Uncategorized

Early initiation of norepinephrine in patients with septic shock: A propensity score-based analysis

Journal logoUnlabelled imageearly initiation of norepinephrine in pa”>American Journal of Emergency Medicine 54 (2022) 287-296

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage:

Early initiation of norepinephrine in patients with septic shock: A propensity score-based analysis

Fei Xu a,b, Rong Zhong b, Shanyang Shi b, Yiqian Zeng c, Zhanhong Tang a,?

a Department of Intensive Care Unit, The First Affiliated Hospital of Guangxi Medical University, 6S Shuangyong Road, Nanning 530021, Guangxi, China

b Department of Intensive Care Unit, The Affiliated Hospital of Guilin Medical University, 15S Lequn Road, Guilin 541001, Guangxi, China

c Department of Intensive Care Unit, Zhuzhou Central Hospital, 116S South Changjiang Road, Zhuzhou 412007, Hunan, China

a r t i c l e i n f o

Article history:

Received 6 December 2021

Received in revised form 22 January 2022 Accepted 27 January 2022

Keywords:

Time factors Norepinephrine Shock, septic Propensity score Survival rate

a b s t r a c t

Background: The use of vasopressors is vital in septic shock. However, the optimal timing of treatment remains unclear. Therefore, we aimed to explore the impact of early norepinephrine initiation on the survival of patients with septic shock.

Methods: We selected 4253 patients from the Medical Information Mart for Intensive Care IV database between 2008 and 2019. The primary outcome was 28-day mortality. propensity score matching (PSM) was applied to minimize between-group imbalances, and a restricted mean survival time was used to quantify the beneficial im- pact of early norepinephrine treatment on survival. Sensitivity analyses were conducted to test the robustness of the study results in multiple cohorts.

Results: In the PSM cohort, 2862 patients were equally assigned to early (receiving norepinephrine within the first 3 h) and delayed (> 3 h) norepinephrine initiation groups. Patients in the early norepinephrine initiation group received significantly less fluid therapy (0 vs. 79 mL/kg), had lower 28-day mortality (30.0% vs. 37.8%), lon- ger survival days (21.89 vs. 20.37 days), shorter duration of intensive care unit (4.9 vs. 7.2 days) and hospital stays (12.4 vs. 13.6 days), shorter duration of supportive norepinephrine and invasive mechanical ventilation, lower incidence of organ failure progression (64.4% vs. 79.2%) within 24 h after shock onset, and higher mean ar- terial pressure within 6 and 24 h after shock onset than patients in the delayed norepinephrine initiation group (p

< 0.05).

Conclusions: Norepinephrine initiation within the first 3 h, regardless of preload dependency, was associated with longer survival time and shorter duration of supportive norepinephrine and invasive mechanical ventilation and may delay or partially reverse rapid onset organ failure.

(C) 2022

  1. Introduction

Sepsis, the most frequent and deadliest disease in critical care med- icine worldwide, is defined as a life-threatening organ dysfunction syn- drome caused by a dysregulated host response to infection [1]. Despite

Abbreviations: CI, confidence interval; ICU, intensive care unit; INR, international normalized ratio; MAP, mean arterial pressure; MICU, medical intensive care unit; MIMIC-IV, Medical Information Mart for Intensive Care IV; OR, odds ratios; PSM, propensity score matching; RMST, restricted mean survival time; SAPS II, Simplified Acute Physiology Score II; SCC, surviving sepsis campaign; SOFA, Sequential Organ Failure Assessment; SICU, surgery intensive care unit; Sep-3, sepsis-3.0.

* Corresponding author at: The First Affiliated Hospital of Guangxi Medical University, Department of Intensive Care Unit, 6S. Shuangyong Road, Qingxiu District, Nanning 530021, Guangxi Province, China.

E-mail addresses: [email protected] (F. Xu), [email protected] (R. Zhong), [email protected] (S. Shi), [email protected] (Y. Zeng), [email protected] (Z. Tang).

the earlier recognition of and therapeutic advancements for sepsis in re- cent years and that the progressive decline in the sepsis mortality rate, the mortality rate remains unacceptably high [2,3]. With the increasing aging population worldwide, sepsis has imposed huge burdens on clin- ical settings, the economy, and the society at large [2-4]. Septic shock is the most severe phase of sepsis, with the highest risk of mortality (ap- proximately 40-50%) [4,5].

Fluid resuscitation treatment is widely accepted as the cornerstone of septic shock therapy [6]. However, in pathophysiology, severe hypo- tension induced by sepsis occurs not only due to relative or absolute hy- povolemia but also due to vasoplegia, which necessitates vasopressor support. Thus, fluid resuscitation alone is insufficient to increase periph- eral perfusion pressure in most patients [7]. In addition, extracellular fluid overload may lead to tissue edema and increased mortality [8- 10]. Therefore, the appropriate use of vasopressors to promote hemody- namic stabilization in septic shock management is crucial [11]. The 2012 Surviving Sepsis Campaign (SCC) guidelines recommend vasopressors

https://doi.org/10.1016/j.ajem.2022.01.063

0735-6757/(C) 2022

for persistently hypotensive patients who are refractory to aggressive volume resuscitation, and norepinephrine has been recommended as the first-choice vasopressor [12]. However, neither the 2016 nor the current 2021 SCC guidelines explicitly stated the appropriate timing of norepinephrine administration [6,13]; thus, its timing remained un- clear. In an anonymous web-based survey between 2016 and 2017, the experts from 86 countries had different answers to the question: “When do you use your vasopressor?” and early use of a vasopressor re- gardless of preload dependency was preferred by only 26% of the ex- perts [14].

In recent years, many studies have attempted to answer this ques- tion. However, these various studies have come to conflicting conclu- sions about the relationship between early norepinephrine initiation after septic shock onset and mortality [15-18]. Furthermore, while the 2018 SSC bundle guidelines recommend vasopressors to restore at least 65 mmHg mean arterial pressure (MAP) during a 1-h bundle, the mechanisms of such treatment remain unclear [19]. Considering that the beneficial effects of early norepinephrine administration are still un- known, the optimal timing of norepinephrine administration requires further study.

To date, no uniform standard of early or delayed norepinephrine ini- tiation in septic shock management has been identified. In this study, early norepinephrine initiation was defined as a time interval <= 3 h from septic shock onset to norepinephrine treatment based on the 3-h bundle therapy instructions [6,13]. We aimed to assess the relationship between early norepinephrine initiation after septic shock onset, regardless of preload dependency, and the prognosis of septic shock patients by utilizing real-world data obtained from a large observational database, the Medical Information Mart for Intensive Care IV (MIMIC- IV) [20].

  1. Methods
    1. Source of data

This was a retrospective observational cohort study based on MIMIC-IV v1.0, a large and freely available database with considerable and high-quality data from patients admitted to an academic tertiary care hospital in the United States between 2008 and 2019. The database included a total of 76,540 intensive care unit admissions of 53,150 patients and 35,010 ICU admissions defined by the sepsis-3.0 (Sep-3) criteria [20]. The author was given access to the database after complet- ing the Collaborative Institutional Training Initiative Program (Record ID Fei Xu) approved by the appropriate Institutional Review Boards. The study was designed and implemented in accordance with all rele- vant guidelines and regulations. The need for informed consent from participants was waived due to the retrospective design of this study.

    1. Study populations

Adult patients (age >= 18 years) who fulfilled the diagnostic criteria for septic shock-3.0 or sepsis-induced acute circulatory failure were screened for inclusion. In accordance with the current Sep-3 diagnostic survival analysis“>criteria, sepsis was clinically defined as follows: patients considered to have suspected or demonstrated infection, with prescriptions for antibi- otics, sampling of bodily fluids for microbiological culture, and a se- quential organ failure assessment (SOFA) score >= 2 points [1]. Septic shock was also diagnosed based on the current international diagnostic consensus, including the presence of Sep-3, the use of vasopressor drugs, MAP <65 mmHg, and hyperlactatemia (Blood lactate levels

>2.0 mmol/L). Meanwhile, patients without hyperlactatemia were reclassified as having sepsis-induced acute circulatory failure.

Norepinephrine was the initial vasoactive agent. The onset time of

septic shock was defined as the first time that a MAP of <65 mmHg was recorded. Based on the time interval between septic shock onset and norepinephrine treatment, patients were allocated into two groups:

early norepinephrine initiation group and delayed norepinephrine initi- ation group. If the time interval was <=3 h, then patients were assigned into the early norepinephrine initiation group. On the contrary, if the time interval was >3 h, they were assigned to the delayed norepineph- rine initiation group. The patient inclusion flowchart is presented in de- tail in Fig. 1.

    1. Data collection

Data extraction was performed using PostgreSQL11.0 (PostgreSQL Global Development Group, Santa Barbara, California, USA). The follow- ing baseline characteristics were extracted from the MIMIC-IV database in the first 24 h following ICU admission: age, sex, ethnicity, weight, ad- mission style, admission year, ICU style, Charlson Comorbidity Index, SOFA, Simplified Acute Physiology Score II (SAPS II), comorbidities, vital signs, laboratory indicators, culture time, culture positivity, sup- portive therapies, onset time of septic shock, norepinephrine, fluid, and Antimicrobial treatments. We used the mean value of the vital signs on the first day. If a laboratory indicator was recorded more than once on the first day, then we used the values related to the greatest se- verity of septic shock. Multiple organ dysfunction was assessed using the SOFA and SAPS II scores. The laboratory indicators included in the SOFA score were no longer compared. Time to norepinephrine and an- tibiotic administrations and culture time were calculated from the onset of septic shock. The amount of fluid therapy was calculated from septic shock onset to norepinephrine treatment. Inotropes included do- butamine and milrinone. Additional vasopressors included vasopressin, epinephrine, and dopamine.

    1. Primary and secondary outcomes

The primary outcome was 28-day mortality following ICU admis- sion. Secondary outcomes were the duration of ICU and hospital stays, maximum norepinephrine rate and cumulative norepinephrine doses, norepinephrine-free days, invasive mechanical ventilation-free days, nadir MAP and mean MAP within 6 and 24 h after the onset of shock, acute kidney injury, and organ failure progression within 24 h after the onset of shock.

Norepinephrine-free days and invasive mechanical ventilation-free days were presented as the number of days out of 28 days following ICU admission, when the patients were alive and not using norepineph- rine or ventilator. If patients died within 28 days, then the end time point was selected as the time of death instead of 28.

We used MAP at shock onset as a reference to assess changes in blood pressure. Acute kidney injury was diagnosed based on the Kidney Disease: Improving Global Outcomes guidelines [21], and patients with end-stage renal disease were not evaluated for acute kidney injury. The progression of organ failure was based on serial changes in SOFA from the onset of septic shock to 24 h thereafter, which was defined as a SOFA difference equal to or greater than two points [22,23]. The highest scores of physiological derangements were used in the SOFA calcula- tions. If patients died in the meantime, SOFA was assigned the maximal 24 points.

    1. Survival analysis

Survival analysis was completed using the R package “survRM2 package” (R Foundation for Statistical Computing, Vienna, Austria). The restricted mean survival time (RMST) measures average survival for each patient up to the selected time, interpreted readily as the area under the survival curve during the follow-up period. Thus, the RMST difference was calculated based on the difference in the areas under the survival curve for the two groups, which can be used to quantify the treatment effect during the specified period [24].

Image of Fig. 1

Fig. 1. Study design flow chart.

ICU, intensive care unit; MAP, mean arterial pressure; MIMIC-IV, Medical Information Mart for Intensive Care IV; Sep-3, sepsis-3.0; PSM, propensity score matching.

    1. Statistical analysis

All numerical variables were presented as medians (lower quartile- upper quartile), whereas categorical variables as numbers and percent- ages. Numerical variables were compared using the Mann-Whitney test, whereas a pseudo-median difference and corresponding 95% con- fidence interval (CI) were calculated using the Hodges-Lehmann esti- mate. Categorical variables were compared using the chi-squared test, whereas the proportion difference and corresponding 95% CI were cal- culated using the Wilson method. Outliers of all variables were handled. As the presence of extensive missing data might lead to bias, variables exceeding 10% missing values were excluded from subsequent analyses. Correspondingly, the expectation maximization algorithm was utilized for variables with missing values of <5%, and multivariate imputation was used for variables with missing values of >5% (Table S1). After im- putation, we selected one of the datasets for illustration (Table S2).

Propensity score matching with a matching ratio of 1:1 and a caliper width of 0.02 was applied to reduce the influence of potential confounders [25] between the two groups (Fig. S1). Baseline character- istics that were significant different between the two groups (Table 1), except for fluid therapy, or baseline characteristics that appear to influ- ence outcome (age; Charlson comorbidity index; history of myocardial ischemia, chronic heart failure and malignant cancer; the time to antibi- otic use; the use of additional vasopressors and inotropes) were consid- ered as candidate variables. Variables were included in the final PSM analysis: SOFA score, SAPS II score, Charlson comorbidity index; dia- stolic blood pressure, MAP, oxygen saturation; white blood cell count, international normalized ratio, bicarbonate, anion gap, lactate; culture positivity; history of myocardial ischemia, chronic heart failure, malig- nant cancer, cirrhosis and stroke; whether to perform antibiotic and mi- crobiologic cultures in the first hour after shock onset; treated with Continuous renal replacement therapy, invasive mechanical ventilation

Baseline characteristics of the original cohort.

