler D (1967): Wechsler Preschool and Primary Scale of Intelligence. New York: The Psychological Corporation.
Distal Phalangeal Hypoplasia in Children With Prenatal Phenytoin Exposure: Results of a Controlled Anthropometric Study Eija Gaily Children's Hospital, University of Helsinki, and Children$ Castle Hospital, Helsinki, Finland Phalangeal and metacarpal bone length was measured from hand radiographs in 111children of epileptic mothers and 96 control children of nonepileptic mothers. Seventy-six children of the study group had been exposed to phenytoin in the first 20 weeks of pregnancy, 21 had been exposed to other antiepileptic drugs excluding phenytoin, and 14 had not been exposed. Distal phalangeal lengths were significantly reduced in phenytoin-exposed children. The second and fifth digits were most affected. Phenytoin exposure was associated with a significantly elevated prevalence (Ilk) of radiologically defined distal phalangeal hypoplasia. "he subgroup of children exposed to phenytoin levels over 40 pmolA showed more prominent effects than did the subgroup exposed to lower or unknown concentrations. These results confirm that early fetal exposure to phenytoin decreases distal phalangeal size, as suggested by several previous studies relying on clinical examination only. Distal phalangeal hypoplasia was not accompanied by other serious abnormalities.
ent studies [Andermann et al., 1982;Rating et al., 1982; Gaily et al., 1988al. Radiological anthropometry offers a more discrete and objective method of measuring the degree of distal digital hypoplasia than does the clinical examination of phalangeal size and dermatoglyphic patterning. There is one previous uncontrolled study [Kelly, 19841, where distal phalangeal hypoplasia was diagnosed on the measurements of hand radiographs. In that study, distal phalangeal hypoplasia was observed in 32% of 47 phenytoin-exposed children and in only one (10%)of ten children not exposed to phenytoin. The aims of the present study were 1)to investigate the effect of prenatal phenytoin exposure on distal phalangeal length measured from hand radiographs, and 2) to estimate the prevalence of radiologically diagnosed distal phalangeal hypoplasia among phenytoin-exposed and nonexposed children. SUBJECTS AND METHODS From December 1975 to December 1979,140epileptic mothers were enrolled in a prospective study during 153 pregnancies at the Obstetric Clinic of the Helsinki University Hospital. The 148 surviving children from these pregnancies were included in the present study. They represent approximately 75% of the children born to epileptic mothers in the Uusimaa province in southern KEY WORDS: maternal epilepsy, phenytoin, Finland during the study period [Gaily et al., 1988bl. teratogenicity The control group comprises 105children of nonepileptic mothers born in the same area during the study period. Details of the recruitment of the study and the control INTRODUCTION children have been given in a previous report [Gaily et Clinical signs of distal phalangeal hypoplasia have al., 1988al. Radiographs of the left hand were obtained a t 5.5 been repeatedly associated with prenatal exposure to phenytoin in prospective studies of children of epileptic years of age in 111(75%)children of the study group and mothers [Andermann et al., 1982; Kelly et al., 1984; in 96 (91%) control children. The mean age of both Rating et al., 1985; Gaily et al., 1988al. However, the groups at the time of the examination was 5.5 years clinical assessment of phalangeal size is indirect and (range 5.2-5.8 yrs). Seventy-six children of the study easily affected by observer bias. Consequently, there is group had been exposed to phenytoin (44 monotherapy), considerable variation in the frequency of distal pha- 21 to other antiepileptic drugs (mostly carbamazepine), langeal hypoplasia in the control populations in differ- and 14 had not been exposed to drugs in the first 20 weeks of pregnancy. At least one determination of maternal phenytoin level from the first 20 weeks of pregReceived for publication March 17, 1989; revision received July nancy was available for 39 children; in eight of them, the 20, 1989. maximum level measured exceeded 40 pmolil (range Address reprint requests to Eija Gaily, M.D., Children's Castle Hospital, Lastenlinnantie 2, SF-00250 Helsinki, Finland. 52-86 pmol/l). 0 1990 Wiley-Liss, Inc.
