- Email: [email protected]
Frequency of Pleural Effusions in Patients With Pulmonary Arterial Hypertension Associated With Connective Tissue Diseases
CHEST
Original Research PULMONARY VASCULAR DISEASE
Frequency of Pleural Effusions in Patients With Pulmonary Arterial Hypertension Associated With Connective Tissue Diseases Yi-feng Luo, MD; Ivan M. Robbins, MD; Mevlut Karatas, MD; Anupama G. Brixey, MD; Todd W. Rice, MD, FCCP; and Richard W. Light, MD, FCCP
Background: Pleural effusions frequently accumulate in patients with left-sided heart failure. However, our recent study in patients with idiopathic and heritable pulmonary arterial hypertension (PAH) demonstrated that pleural effusions frequently occur in patients with isolated rightsided heart failure (RHF). The objective of this study was to determine the frequency of pleural effusions in patients with PAH associated with connective tissue disease (CTD). Methods: We retrospectively studied consecutive patients with PAH associated with CTD who were treated in the Vanderbilt Pulmonary Vascular Center. Pleural effusions were identified by chest radiograph, chest CT scan, thoracic ultrasonography, or autopsy. Results: Thirty-five of 89 patients (39.3%) with PAH associated with CTD had pleural effusions: 23 of 51 (45.1%) with scleroderma, six of 16 (37.5%) with systemic lupus erythematosus, five of 18 (27.8%) with mixed connective tissue disease, and one of two (50.0%) with Sjögren syndrome. There were alternative explanations for the pleural effusions in six of these patients. Of the 29 patients without alternative explanation for their pleural effusions, 28 had RHF. When compared with the patients without pleural effusions, the 29 patients with pleural effusions had significantly higher mean right atrial pressures (11.3 ⫾ 5.1 mm Hg vs 8.3 ⫾ 4.0 mm Hg, P 5 .004) and lower cardiac indices (2.1 ⫾ 0.6 L/min/m2 vs 2.5 ⫾ 0.7 L/min/m2, P 5 .011). The pleural effusions were predominantly trace to small (58.6%) in size and bilateral (51.7%) in distribution. Conclusions: Pleural effusions frequently accumulate in patients with PAH associated with CTD and are associated with RHF. CHEST 2011; 140(1):42–47 Abbreviations: BNP 5 brain natriuretic peptide; CTD 5 connective tissue disease; CXR 5 chest radiograph; ECHO 5 echocardiography; LHF 5 left-sided heart failure; MCTD 5 mixed connective tissue disease; mPAP 5 mean pulmonary arterial pressure; mRAP 5 mean right atrial pressure; PAH 5 pulmonary arterial hypertension; PVR 5 pulmonary vascular resistance; PWP 5 pulmonary wedge pressure; RA 5 rheumatoid arthritis; RAP 5 right atrial pressure; RHC 5 right-sided heart catheterization; RHF 5 right-sided heart failure; SLE 5 systemic lupus erythematosus
is generally accepted that the accumulation of Itpleural fluid in patients with congestive heart fail-
ure is due to left-sided heart failure (LHF) rather than right-sided heart failure (RHF).1,2 However, we have recently reported that pleural effusions are common in patients with idiopathic pulmonary arterial hypertension (PAH) or heritable PAH. The majority of patients in that study with pleural effusions had RHF.3 The purpose of the present study was to determine the frequency of pleural effusions in patients with PAH associated with connective tissue disease (CTD). We hypothesized that the overall frequency of pleural effusions would be higher in the patients with
CTD than in those with idiopathic or heritable PAH, because some effusions may be due to the CTD itself. We also hypothesized that effusions would be more common in patients with RHF. Materials and Methods Study Procedures We retrospectively studied the medical records of consecutive patients seen in the Vanderbilt Pulmonary Vascular Center from September 1996 to August 2009 with a diagnosis of PAH associated with CTD. All patients were seen by rheumatologists, and
42
Original Research
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
the diagnosis of CTD was established using standard criteria. Patients were diagnosed as having PAH associated with CTD according to the Dana Point clinical classification of PAH.4 Although some patients were evaluated prior to the publication of the Dana Point criteria for classification of PAH, all patients included in this study underwent a thorough evaluation to exclude other possible causes of PAH. All patients underwent right-sided heart catheterization (RHC) for the initial diagnosis. PAH associated with CTD was defined as a mean resting pulmonary arterial pressure . 25 mm Hg, a pulmonary wedge pressure (PWP) ⱕ 15 mm Hg, and the absence of other possible causes of PAH, such as PAH associated with congenital systemic to pulmonary shunts, portal hypertension, HIV infection, pulmonary hypertension secondary to left-sided heart disease (including diastolic dysfunction and valvular disease), and chronic thromboembolic PAH.4,5 RHF was defined as being present when there was evidence on physical examination (jugular venous pressure . 8 cm H2O, hepatojugular reflux, hepatomegaly, right-sided S3, ascites, anasarca, or worsening lower extremity edema), moderate or severe dilation of the right atrium or right ventricle with moderate or severe impairment of right ventricular function and a noncollapsing inferior vena cava on echocardiography (ECHO), or an elevated right atrial pressure ( . 10 mm Hg) on RHC.3,6,7 The study protocol was approved by the Vanderbilt University Institutional Review Board. The requirement for informed consent was waived because of the retrospective nature of the study. For each patient, the clinical features, laboratory data, radiographic data, and the presence or absence, size, and distribution of pleural fluid were reviewed. In patients who had a thoracentesis, the laboratory results and analysis of the pleural fluid were recorded. The presence of a pericardial effusion or ascites, the total serum protein and albumin levels, as well as results of RHC, ECHO, and thoracic ultrasonography that were obtained closest in time to when the pleural effusion was noted were recorded. The chest radiograph (CXR), chest CT scan, thoracic ultrasonography, and autopsy results were reviewed by our team for the presence of a pleural effusion. The size of the pleural effusion on the posteroanterior and lateral CXRs was semiquantitated using the following scale8: 0 5 no pleural fluid present; 1 5 blunting of the costophrenic angle only; 2 5 , 25% of hemithorax occupied by pleural fluid; 3 5 25% to 50% of hemithorax occupied; 4 5 50% to 75% of hemithorax occupied; 5 5 . 75% of hemithorax occupied. The size of the pleural effusion on CT scan or ultrasound in the supine position was semiquantitated by measuring the maximum thickness of fluid between the visceral and parietal pleurae. For patients who underwent autopsy, the volume of pleural fluid was recorded. Based on these findings, the pleural Manuscript received January 26, 2010; revision accepted December 11, 2010. Affiliations: From the Department of Pulmonary Medicine (Dr Luo), The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, Guangdong, China; the Division of Allergy, Pulmonary and Critical Care Medicine (Drs Luo, Robbins, Karatas, Brixey, Rice, and Light), Vanderbilt University Medical Center, Nashville, TN; and the Department of Pulmonary Medicine (Dr Karatas), Ahi Evren Thoracic Heart and Cardiovascular Surgery Training and Research Hospital, Trabzon, Turkey. Correspondence to: Yi-feng Luo, MD, Department of Pulmonary Medicine, The First Affiliated Hospital of Sun Yat-sen University, 58 Zhongshan 2nd Rd, Guangzhou, Guangdong Province, 510080, China; e-mail: [email protected] Funding/Support: The authors have reported to CHEST that no funding was received for this study. © 2011 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (http://www.chestpubs.org/ site/misc/reprints.xhtml). DOI: 10.1378/chest.10-0227 www.chestpubs.org
THE BOTTOM LINE How does this work advance the field? It is generally accepted that pleural effusions frequently accumulate in patients with left-sided heart failure. However, this study suggests that pleural effusions may frequently occur in patients with isolated right-sided heart failure.
