Abstract
605
Objectives Ventilation (V) and perfusion (P) SPECT is a high sensitivity procedure for the detection of pulmonary embolism. The application of low dose CT from SPECT-CT increases specificity of V/P SPECT (1). Using CT data from SPECT-CT allows reconstructing the SPECT data with or without attenuation correction (AC). In general AC leads to homogenisation of the tracer distribution. Aim was to quantify the influence of AC on the homogeneity of P-SPECT, as a marked homogenisation should ease recognition of perfusion deficits.
Methods 20 patients (routine V/P SPECT-CT to rule out pulmonary embolism) with normal findings were recruited. Technegas V-SPECT, then P-SPECT after i.v. injection of 179±31 MBq Tc-99m-MAA, subsequently low-dose CT (SymbiaT, Siemens). Iterative Flash 3D reconstruction with and without AC. Analysis of all P-SPECT data with PMOD, contouring of both lungs (15% isocontour), basal sections were omitted. Variation coefficients of count densities for AC and non-AC data were calculated lungwise. T-test was used to test that AC leads to homogenisation. Significance level: p<0.05 after Bonferroni-Holm correction.
Results The variation coefficients of the count densities in the AC reconstructed P-SPECT data were slightly, but significantly (p<0.005) lower than those in the non AC reconstructed perfusion data (right: 38.1±3.8% (AC) vs. 40.6±4.0% (nonAC), left: 38.9±3.1% (AC) vs. 43.1±2.3% (nonAC)).
Conclusions In pulmonary perfusion SPECT AC increases homogeneity of the perfusion image significantly but only very slightly. As such, using attenuation correction during reconstruction does not seem to produce a relevant clinical advantage. Bearing in mind the possibility - in isolated cases - of a misalignment between SPECT and CT - which could lead to over- or under-correction - it is our judgement that the combination of non-AC perfusion SPECT data with low dose CT is the most reliable diagnostic option. 1. Gutte H, J Nucl Med 2009, 1987-9