The Perioperative Management of a Patient with Complex Single Ventricle Physiology and Pheochromocytoma

Abstract

Pheochromocytoma is associated with intense physiologic effects of α- and β-adrenergic stimulation from catecholamine secretion. Perioperative management for these patients includes α-adrenergic receptor blockade, intravascular volume replacement, and, if necessary, β-adrenergic receptor blockade. Significant perioperative changes in preload and afterload, fluid status, heart rate and rhythm, and inotropy can occur and may be contrary to anesthetic management goals for patients with certain conditions of congenital heart disease. We report the perioperative management with doxazosin of a patient with single ventricle physiology and cavo-pulmonary and aorto-pulmonary lung perfusion who presented for resection of a pheochromocytoma.

In Brief

IMPLICATIONS: Both pheochromocytoma and single ventricle physiology are challenging conditions for anesthesiologists. We report the perioperative management of a patient with single ventricle, cavo-pulmonary and aorto-pulmonary lung perfusion who presented for resection of a pheochromocytoma. Careful α-adrenergic receptor blockade using doxazosin contributed to a successful preoperative, intraoperative, and postoperative course.

Patients with pheochromocytoma are traditionally treated with phenoxybenzamine to manage arterial blood pressure and intravascular volume contraction before surgical excision. Phenoxybenzamine, a nonspecific α-adrenergic receptor antagonist (α-blocker), causes vasodilation with decreased systemic vascular resistance (SVR) and decreased venous return requiring fluid administration. Presynaptic α-2 blockade can cause tachycardia, and β-adrenergic blockade may be required for heart rate control. Patients with certain types of palliated congenital heart disease can be very sensitive to changes in preload, afterload, inotropy, or arrhythmias. We report the use of doxazosin, instead of phenoxybenzamine, for preoperative α-blockade in a patient with complex single ventricle physiology presenting for pheochromocytoma resection.

Case Report

The patient was a 27-yr-old female born with double outlet right ventricle, pulmonary atresia, and hypoplastic left ventricle. After multiple cardiac procedures, the last one at age 9 yr, her cardiac anatomy at the time of surgery was characterized by a single right ventricle with discontinuous pulmonary arteries (Fig. 1). Her pulmonary blood flow originated from two sources: a superior cavo-pulmonary (modified “Glenn”) anastomosis to her right upper and middle pulmonary arteries and a central aorto-pulmonary 5 mm Gore-Tex® graft shunt to her right lower and left pulmonary arteries. She had a history of supraventricular tachycardia. Sotalol had resulted in torsade de pointes and her arrhythmia was controlled by atenolol. She was taking furosemide after episodes of pulmonary edema. After peripheral, but no central, thromboembolic events, she had been placed on aspirin to reduce thromboembolic risks.

Figure 1.: The patient's cardiac anatomy. DORV = double outlet right ventricle; hypo LV = hypoplastic left ventricle; LA = left atrium; PA = pulmonary atresia; RA = right atrium; RUL, RML, RLL = right upper, middle, lower lobe pulmonary artery, respectively; SVC = superior vena cava.

The patient had a long history of hypertension treated with losartan. Work-up for acute pulmonary edema associated with severe paroxysmal hypertension demonstrated increased levels of plasma norepinephrine (7147 pg/mL; normal, 70–750 pg/mL), plasma normetanephrine (8.49 mmol/L; normal, <0.9 mmol/L) and urine normetanephrine (3192 μg/24 h; normal, 103–390 μg/24 h). Plasma dopamine levels were slightly increased (88 pg/mL; normal, <30 pg/mL), and plasma epinephrine and metanephrine levels were normal. A 2.3 × 1.9 cm right adrenal mass on her abdominal computed tomography and a ¹³¹metaiodobenzylguanidine scan confirmed the diagnosis of pheochromocytoma.

On the day of admission, the patient's arterial blood pressure was between 116/64 mm Hg and 175/80 mm Hg, her heart rate was between 75 and 87 bpm, and her oxygen saturation between 78% and 83%. Her hematocrit was 55% and her platelet count was 196,000/μL. Coagulation studies were normal. Preoperative echocardiography showed mild regurgitation at her tricuspid, mitral, and aortic valves and mildly depressed right ventricular systolic function. Electrocardiogram revealed sinus rhythm, first-degree atrioventricular and right bundle branch block, and right axis deviation.

The patient was initially admitted to the intensive care unit (ICU) for close monitoring of heart rate and rhythm, arterial blood pressure, oxygen saturation, and fluid balance during initiation of α-blockade. Aspirin, furosemide, and losartan were discontinued and atenolol was reduced. Single daily doses of doxazosin were started at 0.5 mg and increased to 10 mg over 10 days. Systolic blood pressure was reduced from peaks more than 200 mm Hg to less than 150 mm Hg and asymptomatic orthostatic hypotension could be demonstrated. IV hydration with initiation of α-blockade resulted in mild clinical and radiological pulmonary fluid overload, and hydration was subsequently limited to oral intake ad libitum. The patient's weight gain was moderate, from 54 to 56 kg, and her hematocrit remained unchanged.

