Pituitary DOI 10.1007/s11102-014-0622-5

Hypophysitis: a single-center case series Brandon S. Imber • Han S. Lee • Sandeep Kunwar Lewis S. Blevins • Manish K. Aghi



Ó Springer Science+Business Media New York 2014

Abstract Purpose The authors review their treatment experience and summarize clinical outcomes for patients with hypophysitis over the past 15 years. Methods A retrospective analysis was conducted on patients with lymphocytic, granulomatous or IgG4-related hypophysitis treated from 1997 to 2014 at a single academic center. Patients’ medical records were reviewed and binary logistic regression analysis was used to assess whether various clinical parameters were associated with improved outcomes including endocrine function, radiographic appearance and disease recurrence. Results Twenty-one patients (13 women and 8 men) were identified with a median diagnosis age of 37.4 years. All but two patients (90 %) were diagnosed histopathologically and the remaining two were diagnosed clinically with lymphocytic hypophysitis. 16 patients (76 %) had lymphocytic hypophysitis, 3 (14 %) had granulomatous hypophysitis, 1 (5 %) had IgG4-related hypophysitis and 1 Electronic supplementary material The online version of this article (doi:10.1007/s11102-014-0622-5) contains supplementary material, which is available to authorized users. B. S. Imber  S. Kunwar  L. S. Blevins  M. K. Aghi (&) Department of Neurological Surgery, University of California at San Francisco, 505 Parnassus Ave. Room M779, San Francisco, CA 94143, USA e-mail: [email protected]; [email protected] H. S. Lee Neuropathology Division, Department of Anatomic Pathology, University of California at San Francisco, San Francisco, CA, USA S. Kunwar  L. S. Blevins  M. K. Aghi California Center for Pituitary Disorders at The University of California at San Francisco, San Francisco, CA, USA

(5 %) had mixed lymphocytic-granulomatous. Patients presented with various symptoms of expanding sellar mass with most common signs including headache (57 %), polyuria/polydipsia (52 %), vision changes (52 %) and amenorrhea or decreased libido (48 %). Pre-treatment endocrine evaluation revealed that 12 (57 %) patients had complete anterior hypopituitarism, 11 patients (52 %) had diabetes insipidus, ten patients (48 %) had mild hyperprolactinemia and three patients (14 %) had isolated endocrine axis deficiencies with partial gland function. We observed a broad diversity in pre-treatment imaging with common findings including uniform contrast enhancement (62 %), thickened infundibulum (57 %) and loss of hypophysis bright spot on T1 imaging (43 %). Patients were treated with steroids and hormone supplementation as needed. 16 patients (76 %) had recorded post-treatment MRI scans which revealed that half had radiographic improvement and half had stable or worsened post-treatment imaging. Only female gender was found to significantly predict improved odds of post-steroid radiographic improvement. For post-treatment endocrine evaluation, six patients (29 %) did not have an evaluation on record, four patients (19 %) had some improvement in at least one axis, seven patients (33 %) had stable but non-worsened endocrine function and four patients (19 %) had worsened endocrine function post-steroids. Conclusions Hypophysitis is an increasingly recognized diagnosis that can present with a broad array of radiographic and clinical features. Surgical biopsy can be helpful to make definitive diagnosis and may guide treatment decision-making. Keywords Autoimmune hypophysitis  Pituitary inflammation  Sellar mass  Stalk thickening  Granulomatous  Lymphocytic

123

Pituitary

Introduction Hypophysitis is a rare condition marked by inflammatory infiltration of the pituitary gland which can result in tissue destruction, transient or permanent gland dysfunction and endocrinopathies [1, 2]. Hypophysitis has become a fairly complex clinicopathological spectrum as it can be classified in a number of ways, including histologically, anatomically, or etiologically. There are four histological variants of inflammatory hypophysitis: lymphocytic (also known as autoimmune hypophysitis), granulomatous, xanthomatous and necrotizing [3]. There are also cases of mixed histology, with coincident forms of inflammation. Lymphocytic hypophysitis (LH), the most common variant, has an estimated incidence of one in nine million cases per year [2, 4, 5]. 75 % of LH cases occur in women, about 30–70 % are associated with pregnancy or the postpartum period and association with other autoimmune diseases has been reported in as many as 50 % of cases [3, 6]. LH is characterized by focal or diffuse infiltration of predominantly lymphocytes with fewer accompanying plasma cells, eosinophils and fibroblasts [7]. Granulomatous hypophysitis (GH) is rarer, but is also more common in women [3, 8]. Histology reveals discrete granulomas of multinucleated giant cells and histiocytes with surrounding lymphocytes and plasma cells [2]. A small percentage of patients (\5 %) will present with mixed lymphocyticgranulomatous pathology [2]. Anatomically, hypophysitis can be further classified as adenohypophysitis, neurohypophysitis, or infundibulohypophysitis if the anterior gland, posterior gland or pituitary stalk is involved, respectively. Involvement of the entire gland is also possible. A lesion’s anatomical localization can help to predict presenting symptoms (e.g., neurohypophysitis often presents with diabetes insipidus secondary to the loss of anti-diuretic hormone) however all of the sequellae of pituitary dysfunction may be observed. Hypophysitis is classified as primary when no other inflammatory cause can be identified. This is believed to be the most common variant, and is likely autoimmune [2]. Histopathological detection of lymphocytes within the pituitary gland remains the best diagnostic clue, since only between 6 and 57 % of patients with confirmed LH will have positive autoantibodies [9]. However, there have been reports demonstrating specific autoantibodies which seem to be closely related to hypophysitis pathogenesis and may serve as diagnostic surrogates when histopathological examination of the pituitary is not possible [10–12]. For these patients, it is important to use history, clinical examination, imaging and/or biological tests to exclude other forms of primary pituitary failure. Secondary hypophysitis can be attributable to systemic disease (e.g., sarcoidosis, Langerhans cell histiocytosis, disseminated

123

tuberculosis, syphilis), as a response to an underlying pituitary lesion (e.g., Rathke cleft cyst, adenoma) or from certain medications (e.g., ipilimumab) [3, 13]. Recent work has identified an IgG4-related plasmacytic secondary hypophysitis variant, often in patients with IgG4 systemic disease [3, 14]. Relatively few case reports have been published since first descriptions based on clinical suspicion in 2004 [15], and pathological confirmation in 2007 [16], however a recent review suggests that its incidence may be significantly underreported [17]. It remains difficult to predict the clinical course of hypophysitis patients and modern management remains variable despite several multi-patient case series [5, 18–21] and a pooled case review [1]. We sought to contribute by describing our treatment experience of the various hypophysitis subtypes managed over a nearly 20-year time frame.

