Clinical Endocrinology (2014) 80, 419–424
doi: 10.1111/cen.12307
ORIGINAL ARTICLE
Presentation, management and outcomes in acute pituitary apoplexy: a large single-centre experience from the United Kingdom S. Bujawansa*, S. K. Thondam*, C. Steele*, D. J. Cuthbertson*, C. E. Gilkes†, C. Noonan‡, C. W. Bleaney§, I. A. MacFarlane*, M. Javadpour† and C. Daousi* *Department of Endocrinology & Diabetes, University Hospital Aintree, †Department of Neurosurgery, The Walton Centre for Neurology & Neurosurgery, ‡Department of Ophthalmology, University Hospital Aintree and §University of Liverpool Medical School, Liverpool, UK
Summary Objective To study the presentation, management and outcomes and to apply retrospectively the Pituitary Apoplexy Score (PAS) (United Kingdom (UK) guidelines for management of apoplexy) to a large, single-centre series of patients with acute pituitary apoplexy. Design Retrospective analysis of casenotes at a single neurosurgical centre in Liverpool, UK. Results Fifty-five patients [mean age, 524 years; median duration of follow-up, 7 years] were identified; 45 of 55 (81%) had nonfunctioning adenomas, four acromegaly and six prolactinomas. Commonest presenting features were acute headache (87%), diplopia (472%) and visual field (VF) defects (36%). The most frequent ocular palsy involved the 3rd nerve (81%), followed by 6th nerve (346%) and multiple palsies (19%). Twenty-three patients were treated conservatively, and the rest had surgery either within 7 days of presentation or delayed elective surgery. Indications for surgery were deteriorating visual acuity and persistent field defects. Patients presenting with VF defects (n = 20) were more likely to undergo surgery (75%) than to be managed expectantly (25%). There was no difference in the rates of complete/near-complete resolution of VF deficits and cranial nerve palsies between those treated conservatively and those who underwent surgery. Endocrine outcomes were also similar. We were able to calculate the PAS for 46 patients: for the group treated with early surgery mean, PAS was 38 and for those managed conservatively or with delayed surgery was 18. Conclusions Patients without VF deficits or whose visual deficits are stable or improving can be managed expectantly without negative impact on outcomes. Clinical severity based on a
Correspondence: Dr Christina Daousi, Diabetes & Endocrinology Clinical Research Group, Clinical Sciences Centre, University Hospital Aintree, Lower Lane, Liverpool L9 7AL, UK. Tel.: +44 (151) 529 5920; Fax: +44 (151) 529 5888; E-mail: [email protected] © 2013 John Wiley & Sons Ltd
PAS ≥ 4 appeared to influence management towards emergency surgical intervention. (Received 8 May 2013; returned for revision 14 June 2013; finally revised 29 July 2013; accepted 30 July 2013)
Introduction Acute pituitary apoplexy is described as a clinical syndrome consisting of a constellation of symptoms and signs that occur following rapid expansion of the contents of the sella turcica, generally caused by haemorrhage or infarction of a pre-existing pituitary adenoma.1 Common presenting features are sudden onset headache, visual impairment and reduced consciousness.2 Pituitary apoplexy can lead to serious long-term neuro-ophthalmological and endocrine sequelae, and there is variation in the frequency of these complications depending on the management strategy used and the expertise of the centre.3,4 Pituitary apoplexy is relatively rare,5 and there are no randomized controlled trials on the optimal management of this condition. Management has mostly depended on conclusions drawn from several case series and expert opinion. There are several controversies in the key areas of management of this rare neuroendocrine emergency including whether conservative management or surgery should be the first-line strategy and the optimal timing of surgery. Based on case series data, some authors advocate conservative management initially especially in the presence of mild or nonprogressive neuro-ophthalmic deficits.4,6,7 Others promote an early surgical approach especially in the presence of more significant defects.8–10 The United Kingdom National guidelines on the management of pituitary apoplexy were published recently, and they addressed several important issues including steroid replacement during the acute presentation, management and longterm follow-up.2 They also provided a scoring system Pituitary 419
420 S. Bujawansa et al. Apoplexy Score (PAS) that can be used to monitor patients who are being managed conservatively with a degree of objectivity.2 We present our data from one of the largest cohorts of patients with acute pituitary apoplexy managed by a single neuroendocrine/neurosurgical team. Our series is unique as we have one of the largest groups of conservatively managed patients from the UK published to date and with one of the longest periods of follow-up. This retrospective review had the following aims: (i) to describe the presenting symptoms and signs of patients with acute pituitary apoplexy referred to a single UK regional centre, (ii) to help ascertain differences in endocrine and neuro-ophthalmological outcomes between patients treated with emergency decompressive surgery and those managed expectantly and (iii) to apply retrospectively the Pituitary Apoplexy Scoring tool to our large patient cohort and to determine its applicability in the management of these patients in the acute setting.
