- Email: [email protected]
Empyema Thoracis and Infected Pneumonectomy Space: Case for Cyclical Irrigation
Empyema Thoracis and Infected Pneumonectomy Space: Case for Cyclical Imgation Mohsin Hakim, F.R.C.S., M.R.C.P., and B. B. Milstein, F.R.C.S. ABSTRACT Between January, 1974, and July, 1984, 63 cases of pleural empyema were treated at Papworth HOSpital, Cambridge. Twenty-one of these patients had cyclical irrigation of the empyema cavity, and 23 patients underwent decortication. The two groups were comparable in age and sex distribution. There was no significant difference in the duration of empyema between the two groups (7 and 10.4 weeks for the patients having irrigation and decortication, respectively) ( p > 0.05). There was no correlation between the duration of empyema and the length of hospital stay in either group (r = 0.007 and 0.005 for the irrigation and decortication groups, respectively). However, both the mean duration of tube drainage (7 and 13.5 days) and the length of hospital stay (12.3 and 17.3 days) were significantly shorter in the irrigation than in the decortication group ( p < 0.01). There were two failures in the decortication group and three in the irrigation group. Cyclical imgation was also used in 4 patients with infected pneumonectomy space with satisfactory results. We conclude that cyclical irrigation is an effective, simple, and time-saving technique that does not preclude the use of other procedures if it fails. Sixty-three patients with nontuberculous empyema and 12 with an infected pneumonectomy space were treated at Papworth Hospital, Cambridge, between January, 1974, and July, 1984. The mean age of the whole group was 55.5 years (age range, 5 to 79 years). There were 26 female and 49 male patients. The cause of empyema was postpneumonic in the majority; other causes included esophageal surgery or trauma, chest trauma, and pulmonary infarction (Table l). Infection of a pneumonectomy space was associated with a bronchopleural fistula in 8 patients.
Material and Methods Between January, 1974, and July, 1984, 23 patients underwent decortication for empyema. In August, 1981, the method of cyclical irrigation was introduced by one surgical group in our institution and became their method of choice in the treatment of empyema (21 patients) and of infected pneumonectomy space (4 pa-
From the Department of Cardiothoracic Surgery, Papworth Hospital, Cambridge, England. Accepted for publication Feb 27, 1985. Address reprint requests to Mr. Hakim, Department of Cardiothoracic Surgery, Papworth Hospital, Papworth Everard, Cambridge CB3 ERE, England.
85
tients) [l, 21. Other surgeons at Papworth Hospital continued to use the other methods of treatment.
Empyema Thoracis The indications for irrigation and decortication were identical in patients with empyema thoracis. These were an empyema that had been present for at least 1 week and whose resolution was considered unlikely with aspiration or intercostal tube drainage because of a thickened visceral pleura. Unfit elderly patients and those with other diseases, such as rheumatoid arthritis, for whom early mobilization was considered essential were treated by rib resection. This study compares the two methods of treatmentdecortication and cyclical irrigation. The criteria for success of treatment were resolution of empyema and absence of recurrence of infection as indicated by the clinical picture, leukocyte count, and the appearance of serial chest radiographs. DECORTICATION. Decortication was undertaken in 23 patients using standard techniques [3]. CYCLICAL IRRIGATION. A total of 21 patients received this treatment. Initially we inserted two intercostal tubes under local anesthesia into the empyema cavity. One tube was used for instillation of a solution containing noxythiolin 1%,and the other tube, which was positioned in the most dependent part of the cavity, was used for drainage. During the inflow phase of imgation, the outflow tube was elevated to the shoulder level to create a safety valve, thus ensuring complete filling of the cavity and at the same time avoiding the generation of a high pressure that could cause air embolism. This two-tube system was used in the first 12 patients. In the subsequent 9 