- Email: [email protected]
The occurrence of shivering in cardiac arrest survivors undergoing therapeutic hypothermia is associated with a good neurologic outcome
Resuscitation 84 (2013) 626–629
Contents lists available at SciVerse ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
The occurrence of shivering in cardiac arrest survivors undergoing therapeutic hypothermia is associated with a good neurologic outcome夽 Sanjeev U. Nair a,c,∗ , Justin B. Lundbye b,c a b c
Division of Cardiology, Hartford Hospital, Hartford, CT, United States Division of Cardiology, Hospital of Central Connecticut, New Britain, CT, United States University of Connecticut School of Medicine, Farmington, CT, United States
a r t i c l e
i n f o
Article history: Received 24 September 2012 Received in revised form 25 October 2012 Accepted 19 November 2012
Keywords: Therapeutic hypothermia Cardiac arrest Shivering Neurologic outcome
a b s t r a c t Background: The incidence of shivering in cardiac arrest survivors who undergo therapeutic hypothermia (TH) is varied. Its occurrence is dependent on the integrity of multiple peripheral and central neurologic pathways. We hypothesized that cardiac arrest survivors who develop shivering while undergoing TH are more likely to have intact central neurologic pathways and thus have better neurologic outcome as compared to those who do not develop shivering during TH. Methods: Prospectively collected data on consecutive adult patients admitted to a tertiary center from 1/1/2007 to 11/1/2010 that survived a cardiac arrest and underwent TH were retrospectively analyzed. Patients who developed shivering during the cooling phase of TH formed the “shivering” group and those that did not formed the “non-shivering” group. The primary end-point: Pittsburgh Cerebral Performance Category (CPC) scale; good (CPC 1–2) or poor (CPC 3–5) neurological outcome prior to discharge from hospital. Results: Of the 129 cardiac arrest survivors who underwent TH, 34/94 (36%) patients in the “non-shivering” group as compared to 21/35 (60%) patients in the “shivering” group had good neurologic outcome (P = 0.02). After adjusting for confounders using binary logistic regression, occurrence of shivering (OR: 2.71, 95% CI 1.099–7.41, P = 0.04), time to return of spontaneous circulation (OR: 0.96, 95% CI 0.93–0.98, P = 0.004) and initial presenting rhythm (OR: 4.0, 95% CI 1.63–10.0, P = 0.002) were independent predictors of neurologic outcome. Conclusion: The occurrence of shivering in cardiac arrest survivors who undergo TH is associated with an increased likelihood of good neurologic outcome as compared to its absence. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The use of mild therapeutic hypothermia (TH) has been shown to be associated with improved neurologic outcome when used in patients successfully resuscitated from cardiac arrest.1,2 However, the beneficial effect of TH in these patients is also dependent on other clinical factors and thus the outcome is varied in this population. For instance the rhythm leading to cardiac arrest, bystander cardiopulmonary resuscitation (CPR) and time to return of spontaneous circulation (ROSC) have been shown to predict outcomes in cardiac arrest survivors undergoing TH.2–5 Thus, the prognostication of cardiac arrest survivors undergoing TH with regards to neurologic outcome remains challenging.
Shivering is a natural thermoregulatory response of the human body to cooling.6 The occurrence of shivering during TH has been a concern since uncontrolled shivering during this therapy has been associated with longer time to achieve target cooling temperatures and thus may adversely affect outcomes in cardiac arrest survivors.7 However, recent reports suggest that those patients who develop shivering during TH are more likely to have intact neurologic pathways and thus have better neurologic outcomes than those who do not.8,9 Thus, we hypothesized that those cardiac arrest survivors who underwent TH and developed shivering are more likely to have better neurologic outcome than those who do not develop shivering during TH. 2. Methods
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2012.11.018. ∗ Corresponding author at: 80 Seymour Street, Hartford, CT 06102, United States. Tel.: +1 860 545 4398; fax: +1 860 545 4008. E-mail address: doc [email protected] (S.U. Nair). 0300-9572/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.resuscitation.2012.11.018
2.1. Study population Consecutive adult patients who were admitted to the Cardiac Intensive Care Unit (CICU) at Hartford Hospital between January
S.U. Nair, J.B. Lundbye / Resuscitation 84 (2013) 626–629
1, 2007 and November 1, 2010 after successful resuscitation from an out-of-hospital or in-hospital cardiac arrest and underwent TH formed the study cohort. The inclusion criteria were: an age of 18–75 years, Glasgow Coma Scale ≤ 8 after ROSC, an estimated interval of 5–15 min from the patient’s collapse to the first attempt at resuscitation by emergency medical personnel, an interval of no more than 30 min from collapse to ROSC. Patients were excluded if they met any of the following criteria: missing key data, Glasgow Coma Scale > 8 after ROSC, a tympanic-membrane temperature below 30 ◦ C on admission, response to verbal commands after the ROSC and before initiation of hypothermia, evidence of hypotension (mean arterial pressure less than 60 mmHg or systolic blood pressure less than 90 mm of Hg) for more than 30 min after the ROSC and before initiation of hypothermia, evidence of hypoxemia (arterial oxygen saturation, less than 85%) for more than 15 min after the ROSC and before initiation of hypothermia. Patients who were pregnant or had a terminal illness that preceded the arrest did not receive TH.
