Radiol Oncol 2003; 37(1): 13-6.
The accuracy of chest sonography in the diagnosis of small pleural effusion
Igor Kocijanèiè
Department of Radiology, Institute of Oncology, Ljubljana, Slovenia
Background. The aim of the study was to evaluate the accuracy of chest sonography in the radiological di-agnosis of small pleural effusions.
Patients and methods. Patients referred for abdominal and/or chest sonographies for various reasons were examined for sonographic features of pleural effusion. From January 1997 till January 2000, 69 pa-tients were included into the study. Fifty-two patients were found to have pleural effusion not exceeding 15 mm in depth, the rest of them served as controls. Subsequently erect posteroanterior and expiratory lateral decubitus projections were done in all patients.
Results. Compared to radiological examination chest sonography had a positive predictive value of 92% in the diagnosis of small pleural effusions in our study population. The mean thickness of fluid was 9.2 mm on ultrasonography and 7.6 mm on expiratory lateral decubitus views (P<0.01).
Conclusions. Chest sonography showed a high degree of accuracy for demonstrating small pleural effusions and could replace lateral decubitus chest radiographs adequately.
Key words: pleural effusion-ultrasonography; thoracic radiography
Introduction
A small amount of fluid (5-10 ml) is often present in the pleural space of healthy indi-viduals.1 Small pleural effusions are not read-ily identified on conventional radiographic views of the chest.2 Lateral decubitus radi-ographs or chest ultrasonography proved to
Received 23 January 2003 Accepted 7 February 2003
Correspondence to: Igor Kocijanèiè, M.D., M.Sc., Department of Radiology, Institute of Oncology, Zaloška 2, SI - 1000 Ljubljana, Slovenia; Phone: + 386 1 522 39 81; Fax: + 386 1 43 14 180; E-mail: ikocijancic @onko-i.si
be more efficient methods for demonstrating small amounts of free pleural fluid.3-6 The data on the smallest amount of pleural fluid de-tectable vary considerably, but they are es-sentially within the same broad range whether computed tomography, sonography or X-ray examination are used.1,3,6-12
Rigler used lateral decubitus chest radi-ographs for the detection of small pleural ef-fusions.13 Other investigators3,14 have devel-oped the technique and using cadaveric stud-ies15 have shown that volumes of pleural flu-id as little as 5 ml may be detected. Recent reports have proved that minute pleural effu-sions can be detected using chest ultrasonog-raphy.6,7,17 No formal comparison has been
14
Kocijanèiè I / Diagnosing small pleural effusions
made between the thickness of the pleural ef-fusion as seen on sonography with X-ray and the amount of aspirated fluid.
We have compared sonographically de-tected small pleural effusions with expiratory lateral decubitus radiographs.
Patients and methods
Patients referred for abdominal sonography for a variety of clinical conditions were also examined for unsuspected pleural effusion. Small control group was made up of 17 pa-tients, examined only for clinically suspected pleural effusion, which was not sonographi-cally confirmed.
Between January 1997 and January 2000, 69 patients (51 males, 18 females, 28-80 years old, with the mean age of 57.1 years) were in-cluded into the study. Their condition was clinically diagnosed as lung cancer in 30, car-diac failure in 13, metastasis to the lung in 11, pneumonia and pulmonary tuberculosis in 6 and liver cirrhosis in 3 cases.
Following abdominal sonography, the patient was positioned in the lateral decubitus position for 5 minutes; sonography of the lower pleural space was performed with the patient leaning on the elbow.17 During the ex-amination maximal fluid thickness was meas-ured, with the position of the probe perpen-dicular to the thoracic wall.18
A Toshiba SSA-340A ultrasound unit was used with a 3.7 or 6 MHz convex transducer.
Radiological examination followed if sonography showed a small pleural effusion. A 140 kV Siemens unit was used, with a 2 m film-focus distance for the erect views of the chest, and 1.5 m film-focus distance for lateral decubitus views. For these, the patient was put into lateral decubitus position with 100 hip elevations, for 5 minutes prior to exposure. Exposures were taken in expiration, with the central beam aimed at the lateral chest wall and the patient slightly rotated on-Radiol Oncol 2003; 37(1): 13-6.
to the back. The films were evaluated inde-pendently by two experienced radiologists with no knowledge of the sonographic find-ings.
On sonography, the criteria for determin-ing the presence of pleural fluid were: a non/hypo - echogenic zone between the pari-etal and the visceral pleura and/or changing between expiration and inspiration as well as changing with different positions of the patient or fluttering of the pulmonary edge dur-ing respiration.6,9,19,20
On x - ray, the criteria were as follows: minimum 3 mm thick density with horizontal level on lateral decubitus view and costop-hrenic angle density with meniscus sign on
erect views.3,21
Matching pair’s t-test was used for analysis of differences between measurements of the fluid layer thickness on chest sonography and expiratory lateral decubitus projections.
The study was approved by relevant ethic committee.
Results
On erect posteroanterior chest radiographs pleural fluid was demonstrated in only 17 of 52 (33%) patients.
Lateral decubitus views were positive in 48 of 52 patients (ie, a positive predictive value of 92%) with sonographically visible fluid. In two cases pleural effusions detected sono-graphically were confirmed by thoracocente-sis. In one patient sonographically positive re-sult was not confirmed either way. In the last case radiography revealed diagnostic error occurred on sonography (Figure 1). The range of fluid thickness was 3-6 mm in these three patients.
In a small control group of 17 patients pleural fluid was not confirmed sonographi-cally nor radiographically.
The mean thickness of fluid was 9.2 mm (SD= +/- 3.3 mm) on sonography and 7.6 mm
Kocijanèiè I / Diagnosing small pleural effusions
15
(SD= +/- 4.0 mm) on expiratory lateral decu-bitus views (P<0.01). The ranges of fluid thickness on gray-scale sonography and lateral decubitus radiography were 3-15 mm and 3-11 mm, respectively (Figures 1a, 1b).
Figure 1a. A 6-mm-thick hypoechogenic zone (calipers) between the parietal and the visceral pleura suggestive of a small pleural effusion.
Figure 1b. Left lateral decubitus radiograph clearly shows a flat pleural thickness with calcified plaque on the visceral pleura.
Discussion
In the literature we could not find any exact definition of small pleural effusions. So, our term of small pleural effusions includes clini-cally silent effusions, which are usually unex-pected finding on x-ray or sonographic exam-inations undertaken for other reasons.
Rigler13 was the first to use lateral decubi-tus views for pleural fluid demonstration. He
did not use exposure in expiration, however, nor did he expose with central beam aimed at the lateral chest wall, parallel to the expected fluid level. The latter technical improvement was introduced by Hessen3 together with the elevation of the patient’s hip, while obtaining radiograph during expiration was tested in the work of Kocijanèiè et al.17 The amounts of pleural fluid detectable this way have been assessed in cadaveric experiments15 and has been shown to be as little as 5 ml in experi-mental conditions. This is probably less reli-able in practice, because the fluid may not al-ways be completely aspirated with thoraco-centesis.
Figure 2a. Sonograms show a thin fluid layer (6 mm) visible during inspiration (left image, calipers). Pleural effusion became much more apparent during expira-tion and allowed the reliable diagnosis (right image, calipers).
Figure 2b. Lateral expiratory decubitus radiograph, clearly showing a horizontal fluid layer of approxi-mately the same thickness in a 52-year-old male patient with obstructive pneumonia of the right upper lobe due to lung cancer.
Radiol Oncol 2003; 37(1): 13-6.
16
Kocijanèiè I / Diagnosing small pleural effusions
With the advent of sonography it was shown that very small amounts of pleural flu-id can be demonstrated this way.4-8 However no one precisely determined the sonographic criteria that should be fulfilled for reliable di-agnosis of small pleural effusions. In our study population all effusions were ane-chogenic, the only case with hypoechogenic »fluid« turned out to be pleural thickness. Interestingly, the main sign, allowing the demonstration of the smallest effusions on sonography as well as on radiography,17 was changing of the fluid layer during inspiration - expiration (Figures 2a, 2b).
