Cardiac arrest is a rapidly lethal condition and associated with a high mortality rate, even after successful resuscitation. Following the return of spontaneous circulation (ROSC), postcardiac arrest syndrome, a complex combination of pathophysiologic processes that involve cerebral injury, myocardial dysfunction, systemic ischemia-reperfusion injury, and inflammatory response accounts for most hemodynamic instability and mortality during the postcardiac arrest period. Postcardiac arrest syndrome shares many features with severe sepsis and multiple organ failures such as increased cytokine production and release, endotoxinemia (1 , 2), coagulation abnormality (3 , 4), and adrenal dysfunction (5 , 6).
Several studies have shown that relative adrenal insufficiency is common during the postcardiac arrest period (5–9), and may cause impaired vasoregulation and reduced effectiveness of vasopressors, and thus result in postresuscitation shock (5 , 6). Schultz et al (5) reported that circulating cortisol levels in patients with out-of-hospital cardiac arrest (OHCA) were lower than those in patients with other stress conditions. Furthermore, a low serum cortisol level was associated with unstable hemodynamics after ROSC and short-term survival (24 hr). Hékimian et al (6) also reported that OHCA survivors with refractory shock and early mortality had lower cortisol levels than those who died later. Additionally, relative adrenal insufficiency has been reported to be a prognostic factor for early death (within 7 d after admission) and in-hospital mortality (7–9). In two randomized controlled trials conducted by Mentzelopoulos et al (10 , 11), patients who received vasopressin, epinephrine, and methylprednisolone during cardiopulmonary resuscitation (CPR) and stress-dose hydrocortisone during postcardiac arrest shock displayed a higher frequency of ROSC, better hemodynamic stability, less organ dysfunction, and a better rate of survival to hospital discharge with a favorable neurologic status when compared with patients who received epinephrine alone. However, because multiple interventions were used simultaneously, it was difficult to clarify which of the interventions were truly beneficial. Furthermore, it was difficult to discern the isolated effect of glucocorticoid use during postcardiac arrest care on the outcomes, because the ROSC rates in the study groups were significantly different. Conversely, several other studies do not support the use of glucocorticoids during the postcardiac arrest period, because those drugs did not significantly improve survival or neurologic outcomes (12–14).
The benefit of steroid supplement during CPR has been documented in several animal and human studies (15–17). In the current study, we hypothesized that steroid use during the postcardiac arrest period may be associated with improved patient outcomes. Therefore, we analyzed data from the Taiwan National Health Insurance Research Database (NHIRD) to determine potential associations between post-ROSC exposure to steroids and critical cardiac arrest outcomes.
Study Design and Data Sources
This nationwide retrospective cohort study on the effect of postcardiac arrest steroid use was conducted by analyzing health insurance administrative data retrieved from the Taiwan NHIRD for the interval of January 1, 2003, to December 31, 2012. The claims database, which was released by the National Health Research Institute (NHRI) for research purposes, provides comprehensive information on healthcare utilization and demographics (birthdate, gender, insurance status, area of residence, dates of services provided, primary and minor diagnostic codes, primary and minor procedural codes, and itemized expenditures for each medical service rendered) for greater than 99% of the entire Taiwanese population of 23 million people (18). The study protocol was approved by the NHRI and Review Board of the National Taiwan University Hospital, and the need for informed consent was waived. All data were deidentified, and confidentiality assurances were addressed in accordance with data regulations of the National Health Insurance (NHI) Administration.
Our database search identified 174,068 patients who had undergone CPR (procedure code: 47029C) in the emergency department between January 1, 2004, and December 31, 2011. Among those patients, 172,016 received subsequent medical care and were identified as possible candidates for our study. Patients who were less than 18 years old or older at date of presentation (n = 4,041) and those involved in trauma (n = 12,698) were excluded. To identify patients in cardiac arrest during acute hospital care, patients not triaged as level 1 (resuscitation/emergent) (n = 2,436), and patients with an emergency department observation time greater than 6 hours (n = 7,197) were also excluded. After identifying 22,768 nontraumatic adult cardiac arrest survivors, we further excluded patients who received steroid during CPR (n = 1,180), patients who received steroid for greater than 1 month after ROSC (n = 2,398), and patients given prednisolone equivalent dose greater than 1,250 mg/d (equal to methylprednisolone dose of 1 g/d as pulse therapy) (n = 23). This left 19,229 patients available for inclusion in our study (the first sample). Based on whether the patient received steroid during hospitalization, the 19,229 patients were then divided into the steroid group (n = 5,477) and nonsteroid group (n = 13,752). Among the first sample, 6,955 patients with a history of steroid use within 1 year (oral, IV, and intraarticular administration of steroid) prior to cardiac arrest were further selected as the secondary sample (steroid group, n = 2,227; nonsteroid group, n = 4,728). A total of 12,274 patients without a history of steroid within 1 year prior to cardiac arrest constituted the third sample (steroid group, n = 3,250; nonsteroid group, n = 9,024) (Fig. 1).
Definition of Variables
The primary study outcome was defined as survival to discharge, and the secondary outcome was 1-year survival. A death event was identified if the discharge status of index admission was “death,” the patient had a death record in the catastrophic illness registry, or the patient was disenrolled from the NHI program. The main focus of the study was the effect of steroid use during hospitalization following cardiac arrest, and the relevant information was collected from claims data generated by hospital care. The type and quantity of administered steroid during the entire hospitalization after ROSC were obtained from the NHIRD. The allowable steroidal supplements were hydrocortisone, methylprednisolone, prednisolone, triamcinolone, dexamethasone, and betamethasone. The controlled variables were age, gender, presenting complaint, shockable rhythm, epinephrine dosage, total shocks delivered during CPR, and cardiac catheterization. The presenting complaint was recognized from the International Classification of Diseases, 9th Edition code of primary diagnosis at admission. The steroid and nonsteroid groups had two patients with missing urbanization level of residence and geographic distribution, respectively. They were excluded from analysis of propensity score (PS). To clarify whether the steroid dose was associated with outcomes, we attempted to standardize the prednisolone equivalent dose (prednisolone equivalent dose/d), which was calculated as the total dose of prednisolone equivalent given during hospitalization (when hospitalization duration was < 1 mo) or the first month (when hospitalization duration was ≥ 1 mo), divided by the corresponding number of days (either hospitalization duration or 30 d). We then further divided the steroid group into four different strata using quartiles of standardized prednisolone equivalent dose.
Computation and Matching of PS
Propensity scoring was used to reduce selection bias between the steroid and nonsteroid groups. The PS, defined as the probability of receiving steroid, was first estimated by modeling a logistic regression drawn from pertinent characteristics (19). The Hosmer-Lemeshow goodness-of-fit test was then used to assess propensity models’ performance. Results of the logistic regression model used for PS matching are presented in Supplemental Table 1 (Supplemental Digital Content 1, http://links.lww.com/CCM/E67), Supplemental Table 2 (Supplemental Digital Content 2, http://links.lww.com/CCM/E68), and Supplemental Table 3 (Supplemental Digital Content 3, http://links.lww.com/CCM/E69).
The steroid and nonsteroid groups were matched by PS using 8 to 1 digit-based greedy matching algorithm with the nearest available pair matching method at 1:1 ratio (cases:controls) without replacement, as proposed by Parsons (20). The final PS-matched sample consisted of 5,445 patients in each group among total patients, 2,188 patients among patients with prior steroid use, and 3,248 patients among patients without prior steroid use, respectively.
The demographic and clinical characteristics of study subjects were summarized using the mean ± SD for continuous variables and frequencies and percentages for categorical variables. The standardized difference as effect size was calculated to compare group differences and assess the balance of the baseline characteristic of study subjects before and after PS matching (21).The impact of steroid use and the steroid dose on survival to discharge and 1-year survival was determined by the Cox proportional hazard model. A bootstrapping approach by resampling for 100 replications with replacement from the original samples was used to perform internal model validation in assessing survival to discharge (22).The adjusted 1-year survival curves were also plotted via the Cox proportional hazard model. When performing the dose analysis, a receiver operating characteristic (ROC) curve was plotted to determine the optimal steroid dose cut-off value. Subgroup analyses were performed to examine relationships between steroid use, clinical characteristics, and survival to discharge. All computations were performed using standard software (SAS/Stat v9.3 for Windows; SAS Institute, Cary, NC). The components of retrospective review were checked by using the Strengthening the Reporting of Observational Studies in Epidemiology checklist for a cohort study (Supplemental Table 4, Supplemental Digital Content 4, http://links.lww.com/CCM/E70).
