The authors of the paper entitled ‘Time to platelet recovery predicts outcome of patients with de novo acute lymphoblastic leukaemia who have achieved a complete remission’, published in volume 117, number 4, pp. 869–874, wish to point out the presence of an error in Table I, which in turn caused Fig. 3A to be misleading. The estimated survival probabilities in this figure were calculated using an assumed median age of 50 years instead of the actual median age of 38 years. This caused the predicted curves to appear lower than expected. In Table I, the reported median age and range should be changed from 50 (20–80) to 38 (13–78). The corrected figures are presented below. These corrections do not change any other portion of the paper or alter any of its conclusions. In patients treated with hyper-CVAD, the overall survival probability at 5 years is 39%.

Figure 3.

Survival probabilities of Philadelphia chromosome-negative (A) or -positive (B) patients with time to platelet recovery of 12, 24, 36 or 48 d. Number of patients at risk at 52, 104, 156 and 208 weeks are denoted on the predicted survival curve for TPR = 12 d.

The authors of the paper entitled ‘Guidelines for the use of platelet transfusions’ published in volume 121, pp. 10–23 wish to point out the presence of two errors on page 11 under the following sections:

(1) Neonatal platelets

Components should contain > 40 × 109 platelets/unit, not > 40 × 109/l platelets.

(2) Platelets for intrauterine transfusion (IUT)

Components should contain <0·0025 × 109 leucocytes/unit, not <0·0025 × 109/L leucocytes per donation and > 120 × 109 platelets in 60 ml of plasma, not >120 × 109/l platelets in 60 ml plasma.