Fibrinogen consumption is related to intracranial clot burden in acute ischemic stroke: a retrospective hyperdense artery study

Table 5 Association of on-admission fibrinogen, clot burden represented by clot area, final infarct volume and symptom severity at presentation (NIHSS score) with the risk of in-hospital death or survival with NIHSS score at discharge >15: summary of multivariate analysis

Independents	RR (95 % CI)	p value
Models 1–4: variable of interest + default adjustments^a
Model 1—variable of interest: on-admission fibrinogen
On-admission fibrinogen (by 2.718 fold)^b	0.478 (0.247–0.924)	0.028
Model 2—variable of interest: clot area
Clot area (by 10 mm²)	1.057 (1.013–1.104)	0.010
Model 3—variable of interest: final infarct volume
Final infarct volume (by 10 mm³)	1.030 (1.019–1.043)	<0.001
Model 4—variable of interest: NIHSS at presentation
NIHSS at presentation (by 1 score point)	1.097 (1.071–1.123)	<0.001
Model 5—full model: all variables of interest + adjustments
On-admission fibrinogen (by 2.718-fold)^b	0.790 (0.445–1.401)	0.420
Clot area (by 10 mm²)	0.990 (0.942–1.031)	0.559
Final infarct volume (by 10 mm³)	1.015 (0.998–1.032)	0.084
NIHSS at presentation (by 1 score point)	1.087 (1.060–1.114)	<0.001
Age (by 5 years)	1.065 (0.965–1.176)	0.208
Male gender	1.205 (0.815–1.784)	0.353
Symptom onset to imaging (admission) (by 10 min)	0.997 (0.973–1.021)	0.420
C-reactive protein (by 1 mg/L)	1.032 (0.982–1.084)	0.211
Glucose (by 1 mmol/L)	1.151 (1.041–1.273)	0.006
Proximal vs. distal middle cerebral artery (MCA)	0.397 (0.135–1.145)	0.087
Proximal MCA vs. “non-MCA” artery	2.259 (1.136–4.491)	0.020
Thrombectomy with TICI 2b–3 vs. no thrombectomy	0.700 (0.404–1.213)	0.204
Thrombectomy with TICI 2b–3 vs. TICI 0–2a	0.484 (0.268–0.875)	0.016

Data are presented as relative risks (RR) with 95 % confidence intervals
NIHSS National Institutes of Health Stroke Severity scale
TICI thrombolysis in cerebral infarction grading
^aModels 1–4 each consisted of a variable of primary interest and a set of default adjustments based on their independent associations with the variables of primary interest (Tables 1, 2, 3, 4): age, gender, time elapsed since symptom onset to imaging (reflects admission), C-reactive protein and glucose levels on admission, affected vessel (proximal or distal middle cerebral artery or “non-MCA” artery) and performed thrombectomy (none, with perfusion TICI grade 0–2a or grade 2b–3). Model 5 included all variables of primary interest and all adjustments
^bSince on-admission fibrinogen was ln-transformed (as in all previous models), the “effects” are presented as GMRs by 2.718-fold increase

ISSN: 1479-5876