Journal Pre-proof

Delayed vasopressor initiation is associated with increased

mortality in patients with septic shock

Daniel Colon Hidalgo, Jaimini Patel, Dalila Masic, David Park,

Megan A. Rech

PII: S0883-9441(19)30911-6
DOI: https://doi.org/10.1016/j.jcrc.2019.11.004
Reference: YJCRC 53416

To appear in: Journal of Critical Care

Please cite this article as: D.C. Hidalgo, J. Patel, D. Masic, et al., Delayed vasopressor
initiation is associated with increased mortality in patients with septic shock, Journal of
Critical Care(2018), https://doi.org/10.1016/j.jcrc.2019.11.004

This is a PDF file of an article that has undergone enhancements after acceptance, such
as the addition of a cover page and metadata, and formatting for readability, but it is
not yet the definitive version of record. This version will undergo additional copyediting,
typesetting and review before it is published in its final form, but we are providing this
version to give early visibility of the article. Please note that, during the production
process, errors may be discovered which could affect the content, and all legal disclaimers
that apply to the journal pertain.

© 2018 Published by Elsevier.

 Journal Pre-proof

Delayed vasopressor initiation is associated with increased mortality in patients with septic

shock

Authors:

Daniel Colon Hidalgo, MD1

daniel.colonhidalgo@lumc.edu

Jaimini Patel, PharmD2

jaiminip@umich.edu

Dalila Masic, PharmD2

dalila.masic@luhs.org

of
David Park, MD1
david.park@lumc.edu

ro
Megan A. Rech, PharmD, MS, BCPS, BCCCP2,3
mrech@lumc.edu
-p
re
1
Department of Medicine, Loyola University Medical Center, 2160 S 1st Ave Maywood IL, USA
2
Department of Pharmacy, Loyola University Medical Center, 2160 S 1st Ave Maywood IL USA
3
Department of Emergency Medicine, Stritch School of Medicine, Loyola University Chicago,
lP

2160 S 1st Ave Maywood IL, USA

na
ur
Jo

Corresponding Author: Megan A. Rech, PharmD, MS, BCPS, BCCCP, Loyola University

Medical Center, Department of Pharmacy. 2160 S. First Ave, Maywood, IL 60153, USA.

Email: mrech@lumc.edu

This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
 Journal Pre-proof

Abstract

Purpose: Mortality rate for septic shock, despite advancements in knowledge and treatment,

remains high. Treatment includes administration of broad-spectrum antibiotics and stabilization

of the mean arterial pressure (MAP) with intravenous fluid resuscitation. Fluid-refractory shock

warrants vasopressor initiation. There is a paucity of evidence regarding the timing of

vasopressor initiation and its effect on patient outcomes.

Materials and methods: This retrospective, single-centered, cohort study included patients with

of
septic shock from January 2017 to July 2017. Time from initial hypotension to vasopressor

initiation was measured for each patient. The primary outcome was 30-day mortality.

ro
Results: Of 530 patients screened,119 patients were included. There were no differences in

-p
baseline patient characteristics. Thirty-day mortality was higher in patients who received
re
vasopressors after 6 hours (51.1% vs 25%, p<0.01). Patients who received vasopressors within

the first 6 hours had more vasopressor-free hours at 72 hours (34.5 hours vs 13.1, p=0.03) and
lP

shorter time to MAP of 65 mmHg (1.5 hours vs 3.0, p<0.01).

na

Conclusion: Vasopressor initiation after 6 hours from shock recognition is associated with a

significant increase in 30-day mortality. Vasopressor administration within 6 hours was

ur

associated with shorter time to achievement of MAP goals and higher vasopressor-free hours
Jo

within the first 72 hours.

Keywords: Vasopressor; septic shock; sepsis; norepinephrine; time; mortality.

