Dataset Information

Personalized Prediction of Kidney Function Decline and Network Analysis of the Risk Factors after Kidney Transplantation Using Nationwide Cohort Data.

ABSTRACT: We developed a machine-learning-based model that could predict a decrease in one-year graft function after kidney transplantation, and investigated the risk factors of the decreased function. A total of 4317 cases were included from the Korean Organ Transplant Registry (2014−2019). An XGBoost model was trained to predict the recipient’s one-year estimated glomerular filtration rate (eGFR) below 45 mL/min/1.73 m2 using 112 pre- and peri-transplantation variables. The network of model factors was drawn using inter-factor partial correlations and the statistical significance of each factor. The model with seven features achieved an area under the curve of 0.82, sensitivity of 0.73, and specificity of 0.79. The model prediction was associated with five-year graft and rejection-free survival. Post-transplantation hospitalization >25 days and eGFR ≥ 88.0 were the prominent risk and preventive factors, respectively. Donor age and post-transplantation eGFR < 59.8 were connected to multiple risk factors on the network. Therefore, careful donor−recipient matching in older donors, and avoiding pre-transplantation risk factors, would reduce the risk of graft dysfunction. The model might improve long-term graft outcomes by supporting early detection of graft dysfunction, and proactive risk factor control.

SUBMITTER: Hong MS

PROVIDER: S-EPMC8911006 | biostudies-literature | 2022 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Personalized Prediction of Kidney Function Decline and Network Analysis of the Risk Factors after Kidney Transplantation Using Nationwide Cohort Data.

Hong Moongi Simon MS Lee Yu-Ho YH Kong Jin-Min JM Kwon Oh-Jung OJ Jung Cheol-Woong CW Yang Jaeseok J Kim Myoung-Soo MS Han Hyun-Wook HW Nam Sang-Min SM Korean Organ Transplantation Registry Study Group

Journal of clinical medicine 20220225 5

We developed a machine-learning-based model that could predict a decrease in one-year graft function after kidney transplantation, and investigated the risk factors of the decreased function. A total of 4317 cases were included from the Korean Organ Transplant Registry (2014−2019). An XGBoost model was trained to predict the recipient’s one-year estimated glomerular filtration rate (eGFR) below 45 mL/min/1.73 m2 using 112 pre- and peri-transplantation variables. The network of model factors was ...[more]

PMID: 35268350

Similar Datasets

Project description:IntroductionImpaired kidney function is a robust risk factor for cardiovascular mortality. Age-related annual kidney function decline of 1.0 ml/min per 1.73 m2 after age 40 years is doubled in post-myocardial infarction (MI) patients.MethodsWe investigated the impact of the number of cardiovascular risk factors (including unhealthy lifestyle) on annual kidney function decline, in 2426 post-MI patients (60-80 years) of the prospective Alpha Omega Cohort study. Glomerular filtration rate was estimated by serum cystatin C (eGFRcysC) and combined creatinine-cystatin C (eGFRcr-cysC), using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations from 2012. Data were analyzed by multivariable linear and logistic regression.ResultsAt baseline, mean (SD) eGFRcysC and eGFRcr-cysC were 81.5 (19.6) and 78.5 (18.7) ml/min per 1.73 m2, respectively. Of all patients, 79% were men, 19% had diabetes, 56% had high blood pressure (≥140/90 mm Hg), 16% were current smokers, 56% had high serum low-density lipoprotein (LDL of ≥2.5 mmol/l), and 23% were obese (body mass index of ≥30.0 kg/m2). After multivariable adjustment, the additional annual eGFRcysC decline (95% confidence interval) was as follows: in patients with versus without diabetes, -0.90 (-1.23 to -0.57) ml/min per 1.73 m2; in patients with high versus normal blood pressure, -0.50 (-0.76 to -0.24) ml/min per 1.73 m2; in obese versus nonobese patients, -0.31 (-0.61 to 0.01) ml/min per 1.73 m2; and in current smokers versus nonsmokers, -0.19 (-0.54 to 0.16) ml/min per 1.73 m2. High LDL was not associated with accelerated eGFRcysC decline. Similar results were obtained with eGFRcr-cysC.ConclusionIn older, stable post-MI patients without cardiovascular risk factors, the annual kidney function decline was -0.90 (-1.16 to -0.65) ml/min per 1.73 m2. In contrast, in post-MI patients with ≥3 cardiovascular risk factors, the annual kidney function decline was 2.5-fold faster, at -2.37 (-2.85 to -1.89) ml/min per 1.73 m2.

