| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Departments of Radiology (E.M.M., H.P.M.), Medicine (S.M.P.), Biostatistics and Bioinformatics (J.E.H.), and Cancer Center Biostatistics (J.E.H., C.Z.), Duke University Medical Center, Durham, NC; and Department of Radiology, M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 57, Houston, TX 77030 (E.M.M., J.J.E.). Received July 12, 2002; revision requested August 23; revision received September 26; accepted November 18. Address correspondence to E.M.M. (e-mail: emarom@di.mdacc.tmc.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Chest radiographs and computed tomographic (CT) scans obtained in 31 lung transplant recipients who had pleural effusions treated with catheter drainage were retrospectively reviewed. Duration of drainage and volume of fluid drained were recorded. Results were evaluated 1 and 3 months after chest tube removal. There was complete response (CR) when no pleural fluid remained, partial response (PR) when fluid remaining was less than the pretreatment level, and no response (NR) when fluid recurred to a level at or above the pretreatment level. Associations between cause of effusion (empyema, parapneumonic effusion, rejection, other), response (CR, PR, NR), and type of transplantation (unilateral, bilateral) were examined by using 
2 tests.
RESULTS: Of 31 patients, 25 had bilateral effusions; eight of these 25 patients had small-bore catheters inserted bilaterally. Nine patients had multiple sequential catheter insertions. Duration of drainage ranged from 2 to 44 days (median, 6 days). Fluid output was 110–9,726 mL (median, 1,350 mL). One-month follow-up data were available for 31 of 39 treated pleural effusions: 11 (35%) had CR, 18 (58%) had PR, and two (6%) had NR (percentages do not add up to 100% due to rounding). Three-month follow-up data were available for 28 of 39 treated effusions: 22 (79%) had CR, five (18%) had PR, and one (4%) had NR (percentages do not add up to 100% due to rounding). One- and 3-month response rates, respectively, were not related to cause of effusion (P = .82 and .535) or type of transplantation (P = .568 and >.999).
CONCLUSION: Small-bore catheter drainage of persistent pleural effusions in lung transplant recipients is usually successful, but drainage is often prolonged and may require multiple catheter placements.
© RSNA, 2003
Index terms: Lung, transplantation • Pleura, fluid, 66.761 • Pleura, infection, 66.20 • Pleura, interventional procedures, 66.12986
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Underlying diseases were chronic obstructive pulmonary disease (n = 13), cystic fibrosis (n = 7), pulmonary fibrosis (n = 5), 
-1 antitrypsin deficiency (n = 2), agammaglobulinemia (n = 1), primary pulmonary hypertension (n = 1), and retransplantation owing to graft failure (n = 2). Time from transplantation to pleural fluid drainage ranged from 5 days to 36 months (median, 1.8 months). The institutional review board of Duke University Medical Center waived the requirement for patient consent and approved this retrospective study.
Data Collection and Image Review
Medical records were reviewed by one pulmonary physician (S.M.P.) and one thoracic radiologist (E.M.M.), who worked separately, to determine the following: the cause of the pleural effusion(s), the volume of pleural catheter output, the length of time the tube was left in place (hereinafter referred to as tube duration), the presence of any complications of pleural drainage, the use of sclerosing agents or streptokinase, and patient survival. Findings were compared, and any differences were resolved by consensus. The radiology department database was reviewed by one thoracic radiologist (E.M.M.) to assess the number of treated pleural effusions, number of tube insertions, and method of image guidance (computed tomography [CT] or ultrasonography [US]). One thoracic radiologist (E.M.M.) reviewed chest radiographs (available for all 31 [100%] of the patients) and thoracic CT scans (available for 27 [87%] of the patients) obtained before pleural catheter placement and noted the size and location (unilateral or bilateral) of pleural effusions, as well as whether loculation was present or absent in the effusion.
CT scanning was performed without the administration of intravenous contrast material in 15 patients with a HiSpeed Advantage scanner (GE Medical Systems, Milwaukee, Wis), collimation of 10 mm or 7 mm, and pitch of 1.5, and in 12 patients with a QXi Light-Speed scanner (GE Medical Systems), collimation of 7.5 mm, and table speed of 11.25 mm/sec. An effusion was considered small when it filled less than 25% of the hemithorax, moderate when it filled 25%–50% of the hemithorax, and large when it filled more than 50% of the hemithorax.
Catheter Placement
All patients were referred by the in-hospital pulmonary service or the lung transplantation clinic for treatment of pleural effusion. In general, most patients referred for catheter drainage had previously undergone thoracentesis performed by the medical or surgical team. All patients with pleural space infection (as defined by the presence of positive microbiologic culture results, pH of less than 7.20, or frank pus at thoracentesis) were referred for catheter drainage. In addition, patients who had effusions that recurred after initial thoracentesis and were associated with atelectasis and/or reduction in clinical pulmonary function were referred for catheter placement, regardless of the etiology of the effusion (eg, rejection, parapneumonic effusion). In rare cases, patients with large symptomatic effusions with clear atelectasis and reduced lung function were not first treated with thoracentesis but were instead sent directly to the radiology suite for drainage of the effusion.
All pleural catheters were placed by a senior radiology resident and one of seven experienced thoracic radiologists (including E.M.M., J.J.E., and H.P.M.) with either US or CT guidance. When the effusion appeared on chest radiographs or CT scans to be free flowing, the skin insertion site was the middle or posterior axillary line, and the tip of the tube was directed toward the posteroinferior hemithorax. When the effusion appeared to be loculated, the tip of the tube was directed toward the dependent aspect of the loculation. All tubes were placed by using the modified Seldinger technique as follows: An 18-gauge trocar needle was placed into the pleural space, and return of pleural fluid was confirmed. A 0.38 floppy-tipped wire was advanced well into the fluid collection. Sequential dilators were then used to prepare the tube tract. A vanSonnenberg self-retaining catheter (Medi-tech/Boston Scientific, Watertown, Mass) (usually 14 F, though size varied from 8 to 16 F) was advanced over the wire into the pleural fluid. The pigtail catheter was curled and locked.
At the time of catheter placement, up to 1 L of fluid was initially evacuated, depending on patient comfort and symptoms; the pleural catheter was then placed to -20 cm H2O continuous wall suction through a Pleur-evac water-seal device (Deknatel, Fall River, Mass). A chest radiograph was then obtained to document the position of the pleural catheter. All patients were hospitalized for the duration of the procedure, and tube output was recorded on a daily basis.
If the patient’s tube output diminished but follow-up chest radiographs or CT scans revealed an increase in fluid or evidence of loculation, 250,000 IU of streptokinase (Streptase; Aventis Behring, Marburg, Germany) in 100 mL of normal saline was instilled through the catheter by a senior radiology resident at the direction of an attending thoracic radiologist (attending radiologists included E.M.M., J.J.E., and H.P.M.). The catheter was then clamped for 2 hours, and the patient was encouraged to change positions to help distribute the solution uniformly. The tube was then opened and reconnected to suction. If tube output did not increase and fluid remained loculated, an additional dose of streptokinase was instilled on each subsequent day until tube output improved. Tubes were removed once tube output ceased and chest radiographs showed no residual fluid.
Sclerosing Agents
Sclerosing agents were not routinely used unless specifically requested by the treating clinician. When requested (for two patients), talc was used as a sclerosing agent. Immediately prior to talc instillation, patients received 2 mg of intravenous morphine. The sclerosant was administered by means of slow injection through the pleural catheter by a senior radiology resident at the direction of an attending thoracic radiologist. The sclerosing agent was composed of 5 g of sterile talc mixed with 10 mg of 1% lidocaine in 100 mL of normal saline. After talc instillation, the tube was clamped, suction was discontinued for 2 hours, and the patient was encouraged to rotate from the supine to both decubitus positions every 15 minutes to aid distribution of the talc within the pleural space. Suction was then resumed at -20 cm H2O for 24 hours; the tube was then removed.
Response to Drainage
Two experienced thoracic radiologists (E.M.M., H.P.M.) evaluated the response to catheter drainage immediately after (ie, day 0) and on days 30 and 90 after removal of the pleural catheter. To assess response, we compared chest radiographs obtained on day 0, day 30, and day 90 with pretreatment chest radiographs. We recorded a complete response when no pleural fluid remained, a partial response when fluid remaining was less than the pretreatment level, and no response when fluid recurred to a level at or above the pretreatment level. Findings were recorded in consensus.
For the patients who had bilateral pleural effusions, the number of patients with an undrained contralateral effusion was also recorded. The outcome of these conservatively managed pleural effusions was assessed in the same way the outcome of treated effusions was assessed (ie, we evaluated the conservatively managed effusions for complete, partial, or no response on days 0, 30, and 90 after the pleural catheter was removed from the contralateral side of the thorax).