 Journal Pre-proof

Introduction

Significant advancements have been made in the management of septic shock in recent

years. Despite increasing knowledge of this syndrome, mortality rates remain high.1 Septic

shock is characterized by profound systemic vasodilation, resulting in inadequate tissue

perfusion, multi-organ dysfunction, and ultimately organ failure.2 The 2016 Surviving Sepsis

Campaign (SSC) guidelines recommend implementing a strategic sepsis bundle to guide timely

and appropriate interventions such as prompt administration of antibiotics within one hour of

of
sepsis onset and aggressive fluid resuscitation with 30 ml/kg of crystalloids.1 Bundles at three

and six hours allow for structured re-evaluation of a patient’s response to initial therapy. The

ro
need for initiation of vasopressor therapy is assessed if there is ongoing hemodynamic

-p
instability despite adequate fluid resuscitation. In practice, initiation of vasopressor agents can
re
occur at varied time points in a patient’s course. In a rapidly deteriorating patient, vasopressor

therapy may be initiated concomitantly with fluid therapy in an effort to stabilize hemodynamics
lP

to a mean arterial pressure (MAP) of greater than 65 mmHg, which is an accepted marker of

adequate perfusion.2 Despite this guidance, there is limited data elucidating how timing of
na

vasopressor initiation affects patient outcomes.

ur

A variety of hemodynamic and resuscitative endpoints are assessed in patients with

Jo

septic shock. A retrospective study evaluating outcomes associated with hemodynamic

variables in septic shock found that MAP < 65 mmHg and mixed venous oxygen saturation

(SvO2) < 70% are associated with increased mortality.3 Another retrospective study including

6,514 patients found a small increase in mortality in patients who received delayed vasopressor

therapy; however, the increase in mortality was driven by the decile group of patients that

received vasopressors after 14 hours, which is uncommon in routine practice.4 Other studies

have found similar outcomes, including mortality benefit, when vasopressors were administered

early in septic shock.5,6 From the available data, it is apparent that early administration of

vasopressor agents may be associated with positive outcomes in septic shock. However, the
 Journal Pre-proof

optimal vasopressor initiation timeframe remains unknown. The purpose of this study is to

evaluate the impact of the recommendation by the Surviving Sepsis Campaign that

vasopressors should be administered within the first 6 hours of sepsis care on 30-day mortality.

Material and methods

This was a retrospective cohort study that examined septic shock patients at the Loyola

University Medical Center. The study was approved by the Institutional Review Board. Patients

of
admitted between January 1, 2017, and July 30, 2017, for whom vasopressors were ordered

were reviewed. This initial group was identified as patients with shock. Then, patients’ charts

ro
were reviewed to assess if patients met criteria for septic shock as per the sepsis-3 definition.7

-p
Patients ≥18 years with septic shock were included. Exclusion criteria were as follows:
re
vasopressor use for other indications other than septic shock (i.e. other shock states),

pregnancy, burn-related injury, post-operative cardiac surgery, those with MAP goal ≥70 mmHg,
lP

vasopressor initiation in the operating room, outside hospital transfers, and those patients for
na

whom care was withdrawn or who died within 72 hours of vasopressor initiation. Patients who

died within 72 hours were excluded in an attempt to determine that mortality due to vasopressor
ur

delay rather than the severity of sepsis.

Jo

The primary outcome was 30-day mortality. Secondary outcomes included cumulative

vasopressor dosing at 12, 24, and 72 hours; vasopressor doses were expressed as

norepinephrine-equivalents based on previous literature.8 Vasopressor-free hours at 72 hours,

time to MAP greater than 65 mmHg, hospital and ICU length of stay (LOS) were also compared

between groups.

Baseline characteristics including age, sex, body mass index (BMI), race, ethnicity,

comorbidities, the presumed site of infection, baseline MAP (defined as the lowest MAP during

the first six hours of resuscitation), and sequential organ failure assessment (SOFA) scores

were also recorded. Time to vasopressors was defined as the time between the first recorded
 Journal Pre-proof

MAP that was less than 65 mmHg and the time when vasopressors were initiated. Patients were

divided into two groups: time-to-vasopressor ≤6hrs vs >6hrs.

Baseline characteristics were expressed as means (± standard deviation) or median

(interquartile range) for continuous variables or as proportions for categorical variables. The

Shapiro-Wilk test was utilized to determine the normality of continuous data. Normally

distributed continuous variables were analyzed using a t-test and non-normally distributed

continuous variables using the Mann-Whitney U test. Fisher’s exact test was used to analyze

of
categorical variables.

Multivariable logistic regression was performed to determine predictors for 30-day

ro
mortality and to account for potential confounders. Candidate variables were selected from the

-p
univariate analysis based on a p-value ≤ 0.2 or those that were thought to be clinically relevant.9
re
The data were analyzed using Microsoft Excel and SPSS (Version 25.0, IBM Corp, Armonk,

NY).
lP
na

Results

Of the 530 patients that received vasopressors during the study period, a total of 119
ur

(22.4%) patients were included in the analysis. The most common reason for exclusion was
Jo

vasopressors used for an indication other than septic shock (n=100; Figure 1). Of those

included, 76 patients (63.9%) received vasopressors within 6 hours from initial hypotension and

43 patients (36.1%) received vasopressors after 6 hours from initial hypotension. All

demographic characteristics, comorbidities, SOFA scores and intravenous fluid administrated

were well-matched among groups except the prevalence of cirrhosis (14.4% vs 27.9%, p=0.09)

and cancer (21% vs 32.5%, p=0.19; Table 1).

The primary endpoint of 30-day mortality was significantly lower in the group that

received vasopressors within 6 hours compared to patients who received vasopressors after 6

hours (25% vs. 51.1%, p<0.01). There were no differences between groups with regards to use
 Journal Pre-proof

of stress dose steroids, concomitant inotrope therapy, ICU LOS, and hospital LOS (Table 2).

There were no differences in cumulative vasopressor dose and maximum vasopressor dose at

24 hours; however, patients who received vasopressors within the first 6 hours had significantly

more vasopressor-free hours at the 72-hour mark as compared to those that received

vasopressors after 6 hours (34.5 hours vs. 13.1 hours, p=0.03; Table 2). Additionally, the time to

goal MAP was shorter in the early vasopressor group (1.5 [0.6 – 2.5] hours vs. 3 [1 – 6] hours,

p<0.01).

of
The following variables were included in multivariable logistic regression analysis:

vasopressor after 6 hours from initial hypotension, time to antibiotics, history of cirrhosis, and

ro
cancer. Administration of vasopressors after 6 hours from hypotension (odds ratio [OR] 2.9,

-p
95% confidence interval [CI] 1.3 – 7.0) and the presence of cirrhosis (OR 5.7, 95% CI 1.9 –
re
16.7) were independently associated with increased 30-day mortality (Table 3). Multicollinearity

was evaluated and the variance inflation factor for each variable was 1.07 or less, indicating no
lP

significant multicollinearity.
na

Discussion
ur

This study showed that a 6-hour or more delay in the initiation of vasopressors is
Jo

associated with doubling of 30-day mortality in patient with septic shock. This finding was

significant after accounting for potential confounding factors, including liver disease, through a

multivariable logistic regression. Furthermore, initiation of vasopressors within 6 hours was

associated with longer times off vasopressors within the first 72 hours and faster achievement of

MAP goals. These findings suggest that prompt initiation of vasopressors is associated with

improvement in patient outcomes and faster resolution of the state of shock.

Few studies have addressed the question of the optimal timing of vasopressor initiation

and its potential effect on patient outcomes. In a rat animal model, early use of norepinephrine

was associated with improved MAP with better tissue oxygenation when compared with fluid
 Journal Pre-proof

resuscitation alone.10 Another retrospective cohort study from two surgical intensive care units

found that for every 1-hour delay in norepinephrine initiation during the first 6 hours of septic

shock, there was a 5.3% increase in 28-day mortality. Furthermore, MAP and serum lactate

levels were significantly higher and lower, respectively, at various time points for the early-

norepinephrine group.6 A recent web-based survey study revealed that the consensus among

839 physicians from 82 countries of the European Society of Intensive Care Medicine was to

start vasopressors before the completion of full fluid resuscitation.11 While the necessity for

of
earlier initiation of vasopressors is being recognized, the optimal timing is yet still unknown; in

this study, we found that delays of over 6 hours from the onset of hypotension are associated

ro
with worse patient outcomes.

-p
In this study, cirrhosis was also predictive of 30-day mortality. Previous studies have
re
demonstrated the complexity of treating septic shock in patients with cirrhosis. First, it has been

hypothesized that patients with cirrhosis have a higher predisposition to sepsis given their
lP

higher susceptibility to bacterial infections.12 Also, patients with cirrhosis may have higher pro-
na

inflammatory responses to sepsis compared to their non-cirrhotic counterparts, which could be

reflected in worse mortality. 9,13,14 Patients with cirrhosis tend to have a lower baseline MAP due
ur

to the increased action of vasodilatory substances (such as nitric oxide and cyclooxygenase-
Jo

derivatives) and impaired reactivity to vasoconstriction signals (such as vasopressin,

angiotensin II and endothelin-1); both phenomena are seen in portal hypertension and lead to a

decrease in systemic vascular resistance lower arterial blood pressure. 15 Providers may have

attributed lower pressures to the vasodilatory state of chronic liver disease rather than septic

shock, resulting in delayed vasopressor initiation. It is worth noting that liberal intravenous

crystalloid administration has been associated with worsened outcomes through different

mechanisms.16 Earlier studies suggest that fluid bolus administration only lead to transient

increases of intravascular volume and eventually lead to fluid leakage to the extravascular

spaces.17 In addition, repeated administration of intravenous fluids leads to increases in central

