ABSTRACT
Background: Myocardial injury after noncardiac surgery (MINS) is an easily overlooked complication but closely related to postoperative cardiovascular adverse outcomes; therefore, the early diagnosis and prediction are particularly important. Objective: We aimed to develop and validate an explainable machine learning (ML) model for predicting MINS among older patients undergoing noncardiac surgery. Methods: The retrospective cohort study included older patients who had noncardiac surgery from 1 northern center and 1 southern center in China. The data sets from center 1 were divided into a training set and an internal validation set. The data set from center 2 was used as an external validation set. Before modeling, the least absolute shrinkage and selection operator and recursive feature elimination methods were used to reduce dimensions of data and select key features from all variables. Prediction models were developed based on the extracted features using several ML algorithms, including category boosting, random forest, logistic regression, naïve Bayes, light gradient boosting machine, extreme gradient boosting, support vector machine, and decision tree. Prediction performance was assessed by the area under the receiver operating characteristic (AUROC) curve as the main evaluation metric to select the best algorithms. The model performance was verified by internal and external validation data sets with the best algorithm and compared to the Revised Cardiac Risk Index. The Shapley Additive Explanations (SHAP) method was applied to calculate values for each feature, representing the contribution to the predicted risk of complication, and generate personalized explanations. Results: A total of 19,463 eligible patients were included; among those, 12,464 patients in center 1 were included as the training set; 4754 patients in center 1 were included as the internal validation set; and 2245 in center 2 were included as the external validation set. The best-performing model for prediction was the CatBoost algorithm, achieving the highest AUROC of 0.805 (95% CI 0.778-0.831) in the training set, validating with an AUROC of 0.780 in the internal validation set and 0.70 in external validation set. Additionally, CatBoost demonstrated superior performance compared to the Revised Cardiac Risk Index (AUROC 0.636; P<.001). The SHAP values indicated the ranking of the level of importance of each variable, with preoperative serum creatinine concentration, red blood cell distribution width, and age accounting for the top three. The results from the SHAP method can predict events with positive values or nonevents with negative values, providing an explicit explanation of individualized risk predictions. Conclusions: The ML models can provide a personalized and fairly accurate risk prediction of MINS, and the explainable perspective can help identify potentially modifiable sources of risk at the patient level.
Subject(s)
Machine Learning , Postoperative Complications , Humans , Retrospective Studies , Female , China/epidemiology , Aged , Male , Postoperative Complications/epidemiology , Postoperative Complications/etiology , Postoperative Complications/diagnosis , Middle Aged , Risk Assessment/methods , Surgical Procedures, Operative/adverse effectsABSTRACT
BACKGROUND: The triglyceride-glucose index, a reliable surrogate biomarker of insulin resistance, has been reported to be associated with cardiovascular events and atherosclerosis. However, few studies have investigated the association of the triglyceride-glucose index with postoperative infections. This study aimed to study the clinical risk values of the preoperative triglyceride-glucose index in postoperative infection complications in elderly patients undergoing gastrointestinal-related abdominal and pelvic surgery. METHODS: This retrospective cohort study included 3,225 older patients who underwent gastrointestinal-related abdominal and pelvic surgery between 2014 and 2019. The patients were divided into groups of triglyceride-glucose index ≤8.268 and triglyceride-glucose index >8.268 according to the optimal triglyceride-glucose index cut-off value. The outcome of interest was postoperative infections within 30 days after surgery. Primary and subgroup analyses were performed to confirm that preoperative triglyceride-glucose index qualifies as a reliable, independent risk indicator. Propensity score matching analysis was further applied to address covariates' potential residual confounding effect and test the robustness of the results. RESULTS: In this study, the median age was 71 years (interquartile range, 67, 75 years), the proportion of male patients was 66.3%, and 1,058 (32.8%) were infected within 30 days after surgery. A triglyceride-glucose index >8.268 was associated with an increased risk of postoperative infections in multivariate regression analysis (odds ratio, 1.37; 95% confidence interval, 1.15-1.64; P < .001). The correlation between the triglyceride-glucose index and postoperative infections remained significantly robust (odds ratio, 1.52; 95% confidence interval, 1.21-1.92; P < .001) in the propensity score matching analysis. CONCLUSIONS: The triglyceride-glucose index elevation determined by the optimal cutoff value of 8.268 was an independent risk factor for developing postoperative infections.
