Management of pneumothorax in cystic fibrosis

Management of Pneumothorax in Cystic Fibrosis By Samuel R. Schuster, F John McLaughlin, Wallace J. Matthews, Jr, Denise J. Strieder, Kon Tiak Khaw, and Harry Shwachman Boston, Massachusetts 9 The increased longevity of patients w i t h cystic fibrosis has resulted in a concomitant increase in the frequency w i t h which pneumothorax is seen. While several approaches to this problem have been available from both a medical and surgical standpoint, unsettled questions remain regarding the efficacy of various modalities of therapy. A review of o u r o w n experience w i t h 170 episodes of pneumothorax has provided a basis for proposing w h a t appears to be a reasoned approach to therapy. During the past 12 years, 65 patients ages 5 t o 3 2 years experienced 170 episodes of pneumothorax. Of the 211 trials of treatment, all yielded a high rate of initial resolution (70% to 100%) but rates of recurrence w e r e high. The recurrence rates were: observation 60%, thoracentesis 79%, trochar thoracotomy 63%, tetracycline sclerosis 86%, and silver nitrate 43%. Quinacrine sclerosis yielded an acceptable recurrence rate of 12.5% and partial pleurectomy had no recurrence. There was no statistical difference in the pulmonary function parameters determined before pneumothorax and after chemical pleurodesis or partial pleurectomy. Based upon the data obtained in this review, we recommend that initial management of the pneumothorax include evacuation of the pneumothorax using a chest tube and then chemical pleurodesis using quinacrine sclerosis. In those cases where this p l e u r o d e s i s fails, upper-partial pleurectomy w i t h obliteration of pleural blebs via a limited thoracentesis is the t r e a t m e n t of choice. An occasional patient requires a concomitant Iobectomy.

recently, the conventional treatment used chest tube drainage, however, it has become increasingly apparent that this form of conservative therapy is inadequate since recurrence rates of pneumothorax have been reported to be as high a s 830/0. 2,3 As a result, alternate modes of therapy have been undertaken. These include chemical pleural sclerosis, 4'5 pleural abrasion 6 and parietal pleurectomy.2 In an attempt to determine which therapeutic approach has been most effective, we have retrospectively reviewed the results of the various treatment modalities carried out at the Children's Hospital Medical Center in Boston between January, 1969 and July, 1982. We have used all of the above methods of management except pleural abrasion. MATERIALS AND METHODS

Between January, 1969 and July, 1982, there were well over 3,500 admissions of which 159 were for management of pneumothorax in patients with cystic fibrosis. The age range of these patients was from 5 to 32 years and the male-to-female ratio was 1:1. Modes of Therapy

INDEX WORDS: Pneumothorax; cystic fibrosis.

There were five methods of therapy used for the management of pneumothorax.

HE MEAN SURVIVAL AGE of patients T with cystic fibrosis was 4 years in 1950; in 1976 it became 19 years) Concomitant with this

Observation

increase in longevity, there has been an increase in the frequency with which pneumothorax is seen as a complication of cystic fibrosis. Until

From the Departments of Surgery and Medicine, Children's Hospital Medical Center, and Departments of Surgery and Pediatrics, Harvard Medical School, Boston, Mass. Presented before the 31st Annual Meeting of the Surgical Section of the American Academy of Pediatrics, New York, New York, October 23-24, 1982. Address reprint requests to Samuel R. Schuster, MD, Children's Hospital Medical Center, 300 Longwood Ave, Boston, MA 02115. 9 1983 by (;rune & Stratton, Inc. 0022-3468/83/1804--0031/$01.00/0 492

A few patients with a small but stable asymptomatic pneumothorax were simply observed and treated with bed rest, antibiotic therapy, and pulmonary physiotherapy. There was no progression of the pneumothorax in these patients and they were discharged without further treatment. Thoracentesis The occasional patient with a small and minimally symptomatic pneumothorax was hospitalized for observation, but otherwise was not treated for 24 to 48 hours. We have felt that there is an increased likelihood of sealing the air leak when the lung is partially collapsed. Following this period of observation, aspiration of the Journal of Pediatric Surgery, Vol. 18, No. 4 (August), 1983

PNEUMOTHORAX IN CYSTIC FIBROSIS

493

pneumothorax was carried out using a 16-gauge catheter (Angiocath-Deseret Pharmaceuticals, Sandy, Utah) to avoid needle laceration of the lung. The catheter was always removed immediately after the procedure.

