The updated guidelines for the determination of brain

Transcription

1 Developing a Standard Method for Apnea Testing in the Determination of Brain Death for Patients on Venoarterial Extracorporeal Membrane Oxygenation: A Pediatric Case Series Rima J. Jarrah, MD 1 ; Samuel J. Ajizian, MD 1 ; Swati Agarwal, MD 2 ; Scott C. Copus, RRT 3 ; Thomas A. Nakagawa, MD 1 Objective: The revised guidelines for the determination of brain death in infants and children stress that apnea testing is an integral component in determining brain death based on clinical criteria. Unfortunately, these guidelines provide no process for apnea testing during the determination of brain death in patients supported on venoarterial extracorporeal membrane oxygenation. We review three pediatric patients supported on venoarterial extracorporeal membrane oxygenation who underwent apnea testing during their brain death evaluation. This is the only published report to elucidate a reliable, successful method for apnea testing in pediatric patients supported on venoarterial extracorporeal membrane oxygenation. Design: Retrospective case series. Setting: Two tertiary care PICUs in university teaching hospitals. Patients: Three pediatric patients supported by venoarterial extracorporeal membrane oxygenation after cardiopulmonary arrest. Interventions: After neurologic examinations demonstrated cessation of brain function in accordance with current pediatric brain death guidelines, apnea testing was performed on each child while supported on venoarterial extracorporeal membrane oxygenation. Measurements and Main Results: In two of the three cases, the patients remained hemodynamically stable with normal oxygen saturations as venoarterial extracorporeal membrane oxygenation 1 Department of Anesthesiology, Wake Forest School of Medicine, Winston-Salem, NC. 2 Department of Pediatrics, Section of Pediatric Critical Care, Inova Fairfax Hospital for Children, Falls Church, VA. 3 Department of Respiratory Care, Wake Forest Baptist Health, Brenner Children s Hospital, Winston-Salem, NC. Supported by departmental funds. Dr. Ajizian is a consultant for Covidien. Dr. Nakagawa is a consultant for the U.S. Department of Health and Human Services, Health Resource Service Administration, and The Organ Donation and Transplantation Alliance. The remaining authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, rjarrah@wakehealth.edu Copyright 2013 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies DOI: /PCC sweep gas was weaned and apnea testing was undertaken. Apnea testing demonstrating no respiratory effort was successfully completed in these two cases. The third patient became hemodynamically unstable, invalidating the apnea test. Conclusions: Apnea testing on venoarterial extracorporeal membrane oxygenation can be successfully undertaken in the evaluation of brain death. We provide a suggested protocol for apnea testing while on venoarterial extracorporeal membrane oxygenation that is consistent with the updated pediatric brain death guidelines. This is the only published report to elucidate a reliable, successful method for apnea testing in pediatric patients supported on venoarterial extracorporeal membrane oxygenation. (Pediatr Crit Care Med 2014; 15:00 00) Key Words: apnea testing; brain death; end-of-life; extracorporeal membrane oxygenation; pediatrics The updated guidelines for the determination of brain death for infants and children provide important direction for the intensivist tasked with determining death (1). However, the consensus guidelines do not offer a process for apnea testing when a patient is supported on venoarterial (VA) extracorporeal membrane oxygenation (ECMO), as no supporting literature exists for this clinical situation. In fact, the pediatric and adult literature is bereft of reports regarding apnea testing on VA-ECMO. A 1991 review of infants who died on VA-ECMO cited brain death in 29% of their study group (2). This article unfortunately did not specifically address how apnea testing or clinical documentation of brain death was determined. The authors note [i]nfants were diagnosed as having brain death after ultrasound evidence of brain injury forced termination of ECMO (2). This definition does not meet current or past clinical guidelines for determining brain death in infants and children. As early as 1987, the American Academy of Pediatrics published guidelines for brain death determination in children and infants (3). Those guidelines set Pediatric Critical Care Medicine 1

