Objective:To explore the predictive value of very low frequency (VLF) for the efficacy of vitamin D treatment in children with postural orthostatic tachycardia syndrome (POTS).Methods:A retrospective case-control study was conducted. A total of 65 POTS children from the Department of Pediatrics of the Second Hospital of Lanzhou University from November 2023 to May 2024 were selected as the research subjects, while 102 healthy children from the same period were selected as the control group. The 25 (OH) D levels between POTS children and healthy children were compared. The frequency of eight main symptoms, including palpitations, headache, cold sweat, blurred vision, chest tightness, dizziness, fainting, and digestive discomfort in the POTS group were analyzed, the symptom scores of the orthostatic intolerance were calculated, and the parameters related to short-range heart rate variability (HRV) during the upright test, head-up tilt test (HUTT) were collected. The correlation between 25 (OH) D levels and symptom scores of orthostatic intolerance symptoms and HRV parameters in children with POTS were analyzed by Pearson correlation analysis. The total symptom scores of children with POTS were compared for 3 months before treatment and 2 months after treatment with vitamin D 800 U/d. The difference in score before and after treatment was ≥2 as responsive group, and the score difference before and after treatment was <2 as non-responsive group. The demographic characteristics including age, gender, height, weight, body mass index, as well as HRV parameters such as the standard deviation of nn intervals (SDNN), the standard deviation of the average NN intervals (SDANN), the root mean square of successive differences (rMSSD), low-frequency (LF), high-frequency (HF), very-low-frequency (VLF) were compared between the responsive group and non-responsive group. Multivariate Logistic regression was analyzed to screen independent influencing factors of vitamin D treatment efficacy for POTS. The marker molecules related to the efficacy of vitamin D treatment in children with POTS were analyzed through receiver operating characteristic (ROC) curve.Results:A total of 167 children were included in this study. There were 65 children in the POTS group, 38 males and 27 females, with the age of (10.6±2.2) years. There were 102 healthy children in the control group, 53 males and 49 females, with the age of (11.2±1.8) years.There was no statistically significant difference in gender and age between the two groups (both P>0.05). The serum 25 (OH) D level in children in the POTS group was (37±18)nmol/L, lower than that in the control group (61±17)nmol/L ( P<0.05).Pearson correlation analysis showed that serum 25(OH)D was positively correlated with VLF, SDNN and HF ( r=0.43, 0.65, 0.36, respectively, all P<0.05). Among the POTS children, there were 48 cases in responsive group and 17 cases in non-responsive group. Univariate analysis revealed that responsive group exhibited significantly lower values in age, SDNN, SDANN, and VLF compared to the non-responsive group (all P<0.05). Multivariate Logistic regression analysis showed that age ( OR=1.47, 95% CI 1.08-2.01), SDNN ( OR=1.01, 95% CI 1.00-1.22) and VLF ( OR=1.01, 95% CI 1.00-1.15) were independent influencing factors in the effect of vitamin D in POTS children (all P<0.05). ROC curve analysis showed that the area under the curve of VLF predicting response to vitamin D treatment for POTS was 0.72 (95% CI 0.58-0.85, P=0.008). When the cutoff value was 1 272 ms 2, the sensitivity and specificity were both 70%. Conclusion:VLF may help predict the efficacy of vitamin D treatment in children with POTS.