What are the clinical implications? Pleural effusion is a common complication not only in patients with left-sided heart failure but may also frequently occur in patients with right-sided heart failure.
effusion was placed in one of four categories. A patient with more than one study was placed in the highest category achieved on any of the studies: 1. Trace pleural effusion: CXR score 5 1, or fluid thickness , 5 mm on CT scan or ultrasonography, or fluid volume measured at autopsy , 125 mL. 2. Small pleural effusion: CXR score 5 2, or fluid thickness of 5 to 20 mm on CT scan or ultrasonography, or fluid volume measured at autopsy between 125 mL and 500 mL. 3. Moderate pleural effusion: CXR score 5 3, or fluid thickness of 21 to 50 mm on CT scan or ultrasonography, or fluid volume measured at autopsy between 500 mL and 1,250 mL. 4. Massive pleural effusion: CXR score 5 4 or 5, or fluid thickness . 50 mm on CT scan or ultrasonography, or fluid volume measured at autopsy . 1,250 mL. For patients who had received a thoracentesis, Light’s criteria were used to separate transudative from exudative pleural effusions.9 Statistical Analysis Statistical analysis was performed using SigmaStat 2.03 for Windows software (Systat Software, Inc; Richmond, California) and R software 2.11.1. The data are presented as mean ⫾ SD when normally distributed and as median with interquartile ranges in parentheses when not. Independent samples t test was used to compare the means between two groups that were normally distributed. The Mann-Whitney U test was used to assess the equivalence of two distributions from populations that were not normally distributed or that did not have equal variances. This test evaluates whether the observations from one population have a tendency that is larger (or smaller) than the observation from the other population. x2 and Fisher exact tests were used to assess differences in categorical variables between groups. We did not make corrections for multiple comparisons. We calculated the partial correlation coefficient with Pearson correlation test to analyze the relationship among hemodynamics (mean right atrial pressure [mRAP], cardiac index, and pulmonary vascular resistance [PVR]). Ordered logit regression was used to study the effect of the hemodynamics (mRAP, mean pulmonary arterial pressure [mPAP], cardiac index, and PVR) on the size of the effusion. All P values reported were two-tailed and P values , .05 were considered statistically significant.
Results Ninety-nine patients with PAH associated with CTD were evaluated at our pulmonary vascular center CHEST / 140 / 1 / JULY, 2011
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
43
during the study period. Ten patients were excluded from the study: Four patients had diastolic dysfunction, two patients had LHF, two patients never had an RHC, one patient had valvular heart disease, and one patient had chronic thromboembolic disease. The remaining 89 patients, 80 women and nine men, were included in this study. The breakdown of the type of CTD in our patients was as follows: scleroderma, 51; mixed connective tissue disease (MCTD), 18; systemic lupus erythematosus (SLE), 16; Sjögren syndrome, two; polymyositis, one; and rheumatoid arthritis (RA), one. No patients underwent lung transplantation. Overall, 35 of the 89 patients (39.3%) with PAH associated with CTD had pleural effusions during follow-up: 23 of 51 (45.1%) with scleroderma, six of 16 (37.5%) with SLE, five of 18 (27.8%) with MCTD, and one of two (50%) with Sjögren syndrome. There was no significant difference in the frequency of effusions in the different categories (P 5 0.617). Six of the 35 patients with pleural effusions had an alternative explanation for their effusions. Four had radiologic evidence of pneumonia, one had severe hypoalbuminemia, and one had an exudative pleural effusion due to the CTD itself. This latter patient
was a 70-year-old white woman with MCTD, massive bilateral pleural effusions, and concomitant ascites and pericardial effusions. She underwent a thoracentesis at an outside hospital and reportedly had an exudative pleural effusion (the results of the pleural effusion analysis were unavailable). She had Raynaud phenomenon and a positive antinuclear antibody level of 1:320. All patients had RHC and almost all had ECHOs. Of the 29 patients without an alternative explanation for their pleural effusion, 27 had an ECHO, whereas of the remaining 54 patients, 50 had an ECHO. Of the 29 patients without an alternative explanation for their pleural effusions, 28 were diagnosed as having RHF: 25 patients had evidence of RHF on RHC and/or ECHO, and three patients had obvious evidence of RHF on physical examination. The one patient without RHF had trace pleural effusions and had no obvious symptoms or signs of RHF at the time the effusions were identified. Of the 28 patients with RHF and pleural effusions, 20 had RHC or ECHO performed within 1 month of the time the pleural effusion was noted (13 had both RHC and ECHO, three had RHC only, four had ECHO only).
Table 1—Demographics and Clinical Characteristics of Patients With PAH Associated With CTD With or Without Pleural Effusions Patients Without Pleural Effusions (n 5 54)
Variables Age at diagnosis of PAH associated with CTD, y Female (male) Classification of CTD Scleroderma SLE MCTD Sjögren syndrome Polymyositis RA Median follow-up period, mo Time of follow-up until the patient first developed a pleural effusion, mo Death during follow-up Time from PAH diagnosis to death, mo Occurrence of RHF Occurrence of ascitesa Occurrence of pericardial effusionsa Treatmentb Receiving prostacyclin Not receiving prostacyclin Serum total protein, g/dL Serum albumin, g/dL Serum BNP, pg/mL
Patients With Pleural Effusions (n 5 29)
P Value
51.0 ⫾ 13.3
51.0 ⫾ 14.0
.999
49 (5)
25 (4)
.713 .615
28 (51.9) 10 (18.5) 13 (24.1) 1 (1.9) 1 (1.9) 1 (1.9) 16 (5, 47) … 20 (37.0) 22.5 ⫾ 17.2 (n 5 20) 32 (59.3) 1 (1.9) 23 (42.6) 28 (51.9) 26 (48.1) 7.1 ⫾ 0.9 3.8 ⫾ 0.6 301 (99, 748) (n 5 31)
20 (69.0) 5 (17.2) 4 (13.8) 0 (0) 0 (0) 0 (0) 27 (15, 72) 17.7 ⫾ 23.0 19 (65.5) 37.5 ⫾ 28.6 (n 5 19) 28 (96.6) 9 (31.0) 21 (72.4) 10 (34.5) 19 (65.5) 7.0 ⫾ 0.9 3.6 ⫾ 0.6 654 (398, 1,188) (n 5 13)
.032
.025 .053 , .001 , .001 .018 .199
.817 .282 .037
Unless otherwise indicated, data are presented as mean ⫾ SD, median (interquartile range), or No. (%). BNP 5 brain natriuretic peptide; CTD 5 connective tissue disease; MCTD 5 mixed connective tissue disease; PAH 5 pulmonary arterial hypertension; RA 5 rheumatoid arthritis; RHF 5 right-sided heart failure; SLE 5 systemic lupus erythematosus. aThe occurrence of ascites or pericardial effusions was assessed for the entire follow-up period. bThe patients listed as having pleural effusions and receiving prostacyclin all developed effusions after initiation of prostacyclin. Patients who developed effusions prior to initiation of prostacyclin were included in the group “Not receiving prostacyclin.” 44
Original Research
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
When the clinical characteristics of the 54 patients without pleural effusions and the 29 patients with pleural effusions were compared (Table 1), age at diagnosis and the proportion of the patients in the different CTD categories with pleural effusions did not differ significantly. In patients with pleural effusions, the follow-up period was significantly longer, and the mortality was significantly higher as was the occurrence of RHF, ascites, and pericardial effusions. Brain natriuretic peptide (BNP) levels were available in 13 of 29 patients with pleural effusions and 31 of 54 patients without pleural effusions. The median BNP level (654 pg/mL) was significantly higher in the patients with pleural effusions than the level (301 pg/mL) in the patients without pleural effusions (P 5 .037). The frequency of pleural effusions was not significantly different between the patients who received and did not receive prostacyclin (Flolan, Remodulin). The results of RHC (Table 2) showed that the patients with pleural effusions had a significantly higher mRAP, lower cardiac output, lower cardiac index, and higher median PVR than the patients without pleural effusions. There was no significant difference in the mPAP and PWP between the two groups. There was a significant negative correlation between the mRAP and the cardiac index (r 5 20.275, P 5 .012) but there was no significant correlation between the mRAP and the PVR (r 5 0.108, P 5 .334). There was a significant negative correlation between the cardiac index and the PVR (r 5 20.564, P , .001). The pleural effusions were predominantly bilateral (Table 3). Although the majority of effusions were Table 2—Hemodynamics of Patients With PAH Associated With CTD With or Without Pleural Effusions
Hemodynamicsa mRAP, mm Hg RVEDP, mm Hg mPAP, mm Hg PWP, mm Hg CO, L/min Cardiac index, L/min/m2 PVR, Wood units
Patients Without Pleural Effusions (n 5 54) 8.3 ⫾ 4.0 12.2 ⫾ 5.3 45.9 ⫾ 12.1 8.9 ⫾ 3.9 4.4 ⫾ 1.4 2.5 ⫾ 0.7 8.7 (6.8, 12.6)
Patients With Pleural Effusions (n 5 29)
Characteristics
No. (%)
Diagnostic method Chest radiograph CT scan Ultrasonography Autopsy Distribution of pleural effusion Left-sided only Right-sided only Bilateral Amount of pleural fluida Trace Small Moderate Massive
23 (79.3) 13 (44.8) 5 (17.2) 2 (6.9) 6 (20.7) 8 (27.6) 15 (51.7) 6 (20.7) 11 (37.9) 9 (31.0) 3 (10.3)
See Table 1 legend for expansion of abbreviations. aFor bilateral pleural effusions, the classification of fluid amount was determined by the side that had more fluid.
trace to small, 41.4% were moderate or massive. When we studied the effect of hemodynamics on the size of pleural effusion in the 29 patients with no alternative explanation via ordered logit regression, we found that the mRAP (P 5 .918), the mPAP (P 5 .392), the cardiac index (P 5 .337), and the PVR (P 5 .312) had no significant relationship to the size of the effusion. Thoracentesis was performed in three patients (two patients at Vanderbilt University Medical Center and one at an outside hospital). The pleural fluid values were available only for the thoracenteses performed at Vanderbilt University. Both patients had massive pleural effusions that were transudates (pleural fluid to serum lactate dehydrogenase ratio 5 0.301 and 0.503, pleural fluid to serum total protein ratio 5 0.381 and 0.393).
P Value
Discussion
11.3 ⫾ 5.1 13.0 ⫾ 6.3 49.2 ⫾ 10.6 7.7 ⫾ 4.3 3.5 ⫾ 1.1 2.1 ⫾ 0.6
.004 .521 .226 .193 .004 .011
11.7 (9.2, 15.8)
.016
In this study, the frequency of pleural effusions in patients with PAH associated with CTD was 39.3% (35 of 89). Six of these patients had an identifiable process that could have caused the effusion; the other 29 patients (32.6% of the entire cohort) had no alternative explanation for their effusion. Almost all patients (28 of 29) with pleural effusions without an alternative explanation had RHF. In addition, the mRAP and PVR were significantly higher and the cardiac index was significantly lower in patients with pleural effusions compared with those without pleural effusions. Although most of the pleural effusions were small, 12 patients (41.3%) had moderate or large pleural effusions. The frequency of pleural effusions with no alternate explanation in the present study is more than
Data are presented as mean ⫾ SD or median (interquartile range). CO 5 cardiac output; mPAP 5 mean pulmonary arterial pressure; mRAP 5 mean right atrial pressure; PVR 5 pulmonary vascular resistance; PWP 5 pulmonary wedge pressure; RHC 5 right-sided heart catheterization; RVEDP 5 right ventricular end-diastolic pressure. See Table 1 legend for expansion of other abbreviations. aFor patients without pleural effusions, we used the hemodynamic results from the first RHC that the patient received for the diagnosis; for patients with pleural effusions, we used the hemodynamic results from the RHC performed closest in time to the identification of the pleural effusions. www.chestpubs.org
Table 3—Characteristics of Pleural Effusions in Patients With PAH Associated With CTD (n 5 29)
CHEST / 140 / 1 / JULY, 2011
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
45
twice that of the 14.3% (21 of 147) reported in our recent study in patients with idiopathic and heritable PAH. In both studies, almost all of the pleural effusions with no alternate explanation occurred in patients with RHF: 28 of 29 in the present study and 19 of 21 in the prior study.3 We believe RHF is the primary explanation for the high frequency of pleural effusions in patients with PAH, regardless of etiology. This conclusion is in contrast to the study by Wiener-Kronish et al1,2 in which pleural effusions in patients with heart failure were only due to LHF. In their first study of 37 patients with congestive heart failure secondary to myocardial infarction or cardiomyopathy, patients with pleural effusions (n 5 19) had a significantly higher mean PWP and pulmonary arterial pressure than patients without pleural effusions (n 5 18). In contrast, the mRAP was not different between patients with and without pleural effusions. They concluded that an increased mRAP was not associated with the development of pleural effusions.1 In a second study, pleural effusions were not identified in any of 27 patients with chronic right atrial or PAH (or both). They concluded that chronic elevation of right atrial pressure (RAP) or pulmonary arterial pressure (or both) was not a cause of pleural effusion.2 CTDs represent various categories of immunologically mediated inflammatory disorders with multiorgan involvement. At times, the pleura is involved with CTD and can manifest as a pleural effusion.10 This raises the possibility that the CTD itself, rather than RHF, caused many of the pleural effusions in this study. We believe that this is unlikely since the frequency of pleural effusion was similar in those diseases usually associated with pleural effusion and in those diseases not usually associated with pleural effusion. In patients with SLE, RA, and MCTD, as many as 20% to 50% develop pleural effusions.10-12 However, in patients with scleroderma, polymyositis, or Sjögren syndrome, the frequency is , 7%.10-12 If the pleural effusions in our cohort were the result of pleural inflammation from CTD, then we would have expected a higher frequency of pleural effusions in the patients with SLE, RA, and MCTD. In our study, 20 of 50 (40%) patients with scleroderma, Sjögren syndrome, or polymyositis had pleural effusions, whereas only nine of 33 (27%) with SLE, RA, or MCTD had pleural effusions. Therefore, it is unlikely that the effusions in the present series were due to CTD and more likely that they were due to RHF. The hemodynamic data in our patients supports RHF as the etiology, in concordance with the findings in patients with idiopathic and heritable PAH. Patients with pleural effusions had significantly higher mRAP and lower cardiac index than patients without pleural effusions. The level of the mRAP was signifi-
cantly negatively correlated with the cardiac index. Moreover, the occurrence of RHF in patients with pleural effusions (96.6%) was significantly higher than in patients without pleural effusions (59.3%). Additional support for our contention that the effusions were due to RHF is the significantly higher median BNP levels in the patients with pleural effusions. It should be noted that the difference in the mRAP between those with effusions with no other explanation and those with no effusion was only 3 mm Hg. This suggests that other factors, such as increased capillary permeability, decreased lymphatic clearance, or movement of fluid from the peritoneal or pericardial cavities, may have contributed to formation of the pleural effusion. In contradistinction to the studies by WienerKronish and colleagues,1,2 animal studies have demonstrated that RHF with subsequent increased systemic venous pressure can cause pleural effusions.13,14 A study in dogs showed that pleural effusions developed with acute elevation of systemic venous pressure by balloon obstruction of the superior and inferior vena.13 Another study in sheep found that elevations of superior vena cava pressure above 15 mm Hg resulted in formation of pleural effusions.14 Because the ratio of pleural fluid to serum protein increased, they concluded that the effusion was most likely due to obstruction of lymphatic drainage from the elevated systemic venous pressure.14 Another plausible explanation is that elevated RAP could increase systemic venous pressure, which leads to a higher gradient between the intravascular pressure and the pleural pressure, increasing the rate of pleural fluid accumulation according to Starling’s Equation.15 The frequency of pericardial effusions was significantly higher in patients with than those without pleural effusions in this study (72.4% vs 42.6%, P 5 .018). This raises the possibility that the pericardial and pleural effusions were related to pericardial inflammation from CTD. However, we believe that pericardial inflammation is an unlikely cause of the pericardial effusions because the frequency of pericardial effusions in patients with pleural effusions in this study (72.4%) was similar to that in patients with idiopathic and heritable PAH (78.9%).3 The patients with pleural effusions compared with those without pleural effusions had a significantly longer follow-up period. In this study, the time of follow-up until the patient developed a pleural effusion was very long (17.7 ⫾ 23.0 months). This time is similar to the time that the patients without effusions were followed. We believe that the longer a patient was followed, the more likely he/she would develop RHF and a pleural effusion. There was a significantly higher mortality in patients with pleural effusions
46
Original Research
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
than in those without pleural effusions. This is also probably because the patients with effusions were followed longer and developed RHF. It has been reported that pericardial effusions, elevated RAP, and RHF are all markers of poor outcome in patients with PAH.16-18 We believe that the higher mortality in the patients with pleural effusions, like pericardial effusions, is due to the concomitant RHF rather than the effusion per se. It has been accepted that elevation of mPAP cannot directly induce pleural effusions based on the sheep study previously discussed and on our study in patients with idiopathic and heritable PAH.3,14 In the current study, there was no significant difference in the mPAP between patients with and without pleural effusions, which is consistent with earlier studies. Although increased mPAP may not directly cause pleural effusions, it can lead to RHF, which is a risk factor for the development of pleural effusions. Our study has several limitations. The study was retrospective in nature and some data were not available. In some patients the RHC was not performed close in time to when the pleural effusions were noted. Only a few patients underwent thoracenteses and the pleural fluid analyses were available in only two patients. Moreover, the frequency of ascites might likely be underestimated because only 27 of the 83 patients had a CT scan of the abdomen. The autopsy results were available in only a few patients and therefore the amount of pleural fluid at autopsy was rarely measured. Last, when the groups were compared, no allowance was made for multiple comparisons. Therefore, some significant statistic results might be due to chance. Further prospective study is needed to confirm or refute the findings in this study. In conclusion, our study demonstrates that the frequency of pleural effusions without alternate explanation in patients with PAH associated with CTD is 32.6% (29 of 89). Of these 29 patients, 28 (96.6%) had evidence of RHF. Furthermore, patients with pleural effusions had a significantly higher mRAP and PVR and lower cardiac index compared with those without effusions. This suggests that RHF is the cause of the accumulation of pleural fluid. Acknowledgments Author contributions: Dr Luo: contributed to conceiving the study, reviewing the medical records, collecting the data, performing statistical analysis, analyzing and interpreting the data, drafting the manuscript, and critically revising the manuscript for important intellectual content. Dr Robbins: contributed to reviewing the medical records, collecting the data, and critically revising the manuscript for important intellectual content. Dr Karatas: contributed to reviewing the medical records, collecting the data, and critically revising the manuscript for important intellectual content. Dr Brixey: contributed to critically revising the manuscript for important intellectual content. www.chestpubs.org
Dr Rice: contributed to performing statistical analysis and the statistical analysis discussions in the manuscript. Dr Light: contributed to conceiving the study, analyzing and interpreting the data, drafting the manuscript, and critically revising the manuscript for important intellectual content. Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Other contributions: This work was performed at the Vanderbilt University Medical Center.
References 1. Wiener-Kronish JP, Matthay MA, Callen PW, Filly RA, Gamsu G, Staub NC. Relationship of pleural effusions to pulmonary hemodynamics in patients with congestive heart failure. Am Rev Respir Dis. 1985;132(6):1253-1256. 2. Wiener-Kronish JP, Goldstein R, Matthay RA, et al. Lack of association of pleural effusion with chronic pulmonary arterial and right atrial hypertension. Chest. 1987;92(6):967-970. 3. Tang KJ, Robbins IM, Light RW. Incidence of pleural effusions in idiopathic and familial pulmonary arterial hypertension patients. Chest. 2009;136(3):688-693. 4. Simonneau G, Robbins IM, Beghetti M, et al. Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol. 2009;54(1 suppl):S43-S54. 5. Simonneau G, Galiè N, Rubin LJ, et al. Clinical classification of pulmonary hypertension. J Am Coll Cardiol. 2004; 43(12 suppl S):5S-12S. 6. McGee SR. Physical examination of venous pressure: a critical review. Am Heart J. 1998;136(1):10-18. 7. Wood P. Right and left ventricular failure: a study of circulation time and venous blood pressure. Lancet. 1936;228(5888):15-18. 8. Light RW, Rogers JT, Cheng D, Rodriguez RM. Large pleural effusions occurring after coronary artery bypass grafting. Cardiovascular Surgery Associates, PC. Ann Intern Med. 1999;130(11):891-896. 9. Light RW, Macgregor MI, Luchsinger PC, Ball WC Jr. Pleural effusions: the diagnostic separation of transudates and exudates. Ann Intern Med. 1972;77(4):507-513. 10. Bouros D, Pneumatikos I, Tzouvelekis A. Pleural involvement in systemic autoimmune disorders. Respiration. 2008; 75(4):361-371. 11. Crestani B. The respiratory system in connective tissue disorders. Allergy. 2005;60(6):715-734. 12. Joseph J, Sahn SA. Connective tissue diseases and the pleura. Chest. 1993;104(1):262-270. 13. Mellins RB, Levine OR, Fishman AP. Effect of systemic and pulmonary venous hypertension on pleural and pericardial fluid accumulation. J Appl Physiol. 1970;29(5):564-569. 14. Allen SJ, Laine GA, Drake RE, Gabel JC. Superior vena caval pressure elevation causes pleural effusion formation in sheep. Am J Physiol. 1988;255(3 pt 2):H492-H495. 15. Light RW. Pleural Diseases. 5th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2007:7-16. 16. Raymond RJ, Hinderliter AL, Willis PW, et al. Echocardiographic predictors of adverse outcomes in primary pulmonary hypertension. J Am Coll Cardiol. 2002;39(7):1214-1219. 17. Hinderliter AL, Willis PW IV, Long W, et al. Frequency and prognostic significance of pericardial effusion in primary pulmonary hypertension. PPH Study Group. Primary pulmonary hypertension. Am J Cardiol. 1999;84(4):481-484., A10. 18. Eysmann SB, Palevsky HI, Reichek N, Hackney K, Douglas PS. Two-dimensional and Doppler-echocardiographic and cardiac catheterization correlates of survival in primary pulmonary hypertension. Circulation. 1989;80(2):353-360. CHEST / 140 / 1 / JULY, 2011
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
47
Original Research PULMONARY VASCULAR DISEASE
Frequency of Pleural Effusions in Patients With Pulmonary Arterial Hypertension Associated With Connective Tissue Diseases Yi-feng Luo, MD; Ivan M. Robbins, MD; Mevlut Karatas, MD; Anupama G. Brixey, MD; Todd W. Rice, MD, FCCP; and Richard W. Light, MD, FCCP
