Long-Term Outcomes of Pituitary Gland Preservation in Pituitary Macroadenoma Apoplexy: Case Series and Review of the Literature

Abstract

Introduction  For patients presenting with neurological changes from pituitary tumor apoplexy, urgent surgical intervention is commonly performed for diagnosis, tumor resection, and optic apparatus decompression. Although identification and preservation of the pituitary gland during the time of surgery can be challenging, it may lead to improve endocrine outcomes.

Methods  A retrospective case series of all patients with macroadenomas presenting with apoplexy at Loyola University Medical Center from 2016 to 2018 was studied. Demographic, radiographic, and intraoperative characteristics were collected including age, gender, comorbidities, presenting symptoms, preoperative size of pituitary adenoma, Knosp's grade, Hardy's grade, identification and/or preservation of the gland, pre- and postoperative hormonal levels, intraoperative and/or postoperative complications, and follow-up time.

Results  A total of 68 patients underwent endoscopic endonasal surgery for resection of a macroadenoma. Among them, seven (10.2%) presented with apoplexy; five patients were male and two were female and presenting symptoms and signs included headache (100%), endocrinopathies (57%), visual acuity deficit (71%), visual field deficit (71%), and oculomotor palsy (57%). A gross-total resection rate was achieved in 86% of patients. Among them, 71% of patients obtained complete symptomatic neurological improvement. A statistically significant difference between gender and endocrine function was found, as no females and all males required some form of postoperative hormonal supplementation ( p  = 0.047) .

Conclusion  Endoscopic endonasal resection of macroadenomas with sparing of the pituitary gland in the setting of apoplexy is safe and effective. Preservation of the normal gland led to no posterior pituitary dysfunction, and a statistically significant difference between gender and postoperative endocrinopathy was identified. Further studies with larger samples sizes are warranted.

Introduction

Pituitary adenomas occur in 8.2 to 14.7 per 100,000 individuals and are the third most common primary central nervous system tumors after meningiomas and gliomas. ^{1 2 3} Some autopsy studies have shown that up to 25% of the general population may have silent pituitary adenomas. ²

Pituitary apoplexy can be a life-threatening condition that warrants urgent surgical management, given both its endocrinological dysfunction and the mass effect on the surrounding neurological structures. ² The etiology of pituitary apoplexy is attributed to increased infarction risk of the adenomas given its decreased perfusion in the setting of increased intratumoral pressure and decreased vessel density together with its reduced angiogenesis. ⁴ Up to 50% of pituitary apoplexy occurs with nonfunctioning adenomas. ²

Although identification of the normal pituitary gland can be difficult in these situations, we have placed significant emphasis on its intraoperative identification through visual and/or histological confirmation, following preoperative imaging study. Surgical identification and preservation of the pituitary gland may lead to improved endocrine outcomes, especially in the setting of apoplexy. We examine our case series of gland identification and preservation in the setting of macroadenomas presenting with apoplexy to evaluate outcomes in this rare patient population and then compare it to other published series.

Materials and Methods

A prospective case series of all patients with pituitary macroadenomas presenting between 2016 and 2018 with apoplexy at Loyola University Medical Center treated by a single surgical treating team (neurosurgeon and otolaryngologist) was studied. Institutional review board approval was obtained prior to the start of this study (211097). Pituitary macroadenomas were classified as small (10–25 mm diameter), large (26–39 mm diameter), or giant macroadenoma (≥40 mm diameter). Inclusion criteria included patients with histologically confirmed pituitary adenoma, sudden onset of headache, and/or cranial nerve deficit, hemorrhage identified on imaging, and apoplexy identified by a neuropathologist on the resected specimen.

Patient records were reviewed, and demographic and radiographic characteristics were collected including age, gender, comorbidities, presenting symptoms, precipitating factors, preoperative size of pituitary adenoma, size of hemorrhagic component, Knosp's grade, Hardy's grade, pre- and postoperative hormonal levels, intraoperative and postoperative complications, and follow-up time ( Table 1 ). Radiographic features on preoperative imaging were identified by a neuroradiologist.

Table 1. Demographic characteristics (median and IQR).

