Objective To investigate the incidence and types of patient-ventilator asynchrony(PVA)in mechanically ventilated patients within the intensive care unit(ICU),and to identify associated high-risk factors,thereby providing a basis for reducing PVA,enhancing mechanical ventilation efficiency,and refining ventilation strategies.Methods A prospective observational study was conducted among patients admitted to the general ICU of the Affiliated Hospital of Guizhou Medical University from October to December 2024 who were receiving mechanical ventilation.Inclusion criteria were as follows:age ≥18 years and mechanical ventilation duration ≥12 hours.Exclusion criteria included complete controlled mechanical ventilation,palliative care or do-not-resuscitate status,and lack of informed consent.Senior respiratory therapists performed daily bedside observations of ventilator waveforms for 10～15 minutes between 08:00 and 12:00.PVA was diagnosed based on pressure-time and flow-time waveforms,with the types of PVA being recorded.Demographic and clinical data,including age,sex,body mass index(BMI),primary diagnosis,comorbidities,APACHEⅡ score at ICU admission,blood gas analysis,ventila-tion mode and parameters,analgesia and sedation status,duration of mechanical ventilation,and length of ICU stay,were collected.The incidence and types of PVA during the observation period were analyzed.Univariate and multivariate logistic regression analyses were performed to identify high-risk factors for PVA.Clinical outcomes were compared between patients with and without PVA.Results A total of 105 patients and 453 episodes of assisted mechanical ventilation waveforms were analyzed.Among these,60.95％(64/105)experienced at least one episode of PVA.Of the 453 ventilation waveforms assessed,35.76％(162/453)demonstrated PVA.The types of PVA,ranked by incidence,were as follows:cycling mismatch(12.58％,57/453),double triggering(11.92％,54/453),ineffective triggering(9.49％,43/453),flow starvation(5.30％,24/453),and exhalation flow limitation(1.77％,8/453).The incidence of PVA varied significantly across different ventilation modes:45.7％in volume-assist/control ventilation(V-A/C),38.1％in pressure-assist/control ventilation(P-A/C),42.9％in synchronized intermittent mandatory ventilation(SIMV),and 16.7％in pressure support ventilation(PSV)(P<0.001).Multi-variate logistic regression analysis revealed that the mechanical ventilation mode[reference:PSV;V-A/C:OR=4.687,95％CI:2.140～10.263,P<0.001;P-A/C:OR=2.922,95％CI:1.489～5.734,P=0.002;SIMV:OR=4.682,95％CI:1.758～12.466,P=0.002]and actual respiratory rate(OR=1.07,95％CI:1.016～1.127,P=0.011)were significant high-risk factors for PVA.Patients with PVA had a significantly longer duration of mechanical ventilation[8.21(5.35,13.91)days vs.3.00(1.96,5.71)days,P<0.001]compared to those without PVA.Conclusions PVA is commonly observed in ICU patients receiving assisted invasive mechanical ventilation,with cycling mismatch,double triggering,and ineffective triggering being the most prevalent types.The incidence of PVA tends to be lower when using the PSV mode.Clinically,real-time monitoring of patient-ventilator synchrony via ventilator waveforms,along with the optimization of ventilator modes and parameters,should be employed to minimize the occurrence of PVA and enhance the efficiency of mechanical ventilation.

N⁶-methyladenosine (m⁶A) modification plays a critical role in cell cycle regulation, while the mechanism of m⁶A in regulating mitosis remains underexplored. Here, we found that the total m⁶A modification level in cells increased during mitosis by the liquid chromatography-mass spectrometry/mass spectrometry and m⁶A dot blot assays. Silencing methyltransferase-like 3 (METTL3) or METTL14 results in delayed mitosis, abnormal spindle assembly, and chromosome segregation defects by the immunofluorescence. By analyzing transcriptome-wide m⁶A targets in HeLa cells, we identified polo-like kinase 1 (PLK1) as a key gene modified by m⁶A in regulating mitosis. Specifically, through immunoblotting and RNA pulldown, m⁶A modification inhibits PLK1 translation via YTH N⁶-methyladenosine RNA binding protein 1, thus mediating cell cycle homeostasis. Demethylation of PLK1 mRNA leads to significant mitotic abnormalities. These findings highlight the critical role of m⁶A in regulating mitosis and the potential of m⁶A as a therapeutic target in proliferative diseases such as cancer.
