Key words
hypophysitis - primary hypophysitis - hypopituitarism - immunoglobulin G4 - pituitary adenoma - apoplexy
Introduction
Primary hypophysitis (PH) is a rare autoimmune disease caused by inflammatory
infiltration of the pituitary gland, which can result in transient or permanent
endocrine dysfunction. The annual incidence was reported to be 1 in 7–9
million cases, and several studies of retrospective re-analysis of surgical
pituitary specimens confirmed the low overall incidence of 0.4% [1].
Histopathological examination is the gold standard for PH diagnosis, however,
’clinical’ diagnosis can be made by assessment of symptoms, hormonal
profile and magnetic resonance imaging (MRI) characteristics together. It is
reported that about one third of the cases are diagnosed in this manner [2]. Natural course of the disease has a very
large spectrum: either spontaneous resolution or permanent endocrine dysfunction are
possible [3]. Medical therapy (most commonly
glucocorticoids) has been increasingly used by clinicians, therefore the role for
surgery has been gradually decreasing in recent years [4].
Here we aim to contribute to current literature by analyzing clinical features of
primary hypophysitis cases over a time frame of 18 years diagnosed in a
single tertiary medical center which serves approximately 1600 patients with
pituitary disorders annually. We evaluated diagnostic challenges and our therapeutic
management experiences of both histopathologically and ‘clinically’
identified cases.
All procedures performed in the study were in accordance with the ethical standards
of the institutional research committee (GO 17–170) and with the Helsinki
declaration.
Subjects and Methods
Medical records of 20 PH patients admitted to Hacettepe University Hospital
between August 1999 and September 2017 were analyzed retrospectively.
Considering the low incidence of the disease, we included all cases regardless
of follow-up duration. Four out of 20 PH cases were diagnosed according to
clinical, hormonal and radiological findings whereas 16 had histopathological
confirmation. Secondary causes of hypophysitis were ruled out with appropriate
tests in all clinically diagnosed PH patients; granulomatous vasculitis,
sarcoidosis and Langerhans cell histiocytosis were excluded by related
diagnostic criteria [5] and PPD or
Quantiferon test was performed to rule out tuberculosis. None of the subjects
were receiving immune check point inhibitors or other targeted therapies
associated with hypophysitis. Anti-pituitary antibodies were regrettably not
checked. Due to retrospective nature of the study, some of the patients lacked
detailed evaluations for accompanying autoimmune diseases.
MRI and hormone test results were reviewed from patient files. Assessment of
pituitary hormonal disorders were made according to accessible data, dynamic
tests if present. Hormonal replacement therapies for hypocortisolism,
hypothyroidism, and hypogonadism were set correspondingly. Paraffin-embedded
pathology blocks were obtained from pathology archive for both re-examination
and IgG4 immunostaining. IgG4-related disease was evaluated according to related
criteria [6]: (1) Characteristic organ
involvement, (2) Elevated serum IgG4 levels, and (3) Histopathological findings
showing (i) marked lymphocyte and plasmocyte cell infiltration and fibrosis;
(ii) IgG4-positive plasma cell infiltration: IgG4 to IgG
ratio>40% and IgG4-positive plasma cells per high-powered field
(HPF)>10%. According to these criteria, cases meeting all three
are defined as definite, cases meeting first and second criteria as possible,
and cases meeting first and third criteria as probable IgG4-related disease.
We assessed treatment outcomes according to symptom recovery and follow-up MRI as
well as hormonal findings. Disease recurrence was defined as returning symptoms
and/or progression in follow-up MRI.
SPSS-23 program was used for statistical analysis. Normally distributed variables
were given as mean values with standard deviations while non-normally
distributed variables were given as median values with ranges. p-Values less
than 0.05 were considered statistically significant.