The underlying characteristics, CPR events, and socioeconomic status of the enrolled patients are shown in Table 1. When compared with patients in the nonsteroid group, a higher percentage of patients in the steroid group had chronic obstructive pulmonary disease (COPD), asthma, adrenal insufficiency, autoimmune disease, steroid use within 1 year prior to cardiac arrest, and were treated in a tertiary medical center. In contrast, lower percentages of patients in the steroid group had coronary artery disease (CAD) chronic kidney disease (CKD), a low frequency of shockable rhythm, and required only a smaller dose of epinephrine. After matching with the PS, there were 5,445 patients in the steroid and nonsteroid groups, respectively. Supplemental Table 5 (Supplemental Digital Content 5, http://links.lww.com/CCM/E71) and Supplemental Table 6 (Supplemental Digital Content 6, http://links.lww.com/CCM/E72) shows demographic and clinical characteristics of the patients with and without steroid use prior to cardiac arrest, respectively. After PS matching, no difference existed between the steroid and nonsteroid groups in the patients with and without steroid use prior to cardiac arrest, respectively. Baseline characteristic of unmatched patients in each steroid group were listed in Supplemental Table 7 (Supplemental Digital Content 7, http://links.lww.com/CCM/E73).
Among all the samples (total patients, patients with and without prior steroid use before cardiac arrest), steroid use during hospitalization benefitted the following endpoints: survival to discharge and 1-year survival in both unmatched and matched cohorts (Table 2). With 100 bootstrap replications, the optimism-corrected area under the ROC curve ranged from 0.77 to 0.80 for models with either steroid use or steroid dose as a predictor, showing the models have good performance. The adjusted 1-year survival curves shown in Figure 2 showed significant difference between the two groups in total matched patients (adjusted hazard ratio, 0.73; 95% CI, 0.70–0.76; p < 0.0001). We next investigated whether the steroid use during hospitalization may have interacted with individual clinical characteristics to affect the survival to discharge. As shown in Supplemental Figure 1 (Supplemental Digital Content 8, http://links.lww.com/CCM/E74), steroid use during hospitalization was associated with better survival to discharge, regardless of age, gender, underlying diseases (diabetes, COPD, asthma), shockable rhythm, and steroid use within 1 year prior to cardiac arrest.
The outcomes of patients in each quartile with consecutive increases in steroid dose are listed in Table 3. The demographic and clinical characteristics of each quartile of the total patients, patients with and without steroid use prior to cardiac arrest are shown in Supplemental Table 8 (Supplemental Digital Content 9, http://links.lww.com/CCM/E75), Supplemental Table 9 (Supplemental Digital Content 10, http://links.lww.com/CCM/E76), and Supplemental Table 10 (Supplemental Digital Content 11, http://links.lww.com/CCM/E77). Supplemental Table 11 (Supplemental Digital Content 12, http://links.lww.com/CCM/E78) demonstrates the duration of hospitalization in each steroid strata. The duration of hospitalization and prednisolone dose equivalent of each quartile were listed in Supplemental Table 12 (Supplemental Digital Content 13, http://links.lww.com/CCM/E79). In all the samples, patients in the first (the lowest dose) and second quartiles benefited from steroids in terms of survival to discharge and 1-year survival. However, these benefits began to disappear in the third quartile and reversed in the fourth quartile (the highest dose). Therefore, we created a ROC curve of the third quartile which suggested a steroid dose of 50 mg/d as the optimal cut-off value (Supplemental Fig. 2, Supplemental Digital Content 14, http://links.lww.com/CCM/E80). Thus the steroid group was divided into a low steroid group (≤ 50 mg/d) and a high steroid group (> 50 mg/d). Supplemental Table 13 (Supplemental Digital Content 15, http://links.lww.com/CCM/E81) shows that the low steroid group was associated with lower mortality when compared with the nonsteroid group, and the high steroid group was associated with worse outcomes. The estimations of model internal validity in assessing survival to discharge in matched cohorts are shown in Supplemental Table 14 (Supplemental Digital Content 16, http://links.lww.com/CCM/E82).
