Ten year follow up of microprolactinoma treated by transsphenoidal surgery

BMJ 1994; 309 doi: (Published 26 November 1994) Cite this as: BMJ 1994;309:1409
  1. John A Thomson, readera,
  2. David L Davies, senior lecturerb,
  3. Eon H McLaren, consultantc,
  4. Graham M Teasdale, professor of neurosurgeryd
  1. a Department of Medicine, University of Glasgow, Glasgow Royal Infirmary, Glasgow G31 2ER
  2. b Department of Medicine and Therapeutics, Western Infirmary, Glasgow G11 6NT
  3. c Department of Medicine, Stobhill Hospital, Glasgow G21 3UW
  4. d Institute of Neurological Sciences, Southern General Hospital, Glasgow G51 4TF
  1. Correspondence to: Dr Thomson.
  • Accepted 7 September 1994

In 1985 we published the results of transsphenoidal surgery in 61 women with microprolactinoma.1 Early results were satisfactory in over 80% of patients. Only two out of 21 patients with a normal postoperative prolactin concentration had a relapse after five years. This paper documents the 10 year outcome.

Patients, methods, and results

We tried to obtain information about patients who were not currently being followed up in Glasgow by contacting their general practitioner, referring specialist, and the patients themselves, achieving a 10 year follow up in 45 of the 61 patients (74%). We defined cure of hyperprolactinaemia as a plasma prolactin concentration within the reference range (60-360 mU/l), operative failure as a persistently raised prolactin concentration, and recurrence as a sustained increase in prolactin concentration after a normal value had been achieved. In four patients prolactin concentration was variably raised during follow up.

The overall results are shown in the figure. Cure rates at 10 years were similar to those at five years (hyperprolactinaemia 73% (95% confidence interval 58% to 85%), amenorrhoea 76% (60% to 87%), galactorrhoea 87% (73% to 95%), and infertility 82% (68% to 92%)).

Eleven patients were amenorrhoeic at 10 years: three were menopausal, three had primary operative failure, two had had a hysterectomy (for ovarian cancer and menorrhagia), and three had some evidence of postoperative hypopituitarism. All but eight patients became pregnant. Although infertility was a major indication for operation, four decided postoperatively not to attempt pregnancy despite achieving a normal prolactin concentration. One of the remaining four patients had premature ovarian failure, two had hypopituitarism, and one had a hysterectomy for ovarian carcinoma three years after a successful hypophysectomy. The remaining 37 patients achieved 63 pregnancies, of which six miscarried. Only one of these six patients did not subsequently achieve a successful term delivery. Seven patients (all with primary operative failure) were still taking bromocriptine. Four were receiving thyroxine; in two the need was questionable. Two patients were taking steroids, possibly unnecessarily in one case. One patient was receiving hormone replacement therapy.

The true recurrence rate is difficult to assess because of the variability in serum prolactin concentrations in some patients. During the 10 years of follow up there was evidence of recurrence of hyperprolactinaemia in six of the 45 patients (13%) at two, four, seven, eight, and 10 years (two patients). One patient developed a raised serum prolactin concentration three months postoperatively, but defaulted from the study after five years. This gives an overall known recurrence rate of seven out of 61 patients (12%). One relapse at 10 years was marginal (prolactin concentration 550 mU/l). Sustained amenorrhoea occurred in only one out of six patients with recurrent hyperprolactinaemia. In one, hyperprolactinaemia reverted spontaneously after 13 years. In four of the six with the highest prolactin concentrations repeated computed tomography showed no evidence of tumour. Thus at 10 years only four of the 45 patients (9%) had definite recurrence of hyperprolactinaemia— that is, four of the 39 patients (10%) who had a normal postoperative prolactin concentration.


The initial results of transsphenoidal surgery for microprolactinoma are good. In 1983 Serri et al reported an incidence of apparent recurrence of hyperprolactinaemia in 50% of patients (12/24) with microprolactinoma five years postoperatively.2 Our findings1 and those of Scanlon et al3 and Bevan et al4 did not confirm this observation. Other American and European authors, however, have reported higher recurrence rates. Our 10 year follow up showed that transsphenoidal surgery was successful overall in 73% of patients. Relapses did occur, however, and in some patients the prolactin concentration was quite variable.

Serri et al reported a 10 year cure rate for hyperprolactinaemia of 56% and a clinical remission rate of 75% in 57 patients after successful operations.5 We confirm that transsphenoidal surgery for microprolactinoma remains a reasonable treatment with satisfactory long term results in selected patients.

We are grateful to our colleagues in other hospitals and in general practice who provided details of long term outcome of patients. We particularly thank our biochemical colleagues, especially Dr Graham Beastall and Dr Christina Gray of the Institute of Biochemistry, Glasgow Royal Infirmary, who provided most of the hormone measurements. We gratefully acknowledge the statistical skills of Dr Gordon Murray.


View Abstract