| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Pulmonary Medicine and Infectious Diseases, Vanderbilt University, Nashville, TN.
Correspondence to: Richard W. Light, MD, FCCP, Vanderbilt University Medical Center, T-1218 Medical Center North, Nashville, TN 37232; e-mail rlight98{at}yahoo.com
Abstract
Objectives: There have been no controlled studies to test the efficacy of tissue plasminogen activator (tPA) or recombinant human deoxyribonuclease (rhDNase) in the treatment of empyema. In vitro studies show that streptokinase without rhDNase does not liquefy empyemic material from rabbits. However, the combination of streptokinase and streptodornase and rhDNase have been shown to liquefy pus in vitro. The aim of this study was to determine if tPA or rhDNase, or a combination of the two, is more effective than saline solution in the treatment of empyema in rabbits.
Materials and methods: Empyema was induced in rabbits using 109 Pasteurella multicoda organisms in infusion agar injected via a surgically placed chest tube. Once empyema was verified, a blinded investigator administered one of four treatments via the chest tube: 3 mL of saline solution; 4 mg of alteplase (recombinant tPA); 1 mg of rhDNase; or 4 mg of alteplase and 1 mg of rhDNase. The rabbits received a treatment every 12 h following the initial for a total of six treatments. The animals were killed at day 10, and the degree of empyema and pleural peel was scored macroscopically on a scale of 0 to 4.
Results: The combination group had a significantly lower mean empyema score (1.83 ± 0.75) than did the saline solution group (3.86 ± 0.38, p = 0.001), rhDNase group (3.17 ± 0.75, p = 0.012), and alteplase group (3.71 ± 0.49, p = 0.02) [± SD]. The total volume of pleural fluid was markedly higher in the alteplase and rhDNase combination group (142 ± 79.1 mL) or the alteplase group (231 ± 78.0 mL) compared to either the rhDNase group (0.8 ± 1.6 mL) or the saline solution group (5.8 ± 14.0 mL).
Conclusion: The combination of alteplase and rhDNase is more effective in the treatment of rabbit empyema than either agent alone. The intrapleural injection of alteplase alone or in combination with rhDNase leads to the production of large amounts of pleural fluid.
Key Words: empyema • fibrinolytic • human recombinant deoxyribonuclease • pleural • rabbit • therapy • tissue plasminogen activator
The annual incidence of parapneumonic effusions in the United States is approximately 300,000, and approximately 10% of these effusions require aggressive therapy for resolution.1 In one study,2 the mortality rate of pneumonia patients with a parapneumonic effusion was 3.4 times higher than for those patients without an effusion. Pus in the pleural space increases the risk of death even further. Management of empyema involves the use of appropriate antibiotics as well as drainage of the pleural fluid.
Once pleural fluid becomes loculated, its drainage is difficult. One possible therapeutic endeavor to facilitate drainage is the intrapleural injection of fibrinolytics.1 The intrapleural injection of fibrinolytics and deoxyribonuclease (DNase) was first studied > 50 years ago by Tillet et al,3 who concluded that the intrapleural injection of the combination of streptokinase (a fibrinolytic) and streptodornase caused chemical lysis of intrapleural adhesions in empyema patients and facilitated drainage.
The role of fibrinolytics in the management of loculated parapneumonic effusions is controversial. In the past year, one randomized, double-blind, controlled study4 from South Africa with 44 patients concluded that intrapleural streptokinase reduces the need for surgery and improves the clinical treatment success in patients with loculated parapneumonic effusions. In contrast, a randomized, double-blind, saline solution-controlled, multicenter study5 with 454 patients in the United Kingdom was unable to demonstrate that intrapleural streptokinase had any advantages over saline solution. A metaanalysis6 of four randomized studies published prior to the publication of the above two reports concluded that intrapleural fibrinolytic therapy confers significant benefit when compared with normal saline solution; but on the basis of randomized controlled trial evidence alone, the routine use of fibrinolysis was not recommended because the numbers are too small.
A promising alternative for enzymatic debridement in empyema patients is recombinant human DNase (rhDNase) with or without the addition of a fibrinolytic. The purity of this recombinant enzyme makes it more attractive than the crude streptococcal preparation, streptodornase, which is known for its antibody mediated allergic reactions. An in vitro study published in 2000 by Light et al7 showed that streptokinase or urokinase alone had no effect in liquefying empyemic material from rabbits, but the combination of streptokinase and DNase completely liquefied the material within 4 h. Simpson et al also found that rhDNase alone was effective in vivo in decreasing the viscosity of human empyema pus,8 and was effective in one patient.9
We hypothesized that the administration of recombinant tissue plasminogen activator (tPA) [alteplase] alone or in combination with rhDNase would be an effective adjunct in the treatment of empyema. The aim of our study was to assess the in vivo effectiveness of alteplase and rhDNase, both alone and in combination, vs saline solution alone in the treatment of loculated empyema in a rabbit empyema model.
