PeerJ:Internal Medicinehttps://peerj.com/articles/index.atom?journal=peerj&subject=5200Internal Medicine articles published in PeerJCo-differential genes between DKD and aging: implications for a diagnostic model of DKDhttps://peerj.com/articles/170462024-02-292024-02-29Hongxuan DuKaiying HeJing ZhaoQicai YouXiaochun ZhouJianqin Wang
Objective
Diabetic kidney disease (DKD) is a serious complication of diabetes mellitus (DM) that is closely related to aging. In this study, we found co-differential genes between DKD and aging and established a diagnostic model of DKD based on these genes.
Methods
Differentially expressed genes (DEGs) in DKD were screened using GEO datasets. The intersection of the DEGs of DKD and aging-related genes revealed DKD and aging co-differential genes. Based on this, a genetic diagnostic model for DKD was constructed using LASSO regression. The characteristics of these genes were investigated using consensus clustering, WGCNA, functional enrichment, and immune cell infiltration. Finally, the expression of diagnostic model genes was analyzed using single-cell RNA sequencing (scRNA-seq) in DKD mice (model constructed by streptozotocin (STZ) injection and confirmed by tissue section staining).
Results
First, there were 159 common differential genes between DKD and aging, 15 of which were significant. These co-differential genes were involved in stress, glucolipid metabolism, and immunological functions. Second, a genetic diagnostic model (including IGF1, CETP, PCK1, FOS, and HSPA1A) was developed based on these genes. Validation of these model genes in scRNA-seq data revealed statistically significant variations in FOS, HSPA1A, and PCK1 gene expression between the early DKD and control groups. Validation of these model genes in the kidneys of DKD mice revealed that Igf1, Fos, Pck1, and Hspa1a had lower expression in DKD mice, with Igf1 expression being statistically significant.
Conclusion
Our findings suggest that DKD and aging co-differential genes are significant in DKD diagnosis, providing a theoretical basis for novel research directions on DKD.
Objective
Diabetic kidney disease (DKD) is a serious complication of diabetes mellitus (DM) that is closely related to aging. In this study, we found co-differential genes between DKD and aging and established a diagnostic model of DKD based on these genes.
Methods
Differentially expressed genes (DEGs) in DKD were screened using GEO datasets. The intersection of the DEGs of DKD and aging-related genes revealed DKD and aging co-differential genes. Based on this, a genetic diagnostic model for DKD was constructed using LASSO regression. The characteristics of these genes were investigated using consensus clustering, WGCNA, functional enrichment, and immune cell infiltration. Finally, the expression of diagnostic model genes was analyzed using single-cell RNA sequencing (scRNA-seq) in DKD mice (model constructed by streptozotocin (STZ) injection and confirmed by tissue section staining).
Results
First, there were 159 common differential genes between DKD and aging, 15 of which were significant. These co-differential genes were involved in stress, glucolipid metabolism, and immunological functions. Second, a genetic diagnostic model (including IGF1, CETP, PCK1, FOS, and HSPA1A) was developed based on these genes. Validation of these model genes in scRNA-seq data revealed statistically significant variations in FOS, HSPA1A, and PCK1 gene expression between the early DKD and control groups. Validation of these model genes in the kidneys of DKD mice revealed that Igf1, Fos, Pck1, and Hspa1a had lower expression in DKD mice, with Igf1 expression being statistically significant.
Conclusion
Our findings suggest that DKD and aging co-differential genes are significant in DKD diagnosis, providing a theoretical basis for novel research directions on DKD.Prognostic factors and outcomes of invasive pulmonary aspergillosis, a retrospective hospital-based studyhttps://peerj.com/articles/170662024-02-282024-02-28Wei-Che ChenI-Chieh ChenJun-Peng ChenTsai-Ling LiaoYi-Ming Chen
Objective
Invasive pulmonary aspergillosis (IPA) affects immunocompromised hosts and is associated with higher risks of respiratory failure and mortality. However, the clinical outcomes of different IPA types have not been identified.
Methods
Between September 2002 and May 2021, we retrospectively enrolled patients with IPA in Taichung Veterans General Hospital, Taiwan. Cases were classified as possible IPA, probable IPA, proven IPA, and putative IPA according to EORTC/MSGERC criteria and the AspICU algorithm. Risk factors of respiratory failure, kidney failure, and mortality were analyzed by logistic regression. A total of 3-year survival was assessed by the Kaplan-Meier method with log-rank test for post-hoc comparisons.
Results
We included 125 IPA patients (50: possible IPA, 47: probable IPA, 11: proven IPA, and 17: putative IPA). Comorbidities of liver cirrhosis and solid organ malignancy were risk factors for respiratory failure; diabetes mellitus and post-liver or kidney transplantation were related to kidney failure. Higher galactomannan (GM) test optical density index (ODI) in either serum or bronchoalveolar lavage fluid was associated with dismal outcomes. Probable IPA and putative IPA had lower 3-year respiratory failure-free survival compared to possible IPA. Probable IPA and putative IPA exhibited lower 3-year renal failure-free survival in comparison to possible IPA and proven IPA. Putative IPA had the lowest 3-year overall survival rates among the four IPA groups.
