Is surgical intervention frequently required for medically managed macroprolactinomas? A study of spontaneous cerebrospinal fluid rhinorrhea

Prolactinoma extension as a contributing factor in dopamine agonist-induced CSF rhinorrhea: a systematic review of the literature

BACKGROUND

Dopamine agonist-induced cerebrospinal fluid (CSF) rhinorrhea is an uncommon treatment-related complication arising in 6.1% of prolactinoma patients treated with dopamine agonists. Locally invasive prolactinomas may create CSF fistulae through formation of dural and osseous skull base defects. Tumor shrinkage secondary to dopamine agonist therapy unmasks skull base defects, thus inducing CSF rhinorrhea. In these cases, repair of the leak may be achieved through collaborative surgical intervention by rhinologists and neurosurgeons. Multiple variables have been investigated as potential contributors to the risk of CSF rhinorrhea development in medically treated prolactinoma patients, with little consensus.

OBJECTIVE

The primary aim of our study was the characterization of risk factors for CSF rhinorrhea development following dopamine agonist treatment.

METHODS

A systematic review of the literature was conducted to identify cases of CSF rhinorrhea following dopamine agonist treatment of prolactinoma. The clinical history, radiographic findings and treatment outcomes are discussed.

RESULTS

Fifty-four patients with dopamine agonist-induced CSF rhinorrhea were identified across 23 articles published from 1979 to 2019. Description of diagnostic imaging [computed tomography (CT)/magnetic resonance imaging (MRI)] was not provided for 18/54 subjects. For the 36 cases that described prolactinoma appearance on CT or MRI, invasion of the cavernous sinuses was reported in 13 (36.1%) and invasion of the sphenoid sinus was reported in 18 (50%).

CONCLUSION

Based on our systematic review, we propose that CT findings of osseous erosion of the sella or the anterior skull base may predict dopamine agonist-induced CSF rhinorrhea. We recommend obtaining a thin-slice CT of the sinuses in cases with MRI evidence of sphenoid involvement.

Spontaneous Cerebrospinal Fluid Rhinorrhea from a Prolactin-Secreting Pituitary Macroadenoma

Cerebrospinal fluid (CSF) rhinorrhea is a rare complication of macroprolactinomas that, in the vast majority of cases, is subsequent to either medical or surgical intervention. Here, we present the successful management of a rare case of spontaneous, noniatrogenic CSF rhinorrhea in a patient with an untreated macroprolactinoma. A 27-year-old man with no significant medical history presented with six months of persistent CSF rhinorrhea, which was confirmed by testing for beta-2-transferrin. He had had decreased libido since adolescence and impaired growth of secondary sexual characteristics. Workup revealed an elevated prolactin level, and imaging demonstrated erosion of the anterior sellar floor and soft tissue within the sphenoid sinus, concerning for tumor. The patient underwent surgical repair of the CSF leak via a transnasal transsphenoidal approach, with resection and biopsy of tumor material within the sinus. No tumor was noted within the sella itself. The patient tolerated the procedure well and had subsequent normalization of his prolactin level with no further CSF egress. Spontaneous noniatrogenic CSF rhinorrhea, although rare, should be considered in the differential diagnosis of invasive pituitary macroadenomas, especially prolactinomas. The mechanism of CSF leak from a prolactinoma is not completely understood, but the CSF leak should be urgently repaired through a transnasal transsphenoidal approach. Concurrently, tumor resection should be performed and a postoperative lumbar puncture or lumbar drain should be considered to reinforce the skull base reconstruction.

The risks of medical treatment of prolactinoma

First-line treatment of prolactinoma is usually medical, based on dopamine agonists receptors, mainly cabergoline. The classical side-effects of cabergoline (low blood pressure and nausea) have been well known since it was first introduced. Other side-effects, however, are more controversial or simply less frequent, but need to be considered during monitoring. This review will focus on these side-effects: cardiac valvular fibrosis, pleural, pericardial and retroperitoneal fibrosis, addictive/compulsive behaviors, and risks secondary to significantly decreased tumor volume. We will also describe how such side-effects should be monitored and managed. In our opinion, the low prevalence of these side-effects should not cast doubt on the role of cabergoline in the therapeutic algorithm of prolactinoma.

SPONTANEOUS CEREBROSPINAL FLUID LEAK AFTER INITIATION OF DOPAMINE AGONIST THERAPY IN MACROPROLACTINOMAS: TWO CASE REPORTS AND A LITERATURE REVIEW

Objective

To describe 2 spontaneous cerebrospinal fluid (CSF) leaks complicating treatment of macroprolactinoma (MPRL) with dopamine agonist (DA) therapy.