Objective:To explore the predictive value of very low frequency (VLF) for the efficacy of vitamin D treatment in children with postural orthostatic tachycardia syndrome (POTS).Methods:A retrospective case-control study was conducted. A total of 65 POTS children from the Department of Pediatrics of the Second Hospital of Lanzhou University from November 2023 to May 2024 were selected as the research subjects, while 102 healthy children from the same period were selected as the control group. The 25 (OH) D levels between POTS children and healthy children were compared. The frequency of eight main symptoms, including palpitations, headache, cold sweat, blurred vision, chest tightness, dizziness, fainting, and digestive discomfort in the POTS group were analyzed, the symptom scores of the orthostatic intolerance were calculated, and the parameters related to short-range heart rate variability (HRV) during the upright test, head-up tilt test (HUTT) were collected. The correlation between 25 (OH) D levels and symptom scores of orthostatic intolerance symptoms and HRV parameters in children with POTS were analyzed by Pearson correlation analysis. The total symptom scores of children with POTS were compared for 3 months before treatment and 2 months after treatment with vitamin D 800 U/d. The difference in score before and after treatment was ≥2 as responsive group, and the score difference before and after treatment was <2 as non-responsive group. The demographic characteristics including age, gender, height, weight, body mass index, as well as HRV parameters such as the standard deviation of nn intervals (SDNN), the standard deviation of the average NN intervals (SDANN), the root mean square of successive differences (rMSSD), low-frequency (LF), high-frequency (HF), very-low-frequency (VLF) were compared between the responsive group and non-responsive group. Multivariate Logistic regression was analyzed to screen independent influencing factors of vitamin D treatment efficacy for POTS. The marker molecules related to the efficacy of vitamin D treatment in children with POTS were analyzed through receiver operating characteristic (ROC) curve.Results:A total of 167 children were included in this study. There were 65 children in the POTS group, 38 males and 27 females, with the age of (10.6±2.2) years. There were 102 healthy children in the control group, 53 males and 49 females, with the age of (11.2±1.8) years.There was no statistically significant difference in gender and age between the two groups (both P>0.05). The serum 25 (OH) D level in children in the POTS group was (37±18)nmol/L, lower than that in the control group (61±17)nmol/L ( P<0.05).Pearson correlation analysis showed that serum 25(OH)D was positively correlated with VLF, SDNN and HF ( r=0.43, 0.65, 0.36, respectively, all P<0.05). Among the POTS children, there were 48 cases in responsive group and 17 cases in non-responsive group. Univariate analysis revealed that responsive group exhibited significantly lower values in age, SDNN, SDANN, and VLF compared to the non-responsive group (all P<0.05). Multivariate Logistic regression analysis showed that age ( OR=1.47, 95% CI 1.08-2.01), SDNN ( OR=1.01, 95% CI 1.00-1.22) and VLF ( OR=1.01, 95% CI 1.00-1.15) were independent influencing factors in the effect of vitamin D in POTS children (all P<0.05). ROC curve analysis showed that the area under the curve of VLF predicting response to vitamin D treatment for POTS was 0.72 (95% CI 0.58-0.85, P=0.008). When the cutoff value was 1 272 ms 2, the sensitivity and specificity were both 70%. Conclusion:VLF may help predict the efficacy of vitamin D treatment in children with POTS.

Objective To evaluate the surgical treatments of refractory sclerosing peritonitis related peritoneal dialysis.Methods Clinical data of 15 patients with refractory sclerosing peritonitis related to peritoneal dialysis treated in the General Surgery Department of the 900th Hospital of the Joint Logistics Support Force of the People's Liberation Army from June 30,2014 to May 30,2018.Among them,5 cases underwent"open abdomen peritoneal catheter removal+intestinal adhesiolysis+abdominal infection flushing and drainage with catheter",4 cases underwent"laparoscopic peritoneal catheter removal+intestinal adhesiolysis+abdominal infection flushing and drainage with catheter",3 cases underwent"laparoscopic peritoneal dialysis catheter removal+abdominal infection flushing and drainage with catheter",2 cases underwent"open abdomen peritoneal dialysis catheter removal+abdominal infection flushing and drainage with catheter",and 1 case underwent"laparoscopic examination combined with laparotomy exploration and removal of lower abdominal catheter+intestinal adhesiolysis+abdominal infection flushing and drainage with catheter".Age,gender,clinical symptoms,abdominal CT examination,peripheral blood routine,blood biochemistry,blood C-reactive protein(CRP),white blood cells,biochemistry,and aetiology of peritoneal dialysis fluid were collected and followed up,and the therapeutic effect was evaluated.Results 15 patients were transferred to the Department of Surgery after ineffective treatment in the Department of Internal Medicine.Preoperatively(after 5 days of antibiotic treatment)compared to before antibiotic treatment,there were no significant changes in blood WBC,blood NEUT％,CRP,and peritoneal fluid WBC(P＞0.05).Laparoscopic exploration or laparotomy exploration was performed,during which the peritoneal dialysis catheter was removed and the abdominal infection focus was cleared.A pelvic cavity washout drainage tube was left in place postoperatively.Fourteen patients had a good recovery after surgery,with effective control of peritonitis symptoms and no complications such as intestinal obstruction or enterocutaneous fistula.After the removal of the peritoneal dialysis catheter,all patients switched to hemodialysis.A comparison of inflammatory markers before and after surgery showed a significant decrease after surgery.Three days postoperatively compared to before surgery(after 5 days of antibiotic treatment),there were no significant changes in blood WBC,blood NEUT％,CRP,and peritoneal fluid WBC(P＞0.05).Seven days postoperatively compared to before surgery(after 5 days of antibiotic treatment),there was a significant decrease in blood WBC[(7.43±2.65)× 109/L VS(10.17±5.24)× 109/L],blood NEUT％[(88.23±9.02)％VS(85.07±11.57)％],and CRP[(152.88±113.01)mg/L VS(114.49±92.97)mg/L](P＜0.05);the peritoneal fluid WBC at 7 days postoperatively showed no significant change compared to before surgery(after 5 days of antibiotic treatment)(P＞0.05).The cases were followed up for at least 22 months,and 13 patients did not experience peritonitis or intestinal obstruction again.One patient died 39 days after surgery due to multiple organ failure,and one patient died from other causes after a 2-year follow-up.Conclusion For refractory sclerosing peritonitis related peritoneal dialysis that is ineffective in medical conservative treatment,On the basis of reasonable and effective antibiotics to control infection,surgical intervention should be actively carried out and surgical methods such as surgery should be used to control the progress of peritonitis,reduce mortality and improve the cure rate.

Objective To compare the set up errors derived from different registration methods of the X-ray volume imaging (XVI) system for radiotherapy in the treatment of middle/lower-segment esophageal cancer, and to provide a reference for radiation treatment of esophageal cancer. Methods We randomly selected 63 patients with middle/lower-segment esophageal cancer, and obtained their reconstructed XVI images at the first therapy to perform automatic registration with gray-value and bone registration methods. We acquired and compared the three translation errors (along x [left to right], y [head to feet], and z [front to back] axes) and three rotation errors (around the x, y, and z axes) derived from the two registration methods. Results Gray-value registration had significantly smaller translation errors along the x and z axes than bone registration (x azes t = −2.78, z azes t = −2.15, P < 0.05), but there was no significant difference along the y axes (P > 0.05). The rotation errors around the three axes were all smaller than 1°, and were smaller with gray-value registration than with bone registration, but without significant differences (P > 0.05). Conclusion We recommend gray-value registration for radiotherapy in the treatment of middle/lower-segment esophageal cancer. Manual verification or fine-tuning is recommended after automatic registration in clinical practice. Besides translation errors, rotation errors should also be paid attention to.

