Men with low testosterone levels after age 40 have a higher risk of death over a four-year period than those with normal levels of the hormone, a new study suggests.

It's not clear, however, if the two are directly related, and researchers say that it's possible a third unknown factor is responsible for both low testosterone levels and increased mortality.

The study, led by Molly Shores of the VA Puget Sound Health Care System and the University of Washington, Seattle, is detailed in the current issue of Archives of Internal Medicine.

Wild variation

Testosterone levels can vary wildly between different men and even within individuals as a result of illness or stress. However, studies have consistently shown that levels of the hormone decrease as men age.

Testosterone levels peak during a man's late 20's but decline soon after, decreasing about 1.5 percent per year after age 30. The decline is thought to be due to age-related changes in the testes, where the hormone is produced, and in the brain.

Low levels of testosterone can cause decreased muscle mass and bone density, insulin resistance, decreased sex drive, reduced energy, irritability and feelings of depression.

The researchers studied the relationship between testosterone levels and death in a total of 858 male veterans who were age 40 or older. Participants had their testosterone levels checked at least twice between 1994 and 1999. The men were followed for an average of 4.3 years and a maximum of eight years, through 2002.

About 19 percent (166 men) had low testosterone levels; 28 percent (240 men) had varying low and normal levels; and 53 percent (452 men) had normal testosterone levels.

About 20 percent of the men with normal testosterone levels died during the course of the study, compared with 24.6 percent of men with varying levels and 34.9 percent of those with low levels.

Men with low testosterone levels had an 88 percent increase in risk of death compared with those who had normal levels. The effect persisted even after variables such as age, other illnesses and body mass index were controlled for.

The mortality risk for men with low testosterone dropped to 68 percent after the researchers reanalyzed the data to eliminate men who had died within the first year of follow-up. This was done to control for dips in testosterone levels that can occur shortly after surgery, trauma or critical illness.

Uncertain link

That the effect still remained high after excluding early deaths suggests the link between low testosterone and mortality is not simply due to acute illness, the researchers say.

However, because the current study is a retrospective study that relies on old medical records, it can only conclude that low testosterone and increased mortality are linked, not how they are linked. For that, a prospective study that follows a group of individuals over a long period of time is needed.

"The men with low testosterone did have higher death rates, but it may be due to some other factor that we weren't able to measure," Shores told LiveScience.

Shores also points out that the veterans in her study are a unique subset of the population and that future studies should look at more diverse groups of men.

"They were probably more medically ill than most men their age," she said. "They all had an average of about 5 chronic illnesses."

Shores said the results of her study are not conclusive enough for her to recommend testosterone replacement therapy for aging men. Furthermore, the overall risks and benefits of such therapy are not yet known.

"There's no link between testosterone replacement and prostate cancer but the studies that have been done aren't big enough to detect if there's a link or not," Shores said.

Sex Still Satisfying For Men In Their 50sGuys, Put a Cork In It: Tomorrow's ContraceptivesFatherhood at First Sight: Women Can Spot Good Dads By Their FacesBirds Gone Wild: Extra Testosterone Makes Males IrresistibleWhy Men Report More Sex Partners than WomenThe Truth on Longer Life Spans