Therapeutic Hypothermia After Resuscitation From a Non-Shockable Rhythm Improves Outcomes in a Regionalized System of Cardiac Arrest Care

Transcription

1 DOI /s z ORIGINAL ARTICLE Therapeutic Hypothermia After Resuscitation From a Non-Shockable Rhythm Improves Outcomes in a Regionalized System of Cardiac Arrest Care Gene Sung 1 Nichole Bosson 2,3,4 Amy H. Kaji 2,3 Mark Eckstein 5 David Shavelle 6 William J. French 3,7 Joseph L. Thomas 3,7 William Koenig 4 James T. Niemann 2,3 Springer Science+Business Media New York 2015 Abstract Objective Therapeutic hypothermia (TH) improves neurologic outcome in patients resuscitated from ventricular fibrillation. The purpose of this study was to evaluate TH effects on neurologic outcome in patients resuscitated from a non-shockable out-of-hospital cardiac arrest rhythm. Design and Setting This is a retrospective cohort study of data reported to a registry in an emergency medical system in a large metropolitan region. Patients achieving field return of spontaneous circulation are transported to designated hospitals with TH protocols. Patients Patients with an initial non-shockable rhythm were identified. Patients were excluded if awake in the Emergency Department or if TH was withheld due to & Gene Sung gsung@usc.edu Department of Neurology, Keck School of Medicine of the University of Southern California, 1520 San Pablo St, Ste 3000, Los Angeles, CA 90033, USA Department of Emergency Medicine, Harbor-UCLA Medical Center, Los Angeles Biomedical Research Institute at Harbor-UCLA, Torrance, CA, USA David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Los Angeles County Emergency Medical Services Agency, Los Angeles, CA, USA Department of Emergency Medicine, Keck School of Medicine of the University of Southern California, Los Angeles, CA, USA Department of Cardiology, Keck School of Medicine of the University of Southern California, Los Angeles, CA, USA Department of Cardiology, Harbor-UCLA Medical Center, Los Angeles Biomedical Research Institute at Harbor-UCLA, Torrance, CA, USA preexisting coma or death prior to initiation. The decision to initiate TH was determined by the treating physician. Measurements The primary outcome was survival with good neurologic outcome defined by a cerebral performance category of 1 or 2. Main Results Of the 2772 patients treated for cardiac arrest during the study period, there were 1713 patients resuscitated from cardiac arrest with an initial non-shockable rhythm and 1432 patients met inclusion criteria. The median age was 69 years [IQR 59 82]; 802 (56 %) male. TH was induced in 596 (42 %) patients. Survival with good neurologic outcome was 14 % in the group receiving TH, compared with 5 % in those not treated with TH (risk difference = 8 %, 95 % CI 5 12 %). The adjusted OR for a CPC 1 or 2 with TH was 2.9 (95 % CI ). Conclusion Analyzing the data collected from the registry of the standard practice in a large metropolitan region, TH is associated with improved neurologic outcome in patients resuscitated from initial non-shockable rhythms in a regionalized system for post-resuscitation care. Keywords Therapeutic hypothermia Cardiac arrest Resuscitation Introduction Therapeutic hypothermia (TH) is a proven intervention that improves neurologic outcomes after successful resuscitation from out-of-hospital cardiac arrest (OHCA) with initial shockable rhythm. [1, 2] The International Liaison Committee on Resuscitation supports the use of TH in patients after OHCA with an initial rhythm of ventricular fibrillation, and state that, while TH may be beneficial for other rhythms, further study is required in non-shockable

2 rhythms. [3] The proposed mechanism by which TH improves outcomes includes reducing cerebral reperfusion injury by decreasing oxygen demand, glutamate release, and production of free radicals. [1, 3] Cerebral reperfusion injury is applicable to all patients suffering cardiac arrest. However, given the lower survival rates and higher potential for comorbidities in patients with non-shockable rhythms, the degree to which TH may be beneficial for these patients remains unclear. Small observational studies that include patients with non-shockable rhythms demonstrated safety and suggest that there may be a survival benefit in all rhythms. [4 8] The benefit was not consistently seen when non-shockable rhythms were isolated, causing some to question benefit in these cases, though these results were possibly due to small sample size. [9 12] While