The influence of through-plane motion on left ventricular volumes measuredby magnetic resonance imaging: Implications for image acquisition and analysis

In rite evaluation of the left vertricular (LV) function using magnetic res onance imaging (MRI), a stack of parallel short-axis (SA) cine images is ac quired that covers the whole LV. The aim of this study is to quantify the c ontribution to the LV volume parameters, provided by the most basal image p lane that shows the LV wall only in end diastole (ED) but not in end systol e (ES). In 57 healthy volunteers (31 men, mean body surface area 1.87 m(2)) , a complete set of parallel SA images was acquired (10-mm slice distance) by breathhold segmented k-space cine MRI (7 k(y) lines per beat). The LV en d-diastolic volume (EDV), stroke volume (SV), ejection fraction (EF), and c ardiac output (CO) were determined by slice summation. Calculations were pe rformed both with and without inclusion of the most basal slice. With inclu sion of the most basal slice, all parameters were significantly (p < 0.001) larger compared with the values obtained by excluding this slice. EDV was 134 +/- 29 ml versus 113 +/- 26 ml: SV was 93 +/- 18 ml versus 72 +/- 16 ml , EF was 70 +/- 4% versus 64 +/- 4%; and CO was 5.3 +/- 1.4 l/min versus 4. 1 +/- 1.1 l/min. The inclusion of the most basal slice leads to significant ly larger values of LV volume parameters. Thus, this most basal SA image sl ice should be included in calculating the EDV. Whether or not this basal SA slice also contributes to the ES volume should be decided by using anatomi cal criteria on the ES image. The projection line onto the ES image of a lo ng-axis view provides an additional criterion.