Folie 1

1
Reference number
Criteria forICH diagnosis
Adjusted OR (95CI)
Level of adjustment
Age
Sex
NIHSS
BP
HxStrk
AF
INR
DM
Others
Case series Innsbruck
NINDS
?
-
?
-
-
-
-
?
-
13
NINDS
?
?
?
-
-
?
?
-
-
10
Modified SITS-MOST
-
-
-
-
-
-
-
-
-
14
NINDS
?
?
?
-
?
-
-
-
-
15
NINDS
?
?
?
?
?
?
-
?
?
16
N.R.
-
-
-
-
-
-
-
-
-
17
Modified SITS-MOST
-
-
-
-
-
-
-
-
-
11
N.R.
-
-
-
-
-
-
-
-
-
12
ECASS
-
-
?
?
-
-
?
-
?
Pooled estimate (random effects meta-analysis)
2.20 (0.99, 4.88)
P0.052
Pooled estimate (fixed effect meta-analysis)
1.90 (1.20, 3.01)
P0.006
I2 statistic (95CI)
58 (11, 80)


1
.01
.1
1
10
100
Figure e-2 Adjusted random effects and fixed
effect meta-analyses on the association between
sub-therapeutic warfarin pre-treatment (INR1.7)
and risk of intracranial bleedings. In four
studies including 3451 patients intracranial
bleedings were defined according to NINDS
criteria, in one study according to ECASS III
criteria (n212), in two studies according to a
modification of SITS-MOST criteria (parenchymal
bleedings PH1 and PH2 with clinical deterioration
4 points on the NIHSS score) (n391) and in two
studies no information was reported on the
diagnostic criteria used for the classification
of bleedings (N.R.) (802 patients). Level of
adjustment differs between studies and is
detailed in the figure. If no adjusted ORs were
presented in the publications unadjusted ORs were
applied to allow comparison with the pooled
estimates given in Figure 1. Abbreviations AF,
atrial fibrillation BP, blood pressure DM,
diabetes mellitus ECASS, The European
Cooperative Acute Stroke Study HxStrk, history
of stroke INR, international normalized ratio
NIHSS, National Institutes of Health Stroke
scale NINDS, National Institute of Neurological
Disorders and Stroke N.R., not reported
SITS-MOST, Safe Implementation of Thrombolysis in
Stroke Monitoring Study.