2 Jarrah et al forth specific criteria in the areas of history, physical examination, and ancillary testing needed to determine brain death in children. Although intracranial bleeding may certainly lead to brain death, neither ultrasound nor pathologic findings of intracranial hemorrhage are sufficient to diagnose irreversible cessation of brain function. In a comprehensive review of 87 adult patients (15 yr and older) over an 8-year period, patients who lost all brainstem reflexes were evaluated for apnea while supported on VA-ECMO (4). Three patients met criteria. In two cases, apnea testing was deemed too difficult to perform, no further explanation is given. In the third case, support was withdrawn before apnea testing could be performed (4). The updated pediatric brain death guidelines for infants and children address apnea testing for patients undergoing clinical evaluation for brain death. The guidelines recommend baseline and serial arterial blood gas (ABG) measurements to determine Paco 2 levels during apnea testing. The apnea test is continued until Paco 2 rises to greater than or equal to 60 mm Hg and greater than or equal to 20 mm Hg above baseline. The patient is continually monitored for respiratory activity during apnea testing. Any respiratory effort is inconsistent with brain death, and the patient should be placed back on ventilatory support. Testing for brain death by clinical examination and apnea testing may be repeated at a later time. Alternatively, an ancillary study can be pursued to assist with the determination of neurologic death. The updated guidelines also stress apnea testing must be performed safely and should be aborted if the patient desaturates to less than 85%, develops hemodynamic instability, or if a Paco 2 level of 60 mm Hg cannot be safely achieved (1). Unlike patients on mechanical ventilation alone, gas exchange for patients supported on VA-ECMO occurs mainly through the ECMO oxygenator. Carbon dioxide elimination is dependent on the sweep gas flow rate, which must be reduced to allow Paco 2 to rise to appropriate levels to test for apnea. Apnea testing for patients supported on VA-ECMO has never been described in the pediatric literature. We review three cases of children who met clinical examination criteria for brain death and underwent apnea testing while supported on VA-ECMO. Institutional review boards waived the need for approval in our review of these cases. CASE 1 A 2½-year-old, 13.6 kg, African American child with a history of sickle cell anemia was admitted for suspected pain crisis involving the abdomen. One day after admission, the patient developed acute severe congestive heart failure. The patient was treated with milrinone, furosemide, supplemental oxygen, and transfused packed RBCs. Subsequently, the child developed leftward eye deviation with clenching of the jaw and became bradycardic and then asystolic. Pediatric Advanced Life Support (PALS) and cardiopulmonary resuscitation (CPR) were initiated. The patient was cannulated for VA- ECMO (Medtronics, Woods Cross, UT; MAQUET Quadrox D oxygenator, Wayne, NJ). The child had return of spontaneous circulation (ROSC) following 42 minutes of CPR. VA-ECMO was initiated 44 minutes postarrest. The child remained unresponsive and the neurologic examination deteriorated with bilateral fixed and dilated pupils. A computed axial tomographic scan of the brain showed diffuse hypoxic/ischemic insult in a watershed distribution, with herniation. After meeting all prerequisite criteria, brain death testing was initiated in accordance with the updated pediatric brain death guidelines and was consistent with loss of all brainstem reflexes (1). The patient was normothermic (36 C) and acceptable hemodynamics were maintained with VA-ECMO flows of 88 ml/kg/min (Table 1). Prior to initiation of apnea testing, the patient was preoxygenated via the VA-ECMO circuit with an Fio 2 of 1.0. Sweep gas flow rate at the initiation of apnea testing was 0.5 L/min. Following preoxygenation and normalization of Paco 2, ABG analysis showed a ph of 7.48, Paco 2 of 39 mm Hg, and Pao 2 of 596 mm Hg (Table 2). The child was placed on continuous positive airway pressure (CPAP) with an Fio 2 of 1.0 administered through a flow-inflating anesthesia bag connected to the endotracheal tube. VA-ECMO flows were maintained at 88 ml/kg/min and Fio 2 at 1.0 while sweep gas flow was titrated down. Continuous blood gas monitoring via CDI blood parameter monitoring system (CDI 500, Terumo Cardiovascular Systems, Ann Arbor, MI) was observed during apnea testing. Blood gas information during the apnea test as sweep gas flow rate was decreased is shown in Table 2. The patient maintained adequate hemodynamics throughout the testing period with no hypotension (Table 1). No spontaneous respiratory effort during apnea testing was noted, with documented Paco 2 greater than 60 mm Hg and greater than or equal to 20 mm Hg increase above the baseline Paco 2. CASE 2 A 14-year-old, 100 kg, African American child was admitted for open biopsy of a large retroperitoneal mass causing compression of the inferior vena cava. The intraoperative and immediate postoperative course was uncomplicated. On postoperative day 1, the patient had syncope, progressing to lethargy and cardiovascular collapse. The child became bradycardic, followed by asystole. PALS and CPR were initiated, and the decision was made to cannulate for VA-ECMO (Medtronics; MAQUET Quadrox D oxygenator) was made after 60 minutes of resuscitation. The patient progressed to complete unresponsiveness while on VA-ECMO. A head CT revealed diffuse hypoxic ischemic insult with central herniation. After meeting all prerequisite criteria, brain death testing was initiated in accordance with the updated pediatric brain death guidelines and was consistent with loss of all brainstem reflexes (1). The patient was normothermic (37.2 C) and acceptable hemodynamics were maintained with VA-ECMO flows of 38 ml/kg/min. Prior to initiation of apnea testing, the patient was preoxygenated via the VA- ECMO circuit with an Fio 2 of 1.0. Sweep gas flow rate at the initiation of apnea testing was 5.5 L/min. Following preoxygenation, ABG analysis showed a ph 7.5, Paco mm Hg, and Pao mm Hg (Table 2). Oxygen saturation was 100% 2 February 2014 Volume 15 Number 2

3 Online Clinical Investigation Table 1. Summary of Patient Characteristics Characteristics Case 1 Case 2 Case 3 Therapeutic hypothermia? No No Yes. Examination was undertaken 24 hr after rewarming Time on ECMO when examination done 24 hr 22 min 18 hr 45 min > 48 hr Time between last sedation/neuromuscular blockade and examination > 24 hr 8 hr > 24 hr Temperature at the time of examination 36.1 C 37.2 C 36.5 C VA-ECMO flows during apnea test 88 ml/kg/min 38 ml/kg/min 140 ml/kg/min Lowest recorded Spo 2 during apnea test Lowest recorded blood pressure/map during apnea test Head CT Ancillary testing (i.e., EEG, cerebral blood flow) Time from arrest to return of spontaneous circulation 100% (calculated from arterial blood gas) 100% 78% MAP = /50 39/34 Diffuse hypoxic ischemic insult in a watershed distribution, with herniation Diffuse hypoxic ischemic insult with central herniation None None None No evidence of electrical activity on EEG 42 min 76 min (when placed on VA-ECMO) Not available Time from arrest to ECMO 44 min 76 min ~60 min ECMO = extracorporeal membrane oxygenation, VA-ECMO = venoarterial extracorporeal membrane oxygenation, MAP = mean arterial pressure, EEG = electroencephalogram. on Fio 2 of 1.0 via ECMO circuit. The patient was placed on CPAP with an Fio 2 of 1.0 administered through a flow-inflating anesthesia bag connected to the endotracheal tube. VA- ECMO flows were maintained at 38 ml/kg/min and Fio 2 at 1.0 while sweep gas flow was weaned from 5.5 to 1.0 L/min over 60 minutes (Table 2). The child remained hemodynamically stable and normothermic during apnea testing. Continuous blood gas monitoring via CDI monitor was observed during apnea testing. After 60 minutes, an ABG was obtained, with the following results: ph 7.21, Paco 2 62 mm Hg, and Table 2. Results of Weaning Sweep Flow on Arterial Blood Gases for Three Patients on Venoarterial Extracorporeal Membrane Oxygenation Apnea Test Time (min) ECMO Pump Flow (ml/kg/min) ECMO Sweep Flow (Fio 2 1.0) (L/min) Arterial Blood Gas (ph/paco 2 /Pao 2 /Hco 3 ) Case /39/596/ /47/497/ /57/532/ /73/564/30 Case /32/207/ /62/73/25 Case /40/135/ /62.4/390/28 After, the patients were connected to a flow-inflating bag system providing continuous positive airway pressure and Fio 2 of 1.0. Venoarterial extracorporeal membrane oxygenation flows were not changed during apnea testing. Pediatric Critical Care Medicine 3

