Pituitary apoplexy – A single center retrospective study from the neurosurgical perspective and review of the literature

Clinical Neurology and Neurosurgery 163 (2017) 39–45

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro

Full length article

Pituitary apoplexy – A single center retrospective study from the neurosurgical perspective and review of the literature

MARK

Agnieszka Grzywotza, Bernadette Kleista, Lars C. Möllerb, Volkmar H. Hansc, Sophia Göricked, ⁎ Ulrich Surea, Oliver Müllera,1, Ilonka Kreitschmann-Andermahra, ,1 a

Department of Neurosurgery, University Hospital Essen, University of Duisburg-Essen, Hufelandstr. 55, 45147 Essen, Germany Department of Endocrinology and Metabolism and Division of Laboratory Research, University Hospital Essen, University of Duisburg-Essen, Hufelandstr. 55, 45147 Essen, Germany c Institute of Neuropathology, University Hospital Essen, University of Duisburg-Essen, Hufelandstr. 55, 45147 Essen, Germany d Institute of Radiology and Neuroradiology, University Hospital Essen, University of Duisburg-Essen, Hufelandstr. 55, 45147 Essen, Germany b

A R T I C L E I N F O

A B S T R A C T

Keywords: Pituitary apoplexy Hypocortisolism Addisonian crisis Hyponatremia Thunderclap headache

Objective: Thunderclap headache and visual disturbances are typical clinical features of pituitary apoplexy (PA). Because of the acute symptomatology, many patients are referred to a neurosurgical department without prior endocrinological assessment. It is the aim of the present study to analyze initial presenting symptoms, outcome and associated endocrine disturbances in a cohort of patients with pituitary apoplexy primarily seen by neurosurgeons. Patients and methods: Retrospective single-center study in a neurosurgical department. Patients’ records were reviewed for clinical, neuropathological and endocrinological findings. The diagnosis of PA was based on clinical, imaging and histological findings. Results: A total of 60 patients were studied. They were referred most often by neurologists (n = 16), and family physicians (n = 12). Only 2 patients received an endocrinological work-up prior to admission. The most frequently documented presenting symptoms were headache (n = 54), visual field defects (n = 13), reduction of visual acuity (n = 17) and/or diplopia (n = 19). An endocrinological history had rarely been taken and hormone blood tests were oftentimes incomplete or not ordered. At 3-month follow-up 18/44 patients had complete anterior hypopituitarism. At 12 months, 21/60 patients were lost to endocrinological follow-up. Conclusions: The classic neurological symptoms of PA were well documented in our patient cohort, whereas endocrinological symptoms, especially those indicative of pituitary dysfunction were underreported. Neurologists and neurosurgeons need to be aware of the endocrinological sequelae of pituitary apoplexy in order to avoid potentially lethal complications. Patients should be counselled to adhere to long-term endocrinological and neurosurgical follow-up.

1. Introduction In 1912, Harvey Cushing wrote that “there are few subjects in medicine which promise a wider overlap upon the fields of many special workers than this one of hypophyseal disease” [1]. This résumé holds still true in modern medicine, especially for the entity of pituitary apoplexy (PA). The first description of the disease was published by the neurologist Pearce Bailey [2], and the neurosurgeon, Milton Brougham, first coined the term pituitary “apoplexy” in 1950, thus stressing the acute and severe clinical presentation of this event [3]. By definition,

PA is a hemorrhagic or ischemic infarction of a preexisting pituitary adenoma. It is characterized by a sudden severe headache (“thunderclap headache”), visual disturbances and diplopia, and, possibly, a deterioration of consciousness [4]. In addition, nausea and vomiting, hypotension and electrolyte disturbances such as hyponatremia are frequently present [5]. These symptoms can be related to anterior pituitary hormone deficiency, especially adrenocorticotroph hormone (ACTH) deficiency, which may evolve untreated to a life-threatening medical emergency: the Addisonian crisis. Apart from the “classic” PA, clinically less impressive, subacute variants exist, which may, in the

⁎

Corresponding author. E-mail addresses: [email protected] (A. Grzywotz), [email protected] (B. Kleist), [email protected] (L.C. Möller), [email protected] (V.H. Hans), [email protected] (S. Göricke), [email protected] (U. Sure), [email protected] (O. Müller), [email protected] (I. Kreitschmann-Andermahr). 1 Equal contribution. http://dx.doi.org/10.1016/j.clineuro.2017.10.006 Received 12 July 2017; Received in revised form 21 September 2017; Accepted 9 October 2017 Available online 10 October 2017 0303-8467/ © 2017 Elsevier B.V. All rights reserved.

