pleural effusion in ICU.pdf - SASSiT

Pleural effusions in the intensive care unit
Elie Azoulay, MD, PhD
The incidence of pleural effusions in the intensive care unit
varies depending on the screening methods, from
approximately 8% for physical examination to more than 60%
for routine ultrasonography. Several factors contribute to the
occurrence of pleural effusions in intensive care unit patients:
large amounts of intravenous fluid are often administered,
pneumonia is common, and heart failure, atelectasis,
extravascular catheter migration, hypoalbuminemia, or liver
disease are present in many intensive care unit patients. In
surgical intensive care units, cardiac or abdominal surgery is
often followed by pleural effusions, and in trauma patients,
hemothorax is a dreaded event. Because no clinical parameter
excludes pleural infection, and because of the impact of
thoracentesis on diagnosis and treatment, this procedure
should be performed unless contraindicated. Thoracentesis is
safe in mechanically ventilated patients. The author discusses
the following points regarding pleural effusions in the intensive
care unit: screening intensive care unit patients for pleural
effusion, safety of thoracentesis in patients receiving invasive
mechanical ventilation, distinguishing exudates from
transudates, and diagnosing and managing infected pleural
effusions in critically ill patients. Lastly, the author suggests a
research agenda for pleural effusions in intensive care unit
patients.
Keywords
pleural effusions, pleurisy, pneumonia, empyema, thoracentesis
Curr Opin Pulm Med 2003, 9:291–297 © 2003 Lippincott Williams & Wilkins.
Service de Réanimation Médicale, Hôpital Saint Louis et Université Paris,
Assistance Publique–Hôpitaux de Paris, France.
Correspondence to Elie Azoulay, MD, PhD, Service de Réanimation Médicale,
Hôpital Saint-Louis, 1 Avenue Claude Vellefaux, 75010 Paris, France; e-mail:
elie.azoulay@sls.ap-hop-paris.fr
Current Opinion in Pulmonary Medicine 2003, 9:291–297
Abbreviations
ICU intensive care unit
LDH lactate dehydrogenase
ISSN 1070–5287 © Lippincott Williams & Wilkins
When a patient presents with a pleural effusion, three
questions must be asked: (1) Is there a contraindication
to thoracentesis? (2) Is the effusion transudative or exudative?
(3) If it is exudative, what is the cause [1,2]?
Thoracentesis is the key to discovering the cause. The
interstitial lung tissue is the main source of pleural fluid
[3]. Exudates denote a structural pleural abnormality,
and transudates indicate an imbalance between hydrostatic
and oncotic pressures across the normal pleura. In
both cases, the accumulation of fluid in the pleural cavity
indicates that fluid production exceeds the maximum
pleural lymphatic flow [3].
The incidence of pleural effusions in the intensive care
unit (ICU) varies with screening methods, from approximately
8% for physical examination to more than 60% for
routine ultrasonography [4,5]. Several factors contribute
to the occurrence of pleural effusions in ICU patients.
Large amounts of intravenous fluid are often administered
during the first few days to patients admitted for
shock, and pneumonia is common both as a reason for
ICU admission and as a complication of mechanical ventilation.
Heart failure, atelectasis, hypoalbuminemia, and
liver disease are present in many ICU patients. Extravascular
catheter migration [6] and subdiaphragmatic abnormalities
can cause pleural effusion. In surgical ICUs,
cardiac or abdominal surgery is often followed by specific,
large, protracted pleural effusions; and in multiply
injured patients, hemothorax is a dreaded event. Thus,
pleural effusions are common in ICU patients. Nevertheless,
their causes and management are identical to
those in other settings.
This review of the diagnosis and management of pleural
effusions in ICU patients rests on the most recent data
from the literature. The absence of reliable clinical or
laboratory test criteria for determining the cause of pleural
effusions and the potentially devastating consequences
of failing to diagnose and to treat pleural infection
are strong reasons to perform thoracentesis in
patients with clinically detectable pleural effusions and
no contraindication to the procedure. Thoracentesis can
be performed in mechanically ventilated patients.
Screening intensive care unit patients for
pleural effusion
The distribution of etiologies of pleural effusions depends
strongly on the screening method. For instance,
routine ultrasonography detected pleural effusions in
62% of medical ICU patients, with a predominance of
291

292 Diseases of the pleura
transudates [5]. In a study conducted by our group in
medical ICU patients, pleural effusions detected by
physical examination and obscuring at least one third of
the lung field on radiographs were found in only 8.4% of
patients [4] (Tables 1 and 2 [7–9]), and were usually
exudates related to infection (empyema or parapneumonic
effusion). A satisfactory compromise must be
found between a highly sensitive screening method that
detects effusions of little clinical relevance and a less
sensitive method that mainly detects effusions with a
diagnosis that will affect treatment decisions [4].