4 Jarrah et al Pao 2 73 mm Hg (Table 2). Saturations remained 100% during apnea testing. The patient maintained adequate hemodynamics throughout the testing period with no hypotension (Table 1). No spontaneous respiratory effort during apnea testing was noted, with documented Paco 2 greater than 60 mm Hg and greater than or equal to 20 mm Hg increase above the baseline Paco 2. CASE 3 A 5-month-old infant with complex congenital heart disease developed bradycardia followed by cardiac arrest. PALS and CPR were initiated, and after approximately 1 hour of CPR without ROSC, the child was cannulated for VA-ECMO (Medtronics; MAQUET Quadrox D oxygenator). The child was maintained on full VA-ECMO support with flows of 140 ml/kg/min. The patient underwent 24 hours of hypothermia to 34 C postarrest, followed by gradual rewarming on the second day of VA-ECMO. On ECMO day 3, the child was unresponsive despite receiving no sedative or neuromuscular blocking agents for more than 24 hours. After meeting all prerequisite criteria, brain death testing was initiated in accordance with the updated pediatric brain death guidelines and was consistent with loss of all brainstem reflexes (1). Electroencephalogram showed no evidence of electrical activity. The patient was normothermic (36.5 C) and acceptable hemodynamics were maintained with VA-ECMO flows of 140 ml/ kg/min. Prior to initiation of apnea testing, the patient was preoxygenated via the VA-ECMO circuit with an Fio 2 of 1.0. Sweep gas flow rate at the initiation of apnea testing was 0.22 L/min. Following preoxygenation and normalization of Paco 2, ABG analysis showed a ph 7.41, Paco 2 40 mm Hg, and Pao mm Hg (Table 2). The infant was disconnected from mechanical ventilatory support and placed on CPAP with an Fio 2 of 1.0 administered through a flow-inflating anesthesia bag connected to the endotracheal tube. VA-ECMO flows were maintained at 140 ml/kg/min and Fio 2 at 1.0. The patient was continuously monitored and no respiratory effort was noted over 11 minutes. Continuous blood gas monitoring via CDI parameter monitoring system demonstrated that Paco 2 increased to 65 during this time period. Temperature remained unchanged during apnea testing. However, blood pressure gradually decreased to 39/34 mm Hg with a heart rate of 168. Oxygen saturations decreased to 78% although Pao 2 on the CDI was greater than 100. After 11 minutes, an ABG revealed a ph 7.26, Paco mm Hg, and Pao mm Hg, and the apnea test was terminated. DISCUSSION Brain death is characterized by the absence of brainstem reflexes and apnea in the presence of a known cause of irreversible coma (5). Apnea testing is essential to demonstrate the failure of the medullary respiratory centers to respond to hypercapnia and acidosis. Successful apnea testing for patients on VA- ECMO has not been documented in the literature. Apnea testing in this patient population is challenging, as no defined or uniform standards exist. Additionally, apnea testing must be done in a manner that does not compromise the safety of the patient. Attention to pretest conditions can minimize potentially adverse circulatory effects during apnea testing (5, 6). Various techniques have been described for performing apnea testing in ventilated patients (1, 7). A flow-inflating bag system with supplemental oxygen connected to the endotracheal tube is a safe and commonly used method for apnea testing (8). The updated pediatric brain death guidelines recommend removing the patient from mechanical ventilation as newer ventilators will revert to a backup mode of ventilation once apnea is sensed (1). Additionally, false triggering of the ventilator can occur simulating respiration if sensitivity of the ventilator is not adjusted appropriately (1, 5). Tracheal insufflation of oxygen has been used, but caution is warranted to ensure adequate excursion of gases to prevent barotrauma and co 2 washout preventing adequate rise of Paco 2 during apnea testing (1). Oxygenation and ventilation in patients mechanically supported on VA-ECMO relies on blood passing through an oxygenator. Deoxygenated blood in the ECMO circuit passes through the oxygenator and is circulated back to the patient. Ventilation is accomplished by adjusting the flow