Methods

We present the 2 cases of spontaneous, DA-related CSF leaks. Prolactin levels were used to assess hyperprolactinemia. Beta-2 transferrin was tested in rhinorrhea fluid, and magnetic resonance imaging was used to assess the sella.

Results

Case 1 was a 45-year-old woman with a history of MPRL, recently started on bromocriptine at 15 mg/day, presented with clear rhinorrhea, headache, and nuchal rigidity. Magnetic resonance imaging showed a large sellar lesion extending into the cavernous sinuses, posterior sphenoid sinuses, and suprasellar cistern. Computed tomography revealed areas concerning for bony erosion, likely representing leak sites, and the rhinorrhea fluid was positive for beta-2 transferrin, confirming the CSF source. Empiric antibiotics for meningitis were given and she underwent urgent neuroendoscopic, transsphenoidal CSF leak repair and debulking of the pituitary mass. Case 2 was a 55-year-old man with a 10-year history of untreated MPRL who was started on bromocriptine at 5 mg/day 2 weeks prior to admission. He presented with clear rhinorrhea and a metallic taste in his mouth, worse with the Valsalva maneuver. Imaging confirmed clinical suspicion and he was taken for surgery. A high-flow CSF leak was encountered once the tumor was debulked. This was repaired with an abdominal fat pad graft. Both patients developed diabetes insipidus and required postoperative adjuvant DA therapy.

Conclusion

Spontaneous CSF leaks can complicate medical therapy of large MPRL with underlying skull defects, typically within weeks of initiation of DA. This should prompt clinicians to educate patients about the symptoms of potential CSF rhinorrhea and encourage them to promptly report them.

CSF Rhinorrhea Following Medical Treatment for Prolactinoma: Management and Challenges

Objective  Cerebrospinal fluid (CSF) rhinorrhea following medical management of prolactinoma is a rare complication. We evaluated the clinical background of this condition, identify potential risk factors, and discuss the management options and challenges involved. Methodology  We retrospectively reviewed clinical details of patients who were treated for CSF leaks among patients treated for prolactinoma between 2013 and 2017. Results  Seven patients were treated for CSF rhinorrhea in the context of prolactinoma, with the age range between 24 and 56 years. Six patients had CSF leak following initiation of cabergoline, while one patient presented with CSF rhinorrhea. The time of onset of leak following medical treatment ranged from 14 days to 5 years. The mean preoperative serum prolactin level was 12,638 ng/mL (range: 1,000-26,287 ng/mL). All patients underwent repair of skull base defect. (four endoscopic, two microscopic, and one bifrontal craniotomy). The site of defect in the majority of patients was the sellar floor. Two patients who were initially managed with acetazolamide alone, eventually required surgical repair. Three patients were cured of CSF leak with a single procedure. Three patients had to undergo re-exploration and endoscopic repair after their first surgery failed. Two patients required lumboperitoneal (LP) shunt after a failed endoscopic transsphenoidal repair. Conclusion  Surgical management for medically-induced CSF rhinorrhea is necessary; however, it can pose significant issues. Endoscopic repair of the defect should be considered at the earliest. Multiple surgical procedures are often required because of skull base erosion. LP shunt can be considered if CSF leak persists following multiple surgeries.

Surgery for Prolactinomas to Date

While in the past surgery for prolactinomas was a competitive option to medical treatment with dopamine agonists, to date the indications for operations are restricted to only a few specific situations. However, if a patient is intolerant to medical treatment or if the drugs are not sufficiently effective in terms of lowering prolactin levels or tumor shrinkage, surgery should always be considered. In microprolactinomas, operations can achieve excellent prolactin normalization rates. In tumors difficult to treat, the combination of surgical tumor resection and long-term medical treatment supports better control of the disease. The individual indications, results, techniques, and complications of operations for prolactinomas are briefly reviewed in this paper.

Medically induced CSF rhinorrhea following treatment of macroprolactinoma: case series and literature review

PURPOSE

Although several reports have addressed cerebrospinal fluid (CSF) rhinorrhea following dopamine agonist (DA) therapy of macroprolactinomas, further study is warranted for this relatively uncommon entity. Toward this aim, our retrospective series and review of literature further clarifies recommendations in treatment of this rare problem.

METHODS

We retrospectively reviewed all macroprolactinoma cases in our hospital for a 15-year period. Our systematic search of PubMed identified original articles and reviews of all macroprolactinoma cases with an associated medication-induced CSF leak.

