Date of Award


Document Type


Degree Name

Master of Medicine (MMed)


Pathology (East Africa)


Introduction: Lymphoma diagnosis integrates clinical, morphological, immunophenotypical and molecular genetic features as evident from the World Health Organization (WHO) classification of lymphoid malignancies.(1, 2) Although it is standard practice for histopathology laboratories in developed countries to confirm and further sub classify lymphomas using immunohistochemistry and other molecular techniques, the same is not true for developing countries like Kenya. The reasons that prevent the performance of these tests are the prohibitive costs and inadequate trained personnel.

Objectives: The aims of this study were to retrospectively classify lymphomas diagnosed at The Aga Khan University Hospital (AKUHN) Laboratory based on the WHO classification system and to describe the demographic and clinical characteristics of the patients.

Methods: Tissue blocks of 103 consecutive, lymphoma specimens received between 1st January 2007 and 31st December 2008 were retrieved from the archives at the Department of Pathology. Haematoxylin and Eosin slides were prepared and reviewed for purposes of confirmation and morphologic classification. Immunohistochemical expression for various lymphoma markers were assessed and scored. Findings were correlated with relevant demographic, pathologic and clinical data.

Data Management and Analysis: Raw data was entered into Microsoft excel data sheets and subsequently analyzed using Statistical Package for Social Sciences (SPSS) version 17.0 database.

Results: A total of 103 cases were analyzed. 79 cases representing 76.7% of the studied population were classified as Non Hodgkin lymphoma (NHL), 22 cases representing 21.4% of the study population were diagnosed as Hodgkin lymphoma (HL), while 2 cases representing 1.9% of the study population were diagnosed as Non hematological malignancies.

Diffuse large B cell lymphoma (DLBCL) represented 63% of all NHL cases. The second most common subtype was Burkitt lymphoma (BL) with 8 cases representing 10.1% of all NHL cases, followed by chronic lymphocytic leukemia (CLL) and acute lymphocytic leukemia (ALL) with 6 cases each. There were 3 cases of Mucosal associated lymphoid tissue (MALT) lymphoma, 3 cases of Plasmacytoma and 1 case of Follicular lymphoma (FL).

Of the 22 cases of HL, 19 cases (86.4%) were of the classical subtype while 3 (13.6%) were Hodgkin Lymphoma - Nodular Lymphocyte Predominant (HLNLP) subtype. All the 19 cases of classical subtype (100%) showed positivity for LMP1 Protein indicating Epstein Barr Virus (EBV) infection, whereas all the 3 cases of HLNLP were negative for LMP1.

One case which did not take up immunohistochemical stain due to poor tissue fixation was labeled as “favor NHL” by a consensus panel. This case was included in the NHL group

Conclusion: Independent of HIV status, DLBCL was the most common NHL subtype accounting for 64.6% of all NHL cases and 50.5% of all lymphomas.

In contrast to western countries in which it occurs more frequently in the elderly,(2) 49 % of all DLBCL cases in this study occurred in patients aged 40 years or younger. Overall, 60 % of all Lymphomas occurred in patients aged 40 years or under while 50 % of all HL cases occurred in patients aged 30 years or younger.

The results of this study also show that, EBV appears to play a more direct role in the pathogenesis of HL, as it does for endemic Burkitt lymphoma (BL).(3)

This study also highlights the importance of standardized pre analytical, analytical and post analytical variables so that an accurate, timely and well communicated diagnosis can be relayed to the clinician.