Patients and methods Patients We conducted a retrospective study of all patients (n = 55) referred between 1985 and 2010 to a single regional joint neurosurgical/neuroendocrine team at the Walton Centre for Neurology and Neurosurgery in Liverpool with a diagnosis of acute pituitary apoplexy. Permission from the Institution’s Audit Department was obtained, and data were retrieved from the hospital casenotes and electronic records. We diagnosed pituitary apoplexy according to the clinical definition of Randeva et al.3 Only patients who presented with sudden onset severe headache and/or acute visual disturbance [ophthalmoplegia, reduced visual acuity or acute onset visual field defects (VFD)] and confirmed by brain imaging as having haemorrhage/infarction of a pituitary adenoma were included in the study. Details of initial presentation, investigations, management and follow-up were collected. Visual acuity was measured clinically and VFD were assessed initially clinically by simple confrontation testing and subsequently confirmed with formal VF charting within a few days of presentation by a neuro-ophthalmologist, who also documented any ophthalmoplegia and serially monitored progression or resolution. Data were collected on administration of corticosteroid treatment in the acute setting and the basis on which decisions were taken regarding surgical management vs expectant monitoring. We have arbitrarily used a cut-off within 7 days of presentation to define early surgery in keeping with previously published series to help make our data comparable.2,11,12 The PAS was calculated in the subset of patients in whom all relevant parameters were available from the time of presentation. This scoring system uses four parameters (conscious level – Glasgow Coma Scale, visual acuity, VFD and cranial nerve palsies) and gives a score ranging from 0 to 10 with higher scores, suggesting more significant neuro-ophthalmological impairment.
Follow-up After discharge, patients were followed up in a joint clinic conducted by an endocrinologist and a pituitary neurosurgeon. The frequency of follow-up was decided by the managing clinicians depending on the complexity of the case. Data on follow-up endocrine testing and imaging, details on tumour recurrence or endocrine hypo-/hypersecretion were collected. Information regarding the basis on which decisions were taken for further surgery, radiotherapy or medical management was gathered. Tumour recurrence was defined as an increase in tumour dimensions compared with the initial postoperative scan in patients treated surgically or compared with the first follow-up scan in conservatively managed patients. In hormonally active tumours, recurrence of endocrine activity was considered to be present if the prolactin level was increased during follow-up in prolactinomas, or by appropriate growth hormone suppression testing and IGF-1 values in somatotroph adenomas. During the period of follow-up, all patients underwent repeat, serial endocrine assessment following additional/adjuvant treatment for their tumour (elective surgery or repeat surgery, radiotherapy). Statistical analysis Data were analysed using PASW statistics 18. Categorical data were examined with the chi-squared test or the Fisher exact test, as appropriate. Continuous data were examined with the use of the t-test for normally distributed data or the Mann–Whitney U-test for non-normally distributed data. A two-tailed P value < 005 was considered to be statistically significant.
