VIRUSES AND THROMBOTIC MICROANGIOPATHY

We read with interest the case report by Jeejeebhoy FM et al. in the June 27, 1998 issue of Transplantation entitled "Thrombotic microangiopathy in association with cytomegalovirus infection in a renal transplant patient" (1). In this article, the authors suggested a causal association between cytomegalovirus (CMV) infection and thrombotic microangiopathy (TMA) in a renal transplant recipient. Other authors have similarly suggested such an association (2). We propose that in addition to CMV, other viruses may be implicated in the pathogenesis of TMA. For example, we have observed an association between TMA and anticardiolipin antibodies (ACAs) in hepatitis C (HCV)-positive renal allograft recipients (3). Among 18 HCV-positive recipients, 5 patients presented with biopsy-proven renal TMA occurring 5 to 120 days after transplantation. Two of these five patients also had evidence for active CMV infection. All 5 patients had elevated ACAs as compared with only 1 of 13 patients without TMA (P<0.05). In further support of this observation, we are now reporting TMA in the native kidneys of two patients with chronic HCV infection associated with high titers of ACAs. Case 1. A 34-year-old male underwent orthotopic liver transplantation for end-stage liver disease secondary to hepatitis C. His posttransplantation immunosuppressive regimen consisted of cyclosporine, prednisone, and azathioprine. Three years after transplantation, because of elevated liver enzymes, a liver biopsy specimen was obtained, which demonstrated a recurrence of hepatitis C in the allograft. The HCV viral RNA titer was elevated at 8.7 million copies/ml (lower limit of detection=2000 copies/ml) by quantitative polymerase chain reaction (Roche Diagnostics, Somerville, NJ). The HCV genotype was found to be type 1a. He was initiated on alpha interferon therapy at 3 MU three times a week, but no improvement in liver function tests was noted. Eight months later, he was admitted for worsening liver function tests, and his cyclosporine was changed to tacrolimus. One week after starting tacrolimus, he developed acute renal failure in association with microangiopathic anemia, with numerous schistocytes in the peripheral smear, severe thrombocytopenia (6000/mm3), elevated lactate dehydrogenase up to 1800 U/L (normal: 110-210), and decreased haptoglobin to <6 mg/dl (normal: 16-199) along with increased free hemoglobin of 41 mg/dl (normal: 2-7). Endoscopic retrograde cholangiopancreatography was negative, and a repeat liver biopsy specimen revealed fibrosing cholestatic hepatitis and positive immunoperoxidase staining for CMV in rare hepatocytes. He subsequently developed seizures, prompting a computerized tomography of the brain that demonstrated ischemic changes in the left internal capsule. The IgM ACA titer was elevated at 70 phospholipid units/L (normal <15) with normal IgG ACA. Despite treatment with intravenous ganciclovir for 3 weeks, fresh-frozen plasma infusions, and discontinuation of tacrolimus, his clinical condition continued to decline with progressive hepatic and renal failure. He expired 5 weeks after admission due to multiorgan failure. Case 2. A 54-year-old woman with chronic HCV infection was admitted to the hospital because of pneumonia and acute non-Q wave myocardial infarction. One week later, she developed acute oliguric renal failure with a rise in creatinine from 1.7 mg/dl to 5.2 mg/dl, and she required hemodialysis. A renal biopsy specimen demonstrated thrombotic microangiopathy. She also had severe thrombocytopenia (7000/mm3), microangiopathic anemia with peripheral schistocytes, elevated lactate dehydrogenase (752 U/L), and low haptoglobin (<6 mg/dl). A diagnosis of "idiopathic hemolytic uremic syndrome" (HUS) was made. Her course was complicated by spontaneous splenic hemorrhage requiring splenectomy. ACAs were markedly elevated with IgG and IgM ACAs of 364 and 358 phospholipid units/L (normal <15), respectively, and a lupus anticoagulant test was positive. Other laboratory tests included positive antinuclear antibody at 1:100 but negative anti-native DNA antibodies, positive rapid plasma reagent test but negative fluorescent titer antibody-absorption test. An aggressive treatment with daily plasmapheresis for 2 weeks, steroids, and systemic anticoagulation was initiated with a good response. In the ensuing 3 weeks, she came off dialysis and the microangiopathic hemolytic anemia resolved. One year later, she is stable with a creatinine of 2.3 mg/dl. Discussion. HUS or TMA occurring after solid organ or bone marrow transplantation has been associated with the use of cyclosporine or tacrolimus (2, 4). However, HUS/TMA is relatively rare after transplantation, indicating that other cofactors are likely to be important in triggering this syndrome. In our first patient, TMA was associated with the initiation of tacrolimus therapy. Interestingly, this HCV-positive liver transplant recipient had elevated titers of ACA, similar to our previous observation in HCV-positive kidney transplant recipients (3). In addition, he also had evidence of CMV infection in his liver biopsy specimen. The second patient presented with idiopathic HUS but she had chronic HCV infection and markedly elevated ACAs as well. In another study of tacrolimus-associated TMA, one patient had undergone liver transplantation for chronic HCV infection, and the second case was a cardiac transplant recipient who had active CMV infection, also suggesting a possible association between these viruses and TMA (4). The precise underlying mechanisms linking viruses to the HUS/TMA remain to be elucidated; one possibility could be related to exaggerated endothelial cell activation, particularly after transplantation, in patients receiving cyclosporine or tacrolimus. CMV has been shown to infect and activate endothelial cells (5), and similarly ACAs in HCV-infected patients may have detrimental antiendothelial and/or antiplatelet activity (6, 7). Interestingly, ACAs are frequently found both in chronic HCV and HIV-infected patients and TMA can be a complication of both viral infections (3, 7-9). The presence of ACAs may thus explain the occurrence of TMAs during some viral infections. Procoagulant properties of these viruses may also play an important role. In addition, an increased prevalence (65%) of ACAs has been found in classic pediatric HUS/TMA associated with enteric infections (10). Finally, of particular interest is the recent description of antibodies to von Willebrand factor-cleaving protease in patients with acute thrombotic thrombocytopenic purpura, a syndrome that closely resembles HUS (11, 12). Whether ACAs inhibit von Willebrand factor-cleaving protease in some patients with TMA after transplantation will be important to determine in the future. Whatever the precise mechanisms involved, it seems that various viruses may be implicated in the pathogenesis of HUS/TMA in both allografted organs and in native kidneys. Prompt search for an ongoing viral infection is warranted in such instances because early initiation of antiviral therapy in addition to the conventional therapeutic approaches may improve the outcome in these patients. Seema Baid1,2 Manuel Pascual1,3 A. Benedict Cosimi3 Raymond T. Chung4 Robert B. Colvin5 Nina Tolkoff-Rubin3 Renal Unit; Transplantation Unit; Gastrointestinal Unit; Department of Pathology; Massachusetts General Hospital; Boston, Massachusetts