HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Ashis K. Mandal Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mandal, A. K. Articles by Chettipally, U. Search for Related Content PubMed PubMed Citation Articles by Mandal, A. K. Articles by Chettipally, U.

Ann Thorac Surg 1998;66:1782-1786
© 1998 The Society of Thoracic Surgeons

---

Original articles: general thoracic

Outcome of primary empyema thoracis: therapeutic and microbiologic aspects

^a Division of Cardiothoracic Surgery, Martin Luther King Jr. General Hospital, Charles R. Drew University School of Medicine and Sciences, Los Angeles, California, USA

Accepted for publication May 21, 1998.

Address reprint requests to Dr Ashis Mandal, Division of Cardiothoracic Surgery, King/Drew Medical Center, 12021 S Wilmington Ave, Los Angeles, CA 90059

	Abstract

Top Abstract Introduction Patients and methods Results Comment Acknowledgments References

 
Background. This study was undertaken to determine whether all adult patients with primary empyema thoracis need decortication.

Methods. A management algorithm was developed and analyzed in a prospective, longitudinal, nonblinded study of 179 consecutive adult patients. The treatment options included thoracentesis, closed (tube) thoracostomy, image-guided catheter drainage, and decortication. We reviewed the outcomes of these procedures as they related to the pleural fluid cultures isolated and the antibiotic regimens used.

Results. Of the 179 patients, 20 had thoracentesis as the primary procedure, and 18 (90%) were cured. Ninety patients underwent closed thoracostomy as the primary procedure with a cure rate of 62% (56 patients) and a mortality rate of 11% (10 patients), and 24 patients required a secondary procedure. Seventy-six patients underwent decortication as either the primary or the secondary procedure with a cure rate of 88% (67 patients) and a mortality rate of 1.3% (1 patient); 8 patients required conversion to open thoracostomy. Hospital stay for decortication was 14 ± 1 days and for closed thoracostomy, 17 ± 1 days (p < 0.05). Decortication was necessary in 55% of patients with anaerobic infections and in 50% with aerobic infections. Clindamycin in combination with gentamicin sulfate was the most efficacious regimen with a success rate of 82% (51 of 62 patients); only 33% (17 of 52) were cured with penicillin. The overall mortality rate in this study was 6.7% (12 of 179 patients).

Conclusions. Forty-two percent of patients with primary empyema thoracis ultimately require decortication. Decortication is more frequently necessary for anaerobic, tuberculous, staphylococcal, and pneumococcal infections. Although the overall mortality in this study was low, mortality remains high in elderly patients and patients with comorbid disease.

	Introduction

Top Abstract Introduction Patients and methods Results Comment Acknowledgments References

 
Empyema thoracis can be classified as either primary or secondary. Primary empyema thoracis accounts for about half of all empyema cases and includes pleural effusion caused by pleuropulmonary inflammation. Secondary empyema thoracis includes postoperative complications of thoracic operations (25%), sequelae of thoracic trauma (10% to 15%), and extension of a suppurative process from either the neck or the abdomen (8% to 10%).

Treatment of empyema thoracis is twofold: to drain the pus from the pleural cavity to achieve full expansion of the lung and to treat the infection with antimicrobial agents. A number of drainage procedures have been employed: thoracentesis [1], image-guided catheter drainage [2], closed tube thoracostomy, decortication, empyemectomy, thoracoscopic debridement, posterior rib resection, and others [3, 4]. We report our experience with the management of primary or postpneumonic empyema of the thorax in a single institution during the past 24 years.

	Patients and methods

Top Abstract Introduction Patients and methods Results Comment Acknowledgments References

 
This prospective, longitudinal, nonblinded study involves 179 consecutive adult patients seen with a diagnosis of primary bacterial thoracic empyema from July 1973 to July 1997. There were 141 men and 38 women aged 19 to 82 years (mean age, 40 ± 1 years [± standard error of the mean]). There were 131 black patients (73%), 43 Latino patients (24%), and 5 patients of other races (3%). From 1987 when the test for human immunodeficiency virus infection became widely used, 81 consecutive patients with empyema thoracis were tested, and only 2 (25%) were positive for that infection.

