Baseline characteristics of participants
A total of 611 biopsy-proven IgAN patients between May 2009 and April 2013 were initially screened in the study, and 291 patients were enrolled in this retrospective cohort study according to the inclusion and exclusion criteria (shown in Figure. 1). The median follow-up time was 41.2 months. Patients were divided into two groups based on the T score, and the baseline characteristics and histopathological features at the time of renal biopsy were shown in Table 1. Of all 291 patients, 131 (45.02%) were males, and the mean age was 32.50±11.94 years. The mean SBP, MAP, Scr, eGFR and 24h urine protein were 127.18±11.95mmHg, 96.3±12.75mmHg, 92.73±47.14μmol/L, 90.87±69.14mL/min/1.73m2, 1.71±2.51g/day, respectively. Regarding MEST-C oxford scores in all patients, 6.53% were M1, 11% were E1, 66.67% were S1, 40.2% T≥1 and 21.99% C≥1. Compared to the absence of T lesion group, IgAN patients with T had significantly higher levels of SBP, MAP (129.17±16.21mmHg vs 125.84±16.05mmHg, and 98.40±12.97mmHg vs 94.90±12.43mmHg), Scr (115.35±59.31μmol/L vs 77.52±28.05μmol/L), and lower eGFR (80.1±102.43mL/min/1.73m2 vs 98.11±29.11mL/min/1.73m2). There was also a tendency of difference in 24h proteinuria between two groups 1.62 ±2.20g/24h vs 1.77±2.71g/24h). For renal biopsy, 6 (5.13%) patients had M1, 15 (12.82%) had E1, 100 (85.47%) had S1, 30 (25.64%) had C1 or C2 in presence of T group. 13 (7.47%) of absence T patients had M1, 17 (9.77%) had E1, 94 (54.02%) had S1, 34 (19.54%) had C1 or C2, and percentage of S1 was significantly lower compared to the presence of T group. There were no significant differences in age, gender, M, E, C scores between the two groups.
Relationships between coagulation parameters and T score
The association of coagulation parameters, including PT, APTT, TT, FDP, FIB, D-D, AT-Ⅲ, and T lesion were investigated. For all IgAN patients, 21 patients had shorter PT seconds and 16 patients had longer seconds than the reference value. For APTT, 41 of IgAN patients were shorter and 46 patients were longer than the reference. There were 22 patients with both abnormal PT and APTT levels. As is shown in Table 2, PT and APTT seconds of patients with T was significantly shorter (both p<0.001) than that of the without T group. While there were no statistical differences in TT, FDP, FIB, D-D, and AT-III levels. We also explored the relationship between PT, APTT and other oxford classification of IgAN, but no significant differences were found. In addition, we would like to know whether PT and APTT are related to the degree of T rating. In the comparison of the T0 group, PT and APTT were significantly shorter in T1 (p=0.003 and p=0.012, respectively) and T2 group (p<0.001 and p=0.003, respectively). Moreover, PT also showed a statistical difference between T1 and T2 group (p=0.036) (shown in Figure. 2).
Relationships between PT, APTT and renal outcome
Of all the participants, a 50% decline in eGFR was observed in 4.8% (n=14) of the patients, and 2.4% (n=7) of the patients developed ESRD during the follow-up period. In order to investigate the association of PT, APTT and renal progression, ROC curves were firstly constructed (shown in Figure. 3). The areas under the ROC curves (AUCs) for PT and APTT were 0.67 (95%CI 0.54–0.79) and 0.7 (95%CI 0.59–0.79). Based on PT and APTT cut-off value, patients were divided into PT<11.15s and >11.15s groups, or APTT<29.65s and >29.65s groups respectively. Kaplan-Meier survival analysis was performed to examine renal survival between groups (shown in Figure. 4). The cumulative survival rate was significantly higher in patients with longer PT (>11.15s) or APTT (>29.65s) than the lower groups (p=0.008, p=0.027 respectively).
Cox proportional hazards models were constructed to determine the prognostic value of PT, APTT, and clinical/histopathological parameters in IgAN patients (Table 3). Univariate analysis showed higher values of Scr (HR 1.22, 95%CI 1.16–1.28, per 10umol/L increase), SBP (HR 1.34, 95% CI 1.08–1.67, per 10mmHg increase), 24h proteinuria (HR 1.13, 95%CI 1.00–1.28, per 1g increase) and lower values of eGFR (HR 0.52, 95%CI 0.41–0.66, per 1mL/min/1.73m2 increase), PT (HR 3.27, 95%CI 1.30–8.24 versus >11.15s group), APTT (HR 2.59, 95%CI 1.08–6.21 versus >29.65s group) at baseline were significantly associated with a higher risk of renal outcome. The biopsy score S1 (HR 3.68, 95% CI 1.08–12.59) and T≥1 (HR 13.16, 95%CI 3.06–56.67) also significantly correlated to poor prognosis. However, after adjusting for univariate risk factors, the multivariate analysis showed no significant value in PT or APTT (data not shown).
Thus, we assessed the additive effect of PT or APTT and T lesion on the renal outcome (shown in Figure. 5). Patients without T lesion and with PT>11.15s or APTT>29.65s were set as the reference groups. The odd ratios (ORs) were 3 (95%CI: 0.18–49.00; P=0.44) for patients without T and PT<11.15s, 13.35 (95%CI: 1.57–113.37; P=0.018) for patients with T and PT>11.15s, and 41.93 (95%CI: 5.32–330.48; P<0.001) for patients with T and PT<11.15s. This association was consistent with the combination of APTT and T lesion. The corresponding ORs were 5.88 (95%CI: 0.36–97.09; P=0.22), 22.15 (95%CI: 2.81–174.89, P=0.003), and 47.04 (95%CI: 5.64–392.55; P<0.001).