of the sweep gas, which directly influences co 2 removal. In a patient who is fully supported on VA-ECMO, nearly all ventilation is accomplished using the ECMO oxygenator. VA-ECMO presents challenges to determine apnea for brain death testing because oxygenators are highly efficient and may not allow the Paco 2 to rise to appropriate levels. Our patients were placed on a CPAP circuit using a flowinflating anesthesia bag supplied with an Fio 2 of 1.0. Once ventilatory support was terminated and patients were connected to the flow-inflating circuit, the ECMO sweep gas was decreased allowing the Paco 2 to rise. In all cases, sweep gas Fio 2 remained at 100% while supported on VA-ECMO. Paco 2 was continually monitored through the arterial limb of the ECMO circuit using the CDI blood parameter monitoring system for continuous blood gas measurement. An ABG sample was analyzed to confirm the Paco 2 measurement and readings from the CDI monitoring system. In mechanically ventilated patients, Paco 2 rises in a linear fashion by approximately 3 5 mm Hg/min during apnea testing for brain death once a patient is removed from ventilatory support (9 11). In cases of severe brain injury, this increase may be variable because the injured or dead brain does not contribute to co 2 production. In two cases described, the rate of Paco 2 rise was much slower and attributable to weaning of sweep gas flow (Table 2). The rate of Paco 2 rise on ECMO is a function of the co 2 removal rate of the oxygenator, which, in turn, is proportional to ECMO circuit flow, sweep gas flow, and size of the oxygenator. We are not aware of any literature to guide the clinician in the rate of Paco 2 rise while on VA-ECMO. A recent article proposed adding co 2 to the oxygenator to elevate the Paco 2 to the required levels to demonstrate apnea (4). However, the researchers were unable to perform this maneuver as their patients were too unstable for apnea testing. 4 February 2014 Volume 15 Number 2

5 Table 3. Proposed Process for Apnea Testing of Pediatric Patients Supported on Venoarterial Extracorporeal Membrane Oxygenation 1. Normalize Paco 2 by adjusting ECMO sweep gas and obtain a baseline ABG analysis 2. Preoxygenate the patient on VA-ECMO by increasing sweep gas Fio 2 to 1.0 Online Clinical Investigation 3. Remove the patient from mechanical ventilation and connect endotracheal tube to a flow-inflating bag system with adequate continuous positive airway pressure and Fio Maintain the ECMO sweep gas Fio 2 at 1.0. Decrease ECMO sweep gas flow to as low as possible, without precipitating hypoxia. We find that a flow of 0.1 L/min for smaller children and 1 L/min for adult-sized children is well tolerated and allow for successful apnea testing 5. A CDI monitor can assist with determination of Paco 2, but must be confirmed with an ABG. If a CDI monitoring system is not used, serial ABGs should be obtained at 5-min intervals throughout apnea testing until, 1) the Paco 2 reaches the appropriate threshold noted in the current guidelines; 2) the patient shows signs of spontaneous respiratory effort; or 3) the patient becomes hemodynamically unstable 6. If the patient exhibits respiratory effort, desaturates to < 85%, or develops hemodynamic instability, the apnea test should be terminated (1). Maintaining circulatory support in the event of hypotension may be accomplished in patients supported on VA-ECMO by increasing circuit flow or using additional inotropic agents 7. If the apnea test cannot be successfully completed, an ancillary study can be pursued to assist with the determination of brain death or another attempt to test for apnea can be attempted at a later time ECMO = extracorporeal membrane oxygenation, ABG = arterial blood gas, VA-ECMO = venoarterial extracorporeal membrane oxygenation. Criteria for brain death testing rely on successful and safe completion of apnea testing in conjunction with a clinical examination that is consistent with brain death. Prerequisites prior to brain death testing must be met to ensure variables that could interfere with testing have been addressed and treated (1). Brain death can be determined solely on clinical criteria and apnea testing. Normalization and maintenance of blood pressure and oxygenation are essential during apnea testing. If the patient desaturates or becomes hemodynamically unstable during apnea testing, the apnea test should be aborted. An ancillary study to assist with determination of brain death or another attempt to perform apnea testing at a later time can be pursued to see if apnea testing is better tolerated under different circumstances. Ancillary studies such as cerebral blood flow or electroencephalogram can assist with determination of brain death; however, maintaining normal hemodynamic variables is essential when conducting brain death testing. We suggest the following processes when conducting apnea testing for brain death on VA-ECMO (Table 3): 1. Normalize Paco 2 by adjusting ECMO sweep gas and obtain a baseline ABG analysis. 