Renal Vein Thrombosis and Vascular Occlusion by Sickled Erythrocytes with Diffuse Renal Allograft Necrosis Shane Meehan The University of Chicago Medical Center Chicago, IL

Click here to download handout in 1-up pdf format for the current section (18.2 MB) Click here to download handout in 6-up pdf format for the current section (2.37 MB) Clinical History A 58 year old female developed end stage renal disease due to hypertensive nephrosclerosis and received chronic peritoneal dialysis for 4 years. Her past medical history was notable for hypertension, left cerebrovascular accident, asthma, avascular necrosis of the left hip treated by total hip replacement and superior vena cava syndrome related to a central line. She underwent cadaveric renal transplantation in May 2008. The transplant kidney was from a previously well 28 year old donor after cardiac death. The cause of donor death was blunt head trauma from a motorcycle accident. This was the recipient's first kidney transplant, with a six HLA antigen mismatch. Standard and flow cytometric pre-transplant cross matches were negative. Donor warm ischemia time was 13 minutes. Pulsatile cold (4°C) perfusion with University of Wisconsin perfusate was used for 17 hours and 17 minutes. Total cold ischemia time was 24 hours and 29 minutes. Subsequent warm ischemia time at implantation was 56 minutes. At reperfusion the kidney became immediately homogenously pink. Within minutes a mottled blue discoloration developed. At this time the renal artery had a regular strong pulse and the renal vein was soft. The discoloration persisted and a needle biopsy was obtained at approximately 45 minutes and sections are submitted (Slide 1, H&E). Indirect immunofluorescence for C4d on a frozen core was negative. Over the next 24 hours the patient developed oliguria and then anuria. Ultrasound revealed no flow in the renal vein and systolic flow in the renal artery. At this time, the white cell count was 7.1 K/ µl (3.5-11), red cell count 4.2 M/ µl (3.9-5.3), hemoglobin 11.4 g/dl (11.5-15.5), MCV 81.2 fL (81-99), MCHC 33.8 g/dl (32-35), hematocrit 33.8% (36-47), RDW 16.5% (<15% in females) and platelets were 288 K/µl (150-450). The serum creatinine was 12.1 mg/dl (0.5-1.4). The patient had commenced Thymoglobulin induction. On re-exploration the allograft was dark purple and had a palpable thrombus in the renal vein. Allograft nephrectomy was performed. On gross examination the kidney was 290 gm and had firm thrombus filling the arcuate, interlobar and renal vein branches. The cut surface was mottled red-purple. A representative section of the parenchyma is submitted (Slide 2, H&E). Slide 1 Click to view with ImageScope Click to view with a Web-Based Viewer Slide 2 Click to view with ImageScope Click to view with a Web-Based Viewer Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Pathologic Findings 1. Renal transplant biopsy (˜45 minute) LM: 17 glomeruli with diffuse segmental capillary obliteration by fibrin-platelet thrombi and with focal marked glomerular capillary congestion. Focal glomerular capillary pyknotic nuclei, occasional neutrophils and rare eosinophils. No thrombi apparent in arteries, arterioles or peritubular capillaries. Tubules with acute tubular injury. IF: Focal segmental glomerular capillary fibrinogen. C4d negative in peritubular capillaries. Immunoperoxidase: CD61+ platelet aggregates in glomerular and peritubular capillaries, and arterioles. Loss of capillary and arteriolar endothelial CD34 expression. EM: Glomerular endothelial disruption and loss, with fibrin tactoids, platelet aggregates and erythrocyte congestion. Additional Studies Preimplantation biopsy ˜40 unremarkable glomeruli. No thrombi. Acute tubular injury. Immunoperoxidase revealed widespread CD34 in vascular endothelium without evidence of loss of endothelial expression. Electron microscopy revealed segmental endothelial disruption with denuded capillary basement membranes. Differential Diagnosis Donor thrombotic microangiopathy (TMA) Atheroembolism