Correction to: Intensive Care Med (2018) 44:847–856 https://doi.org/10.1007/s00134-018-5222-9

Because of a technical error, the code corresponding to the outcome for the Basir et al. cohort was mis-implemented in the original version of our article. Characteristics of the cohort are in fact the followings:

Observational studies

Inclusion period

Cause of cardiogenic shock

Single/multicenter

Number of patients (n)

Epinephrine-treated patients n (%)

Death n (%)

Mortality endpoint

ECLS (n)

Basir et al., 2018 [1]

2016

Acute coronary syndrome

Single

41

8 (18)

10 (24)

In-hospital

Yes (27)

Yet, this has a slight impact on the estimation of the association of epinephrine with the outcome for this cohort and subsequently to the entire cohort. More precisely, the ORs (with 95% confidence interval) have changed as depicted in the table below:

 

Original version

After erratum

Basir et al. [1]

0.96 (0.16–5.73)

1.04 (0.17–6.22)

All studies

3.33 (2.81–3.94)

3.42 (2.89–4.05)