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

Abstract

Full text

Correction: Mentzelopoulos and Chalkias Critical Care (2024) 28:191 10.1186/s13054-024-04962-8

Following publication of the original article [1], the authors identified an error within row 7 of Table 2. In Table 2, row 7, the lowest percentages of postresuscitation hypotension (i.e. 17% and 15%) actually correspond to the intervention group(s) and the highest (i.e. 28% and 29%) to control.

Table 2 row 7 currently reads:Lowest MAP ≤ 50 mmHg and SAP ≤ 80 mmHg, intervention group(s) versus control (%)28% versus 17%—P = 0.12 and 29% versus 15%; P = 0.03^d^NR but significant difference unlikely^c^

Table 2. Key differences between the Greek VSE trials and the Danish VAM IHCA trialKey characteristicVSE 1 and 2—pooled dataVAM-IHCA (Danish trial)Potential effectTime to study drugs, median (IQR)—min4 (3–5)9 (6–12)Earlier, simultaneous activation of V1A-vasopressin and α1-adrenergic receptors, with maximization of pressor effect and associated probability of prompt ROSC,^a^ likely occurring only in intervention groups (vs. control) of the VSE 1 and 2 RCTsEpinephrine and Vasopressin started and always given simultaneouslyYesNoMedian time lag between first dose of epinephrine and study drugs (min)03–4Significantly shorter median (IQR) time to ROSC (min) in intervention group(s) versus controlYes; 14 (7–24) versus 20 (10–30)—P < 0.001No; 16 (12–25) versus 18 (11–31)—P-value NRAttenuated ischemia / reperfusion injury and lower risk of epinephrine-related adverse effects likely present only in intervention groups (vs. control) of the VSE 1 and 2 RCTsSignificantly lower median (IQR) total dose (mg) of epinephrine in intervention group(s) versus controlYes; 4 (2–6) versus 5 (3–9)—P < 0.001No; 3 (2–5) versus 3 (2–5)—P-value NRSAP < 90 mmHg within 20 min of ROSC, intervention group(s) versus control (%)19% versus 44%—P < 0.001^b^NR but significant difference unlikely ^c^Early postresuscitation hypotension has been consistently associated with increased in-hospital mortality [11]; reported differences may partly explain the long-term benefit observed only in the VSE 1 and 2 RCTsLowest MAP ≤ 50 mmHg and SAP ≤ 80 mmHg, intervention group(s) versus control (%)28% versus 17%—P = 0.12 and 29% versus 15%; P = 0.03^d^NR but significant difference unlikely^c^Postresuscitation use of steroids in intervention group(s) versus control (%)99% versus 22% ^d^26% versus 46% ^e^In VAM-IHCA, there was non-protocolized, more frequent use of potentially beneficial interventions in the control groupPostresuscitation use of ECMO in intervention group(s) versus control (%)0% versus 0% ^d,f^14% versus 30% ^e^VSE, vasopressin-steroids-epinephrine; VAM, vasopressin-adrenaline-methylprednisolone; IHCA, in-hospital cardiac arrest; IQR, interquartile range; ROSC, return of spontaneous circulation; NR, not reported; RCT, randomized clinical trial; SAP, systolic arterial pressure; MAP, mean arterial pressure; ECMO, extracorporeal membrane oxygenationReported data originate from the total of the VSE 1 and 2 or VAM-IHCA participants, unless otherwise specified. For data reported as median (IQR), P-values were determined using the Mann Whitney U test. For data reported as percentages, P-values were determined by Fisher’s exact test^a^, Defined as ROSC after ≤ 3 vasopressor doses^b^, VSE 1 and 2 data originate from the pooled subgroup of survivors for ≥ 4 h with postresuscitation shock and available SAP data (intervention, n = 94; control, n = 72)^c^, Analyses of VAM-IHCA hemodynamic data revealed very similar, early postresuscitation arterial pressures and vasopressor support [2]; consequently, there was likely no significant, between-group difference in the frequency of postresuscitation hypotension^d^, VSE 1 and 2 data originate from the pooled subgroup of survivors for ≥ 24 h (intervention, n = 94; control, n = 69)^e^, VAM-IHCA data originate from survivors for ≥ 24 h (intervention, n = 63; control, n = 61)^f^, There was no postresuscitation use of ECMO in the VSE 1 and 2 studies]