The current study evaluated the association between steroids administered during postcardiac arrest care and the outcomes of cardiac arrest survivors by analyzing data from the NHIRD. The NHIRD showed that 22,768 patients had been successfully resuscitated from nontraumatic cardiac arrest during acute hospital care from 2004 to 2011. Our findings demonstrate that steroid use after ROSC may benefit survival to hospital discharge and 1-year survival. To our best knowledge, this study is the first to use nationwide population-based data to assess the independent association between steroid administered during the postcardiac arrest period and the outcomes of cardiac arrest survivors.
The postcardiac arrest syndrome may profit from glucocorticoid administered after ROSC in several aspects in addition to adrenal insufficiency. Glucocorticoids have been reported to decrease oxidative stress (23), reduce apoptosis (24), and thus ameliorate postresuscitation myocardial dysfunction (25) and cerebral injury (26). Glucocorticoid administration following cardiac arrest also attenuates circulating cytokine levels (27) and leukocyte adhesion (28) and ameliorates endotoxin-mediated myocardial dysfunction and hemodynamic instability (29). Furthermore, glucocorticoids also help to maintain cardiovascular stability by inhibiting catecholamine reuptake, enhancing vascular response to vasopressors (11 , 30) and decreasing nitric oxide-mediated vasodilation (31).
However, the question of whether glucocorticoid use during the postcardiac arrest period is beneficial to cardiac arrest survivors has been debated for years. Jastremski et al (12) found that steroid use within 8 hours after ROSC did not improve survival or neurologic recovery in a retrospective study. However, that study was limited by differences in the etiology of cardiac arrest and underlying characteristics between the study groups that potentially favored the nonsteroid group. In another nonrandomized retrospective study of 458 OHCA survivors, no significant differences in survival or neurologic recovery between patients with and without steroid use were identified (13). Donnino et al (14) conducted a randomized, double-blind, placebo-controlled trial, and found that hydrocortisone did not improve the time to shock reversal or clinical outcomes in patients with refractory shock following cardiac arrest. However, there were only 25 patients in each group, and the timing of first steroid dose was not restricted, which raised the question of whether a therapeutic window may exist. In contrast, two prospective, randomized, double-blind, placebo-controlled, parallel-group trials showed that combination therapy with vasopressors and methylprednisolone during CPR, followed by hydrocortisone after ROSC, improved neurologic outcomes, and the rate of survival to hospital discharge. As previously mentioned, multiple interventions given at the same time and obvious differences in ROSC rates between patient groups make it difficult to identify the isolated benefit of steroid administration after ROSC (10 , 11). Different from the limitation of small patient numbers and design features in these clinical studies, the current study investigated the association between postarrest steroid use and outcome of cardiac arrest survivors by analyzing nationwide population-based data. To avoid the influence of steroid use during CPR in the current study, the patients who received steroid during resuscitation were excluded. Also, the variable “steroid use within 1 year prior to cardiac arrest” was matched to attenuate potential observational bias generated by steroid dependence prior to cardiac arrest. Furthermore, after reassessing the effect of postarrest steroid administration in patients with and without steroid use prior to cardiac arrest, all the results showed a consistent benefit of steroids.
In the current study, the dose analysis demonstrated a benefit of steroid administration only on the outcomes of patients who received a low dose of steroid, and this beneficial effect was reversed in patients who received a high dose of steroid. Higher steroid doses were associated with worse outcomes as compared with the nonsteroid group. Some possible reasons for this observation are: 1) steroid usage in some patients with a poor clinical recovery should not be rapidly tapered, 2) some patients in the high steroid group died too soon to be able to taper their steroid use, and 3) increasing steroid dose was associated with decreasing cardiovascular comorbidity burden (CAD, congestive heart failure, CKD) and increasing pulmonary comorbidity burden (COPD and asthma) potentially leading to important confounding in this analysis. Further animal and human studies are required to identify the true effect of a high steroid dose on outcomes of cardiac arrest survivors.