ABSTRACT
BACKGROUND: Postoperative delirium is a common complication in older patients, with poor long-term outcomes. This study aimed to investigate risk factors and develop a predictive model for postoperative delirium in older patients after major abdominal surgery. METHODS: This study retrospectively recruited 7577 patients aged ≥ 65 years who underwent major abdominal surgery between January 2014 and December 2018 in a single hospital in Beijing, China. Patients were divided into a training cohort (n = 5303) and a validation cohort (n = 2224) for univariate and multivariate logistic regression analyses and to build a nomogram. Data were collected for 43 perioperative variables, including demographics, medical history, preoperative laboratory results, imaging, and anesthesia information. RESULTS: Age, chronic obstructive pulmonary disease, white blood cell count, glucose, total protein, creatinine, emergency surgery, and anesthesia time were associated with postoperative delirium in multivariate analysis. We developed a nomogram based on the above 8 variables. The nomogram achieved areas under the curve of 0.731 and 0.735 for the training and validation cohorts, respectively. The discriminatory ability of the nomogram was further assessed by dividing the cases into three risk groups (low-risk, nomogram score < 175; medium-risk, nomogram score 175~199; high-risk, nomogram score > 199; P < 0.001). Decision curve analysis revealed that the nomogram provided a good net clinical benefit. CONCLUSIONS: We developed a nomogram that could predict postoperative delirium with high accuracy and stability in older patients after major abdominal surgery.
ABSTRACT
BACKGROUND: Myocardial injury after non-cardiac surgery (MINS) is a common and insidious postoperative complication. This study aimed to evaluate the relationship between the triglyceride-glucose index (TyG) and MINS in advanced-age patients. METHODS: We performed a single-center retrospective study including patients ≥ 65 years of age who underwent non-cardiac surgery. The relationship between TyG and MINS was investigated using univariate and multivariate logistic regression analyses. Multivariate logistic regression analysis involved three models: Model I adjusted for preoperative factors, Model II adjusted for surgery-related factors, and Model III adjusted for both preoperative and surgery-related factors. Propensity score matching (PSM) was used to reduce the confounding effects of covariates. Subgroup analyses were then performed to evaluate the relationship between TyG and MINS in various subsamples. RESULTS: A total of 7789 patients were studied, among whom 481 (6.2%) developed MINS. A cut-off value of TyG of 8.57 was determined using a receiver operating characteristic (ROC) curve to be associated with the best predictive performance. Participants with TyG ≥ 8.57 were at a higher risk of developing MINS than those with TyG < 8.57 [n = 273 (7.6%) vs. n = 208 (4.9%), respectively; p < 0.001]. The univariate analysis showed that TyG ≥ 8.57 was significantly associated with MINS in elderly patients [odds ratio (OR): 1.58; 95% confidence interval (95%CI): 1.32-1.91; p < 0.001)]. In multivariate logistic regression, adjustments were made for risk factors including age, sex, body mass index (BMI), hypertension, coronary heart disease, and duration of surgery, etc. The adjusted ORs for TyG ≥ 8.57 were 1.46 (95%CI: 1.17-1.82), p = 0.001; 1.46 (95%CI: 1.19-1.77), p < 0.001; and 1.43 (95%CI: 1.13-1.81), p = 0.003, in the three multivariate models, respectively. The relationship remained after PSM (adjusted OR: 1.35, 95% CI: 1.03-1.78, p = 0.029). Furthermore, the relationship between TyG and MINS remained in a number of subgroups in the sensitivity analyses, but not in participants with peripheral vascular stenosis. CONCLUSIONS: A preoperative high TyG (≥ 8.57) is associated with a higher risk of MINS in advanced-age patients undergoing non-cardiac surgery.
ABSTRACT
Sepsis is a systemic syndrome involving physiological, pathological, and biochemical abnormalities precipitated by infection and is a major global public health problem. Endothelial cells (ECs) dysfunction is a major contributor to sepsis-induced multiple organ failure. This bibliometric analysis aimed to identify and characterize the status, evolution of the field, and new research trends of ECs and sepsis over the past 20 years. For this analysis, the Web of Science Core Collection database was searched to identify relevant publications on ECs in sepsis published between January 1, 2002, and December 31, 2022. Microsoft Excel 2021, VOSviewer software, CiteSpace software, and the online analysis platform of literature metrology (http://bibliometric.com) were used to visualize the trends of publications' countries/regions, institutions, authors, journals, and keywords. In total, 4200 articles were identified and screened, primarily originating from 86 countries/regions and 3489 institutions. The USA was the leading contributor to this research field, providing 1501 articles (35.74 %). Harvard University's scientists were the most prolific, with 129 articles. Overall, 21,944 authors were identified, among whom Bae Jong Sup was the most prolific, contributing 129 publications. Additionally, Levi Marcel was the most frequently co-cited author, appearing 538 times. The journals that published the most articles were SHOCK, CRITICAL CARE MEDICINE, and PLOS ONE, accounting for 10.79 % of the total. The current emerging hotspots are concentrated on "endothelial glycocalyx," "NLRP3 inflammasome," "extracellular vesicle," "biomarkers," and "COVID-19," among others. In conclusion, this study provides a comprehensive overview of the scientific productivity and emerging research trends in the field of ECs in sepsis. The evidence supporting the significant role of ECs in both physiological and pathological responses to sepsis is continuously growing. More in-depth studies of the molecular mechanisms underlying sepsis-induced endothelial dysfunction and EC-targeted therapies are warranted in the future.