Materials and Methods
Empyema Model
The protocol was approved by the animal care committees of Vanderbilt University, Nashville, TN. Empyema was induced in 1.5- to 2.0-kg New Zealand white rabbits in a manner similar to that described previously.1011 After a chest tube was inserted into the right pleural space using methods previously described,12 an empyema was induced by the intrapleural injection of 109 Pasteurella multicoda bacteria in 5 mL of 0.5% brain heart infusion agar.1011 The animals were housed in special, infection-containment, negative-pressure rooms, and isolation precautions were used each time the rabbits were handled.
Twenty-four hours following the injection of the bacteria, a maximum of 2 mL of pleural fluid was removed for analysis to verify that an empyema was present. The protein and lactate dehydrogenase (LDH) levels were determined with an automated analyzer (Johnson & Johnson; Rochester, NY). The lower and upper limits of normal serum LDH are 120 IU/L and 200 IU/L, respectively. Total leukocyte count was measured using an automated counter (Coulter Electronics; Luton, UK), which was calibrated daily. The pleural fluid specimens were analyzed for pH with a blood gas machine and glucose (TheraSense; Alameda, CA). Empyema was diagnosed if the pleural fluid appeared grossly infected, if the glucose level was < 40 mg/dL, or if the pleural fluid pH was < 7.1. All rabbits also underwent an ultrasonic examination of the right pleural space to verify the presence of pleural fluid and loculations. If a rabbit did not have an empyema via analysis of the pleural fluid or did not have pleural fluid with loculations, the animal was killed and was not included in the study. All rabbits received 100,000 U of procaine penicillin (Anthony Products Company; Irwindale, CA) IM q24h starting 24 h after the bacteria were injected until death to prevent systemic infection and early death and to mimic the clinical situation in humans.
Intrapleural Treatment
Once an empyema was verified, animals received an initial treatment via the chest tube by an observer (H.L.M.) blinded to the treatment group. The treatments were prepared by an unblinded person not otherwise involved in the experiment (Y.G.). They were prepared in syringes labeled only with the identification number of the rabbit with a final volume of 3 mL. The syringes contained either saline solution, 1 mg of rhDNase (Genentech; San Francisco, CA), 4 mg of alteplase (Genentech), or 1 mg of rhDNase plus 4 mg of alteplase. Injections were performed every 12 h for 3 days, a total of six injections per rabbit. Prior to each injection, the right pleural space was aspirated dry and the volume of the aspirated fluid was recorded. The chest tubes were left in place and were aspirated at 12-h intervals until the rabbits were killed.
Pleura and Empyema Scoring
All rabbits were killed on day 10 following chest tube placement. Anesthesia was initially induced with IM injection of 35 mg/kg of ketamine hydrochloride and 5 mg/kg of xylazine hydrochloride (Ben Venue Laboratories; Bedford, OH). After this initial anesthesia was achieved, the animal was killed in a closed CO2 chamber. The thorax was dissected from the carcass, and the chest was bisected along a coronal plane from the diaphragm to the neck for examination. An empyema score and a pleura score was then assigned by an examiner (R.W.L.) who has experience with this scoring system and who was also blinded to the treatment that each animal received. The empyema scoring system with scores of 0 to 4 was used to grade the degree of empyema seen grossly at autopsy (Table 1
). The pleura score was obtained by assessing the amount of pleural adhesions and pleural peel (Table 1) on the right side (the left side was used as the control).
|
Results
Chest tubes were placed in 40 rabbits, but 4 rabbits were excluded from further study because the chest tubes inadvertently came out of the pleural space. Empyema developed in the remaining 36 rabbits after the intrapleural injection of 109 P multicoda bacteria. The pleural fluid 24 h after the bacteria were injected was thick and grossly opaque in all. The pH level of the pleural fluid was < 7.10, the glucose level of the pleural fluid was < 40 mg/dL, and the LDH level was > 4,000 IU/L in all 36 rabbits.
