Mean hemorrhage volume was 745 (250.8) mL, representing 18.3 (6.1) % of the estimated total blood volume. Mean SAP at the end of the hemorrhagic phase was 59.8 (14.8) mmHg. Two animals died in the control group, compared with three treated with IM AVP and one in the IM terlipressin group. No animals died in the IV terlipressin group. The difference in mortality was not statistically significant (χ2 = 6.453, df = 3, p = 0.09). One animal was excluded due to intercurrent illness. Median core temperature at baseline was 38.3 ˚C (IQR 37.9; 38.7) and increased to 40.2 ˚C (IQR 39.7; 40.5) by the end of the experiment. The lowest recorded body temperature across all animals was 37.4 ˚C.

Hemodynamics

IV terlipressin significantly increased SAP compared to time of administration, difference of mean to controls: 24.4 mmHg (CI 6.8; 42.1, p = 0.01) (Fig. 2a). The increase in SAP following IV terlipressin was significant within 15 min after administration (p = 0.005). In contrast, SAP remained unaffected by administration of IM AVP and IM terlipressin compared to time of administration, difference of means: -5.6 mmHg (CI -22.5; 11.3, p = 0.5) respectively 5.1 mmHg (CI -11.6; 21.8, p = 0.51)(Fig. 2a). Additionally, IV terlipressin significantly increased MAP after 15 min by a mean of 44.2 (16.1) mmHg, compared to 3.2 (11.1) mmHg in controls (p = 0.001). The effects on MAP persisted for 30 min, and were no longer significant 45 min after administration, p = 0.06. Neither IM AVP nor IM terlipressin affected MAP, difference of means: -6.7 mmHg (CI -19.1; 5.7, p = 0.26), respectively 7.9 mmHg (CI -10.5; 26.3, p = 0.36). SVRI increased following administration of IV terlipressin compared to controls, difference of means: 44,593 dynes*s/cm− 5*kg (CI -17556; 71630, p = 0.004) (Fig. 2b). The increase in SVRI following IV terlipressin was significant within 15 min following administration (p < 0.001). SVRI remained unaffected by administration of IM AVP and IM terlipressin, difference of means: 11,505 dynes*s/cm− 5*kg (CI -11915; 34924, p = 0.3), respectively 20,748 dynes*s/cm− 5*kg (CI -5530; 47027, p = 0.11). In controls, SVRI decreased by the end of the experiment compared to baseline (p = 0.01), and SAP decreased from a baseline median of 124.1 mmHg (IQR 92.0; 127.9) compared to 34.0 mmHg (IQR 26.8; 55.8) at the end of the experiment. CI remained unaffected in all groups after administration compared to controls, difference of means: IM AVP − 0.009 L/min/kg (CI -0.03; 0.0082, p = 0.26), IV terlipressin 0.002 L/min/kg (CI -0.01; 0.02, p = 0.77), IM terlipressin 0.002 L/min/kg (CI -0.02; 0.002, p = 0.82), (Fig. 2c). SV was unaffected by treatment comapred to controls, difference of means: AVP IM -0.5 mL (CI -10.7; 9.6, p = 0.91), IV terlipressin 5.8 mL (CI -2.0; 13.7, p = 0.13), respectively IM terlipressin 4.3 mL (CI -5.9; 14.5, p = 0.37) (Fig. 2d). Heart rate decreased after administration of IV terlipressin and IM terlipressin compared to controls, difference of means: -18.4/min (CI -32.4; 4.3, p = 0.02), respectively − 20.0/min (CI -38.8; 1.3, p = 0.04) (Fig. 2e). However, administration of IM AVP showed no effect on HR compared to controls, with a difference of means of -13.1/min (CI − 41.4; 15.3, p = 0.33) (Fig. 2e). Central venous pressure remained unaffected in all groups after administration compared to controls, differences of mean: IM AVP − 0.9 mmHg (CI -7.9; 6.1, p = 0.77), IV terlipressin − 2.9 mmHg (CI -8.9; 3.0, p = 0.3), and IM terlipressin − 2.0 mmHg (CI -8.4; 4.3, p = 0.49). Pulmonary wedge pressure remained unaffected after administration compared to controls, difference of means: IV terlipressin − 0.7 mmHg (CI -2.2; 0.8, p = 0.31), and IM terlipressin 0.3 mmHg (CI -1.5; 2.1, p = 0.72). Pulmonary wedge pressure remained unaffected by IM AVP when compared to the time of administration (p = 0.86). Intravenous terlipressin increased urine output at the end of the experiment compared to controls, 0.6 mL/kg/h (0.19) vs. 0.12 mL/kg/h (0.14), p = 0.001, (Fig. 2f). Urine output at the end of the experiment was unaffected by IM AVP and IM terlipressin when compared to controls, 0.04 mL/kg/h (0.02), p = 0.2, respectively 0.18 mL/kg/h (0.15), p = 0.51, (Fig. 2f).

Fig. 2The alternative text for this image may have been generated using AI.