Trocar Thoracotomy A single chest tube was inserted using local anesthesia. The tube was positioned anteriorly in the apex of the chest after creating a subcutaneous tunnel over a distance of one or two ribs to ensure a good seal. Pleural Sclerosis Initial treatment was by chest tube drainage and following cessation of the air leak and reexpansion of the lung, one of three agents was instilled in the pleural space through the chest tube; silver nitrate 1% as a single 10 ml instillation; tetracycline in 50 ml of saline as a single instillation, (our tetracycline dose usually averaged 15 mg/kg of body weight with a maximum dose of 26 mg/kg of body weight); or quinacrine 2%, 100 mg in 50 ml isotonic saline instilled daily for three days. Following instillation of these agents, the chest tube was clamped for a 1-hour period during which the patient was placed in various positions; sitting upright, right and left lateral decubitus, and in both the prone and supine head down position. Each of these positions was maintained for 15 minutes to expose as much of the pleural surface as possible to the sclerosing agent. Following this, chest tube drainage was reinstituted. Parietal Pleurectomy An anterior submammary skin incision is made extending from the anterior/axillary line to the medial border of the sternum and the chest is entered through the fifth intercostal space. The only muscles transected are the pectoralis major just above its insertion on the fifth rib, occasionally 1 or 2 cm of the anterior edge of the serratus muscle and the intercostal muscles of the fifth interspace. The chest wall portion of the procedure is portrayed in Fig. 1. All of the parietal pleura extending superiorly from the fifth intercostal space is removed. Pleural blebs in the apex of the upper lobe and/or in the apex of the lower lobe were oversewn with No. 1 chromic catgut sutures using a running-locking suture. When

~

Ref?~ctedcut Majorrn

~tedge 1~7/ "ofIntercosto[rn

Fig. 1. Shows the chest wall component of a limited anterior thoracotomy for apical parietal pleurectomy. Note t h a t in addition to the intercostal muscles, the only major muscle transected is the pectoralis major. Occasionally 1 or 2 cm of the anterior edge of the serratus muscle is cut.

preoperative studies (ventilation perfusion (V/Q) scan) indicated that the upper lobe was not significantly contributing to pulmonary function and there was essentially complete involvement of the upper lobe in bleb formation, a concomitant upper lobectomy was done. There were nine extremely ill patients who

494

SCHUSTER ET AL

died within 1 week of admission or within 1 month of follow-up. The pneumothorax in these terminal patients were invariably treated with chest tube drainage and no further consideration was given to other more definitive therapy. The results were classified according to the following categories: resolution means that a posterior/ anterior chest x-ray at 1 week showed re-expansion or persistence of a stable or diminishing minimal rim of intrathoracic air less than 1 cm in width. Furthermore, there could be no further evidence of continuing air leak. These patients were discharged. Failure refers to continuing air leak after a trial of management, or relapse occurring less than 1 week after re-expansion. Recurrence refers to relapse occuring more than 1 week after re-expansion. Success refers to resolution with no recurrence up to end of the study period or the patient's death from some other cause. Pulmonary function tests were done whenever possible pre- and post-therapy, and included vital capacity, residual volume, total lung capacity, peak flow rate, forced expiratory volume/1 sec, maximal midexpiratory flow rate, and arterial blood gases.

RESULTS There were a total of 72 patients who experienced 180 pneumothoraces of which 98 were first episodes and 79 were recurrences. Forty-five had only a unilateral pneumothorax. There were 219 trials of management used to treat the 180 episodes. Initial resolution occurred in a very large number of patients, however, recurrence rates were high for observation, thoracentesis,

trocarthoracotomy, tetracycline sclerosis, and silver nitrate sclerosis. Quinacrine sclerosis and parietal pleurectomy were clearly the most successful methods of management (Table 1).