2.2. Study design and procedures This was a retrospective study conducted on prospectively gathered data and was approved and conducted as per guidelines of the institutional review board at Hartford Hospital which also includes a formal ethical approval to conduct this study. Patients who formed the study cohort received TH and were admitted to the CICU after successful resuscitation from cardiac arrest. For purpose of the study, patients were grouped into the “non-shivering” and “shivering” groups respectively based on the occurrence of shivering during TH. Patients underwent TH with a goal to target temperature of 32–34 ◦ C using a combination of an infusion of 2 L of cold normal saline at 4 ◦ C administered via a central venous catheter during a period of 20–30 min with an intravenous pressure bag inflated to 300 mmHg, ice-packs and followed by use of an intracaval cooling device. Intracaval cooling was initiated as soon as feasible using an endovascular cooling catheter (ICY Catheter, ZOLL Temperature management, MA, USA) which was inserted into the right or left femoral vein in the CICU. The goal was to reach the target temperature within 4 h after return of spontaneous circulation. The target cooling temperature was defined as a core body temperature of 33 ± 1 ◦ C. Hypothermia was maintained by endovascular cooling for 18 h and the patients were then re-warmed to a target temperature 37.0 ◦ C set to a rate of 0.35 ◦ C/h. Baseline vital signs, including temperature measured by rectal probe, were obtained before and during therapeutic hypothermia therapy. Temperature was measured every 15 min for the first hour followed by hourly measurements and blood pressure as well as heart rate were measured hourly throughout the CICU stay. Baseline laboratory tests included sodium, potassium, chloride, blood urea nitrogen (BUN), creatinine, glucose, hemoglobin, white blood cell count, platelet count, international normalized ratio (INR), lactic acid and arterial blood gases which were measured every 6 h after initiation of cooling with 4 ◦ C normal saline infusion. Sedation (propofol, midazolam, lorazepam) was administered intravenously to all patients who underwent TH. Intravenous meperidine was used exclusively as treatment for shivering in patients as per hospital hypothermia protocol. Paralytic agent (atracurium, cistracurium, and vecuronium) was utilized either during intubation or only when shivering during TH could not be controlled using sedation or meperidine. Patient data were obtained by querying the hospital TH database as well as systematic chart review of cardiac arrest patients admitted to Hartford Hospital during the study period. Data extracted also included location of arrest, time to ROSC, whether the arrest
627
was witnessed, whether bystander cardiopulmonary resuscitation was provided and the initial presenting rhythm. 2.3. Evaluation of shivering As per the TH protocol at Hartford Hospital, the occurrence of shivering during TH (including the rewarming phase) is documented by trained CICU nurses who are involved with the administration of TH. The episodes of shivering were differentiated from seizures by nursing staff, non-neurologist and neurologist physicians using clinical criteria. This differentiation was also supported when feasible by the use of electro-encephalography by a neurologist physician. We also cross-referenced the data on shivering obtained from the TH database with the documentation of administration of intravenous meperidine which is exclusively used as treatment for shivering as per the hospital TH protocol. 2.4. Evaluation of outcome The endpoints of the study were either a good neurologic outcome, defined as a Pittsburgh cerebral-performance category of 1 (good recovery) or 2 (moderate disability) on a five category scale, or a poor neurologic outcome defined as Pittsburgh cerebralperformance categories 3 (severe disability), 4 (a vegetative state), and 5 (death) prior to discharge from the hospital.10–12 The neurological evaluation of the patients was conducted by a consultant neurologist who examined the patient on a daily basis till discharge from hospital or till death. 