Giorgio Busto, Umberto Pensato, Johanna Ospel, Bijoy K Menon, Andrew Demchuk, Ilaria Casetta, Francesco Loverre, Giorgio Umberto Marcello, Francesco Arba, Andrea Ginestroni, Vittorio Miele, Alessandro Padovani, Andrea Morotti, Enrico Fainardi
The gHIR showed a strong association with outcomes and superior prognostic performance compared with other standard CT-based methods for assessing IGR.
BACKGROUND: We introduce a novel perfusion-imaging metric, the granular hypoperfusion intensity ratio (gHIR), to assess the infarct growth rate (IGR).
METHODS: The gHIR was calculated using an equation that weighted the proportions of ischemic volumes across multiple Tmax thresholds (2s to 18s) to yield a value from 0 to 1, with higher values indicating a faster IGR. gHIR was compared with four established IGR metrics: ASPECTS decay, multiphase CT-angiography collaterals, HIR, and rCBF-IGR. The primary outcome was a good functional outcome (90-day modified Rankin Scale ≤2), assessed with multivariable logistic regression. We evaluated prognostic performance with the c-statistic and determined the optimal gHIR cutoff using the Youden Index.
RESULTS: Of 672, 570 (85%) were included: the median age was 77 years [IQR=67-83]; 290 (50.8%) were female; and 272 (47.7%) had a good functional outcome. A good functional outcome was associated with gHIR (adjusted odds ratio [aOR] 0.42 [95%CI=0.34-0.52] per 0.1-unit increase). The AUC of the gHIR for predicting a good functional outcome was 0.75 (95%CI=0.71-0.79). The optimal gHIR cutoff for a good functional outcome was ≤0.30.
CONCLUSION: The gHIR showed a strong association with outcomes and superior prognostic performance compared with other standard CT-based methods for assessing IGR.