Results
Clinical presentation and endocrinological evaluation
Twenty patients (15 females, 5 males) with PH were included ([Table 1]). Mean age at diagnosis were
34.6±15.0 vs. 43.9±12.5 years in male and female patients,
respectively; p=0.19. Median time from symptom onset to diagnosis was 18
months (1–120). Since all data except pathological examination and age
was missing for one patient, 19 patients were included in further analysis. Most
common presenting symptoms were headache (n=12, 63.2%) and
symptoms associated with hypogonadism; such as decreased libido in male and
menstrual irregularities in female patients (n=9, 47.4%). Visual
disturbance (photophobia, hemianopsia, etc.) and polyuria/polydipsia
were also common ([Table 1]). Among six
patients examined for accompanying autoimmune disorders, only one had autoimmune
thyroiditis with normal thyroid functions.
Table 1 Characteristics, pre-treatment endocrine findings,
and magnetic resonance imaging (MRI) findings of primary
hypophysitis patients.
|
|
Total n
|
|
Mean age at diagnosis (years±SD)
|
41.5±13.4
|
20
|
|
Female sex (n, %)
|
15 (75%)
|
20
|
|
Median time to diagnosis (months)
|
18 (1–120)
|
19
|
|
Referral symptoms (n, %)
|
|
19
|
|
Headache
|
12 (63.2%)
|
|
|
Decreased libido/menstrual irregularities
|
9 (47.4%)
|
|
|
Visual disturbance (any)
|
7 (36.8%)
|
|
|
Polyuria/polydipsia
|
6 (31.6%)
|
|
|
Nausea
|
2 (10.5%)
|
|
|
Fatigue
|
2 (10.5%)
|
|
|
Hormonal disorders, any (n, %)
|
12 (63.1%)
|
19
|
|
Panhypopituitarism
|
7 (39%)
|
19
|
|
Hypogonadism
|
12 (66%)
|
18
|
|
Hypothyroidism
|
11 (61%)
|
18
|
|
Hypocortisolism
|
7 (39%)
|
18
|
|
Hyperprolactinemia
|
6 (31.5%)
|
19
|
|
Diabetes insipidus
|
5 (28%)
|
18
|
|
GH deficiency
|
4 (21%)
|
19
|
|
MRI contrast pattern
|
|
17
|
|
Homogenous contrast enhancement
|
4 (23.5)
|
|
|
Heterogeneous contrast enhancement
|
2 (11.7)
|
|
|
Peripheral contrast enhancement
|
2 (11.7)
|
|
|
Infundibular contrast enhancement only
|
1 (5.9)
|
|
|
Hypo-enhancement
|
2 (11.7)
|
|
|
Not reported*
|
6 (35.3)
|
|
|
Other MRI findings
|
|
|
|
Infundibular thickening
|
3 (17.6)
|
|
|
Loss of neurohypophyseal bright spot
|
4 (23.5)
|
|
|
Optic chiasm or nerve compression (any)
|
7 (41.2)
|
|
SD: Standard deviation, GH: Growth hormone. *
Pre-operative MRI was not performed in our center and/or did not
report contrast enhancement pattern.
At least one form of hormonal disorder was present in 12 out of 19 patients
(63.2%), hypogonadism (66.6%) and hypothyroidism (61.1%)
being the most common ([Table 1]). Seven
out of 19 patients had panhypopituitarism (36.8%). Hypocortisolism was
found in 38.9% of the patients (n=7). Posterior hypophyseal
involvement was present in five patients (27.8%): Two of these patients
did not have any concurrent anterior pituitary hormonal deficiencies but
elevated PRL. Overall, six patients had mildly elevated levels of prolactin
(PRL) (31.6%) ranging between 39.0–69.6 ng/ml
(Reference range: 5.18–26.5 ng/ml).
Only one patient was diagnosed subsequent to pregnancy. This 34-year-old patient
was admitted with symptoms of visual disturbance one month after child delivery.
MRI findings suggested pituitary macroadenoma measuring
14×16×24 mm which compressed optic chiasm, normal
adenohypophysis gland was not visible while neurohypophyseal bright spot was in
usual location. The patient was undergone urgent surgery under glucocorticoid
coverage, in which only decompression could be made due to very hard pituitary
tissue structure. Pathology report came back as lymphocytic hypophysitis. The
patient needed prednisolone and levothyroxine replacement postoperatively.
Radiological findings
MRI findings were obtainable for 17, but missing for remaining three patients,
and they are presented in [Table 1].
