Influence of the arm position in myocardial perfusion imaging acquisition

OBJECTIVE: despite the technologic advances in myocardial perfusion imaging, we keep using an uncomfortable and sometimes impracticable patient position - supine with arms raised above the head (U). The purpose of this study was to investigate whether perfusion and functional cardiac gated SPECT scan results of acquisition U are equivalent to another position modality: supine with arms down at the sides of the trunk (D). METHODS AND RESULTS: we performed U acquisition and in sequence D acquisition in 120 patients (pts) using a one-day MPI (rest-gated/stress), with 99mTc-sestamibi (370 MBq and 1110 MBq). Images were processed by the iterative reconstruction method (OSEM). Rest (R) and stress (S) studies were scored using 17-segments model. Functional parameters (left ventricular ejection fraction, and volumes) were automatically obtained by the quantitative gated SPECT (QGS) program. According to the degree of stress defects observed in U study, the patients were categorized in two subgroups: normal (SSS ≤ 3 or < 5%) and abnormal (SSS>3 or ≥ 5%). Shoulder/back pain occurred in 23.3% of U patients and in 5% of D. No significant differences between U and D were found for SSS (p = 0.82) and SRS (p = 0.74) in normal group. In abnormal group, good correlation was found between U and D modes for SSS (Rho = 0.95, p = 0.0001) and SRS (Rho = 0.96 p = 0.0001), but the mean SSS (12.53 ± 7.54) and SRS (10.60 ± 7.08) values of D were significantly lower (p < 0.05) than SSS (13.43 ± 6.81) and SRS (11.33 ± 6.97) of U mode. Function measurements presented good correlations, except for end-diastolic volume (p = 0.0001). CONCLUSION: although D mode appears to be more comfortable and presented a good correlation with U values of SSS and SRS, in abnormal pts, the extent and severity of defects can be underestimated. Considering clinical implications of an accurate perfusion measurement, the acquisition with the arms down should be avoided.