Head-to-Head Prospective Comparison of Quantitative Lung Scintigraphy and Segment Counting in Predicting Pulmonary Function in Lung Cancer Patients Undergoing Video-Assisted Thoracoscopic Lobectomy

Prediction of postoperative pulmonary function in lung cancer patients before tumor resection is essential for patient selection for surgery and is conventionally done with a nonimaging segment counting method (SC) or 2-dimensional planar lung perfusion scintigraphy (PS). The purpose of this study was to compare quantitative analysis of PS to SPECT/CT and to estimate the accuracy of SC, PS, and SPECT/CT in predicting postoperative pulmonary function in patients undergoing lobectomy. Methods: Seventy-five non-small cell lung cancer patients planned for lobectomy were prospectively enrolled (68% male; average age, 68.1 +/- 8 y). All patients completed tests of preoperative forced expiratory volume capacity in 1 s (FEV1) and diffusing capacity of the lungs for carbon monoxide (DLCO), as well as Tc-99m-macroaggregated albumin PS and SPECT/CT quantification. A subgroup of 60 patients underwent video-assisted thoracoscopic lobectomy and measurement of postoperative FEV1 and DLCO. Relative uptake of the lung lobes estimated by PS and SPECT/CT was compared. Predicted postoperative FEV1 and DLCO were derived from SC, PS, and SPECT/CT. Prediction results were compared between the different methods and the true postoperative measurements in patients who underwent lobectomy. Results: Relative uptake measurements differed significantly between PS and SPECT/CT in right lung lobes, with a mean difference of -8.2 +/- 3.8, 18.0 +/- 5.0, and -11.5 +/- 6.1 for right upper, middle, and lower lobes, respectively (P < 0.001). The differences between the methods in the left lung lobes were minor, with a mean difference of -0.4 +/- 4.4 (P > 0.05) and -2.0 +/- 4.0 (P < 0.001) for left upper and lower lobes, respectively. No significant difference and a strong correlation (R = 0.6-0.76, P < 0.001) were found between predicted postoperative lung function values according to SC, PS, SPECT/CT, and the actual postoperative FEV1 and DLCO. Conclusion: Although lobar quantification parameters differed significantly between PS and SPECT/CT, no significant differences were found between the predicted postoperative lung function results derived from these methods and the actual postoperative results. The additional time and effort of SPECT/CT quantification may not have an added value in patient selection for surgery. SPECT/CT may be advantageous in patients planned for right lobectomy, but further research is warranted.