2. Preoxygenate the patient on VA-ECMO by increasing sweep gas Fio 2 to Remove the patient from mechanical ventilation and connect endotracheal tube to a flow-inflating bag system with adequate CPAP and Fio Maintain the ECMO sweep gas Fio 2 at 1.0. Decrease ECMO sweep gas flow to as low as possible, without precipitating hypoxia. We find that a flow of 0.1 L/min for smaller children and 1 L/min for adult-sized children is well tolerated and allows for successful apnea testing. 5. A CDI monitor for continuous blood gas assessment can assist with determination of Paco 2. Measurements must be confirmed with an ABG. If a CDI monitoring system is not used, serial ABGs should be obtained at 5-minute intervals throughout apnea testing until 1) the Paco 2 reaches the appropriate threshold noted in the current guidelines; 2) the patient shows signs of spontaneous respiratory effort; or 3) the patient becomes hemo dynamically unstable. 6. If the patient exhibits respiratory effort, desaturates to less than 85%, or develops hemodynamic instability, the apnea test should be terminated (1). Maintaining circulatory support in the event of hypotension may be accomplished in patients supported on VA-ECMO by increasing circuit flow or using additional inotropic agents. 7. If the apnea test cannot be successfully completed, an ancillary study can be pursued to assist with the determination of brain death or another attempt to test for apnea can be tried at a later time. To our knowledge, we report the only published cases of apnea testing to determine brain death in children supported on VA-ECMO. We describe a method for apnea testing to assist clinicians declaring brain death by clinical criteria for patients supported on VA-ECMO. REFERENCES 1. Nakagawa TA, Ashwal S, Mathur M, et al; Society of Critical Care Medicine; Section on Critical Care and Section on Neurology of the American Academy of Pediatrics; Child Neurology Society: Guidelines for the determination of brain death in infants and children: An update of the 1987 Task Force recommendations. Crit Care Med 2011; 39: Price MR, Galantowicz ME, Stolar CJ: Congenital diaphragmatic hernia, extracorporeal membrane oxygenation, and death: A spectrum of etiologies. J Pediatr Surg 1991; 26: ; discussion American Academy of Pediatrics Task Force on Brain Death in Children. Guidelines for the determination of brain death in children. Pediatrics 1987; 80: Pediatric Critical Care Medicine 5

6 Jarrah et al 4. Muralidharan R, Mateen FJ, Shinohara RT, et al: The challenges with brain death determination in adult patients on extracorporeal membrane oxygenation. Neurocrit Care 2011; 14: Wijdicks EF, Rabinstein AA, Manno EM, et al: Pronouncing brain death: Contemporary practice and safety of the apnea test. Neurology 2008; 71: Goudreau JL, Wijdicks EF, Emery SF: Complications during apnea testing in the determination of brain death: Predisposing factors. Neurology 2000; 55: Wijdicks EF, Varelas PN, Gronseth GS, et al; American Academy of Neurology: Evidence-based guideline update: Determining brain death in adults: Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2010; 74: Lévesque S, Lessard MR, Nicole PC, et al: Efficacy of a T-piece system and a continuous positive airway pressure system for apnea testing in the diagnosis of brain death. Crit Care Med 2006; 34: Outwater KM, Rockoff MA: Apnea testing to confirm brain death in children. Crit Care Med 1984; 12: Rowland TW, Donnelly JH, Jackson AH: Apnea documentation for determination of brain death in children. Pediatrics 1984; 74: Paret G, Barzilay Z: Apnea testing in suspected brain dead children Physiological and mathematical modelling. Intensive Care Med 1995; 21: February 2014 Volume 15 Number 2