Original Investigations

CT Volumetric Analysis of
Pleural Effusions:
A Comparison with Thoracentesis Volumes

David Chiao, MD, MPH, Michael Hanley, MD, Juan M. Olazagasti, MD

Rationale and Objectives: The primary objective of this study was to compare computed tomography (CT) volumetric analysis of pleural
effusions with thoracentesis volumes. The secondary objective of this study was to compare subjective grading of pleural effusion size with
thoracentesis volumes.
Materials and Methods: This was a retrospective study of 67 patients with free-flowing pleural effusions who underwent therapeutic thor-
acentesis. CT volumetric analysis was performed on all patients; the CT volumes were compared with the thoracentesis volumes. In addi-
tion, the subjective grading of pleural effusion size was compared with the thoracentesis volumes.
Results: The average difference between CT volume and thoracentesis volume was 9.4 mL (1.3%)
290 mL (30%); these volumes were
not statistically different (P = .79, paired two-tailed Student’s t-test). The thoracentesis volume of a ‘‘small,’’ ‘‘moderate,’’ and ‘‘large’’
pleural effusion, as graded on chest CT, was found to be approximately 410
260 cc, 770
270 mL and 1370
650 mL, respectively;
the thoracentesis volume of a ‘‘small,’’ ‘‘moderate,’’ and ‘‘large’’ pleural effusion, as graded on chest radiograph, was found to be approx-
imately 610
320 mL, 1040
460 mL, and 1530
830 mL, respectively.
Conclusions: CT volumetric analysis is an accessible tool that can be used to accurately quantify the size of pleural effusions.
Key Words: Pleural effusion; volumetric analysis; quantitative; thoracentesis; computed tomography.
ªAUR, 2015

A
pleural effusion is a pathologic accumulation of fluid
in the pleural space. It is a common finding, which
can occur in response to a variety of insults, including
infection, malignancy, pulmonary embolism, connective tis-
sue disease, congestive heart failure, cirrhosis, and trauma
(1). Although the presence of a pleural effusion is nonspecific,
associated findings such as cardiomegaly, consolidation, and
lung masses often suggest its etiology (2–4).
Pleural effusions can be diagnosed by routine chest radiog-
raphy, ultrasonography, computed tomography, and magnetic
resonance imaging (4,5). Even in cases where pleural effusions
are diagnosed on physical examination alone, imaging is often
warranted to further characterize the pleural effusion and/or
estimate its size. Although ultrasound is well-suited for char-
acterizing pleural effusions, computed tomography (CT) is
superior in evaluating the size of pleural effusions (6–8). In
addition, CT can be used to determine the attenuation of a
pleural effusion and to visualize reactive pleural thickening
Acad Radiol 2015; -:1–6
From the Department of Radiology and Medical Imaging, University of Virginia,
1215 Lee St, PO Box 800170, Charlottesville, VA 22908-0170. Received
October 20, 2014; accepted March 18, 2015. Conflicts of Interest and
Sources of Funding: None. Address correspondence to: D.C. e-mail:
dsc5z@virginia.edu
ªAUR, 2015
http://dx.doi.org/10.1016/j.acra.2015.03.015
or enhancement, which can provide clues to the nature of a
pleural effusion (4).
Pleural effusions can cause symptoms such as chest pain,
dyspnea, cough, exercise intolerance, and poor sleep quality
(9–12). Diagnostic thoracentesis is indicated in patients with
a clinically significant pleural effusion (more than 10-mm
thick on ultrasonography or lateral decubitus radiography)
with no known cause; therapeutic thoracentesis is indicated
in patients with shortness of breath at rest (5). The benefits
of therapeutic thoracentesis have been widely reported,
with clinical improvement in dyspnea, exercise intolerance,
and sleep quality (9–12). Improvements in oxygenation (as
measured by the ratio of partial pressure of oxygen in
systemic arterial blood to inspired fraction of oxygen,
PaO2:FIO2), lung compliance, and end-expiratory volume
have also been reported (13–15). Maximum clinical benefits
of therapeutic thoracentesis are often delayed (>24 hours),
after periprocedural pain and coughing have subsided (9).
Although there are no current recommendations on per-
forming therapeutic thoracentesis based on pleural effusion
size alone, pleural effusion size may be important in future
clinical guidelines. A variety of methods for determining the
size of pleural effusions have been published in the literature.
Early inquiries based on chest radiographs were pursued in the
1980s and 1990s, with the development of chest radiography
prediction rules (16,17). Later endeavors were made with