In the course of our study searching for small pleural effusions of about 200 ml or less,12,18 we have achieved comparable results using sonography and radiography, but sonography appears to assess the thickness of fluid layer more accurately.
References
1.   Felson B. Chest rentgenology. Philadelphia: W.B. Saunders, 1973. p. 352.
2.   Collins JD, Burwell D, Furmanski S, Lorber P, Steckel R. Minimal detectable pleural effusions. A roentgen pathology model. Radiology 1972; 105: 51-3.
3.   Hessen I. Roentgen examination of pleural fluid. A study of the localisation of free effusions, the potentialities of diagnosing minimal quantities of fluid and its existence under physiological condi-tions. Acta Radiol 1951; 86(Suppl 1): 1-80.
4.   Gryminski J, Krakowa P, Lypacewicz G. The diag-nosis of pleural effusion by ultrasonic and radio-logic techniques. Chest 1976; 70: 33-7.
5.   Lipscomb DJ, Flower CDR. Ultrasound in the di-agnosis and management of pleural disease. Br J Dis Chest 1980; 74: 353-61.
6.   Mathis G. Thoraxsonography - part I.: chest wall and pleura. Ultrasound Med Biol 1997; 23: 1131-9.
7.   Eibenberger KL, Dock WI, Ammann ME, Dorffner R, Hörmann MF, Grabenwörger F. Quantification of pleural effussions: sonography versus radiogra-phy. Radiology 1994; 191: 681-4.
8.   Lorenz J, Börner N, Nikolaus HP. Sonographische volumetrie von pleuraergüssen. Ultraschall 1988; 9: 212-5.
9.   Mc Loud TC, Flower CDR. Imaging of the pleura: Sonography, CT and MR imaging. Am J Roentgenol 1991; 156: 1145-53.
10. Mikloweit P, Zachgo W, Lörcher U, Meier-Sydow J. Pleuranage lungenprozesse: diagnostische wertigkeit sonographie versus CT. Bildgebung 1991; 58: 127-31.
11. Leung AN, Muller NL, Miller RR. CT in the differential diagnosis of pleural disease. Am J Roentgenol 1990; 154: 487-92.
12. Maffesanti M, Tommasi M, Pellegrini P. Computed tomography of free pleural effusions. Europ J Radiol 1987; 7: 87-90.
13. Rigler LG. Roentgen diagnosis of small pleural ef-fusions. JAMA 1931; 96: 104-108.
14. Müller R, Löfstedt S. Reaction of pleura in primary tuberculosis of the lungs. Acta Med Scand 1945; 122: 105-33.
15. Moskowitz H, Platt RT, Schachar R, Mellus H. Roentgen visualization of minute pleural effusion. Radiology 1973; 109: 33-5.
16. Reuß J. Sonographic imaging of the pleura: nearly 30 years experience. Eur J Ultrasound 1996; 3: 125-39.
17. Kocijanèiè I, Terèelj M, Vidmar K, Jereb M. The value of inspiratory-expiratory lateral decubitus views in the diagnosis of small pleural effusions. Clin Radiol 1999; 54: 595-7.
18. Eibenberger KL, Dock W, Metz V, Weinstabl C, Haslinger B. Grabenwöger F. Wertigkeit der thorax bettaufnahmen zur diagnostik und quantifizierung von pleuraergüssen - überprüfung mittels sonographie. RöFo 1991; 155: 323-6.
19. Targhetta R, Bourgeois JM, Marty-Double C, Chavagneux R, Proust A, Coste E, et al. Vers une autre approche du diagnostic des masses pul-monaires p�riph�riques. [Towards another diag-nostic approach of peripheral pulmonary masses. Ultrasound-guided puncture] J Radiol 1992; 73: 159-64.
20. Marks WM, Filly RA, Callen PW. Real - time evaluation of pleural lesions: New observations re-garding the probability of obtaining free fluid. Radiology 1982; 142: 163-4.
21. Raasch BN, Carsky EW, Lane EJ, O’Callaghan JP, Heitzman ER. Pleural effusion: explanation of some typical appearances. AJR Am J Roentgenol 1982; 139: 889-904.
Radiol Oncol 2003; 37(1): 13-6.