In addition to physical examination, three screening
methods are used: plain chest radiographs, pleural ultrasonography,
and CT of the chest. Physical examination
(auscultation, palpation, and percussion) is very sensitive
for detecting pleural effusions. Auscultatory percussion
may detect effusions as small as 50 mL [10]. A plain
chest radiograph should be obtained to confirm the diagnosis.
In non-ICU patients, effusions are visible on
lateral views at 50 mL and on conventional posteroanterior
views at 200 mL, whereas obscuring of the hemidiaphragm
occurs only with volumes of more than 500 mL
[11]. Oblique views are recommended as well [12]. In
ICU patients receiving invasive mechanical ventilation, a
daily chest radiograph at the bedside adds to the information
obtained by physical examination in nearly one in
five patients, although poor image quality may lead to
underestimation or overestimation of pleural involvement
[13]. Pleural ultrasonography seems valuable in
ICU patients as a noninvasive investigation that guides,
facilitates, expedites, and improves the safety of thoracentesis
in mechanically ventilated patients. Thus,
Lichtenstein et al. [14] reported that no complications of
thoracentesis occurred when the interpleural distance
was at least 15 mm and was visible over three intercostal
spaces. CT is extremely sensitive for the diagnosis of
pleural effusion, and the development of portable CT
devices seems to hold promise.
Thoracentesis is not contraindicated in
intensive care patients, even those
receiving invasive mechanical ventilation
Provided basic rules are followed, thoracentesis is safe in
ICU patients [4]. The experience of the operator seems
to influence the incidence of complications [15]. In our
study of clinically documented pleural effusions, routine
thoracentesis in patients without contraindications to this
procedure (agitation, severe hypoxia, or unstable hemodynamics)
was followed by pneumothorax in only 7% of
patients, most of whom were receiving mechanical ventilation
with positive end-expiratory pressures greater
than 5 mmHg [4]. Chest tube drainage consistently ensured
resolution of the pneumothorax. Thoracentesis altered
the diagnosis in 45% of patients and altered the
treatment in 33% [4]. Ultrasound guidance reduces the
morbidity of thoracentesis in mechanically ventilated patients
[13,14].
The morbidity associated with drainage in ICU patients
can be reduced by prompt chest tube removal. The optimal
drainage duration for uninfected pleural effusions
has not been established. A reasonable approach may be
to remove the chest tubes when drainage falls to less
Table 1. Characteristics of medical intensive care unit patients without contraindications to thoracentesis
(n = 82; from Fartoukh et al. [4])
Characteristic
All effusions
(n = 82)
Transudates
(n = 20)
Infectious
exudates (n = 35)
Noninfectious
exudates (n = 27)
MICU admission and stay
Admission for pulmonary edema, n (%) 19 (23) 11 (55) † 3 (9)* 5 (19)
SAPS II score at admission, pt (range) 46 (30–56) 51 (38–61) † 48 (28–58)* 40 (26–46)
Length of ICU stay, d (range) 11 (6–19) 10 (6–18) 14 (5–23) 11 (7–16)
Mortality, n (%) 29 (35) 9 (45) 11 (32) 9 (34)
Clinical features at MICU admission
Body temperature, °C (range) 37.7 (37–38.4) 37 (36.6–37.9)* 37.8 (37–38.6) 37.5 (37–38.2)
Bilateral ankle edema, n (%) 40 (49) 17 (85) † 13 (37)* 10 (37)
Unilateral effusion, n (%) 40 (49) 6 (30)* 23 (66) 11 (40)
Documented heart failure, n (%) 13 (16) 9 (45) † 1 (3)* 3 (11)
Laboratory data at thoracentesis 11,800 12,450 11,800 12,600
Blood leukocytes/mm 3 (range) (7375–16,350) (8700–16,300) (7675–16,950) (5725–16,250)
Serum fibrinogen, g/L (range) 4.5 (3.4–6) 4.15 (3.1–5.3) 5.7 (4.2–6.9)* 4 (3.3–5.2) ‡
Characteristics of the effusion
Time from MICU admission to thoracentesis, d (range) 2(0–6) 1 (0–5) 2 (0–10) 2 (0–6)
Clear fluid, n (%) 39 (48) 16 (80)* 10 (29) 13 (48)
Pleural leukocytes/mm 3 , (range) 235 (85–1,180) 131 (42–267)* 1020 (100–2047) † 250 (92–467)
Pleural neutrophils, % (range) 38 (10–80) 25 (11–40)* 80 (63–85) 9 (1–20)
Pleural protein, g/L (range) 26.5 (18–40) 16 (13–21) † 36 (26–50)* 29 (21–40)
Pleural-fluid-to-serum protein ratio (range) 0.47 (0.3–0.65) 0.29 (0.25–0.38) † 0.60 (0.47–0.74)* 0.57 (0.42–0.64)
Pleural fluid-to-serum LDH ratio (range) 1.03 (0.5–2.7) 0.46 (0.33–1.01) † 2.41 (1.06–6.63) ‡ 0.74 (0.45–1.60)
*p < 0.05 between transudates and infectious exudates. † p < 0.05 between transudates and noninfectious exudates. ‡ p < 0.05 between infectious
exudates and noninfectious exudates. ICU, intensive care unit; LDH, lactate dehydrogenase; MICU, medical intensive care unit; SAPS II, Simplified
Acute Physiology Score version II.

