- Email: [email protected]
Pneumothorax in cystic fibrosis: A 26-year experience
Pneumothorax in Cystic Fibrosis: A 26-Year Experience Michael L. Spector, MD, and Robert C. Stern, MD Division of Cardiothoracic Surgery, Department of Surgery, and Cystic Fibrosis Research Center, Department of Pediatrics, Case Western Reserve School of MGic'lne, Cleveland, Ohio
We reviewed the records of all patients with cystic fibrosis and radiologically demonstrated pneumothorax at Rainbow Babies and Childrens Hospital between 1959 and 1987. There occurred 144 pneumothoraces, 71 right and 73 left, in 99 patients, 48 female and 51 male. The median survival from the date of the first pneumothorax was 29.9 months. Primary therapy included the following: observation; tube thoracostomy; tube thoracostomy with instillation of quinacrine hydrochloride, tetracycline, silver nitrate, or talc; or partial pleurectomy. Com-
plications were minimal in each group. The surgical group did significantly better than all! other groups except the group given talc. We conclude that pneumothorax is a late and ominous complication of cystic fibrosis. The primary treatment of pneumothorax should be partial pleurectomy. Talc instillation should be reserved for patients in respiratory failure who are too ill to undergo operation and for the occasional patient in whom surgical intervention fails. (Ann Thorac Surg 1989;47:204-7)
T
Statistical analysis included the use of life tables, the ,$ test, and the Fisher exact test. Many patients in all treatment groups received supplemental oxygen in low to moderate concentrations (ie, an inspired oxygen fraction < 50%). All patients received high doses of antibiotics intravenously beginning no later than the day of admission; some patients already were on a regimen of intravenous administration of antibiotics when the pneumothorax occurred. Other routine therapy for cystic fibrosis was continued and included segmental postural drainage (as tolerated), aerosolized bronchodilators and antibiotics, pancreatic replacement, and supplemental vitamins. Chest tubes were inserted under local anesthesia and, occasionally, sedation with an intravenously administered narcotic. Chemical pleurodesis was accomplished by infusing the sclerosing agent (quinacrine hydrochloride, 100 mg; tetracycline, 500 to 1,000 ing; or talc, 10 g) into the pleural space through the chest tube. The patient was then rotated in various positions to expose the entire surface of the pleura to the sclerosing a:gent. This procedure was performed once a day for three days when quinacrine or tetracycline was used and generally once for talc. Silver nitrate pleurodesis was done in the operating room under general anesthesia. Pleurectomy was performed through an axillary or anterolateral muscle-splitting incision. 17 the upper thorax, the visible parietal pleura was removed; the remainder of the pleural surfaces were abraded. Any active air leak or large blebs were oversewn or stapled. In 1 patient with a bilateral pneumothorax, the surgical approach was through a median sternotomy, which allowed access to both pleural cavities.
he incidence of pneumothorax in patients with cystic fibrosis has increased as improved treatment has resulted in prolonged survival [ 11. This complication tends to occur in older patients with severe pulmonary disease [ 2 , 31. Ideal treatment of the pneumothorax would be quick, safe, and effective, and would result in permanent prevention of recurrence. Treatment has varied from simple observation to open thoracotomy with pleurectomy and abrasion [4-61. We present our results, which include some very long (ie, > 10 year) follow-ups. The data suggest that although pneumothorax is a late and ominous complication of cystic fibrosis, partial pleurectomy is a safe and very effective treatment.
Material and Methods Cystic fibrosis was diagnosed by a positive sweat test and either typical pulmonary or digestive symptoms or a positive farxjly history. All patients with cystic fibrosis who were seen from 1959 to 1987 and who had any radiological evidence of extrapulmonary pleural air were included. Only patients whose pneumothoraces were treated at our center were included. Recurrence was defined as a pneumothorax that occurred after complete radiological resolution of a pneumothorax for at least 1 week after removal of the last chest tube. Treatment failure was defined as persistent substantial pneumothorax or continuing air leak for more than 2 weeks after therapy. All information was obtained by chart review. Clinical scoring was performed by our modification [7] of the method of Shwachman and Kulczcyki (S-K score) [8]. Presented at the Twenty-fourth Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Sep 2628, 1988.
Results
Address reprint requests to Dr Spector, University Hospitals of Cleveland, 2074 Abington Rd, Cleveland, OH 44106.
