- Email: [email protected]
Relative adrenal insufficiency after cardiac arrest: Impact on postresuscitation disease outcome
American Journal of Emergency Medicine (2006) 24, 684 – 688
www.elsevier.com/locate/ajem
Original Contribution
Relative adrenal insufficiency after cardiac arrest: Impact on postresuscitation disease outcomeB Jin Ju Kim MDa, Yong Su Lim MDa,*, Jong Hwan Shin MDa, Hyuk Jun Yang MD, PhDa, Jae Kwang Kim MDa, Sung Youl Hyun MD, PhDb, Ill Rhoo MDc, Sung Youn Hwang MDd, Gun Lee MD, PhDa a
Department of Emergency Medicine, Gachon Medical School Gil Medical Center, Incheon 405-760, South Korea Department of Thoracic Surgery, Gachon Medical School Gil Medical Center, Incheon 405-760, South Korea c Department of Pediatrics, Gachon Medical School Gil Medical Center, Incheon 405-760, South Korea d Department of Emergency Medicine, Masan Samsung Hospital, Sungkyunkwan University, Masan 630-723 South Korea b
Received 25 January 2006; accepted 16 February 2006
Abstract The purpose of this study is to prospectively evaluate the cortisol response and determine the relative adrenal insufficiency after return of spontaneous circulation (ROSC). Methods: This is a prospective cohort study describing the prevalence of relative adrenal insufficiency in consecutive patients admitted to an intensive care unit following ROSC (N24 hours) after cardiac arrest. The study institution is a 1300-bed urban hospital with an annual intensive care unit census of 80. Descriptive statistics is used to report results. Results: This study included 30 patients over a 1-year period. The mean patient age was 52 years, and there were 15 (50%) males. The most common etiology for cardiac arrest was unknown, and the most common initial rhythm was asystole. There were 16 (53%) witnessed arrests. The mean time from arrest to cardiopulmonary resuscitation was 8 minutes, and the mean time from arrest to ROSC was 23 minutes. Relative adrenal insufficiency (incremental response to corticotrophin test b 9 Ag/dL) at initial 12 to 24 hours and N24 to 48 hours occurred in 13 (43%) and 10 (33%) patients, respectively. All basal cortisol values were normal to high. Nineteen (58%) patients survived. The mean cortisol values at the initial 12 to 24 hours and N24 to 48 hours were similar among survivors and nonsurvivors. Relative adrenal insufficiency was present in only 5 (26%) survivors. Conclusion: Relative adrenal insufficiency may be associated with the increased mortality rate in patients with return of circulation after cardiac arrest. Although basal cortisol values were normal to high, treatment of relative adrenal insufficiency with corticosteroids may decrease the mortality rate. Additional studies are needed to confirm this association and to determine the effectiveness of treating relative adrenal insufficiency. D 2006 Elsevier Inc. All rights reserved.
B
This article was presented in The Third Mediterranean Emergency Medicine Congress by poster. * Corresponding author. E-mail address: [email protected] (Y.S. Lim). 0735-6757/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ajem.2006.02.017
Relative adrenal insufficiency after cardiac arrest
1. Introduction Adrenal insufficiency is a common disorder in critically ill patients, and it has been associated with various severe pathogenic conditions such as septic shock, hemorrhagic shock, human immunodeficiency virus infection, traumatic brain injury, and brain-dead patients [1-5]. The recovery of spontaneous circulation after cardiac arrest leads to a whole body ischemia-reperfusion syndrome designated bpostresuscitaion diseaseQ and resembling severe sepsis [1,2]. Postresuscitation disease is also associated with systemic inflammatory response similar to the immunoinflammatory profile observed in patients with severe sepsis [3]. In patients with sepsis, higher cortisol levels and the presence of a relative adrenal insufficiency, defined by an incremental response of less than 9 Ag/dL to the corticotropin test, were associated with poor outcome, and hydrocortisone and fludrocortisone therapy has been shown to decreased mortality and the duration of vasopressor therapy [1].
2. Purpose The objective of this study is to assess the prevalence of relative adrenal insufficiency in patients successfully resuscitated after cardiac arrest and its prognostic value in postresuscitation disease. Furthermore, we also would like to know the optimal timing for performing the corticotrophin test.