	Total ( n  = 7)
Age	57 (49–70)
Gender
Male	5 (71.4%)
Female	2 (28.6%)
Adenoma size	31 (16–34.2)
Small macroadenoma	2 (28.6%)
Large macroadenoma	4 (57.1%)
Giant macroadenoma	1 (14.3%)
Knosp's grade	3 (3–3)
Hardy's grade	3.5 (3–4)
Time to surgery
Symptoms-to-surgery time (d)	4 (2–21)
Admission-to-surgery time (h)	16 (7.25–48)
Follow-up time (mo)	21 (16–29)

Abbreviation: IQR, interquartile range.

Note: values reported as median and IQR. Gender reported as patient counts. Percentages refer to proportion of patients in the current series.

A systematic literature search using PubMed through December 1, 2018 was performed for all articles containing the phrases “pituitary apoplexy” or “pituitary adenoma” or “pituitary macroadenoma.” Abstracts of articles written in the English language were reviewed.

Categorical data were summarized using frequencies and percentages. Continuous data were summarized using median and interquartile range (IQR). Unpaired Wilcoxon's rank-sum test was used to test differences for continuous variables and Fisher's exact test was used for categorical variables. Logistic regression analysis was used to assess if there was association between neurological preoperative symptom resolutions with every one of the preoperative variables aforementioned, as well as to assess association between resolution of the preoperative endocrinopathies with every one of the preoperative variables aforementioned. All reported p -values were two-sided and statistical significance was set at p  < 0.05. All analyses were performed using SAS version 9.3 (SAS Institute, Inc., Cary, North Carolina, United States).

Results

A total of 68 patients undergoing endoscopic endonasal surgery for resection of macroadenoma, 7 (10.2%) of them presented with apoplexy and met inclusion criteria. The median age of these seven patients was 57 years, and the male:female ratio was 5:2. Tumor size was small in two patients, large in four patients, and giant in one patient. Median of tumors' maximum diameter was 31 mm. The median Knosp's grade was 3, while the median Hardy's grade was 3.5. ^{5 6} In one patient, the hemorrhage was noted to be diffuse; in the other six, it was focal. The median size of focal hemorrhage was 10.2 mm. The median symptom onset-to-surgery time was 96 hours, with a median admission-to-surgery time of 16 hours ( Table 1 ).

All patients presented with thunderclap headache, 71% with poor visual acuity, 71% with visual field deficits, and 57% with oculomotor palsy ( Table 2 ). Three of the patients presented with drowsiness or somnolence, and three of the patients endorsed nausea or vomiting. Three patients presented with at least one risk factor for apoplexy those were identified as anticoagulation, hyperlipidemia, and prior radiation therapy. On admission, 14% of patients had abnormally low cortisol levels, and 57% had hypothyroidism. Decreased testosterone was found in 71% of patients (100% of male patients). No females had abnormal preoperative serum hormone levels ( Fig. 1 ).

Table 2. Clinical presentation and neurological outcomes.

Neurological Symptoms	Preoperatively	Complete resolution postoperatively (%)	Improvement postoperatively	Worsened postoperatively
Visual acuity deterioration	5	3 (60)	2 (40%)	–
Oculomotor palsy	4	4 (100)	–	–
Visual field deficit	5	3 (60)	2 (40%)	–
Headache	7	7 (100)	–	–
Somnolence	3	3 (100)	–	–

Note: values listed as counts and refer to number of patients. Percentages listed represent proportion of patients presenting with a symptom who had resolution/improvement of that symptom.

Fig. 1.

Presenting and postoperative endocrine function.

An endoscopic endonasal transsellar approach (with transtubercular/transplanar extension depending on the size of the mass) for resection of the sellar mass was performed on each patient. The normal pituitary gland was able to be identified on preoperative imaging in three patients and intraoperatively at a margin of the tumor in all seven cases. A gross total resection rate was achieved in 86% of patients. Pathology confirmed a diagnosis of adenoma and apoplexy in all patients. Immuohistochemical stains revealed that 86% of adenomas were synaptophysin positive, while 57% demonstrated a stromal pattern of reticulin.

