Dr. Siddiqui raises two important issues. First, if a patient as an apoplectic event with a pituitary adenoma, he or she might require emergent trans-sphenoidal surgery; such patients, even if they have prolactin-secreting tumors, may not be responsive to bromocriptine. However, none of our patients who had chiasmal compression during pregnancy had manifestations of hemorrhage on either magnetic resonance imaging or computed tomography scans. As is well known, microscopic examination of removed adenomas may show hemorrhage when neuroimaging has not. The second point concerns the question of why patients did not have surgical intervention during pregnancy. As shown by the reported outcomes in our patients with visual loss, emergent surgery is not always necessary. In the one patient with progressive visual loss, surgery was clearly indicated. In other patients with minor field loss, surgery can be withheld until the pregnancy is completed. In patients with prolactinomas, reinstitution of bromocriptine is effective. Because this was not a treatment trial, our data do not dictate that all patients be treated in this manner. This approach, however, has worked for our patients.

Dr. Molitch had an important observation about the low frequency of the growth of nonsecretory or growth hormone-secreting tumors. He wondered whether these tumors could enlarge to a size that would cause visual loss during pregnancy. However, 6 of 8 patients developed visual loss during pregnancy; their tumors were 1.2 cm or more in height. We studied such a small group of patients with macroadenomas during pregnancy because we typically advise some sort of therapy (surgery, radiotherapy, or dopamine-agonist treatment) before pregnancy in patients with pituitary adenomas. We believe it would be too risky to study a large group of patients with macroadenomas to determine the exact incidence of visual loss.

It is irrelevant whether the patients were originally evaluated by a neuro-ophthalmologist with an interest in pituitary tumors or by an endocrinologist. We agree that some adenomas will inferiorly grow to a substantial size without ever causing a visual disturbance. However, because the principal neurologic risk with a pituitary tumor is loss of vision, any patient with a macroadenoma should have visual field examinations at baseline and routine follow-up. It is not sufficient to follow these patients with manual confrontation techniques.