Goal-directed fluid therapy combined with dexmedetomidine precipitates neurological outcomes in patients undergoing surgery for traumatic brain injury A randomised controlled trial

Study design and participants

This study was conducted in operating rooms and neurosurgical wards in Second Hospital of Hebei Medical University and has been approved by the Medical Ethics Committee of the hospital, All the investigations were signed an informed consent form with authorized person or family member. We also registered this prospective and observational study at Chinese Clinical Trial Registry (ChiCTR-INR-17012910). We included patients with TBI, aged from 18 to 65 years, gender is not limited, ASA grade II or III, GCS from 6 to10, admitted within 24 hours after injury, there was a clear brain history of trauma, no serious dysfunction of vital organs, no history of malignant arrhythmia, mental illness, cognitive dysfunction, arrhythmia, local skin infection of the internal jugular vein. Patients were randomly divided into four groups: control group (group C), GDFT group (group G), DEX group (group D) and GDFT with DEX group (group GD). Prior to induction of anesthesia, DEX was injected intravenously 1µg/kg for 10 min and then 0.5µg/kg^− 1·h^− 1 to the end of the surgery, and given as needed in NICU in group D and group GD.

Randomization and masking

Patients who met inclusion criteria and provided verbal and written consent were randomized to four groups. Randomization was performed using a predefined computer-generated list, which was kept by the pharmacy responsible for distributing the drug and/or FloTrac Sensor (Edwards Lifesciences, Irvine, CA, USA) for the study. Investigators in patient allocation did not collect data. Outcome adjudicators were blinded for patient allocation. All participants were blind to treatment assignment throughout the trial.

Anaesthesia and post-anesthesia treatment

The patient was placed mask to inhale oxygen at 6 L/min after arriving at the operation room. The non-invasive arterial pressure, pulse oximetry (SpO₂), and three-lead ECG were continuously monitored. Local anesthesia was performed and catheter was placed in the radial artery, and the invasive arterial blood pressure was monitored. Anesthesia was induced with midazolam, propofol, sufentanil and rocuronium. After endotracheal intubation, mechanical ventilation was initiated with a tidal volume 8 ml/kg, an I:E ratio of 1:2, Respiratory rate (RR) was adjusted to maintain end-tidal carbon dioxide (P_ETCO₂) 35 ~ 40 mmHg. Anaesthesia was maintained with continuous infusion of propofol and remifentanil. Intermittent bolus doses of rocuronium were injected to maintain adequate muscle relaxation during surgery. Patients with tracheal catheter reserveing were sent to ICU for observation after the operation.

After anesthesia induction, the FloTrac/Vigileo monitor (Edwards Lifesciences, Irvine, CA, USA) was connected and entered patient’s gender, age, height and weight to monitor the SVV. Guided by ultrasound, the right internal jugular vein or right subclavian vein catheterization was performed for monitoring central venous pressure (CVP) and transfusion, and the retrograde internal jugular vein catheter was placed for blood samples.

Study protocol and fluid therapy regimen

DEX was injected intravenously 1µg/kg within 10 min before anesthesia induction, and then 0.5µg·kg^− 1·h^− 1 to the end of the surgery in group D and group GD. The same amount of normal saline was injected intravenously as the same method with group D or group GD in group C and group G.

All patients in four groups were subjected to compensatory volume expansion (CVE) for acetate Ringer's solution of 5 ~ 7 ml/kg before and during anesthesia induction. The fluid therapy regimen in group C and group D was adopted with the classic infusion protocol, including the CVE, preoperative loss, intraoperative physiological requirements, blood loss and amount of third gap during surgery, to maintain CVP 2 ~ 8 mmHg and mean arterial pressure (MAP) 60 ~ 100 mmHg. Fluids include crystalloid and colloidal solution such as acetate Ringer's solution and succinyl gelatin injection with a crystal to gel ratio of 1 to 2:1, speed of 2 ~ 5 ml/(kg·h). The fluid therapy method in group G and group GD was performed according to the SVV value between 10 and 15%. SVV ≥ 13% indicates insufficient effective circulating blood volume, 250 ml colloidal solution was fastly infused within 10 minutes; when SVV < 13%, a speed of 2 ml/(kg·h) crystalloid solution was continuous infused.

Hemoglobin (Hb) > 80g/l and urine volume > 1ml·kg^− 1·h^− 1 were maintained in four groups. During the operation, when Hb < 80g/l, red blood cells were infused. If the patient has bradycardia (heart rate, HR < 50 beats/min), 0.5mg of atropine was injected; when intraoperative hypotension (MAP < 30% basal value) appeared, vasopressors such as ephedrine or norepinephrine was intravenously injected to maintain a normal or stable MAP.

Observation Indicators

HR and MAP were recorded at after anesthesia induction (T₀), dura mater cut (T₁), dura mater shut (T₂), end of the surgery (T₃) and 12 h after operation (T₄). Total transfusion volume, crystalloid solution infusion volume, colloidal solution infusion volume, urine volume and bleeding volume were recorded in four groups.

Blood samples from the internal jugular vein bulb and radial artery were collected at above five points from T₀ to T₄ for blood gas and biochemical analyses (ABL-80; Radiometer, Copenhagen, Denmark). Blood gas parameters include jugular venous bulb oxygen saturation (SjvO₂), jugular venous bulb partial oxygen pressure (PjvO₂), jugular bulb lactate (jvLac), arterial oxygen saturation (SaO₂), arterial partial oxygen pressure (PaO₂) and Hb. The arterial oxygen content (CaO₂), internal jugular bulb oxygen content (CjvO₂), cerebral arteriovenous oxygen content difference (Da-jvO₂) and cerebral extraction rate of oxygen (CERO₂) were calculated as following.

CjvO₂ = Hb × 1.34 × SjaO₂ + PjaO₂ × 0.0031; CaO₂ = Hb × 1.34 × SaO₂ + PaO₂ × 0.0031; Da-jvO₂ = CaO₂ - CjvO₂; CERO₂ = Da-jvO₂/CaO₂ × 100%;

Blood samples of jugular vein bulb were collected at five time points from T₀ to T₄, and supernatant was gotten after anticoagulation and centrifugation. Serum S100β protein and neuron-specific enolase (NSE) concentration was determined by enzyme linked immunosorbent assay (ELISA). ELISA detection range: 78 pg/ml − 5000 pg/ml, the minimum detectable dose of human S100β is typically less than 19.5 pg/ml, intra - assay precision (precision within an assay): CV% < 8%, inter - assay precision (precision between assays): CV% < 10%, the linear evaluation is the R value of the result standard curve. The days of postoperative length of hospital stay (LOS) were recorded. The Glasgow Prognosis Score (GOS) was performed before discharge to evaluate the quality of life (1 point of death, 2 points of plant survival, 3 points of severe disability, 4 points of mild disability and 5 points of good recovery).

Statistical analysis

Statistical analysis was performed by using SPSS 22.0 software. All data are presented as the mean ± SD (standard deviation)༎One-way ANOVA was used for comparison among groups, and groups were compared using the paired t test. Non-parametric test was used for counting data. P < 0.05 was considered significant༎

Comparison of general data in four groups

Of the 129 enrolled patients, 124 patients met the inclusion criteria and were allocated, and 120 patients were analysed in the study displayed in CONSORT Flow Diagram. There were no significant differences in age, gender, BMI, GCS, ASA grade, anesthesia time and blood transfusion among the four groups (P > 0.05) (Table). Eight patients in group GD and group G, ten patients in group D and fifteen patients in group C had received vasoactive drugs,

Comparison of intraoperative fluid intake and output in four groups (Fig. 1)

Total transfusion volume including crystalloid solution volume and colloidal solution volume decreased in group G (1961 ± 312 ml and 1005 ± 216 ml) and group GD (1999 ± 330 ml and 1029 ± 241 ml) compared with group C (2242 ± 295 ml and 1249 ± 232 ml) and group D (2226 ± 321 ml and 1187 ± 281 ml). The difference was statistically significant (P < 0.05). We found that combination therapy of GDFT + DEX decreased perioperative total transfusion volume including crystalloid and colloidal solution volume than monotherapy.