ABSTRACT
BACKGROUND: Sleep disturbances in the peri-operative period have been associated with adverse outcomes, including postoperative delirium (POD). However, research on sleep quality during the immediate postoperative period is limited. OBJECTIVES: This study aimed to investigate the association between sleep quality on the night of the operative day assessed using the Sleep Quality Numeric Rating Scale (SQ-NRS), and the incidence of POD in a large cohort of surgical patients. DESIGN: A prospective cohort study. SETTING: A tertiary hospital in China. PATIENTS: This study enrolled patients aged 65âyears or older undergoing elective surgery under general anaesthesia. The participants were categorised into the sleep disturbance and no sleep disturbance groups according to their operative night SQ-NRS. MAIN OUTCOME MEASURES: The primary outcome was delirium incidence, whereas the secondary outcomes included acute kidney injury, stroke, pulmonary infection, cardiovascular complications and all-cause mortality within 1 year postoperatively. RESULTS: In total, 3072 patients were included in the analysis of this study. Among them, 791 (25.72%) experienced sleep disturbances on the night of operative day. Patients in the sleep disturbance group had a significantly higher risk of developing POD (adjusted OR 1.43, 95% CI 1.11 to 1.82, P â=â0.005). Subgroup analysis revealed that age 65-75âyears; male sex; ASA III and IV; haemoglobin more than 12âgâl -1 ; intra-operative hypotension; surgical duration more than 120âmin; and education 9âyears or less were significantly associated with POD. No interaction was observed between the subgroups. No significant differences were observed in the secondary outcomes, such as acute kidney injury, stroke, pulmonary infection, cardiovascular complications and all-cause mortality within 1 year postoperatively. CONCLUSIONS: The poor subjective sleep quality on the night of operative day was independently associated with increased POD risk, especially in certain subpopulations. Optimising peri-operative sleep may reduce POD. Further research should investigate potential mechanisms and causal relationships. TRIAL REGISTRY: chictr.org.cn: ChiCTR1900028545.
Subject(s)
Acute Kidney Injury , Cardiovascular Infections , Delirium , Emergence Delirium , Stroke , Aged , Humans , Male , Cardiovascular Infections/complications , Delirium/diagnosis , Delirium/epidemiology , Delirium/etiology , Emergence Delirium/diagnosis , Emergence Delirium/epidemiology , Emergence Delirium/etiology , Postoperative Complications/diagnosis , Postoperative Complications/epidemiology , Postoperative Complications/etiology , Prospective Studies , Risk Factors , Sleep Quality , FemaleABSTRACT
BACKGROUND: Identifying the risk factors associated with perioperative mortality is crucial, particularly in older patients. Predicting 6-month mortality risk in older patients based on large datasets can assist patients and surgeons in perioperative clinical decision-making. This study aimed to develop a risk prediction model of mortality within 6 months after noncardiac surgery using the clinical data from 11 894 older patients in China. MATERIALS AND METHODS: A multicentre, retrospective cohort study was conducted in 20 tertiary hospitals. The authors retrospectively included 11 894 patients (aged ≥65 years) who underwent noncardiac surgery between April 2020 and April 2022. The least absolute shrinkage and selection operator model based on linear regression was used to analyse and select risk factors, and various machine learning methods were used to build predictive models of 6-month mortality. RESULTS: The authors predicted 12 preoperative risk factors associated with 6-month mortality in older patients after noncardiac surgery. Including laboratory-associated risk factors such as mononuclear cell ratio and total blood cholesterol level, etc. Also including medical history associated risk factors such as stroke, history of chronic diseases, etc. By using a random forest model, the authors constructed a predictive model with a satisfactory accuracy (area under the receiver operating characteristic curve=0.97). CONCLUSION: The authors identified 12 preoperative risk factors associated with 6-month mortality in noncardiac surgery older patients. These preoperative risk factors may provide evidence for a comprehensive preoperative anaesthesia assessment as well as necessary information for clinical decision-making by anaesthesiologists.