Clinical Neurology and Neurosurgery 163 (2017) 39–45

A. Grzywotz et al.

case of mild clinical symptoms be diagnosed as “chance findings” either by means of magnetic resonance imaging (MRI), intraoperatively or histopathologically [6,7]. In our clinical experience, the severe headache and the neurological deficits of PA often dominate the clinical picture, while the gastrointestinal symptomatology associated with Addisonian crisis and even the electrolyte disturbances remain unnoticed or are attributed to gastrointestinal infection. Thus, patients are frequently triaged to neurological or neurosurgical departments under the suspicion of stroke, intracranial or subarachnoid hemorrhage, meningitis or other neurological diagnoses [8,9]. Once intracranial imaging confirms the hemorrhagic or infarcted pituitary mass lesion, neurosurgery is frequently performed to alleviate compression of the anatomical structures in the vicinity of the tumor such as the optic chiasm, the cranial nerves in the cavernous sinus or the residual pituitary gland itself. The coincidence of an acute neurological symptom and a hemorrhagic mass lesion in the cranial imaging modality might be sufficient for the endocrinological sequelae of PA to be overlooked, especially by endocrinologically less experienced treating physicians. For example, hyponatremia in the context of PA is frequently an indicator of severe secondary hypocortisolism as cortisol is the strongest inhibitor of antidiuretic hormone (ADH) and loss of ADH inhibition will result in the syndrome of inappropriate ADH secretion (SIADH) [10]. For the present study, we hypothesized that for patients with PA admitted to neurosurgical departments, the neurological symptoms of PA would be well-documented while endocrinological symptoms and diagnostics would be, in comparison to the literature, underreported and less requested. To evaluate this hypothesis, we assessed our own series of patients with PA treated for initial symptoms as documented in medical chart records, referring departments, laboratory parameters, clinical course, histopathology, risk factors for PA and follow-up examinations.

reconstructions from which the tumor size was approximated. Pituitary adenomas were classified as microadenomas (< 1 cm), macroadenomas (> 1 cm) and giant adenomas (> 4 cm). Histopathological diagnosis of the adenomas was assessed if available. As part of the routine work-up, all patients with pituitary pathologies received endocrinological consultations postoperatively at the latest, and were seen again for a routine follow-up at three and twelve months in the endocrinology department of our university. The follow-up data were used to characterize deficient hormone axes or persistent presence of hormonal overproduction. If available, long-term endocrinological data are also reported. The study protocol was reviewed and approved by the local ethics committee of the University Duisburg-Essen (15-6484-BO). 2.3. Statistical analysis SPSS 23.0 (Statistical Package for the Social Sciences, IBM; Armonk/USA) was used for statistical analysis of the data. Metric variables were displayed as mean and standard deviation, categorical variables as frequency and valid percent. Data were tested for normal distribution, using the Kolmogorov-Smirnov test, in addition to Q-Qplots and histograms for every metric variable. Levenés test was conducted to test for homogeneity of variance. Since normal distribution and homogeneity of variance could be assumed, studentś t-test for unpaired variables was used to test for differences in clinical variables between groups in regard to sex, intake of anticoagulants, type of headache, histology and visual disturbances. For categorical variables the Chi-square test or Fisheŕs exact test, if expected frequencies were below five, was conducted. A p value of ≤0.05 was considered significant. 3. Results 3.1. Patient characteristics, referrals and tumor size

2. Patients and methods There were 29 male and 31 female patients (mean age 56.3; standard deviation (SD) ± 20.03 years, range 16.7–90.0 y) with no statistical difference between men and women (t-test n.s.). The referral pathway could be traced back in 51 patients. 32 patients were referred to our neurosurgical department by other hospitals. Of these, 16 (50%) were referred by neurological departments, 6 (18.8%), by departments of internal medicine, and 10 (31.2%) by various other departments (i.e. intensive care units, peripheral neurosurgical departments, endocrinological departments). 19 patients were referred to our department by outpatient medical services. 12 (63.2%) came from family physicians or internists and 2 (10.5%) from neurologists. Five (26.3%) patients were referred by various medical specialists (i.e. gynecologist, ophthalmologist). Patients who presented with nausea and vomiting were significantly more often referred by hospital departments than outpatient medical services (χ2 (1) = 8.51; p = 0.004 and Phi = −0.41). Men were more often referred by hospital departments, whereas women were referred predominantly by outpatient medical services (χ2(1) = 6.25; p = 0.020 and Phi = −0.35). MRI/CT results giving evidence of hemorrhagic/regressively transformed pituitary adenoma were available for tumor size evaluation in 54 patients. Of those, 50 patients had a macroadenoma, one a microadenoma and 3 patients harbored a giant adenoma. The mean tumor size in coronal, sagittal and axial planes was 22.7 ± 7.00 mm, 16.8 ± 5.79 mm, and 21.1 ± 8.61 mm, respectively. Women had significantly smaller adenomas then men in coronal (20.4 ± 7.46 mm vs 24.7 ± 5.93 mm; p = 0.024) and sagittal measurements (14.8 ± 5.45 mm vs. 18.6 ± 5.58 mm; p = 0.018). 53 patients underwent surgery. For 35 patients it was possible to calculate the exact latency between the acute PA and the neurosurgical intervention with 19.7 days (SD ± 21.6 d; range: 0–97 d). The exact timepoint of PA was known for four further patients, who were initially treated conservatively for different medical reasons, with a delayed surgical

2.1. Patient population Inclusion criterion was the diagnosis of hemorrhagic pituitary apoplexy made on the basis of clinical symptomatology and/or a hemorrhagic pituitary tumor on intracranial imaging (MRI or CT), in line with the criteria for PA, set up by Randeva et al. [4]. Patients were identified from the electronic clinical information system using the keyword “pituitary apoplexý́, and from our outpatient clinic. A total of 60 patients met the criteria for PA and were subsequently included in the study. 2.2. Data collection and clinical variables Electronic and paper based chart records were reviewed for clinical, neuropathological and endocrinological findings. If available, hormone levels of the hypothalamo-pituitary hormone axes on admission were used to assess pituitary function, according to the criteria described by Schneider et al. [11]. The diagnosis of corticotroph insufficiency is usually based on endocrinological function tests, which were not available on admission in any of the investigated patients. Thus the corticotroph hormone axis was evaluated as intact at that timepoint in case of a basal cortisol level > 450 nmol/l and not characterized at a value below this level [12]. Low basal cortisol levels in conjunction with hyponatremia were separately noted. Hormone excess (i.e. in the case of prolactinoma, Cushing’s disease and acromegaly) was characterized by the clinical presentation in conjunction with available hormone levels. Endocrinological assessment at follow-up investigations was performed according to established standards [13–15]. The tumor size was measured in the axial, sagittal and coronal plane of the MRI scans. Patients with contraindications for an MRI (e.g. pacemaker) had a cranial CT scan with coronal and sagittal 40

