Is hybrid SPECT/CT necessary for pre-interventional 3D quantification of relative lobar lung function?

Background In pulmonary malignancies pre-interventional 3D estimation of relative lobar perfusion is established to predict post-interventional functional outcome particularly in patients with borderline lung function. Aim was to test whether quantification from SPECT-scanners (non-hybrid) is as acc...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:European Journal of Hybrid Imaging 2018-09, Vol.2 (1), p.1-9, Article 18
Hauptverfasser: Knollmann, Daniela, Avondo, Jerome, Schaefer, Wolfgang M.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Background In pulmonary malignancies pre-interventional 3D estimation of relative lobar perfusion is established to predict post-interventional functional outcome particularly in patients with borderline lung function. Aim was to test whether quantification from SPECT-scanners (non-hybrid) is as accurate as from SPECT/CT-scanners (hybrid) when using dedicated software. Methods Sixty-one patients suffering from pulmonary tumours underwent lung SPECT/CT using Tc-99m MAA to predict postoperative residual lung function prior to surgical treatment. Quantification was done using “HERMES Hybrid 3D–Lung Lobe Quantification”. In the hybrid approach SPECT and combined lowdoseCT/diagnosticCT were used. In the non-hybrid approach SPECT and diagnosticCTs were used, lowdoseCTs were omitted. Bland Altman analysis was done to test for agreement. Results Three hundred five lobes were quantified. Evaluation time was 6:37 ± 0.55 min (hybrid) versus 6:34 ± 0.51 min (non-hybrid). Mean lobar value was 20.0 ± 10.5% (range from 0 to 55%) for hybrid and 20.0 ± 10.6% (range from 0 to 58%) for the non-hybrid approach, mean absolute difference was 1.31%, no significant differences were found when analysing all values ( p  > 0.9). Correlation was excellent ( R  = 0.984, slope of the regression line 1.001 ( p  
ISSN:2510-3636
2510-3636
DOI:10.1186/s41824-018-0036-0