Article

Managing the Complications of Mild Therapeutic Hypothermia in the Cardiac Arrest Patient

Journal of Intensive Care Medicine 1-11 ª The Author(s) 2013 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0885066613516416 jic.sagepub.com

Adam M. Noyes, MD1, and Justin B. Lundbye, MD2

Abstract Mild therapeutic hypothermia (MTH) is used to lower the core body temperature of cardiac arrest (CA) patients to 32 C from 34 C to provide improved survival and neurologic outcomes after resuscitation from in-hospital or out-of-hospital CA. Despite the improved benefits of MTH, there are potentially unforeseen complications associated during management. Although the adverse effects are transient, the clinician should be aware of the associated complications when managing the patient receiving MTH. We aim to provide the medical community comprehensive information related to the potential complications of survivors of CA receiving MTH, as it is imperative for the clinician to understand the physiologic changes that take place in the patient receiving MTH and how to prepare for them and manage them if they do occur. We hope to provide information of how to manage these potential complications through both a review of the current literature and a reflection of our own experience. Keywords therapeutic hypothermia, cardiac arrest, complications, cardiology

Introduction The implementation of mild to moderate therapeutic hypothermia (MTH; 32 C-34 C) has shown improvements in survival and neurologic outcomes that have been demonstrated in patients resuscitated from out-of-hospital cardiac arrest (OHCA) secondary to ventricular fibrillation.1-4 Similar benefits have also been observed in other types of presentations including pulseless electrical activity, asystole, and inpatient populations.5-10 Patients who have cardiac arrest (CA) are usually admitted to a critical care environment for their care. As such, there are numerous immediate and delayed risks of complications that may follow. These complications include increased risk of infection, aspiration pneumonitis and pneumonia and decreased drug metabolism from acute renal and liver failure.11-13 It is likely that both CA survivors by itself and with the addition of therapeutic hypothermia increase patient’s risk of complications from the hypothermia.11,14,15 This includes electrolyte disturbance, infection risk, acid–base disorders, change in drug metabolism, arrhythmia disorders, and shivering. In a review article by Holzer,16 an overall rate of adverse events related to MTH was reported to be 74% (223 events in 300 patients). The author further reported on an observational study involving 986 patients, finding that the most common adverse events associated with MTH were pneumonia (41%), hyperglycemia (37%), cardiac arrhythmias (33%), seizures (24%), and electrolyte disturbances (hypophosphatemia, 19%; hypomagnesemia, 18%; and hypokalemia, 18%).

The aim of this article is to provide evidence-based reporting on the current literature supporting the complication risks and alteration in mechanical, biochemical, and physiological functions associated with therapeutic hypothermia. A summary of potential complications is detailed in Table 1.

Bleeding Complications The relationship between coagulopathy and hypothermia has been well demonstrated to affect hemostasis; the more hypothermic the patient is, the more likely hemorrhage will occur.17-19 It has been shown that patients receiving MTH tend to have more bleeding complications and require blood transfusion more frequently than the control group.20 However, there were no life-threatening bleeding complications in either group, and patients receiving MTH had lower mortality after 1

Department of Medicine, University of Connecticut Medical School, Farmington, CT, USA 2 Division of Cardiology, the Hospital of Central Connecticut, Chief of Cardiology, New Britain, CT, USA

Received July 8, 2013, and in revised form September 26, 2013. Accepted for publication September 27, 2013. Corresponding Author: Adam M Noyes, Department of Medicine, Room L2104 MC 1235, University of Connecticut School of Medicine, 263 Farmington Ave, Farmington, CT, USA 06103. Email: [email protected]

Downloaded from jic.sagepub.com at UNIV OF GEORGIA LIBRARIES on June 12, 2015

2

Journal of Intensive Care Medicine

Table 1. Potential Complications and Management of MTH. Possible Adverse Reaction

Risk

Bleeding

↔

Infection

"

Acid/base disorder

↔

Potassium disorders (hypokalemia followed by hyperkalemia)

"

Magnesium disorders (hypomagnesemia) Phosphate disorder (hypophosphatemia) Glycemic abnormalities

"

Pharmacokinetic changes Seizure activity

" " Unknown "

