Incomplete fissures in severe emphysematous patients evaluated with MDCT: incidence and interobserver agreement among radiologists and pneumologists

Eur J Radiol. 2012 Dec;81(12):4161-6. doi: 10.1016/j.ejrad.2012.06.006. Epub 2012 Jul 6.


Objective: Pulmonary fissures completeness predicts efficacy in endobronchial valves (EBV) implantation, a new lobar volume reduction therapy for severe emphysematous patients. We assessed the incidence of incomplete fissures and the interobserver agreement in its evaluation with MDCT, in severe emphysematous patients prior to EBV implantation.

Materials and methods: Volumetric thin-section CT scans of 35 patients (CODP GOLD 3/4, heterogeneous emphysema) were retrospectively reviewed by 2 pneumologists, 1 general and 2 experienced chest radiologists, independently and blinded for treatment outcome, and the pulmonary fissures were classified as either complete or incomplete. Interobserver agreement was assessed with Kappa index (KI).

Results: Agreement between all readers for the left oblique, right oblique and horizontal fissure was, respectively, moderate (KI=0.53), fair (KI=0.37) and moderate (KI=0.42). Highest agreement (99/105 fissures) was observed among experienced radiologists, being for left oblique, right oblique and horizontal, respectively, almost perfect (KI=0.79), perfect (KI=1.0) and moderate (KI=0.52). These 2 reviewers found that all of 35 patients had at least one incomplete fissure, with a proportion of incomplete fissures assigned as 74/65%, 85/85% and 91/88%, respectively for the left oblique, right oblique and horizontal fissures.

Conclusions: Pneumologists and radiologists agreed fairly to moderately in fissures analysis, while the experienced chest radiologists reached the highest clinically adequate agreement of 94%. We believe that clinical routine visual analysis of the fissures integrity can be done with a good degree of confidence in MDCT images, and experienced readers might be required. Also, a higher than expected incidence of incomplete fissures was described in our studied population.

Publication types

  • Comparative Study

MeSH terms

  • Aged
  • Aged, 80 and over
  • Emphysema / diagnostic imaging*
  • Emphysema / epidemiology*
  • Female
  • Germany / epidemiology
  • Humans
  • Incidence
  • Male
  • Middle Aged
  • Observer Variation
  • Pattern Recognition, Automated / methods*
  • Professional Competence / statistics & numerical data*
  • Pulmonary Medicine / statistics & numerical data*
  • Radiology / statistics & numerical data*
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Tomography, X-Ray Computed / statistics & numerical data*