A study of the pathological outcome of positron emission tomography-computed tomography 2-(18F)-fluro-2-deoxy-D-glucose avid lesion: A 5 years retrospective study.

Abstract

Positron emission tomography-computed tomography (PET-CT) has been commonly used for staging and follow-up in cancer patients. The present study compares radiological and pathological outcomes at all the sites. The benign nonphysiological uptake reduces the specificity of the modality due to high false positive (FP) rate although sensitivity for malignant lesions may be high. To study the sensitivity, specificity, positive and negative predictive value (PPV and NPV) of PET-CT in the detection of malignant lesions for all sites using pathological and final clinical outcome. A retrospective study of 195 cases of PET-CT detected lesions subjected to pathological diagnosis in the form of fine-needle aspiration cytology (FNAC) and/or Tru-cut biopsies were performed on patients with proven or suspected malignancy over a period of 1-year (2009) with a 5 years follow-up. During the same period, 2900 PET-CT imaging studies were performed, of which 195 were suspected to be malignant or benign. Of these, 193 patients were subjected for tissue diagnosis for confirmation. FNAC smears and Tru-cut biopsy were prepared and examined as per standard protocols. Of 195 aspirates in 183 aspirates, a conclusive tissue diagnosis was rendered. The follow-up was available in 79 cases of suspicious PET avid lesions for a period of 1-5 years. The PET-CT correlation with the tissue diagnosis and clinical outcome showed the sensitivity of 97.7% and an overall accuracy of 83% for malignant lesions. However, due to a large number of FP (n = 28) the specificity was only 43% and FP rates were 57%. PPV and NPV for malignant lesions was 82.4% and 87.5%, respectively. PET-CT is a sensitive investigation for detection of malignant lesions in treated and newly diagnosed cases of malignancy.