Three specific facial features define fetal alcohol syndrome (FAS) in every major diagnostic system worldwide: unusually short eye openings (palpebral fissures), a smooth groove between the nose and upper lip (philtrum), and a thin upper lip. These features arise because alcohol disrupts the development of a particular group of embryonic cells during the first trimester, and they serve as visible markers of deeper harm to the brain. But the full picture is more nuanced than a checklist of three traits, involving subtle midface changes, variation across ethnic backgrounds, and a facial appearance that shifts as a child grows.
The Three Cardinal Features
Clinicians worldwide anchor their diagnosis of FAS on the same trio of facial characteristics. Short palpebral fissures refer to the horizontal width of the eye openings, measured from the inner corner to the outer corner. In children with FAS, these openings are narrower than expected for their age. A smooth philtrum means the two vertical ridges that normally run from the base of the nose to the top of the upper lip are flattened or absent, giving the area a featureless look. A thin upper lip vermilion refers to the red part of the upper lip being noticeably narrow. All three must be present, along with growth deficiency and central nervous system involvement, for a full FAS diagnosis.1PubMed Central. Facial dysmorphism across the fetal alcohol spectrum
These three features are sometimes called the “cardinal” facial signs because, although children with prenatal alcohol exposure can have a range of other physical differences, the diagnostic triad is the only set of facial traits formally required for a FAS or partial FAS diagnosis.2PubMed Central. Secondary physical features in children with FASD That distinction matters. Many alcohol-exposed children show some but not all of the cardinal features, placing them elsewhere on the fetal alcohol spectrum rather than in the full FAS category. The face, in other words, is the gatekeeper of the diagnosis.
Beyond the Big Three
While the diagnostic triad dominates clinical attention, researchers have identified a broader set of facial and craniofacial changes linked to prenatal alcohol exposure. A large imaging study of twelve-month-old infants found that alcohol-exposed children showed a general recession of the midface, a shortened nose with an upturned tip, and subtle differences around the eyes, lips, and chin. These shape differences were detectable even in children whose mothers reported only low levels of drinking during the first trimester.3JAMA Pediatrics. Association Between Prenatal Alcohol Exposure and Craniofacial Shape of Children at 12 Months of Age The maximum point-to-point difference between exposed and unexposed groups was about 1.2 millimeters, which sounds tiny but is well within the range that 3D imaging can reliably detect in infant faces.
Other commonly reported secondary features include a flat midface, a short upturned nose, epicanthal folds (small skin folds at the inner corners of the eyes), minor ear anomalies, and a slightly recessed chin. These features do not count toward a formal FAS diagnosis but are frequently noted during clinical evaluations, and their presence alongside the cardinal triad can increase a clinician’s confidence in the diagnosis.2PubMed Central. Secondary physical features in children with FASD
Why These Specific Features
The midface, philtrum, and upper lip all develop from the same pool of embryonic cells called neural crest cells. During the first weeks of pregnancy, these cells migrate from the developing brain region toward the face, where they become the bone, cartilage, and connective tissue of the nose, upper jaw, philtrum, and surrounding structures. Alcohol exposure during this critical window triggers a cascade of damage. It causes premature cell death in the neural crest population and disrupts the cells’ ability to migrate properly, leaving them unable to reach their intended destinations in the face.4PubMed Central. Neural crest development in fetal alcohol syndrome
One of the key molecular players is a signaling molecule called Sonic hedgehog (Shh), which guides the patterning of the face and brain during early development. Alcohol exposure dramatically reduces Shh levels, which in turn starves the neural crest cells of the signals they need to survive and organize. In animal experiments, directly adding Shh back to alcohol-exposed embryos rescued the neural crest cells from dying and partially prevented the craniofacial defects, confirming that Shh loss is a major driver of the characteristic FAS face.5PubMed Central. Sonic hedgehog rescues cranial neural crest from cell death induced by ethanol exposure
This shared developmental origin also explains why facial features and brain abnormalities tend to track together. The same neural crest cells that build the midface contribute to parts of the developing forebrain. When the insult is severe enough to reshape the face, the brain has usually sustained damage too.