Pleural effusions in the ICU Azoulay 293
Table 2. Causes of pleural effusions in intensive care unit patients
Causes of pleural effusions
Light et al. [7],
n = 150
Colt et al. [8],
n = 205
Heffner et al. [9],
n = 1448
Mattison et al. [5],
n = 62*
Fartoukh et al. [4],
n = 113*
Congestive heart failure 39 (26%) 22 (10.7%) 295 (20.4%) 22 (35.5%) 28 (24.8%)
Hepatic hydrothorax 5 (3.3%) / 43 (3%) 5 (8%) 6 (5.3%)
Nephrotic syndrome 3 (2%) 7 (3.4%) 39 (2.7%) 5 (8%) 1 (0.8%)
Parapneumonic effusions 26 (17.3%) 54 (26.3%) 185 (12.8%) 7 (11.3%) 29 (25.6%)
Empyema / 6 (2.9%) / 1 (1.6%) 12 (10.6%)
Tuberculosis 14 (9.3%) 7 (3.4%) 296 (20.4%) / 2 (1.7%)
Malignancies 43 (28.7%) 109 (53.2%) 438 (30.2%) 2 (3.2%) 11 (9.7%)
Pulmonary embolism 5 (3.3%) / 36 (2.5%) / 5 (4.4%)
Hemothorax 2 (1.3%) / 14 (1%) / 4 (3.4%)
Postsurgical effusions / 9 (4.4%) / / 5 (4.4%)
Atelectasis / 9 (4.4%) / 14 (22.6%) 2 (1.7%)
Pancreatitis 6 (4%) / 5 (0.3%) 1 (1.6%) 2 (1.7%)
Other 7 (4.7%) / 97 (6.7%) / /
Unknown / 25 (12.2%) / 5 (8%) 6 (5.3%)
*Series of patients admitted to intensive care units.
than 200 mL per day [16]. Placement of a small indwelling
pleural catheter may reduce morbidity compared
with repeated needle aspiration or chest tube drainage.
In a study of 57 aspirations in 23 patients with large
effusions (defined as opacification of one third of the
hemithorax on the chest radiograph), catheter drainage
seemed effective and was associated with less morbidity
and lower costs than repeated needle aspiration or chest
tube drainage in historical control subjects [17].
Causes of pleural effusions in intensive
care patients
There are no specific causes of pleural effusions in the
ICU. Although complications of invasive procedures [6]
and trauma-related hemothorax [18] are common, all the
causes of pleural effusions can be encountered. Transudative
effusions in ICU patients may be related to congestive
heart failure, hepatic hydrothorax, or nephrotic
syndrome [2]. Exudative effusions may denote an infection
(empyema or parapneumonic effusion) or a noninfectious
condition (surgery, intraabdominal abnormality,
malignancy) [19]. Pleural effusions complicating heart
surgery have been described recently [2,20]. Their incidence
seems dependent on the surgical technique [21].
We conducted a survey among French intensivists to
evaluate their perceptions of pleural effusions in the
ICU. The respondents thought that nearly one in five
ICU patients had a pleural effusion and that congestive
heart failure was the leading cause (57%), followed by
infection (16%). Intensivists who performed thoracentesis
routinely thought that this procedure was warranted
to avoid missing a pleural infection [22].
Table 1 reports the characteristics of the patients included
in our study of pleural effusions in three medical
ICUs over a 1-year period [4]. The effusions were detected
clinically and confirmed by chest radiographs before
thoracentesis. Of the 82 patients with no contraindications
to thoracentesis, 35 had infected exudative
effusions (21 parapneumonic effusions and 14 empyemas),
27 had uninfected exudative effusions, and only
20 had transudates. As shown in Table 2, the screening
method influences the distribution of causes. Our series
shares a far greater resemblance to data obtained outside
the ICU than to data reported by Mattison et al. [5], who
used ultrasonography and found that most pleural effusions
were small transudates. Finally, some pleural effusions
remain undiagnosed despite a careful analysis of
clinical findings and laboratory tests in blood and pleural
fluid (repeated thoracentesis) [23]. In ICU patients, undiagnosed
venous thromboembolism may explain some
of these effusions [23–25].