Statistical Evaluation
Descriptive statistics, including means, SDs, medians, and ranges, were generated for continuous variables; for categorical variables, the frequency and percentage in each category were calculated. Associations between the following categorical variables were examined by using 2 or Fisher exact tests: cause of effusion (empyema, parapneumonic effusion, rejection, other), primary disease (chronic obstructive pulmonary disease, cystic fibrosis, pulmonary fibrosis, other), response (complete response, partial response, no response), treatment-related death (yes, no; treatment-related death was defined as death within 6 months after tube treatment), type of transplantation (unilateral, bilateral), tube duration (≤6 days, >6 days), and effusion size (<25%, 25%–50%, >50%). The Kruskal-Wallis test was used to compare the uncategorized duration of catheter drainage among groups that were defined by cause of effusion. A P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Etiology of Treated Effusions
The etiology of the treated pleural effusion was infection in 16 (52%) patients, 10 (32%) of whom had complicated parapneumonic effusion and six (19%) of whom had frank empyema. The remaining patients had effusions that were due to rejection (n = 7, 23%), mechanical or postoperative complications (n = 5, 16%), heart failure or hypoalbuminemia (n = 2, 6%), or unknown causes (n = 1, 3%). The cause of the effusion was not related to the type of transplantation (P = .953) or to the primary disease (P = .76).
Effusion Characteristics and Treatment Parameters
Twenty-two (56%) of 39 treated effusions were moderate in size, 12 (31%) were small, and five (13%) were large. Duration of catheter drainage ranged from 2 to 44 days (median, 6 days). Tube duration was unrelated to effusion etiology (P = .408). Total fluid output ranged from 110 to 9,726 mL (median, 1,350 mL). Twenty-one (54%) of the treated effusions were loculated. The presence or absence of loculation was not related to effusion etiology (P = .429). Nine patients with 12 treated effusions required placement of multiple catheters for effective drainage; the number of sequentially placed tubes in these patients ranged from two to seven (median, three tubes per patient). Number of tubes (multiple, single) was not related to effusion etiology (P = .082). Streptokinase was administered in nine patients (29%) either once (n = 3), twice (n = 5), or three times (n = 1). Talc sclerotherapy was performed in two patients and resulted in the only tube-related complication encountered— a single case of acute respiratory distress syndrome.
Response
Chest radiographs were obtained in all patients immediately after tube removal. When these were compared with the radiographs obtained before tube placement, 15 (38%) of 39 treated effusions showed a complete response, 22 (56%) showed a partial response, and two (5%) showed no response (treatment failure) (percentages do not add up to 100% due to rounding). Chest radiographs or CT scans obtained 1 month after tube removal were available for only 31 (79%) of 39 treated effusions. One-month follow-up information was not obtained for eight effusions due to patient death (n = 4), interval placement of large-bore surgical tubes (n = 3), or an interval decortication procedure (n = 1). At 1 month, 11 (35%) of 31 treated effusions showed a complete response, 18 (58%) showed a partial response, and two (6%) showed no response (treatment failure) (percentages do not add up to 100% due to rounding).
Chest radiographs or CT scans obtained 3 months after tube removal were available for only 28 (72%) of 39 treated effusions. Three-month follow-up information was not obtained for an additional three effusions owing to patient death (n = 2) or interval hemothorax caused by transbronchial lung biopsy (n = 1). At 3 months, 22 (79%) of 28 treated effusions showed a complete response, five (18%) showed a partial response, and one (4%) showed no response (treatment failure) (percentages do not add up to 100% due to rounding). Immediate and 1- and 3-month response rates, respectively, were not related to the cause of the effusion (P = .81, .82, and .535), nor were they related to the type of transplantation (P = .419, .568, and > .999).
Survival
Sixteen (52%) of the 31 study patients died during the follow-up period, which ranged from 0.6 to 47 months (median, 16 months). However, only nine died within 6 months of pleural intervention. The other seven patients died 11–36 months (median, 19.3 months) after treatment, and because there was evidence of complete radiologic and clinical resolution of pleural disease in these cases, these deaths were considered to be unrelated to the pleural disease or intervention and were not further considered. Among the nine deaths that occurred within 6 months of pleural intervention, there were no statistically significant associations between death and tube duration (P = .233), primary disease that led to transplantation (P > .99), or type of transplantation (P = .7). Although most (seven) of the 10 patients who died had empyema or complicated parapneumonic effusion (Table), this association did not reach statistical significance (P = .07).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
New or persistent pleural effusions that are observed beyond the first few weeks after transplantation are, however, a cause for greater concern. Potential etiologies in this setting include empyema or parapneumonic effusion, acute rejection, organizing pleural hematoma, lymphoproliferative disorder, and cardiac or renal failure. In our study, approximately half of patients with treated pleural effusions were found to have infection—either empyema or complicated parapneumonic effusion—as an etiology. The remainder of effusions in our series were caused by acute lung rejection or cardiac or renal failure.