Timing of Exposure Matters
Not all alcohol exposure during pregnancy produces the same effects. The classic FAS facial features are most strongly linked to drinking during the first trimester, when the face and brain are actively being assembled. Animal and human studies both confirm that craniofacial abnormalities are tied most clearly to early-pregnancy exposure, while behavioral and cognitive problems can arise from drinking at any stage.6PubMed Central. Critical Periods for Prenatal Alcohol Exposure: Evidence From Animal and Human Studies
The pattern of drinking also makes a difference. A study that tracked different drinking trajectories found that the three highest levels of sustained prenatal alcohol exposure each carried a two- to three-fold increased risk of a child having two or more of the cardinal facial features. Binge-level drinking was associated with the most pronounced differences around specific areas like the chin and lower lip. Sustained moderate-to-high exposure produced broader changes across the midface and eye region.7PubMed Central. Patterns of Prenatal Alcohol Exposure and Alcohol-Related Dysmorphic Features This is a practical point: there is no neat cutoff below which alcohol is “safe” for fetal development, but the risk of visible facial effects rises steeply with heavier and more sustained drinking in the first trimester.
How the Face Changes with Age
One of the less widely known aspects of FAS is that the characteristic facial features do not stay the same across a person’s life. They tend to be most recognizable during early childhood and least apparent during and after puberty. A longitudinal study that followed children from childhood through adolescence found that short palpebral fissures and small head circumference became less noticeable with age, while the thin upper lip remained relatively stable. The smooth philtrum fluctuated, sometimes appearing more and sometimes less pronounced at different stages.8PubMed. Evolution of the Physical Phenotype of Fetal Alcohol Spectrum Disorders from Childhood through Adolescence
By adolescence and adulthood, the faces of people with FAS are often described as no longer distinctive in the way they were during early childhood.9JAMA. Fetal Alcohol Syndrome in Adolescents and Adults This happens in part because the growth spurt at puberty changes facial proportions, the nose elongates, and the midface fills out. Studies confirm that most patients outgrow the characteristic facial appearance after puberty.10PubMed Central. A Long-Term Perspective of FAS This has important implications for late diagnosis. If a teenager or adult was never evaluated as a young child, the window for recognizing FAS from facial features alone may have closed, even though the cognitive and behavioral effects persist.
Racial and Ethnic Variation in Facial Features
One of the trickier aspects of FAS diagnosis is that normal facial anatomy varies significantly across racial and ethnic groups. What counts as a “thin” upper lip or a “smooth” philtrum depends partly on the baseline for a given population. About 20% of the variability in the core diagnostic facial features across different study populations is attributable to location and ethnic background.11PubMed Central. Population differences in dysmorphic features among children with fetal alcohol spectrum disorders
To address this, researchers developed separate visual reference guides for scoring lip and philtrum features. The original system includes one guide designed for Caucasians and populations with similar lip thickness, and a second guide for African Americans and populations with naturally fuller lips.12PubMed Central. Comparison of the 4-Digit Code, Canadian 2015, Australian 2016 and Hoyme 2016 fetal alcohol spectrum disorder diagnostic guidelines In South Africa, where FAS rates are among the highest in the world, clinicians found that applying the North American Caucasian guide to the mixed-race Cape Coloured population led to inaccurate diagnoses. A population-specific lip and philtrum guide proved to be more sensitive and specific for that group.13PubMed Central. A South African mixed race lip/philtrum guide for diagnosis of fetal alcohol spectrum disorders
This matters because misdiagnosis in either direction carries consequences. Applying the wrong reference standard could mean missing a diagnosis in a child who genuinely has FAS, or flagging features as abnormal when they fall within the normal range for that child’s background. The existence of multiple reference guides is not widely known outside specialist circles, and in communities where FAS prevalence is high but access to trained dysmorphologists is limited, misclassification remains a real concern.