Clinical Neurology and Neurosurgery 163 (2017) 39–45

A. Grzywotz et al.

Table 1 Initial presenting clinical signs of the patients. Data are presented as frequency (N) and valid percent (%).

Table 2 Anterior pituitary hypopituitarism of the patients at admission and after 3 and 12 months after the apoplexy. Data are presented as frequency (N) and valid percent (%).

Initial symptoms

Number of patients

Insufficiencies

Initial

3 months

12 months

Headache Thunderclap headache Other headache Dizziness Nausea & vomiting Decreased visual acuity Visual field defect Ocular palsy Altered consciousness

54 (91.6) 27 (45.8) 27 (45.8) 6 (10.2) 17 (28.8) 17 (28.8) 13 (22.0) 19 (32.2) 9 (15.3)

Gonadotropic insufficiency

31 (68.9) N = 45 22 (55.0) N = 40 20 (42.6) N = 47 16 (32.0) N = 50 10 (20.4) N = 49

27 (58.7)a N = 46 26 (56.5) N = 46 25 (56.8) N = 44 24 (53.3) N = 45 18 (39.1) N = 46

21 (53.8) N = 39 18 (47.4) N = 38 22 (56.4) N = 39 24 (61.5) N = 39 13 (33.3) N = 39

Corticotropic insufficiency Thyreotropic insufficiency Somatotropic insufficiency Global insufficiencya

a For patients with global insufficiency of the anterior pituitary lobe, the single insufficiencies of the different axes are also listed for the partial insufficiencies.

intervention. The latency from PA to surgery in these four patients was up to 550 days.

context of PA can be a first sign of acute hypocortisolism. Therefore, the initial sodium values (available in 56 (93.3%) patients) were evaluated: 12 (21.4%) patients had hyponatremia, defined as Na+ −levels < 135 mmol/l. Of those, 7 patients presented with severe hyponatremia, defined as sodium < 130 mmol/l. Interestingly, all of the patients with severe hyponatremia were female. Dizziness was noted as a symptom in the chart records in two hypoatremic patients, nausea and vomiting in one, and alterations of consciousness in one further patient. Basal cortisol levels were available only in four patients with hyponatremia. All of them were abnormally low (8, 172, 18, 15 nmol/l, respectively), indicating a possible causal relationship between hypocortisolism and hyponatremia in these patients. However, none of the patients with documented Addisonian crisis presented with hyponatremia, probably because all of them already received hydrocortisone replacement before admission to our department.

3.2. Presenting symptoms and signs The initial presenting clinical signs are summarized in Table 1. Headache was the leading symptom in more than 90% of the patients (n = 54), occurring as acute thunderclap headache in almost half of them (n = 27, 45.8%). 8.5% (n = 5) patients did not have headache, and in one the chart record was inconclusive. In 25 of the 27 patients with thunderclap headache, MRI or CCT gave evidence of an acute intraparenchymal hemorrhage in the pituitary mass. In 2 patients, the imaging diagnostics were missing, but discharge records or surgical records reported a hemorrhagic hypophyseal mass lesion. Neuroophthalmological deficits (visual field deficits, impaired vision, diplopia) constituted the second most frequent documented symptom group, while disturbances of consciousness, reaching from short syncopal episodes to coma, were documented in 15.3% (n = 9) of the patients. Expectedly, patients with any kind of ophthalmologic deficit had significantly larger tumors in axial, coronal and sagittal planes (23.6 × 25 × 18.4 mm vs. 18.2 × 20.1 × 14.8 mm; p < 0.05). They were also significantly older than the other patients in the study group (63.8 ± 16.60 vs. 44.1 ± 19.54 years, p = 0.001). When comparing the patients with thunderclap headache to the patients with any other type of headache (n = 27) there were no differences between groups in regard to the variables age, sex, tumor size, therapy with anticoagulants, sodium < 130 mmol/l pre-surgery and tumor histology (conclusive histology/necrosis).

3.4. Possible predisposing factors for PA The predisposing factors for PA found in our study group are listed in Table 3. In our study population there was no PA after endocrinological function testing, after cardiac surgery or after radiotherapy which are all considered classic risk factors for PA [6,9,16,17]. However, 28.6% of our patients were on anticoagulant medication and 5.4% patients took dopamine agonists, due to giant prolactinomas that did not shrink under medical therapies. Patients on anticoagulant medication were significantly older (76.1 ± 8.89 vs. 49.1 ± 16.68; p < 0.0001) than those patients in the study group who did not take anticoagulants.