Of 1,268 patients with cystic fibrosis who were followed between 1959 and 1987, 99 had at least one episode of
0 1989 by The Society of Thoracic Surgeons
0003-4975189i53.50
Table 1. Pneumothoraces Treated During Each Period
1 .oo
No. of
Years
Pneumothoraces
1960-1965
3
1965-1970
15
1970-1975
21
1975-1980
36
1980-1985
49
1985-1987
20
pneumothorax. Forty-eight patients were female and 51 were male. Thirty-nine patients had eventual involvement on both sides; 1 patient was initially seen with bilateral pneumothoraces. There were 144 episodes (71 right and 73 left) of pneumothorax in these 99 patients. The incidence has been increasing over the years (Table 1). The average age at diagnosis of cystic fibrosis was 2.75 years, and the average age at diagnosis of pneumothorax was 19 years (range, 7 to 35 years; median age, 17 years). Table 2 shows the age distribution in each sex. More male patients were seen at an older age, probably because of their longer survival. The median survival from the date of the first pneumothorax was 29.9 months (range, 0 to 206 months) (Fig 1). Seven primary treatment approaches were used over the study period. They included the following: observation alone; chest tube alone; chest tube with quinacrine, tetracycline, silver nitrate, or talc pleurodesis; or pleurectomy and abrasion (Table 3).
Observation Observation alone was used in 28 patients (average S-K score, 54). As a group, these patients were in stable condition and only minimally compromised by a small pneumothorax. Only 14 patients (50%) experienced primary resolution, and a pneumothorax recurred in 4 patients.
Thoracostorny Alone A varying number of chest tubes (range, 1 to 4) were used as the only treatment in 39 patients (average S-K score, 43). Only 12 (31%) experienced uncomplicated primary resolution; 2 of them had recurrence. Two patients sustained substantial hemorrhage from chest tube insertion, and 1 of them needed a thoracotomy for control of the Table 2. Age and Sex Distribution of Patients With Cystic Fibrosis and Pneumothorax Male
(10 10-20 2C30 30-40
205
SPECTOR AND STERN PNEUMOTHORAX IN CYSTIC FIBROSIS
Ann Thorac Surg 1989;47204-7
Female
3
7
39 30
41
5
2
17
5
.
n
.82
9
.... ..... ........
. . I
9
n
0.00
0.00
25
,
50
.
,
75
.
,
'
I
'
100 125 160
.
-
.
9
1j5 260 2i5 250
Months
Fig I. Percent survival plotted against time from first pneumothorax. Median survival was 29.9 months.
hemorrhage and repair of the lung laceration. The average hospital stay was 22 days.
Quinacrine Pleurodesis Nine patients (average S-K score, 53) were treated with quinacrine pleurodesis. Five experienced uncomplicated resolution with no recurrence. Average length of hospitalization was 15 days. In 4 patients resolution did not occur, and surgical intervention was necessary.
Talc Pleurodesis Five patients were treated with talc as the primary therapy. All had resolution of pneumothorax, but 1 patient died ten days after the instillation of talc.
Tetracycline Pleurodesis Two of the 4 patients treated with tetracycline pleurodesis had resolution; 1 of them had recurrence. In the other 2 patients resolution was not accomplished, and surgical intervention was necessary.
Silver Nitrate Pleurodesis The 2 patients treated with silver nitrate pleurodesis did not have resolution and needed further therapy. Silver Table 3. Initial Therapy
Therapy
No. of No. of No. of Patients Resolutions Recurrences
Observation
28
14
4
Thoracostomy Pleurodesis
39
12
2
Quinacrine
9
Tetracycline
4
5 2
0 1
2
0
...
5
5
0
57
54
6
Silver nitrate Talc Pleurectomy and abrasion
Pleurectomy and abrasion was significantly better ( p < 0.01) than observation, thoracostomy, quinacrine, tetracycline, and silver nitrate.
206
SPECTOR AND STERN PNEUMOTHORAX IN CYSTIC FIBROSIS
nitrate was used early in our experience and has not been used in 20 years.
Pleurectomy and Abrasion Of 57 patients (average S-K score, 53) who underwent surgical intervention, failure to obtain primary resolution occurred in 3. One patient had a continued left-sided air leak after bilateral pleural abrasions performed through a median sternotomy. There were six recurrences (11%) among the remaining 54 patients. One patient died on day 7 of a tension pneumothorax on the nonoperative side. In 1 patient, phrenic nerve paralysis caused considerable difficulty; nerve function improved within 4 weeks, but the patient died of respiratory failure on the 52nd postoperative day. There were no wound complications.
