Abstract

   Fluid replacement is the primary treatment for life-threatening shock
   but is challenging in harsh environments. This study explores
   low-intensity pulsed ultrasound (LIPUS) as a resuscitation strategy.
   Cervical LIPUS stimulation effectively elevated blood pressure in
   shocked rats. It also improved cerebral and multiorgan perfusion.
   Mechanistically, LIPUS activated pathways related to sympathetic nerve
   excitation and vascular smooth muscle contraction, increasing plasma
   catecholamines and stimulating blood pressure–regulating neural nuclei.
   Partial sympathetic nerve transection reduced LIPUS efficacy, while
   complete inhibition of these nuclei abolished the response. Preliminary
   clinical trials demonstrated LIPUS’s ability to raise blood pressure in
   shock patients. The findings suggest that LIPUS enhances sympathetic
   nerve activity and activates blood pressure–regulating nuclei, offering
   a noninvasive, neuromodulation-based approach to shock treatment. This
   method holds potential for improving blood pressure and organ perfusion
   in shock patients, especially in resource-limited environments.
     __________________________________________________________________

   LIPUS elevates blood pressure as a neuromodulated resuscitation
   strategy.

INTRODUCTION

   Shock is a critical medical condition characterized by inadequate blood
   flow to the body’s tissues, resulting in cellular dysfunction and organ
   failure. Hemorrhagic shock, a type of common shock, provoked by severe
   loss of blood causing insufficient tissue perfusion and cellular oxygen
   deficit, is a notable contributor to morbidity and mortality globally
   ([54]1). Common causes encompass traumatic injury, severe burns, major
   surgeries, childbirth, gastrointestinal bleeding, or other severe
   bleeding episodes such as from ruptured blood vessels or aneurysms
   ([55]2). Mortality rates can be as high as 60% in severe cases of
   hemorrhagic shock ([56]3). Septic and cardiogenic shocks are prevalent
   forms of shock. In the United States, approximately 2% of hospital
   admissions are for septic shock, with mortality rates approaching 20 to
   30% ([57]4). Annually, cardiogenic shock affects 40,000 to 50,000
   Americans and carries a mortality rate of about 50% ([58]5, [59]6).
   Despite the type, the fundamental manifestation of shock is the
   difficulty in maintaining adequate circulation, leading to insufficient
   oxygen and nutrient delivery to tissues, which is essential for
   cellular and organ function. This consequence makes immediate and
   effective management critical to avoid progressing down the path of
   multi-organ dysfunction syndrome (MODS) and death ([60]7). The primary
   therapeutic measures for shock involve hemodynamic stabilization and
   restoration of adequate organ perfusion.

   Initial remedial processes for hemorrhagic shock have made a
   revolutionary shift over the years, predominantly focusing on
   hemorrhage control, volume replacement by fluid resuscitation, and
   correction of coagulopathy. Notwithstanding, the mortality rates remain
   high due to the persistent risk of entering an irreversible stage of
   shock ([61]8). Resuscitation strategies conventionally resort to
   substantial intravenous fluids for restoration of lost blood volume
   ([62]9). However, the risk-to-benefit ratio of this approach remains
   controversial, considering its potential downside to incite dilution of
   clotting factors, hypothermia, and consequent hyperinflammatory state
   ([63]10, [64]11). Furthermore, in settings like factories,
   underdeveloped regions, and battlefields, the availability of
   intravenous fluid resuscitation therapy is often limited by the absence
   of both medical professionals and available facilities, leading to
   higher mortality rates from shock. The development of straightforward
   and feasible methods to elevate blood pressure (BP) could enhance
   access to treatment and potentially decrease shock-related mortality.

   Given that circulatory failure is a central characteristic of all types
   of shock, elevating BP remains a critical component in their treatment.
   However, traditional reliance on infusion to raise BP presents
   challenges in implementation, while the use of hemodynamic drugs to
   increase BP can lead to peripheral organ ischemia and is not
   sustainable. Given these challenges, we have focused on neuromodulation
   as a method to elevate BP ([65]12) following shock. An emerging,
   potential game-changer is the use of low-intensity pulsed ultrasound
   (LIPUS), a clinically noninvasive method. LIPUS’s ability to cause
   mechanical effects presents a promising therapeutic tool for a number
   of medical conditions, including neurological disorders, tumors, and
   targeted drug delivery ([66]13). LIPUS essentially uses acoustic energy
   at lower intensities and frequencies, providing a noninvasive and
   effective method of tissue stimulation and manipulation ([67]14). Thus,
   LIPUS can be used as a noninvasive neuromodulatory tool that can have a
   variety of effects, including BP regulation, depending on the target
   site. In neuroscience, the application of LIPUS has gained considerable
   attention, chiefly for its ability to target deeper regions of the
   brain while maintaining a noninvasive approach. This trait of
   neuromodulation, manipulating nerve activity by delivering a
   specifically aimed stimulation, offers a groundbreaking means of
   treating neurological or psychiatric conditions ([68]15). The
   application of 0.5 to 5 min of ultrasound stimulation has been
   demonstrated to yield enduring neuroregulatory effects, persisting from
   several minutes to several hours ([69]16). Clinical research
   spearheaded by Martínez-Fernández et al. ([70]17, [71]18) demonstrated
   the successful implementation of focused ultrasound in managing tremors
   associated with Parkinson’s disease, recording appreciable reductions
   in motor disability scores. Furthermore, in the scope of Alzheimer’s
   disease, promising advancements have been uncovered with LIPUS;
   preclinical trials revealed that repeated ultrasound sessions fostered
   clearance of β-amyloid plaques, consequently enhancing cognitive
   functions ([72]19). Beyond neurodegenerative disease, prolonged
   ultrasound stimulation has been noted to elicit excitation-like effects
   and induce alterations in connectivity in both nonhuman primates (NHPs)
   and human subjects, with effects lasting for over an hour.
   Specifically, stimulation of the amygdala and supplementary motor area
   in NHPs has been shown to influence circuit connectivity, leading to a
   modification in the coupling of functional magnetic resonance imaging
   (fMRI)–observed activity between the focused ultrasound targets and
   other brain regions ([73]20, [74]21). The efficacy of LIPUS is
   contingent upon the stimulation target, with targeted stimulation of
   the autonomic nervous system serving as the theoretical foundation for
   its regulation of vital signs, including BP.

   LIPUS has been proven to be effective in neuromodulation, but there
   have been no studies for increasing BP. Hypotension is a hallmark of
   shock, and our pioneering study investigated the effects and mechanisms
   of noninvasive cervical LIPUS stimulation on BP regulation in rats with
   hemorrhagic shock. Additionally, we executed a small-scale,
   self-controlled, before-and-after clinical trial to evaluate the
   clinical translatability of LIPUS.

RESULTS

LIPUS stimulation increased the BP of rats

   Here, we confirmed the effect of LIPUS stimulation on increasing BP in
   normal and hemorrhagic shock rats through arterial monitoring, plasma
   metabolomics, blood oxygenation level–dependent fMRI (BOLD-fMRI), etc.,
   and explored the mechanisms of LIPUS stimulation elevating BP ([75]Fig.
   1A). A minimum of 5 min of BP monitoring was performed on rats before
   cervical LIPUS stimulation, followed by 10 min of stimulation, and 5
   min of continued monitoring after stopping the stimulation ([76]Fig.
   1B).

Fig. 1. The schematic of the study.

