Computer program supporting the diagnostic accuracy of cellular BALF analysis: a new release

Respir Med. 2001 Oct;95(10):781-6. doi: 10.1053/rmed.2001.1153.

Abstract

Recently we developed a validated computer program based on polychotomous logistic regression analysis using bronchoalveolar avage fluid (BALF) results to distinguish between the three most common interstitial lung diseases (ILD): sarcoidosis, idiopathic pulmonary fibrosis (IPF) and extrinsic allergic alveolitis (EAA) or drug-induced pneumonitis. One of the limitations of this program was that it was not useful in discriminating between infectious disorders and non-infectious disorders. Therefore, we added BALF samples obtained from patients with a confirmed bacterial pulmonary infection based on culture results > or = 10(4) cfum l(-1) (group I: n=31) to the study population mentioned above (group II: n=272). Notably, just one variable, i.e. the percentage of polymorphonuclear neutrophils, allowed us to distinguish between infectious and non-infectious disorders. The agreement of predicted with the actual diagnostic group membership was 99.67% (groups I and II). Additionally, 91.2% of the cases with ILD were correctly classified. In conclusion, this updated Windows version 2000 of the validated computer program provides a very reliable prediction of the correct diagnosis for an arbitrary patient with suspected pneumonia or with ILD given information obtained from BALF analysis results, and is thought to improve the diagnostic power of BALF analysis.

MeSH terms

  • Alveolitis, Extrinsic Allergic / pathology
  • Bacterial Infections / pathology
  • Bronchoalveolar Lavage Fluid / cytology*
  • Diagnosis, Computer-Assisted*
  • Diagnosis, Differential
  • Humans
  • Logistic Models
  • Lung Diseases, Interstitial / pathology*
  • Predictive Value of Tests
  • Pulmonary Fibrosis / pathology
  • Sarcoidosis / pathology