Conclusion
Patients with putative IPA had higher mortality rates than the possible, probable, or proven IPA groups. Therefore, a prompt diagnosis and timely treatment are warranted for patients with putative IPA.
Objective
Invasive pulmonary aspergillosis (IPA) affects immunocompromised hosts and is associated with higher risks of respiratory failure and mortality. However, the clinical outcomes of different IPA types have not been identified.
Methods
Between September 2002 and May 2021, we retrospectively enrolled patients with IPA in Taichung Veterans General Hospital, Taiwan. Cases were classified as possible IPA, probable IPA, proven IPA, and putative IPA according to EORTC/MSGERC criteria and the AspICU algorithm. Risk factors of respiratory failure, kidney failure, and mortality were analyzed by logistic regression. A total of 3-year survival was assessed by the Kaplan-Meier method with log-rank test for post-hoc comparisons.
Results
We included 125 IPA patients (50: possible IPA, 47: probable IPA, 11: proven IPA, and 17: putative IPA). Comorbidities of liver cirrhosis and solid organ malignancy were risk factors for respiratory failure; diabetes mellitus and post-liver or kidney transplantation were related to kidney failure. Higher galactomannan (GM) test optical density index (ODI) in either serum or bronchoalveolar lavage fluid was associated with dismal outcomes. Probable IPA and putative IPA had lower 3-year respiratory failure-free survival compared to possible IPA. Probable IPA and putative IPA exhibited lower 3-year renal failure-free survival in comparison to possible IPA and proven IPA. Putative IPA had the lowest 3-year overall survival rates among the four IPA groups.
Conclusion
Patients with putative IPA had higher mortality rates than the possible, probable, or proven IPA groups. Therefore, a prompt diagnosis and timely treatment are warranted for patients with putative IPA.The effects of selected sedatives on basal and stimulated serum cortisol concentrations in healthy dogshttps://peerj.com/articles/169552024-02-202024-02-20Adam HuntShelly OlinJacqueline C. WhittemoreAlejandro Esteller-VicoCary SpringerLuca Giori
Background
Hormone assessment is typically recommended for awake, unsedated dogs. However, one of the most commonly asked questions from veterinary practitioners to the endocrinology laboratory is how sedation impacts cortisol concentrations and the adrenocorticotropic hormone (ACTH) stimulation test. Butorphanol, dexmedetomidine, and trazodone are common sedatives for dogs, but their impact on the hypothalamic-pituitary-adrenal axis (HPA) is unknown. The objective of this study was to evaluate the effects of butorphanol, dexmedetomidine, and trazodone on serum cortisol concentrations.
Methods
Twelve healthy beagles were included in a prospective, randomized, four-period crossover design study with a 7-day washout. ACTH stimulation test results were determined after saline (0.5 mL IV), butorphanol (0.3 mg/kg IV), dexmedetomidine (4 µg/kg IV), and trazodone (3–5 mg/kg PO) administration.
Results
Compared to saline, butorphanol increased basal (median 11.75 µg/dL (range 2.50–23.00) (324.13 nmol/L; range 68.97–634.48) vs 1.27 µg/dL (0.74–2.10) (35.03 nmol/L; 20.41–57.93); P < 0.0001) and post-ACTH cortisol concentrations (17.05 µg/dL (12.40–26.00) (470.34 nmol/L; 342.07–717.24) vs 13.75 µg/dL (10.00–18.90) (379.31 nmol/L; 275.96–521.38); P ≤ 0.0001). Dexmedetomidine and trazodone did not significantly affect basal (1.55 µg/dL (range 0.75–1.55) (42.76 nmol/L; 20.69–42.76); P = 0.33 and 0.79 µg/dL (range 0.69–1.89) (21.79 nmol/L; 19.03–52.14); P = 0.13, respectively, vs saline 1.27 (0.74–2.10) (35.03 nmol/L; 20.41–57.93)) or post-ACTH cortisol concentrations (14.35 µg/dL (range 10.70–18.00) (395.86 nmol/L; 295.17–496.55); (P = 0.98 and 12.90 µg/dL (range 8.94–17.40) (355.86 nmol/L; 246.62–480); P = 0.65), respectively, vs saline 13.75 µg/dL (10.00–18.60) (379.31 nmol/L; 275.86–513.10).
Conclusion
Butorphanol administration should be avoided prior to ACTH stimulation testing in dogs. Further evaluation of dexmedetomidine and trazodone’s effects on adrenocortical hormone testing in dogs suspected of HPA derangements is warranted to confirm they do not impact clinical diagnosis.
Background
Hormone assessment is typically recommended for awake, unsedated dogs. However, one of the most commonly asked questions from veterinary practitioners to the endocrinology laboratory is how sedation impacts cortisol concentrations and the adrenocorticotropic hormone (ACTH) stimulation test. Butorphanol, dexmedetomidine, and trazodone are common sedatives for dogs, but their impact on the hypothalamic-pituitary-adrenal axis (HPA) is unknown. The objective of this study was to evaluate the effects of butorphanol, dexmedetomidine, and trazodone on serum cortisol concentrations.
Methods
Twelve healthy beagles were included in a prospective, randomized, four-period crossover design study with a 7-day washout. ACTH stimulation test results were determined after saline (0.5 mL IV), butorphanol (0.3 mg/kg IV), dexmedetomidine (4 µg/kg IV), and trazodone (3–5 mg/kg PO) administration.
