Cytomegalovirus

Background: Many lung transplant programs employ lengthy regimens of IV ganciclovir therapy to prevent disease due to cytomegalovirus (CMV). In 1994, we introduced a regimen of delayed ganciclovir prophylaxis for CMV infection. This consisted of 2 weeks of IV ganciclovir therapy, initiated 3 to 4 weeks after transplantation, with subsequent viral monitoring and preemptive therapy as needed. When not receiving ganciclovir, patients received oral acyclovir, 800 mg tid, for 6 months. CMV-seronegative patients with seropositive donors also received four doses of CMV hyperimmune globulin. This study analyzes the CMV outcomes of 54 patients who received the delayed regimen compared to 33 historical control subjects who received only acyclovir prophylaxis (n = 28) or oral acyclovir and 2 to 4 weeks of ganciclovir early after transplantation (n = 5).

Methods: CMV detection was by shell vial culture or IgG seroconversion; after 1996, CMV detection was by blood antigenemia. The diagnosis of CMV disease also required a typical clinical syndrome or pathologic evidence of CMV. The main outcome was the actuarial incidence of CMV infection and disease. In order to account for the effect of other important risk factors for CMV infection, the time to CMV infection and disease was also studied as dependant variables in a Cox proportional-hazard analysis, with the delayed regimen and other important risk factors as independent variables.

Results: The delayed regimen reduced the actuarial incidence of CMV infection from 80 to 48% (p < 0.001) and CMV disease from 31 to 10% (p < 0.01). No seropositive patient receiving the delayed regimen developed CMV disease. Twelve of the 54 patients in the study group required additional IV antiviral treatment, but the total use of ganciclovir averaged only 18 days per patient. In a Cox proportional-hazards model, the use of delayed ganciclovir was the only factor that showed a significant association with freedom from CMV infection (hazard ratio [HR], 0.43; 95% confidence interval [CI], 0.24 to 0.75; p = 0.003) and CMV disease (HR, 0.29; 95% CI, 0.10 to 0.86; p = 0.03).

Conclusion: A regimen of CMV prophylaxis employing 2 weeks of IV ganciclovir initiated 3 to 4 weeks after lung transplantation followed by virologic monitoring and preemptive therapy as needed provides good protection against CMV disease.

Key words: cytomegalovirus; ganciclovir; lung transplantation; prophylaxis; transplantation Abbreviations: CI = confidence interval; CMV = cytomegalovirus; HR = hazard ratio; N-RATS/G = Nashville rabbit antithymocyte serum or globulin

**********

Lung transplant recipients are at high risk for disease due to cytomegalovirus (CMV) because of their pharmacologic immunosuppression and the susceptibility of the transplanted lung to CMV infection. (1,2) Disease due to CMV does not occur uniformly after transplantation, but is usually encountered from 30 to 120 days after transplant surgery. This provides an opportunity to administer prophylaxis against the viral infection in a narrow time frame when patients are at highest risk. Ganciclovir has proven efficacy against CMV and is used for either prophylaxis or treatment. Due to the clinical impact of this infection, many lung transplant centers have employed IV ganciclovir combined with CMV hyperimmune globulin as standard antiviral prophylaxis. These regimens usually start immediately following transplantation. Early studies (3-5) that reported on the use of ganciclovir prophylaxis for brief intervals (10 to 35 days) early after transplantation documented a substantial incidence of CMV pneumonitis after prophylaxis was stopped (38 to 100%). Based on these results, most centers administer a protracted course of ganciclovir ([greater than or equal to] 3 months). (5-8) However, prolonged administration of IV ganciclovir is expensive and may be associated with bone marrow suppression, line infection, or ganciclovir resistance. (5,6) Since CMV disease rarely occurs before the second month after transplantation, we postulated that ganciclovir could be introduced late in the first posttransplant month, thereby focusing antiviral management during the period of highest risk. In 1994, we introduced a standard 2-week course of parenteral ganciclovir beginning 21 to 28 days after lung transplantation in CMV-seropositive patients. The same regimen is also used in seronegative patients with seropositive donors, but combined with doses of CMV hyperimmune globulin administered at 2, 4, 6, and 8 weeks after transplantation. In order to assess the efficacy of this delayed ganciclovir prophylaxis, we reviewed the occurrence of CMV infection and disease in patients receiving this regimen in comparison with patients who underwent transplantation earlier without receiving the regimen. A multivariate analysis was then performed to assess the impact of the delayed regimen on the occurrence of CMV infection and disease with other risk factors, such as patient's CMV serostatus and the intensity of immunosuppression.

