Background

Anti-glomerular basement membrane (GBM) disease is a type of small vessel vasculitis, involving glomerular capillaries, pulmonary capillaries, or both, and presents with anti-GBM autoantibody deposition along the GBM [1]. The outcome of anti-GBM disease has improved with intensive treatment with plasmapheresis, glucocorticoids, and/or immunosuppressive therapy; however, > 60% of patients with anti-GBM disease progress to end-stage renal disease (ESRD), still reaching a mortality rate of approximately 25% at 1 year after diagnosis [2, 3]. Although a retrospective study in China reported that the combination of plasmapheresis and corticosteroids had a favorable effect on both patient survival and renal survival (defined as the time from onset to ESRD) [2], few reports have described the clinical course, treatment, and outcomes of Japanese patients with anti-GBM disease [4, 5]. Here, we report two patients with anti-GBM disease who had distinct clinical courses. In addition, we summarized the clinicopathological findings of 12 patients with anti-GBM disease at our institution and reviewed the literature of mortality in this population.

Case presentation

Case 1

A 72-year-old Japanese man with hypertension and chronic kidney disease due to type 2 diabetes mellitus was referred to our medical center with worsening of the serum creatinine (sCr) level for 2 weeks. He had a baseline sCr of 1.1 mg/dL. The patient never smoked or had no family history of kidney or rheumatic disease. On examination, his respiratory rate was 24 breaths per minute, his heart rate was 73 bpm, his blood pressure was 125/64 mmHg, he had a saturation of 83% in room air, and his body temperature was 38.6 °C. Physical examination was remarkable for labored breathing using respiratory accessory muscles, bilateral coarse crackles, and pretibial pitting edema. There was no rash or purpuric skin lesion. Notable laboratory findings included severe anemia (hemoglobin level, 5.9 g/dL) and a markedly decreased kidney function (blood urea nitrogen (BUN), 182 mg/dL; sCr, 27.3 mg/dL; estimated glomerular filtration rate (eGFR) [6], 1 ml/min/1.73 m2). The immunoglobulin (Ig) levels were normal (IgM, 80 mg/dL; IgA, 196 mg/dL), except for elevation of IgG (1958 mg/dL). Urinalysis showed proteinuria (4.8 g/gCr) with microscopic hematuria (> 100 erythrocytes per high-power field) and a few red blood cell casts. Plain computed tomography showed bilateral pleural effusion and normal-sized kidneys. Antinuclear antibodies and proteinase 3 (PR3) and myeloperoxidase (MPO) antineutrophil cytoplasm antibodies (ANCA) were negative. Anti-GBM antibody was 476 units/mL. PR3-ANCA, MPO-ANCA, and anti-GBM antibody were measured using enzyme-linked immunosorbent assay method. Anti-GBM disease was confirmed by renal biopsy, which showed 10 glomeruli, seven of them had cellular crescent formation (Fig. 1) and three had global sclerosis. A moderate interstitial infiltration composed of monocytes and neutrophils was observed. Immunofluorescence microscopy showed linear staining for IgG along with the GBM (Fig. 2). Along with oral prednisolone (40 mg/day), double-filtration plasmapheresis (DFPP) using human albumin (5%) as the replacement material was initiated. The patient’s anti-GBM antibody decreased from 476 to 18 units/mL with twelve sessions of DFPP, but he underwent hemodialysis because his renal function did not recover. One month after completion of DFPP, however, the patient died from lung abscess (Fig. 3) caused by Pseudomonas aeruginosa.