patients, we employed a simpler system consisting of a single tube with a Y junction to connect the chest drain to the irrigation drip through one arm and to an underwater-seal bottle through the other (Figure). At the same time we substituted normal saline solution for the antiseptic solution. This was done in an attempt to avoid adverse reactions associated with the intrapleural instillation of noxythiolin [4].These reactions are mainly nausea and vomiting, and were noted in one-third of the patients. Irrigation was done in cycles of four hours. The fluid (200 to 500 ml) was run in to fill the cavity, retained for three hours, and then allowed to drain for one hour with suction in a manner similar to that described by Rosenfeldt and associates [2]. A course of systemic antibiotics was given if the patient was pyrexial or when the pleural aspirate revealed pathogenic organisms. Irrigation was continued for 5 to 10 days until the draining fluid became clear and sterile as indicated by bacteriological cul-
86 The Annals of Thoracic Surgery Vol 41 No 1 January 1986
Table 1. Cause and Methods of Treatment of Pleural Empyema in 63 Patients Cause
No. of Patients
Postpneumonic
43
Esophageal
Pulmonary infarction
Chest trauma
4
Insertion of central line Open lung biopsy
1 1
Treatment Irrigation, 19 Decortication, 15 Rib resection, 5 Intercostal tube, 2 Aspiration, 2 Decortication, 4 Rib resection, 4 Intercostal tube, 1 Irrigation, 1 Decortication, 1 Rib resection, 2 Intercostal tube, 1 Irrigation, 1 Decortication, 2 Fenestration, 1 Decortication, 1 Intercostal tube, 1
INFLOW LINE
I
SINGLE INTERCOSTAL TUBE
b
DRAINAGE TUBE
:.:.:.:. ....... .-*..*.*.. .....:+ .. a ..
tures. The cavity was then filled with 100 ml of normal saline solution containing a combination of antibiotics, usually gentamicin sulfate, cefotaxime sodium, and flucloxacillin. The tube was removed and the patient observed for 48 hours particularly for pyrexia and rise of the white cell count. No further diagnostic aspiration of the cavity was carried out. When the patient's condition remained satisfactory, he was discharged home and was seen in the outpatient clinic at regular intervals for follow-up. No adverse reactions were seen as a result of the intrapleural instillation of antibiotics.
Infected Pneumonectomy Space Twelve patients with infected pneumonectomy space were treated. The diagnosis of infection was made on clinical grounds and confirmed by aspiration of pus from the pneumonectomy space. A concomitant bronchopleural fistula was diagnosed on the basis of a descent of the fluid level in the space. Bronchoscopy was carried out in all patients to assess the integrity of bronchial closure. Rib resection was performed in 7 patients and fenestration in 1 patient. These two procedures were carried out using standard techniques. Cyclical irrigation was employed in 4 patients. The treatment policy was based largely on the surgeon's preference. The presence of a bronchopleural fistula was a contraindication to irrigation, though on one occasion a fistula was closed surgically first and irrigation was instituted postoperatively. Initially, irrigation was done with noxythiolin through an 18-gauge intravenous catheter inserted into the second or third space anteriorly; the fluid was allowed to drain through an intercostal tube inserted into the most dependent part of the space. This technique was used for 2 patients. In the subsequent 2 patients, the intercos-
Single-tube system used for cyclical irrigation. During the inflow phase, the drainage tube is raised to the level of the shoulder.
tal tube served for both irrigation and drainage, and normal saline solution was substituted for noxythiolin solution. Irrigation was carried out in a four-hour cycle as already described. Further management also followed the same lines. Treatment by cyclical irrigation was considered successful if the pneumonectomy space was rendered sterile with no evidence of recurrence of infection during the follow-up period (minimum of three months). The clinical condition of the patient, the leukocyte count, and occasionally diagnostic aspiration of the space were the variables used for assessment.
Results The patients were followed for three months to six years.
Empyema Thoracis Two methods of treatment-decortication and irrigation-were compared, particularly in relation to the duration of empyema before treatment, duration of tube drainage, length of hospital stay, and number of failures (Table 2). Both methods share the advantage of a relatively short duration of tube drainage compared with rib resection. There was no significant difference in age and sex distribution between the decortication and irrigation groups. The mean duration of tube drainage was 7 days for the irrigation group and 13.5 days for the decortication group. The mean hospital stay was 12.3 and 17.3 days for the irrigation and decortication groups, respectively. The difference in the duration of tube drainage
87 Hakim and Milstein: Cyclical Irrigation of Pleural Empyema
Table 2. Comparison of Results of Cyclical lrrigation and Other Methods Used to Treat Pleural Empyema”
Sex
Duration of Empyema
Hospital Stay (day)
Duration of Tube Drainage
25-76 (49.4)
M, 11 F, 10
1-24wk (7)
7-21 (12.3)
5-12 day (7)
5-78 (49.5) 54-76 (57.8)
M, 11 F, 12 M, 4 F, 1
2-52 wk (10.4) 2-30wk (10.2)
8-40 (17.3) 12-42 (19.2)
6-21 day (13.5) 4-11 wk (6.5)
54-55 (54.5) 56-79 (65.8)
M, 1 F, 1 M, 9 F, 2
6-8 wk (7) 3-52 wk (9.9)
7-9 3-42 (15.4)
62
M, 1
30 yr
7
Treatmeqt
No. of Patients
Age (yr)
Irrigation
21
Decortication
23
Intercostal tube drainage
5
Aspiration
2
Rib resection
11
Fenestration
1
~
Success Rateb Results
(%)
1 Technical failure due to small bronchopleural fistula; 2 recurrences at 10 day and 3 mo, 1 decortication and 1 rib resection done 2 Failures, 1 rib resection done and 1 intercostal tube placed 4 Resolutions, tube removed 4 to 11 wk; 1 excessive drainage, rib resection done 2 Resolutions
85.7
91.3 80
100
(8)
4-92 wk (31)
...
2 Periop deaths; 3 late deaths within 6 mo; 4 resolutions, 4-8 wk and 1-2 yr; 2 still draining at 2 yr Late death at 2 yr, unrelated cause
I
”Numbers i n parentheses are mean values. qreatment was considered successful if empyema resolved and did not recur.
Table 3. Comparison of Results of Three Methods Used to Treat lnfected Pneumonectorny Spacea
Treatment
No. of Patients’
Age (yr)
Irrigation
4
Rib resection
7
52-67 (60.5) 50-69 (62.5)
Fenestration
1
59.0
Associated Bronchopleural Fistula
Hospital Stay (day)
Duration of Tube Drainage
1
8-19 (13.5) 9-20 (15.6)
5-9 day (6) 6-104 wk (26.2)
6
1
21
...
Results 3 Resolutions; 1 recurrence of infection (see text) 2 Surgical closures of fistula 2 mo later, tube removed; 2 spontaneous closures of fistula 3 and 6 mo later, tube removed; 1 fistula developed as result of recurrent tumor, death 6 mo later; 1 tube in situ 2 yr later; 1 death of unrelated cause at 6 wk Persistent fistula 5 yr later
”Numbers in parentheses are mean values. hAll patients were men.
and in the length of hospital stay was significant ( p < 0.01 by Student’s t test). There was no statistically significant difference between the mean duration of empyema prior to treatment in the two groups (7 weeks in the irrigation and 10.4 weeks in the decortication group). There was no correlation between the duration of empyema before treatment and the length of hospital stay ( r = 0.007 and 0.005 for the irrigation and decortication groups, respectively). The results of using the two protocols of cyclical irrigation were similar. In 2 patients the irrigation method failed. A recurrent empyema developed at 10 days and at three months after the discontinuation of irrigation. The duration of empyema before treatment in these patients was 4 and 3 weeks, respectively. One of the patients subsequently underwent decortication, and the other had a rib resection. In a third patient, it was difficult to achieve satisfac-
tory irrigation because of a small bronchopleural fistula, and long-term drainage had to be instituted. There were two failures in the decortication group. In 1 patient, a loculus of empyema was missed during decortication, and he had a rib resection 3 weeks after the first operation. In the second patient, a purulent discharge persisted from the chest drain site, and a sinogram showed a 20-cm-long track with several narrowings in it. The track was dilated ten months later, and a size 16F Argyle catheter was reinserted. It was possible to remove the drain after six months.