MATERIALS AND METHODS

Patient Population

Between March 1990 and July 1998, 102 patients underwent lung transplantation at Vanderbilt University Medical Center. Seventy-seven patients received single-lung transplants and 25 received double-lung transplants. The hospital charts and pre-transplant CMV serostatus of all patients were reviewed. Fifty-four patients (18 donor-positive, recipient-positive patients; 21 donor-negative, recipient-positive patients; 4 donor-negative, recipient-positive patients) who received delayed ganciclovir prophylaxis and survived at least 28 days served as the study group. Thirty-three patients (13 donor-positive, recipient-positive patients; 11 donor-negative, recipient-positive patients; 6 donor-positive, recipient-negative patients; 3 donor-negative, recipient-negative patients) who received either no prophylaxis (n = 28) or a brief initial course of ganciclovir (n = 5) served as historical control subjects. These latter five patients were included, as we wanted to perform a cohort study. Fifteen patients were excluded either because they died before day 28 (n = 10) or already had active CMV infection, as shown by positive urine culture findings for CMV on the day of transplantation (n = 5).

Immunosuppression

All patients received preoperative immunosuppression consisting of single doses of parenteral cyclosporine, 5 to 7 mg/kg, and azathioprine, 4 mg/kg. A dose of 500 mg of parenteral methylprednisolone was administered in the operating room after release of the pulmonary arterial clamp, followed by three doses of 125 mg of methylprednisolone administered every 8 h. After the operation, all patients were treated with triple immunosuppression (cyclosporine, azathioprine, and methylprednisolone). In addition, all patients received 7 to 14 days (mean, 11 days) of induction therapy with Nashville rabbit antithymocyte serum or globulin (N-RATS/G; Applied Medical Research; Nashville, TN). Two patients did not tolerate N-RATS/G and received equine antithymocyte globulin (ATGAM; Pharmacia and Upjohn; Kalamazoo, MI). Acute rejection episodes were treated with daily methylprednisolone, 1 g, for 3 consecutive days or an oral prednisone taper. A few patients who had rejection episodes unresponsive to steroid therapy and some patients with obliterative bronchiolitis were treated with a subsequent course of N-RATS/G or muromonab-CD3 (Orthoclone OKT3; Ortho Biotech; Raritan, NJ).

Delayed CMV Prophylaxis

Prior to September 1994, the standard CMV prophylaxis of our center was oral acyclovir, 800 mg tid, until 6 months after transplantation. Of the 33 patients who served as historical control subjects, 28 patients received no ganciclovir prophylaxis, whereas the other 5 patients received a limited course of the drug (15 to 29 days) in the first posttransplant month. Subsequently, a regimen of delayed ganciclovir prophylaxis was established whereby the standard acyclovir prophylaxis was interrupted at 21 to 28 days after transplantation for a 2-week course of parenteral ganciclovir. This interval was chosen rather than a specific day as a practical measure to allow drug initiation on a scheduled clinic day. The delayed ganciclovir prophylaxis regimen consisted of 1 week of induction therapy (5 mg/kg q12h) followed by maintenance dosing for an additional 7 days (5 mg/kg q24h). The doses were reduced according to the recommendations of the manufacturer in patients with renal dysfunction. Whenever possible, the ganciclovir was administered outside the hospital through a peripherally inserted central catheter. After completion of ganciclovir prophylaxis, patients were monitored virologically and some received additional courses of preemptive antiviral therapy. These courses were administered either because the patients were receiving IV antirejection therapy or had positive virologic findings (shell vial culture or blood antigenemia). Twelve of the 54 patients in the study group received additional courses of preemptive IV antiviral therapy, 5 patients for positive virologic findings and 7 patients during treatment for acute rejection. Three of these patients also received foscarnet because they had persistent or rising antigenemia while receiving ganciclovir. The mean duration of all IV antiviral therapy (including the standard 14-day course) for the 54 patients was 18 days during the first 100 postoperative days. When not receiving IV antiviral therapy, all patients received oral acyclovir prophylaxis, 800 mg tid, until 6 months after transplantation. Patients who were seronegative for CMV and had seropositive donors (donor-positive, recipient-negative patients) also received 100 mg/kg of parenteral CMV hyperimmune globulin (MedImmune; Gaithersburg, MD) at postoperative weeks 2, 4, 6, and 8.