Results Clinical features at presentation A total of 55 patients were identified (35 men, 67%), median age at presentation 58 years (range, 14–78; mean age, 524 years). Median age at presentation was similar for both sexes. Median duration of follow-up was 7 years (interquartile range, 2–11). Forty-five cases (82%) of apoplexy occurred in clinically nonfunctioning pituitary adenomas (CNFPAs), six tumours (115%) were prolactinomas and four (72%) were growth hormone secreting adenomas. Two patients had multiple endocrine neoplasia (MEN) type 1 syndrome. Prior to the episode of acute apoplexy, only four (72%) of 55 patients had a known diagnosis of pituitary adenoma. Eleven patients (20%) had a history of hypertension, three (55%) were receiving treatment with warfarin and two were taking aspirin. Two patients had coronary artery bypass grafting close to the acute apoplectic event, and another three patients had recently undergone major orthopaedic surgery (one patient had shoulder surgery and two patients had undergone knee surgery). One patient suffered from pituitary apoplexy 5 days after being treated with clomiphene.13 Overall, 11 of 55 (20%) patients had an identified precipitant (aspirin, warfarin, major surgery, clomiphene) for acute pituitary apoplexy other than the presence of hypertension. © 2013 John Wiley & Sons Ltd Clinical Endocrinology (2014), 80, 419–424
Management of acute pituitary apoplexy 421 The commonest presenting symptoms were acute severe headache (87%) and diplopia (38%). Cranial nerve palsies (47%) were the commonest clinical signs at presentation followed by VFD (36%). Other clinical features noted at the time of presentation are listed in Table 1.
total number of n = 32 patients who were managed conservatively during the first week of their presentation. Of these patients, 29 had CNFPAs and three had prolactinomas. Patients with VFD which were improving spontaneously were conservatively managed in the first instance.
Management
Hormonally active tumours. Four of the six prolactinoma cases were treated with early surgery, one with delayed surgery and one lesion was managed conservatively. Both prolactinoma cases not treated with early, emergency surgery received at presentation treatment with dopamine agonists in the form of bromocriptine. All four cases of acromegaly required early transsphenoidal decompressive surgery. All remained biochemically active postoperatively and received medical treatment with dopamine agonists somatostatin analogues, and three required adjuvant radiotherapy. Two patients with acromegaly required multiple surgical procedures, conventional fractionated radiotherapy, stereotactic radiosurgery and subsequently treatment with temozolomide because of the persistent hormonal activity and the invasive and aggressive nature of their tumours.
Surgery. Thirty-three patients (33 of 55, 55%) underwent pituitary surgery at some stage either within 1 week of presentation or elective delayed surgery (>7 days). Indications for surgery were deteriorating visual acuity and persistent or new VF defects. Patients presenting with VF defects (n = 20) were more likely to undergo surgery (15 of 20, 75%) than to be managed expectantly (n = 5/20, 25%). Early emergency surgery (within 7 days of presentation) was performed in 23 patients (16 CNFPAs, four acromegaly and three prolactinomas). Ten of these patients had VFD, 13 had ≥1 cranial nerve palsy and five had both. Eleven patients had only small residual adenoma on latest MRI scan requiring surveillance only, whereas 10 underwent subsequent radiotherapy. Ten patients (182%) underwent delayed elective surgery (after 7 days of presentation, range 8 days to 3 years, mean time to elective surgery 266 days). Indications for elective surgery included further enlargement of the adenoma on subsequent surveillance imaging and new or deteriorating VFD. Nine patients had CNFPAs, and one had a prolactinoma. Three had VFD (two fully/near fully resolved), and five presented with ≥1 cranial nerve palsy (four fully/near fully resolved). Conservative management. Twenty-two patients (22/55, 40%) were managed conservatively and required no surgical intervention throughout their period of follow-up. Ten of these patients had VFD at presentation, which improved spontaneously within the first week of expectant monitoring, 15 presented with ≥1 cranial nerve palsy and two had both. As mentioned previously, ten patients of 55 (182%) underwent delayed elective surgery (after 7 days of presentation), giving a
Table 1. Presenting clinical symptoms and signs Clinical feature
Number (%)
Acute headache Cranial nerve palsy 3rd nerve palsy 6th nerve palsy Multiple palsies Diplopia Visual field defect Vomiting Photophobia Nausea Facial pain/trigeminal neuralgia Collapse
48 26 16 5 5 21 20 14 10 9 8
© 2013 John Wiley & Sons Ltd Clinical Endocrinology (2014), 80, 419–424
