Objective To meet the clinical need for dynamic monitoring on lactate metabolism in septic shock pa-tients,a time-series prediction model based on a long short-term memory(LSTM)network was developed to predict 24-hour lactate clearance rate at admission.Methods A multi-stage retrospective cohort design was adopted to en-roll septic shock patients admitted to the department of critical care medicine of a hospital from January 2018 to Sep-tember 2024.By conducting univariate analysis and LASSO combined feature screening,predictive factors were extracted from multidimensional clinical data.An end-to-end LSTM framework(two-layer 64/32 units,dropout rate=0.3)was constructed.A sliding window strategy(six-hour step size)was adopted for dynamic prediction and compared with traditional logistic model in terms of three dimensions:calibration(Brier score),discrimination(area under the curve[AUC]of time-dependent receiver operating characteristic[ROC]),and clinical practicality(deci-sion curve analysis).Consistency between model prediction result and actual lactate clearance rate was analyzed,and the accuracy of prediction was evaluated.Results A total of 112 septic shock patients were enrolled in the analysis,including 65 males and 47 females,with an average age of(67.35±7.28)years.65 patients were assigned in the lactate good clearance rate group(lactate good clearance rate≥10％)and 47 in the lactate poor clearance rate group(lactate good clearance rate＜10％);78 patients were in the training set and 34 in the validation set.Time-depen-dent AUC analysis revealed that the predictive performance of the LSTM model in the time windows of 6,12,and 24 hours were 0.89(0.85-0.93),0.91(0.88-0.95),and 0.92(0.89-0.96),respectively,superior to the logistic regression model(ΔAUC=0.085,P＜0.01).The core predictive factors included APACHE Ⅱ score(OR=1.38),lactate level at admission(OR=1.65),vasoactive drug dosage(OR=1.42),and 6-hour fluid resuscitation dosage(OR=1.35).The Pearson correlation coefficient between the predicted value of the model and the actual 24-hour lactate clearance rate was 0.83(P＜0.001),with an average absolute error of 8.2％.Decision curve analysis confirmed that when the threshold probability was 15％-60％,the LSTM model could increase clinical net benefits by 27.3％.The validation of each subgroup showed that the model maintained the optimal predictive performance(AUC=0.87)in the lung infection subgroup(n=16).Conclusion The LSTM-based dynamic prediction model for predicting 24-hour lactate clearance rate through integrating early admission indicators demonstrates excellent pre-dictive performance and clinical application value,which can provide important reference for individualized treatment decisions in septic shock patients.

Objective To meet the clinical need for dynamic monitoring on lactate metabolism in septic shock pa-tients,a time-series prediction model based on a long short-term memory(LSTM)network was developed to predict 24-hour lactate clearance rate at admission.Methods A multi-stage retrospective cohort design was adopted to en-roll septic shock patients admitted to the department of critical care medicine of a hospital from January 2018 to Sep-tember 2024.By conducting univariate analysis and LASSO combined feature screening,predictive factors were extracted from multidimensional clinical data.An end-to-end LSTM framework(two-layer 64/32 units,dropout rate=0.3)was constructed.A sliding window strategy(six-hour step size)was adopted for dynamic prediction and compared with traditional logistic model in terms of three dimensions:calibration(Brier score),discrimination(area under the curve[AUC]of time-dependent receiver operating characteristic[ROC]),and clinical practicality(deci-sion curve analysis).Consistency between model prediction result and actual lactate clearance rate was analyzed,and the accuracy of prediction was evaluated.Results A total of 112 septic shock patients were enrolled in the analysis,including 65 males and 47 females,with an average age of(67.35±7.28)years.65 patients were assigned in the lactate good clearance rate group(lactate good clearance rate≥10％)and 47 in the lactate poor clearance rate group(lactate good clearance rate＜10％);78 patients were in the training set and 34 in the validation set.Time-depen-dent AUC analysis revealed that the predictive performance of the LSTM model in the time windows of 6,12,and 24 hours were 0.89(0.85-0.93),0.91(0.88-0.95),and 0.92(0.89-0.96),respectively,superior to the logistic regression model(ΔAUC=0.085,P＜0.01).The core predictive factors included APACHE Ⅱ score(OR=1.38),lactate level at admission(OR=1.65),vasoactive drug dosage(OR=1.42),and 6-hour fluid resuscitation dosage(OR=1.35).The Pearson correlation coefficient between the predicted value of the model and the actual 24-hour lactate clearance rate was 0.83(P＜0.001),with an average absolute error of 8.2％.Decision curve analysis confirmed that when the threshold probability was 15％-60％,the LSTM model could increase clinical net benefits by 27.3％.The validation of each subgroup showed that the model maintained the optimal predictive performance(AUC=0.87)in the lung infection subgroup(n=16).Conclusion The LSTM-based dynamic prediction model for predicting 24-hour lactate clearance rate through integrating early admission indicators demonstrates excellent pre-dictive performance and clinical application value,which can provide important reference for individualized treatment decisions in septic shock patients.