Laboratory Monitoring

Before transplantation, CMV IgG titers were determined for all donors and recipients. Follow-up titers were performed in some patients at the discretion of the attending pulmonary physician. All patients were also screened for active CMV infection on the day of transplantation by viral urine culture. Posttransplant surveillance consisted of weekly blood and urine cultures for 12 weeks followed by cultures every 2 to 4 weeks until 6 months. Thereafter, cultures were obtained for clinical indications. In 1996, CMV blood cultures were replaced with testing for CMV antigenemia (Biotest Diagnostics Corporation; Denville, NJ). Specimens obtained from either bronchoscopy or endoscopy were evaluated for CMV by histology, immunoperoxidase studies, and culture.

Definitions and Monitoring

CMV infection was defined as the presence of either a positive culture finding for CMV from any site, or positive blood CMV antigenemia. Additionally, patients with a fourfold or greater rise in IgG antibody to CMV were deemed to be infected. IgM antibodies were not used to diagnose infection. CMV disease was diagnosed only in patients with CMV infection and was defined as: (1) the presence of pathognomonic cytopathic changes consistent with CMV on biopsy specimens or cytology, or (2) a typical viral syndrome consisting of fever for > 3 days without another source and with at least one of the following: atypical lymphocytosis (> 3%), leukopenia (WBC count < 4,000/[micro]L), thrombocytopenia (platelet count < 100,000/[micro]L), and/or elevated transaminases (alanine transaminase > 40 IU). Additional monitoring (either bronchoscopy or endoscopy) was performed as clinically indicated. Bronchoscopy with BAL and biopsy was performed for clinical indications, such as a fall in FE[V.sub.1], fever associated with pulmonary infiltrates, or new, unexplained hypoxemia. Transbronchial biopsy specimens were graded for acute rejection and obliterative bronchiolitis according to the International Society for Heart and Lung Transplantation criteria. (9)

Outcomes and Variables Assessed

The primary outcomes in this study were the incidence of CMV infection and disease in the first year after transplantation. Pretransplant variables that were evaluated include age, gender, pretransplant diagnosis, type of lung transplant, and donor and recipient CMV serostatus. Posttransplant variables included length of initial posttransplant intubation, ischemia time, number of proved rejection episodes, cumulative dose of steroids used to treat rejection, days of antithymocyte globulin administered, and whether the patient received delayed ganciclovir prophylaxis.

Statistical Analysis

[chi square] or Fisher's Exact Test were used to compare categorical data, and the Student's t test was used for continuous variables. Differences were considered significant if the p value was < 0.05. A Cox proportional-hazard analysis was performed using the times to CMV infection and disease as end points. Patients were censored for loss of follow-up due to any cause. All covariates (Tables 1, 2) were entered into the model in a forward stepwise fashion with p < 0.15 needed to enter. Survival curves were generated by the Kaplan-Meier method, and analysis was performed using the log-rank test. All analyses were performed using statistical software (SAS version 7.0; SAS Institute; Cary, NC).

RESULTS

CMV Infection and Disease

In the first year after transplantation, CMV infection and disease occurred in 57% (50 of 87 patients) and 17% (15 of 87 patients) of the study population, respectively. The rate of infection was 46% (25 of 54 patients) in the delayed ganciclovir group and 76% (25 of 33 patients) in the control group (p < 0.01). The rate of CMV disease was 9% (5 of 54 patients) in the delayed ganciclovir group and 30% (10 of 33 patients) in the control group (p < 0.01). All of the patients with CMV disease had pathologic evidence of CMV on BAL cytology or lung biopsy.

