SUV小于2．5的孤立性肺结节18F—FDGPET／CT诊断价值初探

目的正电子断层显像（PET）通常以标准摄取值（SUV）≥2．5为肺癌诊断标准，但仍有部分SUV〈2．5的孤立性肺结节（SPN〈2.5）为肺癌。本研究旨在分析SPN〈2.5，氟代脱氧葡萄糖（18F—FDG）PET／CT表现及其诊断价值。方法51例18F—FDGPET／CT早期显像SUV平均值（SUVavg）〈2．5的孤立性肺结节病人纳入研究。对SPN放射性摄取测量采用两种方法：视觉分析及半定量分析。测量SPN的早期及延迟显像的SUVavg、SUVmax及其早期一延迟变化率（△SUVavg、△SUVmax）。37例以病理诊断及14例以2年以上临床随诊结果为最终诊断标准。统计学分析采用SPSS16．0软件。结果51例SPN〈2.5患者中，良性11例，恶性40例。虽然良恶性SPN〈2.5患者各项SUV值间均无统计学差异（P〉0．05），但恶性SPN〈2.5，患者各项SUV值均有高于良性的趋势，特别是ASUVavg、ASUVmax。以ASUVmax〉0为诊断阈值，18F—FDGPET／CT诊断恶性SPN的敏感性68．4％、特异性60．0％、准确性66．7％、阳性预测值86．7％及阴性预测值33．3％。视觉分析：无FDG摄取的SPN〈2.5 9例中，5例为良性，4例为恶性；有FDG摄取的SPN〈2.5 42例中，6例良性，36例恶性，两组之间有显著的统计学差异，P=0．02。良恶性SPN〈2.5的CT密度及大小之间均无统计学差异（P〉0．05）。综合PET和CT信息分析SPN〈2.5，18F—FDGPET／CT诊断恶性SPN〈2.5的敏感性97．5％、特异性54．5％、准确性88．2％、阳性预测值88．6％、阴性预测值85．7％。结论单纯以SUV≥2．5为标准鉴别诊断SPN会导致肺癌的漏诊。对于此类结节，ASUV有较高的肺癌阳性预测值；FDG摄取视觉分析能提高结节的良恶性鉴别准确率；综合FDGPET和cT分析的PET／CT能显著提高诊断肺癌的敏感性和准确性，但特异性较低。

18F -FDG PET/CT Characterization of Solitary Pulmonary Nodules with an Initial Average Standard Uptake Value Less Than 2.5

Objective Tradltionly,a FDG standardized uptake value （SUV） of 2.5 is used as a cutoff of pulmonary malignancies.However, the frequency of malignancies with an SUV of 〈 2.5 is significant, and this study aimed to evaluate the value of lS F - FDG PET/ CT for diagnosis of solitary pulmonary nodules （SPNs） with is F - FDG uptake below the initial average standardized uptake value （ SU- Vavg） of 2.5. Methods We retrospectively analyzed 51 patients who had SPNs with an initial SUVavg of 〈 2.5. The uptake of 18F - FDG was graded by semiquantitative methods （SUVavg, SUVmax, A SUVavg, J SUVmax） and a visual method. Besides above, we ana- lyzed the density and size of SPNs. Final classification was based on histopathologic findings or at least 24 months of clinical follow - up. Results We found 40 malignant and 11 benign lesions. Though there were no significant differences between SUV（ including all the SUVs and A SUV） of benign and malignant SPNs,there was a tendency that SUV of malignant SPNs was higher than that of benign ones（ espe- cially a SUV）. When a cutoff of a SUVmax above 0 was used,the sensitivity, specificity, accuracy, positive and negative predictive values of lSF FDG PET/CT in diagnosis SPN were 68.4% ,60.0% , 66.7% ,86.7% ,33.3% ,respectively. There were 5 benign SPNs and 4 ma- lignant SPNs in total 9 absent FDG uptake SPNs. There were 6 benign SPNs and 36 malignant SPNs in total 42 visually evident FDG up- take SPNs （P = 0. 02）. There was no significant difference between benign and malignant SPNs in CT density and diameter （P = 0.71, 0.24）. The sensitivity, specificity, accuracy, the positive and negative predictive values of ~s F - FDG PET/CT in diagnosis SPNs were 97.5% ,54.5% ,88.2% ,88.6% ,85.7% ,rospectively. Conclusion These results suggested that SPNs with low laF -FDG uptake,SU- Vavg and SUVmax did not improve the accuracy of 18F - FDG PET/CT. A SUV might be helpful in differential diagnosis. Visual analysis was more valuable than semiquantitative analysis, 18 F - FDG PET/CT demonstrates an excellent performance in classifying SPNs with an SUVavg of 〈 2.5 as benign or malignant with a highly sensitivity and accuracy,though the specificity was low.