【作者】 邵丹；

【导师】 王淑侠；

【作者基本信息】 汕头大学 ， 影像医学与核医学， 2010， 硕士

【摘要】 第一部分18F-FDG PET/CT鉴别诊断心脏良、恶性病变的可行性研究1.研究目的1.1通过分析18F-FDG PET/CT用于心脏良、恶性病变鉴别诊断的价值，来探讨18F-FDGPET/CT鉴别诊断心脏病变良、恶性的可行性。1.2计算18F-FDG PET/CT鉴别心脏病变良、恶性的最佳临界值（SUVmax及SUVmaxLesion/Blood），同时计算单独CT及PET/CT鉴别诊断心脏病变良、恶性的敏感性、特异性、准确性。综合分析，初步制定心脏病变的18F-FDG PET/CT诊断思路。2.材料与方法2.1研究对象29例经病理或临床证实的心脏良、恶性病变进行18F-FDG PET/CT扫描，其中恶性病变17例，良性病变12例。2.2 18F-FDG PET/CT检查方法2.2.1检查前准备29例中24例患者全身显像前禁食12小时以上，5例患者行PET/CT检查前2餐采用低碳水化合物高脂饮食，餐后3～6小时内进行PET/CT扫描。2.2.2检查方法所有患者均行PET/CT扫描及胸部双期增强扫描。2.2.3数据分析测得病灶、心腔（血本底）SUVmax及SUVmax病灶/SUVmax心腔（SUVmaxLesion/Blood），结果用“均数±标准差”表示，并观察病灶的CT特征及与周围组织关系。选取不同SUVmax及SUVmax Lesion/Blood分别计算其作为心脏良、恶性病变的临界值时的诊断价值，以计算出最佳临界SUVmax及SUVmax Lesion/Blood。同时分别计算单独CT及PET/CT诊断心脏良、恶性病变的敏感性、特异性、准确性。2.2.4统计学方法所有计量资料均先行正态性检验，采用两独立样本t检验/校正t检验（正态分布）或两独立样本非参数检验（Mann-Whitney U检验）（非正态分布）的方法分别分析SUVmax、SUVmax Lesion/Blood在良、恶性病变之间有无统计学差异（P<0.05，差别有统计学意义）。采用配对x2检验，比较PET/CT及单独CT在诊断准确性上的差异。检验标准：P<0.05，差别有统计学意义。所有统计分析经SPSS13.0统计软件处理。3.结果良性病变SUVmax为0.5～7.4，平均为2.26±1.81；恶性病变SUVmax为2.6～21.2，平均为8.37±5.20。SUVmax值在良、恶性病变之间比较采用独立样本t检验，显示差别有高度统计学意义（t = 4.473，P = 0.000），恶性病变的SUVmax值大于良性病变，SUVmax在良、恶性病变两组均数的差值为6.11，差值的95%可信区间为（3.27，8.95）。良性病变SUVmax Lesion/Blood为1.15±1.06，恶性病变SUVmax Lesion/Blood为4.24±2.02。SUVmax Lesion/Blood值在良、恶性病变之间比较采用独立样本t检验，显示差别有高度统计学意义（t = 4.842，P = 0.000），恶性病变的SUVmax Lesion/Blood值大于良性病变，SUVmax Lesion/Blood在良、恶性病变两组均数的差值为3.09，差值的95%可信区间为（1.78，4.40）。SUVmax以3.5或SUVmax Lesion/Blood以1.8～2.2为阈值时，诊断价值最高，为PET鉴别心脏病变良、恶性的最佳临界值。29例心脏病变中，CT平扫及增强单独诊断正确20例，根据最佳临界SUVmax及SUVmax Lesion/Blood同时结合CT平扫及增强图像，即PET/CT诊断正确27例。单独CT及PET/CT鉴别诊断心脏病变良、恶性的敏感性、特异性、准确性分别为64.7%（11/17）、75.0%（9/12）、69.0%（20/29）；100.0%（17/17）、83.3%（10/12）、93.1%（27/29）。CT与PET/CT比较，诊断准确性有统计学差异（P = 0.016）（McNemar检验）。4.结论18F-FDG PET/CT可以较为准确地提示病变的位置、形态、大小、范围、附着点、以及肺部、纵膈（大血管、冠状动脉等）、心包、胸腔的受累情况、同时提供病变的代谢活性，从而能够较为敏感、特异、准确地鉴别心脏、心包病变的良、恶性。1.研究目的测量禁食组及低碳水化合物高脂组两种不同显像前准备方法的患者心肌摄取18F-FDG程度（SUVmax），比较两种方法的患者心肌SUVmax，制定适合心脏病变的18F-FDG PET/CT显像前准备方法的设定。2.材料与方法：2. 1研究对象30例临床怀疑心脏病变患者显像前通过不同的显像前准备，接着进行18F-FDG PET/CT扫描。2.2 18F-FDG PET/CT检查方法2.2.1检查前准备禁食组：共17例患者，均为PET/CT检查前禁食12小时以上，16小时以内。低碳水化合物高脂组：共13例患者，PET/CT检查前2餐采用低碳水化合物高脂饮食，餐后3～6小时内进行PET/CT扫描。2.2.2统计学方法分别测量禁食组与低碳水化合物高脂组两组患者心肌SUVmax及检查前血糖水平。所有数据先行正态分布检验，采用两独立样本t检验/ t’检验（正态分布）或两独立样本非参数检验（Mann-Whitney U检验）（非正态分布）分析SUVmax及血糖值在良性与恶性病变之间有无统计学差异（P<0.05，差别有统计学意义）。3.结果禁食组与低碳水化合物高脂组患者的心肌SUVmax分别为2.2～19.3（中位数为3.1）、0.7～3.4（中位数为2.1）。比较两组SUVmax采用Mann-Whitney U检验，显示禁食组SUVmax的平均秩次为20.18，低碳水化合物高脂组SUVmax的平均秩次为9.38，Z = -3.333,P = 0.001，两者之间差别有统计学意义，低碳水化合物高脂组患者心肌SUVmax低于禁食组患者心肌SUVmax。禁食组血糖水平为3.7～9.0，平均为6.73±1.77；低碳水化合物高脂组的血糖水平为4.2～8.6，平均为5.82±1.05。血糖水平在两组之间比较采用独立样本t检验，显示差别没有统计学意义（t = 1.746，P = 0.092），血糖水平的两组均数的差值为0.52，差值的95%可信区间为（-0.16，1.97）。4.结论患者行PET/CT检查前2餐采用低碳水化合物高脂饮食，餐后3～6小时内进行PET/CT扫描能够较为有效的抑制心肌摄取18F-FDG，使心肌的显影程度不影响诊断，为一种较为合理的适合心脏、心包病变的18F-FDG PET/CT显像前准备方法。更多还原

