v
Search
Advanced

Publications > Journals > Journal of Translational Critical Care Medicine> Article Full Text

  • Original Article
  • OPEN ACCESS

Nonlinear Dose–response Relationship between Norepinephrine and Microcirculatory Perfusion: Identifying the Tipping Point in Septic Shock

  • Yiping Dai1,2,
  • Yao Zhu1,2,
  • Chang Hu1,2,
  • Hui Chen1,2,
  • Zhiyong Peng1,2,*  and
  • Yiming Li1,2,* 
 Author information 

Abstract

Background and objectives

While norepinephrine (NE) is the cornerstone of septic shock resuscitation, “macrocirculation–microcirculation decoupling” at high doses remains a critical clinical challenge. This study aimed to explore the quantitative tipping point where the NE dose shifts from a life-saving vasopressor to a microcirculatory toxin in septic intensive care unit patients.

Methods

In this prospective observational study (January–September 2025), we used handheld vital microscopy to monitor sublingual microcirculation (microvascular flow index [MFI], total vessel density, perfused vessel density, proportion of perfused vessels [PPV], and heterogeneity index [HI]) in adult septic patients within 24 hours of admission and on Day 3. Beyond standard linear analysis, generalized additive models were employed to identify the dose–response thresholds associated with microcirculatory deterioration, adjusted for Acute Physiology and Chronic Health Evaluation II, interleukin-6, and systemic hemodynamics.

Results

Of 144 screened patients, 66 were analyzed. The NE dose showed strong linear correlations with lactate (r = 0.583, P < 0.001) and HI (r = 0.444, P < 0.001), and negative correlations with MFI (r = −0.492, P < 0.001). Crucially, generalized additive models analysis revealed a significant nonlinear “cliff effect”: when the NE dose exceeds the 0.71–0.80 µg/kg/min threshold (PPV: 0.71 µg/kg/min, HI: 0.72 µg/kg/min, MFI: 0.80 µg/kg/min), microcirculatory perfusion parameters deteriorate abruptly (all P < 0.05). Multivariable Cox regression identified an NE dose of 0.80 µg/kg/min as an independent predictor of increased mortality (hazard ratio = 1.32, 95% confidence interval: 1.28–3.10, P = 0.039).

Conclusions

In patients with septic shock, higher NE doses were associated with impaired microcirculatory perfusion and worse outcomes. These findings support individualized vasopressor titration and suggest that microcirculatory monitoring may help identify patients at risk of vasopressor-associated microvascular dysfunction.

Keywords

Sepsis, Norepinephrine, Microcirculation, Dose-response relationship, Mortality, Intensive care units

Introduction

Sepsis is a growing global burden, with accumulating evidence indicating increased long-term morbidity and mortality in both developed and developing countries.1,2 Sepsis is characterized by a dysregulated host response to infection, leading to life-threatening organ dysfunction. This dysregulated host response induced by systemic inflammation induces vasodilation, leading to a relative insufficiency of effective circulating blood volume. This results in microcirculatory disruption and inadequate organ perfusion.3,4 Thus, microcirculatory dysfunction may play a pivotal role in the development of multiple organ failures in critically ill patients.5 Therefore, the primary goal of hemodynamic resuscitation in such patients is to restore microcirculatory perfusion and tissue oxygenation, prevent organ hypoxia, and maintain organ function.

Norepinephrine (NE), an endogenous catecholamine produced by postganglionic sympathetic nerves and the adrenal gland, is considered the first-line vasopressor for septic shock.6–8 NE effectively increases vascular tone and contractility to enhance tissue perfusion,9 and protects cardiac function in the early stage by improving myocardial contractility and diastolic arterial pressure to restore coronary perfusion.10 Studies have shown that NE administration helps ameliorate microcirculatory dysfunction.11,12 However, this effect is dose-dependent.11 With increasing doses, NE may exacerbate microcirculatory impairment, reduce tissue perfusion, and induce organ dysfunction.13,14 High-dose NE may even exert multiple adverse prognostic effects on macro- and microcirculation through immune, metabolic, and coagulation pathways.15,16 This suggests that only an appropriate NE dose is associated with improved tissue perfusion and organ function.

Although macrocirculation serves as the cornerstone of microcirculatory resuscitation, focusing solely on macrocirculatory resuscitation targets while ignoring microcirculatory responses may lead to therapeutic deviations.17 Effective resuscitation relies on the coherence between macrocirculation and microcirculation,18–20 that is, restoring “macrocirculation–microcirculation coupling.” The proportion of perfused vessels (PPV) of microcirculatory perfusion can serve as an early independent predictor of outcomes in patients with sepsis.21 Monitoring microcirculation to verify macrocirculation–microcirculation coupling provides crucial evidence for guiding NE use during hemodynamic resuscitation.22 Based on the principle of sidestream dark field (SDF) imaging,23 handheld non-invasive sublingual microscopes can be used to analyze microcirculatory changes in septic patients by assessing red blood cell velocity and perfused capillary density.24–28

Therefore, this study aimed to investigate the relationship between different NE doses and microcirculation as well as prognosis in septic intensive care unit (ICU) patients. Furthermore, we explored the dose–response relationship between NE and microcirculation, providing a rational reference for the treatment of septic patients and future research.

Materials and methods

Study design

This was a prospective observational study conducted in a 58-bed closed ICU in a tertiary hospital. The study protocol was approved by the Ethics Committee of Zhongnan Hospital of Wuhan University (No. 2025096K). Patients diagnosed with sepsis in the ICU between January 1, 2025, and September 1, 2025, were enrolled. All procedures adhered to the Declaration of Helsinki (as revised in 2024). All enrolled patients or their legally authorized representatives provided written informed consent to participate in the study.