Humans ; Polo-Like Kinase 1 ; Cell Cycle Proteins/metabolism* ; Proto-Oncogene Proteins/metabolism* ; Protein Serine-Threonine Kinases/metabolism* ; Mitosis/physiology* ; HeLa Cells ; Adenosine/genetics* ; Methyltransferases/metabolism* ; RNA, Messenger/metabolism* ; RNA-Binding Proteins/metabolism*

OBJECTIVE To investigate the anti-inflammatory effect of electroacupuncture in mice with chronic obstructive pul-monary disease(COPD)and its underlying mechanisms.METHODS Forty C57BL/6 mice were randomly divided into normal group,model group,electroacupuncture(EA)group,vagotomy group,and vagotomy+EA group,with 8 mice in each group.Except for the normal group,all groups were exposed to cigarette smoke for 12 weeks to establish a COPD model.After model establishment,the vagotomy group and the vagotomy+EA group underwent left cervical vagotomy before EA.EA treatment was performed at the Feishu(BL13)and Zusanli(ST36)acupoints once daily for 20 minutes for a total of 14 days.After EA,the pulmonary ventilation function of the mice was detected by a pulmonary function analysis system;lung tissue pathology was observed by HE staining;the levels of inter-leukin(IL)-6,tumor necrosis factor-α(TNF-α),IL-10,and transforming growth factor-β(TGF-β)in lung tissue were detected by ELISA;the expression of CD86 and CD206 proteins in lung tissue was detected by Western blot;the distribution of F4/80+CD86+(M1 type)and F4/80+CD206+(M2 type)cells in lung tissue was determined by flow cytometry;the expression of CD86 and CD206 in lung tissue was observed by immunofluorescence.RESULTS Compared with the normal group,the model group showed significantly decreased lung function(P<0.01),obvious lung pathological damage,increased M1 proportion,IL-6,TNF-α,CD86 content and expression(P<0.05,P<0.01),and decreased M2 proportion,IL-10,TGF-β,CD206 content and expression(P<0.01).Compared with the model group,the EA group showed varying degrees of improvement in lung function and pathology;the M1 proportion,IL-6,TNF-α,and CD86 were reduced(P<0.05,P<0.01),while the M2 proportion,IL-10,TGF-β,and CD206 were increased(P<0.05,P<0.01).The vagotomy group showed worsened lung function and pathology,increased IL-6,TNF-α,and CD86 content and expression(P<0.05,P<0.01),and decreased IL-10,TGF-β,and CD206 content and expression(P<0.05,P<0.01).Compared with the EA group,the vagotomy+EA group showed increased M1 proportion,IL-6,TNF-α,and CD86 content and expression(P<0.01),and decreased M2 proportion,IL-10,TGF-β,and CD206 content and expression(P<0.05,P<0.01).CONCLUSION EA at Feishu(BL13)and Zusanli(ST36)acupoints can improve lung function and pulmonary inflammation in COPD mice,promoting the polarization of alveolar macrophages from M1 to M2,which is mediated by the vagus nerve.

Epitope vaccine is a novel strategy for vaccine development in recent years,exhibiting ad-vantages such as safety,and the ability to induce specific humoral and cellular immune responses a-gainst pathogens.Among them,the identification and delivery of epitopes are crucial in the design of multi-epitope vaccines.Compared to traditional methods of antigenic epitope identification,vari-ous new techniques such as bioinformatics,ELIspot,and intracellular cytokine staining have been developed.Since the direct immunization of multi-epitope peptides results in low protective effica-cy,efficient delivery systems are urgently needed.The antigenic epitopes commonly used delivery systems include viral vectors,and non-virus vectors such as nanoparticles,etc.Additionally,to en-hance the immunization efficacy of vaccines,antigenic epitopes are usually modified,including mul-tiple epitopes in tandem and epitope targeting modification and other strategies.This review sum-marizes the methods for accurate identification of antigenic epitopes with well immunogenicity as well as effective delivery strategies for antigenic epitopes,providing new insight into the develop-ment of novel multi-epitope vaccines.