   [77]Fig. 1.
   [78]Open in a new tab

   (A) A diagram shows the elevation of BP through cervical LIPUS
   stimulation in both normotensive rats and rats with shock-induced BP.
   The study also explores the effects and potential mechanisms of LIPUS
   in raising BP through arterial monitoring, HRV analysis, plasma
   metabolomics analysis, and BOLD-fMRI. A small-scale self-control
   before-after clinical trial was performed to explore the clinical
   translational potential of LIPUS. (B) The timeline outlines the design
   and progression of animal experiments in the study.

   The BP of the rats varied to different extents after stimulation at 1,
   2, and 3 W/cm^2 spatial peak pulse average intensity (Isppa) ([79]Fig.
   2, A to C). We found that at 1 W/cm^2, the BP and heart rate (HR) of
   the rat had no noticeable change (1 W/cm^2: P > 0.05, pre versus LIPUS;
   [80]Fig. 2D), whereas after stimulation at 2 and 3 W/cm^2 Isppa, there
   was a noticeable rise in the rat’s BP (2 to 3 W/cm^2: P < 0.001, pre
   versus LIPUS; [81]Fig. 2, E and F); the HR of the rat also rose after
   stimulation at 3 W/cm^2 Isppa (3 W/cm^2: P < 0.05, pre versus LIPUS;
   [82]Fig. 2F). At the same time, after stopping the stimulation, the
   rat’s BP would gradually drop back to prestimulation levels (2 to 3
   W/cm^2: P < 0.01, LIPUS versus post; [83]Fig. 2, E and F). On the basis
   of the above results, 2 W/cm^2 Isppa can effectively increase the rat’s
   BP and have no noticeable effect on the HR, and we chose this intensity
   for cervical stimulation in subsequent experiments. To exclude the
   influences of skin temperature rise and pain caused by ultrasound
   thermal effects and mechanical effects on the rat’s BP and HR, we
   performed a continuous 10-min 3-mA 10-Hz electrical stimulation
   ([84]Fig. 2G) and a 40°C heat stimulation ([85]Fig. 2H) on the rat’s
   cervical skin. The monitoring data of BP and HR show that neither skin
   electrical stimulation nor heat stimulation had noticeable effects on
   the rat’s BP and HR (P > 0.05, pre versus LIPUS; [86]Fig. 2, I and J).
   The above results indicate that cervical LIPUS stimulation can
   effectively increase the BP of rats, and it is not related to the
   thermal effects or pain stimulation of ultrasound. To further explore
   the potential physiological mechanisms of cervical LIPUS stimulation in
   increasing BP, we studied the cervical sympathetic nerve (SN) trunk or
   vagus nerve (VN). After the transection of the SN or the VN on the same
   side of the stimulation, we monitored the reactivity of rat BP to LIPUS
   stimulation (fig. S1, A and B). We found that after the cervical SN or
   the VN was severed, the BP still increased after the stimulation (fig.
   S1, A and B), but after cervical SN transection, the extent of BP rise
   [ΔMAP (mean arterial pressure) = average MAP during LIPUS − average MAP
   before LIPUS] was significantly reduced (**P < 0.01, SN transection
   versus control; fig. S1C), whereas after VN transection, there was no
   noticeable change in the extent of the BP rise (P > 0.05, VN
   transection versus control; fig. S1C). After severing SN, the increase
   in BP is diminished, suggesting that the mechanism by which LIPUS
   stimulation elevates BP is linked to SN excitation.

Fig. 2. Cervical LIPUS stimulation elevated BP in rats.

   [87]Fig. 2.
   [88]Open in a new tab

   (A to C) BP monitoring data in Sprague-Dawley rats after cervical LIPUS
   stimulation with different spatial peak pulse average intensity (Isppa)
   of 1, 2, and 3 W/cm^2. (D to F) Effect of cervical LIPUS stimulation
   with different effective ultrasound intensities on BP and HR in
   Sprague-Dawley rats (n = 5; ns., not significant; HR: *P < 0.05; MAP:
   ^##P < 0.01, ^###P < 0.001). (G and H) Arterial BP monitoring data
   after cervical skin electrical stimulation and cervical skin thermal
   stimulation. (I and J) Statistical analysis of MAP and HR data after
   cervical skin electrical stimulation and cervical skin thermal
   stimulation (n = 5).

LIPUS stimulation increases the BP of hemorrhagic shock rats

   We confirmed the effect of cervical LIPUS stimulation in increasing BP
   in normal rats. Furthermore, we explored whether LIPUS has a similar
   BP-elevating effect for shock in the shock rat model. We established
   the hemorrhagic shock model in rats by placing the tube in the rat’s
   femoral artery to monitor their BP, and simultaneously bleeding them
   until their MAP was less than 40 mmHg. Consistent with the timeline of
   previous experiments, rats were monitored for at least 5 min of
   baseline BP before 10 min of cervical LIPUS stimulation, and monitoring
   was continued for 5 min after the stimulation ([89]Fig. 3A). Under the
   stimulation of 2 W/cm^2 Isppa, the BP of normal rats raised
   (P < 0.0001, pre versus LIPUS; [90]Fig. 3B), accompanied by a slight
   elevation in HR (P < 0.05, pre versus LIPUS; [91]Fig. 3B). After
   stopping the stimulation in the normal BP group, the BP gradually
   dropped and returned to the baseline level before the stimulation
   (P < 0.0001, LIPUS versus post; [92]Fig. 3B). At the same time,
   arterial BP monitoring of hemorrhagic shock rats showed that the
   baseline MAP was less than 40 mmHg. However, after stimulation, the BP
   increased significantly (P < 0.001, pre versus LIPUS; [93]Fig. 3C),
   while the HR did not change noticeably (P > 0.05, pre versus LIPUS;
   [94]Fig. 3C). The shock group maintained elevated BP compared to
   baseline for 5 min following the cessation of stimulation (P < 0.05,
   pre versus post; [95]Fig. 3C). After analysis of the HR variability
   (HRV) obtained based on the R-R interval data of the electrocardiogram,
   we obtained the HRV data before and after cervical LIPUS stimulation of
   normal BP and shock rats, including low-frequency power (LFP) related
   to SN excitability, high-frequency power (HFP) related to
   parasympathetic nerve excitability, and LF/HF ratio related to the
   balance of sympathetic and parasympathetic nerves. The results show
   that after cervical LIPUS stimulation, LFP, HFP, and LF/HF of rats in
   the normal BP group raise to varying degrees (LFP: P < 0.01, HFP:
   P < 0.05, LF/HF: P < 0.01; [96]Fig. 3D). At the same time, in shock
   rats, LFP and LF/HF increase significantly after stimulation, and HFP
   does not change noticeably (LFP: P < 0.05, HFP: P > 0.05, LF/HF:
   P < 0.01; [97]Fig. 3E).

Fig. 3. Cervical LIPUS stimulation increased BP in hemorrhagic shock rats.

   [98]Fig. 3.
   [99]Open in a new tab

   (A) Representative BP monitoring data after stimulation in normal BP
   and hemorrhagic shock Sprague-Dawley rats. (B and C) Normal BP group
   (n = 9) and shock group (n = 8) statistical analysis of MAP and HR data
   (*P < 0.05; ^#P < 0.05, ^###P < 0.001). (D and E) Normal BP group
   (n = 9) and shock group (n = 8) HRV in the sympathetic/parasympathetic
   activity–related HFP, LFP, and LF/HF ratio statistical analysis. (F and
   G) Representative H&E-stained images of the cervical skin showed that
   LIPUS stimulation did not affect the superficial skin. (H and I)
   Representative H&E staining of SNs showed that LIPUS had no damage to
   the SNs. (J and K) Representative sections of the VN and the
   accompanying carotid artery (carotid lumen) stained with H&E suggest
   that LIPUS stimulation causes no damage to the VN.