Results
Compared to saline, butorphanol increased basal (median 11.75 µg/dL (range 2.50–23.00) (324.13 nmol/L; range 68.97–634.48) vs 1.27 µg/dL (0.74–2.10) (35.03 nmol/L; 20.41–57.93); P < 0.0001) and post-ACTH cortisol concentrations (17.05 µg/dL (12.40–26.00) (470.34 nmol/L; 342.07–717.24) vs 13.75 µg/dL (10.00–18.90) (379.31 nmol/L; 275.96–521.38); P ≤ 0.0001). Dexmedetomidine and trazodone did not significantly affect basal (1.55 µg/dL (range 0.75–1.55) (42.76 nmol/L; 20.69–42.76); P = 0.33 and 0.79 µg/dL (range 0.69–1.89) (21.79 nmol/L; 19.03–52.14); P = 0.13, respectively, vs saline 1.27 (0.74–2.10) (35.03 nmol/L; 20.41–57.93)) or post-ACTH cortisol concentrations (14.35 µg/dL (range 10.70–18.00) (395.86 nmol/L; 295.17–496.55); (P = 0.98 and 12.90 µg/dL (range 8.94–17.40) (355.86 nmol/L; 246.62–480); P = 0.65), respectively, vs saline 13.75 µg/dL (10.00–18.60) (379.31 nmol/L; 275.86–513.10).
Conclusion
Butorphanol administration should be avoided prior to ACTH stimulation testing in dogs. Further evaluation of dexmedetomidine and trazodone’s effects on adrenocortical hormone testing in dogs suspected of HPA derangements is warranted to confirm they do not impact clinical diagnosis.Risk of circulatory diseases associated with proton-pump inhibitors: a retrospective cohort study using electronic medical records in Thailandhttps://peerj.com/articles/168922024-02-152024-02-15Tanavij PannoiChissanupong PromchaiPenjamaporn ApiromruckSuwikran WongpraphairotYaa-Hui DongChen-Chang YangWen-Chi Pan
Background
Proton-pump inhibitors (PPIs) are prescribed to treat gastric acid-related diseases, while they may also have potential risks to population health. Recent studies suggested that a potential mechanism explaining the association between PPIs and cardiovascular diseases (CVD) includes the inhibition of the nitrate-nitrite-nitric oxide (NO) pathway. However, previous observational studies showed controversial results of the association. In addition, the inhibition of the NO pathway due to PPIs use may lead to peripheral vascular diseases (PVD); however, none of the studies explore the PPI-PVD association. Therefore, this study aimed to evaluate the association of PPIs with circulatory diseases (CVD, ischemic strokes or IS, and PVD).
Methods
We conducted a retrospective hospital-based cohort study from Oct 2010 to Sep 2017 in Songkhla province, Thailand. PPIs and histamine 2-receptor antagonists (H2RAs) prescriptions were collected from electronic pharmacy records, while diagnostic outcomes were retrieved from electronic medical records at Songklanagarind hospital. Patients were followed up with an on-treatment approach. Cox proportional hazard models were applied to measure the association comparing PPIs vs H2RAs after 1:1 propensity-score-matching. Sub-group analysis, multi-bias E-values, and array-based sensitivity analysis for some covariates were used to assess the robustness of associations.
Results
A total of 3,928 new PPIs and 3,928 H2RAs users were included in the 1:1 propensity score-matched cohort. As compared with H2RAs, the association of PPIs with CVD, IS, and PVD, the hazard ratios were 1.76 95% CI = [1.40–2.20] for CVD, 3.53 95% CI = [2.21–5.64] for ischemic strokes, and 17.07 95% CI = [13.82–76.25] for PVD. The association between PPIs and each outcome was significant with medication persistent ratio of over 50%. In addition, the association between PPIs and circulatory diseases was robust to unmeasured confounders (i.e., smoking and alcohol).
Conclusion
PPIs were associated with circulatory diseases, particularly ischemic strokes in this hospital-based cohort study, whereas, the strength of associations was robust to unmeasured confounders.
Background
Proton-pump inhibitors (PPIs) are prescribed to treat gastric acid-related diseases, while they may also have potential risks to population health. Recent studies suggested that a potential mechanism explaining the association between PPIs and cardiovascular diseases (CVD) includes the inhibition of the nitrate-nitrite-nitric oxide (NO) pathway. However, previous observational studies showed controversial results of the association. In addition, the inhibition of the NO pathway due to PPIs use may lead to peripheral vascular diseases (PVD); however, none of the studies explore the PPI-PVD association. Therefore, this study aimed to evaluate the association of PPIs with circulatory diseases (CVD, ischemic strokes or IS, and PVD).
Methods
We conducted a retrospective hospital-based cohort study from Oct 2010 to Sep 2017 in Songkhla province, Thailand. PPIs and histamine 2-receptor antagonists (H2RAs) prescriptions were collected from electronic pharmacy records, while diagnostic outcomes were retrieved from electronic medical records at Songklanagarind hospital. Patients were followed up with an on-treatment approach. Cox proportional hazard models were applied to measure the association comparing PPIs vs H2RAs after 1:1 propensity-score-matching. Sub-group analysis, multi-bias E-values, and array-based sensitivity analysis for some covariates were used to assess the robustness of associations.
