Department of Nuclear Medicine, Saarland University Medical Center, Gebäude 50, 66421, Homburg/Saar, Germany.
Eur J Nucl Med Mol Imaging. 2010 Jun;37(6):1087-94. doi: 10.1007/s00259-010-1387-3. Epub 2010 Feb 6.
(18)F-fluorodeoxyglucose (FDG) PET is the most accurate imaging modality in characterizing a solitary pulmonary nodule (SPN). Besides visual image interpretation, semiquantitative analysis using standardized uptake values (SUV) is performed to improve diagnostic accuracy. Mostly, an SUV threshold of 2.5 is applied to differentiate between benign and malignant lesions. In this study we analysed the use different SUV thresholds to predict the post-test probability of malignancy for the individual patient considering his pre-test probability. Furthermore, we investigated the prognostic value of SUV in SPN for survival.
This retrospective study included 140 consecutive patients who underwent FDG PET for evaluation of SPN. Visual interpretation was performed by two readers. For semiquantitative analysis, maximum SUV (SUV(max)) was measured in all SPN. A final diagnosis was obtained by pathological examination or follow-up of more than 2 years. In a nomogram, positive and negative predictive values (PPV and NPV) were plotted against the hypothetical SUV threshold to determine the optimum SUV threshold. Survival was analysed using the Kaplan-Meier method and log-rank test.
The prevalence of malignancy was 57%. The FDG uptake in malignant SPNs was higher than in benign SPNs (SUV 9.7 +/- 5.5 vs 2.6 +/- 2.5, p < 0.01). More than 90% of SPNs with an SUV below 2.0 were benign (sensitivity, specificity, NPV of 96, 55 and 92%). The highest diagnostic accuracy was achieved with an SUV of 4.0 (sensitivity, specificity and accuracy of 85%). Visual interpretation achieved corresponding values of 94, 70 and 84%, respectively. In lung cancer higher FDG uptake (SUV(max) >or= 9.5) was associated with shorter survival (median survival 20 months) and low FDG uptake with longer survival (>75 months).
FDG PET allows assessment of the individual risk for malignancy in SPNs by considering tumoural SUV and pre-test probability. Higher FDG uptake in lung cancer as measured by SUV analysis is a prognostic factor. In patients with low FDG uptake in an SPN and increased risk during surgery omission of diagnostic thoracotomy may be warranted.
（18）F-氟代脱氧葡萄糖（FDG）PET 是目前用于特征化孤立性肺结节（SPN）的最准确的成像方式。除了进行视觉图像解释外，还使用标准化摄取值（SUV）进行半定量分析以提高诊断准确性。大多数情况下，应用 SUV 阈值 2.5 来区分良性和恶性病变。在这项研究中，我们分析了考虑个体患者的术前概率，使用不同的 SUV 阈值来预测患者的恶性肿瘤术后概率。此外，我们还研究了 SPN 中 SUV 对生存的预后价值。
本回顾性研究纳入了 140 例连续接受 FDG PET 评估 SPN 的患者。由两位读者进行视觉解释。进行半定量分析时，在所有 SPN 中测量最大 SUV（SUV(max)）。通过病理检查或超过 2 年的随访获得最终诊断。在列线图中，针对假设的 SUV 阈值绘制阳性和阴性预测值（PPV 和 NPV），以确定最佳 SUV 阈值。使用 Kaplan-Meier 方法和对数秩检验分析生存情况。
恶性肿瘤的患病率为 57%。恶性 SPN 的 FDG 摄取量高于良性 SPN（SUV 9.7 +/- 5.5 与 2.6 +/- 2.5，p < 0.01）。SUV 低于 2.0 的 SPN 中 90%以上为良性（敏感性、特异性、NPV 分别为 96%、55%和 92%）。SUV 为 4.0 时诊断准确性最高（敏感性、特异性和准确性分别为 85%）。视觉解释分别达到了 94%、70%和 84%。在肺癌中，FDG 摄取量较高（SUV(max) >or= 9.5）与较短的生存时间（中位生存时间 20 个月）相关，而 FDG 摄取量较低与较长的生存时间（>75 个月）相关。
FDG PET 通过考虑肿瘤 SUV 和术前概率，可评估 SPN 中恶性肿瘤的个体风险。SUV 分析中肺癌的 FDG 摄取量较高是预后因素。在 SPN 中 FDG 摄取量较低且手术期间风险增加的患者中，可能无需进行诊断性剖胸术。
