- Email: [email protected]
Modern management of adult thoracic empyema
J
THoRAc CARDIOVASC SURG
90:849-855, 1985
Modern management of adult thoracic empyema Seventy adult patients with thoracic empyema were treated at the Universityof Michigan Medical Center between 1978 and 1982. Twenty-two (31 %) of the empyemas were associated with pneumonia,23 (33 %) occurred as postoperativecomplicatioM, and seven(10%) were iatrogenic. When used as the initial mode of drainage, repeat thoracentesis was successful in only four of 11 cases (36%). Similarly, closed tube thoracostomy, as initial treatment, was successful in only 14 of 40 cases (35% ~ Rib resection, however, provided cure or control in 10 of 11 patients (91%) when employedas the fll'St treatment method. Eight of 12 patients (67%) with parapneumorric empyemas were treated successfully with closed tube thoracostomy, in contrast to only two of 17 patients (12%) with postoperative empyemas so treated. Eventual control or cme of empyema was achieved in 57 patients (81 % ~ whereas 13 (19 %) died (five from their empyema and eight with empyema as an active problem at the time of death). AU of the five empyema-ca~ deaths occurred in patients who underwent chest tube drainage as the most invasive treatment modality. The mortality rate for immunosuppressed patients was 40% (four of 10 patients). This analysis of a large recent series of adult empyemas suggests that chest tube drainage is often inadequate and more aggressivemanagement is likely to result in fewer treatment failures and fewer total procedures. Early rib resection, especiaUy for postoperativeempyemas and those in immunocompromised patients, is recommended.
John H. Lemmer, M.D. (by invitation), Mark J. Botham, M.D. (by invitation), and Mark B. Orringer, M.D., Ann Arbor, Mich.
Empyema, the accumulation of pus in the pleural cavity,continues to be a relatively common and troublesome clinical problem. Recent reports in the medical and surgical literature have suggested that empyemas can be managed effectively with less aggressive treatment than has been advocated in the past. Specifically, closed tube thoracostomy has been recommended as initial treatment for most empyemas by several authors.':' This report reviews the University of Michigan experience with empyema between 1978 and 1982 and has permitted us to formulate our own guidelines for the modem management of this problem. Patients For this study, empyema was defined as the accumulation of thick pus, not merely infected serosanguineous From the Section of Thoracic Surgery, Department of Surgery, University of Michigan Medical Center, Ann Arbor, Mich. Read at the Sixty-fifth Annual Meeting of The American Association for Thoracic Surgery, New Orleans, La., April 29-May 1, 1985. Address for reprints: Mark B. Orringer, MD., Professor and Head, Section of Thoracic Surgery, C7073 Outpatient Building, University Hospital, Ann Arbor, Mich. 48109.
pleural fluid, in the thoracic cavity. Using this definition, we conducted a retrospective review of the medical records of 70 adult patients with thoracic empyema treated at University of Michigan Hospitals between 1978 and 1982. An average of 14 patients per year (range 12 to 17 per year) were treated. Forty-two (60%) of the patients were men and 28 (48%) women. Fifty-nine (84%) of the patients were white and eight (13%) black. The average age of the patients was 46 years (range 16 to 82). Twenty patients were older than 60 years. The most common presenting symptoms were fever in 54 patients (77%), dyspnea in 30 patients (43%), and chest pain in 30 patients (43%). Additional symptoms included weight loss in 10 patients (14%), chills in 10 patients (14%), and night sweats in nine patients (13%). Thirty-nine (56%) of the empyemas involved the right side of the chest and two (3%) were bilateral. Twenty-six (37%) of the empyemas were associated with primary bronchopulmonary infections, of which 22 (31%) were parapneumonic in origin (Table I). Twentythree (33%) were postoperative complications, and seven empyemas (10%) arose from an intra-abdominal source. Seven cases (10%) were considered iatrogenic, occurring 849
850
The Journal of Thoracic and Cardiovascular Surgery
Lemmer, Botham, Orringer
Table n. Common culture isolates
Table I. Etiology of empyema in 70 patients Primary bronchopulmonary infection Parapneumonic Aspiration Pneumococcal Tuberculous Lung infarct Bronchiectasis Postoperative empyema Thoracic procedures Pulmonary resection Esophagectomy Other esophagus Open cardiac Miscellaneous Abdominal procedures Nissen fundoplication Colectomy Intra-abdominal source Subdiaphragmatic abscess Gallbladder empyema Bladder infection Iatrogenic Contamination of sterile effusion Chest tube for pneumothorax Miscellaneous Boerhaave's syndrome Dental infection Spontaneous
26 (37%)
22 7 7 2 I I 23 (33%) 21 9 3 5 2 2
Staphylococcus Streptococcus Enterococcus Bacteroides Pseudomonas Escherichia coli Pneumococcus No growth Various other growth
No.
%
17 16 II 9 8 5 4 4 28
24 23 16 13 II 7 6 6 40
Treatment and results
2
7 (10%) 5 I I 7 (10%) 6 I 4 (6%) I I 2
after diagnostic thoracentesis of a culture-sterile pleural effusion or after chest tube placement for pneumothorax. Empyemas of uncertain origin occurred in two patients (both of whom were immunosuppressed). Forty patients (57%) were initially hospitalized on the internal medicine service and 30 (43%) were surgical service patients. Associated diseases were present in 60 (86%) of the patients treated: neoplasm ill 16 (24%), neurologic disorders in 13 (19%), pulmonary and cardiac disease in 10 each (14%), diabetes mellitus in six (9%), alcohol or drug abuse in six (9%), and inflammatory bowel disease in three (4%). Ten (14%) were immunocompromised patients who were receiving antineoplastic chemotherapy, supraphysiologic doses of steroids, or both, when the empyema developed. Bacteriologic cultures of pus obtained by thoracentesis or at operation were positive in 94% of these patients. Most commonly (39%), a single aerobic organism was identified, although anaerobic bacteria were cultured (either alone or with other organisms) in 26% (Table Il). Forty-one of the empyemas were judged to be "complicated" because of their loculated nature, associated bronchopleural fistula, pulmonary collapse, or persistent esophageal leak.
One patient with empyema and widely metastatic breast cancer refused treatment. All other patients received antibiotics in conjunction with invasive procedures (Table III).
Results of repeat thoracentesis as initial treatment.
Between two and twenty thoracenteses were used in conjunction with antibiotics as the initial method of treatment in 11 patients. This approach was successful in only four patients (11%). After failure of repeated needle drainage, six of the seven patients underwent chest tube drainage. This treatment was successful in two patients, two died of empyema or related causes, and two required further procedures. The remaining patient with unsuccessful repeat thoracentesis underwent subsequent rib resection, which was curative. Mortality in this group of 11 patients treated initially by repeat thoracentesis was 18%.
