0
Clinical Investigations: PLEURA |

Exudative Effusions in Congestive Heart Failure* FREE TO VIEW

Alain A. Eid, MD, FCCP; Jean I. Keddissi, MD; Michel Samaha, MD; Maroun M. Tawk, MD; Kristopher Kimmell, BA; Gary T. Kinasewitz, MD, FCCP
Author and Funding Information

*From the Departments of Pulmonary and Critical Care Medicine (Drs. Eid, Keddissi, Samaha, and Tawk, and Mr. Kimmell) and Physiology and Biophysics (Dr. Kinasewitz), University of Oklahoma Health Sciences Center, Oklahoma City, OK.

Correspondence to: Alain A. Eid, MD, FCCP, Assistant Professor of Medicine, 920 Stanton Young Blvd, WP 1310, Oklahoma City, OK 73104; e-mail: Hammoun@aol.com



Chest. 2002;122(5):1518-1523. doi:10.1378/chest.122.5.1518
Text Size: A A A
Published online

Objectives: Pleural effusions due to congestive heart failure (CHF) typically are transudates, but an occasional patient with CHF is found to have an exudate in the absence of an apparent cause other than CHF. We sought to determine the incidence and clinical significance of such exudative effusions.

Design: Patients with CHF and effusions seen during the 7-year period from January 1994 through December 2000 were identified from their hospital discharge diagnoses and radiographs, while those who had undergone thoracentesis were identified from a review of the laboratory logs. The presenting symptoms and clinical course were determined from a review of the medical records. The effect of RBC contamination on pleural fluid lactate dehydrogenase (LDH) levels was determined by measuring the LDH activity of mock pleural fluid containing known amounts of RBC.

Results: Seven hundred seventy patients had CHF with an effusion, but only 175 patients underwent a thoracentesis. In this select group, 86 patients had transudates and 89 had exudates. A noncardiac cause for the exudate was readily identified in 59 patients by hospital discharge, and 7 more patients had an etiology found during follow-up. Eleven of the remaining 23 patients had undergone coronary artery bypass graft (CABG) surgery ≥ 1 year prior to presentation, and 50% of the effusions in patients who had undergone CABG surgery were exudates. Thus, CHF-related exudates were identified in only 12 patients, and in 4 of these patients the exudates could be explained by RBC contamination of the pleural fluid. The clinical presentation of patients with CHF-associated exudates was similar to that of CHF patients with transudates.

Conclusion: In most patients who have CHF and an exudative effusion, there is a noncardiac cause for the pleural effusion. The high frequency of exudates in patients with a history of CABG indicates a persistent impairment in lymphatic clearance from the pleural cavity. Exudative effusions due solely to CHF are rare.

Figures in this Article

Congestive heart failure (CHF) is the most frequent cause of pleural effusions.1 The typical pleural effusion associated with heart failure has the characteristics of a transudate and resolves when the underlying cardiac problem is treated. CHF is a common clinical condition, so it is not unusual to encounter a patient with heart failure who has an exudative pleural effusion due to a coexisting problem such as pneumonia. However, occasionally a patient with CHF presents with an exudative effusion in the absence of an apparent cause other than heart failure. The significance and natural history of these CHF-associated exudative effusions are unknown.

To address these questions, we identified patients seen at our medical center during a 7-year period with exudative pleural effusions who were discharged home with the diagnosis of CHF as the etiology of their effusion.

All patients with CHF who had a thoracentesis performed at the Oklahoma City Veterans Affairs Medical Center from January 1994 through December 2000 were included in the study. Patients with heart failure were identified from hospital discharge diagnoses, and the presence of a pleural effusion was determined from the chest radiograph. Patients with blunting of the costophrenic angle or a meniscus on either the posteroanterior or lateral chest radiograph were considered to have a pleural effusion. The diagnosis of CHF was established by clinical criteria (presentation and response to therapy) and was verified in most patients by the results of an echocardiogram, cardiac catheterization, and/or nuclear ventriculogram. Those who underwent a thoracentesis were identified by review of laboratory logs.

The following data were collected on the pleural fluids: pH; Pco2; lactate dehydrogenase (LDH) level; total protein level; cell count; RBC count; and microbiological and cytologic results. Simultaneous plasma LDH and protein levels were noted. The upper limit of normal for serum LDH is 200 U/L in our laboratory. When more than one thoracentesis was performed on a patient, only data from the first one were included in the analysis. Effusions were classified as exudative if they met any of the following criteria: (1) pleural fluid-to-serum (PF/S) protein ratio of > 0.5; (2) PF/S LDH ratio of > 0.6; or (3) pleural fluid LDH level of > 133 U/L (two thirds of the upper limit of normal).2

The clinical presentation, medical history, and etiology of the effusion was determined from a review of the medical records. The in-hospital course and the response to therapy were noted. The etiology of the effusion was based on clinical impression and response to therapy. Transudative effusions that improved with diuretic therapy were attributed to heart failure. Exudative effusions that occurred in an appropriate clinical setting (eg, in association with pneumonia) were attributed to the recognized underlying cause. Patients who had undergone thoracotomy or coronary artery bypass graft (CABG) surgery within 3 months of undergoing the thoracentesis were excluded from the study. The diagnosis of an exudative pleural effusion due to heart failure was accepted if no other cause for the effusion was identified during the index hospitalization. For those patients who were discharged from the hospital with this diagnosis, follow-up included a review of subsequent clinical records, radiographic studies, and laboratory data.

To determine the effect of RBC contamination on the pleural LDH measurements, varying numbers of RBCs were added to either normal saline solution or mock pleural fluid that was made by adding one part plasma to two parts saline solution. The LDH activity of these pseudoeffusions was determined on the same autoanalyzer (Dade RxL; Dade Behring, Inc; Newark, DE) that was used to measure clinical samples.

Statistical Analysis

All data are presented as mean ± SEM. The significance of differences between groups was determined by analysis of variance or χ2 analysis for nonparametric data. The relationship between the RBC count and the LDH activity of the mock effusions was determined by least-squares linear regression.

Results

A total of 770 patients who were discharged from the hospital with a diagnosis of CHF between January 1994 and December 2000 had pleural effusions seen on their hospital admission chest radiographs. The majority of these patients (595 [77%]) were treated medically, and their effusions were not tapped. In the remaining 175 patients, a thoracentesis was performed (Fig 1 ). In the latter group of patients, the incidence of transudates and exudates was almost equal. Transudates attributed to heart failure were found in 86 patients, and 89 patients had exudative effusions. Of those 89 patients, 59 had an obvious etiology that was readily recognized and treated during the hospitalization. Infection, malignancy, and uremic pleuritis accounted for most of these effusions.

Thirty patients were discharged to home with the diagnosis of a CHF-associated exudative pleural effusion. In 7 of the 30 patients with an exudative effusion attributed to CHF, an alternative etiology for the exudate was identified within 2 years of follow-up. One patient was readmitted to the hospital within days and an underlying pneumonia was recognized. His infiltrates had failed to respond to therapy targeting CHF, but both the infiltrates in the lung and the associated effusion responded to antibiotic therapy. A second patient was readmitted to the hospital shortly after being discharged with spontaneous bacterial peritonitis in the setting of hepatic cirrhosis with portal hypertension and ascites. Two effusions were subsequently attributed to trauma when follow-up radiographs revealed new callus formation on ribs in the affected hemithorax. Bronchogenic carcinoma was later diagnosed in one patient, and autoimmune diseases (scleroderma, antineutrophil cytoplasmic antibody-positive nephrotic syndrome) were subsequently diagnosed in two patients.

Thus, only 23 patients who had received diagnoses of CHF-related pleural effusion had an exudate and no other clear etiology, despite careful and prolonged follow-up.

