Factors affecting diagnostic accuracy of CT-guided coaxial cutting needle lung biopsy: retrospective analysis of 631 procedures

J Vasc Interv Radiol. 2003 May;14(5):581-8. doi: 10.1097/01.rvi.0000071087.76348.c7.

Abstract

Purpose: To analyze variables affecting diagnostic accuracy of computed tomography (CT)-guided percutaneous coaxial cutting needle biopsy of lung lesions.

Materials and methods: A retrospective analysis of factors affecting diagnostic accuracy of CT-guided percutaneous coaxial cutting needle lung biopsy was performed in 631 consecutive procedures with confirmed final diagnoses. Benign and malignant needle biopsy results were cross-examined with correct and incorrect final outcomes to determine diagnostic accuracy. Factors affecting diagnostic accuracy were determined by multivariate logistic regression analysis of variables thought to affect diagnostic accuracy. A P value less than 0.05 was interpreted as statistically significant.

Results: The overall diagnostic accuracy of CT-guided percutaneous coaxial cutting needle biopsy of lung lesions was 95% (95% CI: 92.7%-96.2%). Sensitivity was 93%, specificity 98%, negative predictive value 6%, positive predictive value 99%, false-positive rate 0.7%, and false-negative rate 15%. The factors affecting diagnostic accuracy were final diagnoses (benign, 86%; malignant, 99%; chi(2) test, P < 0.001) and lesion size (lesions <1.5 cm, 84%; lesions 1.5-5.0 cm, 96%; lesions >5 cm, 93%; chi(2) test, P = 0.06).

Conclusion: Benign lung lesions, lung lesions smaller than 1.5 cm (which pose technical difficulty), and lung lesions larger than 5 cm (which are associated with a higher necrosis rate) affect diagnostic accuracy of CT-guided percutaneous coaxial cutting needle biopsy of lung lesions.

MeSH terms

  • Biopsy, Needle* / methods
  • Female
  • Humans
  • Lung / pathology*
  • Lung Diseases / diagnosis*
  • Lung Diseases / diagnostic imaging
  • Male
  • Middle Aged
  • Predictive Value of Tests
  • Radiography, Interventional*
  • Retrospective Studies
  • Sensitivity and Specificity
  • Tomography, X-Ray Computed*