CushieBlog

Introduction

The incidence of Cushing Disease (CD) is estimated to be between 0.12 to 0.24 cases per 100,00 persons per year^1,2. Of these, 7-23% are macroadenomas (>1 cm)3, 4, 5. Pituitary apoplexy is a potentially life-threatening endocrine and neurosurgical emergency which occurs due to infarction or hemorrhage in the pituitary gland. Apoplexy occurs most commonly in non-functioning macroadenomas with an estimated prevalence of 6.2 cases per 100,000 persons and incidence of 0.17 cases per 100,00 persons per year⁶. Corticotroph macroadenoma presenting with apoplexy is uncommon with only a handful of reports in the literature⁷. We present a case of a sparsely granulated corticotroph (SGCT) which presented with apoplexy leading to remission of hypercortisolism and subsequent central adrenal insufficiency.

Case Presentation

A 33-year-old male who was otherwise healthy and not on any medications presented to a community hospital with sudden and severe headache accompanied by hypotension, nausea, vomiting, bitemporal hemianopsia and diplopia. Computed Tomography (CT) scan of the brain demonstrated a hyperattenuating 2.0 cm x 2.8 cm x 1.5 cm mass at the sella turcica with extension into the right cavernous sinus and encasement of the right internal carotid arteries (Figure 1A). He was transferred to a tertiary care center for neurosurgical management with endocrinology consultation post-operatively.

1. Download : Download high-res image (404KB)
2. Download : Download full-size image

Figure 1. hyperattenuating 2.0 cm x 2.8 cm x 1.5 cm mass at the sella turcica on unenhanced CT (A); MRI demonstrated a 1.9 cm x 3.2 cm x 2.4 cm heterogeneous mass on T1 (B) and T2-weighted imaging (C) showing small hyperintense areas in solid part of the sella mass with flattening of the optic chiasm, remodeling/dehiscence of the floor of the sella and extending into the right cavernous sinus with at least partial encasement of the ICA

In retrospect, he reported a 3-year history of ongoing symptoms of hypercortisolism including increased central obesity, dorsal and supraclavicular fat pad, facial plethora, abdominal purple striae, easy bruising, fatigue, decreased libido and erectile dysfunction. Notably, at the time of presentation he did not have a history of diabetes, hypertension, osteoporosis, fragility fractures or proximal muscle weakness. He fathered 2 children previously. His physical examination was significant for Cushingoid facies, facial plethora, dorsal and supraclavicular fat pads and central obesity with significant axillary and abdominal wide purple striae (Figure 2). Neurological examination revealed bitemporal hemianopsia, right third cranial nerve palsy with ptosis and impaired extraocular movement. The fourth and sixth cranial nerves were intact as was the rest of his neurological exam. These findings were corroborated by Ophthalmology.

1. Download : Download high-res image (477KB)
2. Download : Download full-size image

Figure 2. Representative images illustrating facial plethora (A); abdominal striae (B, C); supraclavicular fat pad (D); dorsal fat pad (E)

Initial laboratory data at time of presentation to the hospital included elevated plasma cortisol of 64.08ug/dL (RR, 2.36 – 17.05), ACTH was not drawn at the time of presentation, normal TSH 0.89 mIU/L (RR, 0.36 – 3.74), free thyroxine 0.91ng/dL (RR, 0.76 – 1.46), evidence of central hypogonadism with low total testosterone 28.8 ng/dL (RR, 219.2 – 905.6) and inappropriately normal luteinizing hormone (LH) 1mIU/mL (RR, 1 – 12) and follicle stimulating hormone (FSH) 3mIU/mL (RR, 1 – 12), low prolactin <1 ng/mL (RR, 3 – 20), and normal insulin growth factor – 1 (IGF–1) 179ng/mL (RR, 82 – 242).

A pituitary gland dedicated MRI was performed to further characterize the mass, which re-demonstrated a 1.9 cm x 3.2 cm x 2.4 cm heterogenous mass at the sella turcica extending superiorly and flattening the optic chiasm, remodeling of the floor of the sella and bulging into the sphenoid sinus and extending laterally into the cavernous sinus with encasement of the right internal carotid artery (ICA). As per the radiologist’s diagnostic impression, this appearance was most in keeping with a pituitary macroadenoma with apoplexy (Figure 1B – C).

