Neurocrit Care DOI 10.1007/s12028-014-0091-8
ORIGINAL ARTICLE
Increased Thyroxin During Therapeutic Hypothermia Predicts Death in Comatose Patients After Cardiac Arrest Mathieu van der Jagt • Saskia Knoops • Margriet F. C. de Jong Martin J. de Jong • Robin P. Peeters • A. B. Johan Groeneveld
•
Ó Springer Science+Business Media New York 2015
Abstract Background The course and prognostic value of pituitary-thyroid axis hormones is not well studied after cardiac arrest. We aimed to study the prognostic role of the pituitary-thyroid axis response to resuscitation from cardiac arrest before, during and after therapeutic hypothermia. Methods We conducted a retrospective cohort study in consecutive comatose patients after out-of-hospital cardiac arrest who were sampled before, during and up to 48 h after a 24-h period of therapeutic hypothermia in the intensive care unit (ICU). Thyroid-stimulating hormone, M. van der Jagt (&) A. B. J. Groeneveld Department of Intensive Care of Erasmus Medical Center, ‘s-Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands e-mail: [email protected] A. B. J. Groeneveld e-mail: [email protected] S. Knoops M. F. C. de Jong A. B. J. Groeneveld Vrije Universiteit Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands e-mail: [email protected]
total and free thyroxine (T4) and triiodothyronine (T3) were determined and compared between ICU outcome groups. Results We included twenty-nine patients. TSH levels were comparable in non-survivors (n = 17) and survivors (n = 12). The free T4 levels were higher in non–survivors than in survivors (P = 0.001), whereas the free T3 levels were comparable. All samples’ results similarly declined in both outcome groups up to 72 h after start of 24 h hypothermia. ROC curves analyses showed a maximum AUC of 0.83 (P = 0.003) for free T4 at the end of hypothermia with an optimal cut off C17.8 pmol/L to obtain 100 % specificity and positive predictive value for non-survival. Conclusion Non-survival after cardiac arrest, coma, and therapeutic hypothermia following successful resuscitation is associated with a transient increase in free T4, most probably due to inhibition of free T4 to T3 conversion. However, before routine clinical application, external validation of our finding to assess generalizability is warranted.
M. F. C. de Jong e-mail: [email protected]
Keywords Cardiac arrest Coma Pituitary–thyroid axis Prognosis Hypothermia Organ donation
S. Knoops Rode Kruis Ziekenhuis, Vondellaan 13, 1942 LE Beverwijk, The Netherlands
Introduction
M. J. de Jong Medisch Centrum Alkmaar, Wilhelminalaan 12, 1815 JD Alkmaar, The Netherlands e-mail: [email protected] R. P. Peeters Department of Internal Medicine, Erasmus Medical Center, ‘s-Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands e-mail: [email protected]
Patients resuscitated from cardiac arrest may exhibit dysregulation of hormonal homeostasis since transient ischemia of brain and endocrine organs, among other factors, may reduce the capacity for hormone production [1, 2]. On the other hand, extreme stress is expected to stimulate specific hypothalamus–pituitary axes and we previously found, for instance, increased adrenocorticotropic hormone and cortisol
123
Neurocrit Care
production after cardiac arrest and treatment with hypothermia, particularly in ultimate non-survivors [3]. In contrast, the hypothalamus–pituitary–thyroid has been reported to be depressed after cardiac arrest and resuscitation, although longitudinal measurements are lacking, and the depression is associated with brain death and non-survival in critically ill patients [1, 2, 4–7]. Although a cause–effect relation in these circumstances is hard to establish, thyroid hormone may protect against ischemia/reperfusion injury of brain and heart [2, 6, 8, 9]. Also, the thyroid hormone status of post arrest patients may affect the quality of remote organs for donation, so that repletion of an alleged thyroid hormone deficit has been tried to optimise donor organs in brain-dead patients after cardiac arrest, with varying success [2, 6, 10–14]. Factors affecting assessment of the thyroid status after cardiac arrest include the measurement of free versus total hormone levels and the time lapsed after cardiac arrest, since critical illness is typically characterised by the so called nonthyroidal illness (NTI) with low triiodothyronine (T3), low to normal (free) thyroxine (T4), and initially high and later low thyroid