0041-1337/99/6701-94$03.00/0 TRANSPLANTATION Vol. 67, 94 –97, No. 1, January 15, 1999 Copyright © 1999 by Lippincott Williams & Wilkins Printed in U.S.A. CYTOMEGALOVIRUS DISEASE RECURRENCE AFTER GANCICLOVIR TREATMENT IN KIDNEY AND KIDNEY-PANCREAS TRANSPLANT RECIPIENTS ABHINAV HUMAR, 1 MARC UKNIS,CASSANDRA CARLONE-JAMBOR,RAINER W. GRUESSNER, DAVID L. DUNN, AND ARTHUR MATAS Department of Surgery, University of Minnesota, Minneapolis, Minnesota 55455 Background. With the introduction of ganciclovir, the clinical pattern of cytomegalovirus (CMV) disease has changed; CMV disease recurrence after successful treatment of the initial episode has emerged as a more common problem. We studied CMV disease recurrence in kidney transplant (KTx) and simultaneous kidney- pancreas transplant (SPK) recipients, and identified risk factors for recurrence. Methods. Between January 1987 and December 1995, of 1272 KTx and 287 SPK recipients, 332 developed CMV disease and were treated with a 14-day course of i.v. ganciclovir, followed by a 10-week course of oral acyclovir. Among these 332 recipients, 103 (31%) devel- oped CMV disease recurrence more than 30 days after treatment for the initial episode; this group was com- pared with those recipients who did not develop re- currence (n229). Risk factors examined were age, presence of diabetes, type of transplant (KTx vs. SPK), donor source (cadaver vs. living donor), treatment for acute rejection, pretransplant CMV serologic status, evidence of tissue-invasive CMV, and treatment of the initial episode with human immune globulin in addi- tion to ganciclovir. Results. Univariate analysis found that patients with recurrence were more likely to be diabetic (70.9% vs. 53.7%; P0.04), to have undergone an SPK (39.8% vs. 20.5%; P0.004), to have received a cadaver organ (78.6% vs. 61.6%; P0.002), and to have received treat- ment for acute rejection (78.6% vs. 59.8%; P0.001). Using multivariate analysis, two statistically signifi- cant risk factors were found: receiving a cadaver or- gan (relative risk [RR]1.90; P0.03) and treatment for acute rejection (RR2.02; P0.008). Diabetes (RR1.44; P0.18) and a cadaver SPK transplant (RR1.55; P0.12) tended toward increased risk for recurrence, but the difference did not reach statistical significance. The remaining variables were not signif- icant. Interestingly, CMV recurrence did not signifi- cantly diminish 5-year graft survival (52.0% vs. 54.4%; P not significant) or patient survival (67.0% vs. 68.3%; P not significant) rates. Conclusions. CMV disease recurs in roughly one- third of KTx and SPK recipients after treatment of the initial episode with ganciclovir. A cadaver organ source and treatment for acute rejection were the most significant clinical risk factors for recurrence. Clinical predictors of recurrence such as these may help to identify those recipients who need more inten- sive therapeutic and prophylactic regimens. Subsequent to the introduction of potent antiviral agents, most notably ganciclovir, the clinical pattern of cytomegalo- virus (CMV*) disease has changed. Formerly, CMV disease was a major source of morbidity and mortality for immuno- suppressed transplant recipients. Some early studies docu- mented that CMV disease was associated with decreased graft and patient survival rates (1, 2). Ganciclovir is an effective treatment for CMV disease in transplant recipients (3) and has dramatically changed the prognosis of this dis- ease. However, CMV disease recurrence after successful treatment of the initial episode has emerged as a common problem. Ganciclovir only temporarily suppresses viral rep- lication, so it is not difficult to see how CMV disease may recur after treatment ends, particularly with ongoing immu- nosuppression. Several studies have reported recurrence rates in abdominal and thoracic organ transplant recipients and have identified a number of risk factors associated with a higher risk for recurrence (4 –14). Similarly, the purpose of this study was to evaluate our current rates of CMV disease recurrence and identify clinical risk factors that place certain recipients at increased risk for recurrence. MATERIALS AND METHODS Only recipients treated with ganciclovir for their first episode of CMV disease were included in this study. Information was obtained from our database (Dataease, Software Solution) for all kidney trans- plant (KTx; n=1272) and simultaneous kidney-pancreas transplant (SPK; n=287) recipients who underwent transplantation between January 1987 and December 1995. During this period, 332 recipients were treated for CMV disease with a 14-day course of i.v. ganciclovir followed by 10 weeks of oral acyclovir. Of these, 103 had disease recurrence more than 30 days after the initial episode, and 229 did not. Data were collected on age, sex, underlying disease, CMV status before transplantation, and characteristics of the initial CMV epi- sode. We defined CMV disease by identification of the virus coupled with the presence of clinical symptoms or signs. Identification meth- ods included (1) culture of the virus using the standard fibroblast technique, (2) positive centrifugation-enhanced shell vial rapid anti- gen detection assay, and (3) histologic observation of characteristic CMV inclusion bodies. Clinical symptoms or signs included fever, chills, leukopenia (white blood cell count3.010 9 /L), malaise, dys- pnea, cough, abdominal pain, nausea, and vomiting. Tissue-invasive CMV disease (TI-CMV) was diagnosed if symptoms were present and the virus was evident in appropriately obtained tissue samples. Specimens considered adequate for this purpose were those obtained 1 Address correspondence to: Abhinav Humar, M.D., Department of Surgery, University of Minnesota, 420 Delaware Street SE, Box 1 Mayo, Minneapolis, MN 55455. * Abbreviations: ALG, antilymphocyte globulin; CMV, cytomega- lovirus; D, donor; HIg, human immune globulin; KTx, kidney trans- plant; R, recipient; RR, relative risk; SPK, simultaneous kidney- pancreas transplant; TI-CMV, tissue-invasive CMV disease. 94