Clinical Endocrinology (2015) 83, 622–628
doi: 10.1111/cen.12813
ORIGINAL ARTICLE
Risk of catecholaminergic crisis following glucocorticoid administration in patients with an adrenal mass: a literature review Catherine Barrett*, Stan H.M. van Uum* and Jacques W.M. Lenders†,‡ *Department of Medicine, Schulich School of Medicine and Dentistry, Western University, London, ON, Canada, †Department of Internal Medicine III, University Hospital Carl Gustav Carus, Technical University Dresden, Germany and ‡Department of Internal Medicine, Radboud University Medical Center, Nijmegen, The Netherlands
Summary Background Glucocorticoids as diagnostic or therapeutic agents have been reported to carry an increased risk of catecholaminergic crisis (CC) in patients with pheochromocytoma or paraganglioma (PPGL). Methods We searched literature databases using the following terms: pheochromocytoma, paraganglioma, adrenal incidentaloma, steroids, glucocorticoids, dexamethasone suppression test (DST), hypertensive crisis, cosyntropin and CRH. From all published case reports (1962–2013), we reviewed medical history, presenting symptoms, dose and route of steroid administration, location and size of adrenal mass, biochemical phenotype and outcome. Results Twenty-five case reports describing a CC were identified. Three patients with an adrenal incidentaloma suffered a CC following high-dose DST, and in one case, this was fatal. In two of these patients, biochemical testing missed the diagnosis, and in the third, a DST was done despite elevated urinary metanephrines. No CC has been reported for patients undergoing a lowdose DST. Three of 16 patients who received therapeutic glucocorticoids and four of six patients following cosyntropin testing died. No specific biochemical phenotype was related to adverse events. Conclusions Although a causal relationship cannot be established from this review, it seems prudent to exclude a PPGL in patients with a large incidentaloma or when high-dose DST is considered in a patient with an incidentaloma of any size. Our literature review does not support the need for biochemical testing for PPGL prior to a low-dose (1 mg) DST. Finally, before starting therapeutic glucocorticoids, any clinical signs or symptoms of a potential PPGL should prompt reliable biochemical testing to rule out a PPGL.
Correspondence: Catherine Barrett, Department of Medicine, Schulich School of Medicine and Dentistry, Western University, St. Joseph’s Health Care, Room B5-120, 268 Grosvenor Street, London, Ontario N6A 4V2, Canada. Tel.: 519-646-6170; Fax: 519-646-6058; E-mail: [email protected]
622
(Received 19 January 2015; returned for revision 12 April 2015; finally revised 26 April 2015; accepted 28 April 2015)
Introduction Pheochromocytomas and paragangliomas (PPGL) are rare neuroendocrine tumours arising from the adrenomedullary chromaffin cells or from paravertebral sympathetic ganglia, respectively. The clinical manifestations of these tumours can vary considerably from no or only minor symptoms to full blown life-threatening crisis, induced by strong release of catecholamines into the blood stream. Such a catecholaminergic crisis (CC) is associated with a high mortality and includes severe hypertension, catastrophic cardiovascular incidents and multiorgan failure.1–4 A CC can be elicited by many potential triggers such as drugs and anaesthesia. A potential source of druginduced CC is exogenous glucocorticoids, and several case reports have described serious and even fatal complications in patients with an occult or overt PPGL.5 A typical clinical scenario of glucocorticoid administration is the patient with a newly diagnosed incidentaloma who is scheduled for low-dose (1 mg) dexamethasone suppression testing (DST). Hormonal evaluation, including an overnight 1 mg DST, is recommended in all cases except when imaging characteristics are unequivocal.6,7 Several case reports have, however, suggested a link between glucocorticoid or cosyntropin administration and a subsequent CC.8–24 As a result, some have suggested that biochemical testing for PPGL should be performed prior to proceeding with a DST in patients with an incidentally discovered adrenal mass.9,25 In this study, we reviewed all cases published in the last fifty years of patients who suffered from a CC after glucocorticoid or cosyntropin administration in the setting of a known metastatic PPGL or in patients subsequently diagnosed with a pre-existing PPGL. In addition, we characterized these patients according to dose and route of drug administration, time course between drug administration and CC, eventual presenting symptoms, biochemical phenotype, tumour location and final outcome. © 2015 John Wiley & Sons Ltd
