Introduction
Immunoglobulin A (IgA) nephropathy (IgAN) is the most common type of glomerulonephritis worldwide, with an overall population incidence of 2.5 per 100,000 and a high prevalence in Asia [1,2]. About 40% of the patients with IgAN develop end-stage kidney disease (ESKD) within 20 years of diagnosis [3]. The major prognostic factors related to the kidney outcome of IgAN include hypertension, proteinuria, decreased estimated glomerular filtration rate (eGFR) at diagnosis, and pathological findings according to the Oxford pathological classification [4–7]. Among these factors, proteinuria is widely used as one of the most important clinical factors for evaluating the prognosis and therapeutic effects in patients with IgAN [7–9].
Hematuria is the most common clinical manifestation of IgAN, resulting from glomerular injury caused by the deposition of polymeric IgA in the mesangium [6,10]. In IgAN, 40% to 50% of patients have macroscopic hematuria, and 30% to 40% of patients present with microscopic hematuria [11,12]. Gross hematuria often resolves over time, whereas microscopic hematuria tends to persist and may resolve with immunosuppressant therapy [6,13]. To date, hematuria is considered benign in patients with IgAN and is not an established risk factor for disease progression. Recurrent macroscopic hematuria is considered a favorable prognostic indicator of long-term outcomes [10,14]. However, recent research has demonstrated that hematuria remission in IgAN is a favorable prognostic factor, whereas persistent hematuria is associated with poor kidney outcomes [15–24].
Since ‘remission of hematuria’ was defined based on the concept of time-averaged (TA) hematuria or median degree of hematuria during the follow-up period in previous studies, it may be considered a rather loose criterion [15–18,20]. The current study compared long-term outcomes between patients with remission of hematuria and those with persistent hematuria, using a stricter yet intuitive definition of “remission of hematuria” than that of previous studies.
Methods
Study population
This retrospective multicenter observational cohort study included patients diagnosed with IgAN through renal biopsy at three tertiary hospitals (Soonchunhyang University Seoul Hospital, Cheonan Hospital, and Bucheon Hospital in Korea). Among the patients enrolled in the cohort, we decided to analyze only those enrolled after 2009, when pathological findings began to be reported using the Oxford classification. A total of 563 patients diagnosed with IgAN through kidney biopsy between January 1, 2010 and December 31, 2016, were reviewed, and 403 patients who had been followed up for more than 3 years and underwent regular check-ups at intervals of 3 or 4 months were enrolled (Fig. 1).
The study was conducted in accordance with the principles of the Declaration of Helsinki. Clinical data were obtained from the electronic medical records with the approval of the Institutional Review Board (IRB) of Soonchunhyang University Bucheon Hospital (No. 2020-12-038-002). Written consent was waived by the IRB because of the retrospective nature of the study, and all data were fully anonymized before access by the researchers.
Clinical and laboratory data
We obtained data on patient demographics, comorbidities, body mass index (BMI), and mean arterial pressure at the time of renal biopsy. Baseline was defined as the date of kidney biopsy. All laboratory data were obtained during the follow-up period, and laboratory data at follow-up intervals of 3 or 4 months after baseline were used in the analysis. The amount of hematuria and proteinuria was determined using the number of red blood cells (RBCs) in the high-power field (HPF) in urine microscope analysis and the protein-creatinine ratio in random urine samples, respectively. We calculated TA proteinuria as the area under the receiver operating characteristic (ROC) curve (AUC) of proteinuria during the follow-up period divided by the total follow-up period, excluding the baseline values. We calculated the eGFR from serum creatinine levels using the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation [25]. Information regarding the type, duration, and total dose of administered medications, including immunosuppressive agents (ISAs) and renin-angiotensin-aldosterone system (RAAS) blockade, during the study period was collected.
We reviewed the pathological findings of kidney biopsies and specifically checked the pathological severity in patients with IgAN according to the Oxford classification [26]. For accurate pathological diagnosis, including the Oxford classification, another pathologist who was not initially involved in the pathological diagnosis reviewed the pathological findings.
Definition of “remission of hematuria” and clinical outcomes
Hematuria remission was defined as the presence of hematuria for at least 3 months before and after the diagnostic biopsy that disappeared four consecutive times in measurements taken at intervals of 3 or 4 months thereafter. The disappearance of hematuria was defined as the absence of RBCs in a HPF in the urine microscopic analysis (<1 RBC/HPF). The newly introduced criterion of four consecutive disappearances of hematuria was conceived based on the observed recurrence rates among the enrolled patients. Hematuria recurred in 20.3% (31/153) and 18.0% (22/122) of cases after two and three consecutive disappearances, respectively, but decreased significantly to 8.0% (8/100) after four consecutive disappearances, making it a reasonably acceptable standard. During the follow-up period, patients who achieved hematuria remission were classified into the remission of hematuria (remission) group, whereas those who did not were classified into the persistent hematuria (persistent) group. We analyzed the mean annual rate of eGFR decline and kidney survival as clinical outcomes. Kidney survival was defined as the absence of progression to ESKD or ≥50% decline in eGFR during the follow-up period.