Background: Pleural effusions frequently accumulate in patients with left-sided heart failure. However, our recent study in patients with idiopathic and heritable pulmonary arterial hypertension (PAH) demonstrated that pleural effusions frequently occur in patients with isolated rightsided heart failure (RHF). The objective of this study was to determine the frequency of pleural effusions in patients with PAH associated with connective tissue disease (CTD). Methods: We retrospectively studied consecutive patients with PAH associated with CTD who were treated in the Vanderbilt Pulmonary Vascular Center. Pleural effusions were identified by chest radiograph, chest CT scan, thoracic ultrasonography, or autopsy. Results: Thirty-five of 89 patients (39.3%) with PAH associated with CTD had pleural effusions: 23 of 51 (45.1%) with scleroderma, six of 16 (37.5%) with systemic lupus erythematosus, five of 18 (27.8%) with mixed connective tissue disease, and one of two (50.0%) with Sjögren syndrome. There were alternative explanations for the pleural effusions in six of these patients. Of the 29 patients without alternative explanation for their pleural effusions, 28 had RHF. When compared with the patients without pleural effusions, the 29 patients with pleural effusions had significantly higher mean right atrial pressures (11.3 ⫾ 5.1 mm Hg vs 8.3 ⫾ 4.0 mm Hg, P 5 .004) and lower cardiac indices (2.1 ⫾ 0.6 L/min/m2 vs 2.5 ⫾ 0.7 L/min/m2, P 5 .011). The pleural effusions were predominantly trace to small (58.6%) in size and bilateral (51.7%) in distribution. Conclusions: Pleural effusions frequently accumulate in patients with PAH associated with CTD and are associated with RHF. CHEST 2011; 140(1):42–47 Abbreviations: BNP 5 brain natriuretic peptide; CTD 5 connective tissue disease; CXR 5 chest radiograph; ECHO 5 echocardiography; LHF 5 left-sided heart failure; MCTD 5 mixed connective tissue disease; mPAP 5 mean pulmonary arterial pressure; mRAP 5 mean right atrial pressure; PAH 5 pulmonary arterial hypertension; PVR 5 pulmonary vascular resistance; PWP 5 pulmonary wedge pressure; RA 5 rheumatoid arthritis; RAP 5 right atrial pressure; RHC 5 right-sided heart catheterization; RHF 5 right-sided heart failure; SLE 5 systemic lupus erythematosus
is generally accepted that the accumulation of Itpleural fluid in patients with congestive heart fail-
ure is due to left-sided heart failure (LHF) rather than right-sided heart failure (RHF).1,2 However, we have recently reported that pleural effusions are common in patients with idiopathic pulmonary arterial hypertension (PAH) or heritable PAH. The majority of patients in that study with pleural effusions had RHF.3 The purpose of the present study was to determine the frequency of pleural effusions in patients with PAH associated with connective tissue disease (CTD). We hypothesized that the overall frequency of pleural effusions would be higher in the patients with
CTD than in those with idiopathic or heritable PAH, because some effusions may be due to the CTD itself. We also hypothesized that effusions would be more common in patients with RHF. Materials and Methods Study Procedures We retrospectively studied the medical records of consecutive patients seen in the Vanderbilt Pulmonary Vascular Center from September 1996 to August 2009 with a diagnosis of PAH associated with CTD. All patients were seen by rheumatologists, and
42
Original Research
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
the diagnosis of CTD was established using standard criteria. Patients were diagnosed as having PAH associated with CTD according to the Dana Point clinical classification of PAH.4 Although some patients were evaluated prior to the publication of the Dana Point criteria for classification of PAH, all patients included in this study underwent a thorough evaluation to exclude other possible causes of PAH. All patients underwent right-sided heart catheterization (RHC) for the initial diagnosis. PAH associated with CTD was defined as a mean resting pulmonary arterial pressure . 25 mm Hg, a pulmonary wedge pressure (PWP) ⱕ 15 mm Hg, and the absence of other possible causes of PAH, such as PAH associated with congenital systemic to pulmonary shunts, portal hypertension, HIV infection, pulmonary hypertension secondary to left-sided heart disease (including diastolic dysfunction and valvular disease), and chronic thromboembolic PAH.4,5 RHF was defined as being present when there was evidence on physical examination (jugular venous pressure . 8 cm H2O, hepatojugular reflux, hepatomegaly, right-sided S3, ascites, anasarca, or worsening lower extremity edema), moderate or severe dilation of the right atrium or right ventricle with moderate or severe impairment of right ventricular function and a noncollapsing inferior vena cava on echocardiography (ECHO), or an elevated right atrial pressure ( . 10 mm Hg) on RHC.3,6,7 The study protocol was approved by the Vanderbilt University Institutional Review Board. The requirement for informed consent was waived because of the retrospective nature of the study. For each patient, the clinical features, laboratory data, radiographic data, and the presence or absence, size, and distribution of pleural fluid were reviewed. In patients who had a thoracentesis, the laboratory results and analysis of the pleural fluid were recorded. The presence of a pericardial effusion or ascites, the total serum protein and albumin levels, as well as results of RHC, ECHO, and thoracic ultrasonography that were obtained closest in time to when the pleural effusion was noted were recorded. The chest radiograph (CXR), chest CT scan, thoracic ultrasonography, and autopsy results were reviewed by our team for the presence of a pleural effusion. The size of the pleural effusion on the posteroanterior and lateral CXRs was semiquantitated using the following scale8: 0 5 no pleural fluid present; 1 5 blunting of the costophrenic angle only; 2 5 , 25% of hemithorax occupied by pleural fluid; 3 5 25% to 50% of hemithorax occupied; 4 5 50% to 75% of hemithorax occupied; 5 5 . 75% of hemithorax occupied. The size of the pleural effusion on CT scan or ultrasound in the supine position was semiquantitated by measuring the maximum thickness of fluid between the visceral and parietal pleurae. For patients who underwent autopsy, the volume of pleural fluid was recorded. Based on these findings, the pleural Manuscript received January 26, 2010; revision accepted December 11, 2010. Affiliations: From the Department of Pulmonary Medicine (Dr Luo), The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, Guangdong, China; the Division of Allergy, Pulmonary and Critical Care Medicine (Drs Luo, Robbins, Karatas, Brixey, Rice, and Light), Vanderbilt University Medical Center, Nashville, TN; and the Department of Pulmonary Medicine (Dr Karatas), Ahi Evren Thoracic Heart and Cardiovascular Surgery Training and Research Hospital, Trabzon, Turkey. Correspondence to: Yi-feng Luo, MD, Department of Pulmonary Medicine, The First Affiliated Hospital of Sun Yat-sen University, 58 Zhongshan 2nd Rd, Guangzhou, Guangdong Province, 510080, China; e-mail: [email protected] Funding/Support: The authors have reported to CHEST that no funding was received for this study. © 2011 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (http://www.chestpubs.org/ site/misc/reprints.xhtml). DOI: 10.1378/chest.10-0227 www.chestpubs.org
THE BOTTOM LINE How does this work advance the field? It is generally accepted that pleural effusions frequently accumulate in patients with left-sided heart failure. However, this study suggests that pleural effusions may frequently occur in patients with isolated right-sided heart failure.
