A clinical stratification tool for chronic kidney disease progression rate based on classification tree analysis

Rethinking Chronic Kidney Disease in the Aging Population

The process of aging population will inevitably increase age-related comorbidities including chronic kidney disease (CKD). In light of this demographic transition, the lack of an age-adjusted CKD classification may enormously increase the number of new diagnoses of CKD in old subjects with an indolent decline in kidney function. Overdiagnosis of CKD will inevitably lead to important clinical consequences and pronounced negative effects on the health-related quality of life of these patients. Based on these data, an appropriate workup for the diagnosis of CKD is critical in reducing the burden of CKD worldwide. Optimal management of CKD should be based on prevention and reduction of risk factors associated with kidney injury. Once the diagnosis of CKD has been made, an appropriate staging of kidney disease and timely prescriptions of promising nephroprotective drugs (e.g., RAAS, SGLT-2 inhibitors, finerenone) appear crucial to slow down the progression toward end-stage kidney disease (ESKD). The management of elderly, comorbid and frail patients also opens new questions on the appropriate renal replacement therapy for this subset of the population. The non-dialytic management of CKD in old subjects with short life expectancy features as a valid option in patient-centered care programs. Considering the multiple implications of CKD for global public health, this review examines the prevalence, diagnosis and principles of treatment of kidney disease in the aging population.

Potential value and impact of data mining and machine learning in clinical diagnostics

Data mining involves the use of mathematical sciences, statistics, artificial intelligence, and machine learning to determine the relationships between variables from a large sample of data. It has previously been shown that data mining can improve the prediction and diagnostic precision of type 2 diabetes mellitus. A few studies have applied machine learning to assess hypertension and metabolic syndrome-related biomarkers, as well as refine the assessment of cardiovascular disease risk. Machine learning methods have also been applied to assess new biomarkers and survival outcomes in patients with renal diseases to predict the development of chronic kidney disease, disease progression, and renal graft survival. In the latter, random forest methods were found to be the best for the prediction of chronic kidney disease. Some studies have investigated the prognosis of nonalcoholic fatty liver disease and acute liver failure, as well as therapy response prediction in patients with viral disorders, using decision tree models. Machine learning techniques, such as Sparse High-Order Interaction Model with Rejection Option, have been used for diagnosing Alzheimer's disease. Data mining techniques have also been applied to identify the risk factors for serious mental illness, such as depression and dementia, and help to diagnose and predict the quality of life of such patients. In relation to child health, some studies have determined the best algorithms for predicting obesity and malnutrition. Machine learning has determined the important risk factors for preterm birth and low birth weight. Published studies of patients with cancer and bacterial diseases are limited and should perhaps be addressed more comprehensively in future studies. Herein, we provide an in-depth review of studies in which biochemical biomarker data were analyzed using machine learning methods to assess the risk of several common diseases, in order to summarize the potential applications of data mining methods in clinical diagnosis. Data mining techniques have now been increasingly applied to clinical diagnostics, and they have the potential to support this field.

Temporal validation of the CT-PIRP prognostic model for mortality and renal replacement therapy initiation in chronic kidney disease patients

Background

A classification tree model (CT-PIRP) was developed in 2013 to predict the annual renal function decline of patients with chronic kidney disease (CKD) participating in the PIRP (Progetto Insufficienza Renale Progressiva) project, which involves thirteen Nephrology Hospital Units in Emilia-Romagna (Italy). This model identified seven subgroups with specific combinations of baseline characteristics that were associated with a differential estimated glomerular filtration rate (eGFR) annual decline, but the model’s ability to predict mortality and renal replacement therapy (RRT) has not been established yet.

Methods

Survival analysis was used to determine whether CT-PIRP subgroups identified in the derivation cohort (n = 2265) had different mortality and RRT risks. Temporal validation was performed in a matched cohort (n = 2051) of subsequently enrolled PIRP patients, in which discrimination and calibration were assessed using Kaplan-Meier survival curves, Cox regression and Fine & Gray competing risk modeling.

