Background and Aims: The role of CTP in thrombolysis treatment decisions is uncertain. We used an online platform to assess observer agreement on thrombolysis decisions in patients with different perfusion deficits. Methods: We selected 24 cases from the IST-3, ATTEST and POSH studies to illustrate various clinical and imaging scenarios. Observers were presented with short clinical vignettes with non-contrast CT, maps of cerebral blood volume and flow, mean transit time, delay time and thresholded maps dichotomised into penumbra and core . A structured questionnaire asked observers to categorise perfusion deficits, and on whether to give thrombolytic therapy. We assessed observer agreement with Krippendorff's-alpha. Univariate predictors of thrombolysis were entered into a multivariate regression analysis. Results: 29 observers contributed 318 treatment decisions. Thrombolysis was recommended in 204 (65%). The most common reason to withhold treatment was perfusion deficit being too extensive (29%). Overall agreement on treatment decisions was low (k-alpha=0.10, 95% CI-0.01-0.20). Univariate factors significantly associated with a decision to give IV rtPA included NIHSS between 5 and 15, time to scan <3 hours, more experience with perfusion imaging and perfusion imaging showing predominantly penumbra[KM2] . In a multivariate model, these factors accounted for 30% of variation in treatment decision (Hosmer and Lemeshow Test=0.35,Nagelkerke R2=0.29) with perfusion deficit being mainly penumbra and more experience with perfusion imaging remaining significant. Conclusions: Agreement on treatment decisions in clinical scenarios with accompanying CTP was low, and while the proportions of core and penumbra were significantly associated with treatment decisions, these factors explained only 30% of variation in decisions
No datasets are available for this submission.
No license information is available for this submission.