Study population

Inclusion criteria

Patients meeting the Sepsis 3.0 diagnostic criteria (confirmed or highly suspected infection with Δ sequential organ failure assessment (SOFA) ≥ 2); who received initial resuscitation within 24 hours of ICU admission; were aged ≥ 18 years; and patients who underwent tracheal intubation and mechanical ventilation.

Exclusion criteria

Pregnant patients or those with peripheral vascular disease; patients unable to undergo microcirculation assessment or obtain high-quality SDF images; patients who died within 24 hours of admission; and patients aged > 80 years.

Study methods and data collection

Microcirculation monitoring

After professional training in microcirculation assessment, the researchers used an SDF imaging device (Microsee V100 Sublingual Microcirculation Imaging System) to visualize the sublingual microcirculation. Following gentle wiping of oral secretions with gauze, the sublingual microcirculation probe was placed lightly under the tongue. For each patient, stable and clear microvascular images were acquired from 5 sites adjacent to the lingual frenulum for at least 20 seconds by adjusting light intensity and focal length. Image quality was evaluated according to the method described by Massey et al.,29 and poor-quality images were excluded. Qualified images were analyzed using computer-aided software (AVA 3.0; MicroVision Medical; Amsterdam, the Netherlands) to assess microcirculatory parameters,24 including the microvascular flow index (MFI), total vessel density (TVD), perfused vessel density (PVD), PPV, and heterogeneity index (HI). Descriptions and calculation methods for all parameters are presented in Supplementary Table 1. In addition, two researchers analyzed all parameters using software to identify small vessels (diameter < 20 µm) through manual verification,30 as they are the primary sites of oxygen exchange.

Data collection

All the microcirculation parameters were collected when the enrolled septic patients had received adequate fluid resuscitation (30 mL/kg), with mean arterial pressure (MAP) ≥ 65 mmHg and no change in vasopressin usage within 2 hours. Two researchers simultaneously performed the assessments within 24 hours of ICU admission and on the third day. During microcirculation assessments, the following data were collected: demographic variables, hemodynamic parameters, laboratory values, vasopressor dose, sedative and analgesic medications, capillary refill time (CRT), and scoring systems, including SOFA score and acute physiology and chronic health evaluation (APACHE) II score as well as mottling score. After discharge, renal replacement therapy, length of ICU stay, and in-hospital mortality were also recorded.

Study outcomes

The primary outcome was 28-day mortality following enrollment. Secondary outcomes included 90-day mortality; mottling scores, lactate levels, and microcirculatory parameters, including MFI, PVD, TVD, PPV, HI, measured on Day 1 and Day 3, as well as their absolute changes between the two time points; NE dose on Day 1 and Day 3 along with the corresponding changes; and length of ICU stay.

Statistical analysis

Based on preliminary experiments and clinical practice, the correlation between NE dose and 28-day mortality was selected as the primary analytical variable.31–33 The sample size was estimated using PASS 15.0 software for the log-rank test. With an effect size f2 = 0.15, α = 0.05, and power (1-β) = 0.80, the minimum sample size was calculated as 60. Considering a 10% attrition rate, 66 patients were planned for enrollment.

The Shapiro–Wilk test was used to assess the normality of the variables. Normally distributed variables were expressed as mean ± standard deviation, non-normally distributed variables as median (IQR), and categorical variables as counts (percentages, %). Intergroup comparisons were performed using an independent samples t-test, Wilcoxon rank-sum test, chi-square test, or Fisher’s exact test based on data distribution characteristics. Pearson and Spearman correlation analyses were used to evaluate associations between variables using the Benjamini–Hochberg method for false discovery rate (FDR) correction. Univariate and multiple linear regression analyses were used to assess the independent association between NE doses and microcirculatory parameters. Nonlinear relationships were analyzed using a GAM model and validated by piecewise regression, allowing preliminary exploration of the dose–response relationship between NE and microcirculatory parameters. Kaplan–Meier curves were plotted to illustrate survival, with a log-rank test for intergroup differences. The Cox proportional hazards regression model was used to assess the relationship between NE doses and mortality risk. Statistical analyses and figure preparation were performed using SPSS software (version 22.0), GraphPad Prism software (version 8.0), and R software (version 4.3.2). A two-tailed P < 0.05 was considered statistically significant.

Results

Cohort characteristics

A total of 144 adult ICU septic patients were consecutively enrolled after ICU admission. As 78 patients were excluded (Fig. 1) from further analysis, 66 adult septic patients receiving NE treatment in the ICU were enrolled. Microcirculatory parameters, NE dose, and relevant laboratory values were collected on Day 1 and Day 3 of ICU admission. Baseline characteristics were compared between the two groups stratified by NE dose of 0.50 µg/kg/min (an NE dose exceeding 0.50 µg/kg/min is commonly considered a high dose) (Table 1).33 Both groups completed the preset fluid resuscitation protocol, with no statistically significant difference in baseline MAP (P = 0.670) and stable vasopressor usage. Compared to the low NE dose group (≤0.50 µg/kg/min, n = 37), patients in the high dose group (>0.50 µg/kg/min, n = 29) had higher Acute Physiology and Chronic Health Evaluation II (APACHE II) scores (29.24 ± 6.46 vs. 25.14 ± 5.94, P = 0.010), serum lactate (Lac) levels (3.20 [2.20, 7.70] vs. 1.40 [1.10, 1.90]) mmol/L, P < 0.001), and heart rates (113.72 ± 15.08 vs. 88.11 ± 18.57 bpm, P < 0.001) compared to the low-dose group. Notably, compared to the low-dose group, microcirculatory parameters were more deteriorated in the high-dose group: mottling scores (0 [0 to 2] vs. 0 [0 to 0], P = 0.001), HI (0.25 [0.11, 0.50] vs. 0.23 [0.11, 0.35], P = 0.033), MFI (2.83 [2.73, 2.98] vs. 2.92 [2.86, 2.98], P = 0.031), PPV (100.00 [97.10 to 100.00] vs. 100.00 [100.00 to 100.00], P = 0.005), and PVD (14.73 ± 3.77 vs. 17.22 ± 4.71 mm/mm2, P = 0.020). Finally, the 28-day mortality rate in the high-dose group was significantly higher than that in the low-dose group (65.5% vs. 29.7%, P = 0.008).

