Objective To develop a neural network model for the evaluation of glomerular filtration rate (GFR) based on multilayer perceptual neural network, and to compare with the improved Chinese based creatinine GFR evaluation formula (C-GFR_cr) and the evaluation formula (EPI-GFR_cr) of the American Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) for the clinical applicability of multilayer perceptual neural network model in evaluating GFR. Methods A total of 684 chronic kidney disease (CKD) patients used for developing a modified version of China′s based creatinine GFR evaluation formula were taken as the research object. The data of 454 patients were randomly selected as the development group and the data of the other 230 patients were as the verification group. The multilayer perceptual neural network GFR evaluation model (M-GFR_cr) was established. With the double plasma GFR as the reference value (rGFR), the correlation, mean difference, mean absolute difference, precision and accuracy of C-GFR_cr, EPI-GFR_cr and M-GFR_cr were compared. Results Among the 684 CKD patients, there were 352 males and 332 females, with age of (49.9±15.8) years. The correlation between M-GFR_cr and rGFR was the highest (Pearson correlation =0.93, P<0.001). The mean difference of M-GFR_cr was lower than that of C-GFR_cr (Z=9.929, P<0.001) and EPI-GFR_cr(Z=10.573, P<0.001). The mean absolute difference of M-GFR_cr was also lower than that of C-GFR_cr(Z=3.953, P<0.001) and EPI-GFR_cr (Z=4.210, P<0.001). The accuracy of ±15% of M-GFR_cr was higher than that of C-GFR_cr ( χ²=26.068, P<0.001) and EPI-GFR_cr ( χ²=23.154, P<0.001). The accuracy of ±30% of M-GFR_cr was also higher than that of C-GFR_cr ( χ²=8.264, P=0.001) and EPI-GFR_cr ( χ²=11.963, P=0.001). The results of different stages of CKD showed that in the early stage of CKD (CKD 1-2), the mean difference of M-GFR_cr was lower than that of C-GFR_cr (Z=7.401, P<0.001) and EPI-GFR_cr (Z=8.096, P<0.001); the mean absolute difference of M-GFR_cr was also lower than that of C-GFR_cr (Z=4.723, P<0.001) and EPI-GFR_cr (Z=4.946, P<0.001); the accuracy of ±15% of M-GFR_cr was higher than that of C-GFR_cr ( χ²=23.547, P<0.001) and EPI-GFR_cr ( χ²=26.421, P<0.001); the accuracy of ±30% of M-GFR_cr was also higher than that of C-GFR_cr ( χ²=12.089, P=0.001) and EPI-GFR_cr ( χ²=16.168, P<0.001). But there was no significant difference in the applicability among C-GFR_cr, EPI-GFR_cr and M-GFR_cr in the advanced stages of CKD (CKD 3-5). Conclusion Compared with the improved Chinese based creatinine GFR evaluation formula C-GFR_cr and CKD-EPI evaluation formula EPI-GFR_cr, the accuracy of multilayer perceptual neural network model to evaluate GFR in CKD patients has been significantly improved, especially in CKD 1-2 stage.

Objective To explore the risk factors and time distribution of renal relapse in patients with lupus nephritis (LN). Methods Clinical, pathological characteristics and long-term outcomes of LN patients who were diagnosed and followed in Jinling Hospital from January 2004 to December 2008 were retrospectively analyzed. The patients were divided into relapse group and non-relapse group. The differences of clinical pathological characteristics between the two groups were compared. The multivariate Cox proportion risk model was used to analyze the risk factors affecting renal relapse in LN. The risk factors and time distribution of renal relapse were analyzed with annual relapse risk-time curve. Results A total of 814 patients with LN were included in the study, with 419 cases (51.5%) of complete remission and 395 cases (48.5%) of partial remission. The age was(30.24±10.90) years old, and there were 112 males (13.8%). There were 367 patients suffering renal relapse. The time of first renal relapse was (3.21±2.70) years. The results of multivariate Cox regression showed that age (HR=0.976, 95% CI 0.966-0.986, P<0.001), renal pathological activity index (AI) score (HR=1.039, 95% CI 1.013-1.065, P=0.003), remission status after induction treatment (HR=0.671, 95% CI 0.504-0.894, P=0.006), 24 h urinary protein quantitation (HR=1.297, 95% CI 1.115-1.508, P=0.001), anti-double strand DNA antibody (A-dsDNA, HR=1.450, 95% CI 1.139-1.846, P=0.003) and complement C3 (HR=0.223, 95% CI 0.128-0.389, P<0.001) were correlated with increasing risk of renal relapse in LN. The annual relapse risk profile was unimodal, with a peak period of the second year after maintenance treatment. Similar patterns of relapse were presented in subgroup analysis. Conclusions Age, renal pathological AI score, remission status after induction therapy, 24 h urine protein, A-dsDNA and blood complement C3 are the influencing factors for relapse of LN patients. The peak period of renal relapse in patients with LN is in the second year of maintenance therapy.

