Reliability of Persian Early Computed Tomography Score in Patients with Brain Infarction

Background: the one-third midd ie cerebral artery (1/3 mca)method and alberta stroke program early computed tomographyscore (aspects) were used to detect s ignificant earlyischemic changes on brain computed tomography (ct) of patientswith acute stroke. we designed the persian early ctscore (pects) and compared it with the above systems .methods: the tomograms were chosen from the stroke databank of ghaem hospital, mashhad, in 2008. the inclusioncriteria were the presence of mca territory infarction and performanceof ct within 6 hours after stroke onset. axial ctswere performed on a third-generation ct scanner (siemens,artx, germany). section thickness above posterior fossawas 10 mm (130 ky, 150 mas). films were made at windowlevel 35 hu. the brain cts were scored by four independentradiologists based on the aspects, 1/3 mca method, andpects. the readers were blinded to the clinical informationexcept the symptom side. cochrane q and kappa tests wereused for statis tica l analysis.results: twenty four ct scans with sufficient quality wereavailable. the difference in distribution of dichotomized < = 7and >7 aspect scores between the four raters was significant;q=13.071 , df=3 , p=0.04. the difference in distributionof dichotomized > 113 and < = 1/3 mca territory involvementbetween 4 raters was also significant; q=13.5, df=3, p=0.004 .distribution of dichotomized <6 and 2.6 scores based onpects system between the four raters was not different;q=6.349, df=3 , p=0.096.conclusion: pects method was more reliable than aspectsand 1/3 mca methods in detecting major earlyischemic changes in stroke patients who were candidated forthrombolysis therapy .