0
Original Research: Disorders of the Pleura |

Accuracy of Fluorodeoxyglucose-PET Imaging for Differentiating Benign From Malignant Pleural EffusionsLabeled PET Imaging For Pleural Effusion Diagnosis: A Meta-analysis FREE TO VIEW

José M. Porcel, MD, FCCP; Paula Hernández, MD; Montserrat Martínez-Alonso, BSc; Silvia Bielsa, MD; Antonieta Salud, MD
Author and Funding Information

From the Pleural Diseases Unit, Departments of Internal Medicine (Drs Porcel, Hernández, and Bielsa), Biostatistics (Ms Martínez-Alonso), and Oncology-Hematology (Dr Salud), Arnau de Vilanova University Hospital, Biomedical Research Institute of Lleida, Lleida, Spain.

CORRESPONDENCE TO: José M. Porcel, MD, FCCP, Department of Internal Medicine, Arnau de Vilanova University Hospital, Avda Alcalde Rovira Roure 80, 25198 Lleida, Spain; e-mail: jporcelp@yahoo.es


FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2015;147(2):502-512. doi:10.1378/chest.14-0820
Text Size: A A A
Published online

BACKGROUND:  The role of fluorodeoxyglucose (FDG)-PET imaging for diagnosing malignant pleural effusions is not well defined. The aim of this study was to summarize the evidence for its use in ruling in or out the malignant origin of a pleural effusion or thickening.

METHODS:  A meta-analysis was conducted of diagnostic accuracy studies published in the Cochrane Library, PubMed, and Embase (inception to June 2013) without language restrictions. Two investigators selected studies that had evaluated the performance of FDG-PET imaging in patients with pleural effusions or thickening, using pleural cytopathology or histopathology as the reference standard for malignancy. Subgroup analyses were conducted according to FDG-PET imaging interpretation (qualitative or semiquantitative), PET imaging equipment (PET vs integrated PET-CT imaging), and/or target population (known lung cancer or malignant pleural mesothelioma). Study quality was assessed using Quality Assessment of Diagnostic Accuracy Studies-2. We used a bivariate random-effects model for the analysis and pooling of diagnostic performance measures across studies.

RESULTS:  Fourteen non-high risk of bias studies, comprising 407 patients with malignant and 232 with benign pleural conditions, met the inclusion criteria. Semiquantitative PET imaging readings had a significantly lower sensitivity for diagnosing malignant effusions than visual assessments (82% vs 91%; P = .026). The pooled test characteristics of integrated PET-CT imaging systems using semiquantitative interpretations for identifying malignant effusions were: sensitivity, 81%; specificity, 74%; positive likelihood ratio (LR), 3.22; negative LR, 0.26; and area under the curve, 0.838. Resultant data were heterogeneous, and spectrum bias should be considered when appraising FDG-PET imaging operating characteristics.

CONCLUSIONS:  The moderate accuracy of PET-CT imaging using semiquantitative readings precludes its routine recommendation for discriminating malignant from benign pleural effusions.

Figures in this Article

Cancer is the most frequent cause of pleural effusions subjected to thoracentesis.1 The diagnosis of malignant effusions can be challenging, since the overall sensitivity of a pleural fluid cytologic examination is, at most, 60%.1,2 In certain tumor types, such as malignant pleural mesothelioma (MPM), this percentage is much lower.1 Therefore, a pleural biopsy through a medical thoracoscopy is often necessary to reach a definitive diagnosis of malignancy. However, medical thoracoscopy is an invasive procedure and is not always available. Obviously, the ability to identify patients with a low probability of malignant effusion would reduce the number of unnecessary thoracoscopic procedures. Conversely, clinicians would also appreciate being able to identify those subjects with a sufficiently high likelihood of pleural malignancy, thereby justifying the need for thoracoscopy. Use of a noninvasive screening test to make a presumptive diagnosis of malignant effusion would help to achieve these goals.

Traditional thoracic imaging (ie, CT scanning and ultrasonography) has limitations for discriminating benign from malignant pleural conditions. Pleural nodules or thickening, being highly suggestive of malignancy, are present in fewer than one-half of cases when detected by operator-dependent ultrasound or CT scan.3

Fluorodeoxyglucose (FDG)-labeled metabolic PET imaging has been increasingly used to stage primary malignancies, but its ability to provide information about the benign or malignant nature of an effusion and/or thickening has not been adequately studied.4 This technique is based on the differential metabolism of normal and abnormal tissues, the uptake of FDG being accelerated in tumor cells. Unfortunately, metabolically active infectious or inflammatory lesions can lead to false-positive results.

The aim of this meta-analysis was to systematically review and summarize the available literature on the diagnostic accuracy of FDG-PET imaging in distinguishing benign from malignant pleural effusions or thickening to evaluate whether this metabolic imaging technique would help reduce the number of unnecessary thoracoscopies.

Search Strategy and Study Selection

The systematic review was performed according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines.5 This study was registered with the International Prospective Register of Systematic Reviews (http://www.crd.york.ac.uk/PROSPERO) as CRD42011001392 (“Accuracy of FDG-PET for diagnosing malignant pleural effusions: a systematic review and meta-analysis”). We searched the Cochrane Library, PubMed/MEDLINE, and EMBASE from inception to June 2013 to identify studies on the diagnostic accuracy of FDG-PET imaging for detecting malignant pleural effusions and/or thickening. The search strategy used “pleural,” “positron emission tomography,” and related terms, without language restrictions (Table 1). The retrieved articles’ accompanying references were also searched.

Table Graphic Jump Location
TABLE 1 ]  Bibliographic Search Strategy: Search Syntax in PUBMED

FDG = fluorodeoxyglucose.