What are the clinical implications? Pleural effusion is a common complication not only in patients with left-sided heart failure but may also frequently occur in patients with right-sided heart failure.
effusion was placed in one of four categories. A patient with more than one study was placed in the highest category achieved on any of the studies: 1. Trace pleural effusion: CXR score 5 1, or fluid thickness , 5 mm on CT scan or ultrasonography, or fluid volume measured at autopsy , 125 mL. 2. Small pleural effusion: CXR score 5 2, or fluid thickness of 5 to 20 mm on CT scan or ultrasonography, or fluid volume measured at autopsy between 125 mL and 500 mL. 3. Moderate pleural effusion: CXR score 5 3, or fluid thickness of 21 to 50 mm on CT scan or ultrasonography, or fluid volume measured at autopsy between 500 mL and 1,250 mL. 4. Massive pleural effusion: CXR score 5 4 or 5, or fluid thickness . 50 mm on CT scan or ultrasonography, or fluid volume measured at autopsy . 1,250 mL. For patients who had received a thoracentesis, Light’s criteria were used to separate transudative from exudative pleural effusions.9 Statistical Analysis Statistical analysis was performed using SigmaStat 2.03 for Windows software (Systat Software, Inc; Richmond, California) and R software 2.11.1. The data are presented as mean ⫾ SD when normally distributed and as median with interquartile ranges in parentheses when not. Independent samples t test was used to compare the means between two groups that were normally distributed. The Mann-Whitney U test was used to assess the equivalence of two distributions from populations that were not normally distributed or that did not have equal variances. This test evaluates whether the observations from one population have a tendency that is larger (or smaller) than the observation from the other population. x2 and Fisher exact tests were used to assess differences in categorical variables between groups. We did not make corrections for multiple comparisons. We calculated the partial correlation coefficient with Pearson correlation test to analyze the relationship among hemodynamics (mean right atrial pressure [mRAP], cardiac index, and pulmonary vascular resistance [PVR]). Ordered logit regression was used to study the effect of the hemodynamics (mRAP, mean pulmonary arterial pressure [mPAP], cardiac index, and PVR) on the size of the effusion. All P values reported were two-tailed and P values , .05 were considered statistically significant.
Results Ninety-nine patients with PAH associated with CTD were evaluated at our pulmonary vascular center CHEST / 140 / 1 / JULY, 2011
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
43
during the study period. Ten patients were excluded from the study: Four patients had diastolic dysfunction, two patients had LHF, two patients never had an RHC, one patient had valvular heart disease, and one patient had chronic thromboembolic disease. The remaining 89 patients, 80 women and nine men, were included in this study. The breakdown of the type of CTD in our patients was as follows: scleroderma, 51; mixed connective tissue disease (MCTD), 18; systemic lupus erythematosus (SLE), 16; Sjögren syndrome, two; polymyositis, one; and rheumatoid arthritis (RA), one. No patients underwent lung transplantation. Overall, 35 of the 89 patients (39.3%) with PAH associated with CTD had pleural effusions during follow-up: 23 of 51 (45.1%) with scleroderma, six of 16 (37.5%) with SLE, five of 18 (27.8%) with MCTD, and one of two (50%) with Sjögren syndrome. There was no significant difference in the frequency of effusions in the different categories (P 5 0.617). Six of the 35 patients with pleural effusions had an alternative explanation for their effusions. Four had radiologic evidence of pneumonia, one had severe hypoalbuminemia, and one had an exudative pleural effusion due to the CTD itself. This latter patient
was a 70-year-old white woman with MCTD, massive bilateral pleural effusions, and concomitant ascites and pericardial effusions. She underwent a thoracentesis at an outside hospital and reportedly had an exudative pleural effusion (the results of the pleural effusion analysis were unavailable). She had Raynaud phenomenon and a positive antinuclear antibody level of 1:320. All patients had RHC and almost all had ECHOs. Of the 29 patients without an alternative explanation for their pleural effusion, 27 had an ECHO, whereas of the remaining 54 patients, 50 had an ECHO. Of the 29 patients without an alternative explanation for their pleural effusions, 28 were diagnosed as having RHF: 25 patients had evidence of RHF on RHC and/or ECHO, and three patients had obvious evidence of RHF on physical examination. The one patient without RHF had trace pleural effusions and had no obvious symptoms or signs of RHF at the time the effusions were identified. Of the 28 patients with RHF and pleural effusions, 20 had RHC or ECHO performed within 1 month of the time the pleural effusion was noted (13 had both RHC and ECHO, three had RHC only, four had ECHO only).
Table 1—Demographics and Clinical Characteristics of Patients With PAH Associated With CTD With or Without Pleural Effusions Patients Without Pleural Effusions (n 5 54)
Variables Age at diagnosis of PAH associated with CTD, y Female (male) Classification of CTD Scleroderma SLE MCTD Sjögren syndrome Polymyositis RA Median follow-up period, mo Time of follow-up until the patient first developed a pleural effusion, mo Death during follow-up Time from PAH diagnosis to death, mo Occurrence of RHF Occurrence of ascitesa Occurrence of pericardial effusionsa Treatmentb Receiving prostacyclin Not receiving prostacyclin Serum total protein, g/dL Serum albumin, g/dL Serum BNP, pg/mL
Patients With Pleural Effusions (n 5 29)
P Value
51.0 ⫾ 13.3
51.0 ⫾ 14.0
.999
49 (5)
25 (4)
.713 .615
28 (51.9) 10 (18.5) 13 (24.1) 1 (1.9) 1 (1.9) 1 (1.9) 16 (5, 47) … 20 (37.0) 22.5 ⫾ 17.2 (n 5 20) 32 (59.3) 1 (1.9) 23 (42.6) 28 (51.9) 26 (48.1) 7.1 ⫾ 0.9 3.8 ⫾ 0.6 301 (99, 748) (n 5 31)
20 (69.0) 5 (17.2) 4 (13.8) 0 (0) 0 (0) 0 (0) 27 (15, 72) 17.7 ⫾ 23.0 19 (65.5) 37.5 ⫾ 28.6 (n 5 19) 28 (96.6) 9 (31.0) 21 (72.4) 10 (34.5) 19 (65.5) 7.0 ⫾ 0.9 3.6 ⫾ 0.6 654 (398, 1,188) (n 5 13)
.032
.025 .053 , .001 , .001 .018 .199
.817 .282 .037
Unless otherwise indicated, data are presented as mean ⫾ SD, median (interquartile range), or No. (%). BNP 5 brain natriuretic peptide; CTD 5 connective tissue disease; MCTD 5 mixed connective tissue disease; PAH 5 pulmonary arterial hypertension; RA 5 rheumatoid arthritis; RHF 5 right-sided heart failure; SLE 5 systemic lupus erythematosus. aThe occurrence of ascites or pericardial effusions was assessed for the entire follow-up period. bThe patients listed as having pleural effusions and receiving prostacyclin all developed effusions after initiation of prostacyclin. Patients who developed effusions prior to initiation of prostacyclin were included in the group “Not receiving prostacyclin.” 44
Original Research
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
When the clinical characteristics of the 54 patients without pleural effusions and the 29 patients with pleural effusions were compared (Table 1), age at diagnosis and the proportion of the patients in the different CTD categories with pleural effusions did not differ significantly. In patients with pleural effusions, the follow-up period was significantly longer, and the mortality was significantly higher as was the occurrence of RHF, ascites, and pericardial effusions. Brain natriuretic peptide (BNP) levels were available in 13 of 29 patients with pleural effusions and 31 of 54 patients without pleural effusions. The median BNP level (654 pg/mL) was significantly higher in the patients with pleural effusions than the level (301 pg/mL) in the patients without pleural effusions (P 5 .037). The frequency of pleural effusions was not significantly different between the patients who received and did not receive prostacyclin (Flolan, Remodulin). The results of RHC (Table 2) showed that the patients with pleural effusions had a significantly higher mRAP, lower cardiac output, lower cardiac index, and higher median PVR than the patients without pleural effusions. There was no significant difference in the mPAP and PWP between the two groups. There was a significant negative correlation between the mRAP and the cardiac index (r 5 20.275, P 5 .012) but there was no significant correlation between the mRAP and the PVR (r 5 0.108, P 5 .334). There was a significant negative correlation between the cardiac index and the PVR (r 5 20.564, P , .001). The pleural effusions were predominantly bilateral (Table 3). Although the majority of effusions were Table 2—Hemodynamics of Patients With PAH Associated With CTD With or Without Pleural Effusions
Hemodynamicsa mRAP, mm Hg RVEDP, mm Hg mPAP, mm Hg PWP, mm Hg CO, L/min Cardiac index, L/min/m2 PVR, Wood units
Patients Without Pleural Effusions (n 5 54) 8.3 ⫾ 4.0 12.2 ⫾ 5.3 45.9 ⫾ 12.1 8.9 ⫾ 3.9 4.4 ⫾ 1.4 2.5 ⫾ 0.7 8.7 (6.8, 12.6)
Patients With Pleural Effusions (n 5 29)
Characteristics
No. (%)
Diagnostic method Chest radiograph CT scan Ultrasonography Autopsy Distribution of pleural effusion Left-sided only Right-sided only Bilateral Amount of pleural fluida Trace Small Moderate Massive
23 (79.3) 13 (44.8) 5 (17.2) 2 (6.9) 6 (20.7) 8 (27.6) 15 (51.7) 6 (20.7) 11 (37.9) 9 (31.0) 3 (10.3)
See Table 1 legend for expansion of abbreviations. aFor bilateral pleural effusions, the classification of fluid amount was determined by the side that had more fluid.
