Measurement of length/stature must be accurate and reproducible in order to correctly assess the degree of interval growth. Children >2 years should be measured supine, children >3 years should be measured standing, and children between ages 2 and 3 years can be measured either way, depending on the child’s ability to stand erect for the time it takes to be measured. Recumbent lengths should be recorded on the length (ages from birth to 36 months) chart and standing heights on the height (ages 2 to 20 years) chart, as they will differ slightly. Recumbent length is best measured with the child’s head against an inflexible board in the Frankfurt plane (wherein the long axis of the trunk is perpendicular to the line formed by connecting the inferior margin of the eyes to the top of the external auditory meatus), legs fully extended, and the feet placed perpendicularly against a movable footboard. To measure height, a stadiometer fixed to the wall is most accurate. The child’s feet are placed together in parallel to each other, and the child’s heels, buttocks, thoracic spine, and back of head should all be up against the stadiometer’s vertical axis. The child should be standing fully erect, heels touching the ground, and the head must be in the Frankfurt plane (see above).

Evaluation of a child’s current height and overall growth pattern occurs within the context of standards typical for the population. Most general pediatric and pediatric endocrine offices use the Centers for Disease Control (CDC) growth charts—percentile curves that illustrate the distribution, in U.S. children, of length or height, weight, head circumference (for ages from birth to 36 months), and body mass index (BMI + weight (kg)/height (m)² for ages 2 to 20 years). The data used to construct the 2000 growth curves came from the National Health and Nutrition Examination Survey. The CDC growth curves were derived from cross-sectional population surveys, rather than longitudinal surveys that follow the same children over a course of years to determine their growth rates over time. Therefore, an individual’s growth pattern may differ somewhat from the standardized pattern, especially during periods of rapid growth (infancy and the pubertal growth spurt).

In April 2006, the World Health Organization (WHO) released new international growth charts for children aged 0 to 59 months. Similar to the 2000 CDC growth charts, these charts describe weight for age, length (or stature) for age, weight for length (or stature), and body mass index for age, but the methodology for the infant growth charts differed. The growth of infants aged 0 to 24 months was tracked longitudinally, although the growth data on older children (up to 71 months of age) was obtained cross-sectionally; thus, the WHO charts for infants aged 0 to 24 months are in fact true growth standards, describing the linear growth of healthy children under optimal conditions. In September 2010, the CDC recommended that clinicians in the United States use the 2006 WHO growth charts, rather than the 2000 CDC growth charts, for tracking linear growth of children aged 0 to 24 months.

WHO charts are available online at www.who.int/childgrowth/standards/en.

Because genetics play a large role in determining both the rate of maturation and the eventual adult height, it is important to consider the heights of the biological mother and father in order to place the child’s current stature and predicted height in proper context. A significant deviation from the familial growth pattern should raise suspicion of possible pathological processes.

Several conditions underlying growth abnormalities cause disproportionate growth (e.g., extremities more shortened than trunk in SHOX insufficiency, skeletal dysplasias, and chondrodysplasias). Thus, measurement of the following body proportions can aid the diagnosis of the underlying defect:

Published standards exist for each of these measurements relative to the child’s age.

The determination of bone age is predicated on the fact that ossification centers appear and progress in a predictable sequence in most children, so that a radiograph may be compared to a standard. The bone age, or skeletal age, is the only quantitative determination of somatic maturation; thus, it mirrors the tempo of growth and maturation and gives an indication of the remaining growth potential. Once the child is beyond infancy, an anteroposterior (AP) radiograph of the left hand and wrist is taken and compared to published standards (Greulich-Pyle or Tanner-Whitehouse method); knee films can be used at ages when hand/wrist films are not yet informative. Factors that impact the rate of skeletal maturation include genetic predisposition, thyroid hormone, growth hormone (GH), insulin-like growth factor (IGF)-1, glucocorticoids, and estrogens (in children of both sexes). Once a bone age is assigned, this value, in combination with the child’s present height, is entered into one of several commonly used algorithms (Bayley-Pinneau, Roche-Wainer-Thissen, or Khamis-Roche method) to generate an adult height prediction. However, as demonstrated in a study published in Pediatrics in October 2010, a fair amount of variability exists between the predicted heights generated by these various methods. Thus, such predictions may be less reliable than was previously believed to be the case.

Both parents’ heights are usually in the lower height percentiles, often below the 10th. The child’s pattern of growth is consistent with that of his/her parents (and often siblings, too). The child’s growth velocity is usually normal, so that the growth curve is low, but parallel to the normal lines. The predicted adult height is in keeping with midparental target height. In the absence of a family history of delayed puberty, the timing of puberty is usually average.

The past medical history, review of systems, and physical examination are typically unremarkable, with no abnormalities of body proportions in either the patient or the parents. As skeletal growth is not delayed, the bone age should be within the normal range. Laboratory workup, including growth factors and a chemistry panel, should be within the normal range. Abnormalities of history, physical examination, or laboratory studies should prompt the investigator to look for other causes of short stature.