Two researchers (J. M. P. and P. H.) independently selected eligible publications and extracted data from the included studies, with disagreements resolved by consensus. The inclusion criteria were as follows: diagnostic study design, use of FDG-PET imaging as the index test, use of pleural cytologic or histologic verification as the reference standard for malignancy, and reporting of sufficient data to construct 2 × 2 contingency tables. A study was excluded if any of the inclusion criteria were not met.

Data Extraction

The following characteristics were extracted from each selected study: year of publication; study design; country; recruitment strategy; sample size; patient demographic characteristics (age, sex); prevalence of pleural malignancy in the study population (pretest probability); causes of benign effusions; primary tumor types in malignant effusions; reference standard; use of stand-alone FDG-PET imaging or combined PET-CT imaging technologies; methods of FDG-PET image analysis (ie, qualitative or visual, semiquantitative using the standardized uptake value [SUV] or dual time point in which changes in maximum SUV [SUVmax] over time between two time-point scans are recorded); Quality Assessment of Diagnostic Accuracy Studies (QUADAS) 2 score; and data for construction of a 2 × 2 table. One author was contacted to provide data on his study published in Chinese.6

Assessment of Risk of Bias

Two researchers (J. M. P. and P. H.) independently assessed methodologic study quality by using the QUADAS-2 checklist.7 QUADAS-2 (University of Bristol; www.quadas.org) is an improved, redesigned tool from the original QUADAS, which comprises four domains: (1) patient selection, (2) index test, (3) reference standard, and (4) flow of patients through the study and timing of both the index text and reference standard (flow and timing). QUADAS-2 assesses each domain in terms of risk of bias (low, high, or unclear). The first three domains are also evaluated regarding concerns pertaining to applicability. Tabular displays helped to summarize QUADAS-2 assessments.

Data Analysis

Continuous variables were expressed as means or medians and categorical variables as frequencies or percentages. Sensitivities, specificities, likelihood ratios (LRs), and diagnostic OR summary estimates of the studies, with their corresponding 95% CIs, were calculated from the 2 × 2 table, using bivariate models with a random-effects approach. For tables in which any cell had a zero value, 0.5 was added to every cell to calculate the LRs. We present the data as forest plots and hierarchical summary receiver operating characteristic curves. The operating characteristics of FDG-PET imaging were examined according to the equipment used (PET or the hybrid PET-CT imaging) and the tracer-uptake interpretation, whether visual (ie, by a qualitative comparison with the level of tracer activity in normal surrounding tissues); semiquantitative, by the SUV; or using a dual time-point technique (where two time-point acquisition images are compared semiquantitatively). In addition, differences in FDG-PET imaging accuracy were explored across different subgroups of patients, namely, those with pleural effusions associated with a confirmed lung cancer or MPM. Calculations were performed after excluding studies having a high risk of bias. A random effects meta-regression analysis allowed for the comparison of the pooled sensitivities and specificities among different FDG-PET imaging reading methodologies and patient subgroups.

The interobserver agreement in the QUADAS-2 assessment was determined by using the unweighted Cohen κ statistic, in which results suggest “good” agreement if > 0.6 and “excellent” agreement if > 0.8. Heterogeneity between studies was addressed using a random-effect approach model with a logit link function. Publication bias was assessed using a funnel plot for positive and negative LRs. Statistical significance was assumed at the P < .05 level. Computations were performed with R-package “mada” statistical software (R-project; http://cran.r-project.org/web/packages/mada/index.html).

Characteristics of the Included Studies

A total of 619 studies were identified, of which 45 were retrieved for full-text review. Only 27 of those, published between 1997 and 2012, met the inclusion criteria.6,833Table 2 summarizes the design, participants, and findings of the initially selected studies. Subsequently, 12 studies with a high risk of bias based on the QUADAS-2 tool9,12,14,1719,2224,26,28,31 and another, which was the sole remaining study based on a dual time-point technique21 were excluded (Fig 1). Consequently, only 14 studies entered the final analysis, of which four were prospective,16,20,25,32 five were retrospective,6,15,27,29,30 and five had an unrevealed design.8,10,11,13,33 These eligible studies comprised 639 patients (66% male patients) with a median age of 62 years, of whom 407 (64%) eventually had malignant effusions (or thickening) and 232 had benign pleural conditions. Specifically, the causes of the malignant pleural effusions included 156 MPMs (98 epithelial, 20 sarcomatoid or mixed, and 38 nonspecified) and 251 pleural metastases, either from lung cancer (135 non-small cell histology, seven small-cell, and 27 nonspecified), nonspecified origin (n = 59), or miscellaneous (n = 23). No data on the precise etiology were reported in 76 cases (33%) of benign effusions or thickening, whereas the remaining ones were attributed to nonspecific pleuritis (n = 40), parapneumonic/empyema (n = 49), benign asbestosis (n = 22), fibrosis (n = 23), TB (n = 13), and benign tumors (n = 9).

Table Graphic Jump Location
TABLE 2 ]  Characteristics of the 14 Included and 13 Excluded Diagnostic Accuracy Studies Evaluating Fluorodeoxyglucose-PET Imaging in Pleural Effusions

FN = false negative; FP = false positive; MPM = malignant pleural mesothelioma; NR = not reported; SUV = standardized update value; TN = true negative; TP = true positive.