   Safety is a basic premise of a treatment method. To clarify the effect
   of cervical LIPUS stimulation on superficial and deep tissues, we
   sliced the superficial skin of the neck, the deep SN trunk, and the VN
   on the stimulation side and used hematoxylin and eosin (H&E) staining.
   The results showed that the histological characteristics of the
   cervical superficial skin tissue without stimulation and after 60 min
   of stimulation ([100]Fig. 3, F and G), and the deep SNs ([101]Fig. 3, H
   and I) and VNs ([102]Fig. 3, J and K) were similar, suggesting that
   there was no noticeable damage after stimulation. Furthermore, Prussian
   blue staining was performed on cervical tissues. The results showed
   that no notable microbleeding was observed in the skin, SN, or VN after
   60 min of LIPUS stimulation (fig. S2). We first evaluated the effect of
   cervical LIPUS stimulation in increasing the BP of shock rats and found
   that the intensity can not only increase BP but also improve cerebral
   perfusion aftershocks, and the safety is secure.

LIPUS stimulation improves cardiac function and multi-organ perfusion of
hemorrhagic shock rats

   The low perfusion of organs after hemorrhagic shock is an important
   cause of poor prognosis. Therefore, it is very important to clarify
   whether LIPUS stimulation improves the function of the heart and the
   perfusion of important organs in rats while raising BP. We detected the
   heart function and blood flow of the liver and kidney in rats by
   ultrasound, and the perfusion of the cerebral cortex by laser speckle
   contrast analysis (LASCA). The heart function results showed that the
   left ventricular ejection fraction (EF) and cardiac output (CO) of the
   shocked rats were significantly reduced compared to the normal group
   (EF: P < 0.05; CO: P < 0.0001; [103]Table 1, [104]Fig. 4A, and fig.
   S3A). However, LIPUS significantly increased the CO of shocked rats
   (P < 0.05; [105]Table 1 and [106]Fig. 4A), but did not significantly
   improve the EF (P = 0.0693; [107]Table 1 and [108]Fig. 4A). For the
   kidneys, the peak systolic velocities (PSVs) and end-diastolic
   velocities (EDVs) of the renal artery and interlobar artery were
   significantly reduced after shock compared to the normal group (renal
   artery PSV and EDV: P < 0.0001; renal interlobar PSV and EDV:
   P < 0.0001; [109]Table 1, [110]Fig. 4B, and fig. S3B), and the renal
   artery resistance index (RI) significantly increased (RI: P < 0.05;
   [111]Table 1 and [112]Fig. 4B), indicating that the renal artery
   resistance increased and kidney perfusion decreased after the shock.
   While LIPUS increased the PSV and EDV of the renal artery (renal artery
   PSV and EDV: P < 0.05; [113]Table 1 and [114]Fig. 4B), there was no
   significant improvement in renal artery RI and interlobular PSV and EDV
   (P > 0.05; [115]Table 1 and [116]Fig. 4B). Similar to the kidney, the
   PSV and EDV of the hepatic artery dropped sharply after shock (hepatic
   artery PSV and EDV: P < 0.05; [117]Table 1 and fig. S3C). Although
   LIPUS stimulation increased the mean value of hepatic artery PSV and
   EDV after shock, there was no statistical difference (P > 0.05;
   [118]Table 1). Furthermore, we noninvasively measured the perfusion of
   the liver and kidneys using arterial spin labeling perfusion MRI
   (ASL-MRI) (fig. S4, A and B). The ASL-MRI results showed that LIPUS
   increased the renal cortical perfusion in both the control group and
   the shock group, with a more pronounced effect in the shock group
   (P < 0.05; fig. S4C). LIPUS also enhanced liver perfusion in the shock
   group (P < 0.05; fig. S4D) but had no significant effect on liver
   perfusion in the control group (P > 0.05; fig. S4D).

Table 1. Effects of LIPUS stimulation on cardiac function and blood flow of
liver and kidney.

   EDV, end-diastolic velocity; PSV, peak systolic velocity; RI,
   resistance index; C, control group; S, shock group;
   S + L = shock + LIPUS group.
   Control (n = 6) Shock (n = 6) Shock + LIPUS (n = 6) P value (C vs. S) P
   value (S vs. S + L)
   Ejection fractions (%) 84.1 ± 6.3 71.8 ± 9.2 82.0 ± 6.0 0.0266 0.0693
   Cardiac output (ml/min) 90.1 ± 16.2 44.0 ± 9.9 62.8 ± 8.7 <0.0001
   0.0409
   Renal artery EDV (mm/s) 195.6 ± 50.0 70.6 ± 22.4 135.7 ± 37.7 <0.0001
   0.0261
   Renal artery PSV (mm/s) 491.2 ± 98.5 260.2 ± 44.3 386.2 ± 48.3 <0.0001
   0.0156
   Renal artery RI 0.601 ± 0.059 0.731 ± 0.060 0.650 ± 0.081 0.0114 0.1285
   Renal interlobar artery EDV (mm/s) 121.8 ± 26.6 58.9 ± 17.1 97.5 ± 33.4
   0.0025 0.0579
   Renal interlobar artery PSV (mm/s) 285.7 ± 51.9 148.6 ± 37.1
   210.0 ± 50.4 0.0004 0.0919
   Renal interlobar artery RI 0.576 ± 0.037 0.603 ± 0.078 0.542 ± 0.102
   0.8226 0.3850
   Hepatic artery EDV (mm/s) 105.6 ± 49.4 23.0 ± 7.4 44.8 ± 10.8 0.0006
   0.4275
   Hepatic artery PSV (mm/s) 252.4 ± 104.6 96.8 ± 54.0 157.8 ± 103.7
   0.0241 0.4900
   Hepatic artery RI 0.585 ± 0.072 0.733 ± 0.086 0.628 ± 0.176 0.1174
   0.3125
   [119]Open in a new tab

Fig. 4. Cervical LIPUS stimulation improved multiple organ perfusion in
hemorrhagic shock rats.

   [120]Fig. 4.
   [121]Open in a new tab

   (A) Representative cross-sectional echocardiography of control, shock,
   and shock + LIPUS group. (B) Representative color Doppler flow images
   of the renal artery and renal interlobar artery. (C) Representative
   cortical CBF images before and after the stimulation in the normal BP
   group and shock group were detected by laser speckle contrast imaging
   instrument. (D and E) Comparative analysis of the brain perfusion in
   the bilateral cortex ROIs of the normal BP group and the shock group.

   To further explore whether stimulation can improve the decline in
   cerebral perfusion aftershocks, we detected the perfusion of the
   cerebral cortex through laser speckle ([122]Fig. 4C). Under normal BP
   conditions, the blood flow of the same side of the brain cortex
   increases after stimulation (P < 0.01; [123]Fig. 4D), and there is no
   noticeable change on the opposite side (P > 0.05; [124]Fig. 4D),
   whereas under shock conditions, the stimulation can significantly
   suggest that the cerebral cortex blood flow on both the stimulation
   side and the opposite side is stimulating (P < 0.01; [125]Fig. 4E).

