Skip Navigation LinksHome > May 2010 - Volume 5 - Issue 5 > A Systematic Review but Systematically Confounded?
Journal of Thoracic Oncology:
doi: 10.1097/JTO.0b013e3181d3524d
Letters to the Editor

A Systematic Review but Systematically Confounded?

Detterbeck, Frank C. MD

Free Access
Article Outline
Collapse Box

Author Information

Section of Thoracic Surgery, School of Medicine, Yale University, New Haven, CT

Disclosure: The author declares no conflicts of interest.

Address for correspondence: Frank C. Detterbeck, MD, Section of Thoracic Surgery, Yale University School of medicine, PO Box 208062, New Haven, CT 06520-8062. E-mail:

Back to Top | Article Outline

To the Editor:

The study by Nair et al1 on the prognostic value of positron emission tomography (PET) intensity in stage I non-small cell lung cancer (NSCLC) is a superbly conducted systematic review, as we have seen consistently from the senior author. However, despite the scientific rigor of the search and the analysis, I worry that the conclusion is flawed due to systematic confounding, emanating from technical issues related to PET imaging that are frequently overlooked.

The intensity of PET images correlates with the amount of fluorodeoxyglucose (FDG) in lesions over 2.5 cm. Below this diameter, lesion size markedly affects the intensity, as demonstrated by phantom studies, in which a constant amount of FDG activity seem less and less intense when placed in a smaller diameter sphere (Figure 1). This means that PET intensity is progressively diminished for tumors less than 2.5 cm, regardless of the metabolic activity or accumulation of FDG in the lesion. This has been confirmed both in patient studies2 and in other investigations of the physics of PET imaging.3 This also underscores a misconception about the term “resolution” of a PET scanner, which is a technical term and not the lesion size that can be accurately imaged. For example, at 1.5 times the resolution, approximately 60% of the true activity is measured, and measurement of 95% of the true activity requires a lesion diameter four times the resolution of the PET scanner.3 The reasons for this are multiple and complex, and partial volume averaging is only part of the problem.

Figure 1
Figure 1
Image Tools

Because the relationship of detected PET intensity and lesion size affects all studies, it results in systematic confounding (particularly for a study of stage I NSCLC!) that no amount of meta-analysis can overcome unless attention is paid to the size of each tumor studied. It is unfortunate that the study by Nair did not list at least the average size or range of the tumors in each study. Perhaps the authors have other information that mitigates my concerns, but there is no mention of possible confounding by size in the discussion.

There is likely also further systematic confounding due to inclusion of both solid and ground glass opacities lesions among the stage I tumors studied. NSCLC that appears as a ground glass opacities usually has a more indolent course,4 and these tumors have much lower PET intensity as well.2

Of course, one can take a global view that an association exists between PET intensity and prognosis, even if the data is confounded. However, application of this observation becomes quite shaky if we are unsure what really is the critical factor. Note that multivariate analyses that treat PET intensity and T stage as independent variables are flawed because of the strong correlation of size and PET intensity. To make an analogy apropos to using poorly understood associations, it would be wrong to mandate a change of music because of an association between listening to hip-hop music and the rate of car accidents, although it is probably true that more teen drivers listen to hip-hop.

It is noteworthy that the only study (prospective, multicenter) that accounted for tumor size could not demonstrate prognostic value to PET intensity over that of stage alone.5 PET intensity is complex, significantly dependent on many factors (Table 1), and requires further study and a sophisticated approach if we are to appropriately base clinical decisions on it.

Table 1
Table 1
Image Tools

Frank C. Detterbeck, MD

Section of Thoracic Surgery

School of Medicine

Yale University

New Haven, CT

Back to Top | Article Outline


1.Nair V, Krupitskaya Y, Gould M. Positron emission tomography 18F-fluorodeoxyglucose uptake and prognosis in patients with surgically treated, stage I non-small cell lung cancer: a systematic review. J Thorac Oncol 2009;4:1473–1479.

2.Detterbeck F, Khandani AH. The role of PET imaging in solitary pulmonary nodules. Clin Pulm Med 2009;16:81–88.

3.Geworski L, Knoop B, Levi de Cabrejas M, et al. Recovery correction for quantitation in emission tomography: a feasibility study. Eur J Nucl Med 2000;27:161–169.

4.Detterbeck F, Gibson C. Turning gray: the natural history of lung cancer over time. J Thorac Oncol 2008;3:781–792.

5.Vesselle H, Freeman JD, Wiens L, et al. Fluorodeoxyglucose uptake of primary non-small cell lung cancer at positron emission tomography: new contrary data on prognostic role. Clin Cancer Res 2007;13:3255–3263.

6.Hubert V, Schmidt RA, Pugsley JM, et al. Lung cancer proliferation correlates with [F-18] fluorodeoxyglucose uptake by positron emission tomography. Clin Cancer Res 2000;3837–3844.

7.Berghmans T, Dusart M, Paesmans M, et al. Primary tumor standardized uptake value (SUVmax) measured on fluorodeoxyglucose positron emission tomography (FDG-PET) is of prognostic value for survival in non-small cell lung cancer (NSCLC): a systematic review and meta-analysis (MA) by the European Lung Cancer Working Party for the IASLC Lung Cancer Staging Project. J Thorac Oncol 2008;3:6–12.

© 2010International Association for the Study of Lung Cancer


Article Tools



Other Ways to Connect



Visit on your smartphone. Scan this code (QR reader app required) with your phone and be taken directly to the site.

 For additional oncology content, visit LWW Oncology Journals on Facebook.