Head-circumference distribution in a large primary care network differs from CDC and WHO curves

To compare currently available head-circumference growth curves to curves constructed from clinical measurements from patients in a large US primary care network (PCN).

We performed a retrospective cohort study of 75 412 patients in an urban-suburban PCN. Patients with a birth weight of <1500 g or gestational age of <33 weeks at birth were excluded. We compared percentile values and the proportion of head-circumference observations above the 95th percentile and below the 5th percentile for the existing and PCN curves.

The PCN curves were most similar to the National Center for Health Statistics (NCHS) curves and were substantially different from the Centers for Disease Control and Prevention (CDC) and World Health Organization (WHO) curves. The overall proportion of observations above the 95th percentile was 4.9% (PCN), 6.2% (NCHS), 8.6% (CDC), and 14.0% (WHO). The proportion below the 5th percentile was 4.4% (PCN), 5.1% (NCHS), 2.9% (CDC), and 2.3% (WHO). When using the CDC curves, the proportion above the 95th percentile increased from 0.2% for children younger than 2 weeks to 11.8% for children 12 months old. When using the WHO curves, the proportion above the 95th percentile was >5% at all ages, with a maximum of 18.0% for children older than 24 months.

The CDC and WHO head-circumference curves describe different distributions than the clinical measurements in our PCN population, especially for children with larger heads. The resulting percentile misclassification may delay diagnosis in children with intracranial pathology in very young infants and spur unnecessary evaluation of healthy children older than 6 months.