¹⁸F-fluorodeoxyglucose PET/CT findings in pleural effusions of patients with known cancer. A cytopathological correlation

Our plain-English summary of ¹⁸F-fluorodeoxyglucose PET/CT findings in pleural effusions of patients with known cancer. A cytopathological correlation, by I. Letovanec, G. Allenbach, A. Mihaescu, et al (Nuklearmedizin. Nuclear Medicine, 2012).

The study

• Authors: I. Letovanec, G. Allenbach, A. Mihaescu, M. Nicod Lalonde, S. Schmidt, R. Stupp, J.W. Fitting, A. Boubaker, H.B. Ris, J.O. Prior
• Journal: Nuklearmedizin. Nuclear Medicine
• Published: 2012;51(5):186-93
• Cited by: 8 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Aim: Pleural effusion is common in cancer patients and to determine its malignant origin is of huge clinical significance. PET/CT with ¹⁸F-FDG is of diagnostic value in staging and follow-up, but its ability to differentiate between malignant and benign effusions is not precisely known.

Patients, Methods: We examined 50 PET/CT from 47 patients (29 men, 18 women, 60 ± 16 years) with pleural effusion and known cancer (24 NSCLC, 7 lymphomas, 5 breasts, 4 GIST, 3 mesotheliomas, 2 head and neck, 2 malignant teratoma, 1 colorectal, 1 oesophageal, 1 melanoma) for FDG uptake in the effusions using SUV(max). This was correlated to cytopathology performed after a median of 21 days (interquartile range -3 to 23), which included pH, relative distribution (macrophages, neutrophils, eosinophils, basophils, lymphocytes, plasmocytes), and absolute cell count.

Results: Malignant cells were found in 17 effusions (34%) (6 NSCLC, 5 lymphomas, 2 breasts, 2 mesotheliomas, 2 malignant teratomas). SUV in malignant effusions were higher than in benign ones [3.7 (95%CI 1.8-5.6) vs. 1.7 g/ml (1.5-1.9), p = 0.001], with a correlation between malignant effusion and SUV (Spearman coefficient r = 0.50, p = 0.001), but not with other cytopathological or radiological parameters (ROC area 0.83 ± 0.06). Using a 2.2-mg/l SUV threshold, 12 PET/CT studies were positive and 38 negative with sensitivity, specificity, positive and negative predictive values of 53%, 91%, 75% and 79%, respectively. For NSCLC only (n = 24), ROC area was 0.95 ± 0.04, 7 studies were positive and 17 negative with a sensitivity, specificity, positive and negative predictive values of 83%, 89%, 71 and 94%, respectively.

Conclusion: PET/CT may help to differentiate the malignant or benign origin of a pleural effusion with a high specificity in patients with known cancer, in particular NSCLC.

