ICVTS Click here for other ICVTS advertising opportunities
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Interact CardioVasc Thorac Surg 2009;9:406-410. doi:10.1510/icvts.2009.207621
© 2009 European Association of Cardio-Thoracic Surgery
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Michel Christodoulou
Hans-Beat Ris
Right arrow Permission Requests
Google Scholar
Right arrow Articles by Petermann, D.
Right arrow Articles by Prior, J. O.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Petermann, D.
Right arrow Articles by Prior, J. O.

Institutional report - Thoracic oncologic

Value of positron emission tomography in full-thickness chest wall resections for malignancies{star}

David Petermanna,*, Gilles Allenbachb, Sabine Schmidtc, Igor Letovanecd, Michel Christodouloua, Angelika Bischof Delaloyeb, Hans-Beat Risa and John O. Priorb,1

a Divisions of Thoracic and Vascular Surgery, Centre Hospitalier Universitaire Vaudois (CHUV) and University of Lausanne, CH-1011 Lausanne, Switzerland
b Department of Nuclear Medicine, Centre Hospitalier Universitaire Vaudois (CHUV) and University of Lausanne, Lausanne, Switzerland
c Department of Radiology, Centre Hospitalier Universitaire Vaudois (CHUV) and University of Lausanne, Lausanne, Switzerland
d Department of Pathology, Centre Hospitalier Universitaire Vaudois (CHUV) and University of Lausanne, Lausanne, Switzerland

Received 17 March 2009; received in revised form 13 May 2009; accepted 18 May 2009

1 John O. Prior is the recipient of an Academic Research Award from the Leenaards Foundation (Lausanne, Switzerland). Back

{star} Presented at the 15th European Conference on General Thoracic Surgery, Leuven, Belgium, June 3–6, 2007.

*Corresponding author. Tel.: +41 21 314 20 45 (business)/+ 41 79 613 61 29 (home); fax: +41 21 314 23 58.

E-mail address: david.petermann{at}bluewin.ch (D. Petermann).


   Abstract
Top
 Abstract
1. Introduction
 2. Materials and methods
 3. Results
 4. Discussion
 Acknowledgements
 References
 
Preoperative imaging for resection of chest wall malignancies is generally performed by computed tomography (CT). We evaluated the role of 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) in planning full-thickness chest wall resections for malignancies. We retrospectively included 18 consecutive patients operated from 2004 to 2006 at our institution. Tumor extent was measured by CT and PET, using the two largest perpendicular tumor extensions in the chest wall plane to compute the tumor surface assuming an elliptical shape. Imaging measurements were compared to histopathology assessment of tumor borders. CT assessment consistently overestimated the tumor size as compared to PET (+64% vs. +1%, P<0.001). Moreover, PET was significantly better than CT at defining the size of lesions >24 cm2 corresponding to a mean diameter >5.5 cm or an ellipse of >4 cmx7.6 cm (positive predictive value 80% vs. 44% and specificity 93% vs. 64%, respectively). Metabolic PET imaging was superior to CT for defining the extent of chest wall tumors, particularly for tumors with a diameter >5.5 cm. PET can complement CT in planning full-thickness chest wall resection for malignancies, but its true value remains to be determined in larger, prospective studies.

Key Words: Chest wall; Positron emission tomography; Computed tomography; Cancer


   1. Introduction
Top
 Abstract
 1. Introduction
2. Materials and methods
 3. Results
 4. Discussion
 Acknowledgements
 References
 
Full-thickness chest wall resection is a well-established technique that can be performed with acceptable morbidity and mortality. Careful preoperative assessment of the extent of the disease is necessary to achieve a radical resection in healthy tissues leading to optimal local control of the disease [1, 2]. The size of the chest wall defect is related to postoperative complications and morbidity due to impaired respiratory function and mechanics of the chest [3]. Therefore, the accuracy of preoperative determination of the extent of resection is crucial to achieve a complete resection without excessive morbidity. This is particularly important for large tumors that require extensive and sometimes complex resections [4].