Objective To analyze the clinicopathological characteristics, treatment responses and kidney outcomes of patients with atypical membranous nephropathy (MN), and to provide information for the clinical practice. Methods The clinical data of patients with atypical MN and synchronous primary MN who were diagnosed, treated and followed up in Peking University First Hospital from January 2008 to June 2020 were retrospectively collected and analyzed. Clinicopathological features, treatment responses and kidney prognosis were compared between the two groups. The expression of phospholipase A2 receptor (PLA2R) in kidney tissues was detected by immunofluorescence. Serum anti-PLA2R antibody was detected by enzyme-linked immunosorbent assay. Clinicopathological indexes were compared between PLA2R-related MN group and non-PLA2R-related MN group. Kaplan-Meier (Log-rank test) survival curve and multivariate Cox regression analysis methods were used to analyze the influencing factors of kidney prognosis in patients with atypical MN. The primary endpoint of renal adverse outcome was renal insufficiency, defined as end-stage renal disease or estimated glomerular filtration rate (eGFR) decline>30% baseline and<60 ml·min^-1·(1.73 m²)^-1. Results A total of 65 atypical MN patients were enrolled in this study. Compared with primary MN (n=324), patients with atypical MN had younger age (Z=-4.229, P<0.001), higher proportion of hematuria ( χ²=5.555, P=0.018), higher level of urinary protein (Z=2.228, P=0.026) and lower level of eGFR (t=-5.108, P<0.001); the proportion of IgG4 deposition in kidneys was lower ( χ²=8.081, P=0.004), and the proportions of IgA ( χ²=16.969, P<0.001) and IgM ( χ²=9.281, P=0.002) deposition were higher. There was no significant difference on gender, serum albumin, positive proportion of anti-PLA2R antibody, anti-PLA2R antibody level and kidney C3/C1q deposition between the two groups (all P>0.05). The proportions of atypical MN patients receiving renin-angiotensin aldosterone system inhibitors (49.3% vs 57.1%), calcineurin inhibitors (27.7% vs 19.1%) and cyclophosphamide (21.5% vs 23.8%) were comparable to those of primary MN patients (all P>0.05). The rates of clinical remission (80.0% vs 77.2%), partial remission (44.6% vs 44.1%), complete remission (35.4% vs 33.1%), spontaneous remission (36.9% vs 42.6%), response to cyclophosphamide (85.7% vs 81.8%), response to calcineurin inhibitor (88.9% vs 79.0%), and relapse (30.8% vs 26.8%) in atypical MN patients were comparable to those in primary MN patients (all P>0.05). During the follow-up 30.0(21.5, 61.5) months, 15 atypical MN patients (23.1%) had eGFR reduction>30%, among whom 7 patients (10.8%) had eGFR reduction>50% and 3 patients (4.6%) had end-stage kidney disease. There was no significant difference on poor kidney prognosis between the two groups (all P>0.05). Kaplan-Meier survival curve showed that patients with age>39 years old ( χ²=10.092, P=0.001), eGFR≤100 ml·min^-1·(1.73 m²)^-1( χ²=5.491, P=0.019), tubular interstitial lesion ( χ²=6.999, P=0.008) and no nephropathy remission ( χ²=22.952, P<0.001) had earlier poor renal prognosis. Multivariate Cox regression analysis showed that no nephropathy remission (HR=12.604, 95%CI 2.691-59.037, P=0.001) was an independent influencing factor for poor renal prognosis in atypical MN patients. Conclusion No significant difference is found between atypical MN and primary MN on treatment responses and kidney prognosis, which implies that clinical practice of atypical MN can be performed by referring to the guidelines and experience of primary MN.