Statistical analysis
Continuous variables with a normal distribution were expressed as mean ± standard deviation and compared using the independent t test. Categorical variables were presented as frequencies and percentages and analyzed using the chi-squared test. The mean annual rate of eGFR decline was determined as the difference between the eGFR at diagnosis and the last eGFR, divided by the follow-up period. The occurrence of progression to ESKD and the cumulative incidence of ≥50% decline in eGFR were estimated using the Kaplan-Meier method. The survival time for each patient was calculated from baseline to the last follow-up. Survival curves were compared using the Breslow test. To assess the impact of several variables on progression to ESKD or ≥50% decline in eGFR, univariate and multivariate Cox proportional hazards models were utilized. Propensity score matching based on logistic regression was performed to balance the baseline characteristics of the two groups. To evaluate the predictive value of remission of hematuria for a ≥50% decline in eGFR, we developed logistic regression models using treatment-related variables during the follow-up period, including ISA use, TA proteinuria, and remission of hematuria. Two models were constructed: one including remission of hematuria as an additional covariate, and one excluding it. The performance of each model was assessed using the AUC. To evaluate whether the inclusion of remission of hematuria significantly improved the predictive performance, AUCs of the two models were compared using DeLong’s test for correlated ROC curves. Statistical significance was set at a p-value less than 0.05. All analyses were performed using IBM SPSS version 25.0 for Windows (IBM Corp.), GraphPad Prism 5 (GraphPad Inc.), or MedCalc 23.2.1 (MedCalc Software Ltd.).
Results
Study population
Of the 403 enrolled patients, 100 were classified into the remission group, whereas the remaining 303 were assigned to the persistent group. Table 1 shows the baseline characteristics of the patients categorized according to the persistence of hematuria. The remission group had a higher proportion of males and a higher BMI than the persistent group. There were no significant differences in the age or blood pressure. Although the remission group tended to have a lower baseline eGFR and higher proteinuria levels, these differences were not statistically significant. The baseline hematuria level in the remission group was lower than that in the persistent group. The use of RAAS blockade and ISAs did not differ between the two groups. Pathological findings were recorded using the Oxford classification system (MEST-C; mesangial hypercellularity, endocapillary hypercellularity, segmental glomerulosclerosis, tubular atrophy/interstitial fibrosis, and cellular/fibrocellular crescents). The remission group had a significantly smaller proportion of segmental glomerulosclerosis and crescents than the persistent group. After propensity score matching, the standardized mean difference for most baseline characteristics was <0.1, indicating an acceptable to excellent balance (Supplementary Table 1, available online).
Remission of hematuria alleviated deterioration of kidney function
Table 2 shows the long-term clinical outcomes of the remission and persistent groups. The mean follow-up duration was longer in the remission group. The mean annual rate of eGFR decline was significantly lower in the remission group than in the persistent group. Similar results were obtained using propensity score matching analysis (Supplementary Table 2, available online). As shown in Table 3, we analyzed 30 patients from the remission group who had a follow-up period >2 years before and after the disappearance of hematuria. Analysis of the mean annual rate of change in eGFR revealed a declining trend in eGFR before the disappearance of hematuria, which shifted to an increasing trend after its disappearance.
Remission of hematuria as a prognostic factor for kidney survival
In the remission group, the proportion of patients who did not experience ESKD progression or ≥50% decline in eGFR was higher than in the persistent group, although statistical significance was observed only for ≥50% decline in eGFR (Fig. 2). The results of the propensity score matching showed a consistent trend; however, this trend was not statistically significant (Supplementary Fig. 1, available online).
Tables 4 and 5 present the results of univariate and multivariate analyses to identify independent prognostic factors for kidney survival. In the multivariate analysis, immunosuppressive treatment was identified as a risk factor for progression to ESKD, whereas a higher baseline eGFR was associated with a reduced risk. Multivariate analysis for ≥50% decline in eGFR showed that hematuria remission and higher baseline eGFR are associated with better kidney survival, while immunosuppressive treatment and mesangial hypercellularity are associated with worse kidney survival.
