| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Medical and Respiratory Intensive Care Unit (Dr. Fartoukh), Antoine Béclère Teaching Hospital; Medical Intensive Care Unit (Drs. Azoulay, Le Gall, and Schlemmer) and Biostatistics Department (Dr. Chevret), Saint Louis Teaching Hospital and Paris 7 University; and Medical Intensive Care Unit (Drs. Galliot and Baud), Lariboisière Teaching Hospital, Paris, France.
Correspondence to: Elie Azoulay, MD, Medical ICU, Saint Louis Teaching Hospital, 1 avenue Claude Vellefaux, 75010 Paris, France; e-mail: elie.azoulay{at}sls.ap-hop-paris.fr
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Design and setting: Prospective, 1-year, three-center study in medical ICU (MICU) patients with physical and radiographic evidence of pleural effusion.
Patients: Of 1,351 patients admitted to three MICUs during the study period, 113 patients had physical and radiographic evidence of pleural effusion, yielding an annual incidence of 8.4%.
Intervention: Routine thoracentesis in 82 patients without contraindications to thoracentesis.
Measurements and results: Twenty patients (24.4%) had a transudate, 35 patients (42.7%) had an infectious exudate (parapneumonic, n = 21; empyema, n = 14), and 27 patients (32.9%) had a noninfectious exudate. Laboratory parameters including the leukocyte count, the neutrophil percentage in pleural fluid, and the fluid/serum protein and lactate dehydrogenase ratios differed significantly among the three groups. Thoracentesis yielded improvements in the diagnosis and/or treatment in 46 patients (56%): the presumptive (prethoracentesis) diagnosis was changed in 37 patients (32 patients with certain benefit and 5 patients with probable benefit from thoracentesis), of whom 27 patients received a change in treatment based on the new diagnosis; 9 other patients received a change in treatment although the diagnosis remained the same. The only complications were pneumothorax in six patients (7%), all with a favorable outcome after drainage.
Conclusion: Infection was the main cause of pleural effusions detected based on physical and radiographic findings in our MICU population. Routine thoracentesis proved a simple and safe means of improving the diagnosis and treatment.
Key Words: chest radiograph • clinical examination • diagnosis • empyema • exudate • ICU • pleural disease • pleural effusion • profitability • thoracentesis • transudate • treatment
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When a pleural effusion is detected, the need for and safety of thoracentesis must be evaluated; if the fluid is an exudate, a cause must be sought.2 Pleural effusions of < 500 mL may produce no more than a subtle haziness over the lower lungs on chest radiographs obtained in the supine or semirecumbent position, as is often the case in MICU patients.3 4 Although chest ultrasound and CT have been proved effective in detecting small pleural effusions,1 5 the clinical relevance of these small effusions in MICU patients remains unclear. Patients with physical and radiographic evidence of pleural effusion may be more likely to benefit from thoracentesis than patients with a pleural effusion detectable only by ultrasound and/or CT.
In a prospective, three-center study of MICU patients with physical and radiographic evidence of pleural effusion, we performed routine thoracentesis to evaluate the contribution of this procedure to the etiologic diagnosis and treatment.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following data were collected: (1) age and sex, comorbidities, reasons for MICU admission, simplified acute physiology (SAPS) II score6 at MICU admission, and vital signs and clinical data on the day of thoracentesis (temperature, ankle edema, cardiogenic pulmonary edema, need for mechanical ventilation, and need for vasopressors); (2) characteristics of the effusion: side, unilateral or bilateral, time from MICU admission to thoracentesis, appearance, and laboratory findings (protein, albumin, glucose, and lactate dehydrogenase [LDH] levels; leukocyte count and proportion of neutrophils; and smear and culture findings for bacteria and mycobacteria); and (3) serum test results on the day of thoracentesis (protein, albumin, LDH, and leukocyte count).