Ten rabbits died prior to their scheduled death on day 10 (Table 2 ). Four rabbits that received alteplase alone or in combination with rhDNase died of respiratory failure with massive bilateral pleural effusions. The average amount of pleural fluid drained from the right hemithorax daily before death was > 45 mL. Two rabbits that received alteplase alone died of a large hemothorax. The other four rabbits died of diarrhea and sepsis. The mortality of rabbits that received alteplase alone or combined with rhDNase (8 of 21 animals) and of rabbits that did not receive alteplase (2 of 15 animals) did not differ significantly (p = 0.142).
|
|
|
|
Discussion
The present study demonstrates that the intrapleural administration of the combination of alteplase and rhDNase results in a significantly lower empyema score in a rabbit model of empyema than either agent alone. Neither agent by itself was effective in reducing the empyema score, although the intrapleural administration of alteplase alone resulted in the greatest amount of pleural drainage. To our knowledge, this is the first controlled study to demonstrate the efficacy of alteplase combined with rhDNase as therapy for empyema in a rabbit model. The present study also demonstrates that the intrapleural administration of alteplase alone or in combination with rhDNase results in the production of large amounts of pleural fluid. In some animals, the large amounts of pleural fluid caused acute fatal respiratory distress.
Management of empyema is frequently problematic. Since pus in the pleural space increases the risk of death, the drainage of the pus is at least as important as the use of appropriate antibiotics in management of empyema. Drainage is more difficult when the pleural fluid is loculated. The reasoning behind the use of fibrinolytics in the management of loculated effusions is that the loculations are produced by fibrin membranes and the fibrinolytics could lyse the fibrin membranes and facilitate drainage of the pleural fluid. Streptokinase and alteplase both achieve fibrinolysis through the activation of endogenous plasminogen. Plasminogen in turn can be activated to plasmin that breaks down fibrin. While alteplase activates plasminogen directly,13 streptokinase actives plasminogen only after forming an intermediate activator complex with plasminogen.14 Drainage is also more difficult when the pus is very thick. The reasoning behind the use of a DNase is that it reduces viscosity by fragmenting the free uncoiled DNA found in pus.78
When a patient with a complicated parapneumonic effusion or empyema is treated, there are three areas in which fibrinolytics or other enzymes might facilitate drainage: (1) a fibrinolytic could break down the fibrin membranes responsible for producing the loculations; (2) the DNase could liquefy the thick pus as shown in the experiments of Simpson et al89 and Light7; and (3) the fibrinolytic and/or the DNase could debride the peel covering the visceral pleura, allowing the underlying lung to expand.
It has been over half a century since Tillet et al3 demonstrated the beneficial effects of intrapleural enzyme treatment. Streptokinase, the most commonly used fibrinolytic agent in the last 50 years, has the potential to cause antibody-mediated systemic side effects. tPA is an effective systemic fibrinolytic agent that is less likely to cause systemic antibody-mediated reactions than streptokinase. As a recombinant tPA with high fibrin specificity, alteplase has some advantages over the first-generation fibrinolytic agents (streptokinase, urokinase). It has been commonly used in the treatment of acute myocardial infarction and stroke,1516171819 and has also been used off-label in the treatment of loculated parapneumonic effusion.202122
The present study demonstrates that alteplase alone was ineffective in decreasing the empyema score or pleural score in our rabbit model of empyema. Our results are in agreement with those of the recently completed multicenter study5 from the United Kingdom, in which streptokinase was no more effective than saline solution in the treatment of loculated parapneumonic effusions in humans. A consensus statement23 by the American College of Chest Physicians that was written before the results of the multicenter study were available concluded that fibrinolytics, thoracoscopy, or thoracotomy were all acceptable options for treating patients with complicated parapneumonic effusions. It should be noted that most of the previous studies on fibrinolytics for complicated parapneumonic effusions have been uncontrolled. However, the results in several randomized controlled studies4242526 with smaller number of patients have been positive. In an editorial concerning fibrinolytics, Heffner27 pointed out some shortcomings of the multicenter UK study,5 but concluded that existing data favor reserving fibrinolytic therapy for patient who are poor surgical candidates or who are in health-care settings where surgical interventions are unavailable. The present study should raise additional doubts about the efficacy of fibrinolytics in the treatment of loculated parapneumonic effusions.
The present study also shows that rhDNase alone was ineffective in decreasing the empyema score or the pleural score in our rabbit model. rhDNase is much purer than streptodornase, which makes it more attractive than the crude streptococcal preparation. Simpson et al8 found that the rhDNase reduced viscosity markedly; rhDNase and the streptokinase-streptodornase combination were effective in liquefying human empyema pus, decreasing the respective viscosities by 93.4% and 94.8% compared to saline solution. Similar results were obtained in rabbit empyema pus using the streptokinase-streptodornase combination.7 Despite the results of the in vitro studies,78 the present study does not provide any evidence that rhDNase alone is effective in the treatment of empyema. It should be noted that there are no controlled studies evaluating the effectiveness of DNase in the treatment of patients with complicated parapneumonic effusions.
Although alteplase or rhDNase alone did not improve the drainage of the intrapleural pus, the combination of alteplase and rhDNase significantly improved the results in our rabbit model of empyema (Table 3, Fig 2). A lower pleural score was also noted in the rhDNase and alteplase combination group than any other group, but the differences did not reach statistical significance. The results of the present study give support to the need for controlled trials evaluating the combination of a fibrinolytic and a DNase in the treatment of complicated parapneumonic effusions.
Similar to published data,18202122 we noted that the intrapleural administration of alteplase with or without rhDNase led to a marked increase in the volume of fluid drained from the pleural space. Indeed, the increased volumes were of sufficient magnitude that they appeared to contribute to the death of four rabbits that received alteplase either alone or in combination with rhDNase and died with massive bilateral pleural effusions. While the intrapleural administration of alteplase results in the generation of large amounts of pleural fluid, the mechanism for this increased production of pleural fluid is unknown. We believe that as a pleiotropic agent, tPA can result in the increased production of pleural fluid via the induction of cytokines such as vascular endothelial growth factor.
There appeared to be a higher mortality rate in rabbits that received alteplase. Although the difference in mortality was not statistically significant (8 of 21 animals vs 2 of 15 animals, p = 0.142), the power of the test to detect a statistically significant difference was only 0.357. Three rabbits that received alteplase alone and one rabbit that received the combination of alteplase and rhDNase died of respiratory failure and bilateral pleural effusions prior to day 10 following empyema induction. Although a large volume of pleural fluid was aspirated from the right hemithorax of these rabbits daily, at autopsy massive bilateral pleural effusions with lung collapse were found. The mechanism for the bilateral pleural effusions in these rabbits is not known. Possibilities include sepsis or rupture of the mediastinum due to the high pleural pressure from the large effusion on the right side. The observation that large volumes of pleural fluid are induced by intrapleural tPA suggests that if intrapleural tPA is administered to humans, it should be in a setting where there is a chest tube in situ, and provisions should be made for the immediate unclamping of the tube if increasing shortness of breath develops.
Two rabbits that received alteplase died from a hemothorax. The hemothorax was possibly related to the intrapleural administration of the alteplase. It should be noted that the intrapleural administration of neither tPA nor DNase is approved in humans, and therefore until controlled studies are conducted evaluating their safety and efficacy they must be used with caution.
It is possible that the production of the large volumes of pleural fluid and the hemothoraces is due to the relatively large dose of alteplase administered (approximately 2 mg/kg per dose). The recommended dose for alteplase in the treatment of pulmonary embolism in a 70-kg adult is 100 mg administered IV in a graded fashion over 2 h (according to the recommendations of Genentech). By these standards, our rabbits received a relatively high dose of alteplase.
There are several limitations to our study. There were only a small number of rabbits in each group. This makes the power of the statistical tests low. This may be particularly pertinent when the mortality rates or the pleural scores are analyzed. We did not save the fluid from the animals with large amounts of pleural fluid. Therefore, it is more difficult to analyze the mechanisms responsible for the large amount of pleural fluid. In addition it remains to be determined whether the results observed in rabbits in the present study can be extrapolated to humans. We did not assess the effects of intrapleural tPA on the systemic bleeding parameters. However, none of the rabbits had any evidence of systemic bleeding. The present model differed from the usual case of the patient with a loculated empyema since most have an underlying pneumonia and in the present model, the primary infection is in the pleural space. However, the significance of this difference is minimized since the main goal of the present study was to assess the effects of the therapeutic agents on pleural drainage.
In conclusion, this study showed the following: (1) intrapleural administration of the combination of alteplase and rhDNase significantly improved the empyema score compared with either agent alone or with saline solution; (2) intrapleural administration of alteplase alone or in conjunction with rhDNase increased the volume of pleural fluid markedly; and (3) intrapleural administration of six doses of 2 mg/kg of alteplase with or without rhDNase was associated with a high mortality in an empyema model of rabbit.
Footnotes
Abbreviations: ANOVA = analysis of variance; DNase = deoxyribonuclease; LDH = lactate dehydrogenase; rhDNase = human recombinant deoxyribonuclease; tPA = tissue plasminogen activator
This study was supported in part by the Saint Thomas Foundation, Nashville, TN.
The alteplase and rhDNase used in these experiments were provided by Genentech, Inc., San Francisco, CA.
Received for publication January 11, 2005. Accepted for publication November 29, 2005.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
