Analysis of clinical predictors of kidney diseases in type 2 diabetes patients based on machine learning

Hui, Dongna; Sun, Yiyang; Xu, Shixin; Liu, Junjie; He, Ping; Deng, Yuhui; Huang, Huaxiong; Zhou, Xiaoshuang; Li, Rongshan

doi:10.1007/s11255-022-03322-1

Cited by 7 publications

(10 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our study showed the same results as the others 27 – 29 . In addition, DR had a close correlation with the presence of DN, which means that patients without DR were more likely to be diagnosed with NDKD 10 . However, there are still some severe DR patients without DN 30 .…”

Section: Discussionmentioning

confidence: 92%

See 1 more Smart Citation

Serum anti-phospholipase A2 receptor antibody in pathological diagnosis of type 2 diabetes mellitus patients with proteinuria

Zhang,

Zhu,

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Patients with diabetes mellitus complicated with proteinuria can be diabetic nephropathy (DN), diabetic complicated with non-diabetic kidney disease (NDKD), or DN with NDKD. Among these membranous nephropathy accounted for a large proportion of DN with NDRD. At present, serum anti-phospholipase A2 receptor (PLA2R) antibody is widely used in the diagnosis and evaluation of therapy in idiopathic membranous nephropathy, our study aimed to investigate the diagnostic significance of anti-PLA2R antibody in type 2 diabetes mellitus (T2DM) patients with proteinuria, providing a method for patients with contraindications of kidney biopsy. Eighty-seven T2DM patients with proteinuria who went on kidney biopsy were divided into the DN group, idiopathic membranous nephropathy (IMN) group, and others group according to their pathological results. In our study, 52.87% and 28.74% of patients were found to have IMN and diabetic nephropathy respectively. The levels of anti-PLA2R antibody, total cholesterol, triglyceride, and estimated glomerular filtration rate (eGFR) were higher in the IMN group, while the prevalence of diabetic retinopathy (DR), systolic blood pressure (SBP) and HbA1c were higher in the DN group. For T2DM patients with proteinuria, anti-PLA2R antibody (AUC = 0.904, 95%CI 0.838–0.970) has a high diagnostic value for IMN. The duration of diabetes (OR = 0.798, P = 0.030), eGFR level (OR = 1.030, P = 0.024), and positive anti-PLA2R antibody (OR = 72.727, P < 0.001) favor the diagnosis of IMN, while DR (OR = 50.234, P < 0.001), SBP (OR = 1.041, P = 0.030), and negative anti-PLA2R antibody (OR = 0.008, P = 0.001) is beneficial to the diagnosis of DN. Our study found that NDKD is not uncommon in patients with T2DM and proteinuria, and IMN was the main pathological type. Positive anti-PLA2R antibody has a strong accuracy in the diagnosis of IMN in patients with T2DM and proteinuria.

show abstract

Section: Discussionmentioning

confidence: 92%

“…In clinical practice, the diagnosis of DN mainly depends on medical history, clinical manifestations, and laboratory indicators, including diabetic retinopathy (DR), the duration of the disease, the levels of glycosylated hemoglobin (HbA1C), and so on 10 , 11 . However, a great part of NDKD is easy to be misdiagnosed.…”

Section: Introductionmentioning

confidence: 99%

Serum anti-phospholipase A2 receptor antibody in pathological diagnosis of type 2 diabetes mellitus patients with proteinuria

Zhang,

Zhu,

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In these diabetic patients, FIB was significantly correlated with UACR, whereas D-D levels showed no significant difference between the groups and lacked association with UACR. Hui et al [ 23 ] also found no significant difference in D-D levels between biopsy-confirmed DKD patients and those with type 2 diabetes without DKD. Several factors might account for the nuanced differences observed in our study compared to others.…”

Section: Discussionmentioning

confidence: 99%

Coagulation Function and Type 2 Diabetic Kidney Disease: A Real-World Observational Study

Liu,

Niu,

Wang

et al. 2023

Journal of Diabetes Research

View full text Add to dashboard Cite

Objectives. Type 2 diabetic kidney disease (DKD), a chronic microvascular complication of diabetes, may exhibit a complex interrelation with coagulation function. This study is aimed at elucidating the association between coagulation function and DKD. Methods. This was a real-world observational study conducted in Beijing, involving 2,703 participants. All patients with diabetes were classified into two groups, viz., DKD and non-DKD groups. Effect magnitudes are denoted as odds ratios (OR) with a 95% confidence interval (CI). To mitigate potential bias in group comparisons, we employed propensity score matching (PSM). Results. After adjusting for variables such as age, gender, systolic blood pressure (SBP), hemoglobin A1c (HbA1c), triglyceride (TG), c-reactive protein (CRP), platelet (PLT), and serum albumin (sALB), it was discerned that fibrinogen (FIB) (OR, 95% CI, P : 1.565, 1.289-1.901, <0.001) and fibrinogen degradation products (FDP) (1.203, 1.077-1.344, 0.001) were significantly correlated with an increased risk of DKD. To facilitate clinical applications, a nomogram prediction model was established, demonstrating commendable accuracy for DKD prediction. Conclusions. Our findings suggest that elevated levels of FIB and FDP serve as potential risk indicators for DKD, and coagulation function may play an important role in the occurrence and development of DKD.

show abstract

“…Three recent studies [21][22][23] have examined the associations of various clinical features with a kidney biopsy diagnosis and have used these associations to develop predictive models. Diabetic retinopathy was the only clinical feature which was included in all predictive models.…”

Section: Discussionmentioning

confidence: 99%

The Diagnostic Performance of a Clinical Diagnosis of Diabetic Kidney Disease

Tan

McDonald

Hoy

2023

Life

View full text Add to dashboard Cite

Background: Diabetic kidney disease (DKD), a common cause of CKD and kidney failure, is usually diagnosed clinically. However, there is little evidence comparing the performance of a clinical diagnosis to biopsy-proven diagnosis. Purpose of the study: Diagnostic performance of a clinical diagnosis was determined in a group of patients with diabetes and chronic kidney disease who underwent kidney biopsy after an initial clinical diagnosis. Methods: A data analysis of 54 patients who were part of a study cohort for a prospective analysis of cardiovascular and kidney outcomes and who had undergone kidney biopsy after an initial clinical diagnosis of DKD or non-DKD (NDKD) at enrolment was used. We determined the sensitivity, specificity, and positive and negative predictive values of a clinical diagnosis of DKD. Results: A total of 37 of 43 patients clinically diagnosed with DKD also had biopsy-proven DKD, whilst only 1 of 11 patients who had clinically diagnosed NDKD had biopsy-proven DKD. Sensitivity was 97.4%, specificity was 62.5%, positive predictive value 86%, and negative predictive value 90.9%. Comparable values were obtained when analysis was restricted to those with primary rather than secondary diagnosis of DKD or when restricted to those with only DKD found at biopsy. Conclusion: A clinical diagnosis of DKD has high sensitivity and is unlikely to overlook cases but may lead to overdiagnosis.

show abstract

Analysis of clinical predictors of kidney diseases in type 2 diabetes patients based on machine learning

Cited by 7 publications

References 34 publications

Serum anti-phospholipase A2 receptor antibody in pathological diagnosis of type 2 diabetes mellitus patients with proteinuria

Serum anti-phospholipase A2 receptor antibody in pathological diagnosis of type 2 diabetes mellitus patients with proteinuria

Coagulation Function and Type 2 Diabetic Kidney Disease: A Real-World Observational Study

The Diagnostic Performance of a Clinical Diagnosis of Diabetic Kidney Disease

Contact Info

Product

Resources

About