Objective To explore the clinical efficacy of lateral osteotomy via the supra-fibular-head approach combined with lateral locking plating for simple posterolateral tibial plateau fractures.Methods From January 2014 to February 2017,11 patients with simple posterolateral tibial plateau fracture were treated at Department of Orthopaedics,General Hospital of Middle Military Command of PLA.They were 8 males and 3 females,aged from 21 to 55 years (average 41.8 years).In all of them the supra-fibular-head approach was used for lateral osteotomy according to the size and orientation of the fracture block.After the fracture was reset,a full bone graft was implanted under direct visualization,followed by fixation of the posterolateral bone mass with a lateral locking plate which was placed as far as possible above the fibular head.The patients were encouraged 6 weeks after surgery to do knee functional exercise within 90°.The therapeutic efficacy was evaluated at the final follow-up according to the Rasmussen's criteria for knee functions.Results The average operative time was 74 minutes (from 55 to 90 minutes);the average intraoperative blood loss was 70 mL (from 40 to 120 mL).All the patients were followed up for 12 to 35 months (average,17.5 months).No neurovascular injury or incision infection was observed.Bony union was achieved in all patients after 10 to 15 weeks (average,12.5 weeks).Follow-ups found no implant loosening/breakage,genu valgum,genu varum,fracture redisplacement,or knee instability.According to the Rasmussen's function evaluation of the knee joint,10 cases were rated as excellent and one as good.Conclusion Lateral osteotomy via the supra-fibular-head approach combined with lateral locking plating is a preferred treatment for simple posterolateral tibial plateau fractures.

Objective To provide anatomic evidence for chock-block technique used for acromioplasty.Methods Thirty cadaveric specimens of intact shoulder fixed by formaldehyde solution were used for this anatomic study.They were chosen from 15 males and 15 females and from 17 left and 13 right sides.First the specimens were dissected for observation of general morphology of the acromion.According to the Bigliani's classification,the acromions were sorted in morphology as type Ⅰ (flat),type Ⅱ (curved) and type Ⅲ (hooked).The shortest distance from the acromion to the humeral head (A-H) was measured.Then the location of subacromial impingement was observed and the thickness of the anterolateral angle was measured preoperatively.At last,acromioplasty using the chock-block technique was performed in the acromions of types Ⅱ and Ⅲ.Postoperatively,the distance A-H and the thickness of the anterolateral angle were measured again for comparison with the preoperative values.Results According to the Bigliani's classification,7 acromions were type Ⅰ,13 type Ⅱ and 10 type Ⅲ.Subacromial impact occurred only in the anterior acromions of types Ⅱ and Ⅲ.The distance A-H was 0.670 ± 0.035 cm in type Ⅰ acromions,significantly longer than the preoperative values of type Ⅱ (0.506 ±0.029 cm) and type Ⅲ (0.334 ±0.023 cm) (P ＜ 0.05),but showed no significant differences from the postoperative values of type Ⅱ (0.692 ±0.025 cm) and type Ⅲ (0.699 ±0.024 cm) (P ＞ 0.05).The anterolateral angular thickness of type Ⅰ acromions was O.413 ± 0.015 cm,significantly thinner than the preoperative values of type Ⅱ (0.607 ±0.014 cm) and type Ⅲ (0.623 ±0.025 cm) (P ＜ 0.05),but showed no significant differences from the postoperative values of type Ⅱ (0.414±0.035 cm) and type Ⅲ (0.412±0.033 cm) (P＞ 0.05).Conclusion The chock-block technique used for acromioplasty is effective because it can enlarge the supraspinatus outlet and deter impingement.