Cardiac arrhythmias

"

Cardiovascular disorder

#

Shivering

"

Monitor

Management

 Serum Hb/Hct every 4 to 6 hours  Serum PT/PTT every 4 to 6 hours

 Control active bleeding  Blood products  Consider aborting MTH if severe or irreversible  Microbial cultures in blood, sputum and urine  Consider prophylactic use of Unasyn with MTH initiation  Leukocytosis and bandemia  Tachycardia  Remove intravenous catheters within 72 hours  "lactic acid  Appropriate antimicrobial coverage in relation to cultures  ABG for pH every 6 hours  Adequate MAP  Lactic acid  Adequate IVF  Glucose levels  Serum potassium every 6 hours  Hypokalemia is transient and will reverse with rewarming  Arrhythmias  Potassium values should be treated while the patient is being cooled with 60 mEq, to be maintained >3.5 mEq/L  Supplementation should be discontinued with rewarming  Serum magnesium  Replete as needed to maintain levels 2.0  Arrhythmias  Serum phosphate  Replete as needed  Serum glucose increases with cooling, and decreases with rewarming  Observe efficacy of medications used  The majority of seizures are nonconvulsive  Obtain CT/MRI brain to evaluated anoxic brain injury

 Insulin infusion to maintain blood glucose around 150  Involve you pharmacist to assist with dosing  Antiepileptic medications  EEG monitoring is warranted  Treat with antiepileptics Consider cooling to 32 C  Bradycardia commonly seen  Heart rate 40 is not a concern in the absence of Prolonged QTc hemodynamic instability  Place an arterial line for blood pressure and  MAP goal of 60 mm Hg vital sign monitoring  Vassopressors may be needed to reach a goal MAP Echocardiogram  Place an esophageal, rectal, or bladder  Paralytics to prevent shivering or opioids such as temperature probe for continuous core meperidine temperature monitoring  Surface cournterwarming  Signs of shivering

Abbreviations: Hb, hemoglobin; Hct, hematocrit; PT, prothrombin Time; PTT, partial thromboplastin time; MTH, mild-to-moderate therapeutic hypothermia; ABG, ARTERIAL blood gas; MAP, mean arterial pressure; IVF, intravenous fluids; QTc, corrected QT interval; CT, computed tomography; MRI, magnetic resonance imaging; EEG, electroencephalography.

6 months and an improved neurologic outcome. In contrast, a small case series of post-CA patients with trauma having significant risk of bleeding revealed no significant coagulopathy or bleeding complications with the use of MTH.21 The hemostatic changes that occur in patients with hypothermia have been shown to significantly affect the clotting parameters with as little as a 0.5 C decrease in body temperature.17 Hypothermia can lead to disorder within the coagulation cascade as well as inducing a reversible platelet dysfunction.17,22,23 In fact, measured alterations in platelet function and enzyme function seen with the use of thrombelastography have been demonstrated in patients with trauma having temperatures of 33.0 C to 33.9 C when compared to nonhypothermic patients (37.0 C).17 Watts et al found that 34 C was the critical point at which enzyme activity slowed

and alteration in platelet activity occurred. Additionally, Michelson et al showed that hypothermia inhibits thrombinand thromboxane-induced platelet aggregation, increased platelet surface expression of P-selectin, and decreased platelet surface expression of the von Willebrand factor receptor glycoprotein Ib–IX complex.24 Fortunately, rewarming has been shown to facilitate optimum function of platelets.25 In addition to hypothermia-disrupting coagulation, the acidosis frequently seen in patients receiving MTH may further potentiate this disorder. Although there are no known human studies to our knowledge, the swine model revealed that hypothermia and acidosis impair thrombin generation. A temperature of 32 C demonstrated inhibition was predominantly involved in the complex of factor (F) VIIa and FIII pathway. Acidosis of pH 7.1 led to inhibition of the activation of FV, VII, IX, X and formation of intrinsic

