AimsDiagnosis of primary membranous nephropathy (PMN) is mainly based on immunofluorescence/immunohistochemistry findings. However, assessment of specific features on optical microscopy can help to estimate the severity of the disease, guide treatment and predict the response. The aim of this study was to identify, classify and grade the precise histological findings in PMN to predict renal function outcome and guide treatment.Methods and resultsHistological parameters, including focal segmental sclerosis (FSGS), tubular atrophy (TA), interstitial fibrosis (IF) and vascular hyalinosis (VH), were re‐evaluated in 752 patients with PMN. Their predictive value was estimated separately, and also in a combination score (FSTIV) graded from 0 to 4. Finally, the impact of histology was assessed in the response to immunosuppressive treatment. Mean age of patients was 53.3 (15–85) years and most presented with nephrotic syndrome. FSGS was present in 32% and VH in 51% of the patients, while TA and IF were graded as stage ≥1 in 52% and 51.4%, respectively. The follow‐up period was 122.3 (112–376) months. FSGS, TA and IF and VH were associated with impaired renal function at diagnosis (P = 0.02, P < 0.0001, P = 0.001 and P = 0.02, respectively) and at the end of follow‐up (P = 0.004, P < 0.0001, P < 0.0001 and P = 0.04, respectively). In multiple regression and binary logistic analysis, the presence of FSGS and degree of TA were the most significant parameters predicting renal function outcome, defined either by eGFR (end), FSGS (r = 0.6, P < 0.0001) and TA (r = 0.6, P < 0.0001), or by the endpoint of >50% eGFR reduction, FSGS (P = 0.001) and TA (P = 0.02). Also, patients presented with FSGS, IF, VH and/or with FSTIV > 1 could benefit from immunosuppression, regardless of clinical presentation.ConclusionsThe presence and degree of four histological indices, FSGS, VH, TA and IF, assessed separately or in combination, and FSTIV score not only predict renal function outcome after long‐term follow‐up, but can also help in the choice of appropriate treatment. Decisions concerning immunosuppressive treatment can be guided by pathology regardless of clinical findings.