Objective To investigate the therapeutic effects of navigation-guided hollow screw fixation on the treatment of elderly patients with the sacroiliac complex injury.Methods Twentyone older patients with sacroiliac complex injury received the three-dimensional images-guided percutaneous hollow screw fixation from March 2015 to May 2016.There were 14 males and 7 females with age of 60-73 years[average(63.5 ± 3.7) years].Traffic accident injury was found in 15 cases,falling injury in 5 cases,others in one case.The pelvic fractures were classified by tile classification:type B2 in 2 cases,type B3 in 6 cases,type C1 in 5 cases,type C2 in 6 cases,type C3 in 2 cases.Intraoperative observation data,the injury of nerve and blood vessel during the screw insertion,postoperative fracture reposition and so on were recorded.The quality of fracture reposition was evaluated by Matta radiological criteria after surgery,and the pelvic function was assessed by Majeed scoring criteria at the last follow-up.Results A total of 30 hollow screws in 21 elderly patients were placed under three-dimensional images navigation,and each screw fixation time was 36-45 min,average (40.5±4.7)min with a small amount of bleeding (10-20)ml.Postoperative examinations showed that S1 vertebral cortical bone screw wear was found in one patient,and the end of the bolt washer was internalized into the outer table of the ilium without breaking the medial iliac plate in another patient.The above two patients had no postoperative complications.Other screws positions were good,without nerve or vascular injury and other complications.Postoperative reposition quality evaluated by the Matta radiological criteria showed that 14 cases had excellent quality,five with good quality,two with fair quality,and no case with poor quality.Therefore,the excellent and good rates were 90.5 ％.Moreover,19 elderly patients were followed up for 7-13 months average(10.3± 2.1)months,and only one patient showed the bilateral sacroiliac joint screws outside the lateral sacroiliac joint 5 mm.However,this patient had good function in the later follow-up without further prolapse.The remaining 18 patients assessed by the pelvic X-ray and CT at the end of the follow-up showed a good fracture healing,no screw breakage,loosening,and prolapse.Meanwhile,pelvic function findings evaluated by the Majeed standard demonstrated that nine cases were with excellent function,ten with good function,and the excellent and good rates were 90.5％.Conclusions Three-dimensional images-guided hollow screw fixation for the treatment of elderly patients with the sacroiliac complex injury is safe and effective with less trauma and high nailing accurateness.However,the placing process should be cautious in the elderly with osteoporosis.

Objective To analysis of risk factors for left atrial thrombosis in patients with rheumatic mitral stenosis.Methods From January 2001 to December 2008, 2277 patients with rheumatic mitral stenosis underwent operations in our hospital. There were 737 males and 1540 female, the age ranged from 19 to 84 years [average (50.9 ±10.2) years]. Left atrial thrombosis group (554 cases) and no thrombosis group (1723 cases) were divided, retrospectively collected data were analyzed with univariate and multivariate Logistic regression. Results 12 bvariables, including age, mitral valve orifice area, left atrial diameter, left ventricular diastole diameter, CRP, gender , degree of mitral stenosis, or regurgitation, degree of bicuspid regurgitation, degree of pulmonary hypertension, atrial fibrillation and heart function had statistic difference between two groups. With multivariate Logistic regression for these 12 factors, age, mitral valve orifice area, left atrial diameter, degree of mitral regurgitation and atrial fibrillation were found to be the affecting factors for left atrial thrombosis in patients with rheumatic mitral stenosis. Conclusion For patients with rheumatic mitral stenosis, age, mitral valve orifice area, left atrial diameter and atrial fibrillation are the risk factors for left atrial thrombosis. Mitral regurgitation is a protective factor for left atrial thrombosis.