Classically, preoperative radiological assessment of the extent of chest wall malignancies is performed with contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI) [5, 6]. Positron emission tomography (PET) and its integration with CT imaging (PET/CT) is nowadays frequently applied to the staging, re-staging and follow-up of patients with neoplasms [7]. The role of PET in planning full-thickness chest wall resection has never been analyzed. The purpose of this retrospective study was to evaluate the ability of PET in estimating tumor size of chest wall malignancies compared to classical morphological CT imaging.


   2. Materials and methods
Top
 Abstract
 1. Introduction
 2. Materials and methods
3. Results
 4. Discussion
 Acknowledgements
 References
 
2.1. Patient selection criteria

We retrospectively evaluated consecutive patients who had full-thickness chest wall resection for malignancies performed at our institution from 2004 to 2006. After review of medical records, 18 patients had preoperative CT and PET imaging and were included (11 female and 7 male patients, mean age 60±9 years, range 37–76). Resection indications were breast cancer recurrence (n=10), chest wall metastasis (n=6), primary chest wall sarcoma (n=1) and desmoid tumor (n=1). Our Institutional Ethics Committee approved this study and patient written consent was waived due to its retrospective nature.

2.2. Pathology

Directly after full-thickness chest wall resection, each surgical specimen was oriented spatially and sent to the pathology unit. Macroscopic and microscopic examinations were performed by one pathologist. Tumor localization was appreciated and the size of the tumor in the chest wall plane by approximating the shape of the tumor by an ellipse. Its two greatest perpendicular axes, defined as the a-axis and the b-axis, were measured. This 2-dimensional measurement method was chosen to reflect the area of chest wall resected in the operating room. As full-thickness resection was always performed, the depth of the tumor across the chest wall had no impact. The tumor relation to the resection margins was also examined. The tumor size derived from the surgical specimen served as the standard of reference used for comparison with tumor size obtained on PET and CT imaging.

2.3. CT

Morphological cross-sectional imaging was performed using an 8- or 16- multidetector CT unit (GE Medical Systems, Milwaukee, Wisconsin). After intravenous (IV) iodinated contrast medium injection (Accupaque®, Iohexol, Amersham Health AG, Waedenswil, Switzerland) axial slices in arterial phase were performed covering the region from the cervicothoracic inlet to the diaphragm (120 kV, 120 mAs, pitch 1.75, slice reconstruction 2.5 mm/2 mm). Two radiologists blinded to the PET and pathological findings analyzed the CT images in consensus using multiplanar image reconstruction of the axial CT slices (Advantage Workstation, GEMS, Milwaukee) using the same definition for tumor size measurement on pathology (Figs. 1 and 2). CT examinations were performed a median of 23 days before the operation [interquartile range (IQR)=25th–75th percentile range=2–41 days).


Figure 1
View larger version (49K):
[in this window]
[in a new window]

  		Fig. 1. Method used for tumor size measurement. PET/CT 3-D reconstruction of a tumor localized in the sternum and the left anterior part of the chest wall (red). The two perpendicular greatest tumor extensions (a and b) were measured similarly on CT imaging and in histopathological assessment.

 

Figure 2
View larger version (30K):
[in this window]
[in a new window]

  		Fig. 2. PET measurements in a 55-year-old woman with chest wall invasion by a breast cancer recurrence (13.5 cmx10.7 cm, surface=113 cm2).

 
2.4. PET

Eleven patients were studied using a hybrid PET/CT system without contrast enhancement (Discovery LS, GEMS, Milwaukee) and seven patients were studied using a PET only system (Advance NXi, GEMS, Milwaukee). In PET/CT studies, the unenhanced CT was only used for attenuation correction and not for CT measurements. Patients were instructed to fast >6 h before the injection of 5 MBq/kg of 18F-fluorodeoxyglucose (FDG). Two nuclear medicine physicians blinded to the CT and pathological findings analyzed PET only images in consensus with knowledge of clinical history and previous radiological imaging study results. Multiplanar image reconstructions were used for image interpretation (Advantage Workstation, GEMS, Milwaukee). On PET images, an increase in FDG uptake above a standardized uptake value (SUV) of 2.0 g/ml was used to define malignancy. The tumor size was measured on PET images by defining its two greatest perpendicular extensions in the chest wall plane, with the same method as used by the pathologist (Figs. 1 and 2). PET examinations were performed a median of 31 (IQR 13–34) days before the operation.

