Protocolized Post-Cardiac Arrest Care with Targeted Temperature Management

Abstract

Improvements in teamwork and resuscitation science have considerably increased the success rate of cardiopulmonary resuscitation. Cerebral injury, myocardial dysfunction, systemic ischemia and reperfusion response, and precipitating pathology after the return of spontaneous circulation (ROSC) constitute post-cardiac arrest syndrome. Because the entire body is involved in cardiac arrest and the early post-arrest period, protocolized post-arrest care consisting of cardiovascular optimization, ventilation and oxygenation adjustment, coronary revascularization, targeted temperature management (TTM), and control of seizures and blood sugar would benefit survival and neurological outcomes. Emergent coronary angiography is suggested for cardiac arrest survivors suspected of having ST-elevation myocardial infarction, however the superiority of culprit or complete revascularization in patients with multivessel coronary lesions remains undetermined. High-quality TTM should be considered for comatose patients who are successfully resuscitated from cardiac arrest, however the optimal target temperature may depend on the severity of their condition. The optimal timing for making prognostication should be no earlier than 72 h after rewarming in TTM patients, and 72 h following ROSC in non-TTM patients. To predict neurological recovery correctly may need the use of several prognostic tools together, including clinical neurological examinations, brain images, neurological studies and biomarkers.

POST-CARDIAC ARREST SYNDROME

Sudden cardiac arrest is a major challenge for clinical physicians. The incidence of sudden cardiac arrest is approximately 300,000/year in the U.S.¹ and 20,000/year in Taiwan.² The success rate of cardiopulmonary resuscitation has increased because of considerable efforts devoted to improving the quality of resuscitation and teamwork. However, the prognosis of cardiac arrest survivors is still unsatisfactory, with the majority of patients not surviving to discharge or being discharged with poor neurological recovery. Approximately 10% of successfully resuscitated patients with out-of-hospital cardiac arrest (OHCA) and less than 20% of those with in-hospital cardiac arrest (IHCA) survive with good neurological function. A large proportion of patients resuscitated from cardiac arrest die of anoxic brain swelling and cardiovascular failure following the resuscitation process in the early post-arrest period.³ For patients who survive the initial few days following return of spontaneous circulation (ROSC), subsequent immune and inflammation reactions, multiorgan failure, and nosocomial infections account for further in-hospital mortality. Post-cardiac arrest syndrome describes the cerebral injuries, myocardial damage, global ischemia/reperfusion injuries and precipitating pathology causing cardiac arrest in the early post-arrest period.^4,5 To alleviate post-cardiac arrest syndrome, bundle care consisting of cardiovascular optimization, ventilation and oxygenation adjustment, coronary revascularization, targeted temperature management (TTM), control of seizures and blood sugar is suggested by the American Heart Association (AHA) and European Resuscitation Council (ERC).^6-8 In addition, the accurate prediction of neurological recovery in patients successfully resuscitated from cardiac arrest is crucial for clinical staff and the patient’s family due to considerations of medical and financial resources (Figure 1).

Figure 1.

Protocolized post-cardiac arrest care. After return of spontaneous circulation, instant management of ventilation, oxygenation and circulation should be implemented, targeting normoxia and normocarbia as well as avoiding hypotension. Post-cardiac arrest evaluations should include electrocardiography and echocardiography. If ST-segment elevation is suspected, emergent coronary angiography and possible PCI are indicated. Other causes of cardiac arrest should be promptly identified and managed. Intensive post-cardiac arrest care also includes targeted temperature management, seizure detection and treatment, and blood sugar control. ICU, intensive care unit; MAP, mean arterial pressure; PaCO₂, arterial carbon dioxide tension; PCI, percutaneous coronary intervention; ROCS, return of spontaneous circulation; SaO₂, arterial oxyhemoglobin saturation; STEMI, ST-elevation myocardial infarction; TV, tidal volume.

