Sir: Length of red blood cell (RBC) storage has been proposed a contributing factor to adverse outcome after RBC transfusion. This may be explained by an impaired ability to transport or deliver oxygen, or the presence of leukocytes in stored RBC preparations producing potential deleterious pro-inflammatory mediators or bioactive lipids [1]. We hypothesized that length of RBC storage might explain our previously observed independent association between RBC transfusion and increased mortality, duration of mechanical ventilation (MV) and length of paediatric intensive care unit (PICU) stay in critically ill children [2].For each time a patient was transfused in our previous study, length of RBC storage was retrieved from our hospital's blood bank [2]. Since we observed a dose-outcome relation between the number of RBC transfusions and mortality, we separately looked at single transfusion and multiple transfusions. The effect of RBC transfusion on oxygenation was assessed by the oxygenation index (OI, mean airway pressure times FiO 2 divided by PaO 2 ) and PaO 2 /FiO 2 ratio. RBC preparations were leukocyte-depleted, but not irradiated. Statistical analysis was performed using the Student t-test; linear regression analysis was applied to study correlations. P \ 0.05 was accepted as statistically significant.Data of 295 patients were studied, of whom 67 (22.7%) were transfused; 39 (58.2%) were transfused only once, the remaining 28 received multiple transfusions at different time intervals (range 2-14) (Table 1). Seventeen (5.8%) patients died. The mean length of RBC storage was 16.7 ± 0.6 days [25-75% interquartile range (IQR) 9-23 days]. Length of RBC storage was comparable between survivors and non-survivors. This remained after comparing single versus multiple transfusions. Differences in OI and PaO 2 /FiO 2 ratio between the first day and fifth day after transfusion, duration of MV and length of PICU stay were not correlated with length of RBC storage.Our findings indicate that our previous reported observed independent association between transfusion of leucocyte-depleted RBC preparations and increased morbidity in critically ill children could not be explained by length of RBC storage. Mortality was also not influenced by length of RBC storage, although our study might not be powered to detect this. To our knowledge, although the issue of RBC transfusion has been extensively studied in critically ill adults, no other paediatric data on length of RBC storage and outcome in critically ill children have been reported. Results from retrospective and prospective Mean length of RBC storage (days ± SEM) 14.7 ± 1.4 13.5 ± 3.0 15.8 ± 1.5 16.0 ± 3.5 DOI difference in oxygenation index, DPaO 2 /FiO 2 difference in PaO 2 /FiO 2 ratio, RBC red blood cell, SEM standard error of the mean Intensive