2.5. Surface calculation and statistical analysis

Bidimensional calculation of tumor surface was performed considering the tumor as an ellipse in the chest wall plane (surface of an ellipse S={pi}[a/2]·[b/2]={pi}/4·a·b). Statistical analyses were performed using Stata 10.0 software (College Station, Texas) using non-parametric Wilcoxon-test, Lin's concordance correlation (a measure of both precision and accuracy [8]), and robust linear regression analysis to compare the tumor size measurements of the surgical specimen with the values obtained with PET and CT. Two-tailed hypothesis testing was used. Robust multivariate analysis was performed to determine how well each imaging modality would predict the tumor surface.

To investigate the possible clinical impact of PET in the planning of full-thickness chest wall resection, we assumed that more accurate determination of the tumor extent would be significant only for larger lesions. Indeed, for small resections, a more accurate determination of tumor extent by PET would not necessarily lead to a significantly different surgical excision than the one determined by CT. On the contrary, larger lesions would incur more extended surgery likely to include chest wall reconstruction and may benefit from the increased accuracy brought by PET. The optimal threshold was defined as the tumor surface measured on PET with the highest accuracy of PET to detect histopathology tumor extent. We determined the ability of each modality to determine if the tumor surface would be above this given threshold using receiver operating characteristics (ROC) curves and areas under the curve (AUC), and computing the corresponding sensitivity, specificity and accuracy with standard formulas.


   3. Results
Top
 Abstract
 1. Introduction
 2. Materials and methods
 3. Results
4. Discussion
 Acknowledgements
 References
 
The mean size of the chest wall defect after resection (measured on the surgical specimen) was 128±94 cm2 (range 25–424 cm2). Complete resection (R0) was achieved in 15 patients (83%) and 3 patients (17%) presented microscopic infiltration (R1) of surgical margins (soft tissue infiltration in two patients and bony infiltration in one patient). The mean tumor surface measured by the pathologist on the surgical specimen was 40±67 cm2 (range 1–265 cm2), corresponding to a mean diameter of 7 cm or an ellipse of 6x8.5 cm (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1 Tumor histology and surface measured on pathology, CT and PET

 
3.1. Comparison of PET and CT to predict tumor extent

Classical morphological cross-sectional imaging significantly overestimated true tumor surface by 16±35 cm2 as compared to pathological measurements (P=0.043). On the contrary, PET measurements overestimated tumor surface by 9±24 cm2 as compared to pathological findings, but the difference was not significant (P=0.23). Fig. 3 shows the Bland–Altman plots with the corresponding limits of agreement. Using robust linear regression, CT tended to significantly overestimate the tumor size by 64% [95% confidence interval (CI) +44 to +90%] in average, while PET was almost equal to histopathological size with a +1% (95% CI –13% to +18%) overestimation in average (Fig. 4). The concordance correlation coefficients between histopathological tumor surface and CT ({rho}c=0.89, 95% CI 0.83–0.96) or PET ({rho}c=0.95, 95% CI 0.91–0.98) were excellent, with slightly higher values for PET, although the difference was not significant. In multivariate robust linear regression, PET was a better predictor of tumor extension measured on the pathological specimen (β=0.61, P<0.001) than CT (β=0.23, P=0.042), while the difference was not significant (P=0.13), however.


Figure 3
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 3. Bland–Altman plots showing the difference vs. mean for each measurement pairs. Upper panel, CT measured surface vs. pathology surface. The mean difference of +16 cm2 (dashed line) was significantly different from 0 (P=0.043). Lower panel, PET measured surface vs. pathology surface, with a mean difference of +9 cm2 (dashed line), not significantly different from 0 (P=0.23). Note the narrower 95% limits of agreement interval for PET as compared to CT (solid lines).

 

Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 4. Linear regression analysis of tumor surface on imaging vs. histopathology. CT overestimates tumor surface by +64% (slope 1.64, thick solid line), which was significantly more than when assessed by PET (+1%, slope 1.01, thin solid line) almost no different from the identity line (dotted line).