Kaplan-Meier Plots

The times to CMV infection and CMV disease in patients who either received or did not receive delayed ganciclovir prophylaxis are shown in Figures 1, 2. Our analysis shows that our prophylactic regimen was associated with a reduced actuarial incidence of CMV infection from 80 to 48% and CMV disease from 31 to 10% in the first year after transplantation. The curves are significantly different by log-rank test (p < 0.001 and p = 0.01, respectively). No deaths due to CMV disease occurred in the delayed ganciclovir group. One patient in the control group had severe primary graft failure, and had fever and CMV inclusions on BAL sample shortly before her death from respiratory failure 43 days after transplantation. The actuarial freedom from bronchiolitis obliterans was not statistically different between the two study' groups (curves not shown; p = 0.94 by log-rank test).

[FIGURES 1-2 OMITTED]

CMV Infection: Univariate Analysis

Table 1 shows the demographic features of the patients studied. Table 2 shows the univariate analysis of variables associated with the development of CMV infection. The receipt of an organ from a CMV-seropositive donor (p = 0.04) and the use of N-RATS/G induction therapy for > 10 days (p = 0.002) were statistically associated with the development of CMV infection. The administration of delayed ganciclovir prophylaxis was inversely associated with the development of CMV infection (p = 0.001). A duration of intubation [greater than or equal to] 3 days was of borderline statistical significance (p = 0.07).

CMV Disease: Univariate Analysis

Table 3 shows the same variables analyzed in relation to the development of CMV disease. In this analysis, donor-positive, recipient-negative serostatus (p = 0.06) and the use of N-RATS/G induction therapy for > 10 days (p = 0.05) were statistically associated with the development of CMV disease. Again, the use of delayed ganciclovir prophylaxis was inversely associated with the development of CMV disease (p = 0.02).

CMV Infection and Disease: Multivariate Analysis

Two Cox proportional-hazards analyses were then performed with time to CMV infection and CMV disease as dependent variables, and the delayed ganciclovir regimen and other potential risk factors for CMV infection and disease as independent variables. These results are shown in Table 4. Only those variables with p < 0.15 on the univariate analysis were included. In this analysis, the only significant variable associated with CMV infection was the use of delayed ganciclovir prophylaxis. The hazard ratio (HR) was 0.43 (95% confidence interval [CI], 0.24 to 0.75; p = 0.003), indicating a lower risk with prophylaxis. A seropositive CMV donor status was of borderline significance (HR, 1.65; 95% CI, 0.91 to 2.97; p = 0.10).

The multivariate analysis of time to CMV disease showed the use of the delayed ganciclovir regimen to be the only statistically significant variable (HR, 0.29; 95% CI, 0.10 to 0.86; p = 0.03). Donor-positive, recipient-negative serostatus was of borderline significance (HR, 2.61; 95% CI, 0.93 to 7.34; p = 0.07).

DISCUSSION

CMV is a cause of serious illness in lung transplant recipients, (2,5,10) and some studies (10-12) suggest a role for CMV in the development of bronchiolitis obliterans. Given the potentially devastating impact of this infection on the lung allograft, most transplant centers administer antiviral prophylaxis for CMV infection. Many centers administer IV ganciclovir for [greater than or equal to] 3 months to suppress CMV infection. (5,6,8) These lengthier regimens have had a greater impact on the rates of CMV disease than shorter courses, but they are expensive, potentially dangerous, and not completely effective. (5) In 1994, our center developed a delayed, abbreviated ganciclovir regimen. Based on the absence of CMV disease in the first transplant month at our center, we thought that delaying the introduction of ganciclovir until 3 to 4 weeks after transplantation would not only be safe, but would potentially permit the host to develop an immune response against CMV. In theory, this emerging immune response could decrease the chance of CMV disease occurring after ganciclovir was discontinued. This early prophylaxis was coupled to a strategy of clinical and virologic monitoring, by which patients could receive additional pre-emptive courses of ganciclovir if they were treated for rejection or if they became positive for CMV by virologic assays.

