Possible significance of hemodynamic and immunomodulatory effects of early stress-dose steroids in cardiac arrest

In an interesting randomized clinical trial (RCT), Donnino et al. [1] studied a mixed out-of-hospital cardiac arrest and in-hospital cardiac arrest (IHCA) population and found no hydrocortisone versus placebo hemodynamic or in-hospital outcome benefit. In the hydrocortisone group, the median time to study intervention was 9.9 h after return of spontaneous circulation (ROSC) [1]. This time lag probably exceeds the therapeutic window for the prevention of detrimental episodes of early post-resuscitation hypotension [2] through a mean arterial pressure (MAP)-stabilizing effect of steroids [3, 4]. 
 
Analyses of pooled post-resuscitation shock data from our IHCA vasopressin-steroids-epinephrine (VSE) RCTs [3, 4] also showed no between-group differences in the time to, or proportions of, discontinuation of vasopressors, and post-ROSC day 1 hemodynamic support (Table 1). However, VSE patients had higher, early post-ROSC systolic arterial pressure (SAP) and MAP during post-resuscitation follow-up [3, 4]. This reflected an improved hemodynamic response to similar vasopressor support titrated to a “wide” MAP range of 70–100 mmHg [4]. 
 
 
 
Table 1 
 
Pooled results (from [3] and [4]) on early post-enrollment hemodynamics in survivors for ≥4 h with post-resuscitation shock 
 
 
 
Recordings of “early post-ROSC SAP >90 mmHg” (i.e., “absence of early post-resuscitation hypotension” [2]) and “≥1 recorded/analyzed, day 1 MAP value of >80 mmHg [2]” were significantly more frequent in VSE patients than controls. Importantly, such SAP/MAP levels corresponded to more frequent survival to hospital discharge with favorable neurological outcome [4] (Table 1). 
 
Early post-resuscitation hemodynamics of VSE patients could be partly attributable to the steroids-vasopressin combination during cardiopulmonary resuscitation (CPR) [3, 4]. However, a previously postulated major CPR-VSE effect, i.e., shorter advanced life support duration [4], possibly leading to attenuated post-resuscitation cardiovascular dysfunction was not clear in the current subgroup analysis (Table 1). Hence, according to the short (i.e., 24 min) half-life of vasopressin, we propose that the more frequent day 1 MAP >80 mmHg was largely due to a post-ROSC steroid-induced augmentation of vascular responsiveness to vasopressors [3, 4]. A mediation analysis of VSE outcome benefit through day 1 MAP is warranted. Analysis of day 1 MAP data from the study by Donnino et al. might causally link between-RCT differences in corticosteroid timing with differences in survival/neurological outcome results [1, 3, 4]. 
 
Post-resuscitation disease is a “sepsis-like” syndrome. In sepsis, acute kidney injury severity is associated with mortality and elevated interleukin (IL)-6. Furthermore, high post-ROSC IL-6 is associated with organ dysfunction and poor long-term outcomes [5]. Notably, post-resuscitation hydrocortisone has been associated with reduced IL-6 levels [1, 3], and VSE patients versus controls had more renal failure-free days [3, 4]. 
 
Conclusively, available evidence prompts toward further evaluation of early, stress-dose steroids in cardiac arrest.

In an interesting randomized clinical trial (RCT), Donnino et al. [1] studied a mixed out-of-hospital cardiac arrest and in-hospital cardiac arrest (IHCA) population and found no hydrocortisone versus placebo hemodynamic or in-hospital outcome benefit. In the hydrocortisone group, the median time to study intervention was 9.9 h after return of spontaneous circulation (ROSC) [1]. This time lag probably exceeds the therapeutic window for the prevention of detrimental episodes of early post-resuscitation hypotension [2] through a mean arterial pressure (MAP)stabilizing effect of steroids [3,4].
Analyses of pooled post-resuscitation shock data from our IHCA vasopressin-steroids-epinephrine (VSE) RCTs [3,4] also showed no between-group differences in the time to, or proportions of, discontinuation of vasopressors, and post-ROSC day 1 hemodynamic support (Table 1). However, VSE patients had higher, early post-ROSC systolic arterial pressure (SAP) and MAP during post-resuscitation follow-up [3,4]. This reflected an improved hemodynamic response to similar vasopressor support titrated to a "wide" MAP range of 70-100 mmHg [4].
Early post-resuscitation hemodynamics of VSE patients could be partly attributable to the steroids-vasopressin combination during cardiopulmonary resuscitation (CPR) [3,4]. However, a previously postulated major CPR-VSE effect, i.e., shorter advanced life support duration [4], possibly leading to attenuated post-resuscitation cardiovascular dysfunction was not clear in the current subgroup analysis (Table 1). Hence, according to the short (i.e., 24 min) half-life of vasopressin, we propose that the more frequent day 1 MAP >80 mmHg was largely due to a post-ROSC steroid-induced augmentation of vascular responsiveness to vasopressors [3,4]. A mediation analysis of VSE outcome benefit through day 1 MAP is warranted. Analysis of day 1 MAP data from the study by Donnino et al. might causally link between-RCT differences in corticosteroid timing with differences in survival/neurological outcome results [1,3,4].
Post-resuscitation disease is a "sepsis-like" syndrome. In sepsis, acute kidney injury severity is associated with mortality and elevated interleukin (IL)-6. Furthermore, high post-ROSC IL-6 is associated with organ dysfunction and poor long-term outcomes [5]. Notably, postresuscitation hydrocortisone has been associated with reduced IL-6 levels [1,3], and VSE patients versus controls had more renal failure-free days [3,4].
Conclusively, available evidence prompts toward further evaluation of early, stress-dose steroids in cardiac arrest.
Abbreviations CPR, cardiopulmonary resuscitation; IHCA, in-hospital cardiac arrest; IL, interleukin; MAP, mean arterial pressure; RCT, randomized clinical trial; ROSC, return of spontaneous circulation; SAP, systolic arterial pressure; VSE, vasopressinsteroids-epinephrine Availability of data and materials For the purpose of the above-mentioned re-analysis protocol (Clinicaltrials.gov identifier, NCT02408939), we extracted individual peri-arrest and follow-up data from survivors for ≥4 h with post-resuscitation shock (n = 191) from an electronic masterfile containing de-identified data from references [3] and [4]. Extracted, de-identified data was saved in a Microsoft Excel datafile. Data will not be shared because we plan to use it in a future mediation analysis mentioned in the fourth paragraph of the current main text.
Authors' contributions SDM is responsible for the conception and drafting of the manuscript, conduct and accuracy of the data analyses, and interpretation of the results. NM, TX, and SGZ contributed to the interpretation of the results, and to critically important revisions of the original draft. All authors read and approved the final manuscript.

Competing interests
The authors declare that they have no competing interests.