Downloaded from jic.sagepub.com at UNIV OF GEORGIA LIBRARIES on June 12, 2015

Noyes and Lundbye

3

FXase and prothrombinase complex.26 Therefore, hypothermia and associated acidemia results in impaired clot formation that leads to disruption of the coagulation pathway and may potentiate the risk of bleeding in the humans. The same authors were also able to show that hypothermia inhibits fibrinogen synthesis, while acidosis accelerates fibrinogen degradation, resulting in a potential fibrinogen deficit.27,28 Furthermore, prothrombin time (PT) and activated partial thromboplastin time (aPTT) were not significantly altered under the combined effects of hypothermia.29 Initiation of MTH commonly begins with the rapid infusion of 4 C cold crystalloids,30 which may lead to acute hemodilution and directly alter coagulation parameters.31 One study sought to explain how the induction of MTH by bolus infusion of cold crystalloids affects the coagulation system reported no change in PT; however, a prolonged aPTT (2.7-fold; P < .01) occurred after 1 hour.22 Furthermore, due to hemodilution, the authors report an average drop in hematocrit of 16% after 1-hour infusion. After infusion cooling, continuation of MTH may be achieved through a variety of methods including external cooling. With external cooling of the skin, a 3- to 4-fold increase in bleeding time was observed, with a reduction in local skin temperature from 35 C to 22 C in non-MTH individuals.25 This was associated with a substantial reduction in thromboxane B2, the stable metabolite of thromboxane A2 that constricts the blood vessels and aggregates the platelets at the bleeding site. Finally, many patients who may require MTH may have also experienced concomitant acute coronary syndromes (ACSs)32,33 that additionally requires early revascularization with either thrombolysis or percutaneous coronary intervention (PCI) to further improve outcomes. Instrumentation such as PCI while patients receive MTH may offer a potential for bleeding, thus increasing morbidity and mortality.26 Moreover, many of these patients receive antithrombotic and antiplatelet medications that may potentiate bleeding rates. In a registry study involving 986 patients receiving MTH, bleeding requiring transfusion took place in 4% of patients, with transfusion significantly increased if PCI was performed (2.8% vs 6.2%, respectively). However, early PCI was also observed to be predictors of good outcome.34 A study by Schefold et al illustrated no difference in bleeding complications, transfusion requirements, or the number of transfusions between patients who had an OHCA due to an acute myocardial infarction (AMI) who were treated with hypothermia and that of a control. They found the benefit of MTH outweighs the risk of bleeding complications and that the optimal therapy for patients with AMI and OHCA should include both thrombolysis or PCI and therapeutic hypothermia.35 Furthermore, initiation of MTH did not result in longer door-toballoon times and did not delay the onset of primary PCI.20 Importantly, patients in 2 randomized control trial did not have significantly increased bleeding associated with therapeutic hypothermia.1,2 Based on these studies, it is our recommendation that patients receiving MTH require careful observation for

bleeding complications, especially when treated with antiplatelet and anticoagulant drugs, and patients with active bleeding be excluded from receiving MTH. Despite laboratory data suggesting a change in the coagulation pathways, MTH does not substantially increase the clinical risk of bleeding, and the small increased risk of bleeds seen in retrospective studies, as well as in laboratory models, does not outweigh the benefits of receiving hypothermic therapy. Patients can continue to be cooled during PCI and may use aspirin, antiplatelet compounds, and thrombolytic agents necessary for a primary cardiac syndrome.

