We compared the incidence of refractory thrombocytopenia (RT) and platelet transfusion requirements (PTR) in 35 children who developed VOD with 35 matched controls who underwent HSCT but did not develop VOD. RT developed in 100% of the VOD patients, at a median of 8 days before VOD diagnosis, as compared to 71.5% of the control group. VOD patients required more
platelets transfusions than controls (median PTR 6.9 ml/kg (range: 0.57-17.59) vs. 3.57 ml/kg (range: 0-14.63) in the control group with the difference being statistically significant (P < 0.0001). Numbers of days with platelet requirement, were significantly higher for VOD patients as compared to controls, (median % of days 68% vs. 39%, P=< .0001). The PTR peaked at approximately 12 mL/kg/day, 2 days before VOD diagnosis, whereas the PTR in the control population was 5 mL/kg/day. The positive predictive value (PPV) of developing VOD was 88.9% (95% CI: 66.5-97%) in patients who were given >7ml/kg/day of platelets during at risk period of days +3 to +13 after transplant. For patients who got >8ml/kg/day pf platelets, the PPV of developing VOD was 86.7% (95%CI: 61.2 -96.4%). There was no difference in the PTR in patients with mild to moderate VOD as compared to those with severe VOD, however PTR was higher in patients whose VOD did not resolve. The median, average, daily PTR after the diagnosis of VOD in 17 patients who got defibrotide as compared to those who did not get defibrotide was 6.04 ml/kg and 5.72 ml/kg respectively and the difference was not statistically significant (p=0.56). On univariate and multivariate analysis use of intravenous immunoglobulin (IVIG) was significantly associated with VOD (p=0.0088), however use of IVIG was not significantly associated with fatal VOD. In conclusion RT occurs in 100% of patients at a median of 8 days before VOD diagnosis. VOD should be suspected in any patient with RT after the exclusion of other causes of consumptive thrombocytopenia especially if they are needing >7ml/kg/day of platelets.