Comment Pneumothorax is an ominous complication of cystic fibrosis. Median survival after the first pneumothorax was 29.9 months. A contralateral pneumothorax eventually occurred in 40% of our patients. Subsequent survival was usually short. One explanation for the grim survival figures for this complication is that patients with cystic fibrosis who experience a pneumothorax generally have very advanced pulmonary disease [l],which has a correspondingly poor prognosis, and would not have survived long anyway. However, it seems reasonable to hypothesize that the pneumothorax itself and related treatment contributed to the poor survival. Morbidity associated with unresolved leaks can be devastating. Pneumothorax and its treatment cause restrictive pulmonary disease, which, no doubt, contributes substantially to the subsequent downhill course. Schuster and colleagues [6] reported that pulmonary function did not change significantly after pleurectomy or pleurodesis. We attempted to look at pulmonary function before and after treatment, but because of the extreme variability over time and in each patient group, no valid statistical analysis could be performed. A very high recurrence rate has been described after observation or tube thoracostomy only [9]. We had 29% and 17% recurrence rates, respectively. Pleurectomy and abrasion, quinacrine pleurodesis, and talc pleurodesis have been reported to have low recurrence rates [l, 4-61. Quinacrine pleurodesis did result in a very low recurrence rate; however, only 5 (56%)of 9 patients achieved initial resolution. In our hands, partial pleurectomy has been the most effective treatment for pneumothorax in patients with cystic fibrosis. There is a low complication rate and a 95% success rate. It is significantly better ( p < 0.01) in achieving primary resolution than observation, thoracostomy alone, or chemical pleurodesis with quinacrine, tetracycline, or silver nitrate. Even patients with severely compromised pulmonary function came through the operation and the postoperative period amazingly well. The vast majority of patients (90%) were extubated in the operating room or within five hours postoperatively. By using a small axillary or anterolateral muscle-splitting incision, postoperative pain was kept to a minimum, and analgesic control with maintenance of cough, particularly
Ann Thorac Surg 1989;472047
after the first postoperative day, was possible. Unlike other series [lo, 111 in which major complications such as bleeding or extrapulmonary infection occurred in up to 10% of the patients, we have not experienced any such problems. An additional advantage of surgical intervention is that in a substantial number of patients, the site of the air leak can be found and oversewn. Also, the chest tubes can be accurately placed under (directvision. Talc poudrage has been very effective, although we have had only a small experience with it to date. At present, United States Pharmacopeia-certified pure talc is free from contamination with asbestos, a known carcinogen [ 11. Concern about asbestos contamination lessened the use of talc in the past. Talc poudrage does have the advantage of not requiring anesthesia or a surgical incision, although the patients still have considerable discomfort after treatment. Patients with activl. air leaks have the risks of talc aspiration and pneumonitis. Adhesions from talc also appear to be very dense and virtually eliminate the possibility of future operation in the event of hemoptysis, abscess, or bronchopleural fistula. Our data indicate that partial pleurcxtomy is the best primary therapy for pneumothorax in patients with cystic fibrosis. Patients in severe respiratory failure who are unable to undergo an operation and patients who refuse an operation should be offered closed thoracostomy tube placement and talc poudrage. This study was supported in part by grant DK 27651 from the National Institutes of Health and grants from the Cystic Fibrosis Foundation and the United Way Services of' Greater Cleveland.
References 1. Tribble G, Selden R, Rodgers B. Talc poudrage in the treatment of spontaneous pneumothoraces in patients with cystic fibrosis. Ann Surg 1986;204:677-80. 2. Boat TF, di Sant'Agnese PA, Warwick WJ, Handwerger SA. Pneumothorax in cystic fibrosis. JAMA 1969;209:1498-504. 3. Tomashefski JF, Bruce M, Stern RC, et al. Pulmonary air cysts in cystic fibrosis. Hum Pathol 1985;16:25?-61. 4. Luck SR, Raffensperger JG, Sullivan HJ, Gibson LE. Management of pneumothorax in children with chronic pulmonary disease. J Thorac Cardiovasc Surg 1977;74:834-9. 5. Penketh A, Knight RK,Hodson ME, Batten JC. Management of pneumothorax in cystic fibrosis. Thorax 1982;37:85&3. 6. Schuster SR, McLaughlin J, Matthews WJ, et al. Management of pneumothorax in adults with cystic fibrosis. J Pediatr Surg 1983;18:492-7. 7. Doershuk CF, Mathews LW, Tucker AS, et al. A five-year clinical evaluation of a therapeutic program for patients with cystic fibrosis. J Pediatr 1964;65:677-93. 8. Shwachman H, Kulczycki LL. Long-term study of one hundred five patients with cystic fibrosis. Am J Dis Child 1958;96: 6-15. 9. Davis P, di Santi'Agnese PA. Diagnosi.5 and treatment of cystic fibrosis: an update. Chest 1984;85:1302-9. 10. Rich RH, Warwick WJ, Leonard AS. Open thoracotomy and pleural abrasion in the treatment of spontaneous pneumothorax in cystic fibrosis. J Pediatr Surg 1978;13:23742. 11. Stowe SM, Boat TF, Mendelsohn H, et al. Open thoracotomy for pneumothorax in cystic fibrosis. Am Rev Respir Dis 1975;111:611-7.