Interaction between hematuria and proteinuria
Subgroup analysis was conducted to examine the correlation between proteinuria, a well-established prognostic factor for kidney survival in IgAN, and hematuria (Fig. 3). Patients were divided into the proteinuria-positive group (TA proteinuria ≥0.75 g/day) and the proteinuria-negative group (TA proteinuria <0.75 g/day) to compare the kidney survival outcomes. Regardless of the hematuria remission status, the proteinuria-negative group demonstrated better kidney survival than the proteinuria-positive group (p < 0.001). Hematuria remission was associated with improved kidney survival in both proteinuria-positive and proteinuria-negative groups. We compared the predictive performance of two logistic regression models for progression to ≥50% decline in eGFR. Both models included treatment-related variables during the follow-up period, such as ISA use and TA proteinuria. The model including remission of hematuria demonstrated a higher AUC (0.848) than the model excluding it (0.808). The difference between the two models was statistically significant according to DeLong’s test (p = 0.04), suggesting that remission of hematuria contributes additional discriminative value (Fig. 4).
Discussion
As far as we know, this study is the first study in Korea to analyze the association between hematuria remission and long-term kidney outcomes in patients with IgAN. Over a mean follow-up of 7.0 ± 2.4 years, hematuria remission occurred in 24.8% of patients (100/403), and after the remission, the mean annual rate of eGFR decline significantly decreased. The mean annual rate of eGFR decline was lower in the remission group than in the persistent group, and this finding remained consistent after propensity score matching. The proportion of ≥50% decline in eGFR was also lower in the remission group than in the persistent group, and multivariate analysis confirmed hematuria remission as an independent prognostic factor for ≥50% decline in eGFR. These results suggest that hematuria remission may be a potent prognostic factor for IgAN.
Hematuria is the most common and characteristic clinical manifestation of IgAN; however, its role in disease progression and prognosis remains unclear [5,6]. In treatment decisions for IgAN, proteinuria and eGFR are considered more important factors than hematuria [4–7]. The association between hematuria and IgAN prognosis has shown inconsistent results in earlier studies [23,27–35]. However, recent research has suggested better outcomes in patients with minimal hematuria or remission of hematuria [15–20]. As is well known, the prognosis of IgAN varies by region, ethnicity, and country. As no domestic studies have been performed to date on the relationship between hematuria and long-term prognosis of IgAN, we conducted this research and designed it to overcome several limitations of previous studies, as outlined below. First, as all previous studies, including the current study, were retrospective, there were significant differences in the baseline characteristics between the groups categorized by hematuria status. However, in this study, the differences were minimal and were addressed using propensity score matching. Second, owing to the inherent limitations of retrospective studies, the variability in follow-up intervals among the participants included in the study and the potentially low data quality may pose issues. To overcome this limitation, our study included a larger number of patients than previous studies [15,17,18,20] and benefited from high-quality data. We minimized missing data by strictly applying the inclusion and exclusion criteria. Third, although a standardized definition of hematuria remission in IgAN has not yet been established and studies on this topic remain limited, a Japanese research group has proposed a set of remission criteria—considered to be at the level of expert opinion—based on a nationwide opinion survey of 193 teaching facilities. They defined hematuria disappearance as <1 RBC/HPF in urinary sediment and remission as this finding on ≥3 consecutive occasions over ≥6 months [36]. In the absence of a universally accepted definition, individual studies have used varying criteria to define remission of hematuria. Some studies have defined it as TA hematuria of ≤5 or ≤28 RBC/HPF during the study period [15–17] or as maintaining ≤5 RBC/HPF for more than 6 months [20]. Bobart et al. [18] used the median degree of hematuria during the follow-up period. These definitions are somewhat complex to apply in clinical practice and do not adequately reflect patients’ conditions. For instance, if a hematuria surge occurs during the follow-up period, the TA or median value may meet the remission criteria used above but cannot be considered indicative of true hematuria resolution, as hematuria aggravation may follow the surge. Therefore, we introduced a new definition of hematuria remission using the concept of hematuria disappearance. The disappearance of hematuria represents a stricter yet more intuitive criterion than TA or median values. This approach effectively reflects the true resolution of hematuria and can be easily applied in clinical practice without requiring additional calculations.
The mean annual rate of eGFR decline was significantly lower in the remission group than that in the persistent group, and similar results were obtained after propensity score matching. This finding, in which the previous study failed to show statistical significance [15], appears to be a result of the larger sample size in our study. Moreover, in the subgroup analysis of the remission group, the mean annual rate of change in eGFR showed a decreasing trend before hematuria remission, which shifted to an increasing trend after remission. Although the transient effects of medical treatments, including immunosuppressive therapies, may have contributed to this result [37], remission of hematuria may be an important predictor of improved kidney outcomes.
The remission group showed a significantly lower proportion of ≥50% decline in eGFR than the persistent group. This finding is consistent with previous studies [15–17,19]. Multivariate analysis also revealed remission of hematuria, along with baseline eGFR, as an independent prognostic factor for ≥50% decline in eGFR. This indicates that persistent hematuria significantly affects the deterioration of kidney function, and hematuria remission could serve as a promising predictor of kidney survival in IgAN. Another kidney survival outcome, the proportion of patients who progressed to ESKD, was not statistically significant. This may be attributed to the lower incidence of ESKD during follow-up, which can be explained by the generally favorable prognosis of the enrolled patients. Ultimately, the baseline eGFR of our patients was higher than that reported in previous studies [15,16]. The relatively short follow-up period may have influenced these results [15,21].