Results of tube thoracostomy as initial treatment. Forty patients underwent closed tube thoracostomy as the initial mode of therapy, with cure or control being achieved by this method alone in only 14 (35%). Eight of these 14 patients required multiple tube placements and repositionings for successful drainage. In 26 patients chest tubes did not allow adequate drainage. Five of the patients in whom closed tube thoracostomy failed had no further procedures and died of empyema or related causes (in four of these patients, the empyema was the direct cause of death). Of the remaining 21 patients, 17 underwent rib resection, with cure or control of the empyema occurring in 12. However, two of them died, and the empyema was an active problem at the time of death. Two of the patients who had initial failure of tube thoracostomy underwent thoracotomy and decortication, with success in one· but repeat decortication necessary for eventual cure in the other. One patient who had unsuccessful chest tube drainage underwent thoracotomy for completion esophagectomy and medi-
Volume 90 Number 6 December, 1985
Thoracic empyema 8 5 1
Table m. Results of initial treatment No. of patients (% of total) Therapeutic thoracentesis Closed tube thoracostomy Rib resection Thoracotomy with decortication. etc. No treatment
Success rate (No. of patients, %)
Mortality (%)
11 (16%)
4 (11%)
18
40 (57%)
14 (35%)
20
11 (16%)
10 (91%)
7 (10%) 1 (1%)
3 (43%)
astinal irrigation but died later of an empyema-related cause. The remaining patient was treated successfully with an Eloesser flap. The mortality in this group of patients who were initially treated by closed tube thoracostomy was 20% (eight of 40 patients). Complicationsas a result of chest tube placement occurred in four patients (severe hemorrhage in two, misplacement into the abdomen. in one, and pleurocutaneous fistula in one). Results of rib resection as initial treatment. Rib resection and drainage of the empyema was the initial procedure in 11 patients, with resulting cure or control being achieved by this method alone in 10 (91%). The patient who did not obtain adequate drainage by rib resection had a tuberculous empyema that was successfully treated by a subsequent Eloesser flap. There were no deaths in this group and there were no complications attributable to the procedure. Results of thoracotomy as initial treatment. Seven patients underwent thoracotomy for empyema drainage in addition to concomitant procedures. In two instances decortication was performed. One thoracotomy was required for control of hemorrhage caused by diagnostic thoracentesis. Other concomitant procedures were resection of gastrobronchopleurocutaneous fistula, repair of esophageal rupture, resection of lung infarct, and completion pneumonectomy in one patient each. Initial thoracotomy successfully controlled the empyema in three patients. The remaining four patients required subsequent drainage operations (usually multiple); two of these patients died of empyema or related causes. The mortality in this group was 29%. Parapneumonic versus postoperative empyemas: Comparative results. Twenty-two patients had empyemas associated with pneumonias. Four of them were treated by repeat thoracentesis. This was successful in three and unsuccessful in one. Although closed tube thoracostomy was later performed, the latter patient died with empyema being an active problem at the time of death. Thirteen patients with parapneumonic empyemas were initially treated by chest tube placement.
o
29
0(0%)
This form of treatment was successful in nine (69%) and unsuccessful in four patients, two dying of the empyema and two requiring other procedures (rib resection, decortication) for control. Rib resection was used as the initial therapy in four patients with parapneumonic empyemas and was successful in all four. One patient, who had had pneumonia 6 months previously, underwent thoracotomy with decortication and was cured of the empyema. In all, three of the 22 patients with parapneumonic empyemas died (mortality 14%). Twenty-three patients (33%) had postoperative empyemas. One of these patients was treated initially by repeat thoracentesis, but this failed and multiple operative procedures were necessary to obtain cure. Seventeen postoperative empyemas were initially treated by closed tube thoracostomy, but only two were treated successfully (12%). Three of the patients with postoperative empyemas who had unsuccessful closed tube thoracostomy drainage underwent no further therapy and died; 11 required further operative drainage. Three of the patients with postoperative empyemas were treated initially with rib resection, and success was achieved in all three. Two patients with complicated postoperative empyemas were treated unsuccessfully by thoracotomy. In all, six of the patients with postoperative empyemas died of empyema or related causes (mortality 26%). Results of treatment of empyemas in immunosuppressed patients. Ten patients were receiving antineoplastic chemotherapy and/or supraphysiologic doses of steroids when the empyema developed. Four of these empyemas were associated with pneumonias, one occurred after lobectomy, two developed after diagnostic thoracentesis of culture-negative pleural effusions, one was associated with lung infarction, and two were of uncertain cause. One of the immunosuppressed patients was treated by thoracentesis and antibiotics with success. Seven were initially treated by closed tube thoracostomy, but in only two of these was adequate empyema drainage obtained. Three of the five with inadequate drainage by chest tube underwent no further procedures and died of empyema
8 5 2 Lemmer, Botham, Orringer
The Journal of Thoracic and Cardiovascular Surgery
or related causes. For two of these three patients, the empyema and associated sepsis was the direct cause of death, whereas the third died of gastrointestinal hemorrhage with undrained pus in the chest. The two other immunosuppressed patients who had unsuccessful empyema drainage by tube thoracostomy underwent subsequent rib resections. In one, cure was achieved. The second died of lung cancer, and an undrained empyema was found at autopsy. Two immunosuppressed patients with empyemas underwent rib resection as the initial method of therapy and cure was achieved in both. The overall mortality in this group of patients was 40%. Overall results. Of the 70 patients, 43 were eventually cured of their empyema and 14 had controlled drainage (81% overall cure-control rate). Thirteen patients died of empyema or related causes (19% mortality). The empyema was the direct cause of death in five of these patients and was an active contributing factor in eight. Late deaths, unrelated to the previous empyema, occurred in at least 12 patients. Of the 10 immunosuppressed patients, four died of empyema or related causes. One of the six patients with diabetes mellitus died.
Discussion The causes of empyema in any large series are a reflection of the nature of the patient population being treated at that center. For example, the fact that only 4% of our empyemas were attributable to trauma is indicative of the low incidence of acute penetrating trauma treated at our hospital. Ten percent of our empyemas were iatrogenic, the result of diagnostic thoracentesis (six cases) or chest tube placement for pneumothorax (one case). Similar low, but significant, incidences of iatrogenic empyemas have been noted by others, and the need for sterile technique in performing these procedures must be emphasized." 4,5 As in other major reports of adult empyema thoracis, serious associated diseases were present in the majority (86%) of our patients. The influence of associated systemic illness upon empyema mortality has been documented." The results of microbiological cultures of the empyema contents in this series differ little from other recent reports, with Staphylococcus, Streptococcus, and Pseudomonas being the most frequently isolated organisms." 2 Anaerobic organisms (predominantly Bacteroides fragilis) were present in 26% of cultures. In our series, no relationship between the type of organism cultured and the result of treatment could be determined.