Objective:To investigate the curative effect of the Pavlik harness and polyester material casting bandage for the early treatment of congenital knee dislocation (CDK).Methods:The clinical data of 12 patients (13 knees) with CDK treated in Provincial Hospital Affiliated to Shandong First Medical University from December 2011 and January 2019 were studied retrospectively.There were 3 males and 9 females.The median age at the initial visit was 3 days ( 1-29 days). Eleven patients had unilateral CDK, and 1 patient had bilateral CDK.There were 5 cases(5 knees) with dislocation and 7 cases(8 knees) with subluxation.Associated musculoskeletal anomalies were observed in 7 patients, which included developmental dysplasia of the hip (DDH) in 6 patients, bilateral DDH and right clubfoot in 1 patient.The Pavlik harness was applied to keep the knee in position for 1.5-2.0 months if passive flexion achieved 90°or more.If passive flexion was below 90°, the polyester material casting bandage was employed to increase knee flexion, and the bandage was changed once a week.After passive flexion reached 90°, the patient was treated with the Pavlik harness for 1.5-2.0 months.Cases with DDH who worn the Pavlik harness were evaluated by ultrasonography in the follow-up period, and cases with clubfoot were treated with Ponseti polyester material casting bandages.Results:Three children (4 knees) were directly fixed with Pavlik harness for 1.5-2.0 months, 9 children (9 knees) were fixed with Pavlik harness for 1.5-2.0 months after 1-8 times (average 2.9 times) of plaster fixation. No plaster and Pavlik harness complications were found during the treatment.Twelve patients were followed up for 29.8 months on average (12.0 to 65.0 months). Ten patients (11 knees) were rated excellent, and 2 patients (2 knees) were rated good.Conclusions:The Pavlik harness and polyester material casting bandage for early treatment of CDK is simple, effective and safe.Associated musculoskeletal anomalies can also be corrected in the treatment process.The knee joint dislocation should be treated at first.

Objective:To investigate the effect of activation of microglia in prefrontal cortex on long-term spatial memory in post-stroke depression mice.Methods:Forty-eight male C57BL/6 mice were divided into sham operation group, stroke group, post-stroke depression group and depression group according to the random number table method with 12 in each group, and 36 mice were divided into solvent group, enrofloxacin group and minocycline group according to the random number table method with 12 in each group.Middle cerebral artery occlusion (MCAO) was use to establish the stroke model, and forced swimming was used to establish the depression model.The post-stroke depression model mice were received MCAO first and then received forced swimming on the 4th day after stroke to establish the model.Mice in enrofloxacin group and minocycline group were treated with enrofloxacin and minocycline injection once a clay for 14 days from the 5th day after stroke, respectively.Forced swimming test and sugar water preference test were used to evaluate the depression of mice in each group, Morris water maze test was used to detect the spatial memory function of mice in each group, and Nissl staining and immunofluorescence staining were used to detect the neuronal function and the number and type of microglia activation.The expression of inflammatory cytokines IL-6 and IL-1β were detected by Western blot.GraphPad Prism 8.0.1 statistical software was used for statistical analysis.The single factor variance analysis was used to compare the difference among multiple groups, and pairwise comparison was performed with SNK- q test. Results:(1) There were statistically significant differences in depression, learning and memory, neuron damage, activation of microglia, inflammatory factors and other indicators in sham operation group, stroke group, post-stroke depression group and depression group ( F=43.58-255.70, all P<0.05). Compared with stroke group, post-stroke depression group