The Role of ECMO in Cardiac Arrest Care

Welcome to PICU Doc On Call, a podcast dedicated to current and aspiring intensivists. My name is Pradip Kamat. My name is Rahul Damania and we come to you from Children’s Healthcare of Atlanta-Emory University School of Medicine.

Today's episode Is part two of our pediatric post-cardiac arrest care syndrome

If you have not yet listened to part one, I would highly encourage you to visit that episode prior to delving into this one.

Part 1 addressed the epidemiology, causes, and pathophysiology of POST CARDIAC ARREST SYNDROME.

Part 2 Today will discuss management and complications related to post-cardiac arrest syndrome in the ICU.

To revisit our index case we had a:

• 11 yo previously healthy M who was admitted to the PICU after cardiac arrest. After stabilization: The patient was taken to head CT which showed diffuse cerebral edema and diffusely diminished grey-white differentiation most pronounced in the basal ganglia. He is now 18-24 hours post-cardiac arrest and the team is dealing with hemodynamic changes, arrhythmias, and difficulty with ventilation. The patient’s neurological exam still remains poor with fixed 5 mm pupils and upper motor neuron signs in the lower extremities.

Let’s get right into it:

• What are some of the principles in management of patients with post cardiac arrest syndrome (PCAS)? Where do we keep the patients blood pressure?
• Hypotension after ROSC is commonly encountered in children with PCAS. Early hypotension occurred in 27% of children after cardiac arrest is associated with lower survival to hospital discharge and unfavorable neurological outcome. When post-cardiac arrest hypotension is present, it is not clear whether increasing the blood pressure through administration of fluids and inotropes/vasopressors can mitigate harm, despite this 41% of patients under 18 receive vasopressor therapy within the first 6 hours after ROSC. Currently, there is no high-quality evidence to support any single specific strategy for post-cardiac arrest hemodynamic optimization in children. Treatment of post-cardiac arrest hypotension and myocardial dysfunction may be assisted by monitoring and evaluating arterial lactate and central venous oxygen saturation. Parenteral fluids, inotropes, and vasoactive drugs are to be used as needed to maintain a systolic blood pressure greater than the fifth percentile for age. Appropriate vasoactive drug therapies should be tailored to each patient and adjusted as needed.
• What about cardiac arrhythmia's such as Vtach seen in our patient?
• The rhythm disturbances observed during the post-cardiac arrest period include premature atrial and ventricular contractions, supraventricular tachycardias, and ventricular tachycardias. Heart block is unusual but can be observed as a manifestation of myocarditis. There is inadequate evidence in adults and no published studies in children to support the routine administration of prophylactic antiarrhythmics after ROSC, but rhythm disturbances during this period may warrant therapy. Treatment depends on the cause and hemodynamic consequences of the arrhythmias. Premature depolarizations, both atrial and ventricular, usually do not require therapy other than maintenance of adequate perfusion and normal fluid and electrolyte balance. Ventricular arrhythmias may signify more serious myocardial dysfunction. QT prolonging agents must be avoided. Many of the vasoactive agents used to support myocardial function can increase myocardial irritability and risk of arrhythmias. Premature atrial or ventricular depolarizations are frequently observed and can be controlled by optimizing the dose of the vasoactive drugs. Bradycardia is frequently seen in TTM and typically requires no therapy. During PCAC, mechanical circulatory support (ECMO) may be considered if significant cardiorespiratory instability persists despite appropriate volume expansion and administration of inotropes, vasopressors, and, if indicated, antiarrhythmics.In a study de Mos N et al (CCM 2006) in a PICU population, the use of ECMO within 24 hours after ROSC was associated with reduced mortality. Case series have documented the role of ECMO88 and ventricular assist device support89,90 in children with refractory cardiogenic shock or acute fulminant myocarditis (Blume Ed et al., J Heart Lung Transplant 2016).
• What about oxygenation and ventilation strategies in our patient with PCAS
• Optimal oxygenation and ventilation of children after ROSC may be hampered by the pathology that precipitated the cardiac arrest (such as drowning with resultant post-pulmonary edema) and by the ensuing post-cardiac arrest pathophysiology. Further management challenges may be caused by aspiration and lung injury occurring during resuscitation efforts as well as ventilator-induced lung injury. Additionally, use of TTM alters the relationship between arterial oxygen saturation and arterial oxygen tension such that, for a given arterial oxygen saturation, the arterial oxygen tension (Pao2) is lower than that observed when the temperature is normal. Hypothermia also decreases the metabolic rate; thus, carbon dioxide production will be lower at any given minute ventilation.
• Post–cardiac arrest blood gas abnormalities are common in children, particularly in the first hours after ROSC as seen in our patient case. Published evidence has failed to demonstrate a consistent effect of post-cardiac arrest hyperoxia or hypoxemia on outcome. After ROSC, it is reasonable to aim for normal PaO2 (or the value appropriate for the child’s condition if the child has, for example, cyanotic heart disease) and to use the lowest possible fraction of inspired oxygen, weaning to maintain an oxygen saturation of 94% to 99% as a guideline. Throughout PCAC, hypoxemia must be avoided whenever possible, particularly during oxygen titration. The 2010 AHA PALS guidelines recommended prompt arterial blood gas analysis as soon as possible after ROSC and within 10 to 15 minutes of establishing initial mechanical ventilation to guide oxygen administration and titration and to support mechanical ventilation.
• Post–cardiac arrest derangements in PaCO2 are common. On the basis of available evidence, after ROSC, it is reasonable to target normocapnia (ie, normal for the child, or PaCO2 35–45 mm Hg) or a PaCO2 specific for the patient’s condition, limiting exposure to severe hypercapnia and hypocapnia. Lung protective strategies such as low TV, high PEEP should be used to minimize VILI.
• Can you comment on targeted temperature management?
• Post–cardiac arrest pyrexia (elevated core body temperature) is common, and persistent hyperthermia is associated with unfavorable neurological outcomes in children (Bambea MM PCCM 2010).During PCAC, fever (≥38°C) should be aggressively treated. To treat the child who remains comatose after OHCA, the 2015 AHA PALS guidelines update recommended that it is reasonable either to maintain continuous normothermia (TTM to 36°C–37.5°C) for 5 days or to maintain 2 days of continuous hypothermia (TTM to 32C°–34°C) followed by 3 days of continuous normothermia (TTM to 36°C–37.5°C).2 Because increased mortality was associated with temperatures <32°C, if TTM to 32°C to 34°C is used, meticulous care must be provided to prevent temperatures <32°C.