Subject(s)
Stroke , Humans , Aged , Retrospective Studies , Risk Assessment/methods , Risk Factors , Clinical Decision-MakingABSTRACT
OBJECTIVE: Prognostic nutritional index (PNI) is an indicator to evaluate the nutritional immune status of patients. This study aimed to assess whether preoperative PNI could predict the occurrence of postoperative POD in aged patients undergoing non-neurosurgery and non-cardiac surgery. METHOD: The aged patients undergoing non-neurosurgery and non-cardiac surgery between January 2014 and August 2019 were included in the retrospective cohort study. The correlation between POD and PNI was investigated by univariate and multivariable logistic regression analysis, propensity score matching (PSM), inverse probability of treatment weighting (IPTW), and subgroup analysis. RESULTS: In the cohort (n = 29,814), the cutoff value of PNI was 46.01 determined by the receiver operating characteristic (ROC) curve. In univariate and three multivariable regression analysis, the ORs of PNI ≤ 46.01 was 2.573(95% CI:2.261-2.929, P < 0.001),1.802 (95% CI:1.567-2.071, P < 0.001),1.463(95% CI:1.246-1.718, P < 0.001),1.370(95% CI:1.165-1.611, P < 0.001). In the PSM model and IPTW model, the ORs of PNI ≤ 46.01 were 1.424(95% CI:1.172-1.734, P < 0.001) and 1.356(95% CI:1.223-1.505, P < 0.001). CONCLUSION: The PNI was found to have a predictive value for POD in patients undergoing non-neurosurgery and non-cardiac surgery. Improving preoperative nutritional status may be beneficial in preventing POD for aged patients.
Subject(s)
Emergence Delirium , Nutrition Assessment , Humans , Aged , Retrospective Studies , Prognosis , Cohort Studies , Nutritional StatusABSTRACT
BACKGROUND: Postoperative delirium (POD) is a common complication in elderly patients after hip fracture surgery. Our study was to investigate whether intraoperative mean arterial pressure variability (MAPV) was associated with POD in elderly patients after hip fracture surgery. METHODS: In this retrospective cohort study, patients aged 65 years and older undergoing hip fracture surgery were included. The correlation between MAPV and POD was investigated using univariate and multivariate logistic regression. Covariate-related confounding effects were eliminated with propensity score matching (PSM) analysis. Then, a subgroup analysis was conducted to further examine the associations between MAPV and POD. RESULTS: Nine hundred sixty-three patients with a median age of 80 years (IQR: 73-84) were enrolled. POD occurred in 115/963 (11.9%) patients within 7 days after surgery. According to multivariate regression analysis, MAPV > 2.17 was associated with an increased risk of POD (OR: 2.379, 95% CI: 1.496-3.771, P < 0.001). All covariates between the two groups were well balanced after PSM adjustment. A significant correlation between MAPV and POD was found in the PSM analysis (OR: 2.851, 95% CI: 1.710-4.746, P < 0.001). CONCLUSIONS: An increased intraoperative MAPV may be a predictor for POD.
Subject(s)
Emergence Delirium , Hip Fractures , Aged , Humans , Aged, 80 and over , Arterial Pressure , Retrospective Studies , Hip Fractures/surgery , Multivariate Analysis , Postoperative Complications/diagnosis , Postoperative Complications/epidemiology , Postoperative Complications/etiology , Risk FactorsABSTRACT
BACKGROUND: The COVID-19 pandemic has a heavy impact on the mental health of elderly surgical patients worldwide. In particular, the elderly patients faced considerable psychological stress due to various environmental and medical factors during the outbreak. This study aims to examine changes in mental health trends among non-cardiac surgical patients aged 65 and above in China during the COVID-19 pandemic. METHODS: This multi-center, convenient sampling, longitudinal observational study was conducted from April 1, 2020 to April 30, 2022. Primary outcome was the prevalence of postoperative depression. Secondary outcome was the prevalence of postoperative anxiety. Follow-up was conducted separately at 7 days and 30 days after surgery. Depression symptoms were assessed using the Patient Health Questionnaire 9 (PHQ-9) scale. Anxiety symptoms were assessed using Generalized Anxiety Disorder-7 (GAD-7) scale, with scores of ≥5 defining positive depression or anxiety symptoms. Multivariate logistic regression analysis was used to investigate risk factors of mental health status in more elderly patients undergoing non-cardiac surgery. RESULTS: A total of 4639 patients were included, of whom 2279 (46.0 %) were male, 752 (15.2 %) were over the age of 75, and 4346 (93.7 %) were married. The monthly prevalence trends demonstrated that compared to the outbreak period, a significant reduction in the prevalence of depression and anxiety symptoms in elderly patients who underwent surgery during the post-pandemic period. In post-pandemic period, a statistically significant decrease in the prevalence of all severity depression and anxiety patients was noted at the 7-day follow-up, but no significant decrease was observed for severe depression and anxiety in the 30-day follow-up. In COVID-19 low-risk area, a significant overall decrease in prevalence of mental health was observed during the post-pandemic period compared to the outbreak period, including 7-day depression, 7-day anxiety, 30-day depression, and 30-day anxiety (all with P < 0.001). Female and patients with ≥2 comorbidities appeared to be more susceptible to postoperative depression and anxiety during the pandemic. LIMITATION: The absence of data from the early days of the COVID-19 outbreak. CONCLUSIONS: This study analyzed the prevalence of depression and anxiety in elderly non-cardiac patients during and after the COVID-19 pandemic, focusing on dimensions such as severity, risk-areas, gender, and comorbidity. Our findings revealed a significant decrease in the prevalence of depression and anxiety in elderly surgery patients during the post-pandemic period.
