Abstract
Objective Adult immunoglobulin A vasculitis (IgAV) is documented to be associated with more renal involvement and poorer renal outcomes compared to children, but adult IgAV nephritis (IgAV-N) data are rather limited. The present study aimed to describe the characteristics of adult IgAV-N and investigate the long-term prognostic factors.
Methods Clinical and morphological data from 106 adult patients with biopsy-proven IgAV-N and follow-up data from 94 patients in a single Chinese center were analyzed in this retrospective study. Median follow-up time interval was 102 months.
Results The median age of patients with IgAV-N at biopsy was 38 (IQR 24-53) years, and 52.8% were male. The median blood pressure was 126/80 mmHg, and 25.5% of patients were hypertensive at baseline. The median initial proteinuria was 1.4 (IQR 0.7-2.2) g/day and estimated glomerular filtration rate (eGFR) was 103 (IQR 84-121) mL/min/1.73 m2. The median time interval of onset to biopsy was 8 (IQR 3-40) weeks. In biopsy, the median percentage of global sclerosis was 5.9% (IQR 0.0-13.8), whereas 45.3% of patients had interstitial fibrosis and tubular atrophy. Further, during follow-up, 7.4% patients died, 4.3% patients progressed to endstage kidney disease (ESKD), and 6.4% patients developed > 30% eGFR reduction from baseline. Multivariate Cox proportional analyses revealed hypertension (HTN) history and > 10% global sclerosis at presentation were independent prognostic factors for poor outcome.
Conclusion The present adult IgAV-N cohort revealed a relatively young onset age, and lower incidence of nephrotic syndrome and ESKD. Moreover, nonimmune factors such as history of HTN and renal chronic histological lesions in biopsy played a crucial role in prognosis of IgAV-N.
IgA vasculitis (IgAV), formerly called Henoch-Schönlein purpura, is a systemic IgA-mediated vasculitis characterized clinically by palpable purpura, joint symptoms, abdominal pain, and renal involvement.1 It is the most common, although rather self-limited and benign, vasculitis disease of childhood. Unlike in children, IgAV in adults is complicated and associated with severe outcomes because of higher incidence of IgAV-related nephritis (IgAV-N) and risk of developing endstage kidney disease (ESKD) within 10 years after being diagnosed with IgAV-N.2,3 Owing to the paucity of adult data, the management of adult IgAV-N is still based on evidence from pediatrics, despite the large disparity between children and adults. Therefore, it is of great value to investigate the clinical and morphological characteristics and prognostic factors of adult IgAV-N.
Pediatric IgAV-N studies have suggested that acute clinical and pathological variables were critical risk factors for poor renal outcome, such as proteinuria at onset and massive endocapillary proliferation or cellular crescents at biopsy, and children usually responded well to immunosuppressive therapy.4-8 However, immunosuppressive therapy in adult patients with IgAV-N was not as efficient as that in pediatric patients.9 This inconsistency might be the result of more concomitant and nonimmune factors in adults, which might contribute to more complicated clinical presentation and severe outcome.10-14 Several retrospective studies demonstrated that chronic histological lesions, such as global sclerosis and interstitial fibrosis, were associated with adverse renal events.11,13 A previous French cohort suggested an association between increasing age and overall IgAV-related mortality.12 As a result, it is necessary to determine whether additional factors, such as hypertension (HTN), diabetes mellitus (DM), obesity, and timing of renal biopsy influence the prognosis of adult IgAV-N despite adequate immunosuppressive therapy. Notably, data on Chinese patients with IgAV-N are scarce. We also noticed a high variability in the prevalence and severity of renal disease in different studies; the percentage of individuals experiencing chronic kidney failure ranged from 8% to 63%,11,15-18 which deserves to be compared and investigated.