Differentiating exudates from transudates
The criteria of Light et al. [7], which are based on the
ratio of protein or lactate dehydrogenase (LDH) levels in
the pleural fluid and blood, differentiate exudates from
transudates with a negative predictive value of 96% and
a sensitivity of 98%. These criteria were validated in
pulmonology but remain just as useful in ICU patients,
despite the common presence of undernutrition, hemodilution,
infection, liver failure, and renal failure [26]. In
patients on diuretic therapy, the albumin difference between
the serum and the pleural fluid is useful for separating
exudates from transudates [27,28]. Other variables
have been suggested as useful adjuncts to the diagnosis
of exudates, including albumin, cholesterol, amylase,
pleural fluid pH, triglycerides, and bilirubin [24,29].
Their value has been challenged, however [30,31]. Furthermore,
in addition to the fluid-to-serum ratio of LDH,
the absolute LDH level in the fluid proved useful in a
study of 212 patients studied by Joseph et al. [32]. Table
1 reports the laboratory test findings in the 82 patients
who underwent thoracentesis in our 2002 study in three
medical ICUs [4]. The data make a convincing case that
thoracentesis benefits both the diagnosis and the treatment
in ICU patients, and they confirm the validity of
the criteria of Light et al. [7] in the ICU. Recently, Romero–Candeira
et al. [33] established that the criteria of

294 Diseases of the pleura
Light et al. [7] were better than clinical examination for
establishing the diagnosis of exudate in 249 non-ICU
patients. In their study [33], the criteria of Light et al. [7]
had a 99.5% sensitivity, and other biochemical parameters
added little to the diagnosis. In difficult cases,
rather than a single test or constellation of tests, the
thresholds for biochemical tests that are likely to influence
the pretest clinical diagnosis of exudate or transudate
deserve consideration [34].
Other investigations have been suggested for differentiating
exudates from transudates. Thickening of the parietal
pleura can be visualized by CT with contrast agent
injection [35]. On pleural sonograms, transudates are
echo free, whereas echoic effusions are likely to be exudates,
particularly when loculation or pleural thickening
is seen, and a homogeneous echoic appearance usually
denotes an empyema or hemothorax [36]. Reagent strips
provide fast and reliable discrimination between transudates
and exudates (protein level) and, among exudates,
between infected and uninfected effusions (leukocyte
esterase level) [26]. Glucose, pH, LDH, and leukocyte
counts in the pleural fluid are useful for differentiating
empyema from uncomplicated parapneumonic effusions
[2,37]. Needle or thoracoscopic pleural lavage is being
evaluated, but currently seems useful only for diagnosing
malignant effusions [38,39]. A novel semirigid pleuroscope
is being evaluated [40]. Finally, cytokine assays in
pleural fluid are useful mainly for elucidating the pathophysiology
and genesis of pleural effusions. Assays of
adenosine deaminase and interferon- are useful for the
diagnosis of tuberculosis [2].
Infected pleural effusions (parapneumonic
effusions and empyema): diagnosis
and management
In response to a pleural infectious insult, the mesothelial
cells initiate a massive inflammatory reaction that includes
extravasation of neutrophils and proteins. Cooperation
between neutrophils and mesothelial cells results
in the production of proinflammatory cytokines, oxidants,
and proteinases [2,37]. The increase in pleural
permeability results in the accumulation of a highly inflammatory
exudative effusion that may contain the organism
responsible for the pneumonia or empyema.
Pleural drainage and antibiotic therapy may be successful
in controlling this acute inflammatory response, leaving
minor pleural fibrosis as the only residual abnormality.
However, moderate to severe pleural fibrosis and
contraction may occur. Early diagnosis and aggressive
treatment are essential to prevent this outcome.
Infection of the pleural cavity can result from surgery,
trauma, or lung infection. Although pleural effusions develop
in more than half the patients with infectious
pneumonia, they usually resolve with antibiotic therapy,
having no impact on the management strategy. In ICU
patients, however, a pleural effusion may develop as a
complication manifesting as persistent fever or respiratory
status deterioration. Even in the absence of major
criteria for drainage, this procedure may improve the patient’s
vital functions. Furthermore, because pneumonia
with parapneumonic effusions and empyema are
common, failure of antibiotic therapy alone with pleural
cavity loculation and recurrence of sepsis is not infrequent
[41].
