Mario Magnone

Magnone, Mario; Holley, Jean L.; Shapiro, Ron; Scantlebury, Velma; McCauley, Jerry; Jordan, Mark; Vivas, Carlos; Starzl, Thomas; Johnson, John P. Author Information

OBJECTIVE: The authors' objective was to identify factors associated with candidemia and candidemia-related death among adult liver transplant recipients. SUMMARY BACKGROUND DATA: Invasive candidiasis is the most common severe fungal infection occurring after liver transplantation and is associated with high morbidity and mortality rates. Although candidemia is not always found during invasive candidiasis, it has been considered as an indicator of invasive candidiasis in immunocompromised patients. METHODS: A time-matched case-control study of 26 patients with candidemia, which was defined as the isolation of Candida from at least one blood culture, and 52 control patients without candidemia was reported. Two control patients were matched with each case patient regarding time of transplantation and duration of follow-up. RESULTS: Between December 1985 and December 1992, candidemia developed in 1.4% of adult liver transplant recipients a median of 25 days after transplantation (range, 2-1690 days). The overall mortality rate among patients with candidemia was 81%, and 71% of these deaths were related to candidemia. Conditional logistic regression analysis was used to identify factors associated with candidemia, which were 1) hyperglycemia treated with insulin up to 2 weeks before candidemia (odds ratio [OR], 16.15; p = 0.002), and 2) exposure to more than three different intravenous antibiotics before development of candidemia (OR, 11.15; p = 0.005). The variables predictive of death related to candidemia were abdominal surgery performed up to 1 week before candidemia (relative risk [RR], 7.25; p = 0.02), high white blood cell count (RR, 1.10; p = 0.01), lower platelet count (RR, 0.99; p = 0.02), and elevated AST with candidemia (RR, 1.001; p = 0.01). CONCLUSIONS: Hyperglycemia that requires insulin and exposure to more than three antibiotics are the factors associated with the development of candidemia in liver transplant recipients. When candidemia develops shortly after abdominal surgery and in patients with elevated AST, high white blood cell count, or low platelet count, it is associated with a high mortality rate.

This study describes nine cases of post-transplant lymphoproliferative disease (PTLD) presenting as renal allograft dysfunction. Onset of symptoms was 34 to 265 days post-transplant, typically (in six of nine cases) after refractory rejection treated with OKT3. Diagnosis was made by histopathologic examination of needle biopsy (three of nine cases) or allograft nephrectomy (six of nine cases) specimens. Disease was confined to the allograft in three patients. The morphology was polymorphic in eight cases and monomorphic in one case. Five cases showed monotypic kappa or lambda light chain expression. Expansile lymphoid infiltrates, serpiginous necrosis, nuclear atypia, and presence of Epstein-Barr virus RNA helped to distinguish PTLD from severe rejection. Tubular damage and venulitis was common in PTLD lesions, but arterial involvement was not prominent. Infiltration of the ureter, hilar adipose tissue, and nerve twigs was frequent in nephrectomy specimens. Reduction of immunosuppression led to resolution of PTLD in two of three cases diagnosed by needle biopsy, but severe acute rejection led to graft loss in one case; the third case progressed to fatal multisystem disease. Among cases diagnosed at nephrectomy, two of six patients died of disseminated PTLD and one of six died of sepsis. The five surviving patients are alive 41 to 99 months after initial diagnosis without evidence of recurrent PTLD.