ABSTRACT
BACKGROUND: Hip fractures are common in elderly patients, and almost all the patients undergo surgery. This study aimed to develop a novel modified lymphocyte C-reactive protein (CRP) score (mLCS) to simply and conveniently predict 3-year mortality in elderly patients undergoing intertrochanteric fracture surgery. METHODS: A retrospective study was conducted on elderly patients who underwent intertrochanteric fracture surgery between January 2014 and December 2017. The mLCS was developed according to the value of CRP and lymphocyte counts. Univariate and multivariate Cox regression analyses were used to identify independent risk factors for 3-year mortality after surgery. The performances of the lymphocyte CRP score (LCS) and mLCS to predict 3-year mortality were then compared using C-statistics, decision curve analysis (DCA), net reclassification index (NRI) and integrated discrimination improvement (IDI). RESULTS: A total of 291 patients were enrolled, of whom 52 (17.9%) died within 3 years after surgery. In the multivariate Cox regression analysis, mLCS (hazard ratio (HR), 5.415; 95% confidence interval (CI), 1.743-16.822; P = 0.003) was significantly associated with postoperative 3-year mortality. The C-statistics of LCS and mLCS for predicting 3-year mortality were 0.644 and 0.686, respectively. The NRI (mLCS vs. LCS, 0.018) and IDI (mLCS vs. LCS, 0.017) indicated that the mLCS performed better than the LCS. DCA also showed that mLCS had a higher clinical net benefit. CONCLUSIONS: mLCS is a promising predictor that can simply and conveniently predict 3-year mortality in elderly patients undergoing intertrochanteric fracture surgery.
Subject(s)
C-Reactive Protein , Hip Fractures , Aged , Humans , Death , Hip Fractures/diagnosis , Hip Fractures/mortality , Hip Fractures/surgery , Lymphocytes , Retrospective StudiesABSTRACT
BACKGROUND AND AIMS: Current existing predictive tools have limitations in predicting major adverse cardiovascular events (MACEs) in elderly patients. We will build a new prediction model to predict MACEs in elderly patients undergoing noncardiac surgery by using traditional statistical methods and machine learning algorithms. METHODS: MACEs were defined as acute myocardial infarction (AMI), ischemic stroke, heart failure and death within 30 days after surgery. Clinical data from 45,102 elderly patients (≥65 years old), who underwent noncardiac surgery from two independent cohorts, were used to develop and validate the prediction models. A traditional logistic regression and five machine learning models (decision tree, random forest, LGBM, AdaBoost, and XGBoost) were compared by the area under the receiver operating characteristic curve (AUC). In the traditional prediction model, the calibration was assessed using the calibration curve and the patients' net benefit was measured by decision curve analysis (DCA). RESULTS: Among 45,102 elderly patients, 346 (0.76%) developed MACEs. The AUC of this traditional model was 0.800 (95% CI, 0.708-0.831) in the internal validation set, and 0.768 (95% CI, 0.702-0.835) in the external validation set. In the best machine learning prediction model-AdaBoost model, the AUC in the internal and external validation set was 0.778 and 0.732, respectively. Besides, for the traditional prediction model, the calibration curve of model performance accurately predicted the risk of MACEs (Hosmer and Lemeshow, p = 0.573), the DCA results showed that the nomogram had a high net benefit for predicting postoperative MACEs. CONCLUSIONS: This prediction model based on the traditional method could accurately predict the risk of MACEs after noncardiac surgery in elderly patients.
Subject(s)
Cardiovascular System , Heart Failure , Ischemic Stroke , Myocardial Infarction , Aged , Humans , Retrospective Studies , Myocardial Infarction/diagnosis , Myocardial Infarction/epidemiologyABSTRACT
Among the elderly, depression is one of the most common mental disorders, which seriously affects their physical and mental health and quality of life, and their suicide rate is particularly high. Depression in the elderly is strongly associated with surgery. In this study, we aimed to explore the risk factors and establish a predictive model of depressive symptoms 1 month after video-assisted thoracoscopic surgery (VATS) in elderly patients. The study participants included 272 elderly patients (age > 65 years) undergoing VATS from April 2020 to May 2021 at 1 of 18 medical centers in China. The patients were divided into a depression group and a nondepression group according to the Chinese version of the nine-item Patient Health Questionnaire (PHQ-9). The patients' pre- and postoperative characteristics and questionnaires were collected and compared. Then, binary logistic regression was used to determine the risk factors that affect postoperative depressive symptoms, and the predictive model was constructed. The prediction efficiency of the model was evaluated by drawing the receiver operating characteristic curve (ROC), and the area under the curve (AUC) was calculated to evaluate the value of the predictive model. Among all of the included patients, 16.54% (45/272) suffered from depressive symptoms after VATS. The results of the univariate analysis showed that body mass index (BMI), chronic pain, leukocyte count, fibrinogen levels, prothrombin time, ASA physical status, infusion volume, anxiety, sleep quality, and postoperative pain were related to postoperative depressive symptoms (all p < 0.05). The results of multivariate logistic regression analysis showed that a high fibrinogen level (OR = 2.42), postoperative anxiety (OR = 12.05), poor sleep quality (OR = 0.61), and pain (OR = 2.85) were risk factors of postoperative depressive symptoms. A predictive model was constructed according to the regression coefficient of each variable, the ROC curve was drawn, and the AUC value was calculated to be 0.889. The prediction model may help medical personnel identify older patients at risk of developing depressive disorders associated with VATS and may be useful for clinical purposes.