The present study examined the characteristics of 106 patients with biopsy-confirmed IgAV-N aged more than 18 years in a single-center Chinese cohort, and 94 patients who completed follow-up were analyzed. We included comprehensive clinical, laboratory, and histological data consisting of detailed medical history and the time interval from purpura onset to proteinuria or renal biopsy. The patients were followed for up to 220 (median 102) months. Predictive factors of poor prognosis were analyzed and compared with selected published adult IgAV-N studies with similar follow-up intervals.
METHODS
Patients. The study enrolled 106 patients aged > 18 years with renal biopsy-proven IgAV-N in the division of nephrology at the Huashan Hospital, Fudan University, from 2002 to 2016. IgAV-N was diagnosed based on the European Alliance of Associations for Rheumatology/Paediatric Rheumatology European Society criteria,19 which is characterized by predominate IgA deposits in mesangium or peripheral capillary walls of glomeruli in kidney biopsy. Clinical, biological, and histopathological data at onset, biopsy, and last follow-up were reviewed thoroughly and collected. The study was approved by the ethics committee of the Huashan Hospital, Fudan University (ethics approval number: KY2016-394) and conducted in accordance with the principles of the Declaration of Helsinki. All patients gave informed consent for the anonymous use of their health data.
Clinical and biological data. Clinical and biological data were recorded at the time of biopsy according to the standard clinical pathway. Extrarenal manifestations of IgAV and the time interval of onset to biopsy were collected at admission. History of HTN and DM was determined by the medical record, only if the onset of which was earlier than the onset of IgAV. At the follow-up medical examination, the use of angiontensin receptor blocker (ARBs)/angiotensin-converting enzyme (ACE) inhibitors or immunosuppressive agents and serum creatinine, urine erythrocytes, and protein were recorded. The estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration formula.
Histopathological variables. All renal biopsies were independently examined by 2 pathologists blinded to clinical features. To calculate the percentage of global sclerosis, the number of lesion glomeruli were divided by the total number of glomeruli in the light microscopy specimen. Interstitial fibrosis and tubular atrophy were scaled by the percentage of cortical area affected; if < 5% of kidney cortex was fibrotic or atrophic, it was graded as none, 5% to 25% was graded as mild, and > 25% was graded as moderate to severe. The presence of crescents, fibrinoid necrosis, and endocapillary proliferation were recorded.
Study endpoints. The outcomes of 94/106 patients were evaluated as of the reference date of June 1, 2021. Eleven patients were lost to follow-up and 1 patient was excluded because of developing concomitant antiglomerular basement membrane disease 1 year after IgAV-N onset. Death, ESKD, or eGFR decrease > 30% from baseline were considered endpoint events. The cause of death and renal status at death were determined for patients who died. The second endpoints were renal insufficiency (moderate was defined as eGFR = 30-60 mL/min/1.73 m2, whereas severe was defined as eGFR = 15-29 mL/min/1.73 m2) or persistent proteinuria and hematuria without renal insufficiency. Proteinuria was defined as proteinuria > 0.3 g/day and hematuria as > 13.6 red cells/μL in the urine.
Statistical analysis. Data were presented as median (IQR), mean (SD), or number of cases (% of total cases). Univariate and multivariate Cox proportional analyses were used to investigate the risk factors of endpoints. The Kaplan-Meier plots were used to illustrate the risk of endpoints and log-rank tests were used for tests of significance. Differences between categorical data were analyzed by Pearson chi-square test, Fisher exact test, or Mantel-Haenszel chi-square test. Differences between quantitative data were analyzed by the Mann-Whitney U test or the Kruskal-Wallis H test. Spearman correlation was used for quantitative and categorical data. All tests were 2-tailed, and P < 0.05 was considered significant. SPSS for Mac version 25 (IBM Corp.) and Prism version 9.0 (GraphPad Software Inc.) were used for statistical analyses.
RESULTS
Baseline data at biopsy. Clinical data from a total of 106 patients were included in our study (Table 1). The renal involvement was confirmed by kidney biopsy and the median time interval of onset to biopsy was 8 (IQR 3-40) weeks. Median initial proteinuria was 1.4 (IQR 0.7-2.2) g/day and 12 (11.3%) patients had nephrotic syndrome. Most patients had a normal eGFR, with a median eGFR of 103 (IQR 84-121) mL/min/1.73 m2. The median blood pressure (BP) was 126/80 mmHg, including 27 (25.5%) patients diagnosed as hypertensive before the onset of IgAV.