CARDIOVASCULAR MONITORING AND OPTIMAL TARGETS

Hemodynamic instability is common after cardiac arrest and can have various causes, such as myocardial dysfunction, vasoplegia and arrhythmias, etc. Although refractory hypotension is associated with higher mortality rates and poor neurological outcomes, whether a specific target of blood pressure improves outcomes remains unclear.^9-15 On the basis of experience managing critically ill patients, clinicians have adopted signs of adequate tissue perfusion, including adequate urine output and mean arterial pressure (MAP), as well as normal or decreasing lactate levels. Guidelines worldwide suggest a urine output of > 0.5 mL/kg/h and MAP ≥ 65 mmHg as being adequate. Two randomized clinical trials (RCTs) compared higher (80-100 mmHg) with lower (65-75 mmHg) blood pressure targets, but did not observe any differences in biomarker levels for brain and myocardial injury between the two groups;^16,17 however, the relationship between blood pressure and patient outcomes has not been demonstrated. Another post-hoc analysis revealed that in post-cardiac arrest patients with shock after acute myocardial infarction, targeting a higher MAP (80-100 mmHg) was associated with smaller myocardial injury but not neurological outcomes.¹⁸ Although the optimal target of blood pressure remains unclear, avoiding hypoperfusion (MAP < 65 mmHg, urine output < 0.5 mL/kg/h, increasing lactic acid levels) is mandatory.

Post-cardiac arrest myocardial dysfunction is common and occurs in approximately 60% of cardiac arrest survivors.^19,20 Echocardiography helps in quantifying this disorder, and it can reveal underlying cardiac pathology and guide clinical management; thus, it should be performed early in the process.^21-23 In hemodynamically unstable post-cardiac arrest patients, serial echocardiography can be helpful, and an invasive hemodynamic monitoring system can provide instant and additional hemodynamic information and assist in clinical judgment, such as cardiac output and systemic vascular resistance. However, whether lower cardiac output is associated with poor outcomes is not clear. Post-cardiac arrest myocardial dysfunction can contribute to lower cardiac output. A sub-study of the TTM trial demonstrated that a low cardiac index may not be associated with outcomes if lactate clearance is maintained.²⁴ In addition, a prospective study revealed that a lower cardiac index was associated with in-hospital mortality but not with neurological outcomes.¹⁰ Whether to use pulmonary artery catherization or other minimally invasive modalities such as Pulse index Contour Cardiac Output (PiCCO) or FloTrac should be considered on a case-by-case basis and requires further investigation. Mixed results regarding the precision of PiCCO during hypothermia have been observed.^25-27

OXYGENATION AND VENTILATOR SETTING

Theoretically, in cardiac arrest patients with ROSC, further hypoxemia should be avoided to prevent ongoing ischemic injury. However, hyperoxia may increase oxidative stress and produce more free radicals, which may cause tissue and neuronal damage. Clinical investigations into how hyperoxia may influence outcomes has yielded mixed results.^28-32 Wang et al. reviewed 14 observational studies and found higher in-hospital mortality in patients with hyperoxia.³³ Another prospective cohort study demonstrated that early exposure to hyperoxia was associated with poor neurological function;²⁹ however, a pilot RCT revealed no difference in biomarkers for neuronal damage between higher and lower oxygen partial pressure.³¹ In general, normoxia should be targeted in post-cardiac arrest patients.

Arterial carbon dioxide tension (PaCO₂) may affect cerebral blood flow. Hypocarbia may cause cerebral vasoconstriction and worsen cerebral ischemia; however, hypercarbia increases cerebral blood flow, cerebral blood volume and intracerebral pressure.^34,35 Several observational studies have demonstrated that arterial hypocarbia was consistently associated with poor neurological outcomes.^33,36-43 Two pilot studies compared normocarbia and mild hypercarbia for neurological injury markers, and found that mild hypercarbia was associated with either similar or lower levels of these markers.^31,44 Both studies demonstrated improved cerebral oxygenation, however the impact of hypercarbia on neurological outcomes remains uncertain. A prospective multicenter cohort study indicated that PaCO₂ had a U-shaped association with neurological outcomes, with mild-to-moderate hypercapnia having the highest probability of achieving good neurological outcomes.⁴⁵ Normocarbia with 35-45 mmHg of PaCO₂ should be considered during post-cardiac arrest care, however whether permissive hypercarbia improves outcomes requires further investigations.