Doxazosin was held on the day of surgery. After premedication with IV midazolam, a radial arterial catheter was placed under local anesthesia, and standard ASA monitors were used. Anesthesia was induced using etomidate, sufentanil, and rocuronium accompanied by a small fluid bolus of 5 mL/kg without significant hypotension. Intubation of the patient's trachea was tolerated without hypertensive response. Anesthesia was maintained with a sufentanil infusion and a small concentration of isoflurane in air and oxygen. The patient was normoventilated. Because of prior cardiac catheterizations, access to the inferior vena cava (IVC) from the femoral vein was unsuccessful, and access to the superior vena cava (SVC) was not attempted to avoid jeopardizing the blood flow through the Glenn anastomosis. Open adrenalectomy was performed with a total operating time of 3 h. Manipulation of the tumor was minimized to avoid bursts of catecholamine secretion with vasoconstriction or arrhythmias. Blood loss was minimal and fluid administration was limited to maintenance plus 5 mL · kg⁻¹ · h⁻¹. Urine output was adequate throughout. Intraoperative hemodynamic variables and oxygen saturation remained very stable. Magnesium sulfate (1g bolus followed by an infusion of 1 g/h) was infused prophylactically and IV fenoldopam (0.02–0.1 μg · kg⁻¹ · min⁻¹) was titrated for minimal increases in arterial blood pressure during dissection of the mass. Catecholamine levels were drawn intraoperatively after induction, during tumor manipulation, and after adrenal vein ligation. Samples after induction were similar to preoperative samples with a minimal increase of dopamine, normal epinephrine, and significantly increased norepinephrine levels. During tumor manipulation, epinephrine levels reached 5.5 times the value of upper normal range, norepinephrine levels increased further, and dopamine level increased to 2.5 times upper normal range. After tumor removal, a prophylactic phenylephrine infusion was briefly initiated but discontinued before the end of surgery. The patient was extubated in the operating room and returned to the ICU. Postoperative hemodynamic variables and oxygen saturation remained stable. Reduced doses of atenolol and furosemide were restarted postoperatively. Postoperative analgesia was maintained through patient-controlled analgesia (PCA) with morphine. The patient was discharged home on the third postoperative day.

Discussion

Perioperative management of patients with pheochromocytomas has been reviewed in the literature (^1–3). α-blockade is used to control hypertension (⁴). Phenoxybenzamine preparation over a time up to 2 weeks preoperatively remains popular but other drugs, such as prazosine and doxazosin, have been used (⁴), and shorter preparation over several days only with shorter acting IV drugs such as urapidil (⁵) has been suggested. As vasoconstriction is relieved, aggressive intravascular fluid administration may become necessary to maintain circulating blood volume. Beta blockers or labetalol may be added after sufficient α-blockade to control reflex tachycardia or direct β-adrenergic stimulation. Intraoperative hypertension is managed preferentially with short-acting IV drugs such as phentolamine, sodium nitroprusside, fenoldopam, calcium channel antagonists, such as nicardipine, or, rarely, with adenosine or prostaglandin E1. Magnesium sulfate infusions have also been used during resection of pheochromocytomas (^6,7) as an adjunct to control hemodynamic variability. After resection of pheochromocytomas, relative catecholamine depletion may result in hypotension requiring additional intravascular fluid administration or temporary vasopressor infusion. In case of preparation with phenoxybenzamine, hypotension may persist for several days after tumor resection.

Reports of anesthetic management in patients with single ventricle physiology and pheochromocytoma are rare (⁸). Our patient was at particular risk from traditional preoperative α-blockade and perioperative management.

Pulmonary perfusion through a cavo-pulmonary connection is passive and depends on maintenance of an adequate transpulmonary perfusion pressure determined by the prepulmonary and postpulmonary filling pressure and the transpulmonary resistance (⁹). Our patient required adequate filling pressure in the SVC with risks of excessive filling documented by her history of pulmonary edema and diuretic dependency. Cavo-pulmonary blood flow would be reduced with decreased preload or venous return. Increased left atrial pressures from ventricular dysfunction or arrhythmias and high intraabdominal, intrathoracic, or airway pressures would reduce passive transpulmonary blood flow. Similarly, increased pulmonary vascular resistance (PVR), such as with hypercarbia, atelectasis, or pulmonary vascular response to stress, could decrease passive pulmonary blood flow through those lobes of her lung supplied by the cavo-pulmonary connection. Decreased pulmonary blood flow could result in desaturation and hypotension.