Recommendations on the medical and surgical treatment
of parapneumonic pleural effusions have been developed
recently by the American College of Chest Physicians
based on a painstaking review of the literature [37].
I will not provide a detailed discussion of the antibiotic
treatment of empyema or parapneumonic pleural effusions.
Unless minimal, a parapneumonic effusion should
lead to thoracentesis for cytologic and bacteriologic studies,
glucose and LDH assays, and pH determination.
Thoracentesis should be performed early [42,43], particularly
in immunocompromised patients. Nonloculated
effusions should be drained. Recurrent effusions that do
not meet criteria for empyema (positive cultures, high
glucose level, acidic pH, or pleural LDH level more than
three times the serum LDH level) can be managed by
close monitoring. However, in patients with loculated
effusions or criteria for empyema, a chest tube should be
inserted and thrombolytic agents administered. When
drainage is incomplete, thoracoscopy with debridement
of the pleural cavity is in order. Thoracostomy with decortication
is indicated only when all else fails. These
recommendations leave open the debate on primary surgical
treatment in complicated parapneumonic effusions
or empyema [37], particularly because Mandal et al. [44]
found that 42% of patients with primary empyema ultimately
required decortication. Pleural drainage has been
used since antiquity [45]. Injection of fibrinolytic agents
into the pleural cavity was introduced in the early 1990s,
and urokinase is now the fibrinolytic of choice [46]. Although
drainage with injection of fibrinolytics is effective
and safe in most patients [47], 10% of patients subsequently
require more invasive procedures such as
video-assisted thoracoscopy. Furthermore, pleural decortication
is more likely to be required in patients with
anaerobic, staphylococcal, or pneumococcal infections
[44]. Bouros et al. [48] recently reported that videoassisted
thoracoscopic surgery was effective in 17 of
20 patients with parapneumonic effusion or empyema
refractory to drainage and urokinase instillation. Three
patients required conversion to open thoracotomy for
pleural decortication [48]. An important goal is early discrimination
between patients who will respond to drainage
and fibrinolytics, and those who will require surgery.
Huang et al. [49] found that drainage for parapneumonic
effusion or empyema failed in 47 of 100 patients and

Pleural effusions in the ICU Azoulay 295
that the variables associated independently with failure
were loculation and a pleural fluid leukocyte count
6,400/mm 3 . In a study of 85 patients with parapneumonic
effusion or empyema, drainage and fibrinolytics
failed in 13 patients, and the only significant predictor
was absence of purulence, which predicted success,
whereas purulence did not predict failure [50]. Survival
was 86% after 4 years [50]. These data seem to identify
patient subsets most likely to benefit from early or primary
surgery. Early surgery lowered mortality in a nonrandomized
study [51]. Nevertheless, because drainage
and fibrinolytics ensure a favorable outcome in most patients,
early surgery may be best reserved for patients at
risk for complications of prolonged drainage (eg, advanced
age, severe comorbidities, or immunodepression)
or with major pleural abnormalities by CT. Further studies
are needed to determine the role for early surgery,
particularly as new and more effective fibrinolytic agents
are being developed [52]. In addition, the use of surgery
is based largely on studies conducted before the intrapleural
fibrinolytic era, and CT data suggest that pleural
abnormalities may improve after recovery from empyema.
Recently developed experimental models will help
us to optimize current treatment strategies [53].
Similarly, the better overall results obtained with videoassisted
thoracic surgery than with conventional surgery
[54], particularly at the fibrinopurulent phase of empyema
[55,56], and the favorable outcomes achieved recently
with video-assisted thoracoscopic decortication
[57,58] indicate that surgical techniques long considered
to be procedures of last resort are evolving into less invasive,
safer variants.
Research agenda for pleural effusions in
intensive care patients
Although the data reviewed here indicate that the
diagnosis and treatment of pleural effusions should
follow the same rules in the ICU as they do elsewhere,
several incompletely resolved issues deserve further
investigation.
Morbidity ascribable to pleural effusions in intensive
care patients
Crude mortality rates may be higher in ICU patients
with pleural effusions than in those without pleural effusions
[4]. I am not aware of any data on the attributable
risk of pleural effusions for morbidity and mortality in
ICU patients. El-Ebiary et al. [59] found that pleural
effusion was associated with a poor prognosis in patients
with Legionella pneumonia. Others reported that parapneumonic
effusions were more common in patients with
greater severity of community-acquired pneumonia, a
longer time to treatment, or positive blood cultures [2]. A
vast epidemiologic study would provide information on
the impact of pleural effusions on ICU outcomes such as
time in the ICU, time on mechanical ventilation, ease of
weaning, nosocomial and iatrogenic adverse events, and
death. These studies will have to take into account and
adjust for the screening method used, with small, usually
transudative effusions detected by ultrasonography or
CT at one end of the spectrum and clinically detectable
effusions with radiographic opacification of at least one
third of the hemothorax at the other end.