Hyperacute rejection Perfusion nephropathy The 45 minute biopsy had TMA with endothelial pyknosis and extensive endothelial loss on CD34 immunostaining and EM. In contrast, the donor biopsy had normal CD34 in the vasculature, indicating that post-perfusion endothelial injury and loss occurred rapidly. Sickled erythrocytes were not evident. Absence of TMA in the donor biopsy excludes transplanted TMA. Atheroemboli were not evident. Antibody mediated (hyperacute) rejection was excluded by negative cross match and absence of peritubular capillary C4d. Ultrastructural evidence of endothelial injury with denudation of capillary basement membranes was seen in the pre-implantation biopsy. Hence the most likely explanation for these findings is perfusion nephropathy. Pathologic Diagnosis Thrombotic microangiopathy with rapid endothelial loss consistent with perfusion nephropathy What is Perfusion Nephropathy? Perfusion nephropathy was described in 1976 in kidney transplants subjected to cold pulsatile perfusion preservation [1] and is characterized by capillary and arteriolar endothelial pyknosis and loss, with obliterative fibrin-platelet thrombi, identified in perioperative (one-hour) biopsies. Twenty-one of 36 perfusion preserved kidneys had TMA in 1 hour ( ± 15 mins) biopsies and the lesions were absent from 41 cryopreserved kidneys. At one month, 9 of 21 (43%) underwent nephrectomy and 9 had serum creatinine levels >2mg/dl. In contrast, 2 of 15 perfusion preserved kidneys without TMA required early nephrectomy. Grossly, nephrectomies had a patchy blue appearance. Microscopically there was extensive cortical necrosis, vascular thrombosis, "proliferative and degenerative vascular changes" and mononuclear infiltrates in non-necrotic zones. There were no significant differences in warm or cold ischemia times, perfusion characteristics (duration, pressure, temperature, composition of perfusate), donor or recipient characteristics, HLA mismatches and DSA (by lymphocytotoxicity assay) between perfusion preserved kidneys with TMA and those without TMA. The perfusate contained cryoprecipitated plasma or modified Ringer's albumin, in contrast to University of Wisconsin perfusate used in the current case, which has no plasma. The cause of TMA was not determined. The possibility of inadvertent exposure to donor reactive antibodies or cold agglutinins in allogeneic plasma [2, 3] was not excluded, although TMA was evident in some kidneys not exposed to plasma. A second study from 1976 [4] extended these observations by identification of ultrastructural evidence of endothelial injury in pre-implantation biopsies of allografts that subsequently developed perioperative TMA. A correlation between the severity of microvascular fibrin deposition in perioperative biopsies and both duration of perfusion time and perfusion pressure was observed. Graft loss was 46% at three months, and the extent of fibrin deposition correlated with graft loss. Other studies documented similar features in perioperative biopsies [5, 6, 7], and nephrectomy specimens had extensive cortical necrosis [6]. More recent larger studies of perfusion preserved kidneys, from 1990 onwards, were notable for the absence of documented TMA and early graft loss [8, 9]. 2. Allograft Nephrectomy : Pathologic findings Gross: 290 gm, cortex and medulla red with petechial hemorrhages, renal veins with abundant firm thrombi. LM: Diffuse cortical and medullary coagulation necrosis. Venous thrombi with abundant refractile erythrocytes. Abundant refractile sickled erythrocytes in glomerular and peritubular capillaries, vasa recta and veins. IF: C4d negative. Non-specific reactivity for IgG, IgA, IgM, and C3. Fibrin – abundant in capillaries and moderate in venous thrombi. Along arterial intimal surface. Immunoperoxidase: CD61 in glomerular capillaries; few in peritubular capillaries and along arterial intima. Platelets were sparse in the occluded veins. EM: abundant sickled erythrocytes containing cytoplasmic fibrils. Differential