Test-retest variability and reliability of 123I-IBZM SPECT measurement of striatal dopamine D2 receptor availability in healthy volunteers and influence of iterative reconstruction algorithms.

OBJECTIVE

123I-IBZM single photon emission computed tomography (SPECT) is a widely used method to measure D(2) receptor availability. However, test-retest variability and reliability have not been reported yet. This study aimed to further characterize 123I-IBZM SPECT in healthy volunteers (HVs), by assessing (1) pseudoequilibrium interval after bolus injection; (2) normal specific uptake ratio (SUR) values using filtered-backprojection (FBP); and the iterative reconstruction algorithm ordered-subsets expectation maximization (OSEM); (3) test-retest variability and reliability (intraclass correlation coefficient); and (4) influence of OSEM on test-retest variability and reliability.

METHODS

Ten HVs (Group A) were scanned twice 48 h apart for test-retest variability and reliability measurements, and n = 4 of them were sequentially scanned over time. Eighteen HVs (Group B) were scanned once at pseudoequilibrium. For reconstruction FBP was used. Test-retest scans were reconstructed in addition using OSEM. SPECT-MRI coregistration was used for region of interest drawing.

RESULTS

Pseudoequilibrium was achieved at 90 min postinjection (p.i.) and maintained until the end of the SPECT session (n = 4), and mean SUR at this time point was 0.96 +/- 0.14 (Groups A + B, n = 28). Mean SUR at test was 0.96 +/- 0.19 and at retest 0.94 +/- 0.19 (Group A, n = 10). Using FBP, test-retest variability was (12.7 +/- 9.6)% and reliability was 0.74. Using OSEM with 18 equivalent iterations, test-retest variability and reliability were improved to (6.5 +/- 5.2)% and 0.84, respectively.

CONCLUSIONS

123I-IBZM SPECT imaging using the bolus injection and a single scan at 90 min p.i. is a reproducible method showing acceptable test-retest variability and reliability. Test-retest variability and reliability can be substantially improved using OSEM with 12-36 equivalent iterations.