ABSTRACT
STUDY OBJECTIVE: To determine the association between intraoperative renal tissue desaturation as measured using near-infrared spectroscopy and increased likelihood of developing postoperative acute kidney injury (AKI) in older patients undergoing hepatectomy. DESIGN: A multicenter prospective cohort study. SETTING: The study was conducted at two tertiary hospitals in China from September 2020 to October 2021. PATIENTS: 157 older patients (≥ 60 years) undergoing open hepatectomy surgery. INTERVENTIONS AND MEASUREMENTS: Renal tissue oxygen saturation was continuously monitored during operation using near-infrared spectroscopy. The exposure of interest was intraoperative renal desaturation, defined as at least 20% relative decline in renal tissue oxygen saturation from baseline. The primary outcome was postoperative AKI, defined using the Kidney Disease: Improving Global Outcomes criteria according to the serum creatinine criteria. MAIN RESULTS: Renal desaturation occurred in 70 of 157 patients. Postoperative AKI was observed in 23% (16/70) and 8% (7/87) of patients with versus without renal desaturation. Patients with renal desaturation were at higher risk of AKI than patients without renal desaturation (adjusted odds ratio 3.41, 95% confidence interval: 1.12-10.36, p = 0.031). Predictive performance was 65.2% sensitivity and 33.6% specificity for hypotension alone, 69.6% sensitivity and 59.7% specificity for renal desaturation alone, and 95.7% sensitivity and 26.9% specificity for combined use of hypotension and renal desaturation. CONCLUSIONS: Intraoperative renal desaturation occurred in >40% in our sample of older patients undergoing liver resection and was associated with increased risk of AKI. Intraoperative near-infrared spectroscopy monitoring enhances the detection of AKI.
Subject(s)
Acute Kidney Injury , Hypotension , Humans , Aged , Prospective Studies , Hepatectomy/adverse effects , Postoperative Complications/diagnosis , Postoperative Complications/epidemiology , Postoperative Complications/etiology , Acute Kidney Injury/diagnosis , Acute Kidney Injury/epidemiology , Acute Kidney Injury/etiology , Hypotension/complications , Liver , Retrospective Studies , Risk FactorsABSTRACT
AIMS: To compare the performance of logistic regression and machine learning methods in predicting postoperative delirium (POD) in elderly patients. METHOD: This was a retrospective study of perioperative medical data from patients undergoing non-cardiac and non-neurology surgery over 65 years old from January 2014 to August 2019. Forty-six perioperative variables were used to predict POD. A traditional logistic regression and five machine learning models (Random Forest, GBM, AdaBoost, XGBoost, and a stacking ensemble model) were compared by the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, and precision. RESULTS: In total, 29,756 patients were enrolled, and the incidence of POD was 3.22% after variable screening. AUCs were 0.783 (0.765-0.8) for the logistic regression method, 0.78 for random forest, 0.76 for GBM, 0.74 for AdaBoost, 0.73 for XGBoost, and 0.77 for the stacking ensemble model. The respective sensitivities for the 6 aforementioned models were 74.2%, 72.2%, 76.8%, 63.6%, 71.6%, and 67.4%. The respective specificities for the 6 aforementioned models were 70.7%, 99.8%, 96.5%, 98.8%, 96.5%, and 96.1%. The respective precision values for the 6 aforementioned models were 7.8%, 52.3%, 55.6%, 57%, 54.5%, and 56.4%. CONCLUSIONS: The optimal application of the logistic regression model could provide quick and convenient POD risk identification to help improve the perioperative management of surgical patients because of its better sensitivity, fewer variables, and easier interpretability than the machine learning model.