patients resuscitated after non-shockable cardiac arrest rhythms may benefit from the neuroprotective properties of cooling, hypothermia has known risks including dysrhythmias, coagulopathy, and infection. In addition, it is resource intensive. Finally, there remains the possibility that TH may improve survival in this group of patients without improving neurologic outcome, resulting in more patients surviving severely dependent or moribund. This study reports the results of the largest series of nonshockable rhythm patients in a registry of one of the largest metropolitan regions in the world. It utilizes the regionalized care system for OHCA in Los Angeles County and its standard practices, with a centralized database of patient outcomes, to assess neurologic outcome in patients treated with TH after successful resuscitation from OHCA with an initial rhythm of pulseless electrical activity (PEA) or asystole compared to those patients not treated with TH. Materials and Methods Los Angeles County is a large metropolis with a population of over 13 million. Emergency medical services (EMS) are provided by 32 municipal fire departments, one law enforcement agency, and 25 private ambulance companies with over 3500 licensed paramedics throughout the county. Patients who call 911 are transported to one of 72 emergency departments in the county. The LA County EMS Agency provides oversight of providers operating within the county, establishes protocols and procedures, and designates specialty care centers. This is a retrospective study of data from a registry maintained by the LA County EMS Agency. The study was reviewed and approved with waiver of informed consent by the institutional review board. The system of regionalized cardiac care in LA County, with 33 designated cardiac arrest receiving centers, has been described previously. [13] Countywide protocols mandate transport of all OOHCA patients of presumed cardiac etiology with ROSC in the field to a cardiac arrest center with an institutionally approved hypothermia protocol. Participating hospitals were encouraged to institute TH (target temperature C) within 6 h of ROSC and to maintain it for a minimum of 20 h. The final decision to initiate or withhold TH was guided by particular institutional policy and at the discretion of the treating physician, and TH was not initiated in the field. Six of the 33 cardiac arrest centers have policies that specifically limit TH to patients presenting with a shockable rhythm. Per LA County EMS policy, resuscitation on scene to achieve ROSC prior to transport is encouraged. For patients meeting criteria, termination of resuscitation in the field is supported by official policy since Termination of resuscitation is based on medical futility determined by paramedics in consultation with the base hospital physician and agreement of immediate family on scene. Since April 2011, all cardiac arrest centers have reported their in-hospital mortality and neurologic outcome to a single registry maintained by the LA County EMS Agency. The database was queried from April 2011 through August 2013, representing all available data at the time of analysis. Patients 18 years or older resuscitated from OOHCA with an initial non-shockable rhythm transported to a designated cardiac arrest center were included. Non-shockable rhythm refers to the first rhythm noted by prehospital personnel and includes PEA and asystole, as well as patients analyzed by an automated external defibrillator that advised no shock. Patients with traumatic cardiac arrest and patients <18 years of age were excluded. Additionally, patients who would not benefit from TH were excluded from the analysis in order to mitigate the selection bias toward or against TH given the observational design of the study. This included the following: patients awake and alert in the emergency department, patients with prior coma, and patients who died prior to consideration for TH. (Fig. 1) Finally, patients with termination of resuscitation in the field were not transported by protocol and, therefore, were not eligible for inclusion in the database. Study variables included age, gender, race/ethnicity, initial cardiac rhythm, arrest location, witness, bystander CPR, vasopressor support, induction of hypothermia, reason hypothermia was withheld (if applicable), and whether the patient received cardiac catheterization with or without PCI. The primary outcome of the study was survival to hospital discharge with good neurologic outcome, as defined by a cerebral performance category (CPC) score at hospital discharge of 1 or 2. A CPC of 1 corresponds to a return to normal or mildly impaired cerebral function and independence with activities of daily living. A CPC of 2 corresponds to moderate cerebral disability but sufficient function to remain independent with activities of daily living. The CPC scores documented by physician