Fig. 1
figure 1

Light microscopy of the kidney in case 1 showed cellular crescent formation (periodic acid-methenamine silver, × 400)

Full size image
Fig. 2
figure 2

Immunofluorescence microscopy of the kidney in case 1 showed linear staining along with the glomerular basement membrane (IgG immunofluorescence, × 400)

Full size image
Fig. 3
figure 3

Plain computed tomography of the chest in case 1 showed cavity lesion in the right lower lobe

Full size image

Case 2

A 32-year-old Japanese man, current smoker, presented with a 3-week history of fever with rigor and macroscopic hematuria. He denied hemoptysis, skin rash, or joint pain. Physical examination revealed his respiratory rate was 16 breaths per minute, his heart rate was 108 bpm, his blood pressure was 108/58 mmHg, and his body temperature was 39.0 °C. Other physical examinations were unremarkable. Laboratory tests showed mild anemia (hemoglobin level, 10.9 g/dL) and a moderately decreased kidney function (BUN, 35 mg/dL; sCr, 4.2 mg/dL; eGFR [6], 14 ml/min/1.73 m2). IgG was elevated (1110 mg/dL), but other immunoglobulin levels were normal (IgM, 68 mg/dL; IgA, 113 mg/dL). Antinuclear antibodies, PR3-ANCA and MPO-ANCA, using enzyme-linked immunosorbent assay method were negative. Urinalysis showed proteinuria (0.9 g/gCr) with microscopic hematuria (> 100 erythrocytes per high-power field). Plain computed tomography showed normal-sized kidneys and no alveolar hemorrhage. Anti-GBM antibody was 265 units/mL (enzyme-linked immunosorbent assay). Renal biopsy showed 12 glomeruli; all of them had cellular crescent formation (Fig. 4) and had no global sclerosis. Immunofluorescence microscopy showed linear staining for IgG along with the GBM (Fig. 5). Intravenous infusion of methyl prednisolone (500 mg/day for three consecutive days) followed by oral prednisolone (50 mg/day) was started. After completion of 20 sessions of DFPP using human albumin (5%), his anti-GBM antibody became undetectable. However, the patient’s kidney function did not recover, and he underwent hemodialysis. One year after the initial diagnosis, the patient’s anti-GBM antibody remained undetectable, and thus, he received living-donor kidney transplantation without recurrence of anti-GBM disease or other complications.

Fig. 4
figure 4

Light microscopy of the kidney in case 2 showed cellular crescent formation (periodic acid-methenamine silver, × 400)

Full size image
Fig. 5
figure 5

Immunofluorescence microscopy of the kidney in case 1 showed linear staining along with the glomerular basement membrane (IgG immunofluorescence, × 400)

Full size image

Discussion and conclusions

We reported two patients with anti-GBM disease to illustrate their distinct patient outcomes. Table 1 summarizes demographic, clinical, and pathologic characteristics of 12 patients with anti-GBM disease from 2000 to 2020 at our institution. At baseline, the median age was 70 years (interquartile range (IQR), 32–77 years), 18% were women, 45% had diabetes mellitus, and 64% had hypertension. Of the 12 patients, 4 patients (36%) had anuria, 3 (25%) had pulmonary interstitial opacities on computed tomography, and 2 (17%) had a pulmonary hemorrhage. The median level of sCr at presentation was 8.5 mg/dL (IQR, 5.2–11.4 mg/dL), and the median eGFR was 5 mL/min/1.73 m2 (IQR, 3–8 mL/min/1.73 m2). The median level of anti-GBM antibody level was 214 international unit (IU)/mL (IQR, 43–16385 IU/mL). ANCA positivity was shown in 3 patients (25%), with PR3-ANCA in 1 and MPO-ANCA in 2. Renal biopsy was performed in all 12 patients with the following results: global sclerosis in 6 (50%); crescent formation in 12 (100%); glomerular necrosis in 10 (83%); and interstitial inflammation in 12 (100%). A total of 8 patients (67%) received plasmapheresis with a total number of administrations between 6 and 20. All patients received oral corticosteroids with median initial doses of 40 mg/day (IQR, 40–60 mg/day), whereas 10 patients (91%) received steroid pulse therapy (methyl prednisolone 500 mg/day for three consecutive days), with the total number of steroid pulse courses between 1 and 3. During the median follow-up period of 2.7 years (IQR, 0.4–8.2 years), 4 patients died, all patients progressed to ESRD, and one of these patients received living-donor kidney transplantation (Fig. 6).