Characteristic

Total populations

Early norepinephrine

Delayed norepinephrine

Difference (95%CI)

P value

(n = 4253)

initiation group (n = 2069)

initiation group (n = 2184)

Age (years), median (IQR)

67 (56-77)

67 (56-77)

67 (56-77)

0.0 (-1.0-1.0)

0.990

Sex, n (%)

0.4 (-2.6-3.4)

0.780

Female

1884 (44.3)

912 (44.1)

972 (44.5)

Male

2369 (55.7)

1157 (55.9)

1212 (55.5)

Ethnicity, n (%)

0.931

White

2791 (65.6)

1357 (65.6)

1434 (65.7)

-0.1 (-2.9-2.8)

Black

325 (7.7)

152 (7.3)

173 (7.9)

-0.6 (-2.2-1.0)

Asian

125 (2.9)

63 (3.0)

62 (2.8)

0.2 (-0.8-1.2)

Other

324 (7.6)

162 (7.8)

162 (7.4)

0.4 (-1.2-2.0)

Unknown

Weight (kg), median (IQR)

688 (16.2)

79.8 (67.0-96.5)

335 (16.2)

80.0 (67.0-96.9)

335 (16.2)

79.0 (67.1-96.0)

0.0 (-2.2-2.2)

0.50 (-0.75-1.85)

0.409

Admission type, n (%)

0.724

Elective

374 (8.8)

178 (8.6)

196 (9.0)

-0.4 (-2.1-1.3)

Emergency

2527 (59.4)

1242 (60.0)

1585 (58.8)

1.2 (-1.8-4.1)

Urgent

1352 (31.8)

649 (31.4)

703 (32.2)

-0.8 (-3.6-2.0)

ICU style, n (%)

0.053

MICU/SICU

2691 (63.2)

1309 (63.3)

1382 (63.3)

-0.0 (-2.9-2.9)

TSICU

479 (11.3)

220 (10.6)

259 (11.9)

-1.2 (-3.1-0.7)

CCU

1042 (24.5)

527 (25.5)

515 (23.6)

1.9 (-0.7-4.5)

NICU

Admission year, n (%)

41 (1.0)

13 (0.6)

28 (1.3)

-0.7 (-1.2–0.0)

1.0 (-1.5-3.5)

0.435

2016 later

924 (21.7)

460 (22.2)

464 (21.2)

Scoring system, median (IQR) SOFA

10 (7-13)

11 (8-13)

9 (6-12)

2.0 (1.0-2.0)

< 0.001

SAPS II

47 (38-58)

49 (39-60)

45 (37-56)

4.0 (3.0-4.0)

< 0.001

CCI

6 (5-8)

6 (4-8)

6 (5-8)

0.0 (0.0-0.0)

0.171

Comorbidity, n (%)

Hypertension

1549 (36.4)

723 (34.9)

826 (37.8)

-2.9 (-5.8-0.0)

0.051

Myocardial ischemia

922 (21.7)

470 (22.7)

452 (20.7)

2.0 (-0.5-4.5)

0.110

Chronic heart failure

1609 (37.8)

788 (38.1)

821 (37.6)

0.5 (-2.4-3.4)

0.740

Diabetes

1336 (31.4)

661 (31.9)

675 (30.9)

1.0 (-1.8-3.8)

0.465

Chronic pulmonary disease

1241 (29.2)

596 (28.8)

645 (29.5)

-0.7 (-3.5-2.0)

0.602

Renal disease

1104 (26.0)

527 (25.5)

577 (26.4)

-0.9 (-3.6-1.7)

0.481

end stage renal disease

266 (6.3)

125 (6.0)

141 (6.5)

-0.4 (-1.9-1.0)

0.577

Malignant cancer

684 (16.1)

312 (15.1)

372 (17.0)

-2.0 (-4.2-0.3)

0.083

Cirrhosis

569 (13.4)

239 (11.6)

330 (15.1)

-3.6 (-5.6–1.5)

< 0.001

Severe live disease

505 (18.9)

206 (10.0)

299 (13.7)

-3.7 (-5.7–1.8)

< 0.001

Stroke

199 (4.7)

74 (3.6)

125 (5.7)

-2.1 (-3.4–0.9)

0.001

Vital sign, median (IQR)

Heart rate (/min)

88 (77-101)

88 (77-101)

88 (77-101)

1.0 (0.0-2.0)

0.204

Respiratory rate (/min)

20 (17-23)

20 (17-23)

20 (17-23)

0.0 (0.0-0.0)

0.416

SBP (mmHg)

106 (100-114)

105 (100-112)

107 (100-116)

-2.0 (-2.0–1.0)

< 0.001

DBP (mmHg)

57 (51-63)

56 (51-62)

57 (52-63)

-1.0 (-2.0–1.0)

< 0.001

MAP (mmHg)

71 (66-76)

70 (66-75)

71 (66-77)

-1.0 (-2.0–1.0)

< 0.001

Oxygen saturation (%)

97 (95-98)

97 (95-98)

96 (95-98)

0.00 (0.00-0.00)

< 0.001

Laboratory indicator, median (IQR) White blood cell (109 /L)

15.7 (11.0-21.7)

16.8 (11.9-23.3)

14.6(10.3-20.1)

2.10 (1.60-2.60)

< 0.001

Hemoglobin (mg/dL)

9.2 (7.8-10.7)

9.0 (7.7-10.5)

9.3(7.9-10.9)

-0.30 (-0.40–0.10)

< 0.001

Urea nitrogen (mg/dL)

30 (19-49)

31 (20-49)

29 (18-48)

1.00 (0.00-2.00)

0.015

Prothrombin time (s)

16.5 (13.8-21.6)

16.8 (14.1-21.8)

16.0 (13.5-21.8)

0.60 (0.30-0.90)

< 0.001

APTT (s)

38.6 (30.7-57.6)

38.2 (30.9-60.6)

38.1 (29.8-59.1)

0.70 (0.00-1.60)

0.052

INR

1.5 (1.2-2.0)

1.5 (1.3-2.0)

1.5 (1.2-2.0)

0.10 (0.00-0.10)

< 0.001

Bicarbonate (mmol/L)

19 (16-22)

19 (15-22)

20 (17-23)

-1.00 (-2.00–1.00)

< 0.001

Anion gap (mmol/L)

18 (15-22)

18 (15-22)

17 (15-21)

1.00 (0.00-1.00)

< 0.001

Base excess (mmol/L)

-5 (-10–1)

-6 (-11–2)

-4 (-8-0)

-2.00 (-3.00–2.00)

< 0.001

Sodium (mmol/L)

140 (137-143)

140 (137-143)

140 (136-143)

0.00 (0.00-1.00)

0.019

Chloride (mmol/L)

106 (102-111)

107 (103-111)

106 (101-110)

1.00 (1.00-2.00)

< 0.001

Potassium (mmol/L)

4.6 (4.2-5.2)

4.7 (4.2-5.3)

4.6 (4.1-5.2)

0.10 (0.00-0.10)

< 0.001

Lactate (mmol/L)

2.7 (1.7-4.9)

3.0 (1.8-5.8)

2.5 (1.6-4.4)

0.40 (0.20-0.50)

< 0.001

Culture positive, n (%)

1899 (44.7)

816 (39.4)

1083 (49.6)

-10.1 (-13.1–7.2)

< 0.001

Culture time, n (%)

<=1 h

3381 (79.5)

1729 (83.6)

1652 (75.6)

8.0 (5.5-10.3)

< 0.001

Septic shock definition, n (%)

6.8 (3.9-9.6)

< 0.001

Acute circulatory failure

1531 (40.0)

673 (32.5)

858 (39.3)

Septic shock-3.0

2722 (64.0)

1396 (67.5)

1326 (60.7)

Fluid therapy (mL/kg), median (IQR)

18 (0-81)

0 (0-1.5)

79 (37-180)

-75.03 (-78.80–71.55)

< 0.001

Additional therapy, n (%)

CRRT

790 (18.6)

325 (15.7)

465 (21.3)

-5.6 (-7.9–3.3)

< 0.001

Invasive mechanical ventilation

3397 (79.9)

1582 (76.5)

1812 (83.1)

-6.6 (-9.1–4.2)

< 0.001

Additional vasopressors

2043 (48.1)

1011 (48.9)

1032 (47.3)

1.6 (-1.4-4.6)

0.293

Inotropes

509 (12.0)

235 (11.4)

274 (12.5)

-1.2 (-3.1-0.8)

0.233

Hydrocortisone

625 (14.7)

339 (16.4)

286 (13.1)

3.3 (1.2-5.4)

0.002

Time to antibiotic use

-0.5 (-3.4-2.4)

0.745

<=1 h

1581 (37.2)

764 (36.9)

817 (37.4)

and hydrocortisone. If the p-value was >0.05, then the baseline charac- teristic was considered as a measure of balance between the two groups.