Objective To evaluate the application of rabbit liver cancer model in teaching interventional medicine for graduate students.Methods A total of 10 first-year master graduate students majoring in Radiological Imaging(Interventional Medicine).who were studying at Zhengzhou University of China,were enrolled in this study.The rabbit liver cancer model was used as the experimental teaching materials.The teaching contents included the establishment of rabbit liver cancer model,the interventional operation of rabbit liver cancer,the method of scientific research and teaching,the evaluation of the teaching effect,and the survey of student satisfaction.Results Under the guidance of teaching tutor,the success rate of VX2 rabbit liver cancer modeling performed by the 10 master graduate students majoring in interventional medicine was 100％,and the mean operational quality assessment score was(11.5±2.0)points.During the operation of interventional surgery,the success rate of femoral artery puncture was also 100％,and the mean score for each interventional operation was(11.8±2.3)points.The students'experimental designs were evaluated by the expert group,the results were as follows:2 cases were rated as excellent,7 cases were rated as good,and one case was rated as moderate.The degree of students'satisfaction with experimental teaching method was high,the specific scores of each item are as follows:the understanding of the rabbit liver cancer model was(4.80±0.40)points,the command of interventional technology was(4.60±0.49)points,and the quality and practicability of teaching materials was(4.90±0.30)points.Conclusion This teaching method of using rabbit liver cancer model experiment can improve the animal experiment ability,interventional operation ability and scientific research innovation ability of graduate students.Animal model teaching method is an innovation of teaching mode for graduate students majoring in interventional medicine.

OBJECTIVE To investigate the anti-inflammatory effect of electroacupuncture in mice with chronic obstructive pul-monary disease(COPD)and its underlying mechanisms.METHODS Forty C57BL/6 mice were randomly divided into normal group,model group,electroacupuncture(EA)group,vagotomy group,and vagotomy+EA group,with 8 mice in each group.Except for the normal group,all groups were exposed to cigarette smoke for 12 weeks to establish a COPD model.After model establishment,the vagotomy group and the vagotomy+EA group underwent left cervical vagotomy before EA.EA treatment was performed at the Feishu(BL13)and Zusanli(ST36)acupoints once daily for 20 minutes for a total of 14 days.After EA,the pulmonary ventilation function of the mice was detected by a pulmonary function analysis system;lung tissue pathology was observed by HE staining;the levels of inter-leukin(IL)-6,tumor necrosis factor-α(TNF-α),IL-10,and transforming growth factor-β(TGF-β)in lung tissue were detected by ELISA;the expression of CD86 and CD206 proteins in lung tissue was detected by Western blot;the distribution of F4/80+CD86+(M1 type)and F4/80+CD206+(M2 type)cells in lung tissue was determined by flow cytometry;the expression of CD86 and CD206 in lung tissue was observed by immunofluorescence.RESULTS Compared with the normal group,the model group showed significantly decreased lung function(P<0.01),obvious lung pathological damage,increased M1 proportion,IL-6,TNF-α,CD86 content and expression(P<0.05,P<0.01),and decreased M2 proportion,IL-10,TGF-β,CD206 content and expression(P<0.01).Compared with the model group,the EA group showed varying degrees of improvement in lung function and pathology;the M1 proportion,IL-6,TNF-α,and CD86 were reduced(P<0.05,P<0.01),while the M2 proportion,IL-10,TGF-β,and CD206 were increased(P<0.05,P<0.01).The vagotomy group showed worsened lung function and pathology,increased IL-6,TNF-α,and CD86 content and expression(P<0.05,P<0.01),and decreased IL-10,TGF-β,and CD206 content and expression(P<0.05,P<0.01).Compared with the EA group,the vagotomy+EA group showed increased M1 proportion,IL-6,TNF-α,and CD86 content and expression(P<0.01),and decreased M2 proportion,IL-10,TGF-β,and CD206 content and expression(P<0.05,P<0.01).CONCLUSION EA at Feishu(BL13)and Zusanli(ST36)acupoints can improve lung function and pulmonary inflammation in COPD mice,promoting the polarization of alveolar macrophages from M1 to M2,which is mediated by the vagus nerve.