All patients had diagnostic and therapeutic thoracentesis, and the algorithm used for drainage procedures is depicted in Figure 1. When the pleural fluid was thin and the effusion was small to moderate (500 to 800 mL) without loculation, drainage was accomplished with thoracentesis alone. If the aspirate was turbid or massive (1,500 to 2,000 mL) or associated with a bronchopleural fistula, closed thoracostomy (tube thoracostomy) was used. Size of the chest tube was 34F to 40F.

View larger version (31K): [in this window] [in a new window]   Fig 1. Algorithm for management of empyema thoracis and final outcome for various drainage procedures in 179 consecutive patients with primary adult empyema thoracis. (CT = computed tomographic.)

Image-guided (computed tomography, ultrasonography, or fluoroscopy) catheter drainage was done as the primary procedure in 9 patients. The following criteria were used: a single loculated collection of fluid compromising lung expansion by not more than 15% on split pulmonary perfusion tests; inaccessibility for draining by tube thoracostomy; reaccumulation of fluid after decortication; or incomplete drainage after tube thoracostomy.

Closed thoracostomy was converted to open thoracostomy in 16 patients when there was continued drainage of pus or persistence of bronchopleural fistula after either tube thoracostomy or decortication. Open thoracostomy was usually done 12 to 14 days after the initial procedure, and the patient was discharged with a colostomy bag covering the cut end of the chest tube. The patient was taught how to irrigate the open thoracostomy with an antibiotic solution (polymyxin B sulfate, 5,000,000 units, and bacitracin, 50,000 units, in 500 mL of normal saline solution).

Cultures were grown from all pleural fluid samples by conventional methods for aerobic bacteria and by Virginia Polytechnic Institute methods and anaerobic glove-box procedures for anaerobic bacteria [5]. Fungi were cultured on Sabouraud’s medium. All bacterial isolates were later tested in vitro for their susceptibility to various antibiotics.

Empiric antibiotic regimens included clindamycin with an aminoglycoside, ticarcillin disodium with clavulanic acid (Timentin), and either penicillin or cefazolin sodium with or without an aminoglycoside [6]. When Staphylococcus aureus was isolated, vancomycin or nafcillin sodium was used. All antibiotics were given intravenously for at least 7 days. A successful outcome was determined by negative cultures, improvement in the chest roentgenogram, and a sense of well-being. Clinical follow-up of these patients was done for up to 6 months to 1 year.

	Results

Top Abstract Introduction Patients and methods Results Comment Acknowledgments References

 
Surgical outcomes
As summarized in Figure 1, 179 patients underwent 214 therapeutic procedures. Of the 20 patients who had thoracentesis, 18 (90%) were cured, and 2 (10%) subsequently required decortication as a secondary procedure. Nine patients were treated by image-guided catheter drainage with a success rate of 89% (8 of 9). The patient in whom this method failed required secondary decortication.

As shown in Figure 1, most of the patients (n = 90) were treated initially with closed thoracostomy. Fifty-six (62%) of them were cured, and 10 (11%) died. The method failed in 24 patients (27%); 13 underwent secondary decortication, 8 required conversion to open thoracostomy because of bronchopleural fistula, and 3 underwent image-guided catheter drainage. Of the 8 patients who had conversion to open thoracostomy, 7 were cured, and 1 died.

Seventy-six patients underwent decortication as a primary or secondary procedure. Of these patients, 67 (88%) were cured, 8 (11%) had development of bronchopleural fistula requiring open thoracostomy, and 1 (1.3%) died. Of the 8 with development of a bronchopleural fistula, 4 were in the primary decortication group and 4 in the secondary group. All 8 patients were cured with open thoracostomy.

