Objective To provide a comprehensive overview of model performance and predictive efficacy of machine learning techniques to predict septic shock in children, in order to target and improve the quality and predictive power of models for future studies. MethodsTo systematically review all studies in four databases (PubMed, Embase, Web of Science, ScienceDirect, CNKI, WanFang Data) on machine learning prediction of septic shock in children before April 1, 2024. Two investigators independently conducted literature screening, literature data extraction and bias assessment, and conducted a systematic review of basic information, research data, study design and prediction models. Model discrimination, which area under the curve (AUC), was pooled using a random-effects model and meta-analysis was performed. Subgroup analyses were performed according to sample sizes, machine learning models, types of predictors, number of predictors, etc. And publication bias and sensitivity analyses were performed for the included literature. Results A total of 11 studies were included, of which 2 were at low risk of bias, 7 were at unknown risk of bias, and 2 were at high risk of bias. The data used in the included studies included both public and non-public electronic medical record databases, and the machine learning models used included logistic regression, random forest, support vector machine, and XGBoost, etc. The predictive models constructed based on different databases appeared to have different results in terms of the characteristic variables, so identifying the key variables of the predictive models requires further validation on other datasets. Meta-analysis showed the pooled AUC of 0.812 (95%CI 0.763 to 0.860, P<0.001), and further subgroup analyses showed that larger sample sizes (≥1 000) and predictor variable types significantly improved the predictive effect of the model, and the difference in AUC was statistically significant (95%CI not overlapping). The funnel plot showed that there was publication bias in the study, and when the extreme AUC values were excluded, the meta-analysis yielded a total AUC of 0.815 (95%CI 0.769 to 0.861, P<0.001), indicating that the extreme AUC values were insensitive. ConclusionMachine learning technology has shown some potential in predicting septic shock in children, but the quality of existing research needs to be strengthened, and future research work should improve the quality of research and improve the prediction effect of the model by expanding the sample size.
Objective To evaluate the subjective outcomes of sleepiness behavior and mood status applying continuous positive airway pressure（CPAP） in adults of elderly and middle-aged with obstructive sleep apnea syndrome（OSAS）. Methods Nine randomized controlled trails comparing nocturnal CPAP with inactive control appliances in adults with OSAS with the use of computerized search in related medical databases（MEDLINE，EMBASE，CBMdisk，etc） were included.The quality of literature was reviewed，and all data were extracted by two reviewers independently.Meta analysis was conducted used RevMan 4.2 software.Results 9 RCT involving 665 patients of elderly and middle-aged met the inclusion criteria.Meta analysis indicated that the score of Epworth sleepiness scale（ESS） and general health questionnaire-28（GHQ-28） declined significantly after CPAP treatment on effectiveness with WMD（random） －2.94，95 %CI －4.68 to －1.20，or WMD（fixed） －2.26，95 %CI －3.79 to －0.72，Plt;0.01.Nevertheless，hospital anxiety and depression scale（HADS） was not significantly different between CPAP and control with WMD（random） －0.89，95%CI －1.98 to 0.20，Pgt;0.05.Conclusion Current clinical evidence suggested that CPAP was effective in improving day-time subjective outcomes of sleepiness behavior and general mental health status in OSAS patients of elderly and middle-aged，although evidence of improving emotion disorder of anxiety and depression was not confirmed.
