Introduction
More than 400,000 people have cardiac arrests in the United States each year, and approximately half of these occur in the hospital. Resuscitation attempts are unsuccessful in the majority of cardiac arrest victims, although as many as 30% to 45% of patients achieve the restoration of circulatory function (return of spontaneous circulation [ROSC]). Even among those who are successfully resuscitated, many still have a guarded prognosis secondary to neurologic injury that often results in crippling long-term disabilities for cardiac arrest survivors. Neurologic outcomes from cardiac arrest have been greatly improved over the past decade by the use of therapeutic hypothermia (increasingly referred to by the more broad term targeted temperature management [TTM]), and protocols for this therapy have been implemented in a wide variety of hospital environments. TTM after cardiac arrest involves three phases of care: induction, maintenance, and rewarming ( Figure 34-1 ), and each phase requires commitment of time and resources as well as potential exposure to specific adverse effects. Therefore risks and benefits of TTM use have to be considered carefully in light of the published evidence.
In this chapter, we will review the therapeutic options and approach to TTM induction and maintenance, discuss the current evidence regarding the use of TTM in patients who have experienced cardiac arrests, and evaluate whether such evidence extends to the perioperative or intraoperative setting.
Options/Therapies
Hypothermia Induction: Pharmacology
Before TTM is induced, patients are typically sedated and paralyzed. Given that postarrest patients are typically intubated, sedation is usually already in place, but neuromuscular blockade is additionally required to prevent shivering. It is generally held that shivering can occur during TTM induction until a patient’s core temperature falls below 33° C, when the brainstem shivering reflex appears to become attenuated. Shivering has a number of deleterious effects, including increased metabolic rate, increased risk of rhabdomyolysis and secondary acute tubular necrosis, and heat generation, which, in turn, can slow the desired rate of therapeutic cooling. A number of pharmacologic options exist for sedation and paralysis, and there are no convincing data that specific agents are preferred over others. An understanding of pharmacology during postarrest TTM is limited, but some data support that benzodiazepine metabolism may be altered with reduction of core body temperature, resulting in higher than expected drug levels in the cooled patient; this may result in more prolonged mechanical ventilation and a longer duration in the critical care environment.
Hypothermia Induction: Cooling
Several cooling options exist, including intravascular and external modalities. Intravascular methods include the use of cooling catheters and the administration of chilled fluids (typically normal saline or lactated Ringer’s solution). External options include the use of icepacks, cold air mattresses, and precooled or water-filled cooling pads that are applied to the skin. A combination of methods can be used; for example, chilled fluids are often used in TTM protocols as an initial “booster” to start the cooling process while either intravascular or external TTM devices are prepared and applied. Observational studies have confirmed that for the majority of postarrest patients (e.g., not including those with renal failure), the bolus administration of 1 to 2 L of chilled saline is safe and carries little risk of pulmonary edema, despite the concern for postarrest myocardial depression. This is a useful approach for accelerating the cooling process because laboratory trials have shown that more rapid attainment of the goal temperature (32° C to 34° C) is associated with improved survival rates and neurologic outcomes. However, chilled saline or ice packs alone have limited utility, except perhaps in the prehospital setting as a bridge to hospital care because temperature control during the required 12- to 24-hour maintenance is problematic. Prior work has shown that overcooling and undercooling are common unless a thermostat-driven device is used to maintain 32° C to 34° C. During induction and maintenance, continuous temperature measurement should be established via an esophageal, bladder, or rectal probe. Most commercially available TTM devices can receive temperature input from such probes and control the rate of cooling or warming accordingly. Some evidence suggests that each of these measurement sites lags behind the core body temperature; whether this makes a practical difference during patient care has not been established and is likely to be, at most, of modest importance.