Flow chart of inclusion and exclusion.
Fig. 1  Flow chart of inclusion and exclusion.

NE, norepinephrine; SDF, sidestream dark field.

Table 1

Baseline characteristics stratified by NE dose of 0.50 µg/kg/min

CharacteristicsOverall n = 66Low NE dose (≤ 0.50 µg/kg/min) n = 37High NE dose (>0.50 µg/kg/min) n = 29P value
Age (years)61.65 (13.70)61.16 (13.54)62.28 (14.11)0.750
Male, n (%)50.0 (75.8)25.0 (67.6)25.0 (86.2)0.140
Body mass index (kg/m2)23.43 (3.27)24.17 (3.07)22.48 (3.34)0.039
Heart rate (bpm)99.36 (21.29)88.11 (18.57)113.72 (15.08)<0.001
Mean arterial pressure (mmHg)81.50 (13.10)82.14 (11.14)80.69 (15.41)0.670
CVP (cmH2O)13.13 (3.99)13.04 (4.11)13.24 (3.90)0.840
Hemoglobin (g/L)124.32 (12.58)126.67 (9.98)122.04 (13.71)0.450
Leukocyte (×109/L)11.80 (5.80, 17.22)11.67 (7.99, 16.60)12.20 (2.98, 22.11)0.930
Lymphocyte (×109/L)0.48 (0.23, 0.93)0.47 (0.23, 0.79)0.49 (0.23, 0.96)0.870
Platelets (×109/L)108.50 (57.00, 170.00)83.00 (57.00, 150.00)119.00 (57.00, 191.00)0.460
Procalcitonin (ng/mL)30.28 (3.14, 140.14)22.59 (1.57, 68.48)34.03 (11.51, 140.14)0.390
IL-6 (pg/mL)533.00 (64.30, 3,081.00)245.16 (39.30, 654.00)1,928.00 (382.00, 7,500.00)<0.001
Serum creatinine (mmol/L)169.85 (113.20, 245.00)146.10 (85.00, 230.10)215.00 (142.00, 287.00)0.029
Urea nitrogen (mmol/L)14.40 (11.06, 21.48)13.50 (11.10, 19.00)15.20 (11.06, 23.50)0.210
pH level7.35 (0.10)7.38 (0.09)7.32 (0.10)0.009
Lactate (mmol/L)1.90 (1.30, 4.10)1.40 (1.10, 1.90)3.20 (2.20, 7.70)<0.001
CRT (s)3.00 (2.00, 7.50)2.00 (2.00, 4.00)4.00 (2.00, 7.50)0.100
Mottling score (points)0.00 (0.00, 0.00)0.00 (0.00, 0.00)0.00 (0.00, 2.00)0.001
Peripheral perfusion index0.64 (0.18, 2.20)1.20 (0.35, 2.60)0.22 (0.10, 1.30)0.017
Microvascular flow index (points)2.92 (2.81, 2.98)2.92 (2.86, 2.98)2.83 (2.73, 2.98)0.031
Total vessel density (mm/mm2)16.01 (13.42, 18.44)16.61 (13.83, 19.83)15.02 (13.30, 16.48)0.048
Perfused vessel density (mm/mm2)16.12 (4.46)17.22 (4.71)14.73 (3.77)0.020
Proportion of perfused vessel (%)100.00 (99.80, 100.00)100.00 (100.00, 100.00)100.00 (97.10, 100.00)0.005
Heterogeneity index0.23 (0.11, 0.36)0.23 (0.11, 0.35)0.25 (0.11, 0.50)0.033
Fluid intake (mL)2,562.25 (1,309.80, 3,139.00)1,930.00 (1,060.90, 2,990.00)2,771.50 (2,290.00, 4,463.00)0.016
Combined vasopressor therapya, n (%)32.0 (48.5)16.0 (43.2)16.0 (55.2)0.480
Epinephrine, n (%)1.0 (1.5)0.0 (0.0)1.0 (3.4)0.900
Dobutamine, n (%)1.0 (1.5)1.0 (2.7)0.0 (0.0)>0.990
Terlipressin, n (%)12.0 (18.2)3.0 (8.1)9.0 (31.0)0.038
Methylene blue, n (%)26.0 (39.4)12.0 (32.4)14.0 (48.3)0.290
CRRT, n (%)49.0 (74.2)26.0 (70.3)23.0 (79.3)0.580
ECMO, n (%)14.0 (21.2)10.0 (27.0)4.0 (13.8)0.320
SOFA (points)9.36 (3.11)9.35 (3.36)9.38 (2.82)0.970
APACHE II (points)26.94 (6.46)25.14 (5.94)29.24 (6.46)0.010
ICU stay (days)10.00 (5.00, 19.00)10.00 (8.00, 19.00)9.00 (3.00, 16.00)0.110
28-day mortality, n (%)30.0 (45.5)11.0 (29.7)19.0 (65.5)0.008
90-day mortality, n (%)34.0 (51.5)15.0 (40.5)19.0 (65.5)0.077

Variables are displayed in mean (standard deviation), median (Q1, Q3) or n (%). aCombined use of NE and other vasoactive drugs, including epinephrine, dobutamine, terlipressin, and methylene blue. APACHE II, Acute Physiology and Chronic Health Evaluation II; CRRT, continuous renal replacement therapy; CRT, capillary refill time; CVP, central venous pressure; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; IL-6, interleukin-6; NE, norepinephrine; Q1, first quartile; Q3, third quartile; SOFA, sequential organ failure assessment.

Correlation between NE dose and microcirculatory parameters

In the total cohort (n = 66), the NE dose was significantly correlated with multiple microcirculatory parameters: negative correlations with MFI and PPV (MFI: r = −0.492, P < 0.001; PPV: r = −0.420, P < 0.001) ; positive correlations were observed with the HI, Lac, and mottling scores (HI: r = 0.444, P < 0.001; Lac: r = 0.583, P < 0.001; mottling score: r = 0.364, P = 0.003) (Fig. 2 and Supplementary Table 2). These associations remained statistically significant after FDR correction (P < 0.05). No significant correlations were found between the NE dose and CRT, PI, TVD, or PVD (P > 0.05).

Scatter plots of the association between NE dose and microcirculatory-related parameters.
Fig. 2  Scatter plots of the association between NE dose and microcirculatory-related parameters.

(a) Association between NE and PPV. (b) Association between NE and MFI. (c) Association between NE and HI. (d) Association between NE and Lac. Spearman rank correlation coefficients were used. HI, heterogeneity index; Lac, lactate; MFI, microvascular flow index; NE, norepinephrine; PPV, proportion of perfused vessels.

Multivariate regression analysis of NE dose and microcirculatory parameters

Univariate linear regression analysis showed that NE dose was significantly positively correlated with the HI, Lac, and mottling score (HI: 0.157 (0.078, 0.237), P < 0.001; Lac: 3.518 (2.294, 4.742), P < 0.001; mottling score: 0.445 (0.161, 0.730), P = 0.003), and significantly negatively correlated with MFI and PPV (MFI: −0.103 (−0.149, −0.058), P < 0.001; PPV: −2.136 (−3.289, −0.983), P < 0.001). Based on baseline grouping and clinical significance, APACHE II scores, heart rates, and interleukin (IL)-6 levels were adjusted for multivariate regression analysis. Except for the mottling score, which no longer reached statistical significance after adjustment (P = 0.115), all other associations remained statistically significant (P < 0.05) (Table 2). GAM analysis was employed to analyze the nonlinear relationship between NE dose and microcirculatory parameters while also identifying inflection points in the dose–response curve. After adjusting for covariates, including APACHE II scores, heart rates, and IL-6 levels, the GAM analysis revealed significant nonlinear relationships between NE dose and microcirculatory parameters (all P < 0.05). Significant inflection points of the NE dose (Fig. 3) were identified within these nonlinear relationships, and piecewise regression was employed to validate the respective inflection points (MFI: 0.80 µg/kg/min, P < 0.001, R2 = 0.532; PPV: 0.71 µg/kg/min, P = 0.031, R2 = 0.301; HI: 0.72 µg/kg/min, P = 0.014, R2 = 0.460; Lac: 1.65 µg/kg/min, P < 0.001, R2 = 0.455; mottling score: 0.25 µg/kg/min, P = 0.029, R2 = 0.123).

Table 2

Univariate and multivariate analysis of NE and microcirculatory parameters

ParametersUnivariate analysis
Multivariate analysis (model-1a)
Regression coefficient (95% CI)P valueRegression coefficient (95% CI)P value
Mottling Score0.445 (0.161, 0.730)0.0030.342 (−0.086, 0.770)0.115
MFI−0.103 (−0.149, −0.058)<0.001−0.099 (−0.163, −0.034)0.003
PPV−2.136 (−3.289, −0.983)<0.001−1.698 (−3.391, −0.004)0.049
HI0.157 (0.078, 0.237)<0.0010.146 (0.031, 0.260)0.013
Lac3.518 (2.294, 4.742)<0.0012.502 (0.700, 4.305)0.007

amodel-1: APACHE II scores, heart rates, and IL-6 levels were adjusted. APACHE II, Acute Physiology and Chronic Health Evaluation II; CI, confidence interval; HI, heterogeneity index; IL-6, interleukin-6; Lac, lactate; MFI, microvascular flow index; NE, norepinephrine; PPV, proportion of perfused vessels.

GAM curves of the nonlinear association between NE dose and microcirculatory-related parameters.
Fig. 3  GAM curves of the nonlinear association between NE dose and microcirculatory-related parameters.

(a) Nonlinear association between NE and MFI. (b) Nonlinear association between NE and PPV. (c) Nonlinear association between NE and HI. (d) Nonlinear association between NE and Lac. Adjusted for APACHE II scores, heart rates, and IL-6 levels, GAM analysis was employed to fit the trend between NE dose and microcirculatory parameters, with the point of maximum curvature on the trend line identified as the NE dose inflection point. APACHE II, Acute Physiology and Chronic Health Evaluation II; GAM, generalized additive model; HI, heterogeneity index; IL-6, interleukin-6; Lac, lactate; MFI, microvascular flow index; NE, norepinephrine; PPV, proportion of perfused vessels.

Relationship between dynamic changes in NE dose and microcirculatory function

Longitudinal analysis was performed using data from Day 1 and Day 3 to evaluate the dynamic relationship between changes in NE dose (ΔNE dose = NE dose D3 – NE dose D1) and changes in microcirculatory parameters (Δmicrocirculatory parameters = microcirculatory parameters D3- microcirculatory parameters D1). The correlation analysis showed that changes in NE dose were significantly positively correlated with changes in Lac and mottling score (ΔLac: r = 0.653, P < 0.001; Δmottling score: r = 0.483, P < 0.001), and significantly negatively correlated with changes in MFI (r = −0.565, P < 0.001). These correlations remained significant after FDR correction (P values for all parameters < 0.001) (Fig. 4). After adjusting for APACHE II scores, heart rates, and IL-6 levels, multivariate regression analysis further revealed significant correlations between changes in NE dose and changes in Lac, MFI, and mottling scores (ΔLac: P < 0.001; ΔMFI: P < 0.001; Δmottling score: P = 0.003). No statistically significant association was observed between NE dose changes and alterations in PPV, PI, HI, TVD, PVD, or CRT (all P > 0.05 for correlation and regression after adjustment) (Fig. 5).

Scatter plots of the association between NE dose changes and microcirculatory-related parameter alterations.
Fig. 4  Scatter plots of the association between NE dose changes and microcirculatory-related parameter alterations.

(a) Association between NE changes and MFI changes. (b) Association between NE changes and Lac changes. Spearman rank correlation coefficients were used. Lac, lactate; MFI, microvascular flow index; NE, norepinephrine.

Forest plot of the association between changes in NE dose and microcirculatory-related parameter alterations.
Fig. 5  Forest plot of the association between changes in NE dose and microcirculatory-related parameter alterations.

(Adjusted for APACHE II scores, heart rates, and IL-6 levels). APACHE II, Acute Physiology and Chronic Health Evaluation II; CI, confidence interval; CRT, capillary refill time; HI, heterogeneity index; IL-6, interleukin-6; Lac, lactate; MFI, microvascular flow index; NE, norepinephrine; PI, perfusion index; PPV, proportion of perfused vessels; PVD, perfused vessel density; TVD, total vessel density.

Relationship between NE dose and prognosis

Survival analysis was conducted based on a universally recognized high NE dose (0.5 µg/kg/min) and NE cutoff dose (0.8 µg/kg/min), with cut-off values determined via GAM analysis (PPV: 0.71 µg/kg/min, HI: 0.72 µg/kg/min, MFI: 0.80 µg/kg/min). Given that MFI serves as the primary indicator for assessing microcirculatory perfusion, this value was selected as the prognostic cutoff point. Kaplan–Meier survival curves and log-rank tests indicated that overall survival rates in the high-dose group were significantly lower than those in the low-dose group in both subgroups (all the log-rank P < 0.05). (Fig. 6). Univariate Cox regression models showed that the risk of death was significantly higher in the high NE dose group compared to the low-dose group (cut-off value = 0.50 µg/kg/min: hazard ratio [HR] = 2.28, 95% confidence interval [CI]: 1.16–4.51, P = 0.017; cut-off value = 0.80 µg/kg/min: HR = 2.52, 95% CI: 1.78–4.99, P = 0.008). Remarkably, after adjusting for confounding factors, including APACHE II scores, heart rates, and IL-6 levels, the independent effect of NE dose (>0.8 µg/kg/min) remained statistically significant (fully adjusted HR = 1.32, 95% CI: 1.28–3.10, P = 0.039), whereas the effect of the high-dose group (>0.5 µg/kg/min) was no longer significant (fully adjusted HR = 1.05, 95% CI: 0.78–2.47, P = 0.083).

Kaplan–Meier survival curves for 28-day survival (a) and 90-day survival (b).
Fig. 6  Kaplan–Meier survival curves for 28-day survival (a) and 90-day survival (b).

Survival probabilities were stratified by different NE doses. The log-rank test was used to compare the survival differences between the groups. NE, norepinephrine.

Discussion

This prospective observational study systematically investigated the relationship between NE dose and microcirculatory perfusion in ICU septic patients through a combination of cross-sectional and longitudinal analysis, and further verified the independent association between NE dose and patient prognosis. The findings not only provide new evidence for understanding the dose-dependent effect of NE on microcirculation but also offer actionable insights for optimizing NE dosage strategies in sepsis management. Correlation and regression analysis consistently showed that NE dose was significantly correlated with microcirculation in ICU septic patients (HI, MFI, PPV, Lac, and mottling score), and this association remained stable after adjusting for multiple confounding factors, with the exception of the mottling score. The most significant finding of this study is the exploration of a nonlinear dose–response relationship between NE and microcirculation. When the NE dose exceeds the threshold range of 0.71–0.80 µg/kg/min, which is based on the GAM-derived inflection points of key microcirculatory parameters (PPV: 0.71 µg/kg/min, HI: 0.72 µg/kg/min, MFI: 0.80 µg/kg/min), microcirculation in septic patients deteriorates significantly. This threshold applies to mechanically ventilated adult patients with septic shock who have completed initial fluid resuscitation. Clinical application requires individualized adjustment based on the patient’s underlying disease and vascular responsiveness. However, this dosage range appears inconsistent in terms of the mottling score and lactate levels. In the GAM fitting model, the R2 value for NE and mottling score was 0.123, potentially due to factors such as subjective bias, temperature management, and sample size constraints. Moreover, when the NE dose exceeds 0.50 µg/kg/min, which is often considered a rather high NE dose, it is frequently combined with methylene blue or other vasopressors in clinical practice.33 This may influence lactate metabolism and NE dosage requirements.34 Longitudinal data further showed that an increase in NE dose was significantly negatively correlated with improvements in microcirculatory perfusion. Furthermore, this study identified an association between NE doses and patient prognosis. The multivariate Cox regression model confirmed that high-dose NE (cut-off value = 0.80 µg/kg/min) may be an independent risk factor for death. As a core catecholamine vasoactive agent, NE may exacerbate oxidative stress by accelerating systemic metabolism and inducing mitochondrial uncoupling.35 Oxidative stress stimulates excessive reactive oxygen species (ROS) production, mediating endothelial cell toxicity. Damaged endothelial cells further exacerbate oxidative stress by releasing proinflammatory mediators (e.g., TNF-α and IL-6) and downregulating antioxidant enzymes (e.g., superoxide dismutase [SOD] and catalase), creating a vicious cycle that ultimately promotes microthrombus formation and induces organ dysfunction.36,37 De Backer et al.38 and Kindermans et al.39 proposed the concept of “macrocirculation–microcirculation uncoupling” in septic shock, noting that normal macrocirculation hemodynamic parameters (e.g. MAP and CVP) do not guarantee adequate microcirculatory perfusion. Our data directly corroborate this perspective: although there was no statistically significant difference in MAP or CVP between the high- and low-dose NE groups, microcirculatory function deteriorated markedly in the high-dose group. This highlights the limitation of overreliance on macrocirculatory parameters to guide NE therapy. Clinicians may mistakenly consider normal MAP a sign of effective resuscitation, while microcirculatory damage persists. This aligns with findings from Auchet et al.40 and Reinikainen et al.,41 collectively emphasizing a key concept: in the treatment of septic shock, NE should not be regarded as a harmless supportive measure but rather a therapeutic drug that requires careful weighing and precise regulation. Further evidence from MARTIN C et al.42 indicated a significant increase in mortality risk when NE doses exceed 0.5–1.0 µg/kg/min, a range similar to our findings. This study attempted to refine the optimal dose threshold by exploring the relationship between microcirculation and NE. Van Genderen et al.43 previously proposed a resuscitation strategy guided by peripheral perfusion. Our findings conclude that dynamically adjusting NE dosage to optimize microcirculatory perfusion, combined with individualized dose titration based on lactate clearance rather than solely targeting MAP ≥ 65 mmHg,44,45 represents a safer and more personalized management strategy for treating septic patients. In summary, microcirculatory dysfunction induced by high-dose NE use may be a key mediating link in the increased risk of death in septic patients. Clinical practice should rationally employ microcirculatory monitoring technologies to achieve refined hemodynamic management of sepsis patients. This approach can prevent excessive NE from further deteriorating microcirculation, ultimately improving patient outcomes.

Admittedly, this study has several limitations. First, as a single-center observational study, residual confounding and bias cannot be entirely excluded, despite multivariate adjustments and multiple analytical methods. Second, the sample size was relatively limited. While meeting statistical requirements, it constrained subgroup analyses and precluded the use of a more appropriate restricted cubic spline analysis to precisely depict dose–response curves when exploring dose–response relationships. Finally, while preliminary exploration of the relationship between NE and microcirculation was conducted, the underlying mechanisms linking NE to microcirculation and mortality require further validation through basic experimental studies. Future research should employ multicenter randomized controlled trials to compare the proposed “microcirculation-guided NE therapy” with traditional mean arterial pressure-guided therapy, verifying its impact on 28-day and 90-day mortality rates. Large-scale cohort studies should be conducted to further define precise NE thresholds across different populations. Finally, integrating basic experiments with clinical research will elucidate the molecular mechanisms by which NE regulates microcirculation, providing a theoretical foundation for developing targeted intervention strategies.

Conclusions

This study showed that higher NE doses were associated with worse microcirculatory dysfunction and poorer prognosis. This indicates that when the NE infusion dose exceeds 0.71–0.80 µg/kg/min (PPV: 0.71 µg/kg/min, HI: 0.72 µg/kg/min, MFI: 0.80 µg/kg/min), microcirculatory dysfunction should be noted. This provides important evidence for the precise regulation of NE doses in sepsis management, emphasizing the significance of integrating microcirculation monitoring to avoid further deterioration caused by excessive NE use and thereby improve patient prognosis.

Supporting information

Supplementary material for this article is available at https://doi.org/10.14218/JTCCM.2026.00004 .

Supplementary Table 1

Summary of sublingual microcirculation parameters.

(DOCX)

Click here for additional data file.

Supplementary Table 2

The spearman correlation coefficients between NE and microcirculatory parameters.

(DOCX)

Click here for additional data file.

Declarations

Acknowledgement

None.

Ethical statement

The study protocol was approved by the Ethics Committee of Zhongnan Hospital of Wuhan University (No. 2025096K). The study was conducted in accordance with the Declaration of Helsinki (as revised in 2024). All enrolled patients or their legally authorized representatives provided written informed consent to participate in the study.

Data sharing statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Funding

This work was supported by the National Natural Science Foundation of China (Nos. 82272208 and 82572484 to ZP).

Conflict of interest

ZP is an associate editor-in-chief of Journal of Translational Critical Care Medicine. The authors declare no other conflicts of interest.

Authors’ contributions

Conceptualization (ZP, YL), methodology (CH), data curation (YD, HC), formal analysis (YD, HC), validation (YZ, HC), visualization (YD), writing—original draft (YD, YZ), writing—review and editing (YD, YZ, CH, YL), supervision (CH, ZP, YL), project administration (ZP), and funding acquisition (ZP). All authors read and approved the final manuscript.

