Objective To describe an improved percutaneous tracheostomy combined with conventional tracheostomy technique with result of less trauma and fewer complications, and to explore its application in the patients for whom conventional tracheostomy is difficult to perform.Methods A prospective study was conducted.Fifty-seven hospitalized patients, in whom ordinal tracheostomy was difficult to perform, and admitted to Department of Critical Care Medicine of the First Affiliated Hospital of Bengbu Medical College from January 2013 to December 2014 were enrolled.According to the random digital table method, the patients were divided into small incision combined with percutaneous tracheostomy group (small puncture incision group, n =25) and conventional tracheostomy group (n =32).Amount of blood loss, postoperative bleeding, incision size, operation time and wound healing time were compared between the groups.Results Compared with traditional surgical tracheostomy group, the blood loss and postoperative bleeding were decreased [blood loss (mL): 11.36 ± 4.25 vs.23.72 ± 7.29, t =-7.201, P =0.000;postoperative bleeding (mL):11.60 ± 6.57 vs.26.77 ± 10.77, t =-5.834, P =0.000], incision size was smaller (cm: 2.20 ±0.63 vs.4.06 ± 1.19,t =-6.806, P =0.000), and operation time and wound healing time were shortened [operative time (minutes): 18.16 ±3.61 vs.29.09±6.77, t =-7.001, P =0.000;incision healing time (days): 4.96± 1.59 vs.7.19±2.35, t =-3.975,P =0.000] in small puncture incision group.Conclusion Compared with the traditional method, small incision puncture tracheostomy is less time consuming, with fewer traumas, and fewer intraoperative and postoperative complications.

Objective To evaluate the use of nasojejunal tube in early enteral nutrition in severe traumatic brain injury (STBI) patients under mechanical ventilation.Methods STBI patients requiring mechanical ventilation in intensive care unit (ICU) of the First Affiliated Hospital of Bengbu Medical College admitted in 2013 were randomly divided into the jejunal tube group (n =15) and gastric tube group (n =19).We compared the 2 groups in terms of the tolerable beginning time of enteral nutrition (EN),the time before reaching target feeding volume,the incidences of gastrointestinal complications and ventilator-associated pneumonia (VAP) during EN,mechanical ventilation time,ICU hospital stay,and 28-day mortality rate.Results The tolerable beginning time of EN [(51.73 ± 9.16) hours vs.(81.11 ± 11.82) hours,t =7.920,P ＜0.05] and the time required to reach target feeding volume [(87.27 ± 9.99) hours vs.(152.05 ± 28.74) hours,t =8.320,P ＜ 0.05] in the jejunal tube group were significantly shorter than those in the gastric tube group.In the process of EN,compared with the gastric tube group,the incidences of gastric retention (6.7％ vs.57.9％,x2 =10.937,P ＜ 0.05),reflux (0％ vs.36.8％,x2 =9.566,P ＜ 0.05),vomiting (20.0％.vs.63.2％,x2 =6.642,P＜0.05),aspiration (6.7％ vs.42.1％,x2 =6.087,P＜0.05),VAP (33.3％ vs.73.7％,x2 =5.536,P ＜ 0.05) in the jejunum tube group were significantly lower.The mechanical ventilation time [(10.73 ± 4.68) days vs.(15.74 ± 2.54) days,t =3.730,P＜0.05] and the ICU hospital stay [(13.60 ± 4.80) days vs.(17.42 ± 4.05) days,t =2.497,P ＜0.05] of the jejunum tube group were significantly shorter than those of the gastric tube group.Comparison of 28-day mortality rate between the two groups revealed no statistically significant difference.Conclusion Early implementation of EN via nasojejunal tube in mechanically ventilated STBI patients can alleviate feeding intolerance,shorten the beginning time of EN and the time required to reach target feeding volume,reduce the incidence of complications,and shorten mechanical ventilation time and hospital stay in ICU.

Objective:To design a Checklist for quality control in intensive care unit and observe the effect of clinical application.Methods:By consulting guidelines and literature, such as Critical care medicine professional medical quality control index (2015 edition), the quality control Checklist of intensive care unit was designed. It included four parts: quality control data collection, medical record quality verification, special diagnosis and treatment, and hospital infection prevention and control supervision. Every month, a doctor with a senior professional title served as the quality control director, and was responsible for the quality control of the department's medical care, including collecting data of the past 24 hours during the morning handover, discussing and registering special diagnosis and treatment behaviors that would be performed on the day, and coordinating with the nursing team leader, controlling the quality of the whole department throughout the day, such as supervising each medical staff if they had unreasonable behaviors, checking the running and discharge medical records, and inspecting the status of the staff on duty. The data in 2018, 2019 (Checklist implemented) and 2017 (Checklist not implemented) were retrospectively analyzed, including the status of admitted patients, department management information, length of intensive care unit (ICU) stay, and the incidence of three-tube infection [ventilator-associated pneumonia (VAP), catheter-related bloodstream infection (CRBSI), catheter-associated urinary tract infection (CAUTI)], and standardized mortality, etc. Results:From 2017 to 2019, the number of patients admitted was 373, 446, and 480, with annual growth of 19.57% and 7.62% in 2018 and 2019, respectively, and an increase of 28.69% in 2019 compared with 2017. There was no statistically significant difference in the average age and acute physiology and chronic health evaluationⅡ (APACHEⅡ) of patients in the three years. Compared with 2017, the length of ICU stay of patients in 2018 and 2019 were significantly shortened (days: 8.99±6.12, 9.14±7.02 vs. 10.20±7.21), and the incidence of VAP, CRBSI and CAUTI were significantly reduced [VAP (cases/1 000 ventilation days): 12.97±3.60, 9.62±3.14 vs. 17.48±4.89, CRBSI (cases/1 000 catheter days): 3.75±2.19, 3.87±1.87 vs. 6.19±3.13, CAUTI (cases/1 000 catheter days): 3.29±2.18, 3.28±1.87 vs. 5.61±3.18]. The standardized mortality were also significantly reduced [(77.27±7.24)%, (70.61±7.49)% vs. (84.41±9.05)%], the number of non-compliance with hospital infection prevention per month decreased significantly (person times: 54.00±6.30, 41.08±10.76 vs. 72.08±19.68), and the number of special diagnosis and treatment per month increased significantly (person times: 1 056.67±235.27, 1 361.75±278.48 vs. 722.25±145.96), the rate of etiology submission before antimicrobial treatment [(93.21±3.68)%, (96.59±2.49)% vs. (87.86±5.28)%] and deep vein thrombosis (DVT) prevention rate [(91.13±6.36)%, (96.23±2.99)% vs. (85.58±7.68)%] were significantly improved, and all the differences were statistically significant (all P < 0.05). All medical records in the three years were Grade A, but the average scores in 2018 and 2019 were higher than those in 2017 (96.82±2.84, 96.73±2.94 vs. 93.70±3.33, both P < 0.01). Compared with 2018, the incidence of VAP, the rate of etiology submission before antimicrobial treatment, the DVT prevention rate, and the standardized mortality rate in 2019 were further improved, and the number of non-compliance with hospital infection prevention per month decreased and the number of special diagnosis and treatment per month increased, and the differences were statistically significant (all P < 0.05). Conclusion:The application of quality control Checklist in intensive care unit can build an effective quality control system, reduce the incidence of three-tube infection, standardized mortality and length of ICU stay, improve the quality control awareness and execution of medical staff, and promote the improvement of medical quality.