Results

In both cohorts mortality risk was higher for subgroups 3 (proteinuric, low eGFR, high serum phosphate) and lower for subgroups 1 (proteinuric, high eGFR), 4 (non-proteinuric, younger, non-diabetic) and 5 (non-proteinuric, younger, diabetic). Risk of RRT was higher for subgroups 3 and 2 (proteinuric, low eGFR, low serum phosphate), while subgroups 1, 6 (non-proteinuric, old females) and 7 (non-proteinuric, old males) showed lower risk. Calibration was excellent for mortality in all subgroups while for RRT it was overall good except in subgroups 4 and 5.

Conclusions

The CT-PIRP model is a temporally validated prediction tool for mortality and RRT, based on variables routinely collected, that could assist decision-making regarding the treatment of incident CKD patients. External validation in other CKD populations is needed to determine its generalizability.

Development of an electronic health record-based chronic kidney disease registry to promote population health management

BACKGROUND

Electronic health record (EHR) based chronic kidney disease (CKD) registries are central to population health strategies to improve CKD care. In 2015, Partners Healthcare System (PHS), encompassing multiple academic and community hospitals and outpatient care facilities in Massachusetts, developed an EHR-based CKD registry to identify opportunities for quality improvement, defined as improvement on both process measures and outcomes measures associated with clinical care.

METHODS

Patients are included in the registry based on the following criteria: 1) two estimated glomerular filtration rate (eGFR) results < 60 ml/min/1.73m2 separated by 90 days, including the most recent eGFR being < 60 ml/min/1.73m2; or 2) the most recent two urine protein values > 300 mg protein/g creatinine on either urine total protein/creatinine ratio or urine albumin/creatinine ratio; or 3) an EHR problem list diagnosis of end stage renal disease (ESRD). The registry categorizes patients by CKD stage and includes rates of annual testing for eGFR and proteinuria, blood pressure control, use of angiotensin converting enzyme inhibitors (ACE-Is) or angiotensin receptor blockers (ARBs), nephrotoxic medication use, hepatitis B virus (HBV) immunization, vascular access placement, transplant status, CKD progression risk; number of outpatient nephrology visits, and hospitalizations.

RESULTS

The CKD registry includes 60,503 patients and has revealed several opportunities for care improvement including 1) annual proteinuria testing performed for 17% (stage 3) and 31% (stage 4) of patients; 2) ACE-I/ARB used in 41% (stage 3) and 46% (stage 4) of patients; 3) nephrotoxic medications used among 23% of stage 4 patients; and 4) 89% of stage 4 patients lack HBV immunity. For advanced CKD patients there are opportunities to improve vascular access placement, transplant referrals and outpatient nephrology contact.

CONCLUSIONS

A CKD registry can identify modifiable care gaps across the spectrum of CKD care and enable population health strategy implementation. No linkage to Social Security Death Master File or US Renal Data System (USRDS) databases limits our ability to track mortality and progression to ESRD.

Predicting outcomes of chronic kidney disease from EMR data based on Random Forest Regression

Chronic kidney disease (CKD) is prevalent across the world, and kidney function is well defined by an estimated glomerular filtration rate (eGFR). The progression of kidney disease can be predicted if the future eGFR can be accurately estimated using predictive analytics. In this study, we developed and validated a prediction model of eGFR by data extracted from a regional health system. This dataset includes demographic, clinical and laboratory information from primary care clinics. The model was built using Random Forest regression and evaluated using Goodness-of-fit statistics and discrimination metrics. After data preprocessing, the patient cohort for model development and validation contained 61,740 patients. The final model included eGFR, age, gender, body mass index (BMI), obesity, hypertension, and diabetes, which achieved a mean coefficient of determination of 0.95. The estimated eGFRs were used to classify patients into CKD stages with high macro-averaged and micro-averaged metrics. In conclusion, a model using real-world electronic medical records (EMR) data can accurately predict future kidney functions and provide clinical decision support.

