The occurrence of cytomegalovirus infection after solid organ transplantation has been correlated with decrease patient and allograft survival. The disease has not been conquered for two majors reasons: the length of time to establish the diagnosis of CMV has been excessive, and suitable, nontoxic antiviral agents have not been available for use. The purpose of this study was to examine the current incidence and impact of tissue-invasive cytomegalovirus (TI-CMV) disease that developed in 93 patients who underwent solid organ transplantation at University of Minnesota Hospitals (3/1/87 and 6/30/89) and who were treated with antiviral agent ganciclovir ( [9-(1,3-dihydroxy-2-2-propoxymethyl)-guanine [DHPG]). During this same period of time 323 patients received kidney transplants and 71 received kidney-pancreas transplants. Three patient groups were defined: (1) no CMV; (2) CMV infection (cultural or serologic evidence of noninvasive CMV infection); and (3) evidence of TI-CMV disease based upon initial complaints of fever, malaise, dyspnea, or abdominal pain, leukopenia (WBC less than 3000/ml), and evidence of a positive CMV rapid antigen test, CMV culture, or the presence of characteristic CMV inclusion bodies upon examination of material obtained by means of bronchoscopy, upper-gastrointestinal endoscopy, colonoscopy, or liver or renal biopsy. Patients with solely fever, leukopenia, but without a rising CMV serum titer, or positive CMV urine or blood cultures were excluded from the study. A multivariate analysis revealed that rejection therapy, age greater than 50 years, and receiving an organ from a seropositive donor were all significant variables that predisposed to TI-CMV. Analysis of patient and kidney allograft survival indicated that asymptomatic CMV infection had little current impact upon patient or allograft survival, while patients who developed TI-CMV exhibited higher rates of allograft loss and mortality, despite DHPG therapy. Comparison with historical group of patients indicated that TI-CMV DHPG-treated patients exhibited a trend toward improved allograft survival that may be relevant because the historical group of patients included patients with mild CMV infection. DHPG therapy was well tolerated and produced minimal toxicity, and excellent 30-day cure rates (89.2%), although 21.2% of patients required retreatment subsequently. We are currently conducting a trial to compare the ability of DHPG administered plus an anti-CMV immune globulin preparation with acyclovir to prevent posttransplant TI-CMV disease.