Methods A retrospective review was performed of UCSF patients with pathologically-confirmed lymphocytic, granulomatous or IgG4-related hypophysitis or high clinical suspicion of hypophysitis between 1997 and 2014. Due to limited incidence, we included all cases irrespective of follow-up duration and included two cases of suspected LH treated without tissue diagnosis. These patients were both peripartum women with radiographic features consistent with hypophysitis and both were presumptively treated per clinical guidelines [4]. All cohort patients were drawn from either a neurosurgical or neuro-pathological database. Since we routinely surgically biopsy our patients, our institution does not typically measure titers of pituitary or hypothalamic autoantibodies. All histologic stains were performed at the UCSF pathology laboratory. This study was approved by the UCSF Committee for Human Research (IRB # 11-05901). Patients’ complete medical records were reviewed and analyzed. For patients with additional autoimmune diseases, relevant autoantibodies were assessed. We included patients who underwent pre- and post-treatment endocrinologic testing at our institution or outside facilities and lab results were not specifically repeated for this study. Relevant testing included TSH and thyroid hormone levels (thyroid axis), IGF-1 (growth hormone axis), LH, FSH and testosterone (gonadotropin axis) and random cortisol or ACTH stimulation test (adrenal axis). Diabetes insipidus (DI) was diagnosed using either the water deprivation test or by confirming decreased urine specific gravity in the context of polyuria and polydipsia. Since recent literature supports radiographic diagnosis whenever possible [22], we analyzed pre-treatment MRI for features previously reported to be associated with hypophysitis.

Pituitary

We assessed three post steroid treatment outcomes: biochemical improvement, radiographic improvement and disease recurrence. For biochemical improvement, the most immediate pre-steroid lab values for all four aforementioned endocrine axes were compared against the most recent labs. Patients were scored as ‘‘improved’’ if pretreatment lab values normalized or if hormone supplementation was stopped, ‘‘stable’’ if pre-treatment endocrinopathies persisted after treatment or ‘‘worse’’ if the patient developed new endocrine axis dysfunction after initiating treatment. Post-treatment MRI reports were compared to pre-steroid baselines to assess for normalization of pituitary features. Disease recurrence was defined as worsening clinical symptoms or MRI evidence of recurrent hypophysitis after initial steroid treatment. Binary logistic regression was performed with various clinical covariates, including gender, age at diagnosis, maximal pre-treatment lesion dimension, presence of pre-treatment anterior hypopituitarism, DI or hyperprolactinemia, and initial treatment with supraphysiologic or physiologic steroid doses. Hypophysitis etiology was not specifically tested as a regression covariate due to small sample sizes. Threshold for statistical significance was 0.05 and statistical analysis was performed using SPSS version 22 (IBM, Armonk, NY).

Results Patient and disease demographics Twenty-one patients (13 women and 8 men) were included (Table 1) with a median age at initial diagnosis (either surgical or clinical) of 37.4 years (women = 34.2, men = 48.3). Nineteen patients (90 %) had histopathologically confirmed hypophysitis whereas the remaining two (10 %) were diagnosed clinically as described. Histologically, 16 patients (76 %) had LH (Fig. 1), 3 (14 %) had GH, one had IgG4-related secondary hypophysitis (Fig. 2) and one had mixed lymphocytic-granulomatous. IgG4related hypophysitis was confirmed based on currently accepted diagnostic criteria of IgG4-related disease, including [10 IgG4 positive plasma cells per high power field, and IgG4/IgG cell ratio of [40 % [23, 24]. Four LH patients presented in the peripartum period. Five patients had autoimmune histories prior to developing hypophysitis, including two with sarcoidosis (Cases 2 and 21), one with type 1 diabetes (Case 18), one with suspected IgG4-related parotiditis (Case 4) and one with hyperthyroidism, likely of autoimmune etiology (Case 13). Two patients had previous pituitary conditions including a prolactinoma surgically resected 25 years prior (Case 18) and a patient with a small presumed Rathke cleft cyst found on MRI 2 years prior (Case 1).

Clinical presentation and endocrinological assessment Table 2 shows the common presenting symptoms in our series; over half presented with headache, DI and/or visual changes. Median time between symptomatic onset and hypophysitis diagnosis was 3.5 months (range 0.5–60 months). Two LH patients (10 %) presented with bitemporal hemianopsia and one LH patient presented with a superior field cut. Prior to steroid treatment, anterior pituitary dysfunction, DI or hyperprolactinemia secondary to stalk effect were assessed (Fig. 3). For the ten patients (48 %) with hyperprolactinemia, serum levels were mild and ranged from 32 to 78.1 lg/L (lab reference 2.8–29.2 lg/L). An additional patient’s serum level of 63 lg/L was attributed to pregnancy. One patient with galactorrea was treated with bromocriptine for 5 years for presumed prolactinoma prior to undergoing resection and re-diagnosis with LH. Since hyperprolactinemia was mild, no patient was initiated on a dopamine agonist after hypophysitis diagnosis. Three patients (14 %), two with LH and one with IgG4-related hypophysitis, presented with isolated DI and one patient presented with DI plus hyperprolactinemia but normal anterior gland function. Three LH patients had pre-treatment partial anterior endocrine dysfunction, all of whom had deficient gonadotropin levels and two of whom had deficient growth hormone (as measured by IGF-1). We did not observe any isolated thyroid or adrenal axis dysfunction in patients with partial gland function. We further investigated the DI patients who presented with multi-dimensional pituitary dysfunction and identified several patterns of symptom progression. Three patients presented with symptoms of anterior lobe dysfunction roughly 2 months before developing DI. Two of these patients had posterior lobe lesions and one had a hyperplastic gland. One patient with a proximal infundibular lesion had symptomatic DI for a year before developing anterior panhypopituitarism, leading to his LH diagnosis. Another patient with an infundibular mass developed simultaneous symptoms of DI and anterior panhypopituitarism with hyperprolactinemia. One patient with a neurohypophysis lesion had symptomatic DI and had only biochemical, but not clinical, evidence of select anterior lobe dysfunction. Finally, a patient with an anterior lobe lesion reported a multi-year history of intermittent polyuria and was found with panhypopituitarism and hyperprolactinemia, but reported headache as chief symptom. Radiological findings Nineteen patients (90 %) had available pre-treatment imaging with a fairly broad diversity of MRI findings, ranging from large sellar and suprasellar masses