Some clinicians have suggested using a tidal volume of 6-8 mL/kg, adapted from lung protective ventilation in critically ill patients. However, limited evidence has been produced to support its use for cardiac arrest survivors,^46,47 with one retrospective study observing superior neurological performances among cardiac arrest survivors with a lower tidal volume.⁴⁸

EMERGENT CORONARY ANGIOGRAPHY IN PATIENTS SUSPECTED OF HAVING ST-ELEVATION MYOCARDIAL INFARCTION

Coronary artery disease is one of the main causes of cardiac arrest.^49,50 Emergent percutaneous coronary intervention (PCI) of the culprit vessel has been used to treat this disease, and this strategy has been supported by several observational studies which demonstrated its association with survival and better neurological outcomes. A systematic review also demonstrated a higher survival rate with emergent cardiac catheterization compared with delayed or no cardiac catheterization in post-cardiac arrest patients with ST-elevation myocardial infarction (STEMI).⁵¹ As a result, several guidelines have suggested the use of 12-lead electrocardiography (ECG) following ROSC to detect cases that require an immediate assessment in a cardiac catheterization laboratory. However, the appropriate timing of ECG is not clear. Performing ECG immediately after ROSC may only reflect the relative ischemic status of the heart and profound metabolic derangement without indicating notable coronary occlusion. A retrospective multicenter cohort study found a higher false-positive rate of 18.5% when ECG was performed within 7 minutes of ROSC.⁵² Therefore, emergent cardiac catheterization and PCI should be used in patients with STEMI after ROSC, and STEMI may be diagnosed with repeated ECG and echocardiography.

Whether to perform emergent coronary angiography (CAG) in patients after ROSC without ST-segment elevation is under debate. Several observational studies and a meta-analysis have reported better neurological outcomes with early rather than with delayed CAG in post-cardiac arrest patients without ST-segment elevation.^53-57 However, an RCT conducted in the Netherlands and also an international RCT conducted in Germany and Denmark found no increase in survival rate.^58,59 A recent systematic review and meta-analysis also failed to demonstrate the benefits of routine CAG in post-cardiac arrest patients without ST-segment elevation.⁶⁰ In hemodynamically and electrically stable post-cardiac arrest patients without ST-segment elevation, a delayed coronary intervention allows for timely post-cardiac arrest care in the intensive care unit, including temperature management, and it also allows clinicians to select patients who are most likely to benefit from CAG. In patients without ST-segment elevation after ROSC, the appropriate timing of CAG should be determined. In addition, electrical and hemodynamic instability should be taken into consideration regarding the timing of CAG.

In patients with cardiogenic arrest, more than one coronary lesion is common,^59,61,62 however it is unclear whether treating the culprit vessel or complete revascularization leads to better outcomes. Whilst several observational studies have demonstrated that patients with STEMI who receive multivessel PCI have an increased number of major cardiovascular events and mortality,^63-66 four RCTs and a systematic review and meta-analysis favored complete revascularization.^67-71 Evidence is lacking for cardiac arrest survivors with multivessel coronary artery disease. Two observational studies revealed improved outcomes with multivessel PCI in cardiac arrest survivors with STEMI;^61,72 however further investigations are required.

TARGETED TEMPERATURE MANAGEMENT

Whereas fever is harmful to post-cardiac arrest patients,^73-76 mild induced hypothermia (32-34 °C) has been proposed to decrease cerebral metabolic demand, cerebral edema, and intracranial pressure.^77,78 In 2002, induced hypothermia was demonstrated to improve neurological outcomes in post-cardiac arrest patients with a shockable rhythm.^79,80 Since then, therapeutic hypothermia has been implemented in patients experiencing OHCA with a shockable rhythm who do not regain consciousness after ROSC. The implementation in patients with a non-shockable rhythm and those with IHCA has also been extensively investigated. A systematic review and meta-analysis revealed consistent survival and neuroprotective benefits of the extended use of therapeutic hypothermia, including in patients with a non-shockable rhythm, lenient downtime, and unwitnessed arrest.⁸¹ One RCT also revealed a better neurological outcome for both OHCA and IHCA patients with non-shockable rhythms undergoing TTM at a target temperature of 33 °C.⁸² TTM may benefit patients regardless of the initial arrest rhythm and location of arrest.