According to the radiologic score defined by Gutenberg et al. [7] (age ≦ 30 years: –1,
relation to pregnancy: –4, pituitary volume ≧
6 cm3:+2, medium/high gadolinium
enhancement: –1, asymmetry:+3, loss of posterior pituitary
bright spot: –2, enlarged stalk size: –5, mucosal
thickening:+2) which suggested scores between –13 to+2
(median –5) indicative of PH, only five of our patients were compatible
with autoimmune hypophysitis. One out of these 5 cases had a score of –4
due to related pregnancy (as described above) but did not meet any of the other
criteria. Remaining four patients had scores ranging –5 to –8,
all of whom were diagnosed ‘clinically’ as primary hypophysitis:
Three cases were given glucocorticoid treatment and one was observed without
intervention (described below).
Notably, ‘possible hypophysitis’ was more frequently reported by
radiologists in the recent years compared to early 2000s. It is also noteworthy
that radiological diagnosis was more accurate in patients whose clinical
information was provided as ‘panhypopituitarism’ and preliminary
diagnosis as ‘hypophysitis’ by clinicians. Loss of posterior
pituitary bright spot also appeared to be important in radiologists’
decision making. Three out of four MRIs showing loss of neurohypophyseal bright
spot were reported as possible hypophysitis.
Pathological examination findings
Pathological assessments of all patients were made by the same experienced
neuropathologist (F.S.). Lymphocytic hypophysitis was the histopathological
subtype in eight out of 16 patients (50%). Six cases were diagnosed as
granulomatous hypophysitis (37.5%) while one patient was reported as
xantho-granulomatous and one as lymphogranulomatous hypophysitis.
Breakdown of the reticulin fiber network was reported in seven cases. Fibrosis,
granuloma formations and multinuclear giant cells were detected in 7
(43.7%), 7 (43.7%) and 8 (50%) patients; respectively.
Neither histopathological subtype of PH nor existence of fibrosis had any
association with hormonal deficiencies. Pathological findings indicating the
presence of accompanying Rathke’s cleft cysts were reported in three
lymphocytic hypophysitis cases (18.7%); none of which was reported in
preoperative MRI.
We have examined the pathology specimens prepared from paraffin embedded blocks
for IgG and IgG4 immunostaining. Out of ten specimens available for this
process, none fulfilled the criteria for IgG4-related hypophysitis. However, two
patients (31-year-old male and 40-year-old female) had>10 IgG4 positive
plasma cells per one high powered area. Male patient with panhypopituitarism and
DI was classified as lymphocytic hypophysitis, pathological findings included
fibrosis and marked stromal infiltration with lymphocytes. Histopathology of the
female patient with DI and mild hyperprolactinemia was compatible with both
granulomatous and lymphocytic hypophysitis; fibrosis was present. Since serum
IgG4 levels were regrettably not available in these two cases, these patients
may be diagnosed as probable IgG4-related hypophysitis.
Treatment
Glucocorticoid treatment was given to four patients, as initial therapy in three
cases who were diagnosed ‘clinically’. Patients treated with
glucocorticoid are presented in Supplementary
Table 1S.
Case 1
Only male patient in this group received methylprednisolone at an initial
dose of 60 mg/day intravenous for two weeks, followed by
60 mg/d orally; the dose was tapered down in a monthly basis
to a final dose of 8 mg/day and was stopped after 6 months.
Radiologic improvement was detected after 6 months; however, hormonal
situation did not improve. Serum IgG4 levels were found to be higher than
normal (IgG4: 1100 mg/l, N: 39–864; IgG:
1480 mg/dl, N: 751–1560). No other organ or system
involvement associated with IgG4-related disease has emerged at diagnosis or
during follow-up. Therefore, this case was considered as possible IgG4
related hypophysitis. Progression was detected on MRI after 72 months of
follow up regarding infundibular thickening (6 mm), although no
further immunosuppressive treatment was planned due to lack of clinical
significance. The patient continued hormone replacement therapy for
panhypopituitarism and DI.
