Effects of metaraminol and norepinephrine on hemodynamics and kidney function in a miniature pig model of septic shock

Ethics statement

All animals (male Guizhou miniature pigs weighing 29–34 kg) were housed in separate cages at 23℃–28℃ and 30%– 60% humidity, with a 12-h light/12-h dark cycle and free access to food and water. All experiments were subjected to approval by the Faculty of Medicine of University of Peking Animal Ethics Committee (LA2019002, 2019/1/14) and performed according to local guidelines.

Septic animal models

This is an open-label, randomized controlled study. Fifteen male Guizhou miniature pigs were used for this study. Experimental protocols were similar to those previously described in similar research,^[19] with slight modifications. Briefly, Pulse-indicator Continuous Cardiac Output (PICCO) Monitoring Kits (PULSION Medical Systems SE, Germany) and triple lumen central venous catheters (CVC) (Arrow International, Inc., Czech Republic) were inserted in the femoral artery and internal jugular vein, respectively, to monitor hemodynamic changes and collect blood samples, and cystostomy was used to drain urine (Figure 1A). The animals were randomly assigned to three groups, using online random number generators (Graphpad Random number calculators, https://www.graphpad.com/quickcalcs/randMenu/). In septic animals (n = 10), peritonitis was induced by 1 g/kg of autologous feces gotten from the colon where it joins the cecum. Autologous feces dissolved in 200 mL glucose 5% was instilled into the abdominal cavity. In sham-operated animals (n = 5), the autologous feces collection was not performed, but received glucose 5% injection. After surgery, the animals were maintained throughout the entire experiment in anesthesia with assisted ventilation, and the temperature of the warming blanket was maintained at 37℃. A combination of compound sodium chloride injection and glucose 5% solutions was administered continuously throughout the experiment at a combined total rate of 5 mL·kg^-1·h^-1. All animals were continuously monitored by two experienced researchers with an intensive care medicine background. Eventually the animals were euthanized under deep anesthesia at 3 h after the shock, blood was collected and stored at -80℃, and the kidney samples were harvested and stored in liquid nitrogen until further analysis.

Figure 1

Arrangements for the operation (A) and protocol timeline (B). Endotracheal intubation. Triple lumen central venous catheter. PICCO. Cystostomy. Access to autologous feces from the colon where it joins ④ ⑤ ② ③ ① the cecum. At the end of the operation and after hemodynamic stabilization, baseline measurements were obtained. Blood pressure and heart rate were monitored continuously. The time point at septic shock was defined as the mean arterial pressure (MAP) < 65 mmHg. After 30 min, fluid resuscitation was started with continued infusion of norepinephrine or metaraminol as previously randomized. The last time point was 3 h after the administration of pressor drugs. PICCO: pulse-indicator continuous cardiac output.

Hemodynamic monitoring and vasopressor treatment

Mean arterial blood pressure (MAP), heart rate (HR), cardiac stroke volume, and oxygen saturation were monitored continuously. When the MAP of the 10 experimental animals dropped below 65 mmHg and was maintained for 30 min, we considered them to be in septic shock and began to give crystalloids (30 mL/kg) and vasopressors. The animals were randomly assigned to receive one of the two drugs, norepinephrine or metaraminol, to restore and maintain MAP ≥ 65 mmHg. The 5 animals were given norepinephrine at an initial dose of 0.1 (μg·kg^-1·min^-1), which was increased by 0.1 unit every 10 min until the mean MAP ≥ 65 mmHg for more than 30 min. The 5 animals in the experimental group were treated with metaraminol at the initial dose of 1 (μg·kg^-1·min^-1), which was increased by 1 unit every 10 min. Monitoring was continued thereafter, and if blood pressure decreased subsequently, vasopressors were continued to be upregulated in the same manner until 3 h after shock (Figure 1B).

Conversion dose ratio and heart rate variability

The conversion dose ratio between metaraminol and norepinephrine refers to a method previously described in another research.^[20] In this study, R 4.0.3 statistical software was used for statistical analysis, MatchIt package was used for propensity score matching, Logistic was used to establish the propensity score model, and the neighbor matching method was used for matching. The matching variables were basal mean arterial pressure, shock time, and basal blood pressure combined with shock time, and the matching caliper value was 0.2. The matching ratio between the experimental group and the control group was 1:1. The conversion dose ratio was calculated using the following formula: metaraminol infusion dose (μg·kg^-1·min^-1)/ norepinephrine infusion dose (μg·kg^-1·min^-1). The HR after administration of the vasoactive drugs was recorded hourly, and the variation of HR was calculated using SPSS 23.0 software.