3 calculated to evaluate each model. Mean CPC score outcomes among hospitals were compared with ANOVA. Results Fig. 1 Study profile assessment at the time of discharge were abstracted from the medical record. Individual centers are responsible for reporting their data. Staff members charged with data entry abstract the data points from the patient s medical record, including prehospital care records. Greater than 90 % of staff responsible for the data extraction and entry are registered nurses in the departments of emergency medicine, cardiology, and quality improvement. Completeness and accuracy of the entered data is continually reviewed by the EMS Agency with verification performed during annual site visits. Bi-annual system-wide meetings are held for data review. All data were entered into Microsoft Excel (Microsoft Corporation, Redmond WA) and transferred to SAS 9.3 (SAS Institute, Cary, NC) for analysis. We report neurologically intact survival as proportions with exact binomial confidence intervals. Adjusted odds ratios (OR) and their p values were calculated using logistic regression and the Chi-square test. Variables in the regression (age, witness, bystander CPR, rhythm, catheterization, PCI, and vasopressor support) were selected based on prior knowledge of their contribution to cardiac arrest outcomes and entered simultaneously into the model. The generalized estimating equation was used to adjust for clustering by hospital. Additionally, a propensity score-adjusted model was performed to confirm the results of the multivariable regression, because of the large differences between the treatment groups. [14] A propensity score for receiving TH for each individual was calculated based on the following observed covariates known to affect outcomes in cardiac arrest: age, witness, bystander CPR, initial rhythm, need for vasopressor support, and treatment hospital. This score was then used in a regression analysis to evaluate the association of TH with neurologic outcome, adjusting for the likelihood of receiving TH. The Hosmer Lemeshow goodness-of-fit statistic was Of the 2772 patients treated for cardiac arrest during the study period, 893 (32 %) had an initial shockable rhythm and 166 (6 %) had an undocumented initial rhythm leaving 1713 consecutive patients resuscitated from cardiac arrest with an initial non-shockable rhythm. For these patients, the median age was 71 years [IQR 59 82] and 975 (57 %) were male. Overall, four-hundred and five (28 %) survived to hospital discharge and 176 (12 %) had documentation of good neurologic outcome. There were 33 survivors without documented CPC score and were not included in the analysis. Of the 1713 patients presenting with an initial nonshockable rhythm, 86 were awake and 195 had prior coma or died, leaving 1432 that could have potentially benefitted from TH. The overall characteristics of the study population are given in Table 1. TH was induced in 596 (42 %) patients. Patient characteristics in the TH group and reference group are given in Table 2. Of note, 159 (27 %) of patients treated with TH and 169 (20 %) of patients who did not receive TH, converted to a shockable rhythm and received at least 1 shock during their resuscitation. For those in the TH group, the target temperature of 33 C was achieved 74 % of the time and ranged from 32 to 34 C. Table 1 Patient characteristics (n = 1432) Characteristics Total N (median) % (IQR*) Age Gender Female Male Race/ethnicity Black Asian Hispanic White Pacific Islander/Hawaiian 8 1 Other/unknown 63 4 Witnessed arrest Bystander CPR Pulseless electrical activity Therapeutic hypothermia Vasopressors Catheterization performed PCI performed 46 3 * Inter-quartile range