Diagnosis Surgical/plumbing problem Thrombophilia - (Protein S/C deficiency, factor V Leiden, anti-thrombin III deficiency, anti-phospholipid and anticardiolipin antibodies) Hyperacute rejection Perfusion nephropathy The nephrectomy specimen was removed within 24 hours of implantation. In addition to diffuse cortical and medullary necrosis there was striking renal vein thrombosis (RVT) and abundant capillary and venous sickled erythrocytes admixed with sparse fibrin and platelets. Electron microscopy confirmed RBC sickling with cytoplasmic filaments in almost all erythrocytes. Subsequent hemoglobin (Hb) analysis (HPLC) confirmed HbAS. There were no surgical technical difficulties noted at vascular anastomosis. Hypercoagulability work-up was negative. Repeat cross match was negative, and the tissue had no peritubular capillary C4d or vascular immunoglobulin deposition. Renal vein occlusive thrombi have not been specifically described in perfusion nephropathy. Additional Studies Hb evaluation by HPLC revealed 25.6% HbS, indicative of sickle cell trait. Final Pathologic Diagnosis 1. Renal vein thrombosis and vascular occlusion by sickled erythrocytes with diffuse renal allograft necrosis (Sickle cell trait). 2. Perfusion nephropathy. Sickle Cell Nephropathies Sickle cell disease is caused by a single nucleotide substitution (GAG ® GTG) in the hemoglobin beta chain gene, with substitution of valine for glutamic acid at position 6 in the b -globin chain of hemoglobin (Hb) A, which then becomes HbS. Heterozygosity is evident in ~8% of Americans of African descent, and in Sub-Saharan Africa the frequency is as high as 40-60%. Deoxygenated HbS polymerizes into long filamentous structures which reduce erythrocyte deformability and impart a sickle-like shape. The severity of sickle cell vaso-occlusive disease is directly proportional to the degree of polymerization of HbS, which depends on the erythrocyte HbS concentration and the duration and severity of deoxygenation. The kidney is frequently affected in sickle cell disease (HbSS) and trait (HbAS), by a variety of different disorders [10, 11, 12, 13] (Table 1). Sickle Cell Disease and Renal Transplants Early graft loss has been reported in sickle cell trait at 2 and 6 weeks in one recipient [14]. A survey of 24 renal allografts in 21 recipients with HbAS disease revealed 35% graft loss by 1 year [15]. Early renal allograft loss with RVT and extensive sickling has been noted at days 3 and 6 post-transplantation [16, 17] in patients with HbSS. Graft infarction in HbSS with microvascular sickling of RBC at 4 months has also been described [18]. Montgomery [19] suggested 2 possible pathways of graft loss (i) acute vaso-occlusive crisis with graft thrombosis and early failure and (ii) recurrent sickle nephropathy with a more indolent progression. Vascular Occlusion The pathogenesis of vascular occlusion in HbSS and HbAS disease is complex. Sluggish blood flow and hypoxia are preconditions for polymerization of HbS and irreversible sickling [11, 20]. Sickled RBCs are rigid and adhere by by a variety of mechanisms including specific cell surface receptor interactions [21], principally to post-capillary venular endothelium [22]. RBC adhesion initiates endothelial activation, resulting in altered secretion of vasoactive compounds, increased adhesiveness for leukocytes and procoagulant activity [21]. In the present case, perfusion nephropathy probably caused mechanical/physical endothelial disruption with loss of adhesion and cell death. Exposure of subendothelial matrix molecules (thrombospondin, laminin, fibronectin, collagen) and endothelial activation may have initiated coagulation and recruitment of inflammatory cells. Reduced microvascular red cell velocity, and prolonged erythrocyte hypoxia, favored polymerization of deoxygenated HbS. Sickled erythrocytes, with increased rigidity and endothelial adhesiveness, further activated coagulation, and accumulation of inflammatory cells, with eventual vascular occlusion. It seems