| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
University of Oklahoma College of Medicine at Tulsa, 5524 E 106th Pl, Tulsa, OK 74137-7089
(Email: jamesfleckenstein{at}cox.net).
I read with interest the article by Bryant and colleagues [1] in which the uptake of 18-F (fluorine) fluorodeoxyglucose (FDG) was measured in mediastinal nodes of lung cancer patients. However, I was confused by the reported median value for maximal standard uptake value (maxSUV) in benign nodes. The median was said to be 0, whereas the range was 0 to 18.8. It is unclear how a value of zero is obtained from the formula used by them.
The equation, maxSUV = C(microCi/mL)/[ID(microCi)/w(kg)], in which "C" is the activity at a pixel within the tissue defined by a region of interest, ID is injected dose and "w" is the body weight in kilograms. To be able to obtain a value of zero there would have to be a complete absence of radioactivity in the body, which is never the case in a positron emission tomographic (PET) scan. Thus, the value of zero for SUV is impossible.
As an example, I am now reading a case with an 8-mm 4R node that shows a maxSUV of 2.2, which is not elevated above blood pool (maxSUV = 2.2). Therefore, the node does not appear to be hypermetabolic on the PET images relative to the background activity and would most likely be benign at biopsy. The authors specifically included only cases in which the PET and computed tomographic (CT) scans showed "suspicious" nodes; therefore, one assumes all nodes had activity above background and certainly more than zero or they would not have been deemed suspicious, and they would not have been biopsied. Some were likely above the benchmark maxSUV of 2.5 and others were possibly below. Was there another criterion for determining a node as suspicious other than FDG uptake, such as size? The conclusion of the authors that a maxSUV of 5.3 is superior to a value of 2.5 is helpful information. However, as a busy reader of PET and CT scans, I am often confronted with nodes on scans showing that uptake is only mildly above the mediastinal background activity (eg, SUVs of 3 to 4); therefore, I was hoping to learn what the mean SUV (± standard deviation) was in suspicious nodes that were ultimately proven to be benign. The choice of the authors to report the median (zero) instead of the mean bypassed an opportunity to provide additional important information. In this regard, the tables and figures that include zero as a value probably need to be reviewed. The receiver operator characteristics (ROC) may also need to be reviewed. It is those nodes with borderline high maxSUVs (eg, near 2.5) that would be most interesting to study on a ROC curve.
Could the authors clarify how they arrived at the data and what the mean SUV was in suspicious nodes that were subsequently determined to be histopathologically benign? In addition, if the data are available, it would be interesting to learn what benign pathologies were found in suspicious nodes when SUVs were low (eg, 2 to 4).
 |
References |
|---|
 Top References |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  R. J. Cerfolio and A. S. Bryant Reply Ann. Thorac. Surg., November 1, 2007; 84(5): 1798 - 1798. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