1
CHIAO ET AL
TABLE 1. Subject Enrollment: Inclusion and Exclusion Criteria
CT, computed tomography.
Figure 1. (a) Pleural effusion tracing in axial, coronal, and sagittal planes. (b) Three-dimensional volumetric reconstruction of the pleural effu-
sion. This provides both a quantitative estimate of the size of the pleural effusion and a visual depiction. RA, right-anterior.
chest CT; for example, Mergo et al. suggested measuring the
greatest depth (d) and length (l) of a pleural effusion,
estimating the volume as: V ¼ d  l2 (18,19). These
prediction rules are limited by the assumption of certain
geometries, which are not found naturally in the thorax.
Currently there is no consensus on the optimal method used
to grade the size of pleural effusions; most radiologists use a sub-
jective categorization such as ‘‘small,’’ ‘‘moderate,’’ and ‘‘large’’
(20). Because of varying thresholds for subjective grading, is-
sues can arise when communicating the perceived size of a
pleural effusion. With the advent of picture archiving and
communication system (PACS)-based volumetric tools, volu-
metric analysis of pleural effusions has become increasingly
quick and easy to perform, often not requiring a separate work-
2
Academic Radiology, Vol -, No -, - 2015
station or software package. Volumetric analysis offers the
advantage of calculating an objective volume without relying
on prediction rules which are prone to geometry limitations.
Although the availability and usage of volumetric analysis has
increased over recent years, a comparison of CT volumes
with thoracentesis volumes has not yet been reported in the
literature. In this study, we compare CT volumes with thora-
centesis volumes; we also compare subjective grading of pleural
effusion size with thoracentesis volumes.
MATERIALS AND METHODS
This was a retrospective cohort study of patients who were
treated with ultrasound-guided therapeutic thoracentesis at a
Academic Radiology, Vol -, No -, - 2015
Figure 2. The percent difference between CT volume and thoracentesis volume. CT, computed tomography.
single academic medical center between January 1, 2012, and
December 31, 2012; the goal of the therapeutic thoracentesis
was complete drainage of the pleural effusion. Of the 207 pa-
tients who were treated, 117 patients were excluded due to
lack of a chest CT within 2 weeks before thoracentesis; an
additional 13 patients were excluded due to loculated pleural
effusions and an additional 5 patients were excluded due to
rapidly enlarging pleural effusions (ie, pleural effusions which
rapidly enlarged in the interval between CT and thoracente-
sis). Four patients were excluded due to unsuccessful thora-
centesis and 1 patient was excluded due to nonimage-
guided thoracentesis before image-guided thoracentesis.
This left 67 patients with free-flowing pleural effusions who
had a chest CT within 2 weeks before thoracentesis
(Table 1). The CT protocol involved standard thin-section
1.25-mm collimation (with or without intravenous contrast
depending on the clinical indication) with a field of view
extending from above the jugular notch through the posterior
phrenic angles. Subjective grading of the pleural effusion size
as ‘‘small,’’ ‘‘moderate,’’ or ‘‘large’’ was performed by one of
three thoracic radiologists as part of the clinical interpretation.
Volumetric analysis of the pleural effusions was performed us-
ing Carestream VuePACS v11.3.2.4051 (Rochester, NY) on a
standard diagnostic workstation by a single radiology resident
without prior formal training in volumetric analysis. Volu-
metric analysis involved manual tracing of the pleural effusion
contour in the axial, coronal, or sagittal plane followed by vi-
sual confirmation and three-dimensional (3D) reconstruction
of the pleural effusion (Fig. 1). Interpolation of the pleural
effusion contour was performed automatically so that only
representative slices required tracing.
Ultrasound-guided thoracentesis was performed by a radi-
ology procedures team which consisted of an attending radi-
CT VOLUMETRIC ANALYSIS OF PLEURAL EFFUSIONS
ologist and a radiology resident or fellow. Standard
ultrasound-guided technique was used using a valved centesis
catheter.
The data collected included the volume of pleural fluid esti-
mated by CT volumetric analysis, the volume of pleural fluid
obtained at thoracentesis, the time between CT and thora-
centesis, and the subjective grade of the pleural effusion by
chest CT and by chest radiograph. The CT volumes were
compared with the thoracentesis volumes using a paired