LIPUS stimulation excites SNs, leading to vascular smooth muscle contraction
and increased BP

   To explore the mechanism of cervical LIPUS stimulation leading to an
   increase in BP, we performed the metabolomic analysis of plasma in
   control and shock rats after stimulation or no stimulation (control
   group, control + LIPUS group, shock group, shock + LIPUS group). The
   relationship model between each sample established by orthogonal
   partial least squares discriminant analysis (OPLS-DA) shows that there
   is a certain difference with good distinguishability between different
   groups ([126]Fig. 5A). The cyclic heatmap shows the 300 metabolites
   with the most significant differences in relative expression between
   each group, suggesting a difference in metabolites between each group
   ([127]Fig. 5B). Furthermore, the complex heatmap shows the multiple
   dimension properties of the top 50 most noticeably different
   metabolites between the four groups, including the ontology, the
   expression heatmap of each sample, the average expression heatmap of
   each group, the P value, and the VIP (variable importance on
   projection) value of each sample ([128]Fig. 5C). Why can LIPUS
   stimulate an increase in BP? This is the focus of our investigation.
   Therefore, we further analyzed the different metabolites of the shock
   group and the shock + LIPUS group. The volcano chart shows 133
   different metabolite products of the shock group and the shock + LIPUS
   group (fold change ≥ 1.5 or fold change ≤ 1/1.5, P < 0.05; [129]Fig.
   5D). [130]Figure 5E shows the top 30 different metabolites and their
   fold change and VIP values, such as phosphatidylcholine (PC)
   (18:0/18:2), 5.6-dihydroxy-2-methylaminotetralin, sphingomyelin (SM)
   (d18:1/24:1), bauerine C, and PC (20:5/16:0). After performing pathway
   enrichment analysis on the 133 different metabolites, we found that
   many pathways related to BP rise were activated, such as vascular
   smooth muscle contraction, renin secretion, and cyclic guanosine
   monophosphate-protein kinase G (cGMP-PKG) signaling pathway ([131]Fig.
   5F). The results of plasma metabolomics analysis suggest that the
   potential mechanism for the rise in BP induced by cervical LIPUS
   stimulation may be the activation of signal pathways related to
   vascular smooth muscle contraction.

Fig. 5. Plasma metabolomics suggested differentially expressed metabolites
and related signaling pathways after cervical LIPUS stimulation.

   [132]Fig. 5.
   [133]Open in a new tab

   (A) OPLS-DA analysis showed the intergroup differences and intra-group
   variations of samples in the control group (n = 9), control + LIPUS
   group (n = 9), shock group (n = 8), and shock + LIPUS group (n = 8).
   (B) Ring heatmap showing the top 300 significant different metabolites
   between the control group, control + LIPUS group, shock group, and
   shock + LIPUS group. (C) The complex heatmap displays information on
   the relative expression levels of the top 50 significantly different
   metabolites among the control group (n = 9), control + LIPUS group
   (n = 9), shock group (n = 8), and shock + LIPUS group (n = 8). The
   heatmap includes relative expression values, mean relative values, P
   values, VIP values, and ontology classification information of the
   metabolites. (D) Volcano plot showing the differentially expressed
   metabolites in plasma before and after LIPUS stimulation in the shock
   group. (E) Top 30 metabolites with VIP values and their fold change in
   plasma before and after LIPUS stimulation in the shock group. (F) The
   functional clustering diagram of differential metabolites displays the
   top 30 pathways based on differential abundance (DA) scores (shock +
   LIPUS versus shock).

The level of plasma catecholamines increased after LIPUS stimulation and was
positively related to SN excitation

   We obtained the level of catecholamine metabolites in the shock group
   and the shock + LIPUS group from the plasma metabolomics data, where
   epinephrine 3-sulfate and dopamine increased significantly after
   stimulation (P < 0.05; [134]Fig. 6A), while epinephrine and
   arachidonoyl dopamine did not change noticeably (P > 0.05; [135]Fig.
   6A). To further determine the relevance of plasma catecholamine levels
   and HRV indicators related to sympathetic or vagal nerves, we analyzed
   the relevance of five variables—epinephrine 3-sulfate, dopamine, LFP,
   HFP, and LF/HF ratio—in each animal in the shock group and the shock +
   LIPUS group. The results hint that epinephrine 3-sulfate and dopamine
   have significant positive correlations with the LF/HF ratio (red,
   P < 0.0001; [136]Fig. 6B), while HFP has a negative correlation with
   epinephrine 3-sulfate, dopamine, and LF/HF ratio, but there is no
   statistical difference (blue, P > 0.05; [137]Fig. 6B). The analysis of
   plasma catecholamine level and its correlation with HRV suggest that
   after cervical LIPUS stimulation, SN excitation is closely related to
   the increase in catecholamine secretion.

Fig. 6. The level of catecholamine metabolites detected by plasma
metabolomics after the shock rat received cervical LIPUS stimulation is
related to the excitement of the SN.

   [138]Fig. 6.
   [139]Open in a new tab

   (A) Catecholamine metabolites detected in plasma (n = 8; P < 0.01,
   P < 0.001; shock-LIPUS versus shock). (B) Correlation matrix of LFP,
   HFP, and LF/HF ratio (HRV), and the plasma epinephrine 3-sulfate and
   dopamine (catecholamine metabolites) of shock Sprague-Dawley rats
   (n = 8; **P < 0.001). (C) Representative BOLD-fMRI images indicate that
   LIPUS activates the hypothalamic nucleus associated with BP regulation,
   including OVLT, Pe, and SFO. (D to F) Paired t test statistical
   analysis of fALFF data from the three ROIs (OVLT, Pe, and SFO) before
   and after LIPUS stimulation (n = 5, *P < 0.05). (G) Flowchart of the
   sympathetic inhibition and SFO nucleotomy. (H) Representative T2WI
   images suggest that 6-OHDA injection successfully disrupted the SFO
   nuclei. (I and J) Cervical LIPUS stimulation had no significant effect
   on MAP or HR after inhibition of the SNs or SFO nucleotomy (n = 5).

LIPUS stimulation activates hypothalamic nuclei associated with BP regulation

   BOLD-fMRI results suggest that LIPUS significantly alters the pattern
   of brain activity in rats, with notable increases in hypothalamic
   activity ([140]Fig. 6C and fig. S5). Using the rat brain atlas, we
   extracted and analyzed fractional amplitude of low-frequency
   fluctuation (fALFF) data from regions associated with BP regulation in
   the hypothalamus, the organum vascularis of the lamina terminalis
   (OVLT), the paraventricular nucleus of the hypothalamus (Pe), and the
   subfornical organ (SFO), designated as regions of interest (ROIs).
   Statistical outcomes revealed that LIPUS significantly activated OVLT
   and SFO (n = 5; P < 0.05; [141]Fig. 6, D and F), while there was an
   increase in Pe activity, although it did not reach statistical
   significance (n = 5; P = 0.1019; [142]Fig. 6E). To further investigate
   the role of the sympathetic nervous system and specific nuclei in
   LIPUS-induced BP elevation, we conducted experiments involving SN
   inhibition and nucleotomy ([143]Fig. 6G). Using T2 weighted imaging
   (T2WI), we confirmed the destruction of the SFO nucleus in rats,
   followed by LIPUS stimulation experiments ([144]Fig. 6H).
   Administration of 6-hydroxydopamine (6-OHDA) effectively inhibited SN
   activity, completely nullifying the LIPUS effect on BP elevation
   (n = 5; P > 0.05; [145]Fig. 6I). Similarly, following SFO nucleotomy,
   while LIPUS stimulation induced a slight increase in BP, the change was
   not statistically significant (n = 5; P > 0.05; [146]Fig. 6J). The fMRI
   results further reveal the neuromodulation mechanisms of LIPUS
   stimulation, suggesting that the elevation of BP is associated with the
   activation of nuclei related to BP regulation. The influence of LIPUS
   on BP elevation was negated following SN inhibition and SFO nucleotomy.
   This indicates that the mechanism behind LIPUS-induced BP elevation
   involves the excitation of the sympathetic nervous system and the
   activation of BP-associated nuclei, such as the SFO.