Results
A total of 3,928 new PPIs and 3,928 H2RAs users were included in the 1:1 propensity score-matched cohort. As compared with H2RAs, the association of PPIs with CVD, IS, and PVD, the hazard ratios were 1.76 95% CI = [1.40–2.20] for CVD, 3.53 95% CI = [2.21–5.64] for ischemic strokes, and 17.07 95% CI = [13.82–76.25] for PVD. The association between PPIs and each outcome was significant with medication persistent ratio of over 50%. In addition, the association between PPIs and circulatory diseases was robust to unmeasured confounders (i.e., smoking and alcohol).
Conclusion
PPIs were associated with circulatory diseases, particularly ischemic strokes in this hospital-based cohort study, whereas, the strength of associations was robust to unmeasured confounders.The causal effects of circulating cytokines on sepsis: a Mendelian randomization studyhttps://peerj.com/articles/168602024-02-012024-02-01Weijun FangChen ChaiJiawei Lu
Background
In observational studies, sepsis and circulating levels of cytokines have been associated with unclear causality. This study used Mendelian randomization (MR) to identify the causal direction between circulating cytokines and sepsis in a two-sample study.
Methods
An MR analysis was performed to estimate the causal effect of 41 cytokines on sepsis risk. The inverse-variance weighted random-effects method, the weighted median-based method, and MR-Egger were used to analyze the data. Heterogeneity and pleiotropy were assessed using MR-Egger regression and Cochran’s Q statistic.
Results
Genetically predicted beta-nerve growth factor (OR = 1.12, 95% CI [1.037–1.211], P = 0.004) increased the risk of sepsis, while RANTES (OR = 0.92, 95% CI [0.849–0.997], P = 0.041) and fibroblast growth factor (OR = 0.869, 95% CI [0.766–0.986], P = 0.029) reduced the risk of sepsis. These findings were robust in extensive sensitivity analyses. There was no clear association between the other cytokines and sepsis risk.
Conclusion
The findings of this study demonstrate that beta-nerve growth factor, RANTES, and fibroblast growth factor contribute to sepsis risk. Investigations into potential mechanisms are warranted.
Background
In observational studies, sepsis and circulating levels of cytokines have been associated with unclear causality. This study used Mendelian randomization (MR) to identify the causal direction between circulating cytokines and sepsis in a two-sample study.
Methods
An MR analysis was performed to estimate the causal effect of 41 cytokines on sepsis risk. The inverse-variance weighted random-effects method, the weighted median-based method, and MR-Egger were used to analyze the data. Heterogeneity and pleiotropy were assessed using MR-Egger regression and Cochran’s Q statistic.
Results
Genetically predicted beta-nerve growth factor (OR = 1.12, 95% CI [1.037–1.211], P = 0.004) increased the risk of sepsis, while RANTES (OR = 0.92, 95% CI [0.849–0.997], P = 0.041) and fibroblast growth factor (OR = 0.869, 95% CI [0.766–0.986], P = 0.029) reduced the risk of sepsis. These findings were robust in extensive sensitivity analyses. There was no clear association between the other cytokines and sepsis risk.
Conclusion
The findings of this study demonstrate that beta-nerve growth factor, RANTES, and fibroblast growth factor contribute to sepsis risk. Investigations into potential mechanisms are warranted.Association between lactic acidosis and multiple organ dysfunction syndrome after cardiopulmonary bypasshttps://peerj.com/articles/167692024-01-312024-01-31Dan ZhengGuo-Liang YuYi-Ping ZhouQiao-Min ZhangChun-Guo WangSheng Zhang
Background
The relationship between hyperlactatemia and prognosis after cardiopulmonary bypass (CPB) is controversial, and some studies ignore the presence of lactic acidosis in patients with severe hyperlactacemia. This study explored the association between lactic acidosis (LA) and the occurrence of multiple organ dysfunction syndrome (MODS) after cardiopulmonary bypass.
Methods
This study was a post hoc analysis of patients who underwent cardiac surgery between February 2017 and August 2018 and participated in a prospective study at Taizhou Hospital. The data were collected at: ICU admission (H0), and 4, 8, 12, 24, and 48 h after admission. Blood lactate levels gradually increased after CPB, peaking at H8 and then gradually decreasing. The patients were grouped as LA, hyperlactatemia (HL), and normal control (NC) based on blood test results 8 h after ICU admission. Basic preoperative, perioperative, and postoperative conditions were compared between the three groups, as well as postoperative perfusion and oxygen metabolism indexes.
Results
There were 22 (19%), 73 (64%), and 19 (17%) patients in the LA, HL, and NC groups, respectively. APACHE II (24h) and SOFA (24h) scores were the highest in the LA group (P < 0.05). ICU stay duration was the longest for the LA group (48.5 (42.5, 50) h), compared with the HL (27 (22, 48) h) and NC (27 (25, 46) h) groups (P = 0.012). The LA group had the highest incidence of MODS (36%), compared with the HL (14%) and NC (5%) groups (P = 0.015). In the LA group, the oxygen extraction ratio (O2ER) was lower (21.5 (17.05, 32.8)%) than in the HL (31.3 (24.8, 37.6)%) and the NC group (31.3 (29.0, 35.4) %) (P = 0.018). In the univariable analyses, patient age (OR = 1.054, 95% CI [1.003–1.109], P = 0.038), the LA group (vs. the NC group, (OR = 10.286, 95% CI [1.148–92.185], P = 0.037), and ΔPCO2 at H8 (OR = 1.197, 95% CI [1.022–1.401], P = 0.025) were risk factor of MODS after CPB.
