Features Importance to Improve Interpretability of Chronic Kidney Disease Early Diagnosis

Author(s):  
Pedro A. Moreno-Sanchez
2020 ◽  
Vol 2020 ◽  
pp. 1-7
Author(s):  
Mostafa Abdelsalam ◽  
A. M. Wahab ◽  
Maysaa El Sayed Zaki ◽  
Mohamad Motawea

Background. Diabetes mellitus is the leading cause of end-stage renal disease worldwide. Microalbuminuria is the cornerstone for the diagnosis of diabetic nephropathy. However, it is an inadequate marker for early diagnosis. MicroRNAs are not only new and promising markers for early diagnosis but also, but they may also play a role in the prevention of disease progression. Methods. This study included ninety patients with type 2 DM in addition to 30 control subjects. MicroRNA-451 expression in blood and plasma using real-time PCR was evaluated in addition to the classic diabetic nephropathy markers (serum creatinine, urinary albumin, and eGFR). Results. There was a significant difference between the studied groups versus control regarding serum creatinine, eGFR, urinary, and plasma microRNA-451 with p=0.0001. Patients with eGFR 60 ml/min/1.73 m2 showed a significantly higher plasma microRNA-451 (29.6 ± 1.6) and significantly lower urinary microRNA-451 (21 ± 0.9) in comparison to patients with eGFR >60 ml/min/1.73 m2 and p=0.0001. eGFR showed a positive correlation with urinary microRNA-451 and negative correlation with both plasma microRNA-451 and urinary albumin. Both plasma and urinary microRNA-451 are highly sensitive and specific markers for chronicity in diabetic nephropathy patients with sensitivity of 90.9% and 95.5% and specificity of 67.6% and 95.6%, respectively. Conclusion. MicroRNA-451 is a promising early biomarker for chronic kidney disease in diabetic nephropathy with high sensitivity and specificity.


2021 ◽  
Vol 36 (Supplement_1) ◽  
Author(s):  
Irina Lousa ◽  
Maria João Valente ◽  
Susana Rocha ◽  
Sofia D. Viana ◽  
Inês Preguiça ◽  
...  

Abstract Background and Aims The conventionally used biomarkers for chronic kidney disease (CKD) diagnosis are not very sensitive for early diagnosis. Their values become clinically significant only when kidney damage is advanced, and a substantial filtration capacity has been lost. The reliance on these biomarkers may result in a long-time lapse in diagnosis, compromising the earlier use of successful therapeutic interventions to prevent CKD progression and reduce the risk of other common comorbidities. The study of earlier and more sensitive biomarkers for CKD diagnosis is an important medical need. Potentially new biomarkers reflecting different pathophysiological processes underlying CKD, such as changes in renal function, tubulointerstitial injury, inflammation and fibrosis, have been proposed. The use of a panel of biomarkers is likely to be synergetic in detecting CKD, since there are several different mechanisms by which CKD can initiate. Our aim was to identify markers of renal damage/dysfunction and evaluate their sensitivity for CKD detection, in patients at the earlier stages of the disease, stages 1 and 2. Method This study included 32 healthy controls and 29 CKD patients at stages 1 and 2, categorized according to the KDIGO guidelines, using the CKD-EPI equation based on serum creatinine to estimate the glomerular filtration rate (GFR). Causes of CKD in the studied patients were diabetes mellitus (n = 19), polycystic kidney disease (n = 1) and of unknown cause (n = 7) or other (n = 2). Circulating levels of creatinine and β-trace protein (BTP), as markers of renal function; interleukin 6 (IL-6), as a marker of inflammation; tissue inhibitor metalloproteinase 1 (TIMP 1), as a marker of tubulointerstitial injury; pro B-type natriuretic peptide (proBNP), as a marker of cardio-renal dysfunction; and cell-free DNA (cfDNA), as a marker of cellular damage, were evaluated. Results Compared to controls, we found significantly higher values of all studied markers in CKD patients (stage 1 and stage 2): BTP, TIMP-1, IL-6, pro-BNP, and cfDNA. In CKD patients, GFR was negatively correlated with circulating levels of pro-BNP (r = -0.610, P = 0.004, n = 20) and cfDNA (r = -0.408, P = 0.028, n = 29); and, microalbuminuria was positively correlated with circulating levels of BTP (r = 0.465, P = 0.013, n = 28). The biomarker cfDNA was positively correlated with TIMP-1 (r = 0.445, P = 0.16, n = 29) , a marker of tubulointerstitial injury, and with IL-6 (r = 0.670, P < 0.001, n = 29), a marker of inflammation. All patients presented at least two of the studied biomarkers with higher values than the median value presented by controls. Of all studied biomarkers, BTP was the one that was most altered in patients (86.2% presented higher values than the highest value presented by controls). Conclusion Our results suggest that the studied biomarkers are sensitive to the primary response to renal injury, being significantly elevated in the earlier stages of CKD, particularly BTP. Pro-BNP and cfDNA correlate well with disease severity assessed by GFR. The use of a panel comprising several biomarkers, related with different pathophysiological mechanisms underlying CKD initiation and progression, may increase the potential to detect patients at risk, when compared with the evaluation of each biomarker alone. Further validation for the use of these new potential biomarkers requires larger studies with standardized analytical methodologies. Acknowledgments: This work was supported by Applied Molecular Biosciences Unit (UCIBIO) and financed by FEDER COMPETE2020 funds UIDB/04378/2020 and UIDP/04539/2020 (CIBB); by POCI-01-0145-FEDER-007440; by FCT doctoral grant SFRH/BD/145939/2019; by funds from Portugal Regional Coordination and Development Commissions (Norte-01-0145-FEDER-000024).


