E. Di Bona et al., Early haemorrhagic morbidity and mortality during remission induction withor without all-trans retinoic acid in acute promyelocytic leukaemia, BR J HAEM, 108(4), 2000, pp. 689-695
A total of 622 consecutive patients with acute promyelocytic leukaemia (APL
) treated within the Gruppo Italiano per re Malattie Ematologiche dell'Adul
to (GIMEMA) group during 1989-97 have been reviewed to assess the clinical
effectiveness of all-trans retinoic acid (ATRA) on the incidence of early h
aemorrhagic deaths and on APL-associated coagulopathy. Of them, 499 were tr
eated with idarubicin plus ATRA (study A) and 123 with Idarubicin alone (st
udy B), In both studies, similar guidelines for supportive treatment were u
sed, Haemorrhagic symptoms were evaluated according to a reproducible score
system. Deaths occurring within 10 d of starting treatment were 19 (3.8%)
in study A and nine (7.3%) in study B (P = 0.09), with 15 (3%) and five (4.
1%) (P not significant) due to haemorrhage, Overall, induction mortality wa
s 7.6% and 16.2% respectively (P < 0.003). In study A, days with platelet c
ounts less than or equal to 20 x 10(9)/l or with fibrinogen less than or eq
ual to 1 g/L were reduced by about 30%, the haemorrhagic score by 50% and t
he consumption of blood products by about 40%, and fewer patients were trea
ted with antihaemorrhagic drugs (39% vs. 61%; P < 0.001). On multivariate a
nalysis, early deaths were influenced by blast count at diagnosis >30 x 10(
9)/l (P < 0.001) in both studies, and by a haemorrhagic score of 3 in study
A (P < 0.001).
Although the reduction of early fatal haemorrhages was not significant, a s
ubstantial clinical improvement was evident in terms of reduction of the se
verity of bleeding symptoms, blood product consumption and overall inductio
n mortality when ATRA was combined with idarubicin.