Case 2
This female patient was given methylprednisolone 60 mg/day
initially, and the dose was tapered down in 3 months. She rapidly became
Cushingoid in appearance with concurrent elevated blood glucose and
rhabdomyolysis. Although the infundibular thickening shrank to 5 mm
at the end of 3 months, panhypopituitarism and DI remained. Due to lack of
improvement and Cushingoid adverse effects, cessation of treatment was
decided while hormone replacement therapy was sustained.
Case 3
The other female patient was treated with pulse methylprednisolone 1
g/day for three days and 60 mg/day deflazacort as
maintenance therapy, which was tapered down to as low as
6 mg/day. Desmopressin was given for accompanying DI. After
one year of treatment, MRI findings of the patient were completely normal,
and both deflazacort and desmopressin therapies were stopped.
Case 4
Glucocorticoid was given as secondary treatment in this female patient
presenting with decreased visual acuity and diplopia. She was given to
surgery, which provided decompression of optic chiasm. Pathology report came
back as granulomatous hypophysitis. Two months after surgery she was
admitted to hospital with headache and relapsed decline in visual acuity of
the right eye and was given pulse glucocorticoid treatment which rapidly
improved symptoms. However, panhypopituitarism remained.
A total number of 16 cases underwent surgery. Preliminary diagnosis was
non-functioning pituitary macroadenoma in nine (56.2%);
macroadenoma, Rathke’s cleft cyst or other kind of sellar mass
(e. g. craniopharyngioma) in five (31.2%) patients (data not
available in one). Surgery related minor complications such as mild
bleeding, cerebrospinal fluid leak, arachnoid membrane rupture was seen in
three cases; all were repairable.
One female patient among clinically diagnosed cases presented with headache
and diagnosed as PH according to symptoms and MRI findings (expanded
pituitary gland with homogenous contrast enhancement and thickened
infundibulum, radiologic score of –5 according to radiologic
criteria by Gutenberg et al. [7] at
the age of 50. Hormonal evaluation was completely normal and she was not
given any treatment. Her medical history revealed several previous
examinations for headache in a time range of 84 months and MRI findings were
stable comparing to one year earlier.
Patient Outcomes
Median follow-up period was 12 months (range: 0–132 months). Among 15
patients who underwent operation radiological improvement and symptom relief
occurred in all, yet none showed hormonal improvement (data not available in
one). New hormonal deficiencies emerged in 4 surgical patients:
Panhypopituitarism emerged in one of the two patients with completely normal
pre-operative endocrine evaluation, and DI in the other. The other two patients
with partial endocrine deficiencies needed additional hormone replacements
post-operatively ([Table 2], Supplemantary Table 1S). Radiologic and
symptomatic relapse arose in 1 patient two months after surgery who was treated
effectively with glucocorticoid.
Table 2 Treatment, outcomes, and follow-up of primary
hypophysitis patients.
|
|
Total n
|
|
Preliminary diagnosis (n, %)
|
|
20
|
|
Hypophysitis
|
4 (20%)
|
|
|
Other
|
16 (80%)
|
|
|
Preoperative endocrinology consultation (n,
%)
|
|
19
|
|
Yes
|
8 (42.1%)
|
|
|
No
|
11 (57.9%)
|
|
|
Surgery (n, %)
|
|
20
|
|
Yes
|
16 (80%)
|
|
|
No
|
4 (20%)
|
|
|
Surgical complications, any (n, %)
|
|
15
|
|
Yes
|
3 (20%)
|
|
|
No
|
12 (80%)
|
|
|
Post-operatively emerging hormonal disorders, any (n,
%)
|
4 (26.6%)
|
15
|
|
Improvement of hormonal disorders in cases treated by
steroid (n, %)
|
1 (20%)
|
4
|
|
Median follow-up (months, range)
|
12 (0–132)
|
|
Three ‘clinically’ diagnosed patients received initial
glucocorticoid therapy but in diverse protocols. Pituitary enlargement regressed
in all, as well as symptoms related to mass effect. The only patient to benefit
from glucocorticoid treatment regarding endocrine outcomes was initially given
pulse glucocorticoid and long-term low dose glucocorticoid as maintenance
therapy. Radiologic recurrence was seen in one patient after 72 months without
symptoms (Supplemantary Table 1S).
