Intraobserver and interobserver agreement of volume perfusion CT (VPCT) measurements in patients with lung lesions

Department of Diagnostic and Interventional Radiology, University Hospital of Tuebingen, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany.
European journal of radiology (Impact Factor: 2.37). 07/2011; 81(10):2853-9. DOI: 10.1016/j.ejrad.2011.06.047
Source: PubMed


To evaluate intraobserver and interobserver agreement of manually encompassed lung lesions for perfusion measurements using volume-perfusion computed tomography (VPCT).
Institutional review board approval and informed consent were obtained. HIPAA guidelines were followed. A 65-s dynamic study was acquired with scan parameters 80 kV, 60 mAs (80 mAs for patients ≥ 70 kg), 128 × 0.6mm collimation. Blood flow (BF), blood volume (BV) and K(trans) parameters were determined by syngo volume perfusion CT body with 88 lesions analyzed retrospectively.
Within-subject coefficients of variation for intraobserver agreement (range 6.59-12.82%) were superior to those for interobserver agreement (range 21.75-38.30%). Size-dependent analysis revealed lower agreements for lesions <4 cm as compared to larger lesions. Additionally, agreements of the upper, middle and lower lung zones were different.
Intraobserver agreement was substantial for VPCT lung cancer perfusion measurements encouraging the use for tumor characterization and therapy response monitoring. Interobserver agreement is limited and unexperienced readers should be trained before using this new method.

16 Reads
  • Source
    • "Currently, the preferred method to minimize respiratory motion is to instruct patients to hold their breath or to use shallow breathing. However, motion correcting software is beginning to emerge both in descriptive as well as in comparative studies (Sauter et al., 2012a,b, 2013). In the future, the use of modern respiratory-gated, 256-or 320-detector row CT may improve misregistration through more extensive coverage , while reducing respiratory artefacts. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Lung cancer represents an increasingly frequent cancer diagnosis worldwide. An increasing awareness on smoking cessation as an important mean to reduce lung cancer incidence and mortality, an increasing number of therapy options and a steady focus on early diagnosis and adequate staging have resulted in a modestly improved survival. For early diagnosis and precise staging, imaging, especially positron emission tomography combined with CT (PET/CT), plays an important role. Other functional imaging modalities such as dynamic contrast-enhanced CT (DCE-CT) and diffusion-weighted MR imaging (DW-MRI) have demonstrated promising results within this field. The purpose of this review is to provide the reader with a brief and balanced introduction to these three functional imaging modalities and their current or potential application in the care of patients with lung cancer.
    Full-text · Article · Dec 2013 · Clinical Physiology and Functional Imaging

  • No preview · Article · Mar 2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate the capability of first-pass volume perfusion computed tomography (PCT) for differentiation of solitary pulmonary nodules (SPNs) and to compare that of combination of PCT and routine CT with CT alone for the differentiation. Our institutional review board approved this study and informed consent was obtained. With nine excluded, 65 consecutive patients having a SPN with histopathologic proof or follow-up underwent a 30s PCT using the deconvolution model were evaluated. Kruskal-Wallis tests and receiver operating characteristics (ROC) analysis were underwent. Four radiologists assessed nodules independently and retrospectively. Diagnostic capability was compared for CT alone and PCT plus CT. ROC analysis, McNemar test, and weighted kappa statistics were performed. Significant differences were found in parameters between malignant and benign nodules (p<0.0001 for blood flow, blood volume, and permeability surface area product), SPNs were more likely to be malignant by using threshold values of more than 55ml/100g/min, 2.5ml/100g, and 10ml/100g/min, respectively. PCT plus CT was significantly better in overall sensitivity (93%, p=0.004) and accuracy (94%, p=0.003) compared to CT alone, not specificity (96%). Area under the curve for ROC analyses of PCT plus CT was significantly larger than that of CT alone (p=0.018). Mean weighted kappa for PCT plus CT was 0.715, that for CT alone was 0.447. Volume first-pass PCT can distinguish SPNs. Using PCT plus routine CT may be more sensitive and accurate for differentiating malignant from benign nodules than CT alone and allows more confidence and constancy.
    No preview · Article · May 2012 · European journal of radiology
Show more