Effect of Previous CABG

Eleven of 23 patients (47%) who were discharged from the hospital with a diagnosis of an exudative effusion related to heart failure had undergone CABG surgery 1 to 25 years prior to presentation with an effusion. In contrast, only 11 of 86 patients (13%) with transudative effusions had a history of CABG surgery. Conversely, 11 of 22 patients (50%) with CHF and a history of CABG surgery had exudative effusions, while 12 of 87 patients (14%) with CHF but no history of CABG surgery had exudative effusions. Thus, patients with a history of CABG surgery were significantly more likely to have exudative effusions than were heart failure patients who had not previously undergone CABG surgery (p < 0.001). The specific criteria for an exudate met by patients with and without a history of CABG surgery are summarized in Table 1 .

Effect of Blood Contamination

Nine of the 12 patients who had not undergone CABG surgery who had an exudative effusion had a pleural fluid RBC count of > 5,000 cells/μL, suggesting possible trauma during the thoracentesis. Seven of these patients had an exudative effusion based solely on LDH criteria. There was a significant correlation between the pleural fluid RBC count and the pleural fluid LDH level (r = 0.77) in these patients (Fig 2 ).

To determine whether RBCs present as a result of a traumatic tap could elevate the LDH level sufficiently to produce a pseudoexudate, we measured the LDH level of mock pleural fluid to which we added variable amounts of RBCs. There was a linear relationship between the RBC count and LDH activity of the mock pleural fluid (r = 0.99; p < 0.01) [Fig 2]. The slope of the relationship was 1.2 LDH IU/L per 1,000 RBCs. The LDH value of the pleural fluid in these patients was corrected according to the formula:

After this correction, the exudates in four patients with no history of CABG could be accounted for solely by RBC contamination during the thoracentesis.

Clinical Characteristics

The clinical presentation was similar in patients with transudates and those with exudates. Dyspnea was the most common presenting symptom, occurring in 70% of those with transudates and 50% of those with exudates. Pleuritic chest pain was noted in 30% of those with transudates and 17% of those with exudates. Only 10% of the transudates and 17% of the exudates were asymptomatic radiographic findings.

By definition, the protein concentration and LDH values of patients with exudative effusions were significantly greater than those for patients with transudative effusions (Table 2 ). The cell count and RBC content of fluid from patients with CHF-related exudates were higher than those from patients with transudates or CABG-related exudative effusions. Patients with transudates were more likely to have bilateral effusions. The ejection fraction was determined by echocardiography in approximately half the patients during the hospital admission. Those with CABG-related exudates had the highest ejection fraction, although the differences were not significant because of the small numbers. Although there was a tendency toward longer survival in the post-CABG group than in the other groups, it did not reach statistical significance (most likely because of the small number of patients).

Diuretic Therapy

All of the patients with exudates related to heart failure had received furosemide for several days before thoracentesis. The average routine daily dose of furosemide during the week prior to hospital admission was 106 ± 28, 34 ± 7, and 60 ± 27 mg, respectively, in the exudative CHF, transudative CHF, and prior CABG groups. The difference in dose between the transudative CHF and the exudative CHF groups was statistically significant (p < 0.05). For most patients with CHF, diuretic therapy was increased on admission to the hospital, but those with CHF exudates still received significantly more furosemide (Table 2).

We found that exudative pleural effusions were common in patients with CHF, accounting for slightly more than half of the pleural effusions in those patients who underwent thoracentesis. However, patients undergoing this procedure were a select group as 77% of the patients identified with CHF and a pleural effusion did not undergo thoracentesis. Presumably, their physician was comfortable with the diagnosis of an effusion related to heart failure on the basis of the patient’s clinical presentation and response to therapy. In most patients with an exudative effusion, an etiology for the exudative effusion was readily identified when thoracentesis was performed. Ultimately, only 23 patients were believed to have a CHF-associated exudative pleural effusion, and 12 of these individuals had previously undergone CABG surgery.

Pleural effusions are extremely common after CABG surgery, occurring in > 80% of patients in the immediate postoperative period.3These effusions appear to be more common after patients undergo internal mammary artery grafting and topical cardiac hypothermia, and generally resolve over a several-week period.46 Yet, occasionally these effusions may persist and, if symptomatic, require decortication to free the underlying lung.7Exudative effusions that appear early after CABG are often bloody with a predominance of polymorphonuclear leukocytes and eosinophils, whereas those that appear later tend to have a predominance of small lymphocytes.89 Lymphocytes accounted for 53% of the nucleated cells in our post-CABG patients. It has been suggested that impaired reabsorption of protein by lymphatics injured during surgery may contribute to the high protein concentration and exudative characteristics of these effusions.9We speculate that mesothelial cell-induced fibrin deposition may block the lacunae through which protein is reabsorbed from the pleural cavity.10 Our data suggest that this problem may persist, as 50% of the effusions in patients who had undergone CABG > 1 year earlier were exudative effusions, vs 13% in patients with no history of chest surgery.

Most of the CHF-associated exudates that we identified were classified as such because of an elevated LDH level. Even a small amount of blood contamination may be sufficient to create a pseudoexudate if only a small amount of fluid is removed during a diagnostic thoracentesis. For example, 50 μL whole blood with an RBC count of 4 million cells/μL in 10 mL pleural fluid (1 part per 200) would raise the LDH value by 24 IU/L and might convert a transudate with borderline values into an exudate. Correcting the LDH by subtracting 1.2 LDH IU/L per 1,000 RBCs/μL would prevent misclassification of effusions with borderline values. Four patients in the present study were found to have pseudoexudates when the effect of RBC contamination was taken into consideration.

Diuretic therapy may also accelerate the reabsorption of water and increase the protein and LDH concentrations of an effusion, thereby converting a transudate into an exudate.1113 All of our patients were receiving diuretic medications at the time of their thoracentesis, and the dose was higher in the CHF patients with exudates. This therapy may have been responsible for the exudative nature of the effusion in some patients.

Interestingly, the clinical presentation of those with CHF-associated exudates was similar to that of patients with transudative effusions related to CHF. There was no difference between patients with transudates and those with exudates in terms of presenting symptoms, radiographic size of the effusion, or ejection fraction. Most importantly, there was no difference in outcome. The survival time of those with exudative effusions was similar to that of patients with transudates. Our findings are consistent with the recent report by Gotsman and colleagues14 that the clinical presentation and outcome of CHF patients with unexplained exudates are similar to those of patients with transudates.

Pulmonary embolism is a potential cause of an exudative effusion in patients with CHF. However, the clinical suspicion of thromboembolism was low in our patients with exudative effusions. Although patients were not systematically evaluated to exclude this possibility, only two patients were discharged from the hospital receiving anticoagulant therapy for cardiac disease, and none had evidence of deep venous thrombosis on follow-up.

In summary, we found that most patients with heart failure and an exudative effusion have an underlying cause for their exudate. A history of CABG may predispose patients to the development of an exudative effusion, even years after the surgery. When the potential confounding effect of increased LDH levels due to RBC contamination is corrected, unexplained exudative pleural effusions are rare in patients with CHF.

Abbreviations: CABG = coronary artery bypass graft; CHF = congestive heart failure; LDH = lactate dehydrogenase; PF/S = pleural fluid-to-serum ratio

Figure Jump LinkFigure 1. Distribution of the different categories of patients with CHF and a pleural effusion.Grahic Jump Location
Table Graphic Jump Location
Table 1. Number of CHF Patients With and Without a History of CABG Meeting Exudative Criteria
Figure Jump LinkFigure 2. Top: relationship between measured (•) and corrected (○) PF/S LDH ratios for patients with CHF and exudative effusions who had not undergone CABG. Bottom: relationship between RBC concentration and LDH activity of pleural fluid in patients with CHF and unexplained exudates who had not previously undergone CABG. Solid line is least-squares regression fitted to LDH activity of mock effusions containing known RBC concentrations (▴). The actual(•) and corrected (○) LDH values are shown.Grahic Jump Location
Table Graphic Jump Location
Table 2. Clinical and Laboratory Data in CHF Patients With Transudative and Exudative Effusions*
* 

Data expressed as mean (SEM) except where noted. PMN = polymorphonuclear leukocyte.

 

p < 0.05 between transudate and CHF exudate.

 

p < 0.01 between transudate and CHF exudate.

§ 

p < 0.001 between CHF exudates/CABG and transudates.

We thank Priscilla Peer for her excellent assistance in the preparation of this manuscript.

Marel, M, Zrustova, M, Stastny, B, et al (1993) The incidence of pleural effusion in a well-defined region: epidemiologic study in central Bohemia.Chest104,1486-1489. [PubMed] [CrossRef]
 
Light, RW Clinical manifestations and useful tests. Retford, DC eds.Pleural diseases 3rd ed.1995,36-74 WIlliams & Wilkins. Baltimore, MD:
 
Vargas, FS, Cukier, A, Hueb, W, et al Relationship between pleural effusion and pericardial involvement after myocardial revascularization.Chest1994;105,1748-1752. [PubMed]
 
Kollef, MH, Peller, T, Knodel, A, et al Delayed pleuropulmonary complications following coronary artery revascularization with the internal mammary artery.Chest1988;94,68-71. [PubMed]
 
Hurlbut, D, Myers, ML, Lefcoe, M, et al Pleuropulmonary morbidity: internal thoracic artery vs saphenous vein graft.Ann Thorac Surg1990;50,959-964. [PubMed]
 
Allen, BS, Buckberg, GD, Rosenkranz, ER, et al Topical cardiac hypothermia in patients with coronary disease.J Thorac Cardiovasc Surg1992;104,626-631. [PubMed]
 
Lee, YC, Vaz, MAC, Ely, KA, et al Symptomatic persistent post-coronary artery bypass graft pleural effusions requiring operative treatment.Chest2001;119,795-800. [PubMed]
 
Miro, AM, Vasudevan, V, Shah, H Ciliary motility in two patients with yellow nail syndrome and recurrent sinopulmonary infections.Am Rev Respir Dis1990;142,890-891. [PubMed]
 
Light, RW, Rogers, JT, Cheng, D-S, et al Large pleural effusions occurring after coronary artery bypass grafting: Cardiovascular Surgery Associates, PC.Ann Intern Med1999;130,891-896. [PubMed]
 
Bottles, KD, Laszik, Z, Morrissey, JH, et al Tissue factor expression in mesothelial cells: induction bothin vivoandin vitro.Am J Respir Cell Mol Biol1997;17,164-172. [PubMed]
 
Chakko, SC, Caldwell, SH, Sforza, PP Treatment of congestive heart failure: its effect on pleural fluid chemistry.Chest1989;95,798-802. [PubMed]
 
Shinto, RA, Light, RW Effects of diuresis on the characteristics of pleural fluid in patients with congestive heart failure.Am J Med1990;88,230-234. [PubMed]
 
Romero-Candeira, S, Fernandez, C, Martin, C, et al Influence of diuretics on the concentration of proteins and other components of pleural transudates in patients with heart failure.Am J Med2001;110,681-686. [PubMed]
 
Gotsman, I, Fridlender, Z, Meirovitz, A, et al The evaluation of pleural effusions in patients with heart failure.Am J Med2001;111,375-378. [PubMed]
 