Our analysis shows that our prophylactic regimen reduced the actuarial incidence of CMV infection from 80 to 48% and the incidence of CMV disease from 31 to 10% in the first year after transplantation (Fig 1, 2). Remarkably, no seropositive patient who received the regimen had CMV disease develop. Finally, the total exposure to IV antiviral therapy was only 18 days per patient in the first 100 days after transplantation, the time period during which patients are at greatest risk for CMV. This finding illustrates that it is not necessary to administer lengthy ganciclovir prophylaxis to most patients, if prophylaxis is timed carefully and frequent virologic monitoring is performed to detect recurrent infection.

To date, most studies involving ganciclovir prophylaxis in lung transplant recipients have focused on the duration of therapy. The majority of regimens described in the literature involve administering oral or parenteral ganciclovir for many weeks, and up to many months. (5,6,8,13,14) This has also been true for recipients of liver transplants, where ganciclovir has been administered for as long as 98 days after transplantation. (15) The use of early ganciclovir restricted to the first posttransplant month has not been uniformly effective in preventing CMV disease after the prophylaxis is stopped, especially in patients with primary infection. For instance, Bailey et al (4) noted an incidence of CMV disease of 86% in high-risk (donor-positive, recipient-negative) patients treated with 2 to 3 weeks of ganciclovir and immune globulin immediately following transplantation. Interestingly, these authors suggested that, in future studies, drug therapy could be delayed to enhance the effect of the drug on actively replicating virus. While it is hazardous to compare our ganciclovir regimen with those described in the literature due to possible differences in definitions and viral monitoring, antiviral regimens administered for 35 to 90 days after lung transplantation were associated with CMV disease rates of 32 to 56%. (6,8,14) Another study (5) reported no cases of CMV disease in a group of 13 lung recipients who received ganciclovir for 90 days after transplantation, but the 95% CIs of that observation range from 0 to 25%. Thus, available data suggest that our regimen produces results that are no worse and possibly better than those achieved with prolonged ganciclovir prophylaxis. In addition, longer ganciclovir prophylaxis is costly and may be associated with problems such as bone marrow suppression or line infection. Antiviral resistance also appears to be an important issue, as Kruger and colleagues (16) showed that 5.2% percent of lung transplant recipients receiving prolonged ganciclovir treatment had viral resistance develop.

Our prophylactic regimen did not consist solely of ganciclovir. We also used high-dose oral acyclovir when patients were not receiving ganciclovir. CMV-seronegative patients with seropositive donors also received CMV Ig. Both acyclovir and CMV Ig have a demonstrated impact on CMV disease. (17,18) However, their antiviral activity is far less potent than ganciclovir. Indeed, CMV disease in transplant populations did not really come under excellent control until ganciclovir became available as a therapeutic and prophylactic agent. Thus, it is likely that the outcomes we describe are primarily due to the ganciclovir regimen.

Our strategy couples a short course of ganciclovir prophylaxis with subsequent preemptive therapy based on antigenemia testing. It may be possible to obtain similar results by omitting the prophylaxis and using a purely preemptive strategy. When we first introduced our regimen, viral cultures were our only tool to monitor patients, but they were not thought to be adequately sensitive. Despite the availability of better monitoring tools, we still adhere to our regimen because it provides certain protection for patients at a time when they are highly vulnerable to CMV and a laboratory error could lead to a severe, yet preventable illness.

Two classic risk factors for CMV disease have been determined in solid-organ transplant recipients: the occurrence of primary infection and the use of high doses of immunosuppression, especially antithymocyte globulin. (19) The results of our univariate analysis are consistent with this, in that CMV disease was more common in patients who received a long course of antithymocyte globulin, and there was a strong trend to significance in CMV-seronegative patients with positive donors. The failure to detect significance in the multivariate analysis is likely only due to the relatively small numbers of patients and the greater stringency of that analysis.