There are few published data on management of pleural complications in lung transplant recipients. The recent trend at our institution is to aggressively diagnose and treat these effusions, often with image-guided small-bore catheter drainage. The results of our study suggest that this mode of treatment is quite effective, with either partial or complete resolution of fluid collections in more than 90% of cases. However, because the effusions were frequently loculated, effective drainage often required the placement of multiple tubes and the initiation of lytic therapy with streptokinase. Drainage was also typically prolonged—up to 44 days in one case—for these effusions, as compared with the drainage time required to treat malignant pleural effusions in our experience (8). The rationale for the aggressive management approach at our institution is three-fold: To enable the exclusion, treatment, or prevention of infection; to improve pulmonary function; and to prevent fibrothorax.
Empyema is a potentially devastating complication in lung transplant recipients. This is especially true for bilateral lung transplant recipients, in whom both pleural spaces may directly communicate (9,10). Empyema in one side of the thorax must be rapidly diagnosed and adequately drained so that infected pleural contents do not contaminate the contralateral pleural space. Mortality is high in lung transplant recipients who develop empyema and ranges from 28% to 43% (11,12). In our series, seven (44%) of 16 patients with pleural fluid caused by infection died within 6 months of pleural intervention. Unlike Herridge et al (12), we did not find a correlation between pleural complications and type of transplantation (single vs bilateral), nor did we find a significantly increased mortality rate in patients with empyema compared with those with parapneumonic effusion or rejection.
Percutaneous drainage of infected pleural fluid collections is clearly the standard of care. However, drainage of noninfected pleural fluid is more debatable. We are frequently asked to drain noninfected pleural fluid in lung transplant recipients to improve pulmonary function. While it is clear that large effusions can compromise hemodynamic function and impair cardiac filling (13,14), there is evidence that even small effusions can cause significant respiratory compromise and hypoxemia owing to intrapulmonary shunting (15). Thus, evacuation of even small amounts of pleural fluid may relieve dyspnea and improve lung function (16). We are also asked to drain noninfected pleural fluid collections because of concern that they might become secondarily infected. Such a phenomenon is reported in immunocompetent hosts and can lead to septicemia (17,18).
A further rationale for drainage is concern that persistent undrained pleural fluid in lung transplant recipients may result in fibrothorax, as can be seen with other clinical entities associated with persistent pleural effusions (19–23) and has been found at autopsy in heart-lung transplant recipients (24). Fibrothorax in lung transplant recipients can cause shortness of breath (20,21), chest pain (20,25), and a decline in forced expiratory volume in 1 second that may be misinterpreted as evidence of rejection (20,21,26,27).
It must be noted, however, that all of these rationales for drainage of noninfected pleural fluid in lung transplant recipients are theoretical. There is no conclusive evidence in the lung transplantation literature that aggressive pleural drainage prevents fibrothorax or secondary infection or improves pulmonary function. Because our study was retrospective, we cannot definitively answer these questions. We also did not conduct extended follow-up to evaluate whether pleural intervention affected the development of fibrothorax. To some degree, we had a small "built-in" control group: 12 patients with undrained contralateral effusions. Interestingly, most of these effusions had either partially or completely resolved at 90 days. These data suggest that these noninfected effusions may well have resolved without pleural intervention. However, since both pleural spaces are in direct communication in bilateral lung transplant recipients, the "untreated" effusions in nine patients may have been partially drained by the contralaterally inserted tube (9,10).
In conclusion, image-guided small-bore catheter drainage is an effective means of evacuating the pleural space in lung transplant recipients with persistent pleural effusions. Treatment is often difficult because drainage is frequently prolonged and may require the insertion of multiple tubes and the initiation of intrapleural lytic therapy. Furthermore, despite the adequate drainage in patients with parapneumonic effusions or empyema that was observed in our study, our results suggest that lung transplant recipients with pleural space infections have substantial short-term mortality. We cannot reach any firm conclusions regarding the clinical utility of this mode of treatment in patients with noninfectious effusions. A randomized controlled study would need to be conducted to determine whether our aggressive approach leads to improved long-term clinical outcomes in patients with noninfectious effusions.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. M. Wahidi, D. A. Willner, L. D. Snyder, J. L. Hardison, J. Y. Chia, and S. M. Palmer Diagnosis and Outcome of Early Pleural Space Infection Following Lung Transplantation Chest, February 1, 2009; 135(2): 484 - 491. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Boffa, D. P. Mason, J. W. Su, S. C. Murthy, J. Feng, A. M. McNeill, M. M. Budev, A. C. Mehta, and G. B. Pettersson Decortication After Lung Transplantation Ann. Thorac. Surg., March 1, 2008; 85(3): 1039 - 1043. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