Comparison of HR and MAP in four groups (Fig. 2)

We found that HR experienced a downward trend from the beginning to the end of surgery, even to 12 h after surgery in group D and group GD compared with group C and group G. Compared with group C, the MAP changed more smoothly at dura mater cut in group D and from the beginning to the end of surgery in group G and group GD; compared with group D, the MAP changed more smoothly at dura mater shut in group GD. The difference was statistically significant (P < 0.05). We found that combination therapy of GDFT + DEX maintained stable hemodynamics.

Comparison of cerebral oxygen metabolism indexes in four groups (Fig. 3)

Compared with group C, the SjvO₂ increased whereas the Da-jvO₂ and the CERO₂ decreased from the beginning to the end of surgery, even to 12 h after surgery in group G, group D and group GD. Compared with group G, the SjvO₂ increased whereas the CERO₂ decreased from the beginning to the end of surgery, and the Da-jvO₂ decreased from the beginning to the end of surgery, even to 12 h after surgery in group GD. Compared with group D, the SjvO₂ increased whereas the CERO₂ decreased from the beginning to the end of surgery, and the Da-jvO₂ decreased at end of the surgery in group GD. The difference was statistically significant (P < 0.05). We found that combination therapy of GDFT + DEX decreased jvLac and improved cerebral oxygen metabolic rate than monotherapy.

Comparison of S100β and NSE in four groups (Fig. 4)

Compared with group C, the S100β and the NSE decreased at dura mater shut, end of the surgery and 12 h after operation in group D and group G, and from the beginning to the end of surgery, even to 12 h after surgery in group GD. Compared with group D and group G, the S100β decreased at dura mater shut, end of the surgery and 12 h after operation in group GD, while the NSE decreased from the beginning to the end of surgery, even to 12 h after surgery in group GD. The difference was statistically significant (P < 0.05). We found that combination therapy of GDFT + DEX decreased biomarkers, i.e, precipitated brain protection than monotherapy.

Comparison of GOS and LOS in four groups (Table 1)

Compared with group C, the GOS score of group GD was higher [75.62 vs 49.02; P = 0.014]. The difference was statistically significant (P < 0.05). Compared with group C, the postoperative LOS of group GD was shorter [18 (8) vs 26 (10) days; P = 0.002]. The difference was statistically significant (P < 0.05). We found that combination therapy of GDFT + DEX had a higher GOS score and a shorter LOS than the control group.

SVV guided GDFT strategies

GDFT refers to an individualized rehydration regimen based on the patient's gender, age, body weight, disease type, preoperative systemic condition, and volume status. Traditional indicators such as central venous pressure and pulmonary wedge pressure (PAWP) are static indicators that are susceptible to anesthesia, stress and circulatory functions, and they are imperfect and lagging. Functional hemodynamic parameters (FHP) monitoring including systolic blood pressure variability (SPV), SVV, pulse pressure variability (PPV), etc., this is a new hemodynamic monitoring method, SPV and PPV also has its limitations. It is required to be applied in a regular heart rhythm, standard ventilation mode, and also affected by factors such as intrathoracic pressure, intra-abdominal pressure or heart failure.⁹ The FloTrac/Vigileo Monitoring System is a new tool for calculating SVV and cardiac output (CO). Studies have shown that SVV has a good correlation with volume changes in major surgery patients undergoing mechanical ventilation, and confirmed its effectiveness and accuracy.^10,11 It is also sensitive and reliable for neurosurgery to predict the fluid reactivity of patients.^12,13 This study was based on the literature [6] and numerical value of the SVV is set to 13%, it as a target-directed liquid therapy.

