Diagnosis and grading of non-Hodgkin's lymphomas on fine needle aspiration cytology

Fine needle aspiration cytology (FNAC) enjoys popularity among clinicians worldwide, as a first line of investigation in all patients with lymphadenopathy and is preferred over biopsy because of its minimally invasive nature and cost-effectiveness. Although non-Hodgkin's lymphomas (NHL) are conventionally diagnosed and graded on biopsy specimens, it may be useful to be able to not only diagnose but also grade these cases on FNAC smears. The WHO and REAL classifications forming the basis of treatment in some centres rely on clinical features, immunocytochemistry and cytogenetics, which are beyond reach of most centres in the developing countries. This study therefore is aimed at diagnosing and grading NHLs on morphological parameters. The cytologic grading accuracy is compared with the histologic grades assigned according to the International Working Formulation (IWF) system which relies solely on morphological features, most important of which is cell size. Ninety five cases were retrieved over a 3 year period (May 2000 to April 2003). These were (i) cases where a cytological diagnosis of NHL or suspicious of NHL was made and corresponding histological sections available and (ii) cases where a diagnosis of NHL was made in histology and corresponding FNAC smears were available irrespective of the cytological diagnosis. The diagnostic accuracy of FNAC for NHLs was determined using histology as the gold standard. Cases were also graded on FNAC smears using a three tier grading system based upon cell size into low, intermediate and high grades. Cytologically assigned grades were correlated with the corresponding histological grades (IWF) to determine grading accuracy. An accurate diagnosis of NHL was thus possible in 67/95 (70.5%) cases. Overall accurate grading was seen in 65/95 (68%) cases using cytological criteria. Accurate cytologic grading was possible in 14/15 (93.33%) low grade, 11/18 (61.11%) intermediate and 40/62 (64.5%) high grade non Hodgkin's lymphomas. Kappa statistics revealed a very good agreement between cytological and histological grades for low grade NHL. The kappa scores for intermediate and high grade NHLs indicated moderate agreement. Using the two-tier system grading the kappa value for high grade lymphomas improved to 0.72, indicating good concordance. This study highlights the utility of FNAC as a morphological tool for diagnosing and grading NHLs in a significant number of cases. This modality may assist clinicians in management of cases of NHLs, especially in centres working within the constraints of limited availability or non availability of ancillary techniques.