Subject(s)
Emergence Delirium , Humans , Aged , Retrospective Studies , Logistic Models , ROC Curve , Machine LearningABSTRACT
Background: Diabetes mellitus (DM) has been reported to be associated with perioperative stroke, but the effects of preoperative hyperglycemia on the risk of perioperative stroke in diabetic patients undergoing non-cardiovascular surgery remain unclear. This study investigated the association between preoperative hyperglycemia and the risk of perioperative ischemic stroke in type 2 diabetic patients undergoing non-cardiovascular surgery. Methods: This retrospective cohort study screened 27,002 patients with type 2 DM undergoing non-cardiovascular surgery with general anesthesia between January 2008 and August 2019 at The First Medical Center of Chinese People's Liberation Army (PLA) General Hospital. The exposure of interest was preoperative hyperglycemia, defined as a fasting plasma glucose (FPG) ≥ 7 mmol/L. The outcome of interest was a new diagnosis of perioperative ischemic stroke within 30 days after surgery. Residual confounding was minimized by controlling for observable patient and intraoperative factors. Logistic regression was conducted in the total and propensity score matched cohorts. In addition, we stratified patients into six subgroups to investigate whether the association between preoperative hyperglycemia and perioperative ischemic stroke differs in these subgroups. Results: The overall incidence of perioperative ischemic stroke was 0.53% (n = 144) in the current cohort. The odds of perioperative ischemic stroke were significantly increased for patients with preoperative hyperglycemia after adjusting for patient- related variables (OR: 1.95; 95% CI: 1.39-2.75; p < 0.001), surgery-related variables (OR: 2.1; 95% CI: 1.51-2.94; p < 0.001), and all confounding variables (OR: 1.78; 95% CI: 1.26-2.53; p < 0.001). The risk of perioperative stroke was significantly increased in patients with preoperative hyperglycemia (OR: 2.51; 95% CI: 1.66-3.9; p < 0.001) in the propensity score matched cohort. Preoperative hyperglycemia was associated with the outcome for all the subgroups except for patients undergoing neurosurgery. Conclusion: Preoperative hyperglycemia is associated with an elevated risk of perioperative stroke in patients with type 2 DM undergoing non-cardiovascular surgery. The effect could be eliminated for patients undergoing neurosurgery, during which specific risk factors should be considered.
ABSTRACT
BACKGROUND: N-terminal pro-B-type natriuretic peptide (NT-proBNP) is often viewed as an indicator for heart failure. However, the prognostic association and the predictive utility of NT-proBNP for postoperative major adverse cardiovascular events (MACEs) and myocardial injury after noncardiac surgery (MINS) among older patients are unclear. METHODS: In this study, we included 5033 patients aged 65 years or older who underwent noncardiac surgery with preoperative NT-proBNP recorded. Logistic regression was adopted to model the associations between preoperative NT-proBNP and the risk of MACEs and MINS. The receiver operating characteristic curve was used to determine the predictive value of NT-proBNP. RESULTS: A total of 5033 patients were enrolled, 63 patients (1.25%) and 525 patients (10.43%) had incident postoperative MACEs and MINS, respectively. Analysis of the receiver operating characteristic curve indicated that the cutoff values of ln (NT-proBNP) for MACEs and MINS were 5.16 (174 pg/mL) and 5.30 (200 pg/mL), respectively. Adding preoperative ln (NT-proBNP) to the Revised Cardiac Risk Index score and the Cardiac and Stroke Risk Model boosted the area under the receiver operating characteristic curves from 0.682 to 0.726 and 0.787 to 0.804, respectively. The inclusion of preoperative NT-proBNP in the prediction models significantly increased the reclassification and discrimination. CONCLUSIONS: Increased preoperative NT-proBNP was associated with a higher risk of postoperative MACEs and MINS. The inclusion of NT-proBNP enhances the predictive ability of the preexisting models.
ABSTRACT
BACKGROUND: Postoperative delirium (POD) is a common complication among elderly patients after surgery. It is unclear whether the systemic immune-inflammation index (SII) can be a predictor of POD. We explored the prognostic value of the SII in predicting POD in elderly patients undergoing non-neurosurgery and non-cardiac surgery in a large retrospective cohort. METHODS: We enrolled elderly patients undergoing non-neurosurgery and non-cardiac surgery between January 2014 and August 2019. Univariate and multivariate logistic regression analyses were performed to explore the correlation between POD and the SII value as both a continuous and categorical variable. Then, propensity score matching (PSM) analysis was applied to eliminate the confounding effect of covariates and prove our results. Subgroup analyses were then performed to discover the association between the SII and POD in different subgroups. RESULTS: A total of 29,608 patients with a median age of 70 years (IQR: 67-74) were enrolled in the retrospective cohort. The cut-off value of the SII was 650, which was determined by the receiver operating characteristic (ROC) curve. The ORs of an SII value > 650 was 2.709 (95% CI:2.373-3.092, P < 0.001), 1.615 (95% CI:1.384-1.882, P < 0.001), 1.855 (95% CI:1.602-2.146, P < 0.001), and 1.302 (95% CI:1.106-1.531, P = 0.001) for prediction of POD in univariate model and three multivariate regression models. After PSM, the OR of an SII value > 650 was 1.301 (95% CI: 1.062-1.598, P = 0.011). The subgroup analysis indicated that the SII indicates a significantly increased risk of POD in patients with Hb < 130 g/L, 4*109/L < WBC ≤ 10*109/L, albumin < 39 g/L, or duration of MAP < 60 mmHg ≥ 5 min. The SII was found to be a useful prognostic predictor of POD for patients of different ages, sexes, and ASA classifications. CONCLUSIONS: The SII had a predictive value for POD in patients undergoing non-neurosurgery and non-cardiac surgery. As an index generated from routine blood tests, the SII has advantages regarding cost and time. After further validation, the SII may provide a new option for POD prediction.