References

  1. Perner A, Gordon AC, De Backer D, Dimopoulos G, Russell JA, Lipman J, et al. Sepsis: frontiers in diagnosis, resuscitation and antibiotic therapy. Intensive Care Med 2016;42(12):1958-1969 View Article PubMed/NCBI
  2. Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, Kievlan DR, et al. Global, regional, and national sepsis incidence and mortality, 1990-2017: analysis for the Global Burden of Disease Study. Lancet 2020;395(10219):200-211 View Article PubMed/NCBI
  3. Hernández G, Ospina-Tascón GA, Damiani LP, Estenssoro E, Dubin A, Hurtado J, et al. Effect of a Resuscitation Strategy Targeting Peripheral Perfusion Status vs Serum Lactate Levels on 28-Day Mortality Among Patients With Septic Shock: The ANDROMEDA-SHOCK Randomized Clinical Trial. JAMA 2019;321(7):654-664 View Article PubMed/NCBI
  4. De Backer D, Ricottilli F, Ospina-Tascón GA. Septic shock: a microcirculation disease. Curr Opin Anaesthesiol 2021;34(2):85-91 View Article PubMed/NCBI
  5. Trzeciak S, McCoy JV, Phillip Dellinger R, Arnold RC, Rizzuto M, Abate NL, et al. Early increases in microcirculatory perfusion during protocol-directed resuscitation are associated with reduced multi-organ failure at 24 h in patients with sepsis. Intensive Care Med 2008;34(12):2210-2217 View Article PubMed/NCBI
  6. 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):1097-1105 View Article PubMed/NCBI
  7. 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):304-377 View Article PubMed/NCBI
  8. Russell JA, Rush B, Boyd J. Pathophysiology of Septic Shock. Crit Care Clin 2018;34(1):43-61 View Article PubMed/NCBI
  9. Annane D, Ouanes-Besbes L, de Backer D, DU B, Gordon AC, Hernández G, et al. A global perspective on vasoactive agents in shock. Intensive Care Med 2018;44(6):833-846 View Article PubMed/NCBI
  10. Hernández G, Teboul JL, Bakker J. Norepinephrine in septic shock. Intensive Care Med 2019;45(5):687-689 View Article PubMed/NCBI
  11. Xiang H, Zhao Y, Ma S, Li Q, Kashani KB, Peng Z, et al. Dose-related effects of norepinephrine on early-stage endotoxemic shock in a swine model. J Intensive Med 2023;3(4):335-344 View Article PubMed/NCBI
  12. Ospina-Tascón GA, Aldana JL, García Marín AF, Calderón-Tapia LE, Marulanda A, Escobar EP, et al. Immediate Norepinephrine in Endotoxic Shock: Effects on Regional and Microcirculatory Flow. Crit Care Med 2023;51(8):e157-e168 View Article PubMed/NCBI
  13. Hamzaoui O, Georger JF, Monnet X, Ksouri H, Maizel J, Richard C, et al. Early administration of norepinephrine increases cardiac preload and cardiac output in septic patients with life-threatening hypotension. Crit Care 2010;14(4):R142 View Article PubMed/NCBI
  14. Uhel F, van der Poll T. Norepinephrine in Septic Shock: A Mixed Blessing. Am J Respir Crit Care Med 2020;202(6):788-789 View Article PubMed/NCBI
  15. Andreis DT, Singer M. Catecholamines for inflammatory shock: a Jekyll-and-Hyde conundrum. Intensive Care Med 2016;42(9):1387-1397 View Article PubMed/NCBI
  16. Stolk RF, van der Poll T, Angus DC, van der Hoeven JG, Pickkers P, Kox M. Potentially Inadvertent Immunomodulation: Norepinephrine Use in Sepsis. Am J Respir Crit Care Med 2016;194(5):550-558 View Article PubMed/NCBI
  17. He HW, Liu DW, Long Y. Shock resuscitation: macrocirculation-microcirculation couple (in Chinese). Zhonghua Yi Xue Za Zhi 2018;98(35):2781-2784 View Article
  18. Ince C. Hemodynamic coherence and the rationale for monitoring the microcirculation. Crit Care 2015;19(Suppl 3):S8 View Article PubMed/NCBI
  19. Kara A, Akin S, Ince C. Monitoring microcirculation in critical illness. Curr Opin Crit Care 2016;22(5):444-452 View Article PubMed/NCBI
  20. Lipinska-Gediga M. Sepsis and septic shock-is a microcirculation a main player?. Anaesthesiol Intensive Ther 2016;48(4):261-265 View Article PubMed/NCBI
  21. De Backer D, Donadello K, Sakr Y, Ospina-Tascon G, Salgado D, Scolletta S, et al. Microcirculatory alterations in patients with severe sepsis: impact of time of assessment and relationship with outcome. Crit Care Med 2013;41(3):791-799 View Article PubMed/NCBI
  22. Damiani E, Carsetti A, Casarotta E, Domizi R, Scorcella C, Donati A, et al. Microcirculation-guided resuscitation in sepsis: the next frontier?. Front Med (Lausanne) 2023;10:1212321 View Article PubMed/NCBI
  23. De Backer D, Ospina-Tascon G, Salgado D, Favory R, Creteur J, Vincent JL. Monitoring the microcirculation in the critically ill patient: current methods and future approaches. Intensive Care Med 2010;36(11):1813-1825 View Article PubMed/NCBI
  24. Ince C, Boerma EC, Cecconi M, De Backer D, Shapiro NI, Duranteau J, et al. Second consensus on the assessment of sublingual microcirculation in critically ill patients: results from a task force of the European Society of Intensive Care Medicine. Intensive Care Med 2018;44(3):281-299 View Article PubMed/NCBI
  25. Bruno RR, Wollborn J, Fengler K, Flick M, Wunder C, Allgäuer S, et al. Direct assessment of microcirculation in shock: a randomized-controlled multicenter study. Intensive Care Med 2023;49(6):645-655 View Article PubMed/NCBI