Conclusions
We speculated that there was correlation between lactic acidosis and MODS after CPB. In addition, LA should be monitored intensively after CPB.
Background
The relationship between hyperlactatemia and prognosis after cardiopulmonary bypass (CPB) is controversial, and some studies ignore the presence of lactic acidosis in patients with severe hyperlactacemia. This study explored the association between lactic acidosis (LA) and the occurrence of multiple organ dysfunction syndrome (MODS) after cardiopulmonary bypass.
Methods
This study was a post hoc analysis of patients who underwent cardiac surgery between February 2017 and August 2018 and participated in a prospective study at Taizhou Hospital. The data were collected at: ICU admission (H0), and 4, 8, 12, 24, and 48 h after admission. Blood lactate levels gradually increased after CPB, peaking at H8 and then gradually decreasing. The patients were grouped as LA, hyperlactatemia (HL), and normal control (NC) based on blood test results 8 h after ICU admission. Basic preoperative, perioperative, and postoperative conditions were compared between the three groups, as well as postoperative perfusion and oxygen metabolism indexes.
Results
There were 22 (19%), 73 (64%), and 19 (17%) patients in the LA, HL, and NC groups, respectively. APACHE II (24h) and SOFA (24h) scores were the highest in the LA group (P < 0.05). ICU stay duration was the longest for the LA group (48.5 (42.5, 50) h), compared with the HL (27 (22, 48) h) and NC (27 (25, 46) h) groups (P = 0.012). The LA group had the highest incidence of MODS (36%), compared with the HL (14%) and NC (5%) groups (P = 0.015). In the LA group, the oxygen extraction ratio (O2ER) was lower (21.5 (17.05, 32.8)%) than in the HL (31.3 (24.8, 37.6)%) and the NC group (31.3 (29.0, 35.4) %) (P = 0.018). In the univariable analyses, patient age (OR = 1.054, 95% CI [1.003–1.109], P = 0.038), the LA group (vs. the NC group, (OR = 10.286, 95% CI [1.148–92.185], P = 0.037), and ΔPCO2 at H8 (OR = 1.197, 95% CI [1.022–1.401], P = 0.025) were risk factor of MODS after CPB.
Conclusions
We speculated that there was correlation between lactic acidosis and MODS after CPB. In addition, LA should be monitored intensively after CPB.Establishment and validation of a heart failure risk prediction model for elderly patients after coronary rotational atherectomy based on machine learninghttps://peerj.com/articles/168672024-01-312024-01-31Lixiang ZhangXiaojuan ZhouJiaoyu Cao
Objective
To develop and validate a heart failure risk prediction model for elderly patients after coronary rotational atherectomy based on machine learning methods.
Methods
A retrospective cohort study was conducted to select 303 elderly patients with severe coronary calcification as the study subjects. According to the occurrence of postoperative heart failure, the study subjects were divided into the heart failure group (n = 53) and the non-heart failure group (n = 250). Retrospective collection of clinical data from the study subjects during hospitalization. After processing the missing values in the original data and addressing sample imbalance using Adaptive Synthetic Sampling (ADASYN) method, the final dataset consists of 502 samples: 250 negative samples (i.e., patients not suffering from heart failure) and 252 positive samples (i.e., patients with heart failure). According to a 7:3 ratio, the datasets of 502 patients were randomly divided into a training set (n = 351) and a validation set (n = 151). On the training set, logistic regression (LR), extreme gradient boosting (XGBoost), support vector machine (SVM), and lightweight gradient boosting machine (LightGBM) algorithms were used to construct heart failure risk prediction models; Evaluate model performance on the validation set by calculating the area under the receiver operating characteristic curve (ROC) curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value, F1-score, and prediction accuracy.
Result
A total of 17.49% of 303 patients occured postoperative heart failure. The AUC of LR, XGBoost, SVM, and LightGBM models in the training set were 0.872, 1.000, 0.699, and 1.000, respectively. After 10 fold cross validation, the AUC was 0.863, 0.972, 0.696, and 0.963 in the training set, respectively. Among them, XGBoost had the highest AUC and better predictive performance, while SVM models had the worst performance. The XGBoost model also showed good predictive performance in the validation set (AUC = 0.972, 95% CI [0.951–0.994]). The Shapley additive explanation (SHAP) method suggested that the six characteristic variables of blood cholesterol, serum creatinine, fasting blood glucose, age, triglyceride and NT-proBNP were important positive factors for the occurrence of heart failure, and LVEF was important negative factors for the occurrence of heart failure.
Conclusion
The seven characteristic variables of blood cholesterol, blood creatinine, fasting blood glucose, NT-proBNP, age, triglyceride and LVEF are all important factors affecting the occurrence of heart failure. The prediction model of heart failure risk for elderly patients after CRA based on the XGBoost algorithm is superior to SVM, LightGBM and the traditional LR model. This model could be used to assist clinical decision-making and improve the adverse outcomes of patients after CRA.