In plain English

This retrospective study from Lausanne, Switzerland, reviewed 50 PET/CT scans of 47 patients with known cancer and fluid around the lungs (pleural effusion). Three patients had mesothelioma. The article does not discuss asbestos. The authors compared uptake of a radioactive sugar tracer (FDG) in the fluid with laboratory tests of fluid drawn from the chest, their reference standard. Malignant cells were found in 17 of 50 samples (34%), including 2 with mesothelioma. Average uptake was higher in malignant fluid (3.7 g/ml, 95% confidence interval 1.8 to 5.6) than in benign fluid (1.7 g/ml, 1.5 to 1.9). With a 2.2 g/ml cut-off, sensitivity was 53% (28 to 77%) and specificity 91% (76 to 98%) overall. Among 24 patients with non-small cell lung cancer, 6 with malignant fluid, sensitivity was 83% (36 to 99%) and specificity 89% (65 to 99%). No statistically significant link was found between inflammatory cells and tracer uptake. The authors concluded that uptake predicted malignant cells with high specificity and called for larger prospective studies. They noted the study was retrospective and its results not checked by biopsy.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Retrospective diagnostic study comparing PET/CT tracer uptake in pleural effusions with cytopathology of fluid drawn from the chest (thoracocentesis), which was used as the reference standard
• Who was studied: 47 patients with known cancer and pleural effusion (29 men, 18 women, mean age 60 years), with 50 PET/CT scans, at Lausanne University Hospital, Switzerland, between June 2004 and December 2006. Cancers: 24 non-small cell lung cancer (NSCLC), 7 lymphoma, 5 breast, 4 GIST, 3 mesothelioma, 2 head and neck, 2 malignant teratoma, 1 colorectal, 1 oesophageal, 1 melanoma
• Exposure: No asbestos or other exposure was studied. The factor examined was uptake of the tracer 18F-fluorodeoxyglucose (FDG) in the pleural fluid on PET/CT, measured as SUVmax, together with cell counts and pH of the fluid and CT findings
• Main results:
  • Of 50 fluid samples, 33 (66%) were benign and 17 (34%) contained malignant cells. The 17 malignant samples were 6 NSCLC, 5 lymphoma, 2 breast cancer, 2 mesothelioma and 2 malignant teratoma
  • Tracer uptake (SUV) was higher in malignant effusions (3.7 g/ml, 95% CI 1.8 to 5.6) than in benign effusions (1.7 g/ml, 95% CI 1.5 to 1.9), p = 0.001. Effusion volume estimate and CT density did not differ significantly between the two groups
  • In all patients, the presence of malignant cells correlated with SUV (Spearman rho = 0.50, p less than 0.001). No statistically significant association was found between SUV and any other cytopathological parameter, including inflammatory cells, or CT density
  • The correlation between SUV and malignant cells was stronger in NSCLC patients (rho = 0.66, p less than 0.0001) and weaker in patients with other cancers (rho = 0.41, p = 0.041)
  • All 50 scans, threshold of 2.2 g/ml or more: 9 true positive, 30 true negative, 3 false positive and 8 false negative. Sensitivity was 53% (95% CI 28 to 77%), specificity 91% (76 to 98%), positive predictive value 75% (43 to 95%) and negative predictive value 79% (63 to 90%). The main text gives the area under the ROC curve as 0.81 ± 0.07
  • NSCLC patients (n = 24, of whom 18 had benign and 6 had malignant effusions), same threshold: 5 true positive, 16 true negative, 2 false positive and 1 false negative. Sensitivity was 83% (95% CI 36 to 99%), specificity 89% (65 to 99%), positive predictive value 71% (29 to 96%) and negative predictive value 94% (71 to 99%). Area under the curve was 0.94 ± 0.05
  • Patients with cancers other than NSCLC: area under the curve 0.74 ± 0.10, sensitivity 36% (11 to 69%), specificity 93% (68 to 99%)
• The authors' conclusion: The authors concluded that tracer uptake in the pleural effusion was a significant and independent predictor of malignant cells in patients with known cancers, with high specificity, especially in NSCLC. They stated that larger prospective, preferably multicentre, studies are warranted to confirm the findings
• Limitations the authors state:
  • The study was retrospective, although the imaging readers were blinded to the cytopathology results
  • In 18 examinations the fluid was drawn before the PET/CT. This could theoretically raise tracer uptake through inflammation. The authors found no significant SUV difference in benign effusions between the two timings
  • The interval of up to 60 days between PET/CT and fluid sampling could be a limitation. The authors reported similar results when the interval was limited to 30 days
  • Cytology results were not verified by thoracoscopic biopsy. The authors cite a diagnostic yield of fluid cytology of about 65% in the literature
  • Positive and negative predictive values depend on how common malignant effusions were in this population (34%) and may not apply to other centres

Topics

The National Library of Medicine indexes this study under subjects our site covers: Pleural plaques and thickening.

Medical terms in this study, explained in our Lexicon: Pleural Effusion, Staging, Mesothelioma, Cytopathology, Macrophage.

Related studies

• Clinical implications of 18F-fluorodeoxyglucose positron emission tomography/computed tomography at delayed phase for diagnosis and prognosis of malignant pleural mesothelioma — Y. Abe, K. Tamura, et al. (2012)
• High performance of F-fluorodeoxyglucose positron emission tomography and contrast-enhanced CT in a rapid outpatient diagnostic program for patients with suspected lung cancer — P. Brocken, H.F. van der Heijden, et al. (2014)
• 18-Fluorodeoxyglucose positron emission tomography as a tool for response prediction in solid tumours — D.J. De Maeseneer, B. Lambert, et al. (2010)
• The association of 18F-FDG PET/CT parameters with survival in malignant pleural mesothelioma — A. Klabatsa, S. Chicklore, et al. (2014)
• Prognostic value of 18F-FDG standard uptake value by integrated PET/CT in the staging of malignant pleural mesothelioma — G. Genestreti, A. Moretti, et al. (2012)
• Pseudomesotheliomatous lung cancer mimicking mesothelioma on ¹⁸F-FDG PET/CT images: report of 2 cases — T. Nakamori, S. Kosuda, et al. (2013)

Browse return to medical research.