Urinary elafin and kidney injury in hematopoietic cell transplant recipients

Abstract

Background and objectives: Graft-versus-host disease (GVHD) is associated with kidney injury after hematopoietic cell transplantation (HCT). Because plasma elafin levels correlate with skin GVHD, this study examined urinary elafin as a potential marker of renal inflammation and injury.

Design, setting, participants, & measurements: Urine was collected prospectively on 205 patients undergoing their first HCT from 2003 to 2010. Collections were done at baseline, weekly through day 100, and monthly through year 1 to measure elafin and urine albumin-to-creatinine ratio (ACR). Associations between urinary elafin levels and microalbuminuria, macroalbuminuria, AKI and CKD, and mortality were examined using Cox proportional hazards or linear regression models. Available kidney biopsy specimens were processed for immunohistochemistry.

Results: Mean urinary elafin levels to day 100 were higher in patients with micro- or macroalbuminuria (adjusted mean difference, 529 pg/ml; P=0.03) at day 100 than in those with a normal ACR (adjusted mean difference, 1295 pg/ml; P<0.001). Mean urinary elafin levels were higher in patients with AKI compared with patients without AKI (adjusted mean difference, 558 pg/ml; P<0.01). The average urinary elafin levels within the first 100 days after HCT were higher in patients who developed CKD at 1 year than in patients without CKD (adjusted mean difference, 894 pg/ml; P=0.002). Among allogeneic recipients, a higher proportion of patients with micro- or macroalbuminuria at day 100 also had grade II-IV acute GVHD (80% and 86%, respectively) compared with patients with a normal ACR (58%; global P<0.01). Each increase in elafin of 500 pg/ml resulted in a 10% increase in risk of persistent macroalbuminuria (hazard ratio, 1.10; 95% confidence interval [95% CI], 1.06 to 1.13; P<0.001) and a 7% increase in the risk of overall mortality (95% CI, 1.02 to 1.13, P<0.01). Renal biopsy specimens from a separate cohort of HCT survivors demonstrated elafin staining in distal and collecting duct tubules.

Conclusion: Higher urinary elafin levels are associated with an increased risk of micro- and macroalbuminuria, AKI and CKD, and death after HCT.

Keywords: chronic kidney disease; hematopoietic cell transplant; microalbuminuria; proteinuria; renal tubular epithelial cells.

Figure 1.

Flow diagram of patient enrollment into study protocol.

Figure 2.

Risk of persistent macroalbuminuria as a function of urinary elafin level in pg/ml, where elafin is modeled as a cubic spline as described in the Materials and Methods. The vertical lines represent the 5th, 25th, 50th, 75th, and 95th percentiles of elafin values after hematopoietic cell transplantation. Dotted line represents the pointwise 95% lower confidence limit of the hazard ratio. Hazard ratios depicted in the curve are relative to a reference urinary elafin level of 300 pg/ml.

Figure 3.

Risk of overall mortality as a function of urinary elafin level in pg/ml, where urinary elafin is modeled as a cubic spline as described in the Materials and Methods. The vertical lines represent the 5th, 25th, 50th, 75th, and 95th percentiles of elafin values after hematopoietic cell transplantation. Dotted line represents the pointwise 95% upper and lower confidence limit of the hazard ratio. Hazard ratios depicted in the curve are relative to a reference urinary elafin level of 300 pg/ml.

Figure 4.

Elafin staining in hematopoietic cell transplantation and control kidney samples. (A and B) Low-power image from case 20 demonstrating the absence of staining in proximal tubules, which are identified by their tall epithelium and basally situated nuclei. Staining of the distal segment is noted in the cortex and medulla, where many profiles of the thin segment of the loop of Henle are clearly evident (3,3′-diaminobenzidine; original magnification, ×100). (C) Intermediate-power image from case 21 shows positive staining in a subset of tubules and negative glomerulus (arrowhead). This case demonstrated several patterns, including diffuse finely (*) and coarsely (upper right and lower left) granular as well as coarse luminal granules (#) (3,3′-diaminobenzidine; original magnification, ×200). (D) Finely granular cytoplasmic staining was most commonly diffusely distributed within the cytoplasm (case 15), whereas coarse granules (E) typically accumulated toward the lumen aspect (case 15) (3,3′-diaminobenzidine; original magnification, ×400). (F) Staining for elafin was rarely restricted to the basal aspect of tubular cells (case 27) (3,3′-diaminobenzidine; original magnification, ×400).