Objective To investigate the incidence and types of patient-ventilator asynchrony(PVA)in mechanically ventilated patients within the intensive care unit(ICU),and to identify associated high-risk factors,thereby providing a basis for reducing PVA,enhancing mechanical ventilation efficiency,and refining ventilation strategies.Methods A prospective observational study was conducted among patients admitted to the general ICU of the Affiliated Hospital of Guizhou Medical University from October to December 2024 who were receiving mechanical ventilation.Inclusion criteria were as follows:age ≥18 years and mechanical ventilation duration ≥12 hours.Exclusion criteria included complete controlled mechanical ventilation,palliative care or do-not-resuscitate status,and lack of informed consent.Senior respiratory therapists performed daily bedside observations of ventilator waveforms for 10～15 minutes between 08:00 and 12:00.PVA was diagnosed based on pressure-time and flow-time waveforms,with the types of PVA being recorded.Demographic and clinical data,including age,sex,body mass index(BMI),primary diagnosis,comorbidities,APACHEⅡ score at ICU admission,blood gas analysis,ventila-tion mode and parameters,analgesia and sedation status,duration of mechanical ventilation,and length of ICU stay,were collected.The incidence and types of PVA during the observation period were analyzed.Univariate and multivariate logistic regression analyses were performed to identify high-risk factors for PVA.Clinical outcomes were compared between patients with and without PVA.Results A total of 105 patients and 453 episodes of assisted mechanical ventilation waveforms were analyzed.Among these,60.95％(64/105)experienced at least one episode of PVA.Of the 453 ventilation waveforms assessed,35.76％(162/453)demonstrated PVA.The types of PVA,ranked by incidence,were as follows:cycling mismatch(12.58％,57/453),double triggering(11.92％,54/453),ineffective triggering(9.49％,43/453),flow starvation(5.30％,24/453),and exhalation flow limitation(1.77％,8/453).The incidence of PVA varied significantly across different ventilation modes:45.7％in volume-assist/control ventilation(V-A/C),38.1％in pressure-assist/control ventilation(P-A/C),42.9％in synchronized intermittent mandatory ventilation(SIMV),and 16.7％in pressure support ventilation(PSV)(P<0.001).Multi-variate logistic regression analysis revealed that the mechanical ventilation mode[reference:PSV;V-A/C:OR=4.687,95％CI:2.140～10.263,P<0.001;P-A/C:OR=2.922,95％CI:1.489～5.734,P=0.002;SIMV:OR=4.682,95％CI:1.758～12.466,P=0.002]and actual respiratory rate(OR=1.07,95％CI:1.016～1.127,P=0.011)were significant high-risk factors for PVA.Patients with PVA had a significantly longer duration of mechanical ventilation[8.21(5.35,13.91)days vs.3.00(1.96,5.71)days,P<0.001]compared to those without PVA.Conclusions PVA is commonly observed in ICU patients receiving assisted invasive mechanical ventilation,with cycling mismatch,double triggering,and ineffective triggering being the most prevalent types.The incidence of PVA tends to be lower when using the PSV mode.Clinically,real-time monitoring of patient-ventilator synchrony via ventilator waveforms,along with the optimization of ventilator modes and parameters,should be employed to minimize the occurrence of PVA and enhance the efficiency of mechanical ventilation.

Epitope vaccine is a novel strategy for vaccine development in recent years,exhibiting ad-vantages such as safety,and the ability to induce specific humoral and cellular immune responses a-gainst pathogens.Among them,the identification and delivery of epitopes are crucial in the design of multi-epitope vaccines.Compared to traditional methods of antigenic epitope identification,vari-ous new techniques such as bioinformatics,ELIspot,and intracellular cytokine staining have been developed.Since the direct immunization of multi-epitope peptides results in low protective effica-cy,efficient delivery systems are urgently needed.The antigenic epitopes commonly used delivery systems include viral vectors,and non-virus vectors such as nanoparticles,etc.Additionally,to en-hance the immunization efficacy of vaccines,antigenic epitopes are usually modified,including mul-tiple epitopes in tandem and epitope targeting modification and other strategies.This review sum-marizes the methods for accurate identification of antigenic epitopes with well immunogenicity as well as effective delivery strategies for antigenic epitopes,providing new insight into the develop-ment of novel multi-epitope vaccines.