2.5. Statistical analysis Continuous variables which did not have a normal distribution were expressed as median and interquartile range. Medians were evaluated for significant differences using the Mann–Whitney U test. Categorical variables were expressed as counts and percentages which were analyzed using the chi-square or Fisher exact test. Cumulative survival curves were obtained using the Kaplan–Meier procedure and compared using the log-rank test. A multivariable analysis using binomial logistic regression was performed to determine the association of therapeutic hypothermia to good and poor neurologic outcomes as two separate end-points. Significant parameters obtained on univariable analysis as well as clinically significant variables with a P < 0.20 were entered into the logistic regression model. A forward stepwise selection procedure, based on the Wald statistic probability, was performed, with a threshold of P ≤ 0.05 and P ≥ 0.1 for variable entry and removal, respectively. Odds ratios (OR) and 95% confidence intervals (CI) were calculated from the model. Goodness of fit was evaluated using the Hosmer–Lemeshow test. A reasonable fit can be assumed if the result has a P value > 0.05. In all statistical analyses, a P < 0.05 was considered significant. All statistical analyses were two-tailed and were performed using SPSS version 17.0 software (SPSS, Chicago, IL). 3. Results The demographics and clinical characteristics of the study cohort are shown in Table 1. Overall the two groups were relatively similar except for male gender and time to reach target hypothermia. There were significantly more number of males in the “shivering” group as compared to the “non-shivering” group (96% vs. 83% respectively, P = 0.02). The time to reach target hypothermia was significantly more in the “shivering” group as compared to the “non-shivering” group (360 min vs. 273 min respectively, P = 0.02). More patients in the “shivering” group received paralytic agents as compared to “non-shivering” group (66% vs. 39% respectively,
628
S.U. Nair, J.B. Lundbye / Resuscitation 84 (2013) 626–629
Table 1 Demographics and clinical characteristics. Characteristics
Non-shivering group (N = 94)
Shivering group (N = 35)
P value
Age (years)a Male White History of CAD History of CM History of HTN History of ESRD History of DM Shockable rhythm Out-of-hospital Time to ROSC (min)a Vasopressor/inotrope GCS on admissiona MAP on admission (mm of Hg)a Total epinephrine (mg)a Time to hypothermia (min)a Use of paralytic agents Lowest core temperature (◦ C) CK-admission (U L−1 )a pH-admissiona S. Lactate-admission (mmol L−1 )a S. Creatinine-admission (mg%)a
59 (51–72) 96 (62%) 57 (60%) 30 (32%) 21 (22%) 66 (70%) 12 (8%) 33 (35%) 43 (46%) 65 (67%) 20 (12–30) 16 (17%) 3 (3–3) 88 (69–100) 2 (1–3) 270 (140–372) 36 (39%) 33 (32–33) 239 (119–547) 7.20 (7.02–7.31) 4.1 (2–8) 1.20 (0.90–1.70)
60 (50–75) 29 (83%) 25 (72) 12 (34%) 12 (34%) 24 (69%) 1 (3%) 11 (31%) 22 (63%) 29 (83%) 16 (10–25) 10 (29%) 3 (3–3) 97 (74–109) 2 (1–2) 363 (180–432) 23 (66%) 33 (33–34) 217 (139–571) 7.25 (7.17–7.32) 3 (2–5) 1.10 (0.90–1.20)
0.81 0.02 0.56 0.80 0.29 0.86 0.30 0.70 0.08 0.12 0.10 0.15 0.10 0.08 0.26 0.02 0.005 0.83 0.77 0.50 0.10 0.09
CAD, coronary artery disease; CK, creatine kinase; CM, cardiomyopathy; DM, diabetes mellitus; ESRD, end-stage renal disease; GCS, Glasgow Coma Scale; HTN, hypertension; MAP, mean arterial pressure; ROSC, return of spontaneous circulation. a Values shown as median (25–75% interquartile range). Table 2 Independent predictors of neurologic outcome in cardiac arrest survivors undergoing TH.
P=0.02
70
Neurologic Outcome (%)
60
Odds ratio
50
Good Neurologic Outcome Poor Neurologic Outcome
40 30 20
Presenting rhythm Occurrence of shivering Time to ROSC
4.0 2.71 0.96
95% Confidence interval 1.63–10.0 1.099–7.41 0.93–0.98
P value 0.002 0.04 0.004
ROSC, return of spontaneous circulation; TH, therapeutic hypothermia.
4. Discussion
10 0 Shivering* Non-shivering*
* = P < 0.05 between good and poor neurologic outcome within the group Fig. 1. Neurologic outcome in cardiac arrest survivors undergoing TH.
P = 0.005). There was no significant difference in the lowest core body temperature achieved in both study groups (33 ◦ C vs. 33 ◦ C respectively, P = 0.83). Of the 129 cardiac arrest survivors who underwent TH, 34/94 (36%) patients in the “non-shivering” group as compared to 21/35 (60%) patients in the “shivering” group had good neurologic outcome (P = 0.02) (Fig. 1). Among those patients who were alive at discharge (CPC 1–4), 34/37 (92%) patients in the “non-shivering” group as compared to 21/24 (88%) in the “shivering” group had good neurologic outcome (P = 0.57). Logistic regression analysis was performed to adjust for potential confounders which were gender, location of cardiac arrest, initial presenting rhythm, time to ROSC, use of vasopressor/inotrope, time to target temperature, mean arterial pressure (MAP) on admission, Glasgow Coma Scale (GCS) on admission, occurrence of shivering, serum creatinine and serum lactate on admission. Initial presenting rhythm (OR: 4.0, 95% CI 1.63–10.0, P = 0.002), occurrence of shivering (OR: 2.71, 95% CI 1.099–7.41, P = 0.04) and time to ROSC (OR: 0.96, 95% CI 0.93–0.98, P = 0.004) were independent predictors of neurologic outcome (Table 2).
Our study shows that in a cohort of cardiac arrest survivors who underwent TH, the occurrence of shivering independently predicts a better neurologic outcome. To our knowledge this is the first published study that shows this association. Shivering is a natural thermoregulatory response of the body to lowering of core temperature and has been shown to require multiple peripheral and central neurologic pathways for its
Lateral parabracheal nucleus (pons/brainstem)
Pre-opc area (hypothalalmus)
Brain
Lateral spinothalamic tract
Motor tracts
Spinal Cord
Peripheral (and central) thermal receptors
Skeletal muscles
Fig. 2. The shivering pathway.15
Periphery
S.U. Nair, J.B. Lundbye / Resuscitation 84 (2013) 626–629
manifestation (Fig. 2).13–15 The primary center of shivering in the brain is located in the pre-optic area of the hypothalamus.6,15 Recently it has also been shown that the lateral parabracheal center in the brainstem is a relay center for this pathway.15 Thus shivering can occur only if these central neurologic pathways are relatively intact which in turn may indicate intact higher brain function. Our study findings are consistent with a recent report presented by May et al. which shows that “shivering episodes” as indicated by a combined index of electromyography (EMG) activity and clinical identification of shivering correlated with CPC at hospital discharge.9 The study by May et al. demonstrated that in 39 patients who underwent TH following successful resuscitation from cardiac arrest, the occurrence of each episode of shivering was associated with a 35% increase in the odds of a good neurological outcome (CPC 1–2, OR 1.35, P = 0.05). In another study by the same group performed in 112 adult cardiac arrest survivors who underwent TH, the investigators found that patients with “little-or-no shivering” as per EMG patterns were less likely to have CPC 1 or 2 on discharge from hospital, and thus reflected more severe brain injury and impaired thermoregulation mechanisms.8 Another possible explanation for our study findings could be that there was an increased utilization of sedatives and paralytics in those patients who had shivering as compared to those that did not. This may have resulted in more efficient cooling in the patients with shivering as compared to those patients who had no shivering and thus affected neurologic outcome. In summary, based on the results of our study as well as those presented by May et al., the occurrence of shivering in cardiac arrest survivors undergoing TH is associated with a better neurologic outcome in comparison to those that do not have shivering. This clinical parameter is a potential variable that can be incorporated into a predictive model for neurologic outcome in cardiac arrest survivors undergoing TH. 5. Limitations This is a retrospective study conducted on prospective data collected at a single large tertiary center and limits the extent to which the results can be generalized. Our study findings may not be applicable to those hypothermia protocols which administer paralytic agents to all cardiac arrest patients undergoing TH as a prophylactic measure to prevent shivering. Retrospective studies are limited by the inability to adjust for confounding factors and referral bias. Although adjustment for significant differences seen between the two study groups was performed, unknown confounders could have influenced the results. The dose and frequency of administration of sedatives and paralytic agents in our study patients is not presently known and this could have influenced neurologic outcomes. Patients made “do-not-resuscitate” or “comfort measures/care withdrawal” could have affected the survival to discharge rates in both groups.
629
6. Conclusions In cardiac arrest survivors undergoing therapeutic hypothermia, the occurrence of shivering is independently associated with a good neurologic outcome as compared to its absence. Shivering can be used as an additional clinical predictor of good neurologic outcome during prognostic evaluation in patients undergoing TH. Disclosures Financial disclosure: Justin Lundbye, speaker for ZOLL Temperature Management. Unlabeled/unapproved use disclosure: ICY Catheter© for induction of therapeutic hypothermia. References 1. The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549–56. 2. 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;346:557–63. 3. Nolan JP, Neumar RW, Adrie C, et al. Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication. A Scientific Statement from the International Liaison Committee on Resuscitation; 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; the Council on Stroke. Resuscitation 2008;79:350–79. 4. 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;123:877–86. 5. 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;53:926–34. 6. Sessler DI. Thermoregulatory defense mechanisms. Crit Care Med 2009;37:S203–10. 7. Badjatia N, Kowalski RG, Schmidt JM, et al. Predictors and clinical implications of shivering during therapeutic normothermia. Neurocrit Care 2007;6: 186–91. 8. May T, Riker R, Seder D, McCrum B, Fraser G. A phenotypic description of shivering pattern during targeted temperature management after cardiac arrest using continuous EMG monitoring. Neurocrit Care 2010;13: S44. 9. May T, Seder D, McCrum B, et al. Association of shivering intensity with severity of brain injury during therapeutic hypothermia after cardiac arrest. Neurocrit Care 2010;13:S45. 10. Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet 1975;1:480–4. 11. [BRCT] Brain Resuscitation Clinical Trial II Study Group. A randomized clinical study of a calcium-entry blocker (lidoflazine) in the treatment of comatose survivors of cardiac arrest. N Engl J Med 1991;324:1225–31. 12. Safar P, Bircer N. Cardiopulmonary cerebral resuscitation: basic and advanced cardiac and trauma life support: an introduction to resuscitation medicine: an introduction to resuscitation medicine. 3rd ed. London: W.B. Saunders; 1988. 13. Boulant JA. Role of the preoptic-anterior hypothalamus in thermoregulation and fever. Clin Infect Dis 2000;31:S157–61. 14. Boulant JA. Neuronal basis of Hammel’s model for set-point thermoregulation. J Appl Physiol 2006;100:1347–54. 15. Nakamura K, Morrison SF. A thermosensory pathway that controls body temperature. Nat Neurosci 2008;11:62–71.