Our current study has several limitations. First, our analysis of the relationship between patient outcomes and the steroid dose administered is limited by the fact that the NHIRD provided the data of the total dose of steroid in a period of hospitalization rather than data on individual doses and their timing. Also, the reason why postarrest steroid was administered was not provided. We attempted to reduce confounding factors by excluding patients who received steroid greater than 1 month and patients receiving pulse steroid therapy. We also standardized the dose equivalent with the duration of hospitalization. Second, we did not adjust therapeutic hypothermia in our outcome analysis, because hypothermia was not covered by the Taiwan NHI system until 2015, and only a few such cases were reported. Besides, the general quality-improvement measures may narrow the difference between groups. Therefore, we attempted to attenuate the potential variance of postcardiac arrest care by matching the steroid and nonsteroid groups for hospital level and the geographic distribution of hospitals. In addition, we also matched epinephrine and shockable rhythm, two important survival-related variables, to reduce the potential confounders. Third, individual patient information, including body weight, smoking history, witnessed collapse, bystander CPR, cause of cardiac arrest, hemodynamics during and after CPR, the reason for steroid use, plasma cortisol and adrenocorticotropic hormone concentrations, causes of death, and neurologic outcomes were not available in the Taiwan NHIRD. Fourth, the effect of unmeasured confounders cannot be controlled using PS matching, such that causality cannot be inferred. Finally, the doses and routes of steroids prior to cardiac arrest were not investigated and worth exploring in the further appropriately designed studies.
Among adult survivors with nontraumatic cardiac arrest during acute hospital care, steroid administration during the postcardiac arrest period may ameliorate survival to discharge and 1-year survival.
1. Adrie C, Adib-Conquy M, Laurent I, et al. Successful cardiopulmonary resuscitation after cardiac arrest
as a “sepsis-like” syndrome. Circulation 2002; 106:562–568
2. Adrie C, Laurent I, Monchi M, et al. Postresuscitation disease after cardiac arrest
: A sepsis-like syndrome? Curr Opin Crit Care 2004; 10:208–212
3. Adrie C, Monchi M, Laurent I, et al. Coagulopathy after successful cardiopulmonary resuscitation following cardiac arrest
: Implication of the protein C anticoagulant pathway. J Am Coll Cardiol 2005; 46:21–28
4. Böttiger BW, Motsch J, Böhrer H, et al. Activation of blood coagulation after cardiac arrest
is not balanced adequately by activation of endogenous fibrinolysis. Circulation 1995; 92:2572–2578
5. Schultz CH, Rivers EP, Feldkamp CS, et al. A characterization of hypothalamic-pituitary-adrenal axis function during and after human cardiac arrest
. Crit Care Med 1993; 21:1339–1347
6. Hékimian G, Baugnon T, Thuong M, et al. Cortisol levels and adrenal reserve after successful cardiac arrest
resuscitation. Shock 2004; 22:116–119
7. Kim JJ, Lim YS, Shin JH, et al. Relative adrenal insufficiency after cardiac arrest
: Impact on postresuscitation disease outcome. Am J Emerg Med 2006; 24:684–688
8. Kim JJ HS, Hwang SY, Jung YB, et al. Hormonal responses upon return of spontaneous circulation after cardiac arrest
: A retrospective cohort study. Crit Care 2011; 15:R53
9. Pene F, Hyvernat H, Mallet V, et al. Prognostic value of relative adrenal insufficiency after out-of-hospital cardiac arrest
. Intensive Care Med 2005; 31:627–633
10. Mentzelopoulos SD, Malachias S, Chamos C, et al. Vasopressin, steroids, and epinephrine and neurologically favorable survival
after in-hospital cardiac arrest
: A randomized clinical trial. JAMA 2013; 310:270–279