Microbiology
Pleural fluid was obtained from all patients and analyzed for organisms. One hundred thirty-six patients (76%) had positive pleural fluid cultures, and 43 (24%) had sterile isolates. The most common group comprised purely anaerobic pleural isolates, and was found in 26 (15%) of all 179 patients. In an additional 16 patients, (9%) pleural fluid cultures grew mixed aerobic and anerobic bacteria. In decreasing order of frequency, the most commonly encountered anaerobes were Bacteroides, Fusobacterium, Peptostreptococcus, Peptococcus, Acidaminococcus, Veillonella, and Actinomyces. Therefore, 42 (23%) of all 179 patients had anaerobic organisms, and 23 (55%) of these 42 required decortication.

The next most common organisms were Mycobacterium tuberculosis in 22 patients (12%) or purely staphylococcal organisms in 21 (12%). Ten (45%) of the patients with empyema secondary to M tuberculosis and 11 (52%) of the patients with purely staphylococcal empyema required decortication.

Gram-negative bacilli, identified in 17 patients, accounted for 9%. The gram-negative isolates were Pseudomonas, Escherichia coli, Klebsiella, Enterobacter, and Haemophilus influenzae. Purely pneumococcal isolates from the pleural fluid were the next most frequent, found in 12 patients (7%), 6 (50%) of whom required decortication.

Seventeen (9%) of the 179 patients had positive blood cultures. Of note, pneumococcus was found in the blood of 5 (42%) of the 12 patients with purely pneumococcal pleural isolates. Interestingly, 1 patient with sterile pleural fluid had bacteremia with E coli and Proteus mirabilis found in the blood. Organisms isolated in the blood of the other 11 patients were as follows: Staphylococcus (3 patients), gram-negative bacilli (3 patients), Streptococcus (2 patients), mixed aerobic and anaerobic (2 patients), and anaerobic (1 patient). A mixture of aerobic gram-positive cocci were found in 13 patients (7%), and 7 patients (4%) had Streptococcus other than pneumococcus. Finally, in both patients with amebic empyema thoracis bacterial superinfections developed: one with S aureus and one with enterococcus.

Antibiotic therapy
A combination of clindamycin and gentamicin sulfate proved to be the most efficacious antibiotic regimen in our study when used as either first-line or second-line therapy. Of the 62 patients treated with agents in combination, 51 (82%) were cured. Twenty-six patients received either a first- or second-generation cephalosporin, and 14 (54%) were cured. Twenty-one patients received ticarcillin disodium in combination with clavulanic acid, and 11 (52%) were cured.

Penicillin proved to be the least efficacious agent. Intravenous penicillin was used as the first-line antibiotic in 52 patients and never as the second-line drug. It failed in 35 (67%) of these 52 patients despite adequate surgical drainage. The most common organisms in cultures from these 35 patients included anaerobic organisms either alone or mixed with aerobic bacteria (10 patients) and Staphylococcus (7 patients). Cultures from the remaining 18 patients grew gram-negative bacilli, enterococcus, or ß-lactamase–producing streptococci.

Hospital stay and mortality
The length of stay for decortication (14 ± 1 days) was significantly different from that for closed thoracostomy (17 ± 1 days) when analyzed by the group t test (p < 0.05). The mean duration of hospitalization between a primary and a secondary procedure (cross-over patients) was 12 ± 1 days.

Over all, there were 12 deaths, a mortality rate of 6.7%. As shown in Figure 1, 11 patients originally in the closed thoracostomy group died and 1 patient in the decortication group. Six patients died within 3 days and the other 6, within 2 weeks. Overwhelming sepsis was the cause of death in 7 patients, and the organisms isolated in these patients included gram-negative bacilli (3 patients), Staphylococcus (1 patient), mixed aerobic and anaerobic bacteria (1 patient), pure anaerobic (1 patient), and 1 patient with none. The other 5 patients died of comorbid disease: cardiomyopathy with renal failure (3 patients) and pulmonary embolism (2 patients).

Of the 12 patients in this study who were 65 years or older, 6 died, a mortality rate of 50%. In contrast, only 6 of the 167 younger patients died, a mortality rate of less than 4%.