ObjectiveTo investigate the clinical efficacy of Tenghuang Jiangu tablets （THJGT） combined with oral non-steroidal anti-inflammatory drugs （NSAIDs） in the treatment of osteoarthritis of the knee and its applicable stage based on real-world data， and provide a basis for the rational clinical use of THJGT. MethodA total of 218 cases treated with THJGT combined with oral NSAIDs included in the "THJGT for knee osteoarthritis case registry" from September 2019 to January 2021 were selected as the observation group， and 126 cases treated with oral NSAIDs alone as the control group （CG）. The data of gender， age， body mass index， Kellgren-Lawrence grading scale （K-L scale） score， visual analogue score （VAS score）， Western Ontario and McMaster Universities Osteoarthritis Index （WOMAC） score， swelling grade， joint fear of cold score， back pain and weakness score， and occurrence of adverse events/reactions of the patients in both groups were used for the evaluation of efficacy with full analysis set. The propensity score matching method was used to exclude the influence of confounding factors between groups， and the sub-data sets were established， with which the repeated measures analysis of variance （ANOVA） was carried out to evaluate the efficacy. Visit points were at registration， 4 weeks and 8 weeks after registration. The data were statistically analyzed in Excel 2019 and SPSS 23.0. ResultThe proportion of females in the observation group was 66.06% （144/218）， which was higher than that （58.73%， 74/126） in the control group （χ²=1.846）. The average age in the observation group was （61.12±7.01） years， which was higher than that ［（59.38±5.99） years］ in the control group （W=19 918.50， P<0.05）. The remission rate in the observation group was 98.17% （214/218）. In the observation group， the proportions of the patients at K-L grades Ⅱ and Ⅲ were 64.22% （144/218） and 25.23% （55/218）， respectively. The effect analysis of the whole data set for enrollment and treatment for 8 weeks showed that the VAS score of the experimental group decreased by （3.27±1.24） points on average， which was better than that of the control group [（2.75±1.20）， W=34 179.00， P<0.05]. The average WOMAC score decreased （23.43±11.46） points， which was better than that of the control group [（16.71±8.86）， W=32 387.00， P<0.05]. The average swelling grade decreased （0.63±0.64）， which was better than the control group [（0.33±0.59）， W=33 847.50， P<0.05]. The average score of joint chills decreased （1.90±1.84）， points， which was better than that of control group [（1.40±1.28）， W=35 165.00， P<0.05]. The average lumbar acid fatigue score decreased by （2.02±1.64） points, which was better than that of the control group [（1.10±1.28）， W=32 986.50， P<0.05]. Efficacy analysis of subdata sets for enrollment， 4 weeks of medication and 8 weeks of medication showed that VAS scores of both groups showed a downward trend after treatment， and the improvement of experimental group was more significant than that of control group at 4 weeks， with statistical significance （P<0.05）. After treatment， the total WOMAC score of both groups showed a downward trend， and the improvement of experimental groups was more significant at 4 weeks and 8 weeks （P<0.05）. After treatment， swelling， cold fear grade and lumbar acid fatigue score of both groups showed a decreasing trend,， and the improvement of experimental group was more significant at 8 weeks （P<0.05）. The therapeutic effect analysis of patients in the attack stage and remission stage of the experimental group showed that the total WOMAC score of the two groups showed a downward trend after treatment， and the trend was basically the same， and there was no statistical difference between the two groups at enrollment， 4 weeks after treatment， and 8 weeks after treatment （t=1.675， t=2.068， t=2.364）. The total WOMAC score of the patients in remission stage in the experimental group with K-L grading between grade 0 and grade Ⅲ had statistical significance at 4 weeks after treatment compared with the time of entry （P<0.05， P<0.01）. Group of adverse event rate was 4.13% （9/218）， lower than the control group 10.32% （13/126） （χ²= 5.109， P<0.05）. ConclusionThe population receiving THJGT combined with oral NSAIDs is mostly female， old， in remission， and with K-L grades Ⅱ and Ⅲ. THJGT can enhance the anti-inflammatory and analgesic effects of oral NSAIDs and keep the drug effect in improving joint function and alleviating fear of cold， swelling， and back pain and weakness. The drug combination can be applied to patients in both attack and remission， and the clinical application should take patient's disease stage and degree of osteoarthritis into account. Furthermore， the combination has the potential to reduce the incidence of adverse events caused by NSAIDs.