had longer floating immobility time ((222.70±29.12) s, (79.25±46.78) s, P<0.05), the preference rate of sugar water was significantly lower ( (49.44±6.19) %, (84.49±4.73) %, P<0.05), and the average value of platform approach after correction was higher((125.00±9.95) mm, (96.79±12.57) mm, P<0.05), Nissl bodies expression was lower ((53.50±15.78) cells /mm 2, (85.67±17.52) cells /mm 2, P<0.05), NeuN positive expression rate was lower ((29.78±3.70) %, (45.73±4.51) %, P<0.05), the percent of M1 microglia expression was significantly higher ((75.55±8.84) %, (58.19±5.69) %, P<0.05), the percent of M2 microglia expression was lower ((43.46±5.11)%, (57.14±5.40)%, P<0.05), and the expression levels of IL-6 ((1.14±0.03), (0.94±0.05), P<0.05) and IL-1β((1.17±0.03), (0.56±0.04), P<0.05) were significantly higher.(2) Depression, learning and memory, neuron injury, activation of microglia, inflammatory factors and other indicators of mice in solvent group, enrofloxacin group and minocycline group were significantly different ( F=7.13-94.35, all P<0.05). Compared with enrofloxacin group, mice in minocycline group had shorter floating immobility time ((169.30±13.04) s, (224.30±22.60) s, P<0.05) and higher sugar water preference rate ((62.81±7.75) %, (47.71±8.11) %, P<0.05), the mean value of platform approach estimation after water maze correction was lower ((97.66±14.56) mm, (120.20±12.08) mm, P<0.05), and the expression level of Nissl bodies was higher ((80.17±10.55) cells /mm 2, (52.00±8.94) cells /mm 2, P<0.05), NeuN expression rate was high ((45.04±3.62) %, (28.88±4.50) %, P<0.05), Iba-1 expression was lower ((97.33±10.67) cells/mm 2, (112.50±6.54)cells/mm 2, P<0.05), the percent of M1 microglia expression was lower ((54.43±5.22) %, (73.82±6.88) %, P<0.05), and the percent of M2 microglia expression was significantly higher ((51.86±6.22) %, (36.30±5.72) %, P<0.05). The expression levels of IL-6 ((0.75±0.06), (1.21±0.07), P<0.05) and IL-1β ((0.61±0.06) (1.09±0.09), P<0.05) were lower. Conclusion:The long-term spatial memory impairment of post-stroke depression mice is aggravated, which is related to the neuron damage caused by increased activation of M1 microglia in PFC area.Inhibition of M1 microglia by minocycline can effectively improve the spatial memory ability of mice.

Objective:To investigate the effect of Spautin-1 (an inhibitor of autophagy) on improving anxiety-like behaviors and its mechanism in mice after traumatic brain injury (TBI).Methods:Thirty-six C57BL/6 mice were randomly divided into sham-operated group, TBI group, and TBI+Spautin-1 group ( n=12); TBI models in the latter two groups were established by modified Feeney free fall epidural impingement method. Mice in TBI+Spautin-1 group were administered with Spautin-1 (2 μL, 10 mmol/L) into the lateral ventricle 10 min after modeling, but mice in the other two groups were only injected with same volume of solvent. Neuropathy Symptom Score (NSS) was used to evaluate the functions of motor, sensory and reflexes of mice on 1 st, 7 th and 14 th d of modeling. On 15 th and 16 th d of modeling, open field test and elevated plus maze test were used to evaluate the anxiety-like behaviors in mice. The number of Nissl bodies in the amygdala of mice was calculated by Nissl staining 16 d after modeling. The numbers of neuron specific nucleoprotein (NeuN) positive cells, interleukin (IL)-18 and IL-1β positive astrocytes in the amygdala were detected by immunofluorescent staining. Western blotting was used to detect the autophagy-and pyrotopic-associated protein expressions in the amygdala region of mice. Results:(1) As compared with the sham-operated group, the TBI group and TBI+Spautin-1 group had significantly increased NSS scores on 1 st and 7 th d of modeling ( P<0.05). (2) Open field test showed that as compared with the sham-operated group, the TBI group and TBI+Spautin-1 group had significantly smaller number of crossing grids, significantly decreased percentage of time spending in the central zone ([central area residence time/total time] × 