This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lecouvet, F. E. Articles by Janne d'Othée, B. Search for Related Content PubMed Articles by Lecouvet, F. E. Articles by Janne d'Othée, B. Related Articles Related Correspondence Social Bookmarking                            What's this?

In Reply

Department of Radiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium

François Jamar

Department of Nuclear Medicine, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium

Bertrand Tombal

Department of Urology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium

Jacques Jamart

Center of Biostatistics, Mont Godinne University Hospital, Yvoir, Belgium

Bertrand Janne d'Othée

Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA

We appreciate the interest of Drs Gemmel and De Geeter in our recent paper.¹ Their comments deserve a point-by-point response.

First, we agree with Gemmel and De Geeter that single-photon emission computed tomography (SPECT) in addition to whole-body mode might increase the sensitivity and specificity of bone scanning for detecting bone metastases. However, our study, as stated in our paper, evaluated the potential use as magnetic resonance imaging (MRI) as a substitute to the standard imaging work-up— performed every day in many institutions—rather than the effectiveness of different modalities of bone scanning by comparison with MRI. Whole-body SPECT was not widely available at the time of study initiation and was therefore not evaluated. Moreover, we are not aware of any prospective study comparing SPECT and MRI (both from head to pelvis) in such a population, and we believe that such a study would be interesting. Anyway, presenting SPECT as "the optimal state-of-the-art technique" is by far excessive.

Second, we are disappointed by the lack of understanding and erroneous statements of Gemmel and De Greeter with regard to our statistical analysis. Here is the information they would have received if their questions had been asked to a statistician.

Of course, during our data analysis, we tested all hypotheses, from the most optimistic –consisting in categorizing equivocal readings as positive in presence of metastasis and negative in the absence of metastasis –to the least optimistic –consisting in categorizing equivocal readings as negative in presence of metastasis and positive in the absence of metastases. This latter approach was kept in the paper not only to fit to the word count limit. Most importantly, this approach that always considers equivocal findings as "wrong" is the most conservative estimation of sensitivity and specificity. Sensitivity is the capacity of a test to discover truly diseased patients, with respect to certainly nondiseased subjects and uncertain results, and specificity is the capacity to exclude the studied disease, with respect to certainly diseased patients and uncertain results. Both nonpositive and non-negative tests have to be considered as failures for both estimations and thus decrease the sensitivity and the specificity. In an often referenced paper on the evaluation of diagnostic tests with imperfect standards, Valenstein² wrote that "when the sensitivity of a test is being evaluated, the effects of misclassification can be reduced by using a particularly rigorous definition of disease, when the specificity is being evaluated a less rigorous definition of disease is more appropriate." Valenstein thus recommends considering uncertain results in a different way for estimating both parameters. This approach is neither "unusual" nor "inappropriate statistical treatment of equivocal results" as Gemmel and De Geeter claim.

Regarding the third remark, it comes to the difference between "suggestive" and "true", and to what extent the clinician still agrees to make important therapeutic decisions on the basis of "suggestions." We do agree that a positive bone scintigram with negative radiographs is highly suggestive of malignant disease, but this is not absolute certainty. Bone scintigram only provides certainty in cases of multiple foci, not in case of an isolated focus, even if radiographs are negative. The only condition to reach certainty then is the availability of a previous bone scan that showed no lesion at this level. But this availability of previous bone scan increases costs, irradiation, and, by definition, is not applicable to the initial work-up. This strengthens the role of MRI in this setting.

Finally, our conclusions are not distorted by the lack of comparison between MRI and whole-body SPECT that, as of today, cannot be considered unanimously as a best comparator. We also confirm that our statistical treatment of data was appropriate.

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

ACKNOWLEDGMENTS

Supported in part by an educational grant from the nonprofit organization Fondation Saint-Luc.

REFERENCES

1. Lecouvet FE, Geukens D, Stainier A, et al: Magnetic resonance imaging of the axial skeleton for detecting bone metastases in patients with high-risk prostate cancer: Diagnostic and cost-effectiveness and comparison with current detection strategies. J Clin Oncol 25:3281-3287, 2007[Abstract/Free Full Text]

2. Valenstein PN: Evaluating diagnostic tests with imperfect standards. Am J Clin Pathol 93:252-258, 1990[Medline]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

Related Correspondence

• Magnetic Resonance Imaging Versus Bone Scan in High-Risk Prostatic Carcinoma: Some Methodological Considerations
  Filip F.A.Y. Gemmel and Frank W. De Geeter
  JCO 2008 26: 1189-1190 [Full Text]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lecouvet, F. E. Articles by Janne d'Othée, B. Search for Related Content PubMed Articles by Lecouvet, F. E. Articles by Janne d'Othée, B. Related Articles Related Correspondence Social Bookmarking                            What's this?