Twnety-six studies were presented in a table, including 21 randomised studies (7 of which were blinded) and 14 observational studies. The exact number included was unclear.
1. Seropositive recipient; donor seropositive or seronegative; immunosuppression with antilymphocytic products 3 RCTs in renal transplant patients): benefit was reported for both ganciclovir and acyclovir. 2 good quality RCTs (171 patients) compared ganciclovir with no therapy and reported RR = 0.27 (95% CI: 0.12, 0.64) and RR = 0.84(95% CI: 0.04, 0.65). Side effect was reported in 1/40 patient in one RCT. One low quality RCT reported benefit in patients treated with high doses of acyclovir (no further details were presented): RR = 0.26 (95% CI: 0.09, 0.75). 5 RCTs in other solid organ transplants supported this.
2. Seronegative recipient; donor seropositive; immunosuppression with antilymphocytic products (2 RCTs in renal transplant patients): Significant benefit was reported for oral acyclovir and a positive but non-significant trend for ganciclovir. One small RCT (32 patients) found no significant difference between ganciclovir compared to no therapy with RR = 0.64 (95% CI: 0.36, 1.16). One RCT (number of patients not reported) found significant benefit for oral acyclovir compared to placebo with RR = 0.26 (95% CI: 0.09, 0.75). No comparison studies were identified.
Other evidence was considered from extrapolation from 2 studies in respondent for CMV positive and donor negative for CMV + and D +/-) renal transplant patients and RCTs in heart and liver transplant patients.
3. Seronegative recipient; donor seropositive; conventional immunosuppression: (1 RCT, 99 renal transplant patients): High exclusion rate post randomisation makes conclusions indeterminate (40 out of 99 patient excluded). Study reported a significant reduction in CMV associated syndromes in patients treated with passive immunisation compared to no-treatment with RR = 0.35. Using a worst case and best case scenario in an intention-to-treat analysis gave RR ranges from 1.75 to 0.10 and NNT ranges from -4 to 1.8.
Risk of disease was estimated at 10% to 40% across 7 trials.
4. Seronegative recipient; donor seronegative; any immunosuppressive regime: No trials of treatment were identified. Low prevalence of disease was reported in 9 observational or cohort studies.
5. Seropositive recipient; donor seropositive or seronegative; conventional immunosuppression: few studies have examined the effect of different regimes.
Sub-group analysis from 1 RCT in renal transplant patients (number of patients not reported) suggested benefit for acyclovir over placebo but the strength of evidence was insufficient to make any recommendation.