Clinical and pathological characteristics among patients with or without FN lesions
A total of 665 primary IgAN patients were enrolled in this study, and 233 of them had FN lesions. Detailed clinical and pathological characteristics of IgAN patients with or without FN lesions were summarized in Table 1. Patients with FN lesions suffered more severe hematuria than those without. In addition, patients with FN lesions showed lower levels of Hb and serum Alb (126.9 ± 19.1 g/L vs 132.8 ± 19.7 g/L, P < 0.001; 36.7 ± 4.9 g/L vs 37.6 ± 5.7 g/L, P = 0.040). While, there was no significant difference in baseline renal function and proteinuria between two groups.
On the hand of pathological characteristic, patients with FN lesions showed higher proportions of M1, E1, C1/C2 and T1/T2 lesions compared with those without FN lesions (all P < 0.001).
The median length of follow-up was 44 months in the whole participants. The rate of renal function decline seems to be more slowly in FN1 groups than FN0 (0.6 [-2.8,5.2] ml/min/1.73m^2/year vs. -0.4 [-3.4,2.2] ml/min/1.73m^2/year, P = 0.002). The percentage of patients suffered from ESKD or composite endpoint showed no statistically significant in patients with or without FN lesions.
Table 1 Clinical and pathological characteristics of patients with or without FN lesions in IgAN.
Abbreviations: FN fibrinoid necrosis lesions, FN0 patients without fibrinoid necrosis lesions, FN1 patients with fibrinoid necrosis lesions, MAP mean arterial pressure, Scr serum creatinine, eGFR estimated glomerular filtration rate, UA serum uric acid, MEST-C MEST-C scores by Oxford classification, ESRD, end stage renal disease, defined by eGFR < 15 ml/min/1.73m^2, Composite endpoint, defined by either a 30% reduction in eGFR, a doubling of the base-line serum creatinine, ESRD, chronic dialysis for at least 6 months or renal transplantation, RASB renin-angiotensin system blocker, IS immunosuppressive agents.
Clinical and pathological characteristics among IgAN patients with different fractions of FN lesions
Due to the differences in clinical and pathological characteristics between IgAN patients with or without FN lesions, we next tried to find the impact of different fractions of FN lesions on baseline characteristics. Firstly, we found there was less correlation between the number of glomeruli and number of FN lesions (r = 0.017, P = 0.659, Fig. 2). Furthermore, the distribution of the percentage of glomeruli with FN lesions was shown in Figure 3. Of the 233 patients with any faction of FN lesions, 74% had FN lesions less than 10%. We then divided the patients with FN lesions into three groups depending on the distribution. With the increase of the fraction with FN lesions, patients showed significantly decrease in eGFR (99.8 [76.0,121.4] ml/min/1.73m^2 vs. 93.8 [66.0,111.5] ml/min/1.73m^2 vs. 84.8 [60.7,109.8] ml/min/1.73m^2, P = 0.044), obviously increase in excretion in proteinuria (1100.0 [647.5,2104.5] mg/24h vs. 1280.0 [888.0,1880.0] mg/24h vs. 1690.5 [1192.5,3539.8] mg/24h, P = 0.004, Table 2) and higher level of MAP (96.1 ± 10.6 mmHg vs. 96.5 ± 10.8 mmHg vs 101.2 ± 13.5 mmHg). Unlikely to baseline clinical characteristics, patients showed less difference in pathological changes with the increase of fraction of FN lesions. There was no difference in renal outcomes between two groups. We found patients in the group with higher fraction of FN lesions were more likely to get immunosuppressive treatment (86.5% vs. 78.8% vs. 95.3%, P = 0.018).
Abbreviations: FN fibrinoid necrosis lesions, FN0 patients without fibrinoid necrosis lesions, FN1 patients with fibrinoid necrosis lesions, MAP mean arterial pressure, Scr serum creatinine, eGFR estimated glomerular filtration rate, UA serum uric acid, MEST-C MEST-C scores by Oxford classification, ESRD, end stage renal disease, defined by eGFR < 15 ml/min/1.73m^2, Combined endpoint, defined by either a 30% reduction in eGFR, a doubling of the base-line serum creatinine, ESRD, chronic dialysis for at least 6 months or renal transplantation, RASB renin-angiotensin system blocker, IS immunosuppressive agents.
IS treatment in IgAN patients with or without FN lesions.
As shown in Table 1, patients with FN lesions were more likely to receive IS treatment (FN1 vs FN0, 86.3% vs 59.5%, P < 0.001). We further evaluated the subsequent use of IS agents in different fraction of FN lesions. 86.3% of patients with any fraction of FN lesions received IS treatment compared with 59.5% of patients without FN lesions. Stepwise examination of increasing fractions of glomeruli with FN lesions revealed a steady increase in the percentage of patients treated with immunosuppression agents (Fig.4).
Kidney outcome and adjusted predictive value of FN lesions.
A total of 111 patients (16.7% of all patients) reached composite kidney endpoint. The 1-year, 3-year, and 5-year survival of the composite endpoint were 91.8%, 40.7%, 20.2% in the FN group and 95.8%, 57.6%, 38.4% in the no-FN group (Fig. 5). These differences were statistically significant (all P < 0.05). By contrast, patients showed no statistical significance on the survival of the composite kidney endpoint with or without FN lesions after IS treatment (Fig. 6).
Univariate analyses followed by multivariate analyses were carried out to examine the predictive value of FN lesions for kidney outcome. Clinicopathological parameters were used in the univariate analyses: gender, baseline Hb and Alb, initial eGFR, proteinuria, hematuria, treatment modalities, MEST-C scores based on the updated Oxford classification and FN lesions. In univariate Cox analyses, FN lesions whatever fraction (HR 1.567, 95% CI 1.041–2.360, P = 0.032), as well as Hb, Alb, eGFR, proteinuria, mesangial hypercellularity, endocapillary hypercellularity, tubular atrophy/interstitial fibrosis and crescents, were strongly associated with renal survival (Table 3). Moreover, after adjustment for all model factors, FN lesion was still an independent predictor for renal survival in multivariate Cox analyses (HR 2.132, 95% CI 1.265–3.593, P = 0.004, Table 4). To evaluate the effect of the FN fractions on the prognosis of IgAN, we carry out univariate and multivariate COX analyses on the FN fractions as well. As shown in Fig. 7, after adjustment for all model factors, when IgAN patients had a fraction of FN more than 1/10, the risk to reaching a composite endpoint would increase 3.927 folds (P < 0.001).
Table 3 Univariate determinants of survival from a combined event.
Abbreviations: eGFR estimated glomerular filtration rate, MEST-C MEST-C scores by Oxford classification, IS immunosuppressive agents, FN fibrinoid necrosis lesions.
Table 4 Multivariate effect of the presence of glomeruli with FN lesions on survival from combined event.
Model 1: adjusted for eGFR and proteinuria; Model 2: adjusted for Model 1 plus Hb and Alb; Model 3: adjusted for Model 2 plus METC lesions; Model 4: adjusted for Model 3 plus IS treatment.