Population-Based Validation of the iScore for Predicting Mortality and Early Functional Outcome in Ischemic Stroke Patients

Abstract

Background: We aimed to determine the accuracy of the iScore for predicting mortality and early disability following ischemic stroke in a French population-based study. Methods: All patients with acute ischemic stroke were identified among residents of the city of Dijon, France, between 2006 and 2011, using a population-based stroke registry. The 30-day iScore and 1-year iScore were calculated. A logistic regression model was used to assess the performance of the iScore for predicting both 30-day and 1-year mortality, and poor functional outcome at discharge (modified Rankin Scale score of 3-6). The discrimination and calibration of the model were assessed using the c-statistic and the Hosmer-Lemeshow goodness-of-fit test, respectively. Results: Among the 1,199 ischemic stroke patients recorded, 107 (8.9%) were excluded because at least one item of data was missing. For the remaining 1,092 patients, the c-statistic was 0.85 (95% CI: 0.82-0.89) for 30-day and 0.84 (0.81-0.87) for 1-year mortality, and calibration was good (p = 0.82 and p = 0.96, respectively, for the Hosmer-Lemeshow test). Similar results were found for disability (c-statistic 0.81, 95% CI: 0.79-0.84, and p = 0.45 for the Hosmer-Lemeshow test). Conclusion: This is the first population-based study to demonstrate the accuracy of the iScore for predicting mortality and early disability in ischemic stroke patients.

Validation of Stroke Prognostic Scores: What Do Clinicians Need to Know?

“…More recently, studies from Greece, Korea, China and other countries revealed good performance in other ethnic groups [16,17,18,19]. The present study provides some additional evidence of its application to a well-defined European population [11]. Together, the aforementioned points illustrate some of the benefits, caveats and limitations for outcome predictions using risk scores.…”

“…Béjot et al [11] evaluated the performance of the iScore in over 1,000 stroke patients, applying a population-based design in France. They found good performance of the iScore when applied to their patients: the C-statistic was 0.85 (95% CI: 0.82-0.89) for 30-day and 0.84 (0.81-0.87) for 1-year mortality.…”

Predicting Functional Outcome and Symptomatic Intracranial Hemorrhage in Patients With Acute Ischemic Stroke

“…8 The ASTRAL score, the DRAGON score, the iScore, the Six Simple Variables score, and the Get With The Guidelines (GWTG) score have been extensively validated in external data sets and sometimes for the prediction of outcomes other than the ones for which they were originally developed. 69,70,[82][83][84][85][86][87][88][89][90][91][92][93][94][95] These extended validations further increase the scores' reliability. The scores that are free of specialized laboratory tests or imaging techniques and include only simple parameters are more feasible for centers with limited resources.…”

Prognostic indices for early mortality in ischaemic stroke - meta-analysis

“…We included 18 relevant studies from 2374 hits that were retrieved through the electronic database search . The flow chart of study selection is shown in Fig.…”

“…As the majority of studies were retrospective in design or post hoc analyses of prospectively collected clinical data, we were seldom able to judge whether the prognostic variables were collected early in the course of the presentation. Treatment pathways were seldom reported, with only three studies explicitly stating that participants received similar care . We recorded more complete reporting of the modality used in neuroimaging (12 studies), as well as amount of missing data (10 studies) (Table S2).…”