	Comment

Top Abstract Introduction Patients and methods Results Comment Acknowledgments References

 
The suppurative process in the pleural space can be diffuse or localized. The pathologic stages are exudative, fibrinopurulent, and organizing and correlate well with the clinical phases—acute, transitional, and chronic. Thus, the choice of drainage procedure should be based on the pathologic and clinical phases of the suppurative processes [6, 7], not on the biochemical variables. During the acute (exudative) phase, thin fluid can be aspirated, and this may be adequate as both a diagnostic and a therapeutic procedure [6, 8]. During needle aspiration, as much of the pleural fluid collection as possible must be drained, and repeated thoracentesis should be discouraged. Fifteen percent of our patients (26 of 179) required no more than simple needle aspiration or image-guided catheter drainage for cure.

Surgical treatment should be considered when the fluid is thick and difficult to aspirate even with a large-bore needle. One hundred fifty patients (84%) had thick pus requiring either closed thoracostomy or decortication. Turbid fluid associated with bronchopleural fistula, which may be a complication of pneumonia, lung abscess, or sepsis, and massive effusion compromising cardiorespiratory function need immediate chest tube thoracostomy. Ninety of our patients (50%) had closed thoracostomy as a primary procedure, and 8 of them also had bronchopleural fistulas.

When this inflammatory process is organized as a fibroelastic membrane layering out as a continuous sheet on both the visceral and the parietal pleura, which usually takes about 8 weeks to develop, decortication can be safely performed. By this time, there is less chance of creating a bronchopleural fistula or of bleeding caused by tearing underlying lung tissue. In contrast, during the fourth to sixth week, removal of an immature visceral pleural peel usually results in tearing the underlying lung tissue because of fibroblast adherence to the lung surface. Accordingly, 60 patients (34%) had primary decortication. It is customary to perform bronchoscopy and contrast computed tomography of the chest to assess the nature of the lesion in the underlying lung and to decide if pulmonary resection is needed with the decortication procedure [9]. In an attempt to avoid excessive bleeding, we do not routinely remove the parietal pleural peel during decortication. This peel is expected to be absorbed with time [10]. However, the parietal pleura may have to be removed in patients with long-standing empyema because it may restrict chest wall movement and prevent expansion of the underlying lung. It is not a cause for worry if the chest roentgenogram does not improve immediately, as radiographic resolution usually lags behind clinical improvements.

In our study, the rate of decortication was around 50% when the microbial etiology of empyema thoracis was anaerobic, staphylococcal, pneumococcal, or tuberculous. Bacteroides, M tuberculosis, and gram-positive cocci elicit an immediate polymorphonuclear response followed by active proliferation of mononuclear cells [11]. The mononuclear cells activate the fibroblast proliferation that eventually results in the formation of a fibroelastic membrane or pleural peel. On the other hand, the aerobic gram-negative bacilli such as Klebsiella and Pseudomonas elicit an acute inflammation associated with systemic toxicity that does not allow enough time for the initial polymorphonuclear cell response to be well organized before monocyte-mediated fibroblast proliferation. Thus, we believe if microbial cultures of pleural fluid yield anaerobic, staphylococcal, pneumococcal, or tuberculous organisms and if the clinical condition of the patient and the chest roentgenogram do not improve, early thoracoscopic debridement or decortication should be considered.

In contrast, decortication was done in only 2 (29%) of 7 patients with streptococcal empyema. This low frequency of decortication can perhaps be explained by the fact that streptococci produce proteolytic enzymes capable of converting plasminogen to plasmin. Plasmin in turn degrades fibrinogen and fibrin clots and thereby prevents development of a pleural peel. Therefore, intrapleural installation of fibrinolytic agents is also effective as an ancillary measure in the treatment of empyema thoracis [12, 13].