Clinical self-control before-after trial suggested that LIPUS stimulation
could effectively increase human BP

   We conducted a small-scale trial of cervical LIPUS stimulation in 12
   shock patients at Tianjin Medical University General Hospital (table
   S1). The trial was approved by the hospital ethics committee
   (IRB2023-YX-267-01), and the written consent of the patient or his/her
   agent was obtained. All patients’ BP was obtained by an arterial
   monitoring device. We applied cervical LIPUS stimulation to shock
   patients for 5 min and recorded arterial BP data ([147]Fig. 7A). The
   self-control before-after trial showed that LIPUS elevated systolic BP
   but had no significant effect on diastolic BP in the shock patients
   ([148]Fig. 7B).

Fig. 7. Small-scale self-control before-after clinical trial showed that
cervical LIPUS stimulation elevated BP in shock patients.

   [149]Fig. 7.
   [150]Open in a new tab

   (A) BP in patients before and after LIPUS stimulation (n = 12 patients;
   prestimulation: 0 to 5 min, during stimulation: 5 to 10 min,
   poststimulation: 10 to 15 min). (B) Change in systolic and diastolic
   BPs of shock patients pre- and post-LIPUS stimulation (n = 12 patients;
   *P < 0.05). SBP: systolic blood pressure; DBP: diastolic blood
   pressure.

   In conscious healthy volunteers, after 5 min of cervical LIPUS
   stimulation, the results suggested that the pre- and poststimulation BP
   and HR of all volunteers were not significantly different from the
   prestimulation period, and no discomfort was reported (table S2).
   Elevated BP was not observed in healthy volunteers following LIPUS
   application, likely due to the limitations of the noninvasive arm cuff.
   The arm cuff, as an indirect means of measuring BP, does not permit
   continuous, real-time, and accurate monitoring. Nevertheless, the
   volunteers did not report any discomfort initially, validating the
   safety of cervical LIPUS stimulation. The results indicated that
   cervical LIPUS stimulation can safely increase BP in shock patients,
   suggesting that LIPUS may be beneficial not only for hemorrhagic shock
   but also for other types of shock.

DISCUSSION

   Here, we observed that LIPUS stimulation of the carotid triangle area
   improves BP in shock rats. BP is one of the main parameters of
   hemodynamics, and our discovery is that noninvasive cervical LIPUS
   stimulation can raise BP, leading to improvement of multiple organ
   perfusion. The circulatory shock caused by low BP is a situation of
   mismatch between oxygen delivery and consumption. If not corrected in
   time, it will lead to high mortality and disability rates ([151]22).
   Hemorrhagic shock can lead to single-organ or multiple-organ failure,
   especially for organs with high oxygen demand ([152]10). Potential
   brain hypoxia damage caused by reduced cerebral perfusion leads to
   consciousness disorders and long-term neurological damage after shock
   ([153]23). Our results suggest that cervical LIPUS stimulation can
   improve cerebral cortical blood flow perfusion aftershock, indicating
   that stimulation not only increases BP but also increases blood flow to
   vital organs, providing a promising strategy for anti-shock. Since
   LIPUS is portable and economical, the results of our study make
   noninvasive neuromodulation therapy for shock possible.

   Because LIPUS has strong and multiple physical and biological
   properties, many studies have tried to use ultrasound stimulation to
   regulate BP. A study found that after the left VN of the hypertensive
   rabbit model was invasively stimulated with low-intensity focused
   ultrasound, the rabbit’s BP and HR dropped ([154]24). Similarly, the
   team lowered the BP, HR, respiratory rate, and LF/HF ratio in rats with
   hypertension by using ultrasound to stimulate the isolated left VN,
   indicating that the parasympathetic nervous system was activated
   ([155]25). In hypertensive animal models, isolating the VN for
   ultrasound-targeted stimulation has been shown to lower BP, consistent
   with the function of the VN. Conversely, our noninvasive cervical LIPUS
   stimulation can increase BP, and the results may be associated with
   targeted SN excitation. Both of the above-reported studies have
   creatively separated the VN and stimulated it, while Cao et al. used
   noninvasive transcranial LIPUS to stimulate the solitary nucleus of
   hypertensive rats, which reduced rat BP. At the same time, the neuronal
   activity of the solitary nucleus, the ventrolateral periaqueductal gray
   (vlPAG) in the midbrain, and the tail lateral area of the caudal
   ventrolateral medulla (CVLM) were also enhanced, but ultrasound
   stimulation did not cause brain tissue damage ([156]26). Thus, the
   effectiveness of ultrasound stimulation is determined by its target.
   Our BOLD-fMRI studies revealed that cervical LIPUS stimulation enhances
   the activity of hypothalamic nuclei involved in BP regulation, such as
   the OVLT and the SFO. In contrast, transection of the cervical
   sympathetic trunk only partially attenuated the BP-raising effect of
   LIPUS, whereas more comprehensive suppression of sympathetic neural
   activity or ablation of BP-regulatory nuclei completely abolished the
   BP response elicited by LIPUS. These results further explain the
   mechanism by which cervical LIPUS stimulation increases BP.
   Additionally, the carotid sinus and carotid body are important
   receptors for BP regulation, and these two receptors are under the
   control of the carotid sinus nerve ([157]27). There is evidence that
   after ultrasound stimulation of the carotid sinus nerve, BP markedly
   increased ([158]24), similar to our results. The carotid sinus nerve is
   a branch of the glossopharyngeal nerve, and the glossopharyngeal nerve
   can rapidly modulate sympathetic and parasympathetic nervous activity,
   which is crucial for regulating BP and HR ([159]28). This suggests a
   correlation between cervical LIPUS stimulation and excitation of the
   glossopharyngeal nerve. The reason why noninvasive cervical LIPUS in
   our study raises BP instead of lowering BP may be that the target nerve
   is the carotid sinus nerve or SN. After the cervical sympathetic trunk
   was severed, the effect of LIPUS on BP increase was reduced; after
   further extensive inhibition of sympathetic nervous system activity,
   LIPUS completely lost its effect of raising BP. At the same time, the
   HRV analysis results show that LFP and LF/HF ratio increased after
   stimulation, both of which imply that the stimulation may increase SN
   excitability to play an anti-shock effect ([160]29).

   LIPUS has been proven to activate mechanoreceptors, integrin protein,
   on the cell membrane, thereby activating intracellular focal adhesion
   kinase (FAK) signaling and promoting its phosphorylation ([161]30). The
   phosphorylation of FAK activates mitogen-activated protein kinase
   (MAPK) and downstream extracellular signal–regulated kinase (ERK), P38,
   and c-Jun N-terminal kinase (JNK) signals, thereby playing an important
   role in cell proliferation, survival, and apoptosis ([162]31). In
   addition, LIPUS activates the phosphatidylinositol 3-kinase
   (PI3K)/Akt/Nrf2 signaling pathway, alleviating alveolar bone absorption
   caused by periodontitis ([163]32). In heart failure mice, LIPUS
   activated cardiomyocyte endothelial nitric oxide synthase
   (eNOS)/nitrogen monoxide (NO)/cGMP-PKG pathway and cardiomyocyte
   Ca^2+-handling system ([164]33). Our findings indicate that LIPUS
   elevates BP by augmenting the excitability of the sympathetic nervous
   system. Prior research has demonstrated that activation of SNs via
   knockdown of ADGRA1 or tumor necrosis factor–α (TNF-α) administration
   leads to the activation of the PI3K/Akt and MAPK kinase (MEK)/ERK
   pathways in the hypothalamus ([165]34, [166]35). This aligns with the
   signal transduction pathways stimulated by LIPUS, reinforcing our
   assertion that LIPUS modulates sympathetic nervous system activity,
   resulting in increased BP. The results also depict the potential
   physiological mechanism of LIPUS-enhanced BP through plasma
   metabolomics. Metabolomics analysis suggests that metabolites related
   to vascular smooth muscle contraction and renin secretion increase
   after LIPUS stimulation; meanwhile, cGMP-PKG, cAMP (adenosine
   3′,5′-monophosphate), and sphingolipid signaling pathway–related
   metabolites are also activated, which is consistent with previous
   research results ([167]33).