stimulating hormone (TSH) levels, which may be associated with mortality [1, 5, 7, 15–21]. This syndrome may result in part from impaired peripheral conversion of T4 to T3 as well as an increased degradation of T4 and T3, caused by tissue hypoxia, among other factors [17, 18, 21, 22]. Hypothermia may variably reduce the activity of the pituitary–thyroid axis [15, 16, 23–27], but data on its effect after cardiac arrest are lacking. Next to these pathophysiological considerations there is a need for validation of prognostic biomarkers in comatose patients after cardiac arrest, especially in those with persisting coma but intact bilateral somatosensory evoked potentials (SSEP) in whom prognostication is cumbersome [28]. Since thyroid hormone determination is readily available to clinicians it could be easily implemented in prognostication. In order to resolve the issues mentioned, we studied the pituitary–thyroid axis hormones and their subsequent course prior to, during and after therapeutic hypothermia in surviving and non-surviving patients after cardiac arrest and determined their prognostic value for survival. The hypothesis is that non-survivors have greater inhibition of the axis than survivors from cardiac arrest, particularly during therapeutic hypothermia.
Patients and Methods This is a two-centre prospective observational study. Consecutive patients admitted to the intensive care unit (ICU) of a university hospital (Vrije Universiteit Medical Centre, VUmc) and an affiliated regional hospital (Medisch
123
Centrum Alkmaar, MCA), in a 1-year period, in whom circulation had been restored after out-of-hospital or inhospital cardiac arrest and in whom therapeutic hypothermia was instituted because of coma (Glasgow Coma Scale (GCS) score B8) after otherwise successful resuscitation were eligible for the study. Patients were excluded if they met the following criteria: age 20/min or the presence of mechanical ventilation. Sepsis was defined as the presence of SIRS with a positive microbiological local (urine, trachea, or other) and/or blood culture. Patients were classified as having septic shock when meeting sepsis criteria and when systolic blood pressure was
ORIGINAL ARTICLE
Increased Thyroxin During Therapeutic Hypothermia Predicts Death in Comatose Patients After Cardiac Arrest Mathieu van der Jagt • Saskia Knoops • Margriet F. C. de Jong Martin J. de Jong • Robin P. Peeters • A. B. Johan Groeneveld
•
Ó Springer Science+Business Media New York 2015
Abstract Background The course and prognostic value of pituitary-thyroid axis hormones is not well studied after cardiac arrest. We aimed to study the prognostic role of the pituitary-thyroid axis response to resuscitation from cardiac arrest before, during and after therapeutic hypothermia. Methods We conducted a retrospective cohort study in consecutive comatose patients after out-of-hospital cardiac arrest who were sampled before, during and up to 48 h after a 24-h period of therapeutic hypothermia in the intensive care unit (ICU). Thyroid-stimulating hormone, M. van der Jagt (&) A. B. J. Groeneveld Department of Intensive Care of Erasmus Medical Center, ‘s-Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands e-mail: [email protected] A. B. J. Groeneveld e-mail: [email protected] S. Knoops M. F. C. de Jong A. B. J. Groeneveld Vrije Universiteit Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands e-mail: [email protected]
total and free thyroxine (T4) and triiodothyronine (T3) were determined and compared between ICU outcome groups. Results We included twenty-nine patients. TSH levels were comparable in non-survivors (n = 17) and survivors (n = 12). The free T4 levels were higher in non–survivors than in survivors (P = 0.001), whereas the free T3 levels were comparable. All samples’ results similarly declined in both outcome groups up to 72 h after start of 24 h hypothermia. ROC curves analyses showed a maximum AUC of 0.83 (P = 0.003) for free T4 at the end of hypothermia with an optimal cut off C17.8 pmol/L to obtain 100 % specificity and positive predictive value for non-survival. Conclusion Non-survival after cardiac arrest, coma, and therapeutic hypothermia following successful resuscitation is associated with a transient increase in free T4, most probably due to inhibition of free T4 to T3 conversion. However, before routine clinical application, external validation of our finding to assess generalizability is warranted.