Discussion
In this retrospective study, we evaluated characteristics, presenting signs and
symptoms, management modalities and treatment outcomes of 20 PH patients diagnosed
in our center. In addition, we examined the pathology specimens for IgG and IgG4
staining. Main findings of our cohort are presented in Table 3 comparably with previous
single-center case series.
Table 3 Characteristics, treatment modalities and treatment
outcomes of PH patients comparably with previous single-center case
series.
|
Leung, GK. 2004 [8]
|
Park, SM. 2014 [11]
|
Khare, S. 2015 [13]
|
Imber, BS. 2015 [12]
|
Chiloiro, S. 2017 [15]
|
Wang, S. 2017 [9]
|
Angelousi, A. 2018 [16]
|
Korkmaz, OP. 2019 [14]
|
Current case series
|
|
Number of patients
|
16
|
22
|
24
|
21
|
21
|
50
|
22
|
17
|
20
|
|
Mean age at diagnosis (years)
|
47.6
|
48
|
31.5
|
37
|
40
|
37
|
42
|
31
|
41.5
|
|
Female to male Ratio
|
1
|
3.4
|
7
|
1.6
|
4.2
|
1.9
|
3.4
|
1.4
|
3
|
|
Pregnancy-related cases (n)
|
3
|
|
1
|
4
|
0
|
8
|
0
|
0
|
1
|
|
Most common presenting symptom/s (%)
|
Headache 75%
Lethargy 62%
Gonadal
dysfunction 50%
|
Polyuria/
polydipsia 82%
Headache
27%
|
Headache 83%
|
Polyuria/
polydipsia 52%
Headache
57%
Visual disturbance 52%
Gonadal
dysfunction 48%
|
Polyuria/
polydipsia 47%
Headache
24%
Gonadal dysfunction 47%
|
Headache and visual disturbance
|
Headache 59%
Gonadal dysfunction
59%
Visual disturbance 32%
|
Headache 53%
Polyuria/
Polydipsia
47%
Fatigue 41%
|
Headache 63%
Gonadal dysfunction
66%
Visual disturbance
37%
Polyuria/
Polydipsia
32%
|
|
Time to diagnosis (months)
|
15.8 (1–60)
|
|
10
|
3.5 (0.5–60)
|
|
4 (1–60)
|
|
12 (1–96)
|
18 (1–120)
|
|
Hormonal deficiencies (%)
|
|
Adrenal
|
58
|
36
|
75
|
|
(Any kind of anterior hormonal deficiency in 81%)
|
26
|
(Any kind of anterior hormonal deficiency in 77%)
|
59
|
39
|
|
Thyroid
|
50
|
36
|
58.3
|
|
|
38
|
|
53
|
61
|
|
Gonadal
|
91.6
|
32
|
50
|
|
|
60
|
|
47
|
66
|
|
GH
|
43
|
23
|
|
|
|
22
|
|
6
|
21
|
|
DI
|
31
|
82
|
16.7
|
14
|
47.6
|
30.5
|
32
|
47
|
28
|
|
Hyperprolactinemia
|
37.5
|
23
|
41.6
|
48
|
42.8
|
48
|
|
41
|
31
|
|
Panhypopituitarism
|
|
|
|
57
|
4
|
10
|
23
|
23.5
|
39
|
|
Co-existing autoimmune diseases (n)
|
5
|
2
|
|
5
|
13
|
3
|
8
|
2
|
2
|
|
Histopathological type %
|
|
LH
|
77
|
73
|
100 (?)