The optimal target temperature has not yet been established. The TTM trial published in 2003 found no difference in survival and neuroprotective benefits between body temperatures of 33 and 36 °C,⁸³ and since then, many clinicians have changed their TTM protocols to a target of 36 °C under the assumption that a similar effect would be observed with fewer adverse events. However, the trial was criticized because the majority of patients had experienced a witnessed arrest, shockable rhythm and bystander cardiopulmonary resuscitation (CPR), which may have led to better neurological outcomes in both groups regardless of the target temperature. Moreover, the longer induction time for the 33 °C target temperature in the trial may have attenuated the neuroprotective effect of TTM. Several observational studies have demonstrated lower adherence to the TTM protocol and worse neurological outcomes after switching the target temperature to 36 °C.^84-88 The TTM-2 trial, which compared patients undergoing targeted hypothermia at 33 °C with those undergoing normothermia at lower than 37.8 °C, demonstrated no increase in survival rate by 6 months.⁸⁹ However, more than 70% of the patients in both study groups had experienced a witnessed collapse, shockable rhythm and bystander CPR, which may have led to better outcomes that could not be differentiated by different target temperatures. Nishikimi et al. suggested that patients with post-cardiac arrest syndrome of moderate severity may benefit more from TTM at lower target temperatures (33-34 °C).⁹⁰ Individualized target temperatures should thus be considered for patients with different severities in the future. Moreover, few studies have used high-quality TTM in terms of timely initiation, core temperature measurement, the speed at which the target temperature is reached, maintenance of the target temperature, and the length and steadiness of the rewarming phase.⁹¹ The current consensus is to achieve the target temperature as soon as possible, maintain a steady and stable target temperature for at least 24 hours, and then rewarm at a rate of 0.25-0.5 °C/h. Whether cooling for a duration of up to 48 hours would improve outcomes has not yet been determined. An RCT that compared 48- and 24-h TTM at 33 °C indicated that the 6-month survival rate did not increase and that neurological outcomes did not improve, although a trend toward longer duration was observed.⁹² Further research might be warranted.

NEUROPROGNOSTICATION

Before the introduction of TTM, only 1 in every 100 patients with cardiac arrest was discharged with a good neurological recovery. After the introduction of TTM and the protocolization of post-arrest care, the proportion of cardiac arrest survivors with favorable neurological outcomes increased substantially. Therefore, determining the correct prognosis for comatose cardiac arrest survivors has become increasingly crucial for primary care medical staff and the patients’ families. Finding a balance between "avoiding early withdrawal" and "preventing hopeless waiting" depends on the predictive ability and timing of assessments for neuroprognostication.⁹³ The optimal timing for making prognostication suggested by the AHA and ERC is no earlier than 72 hours after rewarming completion in patients receiving TTM and 72 hours following ROSC in those without TTM.^6-8 The influence of sedatives and neuromuscular blockers should be eliminated as far as possible. Multimodal prediction including clinical neurological examinations, brain images, neurological studies and biomarkers has been proposed to increase accuracy. The motor component of the Glasgow Coma Scale, pupil light reflex and corneal reflex are suitable methods for screening patients with potentially poor recovery due to their convenience and simplicity.^94,95 However, clinical neurological examinations are easily affected by sedatives and neuromuscular blockers. The gray-white matter ratio (GWR) of brain computed tomography^96-98 and diffusion-weighted imaging (DWI) of brain magnetic resonance imaging⁹⁹ have been reported to be associated with the prognosis, and when the cortex is injured and becomes swollen, the GWR and DWI values decrease. Electroencephalogram (EEG) can be used to detect post-arrest seizures and predict neurological recovery. Status epilepticus, burst suppression and seizures are associated with poor neurological outcomes.^93,100,101 The prevalence of seizures, non-convulsive status epilepticus, and other epileptiform activity after cardiac arrest has been reported to be 12% to 22% in adults and 47% in children.¹⁰² Post-arrest seizures, as the result of post-arrest brain injury, may also cause secondary brain damage,¹⁰³ and can be classified into generalized convulsive seizures and non-convulsive seizures, which can easily be masked by neuromuscular blockage.^102,104 Therefore, prompt frequent or continuous EEG monitoring is recommended.^8,104 The same anticonvulsant regimens for the treatment of status epilepticus caused by other etiologies may be considered for the management of post-arrest seizures.¹⁰⁴ Available evidence does not support the prophylactic administration of anticonvulsant drugs.^103,104 In addition to EEG, the absence of N20 cortical wave on somatosensory evoked potentials has also been used to identify unfavorable neurological outcomes.¹⁰⁵ Serial changes in serum biomarkers, such as neuron-specific enolase, S100B and Tau protein, may help predict neurological recovery.¹⁰⁶

CONCLUSION

The protocolization of post-cardiac arrest care has advanced considerably. Multimodal prognostication for neurological recovery has been proposed and incorporated into post-cardiac arrest care. The next steps are to increase the accuracy of predictive tools and apply psychological and physical rehabilitation after discharge.

DECLERATION OF CONFLICTS OF INTEREST

All the authors declare that there are no conflicts of interest.