3.3. Hormone levels, Addisonian crisis and hyponatremia Upon admission, in 38 patients complete basal hormone levels of the hypothalamo-pituitary axis (TSH, fT3, fT4, LH, FSH, testosterone (males) or estradiol (females), GH, IGF-1, ACTH and cortisol) had been assessed. At least an incomplete hormone assessment had been done in a further 17 patients. The prolactin level was missing in only 6 patients. 31 (53.4%) of 58 of our patients were seen by our consultant endocrinology service shortly after admission. Of these 31 patients (100%), 10 patients (32.3%) were diagnosed with complete anterior pituitary insufficiency. Table 2 shows the frequency of anterior hypopituitarism at the time of admission, at three-month follow-up and at the latest available follow-up. Symptoms indicative of Addisonian crisis (nausea, diarrhea, vomiting, abdominal pain, hyponatremia) in conjunction with the correct diagnosis were documented in the chart records of three patients. One further patient with known primary adrenal insufficiency also developed Addisonian crisis and PA in the wake of a complicated dental procedure. Most probably, in this case the acute hypocortisolism had been caused by insufficient hydrocortisone replacement in the stressful situation. As spelled out in the introduction of this paper, hyponatremia in the

3.5. Histology Histological reports were available in 50 of the investigated patients. In 20 cases, only necrosis, unsuitable for further diagnostic subtyping by immunohistochemistry, was seen. In 28 patients Table 3 Predisposing factors of the pituitary apoplexy in the study group.

41

Predisposing factors

Number of patients (N)

Major surgery Acute severe infection Therapy with dopamine agonists Anticoagulation

4 4 4 16

Other factors Pregnancy Primary adrenal insufficiency with Addisonian crisis Dialysis Hepatitis Type C Severe anaemia

1 1 1 2 1

Clinical Neurology and Neurosurgery 163 (2017) 39–45

A. Grzywotz et al.

4. Discussion

immunostaining could be performed and yielded 13 (26%) gonadotropinomas, 5 (10%) prolactinomas, 4 (8%) ACTH-expressing tumors, 4 (8%) null-cell adenomas and 2 (4%) plurihormonal tumors. Furthermore, one hemorrhagic Rathke’s cleft cyst and one hemorrhagic hypophysitis were diagnosed. There were no significant differences in clinical parameters (age, sex, tumor size, headache, therapy with anticoagulants, sodium < 130 mmol/l pre-surgery) between those patients with conclusive histology vs. those with only necrosis.

In the present study, we report the clinical presentation and clinical course, biochemical, radiological and histological findings of 60 patients with PA, treated at a tertiary neurosurgical center. The overall knowledge about the clinical characteristics and long-term outcome of patients with PA is still limited to date. A PubMed search with the keywords ‘pituitary apoplexy’; combined with “case series”; “clinical features” and “patients”; performed in August 2016 yielded 70 papers of the last 45 years. The number of patients in these studies ranged from 3 to 113. 28 studies examined patients with PA from an endocrinological; and another 19 studies from a neurosurgical perspective. However; the majority of papers; especially the older ones; comprise case reports or a small cohort of patients. An overview of the most recent studies is given in Table 5. In our series, patients suffered from PA at a mean age of 56.3 years with an equal sex distribution. These demographical data differ from those in the literature where a male preponderance and a younger age at the time of the ictus are reported [17,19–24]. Yet, the reason for this male dominance remains obscure as patients with pituitary adenomas prone to apopolexy (i.e. prolactinomas, ACTH-secreting adenomas) are predominantly female. Furthermore, female sex hormone replacement as well as pregnancy are accepted risk factors for PA [25]. The neurological symptoms in our study group are in accordance with those reported in the literature where headache is documented in 63–100% of all investigated patients and neuroophthalmological symptoms in 23–81% [19,21,22,24,26]. The typical thunderclap headache in conjunction with PA, experienced by 45.8% patients in our study group is acknowledged as a separated headache category in the classification of the International Headache Society [27] and is described as a severe acute retroorbital headache, accompanied by at least one of the following symptoms such as nausea and vomiting, fever, alterations of consciousness, pituitary insufficiency, hypotension, diplopia, with all symptoms usually disappearing one month after the acute event. The pathophysiology of this acute headache is still largely unresolved with pathophysiological explanations ranging from chemical meningitis due to leakage of blood into the subarachnoid space [28] to increased intrasellar pressure in combination with a thus mediated rise in meningeal pressure and trigeminal irritation [29]. Ophtalmologic signs can be explained by compression of the optic chiasm and/or the cavernous sinus by the rapidly expanding hemorrhagic lesion. Nausea, vomiting (28.8%) and alterations of consciousness (15.3%) were found in our cohort in a comparable frequency with the literature. Even though, these are well-recognized symptoms of hyponatremia and hypocortisolism [10], and, moreover, hypopituitarism and hyponatremia were present in 22 and 12 of our patients, respectively, these symptoms were only recognized as clear-cut endocrinological sequelae of PA, namely Addisonian crisis, in three of our patients. In the literature, partial or complete anterior hypopituitarism is documented in 50%–86% of the patients [21,26,30], while hyponatremia is initially present in 12–44% [4,22]. Thus, hypopituitarism and hyponatremia occurred in our study group with a similar incidence as in the literature. Yet, the connection between these laboratory results and possible endocrinological sequelae of PA was not made in the majority of patients. As neurological departments or neurologists were the largest group of specialists to refer patients with PA to neurosurgery, it can be speculated that this referral pathway was the reason for a proper evaluation of neurological symptoms whereas endocrinological symptoms were not noted because physicians were not familiar with them. This may also account for the fact that in 23 of the investigated 60 patients endocrinological laboratory findings were incomplete or unavailable in the chart records. Interestingly, the prolactin level was assessed in most cases, missing in only 10% of patients, which may be explained by the fact that neurosurgeons are trained to rule out a prolactinoma prior to elective surgery. Almost every publication on PA reports potential risk factors for this

3.6. Neurological outcome Data available for the three- and/or twelve-month follow-up visits indicated that 7 of 17 patients experienced an improvement of visual acuity. Diplopia persisted in 4 of 19 patients. Visual field defects persisted in 8 of 13 patients. 8 patients reported about persisting headache at follow-up visits. In two of them the symptomatology was characterized as a typical migraine. In 8 patients of the study group absence or presence of headache was not noted in the chart records.

3.7. Endocrinological outcome 14 patients were lost to early follow-up at three months. After 12 months, only 39 patients attended the follow-up visit. Table 3 (see above) shows the endocrinological results of the patients. Please note that the patients who attended the 3-month and the 12-month followup were not necessarily the same. At the time of the last available endocrinological follow-up (three months, 12 months or later), 51/58 (87.9%) patients were on some kind of pituitary replacement therapy. No data were available in 2 patients. Table 4 shows the replaced hormone axes (multiple answers possible). Histology and biochemical results yielded a total of 14 patients with hormone-producing pituitary adenomas. Two females and 4 males were diagnosed with prolactinoma. Of those, 5 received long-term follow-up care (3–8 years) and were treated with dopamine agonists. Of the 4 patients with ACTH-immunostaining adenomas, 2 had been diagnosed with Cushing’s disease (CD) prior to PA, another after the ictus. 2 of the patients had clinical and biochemical signs of CD recurrence 7 and 9 years after PA and were on continuous medical treatment (fluconazole, pasireotide) or radiotherapy, respectively. In one patient with impressive clinical signs of active acromegaly (coarse facial features, macroglossia, enlargement of hands and feet, sweating, snoring, water retention etc.), the clinical symptomatology resolved after PA, but panhypopituitarism evolved instead. Unfortunately, this patient was not available for follow-up examinations. The second patient with histologically confirmed GH secreting pituitary adenoma was still in remission 5 years after PA. Five patients had died at the time of this study. One 91 year-old patient died one week after PA of cardio-respiratory insufficiency. Another patient with extensive subarachnoid hemorrhage accompanying the PA died 6 weeks after uneventful transsphenoidal surgery. The exact cause of death of the other three patients is unknown.

Table 4 Substitution of the different pituitary-axes at the time of the last endocrinological followup. Data are presented as frequency (N) and valid percent (%). Variable

Number of patients (N = 58)

Corticotropic axis (hydrocortisone) Thyreotropic axis (levothyroxine) Somatotropic axis (growth hormone) Gonadotropic axis Men = 29 (testosterone) Women n = 29 (estrogen) Diabetes insipidus (desmopressin)

31 (53.4) 36 (62.1) 3 (5.2) 13 (44.8) 3 (10.3) 7 (12.1)

42

43

49.8/M & F

Randeva et al. [4]

35

40

51/M & F

109

51.2/M & F

Lubina et al. [22]

8

Jho et al. [17]

29–66/ M&F

Elsässer et al. [26]

45

8

49/M & F

Sibal et al. [30]

32

55

Neurosurgical series Giammattei et al. 70/M & F [20]

56.6/M & F

52.4/M & F

Bujawansa et al. [19]

Pal et al. [23]

34/M & F

Sarwar et al. [18]

25

46

50.9/M & F

Vargas et al. [24]

No. of Patients

31

Age (years)/ Sex

Endocrinological series Giritharan et al. 55/M & F [21]

Author/Year

Headache (n = 7), nausea & vomiting (n = 4), visual impairment (n = 2), ophthalmoplegia (n = 6), photophobia (n = 1), decreased vigilance (n = 1) Headache (87%), vomiting (33%), visual impairment (39%), ophthalmoplegia (36%), meningismus (7.3%), decreased vigilance (12.8%), fever (1.8%)

nausea (80%), vomiting (57%), visual field deficit (71%), visual impairment (66%), ophthalmoplegia (69%), photophobia (49%), decreased vigilance (11%), hyponatriaemia (44%), fever (20%)

Headache (78%), nausea (28%), vomiting (47%), visual field deficit (50%), visual impairment (69%), ophthalmoplegia (81%), photophobia (19%), impairment of consciousness (12%), insuffiency of the anterior pituitary lobe (75%) Headache (96%), nausea & vomiting (78%), visual field deficit (48%), visual impairment (46%), ophthalmoplegia (51%), photophobia (33%), impairment of consciousness (22%), insuffiency of the anterior pituitary lobe (initial, 81%), fever (24%) Headache (n = 8), vomiting (n = 1), visual field deficit (n = 3), ophthalmoplegia (n = 5), photophobia (n = 1), one-sided amaurosis (n = 2) insuffiency of the anterior pituitary lobe (initial, n = 14), impairment of consciousness (n = 2) Headache (63%), vomiting (50%), visual field deficit (61%), ophthalmoplegia (40%), hyponatremia (12%), syncope (5%), impairment of consciousness (12%)

Headache (87.3%), nausea (16.4%), vomiting (25.5%), visual field deficit (36.4%), ophthalmoplegia (47.3%), photophobia (18.2%), trigeminal neuralgia (14.5%), collapse (3.6%

Anticoagulant drugs (8%)

