Dual-Energy Subtraction Radiography versus Conventional Radiography: Enhancing Pulmonary Nodule Detection

Purpose: This literature review compared the accuracy and sensitivity of dual-energy subtraction radiography (DESR) to conventional radiography (CR) in the detection of pulmonary nodules. We hope to showcase DESR as an imaging aid for detecting pulmonary nodules.

Methods/Materials: The authors searched Pubmed/MEDLINE using the terms “Dual energy subtraction radiography,” and “Subtraction Radiography.” Only studies involving the diagnosis of pulmonary nodules and those comparing DESR to conventional plain-film chest radiography were included. The following data was recorded: Reader Type (Resident or Attending Physician or Both), Number of radiographs, Number of nodules, diameter of nodules (in mm), and Energies used (Kvp). The quantitative outcomes were the difference of Receiver Operating Characteristic Area under the curve (ROC AUC) and difference in sensitivity between DESR and CR.

Results: A total of 17 studies were analyzed. A total of 1365 patients were included in the review, 465 of whom were controls without any nodules present. A total of at least 1661 nodules were identified among the 900 patients with pulmonary nodules. The mean/median of nodule size ranged from 5 mm to 27 mm.

The difference between DESR sensitivity (mean = 0.594, SD = 0.144) and CR sensitivity (mean = 0.454, SD = 0.914) was 0.14 (95% CI: 0.084 – 0.196), and this difference was statistically significant (t(8) = 5.78, p < 0.05)

The difference between DESR AUC (mean = 0.794, SD = 0.139) and CR ROC AUC (mean = 0.726, SD = 0.122) was 0.068 (95% CI: 0.038 - 0.984), and this difference was statistically significant (t(12) = 4.90, p < 0.05).

Conclusions: DESR showed superior sensitivity and ROC AUC values compared to CR in detecting pulmonary nodules. This shows how DESR is superior to CR in discriminating between classes (higher ROC AUC) and more effective in identifying true positives (higher sensitivity). DESR may enhance radiologist capacity to detect pulmonary nodules and may be used to augment physician training and education.