WSO - March 2024 - 325

Bala et al.
325
Statistical analyses
Baseline characteristics, imaging variables, and outcomes
were compared between intravenous treatment groups
(tenecteplase vs. alteplase) in patients with tandem lesion.
Analyses were divided into patients treated with combined
therapy (EVT + IVT) and patients treated with IVT alone.
Categorical variables were expressed as frequencies and
percentages, quantitative variables as mean (standard deviation
(SD)) for normal distribution, or median (interquartile
range (IQR)) for non-normal distribution. Fisher's exact
test was used for categorical data and Student t-test or
Wilcoxon `rank` sum test for continuous variables.
Adjusted analyses for functional outcomes were conducted
using mixed-effects logistic regression. Fixedeffects
covariates were age, baseline NIHSS score, stroke
onset-to-arterial access time, and occlusion location at
baseline (ICA, M1-MCA, and M2-MCA), and participating
site was included as a random-effects variable to account
for clustering within each site. Thrombolysis treatment by
tandem lesion interaction was assessed by using multiplicative
interaction terms in the mixed-effects logistic models
of mRS 0-1, mRS 0-2, ordinal analysis of mRS, and death
at 90 days. Effect size estimates were reported as adjusted
odds ratios (aORs) with their 95% confidence intervals
(95% CI) or adjusted common OR for a 1-step shift toward
better functional outcome. For ordinal regression, mRS 5
and 6 were combined and the Brant test was used to assess
the proportional odds assumption.
Furthermore, we reported the outcomes in the tandem
EVT group for patients who underwent cervical carotid
stenting versus those who did not, using descriptive statistics,
and assessed for effect modification of stenting on the
relationship between thrombolysis treatment type and functional
outcomes using the Mantel Haenszel test.
No imputation was performed as missing data were minimal.
No correction for multiple testing was done as all secondary
analyses were considered exploratory.
Statistical significance was defined as two-tailed
p < 0.05. All analyses were performed using Stata/MP 17.0
(STATA LLC, Corp).
Results
Among 1577 AcT patients, 682 had intracranial ICA,
M1-MCA or M2-MCA occlusions. Carotid tandem lesions
were seen in 128/682 (18.8%), and 10/128 (7.8%) had
occlusions due to cervical carotid dissection (Figure 1).
Among the 128 tandem lesion patients, 93 (72.7%)
patients underwent EVT. Compared with patients who
underwent EVT, patients who did not undergo EVT were
older (median 77 years (IQR 65-87) vs. median 67 years
(IQR 58-75), p < 0.01), had lower NIHSS scores (median
10 (IQR 6-18) vs. median 17 (IQR 13-22), p < 0.01),
longer delays from stroke symptom onset to thrombolysis
initiation (median 142 (111-189) min vs. median 110
(87-149) min, p < 0.01), more often had M2-MCA (18
(51.4%) vs. 18 (19.3%) compared with ICA and M1 occlusions
(15 (48.6%) vs. 75 (80.6%) p < 0.01), had a higher
prevalence of poor collateral grade at baseline (5 (14.7%)
vs. 9 (9.7%) p = 0.52) and higher ASPECTS scores (median
10 (IQR 8-10) vs. median 8 (IQR 7-9), p < 0.01)
(Supplemental Table 1). In the EVT subgroup, patients with
tandem lesions were younger, more commonly men, and
more often had terminal ICA occlusion and lower ASPECTS
(Supplemental Table 2).
Tandem patients treated with EVT + IVT
(n = 93)
Baseline characteristics were similar between the thrombolysis
treatment groups (Table 1).
Proportions of mRS 0-1 (23 (46.0%) vs. 14 (32.6%),
p = 0.21) and mRS 0-2 (27 (54.0%) vs. 25 (58.1%), p = 0.83)
at 90-120 days were not significantly different between
tenecteplase and alteplase groups, respectively.
In the multivariable analyses adjusted for age, baseline
NIHSS score, time from onset to arterial access, and occlusion
site, tenecteplase was significantly associated with
higher rates of mRS 0-1 (aOR 3.21; 95% CI = 1.06-9.71).
However, tenecteplase was not associated with mRS 0-2
(aOR 1.53; 95% CI = 0.51-4.55) or change along the full
mRS score (adj. common OR 1.34; 95% CI = 0.61-2.94)
(Figure 2).
On first intracranial angiographic acquisition, there was
no significant difference between the tenecteplase and
alteplase groups for successful intracranial recanalization
(8 (16.3%) vs. 12 (28.6), p = 0.21) and successful reperfusion
(4 (8.2%) vs. 8 (19.0%), p = 0.21). Successful reperfusion
on final angiographic run was also similar (41 (83.7%)
vs. 36 (85.7%), p = 0.99) (Table 2).
Death at 90-120 days occurred in 9 (18.0%, tenecteplase
group) vs. 7 (16.3%, alteplase group), p = 0.99, and aOR
was 0.55; 95% CI = 0.13-2.30). There was only one case
of symptomatic ICH, which occurred in the tenecteplase
group. Rates of intracerebral hemorrhages and procedural
complications were comparable between the thrombolysis
treatment groups. Parenchymal hemorrhage occurred in
two (4.0%) patients in the tenecteplase group versus four
(9.3%) in the alteplase group (p = 0.41).
No thrombolysis treatment effect modification by tandem
lesion was found for all outcomes (p interaction = 0.09
for mRS 0-1, p = 0.95 for mRS 0-2, p = 0.99 for change
along the full mRS, and p = 0.74 for death at 90 days)
(Supplemental Table 3).
Acute treatment of the cervical ICA lesion
among EVT patients
Stenting of cervical ICA was performed in 29 of the 93
patients who underwent EVT (31.2%); 23/29 (79%) with
International Journal of Stroke, 19(3)
WSO - March 2024

Table of Contents for the Digital Edition of WSO - March 2024