Thirteen of the 70 patients described in this report died. Five of these deaths were directly caused by the empyema (7% empyema mortality), and in eight the empyema was a contributing factor (19% empyema and empyema-related mortality). Recently reported empyema-related death rates range from 8% to 33%.1.5,6,8,9 All five of our deaths directly related to empyema occurred in patients who were treated only by chest tube drainage. Four of these patients had fatal sepsis as a result of pyothorax and one bled uncontrollably from the empyema cavity. These results suggest that chest tube drainage is often inadequate treatment of an empyema. A significant finding of this retrospective analysis is that chest tube drainage failed in 65% of the 40 cases in which it was used as the initial method of treatment. This failure rate is substantially greater than the 26% to 36% chest tube failure rate reported by others.' The failure of chest tube drainage occurred particularly in patients who had postoperative empyemas, with success being achieved by this method alone in only two of the 17 patients with postoperative empyemas. Several recently reported series have found closed tube thoracostomy to be effective in 62% to 80% of empyemas, but in these reports there has been a much greater proportion of parapneumonic empyemas than in our patients." That postoperative empyema is a more difficult problem is not a new observation. Cohn and Blaisdell'? (1970) emphasized the aggressive use of open drainage (rib resection or Eloesser flap) for post-pulmonary resection empyemas and encouraged open drainage of postpneumonectomy empyema, as has also been advocated by Kirsh and associates." Rib resection was highly successful when used as the initial method of empyema treatment, with cure or control of the empyema occurring in 10 of 11 patients so treated. Four of these patients had para pneumonic empyemas and three had postoperative empyemas. Successful drainage was achieved in all seven of these patients. Of our 10 immunosuppressed patients, two were treated initially by rib resection and both were cured. Seven of the immunosuppressed patients underwent closed tube thoracostomy. Adequate drainage or cure could not be obtained in five, and four of these five died. These deaths potentially may have been prevented by early rib resection rather than chest tube drainage in these high-risk patients. Thus, a more aggressive approach in immunosuppressed patients seems prudent. Fishman and Ellertson" described failure of healing of empyema cavities in two immunosuppressed patients who were treated by rib resection and
Volume 90 Number 6 December, 1985
advocated early pleural decortication for these patients. The advantages of rib resection and empyema tube placement are several. Rib resection allows for accurate dependent positioning of a large-bore, end-hole drainage tube. The intercostal neurovascular bundle of the resected rib segment is divided to minimize intercostal neuralgia caused by the pressure of the tube. Not only is the conventional rigid intercostal chest tube for empyema drainage. more uncomfortable for the patient because of intercostal neuritis, but also the ingrowth of granulation tissue into the side holes may make subsequent tube removal difficult. Our analysis of 70 adult patients with empyema thoracis treated during a recent 5 year period reaffirms the principle of aggressive empyema drainage that has long been advocated by thoracic surgeons. It does not support a conservative approach to this serious clinical problem. Closed tube thoracostomy drainage is often inadequate, and early rib resection is a safer, more efficient approach, particularly in postoperative empyemas and those occurring in immunosuppressed patients.
2
3
4
5
6 7 8 9
10 11
12
REFERENCES LeBlanc KA, Tucker WY: Empyema of the thorax. Surg Gynecol Obstet 158:66-70, 1984 Pezzela AT, Wallis J'T, Curtis 11: Nontuberculous empyema. A clinical experience. Texas Heart Inst 1 10:263268, 1983 Mavroudis C, Symmonds lB, Minagi H, Thomas AN: Improved survival in management of empyema thoracis. 1 THoRAc CARDIOVASC SURG 82:49-57, 1981 Weese WC, Shindler ER, Smith lA, Rabinovich S: Empyema of the thorax then and now. Arch Intern Moo 131:516-520, 1973 Varkey B, Rose HD, Kutty CPK, Politis J: Empyema thoracis during a ten-year period. Arch Intern Med 141:1771-1776,1981 less P, Brynitz S, Moller AF: Mortality in thoracic empyema. Scand 1 Cardiovasc Surg 18:85-87, 1984 de la Rocha AG: Empyema thoracis. Surg Gynecol Obstet 155:839-845, 1982 Benfield GFA: Recent trends in empyema thoracis. Br 1 Dis Chest 75:358-366, 1981 Schachter EN, Kreisman H, Putman C: Diagnostic problems in suppurative lung disease. Arch Intern Med 136:167-171, 1976 Cohn LM, Blaisdel EW: Surgical treatment of nontuberculous empyema. Arch Surg 100:376-381, 1970 Kirsh MM, Rotman H, Behrendt DM, Orringer MB, Sloan H: Complications of pulmonary resection. Ann Thorac Surg 20:210-236, 1975 Fishman NH, Ellertson DG: Early pleural decortication
Thoracic empyema 8 5 3
for thoracic empyema in immunosuppressed patients. 1 THORAC CARDIOVASC SURG 74:537-541, 1977
Discussion DR. MARK BRAIMBRIDGE London. England
Until 5 years ago our experience at St. Thomas' Hospital was similar to that of the authors. Intercostal tube drainage did not work. Rib resection was effective, but many of our empyemas occurred in vagrants and their management over months with an open drainage chest tube was difficult. Decortication would not be tolerated acutely in such persons. Rib resection and drainage is more effective than intercostal tube drainage because of adequate debridement of the cavity, breakdown of loculi, and establishment of dependent drainage. If intercostal tube drainage could be made effective by adequate debridement and drainage, it could be upgraded as a modality of treatment. We now use the abdominal laparoscope, which has a large lumen, to inspect the empyema cavity; to do a full debridement, picking out the fibrin with biopsy forceps and washing out the cavity; to break down loculi with the long laparoscope arm; and then to irrigate the cavity. This is done by putting a No. 32 Argyle drain at the base and a No. 16 Argyle drain at the apex, under vision through the laparoscope, irrigating three times a day, with saline twice and a local antiseptic (such as Noxyflex) and with no antibiotics the third time. The tubes are removed after three successive sterile cultures. Over the past 5 years, this has been our initial therapy in all empyemas. Our few failures have been due to the underlying condition being malignant disease or to total empyema, in which the apex of the lung is separated by a significant degree from the chest wall. I would like to recommend the extra modality of intercostal drainage, but it must be preceded by thoracoscopy and debridement of the cavity and followed by effective irrigation of the residual pus. I would ask whether irrigation played any part in the regimen in the authors' institution. DR. GORDON F. MURRAY s.c. Dr. Lemmer and his associates have accurately emphasized that the increased incidence of major thoracic procedure and the numbers of immunologically compromised hosts have contributed to the continued presence of empyema as a serious and potentially fatal disease. In our institution, these factors have resulted in the formulation and application of the concept of early complete decortication whenever possible. My comments relate to 61 patients at the University of North Carolina who had a pleural infection requiring operative intervention between January, 1970, and lune, 1984. There were 69 operative procedures performed in the 61 patients. In contrast to the series from the University of Michigan, over one third of our patients underwent early decortication with only one operative death. Chapel Hill.
