Interpretive intra- and interobserver reproducibility of rest/stress Tc-99(m)-sestamibi myocardial perfusion SPECT in a consecutive group of male patients with stable angina pectoris before and after percutaneous transluminal angioplasty
A. Johansen et al., Interpretive intra- and interobserver reproducibility of rest/stress Tc-99(m)-sestamibi myocardial perfusion SPECT in a consecutive group of male patients with stable angina pectoris before and after percutaneous transluminal angioplasty, NUCL MED C, 22(5), 2001, pp. 531-537
Background Observer variability of Tc-99(m)-sestamibi myocardial perfusion
imaging (MPI) has rarely been investigated. The aim of our study was to eva
luate the interpretive reproducibility with this technique.
Patients We report on 108 consecutive male patients with stable angina pect
oris, investigated before and after percutaneous transluminal angioplasty (
PTCA).
Methods A 2-day rest/stress Tc-99(m)-sestamibi gated single photon emission
computed tomography (SPECT) protocol was used. MPI was interpreted by two
independent observers without knowledge of clinical data, using a 20-segmen
t scoring model.
Results Intra- and interobserver agreement was found to be good to excellen
t (kappa = 0.71-0.85) with regard to the overall diagnosis as well as the i
ndividual vessel diagnosis (kappa =0.60-0.87). However, agreement was highe
r for left anterior descending coronary artery (LAD) and left circumflex co
ronary artery (LCX) vascular territories than for the right coronary artery
(RCA) territory. Moderate to good intraobserver agreement (kappa =00.54-0.
68) and slightly lower interobserver agreement (kappa =0.52-0.56) was found
for segmental score interpretation. When comparing the interpretive reprod
ucibility before and after PTCA intra- and interobserver agreement was bett
er after PTCA, probably reflecting the increase in normal scans after revas
cularization.
Conclusions In a group of consecutive male patients with stable angina pect
oris interpretive reproducibility (overall and individual vessel diagnosis)
was good to excellent. However, segmental scoring reproducibility was mode
rate to good. ((C) 2001 Lippincott Williams & Wilkins).