The decline was 52.4 percent -- from 99 million to 47 million sperm per millilitre of semen
Sperm count among men in North America, Europe, Australia and New Zealand has halved in 40 years, according to research warning of fertility risk, though outside experts urged caution about the results.
A review of data collected from nearly 43,000 men in 185 previous studies from 1973 to 2011, found a "significant decline" in sperm concentration, according to findings published in the journal Human Reproduction Update.
The decline was 52.4 percent -- from 99 million to 47 million sperm per millilitre of semen, said the authors.
This was still above the range considered "normal" by the World Health Organization -- between 15 million and 200 million sperm per millilitre. Even a concentration under 15 million does not automatically mean infertility.
The research team found no significant decline in South America, Asia, and Africa, where fewer studies had been conducted.
Apart from fertility risks, reduced sperm count could be an indicator of underlying health problems caused by environmental factors such as pesticide poisoning, or stress, smoking, a poor diet and other lifestyle factors.
The topic is a controversial one, as numerous studies in the past have contradicted one another -- some showing a sperm count decline, and others not.
The latest work seeks to address some potentially confounding factors.
It excluded data from men recruited for studies because of fertility problems, and used only studies in which sperm concentration was measured with a hi-tech tool called a haemocytometer.
This sidesteps the problem of comparing samples measured with incomparable technologies -- older tests tended to overestimate sperm count.
The work had other limitations, according to male fertility expert Martin Blomberg Jensen of Denmark's Righospitalet.
It compared samples from different individuals from different countries and laboratories. The best way to confirm changes in sperm count would be through long-term surveillance, known as a prospective cohort study, of a group of healthy males.
The latest figures were not backed by the findings of one such study, conducted from 1996 to 2010, in Denmark, said Blomberg Jensen.
The new work contributed useful data, added Allan Pacey of the University of Sheffield, but "the debate has not yet been resolved and there is clearly much work still to be done."