Microsurgical management of prolactinomas - Clinical and hormonal outcome in a series of 172 cases
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.84332
Source of Support: None, Conflict of Interest: None
Aims and Objectives: To report hormonal outcome after surgery in a special subgroup of prolactinomas in whom medical therapy is not effective or not indicated. Patients and Methods: All patients who underwent surgery for prolactinomas, between December 2002 and December 2009, were analyzed retrospectively. The study group consisted of patients who were either intolerant or noncompliant to dopamine agonist (DA) therapy or those in whom medical treatment could not be given due to various reasons. The surgical results were assessed according to whether patients had microadenoma (<1 cm diameter), macroadenoma (>1 cm diameter), or giant prolactinoma (>4 cm diameter). Initial and final hormonal remission was defined as normalization of serum prolactin levels (<25 ng/ml) at 7 days after surgery and at last follow up, respectively. Results: Of the 172 patients, 133 (77%) were operated by primary transsphenoidal approach and 25 (14.5%) were operated transcranially and 14 patients required reoperation, either transsphenoidally (6 patients) or transcranially (8 patients). Overall, at last follow-up, hormonal remission could be achieved in 44% patients (83% microadenomas, 48% macroadenomas, and 16% of giant adenomas). There were three deaths (1.7%), 12 patients (7%) had single/multiple postoperative treatable complications. 59% of those patients, who did not achieve remission after surgery, finally attained normoprolactinemia with either DA therapy or radiotherapy, at last follow-up. Conclusions: Medical treatment of prolactinomas with DA should be considered as treatment of choice for these tumors. However, there is a major subgroup of patients who respond better to surgery rather than protracted medical therapy.
Keywords: Dopamine agonists, prolactinomas, transsphenoidal surgery, transcranial surgery
Prolactinomas account for 30 to 50% of all pituitary adenomas and are common cause of reproductive and sexual dysfunction.  The aim of treatment is to suppress hyperprolactinemia and its clinical consequences, removal of tumor mass, preservation of normal pituitary, and prevention of tumor recurrence/progression. The treatment modalities for these tumors include medical treatment (dopamine agonists [DA]), surgery with or without radiotherapy, and radiosurgery. The normalization of serum prolactin (PRL) in >80% patients treated with DAs dictates medical therapy as preferred initial choice. Surgery is reserved for tumors not responding to medical management or in patients with visual deficits or pituitary apoplexy. ,, Furthermore, surgery should be considered as a therapeutic option for selected patients with poor drug compliance, intolerance or resistance to DAs, cystic prolactinomas, patients consenting against having lifelong medical treatment, and patients seeking fertility with macroadenomas adjacent to chiasma.  The aim of the study was to analyze results of surgery in terms of hormonal outcome, in this special subgroup of prolactinomas.
All patients undergoing surgery for prolactinomas at our center, between December 2002 and December 2009, were analyzed retrospectively. During this period, a total of 610 patients were operated for pituitary adenomas. Of these patients, 172 patients (28%) were operated for prolactinomas. All these patients were surgically treated by either transsphenoidal or transcranial approach and constituted the study group.
The patients receiving medical treatment only (DAs) were treated on an outpatient basis and were excluded from the study. All patients with residual or recurrent adenomas, who had been previously operated, were also excluded. DAs (bromocriptine or cabergoline) started by the referring physicians were aborted in 158 patients (92%) 6 weeks prior to surgery. The surgical indications in the patients are enlisted in [Table 1]. The diagnosis of prolactinoma was based on the clinical features of hyperprolactinemia (oligomenorrhea, galactorrhea, and gonadal dysfunction), radiological examination (contrast-enhanced magnetic resonance imaging [MRI]), and elevated serum PRL levels >250 ng/ml (normal laboratory value - <20 ng/ml in males and <25 ng/ml in females). The diagnosis of prolactinoma was confirmed postoperatively in all tumors by hematoxylin-eosin staining and immunohistochemical analysis. The study group included 101 males (59%); the mean age was 34.6 years (median age - 32 years), with a range from 17 to 65 years. The mean duration of symptoms was 18 months in females and 34 months in males. The most frequent symptoms were headaches, visual deficits, galactorrhea, amenorrhea, infertility, and impotence.