Objective: To investigate the changing laws of rest energy expenditure (REE) in intensive care unit (ICU) patients and the intervention effect for nutritional support. Methods: A prospective randomized control trial was conducted. Fifty-eight critically ill patients who were expected to be able to receive sustained enteral and (or) parenteral nutrition for more than 7 days admitted to ICU of the First Affiliated Hospital of Bengbu Medical College from December 2016 to June 2017 were enrolled. The patients were divided into REE group (n = 29) and HBREE group (n = 29) according to the random number table. On the 1st to 7th day after ICU admission, the indirect calorimetry and the Harris-Benedict (HB) formula were used to obtain the REE and HBREE values, and nutritional support was given according to REE and HBREE values respectively. The data of hemoglobin (Hb), albumin (Alb), prealbumin (PA), C-reactive protein (CRP), oxygenation index (OI) on 1st, 3rd, 5th, 7th and discharged day, and insulin dosage, vasopressor time, mechanical ventilation time, the length of ICU stay, and 28-day mortality were collected. Results: ① At the beginning, the REE level was high, and then decreased gradually with the extension of hospitalization, and the decline was obvious on the 2nd to 3rd day (kJ/d: 7 088.38±559.41, 6 751.34±558.72 vs. 7 553.44±645.55, both P < 0.05), and was stable from the 5th day, the changing laws showed high at first, then the low, the first rapid decline, then the slow decline, and then reached the steady, there was a 2-day plateau in the middle. During the first 2 days, the REE value was significantly higher than the HBREE value (kJ/d: 7 553.44±645.55 vs. 6759.21±668.14, 7 088.38±559.41 vs. 6 759.21±668.14, both P < 0.01); on the 3rd, 4th day, the REE value was almost the same as the HBREE value (kJ/d: 6 751.34±558.72 vs. 6 759.21±668.14, 6 568.03±760.19 vs. 6 759.21±668.14, both P > 0.05). After that, the REE value was significantly lower than the HBREE value (kJ/d: 6 089.55±560.70 vs. 6 759.21±668.14, 5 992.55±501.82 vs. 6 759.21±668.14, 5 860.84±577.59 vs. 6 759.21±668.14, all P < 0.01). ② After the initiation of nutritional support, Hb in the REE group (the first 3 days) and HBREE group (the first 7 days) all increased slowly in the early stage. It increased obviously on the 5th day in the REE group. Compared with the REE group, Hb increased more slowly in the HBREE group, however, there was no difference between the two groups at the time of discharge (g/L: 113.75±17.28 vs. 110.86±15.35, P > 0.05). PA and OI all enhanced significantly on the 3rd day since the nutritional support was initiated, but the daily increase of the REE group was significantly higher than that of the HBREE group [3rd day, PA (mg/L): 110.38±27.65 vs. 96.28±18.06, OI (mmHg, 1 mmHg = 0.133 kPa): 259.29±49.36 vs. 231.74±28.02, both P < 0.05]. The Alb and CRP in the REE group began to improve on the 3rd day, while the index in the HBREE group was delayed on the 5th day, overall, at the time of discharge, the PA, CRP and OI were lower in the HBREE group than in the REE group [PA (mg/L): 252.28±56.94 vs. 295.86±57.26, CRP (mg/L): 73.14±17.63 vs. 56.52±14.91, OI (mmHg): 353.59±70.36 vs. 417.52±71.58, all P < 0.01]. ③ The vasopressor was used in both groups for less than 3 days, but the REE group was shorter (days: 2.26±0.82 vs. 2.95±1.22, P < 0.05), the insulin dosage in the HBREE group was much more than that in the REE group (U: 101.97±21.05 vs. 84.59±22.21, P < 0.01); compared with the REE group, the time of mechanical ventilation and the length of ICU stay in the HBREE group were longer (hours: 113.07±25.96 vs. 93.41±27.25, days: 10.41±3.11 vs. 8.45±2.44, both P < 0.01). There was no significant difference in the 28-day mortality between the REE group and HBREE group (17.24% vs. 24.14%, P > 0.05). Conclusions Indirect calorimetry can more accurately grasp the changing laws of REE in critically ill patients. Nutritional support with REE value can make relevant nutritional indicators as good as possible, and reduce insulin dosage, shorten vasopressor use time, the length of ICU stay and mechanical ventilation time, but does not change the 28-day mortality.

Peripartum cardiomyopathy (PPCM) often develops in the early postpartum period, especially within one month after delivery. However, in recent years, some cases of pregnancy-related cardiomyopathy with earlier or slightly later onset showed similar features with PPCM entirely. One patient with PPCM was admitted to the Department of Intensive Care Medicine of the First Affiliated Hospital of Bengbu Medical College on February 2022. After admission, the patient was exceptionally hemodynamically unstable and had repeated ventricular fibrillation. Apart from extracorporeal membrane oxygenation (ECMO) support, comprehensive treatment with cardiotonic agents and drugs to inhibit the "ventricular electric storm" was given. The cardiac structure and function were gradually improved. ECMO was successfully withdrawn on the 7th day after the onset of symptoms, and the patient was discharged after a full recovery. Early diagnosis and active intervention are the keys to improve the prognosis of patients with PPCM and avoid irreversible changes in cardiac structure.