The PIRP project (Prevenzione Insufficienza Renale Progressiva): how to integrate hospital and community maintenance treatment for chronic kidney disease

Chronic kidney disease (CKD) represents a global health burden with great economic impact on healthcare and therefore it requires appropriate interventions by Health Care Systems. The PIRP (Prevenzione Insufficienza Renale Progressiva) project is endorsed and funded by the Emilia-Romagna Regional Health Board and involves all the Nephrology Units of the Emilia-Romagna Region (Italy). The project has a predominantly clinical purpose and is expected to bring about a continuous quality improvement in the treatment of patients with CKD. Its aims are to intercept patients in an early phase of CKD, to delay their illness progression and to prevent cardiovascular complications. An integrated care pathway involving nephrologists, general practitioners (GPs) and other specialists has been created to identify patients to whom ambulatory care targeted on effective, efficient pharmaceutical and dietary treatment as well as on lifestyle modifications is subsequently provided. With the cooperation of GPs, in its 13 years of activity the project identified and followed up more than 25,000 CKD patients, who attended the Nephrology units with more than 100,000 visits. The effects of a closer and joint monitoring of CKD patients by GPs and nephrologists can be quantified by the reduction of the mean annual GFR decline (average annual CKD-EPI change: - 0.34 ml/min), and by the decrease in the overall incidence of patients who annually started dialysis in the Emilia-Romagna Region, that dropped from 218.6 (× million) in 2006 to 197.5 (× million) in 2016, corresponding to about 100 cases.

Contemporary rates and predictors of fast progression of chronic kidney disease in adults with and without diabetes mellitus

Background

Chronic kidney disease (CKD) is highly prevalent but identification of patients at high risk for fast CKD progression before reaching end-stage renal disease in the short-term has been challenging. Whether factors associated with fast progression vary by diabetes status is also not well understood. We examined a large community-based cohort of adults with CKD to identify predictors of fast progression during the first 2 years of follow-up in the presence or absence of diabetes mellitus.

Methods

Within a large integrated healthcare delivery system in northern California, we identified adults with estimated glomerular filtration rate (eGFR) 30–59 ml/min/1.73 m² by CKD-EPI equation between 2008 and 2010 who had no previous dialysis or renal transplant, who had outpatient serum creatinine values spaced 10–14 months apart and who did not initiate renal replacement therapy, die or disenroll during the first 2 years of follow-up. Through 2012, we calculated the annual rate of change in eGFR and classified patients as fast progressors if they lost > 4 ml/min/1.73 m² per year. We used multivariable logistic regression to identify patient characteristics that were independently associated with fast CKD progression stratified by diabetes status.

Results

We identified 36,195 eligible adults with eGFR 30–59 ml/min/1.73 m² and mean age 73 years, 55% women, 11% black, 12% Asian/Pacific Islander and 36% with diabetes mellitus. During 24-month follow-up, fast progression of CKD occurred in 23.0% of patients with diabetes vs. 15.3% of patients without diabetes. Multivariable predictors of fast CKD progression that were similar by diabetes status included proteinuria, age ≥ 80 years, heart failure, anemia and higher systolic blood pressure. Age 70–79 years, prior ischemic stroke, current or former smoking and lower HDL cholesterol level were also predictive in patients without diabetes, while age 18–49 years was additionally predictive in those with diabetes.

Conclusions

In a large, contemporary population of adults with eGFR 30–59 ml/min/1.73 m², accelerated progression of kidney dysfunction within 2 years affected ~ 1 in 4 patients with diabetes and ~ 1 in 7 without diabetes. Regardless of diabetes status, the strongest independent predictors of fast CKD progression included proteinuria, elevated systolic blood pressure, heart failure and anemia.

The Patterns, Risk Factors, and Prediction of Progression in Chronic Kidney Disease: A Narrative Review

Chronic kidney disease (CKD) is a global public health problem that is associated with excess morbidity, mortality, and health resource utilization. The progression of CKD is defined by a decrease in glomerular filtration rate and leads to a variety of metabolic abnormalities including acidosis, hypertension, anemia, and mineral bone disorder. Lower glomerular filtration rate also bears a strong relationship with an increased risk of cardiovascular events, end-stage renal disease, and death. Patterns of CKD progression include linear and nonlinear trajectories, but kidney function can remain stable for years in some individuals. Addressing modifiable risk factors for the progression of CKD is needed to attenuate its associated morbidity and mortality. Developing effective risk prediction models for CKD progression is critical to identify patients who are more likely to benefit from interventions and more intensive monitoring. Accurate risk-prediction algorithms permit systems to best align health care resources with risk to maximize their effects and efficiency while guiding overall decision making.

Predicting Renal Failure Progression in Chronic Kidney Disease Using Integrated Intelligent Fuzzy Expert System

Background. Chronic kidney disease (CKD) is a covert disease. Accurate prediction of CKD progression over time is necessary for reducing its costs and mortality rates. The present study proposes an adaptive neurofuzzy inference system (ANFIS) for predicting the renal failure timeframe of CKD based on real clinical data. Methods. This study used 10-year clinical records of newly diagnosed CKD patients. The threshold value of 15 cc/kg/min/1.73 m² of glomerular filtration rate (GFR) was used as the marker of renal failure. A Takagi-Sugeno type ANFIS model was used to predict GFR values. Variables of age, sex, weight, underlying diseases, diastolic blood pressure, creatinine, calcium, phosphorus, uric acid, and GFR were initially selected for the predicting model. Results. Weight, diastolic blood pressure, diabetes mellitus as underlying disease, and current GFR_(t) showed significant correlation with GFRs and were selected as the inputs of model. The comparisons of the predicted values with the real data showed that the ANFIS model could accurately estimate GFR variations in all sequential periods (Normalized Mean Absolute Error lower than 5%). Conclusions. Despite the high uncertainties of human body and dynamic nature of CKD progression, our model can accurately predict the GFR variations at long future periods.

Excess mortality attributable to chronic kidney disease. Results from the PIRP project

Although chronic kidney disease (CKD) has a high mortality rate, the estimation of CKD mortality burden in the general population may be challenging because CKD is not always listed as a cause of death in mortality registries. To overcome this limitation, relative survival was used to estimate the excess mortality attributable to CKD as compared to the general population using data of patients registered in the Prevenzione Insufficienza Renale Progressiva (PIRP) registry since 2005 and were followed up until 2013. Relative survival was the ratio of survival observed in CKD patients to the expected survival of the general population. Multivariate parametric survival analysis was used to identify factors predicting excess mortality. The relative survival of CKD patients at 9 years was 0.708. Survival was significantly lower in CKD patients with cardiovascular comorbidities, proteinuria, diabetes, anemia and high phosphate levels and in advanced CKD stages, males, older patients and those who underwent dialysis. Relative survival is a viable method to determine mortality attributable to CKD. Study limitations are that patients are representative only of CKD patients followed by nephrologists and that our follow-up duration may be relatively short as a model for mortality.

Determinants of cesarean delivery: a classification tree analysis

BACKGROUND

Cesarean delivery (CD) rates are rising in many parts of the world. To define strategies to reduce them, it is important to identify their clinical and organizational determinants. The objective of this cross-sectional study is to identify sub-types of women at higher risk of CD using demographic, clinical and organizational variables.

METHODS

All hospital discharge records of women who delivered between 2005 and mid-2010 in the Emilia-Romagna Region of Italy were retrieved and linked with birth certificates. Sociodemographic and clinical information was retrieved from the two data sources. Organizational variables included activity volume (number of births per year), hospital type, and hour and day of delivery. A classification tree analysis was used to identify the variables and the combinations of variables that best discriminated cesarean from vaginal delivery.

RESULTS

The classification tree analysis indicated that the most important variables discriminating the sub-groups of women at different risk of cesarean section were: previous cesarean, mal-position/mal-presentation, fetal distress, and abruptio placentae or placenta previa or ante-partum hemorrhage. These variables account for more than 60% of all cesarean deliveries. A sensitivity analysis identified multiparity and fetal weight as additional discriminatory variables.

CONCLUSIONS

Clinical variables are important predictors of CD. To reduce the CD rate, audit activities should examine in more detail the clinical conditions for which the need of CD is questionable or inappropriate.