In this retrospective study in patients with LN, we have designed a prognostic index for evolution of renal function in patients with lupus nephritis. The main predictors of poor renal evolution were thrombocytopenia and interstitial fibrosis. Our findings highlight the value of thrombocytopenia and histology to determine renal survival in patients with LN.
In our study, older patients (44,4 ± 19,1 years) had worst evolution renal function compared with younger patients (32,2 ± 13,9 years). Kang et al (13), found similar results, in 117 patients with LN followed during followed-up during a mean of 76.5 months. They divided them into three groups based on aged: juvenile LN (JLN) if < 8 years old, adult LN (ALN) between 18–50 years old and late-onset LN (LLN) if > 50 years old. The study findings showed that the patients with LLN had a higher chronicity index, developed CKD and death higher than JLN and ALN patients.
Several studies have shown that tubular atrophy and interstitial fibrosis were an independent factors for poor renal evolution (14–21) as well as the chronicity index (22). Tang et al (23), have developed and validated a risk score for the development of ESRD in LN, emphasising the importance of tubulointerstitial lesions (tubular atrophy and interstitial fibrosis) than the histological subtype according to the ISN/RPS classification (24). These renal histopathological changes will be considered a chronic change and loss of function of the nephrons and therefore they are related to the poor renal evolution. Ayoub et al (5), have developed a prediction model of treatment in LN, showing that early detection and treatment of NL was essential to achieve good long-term renal outcomes. In this predictive model they have used classical biomarkers (proteinuria, renal glomerular filtration rate) and new urinary biomarkers (cytokines, chemokines). This study has showed that predictive value the proteinuria in LN is complicated because proteinuria may represent acute kidney injury due to inflammation and podocyte dysfunction, or chronic kidney injury due to scarring after inflammation. However, clinical and demographic variables were relatively more important than any novel urine biomarker.
A recent systematic review by our group on the main prognostic factors in the outcome of CKD has shown that the classical biomarkers (proteinuria, GFR and urinary sediment) remain despite advances in the diagnosis and treatment of lupus nephritis (6). One of the main limitations of clinical trials in LN has been considering renal function and proteinuria as the only criteria for assessing response to treatment. However, the concept of a histopathological target emerged from observations that clinical outcome based on proteinuria and/or urinalysis and histopathological outcome based on repeat kidney biopsies are discordant. Recent studies have shown that an activity and chronicity index > 3 correlates with a higher incidence of relapse and CKD, respectively, in lupus nephritis (25). The nuances of histological lesions have become a cornerstone of the evolution of renal function. Several publications have shown that chronic damage in the tubulointerstitial compartment and different kinds of vascular lesions contributed significantly to the association with poor long-term renal function (19–21,26,27). Korbet et al (16), have showed a significant association between the evidence of irreversible kidney damage (renal sclerosis, tubular atrophy, or interstitial fibrosis) with the negative impact on achieving remission.
We recently showed that histological findings in repeat kidney biopsies of LN patients commonly present discordance in relation to clinical expression. At repeat biopsy, chronicity index was more influential over CKD progression than the shift to lower pathological classes. Histological data from repeat kidney biopsies in LN could be useful to guide therapeutic approach (27). For this reason, prospective randomised studies such as "Per-protocol repeat kidney biopsy in incident cases of LN" should shed some more light on the possibility of changing the course of lupus nephritis.
In the present study, the presence of thrombocytopenia below 50,000 cells/mm3 has been identified as an important risk factor for the progression of renal damage. The finding of thrombocytopaenia was not in the context of a manifestation associated with thrombotic microangiopathy (TMA), but as a more severe extrarenal systemic manifestation of systemic lupus erythematosus. The presence of thrombocytopenia at the debut of SLE should alert us to a worse evolution of patients with lupus nephritis and therefore, we should try to be more forceful in our immunosuppressive treatment. Haematological abnormalities, especially thrombocytopenia, are highly prevalent among patients with systemic lupus erythematosus and at the same time it has been reported as a significant prognostic factor of SLE course (28). Several studies have shown that the significance platelet count has a negative correlation with disease activity in SLE patients (arthritis, neurologic manifestations, and nephritis), whatever the associated manifestations, and it should be considered as a prognostic factor, identifying patients with aggressive disease course (28–30).
This study is subject to limitations. Due to the small sample size and its single-centre retrospective nature. However, strengths include that it is a real-world experience in standard clinical practice and a long follow-up time, giving homogeneity to our histological results. Our predictive model, show good discrimination capacity with area under the curve close to 0.9.
Our study suggests that this prognosis index may be useful in clinical practice to detect which patients with lupus nephritis may have a worse renal prognosis and to modify our therapeutic approach to preserve kidney function. In order to stratify patients into different risk grades, future research is needed for internal and external validation with another cohort of patients.
In conclusion, we have developed a prognostic index of poor renal evolution in patients with LN that combines demographic, clinical, analytical and histopathological factors, easy to use in routine clinical practice and that could be an effective tool in the early detection and management of patients.