3. Materials and methods 3.1. Study population The study institution is a 1300-bed urban hospital with an annual intensive care unit (ICU) census of 80. All 30 consecutive patients older than 16 years who were admitted to our ICU between September 2004 and September 2005 after successful out-of-hospital cardiac arrest (OHCA) resuscitation were evaluated prospectively in this study. Following admission to the ICU, patients were treated with therapeutic hypothermia (32-348C), initiated as early as possible on admission and maintained for 24 hours using wet cold wrap and ice pack together with neuromuscular blocking and sedative agents to prevent shivering.
3.2. Clinical evaluation We divided the subjects into 3 groups, and several variables were compared in each group: (1) survivors (patients who had normal neurological function and had disability), (2) early dead from refractory shock (patients who died within 3 days from early refractory shock with multiple organ failure), and (3) late dead from neurological
685 dysfunction (patients who died later from neurological dysfunction or cardiovascular problem, eg, brain death, myocardial infarction, and etc.).
3.3. Laboratory variables We measured the serum cortisol level and performed rapid corticotrophin stimulation test 3 times: initial (between 1 and 12 h after return of spontaneous circulation [ROSC]), the first morning (between 12 and 24 h after ROSC), and the second morning (between 24 and 48 h after ROSC). Tetracosactrin (250 Ag, synacthene) was given intravenously, and blood sample was taken immediately before injection, and 30 and 60 min after injection. Relative adrenal insufficiency was defined as an increase of serum cortisol of not more than 9 Ag/dL.
Table 1
Characteristics of patients (n = 30)
Characteristics
Mean F standard deviation (minimum- maximum) or n (%)
Age, y Sex, M/F Arrest CPR, min Arrest ROSC, min Initial rhythm Asystole V-fibrillation PEA Countershocks, number Epinephrine, mg Cause of cardiac arrest Unknown Respiratory arrest CAOD Cerebral infarction Etc.* Simplified Acute Physiology Score II Acute Physiology and Chronic Health Evaluation II SOFA LOD Lactate Protein Basal cortisol Initial First morning Second morning
51.93 F 20.563 (16-84) 15 (50)/15 (50) 8.00 F 7.896 (1-30) 22.93 F 14.042 (5-70) 18 (60) 6 (20) 6 (20) 1.13 F 2.543 (0-13) 4.18 F 3.692 (0-17) 15 (50) 5 (17) 4 (13) 2 (7) 4 (13) 48.10 F 11.220 (25-69) 23.33 F 4.294 (15-32)
7.30 F 1.705 (4-12) 7.30 F 2.277 (4-15) 7.493 F 2.5906 (1.9-14.1) 6.267 F 0.7078 (5.0-8.0) 41.697 F 17.2007 (17.5-76.4) 29.933 F 18.5567 (10.4-89.1) 27.955 F 14.8993 (5.8-57.5)
Data are presented mean and standard deviation. (CPR: cardiopulmonary resuscitation, ROSC: return of spontaneous circulation, PEA: pulseless electrical activity, CAOD: coronary artery obstructive disease, SAPSII: simplified acute physiology score II, APACHEII: acute physiology and chronic health evaluation II, SOFA: sequential organ failure assessment, LOD: logistic organ dysfunction). * Electrical injury, asphyxia, anaphylaxis, cardiac arrhythmia.
686
J.J. Kim et al.
Table 2 Basal cortisol concentration of patients with or without relative adrenal insufficiency Basal cortisol concentration (Ag/dL)
RAI ( )
RAI (+)
P
Initial First morning Second morning
39.31 F 15.97 21.51 F 7.81 20.70 F 9.92
44.80 F 18.88 40.94 F 22.78 39.56 F 14.47
.40 .003 .01
Note. Mean basal cortisol levels at 1st and 2nd morning were significantly related to relative adrenal insufficiency.
analyzed, and the result of first morning was correlated with each other statistically. There were 19 patients who lived, 4 patients who died lately from neurological and cardiovascular complications, and 7 patients who died from refractory shock leading to irreversible multiple organ failure despite maximal life support within 3 days after ROSC. Table 3 shows the characteristics associated survivors and nonsurvivors. Survivors had a significantly shorter mean interval from cardiac arrest to CPR, lower SOFA scores and lower mean lactate level than
3.4. Statistical analysis The data were analyzed using the SPSS software (version 11.5, SPSS Inc, Chicago, Ill). T test and v 2 tests were used, and single and multiple variable logistic regression model analysis was performed to estimate odds ratio of dying along with 95% confidence intervals (CIs). Statistical significance was defined as P value less than .05.