Enzyme-linked immunosorbent assay (ELISA)

Endothelin-1 (ET-1) was measured by ELISA (EK0945-PO, Boster, Wuhan, China).

Assessment of kidney functions

Concentrations of serum creatinine were measured using a creatinine assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

Assessment of kidney histology

The kidney tissue isolated from the pigs was washed with phosphate-buffered saline (PBS). After fixation with 10% neutral formalin, the kidney tissues embedded in paraffin, and cut into 4-μm-thick sections for haematoxylin and eosin (HE) staining. HE staining of kidney tissues was examined by microscopy, and the images were analyzed by computerised digital image analysis (NDP view 2.0). The tubular injury was examined by the percentage of damaged tubules ^[21]: grade 0, no damage; grade 1, < 25%; grade 2, 25%–49%; grade 3, 50%–75%; and grade 4, > 75%.

Western blotting

The protein was extracted from the kidney tissue with radioimmunoprecipitation (RIPA) lysis buffer containing protease inhibitor in order to protect the protein from degradation. The supernatant was then collected and the protein concentration was measured by a bicinchoninic acid (BCA) protein assay kit. The samples were subjected to SDS-PAGE, and then protein was transferred to blotting membrane. The membrane blocking was then done with 5% nonfat milk and incubated at room temperature for 2 h. The blot was incubated with a specific primary antibody at 4°C overnight, which was followed by incubating the membranes with horseradish-peroxidase (HRP) conjugated secondary antibody for 2 h at ambient temperature. Finally, the membranes were photographed to X-ray film by an enhanced electrochemiluminescence (ECL) substrate kit from Yeasen Biotechnology, Shanghai, China. The relative expression ratios for the control and experimental groups were analyzed based on density by Image J software and the β-Actin signal as a reference. Antibodies were used against KIM-1 (bs-2713R; Bioss, Beijing, China), and β-actin (3700S; CST, Shanghai, China).

qRT PCR assay

RNA extraction and quantitative real-time (qRT) polymerase chain reaction (PCR) total RNA was extracted using TRIzol Reagent (Ambion, Carlsbad, CA), according to the manufacturer’s instructions, and RNA concentration was determined by Nanodrop. RNA was reverse transcribed in Biometra TAdvanced 96 SG (Analytikjena, Germany). The levels of neutrophil gelatinase-associated lipocalin (NGAL), IL-6, TNF-α, IL 33, CXCL10, COL1A1, and GAPDH were determined by Archimed v1.3.3. The primers are described in this paper’s supplemental material (see Table S1). The relative levels of target genes were analyzed by the 2–ΔΔCt method, normalized to the GAPDH, and compared with controls.

Statistical analysis

After testing for normality of distribution by the Shapiro-Wilk test, the continuous data were presented as means ± standard deviation (SD) or medians (25th–75th) depending on the distribution. Independent samples Student’s t-test and Mann-Whitney U test were used to calculate P values for continuous variables. MAP, heart rate, cardiac output (CO), central venous pressure (CVP), urine output, and creatinine in norepinephrine and metaraminal were analyzed using two-way repeated measures analysis of variance (ANOVA). If the normal distribution and homogeneous variance are not satisfied, the Scheirer-Ray-Hare test was used as an alternative method by using R 4.2.2 with the “Rcompanion” package. The analysis was performed using SPSS version 23.0 for windows (IBM Crop., Chicago, USA). A value of P < 0.05 (corrected, 2-sided) was considered to be statistically significant.

Septic shock induction

All 10 animals in the two treatment groups developed severe hypotension (P < 0.01) and tachycardia (P = 0.006) compared to the baseline (Figure 2A and 2B). The mean time to reach the septic shock timepoint was similar in both treatment groups (P = 0.22) (Table 1). There were no statistically significant differences in MAP, HR, CO, global ejection fraction (GEF), or systemic vascular resistance (SVR) between the treatment groups until the administration of the vasopressor (Table 1).