Irrigation of Infected Postpneumonectomy Space The results are shown in Table 3. In the irrigation group, 1 patient had a concomitant bronchopleural fistula. This was repaired surgically first using a pericardial flap. The defects in the pericardium were closed with a Teflon
88
The Annals of Thoracic Surgery Vol 41
No 1 January 1986
patch. The patch harbored a n infection, resulting in failure of the first trial of irrigation, and the empyema recurred 6 weeks later. The patch was subsequently removed, and a second trial of sterilizing the cavity by irrigation was successful.
Comment
Pleural Empyema Although it has become a relatively uncommon disease since the advent of antibiotics, pleural empyema is still responsible for considerable morbidity and mortality. As outlined by Samson [ 5 ] , the spectrum of surgical procedures available for managing thoracic empyema is wide, ranging from simple aspiration to complete decortication and sometimes extensive thoracoplasty. Adequate drainage, obliteration of the pleural space, and expansion of the underlying lung can sometimes be achieved by closed-chest tube thoracostomy. This is used frequently as an emergency measure to reduce toxemia in a seriously ill patient. However, it has its limitations. Without rib resection, it is difficult and at times dangerous to position the tube adequately. The technique of cyclical irrigation with or without antimicrobial solutions has extended the scope of closedchest tube drainage, and it appeared to us that it could effectively supplant decortication in the majority of patients. This method was used exclusively by one surgical group in our institution. The other surgeons continued to use decortication. We were thus able to make a comparison between these two methods of treatment, since the criteria for treatment with one or the other have remained constant. In our experience, cyclical irrigation proved to be an effective procedure. The rates of success and recurrence are similar whether irrigation or decortication is used. However, the duration of tube drainage and the length of hospital stay are significantly shorter with irrigation compared with decortication (p < 0.01). Moreover, cyclical irrigation does not preclude the subsequent use of other types of treatment, for example, decortication or rib resection, if necessary. There was no significant difference in the duration of empyema before treatment between the two groups. Therefore, we believe that the unerring selection, in ad-. vance, of the empyema that will require decortication is beyond the ability of even experienced thoracic surgeons. To subject all empyemas to this procedure is to overlook the inherent ability of the pleura to rehabilitate itself [6]. It also is exposing the patient to increased surgical trauma and to all the potential complications of a major thoracic operation. However, decortication has to be considered under
the following circumstances: the persistence or recurrence of sepsis after an adequate course of cyclical irrigation; the presence of a n underlying lung pathological condition that may have been the cause of the empyema, for example, neoplasm or bronchiectasis; and a concomitant large bronchopleural fistula.
Infected Postp"eutnotiecto'ny Space A major factor in planning a logical therapeutic approach to this problem is the presence of a concomitant bronchopleural fistula. In the absence of a fistula or following its repair, cyclical irrigation proved successful in all 4 patients with infected postpneumonectomy space whom we treated. The recurrence of infection in 1 was due to the presence of a large pericardial Teflon patch. This was removed, and a second course of irrigation succeeded in sterilizing the cavity. This regime has the advantages of a shorter duration of tube drainage and a shorter hospital stay compared with methods of open chest drainage, such as rib resection and fenestration. It is also more convenient for and better tolerated by the patient. We conclude that the method of cyclical irrigation of pleural empyema or infected postpneumonectomy space, when properly instituted, should be the first line of treatment. It is a simple, effective, and time-saving procedure. Furthermore, if this regimen fails, it does not preclude the use of other procedures. We thank Mr. T. A. H. English, Mr. J. Wallwork, and Mr. R. Cory-Pearce for allowing us to report on their patients, and Mrs. P. Norman for the excellent secretarial work.