Distal Phalangeal Hypoplasia and Phenytoin Exposure The length of all phalanges and metacarpal bones was measured with a direct reading caliper taking the maximum length of each bone along its longitudinal axis [Poznanski, 19841, including the epiphysis, and recorded with an accuracy of 0.1 mm. Because of technical reasons, the first distal phalanx of one child (phenytoin exposure with low levels), and the third distal and middle phalanges of another child (carbamazepine monotherapy), could not be reliably measured and were recorded as missing values. All measurements were done by one person (E.G.) without knowledge of group assignment. Metacarpophalangeal profile patterns with bone lengths expressed as standard deviation units (using the control group of this study as the standard) were constructed as described by Poznanski [19841. Kelly [19841 diagnosed distal phalangeal hypoplasia to be present in a digit if two of the three following length ratios-l) dista1:middle phalanx, 2) dista1:proximal phalanx, and 3) distal pha1anx:metacarpal-were below - 2 SD from the mean for age, and in a child, if at least two of four digits (the thumb was excluded) showed distal hypoplasia by the first criterion. In the present study, these criteria were slightly modified. The thumb was included, and the first distal phalanx was considered hypoplastic if both length ratios (dista1:proximal phalanx and distal pha1anx:metacarpal bone) were below - 2 SD. For digits two to five, similar criteria as in Kelly’s [19841study were used. Cutoff points at - 2 SD were defined separately for boys and girls using the control data of this study. Distal phalangeal hypoplasia was deemed to be present in a child if at least two of five digits showed distal hypoplasia as defined above. Other examinations done to the study and control children at 5.5 years included psychometric testing by the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) [Wechsler, 19671, the Leiter International Performance Scale (LIPS) [Levine, 19821, and the measurement of height [Gaily and Granstrom, 19891. Total hand length (middle finger plus palm length, with an accuracy of 1mm) of the parents was measured by the author a t a clinical examination as shown by Smith [19821. Both parents of 76 (68%) study and 51 (53%)control children were available for the measurement. Midparent hand length was defined as the mean of maternal and paternal hand length. One- and two-way analysis of variance was used in the comparison of the bone lengths and the length ratios with distal phalanx as numerator and middleiproximal phalanges and metacarpal bones as denominators. The numbers of children in different categories were compared with the x2 test. The computations were performed with the BMDP statistical package [Dixon et al., 19851. Informed consent was obtained from the parents of the participating children. The study was accepted by the ethical committee of the Pediatric Clinic of the Helsinki University Hospital.
RESULTS Distal phalanges 2-5 were significantly shorter in the study group compared with the controls (Table I). Lengths of other phalanges or metacarpals did not differ
575
significantly between the groups. Males and females differed significantly in three bone lengths: boys had longer fifth middle and proximal phalanges but shorter first metacarpal bone than do girls. One child exposed t o high phenytoin levels (maximum 86 FmoL’l measured at 18 weeks) was observed to have complete aplasia of the distal phalanges of the second and fifth digits, and severe hypoplasia of the third and fourth distal phalanges. His phalangeal and metacarpal lengths are presented separately from the other phenytoin-exposed children in Figure 1. Relative shortness of distal phalanges compared to the other bones is seen also in the other seven children exposed to phenytoin levels over 40 kmolil (Fig. 1).Both in the severely affected child and in the other seven children exposed to phenytoin over 40 kmol/l, the distal phalanges of digits 2 and 5 were the shortest and the middle phalanges were relatively long. The children exposed to phenytoin levels below 40 kmolil or unknown levels had metacarpal and proximal phalangeal lengths that were very close to the control group and somewhat longer middle and shorter