Background
IgAN is one of the most common glomerulonephritis in Asian, accounting for 45.26% of primary glomerular diseases in China, and is also a leading cause of ESRD. Most of the IgAN patients showed good outcomes, but about 30% of them progressed to ESRD within 10–20 years, while in some patients the disease developed rapidly to ESRD within ten years. Clinical and pathological markers, such as proteinuria, hypertension, decreased eGFR at the time of biopsy and the MEST score were found to be associated with renal outcome in previous studies [
1,
2]. It was widely accepted that clinical and pathological markers above were significant factors for IgAN development. Compared with numerous studies on MEST score and clinical data at biopsy, follow-up features, except for urine protein, were paid little attention. Even then those follow-up data were still of great importance [
3,
4]. The main purpose of this study was to assess the prognostic value of these follow-up clinical data, including TA-P, TA-UA, TA-Alb, TA-Hb and TA-TC, on the progression of progressive IgAN and also to reassess the predictive value of MEST score and clinical features at biopsy. This could be practical to the full stage guidance of IgAN treatment.
Discussion
IgAN is the leading cause of adult primary glomerulonephritis progressing to ESRD [
8]. However, there is still a lack of specific therapeutic interventions, because of the unclearness of its etiology and pathogenesis, so early identification of risk factors and timely treatment may be essential to the renal outcome. Progressive IgAN, which progress to ESRD within ten years, only accounted for a small part of IgAN population and normally were studied along with patients under stable condition, for instance, remaining normal renal function and slowly developing sCr, by retrospective cohort study [
6,
9], which means some characteristics of progressive IgAN may be covered by the stable one. In this study, as the main object, progressive IgAN patients were compared with patients under stable renal conditions to identify the risk factors of poor prognosis, aiming to provide a promising guidance for the further treatment.
Impaired renal function, consistent proteinuria with more than 1 g per day and Oxford pathological score T1/T2 were recognized as independent risk factors for IgAN in a couple of previous studies [
5,
7]. However, a great number of those founding was based on clinical and pathological data at the time of biopsy, with the absence of follow-up data. In this study, those factors were further considered with the present of follow-up features. In statistical practice (Additional file
1), variables were put into the logistic regression mode one by one to dig underlying information between them. When only MEST scores were included in the multivariate logistic analysis (Additional file
1, mode 1), M1, T1 and T2 were adverse prognostic factors to ESRD, but the OR value of them decreased and even showed no significance with the entrance of eGFR at biopsy (Additional file
1, mode 2). This could be explained by the strong correlation between MEST score and eGFR (Additional file
2), indicating eGFR may be a better predictor for renal outcome than T2 in severe IgAN. This situation maintained until the participant of demographic and baseline laboratorial data, including gender, age, UA, Hb, Alb, TC and 24 h urinary protein (Additional file
1, mode 8), in which score T and eGFR lost its position and gender (male) appeared to be a strong risk factor to unfavorable prognosis. The complex cross-relation among score T, gender, baseline eGFR, UA, Hb and Alb may contribute to this result (Additional file
2). In addition, there are innate difference between males and females in the normal range of UA and Hb. The situation changed again when follow-up clinical data were enrolled (Additional file
1, mode 11). Then eGFR came back to the predictor group instead of gender. This mode showed the relevance of follow-up clinical features more than those at biopsy. TA-P, though not that credible, showed a significant impact on renal outcome in mode 15 (Additional file
1). In general, stepwise logistic analysis illustrated that correlations between clinical data were inevitable and results of statistical analysis can be various with different variates in the analysis mode, and multivariate analysis was a proper way to distinct the strongest factors.
In previous studies, Oxford pathological score T was highly recognized as an important predictor of developing ESRD [
9,
10], while roles M, E and S played were controversy. The Oxford score of 1026 primary IgAN patients from 18 clinical research centers in China were compared with the data in the original IgAN Oxford classification study [
11]. Then a higher ratio of M1 and E1 and a lower ratio of S1 were observed in Chinese population, while the proportion of T1/T2 between them was similar. In this study, M1, T1 and T2 were identified as risk factors when only Oxford score were included in the logistic regression (Additional file
1, mode 1), but only M1 remained as an independent factor in multivariate analysis. Although T did not maintain a marked role in multivariate analysis, it was also of great importance because of its correlation with most variates in this study (Additional file
2). Combined with the results, we came up with a hypothesis that pathological score M1 might be a risk factor for progressing to ESRD in Chinese IgAN population.
There was an increasing attention on the relationship between hyperuricemia [
12] and chronic kidney disease [
13]. High concentration of serum UA may lead to nephropathy and renal function impairment by directly mechanical damage or indirectly inflammation and cell phenotype inversion, resulting in the reduction of UA excretion and forming a vicious cycle. Normal ranges of Hb and Alb were also reported beneficial to stable internal environment and normal physiological activities [
14,
15]. Theories above could provide our result with support that serum UA and Hb levels during the follow-up can significantly affect the progress of IgAN to ESRD. Although TA-P was excluded from the final multivariate analysis for the accuracy of the mode, the importance of it cannot simply be ignored (Additional file
1, mode 15) and it had also long been recognized as a strong and reliable factor to renal outcome [
16,
17].
Our research did not only come up with a similar result with others that M1, T1/T2 and eGFR at biopsy were strongly related to IgAN outcome, but also first assess the role TA-UA, TA-Hb, TA-Alb and TA-TC played in IgAN progressing and come up with a clinically significant result that TA-UA and TA-Hb were independent factors for developing ESRD. One of the limitation of this study was its retrospective nature, which means the data we collected was restricted. There was also a selection bias: patients enrolled were those who followed up in our hospital for a long term, so they might not be a fully presentative sample of the general population. A long-term prospective study or a clinical trial may be better for the study of IgAN treatment.
Acknowledgements
The authors would like to thank colleagues in the department of nephrology, the First Affiliated Hospital of Wenzhou Medical University, for their great support and selfless help during this study.