 Journal Pre-proof

venous pressure (CVP), which can result in decreased tissue perfusion pressure.17 The higher

proportion of patients with liver disease in the late vasopressor group could have contributed to

delayed initiation of therapy. We attempted to control for this in our multivariable logistic

regression, but future studies could evaluate vasopressor timing in patients with cirrhosis to

further investigate the effect of vasopressor administration delay in this subset of patients.

The timing of antibiotics has been previously linked to improvement in sepsis

outcomes.18 Despite this known concept, our study did not find mortality changes related to

of
antibiotic usage. The most likely reason for this is that a majority of patients within both groups

received antibiotics within the first hour of hypotension. This finding supports the impact

ro
regarding the timing of vasopressor initiation as the time to antibiotic administration was not

confounding these results.

-p
re
This study demonstrated that the significant mortality benefit with early initiation of a

vasopressor is likely secondary to earlier achievement and maintenance of adequate perfusion

lP

pressures, preventing the onset and/or progression of organ dysfunction.3 The use of
na

vasopressors is not without consequences; risk of well-known adverse reactions, such as

arrhythmias, may have been increased for patients with prolonged exposure to vasopressors,
ur

potentially adding to the increased mortality rate, but this study suggests that early initiation of
Jo

vasopressors was not associated with higher cumulative doses and was associated with fewer

hours on vasopressors (Table 2). Additionally, a prior study suggests detrimental effects of

aggressive fluid resuscitation due to imbalances in intracellular and extracellular osmolarity; the

need for such large volume fluid can be lessened with early vasopressor initiation.19

There are several limitations to our study. First, its single-center retrospective nature

limits the ability to generalize its results and adds the possibility for potential confounders.

Moreover, the data were dependent on the accurate charting of patient data including vital signs

and laboratory results that were used to identify time zero as well as documentation on the

timing of intravenous fluids and vasopressor initiation. We believe that a larger, retrospective
 Journal Pre-proof

study would be useful in further characterizing the consequences of delayed vasopressor

initiation.

A second limitation of this study relates to the determination of time zero. Whereas we

defined time zero as the time from initial hypotension, patients could have had widely different

times from the actual time when septic shock developed. This limitation would be present in any

retrospective study of septic patients. Further prospective, in-hospital, studies might be needed

to address this limitation.

of
Furthermore, another potential limitation of the study has to do with its exclusion criteria.

Twenty-nine patients were excluded because they died within 72 hours of presentation. As

ro
stated previously, the rational of this criterion was to exclude those patients who died due to the

-p
severity of their presentation rather than due to timing of vasopressors.6 A previous study
re
excluded patients who died within 24 hours. If we had applied this criterion (exclusion if death

within 24 hours as opposed to 72 hours), nine more patients would have been added to the
lP

study cohort, which we do not believe would have change the study conclusions.
na

Lastly, another limitation is the use of MAP to diagnose shock. Despite MAP being

commonly used in practice, the diagnosis of septic shock also relies on other clinical measures.
ur

Furthermore, a MAP less than 65mmHg does not always translate to cellular injury and organ

underperfusion.4 Nonetheless, MAP remains a widely used parameter in ICUs as a marker of

Jo

shock and its easy access makes it ideal for studies like this one.

Conclusions

In patients with septic shock, early administration of vasopressor (less than 6 hours from

initial hypotension) is associated with decreased mortality, more vasopressor-free hours within

the first 72 hours, and faster achievement of MAP goals.

 Journal Pre-proof

References

1 Rhodes A, Evans LE, Alhazzani W, et al. Surviving Sepsis Campaign: International Guidelines
for Management of Sepsis and Septic Shock: 2016. Intensive Care Med. 2017;43(3):304-377.

2 Gotts JE, Matthay MA. Sepsis: pathophysiology and clinical management. BMJ.
2016;353:i1585.

3 Varpula M, Tallgren M, Saukkonen K, Voipio-pulkki LM, Pettilä V. Hemodynamic variables

related to outcome in septic shock. Intensive Care Med. 2005;31(8):1066-71.

4 Beck V, Chateau D, Bryson GL, et al. Timing of vasopressor initiation and mortality in septic

of
shock: a cohort study. Critical Care 2014:18;1-8

5 Waechter J, Kumar A, Lapinskly SE, et al. Interaction between fluids and vasoactive agents

ro
on mortality in septic shock: a multicenter, observational study. Crit Care Med.
2014;42(10):2158 – 68

-p
6 Bai X, Yu W, Ji W, et al. Early versus delayed administration of norepinephrine in patients with
septic shock. Crit Care. 2014;18(5):532.
re
7 Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions
for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801-10.
lP

8 Brown SM, Lanspa MJ, Jones JP, et al. Survival after shock requiring high-dose vasopressor
therapy. Chest. 2013;143:664 – 671
na

9 Moreau R, Hadengue A, Soupison T, et al. Septic shock in patients with cirrhosis:

hemodynamic and metabolic characteristics and intensive care unit outcome. Crit Care Med.
1992;20(6):746-50.
ur

10 Sennoun N, Montemont C, Gibot S, Lacolley P, Levy B. Comparative effects of early versus

Jo

delayed use of norepinephrine in resuscitated endotoxic shock. Crit Care Med. 2007
Jul;35(7):1736-40.

11 Scheeren TWL, Bakker J, De Backer D, Annane D, Asfar P, Boerma EC, et al. Current use
of vasopressors in septic shock. Ann Intensive Care. 2019;9(1):20.

12 Bonnel AR, Bunchorntavakul C, Reddy KR. Immune dysfunction and infections in patients
with cirrhosis. Clin Gastroenterol Hepatol. 2011;9(9):727-38.

13 Byl B, Roucloux I, Crusiaux A, Dupont E, Devière J. Tumor necrosis factor alpha and
interleukin 6 plasma levels in infected cirrhotic patients. Gastroenterology. 1993;104(5):1492-7.

14 Devière J, Content J, Denys C, et al. Excessive in vitro bacterial lipopolysaccharide-induced

production of monokines in cirrhosis. Hepatology. 1990;11(4):628-34.

15 Bolognesi M, Di pascoli M, Verardo A, Gatta A. Splanchnic vasodilation and hyperdynamic

circulatory syndrome in cirrhosis. World J Gastroenterol. 2014;20(10):2555-63.
 Journal Pre-proof

16 Self WH, Semler MW, Bellomo R, et al. Liberal Versus Restrictive Intravenous Fluid Therapy
for Early Septic Shock: Rationale for a Randomized Trial. Ann Emerg Med. 2018;72(4):457-466.

17 Legrand M, Dupuis C, Simon C, et al. Association between systemic hemodynamics and

septic acute kidney injury in critically ill patients: a retrospective observational study. Crit Care.
2013;17(6):R278.

18 Liu VX, Fielding-singh V, Greene JD, et al. The Timing of Early Antibiotics and Hospital
Mortality in Sepsis. Am J Respir Crit Care Med. 2017;196(7):856-863.

19 Cotton BA, Guy JS, Morris JA Jr, Abumrad NN. The cellular, metabolic, and systemic
consequences of aggressive fluid resuscitation strategies. Shock. 2006 Aug;26(2):115-21.

of
ro
-p
re
lP
na
ur
Jo
 Journal Pre-proof

Tables and Figures

Figure. 1 Study design flow chart

76 patients received
vasopressors within 6
hours

530 patients for whom

vasopressors were 119 patients included
ordered for analysis

of
43 patients received
vasopressors after 6

ro
hours

Exclusions

100
-p
Vasopressors for other diagnosis other than septic shock
re
65 Vasopressors ordered but not started
62 Post-op cardiac surgery
lP

52 Care withdrawn
46 Vasopressors started in the OR
38 Vasopressors started at an outside hospital
29 Patient died within 72 hours
na

11 Missing or incomplete records

4 Burn injury patients
4 MAP goal >70 mmHg
ur
Jo
 Journal Pre-proof

Table 1. Baseline Demographics

Characteristic Vasopressors Vasopressors after p-value
within 6h (n=76) 6h (n=43)

Male sex, n (%) 39 (51.3) 20 (46) 0.25

Age (years), median (IQR) 63.5 (55-75.7) 64.0 (56-70) 0.71

Body mass index (kg/m2), median (IQR) 27 (22-31.8) 28.9 (23.6-34.3) 0.65

Race, n (%)
Caucasian 43 (56.6) 29 (67.4) 0.89
African American 19 (25.0) 10 (23.3) 0.95

of
Other 14 (18.4) 4 (9.3) 0.66

ro
Past medical history, n (%)
Hypertension 39 (51.3) 19 (44.1) 0.57
Diabetes mellitus 26 (34.2) 14 (32.5) >0.99
Coronary artery disease
Atrial fibrillation -p
21 (27.6)
13 (17.1)
8 (18.6)
7 (16.2)
0.37
>0.99
re
Cirrhosis 11 (14.4) 12 (27.9) 0.09
Chronic lung disease 16 (21.0) 6 (13.9) 0.46
Cancer 16 (21.0) 14 (32.5) 0.19
lP

Heart failure 12 (15.7) 9 (20.9) 0.62

Chronic kidney disease 19 (25) 10 (23.2) >0.99

Site of infection, n (%)

na

Pulmonary 31 (40.7) 17 (39.7) >0.99

Genitourinary 18 (23.6) 10 (23.2) >0.99
Intra-abdominal 15 (19.6) 10 (23.2) 0.45
ur

Skin/soft tissue infection 7 (9.2) 4 (9.3) >0.99

Blood 16 (21.0) 10 (23.2) >0.99
Osteomyelitis 1 (1.3) 0 (0) >0.99
Jo

Central nervous system 0 (0) 1 (2.3) 0.36

Unknown 5 (6.5) 2 (4.6) >0.99

Sequential organ failure assessment (SOFA) 9.0 (6.25-12.0) 8.0 (6.0-12.0) 0.98
score on admission, median (IQR)

Baseline MAP (mmHg), median (IQR) 54 (45.5-59.7) 52.0 (49.0-58.0) 0.58

Total volume of fluid within initial 6 hours of 2100 (1000-3000) 2000 (500-2677) 0.51
resuscitation (mL), median (IQR)

Norepinephrine as first choice pressor, n (%) 73 (96.1) 43 (100) 0.55

 Journal Pre-proof

Table 2. Outcomes
Outcome Vasopressors Vasopressors p-value
within 6h (n=76) after 6h (n=43)

Vasopressor free hours at 72h, 34.5 (4-58.6) 13.1 (0-44.4) 0.03

hours, median (IQR)

Time to Goal MAP, hours, median 1.5 (0.66-2.5) 3.0 (1-6) <0.01
(IQR)

Cumulative vasopressor dose at 60.9 (20.2-148-7) 56.7 (21.2-92) 0.76

of
12h (mcg/kg), median (IQR)

ro
Cumulative vasopressor dose at 96.1 (32.3-258) 109.5 (49-254.6) 0.95
24h (mcg/kg), median (IQR)

-p
re
Cumulative vasopressor dose at 179.0 (43.6-428) 231.1 (83.5-565.1) 0.41
72h (mcg/kg), median (IQR)
lP

Max pressor dose at 24h 0.20 (0.1-0.4) 0.20 (0.1-0.2) 0.49

(mcg/kg/min), median (IQR)
na
ur

30 day mortality, n (%) 19 (25 22 (51.1) <0.01

Jo

Stress dose steroids, n (%) 22 (29) 11 (26) 0.832

Inotropes, n (%) 5 (6.5) 2 (4.6) >0.99

ICU LOS days, median (IQR) 7 (3-12) 7 (4-11) 0.89

Hospital LOS ,median (IQR) 15 (7-24) 12 (8-20) 0.42

 Journal Pre-proof

Table 3. Multivariable logistic regression for 30-day mortality

Variable Odds Ratio 95% Confidence interval

Vasopressors after 6 2.9 1.3 7.0

hours

Time to antibiotics 0.9 0.9 1.0

Cirrhosis 5.7 1.9 16.7

of
Cancer 1.6 0.6 4.1

ro
-p
re
lP
na
ur
Jo
 Journal Pre-proof

Highlights

 Delays in vasopressors lead to longer times to shock resolution.

 Early initiation does not result in increases in cumulative vasopressor doses.

 Higher mortality seen when vasopressors started after six hours from hypotension.

of
ro
-p
re
lP
na
ur
Jo
Figure 1