distal phalanges (Fig. 1). The metacarpal and phalangeal length profiles of the children exposed t o antiepileptic drugs other than phenytoin, and of those not exposed to any drugs during pregnancy, followed very closely the control profile. All the length ratios of distal to middle phalanx, and also some of the ratios distal to proximal phalanx and distal phalanx to metacarpal bone ratios, were significantly reduced in phenytoin-exposed children, even when the severely affected child was excluded (Table 11). The length ratios of children exposed to other anticonvulsants and of those not exposed to drugs were similar to controls. The mean (SD)midparent total hand length was 18.5 (0.6) cm in both the study and the control group. The mean values expressed as standard deviation units (control mean = 0) in the drug-exposure subgroups were high phenytoin, + 0.5 SD; low or unknown phenytoin, + 0.1 SD; other drugs, - 0.7 SD; and no drugs, + 0.9 SD. Significantly more phenytoin-exposed children (8175 = 11%) had radiologically diagnosed distal phalangeal hypoplasia than did children not exposed to phenytoin (11130 = 1%) (x2 = 8.9, P = .003). If the thumb was excluded according to the criteria used by Kelly [19841, the prevalence of distal phalangeal hypoplasia was 7175 (9%) in phenytoin-exposed and Oil30 (0%) in nonexposed children (x2 = 9.7, P = .002). Maternal serum levels of phenytoin in the first 20 weeks of pregnancy had been over 40 Fmolil in four, and unknown in the remaining four phenytoin-exposed children with distal phalangeal hypoplasia. The mean intelligence scores of seven of the eight phenytoin-exposed children showing distal phalangeal hypoplasia by radiological criteria were 104.7 (WPPSI) (range 77-125) and 103.1 (LIPS) (from 93 t o 113). The remaining child, who was not tested, was attending normal school with average achievement at age 8 years. The mean height at 5.5 years of the eight children was 112.8 (from 110.0 to 116.9)cm. These mean values were within 1 standard deviation of the mean values (re-
576
Gaily TABLE I. Phalangeal and Metacarpal Lengths of the Study and Control Children by Gender Study
Bones Phalanges Distal
1 2 3 4 5 Middle 2 3 4 5 Proximal 1 2 3 4 5 Metacarpal 1 2 3 4 5
Males (n = 57) Mean SD
Females (n = 541 Mean SD
Control Males Females (n = 43) (n = 351 Mean SD Mean SD
14.5
14.4" 10.3 11.2" 11.3 10.0 15.2 18.5 17.5 11.9 19.2 26.3 29.1a 27.0 20.9 27.9 42.6 40.4 35.9 33.0
14.7 10.5 11.3 11.7 10.4 15.5 18.4 17.7 12.3 19.6 26.6 29.3 27.5 21.4 27.5 42.8 40.4 35.8 32.8
1.1 1.6 1.2 1.3 1.6 0.9 1.1
10.0
10.9 11.2 9.8 15.2 18.4 17.6 12.1 19.3 26.1 28.9 27.0 20.9 27.2 42.6 40.2 35.5 32.9
1.1 1.2 1.1 1.4 1.6 1.6 1.2 1.5 2.4 2.3 2.1 1.9
1.2 0.9 0.9 0.9 0.9 1.5 1.6 1.8 1.6 1.6 1.8 2.0 1.9 1.5 2.2 2.8 2.7 2.4 2.4
0.8 0.8 0.8 0.8 0.7 1.0 1.0
1.1 1.2 1.3 1.7 1.7 1.6 1.2 1.7 2.6 2.3 2.0 1.7
14.5 10.5 11.4 11.7 10.3 14.9 18.1 17.2 11.5 19.0 26.0 28.9 26.8 20.7 27.9 42.8 40.5 36.0 33.1
1.o 0.8 0.9 0.8 0.8 1.1 1.1 1.1 1.3 1.2 1.4 1.4 1.4 1.1 1.n 2.1 2.1 2.1 1.7
Summary of ANOVA Group F
Gender F
1.78 5.66* 4.80* 8.37** 8.36** 0.19 1.51 0.45 0.26 0.29 0.43 0.12 0.26 0.42 0.40 0.33 0.37 0.28 0.01
0.82 1.03 1.35 0.41 0.03 3.46 0.37 2.44 6.27" 3.20 0.92 0.21 2.70 4.2n* 4.13* 0.02 0.28 0.82 0.35
All values are in mm. ANOVA =analysis of variance, SD = standard deviation. = 52.
* P < .05. ** P i.01. Proximal
+Metacarpal+
- *I+
Middle + l e Dlstal
-+
1 2 3 4 5 ' 1 2 3 4 5 ' 2 3 4 5 ' 12 3 4 5 +1
0 h
n rn
v
-1
-2
z
'0 *
P
2 >
n
ALL BONES: MEAN=O, SD=l
-5-
-6-7-8-
PHENYTOIN =+40pmol/l (N=7)
b
PHENYTOIN $40 pmolll OR UNKNOWN (N=68) A NO PHENYTOIN BUT EXPOSED TO OTHER DRUGS [N=21) 0 NO DRUGS (N=14) A CHILD WITH SEVERE DISTAL DIGITAL HYPOPLASIA AND PHENYTOIN 2 4 0 pmol/l (N=l) 0
-'-10-
-11-12-13l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
t
1 2 3 4 5 1 2 3 4 5 2 3 4 5 1 2 3 4 5 +Metacarpal+
Proximal
+ItMdidel++-
Distal
Fig. 1.Metacarpophalangealprofile patterns in study group children with different drug exposures expressed as standard deviation units with the control group as the standard.
ported earlier) of the total study group [Gaily et al., 1988b; Gaily and Granstrom, 19891. The only major malformation observed in these children was very mild hypospadias occurring in the child with severe distal phalangeal hypoplasia.
