Documente Academic
Documente Profesional
Documente Cultură
PII: S0883-9441(19)30911-6
DOI: https://doi.org/10.1016/j.jcrc.2019.11.004
Reference: YJCRC 53416
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.
Delayed vasopressor initiation is associated with increased mortality in patients with septic
shock
Authors:
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
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,
of the mean arterial pressure (MAP) with intravenous fluid resuscitation. Fluid-refractory shock
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
Conclusion: Vasopressor initiation after 6 hours from shock recognition is associated with a
associated with shorter time to achievement of MAP goals and higher vasopressor-free hours
Jo
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
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
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.
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
The primary outcome was 30-day mortality. Secondary outcomes included cumulative
vasopressor dosing at 12, 24, and 72 hours; vasopressor doses were expressed as
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
(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.
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
were well-matched among groups except the prevalence of cirrhosis (14.4% vs 27.9%, p=0.09)
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
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
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
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
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
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.
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
pressures, preventing the onset and/or progression of organ dysfunction.3 The use of
na
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
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
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
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
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.
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
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.
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.
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
76 patients received
vasopressors within 6
hours
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
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
Sequential organ failure assessment (SOFA) 9.0 (6.25-12.0) 8.0 (6.0-12.0) 0.98
score on admission, median (IQR)
Total volume of fluid within initial 6 hours of 2100 (1000-3000) 2000 (500-2677) 0.51
resuscitation (mL), median (IQR)
Table 2. Outcomes
Outcome Vasopressors Vasopressors p-value
within 6h (n=76) after 6h (n=43)
Time to Goal MAP, hours, median 1.5 (0.66-2.5) 3.0 (1-6) <0.01
(IQR)
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
of
Cancer 1.6 0.6 4.1
ro
-p
re
lP
na
ur
Jo
Journal Pre-proof
Highlights
Higher mortality seen when vasopressors started after six hours from hypotension.
of
ro
-p
re
lP
na
ur
Jo
Figure 1