Proprotein convertase subtilisin/kexin type 9 (PCSK9) and clinical outcomes in dialysis patients.

BACKGROUND: Proprotein convertase subtilisin/kexin type 9 (PCSK9), a factor accelerating the degradation of LDL receptors, was associated with a gender-dependent risk for cardiovascular (CV) events in the general population and with all-cause and CV mortality in two relatively small studies in black Africans and South Korean haemodialysis patients. The effect modification by gender was untested in these studies. METHODS: The study enrolled 1188 dialysis patients from the Prospective Registry of The Working Group of Epidemiology of Dialysis Region Calabria (PROGREDIRE) cohort. PCSK9 was measured by colorimetric enzyme-linked immunosorbent assay. The primary outcomes were all-cause and CV mortality. Statistical analysis included Cox regression analysis and effect modification analysis. RESULTS: During a median 2.9-year follow-up, out of 494 deaths, 278 were CV-related. In unadjusted analyses, PCSK9 levels correlated with increased all-cause (HRfor1ln unit increase: 1.23, 95% CI 1.06-1.43, p =.008) and CV mortality (HRfor1ln unit increase: 1.26, 95% CI 1.03-1.54, p =.03). After multivariate adjustment, these associations were no longer significant (all-cause mortality, HRfor 1 ln unit increase: 1.16, 95% CI .99-1.36, p =.07; CV mortality, HRfor1ln unit increase: 1.18, 95% CI .95-1.46, p =.14). However, in fully adjusted interaction analyses, a doubling in the risk of this outcome in women was registered (Women, HRfor1ln unit increase: 1.88, 95% CI 1.27-2.78, p =.002; Men, HRfor1ln unit increase: 1.07, 95% CI .83-1.38, p =.61; p for effect modification: .02). CONCLUSIONS: PCSK9 levels are unrelated to all-cause mortality in haemodialysis patients but, like in studies of the general population, independently of other risk factors, entail a doubling in the risk of CV events in women in this population.

Home Pulse Pressure Predicts Death and Cardiovascular Events in Peritoneal Dialysis Patients.

Increased arterial hypertension represents a prevalent condition in peritoneal dialysis patients that is often related to volume expansion. Pulse pressure is a robust predictor of mortality in dialysis patients, but its association with mortality is unknown in peritoneal patients. We investigated the relationship between home pulse pressure and survival in 140 PD patients. During a mean follow-up of 35 months, 62 patients died, and 66 experienced the combined event death/CV events. In a crude COX regression analysis, a five-unit increase in HPP was associated with a 17% increase in the hazard ratio of mortality (HR: 1.17, 95% CI 1.08-1.26 p < 0.001). This result was confirmed in a multiple Cox model adjusted for age, gender, diabetes, systolic arterial pressure, and dialysis adequacy (HR: 1.31, 95% CI 1.12-1.52, p = 0.001). Similar results were obtained considering the combined event death-CV events as an outcome. Home pulse pressure represents, in part, arterial stiffness, and it is strongly related to all-cause mortality in peritoneal patients. In these high cardiovascular risk populations, it is important to maintain optimal blood pressure control, but it is fundamental to consider all the other cardiovascular risk indicators, such as pulse pressure. Home pulse pressure measurement is easy and feasible and can add important information for the identification and management of high-risk patients.

Fundamentals and Applications of the Receiver Operating Characteristic Curve Analysis in the Research of Endothelial Dysfunction in Chronic Kidney Disease.

Endothelial dysfunction (ED) starts early in chronic kidney disease (CKD) and is the hallmark of atherosclerosis in these patients. During recent years, numerous markers have emerged, aiming to predict the onset of ED in CKD patients. Therefore, there is a need to evaluate and assess the discriminatory ability (or diagnostic accuracy) of such a marker (i.e., the ability to correctly classify individuals as having a given disease or not) and identify the optimal cut-off value. A receiver operating characteristic (ROC) curve analysis has been used in the majority of the research papers evaluating the predictive ability of a marker of ED. It is a graphical plot combining pairs of sensitivity (true positive rate) on the y axis and the complement of specificity (1-specificity, false positive rate) in the x axis, corresponding to several of the cut-off values covering the complete range of possible values that this test/marker might take. Herein, using a series of practical examples derived from clinical studies on ED in the special population of CKD, we address the principles, fundamentals, advantages and limitations regarding the interpretation of the ROC analysis.