Anticoagulant drugs (n = 2), arterial hypertension (n = 5), diabetes mellitus (n = 2)

Arterial hypertension (26%), adenoma (15%), ASS (9%), oral contraceptive (14%)

Traumatic brain injury (n = 2) Heparin (n = 2), general anesthesia (n = 1)

Dopamine agonist (n = 1), clopidogrel/cardiac catheter (n = 1), anticoagulant drugs (ASS n = 1, heparin n = 1), oral contraceptive (n = 1)

Surgery (11%), arterial hypertension (27%), anticoagulant drugs (warfarin 2%, ASS, 7%), oral contraceptive (4%), polycythemia (4%), factor-V-disorder (2%)

Not stated

Macroprolactinoma (64%), diabetes mellitus (8%),arterial hypertension (16%), dynamic pituitary testing (12%), radiotherapy (4%), anticoagulant drugs (12%) Therapy with: bromocriptine (36%), cabergoline (80%), pergolide (4%) Arterial hypertension (20%) Anticoagulant drugs (warfarin n = 3, ASS n = 2), coronary bypass surgery (n = 2), orthopedic surgery (n = 3) Clomiphene (n = 1)

Diabetes mellitus (20%), arterial hypertension (20%), dyslipidemia (33%), anticoagulant drugs (14%), dopamine agonists (18%)

Headache (75%), visual field deficit (73%) ophthalmoplegia (23%)

Headache (28%)

Arterial hypertension (16%), pregnancy (35%) dialysis (with heparin, 3%) anticoagulant drug (dabigatran, 10%)

Possible risk factors

Headache (100%), nausea & vomiting (55%), visual field deficit (58%), visual impairment (81%), ophthalmoplegia (39%), insuffiency of the anterior pituitary lobe (86%)

Clinical symptoms (% of patients)

Table 5 Overview of the largest endocrinological and neurosurgical series of patients with pituitary apoplexy.

(Partial) insuffiency of the anterior pituitary lobe in all patients.

19% of patients with normal pituitary function after surgery. 11% of patients with normal pituitary function after conservative treatment. Global insuffiency of the anterior pituitary lobe (n = 4) Partial insuffiency of the anterior pituitary lobe (n = 2) Gonadotropic insufficiency: 79% Thyreotropic insufficiency: 54% Corticotropic insufficiency: 40% Diabetes Insipidus: 8% Gonadotropic insufficiency: 43% of men Thyreotropic insufficiency: 45% Corticotropic insufficiency: 58% Diabetes Insipidus: 6%

Somatotropic insufficiency: 38,2% Gonadotropic insufficiency: 77.1% (men) Diabetes insipidus: n = 2 Not stated

Corticotropic insufficiency: 72.7% Thyreotropic insufficiency: 52.7%

No improvement of the insuffiencies of the anterior pituitary lobe over time: Somatotropic insufficiency: 74% Corticotropic insufficiency: 68% Gonadotropic insufficiency:60% Persistent visual field deficit (n = 35) Persistent palsy of N. oculomotorius (n = 11) Hypothyroidism (n = 24) Hypocortisolism (n = 24) Hypogonadism (n = 22) Panhypopituitarism (n = 16) anterior pituitary lobe insufficiency in 3 / 23 patients within the clinical course