In the case of incomplete drainage or inadequate reexpansion of the lung after thoracentesis or image-guided catheter drainage or if there is persistent high chest-tube drainage associated with fever, either decortication or thoracoscopic debridement should be considered. Thirteen percent of patients (16 of 119) in this group had secondary decortication, but none of them had thoracoscopy, video-assisted thoracic surgery, or fibrinolytic therapy. Video-assisted thoracic surgical procedures can treat empyema thoracis effectively during the transitional (fibrinopurulent) phase, characterized by a "chicken-fat" type of drainage through the chest tube with incomplete resolution or multiple loculations. Cure can be achieved by lysing fibrinous adhesions, breaking off the loculations, and properly placing the chest tube under direct vision with the thoracoscope. On the other hand, if the underlying lung does not expand, thoracoscopy should be converted to formal thoracotomy for decortication. Video-assisted thoracic surgical procedures have the advantages of being less invasive, eliminating the need of decortication, and producing less postoperative pain; however, they are less helpful during the organizing phase or when the patient cannot tolerate single-lung ventilation.

Decortication has its own merits as well as its own complications, such as recurrence of empyema, bronchopleural fistula, bleeding, and death. Eight (11%) of 76 patients had development of a bronchopleural fistula after decortication. All healed spontaneously with open thoracostomy because postpneumonic bronchopleural fistulas are small and lung reexpansion obliterates the pleural space, unlike the case with postresectional bronchopleural fistulas. The mortality rate in the decortication group was 1.3% (1 of 76), which is very low compared with the reported mortality rates of 4% to 8% [14–16]. Eleven (14%) of 77 (12% of 90) patients died in the thoracostomy group. One could argue that these 11 patients might have survived if they had had a decortication procedure. However, this is highly speculative because of their serious comorbid diseases. It is known that the death rate is as high as 44% when there is an underlying illness in contrast to the low rate of 7% in the absence of any comorbid diseases [15]. In a previous study [17], the mortality rate was reported to be high at both extremes of life—28% for patients less than 12 years old and 75% for those older than 70 years—and 46% for patients between 30 and 49 years. A mortality rate of 8% among patients younger than 50 years and 30% among those older than this was reported in another study [18]. Our mortality rate was 50% for patients 65 years old or older and 3.6% among those younger than 65 years. Our overall mortality rate was 6.7%, which is comparable to the rate of 8% to 9% reported by others [16].

One hundred three patients (58%) were cured with less invasive procedures. Therefore, if we had adopted decortication as the primary drainage procedure for all patients with postpneumonic empyema, we would have done unnecessary surgical procedures on others. We recommend that therapeutic thoracentesis, closed thoracostomy, and thoracoscopic debridement not be abandoned just because of the shorter hospital stay associated with decortication. In our study, decortication was implemented as an alternative drainage procedure within 12 ± 1 days, whereas others [19] reported a waiting period for tube thoracostomy as long as 4 weeks.

Finally, antibiotics used for treatment of primary bacterial empyema should be effective against anerobic bacteria, staphylococci, and polymicrobial infections. Clindamycin plus gentamicin meets this requirement and is the most effective antibiotic regimen to date for us in the treatment of empyema thoracis. In this study, penicillin failed to eradicate ten of 13 anaerobic infections (77%) and all seven S aureus infections. Penicillin is not recommended as an antibiotic for empyema thoracis [6].

	Acknowledgments

Top Abstract Introduction Patients and methods Results Comment Acknowledgments References

 
We thank Ms Charvette Pilchier for her patience and for processing the manuscript, and Ms Mina R. Mandal for her assistance in our literature search.

	References

Top Abstract Introduction Patients and methods Results Comment Acknowledgments References

 