4. Result Thirty patients were hospitalized in the ICU after successful resuscitation from OHCA. The mean patient age was 52 years, and there were 15 (50%) males. The most common etiology for cardiac arrest was unknown, and the most common initial rhythm was asystole. There were 16 (53%) witnessed arrests. The mean time from arrest to cardiopulmonary resuscitation (CPR) was 8 minutes, and the mean time from arrest to ROSC was 23 minutes. Of the 30 patients, 19 patients survived and 11 patients died, 7 patients from refractory shock and 2 patients from late neurologic and cardiovascular dysfunction. There were 22 patients (73%) who had relative adrenal insufficiency at initial, 13 patients (43%) at first morning, and 7 patients (23%) at second morning. Table 1 shows the characteristics of the patients. Arrest CPR is the time to start CPR by bystanders or doctors, and this means no-flow period. ROSC means ROSC over 24 hours. The most common cause of cardiac arrest was unknown because further evaluation was not performed due to the patient’s serious condition or the relative’s refusal (cost problem). After the patient has recovered mentally, various imaging studies were taken, such as brain magnetic resonance imaging, coronary computed tomography, coronary angiogram, electroencephalogram, and so on. Basal cortisol concentration in patients with relative adrenal insufficiency was higher than those without relative adrenal insufficiency (Table 2). Fig. 1 shows the distribution of baseline serum cortisol concentration. Basal serum cortisol level was high at initial and then significantly decreased at first morning in patients without relative adrenal insufficiency compared with that of patients with adrenal insufficiency ( P = .07). All the results at each day (initial, first morning, second morning) were
Fig. 1 Distribution of baseline serum cortisol concentration of initial, 1st morning and 2nd morning in patients with or without adrenal insufficiency (A). Basal cortisol concentration in patients with relative adrenal insufficiency was higher than without relative adrenal insufficiency (P = .001). Basal serum cortisol level was high at initial then significantly decrease at 1st morning in patients without relative adrenal insufficiency compared to those of patients with adrenal insufficiency (P = .07) (B).
Relative adrenal insufficiency after cardiac arrest
687
Table 3 Characteristics associated with survivors and nonsurvivors Characteristics
Age, y Arrest CPR, min Arrest ROSC, min Countershocks, number Epinephrine, mg Simplified Acute Physiology Score II Acute Physiology and Chronic Health Evaluation II SOFA LOD Lactate Protein Basal cortisol Initial First morning Second morning
P Survivors (mean Nonsurvivors (mean F standard F standard deviation) deviation) 53.79 F 19.29 5.42 F 5.79 20.95 F 12.97 1.05 F 1.39
48.73 F 23.21 12.45 F 9.28 26.36 F 15.77 1.27 F 3.90
.53 .02 .31 .82
3.89 F 2.69 45.63 F 10.19
4.68 F 5.11 52.36 F 12.10
.58 .11
22.84 F 3.20
6.84 6.84 6.77 6.28
F F F F
1.26 1.61 2.57 0.67
40.90 F 16.86 25.68 F 12.82 26.82 F 13.54
24.18 F 5.81
8.09 8.09 8.75 6.24
F F F F
2.12 3.05 2.21 0.79
43.06 F 18.53 37.27 F 24.69 31.04 F 18.96
.42
.05 .15 .04 .86 .75 .10 .53
Note. Survivors had a significantly longer mean interval from cardiac arrest to CPR, lower SOFA scores and higher mean lactate level than non-survivors. But there was no statistical difference between basal cortisol levels of survivors and nonsurvivors.
nonsurvivors. However, there was no statistical difference between the basal cortisol levels of survivors and nonsurvivors. Nonsurvivors include early death and late death groups. In the comparison of outcome in patients with or without adrenal insufficiency, patients with relative adrenal insufficiency had a tendency to die early from refractory shock or postresuscitation shock (Table 4; P = .008). In a single variable logistic regression model, several variables were correlated with shock-related (early refractory shock, postresuscitation shock) mortality. In the multiple variables logistic regression model analysis, patients with relative adrenal insufficiency were apt to have a poor outcome (Table 5; odds ratio 13.82, CI 1.22-155.93, P = .03).
Table 4 Comparison of patients with or without relative adrenal insufficiency in early dead (early refractory shock) and late dead and lived RAI ( ) RAI (+)
Early dead
Late dead
Lived
P
1 6
2 2
14 5
.008
RAI, relative adrenal insufficiency, P value was obtained by chi-square tests.
