A critical reappraisal of vasopressin and steroids in in-hospital cardiac arrest

Epinephrine during resuscitation

Advanced life support (ALS) objectives include maximization of coronary perfusion pressure (CPP) for prompt return of spontaneous circulation (ROSC) and minimization of cardiac arrest-associated ischemia–reperfusion injury. Epinephrine, the standard ALS vasopressor, improves ROSC rate, with uncertain effect on neurological outcome [1]. Epinephrine efficacy is limited by its potential to cause arrhythmias, myocardial ischemic contracture, and cerebral microcirculatory dysfunction [2].

Combined stress-hormone approaches

Effectiveness of stress-hormone interventions may depend on timely administration and/or dose for prompt onset of action, and/or use of concurrently administered combinations. In a recent, two-center, randomized clinical trial (RCT) of in-hospital cardiac arrest (IHCA; participants, n = 184) [3], we assessed the effect of methylprednisolone 40 mg, or placebo (plus repeated 1-mg epinephrine) during ALS, followed by postresuscitation hydrocortisone (240 mg/day for 7 days maximum and gradual taper) or placebo on several physiological and long-term outcomes. Neutral results, primarily on early post-ROSC arterial pressure and inflammatory response, strongly implied a resistance to previously well-documented, circulatory and immunomodulatory effects of steroids [3]. Nevertheless, single, high-dose (i.e., 250 mg) methylprednisolone within 5–30 min following successful out-of-hospital ALS has been recently associated with lower postresuscitation interleukin-6, improved postresuscitation hemodynamics and higher probability of survival to discharge [4, 5]; these benefits might reflect a rapid, nongenomic, high-dose methylprednisolone action [4].

A major characteristic of our IHCA-steroids RCT was prolonged median ALS duration, i.e. 25–27 min [3]. Notably, pooled data analyses (n = 368) from our two, prior vasopressin-steroids-epinephrine (VSE) RCTs [6, 7] revealed median ALS duration of just 14 min in VSE patients, as opposed to 20 min in controls [3]. In these RCTs, we added up to five doses of 20-IU vasopressin to epinephrine during ALS to maximize CPP/expedite ROSC by concurrent stimulation of V1A vasopressin and alpha-1 adrenergic receptors [2, 6, 7]; stress-dose steroids were also given during and after ALS for their hypertensive/anti-inflammatory effects. Triple stress-hormone intervention resulted in higher ROSC-rates and postresuscitation arterial pressure, lower serum cytokine concentrations, more organ failure-free days, and improved long-term outcomes [6, 7].

A follow-up, Danish RCT (n = 501) testing vasopressin, adrenaline and methylprednisolone (VAM) during ALS (without postresuscitation steroids) reported improved ROSC-rate and neutral results on long-term outcomes [8]. The results of an individual patient-data meta-analysis (IPDMA) including all 3 RCTs were inconclusive [9], supporting a suggestion against VSE/VAM in IHCA [10]. However, following correction of misclassification of a VSE-2 study participant [7], IPDMA’s adjusted odds ratio (aOR) (95% confidence interval (CI) for cerebral performance category (CPC) score ≤ 2 at discharge rose to 1.80 (1.08–3.01), in favor of VSE/VAM [9].

Correction-associated, main changes in IPDMA results are summarized in Table 1. Key pertinent messages include (1) frailty/fragility of results’ positivity, depending on minor changes in a small absolute difference of approximately 4% in favorable neurological outcome; and (2) lack of power of included RCTs to detect differences in favorable neurological outcome at a level of 4–5%.

Table 1 Correction-induced changes in an IPDMA of three RCTs of vasopressin and steroids in cardiac arrest

Major, corrected IPDMA results were primarily driven by our RCTs [9], which had key differences from the Danish trial [2], as further detailed in Table 2. Regarding time-to-study drugs (T_DRUG), IPDMA data correction revealed significant effect measure modification, with decremental T_DRUG of ≤ 6 min, favoring VSE/VAM as regards survival to discharge and CPC score ≤ 2 at discharge [9]. Danish trial subgroup point estimates for both ROSC and 30-day survival/neurological outcome were also favorable for VAM (ranging within 1.17–1.46) at T_DRUG ≤ 8 min [8], implying a T_DRUG-dependent, favorable response to VAM in 251/501 (50%) of study participants.

Table 2 Key differences between the Greek VSE trials and the Danish VAM IHCA trial

Multi-level VSE effects in IHCA

Transcriptional signaling by glucocorticoids is limited by proteasome degradation of the phosphorylated (oxidized) glucocorticoid receptor (GR) [3, 12]. During ischemia and reperfusion, proteasome mediates removal of oxidized, intracellular proteins [12]. Furthermore, the longer the duration of ischemia, the greater the degradation of adenosine 5′-triphosphate and intracellular accumulation of hypoxanthine [13]. During subsequent reperfusion, hypoxanthine is reconverted to xanthine, by xanthine oxidase, with concurrent production of toxic, reactive oxygen species [12]. Consequently, prolonged ischemia time (followed by reperfusion) is associated with greater, intracellular, oxidative stress [13], and likely, more extensive oxidation and loss of function [12] of various proteins, including the GR. Accordingly, a recent study reported rapid, post-ROSC decline in B/T lymphocyte GR-expression [14].

In our trials [6, 7], VSE patients received post-ROSC stress-dose hydrocortisone (300 mg/day for 7 days maximum and gradual taper). Furthermore, the shorter "low-flow/ischemia" time might have mitigated the ischemia–reperfusion induced oxidation/proteasome degradation of GR, with consequent preservation of vasopressor and anti-inflammatory effects of steroids. These two pharmaco-physiological factors explain the decreased frequency of potentially detrimental, early, postresuscitation hypotension in VSE patients (Table 2). Also, in our trials, steroid treatment was associated with lower postresuscitation cytokine concentrations [6]. This indicates attenuation of cardiac arrest-associated systemic inflammatory response, and partly explains the lesser organ dysfunction in VSE groups [6, 7].

Additional key facts supporting VSE use in IHCA

IHCA nonshockable rhythms’ incidence and long-term outcomes are comparably high and poor (respectively) across Registry studies and control groups of VSE/VAM RCTs [Table 3; 6–8; Additional file 1]. Relatively minor differences can be partly explained by potentially more frequent pseudo-pulseless electrical activity (PEA) (which has better prognosis) in studies with high, overall incidence of PEA; also, in Registry studies, epinephrine was not given to some patients and this was associated with improved survival (Table 3; Additional file 1). Thus, our combined VSE 1 and 2 group results of nonshockable rhythms’ survival to discharge and CPC ≤ 2 at discharge of 17% and 14% (respectively) [6, 7] suggest VSE benefit in the large subgroup of nonshockable IHCA. Accordingly, corrected IPDMA aOR (95%CI) of VSE/VAM versus control for CPC score ≤ 2 at discharge was 2.02 (1.11–3.67) [9].

Table 3 Comparative presentation of frequency/outcomes of nonshockable presenting rhythms across studies, and of do-not-resuscitate practices across countries of study conduct

Regarding VSE practicability, vasopressin/methylprednisolone physical/chemical stability in normal saline solutions has been previously confirmed [6]. Routinely using prefilled syringes for prompt VSE-administration is feasible/effective [6, 7].

Conclusion

In view of the above-presented discussion/evidence and until publication of new evidence from a large, ongoing Swedish RCT (www.clinicaltrials.gov/study/NCT05139849), we suggest that VSE [6, 7] might be considered in IHCA [15]. However, the frailty/fragility of the corrected meta-analysis results needs to be considered.