Post–cardiac arrest derangements in PaCO2 are common. On the basis of available evidence, after ROSC, it is reasonable to target normocapnia (ie, normal for the child, or Paco2 35–45 mm Hg) or a Paco2 specific for the patient’s condition, limiting exposure to severe hypercapnia and hypocapnia. Lung protective strategies such as low TV, high PEEP should be used to minimize VILI.

• What about treatment of seizures in PCAS and can you also comment on sedation , analgesia and the use of NMB in these patients ?
• Seizures occur in 10% to 50% of children who remain encephalopathic after achieving ROSC. (Abend NS et al Neurology 2009). Furthermore, about half of children with post-ROSC seizures experience exclusively non-convulsive (subclinical, EEG only) seizures, which cannot be identified by clinical observation alone. Seizures could not be predicted from any clinical or resuscitation variables. Seizures were associated with unfavorable gross neurological outcomes at discharge but not with higher mortality. Because seizures increase metabolic demand, can worsen metabolic dysfunction, and can increase intracranial pressure, they can contribute to secondary brain injury.
• For these reasons, many clinicians aim to treat seizures, although the approach is generally guided by the child’s overall medical condition and other prognostic indicators. Typical acute clinical or electrographic seizures are often initially treated with benzodiazepines, levetiracetam, or phenytoin. Myoclonic seizures such as those reported in our patient case may be refractory to treatment. (Ostendorf AP et al PCCM 2016) Providers must be alert for potential adverse effects of anticonvulsants such as cardiac arrhythmias, hypotension, and respiratory depression. In addition, sedation induced by anti-seizure drugs may complicate the neurological examination. Pain and discomfort needs to be controlled using opioids (morphine or fentanyl) and sedatives (dexmedetomidine or benzodiazepines). Neuromuscular blocking agents (eg, vecuronium or pancuronium) with analgesia or sedation (or both) may improve oxygenation and ventilation in case of patient-ventilator dyssynchrony or severely compromised pulmonary function. Providers are cautioned, however, that NMB agents can mask seizures and impede neurological examinations.If TTM is used, practitioners must be aware that the pharmacokinetics and pharmacodynamics of sedatives/hypnotics and neuromuscular blocking agents will be altered, resulting in prolonged time to both hepatic and renal clearance.
• What about endocrine dysfunction in PCAS patients ? Can you comment on glucose control and treatment of adrenal dysfunction?
• Both hypoglycemia and hyperglycemia have been associated with unfavorable outcomes in critically ill children and adults. During PCAC, clinicians should avoid and promptly treat hypoglycemia.Severe hyperglycemia can also be problematic because it can lead to uncontrolled osmotic diuresis, which can exacerbate post–cardiac arrest volume depletion and hemodynamic instability. Therefore, it is important to monitor serum glucose concentration, to treat significant hyperglycemia, and to monitor urine volume. There is currently insufficient published evidence to determine the optimal blood glucose concentration during PCAC that will maximize neurological outcome. Approximately 30% of critically ill children have relative adrenal insufficiency, but this has not been evaluated in children resuscitated from cardiac arrest. There is insufficient evidence to support the routine use of corticosteroids after cardiac arrest. Patients should be treated per recommendations for critically ill children.
• How do we manage Renal failure in these patients?
• In a recent retrospective study of 296 children during PCAC, 37% had AKI, 11.5% had severe AKI by Acute Kidney Injury Network criteria, and 6.4% required RRT within 48 hours of ROSC.(Neumayr TM et al. PCCM 2017). Risk factors for severe AKI after cardiac arrest included abnormal baseline creatinine, lack of a chronic lung condition, in-hospital arrest location, higher number of doses of epinephrine during arrest, and worse post–cardiac arrest acidosis.Throughout PCAC, it is important to monitor kidney function, including urine output and creatinine, because patients are at risk for developing AKI, and RRT may be indicated. Nephrotoxic medications and medications excreted by the kidneys should be used with caution, and dose adjustment may be needed. Serum concentrations of nephrotoxic medications should be closely monitored.
• Do we need antibiotics during post cardiac arrest care (PCAC) ? Can you also comment on management of inflammation and coagulation abnormalities?
• Infection is common after pediatric cardiac arrest. Most studies reporting the incidence of infection during PCAC enrolled children treated with THAPCA trial. The incidence varied from < 5 infections per 100 days in IHCA to 11.1 infections per 100 days for OHCA patients. The incidence of culture-proven infection did not differ between patients treated with TTM 32-34 and those treated with TTM to 36-37.5. During ECMO therapy for PCAS, the infection rate was ~ 10%. Monitoring for signs of infection is important during PCAC. The decision to obtain cultures and to initiate empirical antimicrobial coverage should follow local PICU protocols.
• Inflammatory pathways are activated as part of PCAS, including disturbances of the coagulation cascade. The effects of blocking or modulation of these pathways have been studied in adults and in animal models; we identified no studies to date involving infants or children. Extensive animal research into blocking or modifying inflammatory pathways has yielded promising results. However, to date, most attempts to translate this work to humans have been unsuccessful. Because inflammation can alter the coagulation cascade, providers should monitor for signs of bleeding or coagulopathies; this is particularly important for patients receiving ECMO support. At this time, there is insufficient evidence to support specific treatments to modulate inflammatory pathways during PCAC.

To summarize, Infection is common after pediatric cardiac arrest. Inflammatory pathways are activated as part of PCAS, including disturbances of the coagulation cascade. The effects of blocking or modulation of these pathways have been studied in adults and in animal models; we identified no studies to date involving infants or children.

• Can you comment on rehabilitation and recovery after cardiac arrest ?
• Children surviving cardiac arrest are at high risk for physical, cognitive, and emotional disabilities that can affect quality of life, family function, activities of daily living, school performance, and employment. There is little evidence on specific interventions during PCAC that will improve functional outcomes of children after cardiac arrest. Small observational studies of children after critical illness or injury suggest that children with anoxic injury have more severe disability and demonstrate less improvement compared with children with traumatic brain injury. There is insufficient evidence to support specific rehabilitation interventions or the optimal timing of initiation of such interventions. However, on the basis of the benefits of rehabilitation for patients with traumatic brain injury and stroke, it is reasonable for providers to consult rehabilitation experts within the first 72 hours after cardiac arrest to tailor a plan of rehabilitation interventions for survivors of cardiac arrest.
• As we look into the future, What about biomarkers for post arrest prognostication?
• Currently, there is insufficient evidence to support the use of serum biomarker concentrations alone to predict outcome after pediatric cardiac arrest. Although specific biomarkers have shown promise, they have yet to be validated in prospective pediatric studies after cardiac arrest. After cardiac arrest, elevations in lactate concentration may reflect not only severe post–cardiac arrest systemic hypoperfusion but also severe cerebral hypoperfusion. In several pediatric cardiac arrest studies, higher serum lactate concentrations in the first 12 hours after cardiac arrest were associated with increased mortality, and higher concentrations within 12 hours of ROSC were modestly predictive of unfavorable outcome (area under the curve: for IHCA, 0.76; for OHCA, 0.75). (Meert K et al. PCCM 2009; Topjian A PCCM 2013). Numerous other promising biomarkers of neurological injury, systemic inflammation, and genetic polymorphisms are currently under evaluation. An ongoing trial is investigating concentrations of NSE, S100B, glial fibrillary acid protein, and ubiquitin carboxy-terminal hydrolase L1 in the first 72 hours after pediatric OHCA and their association with 1-year neurological outcomes. (Prout AJ et al. Curr Opin Pediatr 2017).

To summarize, In several pediatric cardiac arrest studies, higher serum lactate concentrations in the first 12 hours after cardiac arrest were associated with increased mortality, and higher concentrations within 12 hours of ROSC were modestly predictive of unfavorable outcome

How do we prognosticate PCAS

• Providers must consider multiple variables when attempting to prognosticate outcomes during and after cardiac arrest. Providers must consider multiple variables when attempting to prognosticate outcomes during and after cardiac arrest. Although there are factors associated with better or worse outcomes, no single factor studied predicts outcome with sufficient accuracy to recommend termination or continuation of CPR or to enable prognostication after ROSC (de Caen AR et al. Circulation 2015).
• Several prearrest conditions and therapies have been independently associated with worse survival to discharge and unfavorable neurological outcomes after pediatric cardiac arrest: Worse outcomes from OHCA are associated with decreased age and some causes of arrest, including sudden infant death syndrome and blunt trauma. (Meert KL et al. PCCM 2016. Matos RI et al. Circulation 2013). Factors associated with lower survival after IHCA include older age; presence of preexisting conditions; interventions such as tracheal intubation, mechanical ventilation, and use of vasopressors at the time of arrest; and arrests occurring...