2015 ◽  
Vol 23 (4) ◽  
pp. 223-229
Author(s):  
Luciana Saraiva da Silva ◽  
Rosângela Minardi Mitre Cotta ◽  
Tiago Ricardo Moreira ◽  
Rodrigo Gomes da Silva ◽  
Carla de Oliveira Barbosa Rosa

2021 ◽  
Vol 38 (6) ◽  
pp. 74-82
Author(s):  
Ekaterina Alekseevna Burtseva ◽  
Vitaly Yurievich Mishlanov ◽  
Anna Vladimirovna Anikeeva ◽  
Victoria Ivanovna Selezneva ◽  
Ekaterina Petrovna Koshurnikova ◽  
...  

Objective. To conduct a comparative study of the prevalence of specific symptoms in patients with kidney damage using the automated program "Electronic Polyclinic" to optimize the algorithm for early diagnosis of chronic kidney disease. Materials and methods. 18 patients of the therapeutic unit with kidney lesions, confirmed by laboratory and instrumental studies, as well as 7 healthy persons were examined. The main problems of patients were identified by the method of interactive questioning with the help of the program "Electronic Polyclinic". Further, a statistical analysis of the data and a comparative study with the control group of healthy persons were carried out using STATISTICA 12.0 program. Results. The main symptoms of kidney damage were reliably determined and a low sensitivity of individual symptoms in the diagnosis, in contrast to the syndromic approach, was revealed. It showed that in 100 % of cases the automated program "Electronic Polyclinic" detected the syndromes, that indicates its high efficiency. An algorithm for early diagnosis of chronic kidney disease was proposed, which consists in an initial interactive survey followed by examination formulated by the computer program. Conclusions. The method of interactive survey using the automated program "Electronic Polyclinic" allows you to effectively identify the syndromes of kidney damage, makes a preliminary diagnosis and draws up a plan of examination for further confirmation of the diagnosis. Symptomatic diagnosis has a number of disadvantages including low sensitivity and specificity, so it yields to syndromic diagnosis.


2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Ebrahime Mohammed Senan ◽  
Mosleh Hmoud Al-Adhaileh ◽  
Fawaz Waselallah Alsaade ◽  
Theyazn H. H. Aldhyani ◽  
Ahmed Abdullah Alqarni ◽  
...  

Chronic kidney disease (CKD) is among the top 20 causes of death worldwide and affects approximately 10% of the world adult population. CKD is a disorder that disrupts normal kidney function. Due to the increasing number of people with CKD, effective prediction measures for the early diagnosis of CKD are required. The novelty of this study lies in developing the diagnosis system to detect chronic kidney diseases. This study assists experts in exploring preventive measures for CKD through early diagnosis using machine learning techniques. This study focused on evaluating a dataset collected from 400 patients containing 24 features. The mean and mode statistical analysis methods were used to replace the missing numerical and the nominal values. To choose the most important features, Recursive Feature Elimination (RFE) was applied. Four classification algorithms applied in this study were support vector machine (SVM), k-nearest neighbors (KNN), decision tree, and random forest. All the classification algorithms achieved promising performance. The random forest algorithm outperformed all other applied algorithms, reaching an accuracy, precision, recall, and F1-score of 100% for all measures. CKD is a serious life-threatening disease, with high rates of morbidity and mortality. Therefore, artificial intelligence techniques are of great importance in the early detection of CKD. These techniques are supportive of experts and doctors in early diagnosis to avoid developing kidney failure.


2019 ◽  
Vol 13 (2) ◽  
pp. 125-127 ◽  
Author(s):  
Maria Dolores Sanchez-Niño ◽  
Beatriz Fernandez-Fernandez ◽  
Alberto Ortiz

Abstract Chronic kidney disease (CKD) is one of the fastest growing causes of death worldwide. Only early diagnosis will allow prevention of both CKD progression and the negative impact of CKD on all-cause and cardiovascular mortality. Klotho is a protein produced by the kidneys that has anti-ageing and phosphaturic properties, preventing excess positive phosphate balance. There is evidence that Klotho downregulation is one of the earliest consequences of kidney injury. Thus the development of reliable assays to monitor Klotho levels may allow an early diagnosis of CKD and monitoring the impact of therapies aimed at preserving Klotho expression or at preventing CKD progression. However, the performance of Klotho assays has been suboptimal so far. In this issue of Clinical Kidney Journal, Neyra et al. explore methods to improve the reliability of Klotho assays.


Sign in / Sign up

Export Citation Format

Share Document