plausible that endothelial injury from perfusion nephropathy and TMA together played an important role in development of sickle cell-associated vascular occlusion, RVT and renal allograft infarction in this case. In conclusion, a case of early graft loss attributable to the combined effects of perfusion nephropathy and (clinically occult) sickle cell trait with resultant RVT, vascular occlusion and graft necrosis is presented. Patients with sickle cell trait may be at risk for early graft loss. Table 1. Renal Lesions in Sickle Cell Disorders Pathologic finding AS SS Papillary necrosis + + Medullary scars/VR loss + + Hemosiderosis + + Congestion and RBC impaction + + Cortical infarcts + + Pyelonephritis + + Glomerulomegaly with chronic TMA + + Immune complex GN + + Focal segmental glomerulosclerosis - + Neoplasia Medullary carcinoma + + Other renal cell carcinoma + + Key Words Renal allograft, thrombotic microangiopathy, perfusion nephropathy, renal vein thrombosis, sickle cell trait References Spector D, Limas C, Frost JL, et al.: Perfusion nephropathy in human transplants. N Engl J Med 295:1217-1221, 1976 Sturgill BC, Lobo PI, Bolton WK: Cold-reacting IgM antibody-induced renal allograft failure. Similarity to hyperacute rejection. Nephron 36:125-127, 1984 Light JA, Annable C, Perloff LJ, et al.: Immune injury from organ preservation. A potential cause of hyperacute rejection in human cadaver kidney transplantation. Transplantation 19:511-516, 1975 Hill GS, Light JA, Perloff LJ: Perfusion-related injury in renal transplantation. Surgery 79:440-447, 1976 Gattone VH, 2nd, Filo RS, Evan AP, et al.: Time course of glomerular endothelial injury related to pulsatile perfusion preservation. Transplantation 39:396-399, 1985 Limas C, Spector D, Wright JR: Histologic changes in preserved cadaveric renal transplants. Am J Pathol 88:403-428, 1977 Evan AP, Gattone VH, 2nd, Filo RS, et al.: Glomerular endothelial injury related to renal perfusion. A scanning electron microscopic study. Transplantation 35:436-441, 1983 Plata-Munoz JJ, Muthusamy A, Quiroga I, et al.: Impact of pulsatile perfusion on postoperative outcome of kidneys from controlled donors after cardiac death. Transpl Int 21:899-907, 2008 Polyak MM, Arrington BO, Stubenbord WT, et al.: The influence of pulsatile preservation on renal transplantation in the 1990s. Transplantation 69:249-258, 2000 Bhathena DB: Sickling Disorders (Chapter 28), in Heptinstall's Pathology of the Kidney, edited by Jennette JC, Olson JL, Schwartz MM, Silva FG, 5th ed, Philadelphia, Lippincott-Raven, 1998, pp 1231-1246 Bunn HF: Pathogenesis and treatment of sickle cell disease. N Engl J Med 337:762-769, 1997 Davis CJ, Jr., Mostofi FK, Sesterhenn IA: Renal medullary carcinoma. The seventh sickle cell nephropathy. Am J Surg Pathol 19:1-11, 1995 Saborio P, Scheinman JI: Sickle cell nephropathy. J Am Soc Nephrol 10:187-192, 1999 Chatterjee SN, Lundberg GD, Berne TV: Sickle cell trait: possible contributory cause of renal allograft failure. Urology 11:266-268, 1978 Chatterjee SN: National study on natural history of renal allografts in sickle cell disease or trait. Nephron 25:199-201, 1980 Barber WH, Deierhoi MH, Julian BA, et al.: Renal transplantation in sickle anemia and sickle disease. Clin Transplantation 1: 169-175, 1987 Donnelly PK, Edmunds ME, O'Reilly K: Renal transplantation in sickle cell disease. Lancet 2:229, 1988 O'Rourke EJ, Laing CM, Khan AU, et al.: Allograft dysfunction in a patient with sickle cell disease. Kidney Int 74:1219-1220, 2008 Montgomery R, Zibari G, Hill GS, et al.: Renal transplantation in patients with sickle cell nephropathy. Transplantation 58:618-620, 1994 Kaul DK, Fabry ME: In vivo studies of sickle red blood cells. Microcirculation 11:153-165, 2004 Frenette PS: Sickle cell vaso-occlusion: multistep and multicellular paradigm. Curr Opin Hematol 9:101-106, 2002 Nath KA, Grande JP, Croatt AJ, et al.: Transgenic sickle mice are markedly sensitive to renal ischemia-reperfusion injury. Am J Pathol 166:963-972, 2005