Reported cure rates following trans-sphenoidal surgery for microprolactinoma are variable and recurrence rates in some series are high. We wished to examine the cure rate of trans-sphenoidal surgery for microprolactinoma, and to assess the long-term complications and recurrence rate. A retrospective review of the outcome of trans-sphenoidal surgery for microprolactinoma, performed by a single neurosurgeon at a tertiary referral centre between 1976 and 1997. All thirty-two patients operated on for microprolactinoma were female, with a mean age of 31 years (range 16-49). Indications for surgery were intolerance of dopamine agonists in ten (31%), resistance in six (19%) and resistance and intolerance in four (12.5%). Two patients were from countries where dopamine agonists were unavailable. The mean pre-operative prolactin level was 2933 mU/l (range 1125-6000). All but 1 had amenorrhoea or oligomenorrhoea, with galactorrhoea in 15 (46.9%). Twenty-five (78%) were cured by trans-sphenoidal surgery, as judged by a post-operative serum prolactin in the normal range. During a mean follow-up of 70 months (range 2 months to 16 years) there was one recurrence at 12 years. Post-operatively, one patient became LH deficient, two patients became cortisol deficient and two became TSH deficient. Out of 21 patients tested for post-operative growth hormone deficiency, 6 (28.6%) were deficient. Five patients developed post-operative diabetes insipidus which persisted for greater than 6 months. There were no other complications of surgery. The estimated cost of uncomplicated trans-sphenoidal surgery, and follow-up over 10 years, was similar to that of dopamine agonist therapy. In patients with hyperprolactinaemia due to a pituitary microprolactinoma, transsphenoidal surgery by an experienced pituitary surgeon should be considered as a potentially curative procedure. The cost of treatment over a 10 year period is similar in uncomplicated cases to long-term dopamine agonist therapy.