OBJECTIVE: To compare and analyze the effect of unplanned versus planned admission to the intensive care unit (ICU) on the prognosis of high-risk patients after surgery, so as to provide a clinical evidence for clinical medical staff to evaluate whether the postoperative patients should be transferred to ICU or not after surgery. METHODS: The clinical data of patients who were transferred to ICU after surgery admitted to the Affiliated Hospital of Guizhou Medical University from January to December in 2021 were retrospectively analyzed, including gender, age, body mass index, past history (whether combined with hypertension, diabetes, pulmonary disease, cardiac disease, renal failure, liver failure, hematologic disorders, tumor, etc.), acute physiology and chronic health evaluation II (APACHE II), elective surgery, pre-operative hospital consultation, length of surgery, worst value of laboratory parameters within 24 hours of ICU admission, need for invasive mechanical ventilation (IMV), duration of IMV, length of ICU stay, total length of hospital stay, ICU mortality, in-hospital mortality, and survival status at 30th day postoperative. The unplanned patients were further divided into the immediate transfer group and delayed transfer group according to the timing of their ICU entrance after surgery, and the prognosis was compared between the two groups. Cox regression analysis was used to find the independent risk factors of 30-day mortality in patients transferred to ICU after surgery. RESULTS: Finally, 377 patients were included in the post-operative admission to the ICU, including 232 in the planned transfer group and 145 in the unplanned transfer group (42 immediate transfers and 103 delayed transfers). Compared to the planned transfer group, patients in the unplanned transfer group had higher peripheral blood white blood cell count (WBC) at the time of transfer to the ICU [×10⁹/L: 10.86 (7.09, 16.68) vs. 10.11 (6.56, 13.27)], longer total length of hospital stay [days: 23.00 (14.00, 34.00) vs. 19.00 (12.00, 29.00)], and 30-day post-operative mortality was higher [29.66% (43/145) vs. 17.24% (40/232)], but haemoglobin (Hb), arterial partial pressure of carbon dioxide (PaCO₂), oxygenation index (PaO₂/FiO₂), and IMV requirement rate were lower [Hb (g/L): 95.00 (78.00, 113.50) vs. 98.00 (85.00, 123.00), PaCO₂ (mmHg, 1 mmHg ≈ 0.133 kPa): 36.00 (29.00, 41.50) vs. 39.00 (33.00, 43.00), PaO₂/FiO₂ (mmHg): 197.00 (137.50, 283.50) vs. 238.00 (178.00, 350.25), IMV requirement rate: 82.76% (120/145) vs. 93.97% (218/232)], all differences were statistically significant (all P < 0.05). Kaplan-Meier survival curve showed that the 30-day cumulative survival rate after surgery was significantly lower in the unplanned transfer group than in the planned transfer group (Log-Rank test: χ² = 7.659, P = 0.006). Univariate Cox regression analysis showed that unplanned transfer, APACHE II score, whether deeded IMV at transfer, total length of hospital stay, WBC, blood K⁺, and blood lactic acid (Lac) were associated with 30-day mortality after operation (all P < 0.05). Multifactorial Cox analysis showed that unplanned transfer [hazard ratio (HR) = 2.45, 95% confidence interval (95%CI) was 1.54-3.89, P < 0.001], APACHE II score (HR = 1.03, 95%CI was 1.00-1.07, P = 0.031), the total length of hospital stay (HR = 0.86, 95%CI was 0.83-0.89, P < 0.001), the need for IMV on admission (HR = 4.31, 95%CI was 1.27-14.63, P = 0.019), highest Lac value within 24 hours of transfer to the ICU (HR = 1.17, 95%CI was 1.10-1.24, P < 0.001), and tumor history (HR = 3.12, 95%CI was 1.36-7.13, P = 0.007) were independent risk factors for patient death at 30 days post-operative, and the risk of death was 2.45 times higher in patients unplanned transferred than in those planned transferred. Subgroup analysis showed that patients in the delayed transfer group had significantly longer IMV times than those in the immediate transfer group [hours: 43.00 (11.00, 121.00) vs. 17.50 (2.75, 73.00), P < 0.05]. CONCLUSIONS The 30-day mortality, WBC and total length of hospital stay were higher in patients who were transferred to ICU after surgery, and PaO₂/FiO₂ was lower. Unplanned transfer, oncology history, use of IMV, APACHE II score, total length of hospital stay, and Lac were independent risk factors for patient death at 30 days postoperatively, and patients with delayed transfer to ICU had longer IMV time.
Humans ; Retrospective Studies ; Respiration, Artificial ; Hospitalization ; Prognosis ; Intensive Care Units