Objective
To develop and validate a heart failure risk prediction model for elderly patients after coronary rotational atherectomy based on machine learning methods.
Methods
A retrospective cohort study was conducted to select 303 elderly patients with severe coronary calcification as the study subjects. According to the occurrence of postoperative heart failure, the study subjects were divided into the heart failure group (n = 53) and the non-heart failure group (n = 250). Retrospective collection of clinical data from the study subjects during hospitalization. After processing the missing values in the original data and addressing sample imbalance using Adaptive Synthetic Sampling (ADASYN) method, the final dataset consists of 502 samples: 250 negative samples (i.e., patients not suffering from heart failure) and 252 positive samples (i.e., patients with heart failure). According to a 7:3 ratio, the datasets of 502 patients were randomly divided into a training set (n = 351) and a validation set (n = 151). On the training set, logistic regression (LR), extreme gradient boosting (XGBoost), support vector machine (SVM), and lightweight gradient boosting machine (LightGBM) algorithms were used to construct heart failure risk prediction models; Evaluate model performance on the validation set by calculating the area under the receiver operating characteristic curve (ROC) curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value, F1-score, and prediction accuracy.
Result
A total of 17.49% of 303 patients occured postoperative heart failure. The AUC of LR, XGBoost, SVM, and LightGBM models in the training set were 0.872, 1.000, 0.699, and 1.000, respectively. After 10 fold cross validation, the AUC was 0.863, 0.972, 0.696, and 0.963 in the training set, respectively. Among them, XGBoost had the highest AUC and better predictive performance, while SVM models had the worst performance. The XGBoost model also showed good predictive performance in the validation set (AUC = 0.972, 95% CI [0.951–0.994]). The Shapley additive explanation (SHAP) method suggested that the six characteristic variables of blood cholesterol, serum creatinine, fasting blood glucose, age, triglyceride and NT-proBNP were important positive factors for the occurrence of heart failure, and LVEF was important negative factors for the occurrence of heart failure.
Conclusion
The seven characteristic variables of blood cholesterol, blood creatinine, fasting blood glucose, NT-proBNP, age, triglyceride and LVEF are all important factors affecting the occurrence of heart failure. The prediction model of heart failure risk for elderly patients after CRA based on the XGBoost algorithm is superior to SVM, LightGBM and the traditional LR model. This model could be used to assist clinical decision-making and improve the adverse outcomes of patients after CRA.A retrospective study investigating the clinical significance of body mass index in acute pancreatitishttps://peerj.com/articles/168542024-01-292024-01-29Yuanzhen BaiGuanwen GongReziya AierkenXingyu LiuWei ChengJunjie GuanZhiwei Jiang
Background
Acute pancreatitis is an unpredictable and potentially fatal condition for which no definitive cure is currently available. Our research focused on exploring the connection between body mass index, a frequently overlooked risk factor, and both the onset and progression of acute pancreatitis.
Material/Methods
A total of 247 patients with acute pancreatitis admitted to Jiangsu Provincial Hospital of Chinese Medicine from January 2021 to February 2023 were retrospectively reviewed. After screening, 117 patients with complete height and body weight data were selected for detailed assessment. Additionally, 85 individuals who underwent physical examinations at our hospital during this period were compiled to create a control group. The study received ethical approval from the ethics committee of Jiangsu Province Hospital of Chinese Medicine (Ref: No.2022NL-114-02) and was conducted in accordance with the China Good Clinical Practice in Research guidelines.
Results
A significant difference in body mass index (BMI) was observed between the healthy group and acute pancreatitis (AP) patients (p < 0.05), with a more pronounced disparity noted in cases of hyperlipidemic acute pancreatitis (p < 0.01). A potential risk for AP was identified at a BMI greater than 23.56 kg/m2 (AUC = 0.6086, p < 0.05). Being in the obese stage I (95%CI, [1.11–1.84]) or having a BMI below 25.4 kg/m2 (95%CI, [1.82–6.48]) are identified as risk factors for adverse AP progression. Moreover, BMI effectively predicts the onset of acute edematous pancreatitis and acute necrotizing pancreatitis (AUC = 0.7893, p < 0.001, cut-off value = 25.88 kg/m2). A higher BMI correlates with increased recurrence rates within a short timeframe (r = 0.7532, p < 0.01).
Conclusions
Elevated BMI is a risk factor for both the occurrence and progression of AP, and underweight status may similarly contribute to poor disease outcomes. BMI is crucial for risk prediction and stratification in AP and warrants ongoing monitoring and consideration.
Background
Acute pancreatitis is an unpredictable and potentially fatal condition for which no definitive cure is currently available. Our research focused on exploring the connection between body mass index, a frequently overlooked risk factor, and both the onset and progression of acute pancreatitis.
Material/Methods
A total of 247 patients with acute pancreatitis admitted to Jiangsu Provincial Hospital of Chinese Medicine from January 2021 to February 2023 were retrospectively reviewed. After screening, 117 patients with complete height and body weight data were selected for detailed assessment. Additionally, 85 individuals who underwent physical examinations at our hospital during this period were compiled to create a control group. The study received ethical approval from the ethics committee of Jiangsu Province Hospital of Chinese Medicine (Ref: No.2022NL-114-02) and was conducted in accordance with the China Good Clinical Practice in Research guidelines.