Optimizing the management of parapneumonic
effusions and empyema
As shown earlier, three requirements emerge from the
literature on infected pleural effusions: thoracentesis
should be performed routinely and pleural fluid biochemistry
analyzed adequately to ensure early diagnosis;
antibiotics, drainage, and fibrinolytics should be instituted
as soon as possible; and patients at risk for complications
should be identified (purulence, inflammation,
loculation). The role for early video-assisted thoracoscopic
surgery (debridement, pleural toilette, optimal
drainage) remains to be determined. This procedure may
prove useful either early (after 48 hours of optimal medical
treatment) or as the primary treatment. The first alternative
would leave time to evaluate novel fibrinolytic
agents. This strategy could be reserved for high-risk patients
(immunodepression, advanced age, chronic respiratory
failure) or could be offered to all patients. Evaluation
criteria would have to include lung function data,
lengths of stay in the ICU and hospital, mortality, and
cost of the acute condition.
Does pleural fluid drainage help to optimize
mechanical ventilation in patients with acute
respiratory distress syndrome?
Pleural effusions may develop in patients with acute respiratory
distress syndrome and refractory hypoxemia
[60]. Conceivably, drainage may improve oxygenation
and ventilatory mechanics. Pleural effusions of moderate
size had little effect on oxygenation [61]. However, Talmor
et al. [60] found that drainage improved oxygenation
and dynamic compliance in 17 of 19 patients with severe
acute respiratory distress syndrome, independent from
the volume of fluid removed. Guinard et al. [62] used a
therapeutic optimization strategy including drainage in
36 patients with severe acute respiratory distress syndrome
and found that a response to this strategy was
correlated with survival. In nine patients with pleural
effusions and no other medical conditions, Agusti et al.
[63] reported that drainage had no effects on ventilationto-perfusion
ratios or oxygenation. Clarifying the impact
of pleural drainage in patients with acute respiratory distress
syndrome and identifying factors associated with
beneficial effects may be of interest, given the risks associated
with pleural drainage in these patients. Appropriate
evaluation criteria would include the PaO 2 -to-
FiO 2 ratio, pleural pressure [64,65], esophageal pressure,
lung compliance, and time on mechanical ventilation.

296 Diseases of the pleura
References and recommended reading
1 Light RW: Pleural effusions. Med Clin North Am 1977, 61:1339–1352.
2 Light RW: Clinical practice. Pleural effusion. N Engl J Med 2002, 346:1971–
1977.
3 Miserocchi G: Physiology and pathophysiology of pleural fluid turnover. Eur
Respir J 1997, 10:219–225.
4 Fartoukh M, Azoulay E, Galliot R, et al.: Clinically documented pleural effusions
in medical ICU patients: how useful is routine thoracentesis? Chest
2002, 121:178–184.
5 Mattison LE, Coppage L, Alderman DF, et al.: Pleural effusions in the medical
ICU: prevalence, causes, and clinical implications. Chest 1997, 111:1018–
1023.
6 Strange C: Pleural complications in the intensive care unit. Clin Chest Med
1999, 20:317–327.
7 Light RW, Macgregor MI, Luchsinger PC, et al.: Pleural effusions: the diagnostic
separation of transudates and exudates. Ann Intern Med 1972,
77:507–513.
8 Colt HG, Brewer N, Barbur E: Evaluation of patient-related and procedurerelated
factors contributing to pneumothorax following thoracentesis. Chest
1999, 116:134–138.
9 Heffner JE, Brown LK, Barbieri C, et al.: Pleural fluid chemical analysis in
parapneumonic effusions. A meta-analysis. Am J Respir Crit Care Med 1995,
151:1700–1708.
10 Guarino JR, Guarino JC: Auscultatory percussion: a simple method to detect
pleural effusion. J Gen Intern Med 1994, 9:71–74.
11 Blackmore CC, Black WC, Dallas RV, et al.: Pleural fluid volume estimation:a
chest radiograph prediction rule. Acad Radiol 1996, 3:103–109.
12 Lawson CC, LeMasters MK, Kawas Lemasters G, et al.: Reliability and validity
of chest radiograph surveillance programs. Chest 2001, 120:64–68.
13 Koh DM, Burke S, Davies N, et al.: Transthoracic US of the chest: clinical uses
and applications. Radiographics 2002, 22:E1.
14 Lichtenstein D, Hulot JS, Rabiller A, et al.: Feasibility and safety of ultrasoundaided
thoracentesis in mechanically ventilated patients. Intensive Care Med
1999, 25:955–958.