Monitoring Risk Factors and Improving Adherence to Therapy in Patients With Chronic Kidney Disease (Smit-CKD Project): Pilot Observational Study.

BACKGROUND: Chronic kidney disease is a major public health issue, with about 13% of the general adult population and 30% of the elderly affected. Patients in the last stage of this disease have an almost uniquely high risk of death and cardiovascular events, with reduced adherence to therapy representing an additional risk factor for cardiovascular morbidity and mortality. Considering the increased penetration of mobile phones, a mobile app could educate patients to autonomously monitor cardiorenal risk factors. OBJECTIVE: With this background in mind, we developed an integrated system of a server and app with the aim of improving self-monitoring of cardiovascular and renal risk factors and adherence to therapy. METHODS: The software infrastructure for both the Smit-CKD server and Smit-CKD app was developed using standard web-oriented development methodologies preferring open source tools when available. To make the Smit-CKD app suitable for Android and iOS, platforms that allow the development of a multiplatform app starting from a single source code were used. The integrated system was field tested with the help of 22 participants. User satisfaction and adherence to therapy were measured by questionnaires specifically designed for this study; regular use of the app was measured using the daily reports available on the platform. RESULTS: The Smit-CKD app allows the monitoring of cardiorenal risk factors, such as blood pressure, weight, and blood glucose. Collected data are transmitted in real time to the referring general practitioner. In addition, special reminders improve adherence to the medication regimen. Via the Smit-CKD server, general practitioners can monitor the clinical status of their patients and their adherence to therapy. During the test phase, 73% (16/22) of subjects entered all the required data regularly and sent feedback on drug intake. After 6 months of use, the percentage of regular intake of medications rose from 64% (14/22) to 82% (18/22). Analysis of the evaluation questionnaires showed that both the app and server components were well accepted by the users. CONCLUSIONS: Our study demonstrated that a simple mobile app, created to self-monitor modifiable cardiorenal risk factors and adherence to therapy, is well tolerated by patients affected by chronic kidney disease. Further studies are required to clarify if the use of this integrated system will have long-term effects on therapy adherence and if self-monitoring of risk factors will improve clinical outcomes in this population.

A machine learning-based approach for predicting the outbreak of cardiovascular diseases in patients on dialysis.

BACKGROUND AND OBJECTIVE: Patients with End- Stage Kidney Disease (ESKD) have a unique cardiovascular risk. This study aims at predicting, with a certain precision, death and cardiovascular diseases in dialysis patients. METHODS: To achieve our aim, machine learning techniques have been used. Two datasets have been taken into consideration: the first is an Italian dataset obtained from the Istituto di Fisiologia Clinica of Consiglio Nazionale delle Ricerche of Reggio Calabria; the second is an American dataset provided by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) repository. From each one we obtained 5 datasets, according to the outcome of interest. We tested different types of algorithm (both linear and non-linear), but the final choice was to use Support Vector Machine. In particular, we obtained the best performances using the non-linear SVC with RBF kernel algorithm, optimizing it with GridSearch. The last is an algorithm useful to search the best combination of hyper-parameters (in our case, to find the best couple (C, γ)), in order to improve the accuracy of the algorithm. RESULTS: The use of non-linear SVC with RBF kernel algorithm, optimized with GridSearch, allowed to obtain an accuracy of 95.25% in the Italian dataset and of 92.15% in the American dataset, in a timeframe of 2.5 years,in the prediction of Ischaemic Heart Disease. A worse performance was obtained for the other outcomes. CONCLUSIONS: The machine learning-based approach applied in our study is able to predict, with a high accuracy, the outbreak of cardiovascular diseases in patients on dialysis.