The researchers collected semen and blood samples from hundreds of men at the Massachusetts General Hospital fertility clinic between 2000 and 2017, and asked them about their marijuana use.
The results contradicted expectations, says Jorge Chavarro, an associate professor at the Harvard Chan School of Public Health.
"We found that men who reported using marijuana had higher sperm counts than men who reported never having used marijuana in their lives," Chavarro says.
The study cannot be taken to mean that marijuana is good for male fertility, Chavarro emphasizes; more research on that is needed. But it does underscore how little we really know about the health effects of marijuana, and "knowing that we don't know is important," he says.
Men who reported smoking marijuana also had higher average sperm concentrations in their semen than men who had never smoked, and were much less likely to have abnormally low levels, the study found. It also found higher testosterone levels among heavier marijuana smokers.
The findings, published in the journal Human Reproduction, diverge from those of previous studies — including a large Danish study of young military conscripts, and animal studies — that suggested marijuana use could reduce sperm count or produce mixed effects on fertility.
When the numbers first came in, Chavarro says, the research team double-checked their data and confirmed them.
But cause and effect remain to be determined. It could well be, he says, that the numbers reflect "reverse causation." That is, men with higher testosterone levels are more likely to use marijuana — perhaps in part because they're more willing to take risks — and men with higher testosterone levels also tend to have higher sperm counts.
The study also raises the possibility that marijuana use could affect sperm production via a neurotransmitter system in the body known as the endocannabinoid system.
It's also possible that some previous studies finding harm to men's fertility were confounded by subjects who used multiple drugs as well as marijuana, the study notes.
There are more caveats: It is not clear whether results found in men being seen at a fertility clinic can be generalized to all men. Also, the study depended on men's own reports of their marijuana use, and some of their self-reports may have been inaccurate.
So would Chavarro describe the results as reassuring for men who use marijuana?
The research team agonized over that wording, he says. The paper concludes carefully: "These findings are not consistent with a deleterious role of marijuana smoking on testicular function as initially hypothesized."
"I really think that the most important thing about this paper is highlighting how little we know at the moment," Chavarro says.