Localised prostate cancer, confined to the prostate gland, occurs mainly in men over 65 years of age. The principal management options are watchful waiting, prostatectomy and radiation therapy. Which of these options has the best harm-benefit balance for patients with localised prostate cancer? To answer this question, we conducted a review of the literature using the standard Prescrire methodology. The natural history of localised prostate cancer depends on the extent and histologic grade of the tumour, and pretreatment PSA level. Without immediate treatment, the risk of death from prostate cancer that only one involves one lobe, a Gleason histological score of 7 or less, and a PSA level of 20 ng/ml or lower is less than 0.5% per year. The risk is about 4% per year in patients with larger tumours, poorly differentiated cancer cells (Gleason score above 7), or an elevated PSA level. Most data on radical prostatectomy come from a randomised trial versus watchful waiting in 695 men with localised cancer. Prostatectomy reduced all-cause mortality after a median followup of about 13 years (46% versus 53% without treatment), but this benefit was only seen in patients younger than 65 years at diagnosis. After 4 years of follow-up, prostatectomy was associated with erectile dysfunction in approximately 40% of patients and with incontinence in about 25% of patients. External beam radiation therapy reduced overall mortality to a lesser degree than prostatectomy, but the level of evidence is lower for this modality. Brachytherapy (implantation of a radioactive isotope in the prostate) has not been compared directly with other treatments. Transient radiation proctitis is common after external beam radiation therapy. About 15% of patients treated with external beam radiation therapy and 10% of patients treated with brachytherapy experience long-term intestinal disorders. About half of patients treated with external beam radiation therapy and the majority of patients treated with brachytherapy have transient symptoms of radiation cystitis. In the long term, about 5% of patients treated with radiation therapy have urinary incontinence, versus 12% to 25% of surgical patients. In the long term, about 75% of surgical patients experience erectile dysfunction, compared to about 60% of patients treated with external beam radiation therapy and about 50% of patients who opt for watchful waiting. Brachytherapy appears to cause less erectile dysfunction than external beam radiation therapy. In patients treated with external beam radiation therapy, the addition of hormone therapy for 4 to 6 months reduced all-cause mortality in two randomised trials but caused gynaecomastia, more erectile dysfunction, hot flashes, and hepatitis. Hormone therapy has an unfavourable harm-benefit balance when used alone to treat localised prostate cancer. Further studies of cryotherapy and high-intensity focused ultrasound therapy are needed to determine their respective benefits and harms. In practice, watchful waiting is the most reasonable option for men with low-risk localised prostate cancer and a life expectancy of less than 10 years. In men with low- or intermediate-risk localised prostate cancer and a life expectancy of more than 10 years, there is insufficient data available in early 2012 to show which of the following options is preferable: watchful waiting, radical prostatectomy, external beam radiation therapy, or brachytherapy. Patients should be informed of the risks associated with each of these options and should be actively involved in the choice of treatment. Treatment is often warranted for patients with high-risk localised prostate cancer.The main options are either radical prostatectomy or external beam radiation therapy combined with hormone therapy.