Growth hormone use: How short is too short?

November 8, 2004

The Food & Drug Administration has created a conundrum for short children, their parents, pharmacists, and pediatricians. The agency approved the use of human growth hormone (somatropin recombinant, Humatrope, Eli Lilly) in June 2003 to treat idiopathic shortness. But instead of clarifying which children should be treated for short stature, the FDA may have opened the door to abuse.

The Food & Drug Administration has created a conundrum for short children, their parents, pharmacists, and pediatricians. The agency approved the use of human growth hormone (somatropin recombinant, Humatrope, Eli Lilly) in June 2003 to treat idiopathic shortness. But instead of clarifying which children should be treated for short stature, the FDA may have opened the door to abuse.

"The indication seemed very clear at the time, but it is much more complicated," said Ron Rosenfeld, M.D., senior VP for medical affairs at the Lucile Packard Foundation for Children's Health at Stanford University. "The question has really become, how short is too short?"

Human growth hormone was originally approved in 1985 to treat growth hormone deficiency in children, Rosenfeld told the American Academy of Pediatrics National Conference & Exhibition, last month in San Francisco. And it works. More than 90% of growth hormone-deficient children who receive the drug attain normal adult height, he said. In the United States, the average adult male is 5 ft. 10 in., the average female 5 ft. 5 in. Under the FDA definition, idiopathically short boys would grow to less than 5 ft. 3 in. as adults.

A year after the idiopathic shortness indication was approved, demand shows signs of doubling. The sales boom is easy to understand, said Rosenfeld. All of the clinical indications combined have an incidence of about one child per thousand in the general population. The idiopathic shortness indication includes all children who are at least 2.25 standard deviations shorter than normal, one child in every hundred.

"None of us believe that 1% of children should be treated," Rosenfeld said. "But there are seven drug manufacturers hungering after this market. The idiopathic indication could become a bonanza that fosters questionable 'short centers' for children the way we already have dubious 'antiaging centers' that target older adults."

Candidates for growth hormone treatment are usually identified by pediatricians who see that a child's growth rate is falling below the standard curve for his or her age. Growth rates vary widely by age and by individual, Rosenfeld said, but any child who is growing less than two inches per year should be evaluated. A few children are identified soon after birth, he added, but most patients are not identified until three to four years of age. It is not unusual to find patients who are seven or eight years old before they are diagnosed.

When treating clinical conditions such as growth hormone deficiency, clinical goals are relatively clear, Rosenfeld said. The endpoints are murkier when treating undefined shortness, he said. Is it enough to boost the short child into the bottom end of the normal range? Should clinicians aim for the center of the bell curve that defines height distribution? What role should parental desire play in treatment decisions? Just two decades ago, he noted, parents were asking pediatric endocrinologists to make their tall daughters shorter.

The problem, Rosenfeld explained, is that growth hormone works almost too well. It is highly effective in stimulating growth in children. Recombinant human growth hormone has shown virtually no serious side effects until well after growth plates fuse during adolescence. In theory, he noted, growth hormone could drive children to seven feet or taller. "No responsible physician is going to do that," he said. "But pituitary abnormalities that cause excessive production of growth hormone routinely produce gigantism."

Children who are short for no obvious reason can be just as short as those with specific clinical conditions, but shortness itself is not a disease. Conventional wisdom suggests that taller people are more successful in life, but there are no reliable data on the psychological or social impact of short stature. And there are no data showing that increasing height resolves any psychosocial handicap associated with shortness. "Growth hormone deficiency is clinical," Rosenfeld said. "Short is subjective. Every pediatrician keeps private records on the tallest child whose parents come in asking for growth hormone therapy."

Rosenfeld's personal records included a boy who was 5 ft. 10 in. His father, a physician, insisted that the boy could become a basketball superstar if he were just a little taller. Rosenfeld said he declined to treat the child.