Figure Jump LinkFigure 1 –  Flowchart of search strategy and study selection. QUADAS = Quality Assessment of Diagnostic Accuracy Studies.Grahic Jump Location

The intended use of the index test was to discriminate between benign and malignant pleural effusions (or thickening) either in patients with effusions of uncertain etiology,8,10,11,16,20,32 a previously diagnosed lung cancer,6,15,27,29 or a suspected or confirmed MPM.13,25,30,33 Moreover, FDG-PET imaging metabolic activity was interpreted by a qualitative method in seven studies,8,10,11,13,15,16,32 a semiquantitative method in three studies,20,25,33 and by both in four studies.6,27,29,30 For semiquantitative assessments over a region of interest, SUVmax optimal cutoff discriminating values ranged from 2.2 to 3.5. Examinations with hybrid PET-CT imaging machines were performed in seven studies, which were the most recent (from 2010 onward).6,25,27,29,30,32,33 All studies recorded the FDG uptake in the pleural membranes, regardless of whether they were morphologically abnormal. Yet, in two studies, the tracer uptake of the fluid itself was also assessed.27,30 It should be noted, however, that the great majority of patients evaluated in the studies had pleural thickening and/or nodularity on CT scans.

Diagnostic Accuracy Indexes for Malignant Effusions

Table 3 presents the operating characteristics of FDG-PET imaging for pleural malignancy according to the PET imaging reading interpretation methodology (qualitative or semiquantitative), through the use of either a PET or an integrated PET-CT imaging system. The 11 studies interpreting FDG-PET imaging visually or qualitatively6,8,10,11,13,15,16,27,29,30,32 yielded a pooled sensitivity of 91%, a specificity of 67%, a positive LR of 2.83, a negative LR of 0.14, and an AUC of 0.893 (Fig 2). The corresponding diagnostic performance measures for the seven studies using SUV-based readings6,20,25,27,29,30,33 were as follows: sensitivity, 82%; specificity, 74%; positive LR, 3.24; negative LR, 0.25; and AUC, 0.840 (Fig 3). Overall, the pooled sensitivity of FDG-PET imaging was significantly higher by using qualitative as compared with semiquantitative assessments (P = .026). Studies that used a stand-alone PET imaging system contributed the most to the better sensitivity of qualitative readings (96%). In fact, when only integrated PET-CT imaging studies were considered, accuracy was not significantly different between qualitative and semiquantitative readings (P = .208). Specificity values did not differ among the groups listed in Table 3. Although etiological data for a significant proportion of benign pleural conditions were lacking, it should be noted that five of 13 (38.5%) tuberculous effusions and 21 of 49 (43%) parapneumonic effusions or empyemas exhibited avid FDG-PET imaging uptake.

Table Graphic Jump Location
TABLE 3 ]  Summary Measures of Diagnostic Accuracy for FDG-PET Imaging in the Identification of Malignant Pleural Effusions

AUC = area under the curve; DOR = diagnostic OR; LR = likelihood ratio. See Table 1 legend for expansion of other abbreviation.

Figure Jump LinkFigure 2 –  A, B, Forest plot of the sensitivity (A) and specificity (B) of fluorodeoxyglucose-PET imaging, using a qualitative interpretation, for the diagnosis of malignant pleural effusions. The sensitivity and specificity of individual studies are represented by a square, through which runs a horizontal line (95% CI).Grahic Jump Location
Figure Jump LinkFigure 3 –  A, B, Forest plot of the sensitivity (A) and specificity (B) of fluorodeoxyglucose-PET imaging, using a semiquantitative interpretation, for the diagnosis of malignant pleural effusions. The sensitivity and specificity of individual studies are represented by a square, through which runs a horizontal line (95% CI).Grahic Jump Location

Meta-regression analysis showed that the pleural benign-malignant discriminating capacity of FDG-PET imaging remained unchanged regardless of the target population, whether patients with lung cancer or MPM (P > .4). Figure 4 presents the diagnostic values of the studies in a hierarchical summary receiver operating characteristic graph. The more the curve approaches the left corner, the better the diagnostic accuracy, whereas those closer to the diagonal have poorer diagnostic accuracy.

Figure Jump LinkFigure 4 –  Hierarchical summary receiver-operating characteristic curve plot of visual and semiquantitative fluorodeoxyglucose-PET imaging studies for malignant pleural effusions. Each symbol represents an individual study in the meta-analysis. = visual (or qualitative) interpretation studies; * = standardized uptake value-based readings.Grahic Jump Location
Risk of Bias and Heterogeneity

Figure 5 summarizes the QUADAS-2 assessment for the initial 27 studies. The interobserver correlation for this evaluation was good (κ = 0.77). Fifteen6,8,10,11,13,15,16,20,21,25,27,29,30,32,33 of the 27 studies lacked a high risk of bias in any of the QUADAS-2 domains. Overall, the quality of the studies was modest because none of the 15 was rated as having a low risk of bias in the four domains, and only five15,16,20,25,32 met that risk characteristic in three of the domains of the quality tool.

Figure Jump LinkFigure 5 –  A, B, Graphic display for the methodologic quality of the included studies using the Quality Assessment of Diagnostic Accuracy Studies-2 criteria. F & T = flow and timing; I. = index.Grahic Jump Location

Asymmetric funnel tests for the positive and negative LRs in the studies dealing with qualitative FDG-PET imaging readings (P < .05) were consistent with publication bias. The same applied to the negative LR of the studies using semiquantitative readings (P = .03). In contrast, there was no significant publication bias for either qualitative or semiquantitative readings when the PET imaging system, whether alone or in combination with CT imaging, was taken into account (all P > .10). Finally, there was significant statistical heterogeneity for sensitivity and specificity measures in all subgroups of studies, with the exception being those that dealt with qualitative readings using a PET imaging system (SD of sensitivity, 0.062).