Contents lists available at SciVerse ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
The occurrence of shivering in cardiac arrest survivors undergoing therapeutic hypothermia is associated with a good neurologic outcome夽 Sanjeev U. Nair a,c,∗ , Justin B. Lundbye b,c a b c
Division of Cardiology, Hartford Hospital, Hartford, CT, United States Division of Cardiology, Hospital of Central Connecticut, New Britain, CT, United States University of Connecticut School of Medicine, Farmington, CT, United States
a r t i c l e
i n f o
Article history: Received 24 September 2012 Received in revised form 25 October 2012 Accepted 19 November 2012
Keywords: Therapeutic hypothermia Cardiac arrest Shivering Neurologic outcome
a b s t r a c t Background: The incidence of shivering in cardiac arrest survivors who undergo therapeutic hypothermia (TH) is varied. Its occurrence is dependent on the integrity of multiple peripheral and central neurologic pathways. We hypothesized that cardiac arrest survivors who develop shivering while undergoing TH are more likely to have intact central neurologic pathways and thus have better neurologic outcome as compared to those who do not develop shivering during TH. Methods: Prospectively collected data on consecutive adult patients admitted to a tertiary center from 1/1/2007 to 11/1/2010 that survived a cardiac arrest and underwent TH were retrospectively analyzed. Patients who developed shivering during the cooling phase of TH formed the “shivering” group and those that did not formed the “non-shivering” group. The primary end-point: Pittsburgh Cerebral Performance Category (CPC) scale; good (CPC 1–2) or poor (CPC 3–5) neurological outcome prior to discharge from hospital. Results: Of the 129 cardiac arrest survivors who underwent TH, 34/94 (36%) patients in the “non-shivering” group as compared to 21/35 (60%) patients in the “shivering” group had good neurologic outcome (P = 0.02). After adjusting for confounders using binary logistic regression, occurrence of shivering (OR: 2.71, 95% CI 1.099–7.41, P = 0.04), time to return of spontaneous circulation (OR: 0.96, 95% CI 0.93–0.98, P = 0.004) and initial presenting rhythm (OR: 4.0, 95% CI 1.63–10.0, P = 0.002) were independent predictors of neurologic outcome. Conclusion: The occurrence of shivering in cardiac arrest survivors who undergo TH is associated with an increased likelihood of good neurologic outcome as compared to its absence. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The use of mild therapeutic hypothermia (TH) has been shown to be associated with improved neurologic outcome when used in patients successfully resuscitated from cardiac arrest.1,2 However, the beneficial effect of TH in these patients is also dependent on other clinical factors and thus the outcome is varied in this population. For instance the rhythm leading to cardiac arrest, bystander cardiopulmonary resuscitation (CPR) and time to return of spontaneous circulation (ROSC) have been shown to predict outcomes in cardiac arrest survivors undergoing TH.2–5 Thus, the prognostication of cardiac arrest survivors undergoing TH with regards to neurologic outcome remains challenging.
Shivering is a natural thermoregulatory response of the human body to cooling.6 The occurrence of shivering during TH has been a concern since uncontrolled shivering during this therapy has been associated with longer time to achieve target cooling temperatures and thus may adversely affect outcomes in cardiac arrest survivors.7 However, recent reports suggest that those patients who develop shivering during TH are more likely to have intact neurologic pathways and thus have better neurologic outcomes than those who do not.8,9 Thus, we hypothesized that those cardiac arrest survivors who underwent TH and developed shivering are more likely to have better neurologic outcome than those who do not develop shivering during TH. 2. Methods
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2012.11.018. ∗ Corresponding author at: 80 Seymour Street, Hartford, CT 06102, United States. Tel.: +1 860 545 4398; fax: +1 860 545 4008. E-mail address: doc [email protected] (S.U. Nair). 0300-9572/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.resuscitation.2012.11.018
2.1. Study population Consecutive adult patients who were admitted to the Cardiac Intensive Care Unit (CICU) at Hartford Hospital between January
S.U. Nair, J.B. Lundbye / Resuscitation 84 (2013) 626–629
1, 2007 and November 1, 2010 after successful resuscitation from an out-of-hospital or in-hospital cardiac arrest and underwent TH formed the study cohort. The inclusion criteria were: an age of 18–75 years, Glasgow Coma Scale ≤ 8 after ROSC, an estimated interval of 5–15 min from the patient’s collapse to the first attempt at resuscitation by emergency medical personnel, an interval of no more than 30 min from collapse to ROSC. Patients were excluded if they met any of the following criteria: missing key data, Glasgow Coma Scale > 8 after ROSC, a tympanic-membrane temperature below 30 ◦ C on admission, response to verbal commands after the ROSC and before initiation of hypothermia, evidence of hypotension (mean arterial pressure less than 60 mmHg or systolic blood pressure less than 90 mm of Hg) for more than 30 min after the ROSC and before initiation of hypothermia, evidence of hypoxemia (arterial oxygen saturation, less than 85%) for more than 15 min after the ROSC and before initiation of hypothermia. Patients who were pregnant or had a terminal illness that preceded the arrest did not receive TH.