Results
A significant difference in body mass index (BMI) was observed between the healthy group and acute pancreatitis (AP) patients (p < 0.05), with a more pronounced disparity noted in cases of hyperlipidemic acute pancreatitis (p < 0.01). A potential risk for AP was identified at a BMI greater than 23.56 kg/m2 (AUC = 0.6086, p < 0.05). Being in the obese stage I (95%CI, [1.11–1.84]) or having a BMI below 25.4 kg/m2 (95%CI, [1.82–6.48]) are identified as risk factors for adverse AP progression. Moreover, BMI effectively predicts the onset of acute edematous pancreatitis and acute necrotizing pancreatitis (AUC = 0.7893, p < 0.001, cut-off value = 25.88 kg/m2). A higher BMI correlates with increased recurrence rates within a short timeframe (r = 0.7532, p < 0.01).
Conclusions
Elevated BMI is a risk factor for both the occurrence and progression of AP, and underweight status may similarly contribute to poor disease outcomes. BMI is crucial for risk prediction and stratification in AP and warrants ongoing monitoring and consideration.Single-cell data revealed CD14-type and FCGR3A-type macrophages and relevant prognostic factors for predicting immunotherapy and prognosis in stomach adenocarcinomahttps://peerj.com/articles/167762024-01-222024-01-22Mengling LiMing LuJun LiQingqing GuiYibin XiaChao LuHongchun Shu
Background
Stomach adenocarcinoma (STAD) exhibits profound tumor heterogeneity and represents a great therapeutic challenge. Single-cell sequencing technology is a powerful tool to identify characteristic cell types.
Methods
Single-cell sequencing data (scRNA-seq) GSE167297 and bulk RNA-seq data from TCGA, GTEx, GSE26901 and GSE15459 database were included in this study. By downscaling and annotating the cellular data in scRNA-seq, critical cell types in tumor progression were identified by AUCell score. Relevant gene modules were then identified by weighted gene co-expression network analysis (WGCNA). A prognostic scoring system was constructed by identifying prognostic factors in STAD by Least absolute shrinkage and selection operator (LASSO) COX model. The prognosis and model performance in the RiskScore groups were measured by Kaplan-Meier (K-M) curves and Receiver operating characteristic (ROC) curves. Nomogram was drawn based on RiskScore and prognosis-related clinical factors. In addition, we evaluated patient’s feedback on immunotherapy in the RiskScore groups by TIMER, ESTIMATE and TIDE analysis. Finally, the expression levels of prognostic factors were verified in gastric cancer cell lines (MKN7 and MKN28) and human normal gastric mucosal epithelial cells (GES-1), and the effects of prognostic factors on the viability of gastric cancer cells were examined by the CCK8 assay and cell cycle.
Results
scRNA-seq analysis revealed that 11 cell types were identified, and macrophages exhibited relatively higher AUCell scores and specifically expressed CD14 and FCGR3A. High macrophage scores worsened the prognosis of STAD patients. We intersected the specifically expressed genes in macrophages subgroups (670) and macrophage module genes (2,360) obtained from WGCNA analysis. Among 86 common genes, seven prognostic factors (RGS2, GNAI2, ANXA5, MARCKS, CD36, NRP1 and PDE4A) were identified and composed a RiskScore model. Patients in low Risk group showed a better survival advantage. Nomogram also provided a favorable prediction for survival at 1, 3 and 5 years in STAD patients. Besides, we found positive feedback to immunotherapy in patients with low RiskScore. The expression tendency of the seven prognostic factors in MKN7 and MKN28 was consistent with that in the RNA-seq data in addition to comparison of protein expression levels in the public HPA (The Human Protein Atlas) database. Further functional exploration disclosed that MARCKS was an important prognostic factor in regulating cell viability in STAD.
Conclusion
This study preliminary uncovered a single cell atlas for STAD patients, and Macrophages relevant gene signature and nomogram displayed favorable immunotherapy and prognostic prediction ability. Collectively, our work provides a new insight into the molecular mechanisms and therapeutic approach for LUAD patients.
Background
Stomach adenocarcinoma (STAD) exhibits profound tumor heterogeneity and represents a great therapeutic challenge. Single-cell sequencing technology is a powerful tool to identify characteristic cell types.
Methods
Single-cell sequencing data (scRNA-seq) GSE167297 and bulk RNA-seq data from TCGA, GTEx, GSE26901 and GSE15459 database were included in this study. By downscaling and annotating the cellular data in scRNA-seq, critical cell types in tumor progression were identified by AUCell score. Relevant gene modules were then identified by weighted gene co-expression network analysis (WGCNA). A prognostic scoring system was constructed by identifying prognostic factors in STAD by Least absolute shrinkage and selection operator (LASSO) COX model. The prognosis and model performance in the RiskScore groups were measured by Kaplan-Meier (K-M) curves and Receiver operating characteristic (ROC) curves. Nomogram was drawn based on RiskScore and prognosis-related clinical factors. In addition, we evaluated patient’s feedback on immunotherapy in the RiskScore groups by TIMER, ESTIMATE and TIDE analysis. Finally, the expression levels of prognostic factors were verified in gastric cancer cell lines (MKN7 and MKN28) and human normal gastric mucosal epithelial cells (GES-1), and the effects of prognostic factors on the viability of gastric cancer cells were examined by the CCK8 assay and cell cycle.
