OBJECTIVE: To determine the clinical-histopathologic concordancy in lymph node biopsies as well as the sensitivity and specificity of the clinical diagnoses in our institution. METHODS: We reviewed the files of patients with lymph node biopsy performed from January 1991 to July 1997. We estimated the clinical-histopathologic concordancy with a kappa test. Our institution lacked the immunochemistry and ultrastructural techniques complementing the diagnoses. RESULTS: There were 72 biopsies. Most were from the cervical (N = 27), mesenteric (N = 13) or supraclavicular (N = 12) regions. The supraclavicular lymph nodes had the highest sensitivity for clinical diagnosis (S = 67%) and was mostly associated with neoplasias (lymphoma or metastatic cancer) in 75% of the cases. The global concordancy was fair (kappa 0.37) and the sensitivity and specificity of the clinical diagnoses were 50%. CONCLUSIONS: The concordancy as well as the clinical diagnostic sensitivity and specificity were low in our hospital. This justifies the efforts of continuing medical education in our institution to reduce errors in the diagnostic process.
OBJECTIVE: To determine the clinical-histopathologic concordancy in lymph node biopsies as well as the sensitivity and specificity of the clinical diagnoses in our institution. METHODS: We reviewed the files of patients with lymph node biopsy performed from January 1991 to July 1997. We estimated the clinical-histopathologic concordancy with a kappa test. Our institution lacked the immunochemistry and ultrastructural techniques complementing the diagnoses. RESULTS: There were 72 biopsies. Most were from the cervical (N = 27), mesenteric (N = 13) or supraclavicular (N = 12) regions. The supraclavicular lymph nodes had the highest sensitivity for clinical diagnosis (S = 67%) and was mostly associated with neoplasias (lymphoma or metastatic cancer) in 75% of the cases. The global concordancy was fair (kappa 0.37) and the sensitivity and specificity of the clinical diagnoses were 50%. CONCLUSIONS: The concordancy as well as the clinical diagnostic sensitivity and specificity were low in our hospital. This justifies the efforts of continuing medical education in our institution to reduce errors in the diagnostic process.