2.2. Study design and procedures This was a retrospective study conducted on prospectively gathered data and was approved and conducted as per guidelines of the institutional review board at Hartford Hospital which also includes a formal ethical approval to conduct this study. Patients who formed the study cohort received TH and were admitted to the CICU after successful resuscitation from cardiac arrest. For purpose of the study, patients were grouped into the “non-shivering” and “shivering” groups respectively based on the occurrence of shivering during TH. Patients underwent TH with a goal to target temperature of 32–34 ◦ C using a combination of an infusion of 2 L of cold normal saline at 4 ◦ C administered via a central venous catheter during a period of 20–30 min with an intravenous pressure bag inflated to 300 mmHg, ice-packs and followed by use of an intracaval cooling device. Intracaval cooling was initiated as soon as feasible using an endovascular cooling catheter (ICY Catheter, ZOLL Temperature management, MA, USA) which was inserted into the right or left femoral vein in the CICU. The goal was to reach the target temperature within 4 h after return of spontaneous circulation. The target cooling temperature was defined as a core body temperature of 33 ± 1 ◦ C. Hypothermia was maintained by endovascular cooling for 18 h and the patients were then re-warmed to a target temperature 37.0 ◦ C set to a rate of 0.35 ◦ C/h. Baseline vital signs, including temperature measured by rectal probe, were obtained before and during therapeutic hypothermia therapy. Temperature was measured every 15 min for the first hour followed by hourly measurements and blood pressure as well as heart rate were measured hourly throughout the CICU stay. Baseline laboratory tests included sodium, potassium, chloride, blood urea nitrogen (BUN), creatinine, glucose, hemoglobin, white blood cell count, platelet count, international normalized ratio (INR), lactic acid and arterial blood gases which were measured every 6 h after initiation of cooling with 4 ◦ C normal saline infusion. Sedation (propofol, midazolam, lorazepam) was administered intravenously to all patients who underwent TH. Intravenous meperidine was used exclusively as treatment for shivering in patients as per hospital hypothermia protocol. Paralytic agent (atracurium, cistracurium, and vecuronium) was utilized either during intubation or only when shivering during TH could not be controlled using sedation or meperidine. Patient data were obtained by querying the hospital TH database as well as systematic chart review of cardiac arrest patients admitted to Hartford Hospital during the study period. Data extracted also included location of arrest, time to ROSC, whether the arrest
627
was witnessed, whether bystander cardiopulmonary resuscitation was provided and the initial presenting rhythm. 2.3. Evaluation of shivering As per the TH protocol at Hartford Hospital, the occurrence of shivering during TH (including the rewarming phase) is documented by trained CICU nurses who are involved with the administration of TH. The episodes of shivering were differentiated from seizures by nursing staff, non-neurologist and neurologist physicians using clinical criteria. This differentiation was also supported when feasible by the use of electro-encephalography by a neurologist physician. We also cross-referenced the data on shivering obtained from the TH database with the documentation of administration of intravenous meperidine which is exclusively used as treatment for shivering as per the hospital TH protocol. 2.4. Evaluation of outcome The endpoints of the study were either a good neurologic outcome, defined as a Pittsburgh cerebral-performance category of 1 (good recovery) or 2 (moderate disability) on a five category scale, or a poor neurologic outcome defined as Pittsburgh cerebralperformance categories 3 (severe disability), 4 (a vegetative state), and 5 (death) prior to discharge from the hospital.10–12 The neurological evaluation of the patients was conducted by a consultant neurologist who examined the patient on a daily basis till discharge from hospital or till death. 2.5. Statistical analysis Continuous variables which did not have a normal distribution were expressed as median and interquartile range. Medians were evaluated for significant differences using the Mann–Whitney U test. Categorical variables were expressed as counts and percentages which were analyzed using the chi-square or Fisher exact test. Cumulative survival curves were obtained using the Kaplan–Meier procedure and compared using the log-rank test. A multivariable analysis using binomial logistic regression was performed to determine the association of therapeutic hypothermia to good and poor neurologic outcomes as two separate end-points. Significant parameters obtained on univariable analysis as well as clinically significant variables with a P < 0.20 were entered into the logistic regression model. A forward stepwise selection procedure, based on the Wald statistic probability, was performed, with a threshold of P ≤ 0.05 and P ≥ 0.1 for variable entry and removal, respectively. Odds ratios (OR) and 95% confidence intervals (CI) were calculated from the model. Goodness of fit was evaluated using the Hosmer–Lemeshow test. A reasonable fit can be assumed if the result has a P value > 0.05. In all statistical analyses, a P < 0.05 was considered significant. All statistical analyses were two-tailed and were performed using SPSS version 17.0 software (SPSS, Chicago, IL). 3. Results The demographics and clinical characteristics of the study cohort are shown in Table 1. Overall the two groups were relatively similar except for male gender and time to reach target hypothermia. There were significantly more number of males in the “shivering” group as compared to the “non-shivering” group (96% vs. 83% respectively, P = 0.02). The time to reach target hypothermia was significantly more in the “shivering” group as compared to the “non-shivering” group (360 min vs. 273 min respectively, P = 0.02). More patients in the “shivering” group received paralytic agents as compared to “non-shivering” group (66% vs. 39% respectively,
628
S.U. Nair, J.B. Lundbye / Resuscitation 84 (2013) 626–629
Table 1 Demographics and clinical characteristics. Characteristics
Non-shivering group (N = 94)
Shivering group (N = 35)
P value
Age (years)a Male White History of CAD History of CM History of HTN History of ESRD History of DM Shockable rhythm Out-of-hospital Time to ROSC (min)a Vasopressor/inotrope GCS on admissiona MAP on admission (mm of Hg)a Total epinephrine (mg)a Time to hypothermia (min)a Use of paralytic agents Lowest core temperature (◦ C) CK-admission (U L−1 )a pH-admissiona S. Lactate-admission (mmol L−1 )a S. Creatinine-admission (mg%)a
59 (51–72) 96 (62%) 57 (60%) 30 (32%) 21 (22%) 66 (70%) 12 (8%) 33 (35%) 43 (46%) 65 (67%) 20 (12–30) 16 (17%) 3 (3–3) 88 (69–100) 2 (1–3) 270 (140–372) 36 (39%) 33 (32–33) 239 (119–547) 7.20 (7.02–7.31) 4.1 (2–8) 1.20 (0.90–1.70)
60 (50–75) 29 (83%) 25 (72) 12 (34%) 12 (34%) 24 (69%) 1 (3%) 11 (31%) 22 (63%) 29 (83%) 16 (10–25) 10 (29%) 3 (3–3) 97 (74–109) 2 (1–2) 363 (180–432) 23 (66%) 33 (33–34) 217 (139–571) 7.25 (7.17–7.32) 3 (2–5) 1.10 (0.90–1.20)
0.81 0.02 0.56 0.80 0.29 0.86 0.30 0.70 0.08 0.12 0.10 0.15 0.10 0.08 0.26 0.02 0.005 0.83 0.77 0.50 0.10 0.09
CAD, coronary artery disease; CK, creatine kinase; CM, cardiomyopathy; DM, diabetes mellitus; ESRD, end-stage renal disease; GCS, Glasgow Coma Scale; HTN, hypertension; MAP, mean arterial pressure; ROSC, return of spontaneous circulation. a Values shown as median (25–75% interquartile range). Table 2 Independent predictors of neurologic outcome in cardiac arrest survivors undergoing TH.
P=0.02
70
Neurologic Outcome (%)
60
Odds ratio
50
Good Neurologic Outcome Poor Neurologic Outcome
40 30 20
Presenting rhythm Occurrence of shivering Time to ROSC
4.0 2.71 0.96
95% Confidence interval 1.63–10.0 1.099–7.41 0.93–0.98
P value 0.002 0.04 0.004
ROSC, return of spontaneous circulation; TH, therapeutic hypothermia.
4. Discussion
10 0 Shivering* Non-shivering*
* = P < 0.05 between good and poor neurologic outcome within the group Fig. 1. Neurologic outcome in cardiac arrest survivors undergoing TH.
P = 0.005). There was no significant difference in the lowest core body temperature achieved in both study groups (33 ◦ C vs. 33 ◦ C respectively, P = 0.83). Of the 129 cardiac arrest survivors who underwent TH, 34/94 (36%) patients in the “non-shivering” group as compared to 21/35 (60%) patients in the “shivering” group had good neurologic outcome (P = 0.02) (Fig. 1). Among those patients who were alive at discharge (CPC 1–4), 34/37 (92%) patients in the “non-shivering” group as compared to 21/24 (88%) in the “shivering” group had good neurologic outcome (P = 0.57). Logistic regression analysis was performed to adjust for potential confounders which were gender, location of cardiac arrest, initial presenting rhythm, time to ROSC, use of vasopressor/inotrope, time to target temperature, mean arterial pressure (MAP) on admission, Glasgow Coma Scale (GCS) on admission, occurrence of shivering, serum creatinine and serum lactate on admission. Initial presenting rhythm (OR: 4.0, 95% CI 1.63–10.0, P = 0.002), occurrence of shivering (OR: 2.71, 95% CI 1.099–7.41, P = 0.04) and time to ROSC (OR: 0.96, 95% CI 0.93–0.98, P = 0.004) were independent predictors of neurologic outcome (Table 2).