4 Table 2 Patient characteristics by treatment group (n = 1423) Characteristics TH (n = 596) No TH (n = 827) N % N % Age (median/iqr ) Gender Female Male Race/ethnicity Black Asian Hispanic White Pacific Islander/Hawaiian Other/unknown Witnessed arrest Bystander CPR Pulseless electrical activity Defibrillated at any point Initial GCS after ROSC (median/iqr ) Vasopressors Catheterization performed PCI performed Inter-quartile range Cooling was maintained for a median time of 23 h (IQR 19 24). For patients that did not receive TH, the reason was undocumented in one-third and less than half had documentation of a known contraindication to cooling, Table 3. Survival with good neurologic outcome occurred in 14 % in the group receiving TH, compared with 5 % in the group not treated with TH (risk difference = 8 %, 95 % CI 5 12 %). The adjusted OR for a CPC 1 or 2 with TH was 2.9 (95 %CI ) compared to patients not receiving TH, adjusted for age, witnessed arrest, bystander CPR, arrest rhythm (PEA or asystole), vasopressor support, treatment in the cath lab, whether PCI was performed, and clustering by hospital (Table 4). Frequency of adverse events from TH is in Table 5. The propensity score analysis, which included a variable representing the probability of the patient receiving TH, given the age of the patient, whether the arrest was witnessed and bystander CPR was performed, the initial presenting rhythm, need for vasopressor support, and the hospital at which the patient was treated, yielded identical results to the regression analysis. The overall survival increased from 16 % without TH to 30 % with TH (risk difference = 14 %, 95 %CI %), as well as an increase in the proportion of patients who survived with CPC 3 or 4 13 % with TH versus 9 % without TH (risk difference = 4 %, 95 % CI 1 7 %). Table 3 Primary reason given for no therapeutic hypothermia (n = 827) Reason given Frequency Percent Bleeding complications 60 5 Hypotension and/or dysrhythmia 45 4 Temperature less than 35 C 19 2 Sepsis 32 3 Drug-induced coma 7 1 Chronic renal disease 31 3 Non-shockable rhythm 48 4 Withdrawal of care Other 66 6 Missing Discussion In the LA County regional system for OHCA care, TH improved overall survival and survival with good neurologic outcome in patients resuscitated from cardiac arrest with an initial non-shockable rhythm. To our knowledge, this is the largest cohort to date evaluating use of TH for patients with non-shockable cardiac arrest rhythms and the first to report the results from the general practice of TH in

5 Table 4 Adjusted odds ratios for survival with good neurologic outcome Adjusted odds ratio for good outcome (95 % CI)* Therapeutic hypothermia Witnessed arrest Rhythm (PEA) Bystander CPR Age (per year) Cath lab PCI Vasopressor *Adjusted odds ratio generated by the simultaneous entry of covariates in the logistic regression model Hosmer Lemeshow goodness-of-fit statistics p 0.2, Akaike Information Criterion = 686 Table 5 Reported adverse events from TH Adverse event Shivering 53 Electrolyte abnormality 50 Dysrhythmia 24 Bleeding/coagulopathy 19 Infection 18 Aspiration 5 Thrombotic event 2 Ileus 1 Decubitus 1 Other 34 Frequency the care of cardiac arrest patients in a large metropolitan region. Prior literature has been conflicting on the effectiveness of TH for these patients, possibly due to the limitations of small sample size. [15 17] In our cohort, TH was used in 35 % of patients. Lundbye et al. evaluated 100 patients with non-shockable rhythms before and after implementation of TH protocol and found TH improved both survival to hospital discharge and neurologic outcome. [18] In contrast, two other studies found trends for benefit but did not reach statistical significance. Vaahersalo and colleagues prospectively evaluated the effect of TH on adult OHCA patients treated in intensive care units in Finland, where national guidelines do not recommend TH for patients with non-shockable cardiac arrest rhythms. While the authors found a 3.4 % reduction in poor neurologic outcome for patients with initial non-shockable rhythm, this difference did not reach statistical significance. The outcome in this study differs from ours in that neurologic outcome was assessed at one year rather than at hospital discharge. There were a significant number of deaths from any cause in both groups during the one-year follow-up, which may or may not be related to the initial cardiac event. Similarly, a study by Dumas et al. found a hazard ratio for death of 0.89 (95 % CI ) in patients resuscitated from PEA or asystole who received TH compared to those not treated. Over a period of nine years, only 261 patients with nonshockable rhythm were enrolled of whom 68 were cooled. Several studies have demonstrated a benefit to TH across all ROSC patients irrespective of initial rhythm. [4 8, 19] Because of conflicting