 
3.2. Clinical impact of PET

The optimal threshold with the highest accuracy of PET to detect histopathology tumor extent was found to be about 24 cm2, equivalent to a diameter of 5.5 cm or an ellipse of 4x7.6 cm. At this threshold, PET accuracy was maximized to reach 94%. The corresponding positive and negative predictive values, sensitivity and specificity, and likelihood ratios for PET and CT to predict a tumor diameter of >5.5 cm are given in Table 2. PET was significantly better than CT to determine the need for resection >5.5 cm, with a significantly better positive predictive value (80% vs. 44.4%), specificity and AUC on ROC analysis (Fig. 5).


View this table:
[in this window]
[in a new window]

  		Table 2 Ability of CT and PET for delineating lesions of >5.5 cm in diameter

 

Figure 5
View larger version (35K):
[in this window]
[in a new window]

  		Fig. 5. Receiver operating characteristics (ROC) curves for CT (shaded area) and PET to predict resections of >5.5 cm in diameter using histopathology as gold standard (dashed line=reference line). The area under the curve (AUC) was significantly higher for PET as compared to CT (P=0.026).

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Materials and methods
 3. Results
 4. Discussion
Acknowledgements
 References
 
Full-thickness resection of chest wall malignancies is currently performed in thoracic surgery units. Rewarding results can be obtained with acceptable morbidity even in the case of extended resections by use of synthetic substitutes and of pediculized myocutaneous flaps [3, 4, 9–12]. The cornerstone for long-term disease-free survival is a complete resection and preoperative planning of the extent of resection is usually performed by CT or MRI. In this retrospective study, we showed that PET-based measurements of tumoral extent better correlated with histopathological results than contrast-enhanced CT. In addition, PET allowed a significantly better prediction of histopathological results for large tumors >5.5 cm in diameter as compared to enhanced CT. Thus, PET may be helpful in complementing imaging for the planning of full-thickness chest wall resection for malignant disease.

PET/CT imaging is routinely used in modern oncological patient care, allowing a more accurate staging of tumor than morphological cross-sectional imaging. PET/CT enables in one examination to assess the local tumor extension and the detection of distant metastasis. The utility of PET has been described in evaluating chest wall involvement of lung cancer [13], but its use in planning chest wall resection has not been reported. Similarly, several studies have described the utility of PET/CT for diagnosing breast cancer recurrence. For instance, Radan et al. reported that PET/CT was superior to CT for the diagnosis of recurrence and the assessment of the extent of the disease [14]. Interestingly, Briccoli et al. have reported the superiority of ultrasonography to CT and MRI in determining soft tissue resection margins of malignant chest wall tumors [15]. In their series of 22 patients, the authors obtained 100% of complete resection and showed that ultrasonography was particularly useful in identifying micronodules, but this technique is not applicable for the examination of the chest wall per se. MRI is frequently performed for investigation of tumors, particularly for osseous tumors. In our institution, MRI is rarely performed for chest wall malignancies because of the high quality of modern multidetector CT and the use of PET for local and distant staging of tumors. The role of MRI in this setting would need to be explored prospectively.

A limitation of our study would be the variety of histological types of the chest wall tumors, which may have had very different metabolic behaviors on PET. However, this heterogeneity reflects the daily clinical care of patients with chest wall tumors undergoing resection and reconstruction. Another limitation would be the retrospective design of our study. Thus, our promising results demonstrated the value of PET imaging for planning the extent of full-thickness resection for chest wall tumors. This would need to be confirmed in a larger, prospective study, where the PET information would be systematically integrated in resection planning. Such a study would allow in addition to determine the impact of PET on R0, as well as on the long-term outcome such as overall survival and recurrence rate.

In conclusion, our initial retrospective study suggests that PET might more accurately predict the extension of chest wall tumors than CT and would have a clinical impact in predicting the extent of required resections, especially in large tumors (>5.5 cm in diameter). However, prospective studies are needed to confirm these preliminary results and to determine the impact of delineating chest wall tumors by adding PET-derived information on tumor biology.


   Acknowledgements
Top
 Abstract
 1. Introduction
 2. Materials and methods
 3. Results
 4. Discussion
 Acknowledgements
References
 
The authors are acknowledging the contribution of the physicians involved in the management of the patients, and M. Martins-Favre, MD (Department of Radiology, CHUV) for her contribution, as well as M. Faouzi, PhD (Institute of Social and Preventive Medicine, CHUV) for statistical review.


