Özet:

Aim: Lymphoma staging is important from the prognostic and therapeutic point of view and bone marrow biopsy performed for this purpose makes up a large part of current hematopathology practice. PET-CT is the currently preferred method to evaluate bone marrow involvement by lymphoproliferative disorders as it is non-invasive and practical. The aim of this study was to compare trephine biopsy and PET-CT results as regards determining bone marrow involvement in various lymphoma subtypes. Materials and Methods: A total number of 198 bone marrow biopsies and 185 PET-CT images of cases consisting of various lymphoma subtypes were included in the study. The results of both methods evaluated for bone marrow infiltration were grouped as positive, negative, and suspicious to compare consistency. Statistical agreement was calculated with the kappa coefficient. The sensitivity, specificity, and the positive and negative predictive values were calculated as the diagnostic test measures. Results: Twenty-six cases (15.8%) had lymphoma involvement in trephine biopsies and 36 cases (21.2%) had positive findings for lymphoma involvement of the bone marrow on PET-CT. The two methods had weak statistical agreement (κ= 0.21). Biopsy and PET- CT results were similar in 132 cases (71.20%). PET-CT showed false negative results in 11 cases in which infiltration was observed with biopsy. Twenty-six cases that were negative for lymphoma involvement on biopsy were accepted as positive on PET-CT and 16 of these cases were classical Hodgkin's lymphoma. One case with suspicious bone marrow biopsy was positive on PET-CT, while four cases with suspicious PET-CT results were positive on biopsy. Conclusion: The results showed that both of the methods have advantages and disadvantages as regards lymphoma staging. However, histopathology is globally accepted as the gold standard for a definite diagnosis. We believe that the complementary use of the two methods is more beneficial for correct guidance during clinical practice.

Anahtar Kelimeler: