- Email: [email protected]
Catecholamine Surge during Image-Guided Ablation of Adrenal Gland Metastases: Predictors, Consequences, and Recommendations for Management
CLINICAL STUDY
Catecholamine Surge during Image-Guided Ablation of Adrenal Gland Metastases: Predictors, Consequences, and Recommendations for Management Florian J. Fintelmann, MD, FRCPC, Kemal Tuncali, MD, Stefan Puchner, MD, Debra A. Gervais, MD, Ashraf Thabet, MD, Paul B. Shyn, MD, Ronald S. Arellano, MD, Servet Tatli, MD, Peter R. Mueller, MD, Stuart G. Silverman, MD, and Raul N. Uppot, MD
ABSTRACT Purpose: To identify retrospectively predictors of catecholamine surge during image-guided ablation of metastases to the adrenal gland. Materials and Methods: Between 2001 and 2014, 57 patients (39 men, 18 women; mean age, 65 y ⫾ 10; age range, 41–81 y) at two academic medical centers underwent ablation of 64 metastatic adrenal tumors from renal cell carcinoma (n ¼ 27), lung cancer (n ¼ 23), melanoma (n ¼ 4), colorectal cancer (n ¼ 3), and other tumors (n ¼ 7). Tumors measured 0.7–11.3 cm (mean, 4 cm ⫾ 2.5). Modalities included cryoablation (n ¼ 38), radiofrequency (RF) ablation (n ¼ 20), RF ablation with injection of dehydrated ethanol (n ¼ 10), and microwave ablation (n ¼ 4). Fisher exact test, univariate, and multivariate logistical regression analysis was used to evaluate factors predicting hypertensive crisis (HC). Results: HC occurred in 31 sessions (43%). Ventricular tachycardia (n ¼ 1), atrial fibrillation (n ¼ 2), and troponin leak (n ¼ 4) developed during HC episodes. HC was significantly associated with maximum tumor diameter r 4.5 cm (odds ratio [OR], 26.36; 95% confidence interval [CI], 5.26–131.99; P o .0001) and visualization of normal adrenal tissue on computed tomography (CT) or magnetic resonance (MR) imaging before the procedure (OR, 8.38; 95% CI, 2.67–25.33; P o .0001). No HC occurred during ablation of metastases in previously irradiated or ablated adrenal glands. Conclusions: Patients at high risk of catecholamine surge during ablation of non–hormonally active adrenal metastases can be identified by the presence of normal adrenal tissue and tumor diameter r 4.5 cm on pre-procedure CT or MR imaging.
ABBREVIATIONS GETA = general endotracheal anesthesia, HC = hypertensive crisis, MAC = monitored anesthesia care, RF = radiofrequency, SBP = systolic blood pressure
From the Department of Radiology (F.J.F., S.P., D.A.G., A.T., R.S.A., P.R.M., R.N.U.), Massachusetts General Hospital, 55 Fruit Street, FND-202, Boston, MA 02111; and Department of Radiology (K.T., P.B.S., S.T., S.G.S.), Brigham and Women’s Hospital, Boston, Massachusetts. Received July 27, 2015; final revision received October 12, 2015; accepted November 10, 2015. Address correspondence to F.J.F.; E-mail: ffi[email protected] K.T. received grants from Canon USA (Melville, New York). P.R.M. is a consultant for Cook, Inc (Bloomington, Indiana). None of the other authors have identified a conflict of interest. From the SIR 2014 Annual Meeting. & SIR, 2015 J Vasc Interv Radiol 2015; XX:]]]–]]] http://dx.doi.org/10.1016/j.jvir.2015.11.034
A severe increase in blood pressure after thermal injury of adrenal gland tissue was first described by Onik et al (1) during radiofrequency (RF) ablation of a posterior right lobe liver lesion in close proximity to a normal right adrenal gland. An acute increase in blood pressure 4 180 mm Hg or diastolic blood pressure 4 110 mm Hg, or both, is referred to as hypertensive crisis (HC). HC and cardiac irritability has been reported during thermal ablation of functional primary adrenal tumors and nonfunctional metastases to the adrenal gland with incidence up to 46% (2–9). Yamakado et al (10,11) firmly established the link between the onset of HC during thermal ablation of adrenal gland tumors and
2
’
Catecholamine Surge during Adrenal Gland Metastasis Ablation
catecholamine surge by measuring the catecholamine blood levels before, during, and after RF ablation of adrenal glands in swine. As a result, medication with alpha and beta blockers was suggested before imageguided ablation of functional and nonfunctional adrenal tumors (7–9,12). In cases of functional tumors, this recommendation mirrors guidelines in the surgical literature on pheochromocytomas (13). However, in cases of nonfunctional tumors such as metastases, no guidelines exist. It is unclear how to assess the need for medication before the procedure in a particular patient without better understanding the risk factors for HC. The purpose of this study was to identify predictors of catecholamine surge during image-guided ablation of metastases to the adrenal gland, to report consequences, and to guide management before, during, and after the procedure.
MATERIALS AND METHODS Patient Cohort The institutional review board at two participating academic medical centers approved this retrospective study. Patients included 57 adults (39 men and 18 women) with a mean age of 65 years (SD 10; range, 41–81 y) with adrenal metastases who underwent imageguided ablation at either of two large academic medical centers between January 2001 and March 2014. The series included one patient with bilateral adrenal gland tumors from a previously published case report (14). There were 35 patients (61%) who had received systemic chemotherapy at some point during the course of their disease, and four of these patients (7%) were on systemic therapy at the time of ablation.
Disease Burden and Index Tumors In 21 patients (37%), the adrenal gland was the only site of disease; 36 patients (63%) had metastases to other
Figure 1. Distribution of maximum axial adrenal tumor diameter.
Fintelmann et al
’
JVIR
organs. Three patients had two lesions in the same gland, and four patients had at least one lesion in each adrenal gland. Proof of metastatic involvement was available before ablation based on either biopsy in 45 patients (70%) or documentation of growth and fluorine18 fluorodeoxyglucose uptake on positron emission tomography (PET) combined with computed tomography (CT) in 19 patients (30%) (15). The total number of adrenal tumors was 64, and most metastases were from renal cell cancer and non–small cell lung cancer (Table 1). Mean maximum tumor diameter was 4 cm (SD 2.5; range, 0.7–11.3 cm) (Fig 1).
Ablation Technique There were 72 ablation sessions performed by 10 operators. Treatment indications included cure in 28 cases of solitary metastasis to the adrenal gland (39%), cytoreduction in 28 cases with two or more metastases involving organs other than the adrenal gland (39%), and palliation in 16 cases for large adrenal masses causing abdominal or flank pain (22%). Ten sessions (14%) were performed on lesions in previously irradiated (n ¼ 1) or ablated (n ¼ 9) adrenal glands. Treatment
Table 1 . Adrenal Tumor Histology Adrenal Tumor Histology Renal cell cancer
Values (n ¼ 64 Tumors) 42% (27/64)
Non–small cell lung cancer
36% (23/64)
Melanoma Colorectal cancer
6.3% (4/64) 4.7% (3/64)
Endometrial cancer
3% (2/64)
Breast cancer Ovarian cancer
1.5% (1/64) 1.5% (1/64)
Transitional cell carcinoma
1.5% (1/64)
Malignant fibrous histiocytoma Hepatocellular carcinoma
1.5% (1/64) 1.5% (1/64)
Volume XX
’
Number X
’
Month
’
2015
3
modalities included 38 cryoablations (53%), 20 RF ablations (28%), 10 RF ablation procedures combined with injection of up to 10 mL of dehydrated ethanol (14%), and 4 microwave ablations (6%). Cryoablation was performed using the SeedNet (Galil Medical, Inc, Arden Hills, Minnesota) system. RF ablation was performed using a single straight electrode (Cool-tip; Covidien, Boulder, Colorado) in four sessions, a cluster electrode (Radionics, Inc, Burlington, Massachusetts) in 14 sessions, a multitined Starburst electrode (RITA Medical Systems, Inc, Mountain View, California) in six sessions, and a multitined Leveen electrode (Boston Scientific, Marlborough, Massachusetts) in six sessions. Microwave ablation was performed using the Certus 140 system (NeuWave Medical; Madison, Wisconsin). The tumors were ablated per manufacturer recommendations: temperature-based (Starburst electrode), impedance (Leveen electrode), or both (Cool-tip) as well as time-based for microwave ablation and double-freeze cycle for cryoablation. The goal was to extend the zone of ablation at least 5 mm beyond the tumor margin. In patients referred for ablation of symptomatic large adrenal masses, the goal was to diminish tumor burden as much as possible to decrease pain rather than achieve eradication. After ablation, the applicators were removed.
Image Guidance CT, magnetic resonance (MR) imaging, and PET/CT (Discovery ST; GE Healthcare, Milwaukee, Wisconsin) were used to guide placement of applicators (Table 2). CT scanners had 16 (LightSpeed; GE Healthcare) or 40 (SOMATOM Sensation Open; Siemens Medical Solutions, Forchheim, Germany) detector rows. MR imaging was either 0.5 tesla (Signa SP; GE Healthcare) or 3 tesla (Siemens MAGNETOM Verio; Siemens Healthcare, Erlangen, Germany). Non–contrast enhanced CT or MR images of the treatment area were obtained before, during, and immediately after ablation to plan and monitor treatment and to assess for complications. Image slice thickness was 5 mm on CT and 4 mm on MR imaging examinations. MR imaging sequences were either spoiled gradient recalled or T2-weighted halfFourier acquisition single-shot turbo spin echo. When Table 2 . Treatment Modalities and Image Guidance Used for Ablation Sessions MR CT Imaging PET/CT Totals Cryoablation RF ablation
24 20
12 0
2 0
38 20
RF ablation and EtOH injection
10
0
0
10
Microwave ablation Totals
3 57
0 12
1 3
4
EtOH ¼ dehydrated ethanol; PET/CT = positron emission tomography/computed tomography; RF = radiofrequency.
PET/CT was used for image guidance, the protocol described by Ryan et al (16) was followed.
Care during the Procedure Ablation was performed after induction of general endotracheal anesthesia (GETA) in 57 sessions (79%). Monitored anesthesia care (MAC) was used for 12 sessions (17%), and conscious sedation was used for three sessions (4%). An indwelling radial artery catheter was used for intraprocedural hemodynamic monitoring in addition to a blood pressure cuff in 27 of 69 sessions performed under MAC or GETA. If patients were treated under MAC or GETA, they were admitted to the hospital for overnight observation.
Care before and after Procedure In 20 patients, consultation with an endocrinologist occurred before and after the procedure to establish recommendations for medication before the ablation, periprocedural management of catecholamine surge, and follow-up after the procedure to diagnose and manage any adrenal insufficiency. Adrenergic blockade was administered before 14 sessions (19%). Early in this series, patients were admitted for 3 days before the procedure for gradual administration of alphaadrenergic and beta-adrenergic blockade. Because this medication protocol may result in hypotension, intravenous fluid was administered simultaneously. One of two patients treated this way did not tolerate the medications because of hypotension and could not receive the full dose. The other patient needed diuretics for fluid overload secondary to the medication protocol. Later in this series, low-dose oral adrenergic blockade was administered in the outpatient setting after consultation with an endocrinologist. Doxazosin mesylate (Teva Pharmaceuticals, North Wales, Pennsylvania) 1 mg administered by mouth once daily for 14 days before the procedure and long-acting metoprolol succinate (Wockhardt USA, Parsippany, New Jersey) 25 mg by mouth once daily for 4 days before the procedure does not result in significant side effects in most cases. Technical success was assessed using contrastenhanced MR imaging (71%), CT (15%), or PET/CT (14%) at the first imaging follow-up, which occurred on average 10 days after the procedure (SD 16; range, 0– 78). All but three patients were seen in the interventional radiology clinic shortly after the imaging study. Clinical follow-up beyond this first visit was available through notes in the medical chart for a mean of 26 months (SD 27; range, 1–116). The three patients who did not return were considered lost to follow-up.
Data Analysis A radiologist blinded to the purpose of this study (S.P.; 9 years of experience) measured and recorded the maximum axial diameter of each adrenal tumor on the CT or
4
’
Catecholamine Surge during Adrenal Gland Metastasis Ablation
MR images obtained immediately before the ablation. The same radiologist recorded whether or not any normal adrenal gland tissue was visible adjacent to the tumor as illustrated in Figure 2. Baseline and peak intraprocedural blood pressure and heart rate were extracted from the sedation and anesthesia portion of the medical record (F.J.F.; 6 years of experience) and correlated with the timing of the ablation. Based on the work of Yamakado et al (10), a systolic blood pressure (SBP) spike 4 180 mm Hg during the procedure in close temporal association with onset of cell death (heating during RF ablation and microwave ablation, thawing during cryoablation) was chosen as a surrogate marker for catecholamine-induced HC. To qualify as a spike, the peak SBP had to increase acutely by at least 50 mm Hg above the baseline determined at the beginning of the procedure. All adverse events were categorized using Common Terminology Criteria for Adverse Events Version 4.03 (17) and graded as mild (grade 1), moderate (grade 2), severe (grade 3), or life-threatening (grade 4), with specific parameters according to the organ system involved.
Statistical Methods Statistical analyses were performed using the SAS software package (SAS Institute, Cary, North Carolina). Fisher exact test was used to test for an association between continuous blood pressure monitoring with an arterial line and the detection of HC. Univariate and multivariate logistical regression analysis was used to detect associations between tumor size, visualization of normal adrenal tissue on images obtained before the procedure, ablation modality, medication administered before the procedure, prior treatment of the adrenal gland, concurrent systemic therapy, and intraprocedural
Fintelmann et al
’
JVIR
HC. Odds ratio with corresponding 95% confidence interval and relative risk was calculated. P values are two-tailed. P values o .05 were considered statistically significant.
RESULTS The most common serious adverse event during ablation of adrenal gland metastases was HC in 31 sessions (43%). No HC occurred during the 10 ablations performed on metastases in previously ablated or irradiated adrenal glands. The mean duration of the HC was 4 minutes (SD 4.4; range, 1–15 min). Peak SBP was 182– 320 mm Hg (average 224 mm Hg, SD 33). HC was accompanied by elevated cardiac troponin T in four patients with peak levels of 0.18 ng/mL, 0.4 ng/mL, 0.6 ng/mL, and 0.73 ng/mL, indicating cardiomyocyte damage. Duration of HC in these patients was 1–5 minutes, and the peak SBP was 186–260 mm Hg. Three of these four patients also experienced cardiac arrhythmia. A 64year-old man developed ventricular tachycardia after the onset of HC following the third ablation of a 3.1-cm solitary renal cell cancer metastasis in the adrenal gland using RF. Vital signs remained stable and the cardiac rhythm normalized within 2 minutes without further intervention. The patient was admitted for observation and discharged home in good condition after a negative nuclear cardiac stress test 2 days later. Two patients developed atrial fibrillation after the onset of HC. Both patients returned to normal rhythm within 24 hours after treatment with verapamil hydrochloride (Mylan, Canonsburg, Pennsylvania) and digoxin (GlaxoSmithKline, Philadelphia, Pennsylvania) and had normal echocardiograms and nuclear cardiac stress tests. One of these two patients experienced chest pain and underwent coronary
Figure 2. Axial contrast-enhanced CT image shows a 2-cm tumor (arrow) in the lateral limb of the left adrenal gland (a). Residual normal adrenal gland tissue (arrow) is demonstrated in the same patient at the same time, approximately 3 cm superiorly (b).
Volume XX
’
Number X
’
Month
’
2015
5
angiography for suspected non–ST segment elevation myocardial infarction. Findings included complex left main and three-vessel coronary atherosclerosis. Patients were discharged home in o 24 hours after 54 (75%) sessions. Adverse events triggered unplanned admissions after 18 sessions (25%) with an average length of stay of 3 days (range, 2–10 d). Anesthesiologists were involved in cases performed under MAC and GETA, which allowed for invasive blood pressure monitoring with an arterial line. Fisher exact test revealed a significant association (P ¼ .049) between the use of an arterial line and the detection of HC. Antihypertensive medications administered by anesthesiologists during HC included one or more intravenous beta blockers—labetalol hydrochloride (Hospira, Lake Forest, Illinois), esmolol hydrochloride (Mylan), metoprolol tartrate (Hospira), and propranolol hydrochloride (West-Ward Pharmaceuticals, Eatontown, New Jersey) in order of decreasing frequency—87% of the time (27 of 31 sessions), whereas the alpha blocker phentolamine mesylate (Bedford Laboratories, Bedford, Ohio) was administered only once instead. In addition to the adrenergic blocker, intravenous sodium nitroprusside (Valeant Pharmaceuticals, Bridgewater, New Jersey) was administered 23% (seven of 31 sessions) of the time, and hydralazine hydrochloride (Novation, Irving, Texas) was administered 6% (two of 31 sessions) of the time. HC occurred during ablation of 43% (6 of 14) of tumors in
patients who had received prophylactic adrenergic blockade. One patient who had received medication before the procedure also received long-acting betaadrenergic blockers during the ablation to treat HC. The patient experienced persistent hypotension after the procedure, which triggered admission to the intensive care unit and a 10-day hospital stay. Aspiration pneumonia and hemobilia, both grade 3 events, each occurred in one of 72 sessions. Adrenal insufficiency, a grade 2 event, occurred after ablation of solitary adrenal glands (n ¼ 16) and was successfully managed by replacement therapy using prednisone (Roxane Laboratories, Columbus, Ohio) 5 mg by mouth once daily. Three additional category 2 and two category 1 events affecting the skin, lungs, and gastrointestinal tract were documented (Table 3). Technical success was achieved in 71 sessions (99%). Univariate logistical regression analysis showed no association between HC and the 10 ablations of metastases in previously irradiated or ablated adrenal glands. Concurrent systemic therapy was not significantly associated with an increased risk of complications (P ¼ .629). Analysis of the other 62 sessions revealed that maximum axial tumor diameter r 4.5 cm and the visualization of normal adrenal tissue on CT or MR images obtained before the procedure independently significantly increased the likelihood of HC (Table 4). Multivariate analysis showed that maximum tumor
Table 3 . Adverse Events Encountered during Image-Guided Ablation of Adrenal Gland Metastases, Grouped by Severity According to CTCAE v4.03 Adverse Event (Incidence) Hypertensive crisis (31/72)
System Organ Class Vascular disorders
CTCAE v4.03 Grade Grade 4—hypertensive crisis; urgent intervention indicated
Ventricular tachycardia (1/72)
Cardiac disorders
Grade 3—medical intervention indicated
Cardiac troponin T increased (4/72)
Investigations
Grade 3—levels consistent with myocardial infarction as defined by manufacturer
Aspiration (1/72)
Respiratory, thoracic, and mediastinal
Grade 3—dyspnea and pneumonia symptoms (eg,
Hemobilia (1/72)
disorders Hepatobiliary disorders
aspiration pneumonia); hospitalization indicated Grade 3—transfusion indicated; also had ERCP
Adrenal insufficiency (16/72)
Endocrine disorders
Grade 2—moderate symptoms; medical intervention
New-onset atrial fibrillation (2/
Cardiac disorders
indicated Grade 2—nonurgent medical intervention indicated
72) Nausea (1/72)
Gastrointestinal disorders
Grade 2—oral intake decreased without significant weight loss, dehydration, or malnutrition
Vomiting (1/72)
Gastrointestinal disorders
Grade 2—3–5 episodes (separated by 5 min) in 24 h
Pain (1/72) Pneumothorax (1/72)
General disorders Respiratory, thoracic, and mediastinal
Grade 2—moderate pain; limiting instrumental ADL Grade 1—asymptomatic; intervention not indicated
Pad burn (1/72)
Skin and subcutaneous tissue disorders
disorders Grade 1—asymptomatic; intervention not indicated
Source–Data from U.S. Department of Health and Human Services. Common Terminology Criteria for Adverse Events Version 4.03. 2010. Available at: http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_8.5x11.pdf. Accessed November 27, 2015 (17). ADL ¼ activities of daily living; CTCAE ¼ Common Terminology Criteria for Adverse Events; ERCP ¼ endoscopic retrograde cholangiopancreatography.
6
’
Catecholamine Surge during Adrenal Gland Metastasis Ablation
’
Fintelmann et al
JVIR
Table 4 . Likelihood of Hypertensive Crisis Depending on Tumor Size and Visualization of Normal Adrenal Tissue as Determined by Univariate Logistic Regression Analysis n ¼ 62
Odds Ratio (95% CI)
P Value
Maximum tumor diameter r 4.5 cm Visualization of normal adrenal tissue
26.36 (5.26–131.99) 8.38 (2.67–25.33)
o .0001 o .0001
CI ¼confidence interval.
diameter r 4.5 cm was significantly associated with HC (P ¼ .001), whereas the visualization of normal adrenal tissue on CT or MR images obtained before the procedure did not reach statistical significance (P ¼ .06) when combined with tumor diameter (Table 5). Preoperative adrenergic blockade reduced the risk of catecholamine surge (odds ratio, 0.35; 95% confidence interval, 0.06–1.91; P o .227), albeit not significantly. Relative risk ratio interpretation of multinomial logistic regression (Table 6) revealed a very low likelihood (5%) of HC when the tumor was large and no normal adrenal gland tissue was visible and a high likelihood (78%) when the tumor was small and normal adrenal gland was visible on images obtained before the procedure. No correlation was identified between a particular ablation modality and the risk of HC.
DISCUSSION HC was the most common adverse event during imageguided ablations of non–hormonally active adrenal tumors: It occurred in 43% of ablations and resulted in cardiomyocyte damage and life-threatening arrhythmias. The incidence of HC in this series differs substantially from the incidence reported by Wolf et al (8) but is directly in line with the rate reported in the Mayo Clinic cohort (3). The significant association between use of an arterial line and HC supports the hypothesis that crises with a mean duration of 4 minutes may go undetected
without invasive blood pressure monitoring. Only two other severe complications were observed (3%), which compares favorably with the 9% of severe complications recently reported in a large series (3). The causative mechanism for HC is likely the ablation of normal adrenal medulla surrounding the tumor. Chromaffin cells in the adrenal medulla specialize in the synthesis, storage, and secretion of catecholamines (18). Cytolysis of chromaffin cells is expected to result in release of catecholamines into the systemic circulation. With RF ablation and microwave ablation, cell death and release of catecholamines occurs as heat is applied. In contrast, with cryoablation, cell death and release of catecholamines occurs during the thawing period, when the damaged adrenal tissue that is fixed during freezing is able to release catecholamines (9). Catecholamine surge during the ablation of non– hormonally active adrenal lesions cannot occur if no normal adrenal medulla is present. This explains why no HC occurred during the ablation of 10 lesions in previously irradiated or ablated adrenal glands. Normal adrenal medulla is also unlikely to be present if the metastasis is big because large lesions can replace the entire adrenal gland (19); this explains the significant association of HC with a maximum tumor diameter r 4.5 cm and the visualization of normal adrenal tissue on CT or MR images before the procedure. If both conditions were present, relative risk for HC was 78%. The ability to predict the likelihood of catecholamine surge during image-guided ablation of metastases to the
Table 5 . Likelihood of Hypertensive Crisis Depending on Tumor Size, Visualization of Normal Adrenal Tissue, and Adrenergic Blockade before the Procedure as Determined by Multivariate Logistic Regression Analysis n ¼ 62 Maximum tumor diameter r 4.5 cm Visualization of normal adrenal tissue Adrenergic blockade before procedure
Odds Ratio (95% CI)
P Value
18.08 (3.15–103.72)
.001
4.32 (0.94–19.83) 0.35 (0.06–1.91)
.06 .227
CI ¼confidence interval.
Table 6 . Relative Risk of Hypertensive Crisis Depending on Maximum Tumor Diameter and Visualization of Normal Adrenal Tissue as Determined by Multivariate Logistic Regression Analysis Visualization of Normal Adrenal Tissue n ¼ 62 Maximum tumor diameter
Yes
No
r 4.5 cm
78%
62%
4 4.5 cm
50%
5%
Volume XX
’
Number X
’
Month
’
2015
adrenal glands using images obtained before the procedure has the potential to address the most serious and common side effect of this otherwise well-tolerated procedure. Preventing HC is particularly important in patients who are at increased risk for end-organ damage during a severe acute increase in blood pressure secondary to preexisting coronary artery and cerebrovascular disease. Once identified, patients at high risk of catecholamine surge during ablation of non–hormonally active adrenal tumors can benefit from targeted administration of preoperative adrenergic blockade. Consultation before the procedure and follow-up after the procedure by an endocrinologist is valuable to coordinate administration of medication before the procedure and address the possibility of adrenal insufficiency if the ablation target is in a solitary adrenal gland (20). Medication before the procedure with the irreversible alpha blocker phenoxybenzamine (WellSpring Pharmaceutical Corporation, Sarasota, Florida), with or without the addition of beta blockers, can result in hypotension (20). Medication before the procedure with reversible adrenergic blockers did not significantly reduce the risk of HC in this series. However, medication before the procedure is known to dampen the peak SBP if HC occurs (9) because adrenergic blockers and catecholamines compete for the same receptors. A combination of the selective reversible alpha-1 receptor blocker doxazosin mesylate 1 mg by mouth once daily for 14 days before the procedure and the long-acting selective beta-1 receptor blocker metoprolol succinate 25 mg by mouth once daily for 4 days before the procedure can be administered on an outpatient basis with few side effects. Adrenal insufficiency can be managed with prednisone (Roxane Laboratories) 5 mg by mouth once daily. Although unlikely (5%), HC can occur even if the ablation target is large (4 4.5 cm) and when no normal adrenal gland tissue is visible. Therefore, all adrenal ablations should be performed under either MAC or GETA to allow for optimal blood pressure monitoring with an arterial line (5,7,8). The anesthesia team should be briefed before the procedure about the risk of HC during the case at hand to facilitate the rapid administration of antihypertensive medication, as previously suggested by other authors (5,7). In addition, a verbal warning should be given before releasing catecholamines during heat application (RF ablation, microwave ablation) or thawing (cryoablation) (9,21). Ultra-short-acting beta blockers and intravenous sodium nitroprusside (Valeant Pharmaceuticals) are preferred over longer acting agents for treatment of HC because long-acting beta blockers can result in hypotension when the catecholamine surge created by the ablation subsides (5). This study has some limitations. First, this series includes a variety of tumors, not all of which were pathologically proven; three different ablation modalities; and 10 different operators. Despite this being the
7
largest reported series of non–hormonally active adrenal tumors to date, the sample sizes were still relatively small, which translated into wide confidence intervals. Small numbers precluded conclusions about whether a particular ablation modality was more prone to complications. Also, the study is retrospective. Data were gathered at two institutions with different algorithms for care before, during, and after ablation procedures. However, it included consecutive patients, diminishing the possibility of selection bias. In addition, an interventional radiologist not involved in the ablation procedures and blinded to the study purpose performed the visual analysis of the images obtained before the procedure. In conclusion, tumor diameter o 4.5 cm and visualization of normal adrenal tissue on CT or MR images obtained before the procedure identify patients at high risk for HC and arrhythmia secondary to intraprocedural catecholamine surge during image-guided ablation of non–hormonally active adrenal tumors, unless the adrenal gland has previously been irradiated or ablated. Consultation with an endocrinologist and adrenergic blockade before the procedure are recommended in high-risk patients. Intraprocedural anesthesia support is recommended for all adrenal ablations.
ACKNOWLEDGMENT We thank Dr. Johann Blauth and Dr. Hang Lee for guidance with the statistical analysis. Dr. Lee is funded through the Harvard Catalyst program.
REFERENCES 1. Onik G, Onik C, Medary I, et al. Life-threatening hypertensive crises in two patients undergoing hepatic radiofrequency ablation. AJR Am J Roentgenol 2003; 181:495–497. 2. Keeling AN, Sabharwal T, Allen MJ, Hegarty NJ, Adam A. Hypertensive crisis during radiofrequency ablation of the adrenal gland. J Vasc Interv Radiol 2009; 20:990–991. 3. Welch BT, Callstrom MR, Carpenter PC, et al. A single-institution experience in image-guided thermal ablation of adrenal gland metastases. J Vasc Interv Radiol 2014; 25:593–598. 4. Carrafiello G, Laganà D, Recaldini C, et al. Imaging-guided percutaneous radiofrequency ablation of adrenal metastases: preliminary results at a single institution with a single device. Cardiovasc Intervent Radiol 2008; 31:762–767. 5. Chini EN, Brown MJ, Farrell MA, Charboneau JW. Hypertensive crisis in a patient undergoing percutaneous radiofrequency ablation of an adrenal mass under general anesthesia. Anesth Analg 2004; 99:1867–1869. 6. Mayo-Smith WW, Dupuy DE. Adrenal neoplasms: CT-guided radiofrequency ablation—preliminary results. Radiology 2004; 231:225–230. 7. Atwell TD, Wass CT, Charboneau JW, Callstrom MR, Farrell MA, Sengupta S. Malignant hypertension during cryoablation of an adrenal gland tumor. J Vasc Interv Radiol 2006; 17:573–575. 8. Wolf FJ, Dupuy DE, Machan JT, Mayo-Smith WW. Adrenal neoplasms: effectiveness and safety of CT-guided ablation of 23 tumors in 22 patients. Eur J Radiol 2012; 81:1717–1723. 9. Welch BT, Atwell TD, Nichols DA, et al. Percutaneous image-guided adrenal cryoablation: procedural considerations and technical success. Radiology 2011; 258:301–307. 10. Yamakado K, Takaki H, Yamada T, et al. Incidence and cause of hypertension during adrenal radiofrequency ablation. Cardiovasc Intervent Radiol 2012; 35:1422–1427.
8
’
Catecholamine Surge during Adrenal Gland Metastasis Ablation
11. Yamakado K, Takaki H, Uchida K, Nakatsuka A, Shiraishi T, Takeda K. Adrenal radiofrequency ablation in swine: change in blood pressure and histopathologic analysis. Cardiovasc Intervent Radiol 2011; 34:839–844. 12. Sudheendra D, Wood BJ. Appropriate premedication risk reduction during adrenal ablation. J Vasc Interv Radiol 2006; 17:1367–1368. 13. Pacak K. Preoperative management of the pheochromocytoma patient. J Clin Endocrinol Metab 2007; 92:4069–4079. 14. Lo WK, vansonnenberg E, Shankar S, et al. Percutaneous CT-guided radiofrequency ablation of symptomatic bilateral adrenal metastases in a single session. J Vasc Interv Radiol 2006; 17:175–179. 15. Boland GWL, Dwamena BA, Jagtiani Sangwaiya M, et al. Characterization of adrenal masses by using FDG PET: a systematic review and metaanalysis of diagnostic test performance. Radiology 2011; 259:117–126. 16. Ryan ER, Sofocleous CT, Schöder H, et al. Split-dose technique for FDG PET/CT-guided percutaneous ablation: a method to facilitate lesion
Fintelmann et al
17.
18. 19.
20. 21.
’
JVIR
targeting and to provide immediate assessment of treatment effectiveness. Radiology 2013; 268:288–295. U.S. Department of Health and Human Services. Common Terminology Criteria for Adverse Events Version 4.03. 2010. Available at: http://evs.nci. nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_8.5x11.pdf. Accessed November 27, 2015. Perlman RL, Chalfie M. Catecholamine release from the adrenal medulla. Clin Endocrinol Metab 1977; 6:551–576. Efremidis SC, Harsoulis F, Douma S, Zafiriadou E, Zamboulis C, Kouri A. Adrenal insufficiency with enlarged adrenals. Abdom Imaging 1996; 21: 168–171. Uppot RN, Gervais DA. Imaging-guided adrenal tumor ablation. AJR Am J Roentgenol 2013; 200:1226–1233. Ethier MD, Beland MD, Mayo-Smith W. Image-guided ablation of adrenal tumors. Tech Vasc Interv Radiol 2013; 16:262–268.
Catecholamine Surge during Image-Guided Ablation of Adrenal Gland Metastases: Predictors, Consequences, and Recommendations for Management Florian J. Fintelmann, MD, FRCPC, Kemal Tuncali, MD, Stefan Puchner, MD, Debra A. Gervais, MD, Ashraf Thabet, MD, Paul B. Shyn, MD, Ronald S. Arellano, MD, Servet Tatli, MD, Peter R. Mueller, MD, Stuart G. Silverman, MD, and Raul N. Uppot, MD
ABSTRACT Purpose: To identify retrospectively predictors of catecholamine surge during image-guided ablation of metastases to the adrenal gland. Materials and Methods: Between 2001 and 2014, 57 patients (39 men, 18 women; mean age, 65 y ⫾ 10; age range, 41–81 y) at two academic medical centers underwent ablation of 64 metastatic adrenal tumors from renal cell carcinoma (n ¼ 27), lung cancer (n ¼ 23), melanoma (n ¼ 4), colorectal cancer (n ¼ 3), and other tumors (n ¼ 7). Tumors measured 0.7–11.3 cm (mean, 4 cm ⫾ 2.5). Modalities included cryoablation (n ¼ 38), radiofrequency (RF) ablation (n ¼ 20), RF ablation with injection of dehydrated ethanol (n ¼ 10), and microwave ablation (n ¼ 4). Fisher exact test, univariate, and multivariate logistical regression analysis was used to evaluate factors predicting hypertensive crisis (HC). Results: HC occurred in 31 sessions (43%). Ventricular tachycardia (n ¼ 1), atrial fibrillation (n ¼ 2), and troponin leak (n ¼ 4) developed during HC episodes. HC was significantly associated with maximum tumor diameter r 4.5 cm (odds ratio [OR], 26.36; 95% confidence interval [CI], 5.26–131.99; P o .0001) and visualization of normal adrenal tissue on computed tomography (CT) or magnetic resonance (MR) imaging before the procedure (OR, 8.38; 95% CI, 2.67–25.33; P o .0001). No HC occurred during ablation of metastases in previously irradiated or ablated adrenal glands. Conclusions: Patients at high risk of catecholamine surge during ablation of non–hormonally active adrenal metastases can be identified by the presence of normal adrenal tissue and tumor diameter r 4.5 cm on pre-procedure CT or MR imaging.
ABBREVIATIONS GETA = general endotracheal anesthesia, HC = hypertensive crisis, MAC = monitored anesthesia care, RF = radiofrequency, SBP = systolic blood pressure
From the Department of Radiology (F.J.F., S.P., D.A.G., A.T., R.S.A., P.R.M., R.N.U.), Massachusetts General Hospital, 55 Fruit Street, FND-202, Boston, MA 02111; and Department of Radiology (K.T., P.B.S., S.T., S.G.S.), Brigham and Women’s Hospital, Boston, Massachusetts. Received July 27, 2015; final revision received October 12, 2015; accepted November 10, 2015. Address correspondence to F.J.F.; E-mail: ffi[email protected] K.T. received grants from Canon USA (Melville, New York). P.R.M. is a consultant for Cook, Inc (Bloomington, Indiana). None of the other authors have identified a conflict of interest. From the SIR 2014 Annual Meeting. & SIR, 2015 J Vasc Interv Radiol 2015; XX:]]]–]]] http://dx.doi.org/10.1016/j.jvir.2015.11.034
A severe increase in blood pressure after thermal injury of adrenal gland tissue was first described by Onik et al (1) during radiofrequency (RF) ablation of a posterior right lobe liver lesion in close proximity to a normal right adrenal gland. An acute increase in blood pressure 4 180 mm Hg or diastolic blood pressure 4 110 mm Hg, or both, is referred to as hypertensive crisis (HC). HC and cardiac irritability has been reported during thermal ablation of functional primary adrenal tumors and nonfunctional metastases to the adrenal gland with incidence up to 46% (2–9). Yamakado et al (10,11) firmly established the link between the onset of HC during thermal ablation of adrenal gland tumors and
2
’
Catecholamine Surge during Adrenal Gland Metastasis Ablation
catecholamine surge by measuring the catecholamine blood levels before, during, and after RF ablation of adrenal glands in swine. As a result, medication with alpha and beta blockers was suggested before imageguided ablation of functional and nonfunctional adrenal tumors (7–9,12). In cases of functional tumors, this recommendation mirrors guidelines in the surgical literature on pheochromocytomas (13). However, in cases of nonfunctional tumors such as metastases, no guidelines exist. It is unclear how to assess the need for medication before the procedure in a particular patient without better understanding the risk factors for HC. The purpose of this study was to identify predictors of catecholamine surge during image-guided ablation of metastases to the adrenal gland, to report consequences, and to guide management before, during, and after the procedure.
MATERIALS AND METHODS Patient Cohort The institutional review board at two participating academic medical centers approved this retrospective study. Patients included 57 adults (39 men and 18 women) with a mean age of 65 years (SD 10; range, 41–81 y) with adrenal metastases who underwent imageguided ablation at either of two large academic medical centers between January 2001 and March 2014. The series included one patient with bilateral adrenal gland tumors from a previously published case report (14). There were 35 patients (61%) who had received systemic chemotherapy at some point during the course of their disease, and four of these patients (7%) were on systemic therapy at the time of ablation.
Disease Burden and Index Tumors In 21 patients (37%), the adrenal gland was the only site of disease; 36 patients (63%) had metastases to other
Figure 1. Distribution of maximum axial adrenal tumor diameter.
Fintelmann et al
’
JVIR
organs. Three patients had two lesions in the same gland, and four patients had at least one lesion in each adrenal gland. Proof of metastatic involvement was available before ablation based on either biopsy in 45 patients (70%) or documentation of growth and fluorine18 fluorodeoxyglucose uptake on positron emission tomography (PET) combined with computed tomography (CT) in 19 patients (30%) (15). The total number of adrenal tumors was 64, and most metastases were from renal cell cancer and non–small cell lung cancer (Table 1). Mean maximum tumor diameter was 4 cm (SD 2.5; range, 0.7–11.3 cm) (Fig 1).
Ablation Technique There were 72 ablation sessions performed by 10 operators. Treatment indications included cure in 28 cases of solitary metastasis to the adrenal gland (39%), cytoreduction in 28 cases with two or more metastases involving organs other than the adrenal gland (39%), and palliation in 16 cases for large adrenal masses causing abdominal or flank pain (22%). Ten sessions (14%) were performed on lesions in previously irradiated (n ¼ 1) or ablated (n ¼ 9) adrenal glands. Treatment
Table 1 . Adrenal Tumor Histology Adrenal Tumor Histology Renal cell cancer
Values (n ¼ 64 Tumors) 42% (27/64)
Non–small cell lung cancer
36% (23/64)
Melanoma Colorectal cancer
6.3% (4/64) 4.7% (3/64)
Endometrial cancer
3% (2/64)
Breast cancer Ovarian cancer
1.5% (1/64) 1.5% (1/64)
Transitional cell carcinoma
1.5% (1/64)
Malignant fibrous histiocytoma Hepatocellular carcinoma
1.5% (1/64) 1.5% (1/64)
Volume XX
’
Number X
’
Month
’
2015
3
modalities included 38 cryoablations (53%), 20 RF ablations (28%), 10 RF ablation procedures combined with injection of up to 10 mL of dehydrated ethanol (14%), and 4 microwave ablations (6%). Cryoablation was performed using the SeedNet (Galil Medical, Inc, Arden Hills, Minnesota) system. RF ablation was performed using a single straight electrode (Cool-tip; Covidien, Boulder, Colorado) in four sessions, a cluster electrode (Radionics, Inc, Burlington, Massachusetts) in 14 sessions, a multitined Starburst electrode (RITA Medical Systems, Inc, Mountain View, California) in six sessions, and a multitined Leveen electrode (Boston Scientific, Marlborough, Massachusetts) in six sessions. Microwave ablation was performed using the Certus 140 system (NeuWave Medical; Madison, Wisconsin). The tumors were ablated per manufacturer recommendations: temperature-based (Starburst electrode), impedance (Leveen electrode), or both (Cool-tip) as well as time-based for microwave ablation and double-freeze cycle for cryoablation. The goal was to extend the zone of ablation at least 5 mm beyond the tumor margin. In patients referred for ablation of symptomatic large adrenal masses, the goal was to diminish tumor burden as much as possible to decrease pain rather than achieve eradication. After ablation, the applicators were removed.
Image Guidance CT, magnetic resonance (MR) imaging, and PET/CT (Discovery ST; GE Healthcare, Milwaukee, Wisconsin) were used to guide placement of applicators (Table 2). CT scanners had 16 (LightSpeed; GE Healthcare) or 40 (SOMATOM Sensation Open; Siemens Medical Solutions, Forchheim, Germany) detector rows. MR imaging was either 0.5 tesla (Signa SP; GE Healthcare) or 3 tesla (Siemens MAGNETOM Verio; Siemens Healthcare, Erlangen, Germany). Non–contrast enhanced CT or MR images of the treatment area were obtained before, during, and immediately after ablation to plan and monitor treatment and to assess for complications. Image slice thickness was 5 mm on CT and 4 mm on MR imaging examinations. MR imaging sequences were either spoiled gradient recalled or T2-weighted halfFourier acquisition single-shot turbo spin echo. When Table 2 . Treatment Modalities and Image Guidance Used for Ablation Sessions MR CT Imaging PET/CT Totals Cryoablation RF ablation
24 20
12 0
2 0
38 20
RF ablation and EtOH injection
10
0
0
10
Microwave ablation Totals
3 57
0 12
1 3
4
EtOH ¼ dehydrated ethanol; PET/CT = positron emission tomography/computed tomography; RF = radiofrequency.
PET/CT was used for image guidance, the protocol described by Ryan et al (16) was followed.
Care during the Procedure Ablation was performed after induction of general endotracheal anesthesia (GETA) in 57 sessions (79%). Monitored anesthesia care (MAC) was used for 12 sessions (17%), and conscious sedation was used for three sessions (4%). An indwelling radial artery catheter was used for intraprocedural hemodynamic monitoring in addition to a blood pressure cuff in 27 of 69 sessions performed under MAC or GETA. If patients were treated under MAC or GETA, they were admitted to the hospital for overnight observation.
Care before and after Procedure In 20 patients, consultation with an endocrinologist occurred before and after the procedure to establish recommendations for medication before the ablation, periprocedural management of catecholamine surge, and follow-up after the procedure to diagnose and manage any adrenal insufficiency. Adrenergic blockade was administered before 14 sessions (19%). Early in this series, patients were admitted for 3 days before the procedure for gradual administration of alphaadrenergic and beta-adrenergic blockade. Because this medication protocol may result in hypotension, intravenous fluid was administered simultaneously. One of two patients treated this way did not tolerate the medications because of hypotension and could not receive the full dose. The other patient needed diuretics for fluid overload secondary to the medication protocol. Later in this series, low-dose oral adrenergic blockade was administered in the outpatient setting after consultation with an endocrinologist. Doxazosin mesylate (Teva Pharmaceuticals, North Wales, Pennsylvania) 1 mg administered by mouth once daily for 14 days before the procedure and long-acting metoprolol succinate (Wockhardt USA, Parsippany, New Jersey) 25 mg by mouth once daily for 4 days before the procedure does not result in significant side effects in most cases. Technical success was assessed using contrastenhanced MR imaging (71%), CT (15%), or PET/CT (14%) at the first imaging follow-up, which occurred on average 10 days after the procedure (SD 16; range, 0– 78). All but three patients were seen in the interventional radiology clinic shortly after the imaging study. Clinical follow-up beyond this first visit was available through notes in the medical chart for a mean of 26 months (SD 27; range, 1–116). The three patients who did not return were considered lost to follow-up.
Data Analysis A radiologist blinded to the purpose of this study (S.P.; 9 years of experience) measured and recorded the maximum axial diameter of each adrenal tumor on the CT or
4
’
Catecholamine Surge during Adrenal Gland Metastasis Ablation
MR images obtained immediately before the ablation. The same radiologist recorded whether or not any normal adrenal gland tissue was visible adjacent to the tumor as illustrated in Figure 2. Baseline and peak intraprocedural blood pressure and heart rate were extracted from the sedation and anesthesia portion of the medical record (F.J.F.; 6 years of experience) and correlated with the timing of the ablation. Based on the work of Yamakado et al (10), a systolic blood pressure (SBP) spike 4 180 mm Hg during the procedure in close temporal association with onset of cell death (heating during RF ablation and microwave ablation, thawing during cryoablation) was chosen as a surrogate marker for catecholamine-induced HC. To qualify as a spike, the peak SBP had to increase acutely by at least 50 mm Hg above the baseline determined at the beginning of the procedure. All adverse events were categorized using Common Terminology Criteria for Adverse Events Version 4.03 (17) and graded as mild (grade 1), moderate (grade 2), severe (grade 3), or life-threatening (grade 4), with specific parameters according to the organ system involved.
Statistical Methods Statistical analyses were performed using the SAS software package (SAS Institute, Cary, North Carolina). Fisher exact test was used to test for an association between continuous blood pressure monitoring with an arterial line and the detection of HC. Univariate and multivariate logistical regression analysis was used to detect associations between tumor size, visualization of normal adrenal tissue on images obtained before the procedure, ablation modality, medication administered before the procedure, prior treatment of the adrenal gland, concurrent systemic therapy, and intraprocedural
Fintelmann et al
’
JVIR
HC. Odds ratio with corresponding 95% confidence interval and relative risk was calculated. P values are two-tailed. P values o .05 were considered statistically significant.
RESULTS The most common serious adverse event during ablation of adrenal gland metastases was HC in 31 sessions (43%). No HC occurred during the 10 ablations performed on metastases in previously ablated or irradiated adrenal glands. The mean duration of the HC was 4 minutes (SD 4.4; range, 1–15 min). Peak SBP was 182– 320 mm Hg (average 224 mm Hg, SD 33). HC was accompanied by elevated cardiac troponin T in four patients with peak levels of 0.18 ng/mL, 0.4 ng/mL, 0.6 ng/mL, and 0.73 ng/mL, indicating cardiomyocyte damage. Duration of HC in these patients was 1–5 minutes, and the peak SBP was 186–260 mm Hg. Three of these four patients also experienced cardiac arrhythmia. A 64year-old man developed ventricular tachycardia after the onset of HC following the third ablation of a 3.1-cm solitary renal cell cancer metastasis in the adrenal gland using RF. Vital signs remained stable and the cardiac rhythm normalized within 2 minutes without further intervention. The patient was admitted for observation and discharged home in good condition after a negative nuclear cardiac stress test 2 days later. Two patients developed atrial fibrillation after the onset of HC. Both patients returned to normal rhythm within 24 hours after treatment with verapamil hydrochloride (Mylan, Canonsburg, Pennsylvania) and digoxin (GlaxoSmithKline, Philadelphia, Pennsylvania) and had normal echocardiograms and nuclear cardiac stress tests. One of these two patients experienced chest pain and underwent coronary
Figure 2. Axial contrast-enhanced CT image shows a 2-cm tumor (arrow) in the lateral limb of the left adrenal gland (a). Residual normal adrenal gland tissue (arrow) is demonstrated in the same patient at the same time, approximately 3 cm superiorly (b).
Volume XX
’
Number X
’
Month
’
2015
5
angiography for suspected non–ST segment elevation myocardial infarction. Findings included complex left main and three-vessel coronary atherosclerosis. Patients were discharged home in o 24 hours after 54 (75%) sessions. Adverse events triggered unplanned admissions after 18 sessions (25%) with an average length of stay of 3 days (range, 2–10 d). Anesthesiologists were involved in cases performed under MAC and GETA, which allowed for invasive blood pressure monitoring with an arterial line. Fisher exact test revealed a significant association (P ¼ .049) between the use of an arterial line and the detection of HC. Antihypertensive medications administered by anesthesiologists during HC included one or more intravenous beta blockers—labetalol hydrochloride (Hospira, Lake Forest, Illinois), esmolol hydrochloride (Mylan), metoprolol tartrate (Hospira), and propranolol hydrochloride (West-Ward Pharmaceuticals, Eatontown, New Jersey) in order of decreasing frequency—87% of the time (27 of 31 sessions), whereas the alpha blocker phentolamine mesylate (Bedford Laboratories, Bedford, Ohio) was administered only once instead. In addition to the adrenergic blocker, intravenous sodium nitroprusside (Valeant Pharmaceuticals, Bridgewater, New Jersey) was administered 23% (seven of 31 sessions) of the time, and hydralazine hydrochloride (Novation, Irving, Texas) was administered 6% (two of 31 sessions) of the time. HC occurred during ablation of 43% (6 of 14) of tumors in
patients who had received prophylactic adrenergic blockade. One patient who had received medication before the procedure also received long-acting betaadrenergic blockers during the ablation to treat HC. The patient experienced persistent hypotension after the procedure, which triggered admission to the intensive care unit and a 10-day hospital stay. Aspiration pneumonia and hemobilia, both grade 3 events, each occurred in one of 72 sessions. Adrenal insufficiency, a grade 2 event, occurred after ablation of solitary adrenal glands (n ¼ 16) and was successfully managed by replacement therapy using prednisone (Roxane Laboratories, Columbus, Ohio) 5 mg by mouth once daily. Three additional category 2 and two category 1 events affecting the skin, lungs, and gastrointestinal tract were documented (Table 3). Technical success was achieved in 71 sessions (99%). Univariate logistical regression analysis showed no association between HC and the 10 ablations of metastases in previously irradiated or ablated adrenal glands. Concurrent systemic therapy was not significantly associated with an increased risk of complications (P ¼ .629). Analysis of the other 62 sessions revealed that maximum axial tumor diameter r 4.5 cm and the visualization of normal adrenal tissue on CT or MR images obtained before the procedure independently significantly increased the likelihood of HC (Table 4). Multivariate analysis showed that maximum tumor
Table 3 . Adverse Events Encountered during Image-Guided Ablation of Adrenal Gland Metastases, Grouped by Severity According to CTCAE v4.03 Adverse Event (Incidence) Hypertensive crisis (31/72)
System Organ Class Vascular disorders
CTCAE v4.03 Grade Grade 4—hypertensive crisis; urgent intervention indicated
Ventricular tachycardia (1/72)
Cardiac disorders
Grade 3—medical intervention indicated
Cardiac troponin T increased (4/72)
Investigations
Grade 3—levels consistent with myocardial infarction as defined by manufacturer
Aspiration (1/72)
Respiratory, thoracic, and mediastinal
Grade 3—dyspnea and pneumonia symptoms (eg,
Hemobilia (1/72)
disorders Hepatobiliary disorders
aspiration pneumonia); hospitalization indicated Grade 3—transfusion indicated; also had ERCP
Adrenal insufficiency (16/72)
Endocrine disorders
Grade 2—moderate symptoms; medical intervention
New-onset atrial fibrillation (2/
Cardiac disorders
indicated Grade 2—nonurgent medical intervention indicated
72) Nausea (1/72)
Gastrointestinal disorders
Grade 2—oral intake decreased without significant weight loss, dehydration, or malnutrition
Vomiting (1/72)
Gastrointestinal disorders
Grade 2—3–5 episodes (separated by 5 min) in 24 h
Pain (1/72) Pneumothorax (1/72)
General disorders Respiratory, thoracic, and mediastinal
Grade 2—moderate pain; limiting instrumental ADL Grade 1—asymptomatic; intervention not indicated
Pad burn (1/72)
Skin and subcutaneous tissue disorders
disorders Grade 1—asymptomatic; intervention not indicated
Source–Data from U.S. Department of Health and Human Services. Common Terminology Criteria for Adverse Events Version 4.03. 2010. Available at: http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_8.5x11.pdf. Accessed November 27, 2015 (17). ADL ¼ activities of daily living; CTCAE ¼ Common Terminology Criteria for Adverse Events; ERCP ¼ endoscopic retrograde cholangiopancreatography.
6
’
Catecholamine Surge during Adrenal Gland Metastasis Ablation
’
Fintelmann et al
JVIR
Table 4 . Likelihood of Hypertensive Crisis Depending on Tumor Size and Visualization of Normal Adrenal Tissue as Determined by Univariate Logistic Regression Analysis n ¼ 62
Odds Ratio (95% CI)
P Value
Maximum tumor diameter r 4.5 cm Visualization of normal adrenal tissue
26.36 (5.26–131.99) 8.38 (2.67–25.33)
o .0001 o .0001
CI ¼confidence interval.
diameter r 4.5 cm was significantly associated with HC (P ¼ .001), whereas the visualization of normal adrenal tissue on CT or MR images obtained before the procedure did not reach statistical significance (P ¼ .06) when combined with tumor diameter (Table 5). Preoperative adrenergic blockade reduced the risk of catecholamine surge (odds ratio, 0.35; 95% confidence interval, 0.06–1.91; P o .227), albeit not significantly. Relative risk ratio interpretation of multinomial logistic regression (Table 6) revealed a very low likelihood (5%) of HC when the tumor was large and no normal adrenal gland tissue was visible and a high likelihood (78%) when the tumor was small and normal adrenal gland was visible on images obtained before the procedure. No correlation was identified between a particular ablation modality and the risk of HC.
DISCUSSION HC was the most common adverse event during imageguided ablations of non–hormonally active adrenal tumors: It occurred in 43% of ablations and resulted in cardiomyocyte damage and life-threatening arrhythmias. The incidence of HC in this series differs substantially from the incidence reported by Wolf et al (8) but is directly in line with the rate reported in the Mayo Clinic cohort (3). The significant association between use of an arterial line and HC supports the hypothesis that crises with a mean duration of 4 minutes may go undetected
without invasive blood pressure monitoring. Only two other severe complications were observed (3%), which compares favorably with the 9% of severe complications recently reported in a large series (3). The causative mechanism for HC is likely the ablation of normal adrenal medulla surrounding the tumor. Chromaffin cells in the adrenal medulla specialize in the synthesis, storage, and secretion of catecholamines (18). Cytolysis of chromaffin cells is expected to result in release of catecholamines into the systemic circulation. With RF ablation and microwave ablation, cell death and release of catecholamines occurs as heat is applied. In contrast, with cryoablation, cell death and release of catecholamines occurs during the thawing period, when the damaged adrenal tissue that is fixed during freezing is able to release catecholamines (9). Catecholamine surge during the ablation of non– hormonally active adrenal lesions cannot occur if no normal adrenal medulla is present. This explains why no HC occurred during the ablation of 10 lesions in previously irradiated or ablated adrenal glands. Normal adrenal medulla is also unlikely to be present if the metastasis is big because large lesions can replace the entire adrenal gland (19); this explains the significant association of HC with a maximum tumor diameter r 4.5 cm and the visualization of normal adrenal tissue on CT or MR images before the procedure. If both conditions were present, relative risk for HC was 78%. The ability to predict the likelihood of catecholamine surge during image-guided ablation of metastases to the
Table 5 . Likelihood of Hypertensive Crisis Depending on Tumor Size, Visualization of Normal Adrenal Tissue, and Adrenergic Blockade before the Procedure as Determined by Multivariate Logistic Regression Analysis n ¼ 62 Maximum tumor diameter r 4.5 cm Visualization of normal adrenal tissue Adrenergic blockade before procedure
Odds Ratio (95% CI)
P Value
18.08 (3.15–103.72)
.001
4.32 (0.94–19.83) 0.35 (0.06–1.91)
.06 .227
CI ¼confidence interval.