   After shock occurred, we observed that LIPUS stimulation raised BP and
   plasma catecholamine levels, and catecholamine levels were positively
   correlated with SN activity in HRV analysis. The SNs are excited, and
   the postganglionic neurons of the sympathetic nervous system release
   catecholamines to contract the vascular smooth muscle and increase BP
   ([168]36). Although different types of shock have different treatment
   plans, vascular vasoactive drugs are one of the key treatment methods.
   Shock treatment includes the administration of endogenous
   catecholamines (epinephrine, norepinephrine, and dopamine), as well as
   isoproterenol, phenylephrine, milrinone, and other drugs, which have
   become the mainstays of shock treatment for decades ([169]37). However,
   in patients with hypovolemic shock, excessive use of norepinephrine may
   lead to reduced renal perfusion ([170]38, [171]39). Our findings
   demonstrated that LIPUS stimulation improved cerebral and renal
   perfusion in rats with hemorrhagic shock. This suggests that LIPUS
   stimulation not only raises BP but also ameliorates organ perfusion.

   We were pleasantly surprised to discover that other teams are also
   advocating for advancements in the traditional shock treatment
   paradigm. For decades, this paradigm has remained largely unchanged.
   These teams are exploring innovative and challenging treatment
   strategies. Narayan’s team found that cervical trigeminal nerve
   stimulation can improve the activity of the sympathetic nervous system
   in rats with hemorrhagic shock, and increase cerebral blood flow (CBF)
   perfusion and brain oxygen tension, thereby improving survival
   ([172]12, [173]40). Trigeminal nerve stimulation presents clinical
   translation challenges due to its invasive nature and the necessity for
   precise anatomical positioning. Conversely, LIPUS not only elevates BP
   but also ameliorates multi-organ hypoperfusion. We have demonstrated
   that LIPUS stimulates BP by exciting the sympathetic nervous system,
   with elevated levels of vasoactive substances in plasma indicating its
   potential for sustained BP enhancement. What is encouraging is that our
   early clinical self-control before-after trial has emphasized the
   possibility of translating these findings into clinical applications,
   underscoring its reliability, ease of operation, device portability,
   and potential for clinical application. The trial shows that cervical
   LIPUS stimulation can effectively increase the BP of patients with a
   variety of shock types, not limited to the type of hemorrhagic shock in
   animal models. The BP regulation center is mainly located in the
   medulla oblongata, and the increase in BP is related to the specific
   activation of the rostral ventrolateral medulla (RVLM) ([174]41).
   Therefore, we speculate that LIPUS requires a functional BP regulation
   center, the brainstem, to take effect. In the future, performing larger
   sample clinical trials in hemorrhagic shock patients is crucial for
   validating these positive results and optimizing the clinical use of
   LIPUS.

   In conclusion, cervical LIPUS stimulation represents a noninvasive,
   convenient, and easily controllable technique applicable for
   resuscitating shock. It may serve as an innovative first-aid tool for
   trauma patients and potentially alter the existing paradigm of
   hemorrhagic shock management. On the basis of the principle of
   neuromodulation through LIPUS stimulation, LIPUS could play a role in
   the resuscitation management of patients suffering various types of
   shock. While the parameters and sustained effects of cervical LIPUS
   stimulation still require refinement and optimization for patient
   application, its clinical translation appears imminent. However, this
   clinical trial is limited to adjunctive treatment for shock patients
   with stable vital signs. To establish LIPUS as an emergency
   resuscitation treatment for shock, more extensive research and
   larger-scale clinical trials are needed.

MATERIALS AND METHODS

Study design

   We performed cervical carotid triangle area stimulation using
   noninvasive LIPUS in normal and hemorrhagic shock rats. All animals in
   this study were male Sprague-Dawley rats (HFK Bioscience, Beijing) aged
   6 to 8 weeks. While monitoring vital signs before and after
   stimulation, we also monitored CBF in rats by LASCA, and cardiac
   function and multiorgan perfusion by time-motion mode ultrasound and
   Doppler-mode ultrasound. Furthermore, the mechanism of BP elevation by
   LIPUS stimulation was explored through HRV analysis, peripheral blood
   metabolomics analysis, and fMRI. Cervical sympathectomy, inhibition of
   sympathetic activity, or nucleotomy of the BP-related nuclei have been
   used to elucidate the neuromodulatory mechanisms of LIPUS. A
   small-scale self-control before-after clinical trial was performed to
   explore the clinical translational potential of LIPUS.

Arterial monitoring and hemorrhagic shock model

   Place rats into a chamber containing 5% isoflurane to induce
   anesthesia. Once animals are completely anesthetized, place their nose
   in an anesthesia mask and maintain anesthesia with 2.5% isoflurane.
   Position the rats supinely on a 37°C warming pad and use depilatory
   cream to remove hair from the cervical and groin areas. On the left
   side of the sternohyoid muscle, locate the most prominent pulsation of
   the left carotid artery and mark it for LIPUS stimulation. Disinfect
   the skin in the groin area, and make an approximately 2-cm incision
   along the direction of the femoral artery after palpating the femoral
   artery in the left groin triangle area. Insert an arterial cannula
   filled with heparinized saline solution (40 U/ml) into the left femoral
   artery and connect it to a three-way valve. One end of the three-way
   valve is connected to a pressure transducer for BP monitoring, and the
   other end is connected to a syringe containing heparin for blood
   sampling. The pressure signal is connected to the biological signal
   acquisition and analysis system (BL-420, Taimeng Co. Ltd., Chengdu,
   China), which records the arterial pressure waveform. The limb leads
   for recording the electrocardiogram are inserted subcutaneously into
   the limbs of the rats and connected to the system. Record the baseline
   BP waveform and electrocardiogram of each rat for 5 min, excluding
   animals with initial MAP below 60 mmHg. We established a rat model of
   hemorrhagic shock as previously reported ([175]12). For the rats in the
   shock group or shock + LIPUS group, we slowly extracted blood at a rate
   of 0.5 ml/min. This process used a syringe containing 0.1 ml of heparin
   (6000 U/ml). We continued the extraction until the average arterial
   pressure remained below 40 mmHg for 5 min.

Cervical nerve transection

   To confirm the influence of the cervical sympathetic trunk and the VN
   on BP elevation during LIPUS, we conducted a nerve transection
   experiment. For the surgery of nerve transection, we opened the left
   carotid sheath of the rats and identified and isolated the VN next to
   the carotid artery, and deep in the carotid artery, we identified and
   isolated the SN trunk. Depending on the animal group, we used
   ophthalmic scissors to sever the VN or SN trunk. After removing air
   from the surgical area with 38°C saline solutions, the skin was sealed
   with biological glue.