Project description:To explore the risk factors of developing chronic pancreatitis (CP) in patients with acute pancreatitis (AP) and develop a prediction score for CP. Using the National Health Insurance Research Database in Taiwan, we obtained large, population-based data of 5971 eligible patients diagnosed with AP from 2000 to 2013. After excluding patients with obstructive pancreatitis and biliary pancreatitis and those with a follow-up period of less than 1 year, we conducted a multivariate analysis using the data of 3739 patients to identify the risk factors of CP and subsequently develop a scoring system that could predict the development of CP in patients with AP. In addition, we validated the scoring system using a validation cohort. Among the study subjects, 142 patients (12.98%) developed CP among patients with RAP. On the other hand, only 32 patients (1.21%) developed CP among patients with only one episode of AP. The multivariate analysis revealed that the presence of recurrent AP (RAP), alcoholism, smoking habit, and age of onset of < 55 years were the four important risk factors for CP. We developed a scoring system (risk score 1 and risk score 2) from the derivation cohort by classifying the patients into low-risk, moderate-risk, and high-risk categories based on similar magnitudes of hazard and validated the performance using another validation cohort. Using the prediction score model, the area under the curve (AUC) [95% confidence interval (CI)] in predicting the 5-year CP incidence in risk score 1 (without the number of AP episodes) was 0.83 (0.79, 0.87), whereas the AUC (95%CI) in risk score 2 (including the number of AP episodes) was 0.84 (0.80, 0.88). This result demonstrated that the risk score 2 has somewhat better prediction performance than risk score 1. However, both of them had similar performance between the derivation and validation cohorts. In the study,we identified the risk factors of CP and developed a prediction score model for CP.

Project description:BackgroundLower physical activity, lower alcohol intake, higher protein intake, higher sodium intake, and lower potassium intake related to greater kidney function decline over time, according to previous studies. The present study aimed to analyze the cumulative effects of these factors.MethodsThis prospective, observational, population-based cohort study included 3039 adult examinees of the Gubbio study who participated in the baseline exam and 15-year follow-up exam. Kidney function was evaluated as estimated glomerular filtration rate (eGFR). Habitual physical activity in leisure time and habitual alcohol intake were assessed by questionnaires; dietary intakes of protein, sodium, and potassium were assessed by urinary markers. Based on previous reports, each one of the five modifiable factors was scored 0 for the tertile associated with smaller eGFR decline (low risk), 2 for the tertile associated with greater eGFR decline (high risk), and 1 for the intermediate tertile (intermediate risk). A cumulative score was calculated as the sum of the factor-specific scores and used as the main independent variable.ResultsThe cumulative score ranged from 0 to 10, that is, from low risk for all factors to high risk for all factors (skewness = 0.032, mean ± SD = 5 ± 2). To avoid the bias of low-n analyses, score 0 was re-coded as 1 and score 10 was recoded as 9; after re-coding, the cumulative score ranged from 1 to 9 (skewness = 0.016, mean ± SD = 5 ± 2). The cumulative score related to annualized eGFR change in multi-variable linear regression (slope = -0.027, 95%CI = -0.039/-0.014, p < 0.001); findings were consistent in apparently healthy examinees and other subgroups. De novo incidence of eGFR < 60 mL/min × 1.73 m2 was higher along the cumulative score (p < 0.001). Compared to score 1 (n examinees = 35, adjusted incidence = 2.0%), incidence of low kidney function was 4.5 times higher in score 5 (n examinees = 624, adjusted incidence = 8.9%) and 6.5 times higher in score 9 (n examinees = 86, adjusted incidence = 12.9%). The cumulative score related to incidence of low kidney function in multi-variable logistic regression (odds ratio = 1.19, 95%CI = 1.08/1.32, p < 0.001).ConclusionsThe combination of five modifiable factors predicted large differences in long-term incidence of low kidney function.

Project description:Rationale & objectivePatients with kidney failure have poor physical performance, but its trajectory is less clear. We examined physical function over the course of kidney disease, including the transition to dialysis.Study designObservational cohort.Setting & participantsCommunity-dwelling adults aged ≥45 years in the Brain in Kidney Disease (BRINK) cohort study.PredictorsEstimated glomerular filtration rate (eGFR) and urine albumin to creatinine ratio (UACR).OutcomesChange in physical performance using the Short Physical Performance Battery (SPPB) (primary) and gait speed (secondary).Analytical approachLinear mixed effects regression models.ResultsThe analytical cohort included 562 participants with mean age of 69.3 (SD, 9.8) years followed for up to 63 months. In total, 49.8% were women. In addition, 79.9% self-identified as White, and 15.3% self-identified as Black. In total, 48.8% had diabetes. Mean eGFR at baseline was 48.1 (SD, 24.3) mL/min/1.73 m2. In unadjusted analysis, lower eGFR was associated with greater decline in SPPB score (P trend < 0.001). The decline in SPPB score was larger among participants with lower eGFR, with a gradient from -0.15 (95% CI, -0.23 to -0.07) points per year for participants with eGFR ≥60 mL/min/1.73 m2 to -0.56 (95% CI, -0.84 to -0.27) for participants with eGFR <15 mL/min/1.73 m2 and -0.61 (95% CI, -0.90 to -0.33) after dialysis initiation. In covariate-adjusted models, SPPB did not continue to decline after dialysis initiation. In secondary analyses evaluating change in gait speed, gait speed continued to decline after dialysis initiation. Higher UACR was also associated with a greater decline in SPPB score and gait speed in unadjusted and adjusted models.LimitationsSmall number of participants started dialysis.ConclusionsWe found a graded association of chronic kidney disease stage and albuminuria with decline in physical performance. The decline in SPPB was not accelerated after dialysis initiation in covariate-adjusted models, whereas gait speed continued to decline.

