Answers are grounded in this paper and cite their sources.
Retrospective analysis compares pre-test diagnoses for mediastinal lymphadenopathy, suggesting implications for diagnosis accuracy.
OBJECTIVE To compare pre-test diagnoses of mediastinal lymphadenopathy with histology obtained from cervical mediastinoscopy. METHODS Retrospective study of cervical mediastinoscopy outcomes between January 2009 and January 2011. We classified patients by histology and elicited whether the pre-test diagnosis was confirmed, supported or refuted by histology. The five histological groups were: i) non-caseating granuloma, ii) caseating granuloma, iii) undifferentiated granuloma, iv) malignant cells, and v) other (non-diagnostic tissue). RESULTS A total of 323 patients underwent cervical mediastinoscopy. 36% (n=115) had non-caseating granulomas, 11% (n=34) caseating granulomas, 2% (n=5) undifferentiated granulomas, 27% (n=87) malignant cells and 26% (n=82) were non-diagnostic. Patients with a pre-test diagnosis of tuberculosis or malignancy were more likely to have this confirmed by histology. Malignancy was confirmed in 42.6% of cases, where cervical mediastinoscopy was performed for staging (n=129), and excluded in 57.4%. Relationship between histology and pre-test differential diagnosis / diagnosis. Lymph node histology (n, %) No. (%) Mean age (yrs) Sex (% male) Proportion of patients where the pre-test diagnosis was confirmed by histology (%) (i) Non-caseating granulomas 115 (36%) 51 61.7 52.2 (ii) Caseating granulomas 34 (11%) 46 47 93.9 (iii) Undifferentiated granuloma 5 (2%) 40 40 80 (iv) Malignant cells 87 (27%) 67 64.4 98.6 (v) Other 82 (26%) 65 64.6 1.2 CONCLUSION A pre-test clinical diagnosis of tuberculosis was most predictive of histology supporting such a diagnosis.
No takes yet. Share an insight, caveat, or question.
Woolnough et al. (2013) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: