Validation of the facial photographic method in fetal alcohol spectrum disorder screening and diagnosis

Objective: A prospective study to validate the computer-assisted method of measuring palpebral fissure length and philtrum smoothness using digital patient photographs. These are key diagnostic facial features of Fetal Alcohol Syndrome. Participants: Motherisk Program (including Breaking the Cycle), Hospital for Sick Children, Toronto - a clinical, research and teaching program dedicated to antenatal drug, chemical, and disease risk counseling. 40 children referred for FASD assessment, 21 under 4 years old, 19 were 4 years or older. Methods/ Materials: Facial measurements were obtained directly from the patient by physicians and compared to those obtained by computer software measurement of photographs of the same patient. Outcome Measures: Palpebral fissure length and philtrum smoothness. Results: The photographic measurements showed shorter palpebral fissure length than the direct measurements when analyzing all children (25.4±2.3 vs.23.2±2.4mm; p<0.0001), and children under four (n=21, 24.7±2.4 vs. 21.6±1.6mm; p<0.0001). The difference for older children (n=19) did not reach statistical significance. The computer found four false positive cases and no false negative cases of clinically short palpebral fissure (sensitivity=100%, specificity=64%). Direct measurement scores for philtrum smoothness were different from the computer's measurements using the frontal view (p=0.0012) but not using the 3/4 view. Conclusion: The method of computer-assisted measurement tends to underestimate the true length and, hence, over-diagnose short palpebral fissure, especially in children under four years old. This method may serve as a useful fetal alcohol syndrome screening tool.