Our study population includes 105 patients: 59 (56.19%) males and 46 (43.81%) females. The mean age of patients was 63.3 years (SD ± 14 years, range 46–78 years). 9 patients were excluded because of missing data on biomarkers. Thus, the final study population includes 96 participants. Table
reports descriptive statistics for classical cerebrovascular risk factors stratified by gender. In particular, women are a bit older than man and their levels of erythrocyte sedimentation rate (ESR), high density lipoprotein (HDL), glucose, hypertension prevalence and anterior stroke site distribution are increased as compared to men. In contrast, serum triglycerides (TG), partial thromboplastin time (PTT) and low density lipoprotein (LDL) values of women are lower compared to men. There are no significant differences between the sexes for other variables. The relationship between plasma biomarkers levels and AIS cerebral lesion site and lesion dimension revealed as regression coefficients with relative p-values are shown in Table
AIS variables characteristics in the study population
Independent correlates of principal biomarkers in AIS
As anticipated by significant magnitudes and directions between biomarkers and lesion sites revealed by regression coefficients, given the random variance introduced by different levels of biomarkers among AIS patients, further relationship between biomarker’s quartiles: first quartile (Q1) used as reference to second quartile (Q2), third quartile (Q3), forth quartile (Q4) and lesion location and size were assessed.
ESR serum levels were independently correlated with sex (b = + 0.65, S.E. = 0.24, p = 0.008) and D3 (b = + 0.53, S.E. = 0.27, p = 0.0045) lesion (Table
). In quartile analysis D3 lesion remained associated with Q4 (OR, 5.250; 95% CI, 1.002 - 27.514 ) of ESR (Table
Multivariate testing: OR (95% CI) to the first quartile of biomarkers investigated vs stroke lesion site (anterior vs posterior) & stroke lesion dimensions
Biomarkers OR stratified by AIS lesion site and dimension (D1 - D4). Patients in Q1 were used as a reference in each group. All values were adjusted for age and sex. *Significantly different from Q1.
Fibrinogen serum levels were independently correlated with D3 (b = + 0.58, S.E. = 0.27, p = 0.040) lesion (Table
). Lesion D3 remained associated with Q3 (OR, 5.500; 95% CI, 1.027 - 29.451 p=0.040) of fibrinogen in quartile analysis (Table
). Platelets serum levels were independently correlated with sex (b = + 0.81, S.E. = 0.23, p = 0.001) and D1 (b = + 0.02, S.E. = 0.01, p = 0.003), D2 (b = + 0.68, S.E. = 0.31, p = 0.035) lesions (Table
). In quartile analysis only D3 lesion remained associated with Q3 (OR, 0.059; 95% CI, 0.003 - 1.175 ) of platelets (Table
Albumin levels were independently correlated with site (b = − 0.48, S.E. = 0.23, p = 0.0436) and D1 (b = − 0.03, S.E. = 0.008, p = 0.002), D2 (b = − 0.59, S.E. = 0.25, p = 0.020), D3 (b = −0.55, S.E. = 0.27, p = 0.042), D4 (b = − 0.98, S.E. = 0.47, p = 0.041) size (Table
); lesion D1 size remained associated with Q3 (OR, 5.250; 95% CI, 1.351 - 20.396 ) of serum albumin level, whereas lesion D2 remained associated with Q2 (OR, 0.227; 95% CI, 0.053 - 0.981), Q3 (OR, 0.164; 95% CI, 0.038 - 0.711 ) and Q4 (OR, 0.205 ; 95% CI, 0.048 - 0.870) (Table
TG levels were independently correlated with site (b = + 0.63, S.E. = 0.24, p = 0.008) and D2
(b = + 0.68, S.E. = 0.25, p = 0.009) size (Table
). Lesion D1 was associated with Q3 (OR, 9.000; 95% CI, 2.487 - 32.567) of TG and lesion D2 size remained associated with Q3 (OR, 0.132; 95% CI, 0.037 - 0.471 ) (Table
HDL serum levels were independently correlated with sex (b = + 0.66, S.E. = 0.23, p = 0.005) and glucose levels were independently correlated with D1 (b = + 0.02, S.E. = 0.008, p= 0.018) lesion (Table
In quartile analysis, both HDL and glucose as well as PTT, international normalized ratio (INR), LDL and total cholesterol (TC) (data not shown) were not significantly associated with lesion size dimension.
Finally, high INR (NIHSS ≥ 14, p = 0.01) and high PTT (NIHSS ≥ 7, p = 0.001) were associated with worse clinical outcomes, respectively. In contrast, higher albumin serum level was associated with better clinical outcome (NIHSS < 7, p = 0.006) at 7 days.