Ann Thorac Surg 1989;472C4-7
SPECTOR AND STERN PNEUMOTHORAX IN CYSTIC FIBROSIS
207
DISCUSSION D R BLAIR KEAGY (Chapel Hill, NC): The authors have elegantly discussed a problem that will assume greater clinical importance as the life expectancy of cystic fibrosis victims increases. In their manuscript, however, they d o not elaborate on the pathophysiology of the pneumothorax. In lung biopsy specimens from patients with cystic fibrosis, Braman noted hypertropy of the bronchial glands, mucus plugs, and fibrosis, whereas Tomashefski described distorted elastic fibers, vascular proliferation, columnar vacuolated mesothelial c e l l s t h e so-called Dunhill l e s i o n s a n d air cysts. I would appreciate the authors’ comments regarding the gross and microscopic findings from their operative specimens. Therapeutic decisions are based on the initial failure rate and the frequency of late ipsilateral recurrence. Using data from Table 3, we can examine the percentage of late recurrences only in those patients who resolved with initial therapy. The rate of recurrence after pleurectomy and abrasion was 9%, that of tube thoracostomy was 17%, and that of the chemical pleurodeses grouped together was 8%. There is no statistical difference among these groups. Would the authors therefore agree that an attempt should first be made to resolve the problem nonoperatively? Data from the paper do certainly show a higher initial failure rate with most of the nonoperative modes of therapy. An exception is the talc pleurodesis, with which complete resolution was achieved in all 5 patients in whom the procedure was attempted. This was confirmed in another study reported by Tribble and co-workers from the University of Virginia, who followed 5 patients, aged 9 to 22 years, for up to 4 years after talc pleurodesis without evidence of recurrence. In the present study the authors state that disadvantages of talc pleurodesis include the possibility of talc aspiration and the formation of dense adhesions that prevent future thoracotomy in the event of hemoptysis, abscess, or bronchopleural fistula. To better compare the risks of pleurectomy versus the use of talc, I would appreciate having the authors inform us how often these late operations actually d o occur. In this study, some of the partial pleurectomies and pleurodeses were carried out using an axillary approach. We found this incision to be quite successful in patients with congenital apical blebs. However, it is difficult to excise blebs in other areas of the lung using this approach. Do the authors note the pathology in their series to be primarily in the lung apex? Recent reports such as one in the August 1987 issue of the British Medical Journal describe heart-lung transplantations in individuals with end-stage cystic fibrosis. If this trend continues in the future, do the authors feel that the scarring that results
from either chemical or surgical sclerosis of the pleural cavity will prevent or make more hazardous later organ transplantation? And, finally, with regard to the initial 69% failure rate with tube thoracostomy, did the lung fail to expand due to insterstitial fibrosis or pleural scarring, or was there a persistent air leak from a diseased bronchus? Also, why did 2 of 39 patients have significant hemorrhage during chest tube insertion? Are there vascular adhesions in the pleural space that make thoracostomy tube insertion more hazardous in patients with cystic fibrosis? DR SPECTOR Dr Tomashefski, at Rainbow Babies and Childrens Hospital, did autopsies on the majority of these patients and described the findings and pathophysiology in his paper. Certainly we’ve seen a changing patient profile with cystic fibrosis. Most of the patients are now treated with steroids and are of a much older age group. We d o not feel that nonoperative therapy should be tried first because the period that patients have a chest tube in place or are being treated with another modality is time lost from being out of the hospital and able to undergo adequate postural drainage and other therapy. Therefore, we feel that the best therapy should be the first one. We’ve also noted that the patients whose original therapy failed and who went on to have second therapy of either quinacrine, tetracycline, talc, or operation, had a good resolution rate and fairly low recurrence rate. However, surgical treatment, when performed after a previous therapy, had a slightly higher incidence of complications. One patient had a wound infection. One can postulate that this patient had a chest tube in for a long period of time and was previously treated with Atabrine, which may have increased the chance of wound infection. There were also 2 patients who had substantial bleeding postoperatively, and this may be due to increased adhesion formation after the initial treatment. The surgical approach, a high axillary incision, gives very good exposure to the apex, in which the majority of the blebs are found. The other place that blebs are found is in the superior segment of the lower lobe, which is also easily accessible through this incision. The question of lung transplantation is something that we have thought about, although it’s not really our therapy that’s eliminating patients from the possibility of receiving a transplant. It is the fact that they have a pneumothorax that has to be treated that’s eliminating them from transplantation. Patients with cystic fibrosis have very stiff lungs and often have adhesions. Therefore, chest tube placement is not without its risks, and the lung and adhesions can be torn.