M. F. C. de Jong e-mail: [email protected]
Keywords Cardiac arrest Coma Pituitary–thyroid axis Prognosis Hypothermia Organ donation
S. Knoops Rode Kruis Ziekenhuis, Vondellaan 13, 1942 LE Beverwijk, The Netherlands
Introduction
M. J. de Jong Medisch Centrum Alkmaar, Wilhelminalaan 12, 1815 JD Alkmaar, The Netherlands e-mail: [email protected] R. P. Peeters Department of Internal Medicine, Erasmus Medical Center, ‘s-Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands e-mail: [email protected]
Patients resuscitated from cardiac arrest may exhibit dysregulation of hormonal homeostasis since transient ischemia of brain and endocrine organs, among other factors, may reduce the capacity for hormone production [1, 2]. On the other hand, extreme stress is expected to stimulate specific hypothalamus–pituitary axes and we previously found, for instance, increased adrenocorticotropic hormone and cortisol
123
Neurocrit Care
production after cardiac arrest and treatment with hypothermia, particularly in ultimate non-survivors [3]. In contrast, the hypothalamus–pituitary–thyroid has been reported to be depressed after cardiac arrest and resuscitation, although longitudinal measurements are lacking, and the depression is associated with brain death and non-survival in critically ill patients [1, 2, 4–7]. Although a cause–effect relation in these circumstances is hard to establish, thyroid hormone may protect against ischemia/reperfusion injury of brain and heart [2, 6, 8, 9]. Also, the thyroid hormone status of post arrest patients may affect the quality of remote organs for donation, so that repletion of an alleged thyroid hormone deficit has been tried to optimise donor organs in brain-dead patients after cardiac arrest, with varying success [2, 6, 10–14]. Factors affecting assessment of the thyroid status after cardiac arrest include the measurement of free versus total hormone levels and the time lapsed after cardiac arrest, since critical illness is typically characterised by the so called nonthyroidal illness (NTI) with low triiodothyronine (T3), low to normal (free) thyroxine (T4), and initially high and later low thyroid stimulating hormone (TSH) levels, which may be associated with mortality [1, 5, 7, 15–21]. This syndrome may result in part from impaired peripheral conversion of T4 to T3 as well as an increased degradation of T4 and T3, caused by tissue hypoxia, among other factors [17, 18, 21, 22]. Hypothermia may variably reduce the activity of the pituitary–thyroid axis [15, 16, 23–27], but data on its effect after cardiac arrest are lacking. Next to these pathophysiological considerations there is a need for validation of prognostic biomarkers in comatose patients after cardiac arrest, especially in those with persisting coma but intact bilateral somatosensory evoked potentials (SSEP) in whom prognostication is cumbersome [28]. Since thyroid hormone determination is readily available to clinicians it could be easily implemented in prognostication. In order to resolve the issues mentioned, we studied the pituitary–thyroid axis hormones and their subsequent course prior to, during and after therapeutic hypothermia in surviving and non-surviving patients after cardiac arrest and determined their prognostic value for survival. The hypothesis is that non-survivors have greater inhibition of the axis than survivors from cardiac arrest, particularly during therapeutic hypothermia.
Patients and Methods This is a two-centre prospective observational study. Consecutive patients admitted to the intensive care unit (ICU) of a university hospital (Vrije Universiteit Medical Centre, VUmc) and an affiliated regional hospital (Medisch
123
Centrum Alkmaar, MCA), in a 1-year period, in whom circulation had been restored after out-of-hospital or inhospital cardiac arrest and in whom therapeutic hypothermia was instituted because of coma (Glasgow Coma Scale (GCS) score B8) after otherwise successful resuscitation were eligible for the study. Patients were excluded if they met the following criteria: age 20/min or the presence of mechanical ventilation. Sepsis was defined as the presence of SIRS with a positive microbiological local (urine, trachea, or other) and/or blood culture. Patients were classified as having septic shock when meeting sepsis criteria and when systolic blood pressure was