Table 5 Prognostic factors of shock-related mortality (multiple variables logistic regression model) Characteristics
P
Odds ratio
95% CI (lower-upper)
Arrest CPR SOFA Lactate RAI
.07 .61 .07 .03
1.17 1.20 1.63 13.82
0.98-1.37 0.58-2.47 0.96-2.77 1.22-155.93
5. Discussion Postresuscitation disease is a constellation of disorders related to whole body ischemia and reperfusion syndrome [1,3]. It shares many features with severe sepsis, including plasma cytokine elevation with deregulated cytokine production, the presence of endotoxin in plasma, coagulation abnormalities, and adrenal dysfunction [1]. During severe illness, many factors can impair the normal corticosteroid response affecting the hypothalamic-pituitary-adrenal axis [2]. Adrenal insufficiency is common in critically ill patients and associated with various pathogenic conditions like severe infections, trauma, burns, and so on [5,6]. Cardiac arrest is also an extreme stressful pathogenic condition, but only few studies have investigated on adrenal response after cardiac arrest. Hekimian [4] reported on cortisol levels and adrenal response after cardiac arrest, lower cortisol levels in the survivor group, and the group of patients who died of early refractory shock. Pene et al [7] also reported about relative adrenal insufficiency after OHCA, that baseline cortisol levels were significantly higher in patients with relative adrenal insufficiency, and the presence of relative adrenal insufficiency was predictive of multiple organ failure–related mortality. In our study, relative adrenal insufficiency occurs frequently at initial (immediate after admission to ICU) after ROSC after OHCA, and the basal serum cortisol levels decreased significantly at first morning in patients without relative adrenal insufficiency and the presence of relative adrenal insufficiency at first morning after ROSC was associated significantly with shock-related mortality. Following this study, in patients with relative adrenal insufficiency at first morning after ROSC, the risk of early death from refractory shock could be developed 13 times more. In Pene et al and in our study, basal cortisol levels were not low in patients with refractory shock, even higher than normal, and this is opposite to the results of the previous study. The important thing is not a single basal cortisol concentration but the presence of relative adrenal insufficiency after corticotropin stimulation test. Our study have several limitations. A small size of the study population restricted the result, and the definite level of relative adrenal insufficiency among hypothalamicpituitary-adrenal axis was not proved. Actually, other hormones like adrenocorticotropic hormone and vasopressin were checked in our study but were excluded because of laboratory machine failure.
688 Mild therapeutic hypothermia was performed on most of the patients. Whether mild hypothermia can affect the cortisol levels and adrenal response has been studied but remains controversial. In some studies, there are no significant difference between hypothermia and normothermia [8,9]. During hypothermia, several sedatives and muscle relaxants were used, and this also can affect the adrenal function [9,10]. Consequently, the optimal timing for obtaining the basal and poststimulation serum cortisol concentration may be between 12 and 24 hours after ROSC (8:00 am next morning), and this is helpful to estimate a prognosis of postresuscitation disease. We suggest that although a basal serum cortisol level is normal or even high, if the patients have relative adrenal insufficiency, especially in patients with refractory shock, low-dose corticosteroid therapy may be considered, and more randomized control studies are needed.
References [1] Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients. N Engl J Med 2003;348:727 - 34. [2] Schultz CH, Rivers EP, Feldkamp CS, et al. A characterization of hypothalamic-pituitary-adrenal axis fuction during and after human cardiac arrest. Crit Care Med 1993;21:1339 - 47.
J.J. Kim et al. [3] Adrie C, Laurent I, Monchi M, et al. Postresuscitation disease after cardiac arrest: a sepsis-like syndrome? Curr Opin Crit Care 2004; 10(3):208 - 12. [4] Hekimian G, Baugnon T, Thuong M, et al. Cortisol levels and adrenal reserve after successful cardiac arrest resuscitation. Shock 2004; 22:116 - 9. [5] Offner PJ, Moore EE, Ciesla D. The adrenal response after severe trauma. Am J Surg 2002;184:649 - 54. [6] Cohan P, Wang C, McArthur DL, et al. Acute secondary adrenal insufficiency after traumatic brain injury: a prospective study. Crit Care Med 2005;33:2358 - 66. [7] Pene F, Hyveirnat H, Mallet V, et al. Prognostic value of relative adrenal insufficiency after out-of-hospital cardiac arrest. Intensive Care Med 2005;31:627 - 33. [8] Holmes CL, Walley KR. The evaluation and management of shock. Clin Chest Med 2003;24:775 - 89. [9] Burchard K. A review of the adrenal cortex and severe inflammation: quest of the bEucorticoidQ state. J Trauma 2001;51:800 - 14. [10] Negovsky VA. Postresuscitation disease. Crit Care Med 1988;16: 942 - 6.