(873) (473)
(382) (364) (255) (182) (164) (145)
2 (36)
Steroid treatment. Details of steroid treatment during the acute presentation were available for 31 patients. Twenty-seven patients (87%) received some form of steroid therapy during the acute phase. Nine patients received dexamethasone. Eight of them had very large tumours with optic chiasmal compression or significant extra-/parasellar extension. Two patients in this group were managed conservatively, five underwent emergency surgery and one patient underwent elective surgery. In the five patients who received dexamethasone and underwent emergency surgery, all VFD and cranial nerve palsies resolved, but it is difficult to ascertain if these favourable neuro-ophthalmic outcomes were attributable to the effects of dexamethasone or the effects of emergency decompressive surgery. No evidence of excess mortality or morbidity was noted in the dexamethasonetreated group. All other patients received stress doses of hydrocortisone except one who continued prednisolone, which had been previously prescribed for polymyalgia rheumatica. Pituitary Apoplexy Score. Pituitary Apoplexy Score was calculated retrospectively for 46 patients in whom all the required baseline data were available. The mean PAS for the group which was treated with early surgery was 38 (range, 1–6), while the mean score for the group managed conservatively or with delayed surgery was 18 (range, 0–5) (Table 2). From the group of patients treated with early emergency surgery, 73% had a PAS of four or more, while only 13% of the conservative management/delayed surgery group had a PAS of four or more. Outcomes. There were no deaths related to pituitary apoplexy in our series. Overall, 65% of patients had complete resolution of their VF defects, while 85% had at least partial improvement of their VF defects. When the whole group was considered, cranial nerve palsies resolved completely in 23 patients, and there was at least partial resolution in 26 by the time of their last follow-up.
422 S. Bujawansa et al. Table 2. Total PAS and individual components of the PAS for the two treatment groups
PAS variable Total PAS (mean) Score for GCS Score for visual acuity Score for visual fields Score for cranial nerve palsies
Early surgery group (n = 19)
Delayed surgery/ conservative group (n = 27)
P-value
38 04 11 17 06
18 02 04 05 06
doi: 10.1111/cen.12307
ORIGINAL ARTICLE
Presentation, management and outcomes in acute pituitary apoplexy: a large single-centre experience from the United Kingdom S. Bujawansa*, S. K. Thondam*, C. Steele*, D. J. Cuthbertson*, C. E. Gilkes†, C. Noonan‡, C. W. Bleaney§, I. A. MacFarlane*, M. Javadpour† and C. Daousi* *Department of Endocrinology & Diabetes, University Hospital Aintree, †Department of Neurosurgery, The Walton Centre for Neurology & Neurosurgery, ‡Department of Ophthalmology, University Hospital Aintree and §University of Liverpool Medical School, Liverpool, UK
Summary Objective To study the presentation, management and outcomes and to apply retrospectively the Pituitary Apoplexy Score (PAS) (United Kingdom (UK) guidelines for management of apoplexy) to a large, single-centre series of patients with acute pituitary apoplexy. Design Retrospective analysis of casenotes at a single neurosurgical centre in Liverpool, UK. Results Fifty-five patients [mean age, 524 years; median duration of follow-up, 7 years] were identified; 45 of 55 (81%) had nonfunctioning adenomas, four acromegaly and six prolactinomas. Commonest presenting features were acute headache (87%), diplopia (472%) and visual field (VF) defects (36%). The most frequent ocular palsy involved the 3rd nerve (81%), followed by 6th nerve (346%) and multiple palsies (19%). Twenty-three patients were treated conservatively, and the rest had surgery either within 7 days of presentation or delayed elective surgery. Indications for surgery were deteriorating visual acuity and persistent field defects. Patients presenting with VF defects (n = 20) were more likely to undergo surgery (75%) than to be managed expectantly (25%). There was no difference in the rates of complete/near-complete resolution of VF deficits and cranial nerve palsies between those treated conservatively and those who underwent surgery. Endocrine outcomes were also similar. We were able to calculate the PAS for 46 patients: for the group treated with early surgery mean, PAS was 38 and for those managed conservatively or with delayed surgery was 18. Conclusions Patients without VF deficits or whose visual deficits are stable or improving can be managed expectantly without negative impact on outcomes. Clinical severity based on a
Correspondence: Dr Christina Daousi, Diabetes & Endocrinology Clinical Research Group, Clinical Sciences Centre, University Hospital Aintree, Lower Lane, Liverpool L9 7AL, UK. Tel.: +44 (151) 529 5920; Fax: +44 (151) 529 5888; E-mail: [email protected] © 2013 John Wiley & Sons Ltd