11. Mentzelopoulos SD, Zakynthinos SG, Tzoufi M, et al. Vasopressin, epinephrine, and corticosteroids for in-hospital cardiac arrest
. Arch Intern Med 2009; 169:15–24
12. Jastremski M, Sutton-Tyrrell K, Vaagenes P, et al. Glucocorticoid treatment does not improve neurological recovery following cardiac arrest
. Brain Resuscitation Clinical Trial I Study Group. JAMA 1989; 262:3427–3430
13. Grafton ST, Longstreth WT Jr. Steroids after cardiac arrest
: A retrospective study with concurrent, nonrandomized controls. Neurology 1988; 38:1315–1316
14. Donnino MW, Andersen LW, Berg KM, et al; Collaborating Authors from the Beth Israel Deaconess Medical Center’s Center for Resuscitation Science Research Group: Corticosteroid therapy in refractory shock following cardiac arrest
: A randomized, double-blind, placebo-controlled, trial. Crit Care 2016; 20:82
15. Tsai MS, Chuang PY, Yu PH, et al. Glucocorticoid use during cardiopulmonary resuscitation may be beneficial for cardiac arrest
. Int J Cardiol 2016; 222:629–635
16. Smithline H, Rivers E, Appleton T, et al. Corticosteroid supplementation during cardiac arrest
in rats. Resuscitation 1993; 25:257–264
17. Tsai MS, Huang CH, Chang WT, et al. The effect of hydrocortisone on the outcome of out-of-hospital cardiac arrest
patients: A pilot study. Am J Emerg Med 2007; 25:318–325
19. D’Agostino R Jr. Tutorial in Biostatistics. Propensity score
methods for bias reduction in the comparison of a treatment to a non-randomized control group. Statist Med 1988; 17, 2265–2281
20. Parsons LS. Performing a 1: N Case-Control Match on Propensity Score
. 2004In: 2004 SUGI 29 Proceeding 165-29, Seattle, WA, Ovation Research Group.
21. Yang D, Dalton JE. A unified approach to measuring the effect size between two groups using SAS®. 2012Cleveland, OH, SAS Global Forum 2012, Statistics and Data Analysis.
22. Steyerberg EW, Harrell FE Jr, Borsboom GJ, et al. Internal validation of predictive models: Efficiency of some procedures for logistic regression analysis. J Clin Epidemiol 2001; 54:774–781
23. Hall ED. Neuroprotective actions of glucocorticoid and nonglucocorticoid steroids in acute neuronal injury. Cell Mol Neurobiol 1993; 13:415–432
24. Bishopric NH, Andreka P, Slepak T, et al. Molecular mechanisms of apoptosis in the cardiac myocyte. Curr Opin Pharmacol 2001; 1:141–150
25. Wynsen JC, Preuss KC, Gross GJ, et al. Steroid
-induced enhancement of functional recovery of postischemic, reperfused myocardium in conscious dogs. Am Heart J 1988; 116:915–925
26. Katz L, Vaagenes P, Safar P, et al. Brain enzyme changes as markers of brain damage in rat cardiac arrest
model. Effects of corticosteroid therapy. Resuscitation 1989; 17:39–53
27. Buttini M, Mir A, Appel K, et al. Lipopolysaccharide induces expression of tumour necrosis factor alpha in rat brain: Inhibition by methylprednisolone and by rolipram. Br J Pharmacol 1997; 122:1483–1489
28. Simon D, Borradori L, Simon HU. Glucocorticoids in autoimmune bullous diseases: Are neutrophils the key cellular target? J Invest Dermatol 2013; 133:2314–2315
29. Söderberg E, Lipcsey M, Sjölin J, et al. Counteraction of early circulatory derangement by administration of low dose steroid
treatment at the onset of established endotoxemic shock is not directly mediated by TNF-α and IL-6. Steroids 2012; 77:1101–1106
30. Oppert M, Schindler R, Husung C, et al. Low-dose hydrocortisone improves shock reversal and reduces cytokine levels in early hyperdynamic septic shock. Crit Care Med 2005; 33:2457–2464
31. Matsumura M, Kakishita H, Suzuki M, et al. Dexamethasone suppresses iNOS gene expression by inhibiting NF-kappaB in vascular smooth muscle cells. Life Sci 2001; 69:1067–1077
cardiac arrest; propensity score; steroid; survival; Taiwan National Health Insurance Research Database
Supplemental Digital Content
Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.