Abstract
Background and Aims
Membranous nephropathy (MN) is the most common cause of nephrotic syndrome in older white adults, with an incidence of 12 cases per millions of people per year. Primary MN (PMN, 75%-80% of MNs) is an organ-specific autoimmune disease caused by antibodies anti-PLA2R and anti-THSD7A. Regardless of treatment one third of patients progresses to end-stage renal disease and two third develop non-progressive chronic kidney disease. Renal biopsy is the gold standard for MN diagnosis. Several clinical and biochemical markers have been associated with the risk of progressive loss of kidney function while contrasting results have been obtained by the few studies which have examined the prognostic value of histologic findings.
In this study the clinical outcome of patients with PMN has been considered based on the prognostic value of histological findings.
Method
Forty-nine patients with PMN of our Nephrology Unit at Padova University Hospital from 2003 to 2018 were considered. 16 patients were excluded from the study due to missing data. Age, comorbidities, proteinuria (g/day) and renal function (eGFR, CKD-EPI) were collected. eGFR decline and decrease of proteinuria were used as clinical outcomes. The follow-up was considered from renal biopsy to the last visit (in absence of GFR decline or decrease in proteinuria).
Histological grading (0-3) was assigned to parameters (glomerulosclerosis (GS), tubular atrophy (TA), interstitial fibrosis (IF), vascular hyalinosis (VH)) and were evaluated separately or in combination (as GSTIV score). Morphometric analysis was used to quantify IF and expressed in percentage as the mean of area covered by pixel. Statistical analysis was performed using Fisher’s exact test and Mann-Whitney U-test where appropriate. Cox regression analyses (univariate and multivariate) were performed to identify variables associated with both renal outcomes and p<0.05 was considered as significant.
ROC curves were used to determine interstitial fibrosis cut-off values predictive for both outcomes. Area under the curve (AUC) between 0.8 and 1.0 was considered as significant. Diagnostic accuracy was assessed by Specificity (Sp), Sensibility (Se), positive (PPV) and negative (NPV) predictive values and positive (LR+) and negative (LR-) likelihood ratios.
Results
Patients with no decrease of proteinuria had a greater degree of IF vs those with a full response (p=0.006).
Univariate Cox analyses identified age ≥65 years (HR 4.92), pre-existing CKD (HR 12.98) and IF (HR3.05) as significant predictors of renal function decline in all patients. Multivariate Cox analysis confirmed these variables (age ≥65 years HR 3.05, CKD HR 6.35, IF HR 3.03). In patients without CKD only IF was significantly associated with eGFR decline in both Cox univariate and multivariate analysis (HR 4.34 and 5.05 respectively).
ROC analysis showed that IF threshold of 17.80% identified patients with eGFR reduction (AUC 0.65, Se 0.50, Sp 0.79, PPV 0.75, NPV 0.45, LR+ 2.38, LR-0.63) and IF threshold of 18.04% the lack of proteinuria reduction (AUC 0.78, Se 0.70, Sp 0.83, PPV 0.67, NPV 0.80, LR+ 4.12, LR- 0.36).
Conclusion
Our study shows that IF could be used as a histologic predictor of renal and proteinuria outcomes. Biopsy report should therefore also include quantitative IF data that could be helpful for the choice of a more appropriate therapeutic approach.
Experimental membranous nephropathy redux
