Home > Riviste > Minerva Pneumologica > Fascicoli precedenti > Minerva Pneumologica 2016 December;55(4) > Minerva Pneumologica 2016 December;55(4):85-93

ULTIMO FASCICOLO
 

JOURNAL TOOLS

eTOC
Per abbonarsi
Sottometti un articolo
Segnala alla tua biblioteca
 

ARTICLE TOOLS

Estratti
Permessi
Per citare questo articolo

 

ORIGINAL ARTICLES   

Minerva Pneumologica 2016 December;55(4):85-93

Copyright © 2016 EDIZIONI MINERVA MEDICA

lingua: Inglese

Transbronchial needle aspiration: ThinPrep versus the conventional smear technique in the diagnosis of lung cancer

Valentina CONTI 1, Martina MARCHI 2, Cecilia TURRINI 2, Marco CONTOLI 2, Rosa RINALDI 3, Claudio CAVICCHI 3, Alberto PAPI 3, Franco RAVENNA 4

1 Unit of Pneumology, San Marino Republic State Hospital, Borgo Maggiore, San Marino; 2 Pneumology Unit, Department of Emergency, University Hospital Arcispedale S. Anna of Ferrara, Cona, Ferrara, Italy; 3 Pathology Unit, Department of Diagnostic Imaging and Laboratory Medicine, University Hospital Arcispedale S. Anna of Ferrara, Cona, Ferrara, Italy; 4 Unit of Pneumology, Carlo Poma Hospital, Mantova, Italy


PDF


BACKGROUND: Approximately 70% of patients with lung cancer present at an advanced stage of their disease and are therefore unresectable. Bronchoscopy is often the first procedure performed, and the diagnosis is usually based on small biopsies and cytology specimens. Cytology is assuming an increasingly relevant role, not only to make a pathologic diagnosis but also to obtain a biomolecular profile of the cancer. Liquid-based cytology, such as ThinPrep (TP), unlike the conventional smear (CS) technique, attempts to preserve enough material for ancillary tests, such as immunochemistry and molecular tests. The aim of this study was to evaluate the role of TP compared to CS in the diagnosis of lung cancer using transbronchial needle aspiration (TBNA) to obtain samples.
METHODS: We performed bronchoschopy on 101 consecutive patients with suspicious pulmonary lesions. For each patient, approximately 4 cytological samples were collected using TBNA. The 4 samples were examined, alternating TP and CS techniques. For each patient, the method employed in the examination of the first sample was decided using a randomized list. The cytopathologic results from each method were compared to the definitive diagnosis obtained through bronchoscopy techniques, thoracic surgery, other invasive procedures or clinical follow-up.
RESULTS: In 81 of the 101 cases (81%), lung cancer diagnosis was confirmed. In 20 cases (19%), the disease was benign. CS and TP have shown sensitivities of 57% and 58%, specificities of 100% and 100%, positive predictive values of 100% and 100%, negative predictive values of 39% and 46% and accuracies of 66% and 74%, respectively. ROC curves were then used to compare the two methods; the difference was not statistically significant (P=0.415). Inadequate material was observed in 7 cases with CS and in no cases with TP. In 19 patients (19%), a rapid on-site evaluation (ROSE) was also performed. In these cases, the sensitivity, negative predictive value and accuracy of TBNA were 85%, 75% and 90%, respectively; in the remaining 82 cases in which ROSE was not employed, the CS and TP results were 54%, 33% and 62%. In particular, the difference detected was statistically significant for NPV (P<0.0005) and accuracy (P<0.05) when comparing ROSE to CS or TP alone using Fisher’s Exact Test.
CONCLUSIONS: We demonstrated that the diagnostic yield of TP is substantially equal to that achieved with CS in lung cancer disease. TP offers advantages in terms of reproducibility, easier slide evaluation, and above all, the possibility of molecular testing. However, CS allows the use of ROSE, which considerably improves TBNA accuracy.
We believe that the two methods could be used together in the diagnosis of lung cancer, integrating their respective advantages to choose the most appropriate and personalized treatment for the patient.

inizio pagina