Hemodynamic effects. Systolic arterial pressure (SAP) (a) and systemic vascular resistance index (SVRI) (b) across treatment groups. IV terlipressin induced a rapid and significant increase in both SAP and SVRI within 15 min compared with controls, whereas IM AVP and IM terlipressin showed no meaningful hemodynamic effect. Cardiac index (c), stroke volume (d), and heart rate (e) remained largely unchanged across groups. IV terlipressin significantly increased urine output (f). Abbreviations: AVP, arginine vasopressin; IM, intramuscular; IV, intravenous; SAP, systolic arterial pressure; SVRI, systemic vascular resistance index. *p < 0.05, **p < 0.01

Respiration

PVRI was unaffected between all groups compared to controls, difference of means: IM AVP 5831 dynes*s*cm− 5*kg (CI -10595; 22258, p = 0.45), IV terlipressin − 9620 dynes*s*cm− 5*kg (CI -24003; 4763, p = 0.17), and IM terlipressin − 2789 dynes*s*cm− 5*kg (CI -22289; 16711, p = 0.76) (Fig. 3a). Intravenous terlipressin increased mixed venous oxygen saturation after administration compared to controls, difference of means 14.6% (CI 2.9; 26.4, p = 0.02), while unaffected by IM AVP and IM terlipressin compared to time of administration, difference of means: 2.5% (CI -16.8; 21.8, p = 0.78) respectively − 7.9% (CI -26.2; 10.3, p = 0.36) (Fig. 3b). Systolic pulmonary arterial pressure remained stable when compared to time of administration, differences of means: AVP IM -0.5 mmHg (CI -4.8; 3.9, p = 0.8), IV terlipressin − 1.4 mmHg (CI -6.4; 3.5, p = 0.54), and IM terlipressin 0.5 mmHg (CI -4.4; 5.4, p = 0.81)(Fig. 3c). Oxygen consumption index was unaffected in all groups compared to controls: differences of means: IM AVP − 0.2 mL/min/kg (CI -1.8; 1.3, p = 0.74), IV terlipressin 0.005 mL/min/kg (CI -1.5; 1.5, p = 0.99), and IM terlipressin 0.3 mL/min/kg (CI -1.3; 1.8, p = 0.72) (Fig. 3d). Similarly, oxygen delivery index did not differ between the groups and controls, differences of means: IM AVP − 1.4 mL/min/kg (CI -3.7; 0.9, p = 0.20), IV terlipressin 0.9 mL/min/kg (CI -1.3; 3.1, p = 0.39), and IM terlipressin 0.7 mL/min/kg (CI -2.0; 3.5, p = 0.58) (Fig. 3e). Extraction of oxygen was stable after intervention when compared to time of administration, differences of means: IM AVP 0.009% (CI -0.2; 0.2, p = 0.9), IV terlipressin − 0.09% (CI -0.2; 0.03, p = 0.13), and IM terlipressin 0.02% (CI -0.17; 0.2, p = 0.82) (Fig. 3f). Lung compliance remained unaffected in all groups after administration compared to controls: differences of mean: IM AVP 1.6 mL/cm H2O (CI -4.5; 7.8, p = 0.57), IV terlipressin 0.2 mL/cm H2O (CI -6.1; 6.6, p = 0.94), and IM terlipressin 1.3 mL/cm H2O (CI -7.0; 9.7, p = 0.73).

Fig. 3The alternative text for this image may have been generated using AI.

Respiratory and oxygenation parameters. Pulmonary vascular resistance index (PVRI) (a) did not differ between groups. Mixed venous oxygen saturation (SvO₂) (b) was significantly higher with IV terlipressin compared with controls, with no effect observed after IM AVP or IM terlipressin. Systolic pulmonary arterial pressure (c), oxygen consumption index (d), oxygen delivery index (e), and oxygen extraction ratio (f) were similar across all treatment groups and controls. Abbreviations: AVP, arginine vasopressin; IM, intramuscular; IV, intravenous; *p < 0.05

