By Samuel Nadler, MD, PhD

Critical Care, Pulmonary Medicine, The Polyclinic Madison Center, Seattle; Clinical Instructor, University
of Washington, Seattle

SYNOPSIS: Phenylephrine pushes in septic patients were associated with early hemodynamic stability, but higher intensive care unit mortality.

SOURCE: Hawn JM, Bauer SR, Yerke J, et al. Effect of phenylephrine push before continuous infusion norepinephrine in patients with septic shock. Chest 2021;159:1875-1883.

Patients with septic shock frequently are treated with vasopressor infusions to maintain mean arterial pressure (MAP). However, emergent or transient drops in MAP can be treated with “push-dose” vasopressors. In this study, the authors examined the effects of push-dose phenylephrine (PE) on patients with septic shock who also received vasopressor infusions.

A total of 1,317 patients were included who were 18 years of age or older, admitted to an intensive care unit (ICU) requiring continuous norepinephrine (NE) infusion, and met definitions for sepsis. Of these, 181 patients received a push dose of PE between 60 minutes prior and 120 minutes after the initiation of NE infusion. Patients were excluded if they received PE pushes outside this window. A propensity score was used to match patients receiving PE with a control group that did not receive push-dose PE. The primary outcome was hemodynamic stability at three and 12 hours defined as MAP 65 mmHg for six hours without further increases in vasopressor infusions. Secondary outcomes included mortality, length of stay (LOS), duration of mechanical ventilation, and vasopressor need, among other variables.

Primary and secondary outcomes were reported for both propensity score matched and unmatched samples. Other than a higher incidence of chronic obstructive pulmonary disease (COPD) in the PE push group vs. non-PE push group (29.1% vs. 17.4%, respectively), the propensity score matched groups were similar.

In the matched groups, patients receiving push-dose PE had statistically greater hemodynamic stability at three hours (28.4% vs. 18.8%), but not at 12 hours (58.2% vs. 50.0%). Compared with patients not receiving push-dose PE, those patients who did had higher ICU mortality (22.0% vs. 31.2%), hospital mortality (22.5% vs. 39.0%), ICU LOS (9.3 vs. 12.4 days), and vasoactive infusion duration (2.0 vs. 2.6 days), but not hospital LOS, duration of mechanical ventilation, or maximum NE infusion dose. Multivariate-adjusted models demonstrated that PE push dosing was independently associated with higher three-hour hemodynamic stability (odds ratio [OR], 1.8; 95% confidence interval [CI], 1.09-2.97; P = 0.02). However, PE pushes were associated with higher ICU mortality (OR, 1.88; 95% CI, 1.10-3.21; P = 0.02).


This study examined the effect of push-dose PE on patients with septic shock and hemodynamic instability. As with any retrospective cohort study, there is a risk of unknown, confounding variables that explain the observed differences in the two experimental groups. The need for PE pushes might simply be a marker for greater severity of illness. However, the authors used a propensity score to adjust for known variables, and the two groups appeared to be well-matched.

Notably, Sequential Organ Failure Assessment (SOFA) scores in both unmatched and propensity matched cohorts were similar, and the changes in SOFA score over 72 hours were nearly identical, supporting similar severity of illness between the groups. Furthermore, in the propensity matched groups, lactate levels, total fluid administration, and corticosteroid use also were similar.

The primary outcome in this study was hemodynamic stability at three and 12 hours. Push-dose PE was associated with greater hemodynamic stability at three hours, but not at 12 hours. Although not statistically significant, PE dosing tended to decrease the time to hemodynamic stability within three hours (P = 0.06).

This raises the question: What is the most appropriate outcome? Phenylephrine rapidly supports MAP by increasing cardiac afterload. However, this increase in MAP may not correlate with greater cardiac output and improved perfusion in septic patients with myocardial dysfunction. Indeed, when PE infusions were substituted for NE infusions for the treatment of sepsis during a shortage of NE, mortality worsened.1 Thus, the increase in mortality observed in patients with PE pushes despite improved hemodynamic stability implies PE may be the wrong agent to counteract transient hypotension.

Although push-dose phenylephrine improved hemodynamic numbers, the number of patients who survived the ICU seemed to be less. Caution should be undertaken when considering push-dose PE for transient hypotension in patients with septic shock. 


  1. Vail E, Gershengorn HB, Hua M, et al. Association between US norepinephrine shortage and mortality among patients with septic shock. JAMA 2017;317:1433-1442.