two-tailed Student’s t-test.
The study was compliant with the Health Insurance Porta-
bility and Accountability Act and was approved by the local
institutional review board.
RESULTS
CT volumetric analysis was performed in all 67 subjects.
The average time between chest CT and thoracentesis was
3.2 days. The average CT volume was 900
500 mL,
and the average thoracentesis volume was 910
550 mL.
The average difference between CT volume and thoracent-
esis volume was 9.4 mL (1.3%)
290 mL (30%); the CT
volumes and thoracentesis volumes were not statistically
different (P = .79, paired two-tailed Student’s t-test). The
percentage difference between CT volume and thoracentesis
volume is displayed in Figure 2. There was no clear correla-
tion between differences in volume and days between CT
and thoracentesis (Fig. 3).
The thoracentesis volume of a ‘‘small,’’ ‘‘moderate,’’ and
‘‘large’’ pleural effusion, as graded on chest CT, was found to
be approximately 410
260 mL, 770
270 mL, and
1370
650 mL, respectively; the thoracentesis volume of a
3
CHIAO ET AL
Figure 3. The percent difference between CT volume and thoracentesis volume, by days between CT and thoracentesis. CT, computed
tomography.
Figure 4. The volume obtained at thoracentesis grouped by subjective CT grade (a) and subjective CXR grade (b). Note the large overlapping
ranges, as shown by the wide standard error bars. CT, computed tomography; CXR, chest radiograph.
‘‘small,’’ ‘‘moderate,’’ and ‘‘large’’ pleural effusion, as graded on
chest radiograph, was found to be approximately
610
320 mL, 1040
460 mL, and 1530
830 mL, respec-
tively (Fig. 4).
DISCUSSION
Pleural effusions alter respiratory mechanics by restricting to-
tal lung capacity, forced vital capacity, and forced-expiratory
volume (9,21–23). Gas exchange is also affected to a lesser
degree, with mild improvements in arterial oxygenation and
A-a gradients after therapeutic thoracentesis (24,25). In
addition, large pleural effusions expand the thoracic cage
which in turn shifts the length–tension curve of the
inspiratory muscles to an unfavorable position, contributing
to the sensation of dyspnea (10). Klecka et al. reported a
case in which a patient experienced dramatic dyspnea relief af-
ter large-volume thoracentesis despite absent perfusion to the
affected lung, suggesting that the sensation of dyspnea is more
likely because of the expansion of the thoracic cage rather than
because of hypoxemia (11). In addition, a study by Walden
et al. (26) demonstrated that there was a correlation between
the amount of fluid drained and the effects on oxygenation
with an increase in the PaO2 of 4 mm Hg for each 100 mL
of pleural fluid drained. However, other studies have shown
mixed results, and the clinical importance of pleural effusion
size is not yet fully understood.
4
Academic Radiology, Vol -, No -, - 2015
Most radiologists use a subjective system in grading pleural
effusion size, such as ‘‘small,’’ ‘‘moderate,’’ and ‘‘large.’’
Although an individual radiologist may have personal thresh-
olds for describing a pleural effusion size as ‘‘small,’’ ‘‘moder-
ate,’’ or ‘‘large,’’ this may not clearly translate to the referring
physician or to the patient. Adding a volumetric descriptor,
such as ‘‘1500 mL,’’ provides a more precise description. In
addition, inserting a 3D image of a patient’s pleural effusion
in a radiology report (such as that in Fig. 1), could aid in
communication as well. Our study demonstrates that CT
volumetric analysis accurately represents the volume obtained
at thoracentesis. In contradistinction, subjective grading
resulted in a wide range of thoracentesis volumes for each
grade. Volumetric analysis was straightforward and easily
learned, performed by a radiology resident with no prior
training in volumetric analysis.
Potential pitfalls of volumetric analysis include the acci-
dental inclusion of adjacent soft tissue (especially collapsed
lung and mediastinal structures), tracing in a nonoptimal
plane, and difficulty with manual tracing due to motion and
beam hardening artifacts (Table 2). In addition, complex
pleural effusions with multiple pockets of fluid require addi-
tional time to trace and can be prone to error. Nevertheless,
these pleural effusions can be analyzed, which is a substantial
advantage over current CT prediction rules. From our expe-
rience, the coronal plane was often the optimal plane for
tracing as it captured most of the pleural effusion in a single
continuous trace.
Academic Radiology, Vol -, No -, - 2015 CT VOLUMETRIC ANALYSIS OF PLEURAL EFFUSIONS