Outcome

A. Grzywotz et al.

Clinical Neurology and Neurosurgery 163 (2017) 39–45

Clinical Neurology and Neurosurgery 163 (2017) 39–45

A. Grzywotz et al.

Funding

event [24,25,30–32]. However, the role of these risk factors remains speculative since the underlying pathophysiological mechanisms have not been elucidated. Pathophysiological considerations pertaining to changes in perfusion of a pre-existing pituitary adenoma might add to understanding PA [33], as exemplified in reports of PA after cardiac surgery with extracorporal circulation or after dynamic endocrine function testing with releasing hormones [16,34–36]. In our study group, more than a quarter of all patients (28.6%) was on anticoagulant medication which is almost twice as high as the highest numbers reported in the literature (3–14%) [21,24,30,32]. The high incidence of anticoagulant medications may be explained by the higher overall population age of our patients. Nevertheless, it may hallmark that we are to expect older patients with PA in the years to come. The literature provides little evidence when it comes to the best treatment modality of pituitary apoplexy. A UK guideline on the management of pituitary apoplexy in 2011 [37], highlighted the controversy about role and time point of neurosurgical removal of PA. Up to date no controlled randomized trials exist, comparing neurosurgical treatment to conservative management. Several retrospective studies could show that the results between surgically and conservatively managed patients were not significantly different, if conservative management was restricted to patients with mild and non-progressive ophthalmological deficits with improvement or complete resolution of their visual deficits in 80–100% [19,21,26,30]. The same holds true for neurosurgical series in which early (within the first 7-8 days) operative removal of PA also led to improvement of visual acuity and visual field deficits in 71–93% of patients. However, the hypothesis that pituitary function would also improve, if neurosurgical removal of PA was performed, could not be confirmed with high rates (73–96%) of persisting pituitary dysfunction in surgically as well as conservatively managed patients [19,21,30]. The outcome of our, mostly surgically treated, study group mirrors that of the literature perfectly, with good resolution of visual deteriorations and a poor recovery of pituitary function. Therefore, we may conclude from our data that patients suffering from PA with no or mild visual field deficits or incomplete diplopia could be managed conservatively. We were surprised how many patients were lost to endocrinological (and neurosurgical) follow-up after the initial treatment. This may be explained partially by the fact, that our department is part of a tertiary referral center with a large catchment area. Therefore, some patients may attend follow-up appointments closer to their place of living. However, the high degree of at least partial hypopituitarism of those who attended follow-up appointments and the more than anecdotal reoccurrence of endocrinopathies such as Cushing’s disease years after PA imply that patients should be strongly advised to attend a regular follow-up appointments with their endocrinologist (and neurosurgeon) as well as MRI examinations. Neurosurgeons and neurologists, on the other hand, should be trained more to order endocrine laboratory testing on admission of their patients with PA and to look for the endocrinologial signs and symptoms accompanying this event in order to avoid potentially lethal complications such as Addisonian crisis.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Acknowledgement The data were collected for the doctoral thesis of Agnieszka Grzywotz. References [1] H. Cushing, The Pituitary Body and Its Disorders, Lippincott company, Philadelphia, London, 1912. [2] P. Bailey, Pathological report of a case of acromegaly, with special reference to the lesions in the hypophysis cerebri an in the thyroid galnd; and a case of haemorrhage into the pituitary, Philadelphia Med. J. 1 (1898) 789–792. [3] M. Brougham, A.P. Heusner, R.D. Adams, Acute degenerative changes in adenomas of the pituitary body–with special reference to pituitary apoplexy, J. Neurosurg. 7 (5) (1950) 421–439. [4] H.S. Randeva, J. Schoebel, J. Byrne, M. Esiri, C.B. Adams, J.A. Wass, Classical pituitary apoplexy: clinical features, management and outcome, Clin. Endocrinol. (Oxf.) 51 (2) (1999) 181–188. [5] C. Capatina, W. Inder, N. Karavitaki, J.A. Wass, Management of endocrine disease: pituitary tumour apoplexy, Eur. J. Endocrinol. 172 (5) (2015) R179–R190. [6] P. Maccagnan, C.L. Macedo, M.J. Kayath, R.G. Nogueira, J. Abucham, Conservative management of pituitary apoplexy: a prospective study, J. Clin. Endocrinol. Metab. 80 (7) (1995) 2190–2197. [7] R.N. Nawar, D. AbdelMannan, W.R. Selman, B.M. Arafah, Pituitary tumor apoplexy: a review, J. Intensive Care Med. 23 (2) (2008) 75–90. [8] D. Ni Chroinin, J. Lambert, Sudden headache, third nerve palsy and visual deficit: thinking outside the subarachnoid haemorrhage box, Age Ageing 42 (6) (2013) 810–812. [9] M. Verrees, B.M. Arafah, W.R. Selman, Pituitary tumor apoplexy: characteristics, treatment, and outcomes, Neurosurg. Focus 16 (4) (2004) E6. [10] S. Diederich, N.F. Franzen, V. Bahr, W. Oelkers, Severe hyponatremia due to hypopituitarism with adrenal insufficiency: report on 28 cases, Eur. J. Endocrinol. 148 (6) (2003) 609–617. [11] H.J. Schneider, G. Aimaretti, I. Kreitschmann-Andermahr, G.K. Stalla, E. Ghigo, Hypopituitarism, Lancet 369 (9571) (2007) 1461–1470. [12] T. Deutschbein, N. Unger, K. Mann, S. Petersenn, Diagnosis of secondary adrenal insufficiency: unstimulated early morning cortisol in saliva and serum in comparison with the insulin tolerance test, Horm. Metab. Res. 41 (11) (2009) 834–839. [13] L. Katznelson, E.R. Laws Jr., S. Melmed, M.E. Molitch, M.H. Murad, A. Utz, J.A. Wass, Acromegaly: an endocrine society clinical practice guideline, J. Clin. Endocrinol. Metab. 99 (11) (2014) 3933–3951. [14] S. Melmed, F.F. Casanueva, A.R. Hoffman, D.L. Kleinberg, V.M. Montori, J.A. Schlechte, J.A. Wass, Diagnosis and treatment of hyperprolactinemia: an Endocrine Society clinical practice guideline, J. Clin. Endocrinol. Metab. 96 (2) (2011) 273–288. [15] A. Colao, M. Boscaro, D. Ferone, F.F. Casanueva, Managing Cushing's disease: the state of the art, Endocrine 47 (1) (2014) 9–20. [16] M. Hidiroglu, A. Kucuker, E. Ucaroglu, S.A. Kucuker, E. Sener, Pituitary apoplexy after cardiac surgery, Ann. Thoracic Surg. 89 (5) (2010) 1635–1637. [17] D.H. Jho, B.M. Biller, P.K. Agarwalla, B. Swearingen, Pituitary Apoplexy Large Surgical Series with Grading System, World neurosurgery, (2014). [18] K.N. Sarwar, M.S. Huda, V. Van de Velde, L. Hopkins, S. Luck, R. Preston, B.M. McGowan, P.V. Carroll, J.K. Powrie, The prevalence and natural history of pituitary hemorrhage in prolactinoma, J. Clin. Endocrinol. Metab. 98 (6) (2013) 2362–2367. [19] S. Bujawansa, S.K. Thondam, C. Steele, D.J. Cuthbertson, C.E. Gilkes, C. Noonan, C.W. Bleaney, I.A. Macfarlane, M. Javadpour, C. Daousi, Presentation, management and outcomes in acute pituitary apoplexy: a large single-centre experience from the United Kingdom, Clin. Endocrinol. (Oxf.) 80 (3) (2014) 419–424. [20] L. Giammattei, G. Mantovani, G. Carrabba, S. Ferrero, A. Di Cristofori, E. Verrua, C. Guastella, L. Pignataro, P. Rampini, M. Minichiello, M. Locatelli, Pituitary apoplexy: considerations on a single center experience and review of the literature, J. Endocrinol. Invest. 39 (7) (2016) 739–746. [21] S. Giritharan, K. Gnanalingham, T. Kearney, Pituitary apoplexy – bespoke patient management allows good clinical outcome, Clin. Endocrinol. (Oxf.) 85 (3) (2016) 415–422. [22] A. Lubina, D. Olchovsky, M. Berezin, Z. Ram, M. Hadani, I. Shimon, Management of pituitary apoplexy: clinical experience with 40 patients, Acta Neurochir. (Wien) 147 (2) (2005) 151–157 (discussion 157). [23] A. Pal, C. Capatina, A.P. Tenreiro, P.D. Guardiola, J.V. Byrne, S. Cudlip, N. Karavitaki, J.A. Wass, Pituitary apoplexy in non-functioning pituitary adenomas: long term follow up is important because of significant numbers of tumour recurrences, Clin. Endocrinol. (Oxf.) 75 (4) (2011) 501–504. [24] G. Vargas, B. Gonzalez, G. Guinto, V. Mendoza, B. Lopez-Felix, E. Zepeda, M. Mercado, Pituitary apoplexy in nonfunctioning pituitary macroadenomas: a case-control study, Endocr. Pract. 20 (12) (2014) 1274–1280. [25] D.L. Moller-Goede, M. Brandle, K. Landau, R.L. Bernays, C. Schmid, Pituitary

