Modifiable and non-modifiable risk factors for poor semen quality: a case-referent study.

OBJECTIVE

Are common lifestyle factors associated with low-motile sperm concentration (MSC)?

CONCLUSIONS

Common lifestyle choices make little contribution to the risk of low MSC.

BACKGROUND

Reviews of male subfertility often highlight how aspects of men's adult lifestyle can significantly increase their risk of subfertility but the strength of supporting evidence is weak. In this study, although low MSC was associated with a history of testicular surgery, being in manual work, not wearing loose underwear and black ethnicity, no relation was found to consumption of alcohol, use of tobacco or recreational drugs or high body mass index (BMI). These results suggest that delaying assisted conception to make changes to lifestyle is unlikely to enhance conception.

METHODS

Unmatched case-referent study with 939 cases and 1310 referents. Cases had a low-MSC relative to the time since last ejaculation (<12 × 10(6) for 3 days of abstinence). Exposures included self-reported exposures to alcohol, tobacco, recreational drugs as well as occupational and other factors.

METHODS

Eligible men, aged 18 or above, were part of a couple who had been attempting conception without success following at least 12 months of unprotected intercourse and also had no knowledge of any semen analysis. They were recruited from 14 fertility clinics across the UK during a 37-month period from 1 January 1999.

RESULTS

Risk factors for low MSC, after adjustment for centre and confounding factors, included a history of testicular surgery [odds ratio = 2.39, 95% confidence interval (CI): 1.75, 3.28], being in manual work [odds ratio (OR) = 1.28, 95% CI: 1.07, 1.53] or not working (OR = 1.78, 95% CI: 1.22, 2.59) and having black ethnicity (OR = 1.99, 95% CI: 1.10, 3.63). Conversely, men who wore boxer shorts (OR = 0.76, 95% CI: 0.64, 0.92) or who had a previous conception (OR = 0.71, 95% CI: 0.60, 0.85) were less likely to be a case. No significant association was found with smoking and alcohol consumption, the use of recreational drugs, a high BMI or having a history of mumps or fever.

CONCLUSIONS

Data were collected blind to outcome, and exposure information should not have been subject to reporting bias. Among men attending the various clinics less than half met the study eligibility criteria and among those who did, two out of five were not recruited. It is not known whether any of those who refused to take part did so because they had a lifestyle they did not want subjected to investigation. Although the power of the study was sufficient to draw conclusions about common lifestyle choices, it cannot comment on exposures that are perhaps rare and poorly reported: the finding that use of street drugs was unrelated to low MSC cannot be assumed to apply to all such drugs and all patterns of use. The case definition did not consider sperm morphology or sperm DNA integrity.

CONCLUSIONS

All participating clinics saw patients at no cost (under the UK National Health Service) and the study population may differ from those in countries without such provision. Even within the UK, low-income couples may choose not to undertake any investigation believing that they would subsequently be unable to afford treatment.