Assessment of ventricular wall thickness in vivo by computed transmission tomography.

The present study was aimed at investigating the potential of computed transmission tomography (CT) scanning for quantitative assessment of interventricular septal wall thickness in living dogs. Seven normal beagles and seven litter mates with left ventricular hypertrophy were scanned in order to determine the sensitivity of the method. Left ventricular hypertrophy was created by banding the ascending aorta when the dogs were 6-8 weeks old. The operated dogs were followed, along with their normal litter mates, to the age of 7-9 months, when CT scanning was performed. The interventricular septum was well visualized using i.v. bolus injections (0.5 ml/kg) of meglumine/sodium diatrizoate. Septal wall thickness was measured using a computer region of interest program that allowed myocardial edge detection. After CT scanning, the dogs were sacrificed. The hearts were excised and sectioned at the same levels as the CT scans were obtained. A linear regression for septal wall thickness as measured by CT and by autopsy was calculated CTmm = 1.15 (autopsymm) - 3.64 (r = 0.92 for n = 14). The correlation between CT and autopsy values was excellent, alth-ough CT consistently underestimated the interventricular septal thickness by 10-20%. The explanation for this underestimation is not clear. Nongated CT scans appear to represent a composite image of the heart obtained during several heart beats. Systolic and diastolic variation in wall thickness as well as the method used for boundary detection are potential sources of error in this CT study. However, CT appears to offer a useful noninvasive method for estimating septal wall thickness and has the potential for providing measurements of total myocardial mass. CT scanning should be readily applicable in the clinical situation.