Objective To investigate the efficacy of LDL-C lowering treatment on NSTE-ACS, and to analyze the target LDL-C level for clinical treatment. Methods PubMed, EMbase, the Cochrane Central Register of Controlled Trials, Web of Science databases were searched up to January 2016 for randomized controlled trials assessing the effects of LDL-C lowering therapy on major adverse cardiac events (MACE) in patients with NSTE-ACS. Two reviewers independently screened litertures, extracted data and assessed the risk of bias of included studies, and then meta-analysis was performed by using Stata12.0 and RevMan 5.3 software. Result A total of 12 RCT including 4 702 individuals with NATE-ACS were included. The results of meta-analysis showed that, compared with the control group, the statin group could significantly reduced the risk of MACE (RR=0.68, 95% CI 0.549 to 0.834,P=0.000). With 18.68 months of follow-up, patients in target LDL-C level from over 70 mg/dL to less than 100 mg/dL group had lower risk of MACE than other LDL-C level group. When LDL-C lower 20% to 40% than baseline with 28.99 months follow-up, patients in target of LDL-C level from over 70 mg/dL to less than 100 mg/dL group had lowest risk of MACE (RR=20.143, 95% CI 6.946 to 58.414,P=0.000). Conclusion LDL-C lower treatment can lower the risk of MACE in patients with NSTE-ACS. Patients in target LDL-C level from over 70 mg/dL to less than 100 mg/dL group have relatively low risk of MACE, in which patients who lower 20% to 40% LDL-C than baseline will get more benefits from LDL-C lowering therapy.
Objective To evaluate systematically the effectiveness and safety of procalcitonin ( PCT) -guided therapy in comparison with standard therapy in patients with suspected or confirmed severe bacterial infections in intensive care unit ( ICU) . Methods Five randomized controlled trials ( 927 patients) were included for statistical analysis by the cochrane collaboration′s RevMan5. 0 software. Results PCT-guided therapy was associated with a significant reduction in duration of antibiotic therapy [ MD =- 2. 01, 95% CI ( - 2. 37, - 1. 64) , P lt;0. 00001] , but the mortality [ OR =1. 11, 95% CI ( 0. 83, 1. 49) ,P =0. 47] and length of ICU stay[ MD = 0. 49, 95% CI( - 1. 44, 2. 42) , P = 0. 62] were not significantly different. Conclusions An algorithmbased on serial PCT measurements would allow a more judicious use of antibiotics than currently traditional treatment of patients with severe infections in ICU. It can reduce the use of antibiotics and appears to be safe.
ObjectiveTo investigate the short-term effect of laparoscopic sleeve gastrectomy (LSG) in treatment of obesity in China. MethodsLiteratures about the effect of LSG in treatment of obesity were searched from WanFang, CNKI, PubMed, and Web of Science, then these literatures were filtered according to the inclusive and exclusive criteria and evaluated. Meta analysis was performed based on data extracted from these literatures. ResultsThree hundreds and eighty three cases of LSG were retrieved from 12 literatures. Compared with the preoperative, the body mass index (BMI) was lower in 6 months after LSG (MD=7.56, 95% CI: 5.77-9.34, P < 0.000 01), and compared with 6 months after LSG, the BMI was lower in 12 months after LSG (MD=3.19, 95% CI: 2.20-4.18, P < 0.000 01). Compared with 1 months after LSG, the excess weight loss (EWL%) was higher in 6 months after LSG (MD=-42.77, 95% CI:-45.77--39.77, P < 0.000 01), and compared with 6 months after LSG, EWL% was higher in 12 months after LSG (MD=-19.71, 95% CI:-25.54--13.87, P < 0.000 01). After the LSG, BMI decreased, but EWL% increased over time. ConclusionsIn China, LSG has obviously short-term curative effect in the treatment of obesity.
Objective To analyze the relationship between helicobacter pylori (HP) and gastric cancer. Methods We searched CNKI (Jan.1995-Dec.2005) and Wangfandatabase (Jan.1995-Dec.2005). Case-control studies on relationship of helicobacter pylori infection and gastric cancer were collected. Meta-analysis method was used to sum up the odds ratio (OR) and 95%CI of these studies.Results We identified 14 case-control studies with 11 studies of healthy adults versus gastric cancer patients and 4 studies of gastritis versus gastric cancer patients. The results of subgroup analyses based on patients resource showed: statistical difference was founded between healthy adults and gastric cancer patients with pooled OR 2.00 and 95%CI 1.25 to 3.20; no statistical difference was founded between gastritis patients and gastric cancer patients with pooled OR 1.54 and 95%CI 0.68 to 3.50. The results of subgroup analyses based on locations of gastric cancer showed: statistical difference was founded between the non-cardiac gastric cancer patients and the control with pooled OR 3.60 and 95%CI 1.25 to 10.36; no statistical difference was found between cardiac gastric cancer patients and control with pooled OR 0.88 and 95%CI 0.56 to 1.39.Conclusion HP infection can be associated with gastric cancer, and the different conclusions of the 14 reports may be attributed to the locations of gastric cancer and the selection of controls.