TABLE 2. Pitfalls in Pleural Effusion Tracing

Pitfall Effect Comments

Accidental inclusion of adjacent soft
tissue
Tracing in a nonoptimal plane
Motion and beam hardening
artifacts
Complex or loculated pleural
effusions
Most commonly occurs with
collapsed lung and mediastinal
structures. Results in
overestimation of pleural effusion.
Tracing in a nonoptimal plane can be
challenging and is often prone to
error.
Artifacts can obscure pleural fluid
boundaries. Motion artifact is
frequently worst near the
diaphragms, whereas beam
hardening is frequently worst near
the apices.
Complex pleural effusions can be
tedious and time-consuming to
trace. More prone to tracing error.
IV contrast helps differentiate fluid
from adjacent soft tissue.
Multiplanar reformats can also
help in defining the anatomy.
The optimal plane is the plane where
the pleural effusion can be
smoothly traced without
discontinuity.
Minimize artifacts with optimal CT
scanning technique.
May require multiple traces to fully
analyze a multicomponent pleural
effusion.

CT, computed tomography; IV, intravenous.

There are several limitations of this study. First, the sample
size was relatively small and included only free-flowing
pleural effusions. Loculated pleural effusions, which would
be difficult to completely drain by thoracentesis, were
excluded. Rapidly developing pleural effusions were also
excluded, as any significant increase in pleural effusion size
between the chest CT and the thoracentesis was felt to
skew the results. Second, we assumed that each thoracentesis
had drained the entire pleural effusion; however, a postpro-
cedural CT was not performed for confirmation. In fact,
there was likely some residual pleural effusion left after thor-
acentesis. Third, we measured the size of a pleural effusion
without adjusting for the size of each patient. Anderson
et al. (27) suggest that reporting the size of a pleural effusion
as a percentage of the hemithorax is more clinically relevant
than the absolute volume (because any given pleural effusion
volume will have less physiological impact on a larger pa-
tient). Although this technically could have been performed
in our study, it would have significantly increased the time
required to perform the analysis, potentially decreasing the
attractiveness of volumetric analysis as a clinical tool. Fourth,
the time to perform volumetric analysis was not assessed
rigorously. Even if only a few minutes were needed to
perform a single volumetric analysis, this could easily add
up in a busy radiology practice. This could potentially be
mitigated by only performing volumetric analysis on symp-
tomatic patients or in those who have more than a ‘‘small’’
pleural effusion. In addition, there is a substantial movement
in developing automatic segmentation algorithms (for
example, autosegmentation based on region growing, math-
ematical morphology, curve fitting, and active contour
models); these autosegmentation tools may reduce the time
required to perform volumetric analysis in the future (28).
The imaging evaluation of pleural effusion size has evolved
remarkably over the last 40 years, from the early chest radio-
graph prediction rules to chest CT prediction rules and now
to PACS-based CT volumetric analysis. Volumetric analysis
offers an accurate quantitative estimate of pleural effusion
size. Given the increased accessibility of CT volumetric tools
and the increased speed to perform CT volumetric analysis,
we expect that the reporting of quantitative volumes of pleural
effusions will become more commonly performed in the
future.
REFERENCES
1. Henschke CI, Davis SD, Romano PM, et al. Pleural effusions: pathogen-
esis, radiologic evaluation, and therapy. J Thorac Imaging 1989; 4:49–60.
2. Jimenez D, Diaz G, Gil D, et al. Etiology and prognostic significance of
massive pleural effusions. Respir Med 2005; 99:1183–1187.
3. Light RW. A new classification of parapneumonic effusions and empyema.
Chest 1995; 108:299–301.
4. Evans AL, Gleeson FV. Radiology in pleural disease: state of the art. Re-
spirology 2004; 9:300–312.
5. Light RW. Clinical practice. Pleural effusion. N Engl J Med 2002; 346:
1971–1977.
6. Hooper C, Lee YC, Maskell N, BTS Pleural Guideline Group. Investigation
of a unilateral pleural effusion in adults: British Thoracic Society Pleural
Disease Guideline 2010. Thorax 2010; 65(Suppl 2):ii4–17.
7. Eibenberger KL, Dock WI, Ammann ME, et al. Quantification of pleural ef-
fusions: sonography versus radiography. Radiology 1994; 191:681–684.
8. Kitazono MT, Lau CT, Parada AN, et al. Differentiation of pleural effusions
from parenchymal opacities: accuracy of bedside chest radiography. AJR
Am J Roentgenol 2010; 194:407–412.
9. Cartaxo AM, Vargas FS, Salge JM, et al. Improvements in the 6-min walk
test and spirometry following thoracentesis for symptomatic pleural effu-
sions. Chest 2011; 139:1424–1429.
10. Estenne M, Yernault JC, De Troyer A. Mechanism of relief of dyspnea after
thoracocentesis in patients with large pleural effusions. Am J Med 1983;
74:813–819.
11. Klecka ME, Maldonado F. Symptom relief after large-volume thoracente-
sis in the absence of lung perfusion. Chest 2014; 145:1141–1143.

