Combination of Acanthocyturia and Elevated Serum IgA Level for the Diagnosis of IgA Nephropathy

Background: Serum IgA level (sIgA) has been assessed as a clinical test for the diagnosis of IgA nephropathy (IgAN). Its performance has been modest with sensitivity (SENS) and specificity (SPEC) around 65%. However, studies had tested its performance independently of urinary findings. We hypothesized that idenitfication of acanthocyturia by microscopic examination of the urinary sediment (uSEDI) may increase the pre-test probability of IgAN and, therefore, the diagnostic yield of sIgA may improve when utilized in cases where acanthocyturia is present. Methods: We prospectively collected data of patients seen in nephrology consultation who had a urine specimen subjected to uSEDI as part of the clinical evaluation over a 5-yr period. Within this cohort, we identified cases in which: 1) urinary acanthocytes were identified by uSEDI; 2) a kidney biopsy was perfomed; and 3) sIgA was obtained as part of the diagnostic work up. Normal range of sIga for our hospital laboratory is 40-350 mg/dL. We assessed the diagnostic performance of sIgA. Results: Among 801 patients assessed by uSEDI, 240 (30%) underwent kidney biopsy. Urinary acanthocytes were identified in 96 (40%), and 49 (51%) had sIgA measured. Mean age was 53 years, 45% women, 71% white and 27% black. Median serum creatinine was 2.2 (0.6-19.2) mg/dL. Diagnosis was IgAN in 22 (24%), pauci immune glomerulonephritis (GN) in 18 (20%), lupus GN in 8 (9%), other proliferative or infection-related GN (IRGN) in 9 (10%), and thin basement membrane nephropathy in 5 (5%). One case was IgA-dominant IRGN. Median sIgA was higher in IgAN or IgA-dominant IRGN than in others (360 vs 253 mg/dL, p=0.012). Using the laboratory normal range, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of elevated sIgA for identifying IgAN (or IgA-dominant IRGN) were 53%, 68%, 42% and 77%. The AUROC for sIgA for IgAN diagnosis was 0.74 (95% CI 0.59-0.88, p=0.013). The optimal cuttoff was 327 mg/dL. Notably, all IgAN cases were ≥ 230 mg/dL. Conclusion: sIgA performs modestly as a diagnostic test for IgAN. Adjusting the cutoff to a level within normal range increases the NPV. We suggest that in patients with acanthocyturia, sIgA may be part of the serological work up.