Objective To determine the effectiveness and safety of weekly versus three weekly regimens of taxanes for non-small cell lung cancer (NSCLC). Methods We searched The Cochrane Library (Issue 1, 2008), PubMed (1966 to May 2008), EMbase (1974 to May 2008), and CBM (1978 to May 2008) to identify randomized controlled trials (RCTs) which compared weekly and three weekly regimens of taxanes for NSCLC. Data collection was undertaken by two reviewers independently; the methodological quality was assessed according to the Cochrane Handbook 4.2.6; and the meta-analyses were performed using RevMan 5.0 software. Results Nine RCTs involving 1 438 patients were included. The results of meta-analyses showed: (1) There were no significant differences in the efficacy between weekly and three weekly regimens of taxanes regarding the one-year survival rate (paclitaxel: RR=1.24, 95%CI 0.83 to 1.86; docetaxel: RR=0.80, 95%CI 0.51 to 1.23) and the overall response rate (paclitaxel: RR=1.03, 95%CI 0.72 to 1.49; docetaxel: RR=0.98 95%CI 0.64 to 1.49). (2) The incidence of neutropenia was less serious in the weekly group (paclitaxel: RR=0.74, 95%CI 0.56 to 0.97; docetaxel: RR=0.22, 95%CI 0.16 to 0.30), while no significant differences existed in other adverse effects such as anemia and nausea/vomiting. Conclusion The efficacy of weekly and three weekly regimens of taxanes for the treatment of NSCLC is similar. The incidence of neutropenia is lower in the weekly group while other toxicities show no differences.
ObjectiveTo systematically evaluate the efficacy of high-flow nasal cannula oxygen therapy (HFNC) in Post-extubation acute exacerbation of chronic obstructive pulmonary disease (AECOPD) patients. MethodsThe Domestic and foreign databases were searched for all published available randomized controlled trials (RCTs) about HFNC therapy in post-extubation AECOPD patients. The experimental group was treated with HFNC, while the control group was treated with non-invasive positive pressure ventilation (NIPPV). The main outcome measurements included reintubation rate. The secondary outcomes measurements included oxygenation index after extubation, length of intensive care unit (ICU) stay, mortality, comfort score and adverse reaction rate. Meta-analysis was performed by Revman 5.3 software. ResultA total of 20 articles were enrolled. There were 1516 patients enrolled, with 754 patients in HFNC group, and 762 patients in control group. The results of Meta-analysis showed that there were no significant difference in reintubation rate [RR=1.41, 95%CI 0.97 - 2.07, P=0.08] and mortality [RR=0.91, 95%CI 0.58 - 1.44, P=0.69]. Compared with NIPPV, HFNC have advantages in 24 h oxygenation index after extubation [MD=4.66, 95%CI 0.26 - 9.05, P=0.04], length of ICU stay [High risk group: SMD –0.52, 95%CI –0.74 - –0.30; Medium and low risk group: MD –1.12, 95%CI –1.56- –0.67; P<0.00001], comfort score [MD=1.90, 95%CI 1.61 - 2.19, P<0.00001] and adverse reaction rate [RR=0.22, 95%CI 0.16 - 0.31, P<0.00001]. ConclusionsCompared with NIPPV, HFNC could improve oxygenation index after extubation, shorten the length of ICU stay, effectively improve Patient comfort, reduce the occurrence of adverse reactions and it did not increase the risk of reintubation and mortality. It is suggested that HFNC can be cautiously tried for sequential treatment of AECOPD patients after extubation, especially those who cannot tolerate NIPPV.