123

Gender

F

F

F

M

F

M

M

F

F

F

F

F

F

F

M

F

M

Case

1

123

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

51

50

45

39

37

36

34

32

32

32

29

22

22

71

75

34

31

Age

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

Lymphocytic

IgG4 plasmacytic

Granulomatous

Granulomatous

Granulomatous

Histopath

33.4

14.1

0.7

20.3

32.5

199.0

11.0

81.7

21.2

0

35.5

12.5

172.9

1.2

0.1

37.3

25.2

Followup (mos)

Polydipsia

Headaches, dizziness

Mild headaches with N/V, weight loss, impaired sense of smell, erectile dysfunction, cold intolerance, fatigue, polyuria

Headaches with N/V, hyponatremia, weight loss

Headaches, amenorrhea, polyuria, polydipsia

Postpartum amenorrhea, fatigue and polyuria, polydipsia

Headache, visual changes, photophobia and eye pain

Amenhorrhea, visual changes, headache

Weight loss, severe headache with N/V, episodes of blindess

Galactorrhea

Polyuria, headache, decreased libido, personality changes, visual changes

Polyuria, polydipsia

Progressive vision loss in third trimester of pregnancy

Polyuria, polydipsia

N/V, dizziness, hyponatremia/hypokalemia, fatigue, weight loss

Headache, nausea, amenorrhea, fatigue, visual changes

Headaches

Presenting symptoms

Table 1 Patient cohort demographic and treatment characteristics

N

N

N

N

N

Y

Y

Y

N

N

N

N

Y

N

N

N

N

Associated with preg?

Panhypopituitary, hyperprolactinemia, DI GH deficiency, central hypogonadism, DI

Panhypopituitary, hyperprolactinemia, DI

Panhypopituitary

Panhypopituitary, DI

Panhypopituitary, DI

Panhypopituitary, hyperprolactinemia

Panhypopituitary, hyperprolactinemia

Panhypopituitary, hyperprolactinemia

Central hypogonadism, hyperprolactinemia

Panhypopituitary, DI

DI

Hyperprolactinemia

DI

Panhypopituitary, hyperprolactinemia

Hyperprolactinemia

Hyperprolactinemia, DI

Pre-treatment endocrine assessment

Subtotal

Biopsy

Biopsy

Biopsy

Biopsy

Biopsy

No surg

Biopsy

No surg

Gross total

Biopsy

Failed biopsy

Biopsy

Biopsy

Biopsy

Biopsy

Biopsy

Extent of first surgery

Y

N

N

N

N

Y

N

N

N

N

N

Y

N

N

N

N

N

Multiple surg?

N

Y

N

Y

N/A

N

Y

Y

N

N/A

N

Y

Y

N

N/A

Y

Y

Poststeroid MRI resolution?

Worsened

Improved

N/A

No improvement

Improved

No improvement

No improvement

No improvement

No improvement

N/A

No improvement

N/A

Improved

N/A

N/A

Worsened

Improved

Post-treatment endocrine status

Pituitary

Worsened Panhypopituitary N Headache, blurred vision, dizziness, erectile dysfunction 132.5 Mixed (L/G) M 21

38

66.5 Lymphocytic M 20

56

N N Subtotal

Worsened N

GH deficiency, central hypogonadism

N N Subtotal

No improvement N/A N M 19

54

Lymphocytic

0.4

Polyuria, polydipsia with associated fatigue, minor headache and decreased libido Hypogonadism

N

DI

Biopsy

N/A N/A N F 18

53

Lymphocytic

0.9

Increasing fatigue with N/V, polyuria, polydipsia

N

Panhypopituitary, hyperprolactinemia, DI

Biopsy

Post-steroid MRI resolution? Age Gender Case

Table 1 continued

Histopath

Followup (mos)

Presenting symptoms

Associated with preg?

Pre-treatment endocrine assessment

Extent of first surgery

Multiple surg?

Post-treatment endocrine status

Pituitary

indistinguishable from the normal gland to fairly normal imaging patterns. Figure 4 shows representative pre-treatment MRI images to demonstrate the range of presentations. For patients with discrete masses, the median maximal dimension was 1.6 cm with a range of 0.7–3.7 cm. One GH patient had a cystic lesion (Case 1) whereas all remaining lesions were solid. Eight patients (38 %) had sellar lesions with concomitant suprasellar extension, four patients (19 %) had sellar lesions, four patients (19 %) had purely infundibular lesions, one patient had a multifocal sellar-infundibular lesion and one patient had an infundibular-hypothalamic lesion. Table 3 highlights pre-treatment MRI features. Treatment course All patients were given hormone replacement and/or DDAVP as needed. 19 patients (90 %) underwent at least one surgery for definitive diagnosis. Two patients (Cases 14 and 15) were treated with pre-surgical glucocorticoids. Both experienced symptomatic improvement but underwent biopsy for definitive diagnosis when imaging failed to improve. The two LH patients diagnosed clinically received steroids presumptively. In the surgical cohort, 18 underwent a microscopic endonasal transsphenoidal approach and one patient whose lesion had hypothalamic extension (Case 21) underwent craniotomy. One patient had a gross total resection of the atypical tissue, three patients (16 %) had subtotal resections and 15 patients (79 %) underwent biopsies for pathologic diagnosis. One patient (Case 6) was deemed to have had an unsuccessful transsphenoidal biopsy and then underwent fronto-temporal craniotomy approximately 1 month later for definitive pathology. Specifically, this patient was presumed to have Langerhans cell histiocytosis and his initial biopsy pathological assessment showed mixed inflammatory infiltrate without evidence of atypical histiocytes. We report no significant operative complications. Post-surgically, or after clinical diagnosis, all patients were started on glucocorticoids for treatment of pituitary inflammation as well as adrenal insufficiency as appropriate. We had the initial steroid regimens for 15 patients (71 %) in our series. Ten patients (66 %) were started on dexamethasone, four (27 %) on hydrocortisone and one on prednisone. Eleven patients (73 %) were started on a physiologic replacement dose whereas four (27 %) were started on a supraphysiologic dose. Patients who were found to have granulomatous, mixed or IgG4-related hypophysitis underwent additional workup to determine if their pituitary dysfunction was secondary to systemic disease. One patient (Case 2) had remote history of biopsy-confirmed granulomatous gastritis and was diagnosed with systemic sarcoidosis. Following pituitary biopsy, another patient (Case 3) was found to have hilar lymphadenopathy and a right kidney mass. She was also

