Parenteral Phenoxybenzamine During Congenital Heart Disease Surgery

II. Background

1. Description of the Problem One of the effects of cardiac operations involving cardiopulmonary bypass is reversible myocardial dysfunction lasting a number of days postoperatively. Typically this period of myocardial dysfunction is treated with vasoactive drugs aimed at improving contractility and reducing afterload. Problems exist with the current vasodilator therapy, including: variable response, inadequate response, complications of delivery mechanism with the potential for swings in blood pressure and ongoing dose adjustments. These are necessary due to the patients' autonomic nervous system reactivity changes during the post-cardiopulmonary bypass period. Evidence to support therapies targeting control of systemic vascular resistance include randomized controlled studies looking at the outcome of high risk neonates undergoing open cardiac procedures with very high dose synthetic narcotic anesthesia. This is known to reduce sympathetic nervous system activity. There exist numerous, uncontrolled but widely accepted, studies documenting the adjunctive use of vasodilators such as sodium nitroprusside, nitroglycerin and alpha-adrenergic blockers including phenoxybenzamine in the perioperative period. The results of all of these studies point towards a salutatory effect of measures to control vasoconstrictor responses on blood flow to the organs at risk, primarily the organs in the splenic circulation: the liver, kidneys, and intestines. Ischemia to these organs is responsible for a significant amount of morbidity in the post-bypass period including late onset sepsis and renal dysfunction. In addition, after myocardial injury the natural vasoconstrictor responses that mammalian organisms have to falling cardiac output are counterproductive and may initiate a cascade of responses, necessitating medical interventions culminating in overt myocardial pump failure. This then can lead to patient death or need for institution of extracorporeal circulatory support after the initial cardiopulmonary bypass period.
2. Physiologic Role of Phenoxybenzamine Numerous reports have demonstrated that with nearly complete blockage of alpha-adrenergic receptors (i.e. phenoxybenzamine) that both infusions of exogenous catecholamines and the neural release of endogenous catecholamines result not in vasoconstrictor responses but instead augmented myocardial contractility without increases in afterload. While medical interventions to effect circulatory responses are already in use commonly in the perioperative period, including synthetic narcotic analgesia and the use of short-acting vasodilator agents, these measures alone have not been universally effective in preventing the sort of hemodynamic deterioration described above. Therefore, we propose to use phenoxybenzamine, a drug which irreversibly binds to alpha-adrenergic receptors and some dopamine receptors, as an adjunct to the management of the perioperative vascular tone abnormalities in high risk infants and children undergoing cardiopulmonary bypass for open heart procedures.
3. Existing pharmacologic and clinical data Phenoxybenzamine (Dibenzyline: Wellspring Pharmaceutical Corporation, Bradenton, FL) is a haloalkylamine that irreversibly blocks both α-1 and α-2 adrenergic receptors. The drug exhibits a slightly higher affinity for the α-1 receptor. 12 There exists a body of literature concerning intravenous administration of phenoxybenzamine in the setting of congenital heart disease surgery, specifically in the Norwood repair for hypoplastic left heart syndrome. The use of phenoxybenzamine during cardiac surgery has been demonstrated to facilitate higher pump flow rates during cardiopulmonary bypass (CPB) and is associated with attenuation of postoperative metabolic acidosis. 13 Specifically, phenoxybenzamine has been shown to be more effective than sodium nitroprusside in improving tissue perfusion postbypass (as demonstrated by a comparison of rewarming characteristics), with lower base deficits in patients treated with phenoxybenzamine. 14 Indeed, to date there are many congenital heart disease surgical centers in the United States who, prior to the Norwood procedure, administer 0.25 mg/kg of phenoxybenzamine at the initiation of cardiopulmonary bypass in an attempt to optimize systemic organ perfusion in the perioperative period. Such a protocol, at this dose specifically, has been described to improve systemic oxygen delivery in patients undergoing the Norwood procedure for hypoplastic left heart syndrome, as well as an improvement in survival to future surgical palliations. 10

III. Aims and Objectives

Our general hypothesis is that a blockade of vasoconstrictor responses in the period around cardiopulmonary bypass will result in better organ preservation and improvement in cardiac output postoperatively. More specifically, as serum lactate serves as a surrogate for demonstrating the adequacy of end organ perfusion, we will utilize this continuous variable as our primary endpoint for the purposes of this protocol. Our hypothesis is that the use of phenoxybenzamine in this select population will reduce initial postoperative lactate levels by a clinically-relevant level of 25%, relative to historical controls. Secondary endpoints evaluated as well will include utilization of inotropic support, duration of hospitalization, and time to resolution of postoperative lactic acidosis. Both physiologic and outcome variables will be examined and compared to a matched cohort of patients drawn from our recent experience without the use of phenoxybenzamine.