Clinical Application

According to the Bayesian method, estimates of the posttest probability of malignant effusion in patients who underwent an FDG-PET imaging examination are a function of disease prevalence (pretest probability). For instance, in a population with a malignancy prevalence of 35% among patients presenting with pleural exudates,1 the finding of a positive PET-CT scan using a semiquantitative interpretation (LR positive, 3.22) would increase this probability to 63.5%, whereas a negative result (LR negative, 0.26) would decrease the probability of malignant pleuritis to 12.3%. However, these figures would have changed to 68.7% (LR positive, 4.09) and 3% (LR negative, 0.06), respectively, with a PET scan interpreted with qualitative criteria.

Summary of Evidence

This meta-analysis summarizes 14 non-high risk of bias studies that assessed the diagnostic usefulness of FDG-PET imaging in patients with pleural effusions (or thickening). According to the data, in the best-case scenario, the yield of stand-alone PET imaging systems for labeling the malignant nature of pleural fluid accumulation, when qualitative criteria were applied, was sensitivity of 96%; specificity of 76%; LR positive, 4.09; LR negative, 0.06; and AUC, 0.95.

However, the stand-alone PET imaging system is rarely used today because the hybrid PET-CT imaging systems provide metabolic and anatomic information in a single session. Moreover, visual assessments are prone to subjectivity and lack reproducibility. This limitation is partially mitigated by the somewhat more consistent semiquantitative technique of SUV, which currently dominates the clinical use of hybrid imaging. In this more realistic clinical setting, the pooled sensitivity, specificity, LR positive, and LR negative of integrated PET-CT imaging systems, using a semiquantitative method of interpretation, for identifying pleural malignancy dropped to 81%, 74%, 3.22, and 0.26, respectively.

The question of why visual interpretations provide better sensitivity than SUV-based methods may be related to the many biologic and technological factors which hamper the latter.34 For example, SUV is usually calculated over a small area of the pleura (region of interest), which may not be adequate for the often scattered or diffused pleural involvement that characterizes malignancy. Conversely, the volume of interest (VOI) technique determines the volume concentration of the radiopharmaceutical in the entire tumor, including areas of different SUVmax, and then may better represent the biologic or metabolic behavior of the entire pleural lesion. Unfortunately, none of the studies included in the meta-analysis used quantitative VOI-based analyses. The rationale behind the outperformance of stand-alone PET imaging over fused PET-CT imaging machines is speculative but, other than the publication bias inherent in the older studies, might be related to the greater experience of radiologists interpreting the former at a time when PET imaging was highly restricted to reference centers.

Overall, the ability of integrated PET-CT imaging systems to separate benign from malignant pleural processes is fraught with potential errors and may be considered moderate, as reflected by an AUC of about 0.84. FDG-PET imaging higher negative than positive LR values makes it slightly more useful for ruling out malignant effusions when the test shows no abnormal tracer uptake. In this sense, FDG-PET imaging would theoretically reduce the number of unnecessary invasive procedures in patients with benign pleural effusions. On the other hand, FDG-PET imaging may guide needle or thoracoscopic biopsy to the highest area of uptake within a thickened pleura, thus minimizing sampling errors.

Strength and Limitations of the Review

The strength of this review is its robust search strategy and in-depth statistical analysis. Our calculations were restricted to those studies without a high risk of bias, which allowed a better assessment of the available literature. In fact, the inclusion of the initial 27 studies in the analysis would have resulted in superior diagnostic indices, thus pointing to a potentially misleading conclusion about the discriminative properties of FDG-PET imaging. Specifically, the pooled sensitivity and specificity of qualitative (20 studies), semiquantitative (11 studies), and dual-point (three studies) FDG-PET imaging interpretations would have reached 92% and 81%, 91% and 83%, and 98% and 90%, respectively (data not shown).

The reliability and generalization of the data were limited by the small number of eligible studies and the presence of some degree of methodologic weakness in all of them. Potential sources of bias and heterogeneity added to the shortcomings. Certainly, many studies suffered from a spectrum bias. For example, most patients with malignant effusions had pleural nodularity or thickening, with PET imaging being interpreted as positive when there was an increased uptake by the altered pleural membranes and not by the fluid itself. Nevertheless, such morphologic pleural abnormalities characterize fewer than one-half of pleural malignancies.3 Similarly, common benign conditions, like parapneumonic effusions, TB, or heart failure, were clearly underrepresented or even absent in the comparator branches.

Publication bias was demonstrated for studies that analyzed functional images by either qualitative or semiquantitative methods. The fact that reports on qualitative readings corresponded with the older studies introduces a confounding factor for interpreting their superior sensitivity as compared with those that are semiquantitative.

Potential sources of heterogeneity included study design, target population, varying cutoff values for test positivity in semiquantitative assessments, and differences in PET image acquisition and/or scanning equipment. Some of them could be controlled in the meta-regression analysis.

Comparison With Previous Research

Two meta-analyses on the diagnostic performance of pleural PET imaging for benign-malignant discriminations, from the same research group, have been published.35,36 In the first study,35 comprising 11 articles8,9,11,13,16,17,2224,30,32 and 440 patients, the pooled results for PET and PET-CT imaging, which only considered visual interpretations, were as follows: sensitivity, 95%; specificity, 82%; LR positive, 5.3; LR negative, 0.09; and AUC, 0.95. The second study36 selected eight articles,6,12,14,15,19,26,27,31 which included 360 patients with confirmed cancer (90% of lung origin) who also developed pleural abnormalities. Under the same assumptions (ie, pooled results for PET and PET-CT imaging using visual analysis), the test yielded a sensitivity of 86%, specificity of 80%, LR positive of 3.7, LR negative of 0.18, and AUC of 0.90.36 In both articles, the authors concluded that FDG-PET imaging is an accurate imaging method for the differential diagnosis of pleural effusions.35,36