Objective To explore the risk factors of hypocalcemia and the correlation between calcium supplementation and clinical parameters after parathyroidectomy (PTX) in maintenance hemodialysis patients with secondary hyperparathyroidism (SHPT), and to analyze the effect of calcium supplementation after PTX on the long-term prognosis of patients. Methods This study was a single-center retrospective study. The patients who underwent PTX in maintenance hemodialysis patients with SHPT in the Huashan Hospital affiliated to Fudan University from October 2014 to March 2021 were retrospectively enrolled. Total PTX with auto transplantation or total PTX alone were the surgical procedures. According to the postoperative requirement of calcium in the first week, the patients were divided into two groups: high calcium supplement (>16.05 g/week) group and low calcium supplement group (≤16.05 g/week). According to the average serum calcium level in the first week after operation, the patients were divided into hypocalcemia group (≤2.1 mmol/L) and non-hypocalcemia group (>2.1 mmol/L) and the differences of clinical parameters between the two groups were compared. The correlation between clinical parameters and the postoperative calcium requirement was examined through Pearson or Spearman correlation analysis. The influencing factors for hypocalcemia after PTX were examined through logistic regression analysis. The survival curve was made by Kaplan-Meier method, and the difference of cumulative survival rate between the two groups was compared by log-rank test. Results A total of 98 maintenance hemodialysis patients with SHPT were enrolled. The levels of serum calcium, phosphorus, and intact parathyroid hormone (iPTH) after the operation decreased significantly than those of preoperation (all P<0.05). Multiple linear regression analysis showed age (β=-0.160, P=0.030), iPTH (β=0.004, P=0.025) and C-reactive protein (β=0.186, P=0.011) were correlated with postoperative calcium requirement. Preoperative alkaline phosphatase (OR=1.002, 95% CI 1.000-1.004, P=0.018) and hemoglobin (OR=0.977, 95% CI 0.954-1.000, P=0.048) independently predicted the occurrence risk of postoperative hypocalcemia through multivariate logistic regression analysis. The recurrence rate of high calcium supplement group was higher than that of low calcium supplement group (10.26% vs 0, P=0.023) and there was no significant difference in all-cause mortality between the two groups (17.95% vs 5.08%, P=0.086). The recurrence rate between the hypocalcemia group and non-hypocalcemia group was no significantly different (8.3% vs 1.8%, P=0.451) and there was no significant difference in all cause mortality between the two groups (12.5% vs 12.7%, P=1.000). Kaplan-Meier survival curve showed that the cumulative survival rate between the two groups was no significantly different (log-rank test χ²=0.147, P=0.702). Conclusions PTX is a safe and effective therapeutic method to reduce the level of iPTH and improve the metabolism of calcium and phosphorus in SHPT patients. Age, iPTH and C-reactive protein are correlated with the postoperative requirement of calcium in the first week. Preoperative alkaline phosphatase and hemoglobin are independent risk factors for postoperative hypocalcemia. Correcting preoperative electrolyte disorder, improving infection and anemia can reduce the incidence of hypocalcemia after PTX.

Objective To compare the survival rate of secondary hyperparathyroidism (SHPT) patients with different dialysis modalities after parathyroidectomy (PTX), and analyze the influencing factors of survival prognosis. Methods Clinical data of dialysis patients diagnosed with SHPT and treated with PTX in the First People′s Hospital of Foshan from April 2014 to May 2019 were retrospectively collected and analyzed. The patients were divided into hemodialysis (HD) group and peritoneal dialysis (PD) group according to preoperative dialysis modalities, and the differences in baseline clinical data and cardiac ultrasound results were compared between the two groups. Kaplan-Meier survival analysis was used to compare the difference in cumulative survival rate between the two groups. Multivariate Cox regression model was used to analyze the influencing factors of all-cause death. Receiver operating characteristic curve (ROC curve) was used to predict the risk of all-cause death. Results A total of 99 patients were enrolled in this study, and 94 patients completed follow-up, including 23 patients who died. Compared with PD group (n=45), HD group (n=54) had higher dialysis age, blood pressure, intact parathyroid hormone, alkaline phosphatase, total heart valve calcification rate, mitral valve calcification proportion, interventricular septal thickness (IVST) and left ventricular mass index (all P<0.05). The median follow-up time was 46.00(32.75, 60.25) months. Kaplan-Meier survival analysis showed that there was no significant difference in cumulative survival rate between HD group and PD group (Log-rank test χ²=0.414, P=0.520). Multivariate Cox regression analysis showed that increasing age (HR=1.066, 95%CI 1.017-1.118, P=0.008), systolic blood pressure>140 mmHg (HR=2.601, 95%CI 1.002-6.752, P=0.049) and increasing IVST (HR=1.269, 95%CI 1.036-1.554, P=0.021) were independent influencing factors for all-cause death in dialysis patients after PTX. ROC curve analysis results showed that the cut-off values of age, dialysis age and IVST for predicting all-cause death after PTX were 51.5 years old (AUC=0.673, 95%CI 0.545-0.802, P=0.013) and 75.0 months (AUC=0.654, 95%CI 0.528-0.780, P=0.027) and 13.5 mm (AUC=0.680, 95%CI 0.557-0.803, P=0.010) respectively. The area under the ROC curve for age, dialysis age, IVST, left ventricular hypertrophy in combination with systolic blood pressure>140 mmHg in the prediction of all-cause death after PTX was 0.776(95%CI 0.677-0.875, P<0.001). Conclusions There is no significant difference in cumulative survival rate between HD and PD patients with SHPT after PTX. Increasing age, systolic blood pressure>140 mmHg and increasing IVST are independent risk factors for all-cause death in dialysis patients with SHPT after PTX.

Objective To explore the influencing and prognosis factors of emphysematous urinary tract infection (EUTI). Methods The baseline clinical data of the patients admitted to Shandong University Qilu Hospital (Qingdao) from December 2013 to June 2020 and diagnosed with EUTI were analyzed retrospectively. The patients with non-EUTI (NEUTI) during the same period were selected as the control group. The baseline characteristics between the two groups were compared. Logistic regression analysis method was used to analyze the influencing factors of EUTI. Results (1) 24 EUTI patients and 53 NEUTI patients were included in the present study. Compared with the NEUTI group, the hemoglobin level was lower (t=-5.245, P<0.001) and the levels of blood urine nitrogen (Z=-4.361, P<0.001), serum creatinine (Scr, Z=-4.543, P<0.001), blood glucose (Z=-2.608, P=0.009), and triacylglycerol (Z=-2.408, P=0.016) were higher in the EUTI group. The proportions of diabetes mellitus ( χ²=13.453, P<0.001) and chronic kidney disease ( χ²=17.936, P<0.001) in the EUTI group were higher than those in the NEUTI group. Increasing Scr was the risk factor of EUTI in patients with urinary tract infection (OR=1.011, 95% CI 1.001-1.020, P=0.025). (2) Escherichia coli (E.coli, 14 cases, 58.3%) was the most common causative organism. The other causative organisms included Klebsiella pneumoniae (2 cases, 8.3%), Enterococcus faecium (1 case, 4.2%), Pantoea (1 case, 4.2%), and mixed bacteria of E.coli and Enterococcus faecium (1 case, 4.2%). Ten cases of E.coli were extended-spectrum β-lactamases (ESBL)-positive. (3) Of the 24 patients with EUTI, 4 patients had adverse outcomes. The length of stay (Z=-2.457, P=0.014), blood urea nitrogen (t=2.432, P=0.024), shock (P=0.002), autoimmune disease (P=0.022), and white blood cell count (Z=-2.091, P=0.036) were statistically different between good prognosis group (n=20) and poor prognosis group (n=4). However, logistic regression analysis results showed that neither was the influencing factor of poor prognosis of EUTI. Conclusions The elevated Scr level is the independent influencing factor of EUTI among urinary infection patients. E.coli is the most common pathogenic bacteria, and ESBL-positive bacteria are common.

Objective To investigate whether it is by regulating interleukin 1β (IL-1β) gene expression that androgen receptor (AR) in macrophages affects hyperphosphate-induced vascular smooth muscle cell calcification. Methods The chromatin immunoprecipitation (ChIP) experiment was used to determine whether AR was bound to the androgen receptor element (ARE) sequence of IL-1β promoter in THP-1 cells. Whether the AR regulated IL-1β gene expression was detected by luciferase assay experiments. AR of THP-1 cells was silenced and transfected by lentivirus with vector or shRNA. Flow cytometry was used to select positive transfected cells THP-1ARsc (control) and THP-1ARsi (AR silencing) with fluorescent markers. Western blotting was used to detect AR protein levels of THP-1ARsc (control) and THP-1ARsi cells (AR silencing in monocytes). Macrophages MФARsc (control) or MФARsi (AR silencing) were induced by 50 ng/ml phorbol ester. Enzyme-linked immunosorbent assay was used to detect IL-1β expression levels of MФARsc or MФARsi conditioned medium. The human aortic smooth muscle cells (HASMC) were cultured in MФARsc or MФARsi conditioned medium with phosphate (2.5 mmol/L final concentration of sodium dihydrogen phosphate), and Alizarin red S staining was used to analyze HASMC calcification degree. Western blotting was used to detect the expression levels of RUNX2 (osteoblast marker) and SM22α (HASMC marker), and neutralization assay was performed to test IL-1β-mediating effect of macrophages AR on HASMC calcification. Results AR was bound to ARE sequence of IL-1β promoter and regulated IL-1β gene expression. The expression level of IL-1β protein in conditioned medium of MФARsi cells decreased significantly compared to MФARsc cells (P<0.001). Compared with MФARsc conditioned medium group, HASMC calcium deposition in MФARsi conditioned medium group decreased significantly, RUNX2 protein decreased and SM22α protein increased (all P<0.05). The degree of HASMC calcification in the MФARsi conditioned medium+IgG antibody group decreased than that in the MФARsc conditioned medium+IgG antibody group significantly, and the degree of HASMC calcification in the MФARsc conditioned medium+IL-1β antibody group decreased significantly than that in the MФARsc conditioned medium+IgG antibody group; while the degree of HASMC calcification in the MФARsi conditioned medium+IgG antibody group and MФARsi conditioned medium+IL-1β antibody group decreased than that in the MФARsc conditioned medium+IL-1β antibody group (all P<0.05). Conclusions Macrophage AR regulates IL-1β expression by binding to ARE sequence within IL-1β promoter, and IL-1β mediates the effect of macrophage AR on hyperphosphate-induced HASMC calcification.