Chairman of Urology
Weill Cornell Medical Center
Former President, ASRM
Director, Male Reproductive Medicine
Mount Sinai Medical Center
Male factor infertility can be influenced by lifestyle decisions and here we’ll chronicle what the data offers. A word of caution, the quality of data in this field is poor. Many of the endpoints are based on whether the subject in question (e.g. obesity) impacts the semen analysis.
Why is that a problem? As we discussed in the lesson on the semen analysis, there’s not a perfect correlation between the quality of the semen analysis and a man’s ability to conceive. In fact, men are often still able to conceive naturally when a semen analysis parameter falls below normal. Thus, improving sperm parameters can be an imperfect proxy for investigating how one behavior change impacts a man’s ability to conceive. That said, it’s often the best proxy we have.
Smoking has a negative impact on sperm concentration and morphology. One illustration, shown below, is a retrospective study on the semen parameters of 614 Singaporean men. The data shows men who smoke have statistically lower levels of sperm concentration and morphology. While the study is imperfect (no “live birth rate” endpoint), it does a nice job of tracking men who have few other issues (excessive drinking, a partner who smokes) making it easier to isolate the possible impact of smoking.
And when the investigators sub-segmented out the pool of smokers by their smoking frequency (cigarette years = cigarettes per day x number of years smoking), a similar trend was noted. This implies that best we can tell, any level of smoking hurts a man’s results on the semen analysis.
And in case you think, “well, we can still get pregnant with poor semen parameters by using IVF and ICSI,” in the case of smoking, you’d be wrong. A German team looked just at this issue in 2003 and showed that amongst over 700 couples, IVF cycles, both with and without ICSI, had far lower pregnancy rates when the man smoked (again, they didn’t study live birth rates, which is a shortcoming). You can see the findings below. After correcting for a few variables, the authors determined that a man’s smoking was a major impediment to IVF’s success, even if ICSI was used in the process.
A large, well-designed study of couples in Michigan and Texas looked at how lifestyle choices impact rates of miscarriage. As you can see below, investigators determined that amongst couples who conceive there is a 73% higher rate of miscarriage if the man drinks at least two or more caffeinated beverages per day.
In the context of a couple doing IVF, male consumption of caffeine probably begins to have an impact after around 200 milligrams per day, or about 1.5 – 2.0 caffeinated drinks per day. Below is the work from the team at Massachusetts General Hospital, who observed only a small number of patients, but did a nice job isolating variables (like smoking and alcohol consumption) and looked at the real world endpoint of live birth rate.
We have less conviction on the role alcohol plays in a man’s ability to conceive. Similar to smoking, most of our studies only measure how alcohol consumption impacts semen parameters, not live birth rates. Furthermore, there is a larger number of studies on this subject that make differing observations and come to differing conclusions.
The best study in the field, we believe, measures semen parameters amongst a large number (1,200+) of 18 – 20 year old Danish men. As you can see below, the investigators saw a plausible impact on sperm concentration at 6 – 10 drinks per week, and then a more consistent impact above 25 drinks per week. Indeed, most reproductive urologists believe alcohol inhibits male fertility, but has a “threshold effect” in that only above certain levels does drinking alcohol become problematic.
A natural follow-on question is, how does male alcohol consumption impact the results for a couple who is beginning IVF? With the data we have available today, it’s unclear. The same team at Massachusetts General Hospital that studied caffeine’s effects looked at this issue. The MGH team noted that men in the highest drinking quartile (1.4 drinks per day) had higher live birth rates than the men in the lowest quartile. All the same, the sample set was small, the men in the top quartile still drank in relative moderation (again, 1.4 drinks per day) and we do not believe men should read into this data that increased alcohol consumption is helpful.
A male’s weight likely correlates with his ability to conceive naturally. When a Norwegian team analyzed 20,000 patients of retrospective data, and isolated out confounders (mainly their female companion’s body mass index), men with a higher body mass index were nearly 50% less likely to conceive naturally after 12 months. Typically, men who are of higher weight overproduce estrogen and struggle to produce testosterone, likely impacting their ability to produce sperm.
What’s also interesting here is that men with a lower body mass index (e.g. less than 20) were possibly the least likely to conceive quickly.
While this trial has flaws (e.g. patients self-reported height and weight), we like the sample size, statistical approach and fact that the endpoint of pregnancy is closer to live birth rate than simple changes in semen parameters.
When we look at how male weight, or body mass index, impacts a couple’s success rate during IVF, the answers are inconclusive. One large meta-analysis (a study of studies, which has its own issues) concluded overweight men who participate in IVF have lower success rates than men of normal weight. But again, the overall body of data is conflicting and hard to interpret.
Sedentary men are over ten times more likely to experience erectile difficulties, but the issues may extend beyond that. In the below study, investigators followed 189 18 – 22 year olds and charted their sperm parameters against self-reported levels of physical activity per week. After correcting for other variables, the investigators saw a strong correlation with sperm count, which is a function of concentration.
At the same time, most reproductive urologists believe there is such a thing as too much exercise, and after 90 minutes of daily exertion a man may start doing more harm than good. The underlying premise is that prolonged exertion amplifies heat and pressure in the testicles, which reduces sperm production (this may be the same mechanism of action that impacts varicocele, which you’ll read about in a following lesson).
There is also a fair amount of data that biking in particular, even as little as two hours per week, can negatively impact sperm concentration. Below is an analysis that shows an increased level of exertion can correlate with lower sperm concentration and worse morphology. However, the analysis suffers from a small sample size (45 participants) who do differing activities (e.g. the triathlete group participates in cycling).
Our data on drug use correlates to semen parameters but we have little data on its relationship to live birth rates. What’s more, running prospective trials gauging the effects of drug use are impossible and retrospective studies suffer from the degree to which patients are forthcoming about their drug use and other variables (e.g. smoking). As a result, we have less understanding about the effects of drug use than we do about smoking, drinking and other modifiable factors.
There is data that shows marijuana use impacts semen parameters, with the most rigorous data coming from a 1,200 patient study in Denmark. Conscripted soldiers reported drug use and had their responses charted against their semen analysis. As you can see from the data below, routine use of marijuana coincided with reduced sperm concentration levels by nearly 30%.
As you can see in the data below, these investigators then compared the semen parameters of men who used marijuana once per week with those who used it more often, and those who used it with other recreational drugs (which was not clearly defined). Those who used marijuana and recreational drugs recorded dramatically lower sperm concentrations.
However, data on the subject is not altogether easy to interpret. Data from Massachusetts General Hospital, which we show below, reveals marijuana use may coincide with improved semen parameters and some studies that have looked at "time to pregnancy" show no difference between men who use marijuana and those who do not. Ultimately, this is a tricky subject that requires more study before hard conclusions can be drawn.
There is reasonably persuasive data showing that prescription opioids play a role in reduced testosterone production (which is necessary to generate sperm) and sexual drive. All the same, as we’ll discuss in the next lesson, taking exogenous testosterone does not make up for this effect, and in fact worsens it.
As for cocaine, one study out of Yale revealed that men who had done cocaine within the last two years were nearly twice as likely to have sperm concentrations below 20 million. Alternatively, there is almost no data in humans on the role that MDMA or ecstasy could play in male infertility.
A number of studies suggest that men who use their cellular phones for a longer duration have severely impacted semen parameters. As you can see in the study below conducted on 360 men, concentration, motility and morphology parameters were diminished. While this study doesn’t prove causality (that the phone use itself was responsible), the trend’s strength is compelling and when coupled with the demonstrated impact electromagnetic waves can have on testicular function, it seems to reinforce the notion of a cause-and-effect relationship.