These previous meta-analyses were, we feel, flawed by a number of methodologic limitations, which make their encouraging conclusions questionable. For instance, concerning the largest study35: (1) the search strategy was faulty, as we were able to find and include three additional articles for the same time period for our systematic review10,18,29; (2) the authors did not meta-analyze five studies20,21,25,28,33 because of an apparent lack of data to construct 2 × 2 tables when, in fact, these data could have been easily extracted; and (3) they excluded no studies for quality reasons. However, we felt compelled to eliminate five9,17,2224 of the 11 articles composing their meta-analysis, due to a high risk of bias. Additionally, the 2 × 2 table of one13 of the 11 studies meta-analyzed contained erroneous data, and the authors chose not to analyze studies in which semiquantitative PET imaging readings were used. Even so, one semiquantitative study was misinterpreted and included as a visual.24

The results of the present meta-analysis, based on 14 studies with > 600 patients spanning the last 16 years, suggest that, although of some value, FDG-PET imaging does not seem to change the probability of pleural malignancy sufficiently as to be recommended in the routine workup of effusions of undetermined cause. The FDG-PET imaging technique warrants broader prospective evaluations using: (1) gating acquisition techniques to reduce motion artifacts, (2) VOI analyses to enhance quantitative PET imaging measurements, and (3) an adequate benign comparator group, which should reflect the local epidemiology of pleural effusions. In addition, its role in the most challenging clinical scenario (ie, malignant effusions without pleural nodularity or thickening and/or a false-negative cytologic examination of the pleural fluid) needs to be determined in future studies.

Author contributions: J. M. P. had full access to all of the data in the study, takes responsibility for the integrity of the data and the accuracy of the data analysis, is guarantor for the entire manuscript, contributed to the study concept and design and to drafting the manuscript, and served as principal author. J. M. P. and P. H. contributed to the systematic review. M. M.-A. contributed to the statistical analysis. S. B. and A. S. contributed to revision of the manuscript. J. M. P., P. H., M. M.-A., S. B., and A. S. approved the final version of the manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Other contributions: The authors appreciate the assistance of Cristina Gámez, MD, for critical review of the manuscript.

AUC

area under the curve

FDG

fluorodeoxyglucose

LR

likelihood ratio

MPM

malignant pleural mesothelioma

QUADAS

Quality Assessment of Diagnostic Accuracy Studies

SUV

standardized uptake value

SUVmax

maximum standardized uptake value

VOI

volume of interest

Porcel JM, Esquerda A, Vives M, Bielsa S. Etiology of pleural effusions: analysis of more than 3,000 consecutive thoracenteses. Arch Bronconeumol. 2014;50(5):161-165. [CrossRef] [PubMed]
 
Porcel JM. Pearls and myths in pleural fluid analysis. Respirology. 2011;16(1):44-52. [CrossRef] [PubMed]
 
Porcel JM, Light RW. Pleural effusions. Dis Mon. 2013;59(2):29-57. [CrossRef] [PubMed]
 
Duysinx B, Corhay JL, Larock MP, et al. Contribution of positron emission tomography in pleural disease. Rev Mal Respir. 2010;27(8):e47-e53. [CrossRef] [PubMed]
 
Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264-269, W64. [CrossRef] [PubMed]
 
Liao R, Yang X, Wang S, et al. Clinical role of F-18 FDG PET/CT in differentiating malignant and benign pleural effusion in patients with lung cancer [in Chinese]. Zhongguo Fei Ai Za Zhi. 2012;15(11):652-655. [PubMed]
 
Whiting PF, Rutjes AWS, Westwood ME, et al; QUADAS-2 Group. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011;155(8):529-536. [CrossRef] [PubMed]
 
Bury T, Paulus P, Dowlati A, Corhay JL, Rigo P, Radermecker MF. Evaluation of pleural diseases with FDG-PET imaging: preliminary report. Thorax. 1997;52(2):187-189. [CrossRef] [PubMed]
 
Bénard F, Sterman D, Smith RJ, Kaiser LR, Albelda SM, Alavi A. Metabolic imaging of malignant pleural mesothelioma with fluorodeoxyglucose positron emission tomography. Chest. 1998;114(3):713-722. [CrossRef] [PubMed]
 
Buchmann I, Guhlmann CA, Elsner K, et al. F-18-FDG PET for differential diagnosis of pleural processes [in German]. Nucl Med (Stuttg). 1999;38(8):319-322.
 
Carretta A, Landoni C, Melloni G, et al. 18-FDG positron emission tomography in the evaluation of malignant pleural diseases - a pilot study. Eur J Cardiothorac Surg. 2000;17(4):377-383. [CrossRef] [PubMed]
 
Erasmus JJ, McAdams HP, Rossi SE, Goodman PC, Coleman RE, Patz EF. FDG PET of pleural effusions in patients with non-small cell lung cancer. AJR Am J Roentgenol. 2000;175(1):245-249. [CrossRef] [PubMed]
 
Gerbaudo VH, Sugarbaker DJ, Britz-Cunningham S, Di Carli MF, Mauceri C, Treves ST. Assessment of malignant pleural mesothelioma with (18)F-FDG dual-head gamma-camera coincidence imaging: comparison with histopathology. J Nucl Med. 2002;43(9):1144-1149. [PubMed]
 
Gupta NC, Rogers JS, Graeber GM, et al. Clinical role of F-18 fluorodeoxyglucose positron emission tomography imaging in patients with lung cancer and suspected malignant pleural effusion. Chest. 2002;122(6):1918-1924. [CrossRef] [PubMed]
 