Controversy over the choice of resuscitation fluid.¹⁴ Isotonic crystal solutions can be used for fluid management in patients with craniocerebral injury, and saline is the recommended solution. Lactic acid Ringer's solution is slightly hypotonic, which may reduce serum osmotic pressure and aggravate brain edema during large infusions.¹⁵ Persistent hyperglycemia is an independent risk factor for poor prognosis in patients with severe head injury. In addition, acidosis and free brain water produced by the anaerobic hydrolysis of glucose can make brain edema worse. Therefore, if the patient is not accompanied by hypoglycemia due to lack of nutritional support, the glucose solution should be avoided during anesthesia.¹⁶ Studies have shown that the ability of artificial colloids such as 6% hydroxyethyl starch (HES) 130 / 0.4 and succinyl gelatin to restore and stabilize circulation is better than that of crystal fluids. At the same time, it is beneficial to maintain cerebral oxygen supply and demand balance and reduce inflammation.^17,18 Maintaining the circulation stability of TBI patients during the perioperative period is the cornerstone of fluid therapy. When the patient's circulation cannot be maintained by using the crystal fluid alone, the anesthesiologist should use the colloid fluid reasonably according to the specific circumstances of the patient.Studies have shown that compared with the traditional fluid replacement methods, the FloTrac/Vigileo system provides SVV and SVV guided GDFT fluid management strategies to optimize perioperative volume status, further reducing postoperative complications and facilitating patient surgery and facilitating postoperative recovery.¹⁹ In this study, GDFT strategies decreased intraoperative total transfusion volume, crystalloid solution volume and colloidal solution volume in TBI patients. It indicates that SVV guides the liquid management strategy of GDFT can accurately and reliably reflect the human body's responsiveness to liquid therapy. Individualized infusion according to different or fluctuating fluid needs of patients, and correction of patients with occult circulating blood volume deficiency or excess in advance, so that both vascular tone and cardiac load are in an optimal state.

DEX as a commonly used anesthesia adjuvant and its brain protection research

DEX is a new type of highly selective α₂ - adrenergic receptor agonist. It is a commonly used anesthesia adjuvant in clinical practice. Wang suggested that DEX exerts neuroprotective effects on rats with subarachnoid hemorrhage by activating the EPK pathway.²⁰ DEX has been widely used in neurology.^{21, 22} Our study found that patients with traumatic brain injury were given DEX (1µg/kg) intravenously before anesthesia induction, followed by 0.5µg·kg^− 1·h^− 1 infusion, and the results showed DEX can reduce the serum S100β protein concentration of patients and DEX has brain protection. This study was based on the literature [8], prior to induction of anesthesia, DEX was injected intravenously 1µg/kg for 10 min and then 0.5µg·kg^− 1·h^− 1 to the end of the surgery.

One of the purposes of the GDFT combined with DEX is to optimize capacity management and stabilize hemodynamics.

In this study, after applying DEX, the patient's heart rate has a slowing trend. The reason may be the following two points. DEX has an inhibitory effect on stress and anti - sympathetic nerves. Its main mechanism of action against sympathetic nerves is to activate the α₂A/D receptor on the postsynaptic membrane of the solitary tract nucleus, reducing sympathetic tone, then lowering blood pressure and heart rate; DEX stimulates the α₂A/D receptor on the presynaptic membrane of the sympathetic nerve, inhibits the release of norepinephrine, reduces the concentration of plasma catecholamine, and thus plays a cyclically stable role.^{23, 24} Compared with the traditional fluid replacement methods, the other three groups of patients have a more stable blood pressure at the surgeon to open the dura mater. And compared with DEX alone, the MAP is more stable when GDFT combined with DEX. It indicates that the GDFT can accurately and reliably reflect the human body's responsiveness to liquid therapy. Individualized infusion according to different or fluctuating fluid needs of patients, and correction of patients with occult circulating blood volume deficiency or excess in advance, so that both vascular tone and cardiac load are in an optimal state. So the GDFT combined with DEX is to optimize capacity management and stabilize hemodynamics.

The second purpose of the GDFT combined with DEX is to play a role in brain protection and improve the neurological outcome

