From April 2016 to December 2018, 512 patients were enrolled in this study. One hundred and seventy-eight patients were excluded for the reasons below: posterior circulation infarct (n = 149), systemic inflammatory disease (n = 3), renal failure (n = 16), hepatic failure (n = 4), brain tumor (n = 1) and presence of an active infection (n = 5). A total of 334 subjects (210 men; mean age, 69.0 ± 11.2 years) were included for the final analysis. Among these patients, 229 (68.6%) had hypertension, 67 (20.1%) had diabetes mellitus, 100 (29.9%) had dyslipidemia, 75 (22.5%) had coronary heart disease, and 99 (29.6%) had atrial fibrillation.
Among the 334 patients enrolled, sICH was observed in 37 patients (11.1%). Table 1 showed the comparisons of baseline characteristics in patients with or without sICH after endovascular therapy. Compared with the patients without sICH after endovascular therapy, those patients with sICH were older (P = 0.016) and had higher proportions of hypertension (P = 0.014) as well as previous antiplatelet use (P = 0.003); Higher levels of SBP (P = 0.020), initial NIHSS score (P = 0.043), the number of thrombectomy device passes (P = 0.048), fasting blood glucose (P = 0.001) and HAR (P = 0.001).
Table 1
Characteristics of subgroups based on the presence of sICH
Variable | sICH group(n = 37) | Non-sICH group(n = 297) | P |
Demographic characteristics | | | |
Age, years | 73.2 ± 10.9 | 68.5 ± 11.2 | 0.016 |
Male, % | 23 (62.2) | 187 (63.0) | 0.904 |
Vascular risk factors, % | | | |
Hypertension | 32 (86.5) | 197 (66.3) | 0.014 |
Diabetes mellitus | 9 (24.3) | 58 (19.5) | 0.515 |
Dyslipidemia | 7 (18.9) | 93 (31.3) | 0.132 |
Coronary heart disease | 12 (32.4) | 63 (21.2) | 0.529 |
Atrial fibrillation | 14 (37.8) | 85 (28.6) | 0.076 |
Previous stroke | 9 (24.3) | 70 (23.6) | 0.100 |
Current drinking | 8 (21.6) | 82 (27.6) | 0.421 |
Current smoking | 9 (24.3) | 102 (34.3) | 0.144 |
Clinical data | | | |
Previous antiplatelet, % | 14 (37.8) | 47 (15.8) | 0.003 |
SBP, mmHg | 149.7 ± 22.7 | 139.8 ± 23.9 | 0.017 |
DBP, mmHg | 84.1 ± 17.7 | 85.2 ± 14.6 | 0.668 |
Body mass index, kg/m2 | 24.2 ± 3.8 | 24.1 ± 3.7 | 0.919 |
Initial NIHSS, score | 16 (13, 22) | 15 (11, 19) | 0.043 |
Time from door to puncture, min | 120.0 (90.5, 149.0) | 110.0 (75.0, 140.0) | 0.075 |
Number of thrombectomy device passes | 2 (1, 3) | 2 (1, 3) | 0.048 |
Poor collateral status, % | 23 (62.2) | 135 (45.5) | 0.055 |
Successful recanalization, % | 33 (89.2) | 256 (86.2) | 0.800 |
Stroke subtype, % | | | 0.169 |
LAA | 12 (32.4) | 139 (46.8) | |
CE | 21 (56.8) | 121 (40.7) | |
Other subtype | 4 (10.8) | 37 (12.5) | |
Laboratory data | | | |
TC, mmol/L | 4.2 ± 1.2 | 4.4 ± 1.2 | 0.537 |
TG, mmol/L | 1.1 (0.8, 1.7) | 1.1 (0.8, 1.6) | 0.375 |
LDL, mmol/L | 2.6 (1.9, 3.2) | 2.6 (2.0, 3.2) | 0.469 |
HDL, mmol/L | 1.1 ± 0.3 | 1.1 ± 0.4 | 0.865 |
FBG, mmol/L | 9.0 ± 3.5 | 6.9 ± 2.4 | 0.001 |
Homocysteine, µmol/L | 12.8 (10.7, 16.0) | 12.3 (10.4, 15.8) | 0.588 |
HAR, 10− 3 | 0.54 (0.24, 1.55) | 0.17 (0.09, 0.33) | 0.001 |
Abbreviation: |
NIHSS, national institute of health stroke scale; SBP, Systolic blood pressure; DBP, Diastolic blood pressure; LAA, large-artery atherosclerosis; CE, cardioembolism; TC, total cholesterol; TG, triglyceride; LDL, low-density lipoprotein; HDL, high-density lipoprotein; FBG, fasting blood glucose; HAR, hypersensitive C-reactive protein-albumin ratio. |
Table 2 showed the results of logistic regression analysis for risk factors with sICH after endovascular therapy. Univariate logistic regression analysis exhibited that HAR[odds ratio (OR), 10.994; 95% confidence interval (CI) 4.567–26.463, P = 0.001], age (OR, 1.044; 95% CI 1.008–1.081, P = 0.017), hypertension (OR, 3.249; 95% CI 1.228–1.938, P = 0.018), previous antiplatelet use (OR, 3.225; 95% CI 1.548–6.718, P = 0.002), SBP (OR, 1.017; 95% CI 1.003–1.031, P = 0.019), initial NIHSS score (OR, 1.058; 95% CI 1.011–1.107, P = 0.014), the number of thrombectomy device passes (OR, 1.362; 95% CI 1.002–1.006, P = 0.022) and fasting blood glucose level (OR, 1.273; 95% CI 1.140–1.421, P = 0.001) might be associated with sICH after endovascular therapy. After adjustment for all potential confounders, HAR was identified as an independent risk factor for sICH after endovascular therapy (OR, 12.384; 95% CI 4.379–35.023, P = 0.001). Moreover, fasting blood glucose level (OR, 1.333; 95% CI 1.145–1.552, P = 0.001) and previous antiplatelet use (OR, 4.940; 95% CI 1.920–12.713, P = 0.001) were also related to sICH, independently.
Table 2
Logistic regression analysis for risk factors with symptomatic intracranial hemorrhage after endovascular therapy
Variable | Unadjusted OR (95%CI) | P | | Adjusted OR (95%CI) | P |
Demographic characteristics | | | | | |
Age, years | 1.044 (1.008–1.081) | 0.017 | | 1.028 (0.984–1.074) | 0.215 |
Male | 0.958 (0.473–1.938) | 0.904 | | | |
Vascular risk factors | | | | | |
Hypertension | 3.249 (1.228–8.594) | 0.018 | | 3.478 (0.956–12.657) | 0.059 |
Diabetes mellitus | 1.325 (0.593–2.959) | 0.493 | | | |
Dyslipidemia | 0.512 (0.217–1.208) | 0.126 | | | |
Coronary heart disease | 1.783 (0.849–3.746) | 0.127 | | | |
Atrial fibrillation | 1.518 (0.746–3.089) | 0.249 | | | |
Previous stroke | 1.038 (0.467–2.304) | 0.927 | | | |
Current drinking | 0.720 (0.316–1.640) | 0.434 | | | |
Current smoking | 0.661 (0.278–1.345) | 0.221 | | | |
Clinical data | | | | | |
Previous antiplatelet | 3.225 (1.548–6.718) | 0.002 | | 4.940 (1.920-12.713) | 0.001 |
SBP | 1.017 (1.003–1.031) | 0.019 | | 1.012 (0.995–1.030) | 0.174 |
DBP | 0.995 (0.972–1.018) | 0.667 | | | |
Body mass index | 1.005 (0.916–1.102) | 0.919 | | | |
Initial NIHSS | 1.058 (1.011–1.107) | 0.014 | | 1.033 (0.970–1.101) | 0.306 |
Time from door to puncture | 1.002 (0.998–1.006) | 0.329 | | | |
Number of thrombectomy device passes | 1.362 (1.046–1.772) | 0.022 | | 1.396 (0.982–1.985) | 0.063 |
Poor collateral status | 1.971 (0.976–3.980) | 0.058 | | | |
Successful recanalization | 1.321 (0.445–3.925) | 0.616 | | | |
Stroke subtype, % | | | | | |
LAA | Reference | | | | |
CE | 0.799 (0.243–2.620) | 0.711 | | | |
Other subtype | 1.605 (0.518–4.974) | 0.412 | | | |
Laboratory data | | | | | |
TC | 0.906 (0.663–1.238) | 0.535 | | | |
TG | 1.129 (0.831–1.532) | 0.438 | | | |
HDL | 0.917 (0.338–2.487) | 0.865 | | | |
LDL | 0.906 (0.621–1.322) | 0.609 | | | |
FBG | 1.273 (1.140–1.421) | 0.001 | | 1.333 (0.145–1.552) | 0.001 |
Homocysteine | 1.000 (0.999–1.001) | 0.865 | | | |
HAR | 10.994 (4.567–26.463) | 0.001 | | 12.384 (4.379–35.023) | 0.001 |
Abbreviation: NIHSS, national institute of health stroke scale; SBP, Systolic blood pressure; DBP, Diastolic blood pressure; LAA, large-artery atherosclerosis; CE, cardioembolism; TC, total cholesterol; TG, triglyceride; LDL, low-density lipoprotein; HDL, high-density lipoprotein; FBG, fasting blood glucose; HAR, hypersensitive C-reactive protein-albumin ratio. |
Figure 1 exhibited the ROC curve, which showed the predictive ability of HAR in sICH after endovascular therapy. The optimal cutoff value for HAR as a predictor of sICH was determined as 0.526 × 10− 3 in the ROC curve analysis, which yielded a sensitivity of 54.1% and a specificity of 88.6%, with the AUC at 0.763 (95% CI, 0.714–0.808).
Figure 2 showed the nomogram. The concordance index was 0.890 (P < 0.001). The novel model indicated that higher HAR was an independent indicator of sICH after endovascular therapy. These findings were similar to those findings above, which was obtained previously in the multivariate logistic model and ROC curve analysis.