I note the following sentence statement - "A recent meta-analysis of the accuracy of ultrasonography in emergency departments in North America showed a pooled sensitivity estimate of 99.3% (96.6% to 100%), but a specificity of only 71% (60% to 80%).9 Transabdominal sonography is still widely used in emergency departments in North America, and the low specificity of the test makes it difficult to exclude an ectopic pregnancy."
For any moderate prevalence a test with high sensitivity is suited for exclusion of a diagnosis as it reliably will detect the condition of interest. If the test is negative a high degree of confidence is justified that the condition of interest is not in fact present (i.e. high negative predictive value). Indeed this was the conclusion of the authors cited (stein et al). The issue with ultrasound in the context cited is that its poor specificity means that it will categorise as ectopic a goodly number of non-ectopic gestations - in other words a positive result is not conclusive evidence of the presence of the condition (positive predictive value).
Rapid Response:
Re: Diagnosis and management of ectopic pregnancy
I note the following sentence statement - "A recent meta-analysis of the accuracy of ultrasonography in emergency departments in North America showed a pooled sensitivity estimate of 99.3% (96.6% to 100%), but a specificity of only 71% (60% to 80%).9 Transabdominal sonography is still widely used in emergency departments in North America, and the low specificity of the test makes it difficult to exclude an ectopic pregnancy."
For any moderate prevalence a test with high sensitivity is suited for exclusion of a diagnosis as it reliably will detect the condition of interest. If the test is negative a high degree of confidence is justified that the condition of interest is not in fact present (i.e. high negative predictive value). Indeed this was the conclusion of the authors cited (stein et al). The issue with ultrasound in the context cited is that its poor specificity means that it will categorise as ectopic a goodly number of non-ectopic gestations - in other words a positive result is not conclusive evidence of the presence of the condition (positive predictive value).
Competing interests: No competing interests