  26. Dubin A, Henriquez E, Hernández G. Monitoring peripheral perfusion and microcirculation. Curr Opin Crit Care 2018;24(3):173-180 View Article PubMed/NCBI
  27. Bezemer R, Bartels SA, Bakker J, Ince C. Clinical review: Clinical imaging of the sublingual microcirculation in the critically ill—where do we stand?. Crit Care 2012;16(3):224 View Article PubMed/NCBI
  28. Ince C. The microcirculation unveiled. Am J Respir Crit Care Med 2002;166(1):1-2 View Article PubMed/NCBI
  29. Massey MJ, Larochelle E, Najarro G, Karmacharla A, Arnold R, Trzeciak S, et al. The microcirculation image quality score: development and preliminary evaluation of a proposed approach to grading quality of image acquisition for bedside videomicroscopy. J Crit Care 2013;28(6):913-917 View Article PubMed/NCBI
  30. Ince C. The microcirculation is the motor of sepsis. Crit Care 2005;9(Suppl 4):S13-S19 View Article PubMed/NCBI
  31. Bauer M, Gerlach H, Vogelmann T, Preissing F, Stiefel J, Adam D. Mortality in sepsis and septic shock in Europe, North America and Australia between 2009 and 2019- results from a systematic review and meta-analysis. Crit Care 2020;24(1):239 View Article PubMed/NCBI
  32. Ruslan MA, Baharuddin KA, Noor NM, Yazid MB, Noh AYM, Rahman A. Norepinephrine in Septic Shock: A Systematic Review and Meta-analysis. West J Emerg Med 2021;22(2):196-203 View Article PubMed/NCBI
  33. Oczkowski S, Alshamsi F, Belley-Cote E, Centofanti JE, Hylander Møller M, Nunnaly ME, et al. Surviving Sepsis Campaign Guidelines 2021: highlights for the practicing clinician. Pol Arch Intern Med 2022;132(7-8):16290 View Article PubMed/NCBI
  34. Zhao CC, Zhai YJ, Hu ZJ, Huo Y, Li ZQ, Zhu GJ. Efficacy and safety of methylene blue in patients with vasodilatory shock: A systematic review and meta-analysis. Front Med (Lausanne) 2022;9:950596 View Article PubMed/NCBI
  35. Hartmann C, Radermacher P, Wepler M, Nußbaum B. Non-Hemodynamic Effects of Catecholamines. Shock 2017;48(4):390-400 View Article PubMed/NCBI
  36. Joffre J, Hellman J. Oxidative Stress and Endothelial Dysfunction in Sepsis and Acute Inflammation. Antioxid Redox Signal 2021;35(15):1291-1307 View Article PubMed/NCBI
  37. Srdić T, Đurašević S, Lakić I, Ružičić A, Vujović P, Jevđović T, et al. From Molecular Mechanisms to Clinical Therapy: Understanding Sepsis-Induced Multiple Organ Dysfunction. Int J Mol Sci 2024;25(14):7770 View Article PubMed/NCBI
  38. De Backer D, Creteur J, Dubois MJ, Sakr Y, Koch M, Verdant C, et al. The effects of dobutamine on microcirculatory alterations in patients with septic shock are independent of its systemic effects. Crit Care Med 2006;34(2):403-408 View Article PubMed/NCBI
  39. Kindermans M, Joachim J, Manquat E, Levé C, Hong A, Mateo J, et al. Micro- and macrocirculatory effects of norepinephrine on anaesthesia-induced hypotension: a prospective preliminary study. BMC Anesthesiol 2023;23(1):374 View Article PubMed/NCBI
  40. Auchet T, Regnier MA, Girerd N, Levy B. Outcome of patients with septic shock and high-dose vasopressor therapy. Ann Intensive Care 2017;7(1):43 View Article PubMed/NCBI
  41. Reinikainen M, Delamarre L, Blaser AR, Hollenberg SM, Lobo SM, Rezende E, et al. Association of noradrenaline dose with mortality in critically ill patients: a systematic review and dose-response meta-analysis. Crit Care 2025;29(1):498 View Article PubMed/NCBI
  42. Martin C, Medam S, Antonini F, Alingrin J, Haddam M, Hammad E, et al. NOREPINEPHRINE: NOT TOO MUCH, TOO LONG. Shock 2015;44(4):305-309 View Article PubMed/NCBI
  43. van Genderen ME, Engels N, van der Valk RJ, Lima A, Klijn E, Bakker J, et al. Early peripheral perfusion-guided fluid therapy in patients with septic shock. Am J Respir Crit Care Med 2015;191(4):477-480 View Article PubMed/NCBI
  44. Kato R, Pinsky MR. Personalizing blood pressure management in septic shock. Ann Intensive Care 2015;5(1):41 View Article PubMed/NCBI
  45. De Backer D, Cecconi M, Lipman J, Machado F, Myatra SN, Ostermann M, et al. Challenges in the management of septic shock: a narrative review. Intensive Care Med 2019;45(4):420-433 View Article PubMed/NCBI
Copyright © 2026 Authors. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial 4.0 License (CC BY-NC 4.0), permitting all non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

About this Article

Cite this article
Dai Y, Zhu Y, Hu C, Chen H, Peng Z, Li Y. Nonlinear Dose–response Relationship between Norepinephrine and Microcirculatory Perfusion: Identifying the Tipping Point in Septic Shock. J Transl Crit Care Med. 2026;8(2):e00004. doi: 10.14218/JTCCM.2026.00004.
Copy        Export to RIS        Export to EndNote
Article History
Received Revised Accepted Published
January 31, 2026 February 27, 2026 March 12, 2026 June 29, 2026
DOI http://dx.doi.org/10.14218/JTCCM.2026.00004
  • Journal of Translational Critical Care Medicine
  • pISSN 2665-9190
  • eISSN 2590-3438
  • © 2026 Xia & He Publishing Inc.
Back to Top

Nonlinear Dose–response Relationship between Norepinephrine and Microcirculatory Perfusion: Identifying the Tipping Point in Septic Shock

Yiping Dai, Yao Zhu, Chang Hu, Hui Chen, Zhiyong Peng, Yiming Li
  • Reset Zoom
  • Download TIFF