100%), significantly decreased percentage of frequencies entering into opening arm (OE) (OE/[OE+frequencies of entering closing arm]×100%) and opening arm time (OT) percentage (OT/[OT+time of closing arm]×100%); as compared with the TBI group, TBI+Spautin-1 group had significantly larger number of crossing grids, and significantly increased time percentage spending in the central zone, OE percentage, and OT percentage ( P<0.05). (3) As compared with sham-operated group, the TBI group and TBI+Spautin-1 group had significantly smaller numbers of Nissl bodies and NeuN positive cells in the amygdala of mice ( P<0.05); as compared with TBI group, TBI+Spautin-1 group had significantly larger numbers of Nissl bodies and NeuN positive cells in the amygdala of mice ( P<0.05). (4) As compared with the sham-operated group, the TBI group and TBI+Spautin-1 group had significantly increased percentages of IL-1β and IL-18 positive astrocytes in amygdala of mice ( P<0.05); as compared with the TBI group, the TBI+Spautin-1 group had significantly decreased percentages of IL-1β and IL-18 positive astrocytes in amygdala of mice ( P<0.05). (5) As compared with sham-operated group, the TBI group and TBI+Spautin-1 group had significantly higher protein expressions of NOD-like receptor family protein 3 (NLRP3), activated cysteine aspartate protease-1 (Caspase-1), pore-forming protein D-N terminal fragment (GSDMD-N), ubiquitin specific peptidase (USP) 13 and B-lymphocytoma-2 interacting protein (Beclin1), and statistically higher ratio of microtubule-associated protein 1 light chain (LC)3 II/LC3 I ( P<0.05); as compared with TBI group, the TBI+Spautin-1 group had significantly decreased protein expressions of NLRP3, activated Caspase-1, GSDMD-N, USP13 and Beclin1 in the amygdala, and statistically lower ratio of LC3 II/LC3 I ( P<0.05). Conclusion:Spautin-1 improves the anxiety-like behaviors in mice after TBI, whose mechanism may be associated with the inhibition of astrocytic pyroptosis in the amygdala.

Objective The literature study the setup errors of head and neck, thoracic, abdominal and pelvic tumors by megavoltage fan-beam CT based image guidance in TOMO-HD to provide the margin enlarging from clinic target volume (CTV) to planning target volume (PTV) in treatment planning system of TOMO-HD. Methods 103 patients with head and neck (30 patients), thoracic (42 patients), abdominal and pelvic (31 patients) carcinoma were enrolled. Megavoltage fan-beam CT based image guidance in tomotherapy-HD was used to acquire CT scan before every treatment. The left-right (X), superior-inferior (Y), anterior-posterior (Z) and rotation (Fy) setup errors of patients can be obtained from the tomography image automatically restructured by the system. Calculating the systematic error and the random error in the three dimensions and check whether the setup data accord with the normal distribution or not, then acquire the data expand in the three directions. Results According to 2593 fan-beam CT scans, the shift errors (µ ± s) in X, Y, Z and Fy (rotation) of three study group were [(−0.31 ± 2.16) mm、(1.09 ± 3.56) mm、(2.36 ± 2.27) mm, (0.29 ± 0.96)°] (head and neck tumor), [(−0.98 ± 2.95) mm、(0.45 ± 6.86) mm、(3.79 ± 2.47) mm, (0.18 ± 0.60)°] (thoracic cancer) and [(−0.86 ± 2.85) mm、(−1.59 ± 6.91) mm、(5.77 ± 2.40) mm, (0.20 ± 0.68)°](abdominal and pelvic carcinoma). The systematic errors (∑) and random errors (σ) in X, Y, Z dimensions of patients with head and neck, thoracic, abdominal and pelvic tumors were (1.06 mm and 1.84 mm), (1.93 mm and 3.43 mm), (2.41 mm and 2.71 mm), (1.10 mm and 2.56 mm), (3.79 mm and 5.46 mm), (1.38 mm and 1.99 mm) and (1.39 mm and 0.87 mm), (4.98 mm and 5.69 mm), (1.19 mm and 2.05 mm), respectively. Conclusion It is recommended as a reference for image guidance in TOMO-HD according to the frequency distribution of setup errors, for patients with head and neck, chest and abdominal and pelvic tumors, the maximum range of motion in three dimensions are (5.00, 5.00, 5.00) mm, (6.63, 17.25, 16.00) mm and (6.49, 16.24, 13.60) mm.

Objective::Older patients with comorbidity, such as coronary heart disease (CHD) and malignant gastrointestinal tumors, are at a high risk of bleeding events. However, risk prediction models based on risk factor assessment remain unclear. This study aimed to establish an individualized bleeding risk assessment system based on the analysis of 10-year inpatient clinical big data.Methods::Total clinical data of 56,819 patients with CHD and 25,988 patients with malignant digestive tract tumors (admitted from January 2008 to December 2017) were retrospectively collected at the First and Second Medical Centers of Chinese People’s Liberation Army General Hospital. Among them, 1307 patients with CHD and malignant digestive tract tumors were screened as the derivation cohort. The dependent variable was the occurrence of major clinical bleeding events. Baseline statistics and hypothesis tests of differences were performed for independent variables according to the occurrence of bleeding. Decision Tree, eXtreme Gradient Boosting (XGBoost), logistic regression, and random forest models were used for comparison. The accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC-ROC) were applied as the criteria for evaluating and verifying model performance. To evaluate this developed model, another cohort comprising 454 patients (admitted from January 2018 to December 2019) was prospectively enrolled as the validation cohort based on the same inclusion and exclusion criteria.Results::Among the 64 variables with <50% missing values, the recursive feature elimination method with a random forest model was used to screen the selected variables. The highest accuracy was obtained following the selection of 10 scalars, and the final model was constructed accordingly. XGBoost demonstrated the best performance comprehensively. The AUC-ROC of this model was 0.981, with an accuracy, sensitivity, and specificity of 0.939, 0.950, and 0.927, respectively. In the validation cohort, the AUC-ROC, accuracy, sensitivity, and specificity of the XGBoost model were 0.702, 0.718, 0.636, and 0.725, respectively. The rate of major bleeding events has a positive correlation with the bleeding risk score quintiles. To allow for convenient clinical application, a smartphone application was developed for easy access and calculation (http://fir.master-wx.com/sghr).Conclusion::We successfully established a risk model and score for predicting bleeding events in older patients with comorbidity, such as CHD and gastrointestinal cancer.

Objective::Older patients with comorbidity, such as coronary heart disease (CHD) and malignant gastrointestinal tumors, are at a high risk of bleeding events. However, risk prediction models based on risk factor assessment remain unclear. This study aimed to establish an individualized bleeding risk assessment system based on the analysis of 10-year inpatient clinical big data.Methods::Total clinical data of 56,819 patients with CHD and 25,988 patients with malignant digestive tract tumors (admitted from January 2008 to December 2017) were retrospectively collected at the First and Second Medical Centers of Chinese People’s Liberation Army General Hospital. Among them, 1307 patients with CHD and malignant digestive tract tumors were screened as the derivation cohort. The dependent variable was the occurrence of major clinical bleeding events. Baseline statistics and hypothesis tests of differences were performed for independent variables according to the occurrence of bleeding. Decision Tree, eXtreme Gradient Boosting (XGBoost), logistic regression, and random forest models were used for comparison. The accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC-ROC) were applied as the criteria for evaluating and verifying model performance. To evaluate this developed model, another cohort comprising 454 patients (admitted from January 2018 to December 2019) was prospectively enrolled as the validation cohort based on the same inclusion and exclusion criteria.Results::Among the 64 variables with <50% missing values, the recursive feature elimination method with a random forest model was used to screen the selected variables. The highest accuracy was obtained following the selection of 10 scalars, and the final model was constructed accordingly. XGBoost demonstrated the best performance comprehensively. The AUC-ROC of this model was 0.981, with an accuracy, sensitivity, and specificity of 0.939, 0.950, and 0.927, respectively. In the validation cohort, the AUC-ROC, accuracy, sensitivity, and specificity of the XGBoost model were 0.702, 0.718, 0.636, and 0.725, respectively. The rate of major bleeding events has a positive correlation with the bleeding risk score quintiles. To allow for convenient clinical application, a smartphone application was developed for easy access and calculation (http://fir.master-wx.com/sghr).Conclusion::We successfully established a risk model and score for predicting bleeding events in older patients with comorbidity, such as CHD and gastrointestinal cancer.