RESULTS

Five patients with drug-induced CSF leak were identified; four of these patients received cabergoline therapy an average of 6 weeks before the onset of rhinorrhea and then underwent surgical repair of the CSF leak. Of 35 published studies included, we identified 60 patients with medication-induced CSF leak. Medical therapy included bromocriptine in 34 patients, cabergoline in 21 patients, and use of both DAs in two patients. Three cases did include complete diagnostic and treatment data. Median time from initiation of the DA treatment to occurrence of rhinorrhea was 6 weeks. For CSF rhinorrhea, 49 patients underwent surgical repair (38 by the transnasal approach) and seven patients were treated nonoperatively.

CONCLUSION

Baseline skull base erosion in macroprolactinomas in combination with subsequent tumor shrinkage induced by DA therapy may result in spontaneous CSF rhinorrhea. Therefore, such patients should be advised about and monitored for this potential setback. Once CSF leak is diagnosed, prompt treatment must be carried out to avoid infectious complications. Transnasal surgery appears the most effective therapeutic approach.

Cerebrospinal Fluid Leak Rhinorrhea after Systemic Erlotinib Chemotherapy for Metastatic Lung Cancer: A Familiar Problem from an Unfamiliar Culprit

BACKGROUND

Cerebrospinal fluid (CSF) rhinorrhea after medical therapy for pituitary prolactinoma is a rare but well-described phenomenon. To our knowledge, no CSF leaks have been reported after targeted medical treatment of pituitary or anterior skull base metastases. We report this unusual case to raise awareness of spontaneous CSF leaks in the setting of skull base metastatic disease.

CASE PRESENTATION

A 66-year-old woman presented with epidermal growth factor receptor-mutant stage IV adenocarcinoma of the lung. Headache workup revealed a large sellar and clival lesion consistent with metastatic disease. Systemic erlotinib chemotherapy was initiated with a robust positive response. Approximately 1 week after chemotherapy initiation, the patient noted clear discharge from the right nostril. Her oncologist first diagnosed her with allergic rhinitis, but she presented with meningitis 4 days after diagnosis of CSF leak and was admitted for intravenous antibiotics and definitive repair of a CSF leak via an endoscopic endonasal approach. An erosion of bone and dura was found at the dorsum sellae where tumor had regressed due to the chemotherapy. A multilayer skull base repair was made uneventfully, and she recovered fully with no leakage seen at 2-month follow-up.

CONCLUSIONS

All members of the treatment team should be aware of this possibility of CSF leak after initiation of systemic chemotherapy and tumor regression and urgently refer patients for repair if a leak should develop before the development of meningitis.

Spontaneous and medically induced cerebrospinal fluid leakage in the setting of pituitary adenomas: review of the literature

OBJECT

Spontaneous and medically induced CSF leaks are uncommonly associated with pituitary adenomas. The aim of this study was to further characterize the clinical scenarios most closely associated with this phenomenon.

METHODS

A review of the literature was conducted to identify all cases of nonsurgical CSF leaks associated with pituitary adenomas. The authors aimed to identify associated risk factors and management strategies used to treat these tumors and repair the CSF leak site.

RESULTS

Fifty-two patients with spontaneous or medically induced CSF leaks in the setting of a pituitary adenoma were identified from 29 articles published from 1980 through 2011. In 38 patients (73%), CSF rhinorrhea developed following initiation of medical therapy, whereas spontaneous CSF leakage developed as the presenting symptom in 14 patients (27%). The patients' mean age was 42.8 years (range 23-68 years). There were 35 males and 17 females. Forty-two patients (81%) had a prolactinoma, with the remaining patients having the following tumors: nonfunctioning pituitary adenoma (6 patients), growth hormone-secreting adenoma (2 patients), mammosomatotroph cell adenoma (1 patient), and ACTH-secreting adenoma (1 patient). Infrasellar tumor invasion into the paranasal sinuses was specifically reported in 56% of patients. The medical agents associated with CSF leakage were dopamine agonists (97%) and somatostatin analogs (3%). The average time from initialization of medical treatment to onset of rhinorrhea was 3.3 months (range 3 days-17 months). Nonsurgical management was successful in 4 patients, including 1 who had successful resolution with a temporary lumbar drain. Forty-six patients (88%) underwent surgical intervention to treat the CSF leak and/or resect the tumor. In 2 patients, surgery was not performed due to medical contraindications or patient preference.

CONCLUSIONS

Nonsurgical development of CSF rhinorrhea may occur in the setting of pituitary adenomas, especially following favorable response of invasive prolactinomas to initiation of dopamine agonist therapy. Additional cases have been reported as the presenting symptom of a pituitary adenoma and are likely to be related to decreased tumor volume due to intrinsic infarction or hemorrhage, ongoing invasion, and/or increases in intracranial pressure. Surgical repair, preferentially via a transsphenoidal approach, is the recommended initial treatment for definitive repair of the CSF leak and achievement of maximal tumor resection.