Objective To evaluate the effectiveness and safety of probiotic agents for ulcerative colitis. Methods We searched electronically the Cochrane Central Register of Controlled Trials (Issue 1, 2007), MEDLINE (1978 to 2007), EMBASE (1978 to 2007), OVID Database (1978 to 2007), Chinese Biological Medicine Database (CBM Disc) (1978 to 2007), CNKI (1979 to 2007), Chinese VIP Database (1989 to 2007) and Wanfang Database (1978 to 2007). We also checked the reference lists of retrieved articles and hand-searched 4 kinds of important journals to identify randomized controlled trials of probiotic agents for ulcerative colitis. Meta-analyses were conducted with The Cochrane Collaboration’s RevMan 4.2 software. Results Thirteen trials involving 1146 patients were included. Meta-analyses showed that probiotic agents were not superior to aminosalicylates for the clinical remission rate (OR 0.93, 95% CI 0.53 to 1.66; P=0.82); but the combination of probiotic agents and aminosalicylates were superior to aminosalicylates alone (OR 2.69, 95% CI 1.57 to 4.61; P=0.0003). In terms of the clinical relapse, the rate for probiotic agents was superior to that for placebo (OR 0.03, 95% CI 0.00 to 0.15; Plt;0.0001); but not superior to aminosalicylates (OR 0.95, 95% CI 0.65 to 1.38; P=0.79). The combination of probiotic agents and aminosalicylates was not superior to aminosalicylates alone (OR 0.57, 95% CI 0.24 to 1.32; P=0.19). As for the incidence of adverse effects, probiotic agents were not superior to aminosalicylates (OR 0.85, 95% CI 0.43 to 1.70; P=0.65); and the combination of probiotic agents and aminosalicylates was not superior to aminosalicylates alone (OR 0.30, 95% CI 0.06 to 1.54; P=0.15). Conclusion Probiotic agents are not superior to aminosalicylates based on the evidence in this review, but the combination of probiotic agents and aminosalicylates is superior to aminosalicylates alone in maintaining remission. Probiotic agents are superior to placebo but not superior to aminosalicylates, and the combination of probiotic agents and aminosalicylates is not superior to aminosalicylates alone in preventing relapse. Probiotic agents have good tolerability. However, all these findings should be interpreted with caution and more clinical trials are needed.
ObjectiveTo investigate the short-term efficacy of laparoscopic sleeve gastrectomy (LSG) for obesity with type 2 diabetes mellitus (T2MD) in China. MethodsClinical randomized controlled trial literatures about domestic LSG treatment of obesity with T2MD were obtained from Wanfang Data, China Knowledge Resource Integrated Database, PubMed and Web of Science English Data. The literatures were selected according to the inclusive and exclusive criteria, then evaluated. Methodological quality assessment and meta analysis were evaluated according to the data extracted from those literatures. The short-term efficacy (fasting blood glucose or glycosylated hemoglobin) was evaluated after operation. ResultsOne hundred and seven patients performed LSG were retrieved from 7 literatures. The levels of fasting blood glucose and glycosylated hemoglobin on the postoperative 6 months were all obviously decreased as compared with the levels before operation (fasting blood glucose:MD=2.99, 95% CI 2.39-3.60, P < 0.000 01; glycosylated hemoglobin:MD=2.24, 95% CI 1.43-3.04, P < 0.000 01), which on the postoperative 12 months were all obviously decreased as compared with the levels on the postoperative 6 months (fasting blood glucose:MD=0.56, 95% CI 0.16-0.95, P=0.006; glycosylated hemoglobin:MD=0.52, 95% CI 0.22-0.81, P=0.000 6). The postoperative fasting blood glucose and glycosylated hemoglobin levels in patients accepted LSG showed a downward trend. ConclusionsLSG on obesity with T2MD has an obvious short-term curative effect. But its long-term efficacy still needs to be supported by large samples randomized controlled clinical research data.