We reviewed the records of all patients with cystic fibrosis and radiologically demonstrated pneumothorax at Rainbow Babies and Childrens Hospital between 1959 and 1987. There occurred 144 pneumothoraces, 71 right and 73 left, in 99 patients, 48 female and 51 male. The median survival from the date of the first pneumothorax was 29.9 months. Primary therapy included the following: observation; tube thoracostomy; tube thoracostomy with instillation of quinacrine hydrochloride, tetracycline, silver nitrate, or talc; or partial pleurectomy. Com-
plications were minimal in each group. The surgical group did significantly better than all! other groups except the group given talc. We conclude that pneumothorax is a late and ominous complication of cystic fibrosis. The primary treatment of pneumothorax should be partial pleurectomy. Talc instillation should be reserved for patients in respiratory failure who are too ill to undergo operation and for the occasional patient in whom surgical intervention fails. (Ann Thorac Surg 1989;47:204-7)
T
Statistical analysis included the use of life tables, the ,$ test, and the Fisher exact test. Many patients in all treatment groups received supplemental oxygen in low to moderate concentrations (ie, an inspired oxygen fraction < 50%). All patients received high doses of antibiotics intravenously beginning no later than the day of admission; some patients already were on a regimen of intravenous administration of antibiotics when the pneumothorax occurred. Other routine therapy for cystic fibrosis was continued and included segmental postural drainage (as tolerated), aerosolized bronchodilators and antibiotics, pancreatic replacement, and supplemental vitamins. Chest tubes were inserted under local anesthesia and, occasionally, sedation with an intravenously administered narcotic. Chemical pleurodesis was accomplished by infusing the sclerosing agent (quinacrine hydrochloride, 100 mg; tetracycline, 500 to 1,000 ing; or talc, 10 g) into the pleural space through the chest tube. The patient was then rotated in various positions to expose the entire surface of the pleura to the sclerosing a:gent. This procedure was performed once a day for three days when quinacrine or tetracycline was used and generally once for talc. Silver nitrate pleurodesis was done in the operating room under general anesthesia. Pleurectomy was performed through an axillary or anterolateral muscle-splitting incision. 17 the upper thorax, the visible parietal pleura was removed; the remainder of the pleural surfaces were abraded. Any active air leak or large blebs were oversewn or stapled. In 1 patient with a bilateral pneumothorax, the surgical approach was through a median sternotomy, which allowed access to both pleural cavities.
he incidence of pneumothorax in patients with cystic fibrosis has increased as improved treatment has resulted in prolonged survival [ 11. This complication tends to occur in older patients with severe pulmonary disease [ 2 , 31. Ideal treatment of the pneumothorax would be quick, safe, and effective, and would result in permanent prevention of recurrence. Treatment has varied from simple observation to open thoracotomy with pleurectomy and abrasion [4-61. We present our results, which include some very long (ie, > 10 year) follow-ups. The data suggest that although pneumothorax is a late and ominous complication of cystic fibrosis, partial pleurectomy is a safe and very effective treatment.
Material and Methods Cystic fibrosis was diagnosed by a positive sweat test and either typical pulmonary or digestive symptoms or a positive farxjly history. All patients with cystic fibrosis who were seen from 1959 to 1987 and who had any radiological evidence of extrapulmonary pleural air were included. Only patients whose pneumothoraces were treated at our center were included. Recurrence was defined as a pneumothorax that occurred after complete radiological resolution of a pneumothorax for at least 1 week after removal of the last chest tube. Treatment failure was defined as persistent substantial pneumothorax or continuing air leak for more than 2 weeks after therapy. All information was obtained by chart review. Clinical scoring was performed by our modification [7] of the method of Shwachman and Kulczcyki (S-K score) [8]. Presented at the Twenty-fourth Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Sep 2628, 1988.
Results
Address reprint requests to Dr Spector, University Hospitals of Cleveland, 2074 Abington Rd, Cleveland, OH 44106.
Of 1,268 patients with cystic fibrosis who were followed between 1959 and 1987, 99 had at least one episode of
0 1989 by The Society of Thoracic Surgeons
0003-4975189i53.50
Table 1. Pneumothoraces Treated During Each Period
1 .oo
No. of
Years
Pneumothoraces
1960-1965
3
1965-1970
15
1970-1975
21
1975-1980
36
1980-1985
49
1985-1987
20
pneumothorax. Forty-eight patients were female and 51 were male. Thirty-nine patients had eventual involvement on both sides; 1 patient was initially seen with bilateral pneumothoraces. There were 144 episodes (71 right and 73 left) of pneumothorax in these 99 patients. The incidence has been increasing over the years (Table 1). The average age at diagnosis of cystic fibrosis was 2.75 years, and the average age at diagnosis of pneumothorax was 19 years (range, 7 to 35 years; median age, 17 years). Table 2 shows the age distribution in each sex. More male patients were seen at an older age, probably because of their longer survival. The median survival from the date of the first pneumothorax was 29.9 months (range, 0 to 206 months) (Fig 1). Seven primary treatment approaches were used over the study period. They included the following: observation alone; chest tube alone; chest tube with quinacrine, tetracycline, silver nitrate, or talc pleurodesis; or pleurectomy and abrasion (Table 3).
Observation Observation alone was used in 28 patients (average S-K score, 54). As a group, these patients were in stable condition and only minimally compromised by a small pneumothorax. Only 14 patients (50%) experienced primary resolution, and a pneumothorax recurred in 4 patients.
Thoracostorny Alone A varying number of chest tubes (range, 1 to 4) were used as the only treatment in 39 patients (average S-K score, 43). Only 12 (31%) experienced uncomplicated primary resolution; 2 of them had recurrence. Two patients sustained substantial hemorrhage from chest tube insertion, and 1 of them needed a thoracotomy for control of the Table 2. Age and Sex Distribution of Patients With Cystic Fibrosis and Pneumothorax Male
(10 10-20 2C30 30-40
205
SPECTOR AND STERN PNEUMOTHORAX IN CYSTIC FIBROSIS
Ann Thorac Surg 1989;47204-7
Female
3
7
39 30
41
5
2
17
5
.
n
.82
9
.... ..... ........
. . I
9
n
0.00
0.00
25
,
50
.
,
75
.
,
'
I
'
100 125 160
.
-
.
9
1j5 260 2i5 250
Months
Fig I. Percent survival plotted against time from first pneumothorax. Median survival was 29.9 months.
hemorrhage and repair of the lung laceration. The average hospital stay was 22 days.
Quinacrine Pleurodesis Nine patients (average S-K score, 53) were treated with quinacrine pleurodesis. Five experienced uncomplicated resolution with no recurrence. Average length of hospitalization was 15 days. In 4 patients resolution did not occur, and surgical intervention was necessary.
Talc Pleurodesis Five patients were treated with talc as the primary therapy. All had resolution of pneumothorax, but 1 patient died ten days after the instillation of talc.
Tetracycline Pleurodesis Two of the 4 patients treated with tetracycline pleurodesis had resolution; 1 of them had recurrence. In the other 2 patients resolution was not accomplished, and surgical intervention was necessary.
Silver Nitrate Pleurodesis The 2 patients treated with silver nitrate pleurodesis did not have resolution and needed further therapy. Silver Table 3. Initial Therapy
Therapy
No. of No. of No. of Patients Resolutions Recurrences
Observation
28
14
4
Thoracostomy Pleurodesis
39
12
2
Quinacrine
9
Tetracycline
4
5 2
0 1
2
0
...
5
5
0
57
54
6
Silver nitrate Talc Pleurectomy and abrasion
Pleurectomy and abrasion was significantly better ( p < 0.01) than observation, thoracostomy, quinacrine, tetracycline, and silver nitrate.
206
SPECTOR AND STERN PNEUMOTHORAX IN CYSTIC FIBROSIS
nitrate was used early in our experience and has not been used in 20 years.
Pleurectomy and Abrasion Of 57 patients (average S-K score, 53) who underwent surgical intervention, failure to obtain primary resolution occurred in 3. One patient had a continued left-sided air leak after bilateral pleural abrasions performed through a median sternotomy. There were six recurrences (11%) among the remaining 54 patients. One patient died on day 7 of a tension pneumothorax on the nonoperative side. In 1 patient, phrenic nerve paralysis caused considerable difficulty; nerve function improved within 4 weeks, but the patient died of respiratory failure on the 52nd postoperative day. There were no wound complications.
