Eighteen studies were included in the review (n=3,172 patients, range 22 to 782); none were randomised or blinded. The demographic, disease and transplant characteristics of the two groups of participants in each study appeared comparable at baseline. Duration of follow up ranged from two to seven years.
Disease-free survival (10 studies; n=2,338 patients): Disease-free survival rates after transplantation were significantly higher in the total body irradiation plus cyclophosphamide group (OR 1.53, 95% CI 1.16 to 2.02, I2=46%), with no significant evidence of publication bias. Results differed by disease, with statistically significant benefit from total body irradiation plus cyclophosphamide for acute lymphoblastic leukaemia (OR 1.93, 95% CI 1.42 to 2.64; three studies; I2=0%) and acute myeloblastic leukaemia (OR 1.49, 95% CI 1.01 to 2.20; four studies; I2=52%), but no statistically significant difference between the treatment regimens for chronic myelocytic leukaemia (three studies).
Secondary outcomes: There was a statistically significant benefit in the total body irradiation plus cyclophosphamide treatment for: transplant-related mortality (OR 0.68, 95% CI 0.49 to 0.93; eight studies; I2=54%), veno-occlusive disease (OR 0.42, 95% CI 0.30 to 0.59; nine studies; I2=0%) and haemorrhagic cystitis (OR 0.32, 95% CI 0.19 to 0.54; four studies; I2=0%). However, total body irradiation plus cyclophosphamide was associated with a significantly higher rate of growth and development problems (OR 5.85, 95% CI 1.55 to 22.13; three studies; I2=0%), interstitial pneumonia (OR 1.70, 95% CI 1.24 to 2.32; seven studies; I2=5%) and cataract (OR 12.69, 95% CI 1.72 to 93.32; three studies; I2=65%) than busulphan plus cyclophosphamide treatment. There was no statistically significant difference between the groups in relapse rate (eight studies), graft failure (seven studies), or graft-versus-host disease. Subgroup analysis showed a significantly lower relapse rate for acute myeloblastic leukaemia in the total body irradiation plus cyclophosphamide group (three studies) and for chronic myelocytic leukaemia in the busulphan plus cyclophosphamide group (three studies).
Sensitivity analyses showed similar results to the main findings.
No significant evidence of publication bias was detected.