Table 2 row 7 should read:Lowest MAP ≤ 50 mmHg and SAP ≤ 80 mmHg, intervention group(s) versus control (%)17% versus 28%—P = 0.12 and 15% versus 29%; P = 0.03^d^NR but significant difference unlikely^c^

Table 2. Key differences between the Greek VSE trials and the Danish VAM IHCA trialKey characteristicVSE 1 and 2—pooled dataVAM-IHCA (Danish trial)Potential effectTime to study drugs, median (IQR)—min4 (3–5)9 (6–12)Earlier, simultaneous activation of V1A-vasopressin and α1-adrenergic receptors, with maximization of pressor effect and associated probability of prompt ROSC,^a^ likely occurring only in intervention groups (vs. control) of the VSE 1 and 2 RCTsEpinephrine and Vasopressin started and always given simultaneouslyYesNoMedian time lag between first dose of epinephrine and study drugs (min)03–4Significantly shorter median (IQR) time to ROSC (min) in intervention group(s) versus controlYes; 14 (7–24) versus 20 (10–30)—P < 0.001No; 16 (12–25) versus 18 (11–31)—P-value NRAttenuated ischemia / reperfusion injury and lower risk of epinephrine-related adverse effects likely present only in intervention groups (vs. control) of the VSE 1 and 2 RCTsSignificantly lower median (IQR) total dose (mg) of epinephrine in intervention group(s) versus controlYes; 4 (2–6) versus 5 (3–9)—P < 0.001No; 3 (2–5) versus 3 (2–5)—P-value NRSAP < 90 mmHg within 20 min of ROSC, intervention group(s) versus control (%)19% versus 44%—P < 0.001^b^NR but significant difference unlikely ^c^Early postresuscitation hypotension has been consistently associated with increased in-hospital mortality [11]; reported differences may partly explain the long-term benefit observed only in the VSE 1 and 2 RCTsLowest MAP ≤ 50 mmHg and SAP ≤ 80 mmHg, intervention group(s) versus control (%)17% versus 28%—P = 0.12 and 15% versus 29%; P = 0.03^d^NR but significant difference unlikely^c^Postresuscitation use of steroids in intervention group(s) versus control (%)99% versus 22% ^d^26% versus 46% ^e^In VAM-IHCA, there was non-protocolized, more frequent use of potentially beneficial interventions in the control groupPostresuscitation use of ECMO in intervention group(s) versus control (%)0% versus 0% ^d,f^14% versus 30% ^e^VSE, vasopressin-steroids-epinephrine; VAM, vasopressin-adrenaline-methylprednisolone; IHCA, in-hospital cardiac arrest; IQR, interquartile range; ROSC, return of spontaneous circulation; NR, not reported; RCT, randomized clinical trial; SAP, systolic arterial pressure; MAP, mean arterial pressure; ECMO, extracorporeal membrane oxygenationReported data originate from the total of the VSE 1 and 2 or VAM-IHCA participants, unless otherwise specified. For data reported as median (IQR), P-values were determined using the Mann Whitney U test. For data reported as percentages, P-values were determined by Fisher’s exact test^a^, Defined as ROSC after ≤ 3 vasopressor doses^b^, VSE 1 and 2 data originate from the pooled subgroup of survivors for ≥ 4 h with postresuscitation shock and available SAP data (intervention, n = 94; control, n = 72)^c^, Analyses of VAM-IHCA hemodynamic data revealed very similar, early postresuscitation arterial pressures and vasopressor support [2]; consequently, there was likely no significant, between-group difference in the frequency of postresuscitation hypotension^d^, VSE 1 and 2 data originate from the pooled subgroup of survivors for ≥ 24 h (intervention, n = 94; control, n = 69)^e^, VAM-IHCA data originate from survivors for ≥ 24 h (intervention, n = 63; control, n = 61)^f^, There was no postresuscitation use of ECMO in the VSE 1 and 2 studies]

Table 2 has been updated in this correction and the original article [1] has been corrected.