Schaffler GJ, Wolf G, Schoellnast H, et al. Non-small cell lung cancer: evaluation of pleural abnormalities on CT scans with 18F FDG PET. Radiology. 2004;231(3):858-865. [CrossRef] [PubMed]
 
Duysinx B, Nguyen D, Louis R, et al. Evaluation of pleural disease with 18-fluorodeoxyglucose positron emission tomography imaging. Chest. 2004;125(2):489-493. [CrossRef] [PubMed]
 
Kramer H, Pieterman RM, Slebos DJ, et al. PET for the evaluation of pleural thickening observed on CT. J Nucl Med. 2004;45(6):995-998. [PubMed]
 
Melloni B, Monteil J, Vincent F, et al. Assessment of 18F-fluorodeoxyglucose dual-head gamma camera in asbestos lung diseases. Eur Respir J. 2004;24(5):814-821. [CrossRef] [PubMed]
 
Toaff JS, Metser U, Gottfried M, et al. Differentiation between malignant and benign pleural effusion in patients with extra-pleural primary malignancies: assessment with positron emission tomography-computed tomography. Invest Radiol. 2005;40(4):204-209. [CrossRef] [PubMed]
 
Duysinx BC, Larock MP, Nguyen D, et al. 18F-FDG PET imaging in assessing exudative pleural effusions. Nucl Med Commun. 2006;27(12):971-976. [CrossRef] [PubMed]
 
Mavi A, Basu S, Cermik TF, et al. Potential of dual time point FDG-PET imaging in differentiating malignant from benign pleural disease. Mol Imaging Biol. 2009;11(5):369-378. [CrossRef] [PubMed]
 
Yildirim H, Metintas M, Entok E, et al. Clinical value of fluorodeoxyglucose-positron emission tomography/computed tomography in differentiation of malignant mesothelioma from asbestos-related benign pleural disease: an observational pilot study. J Thorac Oncol. 2009;4(12):1480-1484. [CrossRef] [PubMed]
 
Yamamoto Y, Kameyama R, Togami T, et al. Dual time point FDG PET for evaluation of malignant pleural mesothelioma. Nucl Med Commun. 2009;30(1):25-29. [CrossRef] [PubMed]
 
Orki A, Akin O, Tasci AE, et al. The role of positron emission tomography/computed tomography in the diagnosis of pleural diseases. Thorac Cardiovasc Surg. 2009;57(4):217-221. [CrossRef] [PubMed]
 
Kurata S, Ishibashi M, Azuma K, et al. Preliminary study of positron emission tomography/computed tomography and plasma osteopontin levels in patients with asbestos-related pleural disease. Jpn J Radiol. 2010;28(6):446-452. [CrossRef] [PubMed]
 
Kim BS, Kim IJ, Kim SJ, Pak K, Kim K. Predictive value of F-18 FDG PET/CT for malignant pleural effusion in non-small cell lung cancer patients. Onkologie. 2011;34(6):298-303. [CrossRef] [PubMed]
 
Alkhawaldeh K, Biersack HJ, Henke A, Ezziddin S. Impact of dual-time-point F-18 FDG PET/CT in the assessment of pleural effusion in patients with non-small-cell lung cancer. Clin Nucl Med. 2011;36(6):423-428. [CrossRef] [PubMed]
 
Abe Y, Tamura K, Sakata I, et al. Clinical implications of 18F-fluorodeoxyglucose positron emission tomography/computed tomography at delayed phase for diagnosis and prognosis of malignant pleural mesothelioma. Oncol Rep. 2012;27(2):333-338. [PubMed]
 
Jung MY, Chong A, Seon HJ, et al. Indeterminate pleural metastasis on contrast-enhanced chest CT in non-small cell lung cancer: improved differential diagnosis with (18)F-FDG PET/CT. Ann Nucl Med. 2012;26(4):327-336. [CrossRef] [PubMed]
 
Elboga U, Yılmaz M, Uyar M, Zeki Çelen Y, Bakır K, Dikensoy O. The role of FDG PET-CT in differential diagnosis of pleural pathologies. Rev Esp Med Nucl Imagen Mol. 2012;31(4):187-191. [PubMed]
 
Letovanec I, Allenbach G, Mihaescu A, et al. 18F-fluorodeoxyglucose PET/CT findings in pleural effusions of patients with known cancer. A cytopathological correlation. Nucl Med (Stuttg). 2012;51(5):186-193.
 
Coolen J, De Keyzer F, Nafteux P, et al. Malignant pleural disease: diagnosis by using diffusion-weighted and dynamic contrast-enhanced MR imaging—initial experience. Radiology. 2012;263(3):884-892. [CrossRef] [PubMed]
 
Terada T, Tabata C, Tabata R, et al. Clinical utility of 18-fluorodeoxyglucose positron emission tomography/computed tomography in malignant pleural mesothelioma. Exp Ther Med. 2012;4(2):197-200. [PubMed]
 
Adams MC, Turkington TG, Wilson JM, Wong TZ. A systematic review of the factors affecting accuracy of SUV measurements. AJR Am J Roentgenol. 2010;195(2):310-320. [CrossRef] [PubMed]
 
Treglia G, Sadeghi R, Annunziata S, et al. Diagnostic accuracy of 18F-FDG-PET and PET/CT in the differential diagnosis between malignant and benign pleural lesions: a systematic review and meta-analysis. Acad Radiol. 2014;21(1):11-20. [CrossRef] [PubMed]
 