Pleural effusion was diagnosed based on physical examination whenever possible: absent breath sounds at auscultation, flatness to percussion, and reduced tactile fremitus in patients able to speak. Confirmation of the effusion was sought routinely on a chest radiograph obtained in the supine or semirecumbent position and reviewed independently by three clinicians (two senior and one junior). The size and side of the effusion were noted. Blunting of the costophrenic angle or obscuring of the lower lung defined a small effusion, opacification of the lower and middle lung defined a moderate effusion, and opacification of the entire lung defined a large effusion. When a pleural effusion was detected, thoracentesis was performed in accordance with published guidelines.7 Contraindications to thoracentesis included hemodynamic instability, severe respiratory insufficiency (PaO2 < 50 mm Hg on room air), a small effusion, and severe hemostasis alterations (platelets < 50 G/L, fibrinogen < 2 g/L, prothrombin < 50% of control, or cephalin-activated time more than twice the control).
Clinically documented pneumonia was defined by association of fever, hypoxemia, and pulmonary infiltrate at chest radiography. In all these patients, an invasive (fibroscopy with distal protected specimen) or noninvasive (sputum or tracheal aspiration analysis) microbiological diagnosis of pneumonia was performed.
The cause of pleural effusion was determined based on the physical and radiographic findings (presumptive diagnosis) and on the results of thoracentesis if performed (definitive diagnosis). Clinicians were asked to predict the precise pretap diagnosis (and not only exudate vs transudate) before thoracentesis and their definitive diagnosis with all the results of tests performed on pleural effusion. Exudates were defined by the criteria of Light et al8 : pleural fluid/serum protein ratio > 0.5 and/or pleural fluid/serum LDH ratio > 0.6 and/or pleural fluid LDH greater than two thirds the upper limit of normal for serum LDH (200 IU/L). When the pleural fluid/serum protein ratio was < 0.5 with a pleural fluid/serum LDH ratio > 0.6, the effusion was considered to be a exudate if the serum-pleural fluid albumin ratio was > 1.2.2 9 Three pleural effusion categories were defined based on these criteria: (1) transudative effusion; (2) infectious exudative effusion, ie, empyema or parapneumonic effusion (see definitions below); and (3) noninfectious exudative effusion (exudate with negative bacteriologic culture findings and no evidence of pneumonia). Diagnostic criteria for the causes of pleural effusion were as follows: heart failure (left-sided S3 gallop rhythm; basal crackles; chest radiograph showing cardiomegaly, bilateral alveolar edema, and bilateral effusions of similar size; abnormal echocardiography results; and elevated pulmonary wedge pressure if available)10 ; atelectasis (plate-like changes on the chest radiograph, volume loss, and small ipsilateral effusion); hypoalbuminemia (transudative effusion in patients with serum albumin < 25 g/L); parapneumonic effusion (clinically or microbiologically documented pneumonia with an ipsilateral free-flowing or loculated effusion); empyema (effusion with turbid fluid or pus or a positive pleural fluid Gram’s stain or culture finding)11 ; malignancy (cytologic specimen finding positive for malignant cells)12 ; hemothorax (blood at thoracentesis); and postoperative effusion (effusion occurring after abdominal surgery).13 Pancreatic effusions were defined as pleural effusions associated with clinically and biologically documented pancreatitis with compatible CT findings. Pulmonary emboli were all diagnosed using helicoidal CT. Treatment changes attributed to the results of thoracentesis were recorded; they consisted of introduction of diuretics or vasopressors, adjustment of antibiotics according to microbiological findings, albumin therapy, antitumoral chemotherapy, steroids, and chest tube drainage. In bilateral effusions, only one side was tapped for the purpose of this study. Complications of thoracentesis were also recorded.
Statistical Analysis
Results are reported as medians (quartiles). Comparisons between
groups used the 
2 test for categorical
variables (or Fisher’s Exact Test when appropriate) and the
nonparametric Wilcoxon test for continuous variables (or the
Kruskal-Wallis test when appropriate). The usefulness of thoracentesis
was evaluated separately for the diagnosis and the treatment.
Diagnostic usefulness was defined as a change in the presumptive
diagnosis made by two senior physicians and one junior physician, and
therapeutic usefulness as a change in treatment. Three situations were
identified: (1) certain and (2) probable benefit of thoracentesis
regarding diagnosis assessment, and (3) certain benefit of
thoracentesis regarding change in treatment. In the first and third
situations, thoracentesis directly allowed physicians to perform a new
diagnosis or to begin a new treatment; in the second situation, results
of thoracentesis invalidated the presumptive diagnosis and helped
physicians to reconsider their diagnosis and to identify the accurate
etiology of the pleural effusion. Intraobserver agreement before and
after thoracentesis was assessed using the 
test to compare the same
group of observers before and after thoracentesis; a coefficient
< 0.50 was taken to indicate poor agreement.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patient Characteristics
There were 59 male and 54 female patients (mean age, 59 years;
range, 42 to 68 years; Table 1
). Eight patients (7%) were neutropenic, 8 patients (7%) had
cirrhosis, 17 patients (15%) had documented heart failure, and 35
patients (31%) had hypertension. Respiratory failure was the most
common reason for MICU admission; 68 patients (60%) were receiving
mechanical ventilation at the time of thoracentesis, and 47 patients
(42%) required a positive end-expiratory pressure (PEEP) 
5 cm
H2O; median
PaO2/fraction of inspired oxygen was
186 (quartile range, 94 to 245), and 44 patients (53.7%) had
PaO2/fraction of inspired oxygen
< 200. Pleural effusion was detected at MICU admission in 83 patients
(74%) and > 5 days after MICU admission in 30 patients (26%). Based
on radiographic findings, the effusion was small in 7 patients (6.2%),
moderate in 81 patients (71.7%), and large in 25 patients (22.1%).
|
Patients Without Contraindications to Thoracentesis
The mean time from MICU admission to thoracentesis was 2
days (0 to 6 days). Of the 86 patients who underwent thoracentesis, 4
patients had a dry tap. Fluid was obtained in 82 patients (73%). The
fluid was clear in 39 patients (48%), blood-tinged in 28 patients
(35%), hemorrhagic in 5 patients (6%), cloudy in 5 patients (6%),
and turbid in 5 patients (6%). The effusion was a transudate in 20
patients (24.4%), a noninfectious exudate in 27 patients (32.9%), and
an infectious exudate in 35 patients (42.7%). Most transudates (16 of
20 transudates) were attributable mainly to fluid overload associated
with congestive heart failure; the cause was hypoalbuminemia in four
patients, including two patients with liver failure. Of the 35 patients
with infectious effusions, 21 patients had a parapneumonic effusion and
14 patients had empyema. Causes of the noninfectious exudates were
malignancy (n = 10), pulmonary embolism (n = 4), postoperative
effusion (n = 4); hemothorax (n = 4), and pancreatitis (n = 2);
in three patients, no cause was found. Microorganisms associated with
empyema were Staphylococcus aureus (n = 3), Streptococcus
spp (n = 3), Escherichia coli (n = 2), Klebsiella
pneumoniae (n = 2), Mycobacterium tuberculosis
(n = 2), and Peptostreptococcus (n = 2). In all the 45 patients
with clinically documented pneumonia (fever, hypoxemia, and pulmonary
infiltrate at chest radiography) an invasive (distal protected) or
noninvasive (sputum or tracheal aspiration analysis) microbiological
diagnosis of pneumonia was performed. The results were positive in 25
patients, including 13 patients with parapneumonic pleural effusion.
Microorganisms involved were Pseudomonas aeruginosa
(n = 4), S aureus (n = 3), Streptococcus spp (n = 3),
Streptococcus pneumoniae (n = 2), and Proteus
mirabilis (n = 1).
