The NFL has been testing for banned substances since 1987. (Damian Dovarganes/Associated Press)

A survey of retired NFL players found that steroid use increases the risk of musculoskeletal injuries.

More than nine per cent of the 2,552 former players to respond to the confidential survey, to be published in the March issue of the American Journal of Physical Medicine & Rehabilitation, admitted to using steroids during their pro football careers.

That percentage rose to 20.3 per cent among players from the 1980s, when the survey suggested "it was acceptable" to use steroids.

The NFL began testing for banned substances in 1987 — for informational purposes — and introduced steroid enforcement in 1989, followed by random testing one year later.

Overall, the highest percentage of players to use steroids were offensive linemen (16.3) and defensive linemen (14.8), and, in turn, they registered a higher rate of injuries like neck stingers, herniated disks, and problems with joint ligaments and cartilage in their ankles, elbows, feet and toes.

"Our findings speak to the compounded medical problems that appear with steroid use, with negative effects on joint health, starting a 'snowball effect' that can lead to other chronic diseases later in life," said senior author Kevin Guskiewicz, a professor of exercise and sports science at the University of North Carolina.

'Weakest link in the chain'

The injury patterns suggest that joint ligaments and cartilage, rather than tendons, might not adapt to muscle changes brought about by steroids, making them "weakest link in the chain," said lead author Dr. Scott Horn of the UNC School of Medicine.

The survey indicated no steroid related increase in the risk of injury to the muscle (biceps and triceps) and no increase of injury to shoulders and tendons.

Based on previous studies, the researchers suspected that tendons might be at increased risk to cope with a rapid increase in muscle size and strength.

"More data are needed to further examine the relationship between [steroid] use and musculoskeletal injury," the survey said.

Respondents who admitted to using steroids also showed an increased risk of osteoarthritis, depression and alcohol use later in life, and, for the most part, were less able to maintain a physically active lifestyle in retirement.

"I call it a snowball effect," Guskiewicz wrote in the survey, "with joint injuries contributing to the gradual development of other medical problems such as osteoarthritis, physical inactivity, depression, obesity and diabetes."

The survey was sent to the 3,683 living members of the NFL Retired Players Association, but it lacks key information like the type, dosage and length of time steroids were used.