Our study suffers the usual limitations of a retrospective analysis, and it would be desirable to confirm the results in a prospective trial. To date, however, no large, prospective, randomized trial of antiviral prophylaxis has been performed in lung recipients. It is unlikely that untreated patients or placebo recipients would be used in any such future trial due to the high morbidity of CMV infection in lung transplantation. Yet, the low rate of CMV disease, as seen with current prophylactic and pre-emptive strategies (including ours), means that it would be necessary to enroll very large numbers of patients in future trials to show differences in disease outcomes, if patients in both arms of the study were receiving effective antiviral management. Thus, future trials may be limited to showing equivalence between conventional and new antiviral compounds, or assessing pharmacoeconomic end points. Antiviral trials would also be useful to evaluate the impact of different regimens on long-term outcomes such as graft survival or the incidence of bronchiolitis obliterans. Bronchiolitis obliterans is the most important complication of lung transplantation, and a number of studies (9,11,13,20) have linked its occurrence to CMV infection. For instance, Duncan and colleagues (10) found that the actuarial prevalence of chronic rejection at 2 years was 74% in lung transplant recipients with a history of biopsy-proven CMV pneumonitis compared with 22% among CMV-negative patients (p < 0.038). However, it is not known whether this association is causal or whether CMV prophylaxis can have an impact on the incidence of chronic rejection. (14) Our study provides evidence that an abbreviated "minimalist" approach to CMV prophylaxis is effective and safe, but it did not show any effect on the incidence of bronchiolitis obliterans.

Although the delayed ganciclovir regimen has been effective in our population, its ability to suppress CMV may vary depending on the intensity of immunosuppression, and this in turn may differ between institutions. The success of this regimen is predicated on the absence of CMV disease in the first posttransplant month. We have discovered that some patients who are receiving immunosuppression to treat idiopathic pulmonary fibrosis already have positive urine culture findings for CMV on the day of transplantation. (21) Delayed prophylaxis would not be appropriate for such patients, and they should receive ganciclovir as soon as their infection is diagnosed. Performance of a rapid, sensitive test for CMV infection, such as a shell vial culture of urine, is indicated on the day of transplantation in anyone who has been receiving immunosuppressive medication. Even in the majority of patients without CMV infection at the time of transplantation, it is probably advisable to monitor for CMV during the first post-transplant month as insurance against a rare occurrence. In addition to a viral urine culture on the day of transplantation, we perform blood antigenemia at 2, 3, and 4 weeks after transplantation.

Our study provides evidence that ganciclovir prophylaxis in lung transplant recipients can be delayed and does not need to be prolonged in most patients. Current virologic monitoring makes it possible to follow patients safely and individualize their antiviral management after an initial brief course of prophylaxis. (22) It is not known whether a more intense CMV prophylactic regimen could improve long-term outcomes such as graft survival. This is an important issue that should be addressed in future trials.

REFERENCES

(1) Dummer JS, White LT, Ho M, et al. Morbidity of cytomegalovirus infection in recipients of heart or heart-lung transplants who received cyclosporine. J Infect Dis 1985; 152: 1182-1191

(2) Duncan AJ, Dummer JS, Paradis IL, et al. Cytomegalovirus infection and survival in lung transplant recipients. J Heart Lung Transplant 1991; 10:638-646

(3) Duncan SR, Paradis IL, Dauber JH, et al. Ganciclovir prophylaxis for cytomegalovirus infections in pulmonary allograft recipients. Am Rev Respir Dis 1992; 146:1213-1215

(4) Bailey TC, Trulock EP, Ettinger NA, et al. Failure of high-dose oral acyclovir with or without globulin to prevent primary cytomegalovirus disease in recipients of solid organ transplants. Am J Med 1993; 95:273-278

(5) Duncan SR, Grgurich WF, Iacono AT, et al. A comparison of ganciclovir and acyclovir to prevent cytomegalovirus after lung transplantation. Am J Respir Grit Care Med 1994; 150:146-152

(6) Hertz MI, Jordan C, Savik SK, et al. Randomized trial of daily versus three-times-weekly prophylactic ganciclovir after lung and heart-lung transplantation. J Heart Lung Transplant 1998; 17:913-920