   All animal studies were conducted in accordance with the Guide for the
   Care and Use of Laboratory Animals and were approved by the Tianjin
   Medical University General Hospital Animal Care and Use Committee
   (IRB2023-DW-82).

Cervical LIPUS stimulation

   Using depilatory cream, remove the hair from the rat’s cervical area.
   Locate the most prominent pulsation of the carotid artery on the left
   side of the sternohyoid muscle and mark it (fig. S6). Apply the
   ultrasound coupling agent to the marked area on the skin, and then
   ensure full contact between the ultrasound probe (DJO FRANCE SAS,
   Mouguerre, France) and the coupling agent before starting the
   stimulation. The contact surface of the ultrasound probe is a circular
   shape with a diameter of 1 cm, generating collimated ultrasound (fig.
   S6). Monitor the animal’s BP and baseline electrocardiogram for at
   least 5 min before starting the LIPUS stimulation, which lasts for 10
   min. After the stimulation stops, continue monitoring the BP and
   electrocardiogram for at least 5 min. In the LIPUS dose-response
   experience, rats were randomly divided into groups of the Isppa = 1, 2,
   or 3 W/cm^2, with five rats in each group. Ten rats with qualified BP
   were randomly assigned to the skin electrical stimulation and thermal
   stimulation groups, with five rats in each group. Ten rats were
   randomly assigned to the VN or SN severed groups, with five rats in
   each group. Rats for HRV analysis and plasma metabolomics analysis were
   divided into four groups (control group, control + LIPUS group, shock
   group, and shock + LIPUS group), initially with 10 rats in each group.
   Because of the losses during the surgery or shock, the sample sizes
   were as follows: control group, n = 9; control + LIPUS group, n = 9;
   shock group, n = 8; and shock + LIPUS group, n = 8. All animals were
   monitored for at least 5 min before the stimulation. Subsequently, the
   LIPUS intervention groups (control + LIPUS group and shock + LIPUS
   group) underwent 10 min of stimulation, while the nonstimulation groups
   (control group and shock group) did not activate the ultrasound
   generator. After 10 minutes of stimulation, we collected 0.5 ml of
   arterial blood from each group. The blood was placed in a centrifuge
   tube containing 10 μl of heparin (6000 U/ml). We then centrifuged the
   tube at 2000 rpm for 10 minutes. This process allowed us to obtain
   plasma.

   LIPUS parameters: In the dose experiment, Isppa is set to 1 to 3
   W/cm^2; for the remaining experiments, the parameters used are as
   follows: 2-cm^2 ultrasound generator, Isppa of 2 W/cm^2, fundamental
   frequency of 3 MHz, repetition frequency of 100 Hz, duty cycle of 50%,
   and pulse duration of 1 ms.

Vital sign collection and analysis

   The BL-420 biological signal acquisition and analysis system was used
   to record the BP waveforms and electrocardiograms. Data are collected
   at a frequency of 1000 Hz and filtered using low-pass filters and a
   50-Hz notch filter to remove noise from the original signals. MAP is
   obtained from the BP waveform, and HR and R-R intervals are derived
   from the electrocardiogram data. The software of this system is used to
   analyze the rat’s electrocardiac signal, and HRV analysis is performed
   based on the R-R interval data of the electrocardiogram waveform.
   Frequency domain analysis is performed using the software, and HRV
   parameters are calculated and analyzed. The frequency domain analysis
   parameters for HRV mainly include total power (TP), which represents
   the frequency band from 0 to 3.0 Hz; LFP, which represents the
   frequency band from 0.20 to 0.60 Hz; and HFP, which represents the
   frequency band from 0.60 to 3.0 Hz. LFP and HFP represent the roles of
   the sympathetic and parasympathetic nervous systems, respectively. The
   LF/HF ratio can effectively reflect the balance between the autonomic
   nervous systems ([176]42). LF/HF ratio = LFP/HFP.

H&E staining

   The paraffin-embedded skin and nerve tissue sections are cleaned with
   xylene to remove the wax and then passed through a series of decreasing
   alcohol concentrations to rehydrate the tissue. The sections are
   stained with hematoxylin for 5 min following treatment with acid
   alcohol to remove the excess hematoxylin. Then, the sections are
   stained with eosin for 1 min. The samples are passed through increasing
   alcohol concentrations to dehydrate, followed by xylene to clear the
   sections, and the samples are finally sealed with a coverslip using a
   mounting medium for viewing under a microscope.

Prussian blue staining

   Prussian blue staining was performed with reference to our previous
   literature ([177]43). Briefly, the sections were incubated in a
   solution containing 5% potassium ferricyanide (Solarbio, China) and 5%
   hydrochloric acid (in a 1:1 ratio) for 30 min, followed by rinsing in
   double-distilled water for 10 min. The nuclei were counterstained blue
   with hematoxylin.

CBF analysis

   As previously reported, LASCA was used to monitor CBF. The Pericam PSI
   system (Perimed AB, Sweden) was used for laser speckle imaging in rats.
   For this purpose, each mouse was anesthetized with 1.5%
   tribromoethanol, the skull was shaved, and the skull was exposed and
   cleaned through a midline skin incision. The red cross point of the
   indicator laser (660 nm) was placed at the center of the brain with a
   fixed measurement distance of 10 cm. The PIM software (Perimed AB,
   version 1.5) was used to collect CBF signals at a wavelength of 785 nm,
   and blood perfusion images were obtained. CBF is represented in
   perfusion units (PUs).

Multi-organ ultrasound examination

   All rats undergoing ultrasound examination had their hair shaved from
   the neck, chest, and abdomen after isoflurane anesthesia. The control
   group underwent dissection of the femoral artery without blood loss and
   without activating the cervical ultrasound generator. The shock group
   achieved the criteria for hemorrhagic shock but did not activate the
   cervical ultrasound generator. The shock + LIPUS group achieved the
   criteria for hemorrhagic shock and activated the cervical ultrasound
   generator. The rats were placed on a 38°C heating pad and examined
   using the Vevo 3100 small animal ultrasound system (FUJIFILM, Canada).

   For cardiac function examination, use a phased array transducer (16
   MHz) for echocardiographic examination. For the renal artery, position
   the transducer in the epigastric and paraumbilical region. Measure
   renal artery PSVs and EDVs. Calculated parameters include the resistive
   index (RI) [RI = (PSV − EDV)/PSV]. For the hepatic artery, we placed
   the transducer in the upper right abdomen. After locating the hepatic
   artery, we measured the PSV and EDV of the hepatic artery.

Plasma metabolome and analysis

   Sample (100 μl) was thoroughly mixed with 400 μl of cold methanol
   acetonitrile (v/v, 1:1) via vortexing, and then the mixture was
   processed with sonication for 1 hour in ice baths. The mixture was then
   incubated at −20°C for 1 hour and centrifuged at 4°C for 20 min with a
   speed of 14,000g. The supernatants were then harvested and dried under
   vacuum liquid chromatography–mass spectrometry (LC-MS) analysis.

   The raw MS data were processed using MS-DIAL (version 5) for peak
   alignment, retention time correction, and peak area extraction. The
   metabolites were identified by accuracy mass [mass tolerance <10 parts
   per million (ppm)] and MS/MS data (mass tolerance <0.02 Da), which were
   matched with Human metabolome database (HMDB), massbank, and other
   public databases and our self-built metabolite standard library. In the
   extracted ion features, only the variables having more than 50% of the
   nonzero measurement values in at least one group were kept.