References 1. Dieter RA, Pifarre R, Neville WE, et al: Empyema treated with neomycin irrigation and closed-chest drainage. J Thorac Cardiovasc Surg 59:496, 1970 2. Rosenfeldt FL, McGibney D, Brainbridge MV, Watson DA: Comparison between irrigation and conventional treatment for empyema and pneumonectomy space infection. Thorax 36:272, 1981 3. Maier HC: The pleura. In Sabiston DC, Spencer FC, (eds): Gibbon's Surgery of the Chest. Third edition. Philadelphia, Saunders, 1976, p p 387-388 4. Rosenfeldt FL, Glover JR, Marossy D: Systemic absorption of noxythiolin from the pleural cavity in man and in the rabbit. Thorax 36:278, 1981 5. Samson PC: Empyema thoracis: essentials of present-day management. Ann Thorac Surg 11:210, 1971 6. Langston HT: Empyema thoracis (editorial). Ann Thorac Surg 2:766, 1966
Material and Methods Between January, 1974, and July, 1984, 23 patients underwent decortication for empyema. In August, 1981, the method of cyclical irrigation was introduced by one surgical group in our institution and became their method of choice in the treatment of empyema (21 patients) and of infected pneumonectomy space (4 pa-
From the Department of Cardiothoracic Surgery, Papworth Hospital, Cambridge, England. Accepted for publication Feb 27, 1985. Address reprint requests to Mr. Hakim, Department of Cardiothoracic Surgery, Papworth Hospital, Papworth Everard, Cambridge CB3 ERE, England.
85
tients) [l, 21. Other surgeons at Papworth Hospital continued to use the other methods of treatment.
Empyema Thoracis The indications for irrigation and decortication were identical in patients with empyema thoracis. These were an empyema that had been present for at least 1 week and whose resolution was considered unlikely with aspiration or intercostal tube drainage because of a thickened visceral pleura. Unfit elderly patients and those with other diseases, such as rheumatoid arthritis, for whom early mobilization was considered essential were treated by rib resection. This study compares the two methods of treatmentdecortication and cyclical irrigation. The criteria for success of treatment were resolution of empyema and absence of recurrence of infection as indicated by the clinical picture, leukocyte count, and the appearance of serial chest radiographs. DECORTICATION. Decortication was undertaken in 23 patients using standard techniques [3]. CYCLICAL IRRIGATION. A total of 21 patients received this treatment. Initially we inserted two intercostal tubes under local anesthesia into the empyema cavity. One tube was used for instillation of a solution containing noxythiolin 1%,and the other tube, which was positioned in the most dependent part of the cavity, was used for drainage. During the inflow phase of imgation, the outflow tube was elevated to the shoulder level to create a safety valve, thus ensuring complete filling of the cavity and at the same time avoiding the generation of a high pressure that could cause air embolism. This two-tube system was used in the first 12 patients. In the subsequent 9 patients, we employed a simpler system consisting of a single tube with a Y junction to connect the chest drain to the irrigation drip through one arm and to an underwater-seal bottle through the other (Figure). At the same time we substituted normal saline solution for the antiseptic solution. This was done in an attempt to avoid adverse reactions associated with the intrapleural instillation of noxythiolin [4].These reactions are mainly nausea and vomiting, and were noted in one-third of the patients. Irrigation was done in cycles of four hours. The fluid (200 to 500 ml) was run in to fill the cavity, retained for three hours, and then allowed to drain for one hour with suction in a manner similar to that described by Rosenfeldt and associates [2]. A course of systemic antibiotics was given if the patient was pyrexial or when the pleural aspirate revealed pathogenic organisms. Irrigation was continued for 5 to 10 days until the draining fluid became clear and sterile as indicated by bacteriological cul-
86 The Annals of Thoracic Surgery Vol 41 No 1 January 1986
Table 1. Cause and Methods of Treatment of Pleural Empyema in 63 Patients Cause
No. of Patients
Postpneumonic
43
Esophageal
Pulmonary infarction
Chest trauma
4
Insertion of central line Open lung biopsy
1 1
Treatment Irrigation, 19 Decortication, 15 Rib resection, 5 Intercostal tube, 2 Aspiration, 2 Decortication, 4 Rib resection, 4 Intercostal tube, 1 Irrigation, 1 Decortication, 1 Rib resection, 2 Intercostal tube, 1 Irrigation, 1 Decortication, 2 Fenestration, 1 Decortication, 1 Intercostal tube, 1
INFLOW LINE
I
SINGLE INTERCOSTAL TUBE
b
DRAINAGE TUBE
:.:.:.:. ....... .-*..*.*.. .....:+ .. a ..
tures. The cavity was then filled with 100 ml of normal saline solution containing a combination of antibiotics, usually gentamicin sulfate, cefotaxime sodium, and flucloxacillin. The tube was removed and the patient observed for 48 hours particularly for pyrexia and rise of the white cell count. No further diagnostic aspiration of the cavity was carried out. When the patient's condition remained satisfactory, he was discharged home and was seen in the outpatient clinic at regular intervals for follow-up. No adverse reactions were seen as a result of the intrapleural instillation of antibiotics.
Infected Pneumonectomy Space Twelve patients with infected pneumonectomy space were treated. The diagnosis of infection was made on clinical grounds and confirmed by aspiration of pus from the pneumonectomy space. A concomitant bronchopleural fistula was diagnosed on the basis of a descent of the fluid level in the space. Bronchoscopy was carried out in all patients to assess the integrity of bronchial closure. Rib resection was performed in 7 patients and fenestration in 1 patient. These two procedures were carried out using standard techniques. Cyclical irrigation was employed in 4 patients. The treatment policy was based largely on the surgeon's preference. The presence of a bronchopleural fistula was a contraindication to irrigation, though on one occasion a fistula was closed surgically first and irrigation was instituted postoperatively. Initially, irrigation was done with noxythiolin through an 18-gauge intravenous catheter inserted into the second or third space anteriorly; the fluid was allowed to drain through an intercostal tube inserted into the most dependent part of the space. This technique was used for 2 patients. In the subsequent 2 patients, the intercos-
Single-tube system used for cyclical irrigation. During the inflow phase, the drainage tube is raised to the level of the shoulder.
tal tube served for both irrigation and drainage, and normal saline solution was substituted for noxythiolin solution. Irrigation was carried out in a four-hour cycle as already described. Further management also followed the same lines. Treatment by cyclical irrigation was considered successful if the pneumonectomy space was rendered sterile with no evidence of recurrence of infection during the follow-up period (minimum of three months). The clinical condition of the patient, the leukocyte count, and occasionally diagnostic aspiration of the space were the variables used for assessment.
Results The patients were followed for three months to six years.
Empyema Thoracis Two methods of treatment-decortication and irrigation-were compared, particularly in relation to the duration of empyema before treatment, duration of tube drainage, length of hospital stay, and number of failures (Table 2). Both methods share the advantage of a relatively short duration of tube drainage compared with rib resection. There was no significant difference in age and sex distribution between the decortication and irrigation groups. The mean duration of tube drainage was 7 days for the irrigation group and 13.5 days for the decortication group. The mean hospital stay was 12.3 and 17.3 days for the irrigation and decortication groups, respectively. The difference in the duration of tube drainage
87 Hakim and Milstein: Cyclical Irrigation of Pleural Empyema
Table 2. Comparison of Results of Cyclical lrrigation and Other Methods Used to Treat Pleural Empyema”
Sex
Duration of Empyema
Hospital Stay (day)
Duration of Tube Drainage
25-76 (49.4)
M, 11 F, 10
1-24wk (7)
7-21 (12.3)
5-12 day (7)
5-78 (49.5) 54-76 (57.8)
M, 11 F, 12 M, 4 F, 1
2-52 wk (10.4) 2-30wk (10.2)
8-40 (17.3) 12-42 (19.2)
6-21 day (13.5) 4-11 wk (6.5)
54-55 (54.5) 56-79 (65.8)
M, 1 F, 1 M, 9 F, 2
6-8 wk (7) 3-52 wk (9.9)
7-9 3-42 (15.4)
62
M, 1
30 yr
7
Treatmeqt
No. of Patients
Age (yr)
Irrigation
21
Decortication
23
Intercostal tube drainage
5
Aspiration
2
Rib resection
11
Fenestration
1
~
Success Rateb Results
(%)
1 Technical failure due to small bronchopleural fistula; 2 recurrences at 10 day and 3 mo, 1 decortication and 1 rib resection done 2 Failures, 1 rib resection done and 1 intercostal tube placed 4 Resolutions, tube removed 4 to 11 wk; 1 excessive drainage, rib resection done 2 Resolutions
85.7
91.3 80
100
(8)
4-92 wk (31)
...
2 Periop deaths; 3 late deaths within 6 mo; 4 resolutions, 4-8 wk and 1-2 yr; 2 still draining at 2 yr Late death at 2 yr, unrelated cause
I
”Numbers i n parentheses are mean values. qreatment was considered successful if empyema resolved and did not recur.
Table 3. Comparison of Results of Three Methods Used to Treat lnfected Pneumonectorny Spacea
Treatment
No. of Patients’
Age (yr)
Irrigation
4
Rib resection
7
52-67 (60.5) 50-69 (62.5)
Fenestration
1
59.0
Associated Bronchopleural Fistula
Hospital Stay (day)
Duration of Tube Drainage
1
8-19 (13.5) 9-20 (15.6)
5-9 day (6) 6-104 wk (26.2)
6
1
21
...
Results 3 Resolutions; 1 recurrence of infection (see text) 2 Surgical closures of fistula 2 mo later, tube removed; 2 spontaneous closures of fistula 3 and 6 mo later, tube removed; 1 fistula developed as result of recurrent tumor, death 6 mo later; 1 tube in situ 2 yr later; 1 death of unrelated cause at 6 wk Persistent fistula 5 yr later
”Numbers in parentheses are mean values. hAll patients were men.
and in the length of hospital stay was significant ( p < 0.01 by Student’s t test). There was no statistically significant difference between the mean duration of empyema prior to treatment in the two groups (7 weeks in the irrigation and 10.4 weeks in the decortication group). There was no correlation between the duration of empyema before treatment and the length of hospital stay ( r = 0.007 and 0.005 for the irrigation and decortication groups, respectively). The results of using the two protocols of cyclical irrigation were similar. In 2 patients the irrigation method failed. A recurrent empyema developed at 10 days and at three months after the discontinuation of irrigation. The duration of empyema before treatment in these patients was 4 and 3 weeks, respectively. One of the patients subsequently underwent decortication, and the other had a rib resection. In a third patient, it was difficult to achieve satisfac-
tory irrigation because of a small bronchopleural fistula, and long-term drainage had to be instituted. There were two failures in the decortication group. In 1 patient, a loculus of empyema was missed during decortication, and he had a rib resection 3 weeks after the first operation. In the second patient, a purulent discharge persisted from the chest drain site, and a sinogram showed a 20-cm-long track with several narrowings in it. The track was dilated ten months later, and a size 16F Argyle catheter was reinserted. It was possible to remove the drain after six months.
Irrigation of Infected Postpneumonectomy Space The results are shown in Table 3. In the irrigation group, 1 patient had a concomitant bronchopleural fistula. This was repaired surgically first using a pericardial flap. The defects in the pericardium were closed with a Teflon
88
The Annals of Thoracic Surgery Vol 41
No 1 January 1986
patch. The patch harbored a n infection, resulting in failure of the first trial of irrigation, and the empyema recurred 6 weeks later. The patch was subsequently removed, and a second trial of sterilizing the cavity by irrigation was successful.