www.elsevier.com/locate/ajem
Original Contribution
Relative adrenal insufficiency after cardiac arrest: Impact on postresuscitation disease outcomeB Jin Ju Kim MDa, Yong Su Lim MDa,*, Jong Hwan Shin MDa, Hyuk Jun Yang MD, PhDa, Jae Kwang Kim MDa, Sung Youl Hyun MD, PhDb, Ill Rhoo MDc, Sung Youn Hwang MDd, Gun Lee MD, PhDa a
Department of Emergency Medicine, Gachon Medical School Gil Medical Center, Incheon 405-760, South Korea Department of Thoracic Surgery, Gachon Medical School Gil Medical Center, Incheon 405-760, South Korea c Department of Pediatrics, Gachon Medical School Gil Medical Center, Incheon 405-760, South Korea d Department of Emergency Medicine, Masan Samsung Hospital, Sungkyunkwan University, Masan 630-723 South Korea b
Received 25 January 2006; accepted 16 February 2006
Abstract The purpose of this study is to prospectively evaluate the cortisol response and determine the relative adrenal insufficiency after return of spontaneous circulation (ROSC). Methods: This is a prospective cohort study describing the prevalence of relative adrenal insufficiency in consecutive patients admitted to an intensive care unit following ROSC (N24 hours) after cardiac arrest. The study institution is a 1300-bed urban hospital with an annual intensive care unit census of 80. Descriptive statistics is used to report results. Results: This study included 30 patients over a 1-year period. The mean patient age was 52 years, and there were 15 (50%) males. The most common etiology for cardiac arrest was unknown, and the most common initial rhythm was asystole. There were 16 (53%) witnessed arrests. The mean time from arrest to cardiopulmonary resuscitation was 8 minutes, and the mean time from arrest to ROSC was 23 minutes. Relative adrenal insufficiency (incremental response to corticotrophin test b 9 Ag/dL) at initial 12 to 24 hours and N24 to 48 hours occurred in 13 (43%) and 10 (33%) patients, respectively. All basal cortisol values were normal to high. Nineteen (58%) patients survived. The mean cortisol values at the initial 12 to 24 hours and N24 to 48 hours were similar among survivors and nonsurvivors. Relative adrenal insufficiency was present in only 5 (26%) survivors. Conclusion: Relative adrenal insufficiency may be associated with the increased mortality rate in patients with return of circulation after cardiac arrest. Although basal cortisol values were normal to high, treatment of relative adrenal insufficiency with corticosteroids may decrease the mortality rate. Additional studies are needed to confirm this association and to determine the effectiveness of treating relative adrenal insufficiency. D 2006 Elsevier Inc. All rights reserved.
B
This article was presented in The Third Mediterranean Emergency Medicine Congress by poster. * Corresponding author. E-mail address: [email protected] (Y.S. Lim). 0735-6757/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ajem.2006.02.017
Relative adrenal insufficiency after cardiac arrest
1. Introduction Adrenal insufficiency is a common disorder in critically ill patients, and it has been associated with various severe pathogenic conditions such as septic shock, hemorrhagic shock, human immunodeficiency virus infection, traumatic brain injury, and brain-dead patients [1-5]. The recovery of spontaneous circulation after cardiac arrest leads to a whole body ischemia-reperfusion syndrome designated bpostresuscitaion diseaseQ and resembling severe sepsis [1,2]. Postresuscitation disease is also associated with systemic inflammatory response similar to the immunoinflammatory profile observed in patients with severe sepsis [3]. In patients with sepsis, higher cortisol levels and the presence of a relative adrenal insufficiency, defined by an incremental response of less than 9 Ag/dL to the corticotropin test, were associated with poor outcome, and hydrocortisone and fludrocortisone therapy has been shown to decreased mortality and the duration of vasopressor therapy [1].
2. Purpose The objective of this study is to assess the prevalence of relative adrenal insufficiency in patients successfully resuscitated after cardiac arrest and its prognostic value in postresuscitation disease. Furthermore, we also would like to know the optimal timing for performing the corticotrophin test.
3. Materials and methods 3.1. Study population The study institution is a 1300-bed urban hospital with an annual intensive care unit (ICU) census of 80. All 30 consecutive patients older than 16 years who were admitted to our ICU between September 2004 and September 2005 after successful out-of-hospital cardiac arrest (OHCA) resuscitation were evaluated prospectively in this study. Following admission to the ICU, patients were treated with therapeutic hypothermia (32-348C), initiated as early as possible on admission and maintained for 24 hours using wet cold wrap and ice pack together with neuromuscular blocking and sedative agents to prevent shivering.