Morbidity There was no morbidity with observation or thoracentesis. Trocar thoracotomy, however, was frequently painful. Amongst the sclerosing solutions, quinacrine sclerosis was clearly the most well tolerated since the patients experienced little or no pain and underwent three daily instillations without apprehension. There was one patient who developed an urticarial rash which subsided following standard treatment. Silver nitrate instillations were painful in every case requiring morphine in significant amounts for analgesia. One patient developed a chemical pneumonitis and a large loculated pleural effusion. Tetracycline sclerosis caused initial severe pain which subsided in a few hours. Parietal pleurectomy was well tolerated despite marginal pulmonary functions, particularly the frequent presence of a low FEV 1. None of these patients required mechanical ventilation and all were extubated in the operating room or in the recovery room. One patient developed a pseudomonas aeruginosa empyema postoperatively which required several drainage procedures.

Lung Function Pulmonary function tests before pneumothorax and after pleural sclerosis were available in

Table 1. Results of 219 Trials of Management Pneumothorax in Cystic Fibrosis Method Observation Needle aspiration Trocar thoracotomy Quinacrine sclerosis Silver nitrate sclerosis Tetracycline sclerosis Parietal pleurectomy

Trials

Failure

Resolution

Recurrence

Long-term Success

Follow-up (rno)

46

13(30%)

33(70%)

20/33

13/46(28%)

1-92

31

17(55%)

14(45%)

11/14

3/34(9%)

1-39

94

29(31%)

65(69%)

41/65

26/94(28%)

13

1(8%)

12(92%)

1/13

12/13(92%)

13-98

1- 158

8

1(12.5%)

7(87.5%)

3/7

4/8 (50%)

1-31

8

1(12.5%)

7(87.5%)

6/7

1/8( 12.5 %)

1-36

20

0(0%)

20(100%)

0/20

20/20(100%)

1-48"

*Two patients died within 4 months of pleurectomy due to progressive cardiopulmonary disease without recurrent pneumothorax.

PNEUMOTHORAX IN CYSTIC FIBROSIS

12 of 17 successful trials (70%). Pulmonary function tests were available pre- and post-pleurectomy in 16 of the 20 trials (80%). There was no significant difference in pulmonary function and gas exchange before pneumothorax and after either pleural sclerosis or parietal pleurectomy. In one instance, there was a dramatic fall in all measured values following pleural sclerosis with silver nitrate, but these had all returned to their pretreatment levels within 6 weeks of the initial treatment.

Mortality Forty-seven of our 72 patients (65%) died during the 13-year study period; 82% of these deaths occurred during the 3 years following their initial presentation. In each case, the pneumothorax was under control and not directly responsible for the deaths which were attributable to cardiorespiratory failure. The other 25 patients died 1 to 82 months after their initial episode of pneumothorax. These deaths were believed to result from cardiorespiratory failure secondary to cystic fibrosis. Sixteen patients who had been managed only by trocar thoracotomy were found at autopsy to have variable degrees of pleural adhesions without complete obliteration of the pleural space. There were thick bilateral pleural adhesions in 6 patients, moderate adhesions in 3 and virtually no adhesions in the remaining 8. The patients that died during the study period had presented initially at a younger age and with lower Shwachman scores than the survivors. Furthermore, in the older age group (patients over 21 years of age at the time of initial presentation), those who died had significantly lower Shwachman scores and, therefore, had more severe pulmonary disease than did the survivors. These observations support the clinical impression that death was mainly related to severity of pulmonary disease and the progression of cardiorespiratory insufficiency rather than to age alone. DISCUSSION

The unacceptably high recurrence rate of pneumothorax in cystic fibrosis patients with moderate-to-severe disease has led to a more aggressive approach in the management of the initial episode. These newer methods of treatment have included pleural sclerosis, 4'5 parietal