Arterial blood pressure, cardiac output, and the relative resistance of the systemic and the pulmonary vascular bed determine pulmonary perfusion through an aorto-pulmonary shunt. Pulmonary perfusion decreases with increasing PVR or with decreasing aortic pressure resulting from decreased SVR, hypovolemia, cardiac dysfunction, or arrhythmias. Decreased pulmonary perfusion would result in decreasing oxygen saturation. Pulmonary perfusion could increase with increasing SVR (for example, from acute vasoconstriction resulting from catecholamine release during tumor manipulation) or excessive pulmonary vasodilation (for example, with hyperventilation) and might result in pulmonary edema of those lobes of her lung supplied by the aorto-pulmonary shunt.

Our main concern was the risk from uncontrolled vasodilation with decreased venous return and decreased SVR from preoperative α-blockade. Doxazosin was felt to have advantages over phenoxybenzamine. Phenoxybenzamine is a noncompetitive, nonselective α-blocker with long duration of action and effects potentially for days postoperatively. Doxazosin, on the other hand, is a competitive, selective α-1-blocker with once daily dosing and reportedly rapid postoperative clearance (^10,11). We therefore assumed that desaturation from excessive reduction in SVR could be reversed by α-agonists if necessary. Because of the absence of presynaptic α-2-blockade, the risk of reflex tachycardia and tachyarrhythmia seemed reduced. Also, faster postoperative recovery of receptor function was expected, reducing the need for postoperative intravascular fluid administration in this patient sensitive to fluids and reducing the need for administration of vasopressors with unpredictable effect on her pulmonary perfusion. Very small starting doses for doxazosin were chosen and the dose was escalated slowly over 10 days to minimize hemodynamic changes and the need for aggressive intravascular fluid administration. Alpha blockade using doxazosin was tolerated well hemodynamically and in regard to oxygenation on room air as a marker of pulmonary to systemic perfusion balance.

Although adequate filling pressures were necessary in this patient for her cavo-pulmonary perfusion, excessive intravascular fluid administration was a concern considering her history of pulmonary edema. The cessation of diuretic therapy and IV hydration at the beginning of α-blockade resulted in temporary clinical and radiologic pulmonary fluid overload. Because slow preoperative α-blockade with doxazosin resulted in hemodynamic stability, oral fluid intake was subsequently allowed to regulate fluid requirement. Intraoperative and postoperative fluid administration could be limited because of her stable hemodynamic course and rapid recovery related to use of doxazosin.

The patient was at increased risk for ventricular dysfunction as a result of her long-standing condition of cyanotic heart disease with a single right ventricle and increased cardiac output because of her aorto-pulmonary shunt. There were no signs of catecholamine-induced cardiomyopathy, which is possible in patients with pheochromocytoma. A narcotic-based anesthetic was chosen to minimize negative inotropic effects and moderation of fluid administration was pursued.

Arrhythmias were a concern in this patient because of her arrhythmia history and potential intraoperative catecholamine release. Arrhythmias could have jeopardized her pulmonary perfusion as a result of decreased cardiac output and arterial blood pressure and increased ventricular and atrial end-diastolic pressures. Antiarrhythmic measures included perioperative continuation of β-adrenergic blockade and use of preoperative doxazosin instead of phenoxybenzamine. Magnesium sulfate was given both for its beneficial effect on hemodynamics and catecholamine release in pheochromocytoma and because of its antiarrhythmic quality (¹²), in particular in a patient with long-standing diuretic therapy.

Pressure measurement in both the SVC and the IVC or atrium would have allowed us to monitor perfusion pressure of the cavo-pulmonary lung perfusion and filling pressures for the single ventricle. However, IVC access proved technically impossible secondary to scarring and SVC access was intentionally avoided in order not to jeopardize pulmonary perfusion via the cavo-pulmonary anastomosis.

Epidural analgesia was considered because of its excellent analgesic effect and minimal respiratory depression with expected superior pulmonary perfusion. It was avoided because of concerns regarding potential epidural collaterals in patients with palliated congenital heart disease and the unpredictable effects of epidural-related hemodynamic effects of vasodilation and bradycardia on her pulmonary perfusion.

Laparoscopic adrenalectomy for pheochromocytoma may be associated with less intraoperative blood loss, shorter time to postoperative oral intake and to hospital discharge, and less postoperative pain; but surgery may take longer (¹³) and conversion to an open procedure may be necessary. Open adrenalectomy was chosen to avoid the effects of CO₂–peritoneum on hemodynamics (catecholamine release (^5,14), decreased preload, increased afterload, tachycardia, and hypertension) and ventilation (hypercarbia, need for increased ventilation with increased mean airway, and intrathoracic pressures) with untoward effects on pulmonary perfusion in this patient. Also, potentially faster control of venous drainage from the tumor was expected with an open procedure. Good postoperative analgesia with PCA, rapid postoperative recovery, and discharge home of our patient on postoperative day 3 compare favorably with reported outcomes of laparoscopic techniques.

In summary, this patient with a very complex cardiac anatomy and physiology, including single ventricle, cavo-pulmonary, and aorto-pulmonary lung perfusion, underwent resection of an adrenal pheochromocytoma. Use of doxazosin for preoperative α-blockade may have contributed to an uncomplicated perioperative course.