123

Pituitary

Fig. 1 Lymphocytic hypophysitis. a H and E histology of anterior pituitary effaced by a mononuclear inflammatory infiltrate. b Synaptophysin immunohistochemistry confirms the presence of intact

lobular architecture of anterior pituitary cells underlying the inflammatory infiltrate. c CD3 immunohistochemistry shows most of the infiltrate consists of T-lymphocytes. Bars = 100 lm

Fig. 2 IgG4 hypophysitis. a H and E histology showing plasma cellpredominant inflammatory infiltrate. b IgG4 immunohistochemistry demonstrates numerous positive plasma cells in this high power field.

c IgG immunohistochemistry demonstrates similar numbers of positive plasma cells as IgG4. Bars = 100 lm

123

Pituitary Table 2 Hypophysitis clinical presentation

Headache

Whole series (n = 21)

Lymphocytic (n = 16)

Granulomatous (n = 3)

Mixed (n = 1)

IgG4-related (n = 1)

12 (57 %)

9 (56 %)

2 (66 %)

1

0

Polyuria/polydipsia

11 (52 %)

9 (56 %)

1 (33 %)

0

1

Visual changes

11 (52 %)

8 (50 %)

2 (66 %)

1

0

Hypogonadism/amenorrhea

10 (48 %)

8 (50 %)

1 (33 %)

1

0

Fatigue

8 (38 %)

6 (38 %)

2 (66 %)

0

0

Nausea/vomiting

6 (29 %)

4 (25 %)

2 (66 %)

0

0

Weight loss Electrolyte imabalance

4 (19 %) 2 (10 %)

3 (19 %) 1 (6 %)

1 (33 %) 1 (33 %)

0 0

0 0

Galactorrhea

1 (5 %)

1 (6 %)

0

0

Pre-treatment endocrinopathies Anterior pan- Hyperhypopituitary prolactinemia

Yes Yes No

Yes No No

Diabetes insipidus

Number of cases

Yes

3 (3 LH)

No

3 (2 LH, 1 GH)

Yes

3 (3 LH)

No

3 (2 LH, 1 mixed)

Yes

1 (1 GH)

No

3 (2 LH, 1 GH)

Yes

4 (3 LH, 1 IgG4)

No

1 (1 LH)

Fig. 3 Pre-treatment endocrine assessment

presumed to have disseminated sarcoidosis but was lost to follow-up prior to confirmation. The third case of GH had a more complicated past medical history with transiently elevated angiotensin converting enzyme (ACE), recent PPD conversion and a presumed small Rathke cleft cyst. She was treated with steroids and then methotrexate for suspected sarcoidosis but diagnosis was never confirmed. The patient with mixed pathology (Case 21) had a remote sarcoidosis diagnosis but underwent transsphenoidal biopsy because of strong suspicion that his pituitary process was unrelated. Finally, the IgG4-related hypophysitis patient underwent full body imaging and work-up for additional sites of disease. Several notable cases are included as Supplemental Material. Patient outcomes and disease recurrence Outcomes were monitored following initial steroid therapy for a median follow-up period of 28.9 months

0

(range = 0.4–199 months). Three patients (14 %) were lost to follow-up after biopsy. Three patients had a short interval of follow-up (0.4–1.2 months) however two of these patients had recent transsphenoidal surgery and are still undergoing active care. Radiographically, four patients (19 %) did not have recorded post-treatment scans and have been lost to followup and one patient was recently biopsied (Case 19) and has not yet had post-operative imaging. Of the remaining sixteen patients, eight patients (50 %) had radiographic improvement with reduction in lesion size and/or normalization of sellar appearance as determined by official neuro-radiology reports. Eight (50 %) had stable or worse imaging findings compared to pre-treatment baselines despite steroid therapy, six of which had LH, one had the IgG4-related variant, and one had mixed lymphocyticgranulomatous hypophysitis. Treatment success (based on radiographic improvement) did not vary by hypophysitis subtype. Of note, one patient with limited radiographic improvement was likely non-compliant with steroid therapy. Using binary logistic regression analysis, only female gender was found to significantly predict improved odds of post-steroid radiographic improvement whereas younger age was suggestive of increased odds but did not reach the threshold for significance (Table 4). In terms of postoperative endocrine evaluation, six patients (29 %) were either lost to follow-up or have not yet had post-treatment endocrinological evaluation. Four patients (19 %) had some endocrine function improvement following initial steroid therapy. One patient had solely mild prolactinemia pre-steroids with complete resolution. Two LH patients with pre-treatment anterior panhypopituitarism and DI had persistent DI but regained all anterior pituitary function and required only transient hormone supplementation. Finally, one GH patient was treated with steroids and then methotrexate for possible sarcoidosis had improvement in her pre-treatment DI and hyperprolactinemia. Seven

123

Pituitary

Fig. 4 Representative preoperative T1 post-gadolinium MRI imaging results. a Sagittal view of a 28-year-old M with lymphocytic hypophysitis revealing a fusiform, 7 9 4 mm region of abnormal enhancement involving the proximal pituitary stalk with loss of posterior pituitary intrinsic T1 hyperintensity. b Coronal view of a 53-year-old M with lymphocytic hypophysitis showing normally sized pituitary with abnormally thickened and enhancing infundibulum. c Sagittal view of a 30-year-old F with history of suspected small