1. Storm H.K., Krasnik M., Bang K., et al. Treatment of pleural empyema secondary to pneumonia: thoracocentesis regimen versus tube drainage. Thorax 1992;47:821-824.[Abstract/Free Full Text]
2. Ulmer J.L., Choplin R.H., Reed J.C. Image-guided catheter drainage of the infected pleural space. J Thorac Imaging 1991;6:65-73.[Medline]
3. Samson P.C. Empyema thoracis: essentials of present-day management. Ann Thorac Surg 1971;11:210-221.[Medline]
4. Ridley P.D., Braimbridge M.V. Thoracoscopic debridement and pleural irrigation in the management of empyema thoracis. Ann Thorac Surg 1991;51:461-464.[Abstract]
5. Holdeman L.V., Cato E.P., Moore W.E.C. Anaerobe laboratory manual, 4th ed. Blacksburg, VA: Virginia Polytechnic Institute and State University, 1977:1-149.
6. Mandal A.K., Thadepalli H. Treatment of spontaneous bacterial empyema thoracis. J Thorac Cardiovasc Surg 1987;94:414-418.[Abstract]
7. Wells F.C. Empyema thoracis: what is the role of surgery?. Respir Med 1990;84:97-99.[Medline]
8. Berger H.A., Morganroth M.L. Immediate drainage is not required for all patients with complicated parapneumonic effusions. Chest 1990;97:731-735.[Abstract/Free Full Text]
9. Pugatch R.D., Faling L.J., Robbins A.H., et al. Differentiation of pleural and pulmonary lesions using computed tomography. J Comput Assist Tomogr 1978;2:601-606.[Medline]
10. Neff C.C., van Sonnenberg E., Lawson D.W., et al. CT follow-up of empyemas: pleural peels resolve after percutaneous catheter drainage. Radiology 1990;176:195-197.[Abstract/Free Full Text]
11. Antony V.B., Hadley K.J., Sahn S.A. Mechanism of pleural fibrosis in empyema: pleural macrophage-mediated inhibition of fibroblast proliferation. Chest 1989;95(Suppl 3):230S-231S.[Free Full Text]
12. Tillet W.S., Garner R.L. Fibrinolytic activity of hemolytic streptococci. J Exp Med 1933;58:485-502.[Abstract]
13. Robinson L.A., Moulton A.L., Fleming W.H., Alonso A., Galbraith T.A. Intrapleural fibrinolytic treatment of multiloculated thoracic empyemas. Ann Thorac Surg 1994;57:803-814.[Abstract]
14. Morin J.E., Munro D.D., MacLean L.D. Early thoracotomy for empyema. J Thorac Cardiovasc Surg 1972;64:530-536.[Medline]
15. Mayo P. Early thoracotomy and decortication for non-tuberculous empyema in adults with and without underlying disease. Am Surg 1985;51:230-236.[Medline]
16. Ashbaugh D.G. Empyema thoracis: factors influencing morbidity and mortality. Chest 1991;99:1162-1165.[Abstract/Free Full Text]
17. Finland M., Barnes M.W. Changing ecology of acute bacterial empyema: occurrence and mortality at Boston City Hospital during 12 selected years from 1935–1972. J Infect Dis 1978;237:274-291.
18. Wesse W.C., Shindler E.R., Smith I.M., et al. Empyema of the thorax then and now: a study of 122 cases over four decades. Arch Intern Med 1973;131:516-520.[Abstract/Free Full Text]
19. Miller J.I., Jr Empyema thoracis. Ann Thorac Surg 1990;50:343-344.[Medline]

This article has been cited by other articles:

				J. E. Heffner, J. S. Klein, and C. Hampson Interventional Management of Pleural Infections Chest, October 1, 2009; 136(4): 1148 - 1159. [Abstract] [Full Text] [PDF]

				T. F. Molnar Current surgical treatment of thoracic empyema in adults Eur. J. Cardiothorac. Surg., September 1, 2007; 32(3): 422 - 430. [Abstract] [Full Text] [PDF]

				M. L. C. V Rodriguez and G. T Catalan Outcome of Pediatric Empyema Thoracis Managed by Tube Thoracostomy Asian Cardiovasc Thorac Ann, April 1, 2006; 14(2): 98 - 101. [Abstract] [Full Text] [PDF]

				Managing empyema in adults DTB, March 1, 2006; 44(3): 17 - 21. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Ashis K. Mandal Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mandal, A. K. Articles by Chettipally, U. Search for Related Content PubMed PubMed Citation Articles by Mandal, A. K. Articles by Chettipally, U.