Glucocorticoids in pheochromocytoma 623 Dexamethasone suppression testing
Methods We conducted a search of PubMed, MEDLINE and Embase, spanning a time period from 21 July 1962–11 February 2013 to find English-language case reports or small case series using the following search terms: (pheochromocytoma) OR (paraganglioma) OR (adrenal incidentaloma) AND (steroids) OR (glucocorticoids) OR (dexamethasone suppression test) OR (hypertensive crisis) OR (cosyntropin) OR (ACTH) OR (CRH). We also checked reference lists of all papers that were found. All identified published cases were patients who suffered from a CC following administration of glucocorticoid or cosyntropin. For the purpose of this review, CC refers to any crisis involving uncontrolled severe hypertension, serious cardiovascular accidents (cardiac ischaemia, heart failure, arrhythmias, stroke or transient ischaemic attack), organ dysfunction, shock or death in a patient with PPGLs. Data extraction Two of us (CB, JWML) retrieved data independently from full text papers. Differences in interpretation were resolved by consensus. We retrieved detailed information on medical history, presenting signs and symptoms, dose and route of corticosteroid administration, location and size of adrenal mass, biochemical phenotype and treatment indication. We stored the data of all eligible cases in a database for further analysis.
Results In total, we identified 25 case reports describing a CC following glucocorticoid or cosyntropin administration. A CC occurred in three patients evaluated for an adrenal incidentaloma and in 16 patients with PPGL after therapeutic glucocorticoid administration. Six patients presenting with CC had recently undergone a cosyntropin test. Of the 16 cases who received therapeutic glucocorticoid administration, two patients were known to have metastatic PPGL at the time of steroid administration and the remainder were diagnosed with PPGL following the CC.
Three case reports identified a CC following a high-dose DST (2 mg PO q6 h for 48 h), and one of them, a young woman of 26 years, died 10 h after onset of symptoms (Table 1).9,23 Of these three patients, one had a history of hypertension and another patient was known to have both hypertension and diabetes. No other signs or symptoms of PPGL were documented at presentation in any of these cases. Tumour size was >40 cm in all patients and density was >28 Hounsfield units (HU) in two patients and reported as low attenuation in the third. Urinary levels of total metanephrines or vanillylmandelic acid (VMA) were falsely normal in two patients (fatal in one case). A 24-h urine collection for total metanephrines was elevated in one patient, but a 2-day high-dose DST was carried out regardless as per the institutional protocol. The diagnosis of PPGL was verified by autopsy in the deceased patient and by histological examination of the surgically removed tumour in the other patient. There are no reported cases of CC following a low-dose (1 mg) overnight DST. Therapeutic glucocorticoid administration Sixteen cases of CC following therapeutic glucocorticoid administration were identified (Table 2). The indications for steroid treatment included sensorineural hearing loss, polymyalgia rheumatica/giant cell arteritis, cough/airway congestion/asthma, headache, joint pain and chronic dermatitis. The type of steroid (prednisolone, betamethasone, dexamethasone, prednisone and methylprednisolone) and route (oral, intravenous, intra-articular and intramuscular) varied amongst reports. Past medical history was not fully documented in the majority of cases. Tumour location was in the right adrenal gland in nine patients, left adrenal gland in four, extra-adrenal in two and unknown in one patient. The elapsed time between glucocorticoid administration and development of signs and symptoms of a CC varied widely between 8 and 96 h in most patients. Five patients endorsed symptoms compatible with pheochromocytoma prior to steroid administration. Three patients died.
Table 1. Patients with adrenal incidentaloma and catecholaminergic crisis following dexamethasone suppression testing Age and Gender Yi23
61F
Rosas9
26F
39F
PMHX
Steroid Dose and Route
HTN obesity
Dexamethasone 2 mg PO q6 h
8
Dexamethasone 2 mg PO q6 h
36
Dexamethasone 2 mg PO q6 h
5
DM HTN
Time to CC (h)
Presenting Signs Symptoms at CC
Management and Outcomes
Tumour Site
Size (cm)
Biochemical Profile
N/V, palpitations, anxiety, HTN
Phentolamine and sodium nitroprusside. Survived Intensive therapy. Fatal
RA
31 9 41
E + NE
RA
45
N/A
LA
80
N/A
Anxiety, shock, HTN, tachycardia, CHF hyperglycaemia, metabolic acidosis, N/V, H/A, HTN, CP, tachycardia
IV vasodilators and beta blockers. Survived
CC, catecholaminergic crisis; M, male; F, female; PMHX, past medical history; PO, oral administration; HTN, hypertension; DM, diabetes; N/V, nausea and vomiting; CHF, congestive heart failure; H/A, headache; CP, chest pain; RA, right adrenal; LA, left adrenal; E, adrenalin; NE, noradrenalin, N/A, not available. © 2015 John Wiley & Sons Ltd Clinical Endocrinology (2015), 83, 622–628
624 C. Barrett et al. A 63-year-old male patient with diabetes and hypertension who developed sensorineural hearing loss was treated with intravenous prednisolone. Marked hypertension and tachycardia developed within 10 h and fatal cardiac arrest occurred.8 Retrospectively, an adrenal tumour was present on previous imaging. A 33-year-old female patient with fever and headache was treated with methylprednisolone and dexamethasone. Progressive neurologic deterioration occurred and was fatal 3 days later.15 The last patient was a 46-year-old male patient who developed chest pain, hypertension and tachycardia following steroid administration for chronic dermatitis.15 Two patients with metastatic PPGL developed a CC following glucocorticoid administration. In one patient with metastatic paraganglioma to bone, hypertension and tachycardia developed within 8 h of intravenous methylprednisolone for pretreatment of MIBG therapy.9 A 58-year-old male patient with a history of right adrenalectomy at age 38 for pheochromocytoma and known metastatic involvement of liver and peritoneum was initiated on prednisone, vincristine and cyclophosphamide for a new diagnosis of poorly differentiated lymphocytic lymphoma. Significant hypertension and tachycardia developed within 48 h and resolved following completion of the chemotherapy cycle.16 Cosyntropin Concerning cosyntropin, we found only one case report that describes a CC in a patient after administration of cosyntropin (250 lg synthetic ACTH(1–24)) but along with 1 mg dexamethasone intravenously 19 (Table 3). Five other patients who were published before 1970 received a partially purified animal product. Of these latter five cases, four patients did not survive.20–22
Discussion The present series of published cases that developed a CC after diagnostic or therapeutic administration of pharmacological doses of glucocorticoids or cosyntropin shows a high fatality rate of 32%, supporting the contention that glucocorticoids and PPGL is a dangerous liaison. The presenting signs and symptoms of a CC were quite diverse and included hypertension, palpitations, arrhythmias, headache, nausea, vomiting, cardiac ischaemia, pulmonary oedema and shock. In most cases, admission to the intensive care unit was necessary. Two cases required insertion of an intra-aortic balloon pump for hemodynamic support. The type of glucocorticoid, dose and route of administration were variable as was also the time to symptom onset, suggesting considerable clinical heterogeneity of this interaction. Two factors that may play a role in the risk of a CC in incidentaloma patients are tumour size and the dose used for DST. All three incidentaloma patients with CC had an adrenal tumour size of >4 cm, and in all cases, a high-dose DST was performed. Therefore, it is impossible to delineate the individual relevance of these factors. Conversely, no case reports have been published of patients who underwent a low-dose (1 mg) DST or who had an adrenal tumour size of
doi: 10.1111/cen.12813
ORIGINAL ARTICLE