The patient underwent urgent TSS and decompression with no acute complications. Pathological examination of the pituitary adenoma showed features characteristic of sparsely granulated corticotroph pituitary neuroendocrine tumor (adenoma)⁸, with regional hemorrhage and tumor necrosis (apoplexy). The viable tumor exhibited a solid growth pattern (Figure 3A), t-box transcription factor (T-pit) nuclear immunolabeling (Figure 3B), diffuse cytoplasmic CAM5.2 (low molecular weight cytokeratin) immunolabeling (Figure 3C), and regional weak to moderate intense granular cytoplasmic ACTH immuno-staining (Figure 3D). The tumor was immuno-negative for: pituitary-specific positive transcription factor 1 (Pit-1) and steroidogenic factor 1 (SF-1) transcription factors, growth hormone, prolactin, TSH, FSH, LH, estrogen receptor-alpha, and alpha-subunit. Crooke hyalinization was not identified in an adjacent compressed fragment of non-adenomatous anterior pituitary tissue. Ki-67 immunolabeling showed a 1.5% proliferative index (11 of 726 nuclei).

1. Download : Download high-res image (2MB)
2. Download : Download full-size image

Figure 3. Hematoxylin phloxine saffron staining showing adenoma with solid growth pattern (A); immunohistochemical staining showing T-pit reactivity of tumor nuclei (B); diffuse cytoplasmic staining for cytokeratin CAM5.2 (C); and regional moderately intense granular cytoplasmic staining for ACTH (D). Scale bar = 20 μm

Post-operatively, he developed transient central diabetes insipidus requiring desmopressin but resolved on discharge. His postoperative cortisol was undetectable, ACTH 36 pg/mL (RR, 0 – 81), TSH 0.07 mIU/mL (RR, 0.36 – 3.74), free thyroxine 1 ng/dL (RR, 0.8 – 1.5), LH <1mIU/mL (RR, 1 – 12), FSH 1 mIU/mL (RR, 1 – 12) and testosterone 28.8 ng/dL (RR, 219.2 – 905.6) (Table 1 and Figure 4). One month later, he reported 15 pounds of weight loss and a 5-inch decrease in waist circumference. He also noted a reduction in the dorsal and supraclavicular fat pads, facial plethora, and Cushingoid facies as well as fading of the abdominal stretch marks. His visual field defects and right third cranial nerve palsy resolved on follow up with ophthalmology post-operatively. Repeat MRI six months post-operatively showed minor residual soft tissue along the floor of the sella. He is being followed by Neurosurgery, Ophthalmology, and Endocrinology for monitoring of disease recurrence, visual defects, and management of hypopituitarism.

Table 1. Pre- and post-operative hormonal panel

	POD -1	POD 0	POD1	POD2	POD3	POD16	6 -9 months	Comments
Cortisol(2.4 – 17 ug/dL)	64↓	32↓		11↓	<1.8↓	<1.8↓	1.8↓	HC started POD3 post bloodwork
ACTH(0 – 81 pg/mL)				41↓	36↓	28↓	13↓
TSH(0.36 – 3.74 uIU/mL)	0.89	0.43	0.12↓	0.07↓		0.05↓	0.73
Thyroxine, free(0.8 – 1.5 ng/dL)	0.9	0.9	1.1	1		2.1↑	1	Levothyroxine started POD4
LH(1 – 12 miU/mL)	1↓			<1↓		1↓	3
FSH(1 – 12 mIU/mL)	3↓			1↓		1↓	3
Testosterone(219.2 – 905.6 ng/dL)				28.8↓		<20↓	175.9↓	Testosterone replacement started as outpatient
Testosterone, free(160 – 699 pmol/L)						<5.8↓	137↓
IGF-1(82 – 242 ng/mL)	179					79
GH(fasting < 6 mIU/L)	4.5					<0.3
Prolactin(3 – 20 ng/mL)	<1↓					<1↓

POD, postoperative day; HC, hydrocortisone; ACTH, adrenocorticotropic hormone; TSH, thyroid stimulating hormone; LH, luteinizing Hormone; FSH, follicle stimulating hormone; IGF-1, insulin like growth factor – 1; GH, growth hormone

1. Download : Download high-res image (259KB)
2. Download : Download full-size image

Figure 4. Trend of select pituitary hormonal panel with key clinical events denoted by black arrows.