Our study shows that in a cohort of cardiac arrest survivors who underwent TH, the occurrence of shivering independently predicts a better neurologic outcome. To our knowledge this is the first published study that shows this association. Shivering is a natural thermoregulatory response of the body to lowering of core temperature and has been shown to require multiple peripheral and central neurologic pathways for its
Lateral parabracheal nucleus (pons/brainstem)
Pre-opc area (hypothalalmus)
Brain
Lateral spinothalamic tract
Motor tracts
Spinal Cord
Peripheral (and central) thermal receptors
Skeletal muscles
Fig. 2. The shivering pathway.15
Periphery
S.U. Nair, J.B. Lundbye / Resuscitation 84 (2013) 626–629
manifestation (Fig. 2).13–15 The primary center of shivering in the brain is located in the pre-optic area of the hypothalamus.6,15 Recently it has also been shown that the lateral parabracheal center in the brainstem is a relay center for this pathway.15 Thus shivering can occur only if these central neurologic pathways are relatively intact which in turn may indicate intact higher brain function. Our study findings are consistent with a recent report presented by May et al. which shows that “shivering episodes” as indicated by a combined index of electromyography (EMG) activity and clinical identification of shivering correlated with CPC at hospital discharge.9 The study by May et al. demonstrated that in 39 patients who underwent TH following successful resuscitation from cardiac arrest, the occurrence of each episode of shivering was associated with a 35% increase in the odds of a good neurological outcome (CPC 1–2, OR 1.35, P = 0.05). In another study by the same group performed in 112 adult cardiac arrest survivors who underwent TH, the investigators found that patients with “little-or-no shivering” as per EMG patterns were less likely to have CPC 1 or 2 on discharge from hospital, and thus reflected more severe brain injury and impaired thermoregulation mechanisms.8 Another possible explanation for our study findings could be that there was an increased utilization of sedatives and paralytics in those patients who had shivering as compared to those that did not. This may have resulted in more efficient cooling in the patients with shivering as compared to those patients who had no shivering and thus affected neurologic outcome. In summary, based on the results of our study as well as those presented by May et al., the occurrence of shivering in cardiac arrest survivors undergoing TH is associated with a better neurologic outcome in comparison to those that do not have shivering. This clinical parameter is a potential variable that can be incorporated into a predictive model for neurologic outcome in cardiac arrest survivors undergoing TH. 5. Limitations This is a retrospective study conducted on prospective data collected at a single large tertiary center and limits the extent to which the results can be generalized. Our study findings may not be applicable to those hypothermia protocols which administer paralytic agents to all cardiac arrest patients undergoing TH as a prophylactic measure to prevent shivering. Retrospective studies are limited by the inability to adjust for confounding factors and referral bias. Although adjustment for significant differences seen between the two study groups was performed, unknown confounders could have influenced the results. The dose and frequency of administration of sedatives and paralytic agents in our study patients is not presently known and this could have influenced neurologic outcomes. Patients made “do-not-resuscitate” or “comfort measures/care withdrawal” could have affected the survival to discharge rates in both groups.
629
6. Conclusions In cardiac arrest survivors undergoing therapeutic hypothermia, the occurrence of shivering is independently associated with a good neurologic outcome as compared to its absence. Shivering can be used as an additional clinical predictor of good neurologic outcome during prognostic evaluation in patients undergoing TH. Disclosures Financial disclosure: Justin Lundbye, speaker for ZOLL Temperature Management. Unlabeled/unapproved use disclosure: ICY Catheter© for induction of therapeutic hypothermia. References 1. The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549–56. 2. 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;346:557–63. 3. Nolan JP, Neumar RW, Adrie C, et al. Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication. A Scientific Statement from the International Liaison Committee on Resuscitation; 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; the Council on Stroke. Resuscitation 2008;79:350–79. 4. 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;123:877–86. 5. 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;53:926–34. 6. Sessler DI. Thermoregulatory defense mechanisms. Crit Care Med 2009;37:S203–10. 7. Badjatia N, Kowalski RG, Schmidt JM, et al. Predictors and clinical implications of shivering during therapeutic normothermia. Neurocrit Care 2007;6: 186–91. 8. May T, Riker R, Seder D, McCrum B, Fraser G. A phenotypic description of shivering pattern during targeted temperature management after cardiac arrest using continuous EMG monitoring. Neurocrit Care 2010;13: S44. 9. May T, Seder D, McCrum B, et al. Association of shivering intensity with severity of brain injury during therapeutic hypothermia after cardiac arrest. Neurocrit Care 2010;13:S45. 10. Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet 1975;1:480–4. 11. [BRCT] Brain Resuscitation Clinical Trial II Study Group. A randomized clinical study of a calcium-entry blocker (lidoflazine) in the treatment of comatose survivors of cardiac arrest. N Engl J Med 1991;324:1225–31. 12. Safar P, Bircer N. Cardiopulmonary cerebral resuscitation: basic and advanced cardiac and trauma life support: an introduction to resuscitation medicine: an introduction to resuscitation medicine. 3rd ed. London: W.B. Saunders; 1988. 13. Boulant JA. Role of the preoptic-anterior hypothalamus in thermoregulation and fever. Clin Infect Dis 2000;31:S157–61. 14. Boulant JA. Neuronal basis of Hammel’s model for set-point thermoregulation. J Appl Physiol 2006;100:1347–54. 15. Nakamura K, Morrison SF. A thermosensory pathway that controls body temperature. Nat Neurosci 2008;11:62–71.