Comparison of Patients Based on the Cause of Pleural Effusion
Patient characteristics according to thoracentesis findings are
detailed in Table 2
. Clinical features at MICU admission differed significantly between
patients with and without a transudate. In patients with a transudate,
MICU admission for cardiogenic pulmonary edema, bilateral ankle edema
at presentation, and documented heart failure were more common than in
the patients with infectious or noninfectious exudates. Two clinical
features, namely, fever at MICU admission and a unilateral effusion,
were more common in patients with infectious exudates than in
transudates, but failed to distinguish between infectious and
noninfectious exudates. Eight of the 19 patients admitted to the MICU
for cardiogenic pulmonary edema had an exudate, as did 23 of the 40
patients with ankle edema and 4 of the 13 patients with documented
heart failure. These 35 exudates in patients with pulmonary edema,
pedal edema, or documented heart failure included parapneumonic
effusions (n = 15), empyema (n = 2), hemothorax (n = 3),
positional atelectasis (n = 3), postoperative effusions (n = 3),
pancreatitis (n = 2), metastatic effusions (n = 4), pulmonary
embolism (n = 1), and exudates from unknown origin (n = 2).
Moreover, six patients with cardiogenic pulmonary edema had a
unilateral effusion, including parapneumonic effusion (n = 1),
empyema (n = 1), and pulmonary embolism (n = 1). Although patients
with transudates had a higher mean SAPS II score at MICU admission,
they had no significant differences in mortality or ICU length of stay
as compared with the two other groups. No significant differences were
found among the three pleural effusion groups for age, immunologic
status, blood leukocytosis, presence of nonpleural infection, or need
for vasopressors or mechanical ventilation at the time of
thoracentesis.
|
coefficient for intraobserver agreement was 0.41 (0.16 to 0.84),
indicating a marked improvement in the etiologic diagnosis of pleural
effusion after thoracentesis. Nine patients (11%) received a change in
treatment directly attributable to thoracentesis although their
diagnosis remained unchanged: six patients with empyema were switched
to more active antibiotics based on microbiological pleural fluid
studies, two patients with hypoalbuminemia-related transudates received
albumin, and one patient with pleural cancer received antitumoral
chemotherapy based on cytologic pleural fluid results. On the day of
thoracentesis, none of the study parameters predicted whether
thoracentesis would be useful. Patients in whom thoracentesis was
useful had a significantly shorter time from ICU admission to
thoracentesis (1 day [0 to 3 days] vs 5 days [2 to 11 days],
p = 0.0006) and trends toward a shorter ICU stay (10 days [6 to 16
days] vs 14 days [7 to 23 days], p = 0.11) and a lower mortality
rate (30% vs 40%, p = 0.16).
|
5 cm H2O; all of them resolved satisfactorily
after bedside drainage. No cases of thoracentesis-related hemothorax or
pleural infection were seen. |
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although thoracentesis has been reported to be safe, even in patients receiving mechanical ventilation,14 15 16 few studies have focused on its feasibility and clinical implications in MICU patients. The prospective, three-center, 1-year study reported here was designed to evaluate the usefulness of routine thoracentesis in consecutive MICU patients with physical and radiographic evidence of pleural effusion. Thoracentesis resulted in a change in diagnosis and/or treatment in as many as 46 of the 82 patients without of contraindications to the procedure.
Patients who benefited from thoracentesis presented similar outcomes than those with contraindications to thoracentesis or those who did not benefit from thoracentesis. Nevertheless, we believe that mortality could not represent a relevant variable of interest in our study. Indeed, improvement in diagnostic and therapeutic strategies may have other advantages such as better quality of care. Moreover, patients who benefited from thoracentesis had trends toward a shorter ICU stay.
The incidence of pleural effusion during the 1-year study period in the patients of the three MICUs was 8.4%. This relatively low figure may be an underestimation, since effusions were diagnosed based on physical and radiographic data, which can miss small amounts of fluid. Moreover, clinical diagnosis of pleural effusions could be impaired in some ICU patients in whom auscultation and percussion of the chest may be difficult. Other diagnostic tools such as ultrasound or CT can detect small effusions, but the clinical significance of these remains unclear.1 5 Nevertheless, there are no data to suggest that small effusions detected by ultrasounds are qualitatively different from those detected clinically (examination plus radiograph). Therefore, our findings may be applied to patients with small effusions.
In our patients, infection was the most common cause of pleural effusion, with 43% of cases (35 of 82 patients). Three fifths of the infectious effusions were related to pneumonia, and the remainder were empyemas. These results are at variance with previous reports that transudates dominate in MICU patients.1 For instance, Mattison et al 1 found a 62% prevalence of heart failure-related transudates in a study of 62 patients, of whom only 14 patients underwent thoracentesis; however, some of the untapped effusions in their study may have been uncomplicated parapneumonic effusions or other types of exudates. Moreover, published guidelines2 do not recommend thoracentesis in patients known to have left-heart dysfunction associated with pleural effusion unless the effusion is febrile and/or unilateral and/or associated with chest pain. In MICU patients, this restriction of thoracentesis use in patients with known heart failure may result in overestimation of transudates related to heart failure and, consequently, in misdiagnosis of effusions due to other causes.
Diagnosing infectious effusions is important to improve the treatment and prognosis. In our study, routine thoracentesis provided the correct diagnosis in eight cases of pleural empyema for which the presumptive cause was a parapneumonic effusion (n = 4), pulmonary embolism (n = 1), hemothorax (n = 2), or surgery (n = 1). This shows that commonly used parameters, including body temperature and leukocytosis, may fail to indicate empyema, leaving routine thoracentesis as the best diagnostic tool.
Noninfectious exudates were common in our study. Their causes were diverse, with a noticeable proportion of pleural cancers. Here again, thoracentesis can improve the treatment and prognosis, particularly at discharge from the MICU.
To the best of our knowledge, our study is the first to evaluate the diagnostic and therapeutic usefulness of routine thoracentesis in critically ill patients with pleural effusions. Thoracentesis improved the diagnosis and treatment, although the reductions in length of stay and mortality were nonsignificant as compared with the patients who had no benefit from thoracentesis. As previously reported in ICU patients, routine thoracentesis was safe.14 15 16 Given that more than one half our patients underwent thoracentesis and that neither clinical nor laboratory parameters predicted the usefulness of this procedure, we suggest that thoracentesis should be performed routinely in ICU patients with a pleural effusion, above all in patients recently admitted to the ICU. If our findings are confirmed in larger studies, this safe and cheap procedure may provide large gains in diagnosis, treatment, and even prognosis.
|
Footnotes |
|---|
Received for publication December 28, 2000. Accepted for publication July 19, 2001.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Bass and D. A. White Thoracentesis in Patients With Hematologic Malignancy: Yield and Safety Chest, June 1, 2005; 127(6): 2101 - 2105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Wahidi and A. Ernst Role of the Interventional Pulmonologist in the Intensive Care Unit J Intensive Care Med, May 1, 2005; 20(3): 141 - 146. [Abstract] [PDF]
|
|
|
|
|
|
A. Roch, M. Bojan, P. Michelet, F. Romain, F. Bregeon, L. Papazian, and J.-P. Auffray Usefulness of Ultrasonography in Predicting Pleural Effusions > 500 mL in Patients Receiving Mechanical Ventilation Chest, January 1, 2005; 127(1): 224 - 232. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-Y. Tu, W.-H. Hsu, T.-C. Hsia, H.-J. Chen, K.-D. Tsai, C.-W. Hung, and C.-M. Shih Pleural Effusions in Febrile Medical ICU Patients: Chest Ultrasound Study Chest, October 1, 2004; 126(4): 1274 - 1280. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. H. Mayo, H. R. Goltz, M. Tafreshi, and P. Doelken Safety of Ultrasound-Guided Thoracentesis in Patients Receiving Mechanical Ventilation Chest, March 1, 2004; 125(3): 1059 - 1062. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 How Useful Is Thoracentesis in the ICU? Journal Watch (General), February 22, 2002; 2002(222): 3 - 3. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