Hypothermia Maintenance
Once the goal temperature is achieved, the maintenance phase of TTM begins. The landmark trials of postarrest TTM established that hypothermia maintenance should last a minimum of 12 hours; most protocols use 24 hours of maintenance at the goal temperature, following the duration of TTM from the larger of the two randomized trials. A number of key variables require careful monitoring in the maintenance phase. During TTM maintenance, electrolytes should be checked frequently (every 4 to 6 hours) because hypomagnesemia and hypokalemia are common phenomena related to “cold diuresis.” Hyperglycemia is also common. Bradycardia, often exhibited by a heart rate in the 40 to 50 beats-per-minute range, is frequent but well tolerated; specific treatments such as intravenous atropine or transcutaneous pacing are rarely required. Bleeding is also a concern with hypothermia induction because a reduced core body temperature can lead to both platelet dysfunction and abnormal kinetics of the serologic clotting cascade, seen clinically as prolongation of the partial thromboplastin time (PTT). This is a key issue with respect to postoperative or intraoperative cooling and will be discussed later in this chapter. Key potential adverse effects of TTM are shown in Table 34-1 .
| Physiologic Category | Adverse Effect |
|---|---|
| Neuromuscular | ShiveringSeizures |
| Electrolyte | HypokalemiaHypomagnesemia |
| Hematologic | Coagulopathy |
| Cardiac | Bradycardia |
| Infectious | Increased pneumonia risk |
After the goal temperature has been maintained for 12 to 24 hours, slow rewarming is initiated (0.2° C to 0.4° C per hour) until normothermia is once again attained (36.5° C to 37.5° C). This rewarming process can be accomplished with the use of the same commercial TTM devices used for cooling. After rewarming, hyperthermia (often termed rebound pyrexia ) often occurs, and careful attention is required to treat such temperature elevations because neurologic recovery may be impaired. After rewarming, paralytics and sedative medications can be weaned to assess patient responsiveness. The time during which a patient remains comatose after arrest and TTM treatment is highly variable, and the amount of time that a patient remains comatose does not necessarily correlate with the magnitude of neurologic recovery. Accordingly, the ability to accurately neuroprognosticate in this patient population remains elusive and is an area of intense research focus.
Evidence
Evidence from Randomized Trials
The modern use of TTM after cardiac arrest was established in two landmark randomized trials published in 2002. The Hypothermia after Cardiac Arrest (HACA) trial, a European study coordinated in Vienna, evaluated external cooling after out-of-hospital ventricular fibrillation cardiac arrest, with randomization of patients to either a cooled group (32° C to 34° C for 24 hours) or a normothermic group, in which temperature was not actively managed. This investigation, which included more than 250 subjects, demonstrated significant survival and neurologic outcome benefits from hypothermia treatment: 59% of cooled patients were alive at 6 months compared with only 45% of patients in the normothermic group. The second crucial randomized trial, based in Melbourne, Australia, also evaluated hypothermia treatment in patients after out-of-hospital ventricular fibrillation cardiac arrest. In this investigation, 49% of the patients in the cooled cohort survived to hospital discharge compared with only 26% of patients in the normothermia group.
These two randomized trials were not without key limitations. Although patients were randomly allocated in both studies, blinding was not possible, and therefore other confounders of care in the emergency department or subsequent intensive care unit stay may have theoretically affected outcomes. Another weakness ascribed to these investigations was that mild temperature elevations were common in the normothermic groups; therefore it is possible that aggressive avoidance of fever was the most important therapeutic benefit of the TTM protocols. Despite these concerns, the measured survival benefits were impressive and had a larger effect size than any other cardiac arrest treatment subjected to randomized trials previously. For instance, no pharmacologic treatment in the Advanced Cardiovascular Life Support regimen (e.g., epinephrine, atropine, or amiodarone) has been shown to contribute to long-term survival with a magnitude comparable to that seen with hypothermia.
Additional Evidence beyond Key Randomized Trials
The two landmark randomized trials in support of postarrest hypothermia only enrolled patients who had had out-of-hospital cardiac arrest, which arguably has a number of pathophysiologic differences from in-hospital cardiac arrest (and therefore perioperative arrest). No randomized trials to date have evaluated TTM after in-hospital, intraoperative or perioperative arrest. It is also important to note that the two randomized trials only enrolled patients with initial arrest rhythms of ventricular fibrillation, one of the most common underlying rhythms of out-of-hospital cardiac arrest. Pulseless electrical activity (PEA) and/or asystole, rhythms that are more frequently seen during in-hospital cardiac arrest, have also not been the subject of TTM randomized trials. However, a number of single-center observational cohort investigations with historic control subjects have evaluated cooling after in-hospital arrest and from a variety of initial rhythms. These single-center studies have generally shown a benefit to TTM, although effect sizes have varied and sample sizes are generally small. For example, Oddo and colleagues reported outcomes from a single hospital postarrest cohort and found that, among the patient subset with asystole or PEA, two of 12 patients receiving cooling had good neurologic outcomes, whereas zero of 11 patients who were not cooled had favorable outcomes. A meta-analysis of such studies confirmed that TTM may be generally beneficial after cardiac arrest, regardless of initial rhythm or location of arrest, although the meta-analysis had a lower quality of data. It is important to note that some investigations have provided data that conflict with this conclusion. For example, in a single-center study of 40 patients experiencing in-hospital cardiac arrest, Rittenberger and colleagues found no significant benefit from cooling patients after cardiac arrest. Only one of 13 cooled patients achieved a good neurologic outcome, whereas two of 25 noncooled patients achieved good outcomes. Of note, no patients with an initial rhythm of ventricular fibrillation or ventricular tachycardia survived in either group of this study.