The surgical results were analyzed on the basis of whether patients had microadenoma (<1 cm diameter), macroadenoma (>1 cm diameter), or giant prolactinoma (>4 cm diameter). On the basis of MRI findings, microadenomas were present in 16 patients (9%), macroadenomas in 114 patients (66%), and giant tumors in 42 (25%). Gross total tumor resection was defined when there was no evidence of residual tumor, as assessed by operating surgeon and on postoperative contrast-computed tomography scans with three-dimensional reconstructions, at the time of discharge.
Fasting basal serum levels of PRL, cortisol, growth hormone (GH), testosterone, estradiol, gonadotropins, and thyroid hormones were assessed in all the patients. The endocrinological evaluation was repeated on day 7 postoperatively, and during follow-up visits at 3 months and thereafter annually. The hormonal assessment and immunohistochemistry revealed that 144 patients (84%) had pure prolactinomas while 28 patients (16%) had PRL-GH positivity. Initial hormonal remission was defined as fasting basal serum PRL levels <25 ng/ml on day 7 after surgery, in patients without dopaminergic therapy for at least 6 weeks preoperatively. Follow-up hormonal remission was defined at last follow-up as normalization of basal serum PRL levels (<25 ng/ml) in the absence of any dopaminergic therapy for 3 months.
Of the 172 patients, 133 (77%) were operated by a primary transsphenoidal approach and 25 (14.5%) were operated transcranially. However, 14 patients required reoperation, either transsphenoidally (six patients) or transcranially (eight patients). These patients had giant invasive adenomas and indications for second operation were postoperative hematoma/residual tumor swelling (five patients) or residual tumor and lack of clinical remission (nine patients). Of the 94 patients who had visual deterioration preoperatively, vision improved in 49 patients (52%) at the time of discharge.
The rates of gross total tumor excision were 98% for microprolactinomas, while it was 76% for macroprolactinomas and only 42% for the giant adenomas. Overall, initial remission was achieved in 83 of 172 patients (48%). The initial remission rates were 92% for the microprolactinomas (14/16 patients), 52% for macroadenomas (59/114 patients), and 23% for giant adenomas (10/42 patients) [Table 2].
Fifty patients were lost to follow-up. Thus, follow-up data were available for 12 patients with microadenomas, 83 patients with macroadenomas, and 27 patients with giant adenomas. In these patients, overall initial remission could be achieved in 50% (61/122 patients). The initial remission rate in patients for microadenomas, macroadenomas, and giant adenomas were 90%, 54%, and 21%, respectively [Table 2]. At the last follow-up, overall final hormonal remission was 44% (54/122 patients). The rates of final follow-up hormonal remission were maintained in 83% patients with microadenomas, 48% for macroadenomas, and only 16% for giant adenomas [Table 2].
A total of 12 patients (7%) had single/multiple postoperative complications [Table 3]. Postoperative diabetes insipidus was observed in six patients (3.5%), which was more common in patients with giant adenomas operated transcranially. Postoperative Cerebrospinal fluid (CSF) rhinorrhea occurred in five patients (3%). This responded to conservative treatment with lumbar drainage for a few days in three patients, while required re-exploration and dural repair in the rest. The overall mortality was 1.7% (three patients). All these patients had giant invasive tumors and were admitted in altered sensorium. Of three patients who developed postoperative hematoma with swelling of the residual tumor, two died because of multiple infarcts, despite repeat surgery and decompression. One patient died of fulminant septicemia and chest infection.
The follow-up period ranging from 1 month to 62 months (mean follow-up of 38.6 months) was available for 122 patients. In follow-up, a total of 68 patients (56%) could not achieve hormonal remission and were treated with DAs (35 patients) or radiotherapy (33 patients). Stereotactic radiotherapy was given in 18 patients and Gamma knife therapy in another 15 patients. On further follow-up of these 68 patients, hormonal remission was achieved in 20 patients (57%) on DA therapy, 11 patients (61%) who underwent radiotherapy, and nine patients (60%) who underwent gamma knife therapy.
Medical treatment with DAs (bromocriptine or cabergoline) is the treatment of choice for PRL-secreting adenomas. ,,,,, They rapidly normalize PRL levels, restore the reproductive function, reverse galactorrhea, and decrease tumor size in most of the patients.  Bromocriptine (2.5-5 mg/day) achieves suppression of hyperprolactinemia and tumor shrinkage in >80% patients with microprolactinomas, ,, and reduces serum PRL levels in up to 75% of patients with macroprolactinomas, with reduction of tumor size by >50%. ,,,, However, 10 to 25% of patients are partially or totally resistant to bromocriptine ,,, and about 5 to 10% of patients do not tolerate this drug well.