Figure 2

Hemodynamic variables in the three groups. Changes in MAP (A) and heart rate (B) throughout the experiment. T0.5 is 0.5 h after administration of the drugs, T1 is 1 h, T1.5 is 1.5 h, T2 is 2 h, T2.5 is 2.5 h, and T3 is 3 h. MAP: mean arterial pressure.

Hemodynamics indices

At shock, the MAP was 60 ± 4 mmHg in the metaraminol group, and 58 ± 6 mmHg in the norepinephrine group, which declined significantly than the sham group (Figure 3A). The MAP was restored to 65 mmHg in 30 min with metaraminol and 42 min with NE (Figure 3B), namely the target (MAP ≥ 65 mmHg) blood pressure was reached faster with metaraminol. However, there were no statistically significant differences between the two sepsis groups at every timepoint (F = 1.009, P = 0.435) (Figure 2A). The mean dose required to maintain MAP ≥ 65 mmHg with NE was 0.38 ± 0.13 μg·kg^-1·min^-1, and it was 3.00 ± 1.73 μg·kg^-1·min^-1 with metaraminol. The total dose between the septic shock timepoint and the end of the experiment was 70.6 ± 25.5 μg/kg with NE and 510.0 ± 253.4 μg/kg with metaraminol.

Figure 3

MAP and heart rate in the three groups. A. Comparison of map while in shock in two treatment groups with shams. B. Time to reach MAP ≥ 65 mmHg after shock. C. Comparison of heart rate in shock. D. Changes in heart rate from shock to 3 h after treatment. MAP: mean arterial pressure.

The mean HR at shock was 196 ± 24 beats/min in the metaraminol group and 188±32 beats/min in the NE group, which increased significantly than the sham group (Figure 3C). The mean HR at 3 h after shock was 168 ± 26 beats/min with metaraminol and 176 ± 43 beats/min with NE; there were no statistically significant differences between the two treatment groups at every timepoint (F = 0.007, P = 0.937) (Figure 2B). However, after 3 h of treatment, the HR declined significantly with metaraminol (P = 0.023) (Figure 3D). No arrhythmias such as atrial fibrillation and ventricular fibrillation presented during continuous monitoring.

At shock, the CO was 1.71 ± 0.55 L/min in the metaraminol group, and 1.27 ± 0.91 L/min in the norepinephrine group, which declined significantly than the sham group (P = 0.032) (Figure 4A) and the timepoint of baseline (P = 0.012) (Figure 4B). After treatment, CO was higher in both sepsis groups than in shock. However, there were no statistically significant differences between the two sepsis groups at every timepoint (F = 0.293, P = 0.608) (Figure 4C). For the CVP, there were no statistically significant differences between the sepsis and sham groups (P = 0.794) (Figure 4D), and the time of baseline and shock (P = 0.915) (Figure 4E). There were also no statistically significant differences between the two sepsis groups at every timepoint (H = 0.220, P = 0.639) (Figure 4F). The change of CVP has no time dependent effect (H = 0.094, P = 0.954).

Figure 4

CO and CVP in the three groups. A. Comparison of CO between in sepsis animals and shams. B. Comparison of CO between shock and baseline. C. The dynamic change over time of CO. D. Comparison of CVP in sepsis animals and shams. E. Comparison of CVP between shock and baseline. F. The dynamic change over time of CVP. CO, Cardiac Output; CVP, Central Venous Pressure.

Evaluation of ET-1 expression

The serum ET-1 concentration for the experimental animals in shock was significantly higher than that of the baseline (P = 0.04) (see Figure S1A). There were no statistically significant differences in serum ET-1 concentration between the metaraminol and norepinephrine groups at shock (P = 0.261) (see Figure S1B), and nor at 3 h after treatment (P = 0.101) (see Figure S1C).

Evaluation of kidney injury

The serum creatinine (Cr) at shock was significantly higher than baseline (P < 0.001) (Figure 5A), and was also significantly higher in the sepsis animals than in the shams (P = 0.007) (Figure 5B). There were no statistically significant differences between the metaraminol and norepinephrine group in the serum Cr at shock (P = 0.909) (Figure 5C). The Cr has different trends over time in the two groups (F = 9.890, P = 0.014), and the serum Cr in the metaraminol group was significantly lower than in the norepinephrine group after 3 h of treatment (F = 34.262, P = 0.017) (Figure 5C).