DISCUSSION The phalangeal and metacarpal lengths, length ratios, and standard deviations obtained in our control group are very close to the previously published standards [Poznanski, 19841. Because no consistent differences were observed between boys and girls in our study, no adjustment of the mean values for gender was made. The relatively long metacarpal and proximal phalangeal lengths in the subgroup exposed t o phenytoin levels over 40 pmol/l, and the relative shortness of all measured bones in the subgroup exposed t o other antiepileptic drugs, are probably explained by genetic determination of hand size, as there was close agreement with the mean total hand lengths observed in their respective parents. The results of this study confirm that maternal phenytoin treatment during pregnancy is associated with a reduction of distal phalangeal length in the offspring,as was suggested by the observations from previous clinical [Andermann et al., 1982;Kelly et al., 1984;Rating et al., 1985; Gaily et al., 1988al and an uncontrolled anthropometric study [Kelly, 19841. The effect seems to be dose-related, as the smallest values of distal phalangeal length and the distal to proximal bone length ratios, and the highest prevalence of radiologicallydiagnosed distal phalangeal hypoplasia were seen in the group exposed to the highest phenytoin levels. A significant reduction in distal phalangeal length
Distal Phalangeal Hypoplasia and Phenytoin Exposure
577
TABLE 11. Ratios of Distal Phalanees to the More Proximal Phalanges and Metacamal Bones in the Left Hand High PHT Low PHT Other AED No drugs Control Summary of (n = 7)” (n = 68) (n = 21) (n = 14) (n = 96) ANOVA Ratio Mean SD Mean SD Mean SD Mean SD Mean SD F P Distal/ Middle
2 3 4 5 Proximal 1 2 3 4 5 Metacarpal 1 2 3 4 5
.59*** .57** .59*** .70*** .73 .36*** .38 .40* .43*** 52 .22** .27 .30* .27***
.03 .02 .02 .05 .02 .02 .03 .02 .03 .03 .01 .01 .02 .03
.67” .60*** .64””* .83** .76b .39 .38 .42** .48* .53b .24 27 .32* .30*
.05 .04 .04 .08 .05 .03 .03 .03 .04 .04 .02 .02 .02 .02
.70 .63” .68 .92 .75 .40 .39‘ .42 .49 .51 .25 .28‘ .32 .31
.05 .04 .05 .10 .04
.02 .02
.02 .03 .03 .01 .02 .02 .02
.68 .61 .67 .86 .76 .40 .39 .43 .49 .53 .25 .28 .33 .31
.04
.04 .04 .08 .04 .03 .03 .03 .03 .03 .01
.02 .02 .02
.70 .63 .67 .88
.76 .40 .39 .43 .49 53 .25 .28 .33 .31
.05 .04 .04 .10 .04 .03 .02 .02 .03 .03 .02 .02 .02 .02
11.7 8.7 13.2 11.4 0.9 5.3 2.1 5.4 7.6 1.0 4.5 1.9 4.0 5.6
.OOOO .OOOO .OOOO .OOOO ,4463 ,0004 ,0786 .0003 .OOOO ,4238 ,0017 .I075 ,0037 .OOOO
High vs. low PHT=maximum measured maternal phenytoin level over 40 p,mol/l vs. 540 p,mol/l or unknown in 20 first weeks of pregnancy. AED =antiepileptic drug, ANOVA = analysis of variance, SD = standard deviation. Child with aplasiaisevere hypoplasia of the distal phalanges is excluded. ‘n=67. ***P< ,001, **P< .01,“P < .05 against controls (pairwise t-tests with Bonferroni correction). ~n=20.
was seen in the children exposed to phenytoin levels over 40 pmol/l even after the exclusion of the most severely affected child. The small variance observed in this subgroup suggests that all or nearly all of the exposed children were affected to some degree. As in Kelly’s [1984] study, the most prominent effects were noted in digits 2 and 5. The length ratios between phalangeal and metacarpal bones have been shown to change minimally with age [Poznanski, 19841, and “adult-type” length proportions are attained already at the 13th week of gestation [Garn et al., 19751. This suggests that phenytoin must affect distal phalangeal length very early in pregnancy. Perhaps the vulnerable period is by the 8th-9th gestational week, when the distal phalanges show rapid disproportionate elongation [Garn et al., 19751. The factors determining the degree of distal hypoplasia in individual children are not obvious. It is possible that severe distal phalangeal hypoplasia or aplasia occurs only when phenytoin levels at the 8th-9th week are in the toxic range. In most mothers in this study, including the mother of the severely affected child, the first drug levels during pregnancy were obtained only after the 12th week. Furthermore, genetic differencesin susceptibility to phenytoin teratogenic effect might play a role [Aase, 19741. The prevalence of distal phalangeal hypoplasia among phenytoin-exposed children in this study was lower than that observed by Kelly [19841. As no maternal drug levels were reported in Kelly’s study, it is possible that there were more children with early exposure to high phenytoin concentrations among his subjects than in the present study. Our methods also differed, as epiphyses could not be included in the measurements obtained at approximately 1year of age in Kelly’s study. Significant catch-up growth, however, should not occur,
as the effect of phenytoin on the developing distal phalanges is presumed to be teratogenic. Although irrelevant for the well-being of the child as such, mild distal phalangeal hypoplasia is a manifestation of a prenatal effect of phenytoin, and one may ask if the affected children show other consequences of this exposure. The present results support the findings of Kelly et al. [19841 that mild distal phalangeal hypoplasia in phenytoin-exposed children is not associated with a high risk of serious developmental abnormalities.