5. Conclusion From our data, we suggest that patients with PA should be strongly advised to be followed by both, endocrinologists and neurosurgeons, after the acute event. With the aging of our population and the increasing use of anticoagulant medication, PA may become a subject of greater concern which has to be dealt with respect to its possible endocrinological, ophthalmological and neurosurgical sequelae. Physicians should be aware of associated problems of PA, especially endocrinological emergencies as Addisonian crisis. Conflict of interest None. 44

Clinical Neurology and Neurosurgery 163 (2017) 39–45

A. Grzywotz et al.

[26]

[27] [28] [29]

[30]

[31] N. Cinar, Y. Tekinel, S. Dagdelen, H. Oruckaptan, F. Soylemezoglu, T. Erbas, Cavernous sinus invasion might be a risk factor for apoplexy, Pituitary 16 (4) (2013) 483–489. [32] P.L. Semple, J.A. Jane Jr., E.R. Laws Jr., Clinical relevance of precipitating factors in pituitary apoplexy, Neurosurgery 61 (5) (2007) 956–961 (discussion 961-2). [33] E.H. Oldfield, M.J. Merrill, Apoplexy of pituitary adenomas: the perfect storm, J. Neurosurg. 122 (6) (2015) 1444–1449. [34] A. Masago, Y. Ueda, H. Kanai, H. Nagai, S. Umemura, Pituitary apoplexy after pituitary function test: a report of two cases and review of the literature, Surg. Neurol. 43 (2) (1995) 158–164 (discussion 165). [35] T. Tansel, M. Ugurlucan, E. Onursal, Pituitary apoplexy following coronary artery bypass grafting: report of a case, Acta Chir. Belg. 110 (4) (2010) 484–486. [36] M. Thurtell, M. Besser, G. Halmagyi, Pituitary apoplexy causing isolated blindness after cardiac bypass surgery, Arch. Ophthalmol. 126 (4) (2008) 576–578. [37] S. Rajasekaran, M. Vanderpump, S. Baldeweg, W. Drake, N. Reddy, M. Lanyon, A. Markey, G. Plant, M. Powell, S. Sinha, J. Wass, UK guidelines for the management of pituitary apoplexy, Clin. Endocrinol. (Oxf.) 74 (1) (2011) 9–20.

apoplexy: re-evaluation of risk factors for bleeding into pituitary adenomas and impact on outcome, Eur. J. Endocrinol. 164 (1) (2011) 37–43. P.N. Elsasser Imboden, N. De Tribolet, A. Lobrinus, R.C. Gaillard, L. Portmann, F. Pralong, F. Gomez, Apoplexy in pituitary macroadenoma: eight patients presenting in 12 months, Medicine 84 (3) (2005) 188–196. The international classification of headache disorders, 3rd edition (beta version), Cephalalgia Int. J. Headache 33 (9) (2013) 629–808. A. Glezer, M.D. Bronstein, Pituitary apoplexy: pathophysiology, diagnosis and management, Arch. Endocrinol. Metab. 59 (3) (2015) 259–264. B.M. Arafah, D. Prunty, J. Ybarra, M.L. Hlavin, W.R. Selman, The dominant role of increased intrasellar pressure in the pathogenesis of hypopituitarism, hyperprolactinemia, and headaches in patients with pituitary adenomas, J. Clin. Endocrinol. Metab. 85 (5) (2000) 1789–1793. L. Sibal, S.G. Ball, V. Connolly, R.A. James, P. Kane, W.F. Kelly, P. Kendall-Taylor, D. Mathias, P. Perros, R. Quinton, B. Vaidya, Pituitary apoplexy: a review of clinical presentation, management and outcome in 45 cases, Pituitary 7 (3) (2004) 157–163.

45