Clinical data of adult IgAV-N cohort.
In the biopsy, 66 (62.3%) patients had crescents but none had crescents in > 50% of glomeruli. Endocapillary proliferation was found in 58 (54.7%) patients. The median percentage of sclerosis glomeruli was 5.9% (IQR 0.0-13.8). Since pathological features could be affected by the biopsy timing, we completed our analysis according to the time interval of onset to biopsy as shown in Supplementary Table S1 (available with the online version of this article).
Follow-up and outcome. The median follow-up duration was 102 (IQR 66-135) months, and 94/106 patients completed the follow-up (Table 1). During follow-up, 7 (7.4%) patients died: 3 of infection, 2 of cardiovascular events, 1 of cancer, and 1 of an accident. Four (4.3%) patients progressed to ESKD and 6 (6.4%) patients developed > 30% eGFR reduction from baseline.
Sixty-six percent of patients were prescribed an ARB or ACE inhibitor. Eighty-one (76.4%) patients received steroids or other immunosuppressive agents, among whom 70 patients received steroids alone, 6 patients received steroids with cyclophosphamide, 4 patients received steroids with mycophenolate mofetil, and 1 patient received steroids with azathioprine. Immunosuppressive therapy was commenced in patients with proteinuria > 1 g/day or proteinuria 0.5-1 g with active lesions in kidney biopsy. Sixty-nine of our patients received immunosuppressive therapy immediately as a result of proteinuria > 1 g/day or crescents/fibrinoid necrosis at biopsy. Immunosuppressive therapy was given to 12 patients with persistent proteinuria > 1 g/day during follow-up. Treatment with other immunosuppressive agents was given according to the experience of the nephrologist, as no solid evidence for adult IgAV-N treatment was available.
The time interval from biopsy to ESKD ranged from 38 to 90 months. Fifty-seven (60%) patients had kidney biopsies within 12 weeks following IgAV onset, whereas 19 patients did not undergo the procedure until ≥ 52 weeks, and 36.8% of them reached the endpoints at follow-up (Supplementary Table S2, available with the online version of this article). Further analysis showed significantly greater total incidence of poor outcomes (36.8% vs 13.3%), especially ESKD (15.8% vs 1.3%), in patients who had a biopsy after 52 weeks compared to those within 52 weeks (Supplementary Table S2).
Prediction of endpoint events. All-cause mortality, ESKD, and eGFR reduction > 30% from baseline were considered endpoint events. As shown in Table 2, patients who reached endpoints were older and had a higher incidence of HTN history, but no difference in admission BP. These patients also exhibited higher proteinuria, lower eGFR, more severe anemia, and worse chronic histological lesions at biopsy. In the multivariable model, a history of HTN and > 10% global sclerosis were suggested to be associated with poor outcomes, which was also confirmed by Kaplan-Meier plots (Figure 1).
Univariate and multivariate Cox proportional analysis for poor outcomes.
Kaplan-Meier plots for risk of endpoints. (A) Overall event-free survival for risk of endpoint events. (B) Event-free survival for risk of endpoints according to history of HTN. (C) Event-free survival for risk of endpoints according to GS > 10% or not. All-cause mortality, ESKD, and eGFR decrease > 30% from baseline were considered endpoint events for all 3 Kaplan-Meier plots. eGFR: estimated glomerular filtration rate (by Chronic Kidney Disease Epidemiology Collaboration equation); ESKD: endstage kidney disease; GS: global sclerosis; HTN: hypertension.