Rathke cleft cyst who developed granulomatous hypophysitis, found to have partially cystic enhancing sellar/suprasellar mass measuring 1.7 9 1.5 9 1.2 cm. d Sagittal view of a 70-year-old M with IgG4 plasmacytic hypophysitis revealing heterogeneously enhancing tissue throughout the sella which was indistinguishable from normal enhancing pituitary gland, a thickened infundibulum and loss of normal T1 hyperintensity in the posterior pituitary

patients (33 %) had stable, but non-worsened, endocrine function following treatment. Six of these cases had pretreatment panhypopituitarism and remained on long-term hormone supplementation whereas the final patient had DI and remained on DDAVP and low-dose dexamethasone therapy. Finally, the remaining four patients (19 %) had

worsened post-treatment endocrine function. One patient initially had growth hormone axis deficiency and central hypogonadism but developed anterior panhypopituitarism. One patient with GH initially had isolated hyperprolactinemia but developed adrenal insufficiency and central thyroid dysfunction. One patient had DI and isolated growth

123

Pituitary Table 3 Hypophysitis pretreatment radiographic features

Radiographic feature

Total cohort (n = 19)

Lymphocytic (n = 15)

Granulomatous (n = 3)

IgG4-related (n = 1)

Uniform contrast enhancement

13 (68 %)

10 (66 %)

2 (66 %)

1

Thickened infundibulum

12 (63 %)

9 (60 %)

2 (66 %)

1

Infundibulum midline

10 (53 %)

6 (40 %)

3 (100 %)

1

Loss of neurohypophysis bright spot on T1 imaging

9 (47 %)

6 (40 %)

2 (66 %)

1

Sellar expansion

8 (42 %)

5 (33 %)

3 (100 %)

0

Heterogeneous contrast enhancement

2 (11 %)

1 (7 %)

Cavernous sinus invasion Concomitant arteritis

2 (11 %) 1 (5 %)

1 (7 %) 1 (7 %)

Dural tail

1 (5 %)

1 (7 %)

hormone and gonadotropin deficiencies which evolved into complete panhypopituitarism (and resulted in a second biopsy). The last patient presented with anterior panhypopituitarism but developed worsening endocrine function post-therapy. We did not identify any covariates significantly associated with improved pituitary function compared to stable or worse pituitary function. Six patients (29 %) had recurrent or worsening hypophysitis following initial steroid therapy, five of which had LH and one had mixed lymphocytic-granulomatous. One patient was undergoing a steroid taper trial and re-presented with headaches, nausea, vomiting and somnolence. MRI confirmed that her pituitary mass had enlarged and she was escalated to supraphysiologic dexamethasone doses. One patient had worsening imaging despite presumptive steroid treatment (Case 15), and then underwent transsphenoidal biopsy for definitive diagnosis. Two patients (Cases 12 and 17) underwent a second transsphenoidal procedure following disease recurrence. One of these repeat surgical patients had MRI evidence of focal infundibular recurrence though histopathology of his second biopsy revealed only crush artifact. He was restarted on steroids and has remained clinically stable. The other repeat surgical patient had symptomatic recurrence and underwent transsphenoidal resection of approximately two-thirds of her pituitary gland, rendering her functionally panhypopituitary. The final patient, who initially had mixed pathology demonstrated no clinical or radiographic improvement and had several surveillance MRIs revealing an enlarged lesion despite high-dose steroids. This patient was transitioned to methotrexate, however he slowly declined and eventually died from fulminant liver and bone marrow suppression secondary to hemophagocytic lymphomatoid granulomatosis. On binary logistic analysis, improved post-treatment MRI imaging seemed to reduce risk of recurrent or worsening disease (odds ratio of 0.086, 95 % CI of 0.007–1.084,

0

1

1 (33 %) 0

0 0

0

0

Table 4 Predictors of post-steroid radiographic improvement Covariate

Odds ratio

95 % CI

p value

Female gender (vs. male)

21.00

1.50–293.3

0.024

Age at diagnosis (years)

0.90

0.80–1.02

0.100

Maximal pre-treatment lesion dimension (cm) Presence of pre-treatment pan anterior pituitary dysfunction

1.51

0.28–8.2

0.636

0.60

0.08–4.4

0.615

5.00

0.58–42.8

0.142

Presence of pre-treatment hyperprolactinemia Presence of pre-treatment DI

0.36

0.05–2.7

0.323

Initial treatment with supraphysiologic steroid dose (vs. physiologic)

n/a

n/a

0.99

Endocrine axis improvement (vs. stable/worse)

n/a

n/a

0.99

p = 0.058). In our series, no other covariates significantly reduced odds of recurrence.

Discussion Hypophysitis has classically been defined as a condition affecting peripartum women who present with sequellae of an expanding sellar mass [2, 25, 26]. However, over the past two decades, improved recognition of the condition and increased availability of sellar MR imaging has revealed much broader susceptibility. This is highlighted by our own experience, with only four peripartum cases. The clinicopathological spectrum is now significantly more complicated, and hypophysitis should be considered for any patient with endocrine dysfunction and sellar enlargement lacking a tumor-gland interface and characteristic MRI hypodensity. Differential diagnosis includes non-functional adenomas, other systemic inflammatory or

