HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lee, C.-H. Articles by Lin, Y.-M. Search for Related Content PubMed PubMed Citation Articles by Lee, C.-H. Articles by Lin, Y.-M. Social Bookmarking                      What's this?

Specificity Gains at the Expense of Sensitivity and Vice Versa

Taichung Veterans General Hospital Taichung, Taiwan

 
WEB—This is a Web exclusive article.

We read with interest the well-written article in the October 2006 issue of AJR titled "Use of the `Two-Slice-Touch' Rule for the MRI Diagnosis of Meniscal Tears" by Drs. De Smet and Tuite [1]. With knee arthroscopy results as the gold standard, the authors reported the positive predictive value increased from 91% to 94% (p = 0.37) for medial meniscus tears and from 83% to 96% (p = 0.02) for lateral meniscus tears when using the two-slice-touch rule compared with the standard criteria of diagnosing meniscus tears if one or more images are abnormal on MR findings. However, in Table 2 of their article, we are concerned that, although the authors calculate the accuracy parameters (sensitivity, specificity, and negative predictive value) for the two-slice-touch rule, they have deliberately excluded the number of possibly torn menisci from the denominator.

Theoretically, the number of possibly torn menisci should be added back in the denominator and treated as not torn when performing the calculation. We wondered why the authors excluded them from the denominator. This manipulation produced strange results in that the specificity increased without changes in the sensitivity. We calculated the accuracy parameters for the two-slice-touch rule again and found the following: For the medial meniscus, the correct sensitivity, specificity, and negative predictive value should be 92.5% (98/106), 91.2% (62/68), and 88.6% (62/70), respectively. The p values of differences for sensitivity, specificity, and negative predictive value should be 0.39, 0.29, and 0.49, respectively. For the lateral meniscus, the correct sensitivity, specificity, and negative predictive value should be 74.6% (53/71), 98.1% (101/103), and 84.9% (101/119), respectively. The p values of differences for sensitivity, specificity, and negative predictive value should be 0.69, 0.02, and 1.0, respectively.

We also appreciate the high positive predictive value for the diagnosis of meniscal tears using the two-slice-touch rule. However, there are no perfect tests in the world because the cut point shifts and the specificity increases as the sensitivity decreases or vice versa [2].

References

Top References

 

1. De Smet AA, Tuite MJ. Use of the "two-slice-touch" rule for the MRI diagnosis of meniscal tears. AJR2006; 187:911 –914[Abstract/Free Full Text]
2. Obuchowski NA. Receiver operating characteristic curves and their use in radiology. Radiology 2003;229 : 3–8[Abstract/Free Full Text]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				A. De Smet and M. Tuite Reply Am. J. Roentgenol., June 1, 2007; 188(6): W575 - W575. [Full Text] [PDF]

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lee, C.-H. Articles by Lin, Y.-M. Search for Related Content PubMed PubMed Citation Articles by Lee, C.-H. Articles by Lin, Y.-M. Social Bookmarking                      What's this?