There were no intraoperative or postoperative complications reported. Median follow-up time was 21 months. Postoperatively, 71% of patients experienced full resolution of all presenting symptoms. Full resolution of visual deficits was achieved in 60% of patients ( Table 2 ). Two of the five patients who presented with visual field deficits had partial improvement. Complete resolution of oculomotor impairment was achieved in all patients who presented with ophthalmoplegia.

A total of 71% patients displayed postoperative endocrine abnormalities ( Fig. 1 ). Among them, 43% patients received a single dose of vasopressin only on the first postoperative day for 2 hours of persistently elevated urine output (greater than 300 cc/hour), a urine specific gravity less than 1.005, and a high-normal serum sodium value (142–143 mmol/L); zero patients required permanent vasopressin. Also, 71% of patients presented with decreased testosterone preoperatively, and all of these patients required long-term testosterone supplementation postoperatively. Additionally, 100% (4/4) of patients who presented with hypothyroidism had persistently low thyroid hormones, while one patient developed new hypothyroidism postoperatively. The one patient that experienced preoperative hypocortisolemia continued to require steroid hormone replacement postoperatively. Three patients developed new postoperative decreased cortisol necessitating medical management. There were no statistically significant differences between those patients who showed resolution of neurological symptoms and those who did not show it for age, gender, size of mass, Knosp's grade, Hardy's grade, size of hemorrhage, time between symptom onset, and surgery and time of admission to surgery. There was a statistically significant difference between patients with postoperative endocrinopathies and gender ( p  = 0.047). No females had postoperative endocrinopathies, while all males had at least one postoperative thyroid and/or cortisol endocrinopathy.

Discussion

Pituitary adenomas are diagnosed in 8.2 to 14.7 per 100,000 individuals and account for approximately 10 to 15% of intracranial tumors. ^{1 7} As the third most common intracranial neoplasm, pituitary adenomas have a reported prevalence from 16.9% to as high as 25% at autopsy. ^{3 8 9} Pituitary adenomas can be functional or nonfunctional, and subsequent endocrine abnormalities often dictate symptoms. In one of the largest reported case series in the literature, Jane and Laws found that 65.3% of pituitary adenomas are functional and 34.7% are nonfunctional. ² While adenomas can have varying endocrine functionality, often the most predictive indicator of symptomatology is size of the adenoma.

As pituitary adenomas enlarge, they have the potential to compress the optic chiasm, optic nerves, carotid arteries, and the pituitary gland. ¹⁰ Compression of any of these structures can present with visual disturbances, ophthalmoplegia, endocrine dysfunction, and many other symptoms. ¹⁰ In a mechanism similar to that of pituitary apoplexy, rapid compression of adjacent structures is responsible for the clinical presentation. Larger pituitary adenomas, therefore, pose a greater threat to compress adjacent structures and more often present with more significant symptomology. In the current literature, pituitary macroadenomas are defined as having a maximum diameter ≥10 mm, and giant pituitary adenomas are classified as having a maximum diameter ≥40 mm. ^{11 12} These tumors often grow beyond the sella and compress nearby structures. ¹² In a case series of giant pituitary adenomas by Juraschka et al, the mean diameter was 4.09 cm and over a third of tumors displayed sphenoid sinus invasion. ¹² Furthermore, 94.5% of tumors compressed the optic nerve, causing a visual field deficit in nearly half of all patients in this series. ¹² The most common presentation of giant pituitary adenomas was visual acuity deficit, visual field deficit, and endocrine dysfunction. ¹²

Definitive treatment of large nonsecretory macroadenomas is surgical resection, and many published case series compare efficacy of various surgical approaches. The reported range of gross total resection of giant pituitary adenomas and macroadenomas varies from 24 to 79.7%, and the literature shows comparable rates of resection in treating with endoscopic compared with microscopic transsphenoidal surgery. ^{12 13} Postoperatively, patients often have immediate improvement or, in some cases, resolution of many symptoms. Hofstetter et al reports that 85% of patients with giant pituitary adenomas have improved visual field defects, while 75% of patients with macroadenomas have improved visual field deficits. ¹⁴ Postoperative endocrine abnormalities, however, often persist and can warrant adjuvant medication postoperatively in up to a third of patients with giant pituitary adenomas and macroadenomas. ¹⁴

Pituitary apoplexy can be a life-threatening phenomenon that occurs after infarction or hemorrhage of the pituitary gland. Apoplexy can cause rapid mass effect on surrounding structures. Clinically, patients with pituitary apoplexy may present with a wide range of symptoms. Up to 100% of patients present with a thunderclap headache, most commonly located bifrontally or retroorbitally. ^{15 16 17 18 19} It is suggested that headache is caused by irritation of the meninges or distension of the dura from the extravasated blood. ^{18 19 20} Visual disturbances are also reported in a majority of patients, ranging from 53 to 74%. ^{19 21 22} Visual deficits result from compression of optic nerves or optic chiasm. ²³ In severe apoplexy, patients can present with fever, nuchal rigidity, and photophobia, due to significant meningeal inflammation. ^{19 23}

Treatment for pituitary apoplexy is most commonly transsphenoidal resection of the tumor (when present) and evacuation of the hemorrhage. Surgery has proven to be very effective in restoring visual acuity, visual field deficits, and ophthalmoplegia in 88 to 100% of patients presenting with primary pituitary apoplexy. ¹⁵ To achieve these high rates of symptomatic resolution, Bills et al suggested operating on patients expeditiously, as surgical intervention after 1 week of symptom presentation can significantly impact postoperative visual field improvement. ¹⁵

Pituitary adenomas presenting with apoplexy is an uncommon yet not unseen occurrence, as up to 7% of patients with pituitary adenomas can develop apoplexy. ²¹ The existing literature has indicated that larger tumors are more prone to necrosis and subsequent hemorrhage leading to apoplexy, but there is no study that exclusively examines apoplexy and size of pituitary tumor. ¹⁹ We present a case series of seven patients with pituitary macroadenomas who presented with pituitary apoplexy. In this unique subset of patients, we examined patient demographics, potential precipitating factors for and extent of apoplexy, pre and postoperative symptomology, imaging characteristics, and operative technique.

As demonstrated by the current series, endoscopic transsphenoidal resection of macroadenomas presenting with apoplexy is a safe and effective treatment. In 86% of patients, a gross-total resection was achieved ( Figs. 2 and 3 ). When compared with other case series, our rates of resection are similar. In a series of patients with macroadenomas presenting with apoplexy treated surgically, Ricciuti et al reported a gross-total resection of 92%. ²⁴ Rutkowski et al reported a gross-total resection rate of 67% in a similar cohort of patients. ²⁵ In addition to the high-resection rate in the current series, no patients had any postoperative complications that is similar to that of other case series. ^{2 24} The effectiveness of the endoscopic transsphenoidal resection is best demonstrated in Table 2 . The resolution or significant improvement of all symptoms in all patients confirms the effectiveness of endoscopic transsphenoidal resection in treating headache, visual field deficits, and ophthalmoplegia.

Fig. 2.

(A and B) Preoperative sagittal and axial contrasted T1 MRI demonstrating a 49-mm pituitary adenoma with hemorrhage extending through the floor of the sella into the sinonasal cavity. MRI, magnetic resonance imaging.

Fig. 3.

(A and B) Postoperative sagittal and axial contrasted T1 MRI showing decompression of the optic apparatus and a gross total resection. MRI, magnetic resonance imaging.

Perhaps the most interesting finding in the current series is the relationship between postoperative recovery of endocrine function and gender. Both female patients in the current series retained normal endocrine function postoperatively and did not require supplemental hormones at follow-up ( p  = 0.047). Contrarily, 100% of male patients who presented with endocrine abnormalities did not attain postoperative resolution ( Fig. 1 ). All five male patients had persistently low testosterone and required thyroid and/or steroid hormone replacement postoperatively. The difference in preoperative and postoperative endocrine function according to gender could be due to many unclear reasons. Differences in pituitary anatomy and structure in both males and females could potentially explain the differences in endocrine function, as the ischemia caused by the apoplexy and gland swelling from surgery may have a more pronounced effect in males versus females. Due to the relatively small sample size in displaying the effect of gender on endocrine function, further study is warranted.