Discussion

Microadenomas account for the majority of corticotroph tumors, but 7% – 23% of patients are diagnosed with a macroadenoma3, 4, 5. It is even rarer for a corticotroph macroadenoma to present with apoplexy with only a handful of case reports or series in the literature⁷. Due to its rarity, appropriate biochemical workup on presentation, such as including an ACTH with the blood work, may be omitted especially if the patient is going for emergent surgery. In this case, the undetectable prolactin can reflect loss of anterior pituitary function and also suggest a functioning corticotroph adenoma due to the inhibitory effect of long term serum glucocorticoids on prolactin secretion⁹. After undergoing TSS, the patient developed central adrenal insufficiency, hypothyroidism and hypogonadism requiring hormone replacement. Presumably, the development of adrenal insufficiency demonstrated the remission of hypercortisolism as a result of apoplexy and/or TSS. The ACTH remains detectable likely representing residual tumor that was not obliterated by apoplexy nor excised by TSS given it location near the carotid artery and cavernous sinus. The presence of adrenal insufficiency in the setting of detectable ACTH is not contradictory as the physiological hypothalamic-pituitary-adrenal axis has been suppressed by the long-term pathological production of ACTH. IGF-1 and prolactin also failed to recover post-operatively. In CD where the production of IGF-1 and prolactin are attenuated by elevated cortisol, it would then be expected that IGF-1 and prolactin recover after hypercortisolism remission. However, the absence of this observation in our case is likely a sequalae of the apoplexy and extensive surgery leading to pituitary hypofunction.

We also want to highlight features of the pre-operative radiographical findings which can provide valuable insight into the subsequent histology. Previous literature has shown that, on T2-weight MRI, silent corticotroph adenomas are strongly correlated with characteristic a multimicrocystic appearance while nonfunctional gonadotroph macroadenomas are not correlated with this MRI finding¹⁰. The multimicrocystic appearance is described as small hyperintense areas with hyperintense striae in the solid part of the tumor (Figure 1C)¹⁰. This is an useful predictive tool for silent corticotroph adenomas with a sensitivity of 76%, specificity of 95% and a likelihood ratio of 15.3¹⁰.

The ability to distinguish between silent corticotroph macroadenoma and other macroadenomas is important for assessing rate of remission and recurrence risk. In 2017, the WHO published updated classification for pituitary tumors. In this new classification, corticotroph adenomas are further divided into densely granulated, sparsely granulated and Crooke’s cell tumors¹¹. DGCT are intensely Periodic Acid Schiff (PAS) stain positive and exhibit strong diffuse pattern of ACTH immunoreactivity, whereas SGCT exhibit faintly positive PAS alongside weak focal ACTH immunoreactivity^4,12. Crooke’s cell tumors are characterized by Crooke’s hyaline changes in more than 50% of the tumor cells⁴. In the literature, SGCT account for an estimated 19-29% of corticotroph adenomas13, 14, 15. The clinicopathological relevance of granulation pattern in corticotroph tumors was unclear until recently.

In multiple studies examining granulation pattern and tumor size, SGCT were statistically larger^13,15,16. Hence, we suspect that many of the previously labelled silent corticotroph macroadenomas in the literature were SGCT. The traditional teaching of CD has been “small tumor, big Cushing and big tumor, small Cushing” which reflects the inverse relationship between tumor size and symptomatology¹⁷. This observation appears to hold true as Doğanşen et al. found a trend towards longer duration of CD in SGCT of 34 months compared to 26 months in DGCT based on patient history^13,17. It has been postulated that the underlying mechanism of the inverse relationship between tumor size and symptomatology is impaired processing of proopiomelanocortin resulting in less effective secretion of ACTH in corticotroph macroadenomas³. Doğanşen et al. also found that the recurrence rate was doubled for SGCT, while Witek et al. showed that SGCT were less likely to achieve remission postoperatively^13,16.

Similar to other cases of SGCT, the diagnosis was only arrived retrospective after pathological confirmation¹⁰. Interestingly, the characteristic Crooke’s hyaline change of surrounding non-adenomatous pituitary tissue was not observed as one would expect in a state of prolonged glucocorticoid excess in this case. Although classically described, the absence of this finding does not rule out CD. As evident in a recent retrospective study where 10 out of 144 patients with CD did not have Crooke’s hyaline change¹⁸. In patients without Crooke’s hyaline change, the authors found a lower remission rate of 44.4% compared to 73.5% in patients with Crooke’s hyaline change. Together with the detectable post-operative ACTH, sparsely granulated pattern and absence of Crooke’s hyaline change in surrounding pituitary tissue, the risk of recurrence is increased. These risk factors emphasize the importance of close monitoring to ensure early detection of recurrence.

Declaration of Interests

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Conclusion

We present a case of a sparsely granulated corticotroph macroadenoma presenting with apoplexy leading to remission of hypercortisolism and development of central adrenal insufficiency, hypothyroidism and hypogonadism requiring hormone replacement.