22 Incidence and Risk Factors of Post-Rewarming “Rebound Hyperthermia” in Cardiac Arrest Patients Undergoing Therapeutic Hypothermia

Abstract

Study ObjectivesGuidelines recommend initiating therapeutic hypothermia as soon as possible after return of spontaneous circulation in patients with witnessed cardiac arrest. However, the incidence of complications, such as overcooling and rebound hyperthermia may increase without careful monitoring of temperature. Post-rewarming “rebound hyperthermia”, defined as a temperature of 38.5°C or greater, may worsen survival and outcome. The purpose of this study was to determine the incidence and risk factors associated with post-rewarming rebound hyperthermia.MethodsThis retrospective, cohort study was performed using a database of visits to the ED chest pain unit of a tertiary referral center. During a 4-year study period, all patients undergoing therapeutic hypothermia (to a central target temperature of 33°C, using endovascular cooling) following cardiac arrest were eligible for the study. Patients less than 18 years old, trauma cases, pregnancy and cardiogenic shock cases were excluded. Electronic dispatch, patient care reports, and hospital records were reviewed by 3 independent reviewers. Our main outcome was the incidence of post-rewarming rebound hyperthermia within 48 hours after cooling withdrawal. Analysis of risk factors was performed as follows: 24 potentially relevant risk factors for rebound hyperthermia were assessed by univariate analysis with chi-square test for categorical variables and simple logistic regression for continuous variables. Collected data also included the rate of medical complications, death or severe disability (modified Rankin Scale 4-6).ResultsDuring the study period, 93 consecutive adult patients were treated with therapeutic hypothermia following out-of-hospital cardiac arrest (70% male, age 56.8 + 17 years). Rebound hyperthermia was documented in 23 patients (24.7%; 95% CI 16.0-33.5) within 48 hours after cooling withdrawal. Post-rewarming rebound hyperthermia was associated with an increased risk of death (70% vs 41%, p< 0.001) as well as severe disability (93% vs 68%, P<0.001). Infectious complications were observed in 8 patients with rebound hyperthermia (34.8%; 95% CI 15.3-54.2), but no patient developed severe sepsis or septic shock. The biological changes that occurred during rewarming manifested principally as hypokalemia (< 3.5 mmol/l) in 74% of patients with rebound hyperthermia (p=0.06). No statistical correlation was found between predictor variables and the incidence of rebound hyperthermia in this population.ConclusionPost-rewarming rebound hyperthermia was observed in 25% of patients within 48 hours after cooling withdrawal and was associated with significant disability and mortality. No correlation was found between predictor variables and the incidence of rebound hyperthermia in our population. Study ObjectivesGuidelines recommend initiating therapeutic hypothermia as soon as possible after return of spontaneous circulation in patients with witnessed cardiac arrest. However, the incidence of complications, such as overcooling and rebound hyperthermia may increase without careful monitoring of temperature. Post-rewarming “rebound hyperthermia”, defined as a temperature of 38.5°C or greater, may worsen survival and outcome. The purpose of this study was to determine the incidence and risk factors associated with post-rewarming rebound hyperthermia. Guidelines recommend initiating therapeutic hypothermia as soon as possible after return of spontaneous circulation in patients with witnessed cardiac arrest. However, the incidence of complications, such as overcooling and rebound hyperthermia may increase without careful monitoring of temperature. Post-rewarming “rebound hyperthermia”, defined as a temperature of 38.5°C or greater, may worsen survival and outcome. The purpose of this study was to determine the incidence and risk factors associated with post-rewarming rebound hyperthermia. MethodsThis retrospective, cohort study was performed using a database of visits to the ED chest pain unit of a tertiary referral center. During a 4-year study period, all patients undergoing therapeutic hypothermia (to a central target temperature of 33°C, using endovascular cooling) following cardiac arrest were eligible for the study. Patients less than 18 years old, trauma cases, pregnancy and cardiogenic shock cases were excluded. Electronic dispatch, patient care reports, and hospital records were reviewed by 3 independent reviewers. Our main outcome was the incidence of post-rewarming rebound hyperthermia within 48 hours after cooling withdrawal. Analysis of risk factors was performed as follows: 24 potentially relevant risk factors for rebound hyperthermia were assessed by univariate analysis with chi-square test for categorical variables and simple logistic regression for continuous variables. Collected data also included the rate of medical complications, death or severe disability (modified Rankin Scale 4-6). This retrospective, cohort study was performed using a database of visits to the ED chest pain unit of a tertiary referral center. During a 4-year study period, all patients undergoing therapeutic hypothermia (to a central target temperature of 33°C, using endovascular cooling) following cardiac arrest were eligible for the study. Patients less than 18 years old, trauma cases, pregnancy and cardiogenic shock cases were excluded. Electronic dispatch, patient care reports, and hospital records were reviewed by 3 independent reviewers. Our main outcome was the incidence of post-rewarming rebound hyperthermia within 48 hours after cooling withdrawal. Analysis of risk factors was performed as follows: 24 potentially relevant risk factors for rebound hyperthermia were assessed by univariate analysis with chi-square test for categorical variables and simple logistic regression for continuous variables. Collected data also included the rate of medical complications, death or severe disability (modified Rankin Scale 4-6). ResultsDuring the study period, 93 consecutive adult patients were treated with therapeutic hypothermia following out-of-hospital cardiac arrest (70% male, age 56.8 + 17 years). Rebound hyperthermia was documented in 23 patients (24.7%; 95% CI 16.0-33.5) within 48 hours after cooling withdrawal. Post-rewarming rebound hyperthermia was associated with an increased risk of death (70% vs 41%, p< 0.001) as well as severe disability (93% vs 68%, P<0.001). Infectious complications were observed in 8 patients with rebound hyperthermia (34.8%; 95% CI 15.3-54.2), but no patient developed severe sepsis or septic shock. The biological changes that occurred during rewarming manifested principally as hypokalemia (< 3.5 mmol/l) in 74% of patients with rebound hyperthermia (p=0.06). No statistical correlation was found between predictor variables and the incidence of rebound hyperthermia in this population. During the study period, 93 consecutive adult patients were treated with therapeutic hypothermia following out-of-hospital cardiac arrest (70% male, age 56.8 + 17 years). Rebound hyperthermia was documented in 23 patients (24.7%; 95% CI 16.0-33.5) within 48 hours after cooling withdrawal. Post-rewarming rebound hyperthermia was associated with an increased risk of death (70% vs 41%, p< 0.001) as well as severe disability (93% vs 68%, P<0.001). Infectious complications were observed in 8 patients with rebound hyperthermia (34.8%; 95% CI 15.3-54.2), but no patient developed severe sepsis or septic shock. The biological changes that occurred during rewarming manifested principally as hypokalemia (< 3.5 mmol/l) in 74% of patients with rebound hyperthermia (p=0.06). No statistical correlation was found between predictor variables and the incidence of rebound hyperthermia in this population. ConclusionPost-rewarming rebound hyperthermia was observed in 25% of patients within 48 hours after cooling withdrawal and was associated with significant disability and mortality. No correlation was found between predictor variables and the incidence of rebound hyperthermia in our population. Post-rewarming rebound hyperthermia was observed in 25% of patients within 48 hours after cooling withdrawal and was associated with significant disability and mortality. No correlation was found between predictor variables and the incidence of rebound hyperthermia in our population.