Blood gases

pH remained unaffected in all groups after administration, compared to controls; differences in means: IM AVP − 0.03 (CI -0.13; 0.06, p = 0.41), IV terlipressin 0.06 (CI -0.01; 0.1, p = 0.09), and IM terlipressin 0.03 (CI -0.04; 0.1, p = 0.35). No differences were observed in PaO2 levels, differences of means: IM AVP − 0.1 kPa (CI -1.7; 1.5, p = 0.89), IV terlipressin 0.6 kPa (CI -0.4; 1.7, p = 0.2), and IM terlipressin 0.06 kPa (CI -1.1; 1.2, p = 0.91) (Fig. 4a). Similarly, PaCO2 was unaffected by the interventions, differences of means: IM AVP − 0.5 kPa (CI -1.3; 0.3, p = 0.21), IV terlipressin − 0.2 kPa (CI -0.9; 0.4, p = 0.46), and IM terlipressin 0.02 kPa (CI -0.9; 0.9, p = 0.97) (Fig. 4b). Base excess decreased throughout the experiment but did not differ between groups and controls, differences of means: IM AVP − 3.5 mmol/L (CI -9.5; 2.4, p = 0.21), IV terlipressin 2.3 mmol/L (CI -2.8; 7.5, p = 0.34), and IM terlipressin 1.6 mmol/L (CI -4.3; 7.5, p = 0.56) (Fig. 4c). Lactate levels increased throughout the experiment but did not differ compared to controls, differences of means: IM AVP 2.1 mmol/L (CI -1.4; 5.6, p = 0.21), IV terlipressin − 1.2 mmol/L (CI -3.8; 1.4, p = 0.31), and IM terlipressin − 0.6 mmol/L (CI -3.5; 2.2, p = 0.64) (Fig. 4d). Additionally, there were no differences in sodium levels compared to controls, differences of means: IM AVP 1.0 mmol/L (CI -0.5; 2.5, p = 0.16), IV terlipressin 1.2 mmol/L (CI -1.0; 3.3, p = 0.25), and IM terlipressin 0.4 mmol/L (CI -1.4; 2.1, p = 0.67). Potassium remained unaffected compared to controls, differences of means: IM AVP 0.5 mmol/L (CI -0.8; 1.7, p = 0.44), IV terlipressin − 0.5 mmol/L (CI -1.9; 0.6, p = 0.28), and IM terlipressin − 0.1 mmol/L (CI -1.4; 1.3, p = 0.89). Glucose did not differ between groups and controls, differences of means: IM AVP − 2.3 mmol/L (CI -7.6; 3.0, p = 0.35), IV terlipressin − 0.1 mmol/L (CI -5.6; 5.4, p = 0.96), and IM terlipressin − 0.6 mmol/L (CI -5.7; 4.6, p = 0.8) (Fig. 4e). In total, 12 animals became hypoglycemic (< 4.0 mmol/L) and required glucose substitution, 2 of controls, 4 of IM AVP, 3 of IV terlipressin, and 3 of IM terlipressin. There were no differences between groups in the number of hypoglycemic animals, IM AVP vs. controls (p = 0.24), IV terlipressin vs. controls (p = 0.56), and IM terlipressin vs. controls (p = 0.56). Moreover, there were no differences in the volume of glucose required to treat hypoglycemic animals, controls median 0 mL (IQR 0; 30), IM AVP mean 72 mL (45.5), p = 0.16, IV terlipressin mean 36 mL (39.1), p = 0.43, and IM terlipressin mean 42 mL (50.2), p = 0.43.

Fig. 4The alternative text for this image may have been generated using AI.

Blood gases and endogenous vasopressin. Arterial oxygen tension (a), carbon dioxide tension (b), base excess (c), lactate (d), and blood glucose (e) did not differ significantly following any treatment. Circulatory shock resulted in a significant increase in median endogenous vasopressin (vasopressin-NIIC) across all groups at the end of the experiment compared with baseline (f). Note: For animals that succumbed to trauma before the scheduled end of the observation period, terminal blood samples were drawn at the time of death and are included in the analysis of final endogenous vasopressin. Abbreviations: AVP, arginine vasopressin; IM, intramuscular; IV, intravenous. *p < 0.05, **p < 0.01, ***p < 0.001

Endogenous Vasopressin-NIIC

The median baseline Vasopressin-NIIC levels in all animals were 3894 pg/mL (IQR 3204; 5127), and increased by the end of the experiment in response to hemorrhagic shock to 14673 pg/mL (IQR 11977; 38976), p < 0.0001 (Fig. 4f). Vasopressin-NIIC levels were increased from baseline vs. the end of the experiment: controls 4008 pg/mL (IQR 3627; 4986) vs. 13996 pg/mL (IQR 12234; 33289), p = 0.0006; IM AVP 3773 pg/mL (IQR 2427; 6390) vs. 20799 pg/mL (IQR 10838; 102362, p = 0.02; IV terlipressin 3907 pg/mL (IQR 2966; 4860) vs. 37518 pg/mL (IQR 14400; 104095), p = 0.008; IM terlipressin 3881 pg/mL (IQR 3497; 5970) vs. 13032 pg/mL(IQR 8971; 57139), p = 0.02 (Fig. 4f). There was no difference in Vasopressin-NIIC levels between controls vs.: IM AVP (p = 0.5), IV terlipressin (p = 0.15), and IM terlipressin (p = 0.76).

Systemic absorption vs. hemodynamic effects of exogenous AVP

The median baseline level of exogenous AVP in the IM AVP group was 18.8 pg/mL (IQR 16.2; 25.4). Analysis of the five animals in the IM AVP group revealed four distinct response patterns: (i) non-absorbers (n = 2), in which peak serum levels remained low (25.4 and 47.3 pg/mL), suggesting perfusion failure at the injection site; (ii) uncoupled responder (n = 1), who achieved increased systemic absorption (82.2 pg/mL) but failed to mount a hemodynamic response and died before study completion; (iii) partial responder (n = 1), with one animal demonstrating both high absorption (214.7 pg/mL) and a correlated hemodynamic response, with SAP temporarily rising from 71 to 102 mmHg, yet dying 75 min after intervention began; (iv) responder (n = 1), who achieved both increased systemic uptake (184.1 pg/mL) and a sustained hemodynamic response, with SAP increasing from 68 mmHg to 114 mmHg, and survived to the end of the experiment.