Disagreements Among Diagnostic Systems
Several competing diagnostic frameworks exist for FAS and the broader fetal alcohol spectrum, and they do not always agree on how to score facial features. The 4-Digit Diagnostic Code, developed at the University of Washington, uses photographic lip and philtrum guides that assign scores on a 1-to-5 scale. A different set of guidelines, updated by Hoyme and colleagues, uses its own reference photographs. These two systems’ lip and philtrum guides differ enough that they are not interchangeable. A child scored using one system may receive a different facial severity rating under the other.14PubMed Central. Comparison of the FASD 4-Digit Code and Hoyme et al. 2016 FASD diagnostic guidelines
For families seeking a diagnosis, this can be confusing. The specific diagnostic label a child receives may depend partly on which system a clinician was trained in, not solely on the severity of the child’s features. In practice, the most experienced FASD clinics are aware of these discrepancies and use whichever guide best matches the child’s ethnic background and the diagnostic context. But the lack of a single universal standard is a genuine limitation of the field.
The Face as a Window Into the Brain
Because the face and brain develop from overlapping embryonic cell populations during the same critical period, the severity of facial features in FAS tends to predict the severity of brain involvement. Imaging research has shown that in alcohol-exposed individuals, shorter palpebral fissures correlate with reduced volumes in deep brain structures called the ventral diencephalon. Greater philtrum abnormality predicts smaller volumes in the basal ganglia and related structures. And lower IQ scores are associated with both smaller basal ganglia volumes and more pronounced facial features.15PubMed Central. Regional brain volume reductions relate to facial dysmorphology and neurocognitive function in fetal alcohol spectrum disorders
This face-brain connection extends to subtler measures of brain organization. In healthy brains, certain structures show a characteristic left-right asymmetry. In individuals with FAS, that normal asymmetry is diminished or absent, and a similar but less severe pattern appears in heavily alcohol-exposed individuals who lack the full FAS facial features. The degree of brain asymmetry loss correlates with neurocognitive test performance.16PubMed Central. Combined Face-Brain Morphology and Associated Neurocognitive Correlates in Fetal Alcohol Spectrum Disorders In plain terms, the face is not just a diagnostic marker. It is a rough gauge of how much the brain was affected.
Genetic Vulnerability and Why Some Children Are Affected More Than Others
Two pregnant women can drink similar amounts and have children with very different outcomes. Part of this variability comes down to genetics, particularly the enzymes that metabolize alcohol. Variations in genes like ADH1B and ALDH2, which code for the enzymes that break down ethanol and its toxic byproduct acetaldehyde, influence how quickly a mother’s body clears alcohol and how much reaches the fetus.17PubMed. Genetic and epigenetic determinants of fetal alcohol spectrum disorders: Toward a precision medicine approach
Specific evidence for this comes from studies showing that when both mother and infant lack a particular protective variant of the ADH1B gene (the ADH1B*3 allele), and the mother drinks during pregnancy, the infant’s facial measurements are smaller than in cases where one or both carry the protective variant.18PubMed. Alcohol dehydrogenase 2*3 affects alterations in offspring facial morphology associated with maternal ethanol intake in pregnancy This gene-environment interaction helps explain why the facial features of FAS do not appear in every alcohol-exposed child. The fetus’s own genetic makeup, not just the mother’s drinking, shapes the severity of the outcome.
Epigenetic changes add another layer. Alcohol exposure during early embryonic development alters DNA methylation patterns across dozens of genes involved in development, metabolism, and brain function.19Epigenetics. Alcohol exposure alters DNA methylation profiles in mouse embryos at early neurulation These methylation changes can shift how active or quiet certain genes are without changing the underlying DNA sequence, potentially contributing to the variability seen in FASD outcomes even among siblings exposed during the same pregnancy.
