Quantitative and anatomical assessment of lung perfusion with ^sup 99m^Tc-MAA SPECT/CT in emphysema and lung cancer patients

• Published in: The Journal of nuclear medicine (1978) Vol. 59; p. 516
• Main Authors: Nicolas, Guillaume, Chirindel, Alin, Cachovan, Michal, Vija, Alexander Hans, Tamm, Michael, Lardinois, Didier, Wild, Damian
• Format: Journal Article
• Language: English
• Published: New York Society of Nuclear Medicine 01.05.2018
• Subjects: Computed tomography; Confidence; Correlation coefficient; Correlation coefficients; Emphysema; Feasibility studies; Forecasting; Lobes; Lung cancer; Lungs; Medical imaging; Patient assessment; Patients; Perfusion; Quantitative analysis; Respiratory function; Scintigraphy; Single photon emission computed tomography; Uncertainty
• ISSN: 0161-5505; 1535-5667

Objectives: Prior to lung resection, it is of paramount importance to predict the postoperative lung function. The current clinical practice is to use planar (2D) scintigraphy with its obvious limitations. The aim of this single center study is to prospectively assess the feasibility and potential value of quantitative SPECT/CT for 3D lung perfusion imaging with anatomical lungs /lobes delineation. Methods Planar and SPECT/CT lung perfusion scans were acquired (Intevo, Siemens) for each patient after 99mTc-labelled macroaggregated albumin (99mTc-MAA) administration. The percentage of total lung perfusion was assessed for each lung region using the established 2D-projection method and compared to the 3D-anatomical and quantitative method (CT Pulmo 3D and xSPECT-Quant - Siemens). Whenever necessary, respiratory motion correction was applied based on gated list-mode SPECT data. Matched-pair data were compared using Student’s paired t-test and interobserver variability between 2 readers was assessed by calculating the intraclass correlation coefficient of the 50 first patients. Results We enrolled 105 patients in our single centre, prospective, cohort study from 10/2016 to 10/2017. As expected, the left:right split function was similar between the 2D and the 3D Methods: However, at the lobar level, there was a very significant difference between the two; while 2D systematically underestimated the contribution of the upper lobes, lower lobes were consistently overestimated compared to 3D, mean [±uncertainty at 95% confidence]: -32% ±8 and -34% ±7 for the left and right upper lobes and +25% ±4 and +18% ±4 for the left and right lower lobes, respectively (p<<0.001). The mean absolute difference between reader 1 and 2, in point of % of the total perfusion [±uncertainty at 95% confidence) was: 0.3% ±0.2 for both the left and right lungs and 0.5% ±0.3, 0.4% ±0.3, 0.6% ±0.2, 0.4% ±0.1 and +0.6 ±0.2 for the left upper, left lower, right upper, right middle and right lower lobes, respectively. The intraclass correlation coefficient was 0.854 ±0.091 for both the left and the right lungs and 0.845 ±0.096, 0.998 ±0.002, 0.998 ±0.001, 0.994 ±0.004, and 0.812 ±0.118 for the left upper, left lower, right upper, right middle and right lower lobes, respectively. Conclusion 3D Quantification of lung perfusion is feasible in clinical practice, shows substantial incongruity to the routinely used 2D-method at the lobar level and yields excellent interobserver agreement. Further studies are warranted to assess its clinical value, especially in predicting postoperative outcome.