15 Collins TR, Sahn SA: Thoracocentesis. Clinical value, complications, technical
problems, and patient experience. Chest 1987, 91:817–822.
16 Younes RN, Gross JL, Aguiar S, et al.: When to remove a chest tube? A
randomized study with subsequent prospective consecutive validation. J Am
Coll Surg 2002, 195:658–662.
17 Grodzin CJ, Balk RA: Indwelling small pleural catheter needle thoracentesis in
the management of large pleural effusions. Chest 1997, 111:981–988.
18 Hata N, Tanaka K, Imaizumi T, et al.: Clinical significance of pleural effusion in
acute aortic dissection. Chest 2002, 121:825–830.
19 Yoshii C, Morita S, Tokunaga M, et al.: Bilateral massive pleural effusions
caused by uremic pleuritis. Intern Med 2001, 40:646–649.
20 Lee YC, Vaz MA, Ely KA, et al.: Symptomatic persistent post-coronary artery
bypass graft pleural effusions requiring operative treatment: clinical and histologic
features. Chest 2001, 119:795–800.
21 Payne M, Magovern GJ Jr, Benckart DH, et al.: Left pleural effusion after coronary
artery bypass decreases with a supplemental pleural drain. Ann Thorac
Surg 2002, 73:149–152.
22 Azoulay E, Fartoukh M, Similowski T, et al.: Routine exploratory thoracentesis
in ICU patients with pleural effusions: results of a French questionnaire study.
J Crit Care 2001, 16:98–101.
23 Ansari T, Idell S: Management of undiagnosed persistent pleural effusions.
Clin Chest Med 1998, 19:407–417.
24 Light RW: Pleural effusion due to pulmonary emboli. Curr Opin Pulm Med
2001, 7:198–201.
25 Pastores SM, Halpern NA: Autopsies in the ICU: we still need them! Crit Care
Med 1999, 27:235–236.
26 Azoulay E, Fartoukh M, Galliot R, et al.: Rapid diagnosis of infectious pleural
effusions by use of reagent strips. Clin Infect Dis 2000, 31:914–919.
27 Eid AA, Keddissi JI, Samaha M, et al.: Exudative effusions in congestive heart
failure. Chest 2002, 122:1518–1523.
28 Romero–Candeira S, Fernandez C, Martin C, et al.: Influence of diuretics on
the concentration of proteins and other components of pleural transudatesin
patients with heart failure. Am J Med 2001, 110:681–686.
29 Vaz MA, Marchi E, Vargas FS: Cholesterol in the separation of transudates
and exudates. Curr Opin Pulm Med 2001, 7:183–186.
30 Branca P, Rodriguez RM, Rogers JT, et al.: Routine measurement of pleural
fluid amylase is not indicated. Arch Intern Med 2001, 161:228–232.
31 Vaz MA, Teixeira LR, Vargas FS, et al.: Relationship between pleural fluid and
serum cholesterol levels. Chest 2001, 119:204–210.
32 Joseph J, Badrinath P, Basran GS, et al.: Is the pleural fluid transudate or
exudate? A revisit of the diagnostic criteria. Thorax 2001, 56:867–870.
33 Romero–Candeira S, Hernandez L, Romero–Brufao S, et al.: Is it meaningful
to use biochemical parameters to discriminate between transudative and exudative
pleural effusions? Chest 2002, 122:1524–1529.
34 Heffner JE, Sahn SA, Brown LK: Multilevel likelihood ratios for identifying
exudative pleural effusions. Chest 2002, 121:1916–1920.
35 Aquino SL, Webb WR, Gushiken BJ: Pleural exudates and transudates: diagnosis
with contrast-enhanced CT. Radiology 1994, 192:803–808.
36 Yang PC, Luh KT, Chang DB, et al.: Value of sonography in determining the
nature of pleural effusion: analysis of 320 cases. AJR Am J Roentgenol 1992,
159:29–33.
37 Colice GL, Curtis A, Deslauriers J, et al.: Medical and surgical treatment of
parapneumonic effusions: an evidence-based guideline. Chest 2000,
118:1158–1171.
38 Mohamed KH, Mobasher AA, Yousef AI, et al.: Pleural lavage: a novel diagnostic
approach for diagnosing exudative pleural effusion. Lung 2000,
178:371–379.
39 Noppen M: Normal volume and cellular contents of pleural fluid. Curr Opin
Pulm Med 2001, 7:180–182.
40 Ernst A, Hersh CP, Herth F, et al.: A novel instrument for the evaluation of the
pleural space: an experience in 34 patients. Chest 2002, 122:1530–1534.
41 Strange C, Sahn SA: The definitions and epidemiology of pleural space infection.
Semin Respir Infect 1999, 14:3–8.