123

Pituitary

granulomatous diseases (e.g., Langerhans cell histiocytosis, granulomatosis with polyangiitis), pituitary lymphoma, abscesses or teratomas. Hypophysitis can be classified based on histopathology, anatomy or etiology, though these criteria can simplify the vast diversity of potential presentations. Even in our small, heterogeneous series, it is clear that there is no ‘‘typical’’ hypophysitis patient, with onsets ranging from acute to chronic associated with several radiographic and symptom profiles. Specifically, our review of the symptom trajectory for patients with multi-axis dysfunction suggests that anatomic lesion localization may not always predict how that patient will present clinically. For example, some patients with posterior lobe lesions first presented with symptoms of anterior pituitary endocrinopathies followed several months later by DI, whereas for others, DI was the first symptom. Our cohort is too small to definitively predict the relationship between these seemingly distinct forms of pituitary dysfunction and we did not identify relationships between these forms of dysfunction and the odds of poststeroid biochemical or radiographic improvement. Literature review suggests that the epidemiology is difficult to calculate, but estimates for autoimmune hypophysitis range from 0.24 to 0.88 % of pituitary surgery cases [1, 6, 19, 27–29]. Given that we report 16 cases of LH between 1997 and 2014, during which time our center performed nearly 3,000 pituitary surgeries, we extrapolate our prevalence to be even lower. To highlight a few notable cases, we report a 70-year-old man with IgG4-related hypophysitis who presented similarly to one of the first published reports [30]. This pathology, thought to be a secondary form of LH, is the newest addition to the disease spectrum [3]. The number of published cases remains limited, however the demographics seem to suggest higher prevalence in men in the seventh and eighth decades [3]. Our institution only recently started routine staining for IgG4 so it is possible that our cohort has mis-categorized some LH patients. Now recognized as a member of the IgG4related systemic diseases (RSD), patients with pituitary involvement may have additional sites of inflammation and fibrosis such as the salivary glands, pancreas, liver, retroperitoneum and gallbladder. In fact, our case presented with an undiagnosed chronic sialadenitis roughly 10 years before his pituitary condition which resulted in parotidectomy. While the tissue was not specifically stained for IgG4, it is possible that this represented an early manifestation. Furthermore, this patient was found to have pleural thickening and a 2.8 cm non-calcified pulmonary nodule. There have been other reports of IgG4-related lung pseudotumors [31– 33] thus the patient was maintained on supraphysiologic prednisone despite resolution of his pituitary symptoms, and will be evaluated with a short-interval CT scan. This case, and numerous others in the literature highlight that

123

hypophysitis should be considered in patients with known IgG4-RSD who present with new neurological or endocrine symptoms. Conversely, patients with new diagnoses of seemingly isolated IgG4-related hypophysitis should be screened for other sites of involvement. We also report two cases of patients with prior pituitary abnormalities who developed LH, one with a distantly resected prolactinoma and the other who developed GH 2 years after a suspected small Rathke cleft cyst was discovered on imaging. Hypophysitis associated with prolactinoma [19] and Rathke cleft cyst rupture [34–40] both have precedent in the literature and it is suspected that mucins produced by cells lining the Rathke cyst wall can produce a granulomatous giant cell reaction [36]. The final case is a 38-year-old man with past history of disseminated sarcoidosis found to have mixed lymphocytic and granulomatous hypophysitis. He was transitioned to methotrexate following no response to supra-physiologic glucocorticoids. Ten years later, the patient died of a disseminated lymphomatous process. Pathological analysis favored lymphomatoid granulomatosis (LG), which is a rare Epstein–Barr virus–associated B cell lymphoproliferative condition [41–44]. LG has predominantly pulmonary and vascular involvement however, as many as 30 % of patients report neurological symptoms [45]. Patients with neurological involvement generally have either focal intraparenchmal lesions or leptomeningeal/cranial nerve enhancement [46] and review of the literature for associations with pituitary dysfunction revealed only two examples. One patient had active pulmonary and cutaneous involvement, presented with central DI, and was then found to have a thickened hypophyseal stalk and empty sella on imaging [47]. This patient was successfully treated with CHOP chemotherapy but had persistent DI. Another study reported one LG patient with an enhancing suprasellar mass which infiltrated the pituitary stalk [46]. Thus, pituitary involvement is possible, but rare, and no treatment consensus has emerged. Most of our patients presented with partial or complete anterior pituitary dysfunction, which aligns with estimates from the literature that suggest that 66–97 % of cases will have some endocrinopathy [4]. The majority had persistent dysfunction after steroids. In our series, female patients, and possibly younger patients had improved chances of post-steroid radiographic improvement. Relapse rate in our series was almost 30 %, and there is no standard management approach for recurrent hypophysitis [48]. Treatment options include additional corticosteroids, other immunosuppressive agents such as azathioprine [49–54] or rituximab [55], surgery or radiosurgery [56, 57]. Literature review also suggests that clinicians’ appropriate preference is for non-surgical treatment whenever possible [58]. There have been a number of associated radiographic findings including symmetrical gland or sellar

Pituitary

expansion, homogeneous post-gadolinium enhancement, thickened but midline infundibulum and loss of posterior pituitary ‘‘bright spot’’ on T1 imaging [22]. Gutenberg et al. have developed an algorithm to differentiate adenomas from autoimmune hypophysitis, thereby facilitating prospective treatment. We acknowledge that the high prevalence of biopsy represents a potential bias in our sample, since we utilized neurosurgical and pathological databases to assemble our case series. Thus, we may be excluding some subacute or mild chronic cases. Another limitation of our work is that we cannot assess LH etiology, since our laboratory does not routinely assay for antipituitary antibodies. There have been interesting studies that have linked certain autoantibodies to clinical symptoms [11, 12], suggesting that we may eventually develop more diagnostic assays. For now, we feel that biopsy can be very helpful for management of hypophysitis and recent reviews continue to advocate it as the diagnostic gold standard [6]. Specifically, there were several cases where pre-treatment clinical context or radiographic appearance could not differentiate between inflammatory lesions, sarcoidosis, Langerhans cell histiocytosis, or lymphoma, each of which would necessitate a different treatment approach (e.g., glucocorticoids, focused radiotherapy or chemoradiotherapy, respectively). Given the safety of transsphenoidal surgery [59–61], we felt it was in the patients’ best interest to achieve a definitive diagnosis.

Conclusion We report a modern series of 21 patients who presented with lymphocytic, granulomatous or IgG4-related hypophysitis. This series exhibited a wide range of clinical and radiographic characteristics highlighting the breadth of this increasingly recognized condition. In general, despite steroid therapy, most patients in our series had some persistent endocrinopathy requiring hormone supplementation. While we agree that the goal should be for clinical diagnosis and medical management as possible, we report the continued importance of surgical diagnosis, particularly in patients whose differential diagnosis is broad. Acknowledgments We thank Jeffrey Wagner, BS for his assistance with identification of cases. Conflict of interest of interest.