   R (version 4.0.3) and R packages were used for all multivariate data
   analyses and modeling. Data were mean-centered using Pareto scaling.
   Models were built on principal components analysis (PCA), OPLS-DA, and
   PLS-DA. All the models evaluated were tested for overfitting with
   methods of permutation tests. The descriptive performance of the models
   was determined by R2X (cumulative) [perfect model: R2X (cum) = 1] and
   R2Y (cumulative) [perfect model: R2Y (cum) = 1] values, while their
   prediction performance was measured by Q2 (cumulative) [perfect model:
   Q2 (cum) = 1] and a permutation test (n = 200). The permuted model
   should not be able to predict classes: R2 and Q2 values at the y-axis
   intercept must be lower than those of Q2 and the R2 of the nonpermuted
   model. OPLS-DA allowed the determination of discriminating metabolites
   using the VIP. The VIP score value indicates the contribution of a
   variable to the discrimination between all the classes of samples.
   Mathematically, these scores are calculated for each variable as a
   weighted sum of squares of PLS weights. The mean VIP value is 1, and
   usually, VIP values over 1 are considered significant. A high score is
   in agreement with a strong discriminatory ability and thus constitutes
   a criterion for the selection of biomarkers.

   The discriminating metabolites were obtained using a statistically
   significant threshold of VIP values obtained from the OPLS-DA model and
   two-tailed Student’s t test (P value) on the normalized raw data at the
   univariate analysis level. The P value was calculated by one-way
   analysis of variance (ANOVA) for multiple groups analysis. Metabolites
   with VIP values greater than 1.0 and P values less than 0.05 were
   considered to be statistically significant metabolites. Fold change was
   calculated as the logarithm of the average mass response (area) ratio
   between two arbitrary classes. On the other side, the identified
   differential metabolites were used to perform cluster analyses with the
   R package.

Blood oxygenation level–dependent fMRI

   We acquired transverse-sectional BOLD-fMRI and T2WI images of rats
   using a 9.4-T MRI system (Bruker, USA). We performed the MRI scan
   immediately after LIPUS stimulation outside the shielded room. Five
   rats were scanned before and after LIPUS stimulation for paired
   comparison. BOLD-fMRI was acquired with the following parameters: echo
   time = 20 ms, repetition time = 1500 ms, repetitions = 240,
   slices = 28. T2WI was acquired using the following: effective echo
   time = 33 ms, repetition time = 2500 ms, echo train length = 8,
   slices = 28. The MRI raw data were processed and analyzed using the
   DPABI toolbox ([178]44, [179]45). Functional images were realigned,
   reoriented, normalized, and smoothed after the first 10 time points
   were removed. ALFF and fALFF data were extracted based on the
   preprocessed data. After registering T2WI images to functional images,
   data were extracted for analysis by delineating ROIs in conjunction
   with the rat brain atlas.

Arterial spin labeling perfusion MRI

   We acquired coronal-sectional ASL-MRI and T2WI images of rats using a
   9.4-T MRI system (Bruker, USA). We performed the MRI scan immediately
   after LIPUS stimulation outside the shielded room. ASL-MRI was acquired
   with the following parameters according to the reference: The inversion
   times used in this study were 30, 100, 200, 300, 500, 700, 1000, 1200,
   1500, 2000, 3000, 5000, and 8000 ms. Imaging was performed with a
   matrix of 128 × 128, a field of view of 40 × 40 mm, and a section
   thickness of 2 mm. A single section was acquired. For section-selective
   inversion, an adiabatic frequency-selective inversion pulse with a
   hyperbolic secant shape was applied, featuring a bandwidth of 5190 Hz
   and a section thickness of 5 mm. To quantify renal perfusion, an
   optimized inversion time of 2000 ms was used. Ten pairs of ASL images,
   acquired using selective and nonselective inversion pulses, were
   analyzed to enhance the signal-to-noise ratio.

SN inhibition and nucleotomy

   Here, we administered intraperitoneal injections of a solution
   containing 6-OHDA (20 mg/ml) and 0.1% ascorbic acid to rats over three
   consecutive days to inhibit SN activity. Following this, under
   anesthesia, we performed LIPUS stimulation (Isppa = 3 W/cm^2) and
   monitored BP and HR on the fourth day. Additionally, the SFO nucleus
   was precisely located at the bregma −0.6 mm and 5 mm deep and received
   an injection of 10 μg of 6-OHDA for nucleotomy. Six days
   post-injection, a T2WI scan verified the accuracy of the injection site
   and the extent of nucleus disruption. Subsequent LIPUS (Isppa = 3
   W/cm^2) stimulation under anesthesia and concurrent monitoring of BP
   and HR were performed on the seventh day.

The small-scale self-control before-after clinical trial

   The enrolled patients were admitted during March to June 2024 at the
   General Hospital of Tianjin Medical University. All shock patients were
   prioritized for emergency resuscitation upon admission.

   Inclusion criteria: (i) age >18 years but <80 years; (ii) admission
   diagnosis of shock (all types), admission systolic BP < 90 mmHg; (iii)
   underwent arterial catheterization with real-time BP monitoring; (iv)
   Glasgow coma scale (GCS) score > 8; (v) relatively stable vital signs
   without immediate need for surgery.

   Exclusion criteria: (i) patients in the perioperative period; (ii)
   pregnant and lactating patients; (iii) patients with carotid artery
   plaques; (iv) patients with neck wounds.

   Cervical LIPUS stimulation was administered in the following manner:
   The patient is positioned supine, with the head tilted rightward to
   expose the left cervical region. The most pronounced pulsation of the
   carotid artery is identified anterior to the upper portion of the
   sternocleidomastoid muscle on the left side. After marking the
   pulsation site, the left neck area is cleansed and disinfected. A
   suitable quantity of ultrasonic coupling agent is then applied to
   ensure full contact between the 5-cm^2 ultrasound generator and the
   skin, and the generator is securely positioned. The LIPUS is initiated
   with parameters set at an ultrasound frequency of 3 MHz, a pulse
   frequency of 100 Hz, a duty cycle of 50%, and an Isppa of 3 W/cm^2. All
   participants underwent a 5-min stimulus, with BP measurements recorded
   for a minimum of 5 min both before and after the stimulus.

   To confirm the safety of cervical LIPUS stimulation, we recruited five
   health-conscious individuals as volunteers to investigate the effects
   of LIPUS stimulation on healthy volunteers. BP measurements for healthy
   volunteers were taken noninvasively using an arm-cuff after they had
   rested for 10 min in a quiet environment, and again immediately
   following 5-min LIPUS stimulation, as previously described. BP was
   measured twice consecutively before LIPUS stimulation to obtain an
   average value and then measured twice after 5 min of LIPUS stimulation
   before stopping the stimulation.

   This study was approved by the Ethics Committee of Tianjin Medical
   University General Hospital (IRB2023-YX-267-01). Informed consent was
   obtained from all subjects or their legal representatives after the
   patient regained consciousness. The clinical study is registered with
   the Chinese Clinical Trial Registry (ChiCTR2400093480).

Statistical analysis

   Statistical analysis was performed using the GraphPad Prism software
   (version 9.4). Two independent samples were analyzed using a two-tailed
   t test, and a paired t test was conducted to assess the significant
   difference in the changes in MAP and HR when the LIPUS was turned off
   and on. Datasets are analyzed with one-way or two-way ANOVA followed by
   post hoc Tukey’s test. P < 0.05 was considered as statistically
   significant. All data were presented as mean ± SD.

Acknowledgments