IV. Patient Selection and Clinical Management

Patient selection will be determined by an assessment of the risk of systemic ventricular dysfunction following open cardiac repair in a population of infants undergoing stage I palliation (Norwood procedure) for the diagnosis of either hypoplastic left heart syndrome or similar left-sided obstructive lesions in the setting of single-ventricle physiology. Eligible neonates and infants include those aged 0 days to 6 months. These patients will be evaluated on an individual basis and the decision to give phenoxybenzamine would be determined by the attending surgeon, anesthesiologist, and cardiologist. No other change in the patients' perioperative management will occur. Efficacy will be assessed by evaluating the need for vasodilators and inotropic agents, as well as by evaluating the effects of alpha-blockade on the clinical course and hemodynamics of the patient. Comparison to a matched historical cohort of patients not receiving the drug will be performed. Likewise, the historical cohort population to be included in this population include all infants who have undergone the Norwood procedure for either hypoplastic left heart syndrome or those with a similar univentricular lesion requiring aortic arch reconstruction. Aside from the importance of meeting protocol inclusion criteria, there are no specific criteria to exclude subjects

1. Drug Administration and Safety Monitoring

Parenteral phenoxybenzamine requires an IND for human administration, which has been submitted to the FDA and is pending (see attached). The only significant risk of parenteral phenoxybenzamine has been excessive alpha-adrenergic blockade resulting in diastolic hypotension. This problem has been treated with intravenous adrenergic support in the form of norepinephrine or vasopressin. Precautions in place during the use of phenoxybenzamine will include physician presence during administration, invasive arterial blood pressure monitoring, and inotropic agents available for immediate use. This drug will be obtained from WellSpring Pharmaceuticals and stored in the Investigational Drug Pharmacy.

The drug will be administered in the operating room. After induction of anesthesia and the pIacement of appropriate cardiovascular monitoring lines, an initial loading dose of 0.25 mg/kg will be administered intravenously immediately prior to cardiopulmonary bypass.For up to 72 hours postoperatively, 0.25 mg/kg/day will be administered Based on published pharmacokinetic data these doses should block 90 -95% of alpha-peripheral receptors with a half life of 24 - 36 hours for regeneration. This period of time corresponds nicely with the period of highest hemodynamic vulnerability of this patient population. No alteration in usual perioperative monitoring will occur solely to allow drug administration. Monitoring of anesthesia, preload, afterload, contractility, and cardiac output will be maintained throughout the period of administration and recovery in ways which are at present, performed in this patient population: invasive transthoracic cardiac lines, echocardiography, venous and systemic oximetry, and analysis of acid/base abnormalities. Secondary end-organ function will be monitored including kidneys (urine output, creatinine,) and brain (cerebral oximetry).

Potential toxicity of the drug is related to the effects of excessive vasodilatation. Subjects will be monitored for toxicity through both frequent physical examination as well as continuous blood pressure monitoring in the immediate postoperative period. The ultimate rescue therapy for excessive vasodilation and hypotension would be the institution of high flow extracorporeal circulation, which we would be in a position to provide in a most timely fashion in both the operating room and intensive care unit. Pharmacologic management to counteract adverse effects of alpha-adrenergic blockade would be norepinephrine, administered in the dose 0.01-0.05 mcglkglmin. Epinephrine may be relatively contraindicated in this setting because beta-adrenergic receptors are left unopposed. Therefore, drugs that stimulate both types of receptors (e.g., epinephrine) may produce an exaggerated hypotensive response and tachycardia. Those subjects experiencing hypotension refractory to conventional postoperative management, who are experiencing end-organ dysfunction related to hypotension, or who require norepinephrine reversal of the previous dose of phenoxybenzamine would be strongly considered for withdrawal from the study.

