Open Access Highly Accessed Open Badges Research

Timing of vasopressor initiation and mortality in septic shock: a cohort study

Vance Beck1, Dan Chateau2, Gregory L Bryson1, Amarnath Pisipati3, Sergio Zanotti4, Joseph E Parrillo5, Anand Kumar36* and The Cooperative Antimicrobial Therapy of Septic Shock (CATSS) Database Research Group

Author Affiliations

1 Department of Anesthesiology, Ottawa Hospital/University of Ottawa, 501 Smyth Road, Ottawa, ON K1H-8L6, Canada

2 Department of Community Health Sciences, University of Manitoba, S113 Medical Services Bldg, 750 Bannatyne Ave, Manitoba, MB R3E-0W3, Canada

3 Department of Medical Microbiology, University of Manitoba, Room 543 Basic Medical Sciences Bldg, 745 Bannatyne Ave, Manitoba, MB R3E-0J9, Canada

4 Section of Critical Care Medicine, Cooper University Hospital, Rowan School of Medicine, 1 Cooper Place, Camden, NJ 08103, USA

5 Department of Medicine, Hackensack Medical Center, Rutgers New Jersey Medical School, 30 Prospect St., Newark, NJ 07601, USA

6 Section of Critical Care Medicine, JJ-399, Health Sciences Centre, University of Manitoba, 700 William Ave, Manitoba, MB R3A-1R9, Canada

For all author emails, please log on.

Critical Care 2014, 18:R97  doi:10.1186/cc13868

Published: 12 May 2014



Despite recent advances in the management of septic shock, mortality remains unacceptably high. Earlier initiation of key therapies including appropriate antimicrobials and fluid resuscitation appears to reduce the mortality in this condition. This study examined whether early initiation of vasopressor therapy is associated with improved survival in fluid therapy-refractory septic shock.


Utilizing a well-established database, relevant information including duration of time to vasopressor administration following the initial documentation of recurrent/persistent hypotension associated with septic shock was assessed in 8,670 adult patients from 28 ICUs in Canada, the United States of America, and Saudi Arabia. The primary endpoint was survival to hospital discharge. Secondary endpoints were length of ICU and hospital stay as well as duration of ventilator support and vasopressor dependence. Analysis involved multivariate linear and logistic regression analysis.


In total, 8,640 patients met the definition of septic shock with time of vasopressor/inotropic initiation documented. Of these, 6,514 were suitable for analysis. The overall unadjusted hospital mortality rate was 53%. Independent mortality correlates included liver failure (odds ratio (OR) 3.46, 95% confidence interval (CI), 2.67 to 4.48), metastatic cancer (OR 1.63, CI, 1.32 to 2.01), AIDS (OR 1.91, CI, 1.29 to 2.49), hematologic malignancy (OR 1.88, CI, 1.46 to 2.41), neutropenia (OR 1.78, CI, 1.27 to 2.49) and chronic hypertension (OR 0.62 CI, 0.52 to 0.73). Delay of initiation of appropriate antimicrobial therapy (OR 1.07/hr, CI, 1.06 to 1.08), age (OR 1.03/yr, CI, 1.02 to 1.03), and Acute Physiology and Chronic Health Evaluation (APACHE) II Score (OR 1.11/point, CI, 1.10 to 1.12) were also found to be significant independent correlates of mortality. After adjustment, only a weak correlation between vasopressor delay and hospital mortality was found (adjusted OR 1.02/hr, 95% CI 1.01 to 1.03, P <0.001). This weak effect was entirely driven by the group of patients with the longest delays (>14.1 hours). There was no significant relationship of vasopressor initiation delay to duration of vasopressor therapy (P = 0.313) and only a trend to longer duration of ventilator support (P = 0.055) among survivors.


Marked delays in initiation of vasopressor/inotropic therapy are associated with a small increase in mortality risk in patients with septic shock.