Computer-Aided Recognition and 3D Imaging
Human clinicians, even experienced ones, are not perfectly reliable at scoring facial features. The assessment is subjective, shaped by training, experience, and the reference guides being used. This has driven interest in automated methods. Researchers have developed 3D facial imaging systems that can map the surface of a child’s face and analyze it using algorithms trained on known FAS and control faces. One such system demonstrated successful automated discrimination between FAS and control faces without any human intervention, identifying unique sets of facial regions that differed between groups.20PubMed Central. Automated diagnosis of fetal alcohol syndrome using 3D facial image analysis
Computer-aided recognition tools hold promise for screening in settings where specialist dysmorphologists are scarce. Early identification of children at risk for FASD allows earlier access to developmental support, speech therapy, and educational accommodations, which evidence suggests improves long-term outcomes.21Pediatrics. Computer-Aided Recognition of Facial Attributes for Fetal Alcohol Spectrum Disorders These tools are not yet standard in clinical practice, but they represent a real shift in how facial phenotyping might eventually work, especially in primary care and in regions where FASD prevalence is high but diagnostic expertise is concentrated in specialized centers.
Oral Clefts and Co-occurring Facial Malformations
Beyond the FAS facial triad and the subtler midface changes, heavy prenatal alcohol exposure is also linked to more severe structural birth defects of the face. A population-based study in Norway found that women who reported binge-level drinking (five or more drinks in a sitting) were roughly twice as likely to have an infant with cleft lip, with or without cleft palate. For cleft palate alone, the odds were about two and a half times higher. Among women who binged on three or more occasions, the risk of cleft lip rose to about three times the baseline rate.22PubMed Central. First-trimester maternal alcohol consumption and the risk of infant oral clefts in Norway: a population-based case-control study
Cleft lip and palate are not part of the FAS diagnostic criteria, and most children with FAS do not have them. But the association underscores the broader vulnerability of facial development to alcohol during the first trimester. The same neural crest cell populations disrupted in FAS are the ones responsible for fusing the palatal shelves and forming the upper lip. When the exposure is extreme enough, the result can be a structural gap rather than just a subtle flattening.
Animal Models and What They Add
Much of what we know about the dose-response relationship between alcohol and facial development comes from animal research, which allows controlled exposures that would be unethical to study directly in humans. Mouse studies have shown that alcohol treatments produce measurable changes in facial dimensions, and that different genetic strains of the same species respond differently. In one study using two closely related strains of laboratory mice, alcohol exposure produced significant changes in four out of fifteen facial measurements in one strain but only two measurements in the other. Automated classification correctly identified alcohol-exposed versus unexposed mice with about 80-86% accuracy, depending on the strain.23PubMed Central. Alcohol-induced facial dysmorphology in C57BL/6 mouse models of fetal alcohol spectrum disorder The strain differences parallel the genetic variability seen in humans and reinforce the idea that identical alcohol exposure does not produce identical facial outcomes.
Why Many Cases Go Unrecognized
Despite decades of research, FASD remains widely underdiagnosed. Several factors contribute. The facial features are most visible in early childhood but can fade with puberty, narrowing the diagnostic window. Many children on the fetal alcohol spectrum lack the full facial triad and receive no physical-feature-based diagnosis at all, even though they carry significant cognitive and behavioral effects. Screening tools exist but are not yet standard in pediatric practice.24PubMed. Fetal alcohol spectrum disorders screening tools: A systematic review
There is also a social dimension. Maternal alcohol use carries stigma, and mothers may underreport drinking during pregnancy, leaving clinicians without the exposure history needed to consider FASD. Foster and adoptive parents often have no access to prenatal records at all. In these situations, the face may be the only clue, and if the child’s ethnic background does not match the reference guide being used, or if the clinician has limited FASD training, the features can be missed entirely. The combination of a narrow diagnostic window, population-specific facial norms, competing diagnostic systems, and incomplete exposure histories makes FASD one of the most challenging conditions to catch consistently, despite having facial markers that are, in theory, visible to the trained eye.