We further analyzed patients’ characteristics according to HTN history and global sclerosis percentage. As shown in Figure 2 and Supplementary Table S3 (available with the online version of this article), patients with a history of HTN had significantly higher age, admission BP, BMI, baseline proteinuria, level of serum IgA, lower eGFR, and more severe global sclerosis compared to those without a history of HTN. The global sclerosis percentage correlated negatively with baseline eGFR (R −0.462, P < 0.001) and positively with age (R 0.379, P < 0.001) and grade of interstitial fibrosis and tubular atrophy at biopsy (R 0.564, P < 0.001; Supplementary Figure S1 and Table S4).
Baseline (A) serum IgA level, (B) eGFR, and (C) percentage of global sclerosis were compared between patients with or without a history of HTN. Differences between quantitative data were analyzed by the Mann-Whitney U test. * P < 0.05, ** P < 0.01. eGFR: estimated glomerular filtration rate (by Chronic Kidney Disease Epidemiology Collaboration equation); HTN: hypertension.
To better sum up the current knowledge of prognostic studies of adult IgAV-N, clinical characteristics and prognostic outcomes of our cohort were compared with those of the 6 most cited studies with similar follow-up intervals, as shown in Table 3.
Selected studies investigating long-term prognosis of adult IgAV-N.
DISCUSSION
The goal of this study was to describe the characteristics of adult IgAV-N and investigate the long-term prognostic factors. In this Chinese adult IgAV-N cohort, 11.7% patients died or progressed to ESKD and 6.4% had a greater than 30% reduction of eGFR from baseline after a median 102 months of follow-up. In our cohort, patients who reached endpoints were older adults with more proteinuria, worse kidney function, a higher incidence of a history of HTN despite indiscriminate admission BP, and a higher degree of glomerulosclerosis. Multivariate analysis revealed that HTN history and global glomerulosclerosis greater than 10% were independent risk factors for adult IgAV-N.
IgAV is the most prevalent form of vasculitis in children and adults. Overall prevalence of renal involvement in children is 20% to 28%,10 but that of adults can be as high as 45% to 85%, depending on the criteria.11,20 IgAV is typically self-limiting in children, but poses a greater risk of kidney failure in adults.15,16,21,22 It is reported that approximately one-quarter of adult patients with IgAV-N would progress to severe chronic kidney disease (eGFR < 30 mL/min per 1.73 m2) or ESKD in 15 years,11,16,23 whereas in our study, 4.3% patients developed ESKD and 2.1% patients developed severe chronic kidney disease at approximately 8.5 years. To explain the relatively lower incidence of renal insufficiency in our cohort, our cohort was compared to those of 6 other studies (Table 3). Our patients had less proteinuria, better renal function, and more crescents at biopsy. In this cohort, median proteinuria was 1.4 g/day and nephrotic syndrome occurred in approximately 11.3% of the patients, whereas the European groups reported nearly 30% with nephrotic syndrome.2 Older patients tended to present with a higher level of proteinuria, and the high portion of older patients contributed to the poor prognosis in some studies.11,17 In those cohorts, the median age of patients was 50 and one-third of patients exceeded 60 years of age, whereas our cohort had a median age of 38 years and only 17.9% patients were aged more than 60 years. Genetic background might have also contributed. Further, 66% of patients in our study presented with acute pathological lesions as crescents, whereas crescents affected 18% to 36% of patients in the previous reports.2,13,16,17 Some studies suggested that endocapillary hypercellularity and crescents were independently associated with poor renal outcomes.13,24 Such typical acute renal lesions displayed no effect on the progress of IgAV-N as per the findings of our study, which we speculated, in addition to a younger population of patients with less proteinuria, would be affected by the relatively high portion of immunosuppressive therapy administration in some way. Immunosuppressive therapy was administrated in 76.4% of patients of our cohort, whereas it was administered in 37% to 72% of patients in the other cohorts. Sixty-nine of our patients received immunosuppressive therapy immediately as a result of proteinuria > 1 g/day or crescents/fibrinoid necrosis at biopsy. Immunosuppressive therapy was given to 12 patients with persistent proteinuria > 1 g/day during follow-up. Davin and Coppo25,26 emphasized that the initiation of an effective antiinflammatory treatment targeting the acute episodes of glomerular inflammation should not be delayed. Otherwise, acute damage could turn into chronic lesions. However, immunosuppressive therapy is a double-edged sword. Two patients in our cohort died of infection secondary to immunosuppressive treatment. Currently, immunosuppressive therapy based on steroids is the principal treatment of adult patients with IgAV-N with a high risk of progression. Previous evidence showed new drugs like rituximab as a promising therapeutic option for adult-onset IgAV.27 Further studies on IgAV-N treatment to assess benefits and risks are necessary in adult patients.
