Patient classification and outcome prediction in IgA nephropathy.

OBJECTIVE: IgA Nephropathy (IgAN) is a common kidney disease which may entail renal failure, known as End Stage Kidney Disease (ESKD). One of the major difficulties dealing with this disease is to predict the time of the long-term prognosis for a patient at the time of diagnosis. In fact, the progression of IgAN to ESKD depends on an intricate interrelationship between clinical and laboratory findings. Therefore, the objective of this work has been the selection of the best data mining tool to build a model able to predict (I) if a patient with a biopsy proven IgAN will reach ESKD and (II) if a patient will reach the ESKD before or after 5 years. MATERIAL AND METHODS: The largest available cohort study worldwide on IgAN has been used to design and compare several data-driven models. The complete dataset was composed of 1174 records collected from Italian, Norwegian, and Japanese IgAN patients, in the last 30 years. The data mining tools considered in this work were artificial neural networks (ANNs), neuro fuzzy systems (NFSs), support vector machines (SVMs), and decision trees (DTs). A 10-fold cross validation was used to evaluate unbiased performances for all the models. RESULTS: An extensive model comparison based on accuracy, precision, recall, and f-measure was provided. Overall, the results indicate that ANNs can provide superior performance compared to the other models. The ANN for time-to-ESKD prediction is characterized by accuracy, precision, recall, and f-measure greater than 90%. The ANN for ESKD prediction has accuracy greater than 90% as well as precision, recall, and f-measure for the class of patients not reaching ESKD, while precision, recall, and f-measure for the class of patients reaching ESKD are slightly lower. The obtained model has been implemented in a Web-based decision support system (DSS). CONCLUSIONS: The extraction of novel knowledge from clinical data and the definition of predictive models to support diagnosis, prognosis, and therapy is becoming an essential tool for researchers and clinical practitioners in medicine. The proposed comparative study of several data mining models for the outcome prediction in IgAN patients, using a large dataset of clinical records from three different countries, provides an insight into the relative prediction ability of the considered methods applied to such a disease.

Blood pressure levels, left ventricular mass and function are correlated with left atrial volume in mild to moderate hypertensive patients.

Arterial hypertension is associated with an increased risk of atrial fibrillation (AF), and leads to a pronounced increase in morbidity and mortality. Left atrial volume (LAV) is an important prognostic marker in the older populations. The aim of our study was to identify the clinical and echocardiographic determinants of LAV in middle-aged (<70 years old) essential hypertensive patients.We evaluated cardiac structure and function in 458 patients, 394 treated and untreated mild to moderate essential hypertensives patients (mean+/-s.d. age 48.4+/-11.1 years) with no associated clinical condition and 64 normotensive control participants (age 45.7+/-12.8 years; P=0.12). A multivariate analysis was performed to calculate the relative weight of each of the variables considered able to predict LAV. The LAV index (LAVi) was significantly increased in the essential hypertensive group vs the control group and was significantly dependent on blood pressure levels (SBP and DBP, P<0.05 for both) and body mass index (BMI) (P<0.0001). Considering the left ventricular (LV) variables, the LV mass index (LVMI) (R(2)=0.19, P<0.001) and LAV were increased in essential hypertensive patients with left ventricular hypertrophy (LVH), and patients with enlarged LAV showed lower systolic and diastolic function and an increased LVMI. The LAVi is dependent on blood pressure levels and anthropometric variables (age and BMI). Further structural (LVMI) and functional (systolic and diastolic) variables are related to the LAVi; LVMI is the most important variable associated with LAV in mild to moderate essential hypertensive adult patients. These findings highlight the importance of left atrium evaluation in adult, relatively young, essential hypertensive patients.

Cognitive decline and low blood pressure: the other side of the coin.

Low blood pressure has been found to be associated with cognitive decline and dementia in cross-sectional studies. Two mechanisms have been proposed to interpret this association: blood pressure levels decrease during the course of the dementia process, and low blood pressure induces or accelerates cognitive decline by lowering cerebral blood flow. Results of the prospective studies are contradictory. Low blood pressure and orthostatic hypotension have been found to predict cognitive impairment in the elderly population in some studies only. While hypotension may play a protective role in healthy elderly people, low blood pressure levels in frail elderly patients with associated diseases may cause cerebral hypoperfusion and accelerate cognitive decline.

Baroreflex sensitivity is impaired in essential hypertensives with central obesity.

Recently, much interest has focussed on the potential interaction between sympathetic nervous system and global cardiovascular risk. We investigated how baroreflex sensitivity (BRS), an index of autonomic function, interacts with central obesity (CO) in an essential hypertensive (EH) population. We selected 170 EHs and 43 normotensives (NT), (median age 47.3+/-11.3 and 49.1+/-13 years, respectively). Anthropometric parameters were measured for each and BRS was evaluated by a non-invasive method using Portapres TNO. The BRS evaluation was made using the sequences method. Systolic blood pressure (SBP) and heart rate were significantly higher in EH (P<0.001 and P=0.007, respectively). BRS was significantly greater in NT (P=0.02), and was associated inversely with waist circumference (WC) (P=0.005), but not with SBP or with other metabolic risk factors. Body mass index, total and high-density lipoprotein cholesterol, age and WC were not significantly different between the two groups. These results were confirmed by age pounded analysis. Finally, a separate analysis of the hypertensive group with CO (n=84) demonstrated a significantly lower BRS compared with the other hypertensive patients (n=86) (P<0.001). BRS is associated with WC but not with arterial pressure values and metabolic risk factors. Hypertensive subjects with CO show an impairment of BRS. Owing to its association with abdominal fat distribution and subsequently insulin resistance, BRS could represent a further and reliable index for evaluation of global cardiovascular risk in hypertensive patients.

Accuracy of the blood pressure measurement.

Blood pressure measurement is the cornerstone for the diagnosis, the treatment and the research on arterial hypertension, and all of the decisions about one of these single aspects may be dramatically influenced by the accuracy of the measurement. Over the past 20 years or so, the accuracy of the conventional Riva-Rocci/Korotkoff technique of blood pressure measurement has been questioned and efforts have been made to improve the technique with automated devices. In the same period, recognition of the phenomenon of white coat hypertension, whereby some individuals with an apparent increase in blood pressure have normal, or reduced, blood pressures when measurement is repeated away from the medical environment, has focused attention on methods of measurement that provide profiles of blood pressure behavior rather than relying on isolated measurements under circumstances that may in themselves influence the level of blood pressure recorded. These methodologies have included repeated measurements of blood pressure using the traditional technique, self-measurement of blood pressure in the home or work place, and ambulatory blood pressure measurement using innovative automated devices. The purpose of this review to serve as a source of practical information about the commonly used methods for blood pressure measurement: the traditional Riva-Rocci method and the automated methods.