Sixteen RCTs (3,734 patients) were included. Of these, 6 RCTs (1,373 patients) did not involve any bone marrow transplant patients.
Significant statistical heterogeneity was found across studies for the outcomes of the empiric use of amphotericin-B and the colonisation of T glabrata. A random-effects model was used to pool the data for these outcome, whilst a fixed-effect model was used for all other outcomes.
Fungal-related deaths. Fungal-related deaths were significantly less frequent for those patients receiving fluconazole; the pooled OR was 0.45 (95% CI: 0.29, 0.72). There was no significant difference between the treatment groups in non-bone marrow transplant trials; the pooled OR was 0.91 (95% CI: 0.30, 2.82).
Proven systemic fungal infection. Systemic fungal infections were significantly less frequent in those patients receiving fluconazole; the pooled OR was 0.42 (95% CI: 0.31, 0.57). There was no significant difference between the treatment groups in non-bone marrow transplant trials; the pooled OR was 0.85 (95% CI: 0.47, 1.55). Systematic fungal infection was reduced only in those studies with an incidence of fungal infection in the control group of greater than 15%. The ORs were 0.78 (95% CI: 0.50, 1.21) and 0.23 (95% CI: 0.15, 0.36) for an incidence of less than 15% and greater than 15%, respectively.
Superficial fungal infections.
Superficial fungal infections were significantly less frequent in those patients receiving fluconazole; the pooled OR was 0.23 (95% CI: 0.17, 0.31). Similar ORs were found for non-bone marrow patients and for patients receiving high- and low-dose fluconazole.
Empiric use of amphotericin-B.
The criteria for use varied between studies. Empiric use of amphotericin-B was significantly less frequent for those patients receiving fluconazole; the pooled OR was 0.76 (95% CI: 0.60, 0.96). There was no significant difference between the treatment groups in non-bone marrow transplant trials; the pooled OR was 0.90 (95% CI: 0.56, 1.45), but after excluding one study, this was changed to 0.70 (95% CI: 0.53, 0.92). The use of amophotericin-B was reduced in high- and low-dose fluconazole studies, albeit not significantly.
Systemic fungal infection caused by Aspergillus species.
There was no significant difference between the treatment groups for the following: all studies combined; non-bone marrow studies only; and high- or low-dose fluconazole studies. The overall pooled OR was 1.24 (95% CI: 0.71, 2.18).
Systemic fungal infection caused by C. krusei or T. glabrata.
The overall pooled OR for C. krusei or T. glabrata was 0.88 (95% CI: 0.46, 1.68).
Colonisation by C. krusei and T. glabrata. C. krusei and T. glabrata were significantly more common in patients treated with fluconazole. The pooled OR was 2.01 (95% CI: 1.30, 3.12) for C. krusei and 2.18 (95% CI: 1.17, 4.08) for T. glabrata. There was no significant difference in the detection rates of C. krusei between the treatment groups for high- or low-dose fluconazole trials. T. glabrata was significantly more common in patients treated with low-dose, but not high-dose, fluconazole; the OR for low-dose fluconazole was 6.30 (95% CI: 3.87, 10.35). There was no significant difference between the treatment groups in terms of T. glabrata colonisation; the OR was 2.61 (95% CI: 0.64, 10.72). After excluding one study, the OR for colonisation by T. glabrata in non-bone marrow studies was 6.39 (95% CI: 1.53, 26.6).
Adverse effects.
The definition of adverse effects differed between the trials. Generally, the overall frequency was similar in the control and experimental groups. Fluconazole patients developed abnormalities in liver enzyme tests more frequently, but these were reversible. Patient compliance exceeded 90% in the fluconazole groups, compared with 70 to 90% in the control groups.
The authors reported that the major shortcoming of this review was the heterogeneity between trials in terms of drug dosage and patient populations, especially in the degree of immunosuppression.