Small at Birth, but How Small? The Definition of SGA Revisited

50 years of published discussion of the definition of SGA [3–5] . The 10th percentile was chosen as a cutoff for SGA in the 1960s as a result of multiple studies indicating that infants born at or below the 10th percentile have increased mortality compared to gestational age-matched controls [4] . It is worth noting that these studies were performed on neonates born at high altitudes, who tend to be smaller compared to those born at sea level [4, 5] . Defining SGA as 2 SDs below the mean was also suggested in the 1960s, as it would only define 4.6% of births as abnormal (approximately 2.3% SGA and 2.3% large for gestational age), as opposed to 20% (10% SGA and 10% large for gestational age). Additionally, the 2 SD cutoff roughly corresponded with an earlier study reporting that 2.3% of neonates were born at 25% below the mean weight when controlled for gestational age [5] . In 1995, the World Health Organization published recommendations defining SGA as less than the 10th percentile of weight for gestational age using localized anthropometric newborn curves [3] . In 2007, a consensus meeting that included representatives from seven international pediatric endocrinology societies, as well as a representation from obstetrics, perinatology and neonatology, pediatrics, epidemiology, and pharmacology, recommended that SGA be defined as more than 2 SDs below the mean for weight and/or length [2] . They also recBeing born small has been associated with many shortand long-term health sequelae. Perinatally, it is associated with increased mortality, lung disease, hypotension, necrotizing enterocolitis, poor thermoregulation, hypoglycemia, and polycythemia [1, 2] . Long-term, small infants are at risk for insulin resistance, type II diabetes mellitus, cardiovascular disease, chronic kidney disease, neurodevelopmental and cognitive impairments, developmental delays, behavioral problems, and adult short stature [1] . But how ‘small’ is ‘too small’? The term ‘small for gestational age’ (SGA) describes newborns who have lowerthan-expected weight, length, and/or head circumference when controlled for gestational age and sex. This is different from intrauterine growth retardation, which refers to poor growth in utero evidenced by at least two ultrasound measurements [2] , and prematurity, which is a broad term defining neonates born prior to 37 weeks gestation. Some, but not all, infants with intrauterine growth retardation and/or prematurity may be born SGA. In spite of potential significant health implications, the exact definition of SGA remains elusive. Multiple criteria have been used, including less than the 10th, 5th, and 3rd percentile in weight, length, or head circumference. Another definition of SGA is parameters more than 2 standard deviations (SDs) below the mean, or the 2.3rd percentile [1, 2] . These discrepancies are not novel, with over Received: August 11, 2016 Accepted: August 12, 2016 Published online: September 30, 2016 HORMONE RESEARCH IN PÆDIATRICS