Risk of catecholaminergic crisis following glucocorticoid administration in patients with an adrenal mass: a literature review Catherine Barrett*, Stan H.M. van Uum* and Jacques W.M. Lenders†,‡ *Department of Medicine, Schulich School of Medicine and Dentistry, Western University, London, ON, Canada, †Department of Internal Medicine III, University Hospital Carl Gustav Carus, Technical University Dresden, Germany and ‡Department of Internal Medicine, Radboud University Medical Center, Nijmegen, The Netherlands
Summary Background Glucocorticoids as diagnostic or therapeutic agents have been reported to carry an increased risk of catecholaminergic crisis (CC) in patients with pheochromocytoma or paraganglioma (PPGL). Methods We searched literature databases using the following terms: pheochromocytoma, paraganglioma, adrenal incidentaloma, steroids, glucocorticoids, dexamethasone suppression test (DST), hypertensive crisis, cosyntropin and CRH. From all published case reports (1962–2013), we reviewed medical history, presenting symptoms, dose and route of steroid administration, location and size of adrenal mass, biochemical phenotype and outcome. Results Twenty-five case reports describing a CC were identified. Three patients with an adrenal incidentaloma suffered a CC following high-dose DST, and in one case, this was fatal. In two of these patients, biochemical testing missed the diagnosis, and in the third, a DST was done despite elevated urinary metanephrines. No CC has been reported for patients undergoing a lowdose DST. Three of 16 patients who received therapeutic glucocorticoids and four of six patients following cosyntropin testing died. No specific biochemical phenotype was related to adverse events. Conclusions Although a causal relationship cannot be established from this review, it seems prudent to exclude a PPGL in patients with a large incidentaloma or when high-dose DST is considered in a patient with an incidentaloma of any size. Our literature review does not support the need for biochemical testing for PPGL prior to a low-dose (1 mg) DST. Finally, before starting therapeutic glucocorticoids, any clinical signs or symptoms of a potential PPGL should prompt reliable biochemical testing to rule out a PPGL.
Correspondence: Catherine Barrett, Department of Medicine, Schulich School of Medicine and Dentistry, Western University, St. Joseph’s Health Care, Room B5-120, 268 Grosvenor Street, London, Ontario N6A 4V2, Canada. Tel.: 519-646-6170; Fax: 519-646-6058; E-mail: [email protected]
622
(Received 19 January 2015; returned for revision 12 April 2015; finally revised 26 April 2015; accepted 28 April 2015)
Introduction Pheochromocytomas and paragangliomas (PPGL) are rare neuroendocrine tumours arising from the adrenomedullary chromaffin cells or from paravertebral sympathetic ganglia, respectively. The clinical manifestations of these tumours can vary considerably from no or only minor symptoms to full blown life-threatening crisis, induced by strong release of catecholamines into the blood stream. Such a catecholaminergic crisis (CC) is associated with a high mortality and includes severe hypertension, catastrophic cardiovascular incidents and multiorgan failure.1–4 A CC can be elicited by many potential triggers such as drugs and anaesthesia. A potential source of druginduced CC is exogenous glucocorticoids, and several case reports have described serious and even fatal complications in patients with an occult or overt PPGL.5 A typical clinical scenario of glucocorticoid administration is the patient with a newly diagnosed incidentaloma who is scheduled for low-dose (1 mg) dexamethasone suppression testing (DST). Hormonal evaluation, including an overnight 1 mg DST, is recommended in all cases except when imaging characteristics are unequivocal.6,7 Several case reports have, however, suggested a link between glucocorticoid or cosyntropin administration and a subsequent CC.8–24 As a result, some have suggested that biochemical testing for PPGL should be performed prior to proceeding with a DST in patients with an incidentally discovered adrenal mass.9,25 In this study, we reviewed all cases published in the last fifty years of patients who suffered from a CC after glucocorticoid or cosyntropin administration in the setting of a known metastatic PPGL or in patients subsequently diagnosed with a pre-existing PPGL. In addition, we characterized these patients according to dose and route of drug administration, time course between drug administration and CC, eventual presenting symptoms, biochemical phenotype, tumour location and final outcome. © 2015 John Wiley & Sons Ltd