trace to small, 41.4% were moderate or massive. When we studied the effect of hemodynamics on the size of pleural effusion in the 29 patients with no alternative explanation via ordered logit regression, we found that the mRAP (P 5 .918), the mPAP (P 5 .392), the cardiac index (P 5 .337), and the PVR (P 5 .312) had no significant relationship to the size of the effusion. Thoracentesis was performed in three patients (two patients at Vanderbilt University Medical Center and one at an outside hospital). The pleural fluid values were available only for the thoracenteses performed at Vanderbilt University. Both patients had massive pleural effusions that were transudates (pleural fluid to serum lactate dehydrogenase ratio 5 0.301 and 0.503, pleural fluid to serum total protein ratio 5 0.381 and 0.393).
P Value
Discussion
11.3 ⫾ 5.1 13.0 ⫾ 6.3 49.2 ⫾ 10.6 7.7 ⫾ 4.3 3.5 ⫾ 1.1 2.1 ⫾ 0.6
.004 .521 .226 .193 .004 .011
11.7 (9.2, 15.8)
.016
In this study, the frequency of pleural effusions in patients with PAH associated with CTD was 39.3% (35 of 89). Six of these patients had an identifiable process that could have caused the effusion; the other 29 patients (32.6% of the entire cohort) had no alternative explanation for their effusion. Almost all patients (28 of 29) with pleural effusions without an alternative explanation had RHF. In addition, the mRAP and PVR were significantly higher and the cardiac index was significantly lower in patients with pleural effusions compared with those without pleural effusions. Although most of the pleural effusions were small, 12 patients (41.3%) had moderate or large pleural effusions. The frequency of pleural effusions with no alternate explanation in the present study is more than
Data are presented as mean ⫾ SD or median (interquartile range). CO 5 cardiac output; mPAP 5 mean pulmonary arterial pressure; mRAP 5 mean right atrial pressure; PVR 5 pulmonary vascular resistance; PWP 5 pulmonary wedge pressure; RHC 5 right-sided heart catheterization; RVEDP 5 right ventricular end-diastolic pressure. See Table 1 legend for expansion of other abbreviations. aFor patients without pleural effusions, we used the hemodynamic results from the first RHC that the patient received for the diagnosis; for patients with pleural effusions, we used the hemodynamic results from the RHC performed closest in time to the identification of the pleural effusions. www.chestpubs.org
Table 3—Characteristics of Pleural Effusions in Patients With PAH Associated With CTD (n 5 29)
CHEST / 140 / 1 / JULY, 2011
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
45
twice that of the 14.3% (21 of 147) reported in our recent study in patients with idiopathic and heritable PAH. In both studies, almost all of the pleural effusions with no alternate explanation occurred in patients with RHF: 28 of 29 in the present study and 19 of 21 in the prior study.3 We believe RHF is the primary explanation for the high frequency of pleural effusions in patients with PAH, regardless of etiology. This conclusion is in contrast to the study by Wiener-Kronish et al1,2 in which pleural effusions in patients with heart failure were only due to LHF. In their first study of 37 patients with congestive heart failure secondary to myocardial infarction or cardiomyopathy, patients with pleural effusions (n 5 19) had a significantly higher mean PWP and pulmonary arterial pressure than patients without pleural effusions (n 5 18). In contrast, the mRAP was not different between patients with and without pleural effusions. They concluded that an increased mRAP was not associated with the development of pleural effusions.1 In a second study, pleural effusions were not identified in any of 27 patients with chronic right atrial or PAH (or both). They concluded that chronic elevation of right atrial pressure (RAP) or pulmonary arterial pressure (or both) was not a cause of pleural effusion.2 CTDs represent various categories of immunologically mediated inflammatory disorders with multiorgan involvement. At times, the pleura is involved with CTD and can manifest as a pleural effusion.10 This raises the possibility that the CTD itself, rather than RHF, caused many of the pleural effusions in this study. We believe that this is unlikely since the frequency of pleural effusion was similar in those diseases usually associated with pleural effusion and in those diseases not usually associated with pleural effusion. In patients with SLE, RA, and MCTD, as many as 20% to 50% develop pleural effusions.10-12 However, in patients with scleroderma, polymyositis, or Sjögren syndrome, the frequency is , 7%.10-12 If the pleural effusions in our cohort were the result of pleural inflammation from CTD, then we would have expected a higher frequency of pleural effusions in the patients with SLE, RA, and MCTD. In our study, 20 of 50 (40%) patients with scleroderma, Sjögren syndrome, or polymyositis had pleural effusions, whereas only nine of 33 (27%) with SLE, RA, or MCTD had pleural effusions. Therefore, it is unlikely that the effusions in the present series were due to CTD and more likely that they were due to RHF. The hemodynamic data in our patients supports RHF as the etiology, in concordance with the findings in patients with idiopathic and heritable PAH. Patients with pleural effusions had significantly higher mRAP and lower cardiac index than patients without pleural effusions. The level of the mRAP was signifi-
cantly negatively correlated with the cardiac index. Moreover, the occurrence of RHF in patients with pleural effusions (96.6%) was significantly higher than in patients without pleural effusions (59.3%). Additional support for our contention that the effusions were due to RHF is the significantly higher median BNP levels in the patients with pleural effusions. It should be noted that the difference in the mRAP between those with effusions with no other explanation and those with no effusion was only 3 mm Hg. This suggests that other factors, such as increased capillary permeability, decreased lymphatic clearance, or movement of fluid from the peritoneal or pericardial cavities, may have contributed to formation of the pleural effusion. In contradistinction to the studies by WienerKronish and colleagues,1,2 animal studies have demonstrated that RHF with subsequent increased systemic venous pressure can cause pleural effusions.13,14 A study in dogs showed that pleural effusions developed with acute elevation of systemic venous pressure by balloon obstruction of the superior and inferior vena.13 Another study in sheep found that elevations of superior vena cava pressure above 15 mm Hg resulted in formation of pleural effusions.14 Because the ratio of pleural fluid to serum protein increased, they concluded that the effusion was most likely due to obstruction of lymphatic drainage from the elevated systemic venous pressure.14 Another plausible explanation is that elevated RAP could increase systemic venous pressure, which leads to a higher gradient between the intravascular pressure and the pleural pressure, increasing the rate of pleural fluid accumulation according to Starling’s Equation.15 The frequency of pericardial effusions was significantly higher in patients with than those without pleural effusions in this study (72.4% vs 42.6%, P 5 .018). This raises the possibility that the pericardial and pleural effusions were related to pericardial inflammation from CTD. However, we believe that pericardial inflammation is an unlikely cause of the pericardial effusions because the frequency of pericardial effusions in patients with pleural effusions in this study (72.4%) was similar to that in patients with idiopathic and heritable PAH (78.9%).3 The patients with pleural effusions compared with those without pleural effusions had a significantly longer follow-up period. In this study, the time of follow-up until the patient developed a pleural effusion was very long (17.7 ⫾ 23.0 months). This time is similar to the time that the patients without effusions were followed. We believe that the longer a patient was followed, the more likely he/she would develop RHF and a pleural effusion. There was a significantly higher mortality in patients with pleural effusions