ACKNOWLEDGMENTS I am grateful to Hillevi Ahonen, RN, for the practical arrangements of the study, and t o Marja-Liisa Granstrom, MD, Brian Neville, MD, and Eino Marttinen, MD, for reviewing the manuscript. Vilho Hiilesmaa, MD, and Ali Bardy, MD, examined the mothers of the study children during pregnancy and provided the background data. The Rinnekoti Research Foundation, ESPOO, Finland; the Foundation of Pediatric Research, and the Orion Foundation gave financial support. These results have been presented in part at the 18th International Epilepsy Congress in New Delhi, India, October 17-22, 1989. REFERENCES Aase JM (1974): Anticonvulsant drugs and congenital abnormalities (Letter). Am J Dis Child 127:758. Andermann E, Dansky L, Andermann F, Loughnan PM, Gibbons J (1982): Minor congenital malformations and dermatoglyphic alterations in the offspring of epileptic women: A clinical investigation of the teratogenic effects of anticonvulsant medication. In J a m D, Dam M, Richens A, et al. (eds): “Epilepsy, Pregnancy and the Child.” New York Raven Press, pp 235-249. Dixon WJ, Brown MB, Engelman L, F’rane JW, Hill MA, Jennrich RI, Toporek J D (1985): BMDP Statistical Software. Berkeley, Los Angeles, London: University of California Press.
578
Gaily
Gaily E, Granstrom M-L (1989): A transient retardation of early postnatal growth in drug exposed children of epileptic mothers. Epilepsy Res. 4:147-155. Gaily E, Granstrom M-L, Hiilesmaa V, Bardy A (1988a): Minor anomalies in offspring of epileptic mothers. J Pediatr 112:520-529. Gaily E, Kantola-Sorsa E, Granstrom M-L (198813): Intelligence of children of epileptic mothers. J Pediatr 113:677-684. Garn SM, Burdi AR, Babler WJ, Stinson C (1975): Early prenatal attainment of adult metacarpal-phalangeal rankings and proportions. Am J Phys Anthropol 43:327-332. Kelly TE (1984): Teratogenicity of anticonvulsant drugs. I11 Radiographic hand analysis of children exposed in utero to diphenylhydantoin. Am J Med Genet 19:445-450. Kelly TE, Edwards P, Rein M, Miller JQ, Dreifuss FE (1984): Teratogenicity of anticonvulsant drugs. 11: A prospective study. Am J Med Genet 19:435-443. Levine ME (1982): “Leiter International Performance Scale. A Handbook.” Los Angeles, California: Western Psychological Services.
Poznanski AK (1984):“The Hand in Radiologic Diagnosis,”Vol 1, Ed 2. Philadelphia, London, Toronto, Mexico City, Rio de Janeiro, Sydney, Tokyo: W.B. Saunders Company, pp 31-54. Rating D, Jager-Roman E, Koch S, Gopfert-Geyer I, Helge H (1982): Minor anomalies in the offspring of epileptic parents. In Janz D, Dam M, Richens A, Bossi L, Helge H, Schmidt D (eds): “Epilepsy, Pregnancy and the Child.” New York: Raven Press, pp 263-288. Rating D, Koch S, Jager-Roman E, Jacob S, Helge H (1985): The influence of antiepileptic drugs on minor anomalies in the offspring of epileptic parents. In Marois M (ed): “Prevention of Physical and Mental Congenital Defects, Part C: Basic and Medical Science, Education, and Future Strategies.”New York: Alan R. Liss, Inc., pp 57-60. Smith DW (1982): “Recognizable Patterns of Human Malformation.” Philadelphia, London, Toronto, Mexico City, Rio de Janeiro, Sydney, Tokyo: W.B. Saunders Company, p 603. Wechsler D (1967): Wechsler Preschool and Primary Scale of Intelligence. New York: The Psychological Corporation.
