Ericka Samantha Teleg, MD
Early Neurological Deterioration after thrombolysis, or END, is defined in many ways, and as the authors have emphasized currently, this term is controversial. This is so because researchers have not been consistent in applying the definitions. This is attributed to the many factors that contribute to its variability as they have enumerated, including heterogeneity and difference in the temporal timing of assessments after stroke and the existing definitions of the different mechanisms of ischemic stroke.
Therefore, this paper aims to describe the END rate in patients with ischemic stroke post-thrombolysis and its relation to potential clinical predictors. It uses the Safe Implementation of treatments in Stroke International Stroke Thrombolysis Registry (SITS-ISTR), which is a multinational open registry of patients with acute ischemic stroke after IV rtPA treatment.
This is retrospective in nature wherein patient data obtained from the SITS-ISTR was recorded between January 2010 and June 2017. Their analyses included the following variables: age, sex, total NIHSS, time logistics, medical history, stroke subtype diagnosis and imaging data on admission and 3-month follow-up. Using the criteria and definition of END as an increase in NIHSS score >4 including death within 24 hours from baseline NIHSS, eligible patients were screened for intracerebral hemorrhage using the hemorrhagic infarction definitions: HI1, hemorrhagic infarction type I, type 2; local or remote parenchymal hemorrhage type 1 and 2 on a timeline of between 22 to 36 hours after rtPA treatment. It is important to note that these were included in the study. Important exclusion criterion included: hemorrhagic stroke on admission; history of anticoagulation use; subjects who underwent endovascular treatment, stenting or angioplasty or treatment beyond 4.5 hours.
The main results included 50,726 patients. And the incidence of END is 3415 out of 50,726 (6.7% [95% CI, 6.5%-7.0%]). The investigators used the following statistical methods: chi square test for categorical variables, Mann-Whitney U test for comparison of two groups, binary logistic regression followed by multivariate logistic regression of significant (P<0.05) univariate predictors.
The rate of END was 3,415 (6.7% [95% CI, 6.5-7.0]) in patients undergoing IV rTPA treatment seen exclusively among patients described as having ischemic stroke vs that of TIA or stroke mimics, which was: 3349/3415 (98% [95% CI, 97.6–98.5]) versus 65/3415 (1.9% [95% CI, 1.5–2.4]). Their findings are lower than published evidence of END rates after IV rtPA from 5.8% to 29.8%. The main findings of this paper showed that the diagnosis of ischemic stroke is a significant predictor of END. The risk is less in patients who achieve a faster and a full neurological improvement after thrombolysis. The risk factors of END in the population include the presence of diabetes mellitus, congestive heart failure, hypertension and atrial fibrillation. All these may contribute to the cerebral perfusion, compromising and slowing down blood flow.
The paper was able to establish novel findings. There is a strong association between the occurrence of END in patients with heart failure or atrial fibrillation. Also, their analyses show a negative association between patients admitted on oral antihypertensive drugs and END after thrombolysis (OR, 0.85 [95% CI, 0.76-0.94], P=0.001). There was also a negative association between END and current smoking status (OR 0.89 [95% CI, 0.79-0.99] P=0.05). However, this result is to be interpreted with caution as their cohort disregarded any previous smoking exposure. Lastly, there is a high association of END with disability and mortality at 90 days.
In conclusion, their work enables the readers to anticipate the occurrence of END and mainly investigate several factors that may contribute to worsening of stroke outcomes. Although pathophysiology is not well defined, the stroke milieu is overall contributing to either improvement, worsening or the temporal profile of a fluctuating stroke after thrombolysis. The authors have provided these limitations: Being retrospective in nature, incomplete data from other investigations, such as angiography, echocardiogram, and other radiological evaluations, failed to establish some possible mechanisms. The registry had no final say on clinical outcomes. Hence, selection bias may have influenced the outcome of patients who have died.