46
Original Research
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
than in those without pleural effusions. This is also probably because the patients with effusions were followed longer and developed RHF. It has been reported that pericardial effusions, elevated RAP, and RHF are all markers of poor outcome in patients with PAH.16-18 We believe that the higher mortality in the patients with pleural effusions, like pericardial effusions, is due to the concomitant RHF rather than the effusion per se. It has been accepted that elevation of mPAP cannot directly induce pleural effusions based on the sheep study previously discussed and on our study in patients with idiopathic and heritable PAH.3,14 In the current study, there was no significant difference in the mPAP between patients with and without pleural effusions, which is consistent with earlier studies. Although increased mPAP may not directly cause pleural effusions, it can lead to RHF, which is a risk factor for the development of pleural effusions. Our study has several limitations. The study was retrospective in nature and some data were not available. In some patients the RHC was not performed close in time to when the pleural effusions were noted. Only a few patients underwent thoracenteses and the pleural fluid analyses were available in only two patients. Moreover, the frequency of ascites might likely be underestimated because only 27 of the 83 patients had a CT scan of the abdomen. The autopsy results were available in only a few patients and therefore the amount of pleural fluid at autopsy was rarely measured. Last, when the groups were compared, no allowance was made for multiple comparisons. Therefore, some significant statistic results might be due to chance. Further prospective study is needed to confirm or refute the findings in this study. In conclusion, our study demonstrates that the frequency of pleural effusions without alternate explanation in patients with PAH associated with CTD is 32.6% (29 of 89). Of these 29 patients, 28 (96.6%) had evidence of RHF. Furthermore, patients with pleural effusions had a significantly higher mRAP and PVR and lower cardiac index compared with those without effusions. This suggests that RHF is the cause of the accumulation of pleural fluid. Acknowledgments Author contributions: Dr Luo: contributed to conceiving the study, reviewing the medical records, collecting the data, performing statistical analysis, analyzing and interpreting the data, drafting the manuscript, and critically revising the manuscript for important intellectual content. Dr Robbins: contributed to reviewing the medical records, collecting the data, and critically revising the manuscript for important intellectual content. Dr Karatas: contributed to reviewing the medical records, collecting the data, and critically revising the manuscript for important intellectual content. Dr Brixey: contributed to critically revising the manuscript for important intellectual content. www.chestpubs.org
Dr Rice: contributed to performing statistical analysis and the statistical analysis discussions in the manuscript. Dr Light: contributed to conceiving the study, analyzing and interpreting the data, drafting the manuscript, and critically revising the manuscript for important intellectual content. Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Other contributions: This work was performed at the Vanderbilt University Medical Center.
References 1. Wiener-Kronish JP, Matthay MA, Callen PW, Filly RA, Gamsu G, Staub NC. Relationship of pleural effusions to pulmonary hemodynamics in patients with congestive heart failure. Am Rev Respir Dis. 1985;132(6):1253-1256. 2. Wiener-Kronish JP, Goldstein R, Matthay RA, et al. Lack of association of pleural effusion with chronic pulmonary arterial and right atrial hypertension. Chest. 1987;92(6):967-970. 3. Tang KJ, Robbins IM, Light RW. Incidence of pleural effusions in idiopathic and familial pulmonary arterial hypertension patients. Chest. 2009;136(3):688-693. 4. Simonneau G, Robbins IM, Beghetti M, et al. Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol. 2009;54(1 suppl):S43-S54. 5. Simonneau G, Galiè N, Rubin LJ, et al. Clinical classification of pulmonary hypertension. J Am Coll Cardiol. 2004; 43(12 suppl S):5S-12S. 6. McGee SR. Physical examination of venous pressure: a critical review. Am Heart J. 1998;136(1):10-18. 7. Wood P. Right and left ventricular failure: a study of circulation time and venous blood pressure. Lancet. 1936;228(5888):15-18. 8. Light RW, Rogers JT, Cheng D, Rodriguez RM. Large pleural effusions occurring after coronary artery bypass grafting. Cardiovascular Surgery Associates, PC. Ann Intern Med. 1999;130(11):891-896. 9. Light RW, Macgregor MI, Luchsinger PC, Ball WC Jr. Pleural effusions: the diagnostic separation of transudates and exudates. Ann Intern Med. 1972;77(4):507-513. 10. Bouros D, Pneumatikos I, Tzouvelekis A. Pleural involvement in systemic autoimmune disorders. Respiration. 2008; 75(4):361-371. 11. Crestani B. The respiratory system in connective tissue disorders. Allergy. 2005;60(6):715-734. 12. Joseph J, Sahn SA. Connective tissue diseases and the pleura. Chest. 1993;104(1):262-270. 13. Mellins RB, Levine OR, Fishman AP. Effect of systemic and pulmonary venous hypertension on pleural and pericardial fluid accumulation. J Appl Physiol. 1970;29(5):564-569. 14. Allen SJ, Laine GA, Drake RE, Gabel JC. Superior vena caval pressure elevation causes pleural effusion formation in sheep. Am J Physiol. 1988;255(3 pt 2):H492-H495. 15. Light RW. Pleural Diseases. 5th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2007:7-16. 16. Raymond RJ, Hinderliter AL, Willis PW, et al. Echocardiographic predictors of adverse outcomes in primary pulmonary hypertension. J Am Coll Cardiol. 2002;39(7):1214-1219. 17. Hinderliter AL, Willis PW IV, Long W, et al. Frequency and prognostic significance of pericardial effusion in primary pulmonary hypertension. PPH Study Group. Primary pulmonary hypertension. Am J Cardiol. 1999;84(4):481-484., A10. 18. Eysmann SB, Palevsky HI, Reichek N, Hackney K, Douglas PS. Two-dimensional and Doppler-echocardiographic and cardiac catheterization correlates of survival in primary pulmonary hypertension. Circulation. 1989;80(2):353-360. CHEST / 140 / 1 / JULY, 2011
Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on July 6, 2011 © 2011 American College of Chest Physicians
47