Objective To systematically review and evaluate the effectiveness and adverse effects of the a-antagonist, terazosin, for treating urinary symptoms associated with benign prostatic obstruction (BPO). Methods Studies were sought and included in the review if they were randomized trials of at least 1 month duration, involved men with symptomatic BPO and compared terazosin with placebo or active controls. The study, patient characteristics and outcome data were extracted in duplicate onto standardized forms using a prospectively developed protocol. Results Seventeen studies involving 5151 men met the inclusion criteria, i.e. placebo-controlled (10), ablockers (seven), ®nasteride alone or combined with terazosin and placebo (one), and microwave therapy (one). The study duration was 4±52 weeks; the mean age of the men was 65 years and 82% were white. Baseline urological symptom scale scores and¯ow rates showed that men had moderate BPO. Ef®cacy outcomes were rarely reported in a way that allowed for data pooling, but indicated that terazosin improved symptom scores and¯ow rates more than did placebo or ®nasteride, and similarly to other a-antagonists. The pooled mean percentage improvement for the Boyarsky symptom score was 37% for terazosin and 15% for placebo (four studies). The mean percentage improvement for the American Urological Association symptom score was 38%, compared with 17% and 20% for placebo and ®nasteride, respectively (two studies). The pooled mean improvement in the International Prostate Symptom Score of 40% was similar to that with tamsulosin (43%). Peak urinarȳ ow rates improved more with terazosin (22%) than with placebo (11%) and ®nasteride (15%), but did not differ signi®cantly from the other a-antagonists. The percentage of men discontinuing terazosin was comparable with those receiving placebo and ®nasteride, but greater than with other a-antagonists. Adverse effects were greater than with placebo and included dizziness, asthenia, headache and postural hypotension. Conclusions The available evidence indicates that terazosin improves the symptoms and¯ow rates associated with BPO; it was more effective than placebo or ®nasteride and similar to other aantagonists. Adverse effects were generally mild but more frequent than with other a-antagonists and associated with a two-to four-fold increase in treatment discontinuation.