8 5 4 Lemmer, Botham, Orringer
Of greater importance is the fact that an excellent result-that is, control of infection and complete reexpansion of the lung-was achieved in 86% of the decortications. In addition, decortication was associated with a shorter postoperative confinement and total hospitalization. Early decortication offers elimination of sepsis, obliteration of the dead space, and maximal preservation of pulmonary function. We have concluded that decortication before extensive pleural damage and debilitation will usually achieve shorter hospitalization, lower mortality, and improved results in the management of thoracic empyema. DR. CHARLES W. VAN WAY III Denver. Colo.
I would like to ring another change on what seems to be a common theme. Dr. James Narrod and I reviewed a series of 80 patients with empyema seen at the Denver General Hospital over the last 7 years. All patients underwent tube management initially. In half, tube management failed. We did 40 thoracotomies. Of the 40, 15 were decortications and 25 were early limited thoracotomy. Our preferred technique for early limited thoracotomy is a short lateral thoracotomy through the sixth or seventh intercostal space. We break down the trabeculations and loculations and place two or three chest tubes. The third one, if necessary, is placed immediately below the lung and above the diaphragm. Using this procedure very early in the management of empyema, after failure of the first chest tube, eliminates the need for a later decortication. The overall mortality in our series was two of 80 patients. No patients died among the 40 who had thoracotomy. I would like to ask the authors to discuss the details of their operative procedure, which I suspect is accomplishing much the same goals as our limited thoracotomy. DR. MAURICE R. HOOD New York. N. Y.
At the Bellevue Hospital and Manhattan Veterans Administration Hospital over a 4 year period from 1981 through the first 4 months of this year, more than 200 patients with empyema have been treated surgically. They have included a large population of drug addicts and alcoholics. To date more than 500 patients with acquired immunodeficiency syndrome have been treated at Bellevue, because of the large population of indigent street people and the large immigrant population from all over the world. There has been a significant incidence of tuberculosis in this group. As a result, we have begun doing procedures that I thought we probably would not have to do any longer. We have done three pleuropneumonectomies, two thoracoplasties, and six Eloesser flaps in patients with tuberculosis. In the first year of this period, it was common practice that the procedures would be decided upon and performed either by the internist or by a junior surgical house officer at the instruction of the internist, with little surgical input and certainly no thoracic surgical involvement. As a result, more than 30% of the patients required secondary operations, usually decortication, and another 30% were left with pulmonary disability despite control of the infection.
The Journal of Thoracic and Cardiovascular Surgery
We have tried to reinstitute the idea that empyema is a surgical disease necessitating surgical decision-making. We have had internists who try to treat empyemas as has been discussed, with thoracocentesis. Junior house officers have repeatedly inserted intercostal tubes, despite being uninstructed on determining the size and shape of the empyema space and not understanding dependent drainage. Recently the radiologists have become involved by putting in miniscule chest tubes under computed tomographic and ultrasound guidance. In the past year, with a considerable change, we have operated upon more than 50 patients with empyema, with only three requiring a secondary procedure. A fairly large number have had anaerobic infections, particularly Pseudomonas. and 80% of these patients died. It seems that we are always rediscovering obvious truths. Perhaps I am old enough to act a bit philosophical. The method gains favor for a while, sinks into oblivion and a few years later is rediscovered by some enthusiast who is ignorant of the fact that his method has already had one and perhaps two or three periods of approval followed by disuse. -Evarts Graham. 1935 This quotation is taken from a book in which Graham was discussing the problem of thoracentesis and simple chest tube insertion for the management of empyema. I think the lessonis obvious. However, I am sure that these comments are being directed to the wrong audience.
It is sad to think of the number of lives which are sacrificed annually by the failure to recognize that empyema should be treated as an ordinary abscess, by free incision. -William Osler. 1892 The second quotation shows that at least one internist some years ago understood the problem better than the pulmonary internists of today. DR. LEMMER (Closing) I would like to thank the discussants for their comments. Dr. Braimbridge has described an innovative technique using thoracoscopy as a method of empyema debridement and drainage. We find this to be interesting but have not tried the procedure he describes. Dr. Murray has discussed the role of early decortication, which emphasizes the need for aggressive therapy in these patients. At the University of Michigan, however, we reserve decortication for those patients in whom rib resection has failed. Dr. Van Way describes a failure rate of chest tube management of empyema that is similar to ours and to those reported both in the recent and older literature. The details of our technique of rib resection follow traditional teachings. A vertical incision is made over the inferior aspect of the empyema cavity as demonstrated on the preoperative roentgenogram. A 5 em segment of rib is resected, and the neurovascular bundle is ligated with chromic suture. A 46 Fr.
Volume 90 Number 6 December, 1985
soft rubber tube is placed through the thoracotomy into the cavityand the skin is closed loosely around the tube. The tube is then placed on suction for 7 to 10 days, at which time it is ascertained with a chest roentgenogram if the lung remains against the chest wall when the tube is disconnected from all suction. If the lung remains fully expanded, the tube is cut off close to the skin. The patient then begins daily tube irrigations
Thoracic empyema 8 5 5
assisted by home nursing care. After 2 weeks, the tract is well established so that the patient can take the tube out, wash it off and then easily slip it back into the cavity. Periodic contrast studies are obtained, and as the empyema cavity closes in, the tube is gradually advanced outward. I appreciate Dr. Hood's comments and his impressive experience with empyemas.
Bound volumes available to subscribers Bound volumes of THE JOURNAL OFTHORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1985 issues from the Publisher, at a cost of $61.45 ($77.10 international) for Vol. 89 (January-June) and Vol. 90 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 30 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact The C. V. Mosby Company, Circulation Department, 11830 Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.