(7) Gutierrez CA, Chaparro C, Krajden M, et al. Cytomegalovirus viremia in lung transplant recipients receiving ganciclovir and immune globulin. Chest 1998; 113:924-932

(8) Maurer JR, Snell G, DeHoyos A, et al. Outcomes of lung transplantation using three different cytomegalovirus prophylactic regimens. Transplant Proc 1993; 25:1434-1435

(9) Yousem SA, Berry GJ, Cagle PT, et al. Revision of the 1990 working formulation for the classification of pulmonary allograft rejection: Lung Rejection Study Group. J Heart Lung Transplant 1996; 15:1-15

(10) Duncan SR, Paradis IL, Yousem SA, et al. Sequelae of cytomegalovirus pulmonary infections in lung allograft recipients. Am Rev Respir Dis 1992; 146:1419-1425

(11) Boehler A, Kesten S, Weder W, et al. Bronchiolitis obliterans after lung transplantation. Chest 1998; 114:1411-1426

(12) Keenan RJ, Lega ME, Dummer JS, et al. Cytomegalovirus serologic status and postoperative infection correlated with risk developing chronic rejection after pulmonary transplantation. Transplantation 1991; 51:433-438

(13) Speich R, Thurnheer R, Gaspert A, et al. Efficacy and cost effectiveness of oral ganciclovir in the prevention of cytomegalovirus disease after lung transplantation. Transplantation 1999; 67:315-320

(14) Soghikian MV, Valentine VG, Berry GJ, et al. Impact of ganciclovir prophylaxis on heart-lung and lung transplant recipients. J Heart Lung Transplant 1996; 15:881-887

(15) Gane E, Saliba F, Valdecasas GJC, et al. gandomised trial of efficacy and safety of oral ganciclovir in the prevention of cytomegalovirus disease in liver-transplant recipients. Lancet 1997; 350:1729-1733

(16) Kruger RM, Shannon WD, Arens MQ, et al. The impact of ganciclovir-resistant cytomegalovirus infection after lung transplantation. Transplantation 1999; 68:1272-1279

(17) Balfour HH Jr, Chace BA, Stapleton JT, et al. A randomized, placebo controlled trial of oral acyclovir for the prevention of cytomegalovirus in recipients of renal allografts. N Engl J Med 1989; 320:1381-1387

(18) Snydman DR, Werner BG, Heinze-Lacey B, et al. Use of cytomegalovirus immune globulin to prevent cytomegalovirus disease in renal transplant recipients. N Engl J Med 1987; 317:1049 -1054

(19) Hibberd PL, Snydman DR. Cytomegalovirus infection in organ transplant recipients. Infect Dis Clin North Am 1995; 9:863-877

(20) Girgis RE, Tu I, Berry GJ, et al. Risk factors for the development of obliterative bronchiolitis after lung transplantation. J Heart Lung Transplant 1996; 15:1200-1208

(21) Milstone AP, Brumble LM, Loyd JE, et al. Active cytomegalovirus infection on the day of transplantation: risk factors and clinical implications. J Heart Lung Transplant 2000; 19:744-750

(22) Boeckh M, Boivin G. Quantitation of cytomegalovirus: methodologic aspects and clinical applications. Clin Microbiol Rev 1998; 11:533-554

* From the Divisions of Infectious Diseases (Dr. Brumble), Mayo Clinic, Jacksonville, FL; and Allergy, Pulmonary, and Critical Care Medicine (Drs. Milstone, Loyd, and Ely), Cardiac and Thoracic Surgery (Dr. Pierson), Preventive Medicine and Biostatistics (Dr. Gautam), and Infectious Diseases (Dr. Dummer), Vanderbilt University Medical Center, Nashville, TN.

Manuscript received February 22, 2001; revision accepted July 26, 2001.

Correspondence to: J. Stephen Dummer, MD, Division of Infectious Diseases, Vanderbilt University Medical Center, 911 Oxford House, Nashville, TN 37232; e-mail: Stephen. Dummer@mcmail. vanderbilt.edu

COPYRIGHT 2002 American College of Chest Physicians
COPYRIGHT 2002 Gale Group