Improve cerebral oxygen metabolism

The blood in the internal jugular vein is returned from the venous sinus to the venous ball, and there is generally no mixed extracranial venous blood. Therefore, this study collected blood samples from this site for blood gas analysis and observation indicators. SjvO₂ is the first clinically used indicator to evaluate the oxygen metabolism of brain tissue, reflecting the changes in oxygen supply and oxygen consumption throughout the brain. According to Fick theory, SjvO₂ and Da-jvO₂ may reflect the relationship between cerebral blood flow and cerebral oxygen consumption- that is oxygen balance. Therefore, collect blood from jugular bulb to detect SjvO₂ and calculate Da-jvO₂ is very important to accurately assess the whole-brain blood flow and metabolism.²⁵ CERO₂ can reflect organic respiratory situation and organic perfusion, which is tightly relative to microcirculation perfusion. Thus, oxygen supply and consumption in brain tissue can be reflected by CERO₂ respectively. Therefore, the combination of SjvO₂, Da-jvO₂ and CERO₂ in this study can more accurately and reliably assess the oxygen supply and demand of the brain. And study has shown that DEX-induced sedation reduces cerebral blood flow (CBF), which may be due to direct alpha(2) - receptor cerebral smooth muscle vasoconstriction or DEX-induced reduction in brain metabolic rate the CBF changes, thus maintaining the balance of cerebral oxygen supply and demand.²⁶ In this study, the lactic acid value was directly measured by collecting the blood of the internal jugular vein, and the brain tissue perfusion was understood from the aspect of energy metabolism. Blood lactic acid and urine volume were important indicators for evaluating microcirculation perfusion.²⁷ And the study has confirmed that the amount of lactic acid in the blood is directly related to the prognosis and mortality of patients, which is helpful to evaluate the prognosis of patients with major surgery.²⁸ In this study, patients in the DEX group or the GDFT group had higher SjvO₂, lower Da-jvO₂ and CERO₂ than the traditional fluid replacement method, this is suggested that the use of DEX or GDFT can improve the patient's cerebral oxygen metabolism. Patients in the GDFT combined with DEX group had higher SjvO₂, lower Da-jvO₂ and CERO₂ than the other three groups, and the group had blood lactate concentrations decreased in the surgery and surgery after 12 hours. It is indicated that the GDFT combined with DEX further improves cerebral oxygen metabolism and brain tissue perfusion.

Decrease the S100β protein and NSE concentration

S100β is a calcium-binding protein that is secreted mainly by glial cells in the brain.The study has shown that S100β protein is a sensitive neurobiochemical marker of brain tissue damage.^{29, 30} NSE is a key protease, it involved in the cytoplasmic glycolytic pathway of neuroendocrine cells and brain neurons. It is mainly found in neuroendocrine cells and neuronal cytoplasm. The glial cells and related nervous tissues in the brain do not contain this. Enzymes, body fluids are also less abundant, so NSE has neurological specificity. The study has shown that NSE concentration is significantly elevated in patients with acute traumatic brain injury, and can be used as a biomarker to reflect damage to nerve tissue.^{31, 32} In this study, the professional enzymes used the classical enzyme-linked immunosorbent assay (ELISA) to determine the concentration of S100β protein and NSE in serum, which ensured the objectivity and reliability of the collected data. In this study, compared with the traditional fluid replacement method, S100β protein concentration and NSE decreased in patients treated with DEX or goal-directed therapy at certain time points, whereas application of GDFT combined with DEX decreased at each time point, and the GDFT combined with the DEX group was further reduced at some time points than the S100β protein concentration and NSE. It is suggested that the application of DEX or GDFT can reduce craniocerebral injury, and GDFT combined with DEX further reduces the extent of craniocerebral injury.

Improve clinical outcome

The Glasgow Coma Scale (GCS) is a method for determining the severity of craniocerebral injury based on the patient's condition of consciousness disorder. It is a well-established method for determining the severity of craniocerebral injury, and a simple and practical scale.³³ It performs 15 tests in three different reactions of blinking, speech and exercise. A total of 15 points, 15 points indicating waking, and 8 points and below indicating conscious coma. The GOS score scored the patient's language, recovery of limb motor function, and assessed the quality of life 1 week after surgery. The enduring appeal of the GOS is linked to its simplicity, short administration time, reliability and validity, stability, flexibility of administration cost-free availability and ease of access.³⁴ The Glasgow Outcome Scale (GOS) is the most widely cited assessment of outcome in the community after brain injury.³⁵ In this study, compared with the traditional fluid replacement methods, the days of postoperative length of hospital stay (LOS) and the GOS was higher in the GDFT combined with DEX. It is suggested that GDFT combined with DEX can reduce the postoperative length of hospital stay and improve their prognosis.

The limitations of this study are as follows. First, this study is a single-center study with a limited number of cases included, therefore, this study may have a bias in patient selection, and the results may not be universally applicable. Second, long-term outcome of patients was not observed and followed up.

In summary, GDFT combined with DEX can better optimize fluid therapy strategy and improve neurological outcomes in patients undergoing surgery for TBI.