Treglia G, Sadeghi R, Annunziata S, et al. Diagnostic performance of fluorine-18-fluorodeoxyglucose positron emission tomography in the assessment of pleural abnormalities in cancer patients: a systematic review and a meta-analysis. Lung Cancer. 2014;83(1):1-7. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1 –  Flowchart of search strategy and study selection. QUADAS = Quality Assessment of Diagnostic Accuracy Studies.Grahic Jump Location
Figure Jump LinkFigure 2 –  A, B, Forest plot of the sensitivity (A) and specificity (B) of fluorodeoxyglucose-PET imaging, using a qualitative interpretation, for the diagnosis of malignant pleural effusions. The sensitivity and specificity of individual studies are represented by a square, through which runs a horizontal line (95% CI).Grahic Jump Location
Figure Jump LinkFigure 3 –  A, B, Forest plot of the sensitivity (A) and specificity (B) of fluorodeoxyglucose-PET imaging, using a semiquantitative interpretation, for the diagnosis of malignant pleural effusions. The sensitivity and specificity of individual studies are represented by a square, through which runs a horizontal line (95% CI).Grahic Jump Location
Figure Jump LinkFigure 4 –  Hierarchical summary receiver-operating characteristic curve plot of visual and semiquantitative fluorodeoxyglucose-PET imaging studies for malignant pleural effusions. Each symbol represents an individual study in the meta-analysis. = visual (or qualitative) interpretation studies; * = standardized uptake value-based readings.Grahic Jump Location
Figure Jump LinkFigure 5 –  A, B, Graphic display for the methodologic quality of the included studies using the Quality Assessment of Diagnostic Accuracy Studies-2 criteria. F & T = flow and timing; I. = index.Grahic Jump Location

Tables

Table Graphic Jump Location
TABLE 1 ]  Bibliographic Search Strategy: Search Syntax in PUBMED

FDG = fluorodeoxyglucose.

Table Graphic Jump Location
TABLE 2 ]  Characteristics of the 14 Included and 13 Excluded Diagnostic Accuracy Studies Evaluating Fluorodeoxyglucose-PET Imaging in Pleural Effusions

FN = false negative; FP = false positive; MPM = malignant pleural mesothelioma; NR = not reported; SUV = standardized update value; TN = true negative; TP = true positive.

Table Graphic Jump Location
TABLE 3 ]  Summary Measures of Diagnostic Accuracy for FDG-PET Imaging in the Identification of Malignant Pleural Effusions

AUC = area under the curve; DOR = diagnostic OR; LR = likelihood ratio. See Table 1 legend for expansion of other abbreviation.

References

Porcel JM, Esquerda A, Vives M, Bielsa S. Etiology of pleural effusions: analysis of more than 3,000 consecutive thoracenteses. Arch Bronconeumol. 2014;50(5):161-165. [CrossRef] [PubMed]
 
Porcel JM. Pearls and myths in pleural fluid analysis. Respirology. 2011;16(1):44-52. [CrossRef] [PubMed]
 
Porcel JM, Light RW. Pleural effusions. Dis Mon. 2013;59(2):29-57. [CrossRef] [PubMed]
 
Duysinx B, Corhay JL, Larock MP, et al. Contribution of positron emission tomography in pleural disease. Rev Mal Respir. 2010;27(8):e47-e53. [CrossRef] [PubMed]
 
Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264-269, W64. [CrossRef] [PubMed]
 
Liao R, Yang X, Wang S, et al. Clinical role of F-18 FDG PET/CT in differentiating malignant and benign pleural effusion in patients with lung cancer [in Chinese]. Zhongguo Fei Ai Za Zhi. 2012;15(11):652-655. [PubMed]
 
Whiting PF, Rutjes AWS, Westwood ME, et al; QUADAS-2 Group. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011;155(8):529-536. [CrossRef] [PubMed]
 
Bury T, Paulus P, Dowlati A, Corhay JL, Rigo P, Radermecker MF. Evaluation of pleural diseases with FDG-PET imaging: preliminary report. Thorax. 1997;52(2):187-189. [CrossRef] [PubMed]
 
Bénard F, Sterman D, Smith RJ, Kaiser LR, Albelda SM, Alavi A. Metabolic imaging of malignant pleural mesothelioma with fluorodeoxyglucose positron emission tomography. Chest. 1998;114(3):713-722. [CrossRef] [PubMed]
 
Buchmann I, Guhlmann CA, Elsner K, et al. F-18-FDG PET for differential diagnosis of pleural processes [in German]. Nucl Med (Stuttg). 1999;38(8):319-322.
 
Carretta A, Landoni C, Melloni G, et al. 18-FDG positron emission tomography in the evaluation of malignant pleural diseases - a pilot study. Eur J Cardiothorac Surg. 2000;17(4):377-383. [CrossRef] [PubMed]
 
Erasmus JJ, McAdams HP, Rossi SE, Goodman PC, Coleman RE, Patz EF. FDG PET of pleural effusions in patients with non-small cell lung cancer. AJR Am J Roentgenol. 2000;175(1):245-249. [CrossRef] [PubMed]
 
Gerbaudo VH, Sugarbaker DJ, Britz-Cunningham S, Di Carli MF, Mauceri C, Treves ST. Assessment of malignant pleural mesothelioma with (18)F-FDG dual-head gamma-camera coincidence imaging: comparison with histopathology. J Nucl Med. 2002;43(9):1144-1149. [PubMed]
 
Gupta NC, Rogers JS, Graeber GM, et al. Clinical role of F-18 fluorodeoxyglucose positron emission tomography imaging in patients with lung cancer and suspected malignant pleural effusion. Chest. 2002;122(6):1918-1924. [CrossRef] [PubMed]
 
Schaffler GJ, Wolf G, Schoellnast H, et al. Non-small cell lung cancer: evaluation of pleural abnormalities on CT scans with 18F FDG PET. Radiology. 2004;231(3):858-865. [CrossRef] [PubMed]
 