We cannot entirely exclude the possibility that the longer follow-up duration in the remission group, compared to the persistent group, may have influenced the clinical outcomes. The longer follow-up duration observed in the remission group is likely attributable to our definition of remission of hematuria as “disappearance in four consecutive measurements taken at intervals of 3 or 4 months,” which inherently increases the probability of identifying remission events in patients with extended observation periods. However, since the mean annual rate of eGFR decline is a rate-based measure, it is unlikely to have been affected by the length of follow-up, and considering that events such as progression to ESKD or ≥50% decline in eGFR are more likely to occur over longer follow-up periods, the extended follow-up in the remission group may have actually exerted a negative influence on their clinical outcomes. Therefore, the difference in follow-up duration between the two groups is more likely to be a consequence of the definition of remission rather than a source of bias, and we consider it unlikely that this difference positively affected the favorable outcomes observed in the remission group.
Multivariate analysis revealed that the M1 score was an independent prognostic factor for ≥50% decline in eGFR, aligning with the established consensus that histological lesions included in the Oxford classification affect the prognosis of IgAN [4,5]. Additionally, crescent counts and segmental glomerulosclerosis were lower in the remission group than in the persistent group. Crescent and glomerulosclerosis represent glomerular inflammation and podocyte damage, respectively. The prevalence of crescents is explained by prior findings that persistent microscopic hematuria contributes to the formation of focal and segmental crescents due to ongoing pathological activity [38]. Glomerulosclerosis can be attributed to the release of hemoglobin and other substances from erythrocytes into the urine, which induces oxidative stress and leads to podocyte injury [39,40]. These results are consistent with previous studies showing an association between MEST-C scores and hematuria remission [15–18]. On the other hand, immunosuppressive treatment was also identified as a risk factor for both progression to ESKD and ≥50% decline in eGFR. This result appeared to be due to a confounding effect, as patients receiving immunosuppressive treatment had a poorer prognosis. Therefore, it cannot be concluded that immunosuppressive treatment worsens kidney outcomes. Although RAAS blockade was more frequently used at baseline in the remission group, it was administered to nearly all patients in both groups during follow-up (97.0% vs. 97.4%, p > 0.99). Despite the lack of a statistically significant difference in its use during follow-up, we included RAAS blockade use in the multivariable analysis, given its well-established reno-protective effect in IgAN. As expected, however, it had no meaningful impact on the results and was therefore excluded from the final model to simplify the analysis and reduce unnecessary complexity.
Considering that proteinuria is a major prognostic factor of IgAN [5–7], we examined the interaction between hematuria and proteinuria on kidney survival. In our analysis, the proteinuria-negative group showed better kidney survival regardless of hematuria persistence, whereas the hematuria remission group showed better kidney survival regardless of proteinuria severity. This aligns with the results of previous studies [15,17]. Despite the strict definition of hematuria remission in our study, proteinuria had a greater impact on kidney survival than hematuria, which is consistent with the established consensus [5,6]. Nonetheless, a notable finding was that the remission group demonstrated better kidney outcomes, although they had higher proteinuria levels and lower eGFR at baseline than the persistent group. This result suggests that even in the presence of conventional risk factors, achieving hematuria remission, whether through treatment or spontaneous resolution, leads to an improved prognosis.
Our study has a few limitations. This study included participants of a single ethnic background. Additionally, this was a retrospective multicenter study, and since treatment policies differed across centers, the strategies for immunosuppressive therapy and RAAS blockade could also have varied. Consequently, the potential impact of these medications on clinical outcomes cannot be controlled. Additionally, the relatively short follow-up period prevented the identification of significant differences in the hard outcome of progression to ESKD. This may have been influenced by the higher baseline eGFR of the patients enrolled in this study than in previous studies. Lastly, although the inclusion of hematuria remission improved the performance of the prediction model, it remains unclear whether remission of hematuria serves as an independent prognostic factor or merely reflects the effects of treatment-related confounders. As this study was not designed to assess causality, and the observed improvement was modest, these findings should be interpreted with caution. Further large-scale, prospective cohort studies are warranted to determine whether hematuria remission independently contributes to long-term kidney outcomes.
In conclusion, remission of hematuria has a favorable impact on kidney outcomes in IgAN, and because it can be assessed in a noninvasive and cost-effective manner, it serves as a valuable marker for patient monitoring. Moreover, the newly proposed stricter definition of hematuria remission provides an intuitive tool for predicting the prognosis of IgAN in clinical practice.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement table 1
Print