495

pleurectomy, 2 or pleural abrasion 6 of the affected side, and there was even a recommendation by Rich et al., that prophylactic contralateral surgery for pleurectomy be undertaken with a first pneumothorax. We undertook this study because of the confusion and uncertainty that has arisen as a result of the recent rapid change in the treatment of pneumothorax in these patients. Our results have confirmed the high recurrence rates noted by others with conservative management. Recurrence rates in our series were 60% for observation, 75% for needle aspiration, and 64% for trocar thoracotomy. Those patients with a small pneumothorax and no respiratory distress that were admitted for observation faired no better with respect to recurrence than those whose larger and symptomatic pneumothorax required aspiration or chest tube drainage. Because of the high recurrence rate noted above, the initial clinical presentation should not direct the choice of management since the goal is definitive treatment to avoid the almost inevitable recurrence. Although silver nitrate has been reported as an effective sclerosing agent, 2 our experience with it showed that it was always associated with severe pleuritic pain. Furthermore, it was effective in only three of eight trials, and we, therefore, discontinued its use. Tetracycline has been reported to be effective for spontaneous pneumothorax, ~ but was ineffective in our use; however, we may have been using inappropriate doses. Sahn et al, ~5 showed that for tetracycline to be an effective sclerosing agent, doses of 35 mg/kg body weight should be used. The doses used in our patients were well below those recommended values and may explain our high recurrence rate. Despite this, however, the pleuritic pain associated with tetracycline was greater than that which we saw with quinacrine. It is of course possible that the use of xylocaine in the pleural space prior t o 16 o r mixed with ~7 the tetracycline instillation might diminish or totally do away with the pleuritic pain associated with its use. This study, as well as others, 2's'~8demonstrated that quinacrine is an effective and well-tolerated sclerosing agent. It was our treatment of choice up until 1977 when the manufacturer (Winthrop Laboratories, New York) discontinued production. However, Larrieu et al, 19 used another quinacrine preparation suitable for pleural instil-

496

lation (Sigma Chemical Company, St. Louis, Missouri) and have shown it to be effective in the treatment of spontaneous pneumothorax in otherwise healthy adults with follow-up periods of 6 months to 2 years. In our series, it was the most well-tolerated agent and in extended follow-up periods was the most effective sclerosing agent for the prevention of recurrent pneumothoraces. Parietal pleurectomy has been shown by others 2'~6"2~to be effective in 90% of cases. In our own series, it was effective in 100% of our cases, however, since it was used largely during the latter part of the study, the follow-up periods have not been as long. We have been very pleasantly surprised to find that despite advanced disease, these patients seem to tolerate the limited anterior thoracotomy and pleurectomy very well. Most have a low FEV1 but despite this, they tolerated the procedure quite well and have not required assisted ventilation postoperatively. The endotracheal tube in all of these patients has been removed either in the operating room or in the recovery room. In order to make this a tolerable procedure, it is important to avoid transection of many major thoracic muscles. The degree of postoperative discomfort is directly related to the patient's willingness to cough and thus clear the airway of the large amount of purulent secretions usually present. There should be a limited skin incision extending only from the anterior axillary line to the medial border of the sternum. The only muscles transected are the pectoralis major, the intercostal muscles of the fifth interspace and, occasionally, the anterior edge of the serratus muscle. The lattismus dorsi does not need to be transected. Thoracotomy has the additional advantage of allowing direct visualization of the lung and, therefore, the large blebs usually present in the apex of the upper lobe can be oversewn. In a few instances, it is advisable to carry out an upper lobectomy when that lobe is no longer contributing to the overall pulmonary function and has been largely replaced by blebs. In this and other series, 2'6 2% to 10% of the patients develop major postoperative complications following parietal pleurectomy. These include bleeding, persistent atelectasis, or chest wall infections. We have had no experience with pleural abra-

SCHUSTER ET AL

sion which has been reported to be equally effective but shares similar complications. We do, however, find it difficult to believe that one can as effectively obliterate the pleural space in the upper portion of the chest by simple abrasion as compared with parietal pleurectomy. The only recurrence that was seen following pleurectomy in our cystic fibrosis clinic was in a patient whose pleurectomy was done at another hospital through a 2-inch minithoracotomy incision in the axilla. Undoubtedly, this pleurectomy was very limited and, therefore, resulted in recurrence. This can only be avoided by carrying out an adequate thoracotomy which allows resection of the parietal pleura of the entire apical portion of the pleural space. Of our patients, 60% had a pneumothorax limited to one side and, therefore, we find it difficult to endorse the suggestion that prophylactic surgical intervention on the opposite side should be considered at the time of admission for a first episode. It might, however, be reasonable to consider quinacrine instillation using an angiocatheter on the opposite side as a prophylactic measure. Our results show little change in pulmonary function before and after pleural sclerosis or parietal pleurectomy. Similar results have been reported by others, however, if the postoperative evaluation is performed shortly after a procedure, pain or pleural effusion does interfere with the evaluation. If pulmonary function studies are done 6 months or more after a procedure, then the natural progression of the disease may be the cause of increased airway obstruction and further deterioration of pulmonary function. Our results show no evidence of a deleterious effect attributable to pleural sclerosis or parietal pleurectomy. CONCLUSIONS

The recurrence rate of conservatively treated pneumothorax in patients with cystic fibrosis varies from 60% to 83%. Once the air leak has stopped and the lung has re-expanded, both parietal pleurectomy and quinacrine pleural sclerosis appear equally effective in preventing a recurrent pneumothorax. A limited but adequate anterior thoracotomy is the best surgical approach for parietal pleurectomy. Adequate pleurectomy cannot be done through a minitho-

PNEUMOTHORAX IN CYSTIC FIBROSIS

r a c o t o m y . N e i t h e r c h e m i c a l p l e u r a l s c l e r o s i s or p a r i e t a l p l e u r e c t o m y s e e m to h a v e a s i g n i f i c a n t deleterious effect on pulmonary function. REFERENCES

1. Cystic Fibrosis Foundation: Report of the Patient Registry, 1978 2. Stowe SM, Boat TF, Mendelson H, et al: Open thoracotomy for pneumothorax in cystic fibrosis. Am Rev Respir Dis 111:611-617, 1975 3. Knight RK, Batten JC: Pneumothorax in cystic fibrosis. Eighth International Cystic Fibrosis Conference Report, 1980. pp 376-381 4. Cattaneo SM, Sirak HD, Klassen KP: Recurrent spontaneous pneumothorax in the high-risk patient. J Thorac Cardiovac Surg 66:467471, 1973 5. Jones RE, Giammona ST: Intrapleural infection of Quinacrine for pneumothorax in a child with cystic fibrosis. Am J Dis Child 130:777-778, 1976 6. Rich RH, Warwick W J, Leonard AS: Open thoracotomy and pleural abrasion in the treatment of spontaneous

497

pneumothorax in cystic fibrosis. J Pediatr Surg 13:237-242, 1978 7. Goldszer RD, Bennett J, Van Campen J. and et al: Intrapleural tetracycline for spontaneous pneumothorax. JAMA 241:724-725, 1979 8. Sahn SA, Ports DE: The effect of tetracycline on the rabbit pleura, Am Rev Respir Dis 117:493-499, 1978 9. Luck SR, Raffensperger JG, Sullivan H J, et al: Management of pneumothorax in children with chronic pulmonary disease. J Thorac Cardiovasc Surg 74:834-839, 1977 10. Wallach HW: Enterpleural therapy with tetracycline and Lidocaine for malignant pleural effusions. Chest 73:(No. 2)p 246, 1978 (letter) 11. Kattwinkel J, Taussig LM, Mclntosh CL, et al: Intrapleural instillation of quinarine for recurrent pneumothorax, JAMA 226:557-559, 1973 12. Larrieu A J, Tyers FO, Williams EH, et al: Intrapleural instillation of quinacrine for treatment of spontaneous pneumothorax. Ann Thorac Surg 28:146-150, 1979 13. Mitchell-Heggs PF, