OBJECTIVE: The objective of this study was to evaluate the accuracy of preoperative clinical staging with computed tomography in predicting the correct pathological stage. METHODS: Medical records of non-small cell lung cancer (NSCLC) patients treated, from 1990 to 2005 were reviewed. Clinical stage was based on routine preoperative clinical and imaging evaluation. Positron emission tomography was not routinely performed. Suspected lesions, that would preclude a surgical resection, were pathologically confirmed. The pathological stage was based on final postoperative or biopsy pathological assessment. A correlation table between clinical and pathological stages was generated. Cohen's kappa index, sensitivity, specificity, positive and negative predictive values and accuracy were calculated. RESULTS: Records of 291 patients were reviewed. Clinical stages Ia, Ib, IIa, IIb, IIIa, IIIb and IV were found respectively in 8.9%, 31.9%, 0.3%, 18.6%, 25.4%, 11% and 3.8%. Pathological staging was different from clinical staging in 33% (15% were upstaged and 18% downstaged). Sensitivity, specificity, positive and negative predictive values and accuracy for clinical staging were 78%, 69%, 82%, 64% and 67%, respectively. Cohen's kappa index was 0.574 (P < 0.001). CONCLUSION: Preoperative clinical staging presents limited efficacy for the correct staging of NSCLC patients from this sample of Brazilian population.
Lung neoplasms; Neoplasm staging; Tomography; Spiral computed