   References
Top
 Abstract
 1. Introduction
 2. Materials and methods
 3. Results
 4. Discussion
 Acknowledgements
 References
 

  1. Incarbone M, Pastorino U. Surgical treatment of chest wall tumors. World J Surg 2001;25:218–230.[CrossRef][Medline]
  2. Mansour KA, Thourani VH, Losken A, Reeves JG, Miller JI Jr, Carlson GW, Jones GE. Chest wall resections and reconstruction: a 25-year experience. Ann Thorac Surg 2002;73:1720–1725; discussion 1725–1726.[Abstract/Free Full Text]
  3. Weyant MJ, Bains MS, Venkatraman E, Downey RJ, Park BJ, Flores RM, Rizk N, Rusch VW. Results of chest wall resection and reconstruction with and without rigid prosthesis. Ann Thorac Surg 2006;81:279–285.[Abstract/Free Full Text]
  4. Novoa N, Benito P, Jimenez MF, de Juan A, Luis Aranda J, Varela G. Reconstruction of chest wall defects after resection of large neoplasms: ten-year experience. Interact CardioVasc Thorac Surg 2005;4:250–255.[Abstract/Free Full Text]
  5. Jeung MY, Gangi A, Gasser B, Vasilescu C, Massard G, Wihlm JM, Roy C. Imaging of chest wall disorders. Radiographics 1999;19:617–637.[Abstract/Free Full Text]
  6. Tateishi U, Gladish GW, Kusumoto M, Hasegawa T, Yokoyama R, Tsuchiya R, Moriyama N. Chest wall tumors: radiologic findings and pathologic correlation: part 2. Malignant tumors. Radiographics 2003;23:1491–1508.[Abstract/Free Full Text]
  7. Blodgett TM, Meltzer CC, Townsend DW. PET/CT: form and function. Radiology 2007;242:360–385.[Abstract/Free Full Text]
  8. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics 1989;45:255–268.[CrossRef][Medline]
  9. Athanassiadi K, Kalavrouziotis G, Rondogianni D, Loutsidis A, Hatzimichalis A, Bellenis I. Primary chest wall tumors: early and long-term results of surgical treatment. Eur J Cardiothorac Surg 2001;19:589–593.[Abstract/Free Full Text]
  10. Lardinois D, Muller M, Furrer M, Banic A, Gugger M, Krueger T, Ris HB. Functional assessment of chest wall integrity after methylmethacrylate reconstruction. Ann Thorac Surg 2000;69:919–923.[Abstract/Free Full Text]
  11. Veronesi G, Scanagatta P, Goldhirsch A, Rietjens M, Colleoni M, Pelosi G, Spaggiari L. Results of chest wall resection for recurrent or locally advanced breast malignancies. Breast 2007;16:297–302.[CrossRef][Medline]
  12. Warzelhan J, Stoelben E, Imdahl A, Hasse J. Results in surgery for primary and metastatic chest wall tumors. Eur J Cardiothorac Surg 2001;19:584–588.[Abstract/Free Full Text]
  13. Erasmus JJ, McAdams HP, Patz EF Jr, Goodman PC, Coleman RE. Thoracic FDG PET: state of the art. Radiographics 1998;18:5–20.[Abstract]
  14. Radan L, Ben-Haim S, Bar-Shalom R, Guralnik L, Israel O. The role of FDG-PET/CT in suspected recurrence of breast cancer. Cancer 2006;107:2545–2551.[CrossRef][Medline]
  15. Briccoli A, Galletti S, Salone M, Morganti A, Pelotti P, Rocca M. Ultrasonography is superior to computed tomography and magnetic resonance imaging in determining superficial resection margins of malignant chest wall tumors. J Ultrasound Med 2007;26:157–162.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Michel Christodoulou
Hans-Beat Ris
Right arrow Permission Requests
Google Scholar
Right arrow Articles by Petermann, D.
Right arrow Articles by Prior, J. O.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Petermann, D.
Right arrow Articles by Prior, J. O.

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