Cytomegalovirus (CMV*), the first reported transplant-associated opportunistic virus (1), has remained the most frequent single cause or co-cause of infections throughout the ensuing three decades(2). Because prophylactic drug and hyperimmune globulin therapy has not been effective in these CMV-infected patients(2-4), Rubin (5) suggested with-holding treatment until the phase of rapid viral replication(preemptive therapy). Early diagnosis of CMV infection, upon which this strategy depends, can be accomplished in the presymptomatic phase(6, 7) with a rapid quantitative direct demonstration of CMV antigen pp65 in cytospin preparations of peripheral blood leukocytes (PBL) (8, 9). This method has aided in the management of 20 CMV seronegative recipients of livers from seropositive donors, a circumstance that carries an 85-90% risk of virus transmission of CMV disease, presumably from migratory monocytes from the graft(10) that are known to harbor the latent virus in healthy seropositive individuals (11). The 20 recipients under tacrolimus/prednisone immunosuppression were followed for 308±100.5 days (range, 63-520 days) after transplantation; a death at 63 days unrelated to CMV accounted for the only follow-up of <3 months. Heparinized blood samples were obtained by schedule during the first 3 months or when clinically indicated during this period and later. The pp65 antigen was detected in duplicate leukocyte spots (2×105 cells in 200 μl/glass slide) with the technique of Van der Bij et al.(8), using immunofluorescence (12) instead of immunoperoxidase staining (9). The anti-pp65 mouse monoclonal antibody was clone 1C3 (Biosoft, Argene, France). Fluorescein-isothiocyanate-labeled goat anti-mouse Fab2 fragment (Cappel, Durham, NC) was applied as the conjugate. Slides were counterstained with Evans blue dye and observed under 400× magnification for the typical green nuclear fluorescence. A conventional shell vial culture assay was done on the same blood specimens. Anti-CMV antibodies (IgG and IgM) were measured before and after transplantation using a semiautomated immunofluorescence test (FLAX test system, Whittaker Bioproducts, Inc., Walkersville, MD). IgG titers >20 were considered positive. CMV IgG was also the routine test used for donor screening. Intravenous ganciclovir therapy (Cytovene; Syntex Pharmaceutical Ltd., Palo Alto, CA) was started only with the first detection of pp65 antigenemia, and continued until pp65 clearance. The ganciclovir dose was governed by creatinine clearance: >50 ml/min → 5 mg/kg of ganciclovir twice daily, 10-50 ml/min → 5 mg/kg every 24-48 hr, and <10 ml/min → 5 mg/kg every 48-96 hr. As expected, 17 (85%) of the 20 patients had positive pp65 tests (CMV antigenemia) after a median posttransplant interval of 34 days (range, 9-83 days). Positive shell vial CMV cultures (viremia) were obtained at the same time as the first antigenemia diagnosis in only 5 of these 17 cases. Twelve of the 17 patients were asymptomatic when the first positive pp65 test was obtained, including a recipient who was diagnosed only 9 days after transplantation (Fig. 1). The pp65 levels were generally low: 20±28 (SD) per 2×105 leukocytes. However, 6 of the 12 had further rises in antigenemia for 1 to 12 days after therapy was started, to a median peak of 73 stained PBLs (range, 29-446 PBLs), before antigenemia began to clear. An example is shown in Figure 1. Ganciclovir instituted at the time of the first antigenemia prevented clinical disease in 11 of the 12 infected patients; the exceptional patient had a transient fever. The five other patients had CMV hepatitis at the time antigenemia was detected 12-83 days after surgery (median, 34 days). The median number of pp65-positive cells was 475 (range, 177-872). Antigenemia was cleared with intravenous ganciclovir therapy in all 17 patients but recurred in 10 (59%) patients 5-174 days (median, 57 days) after the drug had been discontinued. Recurrence was independent of IgG conversion(P=0.26). The only patient who was symptomatic with recurrence had 430 pp65-stained cells, which cleared after resumption of ganciclovir. The nine asymptomatic patients had 3-101 pp65-stained cells. Ganciclovir retreatment promptly cleared the antigenemia in the recipient with 101 stained cells. Retreatment was withheld in the other eight patients whose samples had<100 stained cells, and the antigenemia resolved spontaneously in all. In these cases, the antigen assay was useful as a signal to begin preemptive antiviral treatment, as envisioned by Rubin(2), as a means to decide whether to reinstitute ganciclovir in the event of a recurrence (previously emphasized by The et al.[6] and as a guide to individualize the duration of a ganciclovir course in all of these circumstances. The assay also can be an important differential diagnostic tool. The persistence of symptoms despite clearance of the CMV pp65 antigen is an unambiguous warning of co-infection by another pathogen. Conversely, failure of antigen clearance should raise suspicion of ganciclovir-resistant CMV mutants. The pp65 findings are specific. With a novel method for quantitation of CMV DNA in leukocytes, a highly significant correlation was demonstrated between leukoDNAemia and the presence and quantity of pp65 antigenemia (13). The sensitivity and quantitation nature of the pp65 test allowed new observations about response to therapy, such as the reason for increases in pp65 levels seen during the first few days after beginning ganciclovir in half of the preemptively treated patients. This may have reflected a delay in suppression of viral replication or, alternatively, phagocytosis by PBLs of degrading CMV matrix protein from lysed, infected leukocytes(14) or from the ballooned and infected endothelial cells that shed into the circulation (15). Acknowledgments. We thank Elena Percivalle for assistance in developing the antigenemia assay, Drs. Giuseppe Gerna (Pavia) and Robert Rubin(Boston) for helpful consultations, and the clinical virology staff for technical support.Figure 1: Preemptive ganciclovir treatment guided with CMV antigenemia in an asymptomatic, previously seronegative liver transplantation recipient whose donor was CMV positive.

Coagulopathy is a life-threatening complication of liver cirrhosis. We describe the effect of molecular adsorbent recirculating system (MARS), a cell-free dialysis technique, on the blood coagulation of cirrhotic patients. From February 2002 to July 2002, nine patients--five males (55.5%) and four females (44.4%), age 47-70 yr (median 56)--underwent 12 courses (4-7 sessions each) of MARS. Patients were treated for the following indications: six (66.6%) acute-on-chronic hepatic failure, three (33.3%) intractable pruritus. Platelet count, prothrombin time (PT), international standardized ratio and thromboelastography were measured before and after each MARS session. Coagulation factors II, V, VII, VIII, IX, X, XI, XII, XIII, von Willebrand, lupus anticoagulant, protein C, protein S, antithrombin III, plasminogen, alpha 2 antiplasmin, D-dimer, fibrin monomers, complement, and C(1) inactivator were measured before and at the end of each MARS treatment. We found a statistically significant difference (p < 0.05) in the platelet count, PT, all the thromboelastograph variables (reaction and constant time, alpha angle, and maximal amplitude), factor VIII, von Willebrand, and D-dimer, when measured before and after MARS. Previous reports have shown amelioration of blood coagulation following MARS treatments. However, we document that MARS induces coagulopathy through a platelet-mediated mechanism, whereby platelet may be mechanically destroyed during the passage of blood through the filters and lines. An alternative postulated mechanism is an immune-mediated platelet disruption - coagulopathy.