THE EFFECT OF TRAINING ON THE INTERPRETATION OF TC-99(M)-SESTAMIBI MYOCARDIAL PERFUSION SPET IN PATIENTS WITH SUSPECTED CORONARY-ARTERY DISEASE

The aim of this study was to examine the effect of a period of concent rated training in nuclear cardiology on the accuracy of reporting Tc-9 9(m)-sestamibi (Tc-99(m)-MIBI) single photon emission tomographic (SPE T) images. Two visiting cardiologists, with no previous experience in nuclear cardiology, were asked to report blindly 60 Tc-99(m)-MIBI SPET scans after 2 weeks of training in nuclear cardiology. One (observer 2) reported the same scans blindly after 2 months of further training. The results were compared with the assessment made by two experienced nuclear cardiologists and by using kappa statistics. Kappa values for the overall interpretation of the scan (normal or abnormal), segmenta l analysis (normal, ischaemic, fixed or mixed) and the three arterial territories were 0.7, 0.58 and 0.67 respectively. Following 2 months o f further intensive training of observer 2, the kappa values were 0.85 7, 0.78 and 0.91 respectively. The difference between the two readings of observer 2 was significantly different for the segmental analysis (P < 0.001) and arterial territories (P = 0.006) but it did not reach statistical significance for the overall interpretation (P = 0.7). Thu s, cardiologists without previous interpretation skills in nuclear car diology required about 2 months of intensive training to achieve good accuracy in the interpretation of Tc-99(m)-MIBI SPET images. According ly, these techniques can be established in centres other than tertiary sites.