Comment Pneumothorax is an ominous complication of cystic fibrosis. Median survival after the first pneumothorax was 29.9 months. A contralateral pneumothorax eventually occurred in 40% of our patients. Subsequent survival was usually short. One explanation for the grim survival figures for this complication is that patients with cystic fibrosis who experience a pneumothorax generally have very advanced pulmonary disease [l],which has a correspondingly poor prognosis, and would not have survived long anyway. However, it seems reasonable to hypothesize that the pneumothorax itself and related treatment contributed to the poor survival. Morbidity associated with unresolved leaks can be devastating. Pneumothorax and its treatment cause restrictive pulmonary disease, which, no doubt, contributes substantially to the subsequent downhill course. Schuster and colleagues [6] reported that pulmonary function did not change significantly after pleurectomy or pleurodesis. We attempted to look at pulmonary function before and after treatment, but because of the extreme variability over time and in each patient group, no valid statistical analysis could be performed. A very high recurrence rate has been described after observation or tube thoracostomy only [9]. We had 29% and 17% recurrence rates, respectively. Pleurectomy and abrasion, quinacrine pleurodesis, and talc pleurodesis have been reported to have low recurrence rates [l, 4-61. Quinacrine pleurodesis did result in a very low recurrence rate; however, only 5 (56%)of 9 patients achieved initial resolution. In our hands, partial pleurectomy has been the most effective treatment for pneumothorax in patients with cystic fibrosis. There is a low complication rate and a 95% success rate. It is significantly better ( p < 0.01) in achieving primary resolution than observation, thoracostomy alone, or chemical pleurodesis with quinacrine, tetracycline, or silver nitrate. Even patients with severely compromised pulmonary function came through the operation and the postoperative period amazingly well. The vast majority of patients (90%) were extubated in the operating room or within five hours postoperatively. By using a small axillary or anterolateral muscle-splitting incision, postoperative pain was kept to a minimum, and analgesic control with maintenance of cough, particularly
Ann Thorac Surg 1989;472047
after the first postoperative day, was possible. Unlike other series [lo, 111 in which major complications such as bleeding or extrapulmonary infection occurred in up to 10% of the patients, we have not experienced any such problems. An additional advantage of surgical intervention is that in a substantial number of patients, the site of the air leak can be found and oversewn. Also, the chest tubes can be accurately placed under (directvision. Talc poudrage has been very effective, although we have had only a small experience with it to date. At present, United States Pharmacopeia-certified pure talc is free from contamination with asbestos, a known carcinogen [ 11. Concern about asbestos contamination lessened the use of talc in the past. Talc poudrage does have the advantage of not requiring anesthesia or a surgical incision, although the patients still have considerable discomfort after treatment. Patients with activl. air leaks have the risks of talc aspiration and pneumonitis. Adhesions from talc also appear to be very dense and virtually eliminate the possibility of future operation in the event of hemoptysis, abscess, or bronchopleural fistula. Our data indicate that partial pleurcxtomy is the best primary therapy for pneumothorax in patients with cystic fibrosis. Patients in severe respiratory failure who are unable to undergo an operation and patients who refuse an operation should be offered closed thoracostomy tube placement and talc poudrage. This study was supported in part by grant DK 27651 from the National Institutes of Health and grants from the Cystic Fibrosis Foundation and the United Way Services of' Greater Cleveland.
References 1. Tribble G, Selden R, Rodgers B. Talc poudrage in the treatment of spontaneous pneumothoraces in patients with cystic fibrosis. Ann Surg 1986;204:677-80. 2. Boat TF, di Sant'Agnese PA, Warwick WJ, Handwerger SA. Pneumothorax in cystic fibrosis. JAMA 1969;209:1498-504. 3. Tomashefski JF, Bruce M, Stern RC, et al. Pulmonary air cysts in cystic fibrosis. Hum Pathol 1985;16:25?-61. 4. Luck SR, Raffensperger JG, Sullivan HJ, Gibson LE. Management of pneumothorax in children with chronic pulmonary disease. J Thorac Cardiovasc Surg 1977;74:834-9. 5. Penketh A, Knight RK,Hodson ME, Batten JC. Management of pneumothorax in cystic fibrosis. Thorax 1982;37:85&3. 6. Schuster SR, McLaughlin J, Matthews WJ, et al. Management of pneumothorax in adults with cystic fibrosis. J Pediatr Surg 1983;18:492-7. 7. Doershuk CF, Mathews LW, Tucker AS, et al. A five-year clinical evaluation of a therapeutic program for patients with cystic fibrosis. J Pediatr 1964;65:677-93. 8. Shwachman H, Kulczycki LL. Long-term study of one hundred five patients with cystic fibrosis. Am J Dis Child 1958;96: 6-15. 9. Davis P, di Santi'Agnese PA. Diagnosi.5 and treatment of cystic fibrosis: an update. Chest 1984;85:1302-9. 10. Rich RH, Warwick WJ, Leonard AS. Open thoracotomy and pleural abrasion in the treatment of spontaneous pneumothorax in cystic fibrosis. J Pediatr Surg 1978;13:23742. 11. Stowe SM, Boat TF, Mendelsohn H, et al. Open thoracotomy for pneumothorax in cystic fibrosis. Am Rev Respir Dis 1975;111:611-7.