Table 6 . Relative Risk of Hypertensive Crisis Depending on Maximum Tumor Diameter and Visualization of Normal Adrenal Tissue as Determined by Multivariate Logistic Regression Analysis Visualization of Normal Adrenal Tissue n ¼ 62 Maximum tumor diameter
Yes
No
r 4.5 cm
78%
62%
4 4.5 cm
50%
5%
Volume XX
’
Number X
’
Month
’
2015
adrenal glands using images obtained before the procedure has the potential to address the most serious and common side effect of this otherwise well-tolerated procedure. Preventing HC is particularly important in patients who are at increased risk for end-organ damage during a severe acute increase in blood pressure secondary to preexisting coronary artery and cerebrovascular disease. Once identified, patients at high risk of catecholamine surge during ablation of non–hormonally active adrenal tumors can benefit from targeted administration of preoperative adrenergic blockade. Consultation before the procedure and follow-up after the procedure by an endocrinologist is valuable to coordinate administration of medication before the procedure and address the possibility of adrenal insufficiency if the ablation target is in a solitary adrenal gland (20). Medication before the procedure with the irreversible alpha blocker phenoxybenzamine (WellSpring Pharmaceutical Corporation, Sarasota, Florida), with or without the addition of beta blockers, can result in hypotension (20). Medication before the procedure with reversible adrenergic blockers did not significantly reduce the risk of HC in this series. However, medication before the procedure is known to dampen the peak SBP if HC occurs (9) because adrenergic blockers and catecholamines compete for the same receptors. A combination of the selective reversible alpha-1 receptor blocker doxazosin mesylate 1 mg by mouth once daily for 14 days before the procedure and the long-acting selective beta-1 receptor blocker metoprolol succinate 25 mg by mouth once daily for 4 days before the procedure can be administered on an outpatient basis with few side effects. Adrenal insufficiency can be managed with prednisone (Roxane Laboratories) 5 mg by mouth once daily. Although unlikely (5%), HC can occur even if the ablation target is large (4 4.5 cm) and when no normal adrenal gland tissue is visible. Therefore, all adrenal ablations should be performed under either MAC or GETA to allow for optimal blood pressure monitoring with an arterial line (5,7,8). The anesthesia team should be briefed before the procedure about the risk of HC during the case at hand to facilitate the rapid administration of antihypertensive medication, as previously suggested by other authors (5,7). In addition, a verbal warning should be given before releasing catecholamines during heat application (RF ablation, microwave ablation) or thawing (cryoablation) (9,21). Ultra-short-acting beta blockers and intravenous sodium nitroprusside (Valeant Pharmaceuticals) are preferred over longer acting agents for treatment of HC because long-acting beta blockers can result in hypotension when the catecholamine surge created by the ablation subsides (5). This study has some limitations. First, this series includes a variety of tumors, not all of which were pathologically proven; three different ablation modalities; and 10 different operators. Despite this being the
7
largest reported series of non–hormonally active adrenal tumors to date, the sample sizes were still relatively small, which translated into wide confidence intervals. Small numbers precluded conclusions about whether a particular ablation modality was more prone to complications. Also, the study is retrospective. Data were gathered at two institutions with different algorithms for care before, during, and after ablation procedures. However, it included consecutive patients, diminishing the possibility of selection bias. In addition, an interventional radiologist not involved in the ablation procedures and blinded to the study purpose performed the visual analysis of the images obtained before the procedure. In conclusion, tumor diameter o 4.5 cm and visualization of normal adrenal tissue on CT or MR images obtained before the procedure identify patients at high risk for HC and arrhythmia secondary to intraprocedural catecholamine surge during image-guided ablation of non–hormonally active adrenal tumors, unless the adrenal gland has previously been irradiated or ablated. Consultation with an endocrinologist and adrenergic blockade before the procedure are recommended in high-risk patients. Intraprocedural anesthesia support is recommended for all adrenal ablations.
ACKNOWLEDGMENT We thank Dr. Johann Blauth and Dr. Hang Lee for guidance with the statistical analysis. Dr. Lee is funded through the Harvard Catalyst program.
REFERENCES 1. Onik G, Onik C, Medary I, et al. Life-threatening hypertensive crises in two patients undergoing hepatic radiofrequency ablation. AJR Am J Roentgenol 2003; 181:495–497. 2. Keeling AN, Sabharwal T, Allen MJ, Hegarty NJ, Adam A. Hypertensive crisis during radiofrequency ablation of the adrenal gland. J Vasc Interv Radiol 2009; 20:990–991. 3. Welch BT, Callstrom MR, Carpenter PC, et al. A single-institution experience in image-guided thermal ablation of adrenal gland metastases. J Vasc Interv Radiol 2014; 25:593–598. 4. Carrafiello G, Laganà D, Recaldini C, et al. Imaging-guided percutaneous radiofrequency ablation of adrenal metastases: preliminary results at a single institution with a single device. Cardiovasc Intervent Radiol 2008; 31:762–767. 5. Chini EN, Brown MJ, Farrell MA, Charboneau JW. Hypertensive crisis in a patient undergoing percutaneous radiofrequency ablation of an adrenal mass under general anesthesia. Anesth Analg 2004; 99:1867–1869. 6. Mayo-Smith WW, Dupuy DE. Adrenal neoplasms: CT-guided radiofrequency ablation—preliminary results. Radiology 2004; 231:225–230. 7. Atwell TD, Wass CT, Charboneau JW, Callstrom MR, Farrell MA, Sengupta S. Malignant hypertension during cryoablation of an adrenal gland tumor. J Vasc Interv Radiol 2006; 17:573–575. 8. Wolf FJ, Dupuy DE, Machan JT, Mayo-Smith WW. Adrenal neoplasms: effectiveness and safety of CT-guided ablation of 23 tumors in 22 patients. Eur J Radiol 2012; 81:1717–1723. 9. Welch BT, Atwell TD, Nichols DA, et al. Percutaneous image-guided adrenal cryoablation: procedural considerations and technical success. Radiology 2011; 258:301–307. 10. Yamakado K, Takaki H, Yamada T, et al. Incidence and cause of hypertension during adrenal radiofrequency ablation. Cardiovasc Intervent Radiol 2012; 35:1422–1427.
8
’
Catecholamine Surge during Adrenal Gland Metastasis Ablation
11. Yamakado K, Takaki H, Uchida K, Nakatsuka A, Shiraishi T, Takeda K. Adrenal radiofrequency ablation in swine: change in blood pressure and histopathologic analysis. Cardiovasc Intervent Radiol 2011; 34:839–844. 12. Sudheendra D, Wood BJ. Appropriate premedication risk reduction during adrenal ablation. J Vasc Interv Radiol 2006; 17:1367–1368. 13. Pacak K. Preoperative management of the pheochromocytoma patient. J Clin Endocrinol Metab 2007; 92:4069–4079. 14. Lo WK, vansonnenberg E, Shankar S, et al. Percutaneous CT-guided radiofrequency ablation of symptomatic bilateral adrenal metastases in a single session. J Vasc Interv Radiol 2006; 17:175–179. 15. Boland GWL, Dwamena BA, Jagtiani Sangwaiya M, et al. Characterization of adrenal masses by using FDG PET: a systematic review and metaanalysis of diagnostic test performance. Radiology 2011; 259:117–126. 16. Ryan ER, Sofocleous CT, Schöder H, et al. Split-dose technique for FDG PET/CT-guided percutaneous ablation: a method to facilitate lesion
Fintelmann et al
17.
18. 19.
20. 21.
’
JVIR
targeting and to provide immediate assessment of treatment effectiveness. Radiology 2013; 268:288–295. U.S. Department of Health and Human Services. Common Terminology Criteria for Adverse Events Version 4.03. 2010. Available at: http://evs.nci. nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_8.5x11.pdf. Accessed November 27, 2015. Perlman RL, Chalfie M. Catecholamine release from the adrenal medulla. Clin Endocrinol Metab 1977; 6:551–576. Efremidis SC, Harsoulis F, Douma S, Zafiriadou E, Zamboulis C, Kouri A. Adrenal insufficiency with enlarged adrenals. Abdom Imaging 1996; 21: 168–171. Uppot RN, Gervais DA. Imaging-guided adrenal tumor ablation. AJR Am J Roentgenol 2013; 200:1226–1233. Ethier MD, Beland MD, Mayo-Smith W. Image-guided ablation of adrenal tumors. Tech Vasc Interv Radiol 2013; 16:262–268.