The RMST analysis was used to analyze the correlations between early norepinephrine therapy and 28-day mortality, and the survival time differences between the two groups in the original and PSM co- horts. We recorded and evaluated the differences in RMST with 95% CI for 28 days.

Multivariate logistic regression analysis was performed to adjust for potential confounders and identify the protective and risk factors of 28- day mortality. Variables with a p-value <0.1 in the univariate analysis or variables considered to be clinically relevant were considered as candi- date variables. The method was forward likelihood ratio. Adjusted odds ratios (ORs) of 28-day mortality with their corresponding 95% CIs were calculated. In addition, the timing of norepinephrine initiation was regarded as a categorical variable in the analysis.

Sensitivity analysis was conducted using the RMST analysis and mul- tivariable logistic regression to assess the early norepinephrine initia- tion effects on multiple cohorts: those with missing data (original cohort), those with complete data after imputation (complete cohort), those after PSM (PSM cohort), those diagnosed with septic shock-3.0 (septic shock cohort), those from the medical ICU (MICU) or surgery ICU (SICU) (MICU/SICU cohort), those without liver cirrhosis and Liver failure (non-liver cirrhosis/failure cohort), and those without end stage renal disease (non-dialysis cohort).

Statistical significance was defined as a two-sided p-value of <0.05. All statistical analyses were performed using SPSS software (17.0; SPSS Inc., Chicago, IL, USA), Stata software (12.0; StatCorp LLC, College Sta- tion, TX, USA), and R software (4.1.0; R Foundation for Statistical Com- puting).

  1. Results
    1. Cohort characteristic

A total of 4253 patients were enrolled in the study (Fig. 1). Among these, 2722 patients (64.0%) were diagnosed with septic shock, whereas 1531 patients (36.0%) were diagnosed with sepsis-related acute circula- tory failure. The median onset time of septic shock in all patients was 2 (1-6) h following ICU admission; 99% of them had shock within 4 days following ICU admission, and only two patients had shock within 1 h be- fore ICU admission. Moreover, in the early norepinephrine initiation group, 1227 patients had received norepinephrine treatment simulta- neously with the onset of septic shock, and 473 had received norepi- nephrine within 1 h of septic shock onset. In this study, 28-day mortality was 35.5%, whereas the duration of ICU and hospital stays were 6.4 (3.3-12.9) and 13.4 (7.1-23.4) days, respectively. The baseline characteristics of all patients are summarized in Table 1.

In the original cohort, there were significant differences between the

two groups in SOFA and SAPS II score; cirrhosis, severe liver disease, and stroke; systolic and diastolic blood pressure; MAP; the Positive culture rate and culture time; the proportion of patients receiving hydrocorti- sone, continuous renal replacement therapy, and invasive mechanical ventilation; the amount of fluid therapy; and laboratory indicators, in- cluding white blood cell count; hemoglobin, urea nitrogen, electrolyte, and lactate levels; prothrombin time; and acid-base disturbances (Table 1). Compared to the delayed norepinephrine initiation group, pa- tients with higher SOFA and SAPS II score, lower blood pressure, more severe acid-base disorders, and higher lactate levels were more likely to receive norepinephrine treatment within the first 3 h. In the PSM co- hort, 1431 patients who initiated norepinephrine within the first 3 h were matched with 1431 those who initiated norepinephrine 3 h after shock recognition (Fig. S1). The baseline characteristics were similar across the two groups (P > 0.05), except for fluid therapy: patients in the early norepinephrine initiation group received significantly less fluid. The details are described in Table 2.

    1. Primary and secondary outcome

As shown in Table 3, patients in the early norepinephrine initiation group had a significantly lower 28-day mortality, shorter duration of ICU and hospital stay, shorter duration of supportive norepinephrine and invasive mechanical ventilation, lower incidence of acute kidney in- jury, and lower proportion of organ failure progression than patients in the delayed norepinephrine initiation group in the PSM cohort. In addi- tion, we found that the maximum norepinephrine rate and cumulative norepinephrine doses were not significantly different between the two groups (P > 0.05). Meanwhile, we found that patients in the early group with lower MAP at shock onset had similar nadir MAP within 6 h after shock onset, higher nadir MAP within 24 h after shock onset, and higher mean MAP within 6 h and 24 h after shock onset. This means that pa- tients in the delayed group had more prolonged hypotension.

Although we did not obtain the same differences in the maximum norepinephrine rate, MAP changes, and incidence of acute kidney injury within 24 h after shock onset in the original cohort, they were correlated with more severe Disease states in the early norepinephrine initiation group.

    1. Primary analysis

The RMST indicated a beneficial effect in the early norepinephrine initiation group during the 28 days following ICU admission in the orig- inal cohort (Fig. 2a) and PSM cohort (Fig. 2b). In both the original and the PSM cohorts, patients in the early norepinephrine initiation group had longer survival days than those in the delayed norepinephrine ini- tiation group. The details are described in Table 4.

In the multivariate logistic regression analysis, early norepinephrine initiation was associated with a reduction in the risk of 28-day mortality compared with delayed norepinephrine initiation in the original cohort or in the complete cohort (Fig. 3). In addition, higher SAPS II score, higher International normalized ratio , and the use of additional vasopressors were identified as independent Prognostic risk factors for 28-day mortality in the complete cohort (Table 5).

    1. Sensitivity analysis

Sensitivity analyses were conducted to assess the robustness of our study results. As shown in Fig. 2, the results indicated a beneficial sur- vival effect in the early norepinephrine initiation group within 28 days following ICU admission in the original, PSM, septic shock, MICU/SICU, non-liver cirrhosis/failure, and non-dialysis cohorts (P < 0.05). Further details are shown in Table 4. Moreover, early norepinephrine initiation was also associated with a significant reduction in the risk of 28-day mortality in the multiple cohorts (Fig. 3).

  1. Discussion

hemodynamic support is crucial for the prognosis of patients with sepsis and septic shock [11]. However, the appropriate timing for nor- epinephrine administration remains unclear. Our study demonstrated a correlation between norepinephrine initiation within the first 3 h after shock recognition, regardless of preload dependency, and a signif- icantly decreased risk of 28-day mortality (OR 0.545, P < 0.001). In ad- dition, the RMST analysis results also suggested an association between early initial norepinephrine therapy (<= 3 h) and increased 28-day sur- vival time. Compared with patients in the delayed norepinephrine group, those in the early norepinephrine group had longer survival dur- ing the 28 days following ICU admission, in the original and the PSM co- hort. Furthermore, despite applying imputation, populations defined as septic shock, only from MICU/SICU, without liver cirrhosis/failure, and without end stage renal disease, early norepinephrine intervention can still cause a decreased risk of mortality and a longer survival time at 28 days.