Ann Thorac Surg 1989;472C4-7
SPECTOR AND STERN PNEUMOTHORAX IN CYSTIC FIBROSIS
207
DISCUSSION D R BLAIR KEAGY (Chapel Hill, NC): The authors have elegantly discussed a problem that will assume greater clinical importance as the life expectancy of cystic fibrosis victims increases. In their manuscript, however, they d o not elaborate on the pathophysiology of the pneumothorax. In lung biopsy specimens from patients with cystic fibrosis, Braman noted hypertropy of the bronchial glands, mucus plugs, and fibrosis, whereas Tomashefski described distorted elastic fibers, vascular proliferation, columnar vacuolated mesothelial c e l l s t h e so-called Dunhill l e s i o n s a n d air cysts. I would appreciate the authors’ comments regarding the gross and microscopic findings from their operative specimens. Therapeutic decisions are based on the initial failure rate and the frequency of late ipsilateral recurrence. Using data from Table 3, we can examine the percentage of late recurrences only in those patients who resolved with initial therapy. The rate of recurrence after pleurectomy and abrasion was 9%, that of tube thoracostomy was 17%, and that of the chemical pleurodeses grouped together was 8%. There is no statistical difference among these groups. Would the authors therefore agree that an attempt should first be made to resolve the problem nonoperatively? Data from the paper do certainly show a higher initial failure rate with most of the nonoperative modes of therapy. An exception is the talc pleurodesis, with which complete resolution was achieved in all 5 patients in whom the procedure was attempted. This was confirmed in another study reported by Tribble and co-workers from the University of Virginia, who followed 5 patients, aged 9 to 22 years, for up to 4 years after talc pleurodesis without evidence of recurrence. In the present study the authors state that disadvantages of talc pleurodesis include the possibility of talc aspiration and the formation of dense adhesions that prevent future thoracotomy in the event of hemoptysis, abscess, or bronchopleural fistula. To better compare the risks of pleurectomy versus the use of talc, I would appreciate having the authors inform us how often these late operations actually d o occur. In this study, some of the partial pleurectomies and pleurodeses were carried out using an axillary approach. We found this incision to be quite successful in patients with congenital apical blebs. However, it is difficult to excise blebs in other areas of the lung using this approach. Do the authors note the pathology in their series to be primarily in the lung apex? Recent reports such as one in the August 1987 issue of the British Medical Journal describe heart-lung transplantations in individuals with end-stage cystic fibrosis. If this trend continues in the future, do the authors feel that the scarring that results
from either chemical or surgical sclerosis of the pleural cavity will prevent or make more hazardous later organ transplantation? And, finally, with regard to the initial 69% failure rate with tube thoracostomy, did the lung fail to expand due to insterstitial fibrosis or pleural scarring, or was there a persistent air leak from a diseased bronchus? Also, why did 2 of 39 patients have significant hemorrhage during chest tube insertion? Are there vascular adhesions in the pleural space that make thoracostomy tube insertion more hazardous in patients with cystic fibrosis? DR SPECTOR Dr Tomashefski, at Rainbow Babies and Childrens Hospital, did autopsies on the majority of these patients and described the findings and pathophysiology in his paper. Certainly we’ve seen a changing patient profile with cystic fibrosis. Most of the patients are now treated with steroids and are of a much older age group. We d o not feel that nonoperative therapy should be tried first because the period that patients have a chest tube in place or are being treated with another modality is time lost from being out of the hospital and able to undergo adequate postural drainage and other therapy. Therefore, we feel that the best therapy should be the first one. We’ve also noted that the patients whose original therapy failed and who went on to have second therapy of either quinacrine, tetracycline, talc, or operation, had a good resolution rate and fairly low recurrence rate. However, surgical treatment, when performed after a previous therapy, had a slightly higher incidence of complications. One patient had a wound infection. One can postulate that this patient had a chest tube in for a long period of time and was previously treated with Atabrine, which may have increased the chance of wound infection. There were also 2 patients who had substantial bleeding postoperatively, and this may be due to increased adhesion formation after the initial treatment. The surgical approach, a high axillary incision, gives very good exposure to the apex, in which the majority of the blebs are found. The other place that blebs are found is in the superior segment of the lower lobe, which is also easily accessible through this incision. The question of lung transplantation is something that we have thought about, although it’s not really our therapy that’s eliminating patients from the possibility of receiving a transplant. It is the fact that they have a pneumothorax that has to be treated that’s eliminating them from transplantation. Patients with cystic fibrosis have very stiff lungs and often have adhesions. Therefore, chest tube placement is not without its risks, and the lung and adhesions can be torn.