Duysinx B, Nguyen D, Louis R, et al. Evaluation of pleural disease with 18-fluorodeoxyglucose positron emission tomography imaging. Chest. 2004;125(2):489-493. [CrossRef] [PubMed]
 
Kramer H, Pieterman RM, Slebos DJ, et al. PET for the evaluation of pleural thickening observed on CT. J Nucl Med. 2004;45(6):995-998. [PubMed]
 
Melloni B, Monteil J, Vincent F, et al. Assessment of 18F-fluorodeoxyglucose dual-head gamma camera in asbestos lung diseases. Eur Respir J. 2004;24(5):814-821. [CrossRef] [PubMed]
 
Toaff JS, Metser U, Gottfried M, et al. Differentiation between malignant and benign pleural effusion in patients with extra-pleural primary malignancies: assessment with positron emission tomography-computed tomography. Invest Radiol. 2005;40(4):204-209. [CrossRef] [PubMed]
 
Duysinx BC, Larock MP, Nguyen D, et al. 18F-FDG PET imaging in assessing exudative pleural effusions. Nucl Med Commun. 2006;27(12):971-976. [CrossRef] [PubMed]
 
Mavi A, Basu S, Cermik TF, et al. Potential of dual time point FDG-PET imaging in differentiating malignant from benign pleural disease. Mol Imaging Biol. 2009;11(5):369-378. [CrossRef] [PubMed]
 
Yildirim H, Metintas M, Entok E, et al. Clinical value of fluorodeoxyglucose-positron emission tomography/computed tomography in differentiation of malignant mesothelioma from asbestos-related benign pleural disease: an observational pilot study. J Thorac Oncol. 2009;4(12):1480-1484. [CrossRef] [PubMed]
 
Yamamoto Y, Kameyama R, Togami T, et al. Dual time point FDG PET for evaluation of malignant pleural mesothelioma. Nucl Med Commun. 2009;30(1):25-29. [CrossRef] [PubMed]
 
Orki A, Akin O, Tasci AE, et al. The role of positron emission tomography/computed tomography in the diagnosis of pleural diseases. Thorac Cardiovasc Surg. 2009;57(4):217-221. [CrossRef] [PubMed]
 
Kurata S, Ishibashi M, Azuma K, et al. Preliminary study of positron emission tomography/computed tomography and plasma osteopontin levels in patients with asbestos-related pleural disease. Jpn J Radiol. 2010;28(6):446-452. [CrossRef] [PubMed]
 
Kim BS, Kim IJ, Kim SJ, Pak K, Kim K. Predictive value of F-18 FDG PET/CT for malignant pleural effusion in non-small cell lung cancer patients. Onkologie. 2011;34(6):298-303. [CrossRef] [PubMed]
 
Alkhawaldeh K, Biersack HJ, Henke A, Ezziddin S. Impact of dual-time-point F-18 FDG PET/CT in the assessment of pleural effusion in patients with non-small-cell lung cancer. Clin Nucl Med. 2011;36(6):423-428. [CrossRef] [PubMed]
 
Abe Y, Tamura K, Sakata I, et al. Clinical implications of 18F-fluorodeoxyglucose positron emission tomography/computed tomography at delayed phase for diagnosis and prognosis of malignant pleural mesothelioma. Oncol Rep. 2012;27(2):333-338. [PubMed]
 
Jung MY, Chong A, Seon HJ, et al. Indeterminate pleural metastasis on contrast-enhanced chest CT in non-small cell lung cancer: improved differential diagnosis with (18)F-FDG PET/CT. Ann Nucl Med. 2012;26(4):327-336. [CrossRef] [PubMed]
 
Elboga U, Yılmaz M, Uyar M, Zeki Çelen Y, Bakır K, Dikensoy O. The role of FDG PET-CT in differential diagnosis of pleural pathologies. Rev Esp Med Nucl Imagen Mol. 2012;31(4):187-191. [PubMed]
 
Letovanec I, Allenbach G, Mihaescu A, et al. 18F-fluorodeoxyglucose PET/CT findings in pleural effusions of patients with known cancer. A cytopathological correlation. Nucl Med (Stuttg). 2012;51(5):186-193.
 
Coolen J, De Keyzer F, Nafteux P, et al. Malignant pleural disease: diagnosis by using diffusion-weighted and dynamic contrast-enhanced MR imaging—initial experience. Radiology. 2012;263(3):884-892. [CrossRef] [PubMed]
 
Terada T, Tabata C, Tabata R, et al. Clinical utility of 18-fluorodeoxyglucose positron emission tomography/computed tomography in malignant pleural mesothelioma. Exp Ther Med. 2012;4(2):197-200. [PubMed]
 
Adams MC, Turkington TG, Wilson JM, Wong TZ. A systematic review of the factors affecting accuracy of SUV measurements. AJR Am J Roentgenol. 2010;195(2):310-320. [CrossRef] [PubMed]
 
Treglia G, Sadeghi R, Annunziata S, et al. Diagnostic accuracy of 18F-FDG-PET and PET/CT in the differential diagnosis between malignant and benign pleural lesions: a systematic review and meta-analysis. Acad Radiol. 2014;21(1):11-20. [CrossRef] [PubMed]
 
Treglia G, Sadeghi R, Annunziata S, et al. Diagnostic performance of fluorine-18-fluorodeoxyglucose positron emission tomography in the assessment of pleural abnormalities in cancer patients: a systematic review and a meta-analysis. Lung Cancer. 2014;83(1):1-7. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
Guidelines
Intermetatarsal neuroma.
Academy of Ambulatory Foot and Ankle Surgery | 1/16/2004
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543