outcomes, some investigators have questioned the use of TH in general. [20, 21] A recent randomized control study of adults resuscitated from cardiac arrest regardless of initial rhythm calls into question whether it is the induced hypothermia or the careful temperature modulation alone that may prove beneficial. [22] In this study, patients were randomized to a target temperature of 36 or 33 C and no difference in death or neurologic outcome at 180 days was noted. Since the study was powered to detect a 20 % difference in the groups for the primary outcome of mortality, it does not eliminate the possibility of a smaller but substantial benefit to TH. While further study may be needed to determine the optimum target temperature, hyperthermia is known to be detrimental after cardiac arrest. [23] Post-cardiac arrest syndrome, including brain injury and reperfusion response, is common to all victims resuscitated from cardiac arrest. [24] It follows that interventions demonstrated effective for one rhythm are likely generalizable to all cardiac arrest patients. Our results support the use of TH in patients resuscitated from cardiac arrest with initial non-shockable rhythms, but have several limitations. Our study is retrospective and the validity of the results depends on the accuracy of the data entered in the registry by the cardiac arrest centers. Although all designated hospitals were required to meet criteria set by the LA County EMS Agency and were subject to ongoing quality improvement, there was variety in the institutional protocols and devices used for TH. However, this makes the results more generalizable to other systems with varying methods and target parameters for cooling. The reason TH was withheld was missing in one-third of patients. We were unable to adjust for downtime (time from cardiac arrest to ROSC), which is known to affect outcomes [25], as this was missing a large number of patients and is likely to be inaccurate when reported [26 28]. In addition, there are other uncontrolled factors that may contribute to patient outcomes that we did not measure in this study. In particular, selection bias is possible due to the observational design of the study, despite exclusion of patients in whom TH was withheld for futility and adjustment for measured confounders. Although the

6 propensity score analysis helps account for large differences between groups, it too is limited by inclusion of only the known covariates. Use of additional interventions varied between centers, however, we did adjust for hospital in order to partially account for clustering by center. Misclassification bias is possible if paramedics misinterpreted the initial presenting rhythm. Finally, the ability to obtain information regarding long-term follow-up was not possible, as the database is limited to prehospital and inpatient data through hospital discharge only. Conclusion Patients resuscitated from non-shockable cardiac arrest rhythms may benefit from TH. In our cohort, TH was associated with improved neurologic function at hospital discharge and our results support current international guidelines. Acknowledgments The authors would like to thank all the cardiac arrest center participants and the Los Angeles County EMS Agency staff, in particular Paula Rashi, Richard Tadeo, and Deidre Gorospe, who contributed to the ROSC registry and whose dedicated work provided the necessary data for this analysis. References 1. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med. 2002;346: The Hypothermia After Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med. 2002;346: Nolan JP, Morley PT, Vanden Hoek TL, et al. Therapeutic hypothermia after cardiac arrest: an advisory statement by the advanced life support task force of the International Liaison Committee on Resuscitation. Circulation. 2003;108: Hachimi-Idrissi S, Corne L, Ebinger G, Michotte Y, Huyghens L. Mild hypothermia induced by a helmet device: a clinical feasibility study. Resuscitation. 2001;51: Nielsen N, Hovdenes J, Nilsson F, et al. Outcome, timing and adverse events in therapeutic hypothermia after out-of-hospital cardiac arrest. Acta Anaesthesiol Scand. 2009;53: Bernard SA, Jones BM, Horne MK. Clinical trial of induced hypothermia in comatose survivors of out-of-hospital cardiac arrest. Ann Emerg Med. 1997;30: Busch M, Soreide E, Lossius HM, Lexow K, Dickstein K. Rapid implementation of therapeutic hypothermia in comatose out-ofhospital cardiac arrest survivors. Acta Anaesthesiol Scand. 