DOES HEAD SIZE CORRELATE WITH NEUROLOGIC OUTCOME?
A correlation between small head circumference, or microcephaly, and mental retardation has been recognized since the pioneering work of Kind in 1876 (7) and Tarbell in 1883 (8). This relationship was established primarily through studies that documented a high incidence of microcephaly in groups of mentally retarded or neurologically at-risk individuals. For example, in 1960, measurements were recorded for 2,472 inmates of the California State Institution for the Mentally Deficient. Head circumference means for the total group were below normal for virtually the entire age range (9). Another study reviewed the clinical records of 247 children from 1 to 15 years of age seen between 1956 and 1961 at the Mayo Clinic with small head size, dwarfism, mental retardation, microcephaly or developmental retardation. The authors concluded that "children with a head circumference below minus 2 standard deviations from the mean are, probably with few exceptions, mentally subnormal" (10).
A review of 212 patients with microcephaly seen in either a birth defects or child development clinic documented a direct relationship between IQs and head size. Intelligence quotients decreased from 63 in children with head circumferences between 2.0 and 2.1 standard deviations below the mean to 20 for children whose OFCs were between 5.1 and 7 standard deviations below the mean (11). A study of 73 learning-disabled children revealed a significantly greater prevalence of abnormal head circumferences when compared with 71 controls (12). A review of 202 microcephalic children evaluated at a child development center between 1963 and 1967 demonstrated that only 27 children (13%) had normal intelligence. However, while these children had IQs in the normal range, more than half had evidence of brain damage or dysfunction (13). Finally, evaluation of 360 patients seen between 1976 and 1981 in a child development clinic established a significant direct correlation between head circumference and IQ. Of the children with an OFC =< -2 SD, 18/21 (86%) had a full scale IQ score < 80 (14).
These studies imply that children with a head circumference more than two standard deviations below the mean are almost inevitably neurologically damaged. In terms of international adoptees, review of 268 children referred in early 1998 from Russian orphanages documented that, at the time of referral, 15% of children had a head circumference between 2 and 3 standard deviations below the mean and 7% had an OFC more than 3 standard deviations below the mean. Therefore, almost a quarter of referrals, an ominous number, appear to be at grave neurologic risk.
Conclusions reached to this point are based on data collected from small, very select, at-risk groups of children and not from the pediatric population as a whole. Therefore, it is not surprising that children with microcephaly who seek care in a developmental or genetics clinic have a high rate of neurologic dysfunction. The real question is whether there are children in the general population who are microcephalic but are functioning within the normal range. Such studies are difficult to do since they require a large number of children who have not only sequential head measurements, but data on their cognitive abilities as well.
Over thirty years ago, the National Institutes of Health funded a monumental study, the US National Collaborative Perinatal Project, which enrolled more than 54,000 pregnant women at 12 cooperating institutions across the country. Prenatal, perinatal and neonatal information was collected on the women and their infants, and the children were subsequently followed for up to seven years. Each child was examined at birth, four months, one year, four years and seven years, with physical measurements being taken at each of these ages. Full batteries of developmental tests, including the Wechsler Intelligence Scale, were administered at regular intervals. Using a sample of 35,704 children drawn from this study, Camp et al. (15) identified that microcephaly at birth was associated with a 2.35-fold increase in mental retardation. However, this risk was only observed in African American infants in families with low socioeconomic status.
Persistent microcephaly was a much more ominous finding. Using data from 28,820 term infants with at least three head circumference measurements between birth and seven years and excluding infants with neural tube defects, hydranencephaly, craniosynostosis or Down syndrome, Dolk (16) studied the outcome of infants who had OFCs greater than 2 standard deviations below the mean. The children who fared worst were those with a head circumference consistently more than 3 standard deviations below the mean. Of these children, 51% were mentally retarded (IQ =< 70) and 17% had an IQ of 71-80. Within this group of children, of those with some further specification of pathology such as a syndrome, congenital infection, etc., 94% were mentally retarded. Among those with microcephaly as the only diagnosis, only 25% were mentally retarded. Head circumference consistently between 2 to 3 standard deviations below the mean was associated with retardation in 11% of cases and IQ of 71-80 in 28%. Only 2.6% of children with head circumferences within the normal range were mentally retarded and 7.4% had IQs of 71-80. Another study, relying on head growth of 9,379 children enrolled in the study, looked at the ability of head size at one year to predict IQ at four years (17). At four years, IQ varied directly with head circumference. Head size at one year, at less than 43 (-2.3 SD) for males and 42 (-2.4 SD) for females, was associated with an IQ < 80 in approximately half the cases. None of these children achieved an IQ of 120 or greater at four years of age (17).
School systems have also been exploited in trying to determine the functional level of children with microcephaly. In 1,006 Seattle school children ranging in age from 5 to 18 years, 1.9% were found to have OFCs 2 or more standard deviations below the mean. None were greater than 3 standard deviations below the mean. No significant difference was found between mean IQs of the microcephalic subjects and controls (99.5 vs. 105); however, mean academic achievement scores were significantly lower in the microcephalic subjects (49 vs. 70) (18). In 334 grade-school boys between 8 and 9.5 years of age from St. Louis public schools, head circumference was an excellent predictor of IQ. In subjects with the smallest head circumferences, only five were in the average IQ range and none were over 105 (19).
Many physicians believe in the concept of proportionality of head size.
In other words, one interprets head size in relationship to body size
(e.g., a child with a small OFC who is also short is probably okay). There
are some data to support this concept. The Seattle school study showed
that in children whose head circumference was proportional, mean academic
achievement scores were significantly higher than subjects whose head
size was relatively small; IQ scores did not differ (18). Other studies
contradict this opinion. O'Connell et al. (10) found that children who
had normal intelligence but growth failure all had normal head circumferences.
In the Nelson study (17), which involved the largest number of children,
not only did IQ at four years of age vary with head circumference at one
year, it also varied directly with body length at one year. At a given
small head circumference, four-year IQ rose with increasing body length.
Thus, for a child with a small head, it would seem that the less
proportional the head is to body length, the better.
|This page last updated June 3, 2003 1:36 AM EST|