Objective To investigate the target blood pressure level of restrictive fluid resuscitation in patients with traumatic hemorrhagic shock. Methods Sixty patients with traumatic hemorrhagic shock admitted to the First Affiliated Hospital of Bengbu Medical College from January 2016 to December 2018 were enrolled. All patients were resuscitated with sodium acetate ringer solution after admission. According to the difference of mean arterial pressure (MAP) target, the patients were divided into low MAP (60 mmHg ≤ MAP < 65 mmHg, 1 mmHg = 0.133 kPa), middle MAP (65 mmHg ≤ MAP < 70 mmHg) and high MAP (70 mmHg ≤ MAP < 75 mmHg) groups by random number table using the admission order with 20 patients in each group. Those who failed to reach the target MAP after 30-minute resuscitation were excluded and supplementary cases were deferred. The restrictive fluid resuscitation phase was divided into three phases: before fluid resuscitation, liquid resuscitation for 30 minutes and 60 minutes. The most suitable resuscitation blood pressure level was further speculated by monitoring the inflammatory markers and hemodynamics in different periods in each group of patients. Pearson correlation analysis was used to detect the correlation of variables. Results Before fluid resuscitation, there was no significant difference in hemodynamics or expressions of serum cytokines among the three groups. Three groups of patients were resuscitated for 30 minutes to achieve the target blood pressure level and maintain 30 minutes. With the prolongation of fluid resuscitation time, the central venous pressure (CVP), cardiac output (CO) and cardiac index (CI) were increased slowly in the three groups, and reached a steady state at about 30 minutes after resuscitation, especially in the high MAP group and the middle MAP group. The expressions of serum inflammatory factors in the three groups were gradually increased with the prolongation of fluid resuscitation time. Compared with the low MAP group and the high MAP group, after 30 minutes of resuscitation the middle MAP group was superior to the other two groups in inhibiting the expressions of pro-inflammatory factors tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6) and promoting anti-inflammatory factors IL-10 [TNF-α mRNA (2-ΔΔCt):0.21±0.13 vs. 0.69±0.34, 0.57±0.35; IL-6 mRNA (2-ΔΔCt): 0.35±0.31 vs. 0.72±0.39, 0.59±0.42; IL-10 mRNA (2-ΔΔCt): 1.25±0.81 vs. 0.61±0.46, 0.82±0.53; all P < 0.05], but there was no significant difference in promoting the expression of IL-4 mRNA among three groups. At 60 minutes of resuscitation, compared with the low MAP group and the high MAP group, the middle MAP group could significantly inhibit the expressions of TNF-α, IL-6 and promote IL-10 [TNF-α mRNA (2-ΔΔCt): 0.72±0.35 vs. 1.05±0.54, 1.03±0.49; IL-6 mRNA (2-ΔΔCt): 0.57±0.50 vs. 1.27±0.72, 1.01±0.64; IL-10 mRNA (2-ΔΔCt): 1.41±0.90 vs. 0.81±0.48, 0.94±0.61; all P < 0.05]. Compared with the high MAP group, the middle MAP group had significant differences in promoting the expression of IL-4 mRNA (2-ΔΔCt: 1.32±0.62 vs. 0.91±0.60, P < 0.05). There was no significant difference in serum cytokine expressions at different time points of resuscitation between the low MAP group and the high MAP group (all P > 0.05). Correlation analysis showed that there was a strong linear correlation between MAP and mRNA expressions of TNF-α, IL-6, IL-10 in the middle MAP group (r value was 0.766, 0.719, 0.692, respectively, all P < 0.01), but had no correlation with IL-4 (r = 0.361, P = 0.059). Fitting linear regression analysis showed an increase in 1 mmHg per MAP, the expression of TNF-α mRNA increased by 0.027 [95% confidence interval (95%CI) = 0.023-0.031, P < 0.001], IL-6 mRNA increased by 0.021 (95%CI = 0.017-0.024, P < 0.001), and IL-10 mRNA increased by 0.049 (95%CI = 0.041-0.058, P < 0.001). Conclusions When patients with traumatic hemorrhagic shock received restrict fluid resuscitation at MAP of 65-70 mmHg, the effect of reducing systemic inflammatory response and improving hemodynamics is better than the target MAP at 60-65 mmHg or 70-75 mmHg. It is suggested that 65-70 mmHg may be an ideal target MAP level for restrictive fluid resuscitation.

Objective:To analyze the relationship between the expression of microRNA-126 (miR-126) in peripheral blood lymphocytes with apoptosis and prognosis in patients with sepsis, and to explore its potential regulatory mechanism.Methods:Thirty patients with general infection and 20 patients with sepsis admitted to the department of intensive care unit (ICU) of the First Affiliated Hospital of Bengbu Medical College from January to December 2019 were enrolled. Peripheral blood was taken to separate lymphocytes, and the expressions of miR-126 and caspase-3 were detected by reverse transcription-polymerase chain reaction (RT-PCR). At the same time, the liver and kidney function and other laboratory indexes were measured, and the sequential organ failure assessment (SOFA) and acute physiology and chronic health evaluation Ⅱ (APACHEⅡ) scores were calculated. The 28-day prognosis was observed. Pearson method was used to analyze the correlation between miR-126 and caspase-3, APACHEⅡ score. Receiver operating characteristic (ROC) curve was used to analyze the predictive value of miR-126 on prognosis; at the same time, according to the best cut-off value of miR-126 in predicting prognosis, the patients were divided into two groups, and the 28-day Kaplan-Meier survival curve was drawn.Results:The expression of miR-126 in peripheral blood lymphocytes of patients with sepsis was lower than that of patients with general infection [miR-126 mRNA (2 -ΔCt): 1.239±0.134 vs. 1.599±0.110, P < 0.01], while the expression of caspase-3 and APACHEⅡ score were significantly increased [caspase-3 mRNA (2 -ΔCt): 1.172±0.132 vs. 0.901±0.143, APACHEⅡ: 19.75±3.74 vs. 12.63±3.94, both P < 0.01]. Pearson correlation analysis showed that the expression of miR-126 was negatively correlated with the expression of caspase-3 ( r = -0.678, P < 0.001) and APACHEⅡ score ( r = -0.581, P < 0.001). ROC curve analysis showed that the area under the ROC curve (AUC) for predicting the prognosis by miR-126 expression in peripheral blood lymphocytes was 0.823 ( P < 0.001). When the best cut-off value was 1.395, the sensitivity was 75.0%, the specificity was 71.4%, the positive predictive value was 81.1%, the negative predictive value was 63.6%, the positive likelihood ratio was 2.622, and the negative likelihood ratio 0.350. In addition, the patients were divided into high miR-126 group (miR-126 > 1.395, n = 31) and low miR-126 group (miR-126 ≤ 1.395, n = 19) according to the best cut-off value of miR-126. Kaplan-Meier survival curve analysis showed that the 28-day cumulative survival rate of high miR-126 group was higher than that of low miR-126 group (Log-Rank: χ 2 = 11.702, P = 0.001). Conclusion:miR-126 in peripheral blood lymphocytes of patients with sepsis may affect immune status by promoting apoptosis of lymphocytes, and its expression level can reflect the severity and prognosis of sepsis.

Objective:To observe the protective effect of Angong Niuhuang pill on brain function of rats with sepsis, explore its protective mechanism, and provide the experimental basis for clinical application of Angong Niuhuang pill in the treatment of sepsis-associated encephalopathy (SAE).Methods:Thirty male Sprague-Dawley (SD) rats were divided into sham operation group, sepsis model group and Angong Niuhuang pill group according to random number table method, with 10 rats in each group. The sepsis model was established by cecal ligation and puncture (CLP); rats in sham operation group received open and closed abdomen. The rats in the Angong Niuhuang pill group were given Angong Niuhuang pill (0.3 g/kg) by gastric irrigation daily for 3 days before CLP, and the drugs were administrated 12 hours after modeling again. After 24 hours of CLP, the neuroreflex scores were evaluated, white blood cell count (WBC), the levels of serum neuron-specific enolase (NSE) and S100β were detected. Then the brain tissue was harvested. After hematoxylin-eosin (HE) staining, the pathological changes of brain tissue were observed under the light microscope. The mRNA expressions of interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) in brain tissue were detected by polymerase chain reaction.Results:Compared with the sham operation group, the total score of neuroreflex scores in the sepsis model group and the Angong Niuhuang pill group were significantly reduced (4.43±1.40, 6.57±1.90 vs. 9.40±0.84, both P < 0.05), WBC, serum NSE, S100β were significantly increased [WBC (×10 9/L): 8.07±1.32, 5.84±0.94 vs. 3.60±0.32; NSE (μg/L): 1.04±0.14, 0.61±0.07 vs. 0.16±0.04; S100β (ng/L): 255.624±30.25, 97.72±15.41 vs. 46.88±12.03, all P < 0.05], and the mRNA expressions of IL-6 and TNF-α in brain tissue were significantly increased [IL-6 mRNA (2 -ΔΔCt): 5.668±2.195, 3.605±1.014 vs. 0.997±0.329; TNF-α mRNA (2 -ΔΔCt): 18.996±0.913, 1.746±0.710 vs. 0.674±0.132, all P < 0.05]. Compared with the sepsis model group, the total score of neuroreflex scores in the Angong Niuhuang pill group was significantly increased (6.57±1.90 vs. 4.43±1.40, P < 0.05), WBC, serum NSE, S100β concentration, and the mRNA expressions of IL-6 and TNF-α in the brain were significantly reduced [WBC (×10 9/L): 5.84±0.94 vs. 8.07±1.32, NSE (μg/L): 0.61±0.07 vs. 1.04±0.14, S100β (ng/L): 97.72±15.41 vs. 255.62±30.25, IL-6 mRNA (2 -ΔΔCt): 3.605±1.014 vs. 5.668±2.195, TNF-α mRNA (2 -ΔΔCt): 1.746±0.710 vs. 18.996±0.913, all P < 0.05]. Brain histopathological observation showed that the hippocampal neurons in the sepsis model group were disordered arrangement, a large number of neuronal nuclei were contracted, and the tissue was loose with obvious edema. Compared with the sepsis model group, the Angong Niuhuang pill group had less nuclear shrinkage and tissue edema. Conclusions:The pretreatment of the Angong Niuhuang pill can improve the brain dysfunction of septic rats and reduce the expression of pro-inflammatory cytokines in the brain. It is speculated that the Angong Niuhuang pill can protect the brain function in sepsis by inhibiting the inflammatory reaction in the brain.

Objective:To study the clinical characteristics of different types of neonatal sepsis.Methods:From January 2012 to December 2019, neonates with confirmed sepsis from 5 neonatal centers of central-south China were reviewed. The neonates were assigned into early-onset sepsis (EOS) and late-onset sepsis (LOS) group, and the latter was further subgrouped into hospital-acquired LOS (hLOS) group and community-acquired LOS (cLOS) group. The etiological and clinical characteristics were analyzed. SPSS 26.0 was used for statistical analysis.Results:A total of 580 neonates were enrolled, including 286 (49.3%) in the EOS group and 294 (50.7%) in the LOS group. In LOS group, 147 were in hLOS group and 147 were in cLOS group. The gestational age and birth weight of hLOS group were significantly lower than the other two groups [(32.7±3.6) weeks vs. (37.1±3.7) weeks and (37.7±3.0) weeks, (1 810±717) g vs. (2 837±865) g and (3 024±710) g] ( P<0.05). The common pathogens in EOS and cLOS groups were coagulase-negative staphylococci and Escherichia coli, while Klebsiella pneumoniae was common in hLOS group. Carbapenems usage in the hLOS group was significantly higher than the other two groups [62.6% vs. 28.7% and 16.2%] ( P<0.05). Antibiotics duration in the hLOS group was longer than the other two groups [19 (14, 27) d vs. 15 (12, 20) d and 14 (12, 19) d] ( P<0.05). Conclusions:The clinical characteristics of neonatal sepsis vary among different types of infections, and it is necessary to establish appropriate prevention, control, diagnosis and treatment protocols.

OBJECTIVE: To investigate the effects of restrictive fluid management in patients with severe traumatic brain injury (sTBI). METHODS: Between January, 2019 and June, 2020, we randomly assigned 51 postoperative patients (stay in the ICU of no less than 7 days) with sTBI into treatment group ( RESULTS: The cumulative fluid balance of the two groups were positive on day 1 and negative on days 3 and 7 after ICU admission; at the same time points, the patients in the treatment group had significantly greater negative fluid balance than those in the control group ( CONCLUSIONS Restrictive fluid management can reduce cerebral edema and improve the prognosis but does not affect the 28-day mortality of patients with sTBI.
Brain Injuries, Traumatic/therapy* ; Fluid Therapy ; Humans ; Prognosis ; Respiration, Artificial ; Treatment Outcome