Table 2

Baseline characteristics of the propensity score matching cohort.

Characteristic

Early norepinephrine initiation group (n = 1431)

Delayed norepinephrine initiation group (n = 1431)

Difference (95%CI)

P value

Age (years)

67 (56-77)

67 (56-78)

0.0 (-1.0-1.0)

0.824

Sex, n (%)

-0.3 (-3.9-3.4)

0.880

Female

628 (43.9)

624 (43.6)

Male

803 (56.1)

807 (56.4)

Ethnicity, n (%)

0.412

White

959 (67.0)

928 (64.8)

2.3 (-1.3-5.6)

Black

97 (6.8)

118 (8.2)

-1.5 (-3.4-0.5)

Asian

40 (2.8)

42 (2.9)

-0.1 (-1.4-1.1)

Other

98 (6.8)

114 (8.0)

-1.1 (-3.0-0.8)

Unknown

Weight (kg), median (IQR)

237 (16.6)

80.3 (67.5-96.7)

229 (16.0)

79.2 (66.7-95.7)

0.6 (-2.1-3.3)

0.75 (-0.80-2.30)

0.351

Admission type, n (%)

0.651

Elective

124 (8.7)

121 (8.5)

0.2 (-1.8-2.3)

Emergency

842 (58.8)

866 (60.5)

-1.7 (-5.3-1.9)

Urgent

465 (32.5)

444 (31.0)

1.5 (-1.9-4.9)

ICU style, n (%)

0.395

MICU/SICU

884 (61.8)

907 (63.4)

-1.6 (-5.2-1.9)

TSICU

166 (11.6)

158 (11.0)

0.6 (-1.8-2.9)

CCU

371 (25.9)

349 (24.4)

1.5 (-1.6-4.7)

NICU

Admission year, n (%)

10 (0.7)

17 (1.2)

-0.5 (-1.2-0.2)

1.4 (-1.6-4.4)

0.357

2016 later

308 (21.5)

288 (20.1)

Scoring system, median (IQR) SOFA

10 (8-13)

10 (8-13)

0.0 (0.0-0.0)

0.218

SAPS II

48 (38-59)

48 (39-59)

0.0 (-1.0-1.0)

0.695

CCI

6 (5-9)

6 (5-8)

0.0 (0.0-0.0)

0.330

Comorbidity, n (%)

Hypertension

492 (34.4)

528 (36.9)

-2.5 (-6.0-1.0)

0.160

Myocardial ischemia

318 (22.2)

306 (21.4)

-0.8 (-2.2-3.9)

0.587

Chronic heart failure

558 (39.0)

534 (37.3)

1.7 (-1.9-5.2)

0.356

Diabetes

451 (31.5)

431 (30.1)

1.4 (-2.0-4.8)

0.418

Chronic pulmonary disease

432 (30.2)

405 (28.3)

1.9 (-1.4-5.2)

0.267

Renal disease

366 (25.6)

388 (27.1)

-1.5 (-4.8-1.7)

0.351

End stage renal disease

92 (6.4)

92 (6.4)

0 (-1.8-1.8)

1.000

Malignant cancer

239 (16.7)

229 (16.4)

0.7 (-2.0-3.4)

0.613

Cirrhosis

191 (13.3)

186 (13.0)

0.3 (-2.1-2.8)

0.782

Severe live disease

163 (11.4)

186 (13.0)

-1.6 (-4.0-0.8)

0.189

Stroke

69 (4.8)

76 (5.3)

-0.5 (-2.1-1.1)

0.551

Vital sign, median (IQR)

Heart rate (/min)

88.0 (77-101)

88.0 (77-101)

1.0 (-1.0-2.0)

0.411

Respiratory rate (/min)

20 (17-23)

20 (17-23)

0.0 (0.0-0.0)

0.640

SBP (mmHg)

106 (100-112)

105 (99-114)

0.0 (-1.0-1.0)

0.937

DBP (mmHg)

56 (51-62)

56 (51-62)

0.0 (-1.0-0.0)

0.349

MAP (mmHg)

70 (66-76)

70 (66-76)

0.0 (-1.0-0.0)

0.986

Oxygen saturation (%)

97 (95-98)

97 (95-98)

0.0 (0.0-0.0)

0.283

Laboratory indicator, median (IQR) White blood cell (109 /L)

15.8(11.4-21.8)

15.6 (10.8-21.5)

0.40 (-0.20-1.00)

0.155

Hemoglobin (mg/dL)

9.0 (7.8-10.5)

9.2 (7.8-10.8)

-0.10 (-0.30-0.00)

0.084

Urea nitrogen (mg/dL)

30 (19-48)

30 (19-49)

0.00 (-1.00-1.00)

0.814

Prothrombin time (s)

16.6 (13.9-21.6)

16.7 (13.7-22.0)

0.10 (-0.30-0.40)

0.647

APTT (s)

38.4 (30.8-57.2)

39.4 (30.9-60.0)

-0.20 (-1.30-0.60)

0.505

INR

1.5 (1.3-2.0)

1.5 (1.2-2.0)

0.0 (0.0-0.0)

0.594

Bicarbonate (mmol/L)

19 (16-22)

19 (16-22)

0.0 (0.0-0.0)

0.640

Anion gap (mmol/L)

18 (14-22)

18 (15-22)

0.0 (0.0-0.0)

0.908

Base excess (mmol/L)

-5 (-10- -1)

-5 (-9- -1)

0.0 (0.0-0.0)

0.719

Sodium (mmol/L)

140 (137-143)

140 (137-143)

0.0 (0.0-0.0)

0.849

Chloride (mmol/L)

107 (102-111)

107 (102-111)

0.00 (-1.00-0.00)

0.944

Potassium (mmol/L)

4.6 (4.2-5.2)

4.6 (4.2-5.2)

0.00 (0.00-0.10)

0.636

Lactate (mmol/L)

2.7 (1.7-5.2)

2.8 (1.7-4.8)

0.00 (-0.10-0.10)

0.993

Culture positive, n (%)

631 (44.1)

610 (42.6)

1.5 (-2.2-5.1)

0.428

Culture time, n (%)

<=1 h

1167 (81.6)

1163 (81.3)

-0.6 (-3.5-2.3)

0.847

Septic shock definition, n (%)

0.1 (-3.4-3.6)

0.169

Acute circulatory failure

515 (36.0)

480 (33.6)

Septic shock-3.0

915 (64.0)

950 (66.4)

Fluid therapy (mL/kg), median (IQR)

0 (0-2)

70 (33-149)

-66.64 (-70.78–62.53)

< 0.001

Additional therapy, n (%)

CRRT

259 (18.1)

244 (17.1)

1.0 (-1.7-3.8)

0.461

Invasive mechanical ventilation

1123 (78.5)

1118 (78.1)

0.3 (-2.7-3.4)

0.821

Additional vasopressors

672 (47.0)

688 (48.1)

-1.1 (-4.8-2.5)

0.549

Inotropes

156 (10.9)

178 (12.4)

-1.5 (-3.9-0.8)

0.200

Hydrocortisone

212 (14.8)

210 (14.7)

0.1 (-2.5-2.7)

0.916

Time to antibiotic use

-1.1 (-4.7-2.5)

0.540

<=1 h

549 (38.4)

565 (39.5)

Table 3

Outcome comparison.

Outcome

Early norepinephrine initiation group

Delayed norepinephrine initiation group

Difference (95%CI)

P value

Original cohort

28-day mortality, n (%)

640 (30.9)

871 (39.9)

-9.0 (-11.8–6.1)

< 0.001

Duration of ICU stays (days), median (IQR)

4.8 (2.6-9.7)

8.3 (4.3-15.1)

-2.7 (-3.0–2.3)

< 0.001

duration of hospital stays (days), median (IQR)

11.8 (6.2-20.2)

15.0 (8.2-26.0)

-3.04 (-3.67–2.42)

< 0.001

Maximum norepinephrine rate (ug/kg.min), median (IQR)

0.28 (0.14-0.50)

0.25 (0.12-0.50)

0.00 (0.00-0.02)

0.038

Cumulative norepinephrine doses (mg), median (IQR)

18.1 (6.4-49.6)

16.9 (5.6-49.1)

0.58 (-3.33-1.51)

0.213

Norepinephrine-free days, median (IQR)

25.1 (4.7-26.9)

20.6 (4.1-26.5)

0.50 (0.25-0.75)

< 0.001

Invasive mechanical ventilation-free days, median (IQR)

23.7 (3.1-27.4)

14.2 (1.8-25.4)

1.79 (1.25-2.38)

< 0.001

Acute kidney injury within 24 h, n (%)

1433 (69.3)

1568 (71.8)

-2.5 (-5.3-0.2)

0.069

Organ failure progression, n (%)

1365 (66.4)

1562 (71.5)

-5.5 (-8.3–2.8)

< 0.001

MAP at shock onset (mmHg), median (IQR)

57 (51-61)

60 (55-63)

-2.00 (-2.00–2.00)

< 0.001

Nadir MAP within 6 h (mmHg), median (IQR)

59 (53-64)

59 (53-66)

-1.00 (-1.00-0.00)

0.010

Nadir MAP within 24 h (mmHg), median (IQR)

54 (48-59)

53 (46-59)

0.00 (0.00-0.50)

0.004

Mean MAP within 6 h (mmHg), median (IQR)

70 (64-75)

69 (64-75)

0.40 (-0.11-0.97)

0.120

Mean MAP within 24 h (mmHg), median (IQR)

71 (66-75)

70 (66-75)

0.38 (-0.05-0.82)

0.085

Propensity score matching cohort

28-day mortality, n (%)

430 (30.0)

541 (37.8)

-7.8 (-11.2–4.3)

< 0.001

Duration of ICU stays (days), median (IQR)

4.9 (2.8-9.8)

7.2 (3.9-13.7)

-1.71 (-2.04–1.33)

< 0.001

Duration of hospital stays (days), median (IQR)

12.4 (6.6-20.8)

13.6 (7.3-24.0)

-1.25 (-2.00–0.542)

0.007

Maximum norepinephrine rate (ug/kg.min), median (IQR)

0.25 (0.12-0.50)

0.23 (0.12-0.49)

0.01 (0.00-0.02)

0.104

Cumulative norepinephrine doses (mg), median (IQR)

16.4 (5.6-46.6)

15.5 (5.5-44.7)

0.53 (-0.50-1.57)

0.569

Norepinephrine-free days, median (IQR)

25.3 (6.2-27.0)

22.1 (3.7-26.7)

0.42 (0.17-0.67)

< 0.001

Invasive mechanical ventilation-free days, median (IQR)

24.0 (4.3-27.4)

16.4 (1.7-26.1)

1.38 (0.79-2.04)

< 0.001

Acute kidney injury within 24 h, n (%)

982 (68.6)

1043 (72.9)

-4.3 (-7.6–0.9)

0.012

Organ failure progression, n (%)

922 (64.4)

1134 (79.2)

-14.8 (-18.1–11.6)

< 0.001

MAP at onset (mmHg), median (IQR)

57 (51-61)

60 (55-63)

-2.00 (-2.00–1.50)

< 0.001

Nadir MAP within 6 h (mmHg), median (IQR)

59 (53-64)

58 (52-65)

0.0 (0.0-1.0)

0.396

Nadir MAP within 24 h (mmHg), median (IQR)

54 (48-59)

52 (46-58)

1.00 (0.00-1.50)

< 0.001

Mean MAP within 6 h (mmHg), median (IQR)

70 (64-75)

69 (63-75)

0.98 (0.32-1.60)

0.004

Mean MAP within 24 h (mmHg), median (IQR)

71 (67-75)

70 (65-75)

0.85 (0.33-1.36)

0.001

ICU, intensive care unit; MAP, mean arterial pressure; CI, confidence interval.

These findings are consistent with many previous studies, which have shown a correlation between early norepinephrine administration and reduced mortality. A retrospective cohort study, including 213 pa- tients, showed that 28-day mortality was significantly higher when the time interval from septic shock onset to norepinephrine use was

>=2 h compared to <2 h, and for every 1 h delay within 6 h, the mortality increased by 5.3% [15]. Recently, Ospina-Tascon GA et al. also confirmed that the start of vasopressor drugs within the hour after the first fluid load or even before, was associated with a significant reduction in the risk of 28-day mortality compared with starting >1 h after the first fluid preload dependency [17]. In addition, a recent meta-analysis also demonstrated that early intervention with norepinephrine was associ- ated with decreased short-term mortality [26]. Thus, the following fac- tors could lower the risk of mortality: early use of norepinephrine can prevent prolonged severe hypotension, increase cardiac output, im- prove microcirculation perfusion, and prevent harmful fluid overload [7,27]. Our research also confirmed that patients in the delayed group had more prolonged hypotension.

In addition, we found that early norepinephrine initiation could de- crease the duration of ICU and hospital stays and invasive mechanical ventilation, which reduces the Economic burden. Studies show that early norepinephrine initiation could minimize the need for intrave- nous injection of fluids to stabilize macrohemodynamic parameters [15,16], reducing the incidence of congestive heart failure, pulmonary edema and accompanying the improvement of Respiratory function si- multaneously.

There are ongoing debates about the effects of early norepinephrine initiation in septic shock. The main controversy is that high doses or fre- quent administration of norepinephrine can result in many adverse ef- fects, such as arrhythmias and acute kidney injury [28]. In our study, patients in the early initiation group not only had a shorter duration of norepinephrine use but also had a nearly equivalent maximum rate and drug dosage. Many studies also found the same phenomenon: fluid resuscitation might paradoxically increase vasopressor demands

compared with the early use of vasopressors alone, whether in labora- tory [29] or clinical studies [17]. In addition, the population in the early initiation group had a lower incidence of acute kidney injury in our study. This finding has also been confirmed in the literature [18]. Another main controversy is that the early administration of norepi- nephrine can induce microcirculatory perfusion disorders, resulting in organ dysfunction and accelerated deterioration. Many studies have found that Serum lactate levels at varioUS time points were significantly decreased in the early norepinephrine group, and the delayed use of a vasopressor drug is an independent risk factor for the progression of organ failure [15,30]. Our results also indicate that its early use can in- crease MAP, improve organ perfusion, and delay organ failure with rapid deterioration or reversal to a certain extent. However, microcircu- latory indices, such as Lactate clearance, were not included in this analysis.

Our study demonstrated that higher SAPS II score, higher INR index, and additional vasopressors were independent risk factors for 28-day mortality in septic shock patients. These findings have been confirmed by many previously published studies [16,17,31]. The INR index repre- sents prothrombin time, and it can be normally used to evaluate the se- verity of the liver disease. This can be associated with a decrease in systemic vascular resistance due to increased vasodilatory substances [32]. In addition, the requirements for additional vasopressor support were similar in both groups. Nevertheless, despite more serious clinical presentations, early norepinephrine intervention still improved 28-day mortality (30.9% vs. 39.9%, P < 0.001) and was an independent protec- tive factor for 28-day mortality (adjusted OR 0.545). These results con- firm this conclusion from another perspective.

The strengths of our study are as follows. First, we used a large sam- ple and real-world data obtained from an academic tertiary care hospi- tal in the United States, and the patients were enrolled based on the most recent definition of septic shock (3.0). Second, we applied PSM to reduce selection bias, which makes the two groups comparable. Third, we used RMST analysis to evaluate the association between

Image of Fig. 2

Fig. 2. Restricted mean survival time survival curves in multiple cohorts.

The difference in restricted mean survival time reflects the differences in the areas of the Kaplan-Meier curves for the two groups for 28 days in the original cohort (a), PSM cohort (b), septic shock cohort (c), MICU/SICU cohort (d), non-liver cirrhosis/failure cohort (e), and non-dialysis cohort (f). Early norepinephrine group: patients who received norepinephrine treat- ment within the first 3 h after septic shock onset. Delayed norepinephrine group: patients who received norepinephrine treatment more than 3 h after septic shock onset.

MICU, medical intensive care unit; SICU, surgery intensive care unit; PSM, propensity score matching.

early initiation of norepinephrine therapy and increased 28-day sur- vival time; therefore, the proportional hazards assumption could not be taken into consideration [24].

However, our study has limitations. First, the sample size of our study population was substantially reduced after the use of PSM, result- ing in possible selection bias that should be considered. Second, residual

Table 4

Restricted mean survival time analysis for 28-day mortality in multiple cohorts.

Cohort

Early norepinephrine initiation

Delayed norepinephrine initiation

Difference (95%CI)

P value

Value 95%CI

Value 95%CI

Original cohort (days), median (IQR)

21.56 21.12-22.00

20.40 19.97-20.83

1.16 (0.54-1.78)

< 0.001

PSM cohort (days), median (IQR)

21.89 21.37-22.41

20.37 19.82-20.91

1.53 (0.77-2.28)

< 0.001

Septic shock cohort (days), median (IQR)

20.69 20.12-21.26

19.22 18.64-19.80

1.47 (0.66-2.28)

< 0.001

MICU/SICU cohort (days), median (IQR)

20.98 20.41-21.55

19.68 19.14-20.23

1.30 (0.51-2.09)

0.001

Non-liver cirrhosis/failure cohort (days), median (IQR)

21.56 21.12-22.01

20.40 19.97-20.83

1.16 (0.54-1.78)

< 0.001

Non-dialysis cohort (days), median (IQR)

21.62 21.16-22.08

20.31 19.87-20.78

1.30 (0.67-1.94)

< 0.001

PSM, propensity score matching; MICU, medical intensive care unit; SICU, surgery intensive care unit; CI, confidence interval.

Multiple cohorts

No.of patients

Adjusted OR

P value

Original cohort Complete cohort PSM cohort

Spetic shock cohort MICU/SICU cohort

Non-liver cirrhosis/failure cohort Non-dialysis cohort

4253 0.537(0.465-0.621)

4253 0.545(0.475-0.625)

2861 0.667(0.563-0.789)

2722 0.511(0.431-0.605)

2691 0.544(0.459-0.646)

3591 0.560(0.482-0.651)

3987 0.523(0.454-0.603)

<0.001

<0.001

<0.001

<0.001

<0.001

<0.001

<0.001

0 0.5 1 1.5 2

Fig. 3. Forest plot of multiple cohorts demonstrating effect of early norepinephrine initiation for 28-day mortality. PSM, propensity score matching; MICU, medical intensive care unit; SICU, surgery intensive care unit; OR, odds ratios.

Table 5

Multivariate logistic regression analysis for 28-day mortality in complete cohort.

Factor Adjusted odds ratio 95% CI P value Early norepinephrine initiation 0.545 0.475-0.625 <0.001

SAP II 1.033 1.028-1.038 <0.001

INR 1.398 1.291-1.514 <0.001

Additional vasopressors 1.958 1.707-2.245 <0.001

Hosmer-Lemeshow goodness-of-Fit test, P > 0.05 (0.971).

SAP II, simplified acute physiology score II; INR, international normalized ratio; CI, confi- dence interval.

confounders may still exist, which cannot be excluded in this study de- spite controlling many potential confounders using the PSM and multi- ple logistic regression. Third, the data of patients included in the analysis were from 2008 to 2019 with several SCC guidelines revisions, resulting in additional therapy strategies, which may have also affected the re- sults. Fourth, this was a single-centered retrospective study, although this was based on comprehensive and high-quality data from an aca- demic tertiary care hospital in the United States.

  1. Conclusions

Norepinephrine initiation within the first 3 h, regardless of preload dependency, was associated with longer survival time, shorter duration of supportive norepinephrine and invasive mechanical ventilation, and may delay or partially reverse rapid onset organ failure. The results show that norepinephrine initiation should be administered with the fluid resuscitation in patients with septic shock. Further large-scale ran- domized controlled trials are needed to confirm these conclusions.

Sources of support

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

Previous presentation

None.

Availability of data and materials

The data that support the findings of this study are available from the corresponding author upon reasonable request.

CRediT authorship contribution statement

Fei Xu: Writing – review & editing, Writing – original draft, Method- ology, Investigation, Conceptualization. Rong Zhong: Validation, Data curation. Shanyang Shi: Visualization. Yiqian Zeng: Software. Zhanhong Tang: Supervision, Project administration, Funding acquisi- tion.

Declaration of Competing Interest

We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any na- ture or kind in any product, service and/or company that could be con- strued as influencing the position presented in, or the review of, the manuscript entitled, “Early initiation of norepinephrine in patients with septic shock: A propensity score-based analysis.”

Acknowledgements

We would like to thank an English language editor from Editage (www.editage.cn) for revising the grammar, style, and syntax of the manuscript. We also appreciate the excellent work of the MIMIC team for making the data they collect freely available online.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2022.01.063.

References

  1. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA. 2016;315(8):801-10. https://doi.org/10.1001/jama.2016.0287.
  2. Dupuis C, Bouadma L, Ruckly S, Perozziello A, Van-Gysel D, Mageau A, et al. Sepsis and septic shock in France: incidences, outcomes and Costs of care. Ann Intensive Care. 2020;10(1):145. https://doi.org/10.1186/s13613-020-00760-x.
  3. Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, Kievlan DR, et al. Global, re- gional, and national sepsis incidence and mortality, 1990-2017: analysis for the global burden of disease study. Lancet. 2020;395(10219):200-11. https://doi.org/ 10.1016/s0140-6736(19)32989-7.
  4. Howell MD, Davis AM. Management of Sepsis and Septic Shock. JAMA. 2017;317(8):

847-8. https://doi.org/10.1001/jama.2017.0131.