Long-term endocrine function after the endoscopic approach to pituitary macroadenomas presenting with apoplexy is understudied. The existing literature details postoperative hormonal insufficiency rates of 13% in endoscopic pituitary macroadenoma resection, but it does not address apoplectic macroadenomas. ²⁶ Our rate of 71% of patients with persistent postoperative endocriniopathies suggests that apoplexy in the setting of a macroadenoma may increase endocrine dysfunction, even when identifying and preserving the pituitary gland ( Fig. 1 ). Despite persistent postoperative endocrinopathies in several patients, preservation of posterior pituitary gland function was seen in all patients.

At our institution, we place significant emphasis on determining the location of the normal pituitary gland and stalk through careful study of the preoperative imaging; however, sometimes it is not identifiable given the large size of the hemorrhagic tumor. Previously published studies have described varying and inconsistent results with the use of optical fluorescence agents. ²⁷ Subtle differences in the color, tendency to bleed, and consistency between the hemorrhagic tumor and normal gland can help to identify and potentially preserve the gland ( Fig. 4 ). In this series, based on these differences, the gland could always be identified intraoperatively at the margin of the tumor resection site and confirmed histologically. With this technique, no patients experienced permanent postoperative diabetes insipidus. Forty-three percent of patients (3/7) required a one-time dose of desmopressin for increased urinary output immediately after surgery. Each of these patients' increased urinary output resolved after one dose of DDAVP. This suggests the etiology of the diabetes insipidus was likely due to traction of the posterior pituitary gland and/or mild swelling perioperatively. Due to the resolution of the diabetes insipidus, there was likely no permanent dysfunction of posterior pituitary function postoperatively. In the series by Ricciuti et al, it was unclear what percentage of patients develop diabetes insipidus. ²⁴

Fig. 4.

Intraoperative photo depicting the margin between the apoplectic tumor and normal pituitary gland.

Additionally, when examining time from symptoms to surgery and admission to surgery, it appears that a longer time to surgery is associated with a worse outcome. Out of all seven patients, there were two patients who had both persistent visual acuity deficits and visual field deficits postoperatively. While both symptoms were greatly improved when compared with preoperative neurological exam, the symptoms-to-surgery time for these patients was 18 and 21 days. With a median symptoms-to-surgery time of 4 days, this delay in treatment could likely explain the lack of resolution of visual acuity and visual field deficits in these two patients.

While the current case series supports the safety and efficacy of endoscopic transsphenoidal resection of pituitary macroadenomas presenting with apoplexy, there are several limitations of the current study. Primarily, the small sample size hinders the ability to draw statistically significant conclusions from the data. In particular, the relationship between postoperative endocrine function and gender was found to be statistically significant but may be influenced by the number of patients included in the study. Pituitary macroadenomas presenting with apoplexy is a very rare finding, and thus it is difficult to gather a large sample size. Additional studies with larger sample sizes are necessary for confirmation of the findings in the current study. An additional limitation is the retrospective, nonrandomized nature of the study. We were unable to eliminate selection bias from the patient population. A final limitation is that the study was conducted at a single–tertiary institution, where one surgical team performed every operation. While this ensures that each patient received the same management and surgical care, it presents a lack of external validity of the data to patients receiving care from other surgeons with different levels of surgical skill at different hospitals.

Conclusion

Endoscopic transsphenoidal adenoma resection with gland preservation for treatment of pituitary apoplexy yielded a gross-total resection in 86% of patients with improvement of presenting neurological symptoms in 100% of patients and complete resolution of preoperative neurological deficits in 71% of them. No patients achieved complete postoperative resolution of presenting endocrinopathies.

Rates of resection and symptomatic improvement rival those in the few published macroadenoma series with apoplexy. Identification and preservation of the normal gland still led to anterior but no posterior pituitary dysfunction. Our study showed a statistically significant difference between gender and postoperative endocrinopathy resolution. Further studies with larger sample sizes are necessary to further confirm these findings.