Diagnostic practice and the estimated prevalence of craniosynostosis in Colorado

B. W. Alderman, S. K. Fernbach, C. Greene, E. J. Mangione and S. W. Ferguson
Department of Epidemiology, University of Washington, Seattle, USA.

BACKGROUND: In the late 1980s, evidence of an epidemic of craniosynostosis in Colorado included reports of clusters from selected high-altitude communities and an investigation showing the high and rapidly rising rates of surgically corrected synostosis. Some evidence suggested that local diagnostic practice could account for the epidemic. OBJECTIVE: To determine the contributions of any excess rates of disease occurrence, surgery-based ascertainment, and diagnosis to the reported epidemic. DESIGN: Population-based birth prevalence study with diagnostic evaluation. SETTING: The Colorado Department of Health, April 15, 1986, to July 14, 1989. PATIENTS OR OTHER PARTICIPANTS: Children in the Craniosynostosis Registry or state birth record files. MAIN OUTCOME MEASURES: Birth prevalence was estimated from registry and birth record data; case classification by suture type and malformation patterns were determined by review of radiographs and medical records. RESULTS: The period birth prevalence of radiographically confirmed nonsyndromic synostosis was 14.1 per 10,000 live births. Of a total of 605 children, 307 (51%) had definite radiographic evidence of synostosis, for which the intrarater reliability was good (except for the coronal suture on plain films) and the interrater reliability was fair or good (except for the metopic suture on plain films). Between the first and third years, case reports fell from 347 to 103. CONCLUSIONS: Diagnostic criteria strongly influenced the rate of synostosis. The rate of radiographically confirmed synostosis was within the range of published estimates. Low diagnostic thresholds, which changed over time, created the semblance of a severe statewide epidemic and may have obscured excess rates of disease at high altitude.