Also, biopsy timing is critical for histologic lesions. Pediatric studies suggest that the renal histologic lesions would occur in 85% of cases within 4 weeks following the diagnosis of IgAV, 91% within 6 weeks, and 97% within 6 months, if patients developed hematuria or proteinuria.28 As shown by a large German collaborative study of children with IgAV-N, cellular crescents occur in biopsy specimens as early as 24 days following onset, whereas chronic lesions emerge if the biopsy was performed 57 days later.29 Such data in adults were lacking. Thus, we performed further analysis according to the time interval of onset to biopsy. More than half of the patients in this cohort had biopsies within 12 weeks after the onset of purpura, whereas 40.4% waited longer than 12 weeks, including 21.3% who waited longer than a year. As displayed in Supplementary Table S1 (available with the online version of this article), more than 80% of adults developed acute morphological characteristics within 3 weeks and the acute lesions tended to affect fewer patients with a longer biopsy interval. Notably, global sclerosis and interstitial fibrosis appeared in patients with biopsy taken as early as 3 weeks. Although biopsy intervals were indifferent between patients with or without poor outcomes, patients who postponed kidney biopsy for 52 weeks following purpura appeared to have a remarkably high rate of ESKD (Supplementary Table S2). Taken together, chronic lesions in adult patients with IgAV-N were multifactorial. The immune mechanism of prolonged standing IgA deposition and inflammation in the kidneys, but also nonimmune factors, like primary HTN, could contribute to global sclerosis and be involved in the progression of IgAV-N.
Our study revealed that patients with a history of HTN before initiating IgAV were at higher risk of poor long-term outcomes. Nonetheless, BP at admission was indistinguishable, regardless of the outcome. In our cohort, however, patients with a history of HTN were older and more likely to have glomerulosclerosis and higher serum IgA levels. Several cross-sectional studies have reported an increased level of inflammation markers in patients with HTN, such as C-reactive protein, interleukin (IL)-6, and tumor necrosis factor (TNF).30 Moreover, it has also been reported that IL-6, TNF, and other inflammatory factors were significantly elevated in sera of adult patients with IgAV compared to healthy controls.31 We speculated that inflammatory responses induced by IgAV might also promote vascular wall sclerosis in patients with HTN and contribute to kidney deterioration.
The single-center cohort and sample size were the main limitations of this study. Patients with DM with kidney disease have been reported to have worse kidney outcomes.32,33 We found in our cohort that the prevalence of DM among patients with endpoints was almost 4 times higher than that of those without endpoints (23.5% vs 6.4%), but because of the limited number of patients, medical history of DM did not reach statistical significance. Further, we did not conclude information regarding the response to different treatments.
Different cohort studies presented with varying distributions of clinical characteristics, and thus had different prognoses. The relatively lower incidence of renal insufficiency in our cohort may be related to the younger age of patients, lower incidence of nephrotic syndrome, and higher rate of immunosuppressive therapy administration. The present cohort showed a history of HTN and glomerulosclerosis greater than 10% as independent risk factors. Our study suggested that comorbidities and renal reserve should be evaluated and handled with caution in adult patients with IgAV-N. Further studies to develop a prediction model of risk factors for IgAV-N progression might be useful for patient-specific risk stratification and treatment decisions.
Footnotes
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication April 20, 2023.
- Copyright © 2023 by the Journal of Rheumatology