THoRAc CARDIOVASC SURG
90:849-855, 1985
Modern management of adult thoracic empyema Seventy adult patients with thoracic empyema were treated at the Universityof Michigan Medical Center between 1978 and 1982. Twenty-two (31 %) of the empyemas were associated with pneumonia,23 (33 %) occurred as postoperativecomplicatioM, and seven(10%) were iatrogenic. When used as the initial mode of drainage, repeat thoracentesis was successful in only four of 11 cases (36%). Similarly, closed tube thoracostomy, as initial treatment, was successful in only 14 of 40 cases (35% ~ Rib resection, however, provided cure or control in 10 of 11 patients (91%) when employedas the fll'St treatment method. Eight of 12 patients (67%) with parapneumorric empyemas were treated successfully with closed tube thoracostomy, in contrast to only two of 17 patients (12%) with postoperative empyemas so treated. Eventual control or cme of empyema was achieved in 57 patients (81 % ~ whereas 13 (19 %) died (five from their empyema and eight with empyema as an active problem at the time of death). AU of the five empyema-ca~ deaths occurred in patients who underwent chest tube drainage as the most invasive treatment modality. The mortality rate for immunosuppressed patients was 40% (four of 10 patients). This analysis of a large recent series of adult empyemas suggests that chest tube drainage is often inadequate and more aggressivemanagement is likely to result in fewer treatment failures and fewer total procedures. Early rib resection, especiaUy for postoperativeempyemas and those in immunocompromised patients, is recommended.
John H. Lemmer, M.D. (by invitation), Mark J. Botham, M.D. (by invitation), and Mark B. Orringer, M.D., Ann Arbor, Mich.
Empyema, the accumulation of pus in the pleural cavity,continues to be a relatively common and troublesome clinical problem. Recent reports in the medical and surgical literature have suggested that empyemas can be managed effectively with less aggressive treatment than has been advocated in the past. Specifically, closed tube thoracostomy has been recommended as initial treatment for most empyemas by several authors.':' This report reviews the University of Michigan experience with empyema between 1978 and 1982 and has permitted us to formulate our own guidelines for the modem management of this problem. Patients For this study, empyema was defined as the accumulation of thick pus, not merely infected serosanguineous From the Section of Thoracic Surgery, Department of Surgery, University of Michigan Medical Center, Ann Arbor, Mich. Read at the Sixty-fifth Annual Meeting of The American Association for Thoracic Surgery, New Orleans, La., April 29-May 1, 1985. Address for reprints: Mark B. Orringer, MD., Professor and Head, Section of Thoracic Surgery, C7073 Outpatient Building, University Hospital, Ann Arbor, Mich. 48109.
pleural fluid, in the thoracic cavity. Using this definition, we conducted a retrospective review of the medical records of 70 adult patients with thoracic empyema treated at University of Michigan Hospitals between 1978 and 1982. An average of 14 patients per year (range 12 to 17 per year) were treated. Forty-two (60%) of the patients were men and 28 (48%) women. Fifty-nine (84%) of the patients were white and eight (13%) black. The average age of the patients was 46 years (range 16 to 82). Twenty patients were older than 60 years. The most common presenting symptoms were fever in 54 patients (77%), dyspnea in 30 patients (43%), and chest pain in 30 patients (43%). Additional symptoms included weight loss in 10 patients (14%), chills in 10 patients (14%), and night sweats in nine patients (13%). Thirty-nine (56%) of the empyemas involved the right side of the chest and two (3%) were bilateral. Twenty-six (37%) of the empyemas were associated with primary bronchopulmonary infections, of which 22 (31%) were parapneumonic in origin (Table I). Twentythree (33%) were postoperative complications, and seven empyemas (10%) arose from an intra-abdominal source. Seven cases (10%) were considered iatrogenic, occurring 849
850
The Journal of Thoracic and Cardiovascular Surgery
Lemmer, Botham, Orringer
Table n. Common culture isolates
Table I. Etiology of empyema in 70 patients Primary bronchopulmonary infection Parapneumonic Aspiration Pneumococcal Tuberculous Lung infarct Bronchiectasis Postoperative empyema Thoracic procedures Pulmonary resection Esophagectomy Other esophagus Open cardiac Miscellaneous Abdominal procedures Nissen fundoplication Colectomy Intra-abdominal source Subdiaphragmatic abscess Gallbladder empyema Bladder infection Iatrogenic Contamination of sterile effusion Chest tube for pneumothorax Miscellaneous Boerhaave's syndrome Dental infection Spontaneous
26 (37%)
22 7 7 2 I I 23 (33%) 21 9 3 5 2 2
Staphylococcus Streptococcus Enterococcus Bacteroides Pseudomonas Escherichia coli Pneumococcus No growth Various other growth
No.
%
17 16 II 9 8 5 4 4 28
24 23 16 13 II 7 6 6 40
Treatment and results
2
7 (10%) 5 I I 7 (10%) 6 I 4 (6%) I I 2
after diagnostic thoracentesis of a culture-sterile pleural effusion or after chest tube placement for pneumothorax. Empyemas of uncertain origin occurred in two patients (both of whom were immunosuppressed). Forty patients (57%) were initially hospitalized on the internal medicine service and 30 (43%) were surgical service patients. Associated diseases were present in 60 (86%) of the patients treated: neoplasm ill 16 (24%), neurologic disorders in 13 (19%), pulmonary and cardiac disease in 10 each (14%), diabetes mellitus in six (9%), alcohol or drug abuse in six (9%), and inflammatory bowel disease in three (4%). Ten (14%) were immunocompromised patients who were receiving antineoplastic chemotherapy, supraphysiologic doses of steroids, or both, when the empyema developed. Bacteriologic cultures of pus obtained by thoracentesis or at operation were positive in 94% of these patients. Most commonly (39%), a single aerobic organism was identified, although anaerobic bacteria were cultured (either alone or with other organisms) in 26% (Table Il). Forty-one of the empyemas were judged to be "complicated" because of their loculated nature, associated bronchopleural fistula, pulmonary collapse, or persistent esophageal leak.
One patient with empyema and widely metastatic breast cancer refused treatment. All other patients received antibiotics in conjunction with invasive procedures (Table III).
Results of repeat thoracentesis as initial treatment.
Between two and twenty thoracenteses were used in conjunction with antibiotics as the initial method of treatment in 11 patients. This approach was successful in only four patients (11%). After failure of repeated needle drainage, six of the seven patients underwent chest tube drainage. This treatment was successful in two patients, two died of empyema or related causes, and two required further procedures. The remaining patient with unsuccessful repeat thoracentesis underwent subsequent rib resection, which was curative. Mortality in this group of 11 patients treated initially by repeat thoracentesis was 18%.
Results of tube thoracostomy as initial treatment. Forty patients underwent closed tube thoracostomy as the initial mode of therapy, with cure or control being achieved by this method alone in only 14 (35%). Eight of these 14 patients required multiple tube placements and repositionings for successful drainage. In 26 patients chest tubes did not allow adequate drainage. Five of the patients in whom closed tube thoracostomy failed had no further procedures and died of empyema or related causes (in four of these patients, the empyema was the direct cause of death). Of the remaining 21 patients, 17 underwent rib resection, with cure or control of the empyema occurring in 12. However, two of them died, and the empyema was an active problem at the time of death. Two of the patients who had initial failure of tube thoracostomy underwent thoracotomy and decortication, with success in one· but repeat decortication necessary for eventual cure in the other. One patient who had unsuccessful chest tube drainage underwent thoracotomy for completion esophagectomy and medi-
Volume 90 Number 6 December, 1985
Thoracic empyema 8 5 1
Table m. Results of initial treatment No. of patients (% of total) Therapeutic thoracentesis Closed tube thoracostomy Rib resection Thoracotomy with decortication. etc. No treatment
Success rate (No. of patients, %)
Mortality (%)
11 (16%)
4 (11%)
18
40 (57%)
14 (35%)
20
11 (16%)
10 (91%)
7 (10%) 1 (1%)
3 (43%)
astinal irrigation but died later of an empyema-related cause. The remaining patient was treated successfully with an Eloesser flap. The mortality in this group of patients who were initially treated by closed tube thoracostomy was 20% (eight of 40 patients). Complicationsas a result of chest tube placement occurred in four patients (severe hemorrhage in two, misplacement into the abdomen. in one, and pleurocutaneous fistula in one). Results of rib resection as initial treatment. Rib resection and drainage of the empyema was the initial procedure in 11 patients, with resulting cure or control being achieved by this method alone in 10 (91%). The patient who did not obtain adequate drainage by rib resection had a tuberculous empyema that was successfully treated by a subsequent Eloesser flap. There were no deaths in this group and there were no complications attributable to the procedure. Results of thoracotomy as initial treatment. Seven patients underwent thoracotomy for empyema drainage in addition to concomitant procedures. In two instances decortication was performed. One thoracotomy was required for control of hemorrhage caused by diagnostic thoracentesis. Other concomitant procedures were resection of gastrobronchopleurocutaneous fistula, repair of esophageal rupture, resection of lung infarct, and completion pneumonectomy in one patient each. Initial thoracotomy successfully controlled the empyema in three patients. The remaining four patients required subsequent drainage operations (usually multiple); two of these patients died of empyema or related causes. The mortality in this group was 29%. Parapneumonic versus postoperative empyemas: Comparative results. Twenty-two patients had empyemas associated with pneumonias. Four of them were treated by repeat thoracentesis. This was successful in three and unsuccessful in one. Although closed tube thoracostomy was later performed, the latter patient died with empyema being an active problem at the time of death. Thirteen patients with parapneumonic empyemas were initially treated by chest tube placement.
o
29
0(0%)
This form of treatment was successful in nine (69%) and unsuccessful in four patients, two dying of the empyema and two requiring other procedures (rib resection, decortication) for control. Rib resection was used as the initial therapy in four patients with parapneumonic empyemas and was successful in all four. One patient, who had had pneumonia 6 months previously, underwent thoracotomy with decortication and was cured of the empyema. In all, three of the 22 patients with parapneumonic empyemas died (mortality 14%). Twenty-three patients (33%) had postoperative empyemas. One of these patients was treated initially by repeat thoracentesis, but this failed and multiple operative procedures were necessary to obtain cure. Seventeen postoperative empyemas were initially treated by closed tube thoracostomy, but only two were treated successfully (12%). Three of the patients with postoperative empyemas who had unsuccessful closed tube thoracostomy drainage underwent no further therapy and died; 11 required further operative drainage. Three of the patients with postoperative empyemas were treated initially with rib resection, and success was achieved in all three. Two patients with complicated postoperative empyemas were treated unsuccessfully by thoracotomy. In all, six of the patients with postoperative empyemas died of empyema or related causes (mortality 26%). Results of treatment of empyemas in immunosuppressed patients. Ten patients were receiving antineoplastic chemotherapy and/or supraphysiologic doses of steroids when the empyema developed. Four of these empyemas were associated with pneumonias, one occurred after lobectomy, two developed after diagnostic thoracentesis of culture-negative pleural effusions, one was associated with lung infarction, and two were of uncertain cause. One of the immunosuppressed patients was treated by thoracentesis and antibiotics with success. Seven were initially treated by closed tube thoracostomy, but in only two of these was adequate empyema drainage obtained. Three of the five with inadequate drainage by chest tube underwent no further procedures and died of empyema
8 5 2 Lemmer, Botham, Orringer
The Journal of Thoracic and Cardiovascular Surgery
or related causes. For two of these three patients, the empyema and associated sepsis was the direct cause of death, whereas the third died of gastrointestinal hemorrhage with undrained pus in the chest. The two other immunosuppressed patients who had unsuccessful empyema drainage by tube thoracostomy underwent subsequent rib resections. In one, cure was achieved. The second died of lung cancer, and an undrained empyema was found at autopsy. Two immunosuppressed patients with empyemas underwent rib resection as the initial method of therapy and cure was achieved in both. The overall mortality in this group of patients was 40%. Overall results. Of the 70 patients, 43 were eventually cured of their empyema and 14 had controlled drainage (81% overall cure-control rate). Thirteen patients died of empyema or related causes (19% mortality). The empyema was the direct cause of death in five of these patients and was an active contributing factor in eight. Late deaths, unrelated to the previous empyema, occurred in at least 12 patients. Of the 10 immunosuppressed patients, four died of empyema or related causes. One of the six patients with diabetes mellitus died.
Discussion The causes of empyema in any large series are a reflection of the nature of the patient population being treated at that center. For example, the fact that only 4% of our empyemas were attributable to trauma is indicative of the low incidence of acute penetrating trauma treated at our hospital. Ten percent of our empyemas were iatrogenic, the result of diagnostic thoracentesis (six cases) or chest tube placement for pneumothorax (one case). Similar low, but significant, incidences of iatrogenic empyemas have been noted by others, and the need for sterile technique in performing these procedures must be emphasized." 4,5 As in other major reports of adult empyema thoracis, serious associated diseases were present in the majority (86%) of our patients. The influence of associated systemic illness upon empyema mortality has been documented." The results of microbiological cultures of the empyema contents in this series differ little from other recent reports, with Staphylococcus, Streptococcus, and Pseudomonas being the most frequently isolated organisms." 2 Anaerobic organisms (predominantly Bacteroides fragilis) were present in 26% of cultures. In our series, no relationship between the type of organism cultured and the result of treatment could be determined.
Thirteen of the 70 patients described in this report died. Five of these deaths were directly caused by the empyema (7% empyema mortality), and in eight the empyema was a contributing factor (19% empyema and empyema-related mortality). Recently reported empyema-related death rates range from 8% to 33%.1.5,6,8,9 All five of our deaths directly related to empyema occurred in patients who were treated only by chest tube drainage. Four of these patients had fatal sepsis as a result of pyothorax and one bled uncontrollably from the empyema cavity. These results suggest that chest tube drainage is often inadequate treatment of an empyema. A significant finding of this retrospective analysis is that chest tube drainage failed in 65% of the 40 cases in which it was used as the initial method of treatment. This failure rate is substantially greater than the 26% to 36% chest tube failure rate reported by others.' The failure of chest tube drainage occurred particularly in patients who had postoperative empyemas, with success being achieved by this method alone in only two of the 17 patients with postoperative empyemas. Several recently reported series have found closed tube thoracostomy to be effective in 62% to 80% of empyemas, but in these reports there has been a much greater proportion of parapneumonic empyemas than in our patients." That postoperative empyema is a more difficult problem is not a new observation. Cohn and Blaisdell'? (1970) emphasized the aggressive use of open drainage (rib resection or Eloesser flap) for post-pulmonary resection empyemas and encouraged open drainage of postpneumonectomy empyema, as has also been advocated by Kirsh and associates." Rib resection was highly successful when used as the initial method of empyema treatment, with cure or control of the empyema occurring in 10 of 11 patients so treated. Four of these patients had para pneumonic empyemas and three had postoperative empyemas. Successful drainage was achieved in all seven of these patients. Of our 10 immunosuppressed patients, two were treated initially by rib resection and both were cured. Seven of the immunosuppressed patients underwent closed tube thoracostomy. Adequate drainage or cure could not be obtained in five, and four of these five died. These deaths potentially may have been prevented by early rib resection rather than chest tube drainage in these high-risk patients. Thus, a more aggressive approach in immunosuppressed patients seems prudent. Fishman and Ellertson" described failure of healing of empyema cavities in two immunosuppressed patients who were treated by rib resection and
Volume 90 Number 6 December, 1985
advocated early pleural decortication for these patients. The advantages of rib resection and empyema tube placement are several. Rib resection allows for accurate dependent positioning of a large-bore, end-hole drainage tube. The intercostal neurovascular bundle of the resected rib segment is divided to minimize intercostal neuralgia caused by the pressure of the tube. Not only is the conventional rigid intercostal chest tube for empyema drainage. more uncomfortable for the patient because of intercostal neuritis, but also the ingrowth of granulation tissue into the side holes may make subsequent tube removal difficult. Our analysis of 70 adult patients with empyema thoracis treated during a recent 5 year period reaffirms the principle of aggressive empyema drainage that has long been advocated by thoracic surgeons. It does not support a conservative approach to this serious clinical problem. Closed tube thoracostomy drainage is often inadequate, and early rib resection is a safer, more efficient approach, particularly in postoperative empyemas and those occurring in immunosuppressed patients.
2
3
4
5
6 7 8 9
10 11
12
REFERENCES LeBlanc KA, Tucker WY: Empyema of the thorax. Surg Gynecol Obstet 158:66-70, 1984 Pezzela AT, Wallis J'T, Curtis 11: Nontuberculous empyema. A clinical experience. Texas Heart Inst 1 10:263268, 1983 Mavroudis C, Symmonds lB, Minagi H, Thomas AN: Improved survival in management of empyema thoracis. 1 THoRAc CARDIOVASC SURG 82:49-57, 1981 Weese WC, Shindler ER, Smith lA, Rabinovich S: Empyema of the thorax then and now. Arch Intern Moo 131:516-520, 1973 Varkey B, Rose HD, Kutty CPK, Politis J: Empyema thoracis during a ten-year period. Arch Intern Med 141:1771-1776,1981 less P, Brynitz S, Moller AF: Mortality in thoracic empyema. Scand 1 Cardiovasc Surg 18:85-87, 1984 de la Rocha AG: Empyema thoracis. Surg Gynecol Obstet 155:839-845, 1982 Benfield GFA: Recent trends in empyema thoracis. Br 1 Dis Chest 75:358-366, 1981 Schachter EN, Kreisman H, Putman C: Diagnostic problems in suppurative lung disease. Arch Intern Med 136:167-171, 1976 Cohn LM, Blaisdel EW: Surgical treatment of nontuberculous empyema. Arch Surg 100:376-381, 1970 Kirsh MM, Rotman H, Behrendt DM, Orringer MB, Sloan H: Complications of pulmonary resection. Ann Thorac Surg 20:210-236, 1975 Fishman NH, Ellertson DG: Early pleural decortication
Thoracic empyema 8 5 3
for thoracic empyema in immunosuppressed patients. 1 THORAC CARDIOVASC SURG 74:537-541, 1977
Discussion DR. MARK BRAIMBRIDGE London. England
Until 5 years ago our experience at St. Thomas' Hospital was similar to that of the authors. Intercostal tube drainage did not work. Rib resection was effective, but many of our empyemas occurred in vagrants and their management over months with an open drainage chest tube was difficult. Decortication would not be tolerated acutely in such persons. Rib resection and drainage is more effective than intercostal tube drainage because of adequate debridement of the cavity, breakdown of loculi, and establishment of dependent drainage. If intercostal tube drainage could be made effective by adequate debridement and drainage, it could be upgraded as a modality of treatment. We now use the abdominal laparoscope, which has a large lumen, to inspect the empyema cavity; to do a full debridement, picking out the fibrin with biopsy forceps and washing out the cavity; to break down loculi with the long laparoscope arm; and then to irrigate the cavity. This is done by putting a No. 32 Argyle drain at the base and a No. 16 Argyle drain at the apex, under vision through the laparoscope, irrigating three times a day, with saline twice and a local antiseptic (such as Noxyflex) and with no antibiotics the third time. The tubes are removed after three successive sterile cultures. Over the past 5 years, this has been our initial therapy in all empyemas. Our few failures have been due to the underlying condition being malignant disease or to total empyema, in which the apex of the lung is separated by a significant degree from the chest wall. I would like to recommend the extra modality of intercostal drainage, but it must be preceded by thoracoscopy and debridement of the cavity and followed by effective irrigation of the residual pus. I would ask whether irrigation played any part in the regimen in the authors' institution. DR. GORDON F. MURRAY s.c. Dr. Lemmer and his associates have accurately emphasized that the increased incidence of major thoracic procedure and the numbers of immunologically compromised hosts have contributed to the continued presence of empyema as a serious and potentially fatal disease. In our institution, these factors have resulted in the formulation and application of the concept of early complete decortication whenever possible. My comments relate to 61 patients at the University of North Carolina who had a pleural infection requiring operative intervention between January, 1970, and lune, 1984. There were 69 operative procedures performed in the 61 patients. In contrast to the series from the University of Michigan, over one third of our patients underwent early decortication with only one operative death. Chapel Hill.