Figures

Figure Jump LinkFigure 1. Distribution of the different categories of patients with CHF and a pleural effusion.Grahic Jump Location
Figure Jump LinkFigure 2. Top: relationship between measured (•) and corrected (○) PF/S LDH ratios for patients with CHF and exudative effusions who had not undergone CABG. Bottom: relationship between RBC concentration and LDH activity of pleural fluid in patients with CHF and unexplained exudates who had not previously undergone CABG. Solid line is least-squares regression fitted to LDH activity of mock effusions containing known RBC concentrations (▴). The actual(•) and corrected (○) LDH values are shown.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1. Number of CHF Patients With and Without a History of CABG Meeting Exudative Criteria
Table Graphic Jump Location
Table 2. Clinical and Laboratory Data in CHF Patients With Transudative and Exudative Effusions*
* 

Data expressed as mean (SEM) except where noted. PMN = polymorphonuclear leukocyte.

 

p < 0.05 between transudate and CHF exudate.

 

p < 0.01 between transudate and CHF exudate.

§ 

p < 0.001 between CHF exudates/CABG and transudates.

References

Marel, M, Zrustova, M, Stastny, B, et al (1993) The incidence of pleural effusion in a well-defined region: epidemiologic study in central Bohemia.Chest104,1486-1489. [PubMed] [CrossRef]
 
Light, RW Clinical manifestations and useful tests. Retford, DC eds.Pleural diseases 3rd ed.1995,36-74 WIlliams & Wilkins. Baltimore, MD:
 
Vargas, FS, Cukier, A, Hueb, W, et al Relationship between pleural effusion and pericardial involvement after myocardial revascularization.Chest1994;105,1748-1752. [PubMed]
 
Kollef, MH, Peller, T, Knodel, A, et al Delayed pleuropulmonary complications following coronary artery revascularization with the internal mammary artery.Chest1988;94,68-71. [PubMed]
 
Hurlbut, D, Myers, ML, Lefcoe, M, et al Pleuropulmonary morbidity: internal thoracic artery vs saphenous vein graft.Ann Thorac Surg1990;50,959-964. [PubMed]
 
Allen, BS, Buckberg, GD, Rosenkranz, ER, et al Topical cardiac hypothermia in patients with coronary disease.J Thorac Cardiovasc Surg1992;104,626-631. [PubMed]
 
Lee, YC, Vaz, MAC, Ely, KA, et al Symptomatic persistent post-coronary artery bypass graft pleural effusions requiring operative treatment.Chest2001;119,795-800. [PubMed]
 
Miro, AM, Vasudevan, V, Shah, H Ciliary motility in two patients with yellow nail syndrome and recurrent sinopulmonary infections.Am Rev Respir Dis1990;142,890-891. [PubMed]
 
Light, RW, Rogers, JT, Cheng, D-S, et al Large pleural effusions occurring after coronary artery bypass grafting: Cardiovascular Surgery Associates, PC.Ann Intern Med1999;130,891-896. [PubMed]
 
Bottles, KD, Laszik, Z, Morrissey, JH, et al Tissue factor expression in mesothelial cells: induction bothin vivoandin vitro.Am J Respir Cell Mol Biol1997;17,164-172. [PubMed]
 
Chakko, SC, Caldwell, SH, Sforza, PP Treatment of congestive heart failure: its effect on pleural fluid chemistry.Chest1989;95,798-802. [PubMed]
 
Shinto, RA, Light, RW Effects of diuresis on the characteristics of pleural fluid in patients with congestive heart failure.Am J Med1990;88,230-234. [PubMed]
 
Romero-Candeira, S, Fernandez, C, Martin, C, et al Influence of diuretics on the concentration of proteins and other components of pleural transudates in patients with heart failure.Am J Med2001;110,681-686. [PubMed]
 
Gotsman, I, Fridlender, Z, Meirovitz, A, et al The evaluation of pleural effusions in patients with heart failure.Am J Med2001;111,375-378. [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
Guidelines
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543