The authors declare that they have no conflict

References 1. Caturegli P, Newschaffer C, Olivi A et al (2005) Autoimmune hypophysitis. Endocr Rev 26:599–614

2. Caturegli P, Lupi I, Landek-Salgado M et al (2008) Pituitary autoimmunity: 30 years later. Autoimmun Rev 7:631–637 3. Leporati P, Landek-Salgado MA, Lupi I et al (2011) IgG4-related hypophysitis: a new addition to the hypophysitis spectrum. J Clin Endocrinol Metab 96:1971–1980 4. Rivera J-A (2006) Lymphocytic hypophysitis: disease spectrum and approach to diagnosis and therapy. Pituitary 9:35–45 5. Khare S, Jagtap VS, Budyal SR et al (2013) Primary (autoimmune) hypophysitis: a single centre experience. Pituitary 6. Falorni A, Minarelli V, Bartoloni E et al (2014) Diagnosis and classification of autoimmune hypophysitis. Autoimmun Rev 13:412–416 7. Goudie RB, Pinkerton PH (1962) Anterior hypophysitis and Hashimoto’s disease in a young woman. J Pathol Bacteriol 83:584–585 8. Shi J, Zhang J, Wu Q et al (2009) Granulomatous hypophysitis: two case reports and literature review. J Zhejiang Univ Sci B 10:552–558 9. Hamnvik O-PR, Laury AR, Laws ER, Kaiser UB (2010) Lymphocytic hypophysitis with diabetes insipidus in a young man. Nat Rev Endocrinol 6:464–470 10. Bellastella A, Bizzarro A, Coronella C et al (2003) Lymphocytic hypophysitis: a rare or underestimated disease? Eur J Endocrinol Eur Fed Endocr Soc 149:363–376 11. De Bellis A, Sinisi AA, Conte M et al (2007) Antipituitary antibodies against gonadotropin-secreting cells in adult male patients with apparently idiopathic hypogonadotropic hypogonadism. J Clin Endocrinol Metab 92:604–607 12. De Bellis A, Dello Iacovo A, Bellastella G et al (2014) Characterization of pituitary cells targeted by antipituitary antibodies in patients with isolated autoimmune diseases without pituitary insufficiency may help to foresee the kind of future hypopituitarism. Pituitary 17:457–463 13. Torino F, Barnabei A, De Vecchis L et al (2012) Hypophysitis induced by monoclonal antibodies to cytotoxic T lymphocyte antigen 4: challenges from a new cause of a rare disease. Oncologist 17:525–535 14. Shimatsu A, Oki Y, Fujisawa I, Sano T (2009) Pituitary and stalk lesions (infundibulo-hypophysitis) associated with immunoglobulin G4-related systemic disease: an emerging clinical entity. Endocr J 56:1033–1041 15. Van der Vliet HJJ, Perenboom RM (2004) Multiple pseudotumors in IgG4-associated multifocal systemic fibrosis. Ann Intern Med 141:896–897 16. Wong S, Lam WY, Wong WK, Lee KC (2007) Hypophysitis presented as inflammatory pseudotumor in immunoglobulin G4related systemic disease. Hum Pathol 38:1720–1723 17. Bando H, Iguchi G, Fukuoka H et al (2014) The prevalence of IgG4-related hypophysitis in 170 consecutive patients with hypopituitarism and/or central diabetes insipidus and review of the literature. Eur J Endocrinol Eur Fed Endocr Soc 170:161–172 18. Honegger J, Fahlbusch R, Bornemann A et al (1997) Lymphocytic and granulomatous hypophysitis: experience with nine cases. Neurosurgery 40:713–722. Discussion 722–723 19. Leung GKK, Lopes M-BS, Thorner MO et al (2004) Primary hypophysitis: a single-center experience in 16 cases. J Neurosurg 101:262–271 20. Gutenberg A, Hans V, Puchner MJA et al (2006) Primary hypophysitis: clinical-pathological correlations. Eur J Endocrinol Eur Fed Endocr Soc 155:101–107 21. Menon SK, Sarathi V, Bandgar TR et al (2009) Autoimmune hypophysitis: a single centre experience. Singap Med J 50:1080–1084 22. Gutenberg A, Larsen J, Lupi I et al (2009) A radiologic Score to distinguish autoimmune hypophysitis from nonsecreting pituitary adenoma preoperatively. Am J Neuroradiol 30:1766–1772 23. Leporati P, Landek-Salgado MA, Lupi I et al (2011) IgG4-related hypophysitis: a new addition to the hypophysitis spectrum. J Clin Endocrinol Metab 96:1971–1980