8 5 4 Lemmer, Botham, Orringer
Of greater importance is the fact that an excellent result-that is, control of infection and complete reexpansion of the lung-was achieved in 86% of the decortications. In addition, decortication was associated with a shorter postoperative confinement and total hospitalization. Early decortication offers elimination of sepsis, obliteration of the dead space, and maximal preservation of pulmonary function. We have concluded that decortication before extensive pleural damage and debilitation will usually achieve shorter hospitalization, lower mortality, and improved results in the management of thoracic empyema. DR. CHARLES W. VAN WAY III Denver. Colo.
I would like to ring another change on what seems to be a common theme. Dr. James Narrod and I reviewed a series of 80 patients with empyema seen at the Denver General Hospital over the last 7 years. All patients underwent tube management initially. In half, tube management failed. We did 40 thoracotomies. Of the 40, 15 were decortications and 25 were early limited thoracotomy. Our preferred technique for early limited thoracotomy is a short lateral thoracotomy through the sixth or seventh intercostal space. We break down the trabeculations and loculations and place two or three chest tubes. The third one, if necessary, is placed immediately below the lung and above the diaphragm. Using this procedure very early in the management of empyema, after failure of the first chest tube, eliminates the need for a later decortication. The overall mortality in our series was two of 80 patients. No patients died among the 40 who had thoracotomy. I would like to ask the authors to discuss the details of their operative procedure, which I suspect is accomplishing much the same goals as our limited thoracotomy. DR. MAURICE R. HOOD New York. N. Y.
At the Bellevue Hospital and Manhattan Veterans Administration Hospital over a 4 year period from 1981 through the first 4 months of this year, more than 200 patients with empyema have been treated surgically. They have included a large population of drug addicts and alcoholics. To date more than 500 patients with acquired immunodeficiency syndrome have been treated at Bellevue, because of the large population of indigent street people and the large immigrant population from all over the world. There has been a significant incidence of tuberculosis in this group. As a result, we have begun doing procedures that I thought we probably would not have to do any longer. We have done three pleuropneumonectomies, two thoracoplasties, and six Eloesser flaps in patients with tuberculosis. In the first year of this period, it was common practice that the procedures would be decided upon and performed either by the internist or by a junior surgical house officer at the instruction of the internist, with little surgical input and certainly no thoracic surgical involvement. As a result, more than 30% of the patients required secondary operations, usually decortication, and another 30% were left with pulmonary disability despite control of the infection.
The Journal of Thoracic and Cardiovascular Surgery
We have tried to reinstitute the idea that empyema is a surgical disease necessitating surgical decision-making. We have had internists who try to treat empyemas as has been discussed, with thoracocentesis. Junior house officers have repeatedly inserted intercostal tubes, despite being uninstructed on determining the size and shape of the empyema space and not understanding dependent drainage. Recently the radiologists have become involved by putting in miniscule chest tubes under computed tomographic and ultrasound guidance. In the past year, with a considerable change, we have operated upon more than 50 patients with empyema, with only three requiring a secondary procedure. A fairly large number have had anaerobic infections, particularly Pseudomonas. and 80% of these patients died. It seems that we are always rediscovering obvious truths. Perhaps I am old enough to act a bit philosophical. The method gains favor for a while, sinks into oblivion and a few years later is rediscovered by some enthusiast who is ignorant of the fact that his method has already had one and perhaps two or three periods of approval followed by disuse. -Evarts Graham. 1935 This quotation is taken from a book in which Graham was discussing the problem of thoracentesis and simple chest tube insertion for the management of empyema. I think the lessonis obvious. However, I am sure that these comments are being directed to the wrong audience.
It is sad to think of the number of lives which are sacrificed annually by the failure to recognize that empyema should be treated as an ordinary abscess, by free incision. -William Osler. 1892 The second quotation shows that at least one internist some years ago understood the problem better than the pulmonary internists of today. DR. LEMMER (Closing) I would like to thank the discussants for their comments. Dr. Braimbridge has described an innovative technique using thoracoscopy as a method of empyema debridement and drainage. We find this to be interesting but have not tried the procedure he describes. Dr. Murray has discussed the role of early decortication, which emphasizes the need for aggressive therapy in these patients. At the University of Michigan, however, we reserve decortication for those patients in whom rib resection has failed. Dr. Van Way describes a failure rate of chest tube management of empyema that is similar to ours and to those reported both in the recent and older literature. The details of our technique of rib resection follow traditional teachings. A vertical incision is made over the inferior aspect of the empyema cavity as demonstrated on the preoperative roentgenogram. A 5 em segment of rib is resected, and the neurovascular bundle is ligated with chromic suture. A 46 Fr.
Volume 90 Number 6 December, 1985
soft rubber tube is placed through the thoracotomy into the cavityand the skin is closed loosely around the tube. The tube is then placed on suction for 7 to 10 days, at which time it is ascertained with a chest roentgenogram if the lung remains against the chest wall when the tube is disconnected from all suction. If the lung remains fully expanded, the tube is cut off close to the skin. The patient then begins daily tube irrigations
Thoracic empyema 8 5 5
assisted by home nursing care. After 2 weeks, the tract is well established so that the patient can take the tube out, wash it off and then easily slip it back into the cavity. Periodic contrast studies are obtained, and as the empyema cavity closes in, the tube is gradually advanced outward. I appreciate Dr. Hood's comments and his impressive experience with empyemas.
Bound volumes available to subscribers Bound volumes of THE JOURNAL OFTHORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1985 issues from the Publisher, at a cost of $61.45 ($77.10 international) for Vol. 89 (January-June) and Vol. 90 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 30 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact The C. V. Mosby Company, Circulation Department, 11830 Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.