|
76
|
|
100
|
86
|
75
|
50
|
|
GH
|
23
|
18
|
|
14
|
|
|
|
|
37.5
|
|
XGH
|
|
9
|
|
|
|
|
6
|
12.5
|
6
|
|
Mixed
|
|
|
|
|
|
|
|
12.5
|
6
|
|
IgG-4 related
|
1
|
|
|
5
|
|
|
6
|
|
|
|
Clinically diagnosed cases (n/%)
|
3
|
11
|
15
|
2
|
19
|
22
|
8
|
9
|
4
|
|
Surgery (n)
|
13
|
5
|
5
|
19
|
2
|
15
|
9
|
7
|
16
|
|
Steroid treatment (n)
|
8
|
5
|
4
|
All
|
All
|
26
|
8*
|
5**
|
4
|
|
Observation (n)
|
|
12
|
11
|
None
|
None
|
9
|
5
|
10
|
1
|
|
Duration of follow-up (months)
|
30 (2–107)
|
57 (7–138)
|
18
|
28.9
|
12
|
≥6
|
60
|
24 (6–84)
|
12 (0–132)
|
|
Relief of symptoms after treatment %
|
|
|
|
|
|
|
|
|
|
|
Surgery
|
100
|
100
|
100
|
|
72.2% in all group
|
100
|
|
|
100
|
|
Medical
|
60
|
100
|
|
|
|
100
|
|
|
100
|
|
Observation
|
|
|
|
|
|
22.2
|
|
|
|
|
Radiological improvement in non-surgical patients
|
NA
|
Pituitary mass reduction in 4 steroid treated patients
(80%)
|
Pituitary mass regressed in all, both steroid and observational
groups
|
Mass reduction in 8 (50%)
|
Radiologic improvement in 13 (72.2%)
|
Mass reduction in all steroid treated patients
|
Normal imaging findings in 27% of treatment group
|
Improvement in 40% and 30% of steroid and
observation groups, respectively
|
Pituitary mass regressed in all steroid treated patients
|
|
|
No change in observation group
|
Persistent stalk thickening in 13 (68.4%)
|
Stable or worse in 50%
|
(All group evaluated, regardless of surgery)
|
Spontaneous mass reduction in 22.2% of observation
group
|
|
|
|
|
|
|
Near normal MRI findings in 2 (10.5%)
|
(All group evaluated, regardless of surgery)
|
|
|
|
|
|
|
|
|
Empty sella with thinning of stalk in 4 (21%)
|
|
|
|
|
|
|
|
Endocrine improvement, %
|
|
|
|
19% in all group
|
85.7% in all group
|
|
|
|
|
|
Surgery
|
|
|
25
|
|
|
0
|
16.6
|
14.3
|
0
|
|
Medical
|
0
|
40
|
100
|
|
|
40.9
|
44.4
|
20
|
20
|
|
Observation
|
|
17
|
100
|
|
|
22
|
5.5
|
10
|
NA
|
|
Recurrence (n)
|
2
|
3
|
|
6
|
|
46% in medical, 11% in surgical group
|
4
|
1
|
2
|
* Five out of 8 patients received other immunosuppressive
agents. ** None received steroid as initial
treatment. NA: Not available.
Most common presenting symptom related to PH has indisputably been headache, as in
our series. Visual disturbances according to mass effect are also common, while
other symptoms vary depending mainly on existing hormonal deficiencies. Hypogonadism
and hypothyroidism were the most common endocrine deficiencies in our series,
similar to some previous case series [8]
[9]; but more common than that reported in a
recent PH case review (66% and 61% vs. 55% and 52%,
respectively) [10]. Although hypocortisolism
has generally been defined as the most common endocrine disorder in PH (60%)
[10], some case series reported lower
incidence as in our cohort [9]
[11]. DI, growth hormone deficiency and mild
hyperprolactinemia were also less common in our series than general [10]. Median time from onset of symptoms to
diagnosis ranged from 3.5 to 16 months in previous case series [8]
[9]
[12]
[13]
[14]
[15], while it was 18 months in ours (Table 3).
Formerly higher female to male ratios in PH has recently been given as 2.5:1, which
was 3 in our case series. Mean age at the time of diagnosis (41.5 years) was also
consistent with literature [10]. In our
series, female cases were older than males at the time of diagnosis; which is a
finding mostly incompatible with previous publications [2]
[10]
[16]. Although female patients were also older
in the case series reported by Khare et al., mean age at diagnosis of overall group
was younger than ours (31.5 years) [13]. Even
though PH is highly associated with pregnancy [2], only one of our cases was pregnancy-related.
Other autoimmune diseases co-exist with PH up to 18% of cases, autoimmune
thyroiditis being the most common [2].