The rationale for applying TTM to perioperative or in-hospital arrest patients is largely based on extrapolation from the two randomized trials of cooling after out-of-hospital arrest, already described. However, there are important differences between out-of-hospital and in-hospital arrests that might make this extrapolation problematic. For example, the majority of out-of-hospital arrests have cardiac causes of arrest; that is, lethal arrhythmias occur secondary to myocardial ischemia or at the site of prior myocardial infarction. Related to this point, the most common initial arrest rhythms for these out-of-hospital events are either ventricular fibrillation or ventricular tachycardia. In contrast, perioperative and in-hospital cardiac arrest has a broader set of causes, including medication reactions, fluid or electrolyte derangements, or primary respiratory failure. In these conditions, PEA is the most frequent initial arrest rhythm, and defibrillation is less commonly required. At the present time, there are no randomized trial data evaluating TTM for perioperative or in-hospital cardiac arrest or for PEA arrest in any location.
The notion that TTM can be safely used in concert with surgical interventions has largely been inferred from indirect evidence. For example, several case reports have demonstrated successful use of TTM after postoperative arrest with good outcomes. Another case report demonstrated the successful use of TTM after surgery complicated not by arrest but rather marked hypotension and cerebral ischemia, a situation frequently faced by surgeons and anesthesiologists. While not representing a perioperative arrest but rather perioperative cooling, another case report described the successful use of TTM in a pregnant woman who was delivered of her infant via caesarean section with resulting good outcomes for both mother and infant.
A number of laboratory studies have used hypothermia techniques in the setting of massive pulmonary embolism and cardiopulmonary bypass or traumatic exsanguination arrest with surgical repair. Direct clinical evidence at the cohort level supporting TTM after perioperative arrest is, however, lacking. Case series of postoperative arrest with cooling would be a welcome addition to the growing literature on TTM. Randomized trials of perioperative arrest cooling would be difficult for a number of reasons, including the challenge of patient accrual for such an uncommon event. Illustrating this point, studies from two large tertiary care referral centers reporting on intraoperative cardiac arrests revealed a combined total of 246 events from 736,578 surgical procedures. This equates to 3.3 arrests per 10,000 cases. Resuscitation was successful in approximately 40% of cases across both studies; thus less than two patients per 10,000 procedures would be available for postarrest care investigation. Neither of these studies examined functional or neurologic outcomes after the arrest.
Pathophysiologic Rationale for Targeted Temperature Management
After global ischemia and subsequent reperfusion, a number of pathophysiologic mechanisms become active and lead to clinical morbidity and mortality ( Figure 34-2 ). This set of injuries is often termed postreperfusion syndrome or postarrest syndrome and has clinical features that resemble septic shock. Patients often exhibit marked lowering of systemic vascular resistance, hypotension, and organ dysfunction (e.g., acute tubular necrosis, acute lung injury, and hepatic necrosis). Cerebral edema can also occur and may serve as a precursor to cerebral herniation, which serves as a common cause of postresuscitation death. A hallmark of postarrest syndrome, both in laboratory models and in the clinical environment, is the rapidity of onset, often within several hours of resuscitation.
Related posts:
Is a Preoperative Screening Clinic Cost-Effective?
Which Patient Should Have a Preoperative Cardiac Evaluation (Stress Test)?
What Is the Optimal Airway Management in Patients Undergoing Gastrointestinal Endoscopy?
What Works in a Patient with Acute Respiratory Distress Syndrome?
Are There Special Techniques in Obese Patients?
Is Propofol Safe If Given by Nonanesthesia Providers?
Full access? Get Clinical Tree
