| Objective: 18F-FDG PET/CT is superior to the traditional imageology in differentiating malignant from benign pulmonary nodules, particularly for early diagnosis. Now many hospitals and institutes have been able to diagnosis pulmonary tumors using PET/CT. 18F-FDG is not specialized radiotracer for tomor, which induce the decrease of accuracy of diagnosis lung cancer. In this study, all diagnostic reference of 18F-FDG PET/CT in differentialing diagnosis pulmonary nodules, were been statistically analyzed. The study also proved good clinical value of differential diagnosis pulmonary nodules using 18F-FDG PET/CT by compaired with simple PET or CT in the diagnostic of pulmonary tumors. Finally, it can provids scientific and standard diagnostic criterion for diagnosis lung cancer.Methods:1 All cases came from our hospital. 65 patients were studied, who seek help for diagnosing pulmonary diseases. Their average age was 59.78±13.28, which ranged from 30 to 87. The count of total nodules from 65 patients was 80. The nodules were defined by pathological or clinical follow-up results.2 Imaging instrument was Gemini GXL16 PET/CT produced by the Holland company Phillips. CT acquisition: tube voltage: 120KV, current: 250mA, pitch: 0.813, matrix: 512×512, rotate speed of tube: 0.5 second/rev, thickness: 5mm. PET acquisition and reconstruction: 3D acquisition, Duration: 2.5min/bed, scanning ranged from floor of skull to superior extremity of femur, 8～10 bed in all. 3D reconstruction method: line of response. And used X-ray for attenuation. CT used standard reconstruction, reconstruction thickness: 5mm. PET reconstruction thickness: 5mm, interval: 5mm. Fused the CT and attenuated PET imaging, we obtained CT, PET and PET/CT imaging of transverse plane, sagittal plane and coronal plane, respectively. The delay phase scans were performed 2h～3h after the injection of the 18F-FDG dose. Thoracic part was only scanned, and other acquisition condition same to early time scans. PET, CT and PET/CT imaging were evaluated by at least two experienced PET/CT doctor, respectively.3 The early time maximum standardized uptake value (early time SUVmax), and the change in SUVmax of early and delay phase (dual-time pointΔSUVmax), were the two mainly diagnostic references of 18F-FDG PET, which were analyzed in this study. The statistical distribution of the early time SUVmax and dual-time pointΔSUVmax in 80 pulmonary nodules, the statistical difference of the early time SUVmax and dual-time pointΔSUVmax between the malignant and benign nodule groups, and the best cutoff value of the early time SUVmax and dual-time pointΔSUVmax in diagnosing pulmonary nodules were researched. Finally the diagnostic value of the two references of PET had been evaluated.4 In diagnosis of lung cancer, it was principally refered to the early time SUVmax and dual-time pointΔSUVmax of PET, and morphological characteristic of CT. The final diagnoses were proved by histopathology or by clinical follow-up. Comparison of the diagnostic value between PET/CT and PET or CT, including the values of sensitivity, specificity, accuracy, positive predictive value and negative predictive value in detection of lung cancer, and in order to evaluate the clinical value of 18F-FDG PET/CT in differentiating malignant from benign pulmonary nodules.5 Statistical analysis: The SPSS11.5 statistical soft was used. The early time SUVmax, dual time pointΔSUVmax of benign and malignant were expressed as mean±SD, and the t-test was used when comparing the benign and malignant tomour groups. The receiver Operating characteristic (ROC) curve was used to evaluate the diagnostic value of SUVmax andΔSUVmax for pulmonary nodules. The sensitivity, specificity, accuracy, positive predictive value and negative predictive value of PET/CT, PET and CT were compared by x2 test, respectively. For all analyses, p-values of less than 0.05 were considered statistically significant.Results:1 80 pulmonary nodules included 38 benign nodules and 42 malignant nodules.2 The mean of early time SUVmax in the benign and malignant nodule groups was 2.49±1.64 and 5.44±3.02, respectively. And we found significant differences in the mean of two groups(t=-5.364, p=0.0001)。3 The mean of dual time pointΔSUVmax in the benign and malignant nodule groups was 16.41%±30.06% and 31.76%±25.83%, respectively. And we found significant differences in the mean of two groups(t=-2.456, p=0.016)。4 By analysis of ROC, the best cutoff value of early time SUVmax and dual time pointΔSUVmax in diagnosing pulmonary nodules was >2.45 and >24.66%, respectively. When early time SUVmax>2.45 was taken as the cutoff point, the sensitivity, specificity, accuracy, positive predictive value and negative predictive value in detection of lung cancer were 88.10%(37/42), 63.20%(24/38), 76.25%(61/80), 72.55% (37/51)and 82.76%(24/29), respectively. When dual time pointΔSUVmax>24.66% was taken as the cutoff point, the sensitivity, specificity, accuracy, positive predictive value and negative predictive value in detection of lung cancer were 64.30%(27/42), 71.10%(27/38), 67.50%(54/80), 71.05%(27/38) and 64.29%(27/42), respectively.5 There are 56 patients (70 lesions) diagnosised correctly in 65 patients (80 lesions), and the false positive rate and false negative rate of PET/CT were 18.42%(7/38) and 7.14%(3/42) in differentiating malignant from benign pulmonary nodules, respectively. The sensitivity of PET/CT, PET and CT were 92.86%(39/42), 85.71% (36/42) and 76.19%(32/42); the specificity were 81.58%(31/38), 63.16%(24/38) and 76.32%(29/38); the accuracy were 87.50%(70/80), 75.00%(60/80) and 76.25%(61/80); the positive predictive value were 84.78%(39/46), 72.00%(36/50) and 78.05%(32/41); the negative predictive value were 91.18%(31/34), 80.00%(24/30) and 74.36%(29/39), respectively. In diagnosing lung cancer, PET/CT has higher accuracy than PET and CT (x2=4.103, p=0.043; x2=4.103, p=0.043), and has higher sensitivity than CT (x2=4.459, p=0.035). There was no significant difference in sensitivity between PET/CT and PET(x2=0.498, p=0.48). In specificity, positive predictive value and negative predictive value, there were also no significant difference (x2=3.224, 2.291, 0.852; 0.389, 0.655, 3.510. p=0.073, 0.13, 0.356; 0.533, 0.418, 0.061).Conclusions:1 Using 18F-FDG PET/CT, the best cutoff value of early time SUVmax and dual time pointΔSUVmax in diagnosing pulmonary nodules was >2.45 and >24.66%, respectively, which can be taken as reference of 18F-FDG PET/CT in detection of lung cancer.2 In diagnosing pulmonary nodules, PET/CT had higher accuracy than PET and CT, and had higher sensitivity than CT.3 Integrating of data from PET and CT, the accuracy of 18F-FDG PET/CT diagnosis pulmonary nodules was highest. When 18F-FDG PET/CT in differentiating malignant from benign pulmonary nodules on the basis of grasps clinical information thoroughly, it is necessary to in accordance with refer the early time SUVmax or dual time pointΔSUVmax of 18F-FDG PET metabolism imaging and the morphology characters of CT. |
