Semiquantitative Assessment of Iodine Extravasation in Acute Ischemic Stroke After Mechanical Thrombectomy
Iodine extravasation is a common observation in stroke patients following successful mechanical thrombectomy. This phenomenon is increasingly recognized for its potential association with patient prognosis, particularly in relation to the development of severe hemorrhagic transformation (HT). The ability to identify early hemorrhage after intra-arterial therapy and the imaging indicators that can predict late hemorrhage transformation is crucial. Various studies have employed different quantitative and non-quantitative methods to evaluate iodine extravasation, yet their conclusions have been inconsistent. The aim of this study was to propose two practical and straightforward semiquantitative methods to assess iodine extravasation and its relationship with hemorrhage transformation.
The study was conducted with the approval of the Ethics Committee of People’s Hospital of Deyang City, and all participants provided informed consent. Ninety-three consecutive patients with anterior circulation obstruction who underwent mechanical thrombectomy were enrolled. The inclusion criteria for mechanical thrombectomy included stroke onset within six hours, age of 18 years or older, a National Institutes of Health Stroke Scale (NIHSS) score of six or higher, an Alberta Stroke Program Early Computed Tomography (CT) Score (ASPECTS) of six or higher, and no hemorrhage on CT. All patients underwent a dual-energy CT scan immediately after mechanical thrombectomy and a follow-up conventional CT or MRI scan within two days. Further follow-up examinations were conducted based on the clinical status of the patients.
Patients were divided into two groups based on the presence or absence of HT, as determined by dual-energy CT and follow-up examinations. Exclusion criteria included a previous history of stroke, failure of mechanical thrombectomy, failure of image collection, and other situations that precluded inclusion, such as the presence of posterior circulation hematoma immediately after mechanical thrombectomy.
Clinical data collected included age, gender, baseline NIHSS score, hypertension, diabetes mellitus, atrial fibrillation, cigarette smoking, time from stroke onset to recanalization, and time from puncture to recanalization. The modified Rankin Scale (mRS) at 90 days was evaluated, with a good clinical outcome defined as an mRS score of two or lower.
Dual-energy brain scans were performed using a third-generation dual-source CT scanner. The dual-energy CT protocol included a caudocranial scan direction, tube voltages of 80/150 kV, a mAs reference of 310/207, 64 x 0.6 mm collimation on both detectors, a rotation time of 1.0 second, a pitch of 0.7, and iterative reconstructions with a strength level of three. The raw spiral data were reconstructed into three different series: virtual non-contrast images (VNC), iodine overlay map (IOM) images, and a reconstructed 120 kV conventional CT.
The VNC and IOM images were first used to identify the presence of iodine extravasation and/or hemorrhage. For patients with iodine extravasation, two semiquantitative methods were employed to evaluate the extent of extravasation. The first method involved the use of the modified Alberta Stroke Program Early CT Score (mASPECTS) on the reconstructed 120 kV conventional CT. Instead of evaluating hypoattenuation areas, the mASPECTS scored hyperattenuation areas using the ASPECTS method. A score of 10 indicated no hyperattenuation areas, while a score of 0 indicated diffuse high density throughout the ASPECTS areas. The second method assessed the degree of iodine extravasation by rating it as 0 or 1, with 0 indicating no dense iodine extravasation and 1 indicating dense iodine extravasation. Dense iodine extravasation was defined as involving at least two ASPECTS areas with no normal brain tissues observed in the regions with extravasated iodine contrast.
Interobserver agreement was calculated using weighted Kappa statistics, with excellent agreement defined as k = 0.81-1.00, good agreement as k = 0.61-0.80, and less good or poor agreement as k < 0.60. The mean mASPECTS values were obtained by two readers, and disagreements in the assessment of dense iodine extravasation were resolved by a senior radiologist.
The European Cooperative Acute Stroke Study (ECASS) classification was used to evaluate the type of hemorrhage. Two readers reviewed the follow-up examination results, with disagreements resolved by a senior radiologist. Brain hemorrhage was defined by the presence of persisting and/or increasing high-density parenchyma on follow-up CTs or low-intensity parenchyma in the T2-weighted image upon follow-up MRI.
Statistical analysis was performed using IBM SPSS Statistics version 22 and MedCalc Statistical Software version 19.0.7. Student’s t-test, Mann-Whitney U test, or chi-square/Fisher exact test was used to detect differences between the HT and non-HT groups. Univariate logistic regression analysis was used to evaluate associations between clinical variables, imaging parameters, and hemorrhage transformation. Variables with P < 0.1 in the univariate analysis were entered into the multivariate model. Receiver operating characteristics (ROC) curve analysis was used to evaluate the predictive value of mASPECTS, dense iodine extravasation, and their combination for predicting hemorrhage transformation.
A total of 72 patients were included in the final analysis, with 29 patients in the HT group and 43 patients in the non-HT group. Significant differences were observed between the two groups in mRS scores, mASPECTS, and the sign of dense iodine extravasation. The Kappa analysis demonstrated good interobserver agreement for both mASPECTS and dense iodine extravasation.
Univariate logistic regression analysis showed that mASPECTS and dense iodine extravasation were significantly associated with hemorrhage transformation. Multivariate logistic regression analysis confirmed that dense iodine extravasation was significantly associated with hemorrhage transformation. ROC curve analysis revealed that the areas under the curves (AUCs) for mASPECTS, dense iodine extravasation, and their combination were 0.833, 0.839, and 0.850, respectively. The cut-off for mASPECTS was six, with a sensitivity of 75.86% and a specificity of 81.40%. The sign of dense iodine extravasation showed a sensitivity of 72.41% and a specificity of 88.37%. The combination of the two CT indicators showed a sensitivity of 79.31% and a specificity of 83.72%.
In conclusion, the study demonstrated that semiquantitative methods for assessing iodine extravasation in acute ischemic stroke after successful mechanical thrombectomy are practical and effective. Patients with an mASPECTS of six or lower and/or a positive sign of dense iodine extravasation are more likely to develop hemorrhage, which is associated with poor outcomes. These methods, based on evaluating simulated conventional CT reconstructed by dual-energy CT, may also be applicable to conventional single-energy CT.
doi.org/10.1097/CM9.0000000000001236
Was this helpful?
0 / 0