Previous studies had shown that even with normal early development, preterm children at age six still have executive function deficits, including planning, cognitive flexibility, and nonverbal working memory. The present study aims to discuss further the correlation between IQ and EF of preterm children with different birthweight in order to clarify the potential influence of birthweight.

The preterm children were recruited from the Regional Cohort Network for premature infants who were admitted to neonatal intensive care units. Inclusion criteria were their scores of Bayley Scales of Infant and Toddler Development, second or third edition at 12 and 24 months, and Wechsler Preschool and Primary Scale of Intelligence, Revised Edition at 5 years old were higher than 70. Meanwhile, their FSIQ of Wechsler Intelligence Scale for Children, Fourth Edition were higher than 85 at age 6. Exclusion criteria were visual impairment, hearing impairment, and cerebral palsy. There was a total of 251 preterm children recruited in the present study. Preterm children were then divided into very low birthweight (VLBW) and extremely low birth weight (ELBW) groups. The VLBW group included 183 preterm children, whose birthweight is between 1000-1500g, and gestational age is less than 37 weeks. The ELBW group included 68 preterm children, whose birthweight is less than 1000g and gestational age is less than 37 weeks. Four types of executive function were assessed. Inhibition was assessed through Comprehensive Nonverbal Attention Test Battery (CNAT), cognitive flexibility was assessed through Wisconsin Card Sorting Test (WCST), verbal working memory was assessed through the Longest Digit Span Forward (LDSB) index of Digit Span Subtest of Wechsler Intelligence Scale for Children-IV (WISC-IV), and nonverbal working memory was assessed through Knox's Cube Test (KCT), and planning ability was assessed through Tower of London (ToL). Data were analyzed with independent T-test and Pearson Correlation.

In VLBW preterm group, results showed that there were significant correlations (p<.05) between FSIQ and EF indexes in five out of six indexes of WCST, LDSB of Digit Span of WISC-IV, Backward score of Knox's Cube Test, and three out of six indexes of ToL. As to the CNAT, there was no index of CNAT that was significantly correlated with FSIQ in the VLBW preterm group. In ELBW preterm group, results showed that there were significant correlations (p<.05) between FSIQ and EF indexes in one out of five indexes of CNAT, one out of six indexes of WCST, Backward score of Knox's Cube Test, and two out of six indexes of ToL. And the LDSB of Digit Span of WISC-IV was not significantly correlated with FSIQ in ELBW preterm group.

There were fewer EF indexes significantly correlated with FSIQ in ELBW preterm children with normal early development than VLBW preterm children with normal early development, suggesting that even with an IQ higher than 85, lower the birthweight, especially lower than 1000g, higher the EF performance should be concerned. Therefore, in the clinical setting, it is very important to assess the EF independently. And birthweight may be a crucial factor in preterm children's prefrontal cortex maturity.