^18F-FDG PET／CT双时相显像对肺部病灶的定性诊断价值

目的探讨^18F-脱氧葡萄糖（FDG）PET／CT双时相显像在肺部病灶良恶性鉴别诊断中的临床应用价值。方法78例临床疑诊肺癌患者，均行早期和延迟^18F—FDG PET／CT显像。计算早期及延迟显像最大标准摄取值（SUVmax），并计算2次显像SUVmax变化率（ASUV）。以SUVmax〉2．5和ASUV〉20％作为诊断肺癌的标准。患者最终诊断均经组织病理学、细菌学或治疗后随访证实。采用SPSS13．0软件，SUV组间比较用t检验，用受试者工作特征（ROC）曲线评价SUVmax、ASUV对肺部病灶的定性诊断价值。结果（1）78例患者中肺癌60例，良性病变18例（增殖性病变16例，占88．89％）。肺癌组（9．92±5．33和7．94±4．17，t=10．19）和良性病变组（8．54±6．61和7．21±5．74，t=8．23，P均〈0．01）延迟SUVmax均明显高于早期SUVmax；肺癌组与良性病变组间早期和延迟显像SUVmax差异均无统计学意义（t=0．60和-0．91，P均〉0．05）；肺癌组与良性病变组ASUV差异也无统计学意义（26．04±14．73）％和（18．09±24．09）％，t=1．67，P〉0．05]。18例良性病变患者中有2例延迟显像SUVmax较早期减低，而肺癌患者延迟显像SUVmax均无减低。（2）以SUVmax〉2．5和ASUV〉20％为诊断肺癌的标准，其灵敏度、特异性、准确性、阳性预测值和阴性预测值分别为93．33％（56／60）和63．33％（38／60）、22．22％（4／18）和50．00％（9／18）、76．92％（60／78）和60．26％（47／78）、80．00％（56／70）和80．85％（38／47）、50．00％（4／8）和70．97％（22／31）；根据SUVmax和ASUV得到的ROC曲线下面积分别为0．61（Z=1．38，P〉0．05）和0．56（Z=0．65，P〉0．05），差异均无统计学意义。结论对肺部病灶临床疑诊为肺癌的患者，如良性病变以增殖性病变为主，则^18F—FDG PET双时相显像良恶性鉴别诊断临床应用价值不大；但延迟SUVmax减低可能?

The value of dual-time point 18F-FDG PET imaging for the differentiation between pulmonary malignant and benign lesions

Objective The aim of this study was to evaluate the value of dual-time point 18F-fluorodeoxyglucose (FDG) PET imaging for the differentiation between pulmonary malignant and benign lesions.Methods Seventy-eight patients with clinically suspected lung cancer underwent dual-time point 18F-FDG PET imaging.The maximum standardized uptake value(SUVmax)was calculated for PET imaging of both time points,and the change in SUVmax(△SUV)was defined as the ratio of the increase in SUVmax between early and delayed scans to the SUVmax in the early scan.The final diagnoses were confirmed by histopathology,bacteriology or clinical follow-up.The t-test was used to compare SUVmax,△SUV of benign and malignant groups.The receiver operating characteristic(ROC)curve was used to evaluate the diagnostic value of SUVmax and △SUV for pulmonary lesions.Results (1)There were 60/78 patients with pulmonary malignant lesions and 18/72 patients with benign lesions including 16 with hyperplasia lesions.There was significant difference in the SUVmax between early scan and delayed scan in both malignant group(7.94±4.17 and 9.92±5.33,respectively,t=10.19,P＜0.01)and benign group(7.21±5.74 and 8.54±6.61,respectively,t=8.23,P＜0.01).There was no significant difference in the SUVmax between malignant group and benign group in both early scan(t=0.60,P＞0.05)and delayed scan(t=0.91,P＞0.05).The △SUV was not significantly difierent between malignant and benigh groups (26.04± 14.73)%and(18.09±24.09)%,respectively,t=1.67,P＞0.05].The SUVmax decreased in only 2 of 18 benign lesions,and no SUVmax decrease in all of the malignant lesions.(2)The sensitivity,specificity,accuracy,positive predictive value and negative predictive value with a threshold of SUVmax ＞2.5 and △SUV＞20%were respectively 93.33%(56/60),22.22%(4/18),76.92%(60/78),80.00%(56/70)and 50.00%(4/8);and 63.33%(38/60),50.00%(9/18),60.26%(47/78),80.85%(38/47)and 70.97%(22/31).The areas under the ROC curves were 0.61(Z=1.38,P＞0.05)and 0.56(Z=0.65,P＞0.05)according to the SUVmax and △SUV respectively,and the difference was not statistically significant.Condusiom Dual-time point 18F-FDG PET imasing is not useful for differentiating pulmonary malignancy from the benign lesions of hyperplasia.However,if SUVmax decreases on delayed scan,the lesion is more likely to be benign.