3.2. Clinical evaluation We divided the subjects into 3 groups, and several variables were compared in each group: (1) survivors (patients who had normal neurological function and had disability), (2) early dead from refractory shock (patients who died within 3 days from early refractory shock with multiple organ failure), and (3) late dead from neurological
685 dysfunction (patients who died later from neurological dysfunction or cardiovascular problem, eg, brain death, myocardial infarction, and etc.).
3.3. Laboratory variables We measured the serum cortisol level and performed rapid corticotrophin stimulation test 3 times: initial (between 1 and 12 h after return of spontaneous circulation [ROSC]), the first morning (between 12 and 24 h after ROSC), and the second morning (between 24 and 48 h after ROSC). Tetracosactrin (250 Ag, synacthene) was given intravenously, and blood sample was taken immediately before injection, and 30 and 60 min after injection. Relative adrenal insufficiency was defined as an increase of serum cortisol of not more than 9 Ag/dL.
Table 1
Characteristics of patients (n = 30)
Characteristics
Mean F standard deviation (minimum- maximum) or n (%)
Age, y Sex, M/F Arrest CPR, min Arrest ROSC, min Initial rhythm Asystole V-fibrillation PEA Countershocks, number Epinephrine, mg Cause of cardiac arrest Unknown Respiratory arrest CAOD Cerebral infarction Etc.* Simplified Acute Physiology Score II Acute Physiology and Chronic Health Evaluation II SOFA LOD Lactate Protein Basal cortisol Initial First morning Second morning
51.93 F 20.563 (16-84) 15 (50)/15 (50) 8.00 F 7.896 (1-30) 22.93 F 14.042 (5-70) 18 (60) 6 (20) 6 (20) 1.13 F 2.543 (0-13) 4.18 F 3.692 (0-17) 15 (50) 5 (17) 4 (13) 2 (7) 4 (13) 48.10 F 11.220 (25-69) 23.33 F 4.294 (15-32)
7.30 F 1.705 (4-12) 7.30 F 2.277 (4-15) 7.493 F 2.5906 (1.9-14.1) 6.267 F 0.7078 (5.0-8.0) 41.697 F 17.2007 (17.5-76.4) 29.933 F 18.5567 (10.4-89.1) 27.955 F 14.8993 (5.8-57.5)
Data are presented mean and standard deviation. (CPR: cardiopulmonary resuscitation, ROSC: return of spontaneous circulation, PEA: pulseless electrical activity, CAOD: coronary artery obstructive disease, SAPSII: simplified acute physiology score II, APACHEII: acute physiology and chronic health evaluation II, SOFA: sequential organ failure assessment, LOD: logistic organ dysfunction). * Electrical injury, asphyxia, anaphylaxis, cardiac arrhythmia.
686
J.J. Kim et al.
Table 2 Basal cortisol concentration of patients with or without relative adrenal insufficiency Basal cortisol concentration (Ag/dL)
RAI ( )
RAI (+)
P
Initial First morning Second morning
39.31 F 15.97 21.51 F 7.81 20.70 F 9.92
44.80 F 18.88 40.94 F 22.78 39.56 F 14.47
.40 .003 .01
Note. Mean basal cortisol levels at 1st and 2nd morning were significantly related to relative adrenal insufficiency.
analyzed, and the result of first morning was correlated with each other statistically. There were 19 patients who lived, 4 patients who died lately from neurological and cardiovascular complications, and 7 patients who died from refractory shock leading to irreversible multiple organ failure despite maximal life support within 3 days after ROSC. Table 3 shows the characteristics associated survivors and nonsurvivors. Survivors had a significantly shorter mean interval from cardiac arrest to CPR, lower SOFA scores and lower mean lactate level than
3.4. Statistical analysis The data were analyzed using the SPSS software (version 11.5, SPSS Inc, Chicago, Ill). T test and v 2 tests were used, and single and multiple variable logistic regression model analysis was performed to estimate odds ratio of dying along with 95% confidence intervals (CIs). Statistical significance was defined as P value less than .05.
