Quantifying variation in normal cardiac wall-motion curves

Objectives SPECT RNA can be used to quantify wall motion, and has been investigated for its potential at predicting cardiac resynchronization therapy response (CRT). However, the quantitative analysis of wall-motion curves, which is currently qualitative in nature, may hold potential for other applications as well. However, before abnormal wall-motion can be assessed, normal wall motion must be quantified. In this work, intra-segment and inter-subject variation in wall-motion curves was assessed in a population with normal cardiac function.

Methods Using an in-house developed SPECT RNA program, 8-gate wall-motion curves were obtained at 568 locations around the myocardium for each of 50 subjects with normal cardiac function (LVEF>55%, QRS<120 ms). The wall-motion curves were normalized to the average activity for each subject. For each subject, the mean and standard deviation (SD) wall-motion curve was calculated using a 5-segment model to produce a single wall-motion curve for each of the 5 segments and for each subject. Intra-segment variation was determined as the mean SD, within a segment, averaged over all patients. The inter-subject variation was determined as the SD of the wall-motion curves averaged over all segments and subjects.

Results Normalized time-activity curves were successfully produced for all 50 patients. Mean intra-segment variation was found to be 34% and the mean inter-subject variation was found to be 11% averaged over all gates. The greatest intra-segment variations occurred in the end-systolic gates.

Conclusions Normal cardiac mean variations and time-activity curves were obtained. From this data, a normal database of wall-motion curves can be developed for use in assessing abnormal regions, which may have applications in the prediction of CRT response