Warfarin-related nephropathy modeled by nephron reduction and excessive anticoagulation

Abstract

An acute increase in international normalized ratio (INR) to >3.0 in patients with chronic kidney disease (CKD) can associate with an unexplained acute increase in serum creatinine and accelerated progression of CKD. A subset of these patients have renal tubular obstruction by casts of red blood cells, presumably the dominant mechanism of the acute kidney injury described as warfarin-related nephropathy. Here, we developed an animal model of this acute kidney injury that is based on the 5/6-nephrectomy model to aid future investigation of the pathogenesis of this condition. We found that acute excessive anticoagulation with brodifacoum ("superwarfarin") increased serum creatinine levels and hematuria in 5/6-nephrectomized rats but not in controls. In addition, morphologic findings in 5/6-nephrectomized rats included glomerular hemorrhage, occlusive red blood cell casts, and acute tubular injury, similar to the biopsy findings among affected patients. Furthermore, in the rat model, we observed an increase in apoptosis of glomerular endothelial cells. In summary, the 5/6-nephrectomy model combined with excessive anticoagulation may be a useful tool to study the pathogenesis of warfarin-related nephropathy.

Figure 1.

Treatment with brodifacoum (superwarfarin) results in increased SC levels in 5/6-nephrectomy rats but not in controls. Changes in PT are associated with SC level increase in 5/6-nephrectomy rats only. (A) Change in SC after brodifacoum (superwarfarin) treatment. Control rats (n = 9, ■), 5/6-nephrectomy rats at 3 weeks (n = 8, ▴), and 5/6-nephrectomy rats at 8 weeks (n = 10, ●) after the ablative surgery were treated with a single dose of brodifacoum (BF). SC levels were measured before (day 0) and daily after the treatment with brodifacoum (arrow). (B) Animals were treated with different doses of warfarin in drinking water. PT was measured before and after the treatment. A “surrogate” INR was used by comparing PT time after and before the treatment. The average PT time in 50 rats (25 control and 25 5/6-nephrectomy rats) was used as the normal PT time. Changes in SC were calculated from baseline in the same animal. Squares represent 5/6-nephrectomy rats; triangles represent control rats.

Figure 2.

The ablative surgery results in spontaneous progressive hematuria, which is significantly increased by brodifacoum (superwarfarin) treatment. (A) Changes in hematuria in rats after the ablative surgery (5/6-nephrectomy). Animals at 3 and 8 weeks after the ablative surgery were used for the study. Hematuria was measured using DiaScreen reagent strips. (B) Hematuria was measured before (day 0) and daily after the treatment with brodifacoum (BF, arrow) using DiaScreen reagent strips in control (n = 9, ■), 5/6-nephrectomy 8 weeks after the ablative surgery (n = 10, ●), and 5/6-nephrectomy 3 weeks after the ablative surgery (n = 8, ▴) rats. Hematuria was graded using a semiquantitative scale of 0 to 3+. Score 0 was designated for negative hematuria, score 1+ for mild hematuria, score 2+ for moderate hematuria, and score 3+ for large hematuria. *P < 0.05 compared with control.

Figure 3.

Morphologic findings in the kidneys obtained from experimental animals treated with brodifacoum are similar to those in patients with WRN. (A) Control and (B) 5/6-nephrectomy (ablative nephropathy, 8 weeks postsurgery) rats were treated with brodifacoum. The kidneys were obtained on day 4 post-treatment. The morphologic findings in control rats treated with brodifacoum were mild and nonspecific. In contrast, 5/6-nephrectomy animals had RBCs in Bowman's space and RBC casts in the corresponding tubules. (C) For comparison, the morphologic findings in a kidney biopsy from a patient with WRN are shown. Numerous RBCs and RBC occlusive casts were noticed in tubules and in Bowman's space. Magnification, 200×. Hematoxylin and eosin stain.

Figure 4.

Animals with WRN have an increased number of apoptotic cells in the kidney. Kidneys were obtained from animals treated with brodifacoum (only 5/6-nephrectomy rats are shown) and were stained with antibodies against CD31 (red), with the terminal deoxynucleotidyl transferase dUTP nick-end labeling enzymatic labeling assay (green) and Hoechst (blue). Apoptotic cells were detected in (A) glomeruli, (B) peritubular capillaries, and (C) tubules. Magnification, ×400. (D) The number of apoptotic cells was counted in different compartments of the kidney and by their origin. Treatment with brodifacoum resulted in an increased number of apoptotic endothelial cells in glomeruli (Glom EnC) and tubular epithelial cells (Tub EpC) in 5/6-nephrectomy rats 3 and 8 weeks after the ablative surgery (5/6 NE 3 w [n = 8]; and 5/6 NE 8 w [n = 10], respectively), but not in controls (n = 9). No significant changes in the number of apoptotic endothelial cells in peritubular capillaries (PTC EnC) were noted between different experimental groups. *P < 0.05 as compared with control.

Figure 5.

Experimental protocol of the study. Two- to-3 month-old Sprague–Dawley rats were subjected to 5/6-nephrectomy (ablative nephropathy). SC, proteinuria, and hematuria were measured weekly after the ablative surgery. Animals were treated with brodifacoum (superwarfarin) 3 and 8 weeks after the ablative surgery. SC, proteinuria, and hematuria were measured daily after the treatment. All animals died by day 4 after the treatment with brodifacoum was initiated.