Postoperative surveillance of clinically nonfunctioning pituitary macroadenomas: markers of tumour quiescence and regrowth.

BACKGROUND: Postoperative management of clinically nonfunctioning pituitary adenomas (NFPA) presents difficult challenges. There are no good serum markers for presence or growth of the tumour, medical treatment is not effective and radiotherapy carries the risk of significant side-effects.

OBJECTIVE: The purpose of this study was to investigate the natural history and biological behaviour of surgically treated NFPA, with a special effort to identify characteristics indicative of a more aggressive course that could assist in the clinical decision-making process.

STUDY DESIGN: Patients operated on at our institution for NFPA undergo uniform routine clinical follow-up at the endocrine clinic. Magnetic resonance imaging (MRI) studies are performed 3, 6 and 12 months after transsphenoidal surgery and yearly thereafter for the first 5 years. Subsequently, imaging is performed once every 2 years or as clinically indicated. From 1992 onwards, no patient received immediate postoperative radiation therapy.

PATIENTS: One hundred and twenty-two patients (78M/45F) operated on at our institution since 1989 and with a minimal follow-up of 1 year comprised the study group.

MEASUREMENTS: Tumour size and characteristics were determined by MRI using a modification of Hardy's and Wilson's classifications. Maximal tumour height was also recorded and the information was routinely stored in a computerized database.

RESULTS: Mean (+/- SD) follow-up was 51 +/- 31 months. Fourteen patients received postoperative radiation therapy. Subsequent tumour growth was observed in five of them, reduction in tumour size in four and no size changes in five. One hundred and eight patients did not receive postoperative radiation. Tumour enlargement occurred in 41 of 78 and in six of 30 patients with and without residual tumour after operation (P = 0.0024). The presence of cavernous sinus invasion before surgery [P = 0.02, odds ratio (OR) 2.72; confidence interval (CI) 1.1-6.43] and the extent of suprasellar extension in the postoperative tumour remnant (P = 0.0054 for presence of stage A, OR 4.4; 95% CI 1.5-12.5; and P = 0.012 for presence of stages B or C, OR 16.2; CI 1.8-144) were strong independent predictors of tumour enlargement.

CONCLUSION: Our data may ease the selection of patients in whom radiation therapy is likely to be necessary for tumour control, and confirms that close postoperative follow-up is an adequate primary approach in low-risk patients.