  1. Liu V, Escobar GJ, Greene JD, Soule J, Whippy A, Angus DC, et al. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA. 2014;312(1):90-2. https://doi.org/10.1001/jama.2014.5804.
  2. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, et al. Surviving Sepsis campaign: international guidelines for Management of Sepsis and Septic Shock: 2016. Intensive Care Med. 2017;43(3):304-77. https://doi.org/10.1007/ s00134-017-4683-6.
  3. Hamzaoui O, Georger JF, Monnet X, Ksouri H, Maizel J, Richard C, et al. Early admin- istration of norepinephrine increases cardiac preload and cardiac output in septic patients with life-threatening hypotension. Crit Care. 2010;14(4):R142. https://doi. org/10.1186/cc9207.
  4. Acheampong A, Vincent JL. A positive fluid balance is an independent prognostic fac- tor in patients with sepsis. Crit Care. 2015;19:251. https://doi.org/10.1186/s13054- 015-0970-1.
  5. Boyd JH, Forbes J, Nakada TA, Walley KR, Russell JA. Fluid resuscitation in septic shock: a positive fluid balance and elevated central venous pressure are associated with increased mortality. Crit Care Med. 2011;39(2):259-65. https://doi.org/10. 1097/CCM.0b013e3181feeb15.
  6. Marik PE, Linde-Zwirble WT, Bittner EA, Sahatjian J, Hansell D. Fluid administration in severe sepsis and septic shock, patterns and outcomes: an analysis of a large na- tional database. Intensive Care Med. 2017;43(5):625-32. https://doi.org/10.1007/ s00134-016-4675-y.
  7. Parrillo JE. Septic shock-vasopressin, norepinephrine, and urgency. N Engl J Med. 2008;358(9):954-6. https://doi.org/10.1056/NEJMe0800245.
  8. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving Sepsis campaign: international guidelines for management of severe sepsis and sep- tic shock, 2012. Intensive Care Med. 2013;39(2):165-228. https://doi.org/10.1007/ s00134-012-2769-8.
  9. Evans L, Rhodes A, Alhazzani W, Antonelli M, Coopersmith CM, French C, et al. Sur- viving sepsis campaign: international guidelines for management of sepsis and sep- tic shock 2021. Intensive Care Med. 2021;47(11):1181-247. https://doi.org/10. 1007/s00134-021-06506-y.
  10. 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. https://doi. org/10.1186/s13613-019-0498-7.
  11. Bai X, Yu W, Ji W, Lin Z, Tan S, Duan K, et al. Early versus delayed administration of norepinephrine in patients with septic shock. Crit Care. 2014;18(5):532. https://doi. org/10.1186/s13054-014-0532-y.
  12. Beck V, Chateau D, Bryson GL, Pisipati A, Zanotti S, Parrillo JE, et al. Timing of vaso- pressor initiation and mortality in septic shock: a cohort study. Crit Care. 2014;18

(3):R97. https://doi.org/10.1186/cc13868.

  1. Ospina-Tascon GA, Hernandez G, Alvarez I, Calderon-Tapia LE, Manzano-Nunez R, Sanchez-Ortiz AI, et al. Effects of very early start of norepinephrine in patients with septic shock: a propensity score-based analysis. Crit Care. 2020;24(1):52. https://doi.org/10.1186/s13054-020-2756-3.
  2. Permpikul C, Tongyoo S, Viarasilpa T, Trainarongsakul T, Chakorn T, Udompanturak

S. Early use of norepinephrine in septic shock resuscitation (CENSER). A randomized trial. Am J Respir Crit Care Med. 2019;199(9):1097-105. https://doi.org/10.1164/ rccm.201806-1034OC.

  1. Levy MM, Evans LE, Rhodes A. The surviving Sepsis campaign bundle: 2018 update. Intensive Care Med. 2018;44(6):925-8. https://doi.org/10.1007/s00134-018-5085- 0.
  2. Johnson ABL, Pollard T, Horng S, Celi LA, Mark R. MIMIC-IV, v1. PhysioNet. 2021. https://doi.org/10.13026/s6n6-xd98.
  3. Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, et al. Acute kid- ney injury network: report of an initiative to improve outcomes in acute kidney in- jury. Crit Care. 2007;11(2):R31. https://doi.org/10.1186/cc5713.
  4. Levy MM, Macias WL, Vincent JL, Russell JA, Silva E, Trzaskoma B, et al. Early changes in Organ function predict eventual survival in severe sepsis. Crit Care Med. 2005;33 (10):2194-201. https://doi.org/10.1097/01.ccm.0000182798.39709.84.
  5. Vincent JL, de Mendonca A, Cantraine F, Moreno R, Takala J, Suter PM, et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Working group on “sepsis-re- lated problems” of the European Society of Intensive Care Medicine. Crit Care Med. 1998;26(11):1793-800. https://doi.org/10.1097/00003246-199811000-00016.
  6. Royston P, Parmar MK. The use of restricted mean survival time to estimate the treatment effect in randomized clinical trials when the proportional hazards as- sumption is in doubt. Stat Med. 2011;30(19):2409-21. https://doi.org/10.1002/ sim.4274.
  7. Zhang Z, Kim HJ, Lonjon G, Zhu Y. Written on behalf of AMEB-DCTCG. Balance diag- nostics after propensity score matching. Ann Transl Med. 2019;7(1):16. https://doi. org/10.21037/atm.2018.12.10.
  8. Li Y, Li H, Zhang D. Timing of norepinephrine initiation in patients with septic shock: a systematic review and meta-analysis. Crit Care. 2020;24(1):488. https://doi.org/ 10.1186/s13054-020-03204-x.
  9. Hamzaoui O, Shi R. Early norepinephrine use in septic shock. J Thorac Dis. 2020;12 (Suppl. 1):S72-7. https://doi.org/10.21037/jtd.2019.12.50.
  10. Domizi R, Calcinaro S, Harris S, Beilstein C, Boerma C, Chiche JD, et al. Relationship between norepinephrine dose, tachycardia and outcome in septic shock: a multicen- tre evaluation. J Crit Care. 2020;57:185-90. https://doi.org/10.1016/j.jcrc.2020.02. 014.
  11. Byrne L, Obonyo NG, Diab SD, Dunster KR, Passmore MR, Boon AC, et al. Unintended consequences: fluid resuscitation worsens shock in an ovine model of Endotoxemia. Am J Respir Crit Care Med. 2018;198(8):1043-54. https://doi.org/10.1164/rccm. 201801-0064OC.
  12. Hernandez G, Teboul JL. Is the macrocirculation really dissociated from the microcir- culation in septic shock? Intensive Care Med. 2016;42(10):1621-4. https://doi.org/ 10.1007/s00134-016-4416-2.
  13. Sato R, Ariyoshi N, Hasegawa D, Crossey E, Hamahata N, Ishihara T, et al. Effects of inotropes on the mortality in patients with septic shock. J Intensive Care Med. 2021;36(2):211-9. https://doi.org/10.1177/0885066619892218.
  14. 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. https://doi.org/10.3748/wjg.v20.i10.2555.