Glucocorticoids in pheochromocytoma 623 Dexamethasone suppression testing
Methods We conducted a search of PubMed, MEDLINE and Embase, spanning a time period from 21 July 1962–11 February 2013 to find English-language case reports or small case series using the following search terms: (pheochromocytoma) OR (paraganglioma) OR (adrenal incidentaloma) AND (steroids) OR (glucocorticoids) OR (dexamethasone suppression test) OR (hypertensive crisis) OR (cosyntropin) OR (ACTH) OR (CRH). We also checked reference lists of all papers that were found. All identified published cases were patients who suffered from a CC following administration of glucocorticoid or cosyntropin. For the purpose of this review, CC refers to any crisis involving uncontrolled severe hypertension, serious cardiovascular accidents (cardiac ischaemia, heart failure, arrhythmias, stroke or transient ischaemic attack), organ dysfunction, shock or death in a patient with PPGLs. Data extraction Two of us (CB, JWML) retrieved data independently from full text papers. Differences in interpretation were resolved by consensus. We retrieved detailed information on medical history, presenting signs and symptoms, dose and route of corticosteroid administration, location and size of adrenal mass, biochemical phenotype and treatment indication. We stored the data of all eligible cases in a database for further analysis.
Results In total, we identified 25 case reports describing a CC following glucocorticoid or cosyntropin administration. A CC occurred in three patients evaluated for an adrenal incidentaloma and in 16 patients with PPGL after therapeutic glucocorticoid administration. Six patients presenting with CC had recently undergone a cosyntropin test. Of the 16 cases who received therapeutic glucocorticoid administration, two patients were known to have metastatic PPGL at the time of steroid administration and the remainder were diagnosed with PPGL following the CC.
Three case reports identified a CC following a high-dose DST (2 mg PO q6 h for 48 h), and one of them, a young woman of 26 years, died 10 h after onset of symptoms (Table 1).9,23 Of these three patients, one had a history of hypertension and another patient was known to have both hypertension and diabetes. No other signs or symptoms of PPGL were documented at presentation in any of these cases. Tumour size was >40 cm in all patients and density was >28 Hounsfield units (HU) in two patients and reported as low attenuation in the third. Urinary levels of total metanephrines or vanillylmandelic acid (VMA) were falsely normal in two patients (fatal in one case). A 24-h urine collection for total metanephrines was elevated in one patient, but a 2-day high-dose DST was carried out regardless as per the institutional protocol. The diagnosis of PPGL was verified by autopsy in the deceased patient and by histological examination of the surgically removed tumour in the other patient. There are no reported cases of CC following a low-dose (1 mg) overnight DST. Therapeutic glucocorticoid administration Sixteen cases of CC following therapeutic glucocorticoid administration were identified (Table 2). The indications for steroid treatment included sensorineural hearing loss, polymyalgia rheumatica/giant cell arteritis, cough/airway congestion/asthma, headache, joint pain and chronic dermatitis. The type of steroid (prednisolone, betamethasone, dexamethasone, prednisone and methylprednisolone) and route (oral, intravenous, intra-articular and intramuscular) varied amongst reports. Past medical history was not fully documented in the majority of cases. Tumour location was in the right adrenal gland in nine patients, left adrenal gland in four, extra-adrenal in two and unknown in one patient. The elapsed time between glucocorticoid administration and development of signs and symptoms of a CC varied widely between 8 and 96 h in most patients. Five patients endorsed symptoms compatible with pheochromocytoma prior to steroid administration. Three patients died.