However, we could only define one case with concomitant autoimmune thyroiditis.
Lymphocytic hypophysitis is the most common histological subtype of PH, reported in
68% of cases. Granulomatous hypophysitis is the second most common variant
(20%) while IgG4-related and xanthomatous hypophysitis are reported in 4 and
3% of patients, respectively. Mixed pathological characteristics were
reported in 4% [10]. Our findings were
consistent with these data. Notably, concurrent Rathke’s cleft cysts were
defined in three lymphocytic hypophysitis cases in our series. None of these cases
had cystic appearance on MRI, but recognized by pathological examination. Simply
explained by the triggered inflammatory reaction occurring secondary to cyst
rupture, this co-existence was previously reported in other cases of both
lymphocytic and xantho-granulomatous hypophysitis [17]
[18]. Adrenal insufficiency was reported to be
common in patients with Rathke’s cleft cysts [19]; however, only one out of three of our
cases had adrenal hormone deficiency. Although we performed IgG4 immunostaining in
available pathology specimens, it was not achievable to fully evaluate patients
regarding IgG4-related hypophysitis due to the retrospective nature of our
study.
Previous studies suggested that histopathological confirmation is not always
mandatory to diagnose PH, especially if no alternative diagnoses are considered
which could change treatment [20]. Cases of
‘clinically diagnosed’ PH have been reported more frequently in the
literature in recent years, possibly due to expanded awareness. However, the
majority of patients are still diagnosed after surgical procedures [10], as in our series. The most common
indications for surgery in PH have been symptoms related to mass effect, such as
headache and visual disturbances; but also, diagnostic difficulties. Our study
exposed that radiological diagnosis was more accurate when clinicians provide more
clinical knowledge for radiologists. Also, loss of posterior pituitary bright spot
and thickened infundibulum were associated with more precise diagnosis of PH, like
previously been identified as indicative radiological findings for PH [21]. Nevertheless, yet the best imaging
technique available, pituitary MRI alone may not be enough to distinguish PH from
pituitary adenomas [22].
Treatment of PH targets both hormonal deficiencies and symptoms associated with mass
effect. Conservative management has been suggested in asymptomatic patients [2], which was appropriate for one case in our
series. While glucocorticoids are the first choice of medical treatment in PH,
surgery should be reserved for severe cases having neuro-ophthalmologic findings or
cases who are non-responsive to medical treatment [22]. Overall, mass reducing treatments, either surgical or medical, have
been reported to improve symptoms and radiological findings in
70–80% of patients but endocrine deficiencies in only 18%
[2]
[10]
[23], compatible with our findings. Used in
variable doses and duration, Honegger et al. reported radiologic and hormonal
improvement in 65% and 18% of cases with glucocorticoid treatment,
respectively [23]. All glucocorticoid treated
patients in our series showed radiologic response. Although some previous case
series reported better hormonal response rates to glucocorticoid therapy than ours
[9]
[11]
[13]
[16], our rates were similar with others [12]
[14]
[23] and better than one [8] (Table
3). Out of 16 surgical patients in our series, seven had optic chiasm
compression. Although 14 to 25% recovery of endocrine function in surgical
cases was reported in some previous reports [8]
[13]
[16], none of our patients showed improvement
in this manner. There was only one case relapsed after surgery, though Honegger et
al. reported 25% recurrence in surgically treated PH cases [23]. None of our subjects received third-line
treatments such as other immunosuppressive agents or radiotherapy.
Our study has some limitations; first of which is the rather high surgery rate. This
is mainly a consequence of high occurrence of neuro-ophthalmological involvement,
but also of diagnostic difficulties. Another limitation is that glucocorticoid
treated patients received different treatment regimes. However, we think that the
distinguished regimes may be explicable by the retrospective nature of the
study.
In conclusion, we have reported a series of PH including 20 cases mostly confirmed by
histopathology. We have reviewed clinical characteristics and presentations,
radiological and endocrine findings as well as disease management and outcomes of
the patients comparatively with previous reports. Although few patients received
glucocorticoid therapy, we have discussed each case in detail regarding treatment
modalities and therapy responsiveness.