Comment
Pleural Empyema Although it has become a relatively uncommon disease since the advent of antibiotics, pleural empyema is still responsible for considerable morbidity and mortality. As outlined by Samson [ 5 ] , the spectrum of surgical procedures available for managing thoracic empyema is wide, ranging from simple aspiration to complete decortication and sometimes extensive thoracoplasty. Adequate drainage, obliteration of the pleural space, and expansion of the underlying lung can sometimes be achieved by closed-chest tube thoracostomy. This is used frequently as an emergency measure to reduce toxemia in a seriously ill patient. However, it has its limitations. Without rib resection, it is difficult and at times dangerous to position the tube adequately. The technique of cyclical irrigation with or without antimicrobial solutions has extended the scope of closedchest tube drainage, and it appeared to us that it could effectively supplant decortication in the majority of patients. This method was used exclusively by one surgical group in our institution. The other surgeons continued to use decortication. We were thus able to make a comparison between these two methods of treatment, since the criteria for treatment with one or the other have remained constant. In our experience, cyclical irrigation proved to be an effective procedure. The rates of success and recurrence are similar whether irrigation or decortication is used. However, the duration of tube drainage and the length of hospital stay are significantly shorter with irrigation compared with decortication (p < 0.01). Moreover, cyclical irrigation does not preclude the subsequent use of other types of treatment, for example, decortication or rib resection, if necessary. There was no significant difference in the duration of empyema before treatment between the two groups. Therefore, we believe that the unerring selection, in ad-. vance, of the empyema that will require decortication is beyond the ability of even experienced thoracic surgeons. To subject all empyemas to this procedure is to overlook the inherent ability of the pleura to rehabilitate itself [6]. It also is exposing the patient to increased surgical trauma and to all the potential complications of a major thoracic operation. However, decortication has to be considered under
the following circumstances: the persistence or recurrence of sepsis after an adequate course of cyclical irrigation; the presence of a n underlying lung pathological condition that may have been the cause of the empyema, for example, neoplasm or bronchiectasis; and a concomitant large bronchopleural fistula.
Infected Postp"eutnotiecto'ny Space A major factor in planning a logical therapeutic approach to this problem is the presence of a concomitant bronchopleural fistula. In the absence of a fistula or following its repair, cyclical irrigation proved successful in all 4 patients with infected postpneumonectomy space whom we treated. The recurrence of infection in 1 was due to the presence of a large pericardial Teflon patch. This was removed, and a second course of irrigation succeeded in sterilizing the cavity. This regime has the advantages of a shorter duration of tube drainage and a shorter hospital stay compared with methods of open chest drainage, such as rib resection and fenestration. It is also more convenient for and better tolerated by the patient. We conclude that the method of cyclical irrigation of pleural empyema or infected postpneumonectomy space, when properly instituted, should be the first line of treatment. It is a simple, effective, and time-saving procedure. Furthermore, if this regimen fails, it does not preclude the use of other procedures. We thank Mr. T. A. H. English, Mr. J. Wallwork, and Mr. R. Cory-Pearce for allowing us to report on their patients, and Mrs. P. Norman for the excellent secretarial work.
References 1. Dieter RA, Pifarre R, Neville WE, et al: Empyema treated with neomycin irrigation and closed-chest drainage. J Thorac Cardiovasc Surg 59:496, 1970 2. Rosenfeldt FL, McGibney D, Brainbridge MV, Watson DA: Comparison between irrigation and conventional treatment for empyema and pneumonectomy space infection. Thorax 36:272, 1981 3. Maier HC: The pleura. In Sabiston DC, Spencer FC, (eds): Gibbon's Surgery of the Chest. Third edition. Philadelphia, Saunders, 1976, p p 387-388 4. Rosenfeldt FL, Glover JR, Marossy D: Systemic absorption of noxythiolin from the pleural cavity in man and in the rabbit. Thorax 36:278, 1981 5. Samson PC: Empyema thoracis: essentials of present-day management. Ann Thorac Surg 11:210, 1971 6. Langston HT: Empyema thoracis (editorial). Ann Thorac Surg 2:766, 1966