Clinical observation and analysis of patients with idiopathic membranous nephropathy combined with low 25(OH)D3 levels

Background: To investigate correlations between 25-hydroxyvitamin D 3 [25(OH)D 3 ] levels and selected clinical indicators and renal pathology at initial onset of idiopathic membranous nephropathy (IMN) and to analyse changes in the main quantiable indicators in IMN patients with different 25(OH)D 3 levels. Methods: A total of 109 patients admitted to the Department of Nephrology of the First Aliated Hospital of Guangxi Medical University between August 2015 and April 2019 and diagnosed with IMN were reviewed. Comparative analysis of the effects of different 25(OH)D 3 levels on clinical and pathological indicators of IMN was performed, and changes in serum albumin, the estimated glomerular ltration rate (eGFR), serum creatinine and other clinical indicators in these patients during the follow-up were analysed. Results: The 109 patients with IMN had 25(OH)D 3 deciency [25(OH)D 3 <70 nmol/L]. The incidence of nephrotic syndrome, serum phosphorus levels, high-sensitivity C-reactive protein (hs-CRP) levels, and erythrocyte sedimentation rates (ESRs) were higher in the 25(OH)D 3 <25 nmol/L group than in the 25(OH)D 3 ≥ 25 nmol/L group. In contrast, total protein (TP), albumin (ALB), serum calcium, and IgG levels were lower in the 25(OH)D 3 <25 nmol/L group (P<0.05). No pathological parameters differed signicantly between the two groups (P>0.05). Bivariate correlation analysis showed that 25(OH)D 3 was positively correlated with nephrotic syndrome, TP levels, ALB levels, the eGFR, serum calcium levels and IgG levels (P<0.05) and negatively correlated with age, blood pressure, Scr levels, 24-hour urinary protein (UP) quantitation, serum phosphorus levels, hs-CRP levels and the ESR (P<0.05). In single-factor repeated measures analysis of variance, the ALB levels in the 25(OH)D 3 <25 nmol/L and 25(OH)D 3 ≥ 25 nmol/L groups showed an increasing trend but were lower in the 25(OH)D 3 <25 nmol/L group than in the 25(OH)D 3 ≥ 25 nmol/L group at each follow-up time point (P<0.001). Conclusion: Reduced 25(OH)D 3 levels are prevalent among IMN patients. During the follow-up, serum albumin increased signicantly in patients receiving vitamin D supplementation. 25(OH)D 3 may play an important role in IMN development. of studies showed that the incidence of low 25(OH)D 3 levels in patients with nephrotic syndrome exceeds 90% [13] . The results of our study showed that IMN patients generally have low 25(OH)D 3 levels (a detection rate of 100% and a rate of <25 nmol/L as high as 75.41%), with an average level of 12.27 (range: 6.79-24.91) nmol/L. This study retrospectively analysed the correlations of 25(OH)D 3 levels with clinical and pathological indicators in 109 patients diagnosed with IMN, and repeated measures analysis of variance was used to analyse changes in selected clinical indicators at 6 months after treatment initiation. inhibit the protein expression of desmin, and reduce podocyte damage [17] . Active vitamin D 3 regulates the Wnt/β-catenin pathway in podocytes and inhibits podocyte epithelial-mesenchymal transition (EMT). Active vitamin D 3 regulates TRPC6 expression to reduce podocyte damage, urinary protein excretion, etc. A high level of proteinuria also exacerbates the lack of vitamin D. Mechanistically, a high level of proteinuria leads to loss of the vitamin D binding protein [18] , and a decrease in the Megalin receptor content leads to the decrease in reabsorption of 25(OH)D 3 in renal proximal tubules [19] . In this study, 24-hour urinary protein quantitation was negatively correlated with the 25(OH)D 3 level, and the serum albumin level was positively correlated with the 25(OH)D 3 level. Serum albumin levels were signicantly different in patients with different levels of 25(OH)D 3 : in patients with 25(OH)D 3 <25 nmol/L, the serum albumin level was signicantly lower than that in patients with 25(OH)D 3 ≥ 25 nmol/L. However, the difference in 24-hour urinary protein quantication was not statistically signicant. This discrepancy may have occurred because of an incorrect methodology for retaining the 24-hour urine sample, leading to experimental errors, a large uctuation in the daily urinary protein quantitation of patients with MN (possibly differing according to the patient’s protein intake, body position and activity level) and other factors. The current study showed that among IMN patients, the hs-CRP level in the 25(OH)D 3 <25 nmol/L group was signicantly higher than that in the 25(OH)D 3 ≥ 25 nmol/L group. hs-CRP and 25(OH)D 3 levels were negatively correlated, indicating that the hs-CRP level was correlated with the D 3 level and may aggravate proteinuria through an inammatory reaction. HGF, and HGF can antagonize the TGF-β/Smad pathway through various mechanisms and protect against interstitial brosis [28, 29] . (6) Active vitamin D can also inhibit EMT, and EMT blockade can reduce renal interstitial brosis In this study, the difference in the proportion of renal interstitial brosis in pathological tissues from the patients in the two groups stratied by 25(OH)D 3 level was not statistically signicant, possibly because of an insucient sample size.


Background
Membranous nephropathy (MN) refers to a group of glomerular diseases with epithelial-side immune complex deposition and diffuse thickening of the glomerular basement membrane as pathological features. MN of unknown aetiology is called idiopathic membranous nephropathy (IMN). IMN is a common glomerular disease that causes nephrotic syndrome in the elderly population, accounting for approximately 25% of all cases of nephrotic syndrome, and is more common in men than in women (1.1~2.4:1) [1] . In most patients, the onset is insidious, and approximately 80% of patients have nephrotic syndrome, 20% to 25% have asymptomatic proteinuria, 30% to 50% have microscopic haematuria, and 20% to 40% have varying degrees of hypertension and renal function injury. In recent years, clinical studies have found that the incidence of IMN is increasing annually, that the age of onset is gradually becoming lower, and that the prognosis vastly differs. A 2010 multicentre study from Spain suggested that the spontaneous response rate of IMN patients was 31.7%, and that 29.4% of patients developed end-stage renal disease (ESRD) or died [2] .
Vitamin D 3 is a naturally occurring fat-soluble vitamin also called cholecalciferol. It is a ring-opened form of cholesterol and a steroid hormone. Two hydroxylation reactions are required for its biological activity as a hormone. First, the reaction producing 25-hydroxyvitamin D 3 [25(OH)D 3 ] from vitamin D 3 is catalysed in the liver by 25-hydroxylase, and 1α-hydroxylase then catalyses the production of 1,25dihydroxyvitamin D 3 from 25(OH)D 3 in the proximal tubular epithelial cells of the kidney. 25(OH)D 3 binds to nuclear receptors in target cells and produces effects by regulating gene expression. Vitamin D receptors are widely present in cells of human tissues, including podocytes in the kidney [3] . The kidney is not only an important location of active vitamin D 3 synthesis but also an important organ affecting the metabolism of active vitamin D 3 . The classic physiological role of vitamin D 3 is regulation of calcium and phosphorus metabolism. In addition, vitamin D 3 also has nonclassical physiological effects, such as regulating immune function, cell growth and differentiation.
In clinical work, we found that patients with IMN often have hypocalcaemia and low serum 25(OH)D 3 levels. This pattern suggests that the pathophysiology of IMN is related to active vitamin D 3 . However, current research on active vitamin D 3 and kidney diseases focuses mostly on chronic kidney disease and diabetic nephropathy.

Patients
We performed a retrospective analysis of 109 patients who were hospitalized in the Department of Nephrology of the First A liated Hospital of Guangxi Medical University between August 2015 and April 2019. All patients were diagnosed with MN by renal biopsy and pathology, and all had a follow-up time of longer than 6 months. We excluded patients with secondary MN caused by aetiologies such as infection (hepatitis virus, HIV, etc.), autoimmune disease (systemic lupus erythematosus, etc.), metabolic disease (diabetes, etc.), malignant tumours, drug or heavy metal exposure, and familial or other factors. In addition, patients with clear or suspected acute kidney injury, vitamin D supplementation before renal biopsy, and failure to follow the 2012 Kidney Disease Improving Global (KDIGO) guidelines were excluded. Of the patients, 67 were male (61.47%), and 42 were female (38.53%). The male:female ratio was 1.60:1, the age range was 15 to 72 (43.61 ± 14.21) years, and 87 patients (79.82%) were older than 30 years. Among the patients, 54 cases (49.54%) of nephrotic syndrome were documented.

Data collection
The fasting venous blood 25(OH)D 3 level, routine blood count, serum total protein level, albumin level, renal function, 24-hour urinary protein (UP) quanti cation, blood calcium and phosphorus levels, the high-sensitivity C-reactive protein (hs-CRP) level, the erythrocyte sedimentation rate (ESR), and the immunoglobulin G (IgG) level at admission were collected. The glomerular ltration rate (GFR) was calculated by the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula according to the patient's sex, age, and serum creatinine level.
Each patient underwent renal biopsy within 3 days after admission. The renal pathological results of all patients, including the Ehrenreich-Churg stage (divided into stages I, II, III, and IV; if two or more stages existed simultaneously, the highest stage was recorded), glomerular sclerosis ratio, and qualitative tubule lesion assessment results (renal interstitial brosis and renal tubular atrophy), were recorded.
According to the Kidney Disease Outcomes Quality Initiative (K/DOQI) guidelines [4] , a 25(OH)D 3 concentration ≥75 nmol/L indicates su ciency, a concentration <75 nmol/L and ≥25 nmol/L indicates insu ciency, and a concentration <25 nmol/L indicates de ciency. At admission, the 25(OH)D 3 level was <75 meg/L in all patients; thus, daily vitamin D supplementation (600 IU) was initiated in all patients on the third day after admission, and the patients were divided into two groups: the 25(OH)D 3 ≥25 nmol/L group and the 25(OH)D 3 <25 nmol/L group. All patients were treated with oral irbesartan. Glucocorticoid and immunosuppressive agents were added after the renal pathological results were reported.Follow-up data were collected from the medical records of the inpatients and outpatients.
Statistical methods SPSS 17.0 software was used for statistical analysis. The statistical results for normally distributed data are presented as x ± s values, and the t-test was used to compare the means of the two groups. Nonnormally distributed data are presented as M (P25, P75), and nonparametric tests were used for comparisons between the two groups. Repeated measures data were analysed by repeated measures analysis of variance; count data are expressed as the numbers of cases (percentages), and comparisons between groups were evaluated by c 2 tests. Correlations between two continuous variables were analysed by Pearson correlation analysis, and correlations between categorical variables or graded data were analysed by Spearman correlation analysis. P <0.05 was considered statistically signi cant.

Clinical characteristics of the IMN patients
All 109 patients had insu ciency or a lack of 25(OH)D 3 , although the degree of kidney disease was different. Fifty-four cases (49.54%) of nephrotic syndrome but 91 (83.48%) cases of hypoproteinaemia were identi ed; some patients exhibited a variable decrease in the estimated GFR (eGFR), an increase in serum creatinine, and dysregulation of calcium and phosphorus metabolism at onset.

Clinical characteristics of the IMN patients in different pathological stages
The renal pathological stages of the 109 patients ranged from 1-3, and patients were grouped according to the stage of MN. Because only 1 case had stage 3 IMN, this category was not included in the analysis. Except for the combination of hypertension, the difference in renal pathological stage was not statistically signi cant (Table 1). The proportion of patients with nephrotic syndrome, the blood phosphorus level, the hs-CRP level, and the ESR level were higher, but the total protein (TP) level, albumin (ALB) level, blood calcium level and IgG level were lower in the 25(OH)D 3 <25 nmol/L group than in the 25(OH)D 3 ≥25 nmol/L group (P <0.05). No statistically signi cant difference in pathological indicators was observed between the two groups ( Table 2).

Correlations between 25(OH)D 3 levels and clinical indicators in IMN patients
Correlation analysis showed that 25(OH)D 3 was correlated with multiple clinical indicators. Speci cally, the 25(OH)D 3 level was positively correlated with nephrotic syndrome, the TP level, the ALB level, the eGFR, the blood calcium level and the IgG level (all P <0.05) and negatively correlated with age, blood pressure, the serum creatinine level, 24-h UP quantitation, the blood phosphorus level, the hs-CRP level and the ESR (all P <0.05) ( Table 3).  Changes in the eGFR (Table 4, Figure 2) No statistically signi cant difference in the eGFR was found between the two groups (F group =-0.218, P group <0.828), and no interaction was observed between the grouping and the follow-up time (F interaction =1.258, P interaction =0.283).
Changes in serum creatinine (Table 4, Figure 3) No statistically signi cant difference in the serum creatinine level was noted between the two groups (F group =-0.441, P group =0.659), and no interaction was observed between the grouping and the follow-up time (F interaction =2.106, P interaction =0.074).

Discussion
In recent years, the incidence and prevalence of IMN have shown increasing trends. ZHU [5] found that the incidence of IMN among patients with primary glomerular diseases increased from 16.8% in 2003-2007 to 29.35% in 2008-2012. In China, Yu Xiaofang [6] also found that the detection rate of IMN in renal biopsy was 11.1%, demonstrating a signi cant increase from the previous rate. The pathogenesis of IMN is not yet clear. Newly discovered autoantigens, such as PLA2R [7] and THSD7A [8] , and complement system activation have been found to be involved in the pathogenesis of MN [9] , providing us with a better understanding of the causes of IMN. However, the effects of IMN treatment differ substantially, and many researchers are constantly seeking biological indicators for the disease status and prognosis of IMN.
Vitamin D is a steroid derivative that rst undergoes a reaction catalysed by 25-hydroxylase in the liver to produce 25(OH)D 3 and then undergoes a reaction catalysed by 1-α-hydroxylase in the kidney to produce active vitamin D 3 , which binds to intracellular vitamin D 3 receptors. Vitamin D exerts biological effects, participating in the regulation of calcium and phosphorus in the body and the differentiation, growth and function of immune cells [10] . Current research has shown that vitamin D 3 receptors are expressed in more than 30 cells, tissues and organs in humans [11] ; thus, active vitamin D 3 is associated with diseases of many body systems, such as the circulatory system and endocrine system. 25(OH)D 3 that enters the blood from the liver is a precursor of active vitamin D 3 . Because of its high concentration in blood, good stability and long half-life, 25(OH)D 3 can accurately re ect the nutritional vitamin D status in the body and is thus generally used as a clinical indicator for evaluating vitamin D su ciency [12] . In clinical work, we found that not only patients with chronic kidney disease but also patients with nephrotic syndrome often have low 25(OH)D 3 levels. Earlier studies showed that the incidence of low 25(OH)D 3 levels in patients with nephrotic syndrome exceeds 90% [13] . The results of our study showed that IMN patients generally have low 25 A total of 109 patients with IMN were enrolled in this study, with a male:female ratio of 1.60:1. Of these patients, 79.82% were older than 30 years, and approximately 41.28% had hypertension, similar to the percentages in other domestic and international reports. In addition, 49.54% of the patients had nephrotic syndrome, which is lower than the rates reported in Japan and Western countries [14] ,but is similar to some domestic reports [1] . Pathological analysis of the kidneys showed that glomerular lesions were mainly stage I and II. Grouping the patients by IMN stage and comparing differences in related clinical indicators showed that patients with stage II IMN had a higher risk of developing hypertension, which is consistent with the results of previous studies [2] . Grouping patients by 25(OH)D 3

level revealed signi cant differences in the levels of serum albumin, blood lipids, CRP and other indicators. Correlation analysis between the 25(OH)D 3 level and clinical indicators
showed that 25(OH)D 3 was correlated with age, hypertension, nephrotic syndrome, 24-hour urinary protein, serum albumin levels, blood lipid levels, CRP levels and other indicators.
High levels of proteinuria and hypoproteinaemia are necessary for the diagnosis of nephrotic syndrome. A high level of proteinuria is also the main predictor chronic kidney disease progression. Low serum albumin is an independent risk factor for a poor response to immunosuppressive therapy in IMN patients [6] . Many studies have shown that a high level of proteinuria is associated with vitamin D de ciency [15,16] .Vitamin D de ciency aggravates urinary protein excretion and causes a further decrease in serum albumin. A high level of proteinuria in IMN patients is related to podocyte damage. Active vitamin D 3 can protect podocytes through a variety of mechanisms as follows. Active vitamin D 3 can inhibit the renin-angiotensin system (RAS), negatively regulate the expression of angiotensin II and renin, and reduce podocyte shedding and damage. Active vitamin D 3 can inhibit TGF-β/Smad pathway signalling, activate the BMP-7/Smad pathway, inhibit the protein expression of desmin, and reduce podocyte damage [17] . Active vitamin D 3 regulates the Wnt/β-catenin pathway in podocytes and inhibits podocyte epithelial-mesenchymal transition (EMT). Active vitamin D 3 regulates TRPC6 expression to reduce podocyte damage, urinary protein excretion, etc. A high level of proteinuria also exacerbates the lack of vitamin D. Mechanistically, a high level of proteinuria leads to loss of the vitamin D binding protein [18] , and a decrease in the Megalin receptor content leads to the decrease in reabsorption of 25(OH)D 3 in renal proximal tubules [19] . In this study, 24-hour urinary protein quantitation was negatively correlated with the reactive protein is an acute-phase protein synthesized by the liver under stimulation by in ammatory factors such as IL-6. hs-CRP is an extremely sensitive in ammatorymarker. When the body experiences immune injury or infection, the hs-CRP level in the blood increases.
Chronic in ammation of the kidney is an important factor in the progression of renal impairment in CKD patients. Recent studies have shown that vitamin D de ciency can also lead to increased CRP levels, and that supplementation with paricalcitol can reduce urinary protein excretion and downregulate hs-CRP [20,21] . The current study showed that among IMN patients, the hs-CRP level in the 25(OH)D 3 <25 nmol/L group was signi cantly higher than that in the 25(OH)D 3 ≥25 nmol/L group. hs-CRP and 25(OH)D 3 levels were negatively correlated, indicating that the hs-CRP level was correlated with the vitamin D 3 level and may aggravate proteinuria through an in ammatory reaction.
In terms of pathological changes, the current study indicates that the stage of MN is not directly related to the prognosis of the disease [22] , and that renal interstitial brosis and renal tubular atrophy are in uencing factors for the prognosis of MN [23,24] . In this study, no statistically signi cant differences were observed in the clinical indicators (including 24-hour urinary protein, total blood cholesterol, and other indicators affecting the prognosis of MN) according to the stage of MN. As the 25(OH)D 3 level decreased, the proportion of pathological manifestations, such as glomerular sclerosis, renal tubular atrophy, and renal interstitial brosis, increased, but the difference was not statistically signi cant. In recent years, many studies have indicated that 25(OH)D 3 can exert renoprotective effects by antagonizing renal brosis. The main mechanism may act as follows: (1) Active vitamin D 3 may delay or prevent TGF-β-mediated tubular epithelial-myo broblast transdifferentiation (TEMT) by inhibiting TGF-β expression, thereby delaying renal tubular interstitial brosis [25] . (2) Active vitamin D achieves anti-in ammatory effects by inhibiting the pathway initiated by the transcription factor NF-κB, thereby preventing renal tubular interstitial brosis [26] . (3) Active vitamin D can inhibit an overactivated RAS, delay the progression of chronic kidney disease, and reduce renal tubular interstitial brosis [27] . (4) Active vitamin D can reduce urinary protein excretion, decrease renal interstitial in ammation, and thus reduce renal tubular interstitial brosis. (5) Active vitamin D can induce the expression of HGF, and HGF can antagonize the TGF-β/Smad pathway through various mechanisms and protect against interstitial brosis [28,29] . (6) Active vitamin D can also inhibit EMT, and EMT blockade can reduce renal interstitial brosis [30] . In this study, the difference in the proportion of renal interstitial brosis in pathological tissues from the patients in the two groups strati ed by 25(OH)D 3 level was not statistically signi cant, possibly because of an insu cient sample size.
Therefore, we speculate that 25(OH)D 3 may be related to the e cacy of IMN treatment. In this study, by comparing the serum albumin levels, serum creatinine levels, and eGFRs between two groups of patients with different baseline levels of 25(OH)D 3 across a 6-month treatment course, we found that during the follow-up, the serum albumin level gradually increased in both groups but was lower in the 25(OH)D 3 <25 nmol/L group than in the 25(OH)D 3 ≥25 nmol/L group, indicating that the overall disease severity in the patients in the 25(OH)D 3 ≥25 nmol/L group was milder than that in the patients in the 25(OH)D 3 <25 nmol/L group. However, no statistically signi cant difference in the serum creatinine level or the eGFR was observed between the two groups, and no grouping by time interaction effect was noted, possibly because MN is characterized by a long natural course, but the follow-up time of this study was short; thus, the serum creatinine levels and eGFRs in the follow-up data are insu cient to re ect renal function progress. If we extend the follow-up time to obtain more data, the results may change signi cantly.
In summary, vitamin D de ciency is common in IMN patients, and approximately 75. The study protocol was approved by the Ethics Committee of the First A liated Hospital of Guangxi Medical University (Approval Number: 2019KY-E-067). The data are anonymous, and the requirement for informed consent was therefore waived.

Consent for publication
Consent for publication was not required.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Competing interests
None of the authors have any con icts of interests to declare.

Funding
Not applicable.
Authors' contributions