4. Result Thirty patients were hospitalized in the ICU after successful resuscitation from OHCA. The mean patient age was 52 years, and there were 15 (50%) males. The most common etiology for cardiac arrest was unknown, and the most common initial rhythm was asystole. There were 16 (53%) witnessed arrests. The mean time from arrest to cardiopulmonary resuscitation (CPR) was 8 minutes, and the mean time from arrest to ROSC was 23 minutes. Of the 30 patients, 19 patients survived and 11 patients died, 7 patients from refractory shock and 2 patients from late neurologic and cardiovascular dysfunction. There were 22 patients (73%) who had relative adrenal insufficiency at initial, 13 patients (43%) at first morning, and 7 patients (23%) at second morning. Table 1 shows the characteristics of the patients. Arrest CPR is the time to start CPR by bystanders or doctors, and this means no-flow period. ROSC means ROSC over 24 hours. The most common cause of cardiac arrest was unknown because further evaluation was not performed due to the patient’s serious condition or the relative’s refusal (cost problem). After the patient has recovered mentally, various imaging studies were taken, such as brain magnetic resonance imaging, coronary computed tomography, coronary angiogram, electroencephalogram, and so on. Basal cortisol concentration in patients with relative adrenal insufficiency was higher than those without relative adrenal insufficiency (Table 2). Fig. 1 shows the distribution of baseline serum cortisol concentration. Basal serum cortisol level was high at initial and then significantly decreased at first morning in patients without relative adrenal insufficiency compared with that of patients with adrenal insufficiency ( P = .07). All the results at each day (initial, first morning, second morning) were
Fig. 1 Distribution of baseline serum cortisol concentration of initial, 1st morning and 2nd morning in patients with or without adrenal insufficiency (A). Basal cortisol concentration in patients with relative adrenal insufficiency was higher than without relative adrenal insufficiency (P = .001). Basal serum cortisol level was high at initial then significantly decrease at 1st morning in patients without relative adrenal insufficiency compared to those of patients with adrenal insufficiency (P = .07) (B).
Relative adrenal insufficiency after cardiac arrest
687
Table 3 Characteristics associated with survivors and nonsurvivors Characteristics
Age, y Arrest CPR, min Arrest ROSC, min Countershocks, number Epinephrine, mg Simplified Acute Physiology Score II Acute Physiology and Chronic Health Evaluation II SOFA LOD Lactate Protein Basal cortisol Initial First morning Second morning
P Survivors (mean Nonsurvivors (mean F standard F standard deviation) deviation) 53.79 F 19.29 5.42 F 5.79 20.95 F 12.97 1.05 F 1.39
48.73 F 23.21 12.45 F 9.28 26.36 F 15.77 1.27 F 3.90
.53 .02 .31 .82
3.89 F 2.69 45.63 F 10.19
4.68 F 5.11 52.36 F 12.10
.58 .11
22.84 F 3.20
6.84 6.84 6.77 6.28
F F F F
1.26 1.61 2.57 0.67
40.90 F 16.86 25.68 F 12.82 26.82 F 13.54
24.18 F 5.81
8.09 8.09 8.75 6.24
F F F F
2.12 3.05 2.21 0.79
43.06 F 18.53 37.27 F 24.69 31.04 F 18.96
.42
.05 .15 .04 .86 .75 .10 .53
Note. Survivors had a significantly longer mean interval from cardiac arrest to CPR, lower SOFA scores and higher mean lactate level than non-survivors. But there was no statistical difference between basal cortisol levels of survivors and nonsurvivors.
nonsurvivors. However, there was no statistical difference between the basal cortisol levels of survivors and nonsurvivors. Nonsurvivors include early death and late death groups. In the comparison of outcome in patients with or without adrenal insufficiency, patients with relative adrenal insufficiency had a tendency to die early from refractory shock or postresuscitation shock (Table 4; P = .008). In a single variable logistic regression model, several variables were correlated with shock-related (early refractory shock, postresuscitation shock) mortality. In the multiple variables logistic regression model analysis, patients with relative adrenal insufficiency were apt to have a poor outcome (Table 5; odds ratio 13.82, CI 1.22-155.93, P = .03).
Table 4 Comparison of patients with or without relative adrenal insufficiency in early dead (early refractory shock) and late dead and lived RAI ( ) RAI (+)
Early dead
Late dead
Lived
P
1 6
2 2
14 5
.008
RAI, relative adrenal insufficiency, P value was obtained by chi-square tests.
