Characterization of Hilar and Mediastinal Foci on ¹⁸F-FDG PET: New Variables

TO THE EDITOR: The article by Karam et al. (1) on the features of ¹⁸F-FDG–concentrating bilateral hilar foci associated with benign or malignant pathologies draws the reader's attention to an important and relevant issue of current PET practice in oncology. The need to develop a logical decision-making algorithm in this confounding situation and educate the interpreting physicians about such an approach to eliminate errors can hardly be overemphasized. I would like to congratulate the authors for their well-done analysis that attempts a solution to this common diagnostic dilemma. Despite significant recent progress in PET technology (e.g., emergence of fusion PET/CT and novel quantitative approaches), accurate characterization of ¹⁸F-FDG–concentrating mediastinal and hilar nodes continues to pose a major diagnostic challenge worldwide. This challenge also has geographic relevance and is of major concern in Asian countries, including India, where tuberculosis has a high prevalence, and hence, the possibility of encountering false-positive ¹⁸F-FDG PET lesions is higher than in the West. Tuberculous lesions can demonstrate variable ¹⁸F-FDG uptake as determined by inflammatory activity (2–5). ¹⁸F-FDG PET is being investigated for its potential in the assessment of disease activity in a wide array of infectious and aseptic inflammatory conditions (5,6). Hence, scientific analyses of the patient data of the past 2 decades are much needed. Analysis of data that can give important clues about the diagnosis will be of help in devising an optimal approach, ultimately obviating invasive biopsies and reducing patient anxiety.

As the authors indicate (1), their study probably represents the first attempt to scientifically correlate multiple PET variables with ultimate outcome. The results of their retrospective analysis reaffirm certain traditional notions about the significance of variables such as symmetry, maximum standardized uptake value, node size on CT, and stability of ¹⁸F-FDG uptake during the course of the disease—variables that are being followed in several PET centers in various countries. All these characteristics are important and performed as expected in determining the nature of the lesions in the study of Karam et al. In addition, the results underscore the importance of the absence or presence of ¹⁸F-FDG–avid foci in nonhilar mediastinal nodes, a variable (which they term the “purity” of the lesion) that was found to be an independent determinant after multivariate analysis. Seventy-nine percent of impure scans, versus 18% of pure scans, represented malignancy in the examined population. Also, the significant dependence of the nature of the lesions on the primary malignancy is important to note (75% of patients with colorectal carcinoma were found to harbor benign lesions, whereas the corresponding figures for patients with breast carcinoma, lymphoma, and other malignancies were 34%, 49%, and 37%, respectively). Such retrospective analyses from other centers, especially those from the other parts of the world, are urgently required, as new PET variables that are important for clinical decision making in this setting may arise. It will be important to observe the impact of these new variables in multicenter analyses. If the impact is confirmed, the results will be a firm basis for developing an algorithm for the interpretation of ¹⁸F-FDG PET images. A separate analysis along similar lines in patients with lung carcinoma will also be of considerable value in clinical decisions about this important malignancy.