| Objective: Investigate the reasons of false positive and false negative results of 18F-FDG PET/CT imaging,in order to improve diagnostic accuracy and to reduce the misdiagnosis rate.Methods:The first diagnosis misdiagnosed seventy-two cases are screened out from 7,800 cases which were got 18F-FDG PET/CT examination from Sep.2005 to Jun.2014. The reasons of false positive and false negative results are investigated. All data were analyzed statistically by SPSS13.0 statistical software.Results: The total rate of false positives false negatives was 0.92%(72/7800). The followings are features.1. Twenty-four cases misdiagnosed with benign lesions and forty-eight cases misdiagnosed with malignant tumors by PET combined with CT. Location of involvement: in 72 misdiagnosed cases, there were 21 in lung(29.2%), 8 in generalized lymph nodes(11.1%), 7 in kidney(9.7%), 5 in stomach(6.9%), 5 in liver(6.9%), 5 in intestinal tract(6.9%) and 18 in other parts(25.0%).2. The reasons of false positive and false negative results of 18F-FDG PET/CT imaging are as follows:2.1 In the 24 cases of false positive : There were 7 cases due to tumor pathological type that influence the uptake of 18F-FDG;②There were 2 cases due to the high physiologic uptake of 18F-FDG that cower up tumor’s uptake; ③ There were 5 cases due to microscopic-sized tumors that influence the uptake of 18F-FDG;④There were 10 cases due to tumor’s nontypical appearance of imageology and nonspecific clinical manifestation.2.2 In the 48 cases of false negative:①There were 33 cases due to non-specific uptake of 18F-FDG:13 cases were hyperplasia and hybrid tuberculosis; 16 cases were inflammatory granuloma and nonspecificinflammatio;2 cases were fungal infection; 2 cases were changes after treatmen,such as aspiration、operation and radiotherapy。 ②There were 12 cases due to high physiological uptake or some benign tumors。③ There were 3 cases due to tumor’s nontypical appearance of imageology、combined with clinical closely not enough and limited experience of diagnosis。3. Results using the SUVmax 2.5 as criteria(SUVmax<2.5 benign and SUVmax≥2.5 malignance) are as follows: the sensitivity, specificity, accuracy, omission diagnostic rate, false positive rate, positive predictive value, negative predictive were 51.2%, 14.6%, 27.8%, 45.8%, 85.4%, 24.1% and 38.9% respectively. The most of the false positive is tuberculosis(7 cases), and the most of false negative is gastric adenocarcinoma(3 cases).4. The SUVmax is 6.5±5.76 in 24 malignant tumor, and 6.6±4.11 in 48 benign lesions. The difference was not statistically significant by t-test between two groups(t=0.932127, P>0.05).5. The difference was not statistically significant by t-test between the sizes or ages of two groups(P>0.05).6. The relations between tumor markers and metabolism: there was no difference between the SUVmax of groups statistically significant by t-test(P>0.05), which were grouped by the level of blood serum CEA, Ca125 and Ca199.Conclusions: The cases of false positive and false negative were most special pathological pattern, had atypical imaging features, or lack features of benign lesion or malignant tumor.The criteria of SUVmax 2.5 was not work in the 72 cases, the sensitivity and specificity were low. And two phases △SUVmax was the same. The most cause of the false-negative result were the small size and special pathologic type, especially in adenocarcinoma. There were tuberculosis, sarcoidosis, granuloma, and necrotizing lymphadenitis mostly misdiagnosed with malignant tumor. It is necessary to develop new tracer agent and need comprehensive analysis to reduce misdiagnosis. |
PDF Full Text Request