This two-center observational study included anterior circulation LVO-AIS patients who underwent EVT and achieved mTICI 2c-3. Initial and final infarct volumes(FIV) were defined using admission CTP defined as rCBF<30%, and DWI-MRI at 24 hours post-EVT. We categorized IGR patterns as exponential(ExpIGR) and Non-exponential(NonExp) based on their growth curves. We then dichotomized ExpIGR clinical significance based on the association of infarct growth with 90-day mRS as ExpIGR-A(>13 ml) and ExpIGR-B(<13 ml). Intraprocedural BP drops were calculated as the difference between MAP at admission and the lowest intraprocedural MAP reading before recanalization, and the area between admission MAP threshold and all lower measurements of intraprocedural MAP. Multivariable regression was used to investigate associations between variables of interest.

Of 159 TICI 2c-3 patients, 36% demonstrated ExpIGR-A, 31% ExpIGR-B, and 32.7% NonExp patterns. The Exp-A and Exp-B groups differed significantly in NIHSS, ASPECTS, glucose, and FIV. The Exp-A and NonExp groups differed in rCBF<30% volume, and stroke onset-to-admission CTP-time. The Exp-B and NonExp groups in NIHSS, rCBF<30%, Tmax<6s, collateral flow measured by hypoperfusion intensity ratio(HIR), and FIV. Hypotensive MAP area(HMA) was independently associated with an ExpIGR-A pattern. Infarct volume increased by 1ml per 100 units of hypotensive area and 4.2ml per 0.1 units of HIR, with a significant interaction between both variables.

After an LVO-AIS, IGR can be differentiated into exponential and non-exponential patterns. A subgroup of patients with the exponential experienced clinically significant IGR increase between CTP acquisition and reperfusion and seemed to be vulnerable to sustained intraprocedural BP drops during EVT.