ABSTRACT
INTRODUCTION: Tuberculosis (TB) is a common disease with profound morbidity, mortality and effects on global public health. The differential diagnosis of lymphadenopathy is wide, particularly in areas where HIV is prevalent. Most hospitals in sub-Saharan Africa and across the developing world have limited, if any, histology facilities. This study will assess the validity of the assessment of the lymph nodes by their macroscopic appearance for the diagnosis of TB. Its sensitivity and specificity will be compared to full histological examination. METHOD: This is a single-centre prospective study conducted in a remote rural district hospital in Zambia over a time period of 16 months. All patients with palpable lymphadenopathy where TB was considered in the differential diagnosis were included. The patients underwent an excision lymph node biopsy. The cut surface was judged by the operating surgeon as to the presence of caseation. The excised nodes were then sent for histological examination. RESULTS: In total, 59.8% of patients (64 of 107 patients) in this group had a final histological diagnosis of TB. This is equivalent to the TB disease prevalence in this group of patients that have palpable lymphadenopathy in this population. The diagnostic sensitivity based on macroscopic appearance of the lymph node was 81.25% (95% CI, 69.5-89.9%). The specificity was 97.67% (95% CI, 87.7-99.6%). CONCLUSION: This study has shown that lymph node appearance is a useful diagnostic test even without laboratory histopathological facilities in the diagnosis of TB. This assessment of the macroscopic appearance is both sensitive and specific.
