A prospective cohort study of vasectomy and prostate cancer in US men
E. Giovannucci, A. Ascherio, E. B. Rimm, G. A. Colditz, M. J. Stampfer and W. C. Willett
Channing Laboratory, Department of Medicine, Harvard Medical School, Boston, MA 02115.
OBJECTIVE--To examine prospectively the relationship between vasectomy and
prostate cancer. DESIGN--Cohort study. SETTING--Health professionals
(dentists, veterinarians, osteopaths, optometrists, pharmacists, and
podiastrists) in the United States. PARTICIPANTS--There were 10,055 male
members of the Health Professionals Follow-up Study, aged 40 to 75 years,
who had had a vasectomy, and 37,800 members who had not had a vasectomy at
the time of study entry in 1986. These participants had provided detailed
information on various life-style variables including diet. MAIN OUTCOME
MEASURE--Diagnosis of prostate cancer. RESULTS--Between 1986 and 1990, 300
new cases of prostate cancer were diagnosed in participants who were
initially free of diagnosed cancer. Vasectomy was associated with an
elevated risk of prostate cancer (age-adjusted relative risk, 1.66; 95%
confidence interval, 1.25 to 2.21; P = .0004). This elevated risk persisted
after excluding 21 stage A1 cases (age-adjusted relative risk, 1.56; 95%
confidence interval, 1.15 to 2.11; P = .004). Among men who had their
vasectomy at least 22 years in the past (before 1965), the risk of prostate
cancer was even higher (relative risk, 1.85; 95% confidence interval, 1.26
to 2.72; P = .002). This elevated risk among men with vasectomy did not
appear to be caused by detection bias and persisted when we controlled for
diet, level of physical activity, smoking, alcohol consumption, educational
level, body mass index, and geographical area of residence.
CONCLUSIONS--These results support evidence from other epidemiologic
studies that vasectomy increases risk of prostate cancer. The consistency
of results among various epidemiologic studies, the increase of risk over
time following vasectomy, the apparent lack of confounding or bias, and the
existence of physiological changes in the prostate following vasectomy
suggest that the association may be causal.