Infectious Implications Previous studies have demonstrated that survivors of CA have an increased risk of infectious complications during hospitalization.4 Infectious complications were found in 67% of patients receiving MTH in a study reported in Critical Care Medicine.36 As CA survivors undergoing MTH are admitted to the intensive care unit (ICU), they may carry an even greater risk of infection as their immune response is attenuated from slowed inflammatory processes and impaired wound healing. This is secondary to reduced defense mechanisms such as chemotactic and phagocytic activity and has been reported in patients with temperatures of 33 C.37 Additionally, hypothermia was shown to reduce tumor necrosis factor a concentrations and its messenger RNA (mRNA) expression levels along with delayed peak productions of interleukin b and interleukin 6, and mRNA expressions, as well as suppression of the secretion of proinflammatory cytokines, impaired leukocyte migration, and phagocytosis, all of which increases the risk of infection.37 With implementation of MTH in the ICU patient, suppression or masking of fever creates difficulty for the detection of infections. A large meta-analysis showed a higher incidence of sepsis in patients subjected to hypothermia,38 and the Hypothermia After Cardiac Arrest Study Group2 found nearly double the incidence of sepsis (7% vs 13%) in a group randomized to hypothermia compared to normothermic patients, although this difference was not statistically significant. The lung was the most common infection site (85%), with bloodstream infections as the second most frequent source (8%). Although gram-negative bacteria were the most frequently isolated infectious microbe (64%), Staphylococcus aureus was found as the main causative agent of infection. The investigators found that the duration of MTH was a significant factor in the infected patients (odds ratio [OR], 1.01; 95% confidence interval [CI],1-1.02). A large retrospective cohort of 642 patients successfully resuscitated from OHCA, who underwent MTH, showed an associated increased risk of early-onset pneumonia after OHCA (OR, 1.90; 95% CI, 1.28-2.80), with MTH as the single independent risk factor for this complication (OR, 1.90; 95% CI, 1.28-2.80; P ¼ .001). The most frequent pathogens grown from lung secretions were S aureus, followed by Haemophilus influenza, Streptococcus pneumoniae, and other Enterobacteriaceae. Anaerobes were not found.39 In

Downloaded from jic.sagepub.com at UNIV OF GEORGIA LIBRARIES on June 12, 2015

4

Journal of Intensive Care Medicine

addition, 28% of those with early-onset pneumonia resulted in an increased length of mechanical ventilation and ICU stay, but this did not influence favorable neurologic outcome or mortality.39 It is altogether possible that many patients who survive a CA may arrive with chemical or acid-induced pneumonitis, since aspiration of the gastric fluid is almost constant after CA.40 Preventive measures such as culture of bodily fluid, use of prophylactic antibiotics, and removal of unnecessary catheters may decrease the incidence of infection. Therefore, infectious complications should not sway physicians from implementing MTH. In fact, in a large, multicenter, international registry of patients receiving MTH, pneumonia and sepsis were not associated with increased mortality, and there was an inverse relationship between infection and mortality.41 Additionally, MTH has been shown to significantly shorten the time of mechanical ventilation in survivors after OHCA and also reduce the ICU stay.3 In our center,42 we prophylactically place our patients on ampicillin and sulbactam (Unasyn) during the initiation of MTH and remove our intravascular cooling catheters within 72 hours to reduce the incidence of infection.

Metabolic and Electrolytes Changes Hypothermia alters the normal physiologic state. Specifically, MTH has been directly associated with an increase in sympathetic drive, resulting in peripheral vasoconstriction, endorgan hypoperfusion, and metabolic acidosis due to a shift from aerobic to anaerobic metabolism.43 During the cooling period, patients receiving MTH were found to have an average pH decrease from 7.37 to 7.31.44 Hypothermia produces a metabolic respite by decreasing both oxygen consumption and the production of carbon dioxide, thereby decreasing metabolic demands (+8% per  C drop in core temperature).45 This may modify the interpretation of carbon dioxide and bicarbonate values when rewarming occurs in the laboratory, leading to potentially incorrect estimates in patient management, in particular ventilator settings, which may ultimately lead to cerebral vasoconstriction due to hyperventilation modes.45 Although hypocapnia-induced vasoconstriction minimizes cerebral swelling, overshooting due to incorrect laboratory values may increase the risk of inducing cerebral ischemia. Hypothermia has been shown to frequently increase serum lactate levels (5-7 mmol/L) and increase the synthesis of glycerol, free fatty acids, ketonic acids and lactate, which in turn leads to mild metabolic acidosis.46 These particular changes are considered normal metabolic changes of hypothermia and generally do not require intervention, although metabolic acidosis in the setting of hypothermia has been reported to exacerbate bleeding disorders.47 In addition, MTH commonly induces hyperglycemia due to insulin sensitivity because of decreased insulin secretion by the pancreas45 not only resulting in increased insulin requirements but also may further increase the risk of infection. Observation of sustained hyperglycemia was common, whereas

