We conducted a retrospective, observational, single-center study of all patients diagnosed with anti-GBM disease at the Unidade Local de Saúde São João (ULSSJ), Porto, between January 2003 and December 2023. Follow-up was updated through January 2025.

Anti-GBM disease was defined by a compatible clinical presentation plus the presence of circulating anti-GBM antibodies and/or biopsy-proven crescentic GN with linear IgG deposition along the GBM on immunofluorescence (IF). Immunofluorescence findings were evaluated by an experienced renal pathologist and linear IgG deposition was assessed semi-quantitatively using a standard 0–3+ grading scale. Patients with dual positivity for anti-GBM antibodies and ANCA (MPO or PR3) were included if they lacked clinical features of systemic vasculitis. Individuals with isolated ANCA positivity in the absence of circulating anti-GBM antibodies or linear GBM staining were excluded, as they did not fulfill established diagnostic criteria for anti-GBM disease.

Circulating anti-GBM antibodies were measured by fluorometric enzyme immunoassay (FEIA) using manufacturer-standardized units (U/mL). Values >8U/mL were considered positive according to the validated cut-off applied at our institution during the study period. ANCA testing was performed by antigen-specific ELISA for MPO and PR3.

Demographic, clinical, laboratory and therapeutic data were obtained from electronic medical records. Baseline variables included demographics; symptoms at presentation; AKI severity; presence of RPGN features; serum creatinine (sCr); estimated glomerular filtration rate (eGFR) calculated using the 2009 CKD-EPI equation; proteinuria (g/24h); anti-GBM and ANCA serology; and need for renal replacement therapy (RRT). Therapeutic data included use and route of CS and CP, number of PLEX sessions and Intensive Care Unit (ICU) admissions. Corticosteroid therapy typically consisted of intravenous methylprednisolone pulses followed by oral prednisone. Cyclophosphamide was administered either intravenously or orally according to standard regimens. Plasmapheresis was performed with albumin replacement (fresh frozen plasma if alveolar hemorrhage), daily, using one plasma-volume exchange, until antibody negativation and when indicated according to KDIGO guidelines. Overall and renal survival, relapse events, and kidney transplantation status were collected up to January 2025.

Kidney biopsy was processed for optic microscopy (OM) using routine staining (H&E, PAS, silver, Masson). IF studies included staining for IgG, IgA, IgM, C3, C1q, κ, and λ. Histologic variables collected included total number of glomeruli; percentage of normal glomeruli; crescent type and proportion; presence of fibrinoid necrosis or Bowman's capsule rupture; and degree of interstitial fibrosis and tubular atrophy (IFTA). Electron microscopy was not routinely performed.

Several operational definitions were applied to ensure consistency in the characterization of clinical and renal outcomes. RRT at presentation referred to the initiation of dialysis within 72h of hospital admission. End-stage kidney disease (ESKD) was defined as persistent dialysis dependence without subsequent renal recovery. Renal recovery was considered in patients who were initially dialysis-dependent and subsequently remained dialysis-independent for at least 12 weeks during follow-up. Patient survival was defined as time from diagnosis to death from any cause. Renal survival was defined as the time from diagnosis to initiation of RRT.

Crescentic GN was defined as crescents involving ≥50% of glomeruli. RPGN was defined by rapidly worsening kidney function accompanied by an active urinary sediment and/or crescentic GN on kidney biopsy. Pulmonary hemorrhage was diagnosed based on overt hemoptysis and/or radiologic evidence of alveolar hemorrhage on chest computed tomography (CT). Relapse was defined as pulmonary or renal deterioration occurring more than three months after diagnosis, in association with rising anti-GBM antibody titers and/or biopsy evidence of active GN, as defined by Marques et al.8 Normal glomeruli were defined histologically as glomeruli without crescents, fibrinoid necrosis, or chronic scarring.

Continuous variables were expressed as mean±SD (standard deviation) or median (interquartile range, IQR), according to distribution. Categorical variables were summarized as absolute and relative frequencies. Group comparisons used chi-square or Fisher's exact test for categorical variables, and Student's t-test or Mann–Whitney test for continuous variables. Survival analysis was performed using Kaplan–Meier curves and the log-rank test after stratification according to predefined clinical and histopathological variables. Statistical significance was defined as p<0.05. Analyses were conducted using IBM SPSS Statistics version 28.0.

A total of 27 patients met diagnostic criteria for anti-GBM disease. The mean age at presentation was 58.7±22.0 years and 59.3% were female. Hypertension was present in 33.3% of patients, while 40.7% had a history of smoking. No patient had a previous autoimmune or chronic inflammatory disorder. Baseline demographic and clinical characteristics are summarized in Table 1.

Fatigue was the most frequently reported symptom (77.8%), followed by oliguria or anuria (33.3%), hematuria (63.0%) and respiratory complaints such as dyspnea or cough (25.9%). Flu-like symptoms in the preceding weeks were recalled by 25.9% of patients, while fever, arthralgia, and weight loss were less common.

Pulmonary involvement consistent with alveolar hemorrhage was present in 9 patients (33.3%). An additional 5 patients (18.5%) displayed non-specific pulmonary abnormalities on chest CT, including ground-glass opacities or features suggestive of infection or bronchiectasis, but without evidence of alveolar hemorrhage. Respiratory failure occurred in 6 patients (22.2%), four of whom had confirmed alveolar hemorrhage.

All patients exhibited renal involvement. Mean serum creatinine at presentation was 11.95±10.18mg/dL, corresponding to an eGFR of 5.2mL/min/1.73m2 (IQR 2.5–8.9). The mean 24-h proteinuria was 3.8±2.1g/day, with nephrotic-range proteinuria detected in 7 patients (25.9%). Laboratory and histopathologic features are summarized in Table 2.

Circulating anti-GBM antibodies were detected in 25 patients (92.6%), with a mean titer of 308.2±296.6U/mL; 17 (70.8%) had titers >100U/mL. The median time to anti-GBM antibodies clearance was 17 days (IQR 14–21). Concurrent ANCAs with MPO specificity were identified in 8 patients (29.6%), including one with dual MPO and PR3 positivity. Two patients (7.4%) were anti-GBM seronegative and were diagnosed based on a compatible clinical presentation and biopsy-proven crescentic glomerulonephritis with diffuse linear IgG deposition along the GBM, after exclusion of alternative causes of linear IgG staining. Neither of these last patients had diabetes mellitus. All patients tested negative for hepatitis B and C, EBV, and CMV.

Kidney biopsy was performed in 15 patients (55.6%). A median of 22 glomeruli (IQR 19–26) was available per specimen. Crescentic GN was present in 12 patients (80%). The mean percentage of crescents across all biopsies was 60.9±32.6%, predominantly cellular, with fewer fibrocellular crescents. Fibrinoid necrosis and Bowman's capsule rupture were identified in 80.0% and 33.3% of patients, respectively. IFTA was present in most biopsies with a mean involvement of 33.9±21.5%. The percentage of normal glomeruli was low with a median of 0% (IQR 0–1). IF demonstrated linear IgG deposition in 12 of 14 evaluable biopsies; the two patients without linear IgG staining had pauci-immune patterns. PLA2R staining was positive in 1 of the 2 patients evaluated.

Therapeutic data were available for 25 patients. Twenty-three (92%) received immunosuppressive therapy most commonly the combination of CS, CP, and PLEX (68%). The median number of PLEX sessions was 12.5 (IQR 7–14). Corticosteroid pulses were administered in 84%, and CP was given intravenously in 10 patients (40%) and orally in 8 (32%). Immunosuppressive therapy and PLEX were withheld in 2 patients due to extended chronicity and absence of pulmonary involvement. Six patients (22.2%) required ICU admission during the acute presentation.

At presentation, 25 patients (92.6%) required RRT shortly after admission (Fig. 1). Among these, 5 patients (18.5%) recovered kidney function within the first 6 months (median time to dialysis independence, 34 days [IQR 25–64]), while 17 progressed to ESKD. One additional patient who initially did not require RRT subsequently progressed to ESKD within the first month, resulting in a total of 18 patients (66.7%) reaching ESKD. Only one patient remained dialysis-independent throughout the entire follow-up; this individual had anti-GBM positivity, alveolar hemorrhage, mild kidney impairment and linear IgG deposition along the glomerular basement membrane without crescents. The overall clinical course is summarized in Table 3 and patient trajectories are depicted in Fig. 2.

Fig. 1.

Kaplan–Meier estimates of renal survival in patients with anti-GBM disease (A), stratified according to baseline serum creatinine (B) and the presence of normal glomeruli on kidney biopsy (C). Results of univariate analyses using the log-rank test are shown for each variable. Renal recovery occurred in 5 patients (median time to dialysis independence, 34 days [IQR 25–64]). Abbreviations: IQR, interquartile range; sCR, serum creatinine.

Fig. 2.

Flowchart of the anti-GBM patient's outcome. Abbreviations: ESKD, end-stage kidney disease; GBM, glomerular basement membrane.

Renal outcomes at 6 months were further analyzed by comparing dialysis-independent patients (n=6) with those who progressed to ESKD (n=18), as shown in Table 4. Patients who developed ESKD had higher serum creatinine at presentation (12.1±7.9 vs. 5.3±3.0mg/dL; p=0.009), fewer normal glomeruli (0% vs. 5%; p=0.028), and more frequent crescentic involvement >40% (66.7% vs. 13.3%; p=0.022). Although the mean percentage of crescents was higher in the ESKD group, this difference was not statistically significant. High anti-GBM antibody titers also showed a trend toward worse renal outcomes: both patients who died early had titers >100U/mL, and all patients with titers above this threshold required dialysis at presentation. High titers (>100U/mL) were more common among patients who progressed to ESKD (76.4% vs. 40.0%), although this association did not reach statistical significance.

All 9 patients who presented with pulmonary-renal syndrome experienced resolution of alveolar hemorrhage; however, only two recovered kidney function. ANCA status did not correlate with renal survival; among the 6 surviving ANCA-positive patients, all ultimately progressed to ESKD.

Overall patient survival at 1, 3 and 5 years was 85.2% (95% CI, 71.8–98.6%), 77.8% (95% CI, 62.1–93.5%), and 74.1% (95% CI, 57.5–90.6%), respectively. Three patients (11.1%) died during hospitalization or within 30 days of discharge. During long-term follow-up, 8 patients (33.3%) died (median time to death of 35.5 months). Patient survival did not differ significantly according to baseline serum creatinine or the presence of normal glomeruli on kidney biopsy (Fig. 3).

Fig. 3.

Kaplan–Meier patient survival curves for anti-GBM disease during a 60-month follow-up (A), stratified according to baseline serum creatinine (B) and the presence of normal glomeruli on kidney biopsy (C). Results of the univariate analyses using the log-rank test are given for each variable.

Five patients (18.5%) received a kidney transplant. No graft failure was observed during follow-up; one patient died seven years post-transplantation with a functioning graft. No cases of recurrent anti-GBM disease were observed.