Table 1 Demographic, clinical, and pathologic characteristics from 12 patients at our institution
Full size table
Fig. 6
figure 6

Kaplan–Meier curves of mortality in the alive patients and the deceased patients at our institution

Full size image

This study discusses three clinical observations. First, prior studies have reported the survival rates of patients with anti-GBM disease, ranging from 80 to 95% at 1-year after diagnosis and 24–92% at 5-year after diagnosis, respectively (Table 2) [2,3,4,5, 7,8,9,10,11,12,13,14,15,16]. Our cohort had a similar 1-year survival rate, but a poorer 5-year survival rate than previous studies even with the intensive immunosuppressive treatments and plasmapheresis. The reasons remain uncertain, but one possible explanation may be related to the fact that we had older patients: The median age (70 years) in our cohort was relatively higher than that in other reports [2,3,4,5, 7,8,9,10,11,12,13,14,15,16]. Indeed, several studies demonstrated that older age at diagnosis was associated with a higher patient mortality [11, 12, 14, 15]. Another possibility may be related to the inclusion of more severe cases in our cohort: The baseline sCr levels (8.5 mg/dL) and all patients utilized renal replacement therapy. McAddo et al. [14] showed that renal replacement therapy at presentation predicted mortality in patients with anti-GBM disease. Of note, an epidemiological study in Australia and New Zealand reported that patients with anti-GBM disease had comparable survival on dialysis compared to that with other causes of ESRD [17]. To evaluate prognosis in Japanese dialysis patients with anti-GBM disease, large-scale nationwide studies are required.

Table 2 Summary of the results of prior studies in patients with anti-GBM disease between 2000 and 2020
Full size table

Second, the Kidney Disease: Improving Global Outcomes guidelines [18] do not strongly recommend intensive treatments, including plasma exchange, corticosteroids, and cyclophosphamide, in patients with anti-GBM disease if they presented with dialysis dependency, had no lung hemorrhage, and had 100% cellular crescents in the renal biopsy, which was based on the results of previous studies [7, 11]. A recent study demonstrated that low percentage of normal glomeruli and large extent of interstitial infiltrate were associated with poorer renal survival in patients with anti-GBM disease [19]. Indeed, renal survival of our cohort was very poor. One of the reasons may be related to the fact that no patient had normal glomeruli, whereas all patients had crescents and interstitial inflammation in our cohort.

Third, previous studies recommended performing plasmapheresis until aGBM levels become undetectable or are at least close to the lower limit of detection; however, these recommendations are supported only by expert opinion [3, 20]. Additionally, prior studies demonstrated that the combination of plasmapheresis, corticosteroids, and/or cyclophosphamide yielded a better patient survival versus treatment without plasmapheresis; however, these studies did not investigate the association between post-treatment anti-GBM levels and patient outcomes [20, 21]. In our cohort, 67% of patients became negative for anti-GBM levels after treatment, which was a higher percentage than that in a previous report (52%) [2]. The reasons for this discrepancy remain uncertain, although the finding may be due to the lower proportion of patients with a pulmonary hemorrhage in the current study versus the previous report [2] (45% vs. 17%).

This case series has several limitations. The largest and most apparent is the small sample size and data from a single center; therefore, the study results should be interpreted with caution. The small sample size also did not allow us to perform statistical analyses for the patient outcomes. Second, DFPP was performed for all patients because a recent study reported that anti-GBM patients with DFPP had similar patient and renal survival to those with immunoadsorption [21, 22]. Finally, the results of our study cannot be directly compared with those in the previous studies because the measurement methods of MPO-ANCA, PR3-ANCA, and anti-GBM antibody varied among studies.

In conclusion, we report two cases of anti-GBM disease who had different patient outcomes and summarize clinicopathological characteristics and survival of patients with anti-GBM disease at our medical center. Larger studies would be ideal, but may be difficult to conduct because anti-GBM disease is a rare disease, which has an estimated incidence between 0.5 and 1.6 case per million per year [1]. In this context, we believe that, despite the limited sample size, this work may add to the literature by presenting detailed clinicopathological characteristics and outcomes in Japanese patients with anti-GBM disease.