PAS ≥ 4 appeared to influence management towards emergency surgical intervention. (Received 8 May 2013; returned for revision 14 June 2013; finally revised 29 July 2013; accepted 30 July 2013)
Introduction Acute pituitary apoplexy is described as a clinical syndrome consisting of a constellation of symptoms and signs that occur following rapid expansion of the contents of the sella turcica, generally caused by haemorrhage or infarction of a pre-existing pituitary adenoma.1 Common presenting features are sudden onset headache, visual impairment and reduced consciousness.2 Pituitary apoplexy can lead to serious long-term neuro-ophthalmological and endocrine sequelae, and there is variation in the frequency of these complications depending on the management strategy used and the expertise of the centre.3,4 Pituitary apoplexy is relatively rare,5 and there are no randomized controlled trials on the optimal management of this condition. Management has mostly depended on conclusions drawn from several case series and expert opinion. There are several controversies in the key areas of management of this rare neuroendocrine emergency including whether conservative management or surgery should be the first-line strategy and the optimal timing of surgery. Based on case series data, some authors advocate conservative management initially especially in the presence of mild or nonprogressive neuro-ophthalmic deficits.4,6,7 Others promote an early surgical approach especially in the presence of more significant defects.8–10 The United Kingdom National guidelines on the management of pituitary apoplexy were published recently, and they addressed several important issues including steroid replacement during the acute presentation, management and longterm follow-up.2 They also provided a scoring system Pituitary 419
420 S. Bujawansa et al. Apoplexy Score (PAS) that can be used to monitor patients who are being managed conservatively with a degree of objectivity.2 We present our data from one of the largest cohorts of patients with acute pituitary apoplexy managed by a single neuroendocrine/neurosurgical team. Our series is unique as we have one of the largest groups of conservatively managed patients from the UK published to date and with one of the longest periods of follow-up. This retrospective review had the following aims: (i) to describe the presenting symptoms and signs of patients with acute pituitary apoplexy referred to a single UK regional centre, (ii) to help ascertain differences in endocrine and neuro-ophthalmological outcomes between patients treated with emergency decompressive surgery and those managed expectantly and (iii) to apply retrospectively the Pituitary Apoplexy Scoring tool to our large patient cohort and to determine its applicability in the management of these patients in the acute setting.
Patients and methods Patients We conducted a retrospective study of all patients (n = 55) referred between 1985 and 2010 to a single regional joint neurosurgical/neuroendocrine team at the Walton Centre for Neurology and Neurosurgery in Liverpool with a diagnosis of acute pituitary apoplexy. Permission from the Institution’s Audit Department was obtained, and data were retrieved from the hospital casenotes and electronic records. We diagnosed pituitary apoplexy according to the clinical definition of Randeva et al.3 Only patients who presented with sudden onset severe headache and/or acute visual disturbance [ophthalmoplegia, reduced visual acuity or acute onset visual field defects (VFD)] and confirmed by brain imaging as having haemorrhage/infarction of a pituitary adenoma were included in the study. Details of initial presentation, investigations, management and follow-up were collected. Visual acuity was measured clinically and VFD were assessed initially clinically by simple confrontation testing and subsequently confirmed with formal VF charting within a few days of presentation by a neuro-ophthalmologist, who also documented any ophthalmoplegia and serially monitored progression or resolution. Data were collected on administration of corticosteroid treatment in the acute setting and the basis on which decisions were taken regarding surgical management vs expectant monitoring. We have arbitrarily used a cut-off within 7 days of presentation to define early surgery in keeping with previously published series to help make our data comparable.2,11,12 The PAS was calculated in the subset of patients in whom all relevant parameters were available from the time of presentation. This scoring system uses four parameters (conscious level – Glasgow Coma Scale, visual acuity, VFD and cranial nerve palsies) and gives a score ranging from 0 to 10 with higher scores, suggesting more significant neuro-ophthalmological impairment.