Table 5 Prognostic factors of shock-related mortality (multiple variables logistic regression model) Characteristics
P
Odds ratio
95% CI (lower-upper)
Arrest CPR SOFA Lactate RAI
.07 .61 .07 .03
1.17 1.20 1.63 13.82
0.98-1.37 0.58-2.47 0.96-2.77 1.22-155.93
5. Discussion Postresuscitation disease is a constellation of disorders related to whole body ischemia and reperfusion syndrome [1,3]. It shares many features with severe sepsis, including plasma cytokine elevation with deregulated cytokine production, the presence of endotoxin in plasma, coagulation abnormalities, and adrenal dysfunction [1]. During severe illness, many factors can impair the normal corticosteroid response affecting the hypothalamic-pituitary-adrenal axis [2]. Adrenal insufficiency is common in critically ill patients and associated with various pathogenic conditions like severe infections, trauma, burns, and so on [5,6]. Cardiac arrest is also an extreme stressful pathogenic condition, but only few studies have investigated on adrenal response after cardiac arrest. Hekimian [4] reported on cortisol levels and adrenal response after cardiac arrest, lower cortisol levels in the survivor group, and the group of patients who died of early refractory shock. Pene et al [7] also reported about relative adrenal insufficiency after OHCA, that baseline cortisol levels were significantly higher in patients with relative adrenal insufficiency, and the presence of relative adrenal insufficiency was predictive of multiple organ failure–related mortality. In our study, relative adrenal insufficiency occurs frequently at initial (immediate after admission to ICU) after ROSC after OHCA, and the basal serum cortisol levels decreased significantly at first morning in patients without relative adrenal insufficiency and the presence of relative adrenal insufficiency at first morning after ROSC was associated significantly with shock-related mortality. Following this study, in patients with relative adrenal insufficiency at first morning after ROSC, the risk of early death from refractory shock could be developed 13 times more. In Pene et al and in our study, basal cortisol levels were not low in patients with refractory shock, even higher than normal, and this is opposite to the results of the previous study. The important thing is not a single basal cortisol concentration but the presence of relative adrenal insufficiency after corticotropin stimulation test. Our study have several limitations. A small size of the study population restricted the result, and the definite level of relative adrenal insufficiency among hypothalamicpituitary-adrenal axis was not proved. Actually, other hormones like adrenocorticotropic hormone and vasopressin were checked in our study but were excluded because of laboratory machine failure.
688 Mild therapeutic hypothermia was performed on most of the patients. Whether mild hypothermia can affect the cortisol levels and adrenal response has been studied but remains controversial. In some studies, there are no significant difference between hypothermia and normothermia [8,9]. During hypothermia, several sedatives and muscle relaxants were used, and this also can affect the adrenal function [9,10]. Consequently, the optimal timing for obtaining the basal and poststimulation serum cortisol concentration may be between 12 and 24 hours after ROSC (8:00 am next morning), and this is helpful to estimate a prognosis of postresuscitation disease. We suggest that although a basal serum cortisol level is normal or even high, if the patients have relative adrenal insufficiency, especially in patients with refractory shock, low-dose corticosteroid therapy may be considered, and more randomized control studies are needed.
References [1] Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients. N Engl J Med 2003;348:727 - 34. [2] Schultz CH, Rivers EP, Feldkamp CS, et al. A characterization of hypothalamic-pituitary-adrenal axis fuction during and after human cardiac arrest. Crit Care Med 1993;21:1339 - 47.
J.J. Kim et al. [3] Adrie C, Laurent I, Monchi M, et al. Postresuscitation disease after cardiac arrest: a sepsis-like syndrome? Curr Opin Crit Care 2004; 10(3):208 - 12. [4] Hekimian G, Baugnon T, Thuong M, et al. Cortisol levels and adrenal reserve after successful cardiac arrest resuscitation. Shock 2004; 22:116 - 9. [5] Offner PJ, Moore EE, Ciesla D. The adrenal response after severe trauma. Am J Surg 2002;184:649 - 54. [6] Cohan P, Wang C, McArthur DL, et al. Acute secondary adrenal insufficiency after traumatic brain injury: a prospective study. Crit Care Med 2005;33:2358 - 66. [7] Pene F, Hyveirnat H, Mallet V, et al. Prognostic value of relative adrenal insufficiency after out-of-hospital cardiac arrest. Intensive Care Med 2005;31:627 - 33. [8] Holmes CL, Walley KR. The evaluation and management of shock. Clin Chest Med 2003;24:775 - 89. [9] Burchard K. A review of the adrenal cortex and severe inflammation: quest of the bEucorticoidQ state. J Trauma 2001;51:800 - 14. [10] Negovsky VA. Postresuscitation disease. Crit Care Med 1988;16: 942 - 6.