123

Pituitary 24. Umehara H, Okazaki K, Masaki Y et al (2012) Comprehensive diagnostic criteria for IgG4-related disease (IgG4-RD), 2011. Mod Rheumatol Jpn Rheum Assoc 22:21–30 25. Asa SL, Bilbao JM, Kovacs K et al (1981) Lymphocytic hypophysitis of pregnancy resulting in hypopituitarism: a distinct clinicopathologic entity. Ann Intern Med 95:166–171 26. Landek-Salgado MA, Gutenberg A, Lupi I et al (2010) Pregnancy, postpartum autoimmune thyroiditis, and autoimmune hypophysitis: intimate relationships. Autoimmun Rev 9:153–157 27. Sautner D, Saeger W, Lu¨decke DK et al (1995) Hypophysitis in surgical and autoptical specimens. Acta Neuropathol (Berl) 90:637–644 28. Fehn M, Sommer C, Ludecke DK et al (1998) Lymphocytic hypophysitis: light and electron microscopic findings and correlation to clinical appearance. Endocr Pathol 9:71–78 29. Buxton N, Robertson I (2001) Lymphocytic and granulocytic hypophysitis: a single centre experience. Br J Neurosurg 15:242–245. Discussion 245–246 30. Yamamoto M, Takahashi H, Ohara M et al (2006) A case of Mikulicz’s disease (IgG4-related plasmacytic disease) complicated by autoimmune hypophysitis. Scand J Rheumatol 35:410–411 31. Zen Y, Kitagawa S, Minato H et al (2005) IgG4-positive plasma cells in inflammatory pseudotumor (plasma cell granuloma) of the lung. Hum Pathol 36:710–717 32. Fujiu K, Sakuma H, Miyamoto H, Yamaguchi B (2010) Immunoglobulin G4-related inflammatory pseudotumor of the lung. Gen Thorac Cardiovasc Surg 58:144–148 33. Nishikawa G, Nakamura K, Yamada Y et al (2011) Inflammatory pseudotumors of the kidney and the lung presenting as immunoglobulin G4-related disease: a case report. J Med Case Rep 5:480 34. Albini CH, MacGillivray MH, Fisher JE et al (1988) Triad of hypopituitarism, granulomatous hypophysitis, and ruptured Rathke’s cleft cyst. Neurosurgery 22:133–136 35. Wearne MJ, Barber PC, Johnson AP (1995) Symptomatic Rathke’s cleft cyst with hypophysitis. Br J Neurosurg 9:799–803 36. Roncaroli F, Bacci A, Frank G, Calbucci F (1998) Granulomatous hypophysitis caused by a ruptured intrasellar Rathke’s cleft cyst: report of a case and review of the literature. Neurosurgery 43:146–149 37. Hama S, Arita K, Tominaga A et al (1999) Symptomatic Rathke’s cleft cyst coexisting with central diabetes insipidus and hypophysitis: case report. Endocr J 46:187–192 38. Daikokuya H, Inoue Y, Nemoto Y et al (2000) Rathke’s cleft cyst associated with hypophysitis: MRI. Neuroradiology 42:532–534 39. Nishikawa T, Takahashi JA, Shimatsu A, Hashimoto N (2007) Hypophysitis caused by Rathke’s cleft cyst case report. Neurol Med Chir (Tokyo) 47:136–139 40. Komatsu F, Tsugu H, Komatsu M et al (2010) Clinicopathological characteristics in patients presenting with acute onset of symptoms caused by Rathke’s cleft cysts. Acta Neurochir (Wien) 152:1673–1678 41. Myers JL, Kurtin PJ, Katzenstein AL et al (1995) Lymphomatoid granulomatosis. Evidence of immunophenotypic diversity and relationship to Epstein–Barr virus infection. Am J Surg Pathol 19:1300–1312 42. Jaffe ES, Wilson WH (1997) Lymphomatoid granulomatosis: pathogenesis, pathology and clinical implications. Cancer Surv 30:233–248

123

43. Roschewski M, Wilson WH (2012) Lymphomatoid granulomatosis. Cancer J Sudbury Mass 18:469–474 44. Tagliavini E, Rossi G, Valli R et al (2013) Lymphomatoid granulomatosis: a practical review for pathologists dealing with this rare pulmonary lymphoproliferative process. Pathologica 105:111–116 45. Katzenstein AL, Carrington CB, Liebow AA (1979) Lymphomatoid granulomatosis: a clinicopathologic study of 152 cases. Cancer 43:360–373 46. Patsalides AD, Atac G, Hedge U et al (2005) Lymphomatoid granulomatosis: abnormalities of the brain at MR imaging. Radiology 237:265–273 47. Bushunow PW, Casas V, Duggan DB (1996) Lyphomatoid granulomatosis causing central diabetes insipidus: case report and review of the literature. Cancer Invest 14:112–119 48. Carmichael JD (2012) Update on the diagnosis and management of hypophysitis. Curr Opin Endocrinol Diabetes Obes 19:314–321 49. Lecube A, Francisco G, Rodrı´guez D et al (2003) Lymphocytic hypophysitis successfully treated with azathioprine: first case report. J Neurol Neurosurg Psychiatry 74:1581–1583 50. Curto` L, Torre ML, Cotta OR et al (2010) Lymphocytic hypophysitis: differential diagnosis and effects of high-dose pulse steroids, followed by azathioprine, on the pituitary mass and endocrine abnormalities—report of a case and literature review. Sci World J 10:126–134 51. Lu Z, Yang G, Ba J et al (2011) Spontaneous pregnancy in a patient with a relapse of lymphocytic hypophysitis successfully treated with azathioprine and glucocorticoids. Endocrine 39:296–297 52. Papanastasiou L, Pappa T, Tsiavos V et al (2011) Azathioprine as an alternative treatment in primary hypophysitis. Pituitary 14:16–22 53. Yang G-Q, Lu Z-H, Gu W-J et al (2011) Recurrent autoimmune hypophysitis successfully treated with glucocorticoids plus azathioprine: a report of three cases. Endocr J 58:675–683 54. Caputo C, Bazargan A, McKelvie PA et al (2014) Hypophysitis due to IgG4-related disease responding to treatment with azathioprine: an alternative to corticosteroid therapy. Pituitary 17:251–256 55. Schreckinger M, Francis T, Rajah G et al (2012) Novel strategy to treat a case of recurrent lymphocytic hypophysitis using rituximab. J Neurosurg 116:1318–1323 56. Selch MT, DeSalles AAF, Kelly DF et al (2003) Stereotactic radiotherapy for the treatment of lymphocytic hypophysitis: report of two cases. J Neurosurg 99:591–596 57. Ray DK, Yen CP, Vance ML et al (2010) Gamma knife surgery for lymphocytic hypophysitis. J Neurosurg 112:118–121 58. Howlett TA, Levy MJ, Robertson IJ (2010) How reliably can autoimmune hypophysitis be diagnosed without pituitary biopsy. Clin Endocrinol (Oxf) 73:18–21 59. Swearingen B, Barker FG 2nd, Katznelson L et al (1998) Longterm mortality after transsphenoidal surgery and adjunctive therapy for acromegaly. J Clin Endocrinol Metab 83:3419–3426 60. Jahangiri A, Wagner JR, Pekmezci M et al (2013) A comprehensive long-term retrospective analysis of silent corticotrophic adenomas vs hormone-negative adenomas. Neurosurgery 73:8–17. Discussion 17–18 61. Jahangiri A, Wagner J, Han SW et al (2014) Morbidity of repeat transsphenoidal surgery assessed in more than 1000 operations. J Neurosurg 121:67–74

Hypophysitis: a single-center case series.

The authors review their treatment experience and summarize clinical outcomes for patients with hypophysitis over the past 15 years...
2MB Sizes 6 Downloads 6 Views