【Abstract】 1.Objective1.1 To probe the clinical utilization of Fluorine-18 fluorodeoxyglucose positron emissiontomography CT ( 18F-FDG PET/CT ) in diagnosis of malignant from nonmalignant heart andpericardium lesions, in order to assess the feasibility of 18F-FDG PET-CT for the differentiationof malignancy from benign manifested as heart and pericardium lesions,1.2 To find out the optimal cut-off value of maximum standard uptake values ( SUVmax ) andSUVmax lesion/blood for diagnosing the heart and pericardium lesions, than calculate thediagnostic sensitivity, specificity, accuracy of CT and PET-CT respectively. To make a 18F-FDGPET/CTdiagnostic criteria for heart and pericardium lesions by comprehensive analysis.2.Materials and Methods2.1 Object of studyA total of 29 cases ( malignancy: benign = 17: 12 ) with cardiac and pericardial lesions wereanalyzed in the present study, all of which were pathologically or clinically confirmed.2.2 FDG-PET/CT scanning2.2.1 Preparation for examination2.2.2 Scanning methodsAll patients were examined by 18F-FDG PET/CT and enhancement CT scan.2.2.2 Image analysisAll lesions were evaluated semi-quantitatively using maximum standard uptake values (SUVmax) and SUVmax lesion/blood, and the density of the heart and pericardium lesions andthe relation with surrounding tissues were evaluated. To find out the optimal cut-off value ofmaximum standard uptake values ( SUVmax ) and SUVmax lesion/blood for diagnosing theheart and pericardium lesions, Than calculate the diagnostic sensitivity, specificity, accuracy ofCT and PET-CT respectively.2.2.3 Statistical MethodsAll the measurement data was performed by the Normal Test. The differences of SUVmaxand SUVmax lesion/blood of benign between malignant lesions were analyzed byIndependent-Samples T Test or T’Test ( Normal ), or Mann-Whitney U Test ( Non-normaldistribution ). The statistic difference in sensitivity, specificity, accuracy, positive predictivevalue and negative predictive value of CT and PET/CT were compared by x2 test. Thestatistically significant difference was set at P<0.05. All the statistical analysis was made usingSPSS 13.0 software package.3.ResultsThe SUVmax of the benign lesions ranged from 0.5 to 7.4 (mean 2.26±1.82), and theSUVmax of the malignancy ranged from 2.6 to 21.2 (mean 8.37±5.20). The showed significantdifference ( t = 3.473, P = 0.000), the SUVmax Lesion/Blood of malignancy 1.15±1.06 andbenign 4.24±2.02 also prohibited significant difference(t = 4.842, P = 0.000).The optimal cut-off value of SUVmax is 3.5 and SUVmax Lesion/Blood is 1.8～2.2.The sensitivity, specificity, accuracy of CT and PET-CT were 64.7% ( 11/17 ), 75.0%( 9/12 ), 69.0% ( 20/29 ) and 100.0% ( 17/17 ), 83.3% ( 10/12 ), 93.1% ( 27/29 ) respectively.There was a significant differenec between CT and PET/CT in accuracy (P = 0.016)(McNemarTest).4.Conlusion18F-FDG PET-CT can correctly predict benignity or malignancy of cardiac and pericardiallesions. 1. ObjectiveMyocardial 18F-FDG uptake in PET scans in patients prepared by the usual fasting protocolmay result in difficulties in interpretation because variable uptake may yield false-positive resultsregarding mediastinal abnormalities. We aimed to analyze, retrospectively, the effect of diet onmyocardial FDG uptake.2. Materials and Methods2.1 Object of studyA total of 30 cases with cardiac and pericardial lesions were performed by PET/CT.2.2 FDG-PET/CT scanning2.2.1 Preparation for examinationThe "fasting" group comprised 17 consecutive patients before a clinical change in thepatient preparation protocol. The "new diet" group comprised 60 consecutive patients after theclinical protocol change who were directed to consume a very high-fat, low-carbohydrate,protein-permitted diet before FDG injection. All patients were given a questionnaire that wasused to verify diet adherence.2.2.2 Statistical MethodsAll the measurement data was performed by the Normal Test. The differences of SUVmaxand the blood glucose level of "fasting" group between The "new diet" group were analyzed byIndependent-Samples T Test or T’Test ( Normal ), or Mann-Whitney U Test ( Non-normaldistribution ). All the statistical analysis was made using SPSS 13.0 software package. 3. ResultsThe SUVmax of the "fasting" group ranged from 2.2 to 19.3 (median 3.1), and the SUVmaxof the"new diet" group ranged from 0.7 to 3.4 (median 2.1). The showed significant differenceby Mann-Whitney U Test ( Z = -3.333, P = 0.001<0.01),The blood glucose level of the "fasting" group ranged from 3.7 to 9.0 (mean 6.73±1.77),and the lood glucose level of the"new diet" group ranged from 4.2 to 8.6 (mean 5.82±1.05). Theshowed significant difference by t Test ( t = 1.746，P = 0.092),4. ConlusionA very high-fat, low-carbohydrate, protein-permitted diet before FDG injection suppressesmyocardial FDG uptake. This should facilitate definition of mediastinal abnormalities on FDGPET. It is a more reasonable patient preparation protocol for the patients with heart andpericardium lesions.更多还原