Adverse reactions reported in oral administration include nasal congestion, mycosis, tachycardia, drowsiness and fatigue. This drug has been FDA approved for oral administration in treatment of pheochromocytoma to control episodes of hypertension and sweating. It has also shown efficacy in micturition disorders resulting from neurogenic bladder, functional outlet obstruction and partial prosthetic obstruction.

Subjects will be followed throughout their hospital course until the time of discharge home. There will be no further follow up asked of subjects related to this study. Historical cohort analysis will include analysis of clinically-relevant endpoints as noted below, as well as laboratory values and vital signs during the perioperative period and until time of discharge home.

A three-physician panel, comprised of those involved in operative and postoperative care of this particular patient population (but not an investigator in this study) will be appointed to review subject cases. The panel will meet at 5-subject intervals to review the data and safety profile from each case.

VI. Data Storage, Acquisition, and Analysis An electronic database, which is password-protected and available only to the PI and co-investigators will be utilized for the purposes of data analysis. A copy of the data acquisition form accompanies this application. The data will be destroyed 5 years following completion of the study and analysis of the resulting data. Data acquired for the study will be stored in a password-protected database. Only the PI and the co-investigators will have access to this password and the database. Upon identification of eligible historical cohorts, their charts will be reviewed with pertinent data entered in the above-described database. The data of historical cohorts will be de-identified as an attempt to maintain confidentiality. Subject ID numbers in the historical cohort group will be assigned only to differentiate between subjects for the purpose of data analysis, and will not specifically identify subjects.

Data analysis will include several clinically-relevant endpoints, including overall surgical mortality (survival to discharge), time to initial extubation and cessation of mechanical ventilation, and length of hospital stay. End-organ function status post cardiopulmonary bypass will be assessed by evaluating lab values currently studied, including serum lactate. Monitoring of hemodynamic values including blood pressure and left atrial pressure will be utilized to indirectly assess cardiac output. Other end organs evaluated will include kidney (urine output and other tests of renal function such as creatinine) and brain (clinical evaluation and tests ordered as indicated, as well as cerebral oximetry). We will also evaluate the need for inotropic and/or vasodilator support (both dose and duration). The study will not require additional lab or radiology testing based on patient inclusion. Data reviewed from historical cohorts include those as specified in the data collection sheet, namely the above-described clinically-relevant endpoints, as well as laboratory values (serum lactate, serum creatinine), vital signs (blood pressure, cerebral oximetry, and left atrial pressure), evidence of end organ perfusion (urine output), and postoperative inotrope requirements.

As serum lactate serves as a surrogate for demonstrating the adequacy of end organ perfusion, we will utilize this continuous variable as our primary endpoint for the purposes of this protocol. Our hypothesis is that the use of phenoxybenzamine in this select population will reduce initial postoperative lactate levels by a clinically-relevant level of 25%, relative to historical controls. Secondary endpoints evaluated as well will include utilization of inotropic support, duration of mechanical ventilation and overall hospitalization, and time to resolution of postoperative lactic acidosis.

Recent experience with the Norwood procedure, the conventional palliative surgical therapy for hypoplastic left heart syndrome, demonstrated a mean initial serum lactate of 7.3 ± 3.3 mmol/L in 21 neonates. 15 Assuming a clinically significant reduction by 25% to 5.5 mmol/L (at 80% power and an assumed p ≤ 0.05 demonstrating statistical significance), one would require a total of 54 study subjects. Given an anticipated surgical volume of approximately 20 Norwood procedures annually, along with a 15% dropout or refusal rate, we would expect to accrue a total required volume of 62 subjects in approximately 3 years. We will also compare data with data that already exists within our cardiac registry for further evaluation.

VII. Consent and Peer Judgement

The drug will not be administered without the knowledge of attending surgeon and anesthesiologist and after discussion with the perfusionists in order to ensure safety and advisability. In the postoperative period the drug will not be administered without discussion with all relevant participants in the patient's postoperative care. Informed consent will be obtained from the patients family and documented in the chart prior to administration of the drug. Because this drug is not yet FDA approved in the United States, specific consent is necessary. An investigational new drug application has been filed and permission from the FDA is pending. Significant adverse effects will be reported to the Food and Drug Administration and to the Vanderbilt University Institutional Review Board Health Sciences Committee within 24 hours, and annual status reports will be filed with Food and Drug Administration according to their policies.