Subject(s)
Delirium , Aged , Cohort Studies , Delirium/diagnosis , Delirium/etiology , Humans , Inflammation/diagnosis , Inflammation/etiology , Prognosis , Retrospective StudiesABSTRACT
Background: Although a variety of data showing that diabetes mellitus (DM) (Type 1 or Type 2) is associated with postoperative complication, there is still a lack of detailed studies that go through the specific diabetic subgroups. The goal of this meta-analysis is to assess the relationship between DM and various complications after non-cardiac surgery. Methods: We searched articles published in three mainstream electronic databases (PubMed, EMBASE, Web of science) before November, 2020. A random effects model was conducted since heterogeneity always exist when comparing results between different types of surgery. Results: This paper included 125 studies with a total sample size of 3,208,776 participants. DM was a risk factor for any postoperative complication (Odds ratio (OR)=1.653 [1.487, 1.839]). The risk of insulin-dependent DM (OR=1.895 [1.331, 2.698]) was higher than that of non-insulin-dependent DM (OR=1.554 [1.061, 2.277]) for any postoperative complication. DM had a higher risk of infections (OR=1.537 [1.322, 1.787]), wound healing disorders (OR=2.010 [1.326, 3.046]), hematoma (OR=1.369 [1.120, 1.673]), renal insufficiency (OR=1.987 [1.311, 3.013]), myocardial infarction (OR=1.372 [0.574, 3.278]). Meanwhile, DM was a risk factor for postoperative reoperation (OR=1.568 [1.124, 2.188]), readmission (OR=1.404 [1.274, 1.548]) and death (OR=1.606 [1.178, 2.191]). Conclusions: DM is a risk factor for any postoperative complications, hospitalization and death after non-cardiac surgery. These findings underscore the importance of preoperative risk factor assessment of DM for the safe outcome of surgical patients.
Subject(s)
Diabetes Mellitus, Type 1 , Diabetes Mellitus, Type 2 , Diabetes Mellitus, Type 1/complications , Diabetes Mellitus, Type 1/surgery , Diabetes Mellitus, Type 2/complications , Humans , Odds Ratio , Postoperative Complications/epidemiology , Postoperative Complications/etiology , Risk FactorsABSTRACT
BACKGROUND: Postoperative acute kidney injury (AKI) is associated with poor clinical outcomes. Early identification of high-risk patients of developing postoperative AKI can optimize perioperative renal management and facilitate patient survival. The present study aims to develop and validate a nomogram to predict postoperative AKI after liver resection in older patients. METHODS: A retrospective observational study was conducted involving data from 843 older patients scheduled for liver resection at a single tertiary high caseload general hospital between 2012 and 2019. The data were randomly divided into training (70%, n = 599) and validation (30%, n = 244) datasets. The training cohort was used to construct a predictive nomogram for postoperative AKI with the logistic regression model which was confirmed by a validation cohort. The model was evaluated by receiver operating characteristic (ROC) curve, calibration plot, and decision curve analysis in the validation cohort. A summary risk score was also constructed for identifying postoperative AKI patients. RESULTS: Postoperative AKI occurred in 155 (18.4%) patients and was highly associated with in-hospital mortality (5.2% vs. 0.7%, P < 0.001). The six predictors selected and assembled into the nomogram included age, preexisting chronic kidney disease (CKD), non-steroidal anti-inflammatory drugs (NSAIDs) usage, intraoperative hepatic inflow occlusion, blood loss, and transfusion. The predictive nomogram performed well in terms of discrimination with area under ROC curve (AUC) in training (0.73, 95% confidence interval (CI): 0.68-0.78) and validation (0.71, 95% CI: 0.63-0.80) datasets. The nomogram was well-calibrated with the Hosmer-Lemeshow chi-square value of 9.68 (P = 0.47). Decision curve analysis demonstrated a significant clinical benefit. The summary risk score calculated as the sum of points from the six variables (one point for each variable) performed as well as the nomogram in identifying the risk of AKI (AUC 0.71, 95% CI: 0.66-0.76). CONCLUSION: This nomogram and summary risk score accurately predicted postoperative AKI using six clinically accessible variables, with potential application in facilitating the optimized perioperative renal management in older patients undergoing liver resection. TRIAL REGISTRATION: NCT04922866 , retrospectively registered on clinicaltrials.gov on June 11, 2021.