Results
scRNA-seq analysis revealed that 11 cell types were identified, and macrophages exhibited relatively higher AUCell scores and specifically expressed CD14 and FCGR3A. High macrophage scores worsened the prognosis of STAD patients. We intersected the specifically expressed genes in macrophages subgroups (670) and macrophage module genes (2,360) obtained from WGCNA analysis. Among 86 common genes, seven prognostic factors (RGS2, GNAI2, ANXA5, MARCKS, CD36, NRP1 and PDE4A) were identified and composed a RiskScore model. Patients in low Risk group showed a better survival advantage. Nomogram also provided a favorable prediction for survival at 1, 3 and 5 years in STAD patients. Besides, we found positive feedback to immunotherapy in patients with low RiskScore. The expression tendency of the seven prognostic factors in MKN7 and MKN28 was consistent with that in the RNA-seq data in addition to comparison of protein expression levels in the public HPA (The Human Protein Atlas) database. Further functional exploration disclosed that MARCKS was an important prognostic factor in regulating cell viability in STAD.
Conclusion
This study preliminary uncovered a single cell atlas for STAD patients, and Macrophages relevant gene signature and nomogram displayed favorable immunotherapy and prognostic prediction ability. Collectively, our work provides a new insight into the molecular mechanisms and therapeutic approach for LUAD patients.A machine learning-based model for clinical prediction of distal metastasis in chondrosarcoma: a multicenter, retrospective studyhttps://peerj.com/articles/164852023-12-182023-12-18Jihu WeiShijin LuWencai LiuHe LiuLin FengYizi TaoZhanglin PuQiang LiuZhaohui HuHaosheng WangWenle LiWei KangChengliang YinZhe Feng
Background
The occurrence of distant metastases (DM) limits the overall survival (OS) of patients with chondrosarcoma (CS). Early diagnosis and treatment of CS remains a great challenge in clinical practice. The aim of this study was to investigate metastatic factors and develop a risk stratification model for clinicians’ decision-making.
Methods
Six machine learning (ML) algorithms, including logistic regression (LR), plain Bayesian classifier (NBC), decision tree (DT), random forest (RF), gradient boosting machine (GBM) and extreme gradient boosting (XGBoost). A 10-fold cross-validation was performed for each model separately, multicenter data was used as external validation, and the best (highest AUC) model was selected to build the network calculator.
Results
A total of 1,385 patients met the inclusion criteria, including 82 (5.9%) patients with metastatic CS. Multivariate logistic regression analysis showed that the risk of DM was significantly higher in patients with higher pathologic grades, T-stage, N-stage, and non-left primary lesions, as well as those who did not receive surgery and chemotherapy. The AUC of the six ML algorithms for predicting DM ranged from 0.911–0.985, with the extreme gradient enhancement algorithm (XGBoost) having the highest AUC. Therefore, we used the XGB model and uploaded the results to an online risk calculator for estimating DM risk.
Conclusions
In this study, combined with adequate SEER case database and external validation with data from multicenter institutions in different geographic regions, we confirmed that CS, T, N, laterality, and grading of surgery and chemotherapy were independent risk factors for DM. Based on the easily available clinical risk factors, machine learning algorithms built the XGB model that predicts the best outcome for DM. An online risk calculator helps simplify the patient assessment process and provides decision guidance for precision medicine and long-term cancer surveillance, which contributes to the interpretability of the model.
Background
The occurrence of distant metastases (DM) limits the overall survival (OS) of patients with chondrosarcoma (CS). Early diagnosis and treatment of CS remains a great challenge in clinical practice. The aim of this study was to investigate metastatic factors and develop a risk stratification model for clinicians’ decision-making.
Methods
Six machine learning (ML) algorithms, including logistic regression (LR), plain Bayesian classifier (NBC), decision tree (DT), random forest (RF), gradient boosting machine (GBM) and extreme gradient boosting (XGBoost). A 10-fold cross-validation was performed for each model separately, multicenter data was used as external validation, and the best (highest AUC) model was selected to build the network calculator.
Results
A total of 1,385 patients met the inclusion criteria, including 82 (5.9%) patients with metastatic CS. Multivariate logistic regression analysis showed that the risk of DM was significantly higher in patients with higher pathologic grades, T-stage, N-stage, and non-left primary lesions, as well as those who did not receive surgery and chemotherapy. The AUC of the six ML algorithms for predicting DM ranged from 0.911–0.985, with the extreme gradient enhancement algorithm (XGBoost) having the highest AUC. Therefore, we used the XGB model and uploaded the results to an online risk calculator for estimating DM risk.
Conclusions
In this study, combined with adequate SEER case database and external validation with data from multicenter institutions in different geographic regions, we confirmed that CS, T, N, laterality, and grading of surgery and chemotherapy were independent risk factors for DM. Based on the easily available clinical risk factors, machine learning algorithms built the XGB model that predicts the best outcome for DM. An online risk calculator helps simplify the patient assessment process and provides decision guidance for precision medicine and long-term cancer surveillance, which contributes to the interpretability of the model.