Project description:ObjectiveMost patients with uraemia must undergo chronic dialysis while awaiting kidney transplantation; however, the role of the pretransplant dialysis modality on the outcomes of kidney transplantation remains obscure. The objective of this study was to clarify the associations between the pretransplant dialysis modality, namely haemodialysis (HD) or peritoneal dialysis (PD), and the development of post-transplant de novo diseases, allograft failure and all-cause mortality for kidney-transplant recipients.DesignRetrospective nationwide cohort study.SettingData retrieved from the Taiwan National Health Insurance Research Database.ParticipantsThe National Health Insurance database was explored for patients who received kidney transplantation in Taiwan during 1998-2011 and underwent dialysis >90 days before transplantation.Outcome measuresThe pretransplant characteristics, complications during kidney transplantation and post-transplant outcomes were statistically analysed and compared between the HD and PD groups. Cox regression analysis was used to evaluate the HR of the dialysis modality on graft failure and all-cause mortality. The primary outcomes were long-term post-transplant death-censored allograft failure and all-cause mortality started after 90 days of kidney transplantation until the end of follow-up. The secondary outcomes were events during kidney transplantation and post-transplant de novo diseases adjusted by propensity score in log-binomial model.ResultsThere were 1812 patients included in our cohort, among which 1209 (66.7%) and 603 (33.3%) recipients received pretransplant HD and PD, respectively. Recipients with chronic HD were generally older and male, had higher risks of developing post-transplant de novo ischaemic heart disease, tuberculosis and hepatitis C after adjustment. Pretransplant HD contributed to higher graft failure in the multivariate analysis (HR 1.38, p<0.05) after adjustment for the recipient age, sex, duration of dialysis and pretransplant diseases. There was no significant between-group difference in overall survival.ConclusionsPretransplant HD contributed to higher risks of death-censored allograft failure after kidney transplantation when compared with PD.

Project description:Concussion, also known as mild traumatic brain injury (mTBI), commonly causes transient neurocognitive symptoms, but in some cases, it causes cognitive impairment, including working memory (WM) deficit, which can be long-lasting and impede a patient's return to work. The predictors of long-term cognitive outcomes following mTBI remain unclear, because abnormality is often absent in structural imaging findings. Previous studies have demonstrated that WM functional activity estimated from functional magnetic resonance imaging (fMRI) has a high sensitivity to postconcussion WM deficits and may be used to not only evaluate but guide treatment strategies, especially targeting brain areas involved in postconcussion cognitive decline. The purpose of the study was to determine whether machine learning-based models using fMRI biomarkers and demographic or neuropsychological measures at the baseline could effectively predict the 1-year cognitive outcomes of concussion. We conducted a prospective, observational study of patients with mTBI who were compared with demographically matched healthy controls enrolled between September 2015 and August 2020. Baseline assessments were collected within the first week of injury, and follow-ups were conducted at 6 weeks, 3 months, 6 months, and 1 year. Potential demographic, neuropsychological, and fMRI features were selected according to their significance of correlation with the estimated changes in WM ability. The support vector machine classifier was trained using these potential features and estimated changes in WM between the predefined time periods. Patients demonstrated significant cognitive recovery at the third month, followed by worsened performance after 6 months, which persisted until 1 year after a concussion. Approximately half of the patients experienced prolonged cognitive impairment at the 1-year follow up. Satisfactory predictions were achieved for patients whose WM function did not recover at 3 months (accuracy = 87.5%), 6 months (accuracy = 83.3%), and 1 year (accuracy = 83.3%) and performed worse at the 1-year follow-up compared to the baseline assessment (accuracy = 83.3%). This study demonstrated the feasibility of personalized prediction for long-term postconcussive WM outcomes based on baseline fMRI and demographic features, opening a new avenue for early rehabilitation intervention in selected individuals with possible poor long-term cognitive outcomes.

Dataset Information

Personalized Prediction of Kidney Function Decline and Network Analysis of the Risk Factors after Kidney Transplantation Using Nationwide Cohort Data.

Publications

Personalized Prediction of Kidney Function Decline and Network Analysis of the Risk Factors after Kidney Transplantation Using Nationwide Cohort Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets