Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020;7(4):44.
Epub 2020 Nov 6.

High facial specificity and positive predictive value are required to diagnose fetal alcohol syndrome when prenatal alcohol exposure is unknown

Affiliations

High facial specificity and positive predictive value are required to diagnose fetal alcohol syndrome when prenatal alcohol exposure is unknown

Susan J Astley Hemingway. Adv Pediatr Res. 2020.

Abstract

Background: Facial criteria with high specificity and positive predictive value (PPV) to prenatal alcohol exposure (PAE) are required to diagnose fetal alcohol syndrome (FAS) when documentation of PAE is unavailable. Not all fetal alcohol spectrum disorder (FASD) diagnostic guidelines appear to meet these criteria.

Methods: A dataset generated from a 10-year FAS screening of 1,602 children in fostercare conducted by the University of Washington FAS Diagnostic & Prevention Network was used to determine how well the FAS facial phenotype, microcephaly and growth deficiency (individually and in combination at varying levels of magnitude) predicted PAE.

Results: The 4-Digit-Code Rank 4 FAS facial phenotype was the only outcome that provided sufficient PPV and specificity to PAE (100%) to allow the facial phenotype to serve as confirmation of PAE in a diagnostic setting when PAE is unknown. Even minimal relaxation of the phenotype (e.g., Face Rank 3) resulted in PPV (35%) and specificity (88.7%) values too low to use as confirmation of PAE. Further relaxation of the facial criteria, as defined by the Hoyme et al., 2016 FASD guidelines, resulted in even lower PPV (17.9%) and specificity (76.6%); both too low to serve as confirmation of PAE in a diagnostic setting. The presence of all three physical features of FAS (Hoyme et al FAS facial phenotype, growth and OFC ≤10th percentile) did not increase PPV beyond chance (52%).

Conclusions: FASD diagnostic guidelines that use relaxed criteria for the FAS facial phenotype risk misdiagnosing and over-diagnosing FAS and partial FAS when PAE is unknown.

Keywords: FAS facial phenotype; Fetal Alcohol Spectrum Disorder; Positive Predictive Value.

PubMed Disclaimer

Conflict of interest statement

COMPETING INTERESTS The author does not have any competing interests.

Figures

Figure 1.
Figure 1.. Demonstration: The ability of the Rank 4 FAS facial phenotype to correctly predict prenatal alcohol exposure.
Among 1,602 children in foster care, 14.2% had documented prenatal alcohol exposure (PAE). If the 4-Digit Code Rank 4 FAS facial phenotype is to be used to confirm PAE when a written or verbal history of exposure is not available, then the FAS facial phenotype must have a high PPV (all individuals with the facial phenotype have a documented PAE) and high specificity (the Rank 4 FAS face is never present in an individual with confirmed absence of PAE). • PPV: Probability that a subject with the FAS facial phenotype has PAE. • NPV: Probability that a subject without the FAS facial phenotype does not have PAE. • Sensitivity: Probability that a subject with PAE has the FAS facial phenotype. • Specificity: Probability that a subject without PAE does not have the FAS facial phenotype.
Figure 2.
Figure 2.. 4-Digit-Code method for measuring the FAS facial phenotype using the FAS Facial Photographic Analysis Software [3, 10, 23].
The Rank 4 FAS facial phenotype is defined by 3 features: short PFLs ≤ −2 SDs, smooth philtrum Rank 4 or 5 and thin upper lip Rank 4 or 5. A) Three digital facial photographs are obtained with a ¾ inch adhesive sticker serving as an internal measure of scale. B) The PFL is measured in mm by clicking the mouse on the inner and outer corners of each eye. Using the Face Tables, the software converts the PFL in mm to a z-score and then to a PFL ABC-Score. The red perimeter of the upper lip is traced with the mouse to compute lip circularity (perimeter2/area). The software converts lip circularity to lip rank and then to a lip ABC-Score. Finally, the software converts the philtrum rank to a philtrum ABC-Score. The 3 individual ABC-scores are combined in the order PFL-Philtrum-Lip to create the overall Facial ABC-Score. Highlighted in red font is an example of a child with PFL −2.5 SDs, philtrum Rank 5 and lip circularity 183. These three features produce a facial ABC-Score of CCC, representing the severe expression of the FAS facial phenotype (Face Rank 4). The FAS facial phenotype presents on a continuum: Rank 1: none of the 3 features present; Rank 2: 1 or 2 features present; Rank 3: 2.5 of the 3 features present; and Rank 4: all 3 features present. A video demonstration of the facial software is presented at this weblink: http://depts.washington.edu/fasdpn/movie/software1024-768cd2.mp4 Copyright Susan Astley Hemingway University of Washington
Figure 3.
Figure 3.. FASD 4-Digit Code growth tables for converting height and weight percentiles into Growth Ranks.
The FASD 4-Digit Code documents growth deficiency on a 4-point Likert scale from Rank 1 normal growth to Rank 4 severe growth deficiency. Highlighted in red font is an example of an individual who presented with a height at the 8th percentile and weight at the 2nd percentile. Using the Growth Tables printed on the backside of the Lip-Philtrum Guides, these growth percentiles would translate into a Growth Rank 3, moderate growth deficiency.

Similar articles

Cited by

References

    1. Stratton K, Howe C, Battaglia F. Fetal Alcohol Syndrome: Diagnosis Epidemiology Prevention and Treatment. Institute of Medicine Washington D C National Academy Press; 1996.
    1. Astley S, Clarren S. Diagnosing the full spectrum of fetal alcohol exposed individuals: Introducing the 4-Digit Diagnostic Code. Alcohol Alcohol. 2000;35:400–10. - PubMed
    1. Astley S Diagnostic Guide for Fetal Alcohol Spectrum Disorders: The 4-Digit Diagnostic Code. 3rd ed. Seattle: University of Washington Publication Services; 2004.
    1. Cook J, Green C, Lilley C, Anderson S, Baldwin M, Chudley A, et al. Fetal alcohol spectrum disorder: a guideline for diagnosis across the lifespan. CMAJ. 2016;188(3):191–7. Epub 2015/12/17. doi: 10.1503/cmaj.141593. - DOI - PMC - PubMed
    1. Bower C, Elliot E. Report to the Australian Government Department of Health: Australian Guide to the diagnosis of fetal alcohol spectrum disorder. In: Health AGDo, editor. 2016.

LinkOut - more resources