5
CHIAO ET AL
12. Marcondes BF, Vargas F, Paschoal FH, et al. Sleep in patients with large
pleural effusion: impact of thoracentesis. Sleep Breath 2012; 16:483–489.
13. Talmor M, Hydo L, Gershenwald JG, et al. Beneficial effects of chest tube
drainage of pleural effusion in acute respiratory failure refractory to positive
end-expiratory pressure ventilation. Surgery 1998; 123:137–143.
14. Goligher EC, Leis JA, Fowler RA, et al. Utility and safety of draining pleural
effusions in mechanically ventilated patients: a systematic review and
meta-analysis. Crit Care 2011; 15:R46.
15. Razazi K, Thille AW, Carteaux G, et al. Effects of pleural effusion drainage
on oxygenation, respiratory mechanics, and hemodynamics in mechani-
cally ventilated patients. Ann Am Thorac Soc 2014; 11:1018–1024.
16. Blackmore CC, Black WC, Dallas RV, et al. Pleural fluid volume estimation:
a chest radiograph prediction rule. Acad Radiol 1996; 3:103–109.
17. Woodring JH. Recognition of pleural effusion on supine radiographs: how
much fluid is required? AJR Am J Roentgenol 1984; 142:59–64.
18. Mergo PJ, Helmberger T, Didovic J, et al. New formula for quantification of
pleural effusions from computed tomography. J Thorac Imaging 1999; 14:
122–125.
19. Mironov O, Ishill NM, Mironov S, et al. Pleural effusion detected at CT prior
to primary cytoreduction for stage III or IV ovarian carcinoma: effect on sur-
vival. Radiology 2011; 258:776–784.
20. Moy MP, Levsky JM, Berko NS, et al. A new, simple method for estimating
pleural effusion size on CT scans. Chest 2013; 143:1054–1059.
Academic Radiology, Vol -, No -, - 2015
21. Brown NE, Zamel N, Aberman A. Changes in pulmonary mechanics and
gas exchange following thoracocentesis. Chest 1978; 74:540–542.
22. Wang JS, Tseng CH. Changes in pulmonary mechanics and gas exchange
after thoracentesis on patients with inversion of a hemidiaphragm second-
ary to large pleural effusion. Chest 1995; 107:1610–1614.
23. Light RW, Stansbury DW, Brown SE. The relationship between pleural
pressures and changes in pulmonary function after therapeutic thoracent-
esis. Am Rev Respir Dis 1986; 133:658–661.
24. Perpina M, Benlloch E, Marco V, et al. Effect of thoracentesis on pulmo-
nary gas exchange. Thorax 1983; 38:747–750.
25. Ahmed SH, Ouzounian SP, Dirusso S, et al. Hemodynamic and pulmo-
nary changes after drainage of significant pleural effusions in critically
ill, mechanically ventilated surgical patients. J Trauma 2004; 57:
1184–1188.
26. Walden AP, Garrard CS, Salmon J. Sustained effects of thoracocentesis
on oxygenation in mechanically ventilated patients. Respirology 2010;
15:986–992.
27. Anderson SA, Danelson KA, Mammarappallil JG, et al. Semiautomatic
method of quantifying pleural effusions using computed tomography
scans - Biomed 2013. Biomed Sci Instrum 2013; 49:13–19.
28. Yao J, Han W, Summers RM. Computer aided evaluation of pleural effu-
sion using chest CT images. Proc IEEE Int Symp Biomed Imaging 2009;
2009:241–244.