hypoglycemia was uncommon in one recent study. A reported sustained hyperglycemia (OR 2.5, 95% CI 1.8-3.4) and hypoglycemia (OR 2.3, 95% CI 1.1-4.9) were both associated with increased mortality.41 In an attempt to establish recommendations, 1 study found that strict glucose (72-108 mg/ dL) versus moderate glucose (108-144 mg/dL) resulted with no difference in mortality in patients receiving MTH, although episodes of hypoglycemia occurred in 18% of the strict glucose control group versus 2% of the moderate glucose control group (P ¼ .008).48 Electrolyte disturbances, most notably hypokalemia, are commonly observed during the active cooling phase of MTH. Analysis of 94 patients with OHCA who underwent MTH were found to have an average decrease in potassium of 0.7 mmol 10 hours after initiation of cooling. This can lead to arrhythmias such as polymorphic ventricular tachycardia (as discussed in the section on cardiovascular effect and hypothermia). Experimental data suggest that hypothermia leads to transient potassium shifts into cells,49 resulting from changes in the sodium potassium exchanger activity,50,51 and drive through an increased adrenergic state.43 However, it should also be noted that MTH is also correlated with an increase in electrolyte excretion in the urine during the cooling period and has been described as hypothermia-induced diuresis.44 This study found that urine production increased significantly during cooling, from 219 + 70 to 485 + 209 mL/h and was associated with increased concentrations of urinary potassium, magnesium, and phosphate. This finding may have also been influenced by medication-induced tubular dysfunction.44 In addition to hypokalemia occurring during MTH, hyperkalemia may also be seen during the rewarming phase due to the release of potassium from the cell and appears to correlate with the speed of rewarming, and hence it is suggested that rewarming is done slowly and tightly controlled. Catecholamine levels may increase during the stress response of the active cooling phase and may further potentiate the decrease in serum potassium by shifting potassium in to the cell.52 By giving the kidneys time to excrete excess potassium, hyperkalemia should not develop if rewarming is slow and if renal function is not significantly impaired.44 Additionally, MTH may require magnesium management. Both magnesium and potassium are critical for stabilizing membrane potential and decreasing cell excitability. Magnesium deficiency will not only exacerbate potassium wasting but also aggravate the adverse effects of hypokalemia on target tissues. Furthermore, magnesium deficiency has been shown to be an inciting event for coronary spasm,53 which can promote myocardial infarction and arrhythmia. Thus, frequent measurement and repletion of deficit serum magnesium should be supported. When treating patient receiving MTH, we recommend close monitoring of metabolic and electrolyte changes with appropriate intervention. Importantly, understanding the mechanism behind the changes in serum electrolytes, particularly potassium, is essential in preventing potential complications. Replacing potassium to high normal values during the cooling phase should be avoided.

Downloaded from jic.sagepub.com at UNIV OF GEORGIA LIBRARIES on June 12, 2015

Noyes and Lundbye

5

Pharmacokinetic Changes In addition to impaired drug metabolism due to acute renal and liver injury that follows CA, the reduction in body temperature also has the potential to alter physiological processes and thus several classes of drugs may become affected by hypothermia. The lowered body temperature leads to changes in the drug function through modification in drug absorption, metabolism, and excretion, resulting in possible underperformance of the drug or reduced plasma clearance with subsequent increased drug effects and potential toxicity. The central underlying mechanism in many of the cases of altered metabolism is due to reduced cytochrome P (CYP) 450 enzyme activity that is thought to be secondary to diminished hepatic blood flow and reduced gastrointestinal absorption.54 A systematic review found that P450 activity was reduced between 7% and 22% per  C below 37 C.55 The number of medications potentially affected are extensive, thus we will separate this section into drug classes. Central Nervous System. Sedatives of varying drug classes are administered during management of patient receiving MTH, however, understanding the changes in drug clearance can allow the medical provider the time it may take to allow adequate resolution of sedative effect on the patient. This is essential if neurologic recovery is determined after rewarming. The ICU sedative midazolam, a benzodiazepine metabolized by the CYP3A4 and CYP3A5 enzyme system, was studied in 8 patients receiving MTH against 7 controls. The hypothermia group showed a 5-fold increase in midazolam concentrations and a >100-fold decrease in systemic clearance of midazolam when body temperature was