Figure 5

Expression of serum Cr and by shock and treatment. A. Comparison of Cr between shock and baseline. B. Comparison of Cr between in sepsis animals and shams. C. The dynamic change over time of creatinine. D. Comparison of the values of urine output between shock and baseline. E. Comparison of the values of urine output in sepsis animals and shams. F. The dynamic change over time of urine output. Cr, creatinine.

Urine output is very crucial for evaluating renal function. The values of urine output decreased significantly over time (F = 30.756, P < 0.001). The values of urine output at shock were significantly reduced than baseline (P = 0.005) (Figure 5D) and sham group (P = 0.018) (Figure 5E), but there were no statistically significant differences between the metaraminol and norepinephrine group in the values of urine output at shock (P = 0.251) and treatment (P = 0.463) (Figure 5F).

HE staining of the cortex showed that kidney injury with intraepithelial vacuolar degeneration and nucleolus had disappeared and that the inflammatory cells were mainly distributed in the renal sinuses rather than in the renal parenchyma at shock (Figure 6A-6C). Quantitative analysis of the tubular injury score of metaraminol and norepinephrine group renal tissues was 2.2 ± 0.45 and 1.8 ± 0.84, compared with 0.8 ± 0.45 for sham group kidneys. The tubular injury score was significantly higher in the sepsis animals than in the shams (P = 0.045 and P = 0.005), but there were no statistically significant differences between the metaraminol and norepinephrine group (P = 0.343) (Figure 6D).

Figure 6

Cardinal features of acute kidney injury. Representative histology and pathological scores of tubular injury in kidney cortex by HE staining in sham group (A), norepinephrine group (B) and metaraminol group (C). The arrows indicate magnified views of the tubular injury. D. The tubular injury score in three groups.

The expression of kidney injury molecule 1 (KIM-1) protein, a marker of renal tubular injury, was detected by western blot. The results showed that the expression of KIM-1 protein in the kidney tissue of the sepsis animals was increased compared with that of the shams (P = 0.025), but there was no significant difference in the expression of KIM-1 protein between metaraminol and norepinephrine groups (P = 0.665) (Figure 7). Shock significantly increased the mRNA expression of NGAL in pig kidney tissue, suggesting the occurrence of acute kidney injury, but there was no statistically significant difference between the two treatment groups (P = 0.564). Similarly, mRNA expressions of COL1A1, TNF, IL6, IL33, and CXCL10 in kidney tissues showed no statistically significant differences between the treatment groups (P > 0.05) (Figure 8A-8F).

Figure 7

KIM-1 expression in kidney.

Figure 8

Renal tissue relative mRNA expression of NGAL (A), IL6 (B), TNF (C), COL1A1 (D), IL33 (E), and CXCL10 (F). Relative quantification was achieved using the comparative 2-ΔΔCt method by normalization with the GAPDH. Results are expressed as relative fold increase above the mean value of renal tissue mRNA expression of the sham group, arbitrarily fixed at 1. NGAL: neutrophil gelatinase-associated lipocalin; IL6: interleukin-6; TNF: tumor necrosis factor; COLIA1: collagen Iα1; IL33: interleukin-33; CXCL: chemokine (C-X-C motif) ligand.

Conversion dose ratio

The median infusion speed to reach MAP ≥ 65 mmHg was 2 μg·kg^-1·min^-1 for metaraminol treatment and 0.4 μg·kg^-1·min^-1 for norepinephrine. Based on the basal MAP, the result of the propensity scores of the experimental group and the control group after matching were 5 and 5. The median conversion dose ratio between metaraminol and norepinephrine was 6 (IQR 4–20). Based on the time to reach shock, the scores of the matching were 4 and 4, and the median conversion dose ratio of 7 (IQR 4–30). Based on both indicators, the propensity scores of the matching were 3 and 3, and the median conversion dose ratio was 6 (IQR 4–6.67). Thus, the median ratio is estimated to be between 6 and 7. Pragmatically, this can be estimated to a ratio of 6: 1 for ease of calculations and can be considered appropriate for the dose equivalence calculation of metaraminol (μg·kg^-1·min^-1): norepinephrine (μg·kg^-1·min^-1).