IMAGING OF ONCOLOGIC PATIENTS - BENEFIT OF COMBINED CT AND FDG PET INTHE DIAGNOSIS OF MALIGNANCY

OBJECTIVE. The purpose of this study was to assess the benefit of comb ined CT and F-18-fluorodeoxyglucose (FDG) positron emission tomography (PET) in diagnosing malignancy. MATERIALS AND METHODS. The records of 26 patients with intraabdominal and intrathoracic OBJECTIVE. The purp ose of this study was to assess the benefit of combined CT and F-18-fl uorodeoxyglucose (FDG) positron emission tomography (PET) in diagnosin g malignancy. MATERIALS AND METHODS. The records of 26 patients with i ntraabdominal and intrathoracic neoplasms who underwent CT and FDG PET between January 1995 and September 1996 were retrospectively reviewed . Most of these patients had inconclusive findings on prior CT for the diagnosis of malignancy. Only sites of potential malignant disease we re included in the data danalysis. Presence or absence of malignancy w as confirmed by histopathology or follow-up CT. Three observers experi enced in abdominal imaging used CT findings alone to estimate level of suspicion (1 = definitely not malignant to 5 = definitely malignant) for primary or recurrent neoplasms (n = 21), distant metastases (n = 2 5), and neoplastic nodal involvement (n = 18). Six weeks later the thr ee observers reviewed the same CT examinations supplemented with FDG P ET and reestimated suspicion of malignancy. Receiver operating charact eristic methodology was used to analyze the results. Sensitivity, spec ificity, positive and negative predictive values, and accuracy in diag nosis of malignant disease were calculated using level 4 (probable mal ignancy) as the cutoff for the presence of disease. RESULTS. The mean area under the receiver operating characteristic curve, indicating suc cessful diagnosis of malignancy, was .82 for CT alone and .92 for CT w ith FDG PET (p<.05). The accuracies for diagnosis of primary or recurr ent neoplasms, distant metastases, and neoplastic nodal involvement we re 62%, 68%, and 83%, respectively, for CT alone and 81% (p =.06), 88% (p =.03), and 89% (p >.25), respectively, for CT with FDG PET. Also, supplemental FDG PET imaging improved observer confidence and accuracy in diagnosing recurrent neoplasm in four (36%) of Ii patients who had undergone surgery or chemoradiation and in diagnosing four (29%) of 1 4 extrahepatic sites that had potential metastases. CONCLUSION. Diagno sis of malignancy in oncologic patients is significantly improved when CT is supplemented with FDG PET. Combined imaging is particularly hel pful in the evaluation of potential recurrence in previously treated p atients and for diagnosing extrahepatic lesions that may be distant me tastases.