However, transsphenoidal microsurgery is accepted only as a second-line treatment of choice in a subgroup of prolactinomas. ,,, The surgical cure rates may vary from 80 to 90% for microadenomas to <50% for macroadenomas,  and when performed by experienced neurosurgeons, the associated mortality and morbidity rates are extremely low (0.2% and 1.4%, respectively).  Nonetheless, as is seen in macroadenomas and giant adenomas, even if surgery is not curative, tumor cytoreduction frequently increases the responsiveness of DAs and thereby lowers the required dosage. ,,,
According to the guidelines published by the pituitary society for the management of prolactinomas, surgery is indicated in patients not responding to medical therapy, patients with intolerable side effects of DAs, CSF fistula with DAs, and in patients with rapidly progressive neurological deficits.  Our data also demonstrate that this subgroup of patients respond better to surgery rather than protracted medical therapy. Overall, initial hormonal remission could be achieved in 48% patients. At last follow-up, 44% patients were in hormonal remission (83% in microadenomas, 48% in macroadenomas, and 16% in giant adenomas). The results of our series are very well comparable with the best published rates for hormone remission with medical therapy. ,,,,
Surgery is rarely curative in macroadenomas and giant prolactinomas and may not be able to normalize the hyperprolactinemia because of incomplete tumor excision or disruption of pituitary hypothalamic axis. Multimodal therapy with surgical debulking and subsequent adjuvant therapy (stereotactic radiosurgery or medical therapy) may be an effective strategy, especially in invasive prolactinomas. However, recently many authors have reported complete tumor removal in giant prolactinomas and advances in microneurosurgical methods have played an important role in achieving gratifying results.  In present series, overall hormonal remission was achieved in 48% patients with macroadenoma at the last follow-up. The normalization of hyperprolactinemia was the least in patients with giant prolactinomas (16%). Although some surgeons have reported that prior long-term treatment with DAs alters tumor consistency and hinders resection,  we have not found this to be the case. In contrast, in patients with giant and invasive prolactinomas, pretreatment with a DA may improve safety and success of subsequent surgery. 
The incidence of surgical complications ranges from 6.5 to 29% and the reported mortality varies from 0.27 to 6.5% in literature. ,,,, The mortality and morbidity in the present series were 1.7% and 7%, respectively. This seems plausible as majority of patients in our study had macroadenomas (66%) and giant invasive adenomas (25%), most of which required transcranial approaches for tumor excision. The complications were less commonly seen with the less-invasive transsphenoidal approach as compared with the transcranial approach. Postoperative residual tumor swelling and infarction was present in three patients, of which two died despite re-exploration, because of multiple infarcts. Goel et al. also reported postoperative pituitary apoplexy in five of 30 patients, leading to intraoperative and postoperative swelling and acute elevation of intracranial pressure. Symon et al. also came across this complication in three of the eight patients and all of these patients had a fatal outcome.
Radiation therapy has been reported to be effective in producing clinical or hormonal response in treatment of these tumors, with control rates of 47 to 90% being documented in different series. ,,,, Clarke et al. reported an overall response rate of 71% in a series of 44 functional adenomas and found that prolactinomas were most responsive group among functional adenomas. Bronson et al. reported that postoperative radiotherapy decreased the recurrence rate from 22 to 8% in all types of pituitary adenomas. In our opinion, postoperative radiotherapy is indicated for all patients with persistent clinical signs and neuroradiological evidence of residual tumor in follow-up. Postoperative radiotherapy was given in 18 patients in our series in whom hormonal remission could not be achieved. 61% of these patients were found to be in remission at the last follow-up.
Landolt et al. subjected 20 patients to Gamma knife treatment after failed microsurgery or medical therapy. In five patients, PRL levels reached normal values, while 11 patients experienced partial improvement. The treatment failed in four patients who were on DA therapy at the time of radiosurgery, thereby suggesting some radioprotective effect of DAs.  Pan et al. reported 164 patients treated with primary radiosurgery. Hormonal remission was reported in 41% patients at a mean follow-up of 33.2 months. In our study, postoperative gamma knife therapy was able to normalize serum PRL in 60% of patients who could not achieve hormonal remission with surgery at last follow-up.
In conclusion, our series shows that surgical treatment is an effective form of therapy in a special subgroup of prolactinomas, for the indications mentioned above. The excellent hormonal remission rates in our series emphasize the importance of microsurgical excision as a worthy alternative to medical therapy in these patients.
[Table 1], [Table 2], [Table 3]