Follow-up After discharge, patients were followed up in a joint clinic conducted by an endocrinologist and a pituitary neurosurgeon. The frequency of follow-up was decided by the managing clinicians depending on the complexity of the case. Data on follow-up endocrine testing and imaging, details on tumour recurrence or endocrine hypo-/hypersecretion were collected. Information regarding the basis on which decisions were taken for further surgery, radiotherapy or medical management was gathered. Tumour recurrence was defined as an increase in tumour dimensions compared with the initial postoperative scan in patients treated surgically or compared with the first follow-up scan in conservatively managed patients. In hormonally active tumours, recurrence of endocrine activity was considered to be present if the prolactin level was increased during follow-up in prolactinomas, or by appropriate growth hormone suppression testing and IGF-1 values in somatotroph adenomas. During the period of follow-up, all patients underwent repeat, serial endocrine assessment following additional/adjuvant treatment for their tumour (elective surgery or repeat surgery, radiotherapy). Statistical analysis Data were analysed using PASW statistics 18. Categorical data were examined with the chi-squared test or the Fisher exact test, as appropriate. Continuous data were examined with the use of the t-test for normally distributed data or the Mann–Whitney U-test for non-normally distributed data. A two-tailed P value < 005 was considered to be statistically significant.
Results Clinical features at presentation A total of 55 patients were identified (35 men, 67%), median age at presentation 58 years (range, 14–78; mean age, 524 years). Median age at presentation was similar for both sexes. Median duration of follow-up was 7 years (interquartile range, 2–11). Forty-five cases (82%) of apoplexy occurred in clinically nonfunctioning pituitary adenomas (CNFPAs), six tumours (115%) were prolactinomas and four (72%) were growth hormone secreting adenomas. Two patients had multiple endocrine neoplasia (MEN) type 1 syndrome. Prior to the episode of acute apoplexy, only four (72%) of 55 patients had a known diagnosis of pituitary adenoma. Eleven patients (20%) had a history of hypertension, three (55%) were receiving treatment with warfarin and two were taking aspirin. Two patients had coronary artery bypass grafting close to the acute apoplectic event, and another three patients had recently undergone major orthopaedic surgery (one patient had shoulder surgery and two patients had undergone knee surgery). One patient suffered from pituitary apoplexy 5 days after being treated with clomiphene.13 Overall, 11 of 55 (20%) patients had an identified precipitant (aspirin, warfarin, major surgery, clomiphene) for acute pituitary apoplexy other than the presence of hypertension. © 2013 John Wiley & Sons Ltd Clinical Endocrinology (2014), 80, 419–424
Management of acute pituitary apoplexy 421 The commonest presenting symptoms were acute severe headache (87%) and diplopia (38%). Cranial nerve palsies (47%) were the commonest clinical signs at presentation followed by VFD (36%). Other clinical features noted at the time of presentation are listed in Table 1.
total number of n = 32 patients who were managed conservatively during the first week of their presentation. Of these patients, 29 had CNFPAs and three had prolactinomas. Patients with VFD which were improving spontaneously were conservatively managed in the first instance.
Management
Hormonally active tumours. Four of the six prolactinoma cases were treated with early surgery, one with delayed surgery and one lesion was managed conservatively. Both prolactinoma cases not treated with early, emergency surgery received at presentation treatment with dopamine agonists in the form of bromocriptine. All four cases of acromegaly required early transsphenoidal decompressive surgery. All remained biochemically active postoperatively and received medical treatment with dopamine agonists somatostatin analogues, and three required adjuvant radiotherapy. Two patients with acromegaly required multiple surgical procedures, conventional fractionated radiotherapy, stereotactic radiosurgery and subsequently treatment with temozolomide because of the persistent hormonal activity and the invasive and aggressive nature of their tumours.
Surgery. Thirty-three patients (33 of 55, 55%) underwent pituitary surgery at some stage either within 1 week of presentation or elective delayed surgery (>7 days). Indications for surgery were deteriorating visual acuity and persistent or new VF defects. Patients presenting with VF defects (n = 20) were more likely to undergo surgery (15 of 20, 75%) than to be managed expectantly (n = 5/20, 25%). Early emergency surgery (within 7 days of presentation) was performed in 23 patients (16 CNFPAs, four acromegaly and three prolactinomas). Ten of these patients had VFD, 13 had ≥1 cranial nerve palsy and five had both. Eleven patients had only small residual adenoma on latest MRI scan requiring surveillance only, whereas 10 underwent subsequent radiotherapy. Ten patients (182%) underwent delayed elective surgery (after 7 days of presentation, range 8 days to 3 years, mean time to elective surgery 266 days). Indications for elective surgery included further enlargement of the adenoma on subsequent surveillance imaging and new or deteriorating VFD. Nine patients had CNFPAs, and one had a prolactinoma. Three had VFD (two fully/near fully resolved), and five presented with ≥1 cranial nerve palsy (four fully/near fully resolved). Conservative management. Twenty-two patients (22/55, 40%) were managed conservatively and required no surgical intervention throughout their period of follow-up. Ten of these patients had VFD at presentation, which improved spontaneously within the first week of expectant monitoring, 15 presented with ≥1 cranial nerve palsy and two had both. As mentioned previously, ten patients of 55 (182%) underwent delayed elective surgery (after 7 days of presentation), giving a
Table 1. Presenting clinical symptoms and signs Clinical feature
Number (%)
Acute headache Cranial nerve palsy 3rd nerve palsy 6th nerve palsy Multiple palsies Diplopia Visual field defect Vomiting Photophobia Nausea Facial pain/trigeminal neuralgia Collapse
48 26 16 5 5 21 20 14 10 9 8
© 2013 John Wiley & Sons Ltd Clinical Endocrinology (2014), 80, 419–424
(873) (473)
(382) (364) (255) (182) (164) (145)
2 (36)
Steroid treatment. Details of steroid treatment during the acute presentation were available for 31 patients. Twenty-seven patients (87%) received some form of steroid therapy during the acute phase. Nine patients received dexamethasone. Eight of them had very large tumours with optic chiasmal compression or significant extra-/parasellar extension. Two patients in this group were managed conservatively, five underwent emergency surgery and one patient underwent elective surgery. In the five patients who received dexamethasone and underwent emergency surgery, all VFD and cranial nerve palsies resolved, but it is difficult to ascertain if these favourable neuro-ophthalmic outcomes were attributable to the effects of dexamethasone or the effects of emergency decompressive surgery. No evidence of excess mortality or morbidity was noted in the dexamethasonetreated group. All other patients received stress doses of hydrocortisone except one who continued prednisolone, which had been previously prescribed for polymyalgia rheumatica. Pituitary Apoplexy Score. Pituitary Apoplexy Score was calculated retrospectively for 46 patients in whom all the required baseline data were available. The mean PAS for the group which was treated with early surgery was 38 (range, 1–6), while the mean score for the group managed conservatively or with delayed surgery was 18 (range, 0–5) (Table 2). From the group of patients treated with early emergency surgery, 73% had a PAS of four or more, while only 13% of the conservative management/delayed surgery group had a PAS of four or more. Outcomes. There were no deaths related to pituitary apoplexy in our series. Overall, 65% of patients had complete resolution of their VF defects, while 85% had at least partial improvement of their VF defects. When the whole group was considered, cranial nerve palsies resolved completely in 23 patients, and there was at least partial resolution in 26 by the time of their last follow-up.
422 S. Bujawansa et al. Table 2. Total PAS and individual components of the PAS for the two treatment groups
PAS variable Total PAS (mean) Score for GCS Score for visual acuity Score for visual fields Score for cranial nerve palsies
Early surgery group (n = 19)
Delayed surgery/ conservative group (n = 27)
P-value
38 04 11 17 06
18 02 04 05 06