Table 1. Patients with adrenal incidentaloma and catecholaminergic crisis following dexamethasone suppression testing Age and Gender Yi23
61F
Rosas9
26F
39F
PMHX
Steroid Dose and Route
HTN obesity
Dexamethasone 2 mg PO q6 h
8
Dexamethasone 2 mg PO q6 h
36
Dexamethasone 2 mg PO q6 h
5
DM HTN
Time to CC (h)
Presenting Signs Symptoms at CC
Management and Outcomes
Tumour Site
Size (cm)
Biochemical Profile
N/V, palpitations, anxiety, HTN
Phentolamine and sodium nitroprusside. Survived Intensive therapy. Fatal
RA
31 9 41
E + NE
RA
45
N/A
LA
80
N/A
Anxiety, shock, HTN, tachycardia, CHF hyperglycaemia, metabolic acidosis, N/V, H/A, HTN, CP, tachycardia
IV vasodilators and beta blockers. Survived
CC, catecholaminergic crisis; M, male; F, female; PMHX, past medical history; PO, oral administration; HTN, hypertension; DM, diabetes; N/V, nausea and vomiting; CHF, congestive heart failure; H/A, headache; CP, chest pain; RA, right adrenal; LA, left adrenal; E, adrenalin; NE, noradrenalin, N/A, not available. © 2015 John Wiley & Sons Ltd Clinical Endocrinology (2015), 83, 622–628
624 C. Barrett et al. A 63-year-old male patient with diabetes and hypertension who developed sensorineural hearing loss was treated with intravenous prednisolone. Marked hypertension and tachycardia developed within 10 h and fatal cardiac arrest occurred.8 Retrospectively, an adrenal tumour was present on previous imaging. A 33-year-old female patient with fever and headache was treated with methylprednisolone and dexamethasone. Progressive neurologic deterioration occurred and was fatal 3 days later.15 The last patient was a 46-year-old male patient who developed chest pain, hypertension and tachycardia following steroid administration for chronic dermatitis.15 Two patients with metastatic PPGL developed a CC following glucocorticoid administration. In one patient with metastatic paraganglioma to bone, hypertension and tachycardia developed within 8 h of intravenous methylprednisolone for pretreatment of MIBG therapy.9 A 58-year-old male patient with a history of right adrenalectomy at age 38 for pheochromocytoma and known metastatic involvement of liver and peritoneum was initiated on prednisone, vincristine and cyclophosphamide for a new diagnosis of poorly differentiated lymphocytic lymphoma. Significant hypertension and tachycardia developed within 48 h and resolved following completion of the chemotherapy cycle.16 Cosyntropin Concerning cosyntropin, we found only one case report that describes a CC in a patient after administration of cosyntropin (250 lg synthetic ACTH(1–24)) but along with 1 mg dexamethasone intravenously 19 (Table 3). Five other patients who were published before 1970 received a partially purified animal product. Of these latter five cases, four patients did not survive.20–22
Discussion The present series of published cases that developed a CC after diagnostic or therapeutic administration of pharmacological doses of glucocorticoids or cosyntropin shows a high fatality rate of 32%, supporting the contention that glucocorticoids and PPGL is a dangerous liaison. The presenting signs and symptoms of a CC were quite diverse and included hypertension, palpitations, arrhythmias, headache, nausea, vomiting, cardiac ischaemia, pulmonary oedema and shock. In most cases, admission to the intensive care unit was necessary. Two cases required insertion of an intra-aortic balloon pump for hemodynamic support. The type of glucocorticoid, dose and route of administration were variable as was also the time to symptom onset, suggesting considerable clinical heterogeneity of this interaction. Two factors that may play a role in the risk of a CC in incidentaloma patients are tumour size and the dose used for DST. All three incidentaloma patients with CC had an adrenal tumour size of >4 cm, and in all cases, a high-dose DST was performed. Therefore, it is impossible to delineate the individual relevance of these factors. Conversely, no case reports have been published of patients who underwent a low-dose (1 mg) DST or who had an adrenal tumour size of