42 Heffner JE, McDonald J, Barbieri C, et al.: Management of parapneumonic
effusions. An analysis of physician practice patterns. Arch Surg 1995,
130:433–438.
43 Sasse S, Nguyen TK, Mulligan M, et al.: The effects of early chest tube placement
on empyema resolution. Chest 1997, 111:1679–1683.
44 Mandal AK, Thadepalli H, Chettipally U: Outcome of primary empyema thoracis:
therapeutic and microbiologic aspects. Ann Thorac Surg 1998,
66:1782–1786.
45 Aboud FC, Verghese AC: Evarts Ambrose Graham, empyema, and the dawn
of clinical understanding of negative intrapleural pressure. Clin Infect Dis
2002, 34:198–203.
46 Bouros D, Schiza S, Tzanakis N, et al.: Intrapleural urokinase versus normal
saline in the treatment of complicated parapneumonic effusions and empyema.
A randomized, double-blind study. Am J Respir Crit Care Med 1999,
159:37–42.
47 Sahn SA: Use of fibrinolytic agents in the management of complicated parapneumonic
effusions and empyemas. Thorax 1998, 53:S65–S72.
48 Bouros D, Antoniou KM, Chalkiadakis G, et al.: The role of video-assisted
thoracoscopic surgery in the treatment of parapneumonic empyema after the
failure of fibrinolytics. Surg Endosc 2002, 16:151–154.
49 Huang HC, Chang HY, Chen CW, et al.: Predicting factors for outcome of
tube thoracostomy in complicated parapneumonic effusion for empyema.
Chest 1999, 115:751–756.
50 Davies CW, Kearney SE, Gleeson FV, et al.: Predictors of outcome and longterm
survival in patients with pleural infection. Am J Respir Crit Care Med
1999, 160:1682–1687.
51 Lim TK, Chin NK: Empirical treatment with fibrinolysis and early surgery reduces
the duration of hospitalization in pleural sepsis. Eur Respir J 1999,
13:514–518.
52 Light RW, Nguyen T, Mulligan ME, et al.: The in vitro efficacy of Varidase
versus streptokinase or urokinase for liquefying thick purulent exudative material
from loculated empyema. Lung 2000, 178:13–18.
53 Sasse SA, Causing LA, Mulligan ME, et al.: Serial pleural fluid analysis in a
new experimental model of empyema. Chest 1996, 109:1043–1048.
54 Johna S, Alkoraishi A, Taylor E, et al.: Video-assisted thoracic surgery: applications
and outcome. JSLS 1997, 1:41–44.
55 Cassina PC, Hauser M, Hillejan L, et al.: Video-assisted thoracoscopy in the
treatment of pleural empyema: stage-based management and outcome.
J Thorac Cardiovasc Surg 1999, 117:234–238.

Pleural effusions in the ICU Azoulay 297
56 Striffeler H, Gugger M, Im Hof V, et al.: Video-assisted thoracoscopic surgery
for fibrinopurulent pleural empyema in 67 patients. Ann Thorac Surg 1998,
65:319–323.
57 Cheng YJ, Wu HH, Chou SH, et al.: Video-assisted thoracoscopic surgery in
the treatment of chronic empyema thoracis. Surg Today 2002, 32:19–25.
58 Waller DA, Rengarajan A: Thoracoscopic decortication: a role for videoassisted
surgery in chronic postpneumonic pleural empyema. Ann Thorac
Surg 2001, 71:1813–1816.
59 El-Ebiary M, Sarmiento X, Torres A, et al.: Prognostic factors of severe Legionella
pneumonia requiring admission to ICU. Am J Respir Crit Care Med
1997, 156:1467–1472.
60 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.
61 Gillespie DJ, Rehder K: Effect of positional change on ventilation–perfusion
distribution in unilateral pleural effusion. Intensive Care Med 1989, 15:266–
268.
62 Guinard N, Beloucif S, Gatecel C, et al.: Interest of a therapeutic optimization
strategy in severe ARDS. Chest 1997, 111:1000–1007.
63 Agusti AG, Cardus J, Roca J, et al.: Ventilation–perfusion mismatch in patients
with pleural effusion: effects of thoracentesis. Am J Respir Crit Care
Med 1997, 156:1205–1209.
64 Light RW, Stansbury DW, Brown SE: The relationship between pleural pressures
and changes in pulmonary function after therapeutic thoracentesis.Am
Rev Respir Dis 1986, 133:658–661.
65 Villena V, Lopez–Encuentra A, Pozo F, et al.: Measurement of pleural pressure
during therapeutic thoracentesis. Am J Respir Crit Care Med 2000,
162:1534–1538.