2006;50: Storm C, Steffen I, Schefold JC, et al. Mild therapeutic hypothermia shortens intensive care unit stay of survivors after out-of-hospital cardiac arrest compared to historical controls. Crit Care. 2008;12:R Don CW, Longstreth WT Jr, Maynard C, et al. Active surface cooling protocol to induce mild therapeutic hypothermia after out-of-hospital cardiac arrest: a retrospective before-and-after comparison in a single hospital. Crit Care Med. 2009;37: Oddo M, Schaller MD, Feihl F, Ribordy V, Liaudet L. From evidence to clinical practice: effective implementation of therapeutic hypothermia to improve patient outcome after cardiac arrest. Crit Care Med. 2006;34: Vaahersalo J, Hiltunen P, Tiainen M, et al. Therapeutic hypothermia after out-of-hospital cardiac arrest in Finnish intensive care units: the FINNRESUSCI study. Intensive Care Med. 2013;39: Dumas F, Grimaldi D, Zuber B, et al. Is hypothermia after cardiac arrest effective in both shockable and nonshockable patients? Insights from a large registry. Circulation. 2011;: Bosson N, Kaji AH, Niemann JT, et al. Survival and neurologic outcome after out-of-hospital cardiac arrest: results one year after regionalization of post-cardiac arrest care in a large metropolitan area. Prehosp Emerg Care. 2014;18: Newgard CD, Hedges JR, Arthur M, Mullins RJ. Advanced statistics: the propensity score a method for estimating treatment effect in observational research. Acad Emerg Med. 2004;11: Arrich J, Holzer M, Havel C, Mullner M, Herkner H. Hypothermia for neuroprotection in adults after cardiopulmonary resuscitation. Cochrane Database Syst Rev. 2012;9:CD Kim YM, Yim HW, Jeong SH, Klem ML, Callaway CW. Does therapeutic hypothermia benefit adult cardiac arrest patients presenting with non-shockable initial rhythms? A systematic review and meta-analysis of randomized and non-randomized studies. Resuscitation. 2012;83: Walters JH, Morley PT, Nolan JP. The role of hypothermia in post-cardiac arrest patients with return of spontaneous circulation: a systematic review. Resuscitation. 2011;82: Lundbye JB, Rai M, Ramu B, et al. Therapeutic hypothermia is associated with improved neurologic outcome and survival in cardiac arrest survivors of non-shockable rhythms. Resuscitation. 2012;83: Sunde K, Pytte M, Jacobsen D, et al. Implementation of a standardised treatment protocol for post resuscitation care after outof-hospital cardiac arrest. Resuscitation. 2007;73: Nielsen N, Friberg H, Gluud C, Herlitz J, Wetterslev J. Hypothermia after cardiac arrest should be further evaluated a systematic review of randomised trials with meta-analysis and trial sequential analysis. Int J Cardiol. 2011;151: Fisher GC. Hypothermia after cardiac arrest: feasible but is it therapeutic? Anaesthesia 2008;63:885 6; author reply Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33 degrees C versus 36 degrees C after cardiac arrest. N Engl J Med. 2013;369: Zeiner A, Holzer M, Sterz F, et al. Hyperthermia after cardiac arrest is associated with an unfavorable neurologic outcome. Arch Intern Med. 2001;161: Neumar RW, Nolan JP, Adrie C, et al. Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication. A consensus statement from the International Liaison Committee on Resuscitation (American Heart Association, Australian and New Zealand Council on Resuscitation, European Resuscitation Council, Heart and Stroke Foundation of Canada, Inter American Heart Foundation, Resuscitation Council of Asia, and the Resuscitation Council of Southern Africa); the American Heart Association Emergency Cardiovascular Care Committee; the Council on Cardiovascular Surgery and Anesthesia; the Council on Cardiopulmonary, Perioperative, and Critical Care; the Council on Clinical Cardiology; and the Stroke Council. Circulation. 2008;118: Testori C, Sterz F, Holzer M, Losert H, Arrich J, Herkner H, et al. The beneficial effect of mild therapeutic hypothermia depends on the time of complete circulatory standstill in patients with cardiac arrest. Resuscitation. 2012;83:

7 26. Isaacs E, Callaham ML. Ability of laypersons to estimate short time intervals in cardiac arrest. Ann Emerg Med. 2000;35: Hallstrom AP. Should time from cardiac arrest until call to emergency medical services (EMS) be collected in EMS research? Crit Care Med. 2002;30:S Sawyer KN, Kurz MC. Caution when defining prolonged downtime in out of hospital cardiac arrest as extracorporeal cardiopulmonary resuscitation becomes accessible and feasible. Resuscitation. 2014;85: