This study showed the distribution of demographic, clinical and laboratory data and histopathological diagnoses of 1051 patients who underwent renal biopsies in the years 2000 to 2018 at our center. The histological diagnosis showed that secondary glomerulopathies represented more than half of the sample (52.4%) followed by primary glomerulonephritis (29.6%), and in sequence other pathologies, tubulo-interstitial disease and hereditary nephropathies.
Nephrotic syndrome was prevalent in clinical presentation and indicated renal biopsy mainly in primary forms, but also in some secondary glomerulopathies. Studies have shown a higher frequency of nephrotic syndrome in different glomerulopathies, mainly MCD, FSGS and MN2,6,12,21,22 data also described in systematic reviews17, and mainly as a clinical presentation of primary glomerulopathies16,18. In IgAN, nephritic syndrome and proteinuria/hematuria prevailed, in agreement with other studies6,18,19,21. In secondary forms, nephrotic syndrome and proteinuria/hematuria were prevalent in LN, nephrotic syndrome in DKD, nephrotic syndrome and proteinuria with loss of renal function in 2nd FSGS and monoclonal gammopathies, and nephritic syndrome in systemic vasculitis. These findings are in disagreement with those described in some studies6,12,14,19,23, what can be associated with different clinical and laboratory criteria for indicating renal biopsy between countries or states of a country, in which the prevalence of primary and secondary glomerulopathies has a significant variability.
In primary glomerulonephritis, FSGS was the prevalent histology, followed by IgAN, MN, MCD and MPGN, with frequencies similar to those described in Brazilian registries12–14,24 and in North and South American countries4,5,16,25. In the study by Polito et al12, the prevalence of FSGS, MN, IgAN, MCD and MPGN were 24.6%, 20.7%, 20.1%, 15.5% and 4.2%, respectively, which was similar to that described by Malafronte et al13 who reported frequencies of 29.7%, 20.7%, 17.8%, 9.1% and 7% for the same pathologies. Sim et al16 described a higher frequency of FSGS, followed by MN, MCD and IgAN, with an increase in FSGS and IgAN over 12 years in the follow-up. Barrera-Herrera et al25 also reported a higher prevalence of FSGS and IgAN in the spectrum of primary glomerular diseases.
IgA nephropathy has a distinct epidemiological profile in different countries and continents. In Asia, its predominance is notorious, as described in several studies8,9,18,19. Yang et al8 reported a higher prevalence of IgAN followed by MPGN, MN and MCD, with FSGS described in only 4.6% of cases. In South Korea9 the data are similar with IgAN in 28.3%, and in Taiwan Chiu et al18 also reported this highest prevalence (26%). In Japan26 IgAN predominated as well, clinically manifested by chronic nephritic syndrome. In Europe, IgAN is also described as prevalent in some countries, such as Italy27, Czech Republic11 and Poland10.
Lupus nephritis was the prevalent etiology in secondary glomerulopathies, as shown in Fig. 3 and Fig. 6, representing 41.1% of the cases, followed by DKD and systemic vasculitis. Analyzing the histological class of NL, classes III and IV (63%) and class V (17.8%) predominated, in a total of 80.8% of cases. The prevalence of LN in our study did not differ from publications of other countries5,11,19,28. In Brazil, Polito et al12 analyzed 9617 renal biopsies and LN was found in 42.3%, followed by post-infectious glomerulonephritis (20.4%) and DKD (10.1%). In the publication by Malafonte et al13, LN was reported in 66.2% in secondary forms, with post-infectious glomerulonephritis (12.5%) and DKD (6.2%) in sequence. These data reflect the higher frequency of LN in several countries and also in Brazil, showing similarity with our findings concerning the prevalence of secondary glomerulopathies.
Diabetic kidney disease was the second pathology in terms of frequency in secondary glomerulopathies. Usually, renal biopsy was indicated in patients with type 2 diabetes mellitus in the presence of atypical conditions to rule out a non-diabetic kidney disease, thus defining another pathology with a potential indication of immunosuppression. These criteria included abrupt nephrotic proteinuria, acute loss of renal function not compatible with the evolution of diabetes, short-term diabetes (less than 5 years), absence of diabetic retinopathy, and/or signs and symptoms of a systemic disease. Fiorentino et al29 in a pooled meta-analysis assessing 48 studies who analysed renal biopsies in patients with diabetes, reported that the prevalence of DKD, non-diabetic renal disease and mixed forms ranged from 6.5 to 94%, 3 to 82.9%, and 4 to 45.5% of the overall diagnosis, respectively. The most frequent non-diabetic nephropathy was IgAN, with a variation between 3% and 59% followed by MN (7–35%), FSGS (17-37.7%) and ATIN (18-48.8%).
In a recent study, O'Shaughnessy et al5 evaluated 29 centers involving 42,603 patients, showing a higher prevalence of FSGS and DKD in the USA/ Canada, IgAN and FSGS predominated in Europe, in Asia IgAN and LN, and in Latin America LN (38.1%) and FSGS (15.8%). Our results showed similarity with Latin America for LN (41.1%), but the prevalence of FSGS was much higher in our sample (37.3%). Considering the temporal variation that we found over three periods, DKD is in agreement with the findings of the USA/Canada, and IgAN with those described in the European and Asian continents.
Demographic data showed a predominance of caucasians (85.5%) over afrodescendants (14.1%), which was attributed to the ethnic profile of the southern region of Brazil. It is well know that there are differences in race in different regions of Brazil, and also worldwide. As an example, O'Shaughnessy et al5 reported the frequency of caucasians and afrodescendants in the USA/ Canada (54.2% and 31.4%) and in Europe (98.2% and 0.8%), pointing to variations in countries on different continents. The number of centers involved in these registrations was wide, with 10 centers in the USA/Canada and 14 centers in Europe. However, data on ethnicity in Asia and Latin America were more restricted, with only 2 centers in Asia and 3 centers in Latin America, which limits comparisons in these regions.
The prevalence of each glomerulopathy varies by age group in different countries6,15,18,22. The frequencies described in these studies for primary glomerulonephritis are IgAN and FSGS between 18–50 years, and FSGS and MN between 51–65 and > 65 years. In the secondary forms LN was prevalent between 18–35 years, DKD and 2nd FSGS between 36–60 years, and systemic vasculitis and monoclonal gammopathies over 65 years of age. These prevalences are compatible with those described in our study, for all age groups and for each age group analyzed apart.
The temporal variation of glomerulopathies is widely explored in many studies, usually assessed in the last four decades. Over three time periods, we observed a significant reduction in FSGS and an increase in IgAN in the primary forms, and in the secondary, there was a reduction in LN and an increase in DKD, with a minor and non-significant variation in the other glomerulopathies. Studies have shown different temporal trends in both primary and secondary glomerulopathies. The data described by Sim et al16 between 2000–2011 showed an increase in the frequency of FSGS and IgAN, and minor temporal variations with stabilization in pauci-immune vasculitis, MPGN, MCD and MN. In the study by Garau et al6, an increase in IgAN, MN, LN and DKD was observed between 1990–2014, with a decrease in FSGS after 2004. Brkovic et al30 comparing the periods 1987–2006 and 2007–2014 reported an increase in MN and reduction in non-IgA mesangioproliferative nephropathy, with no change in other primary glomerulopathies. Dos-Santos et al31 showed an increase in the prevalence of FSGS between 1975–2006, and of LN and MN between 2006–2015, with a tendency for FSGS to decrease in this last period. It is likely that these temporal variations correlate with the heterogeneity of biopsy indication in different centers, changes in the population profile (for example, increasing age and ethnic variations), changes in lifestyle and socioeconomic diversity, influencing the epidemiological status of glomerulopathies and their variation over the years.
The degree of chronicity in the histopathology also has a significant prognostic value. Adjusting for demographic, laboratory data and clinicopathological diagnosis, Srivastava et al32 showed a higher risk of kidney disease progression in association with inflammation in the non-fibrous interstitium, moderate to severe tubular atrophy, interstitial fibrosis and global glomerulosclerosis, and also with arterial and arteriolar sclerosis. In our study, we did not perform this analysis based on the percentage of IFTA, global glomerulosclerosis or arteriolar damage described in renal histologies, something to be explored in a future study.
Our study has limitations. As this is a retrospective study, the review of medical records in some patients showed incomplete demographic, clinical and laboratory data. It is a single center study, so it may not represent the real prevalence of other centers. In addition, our center performs kidney biopsy for patients referred from other centers, limiting the access to information in some cases, such as the clinical syndrome that indicated the biopsy and the clinical and laboratory data. However, of the 1051 patients included in this registry, the majority had accessible data in our center medical records, or by data reported by physicians of external centers. In addition, all biopsies were interpreted and reported by experienced nephropathologists, allowing an etiological diagnosis that defined the type of glomerulopathy, or its non-representativeness.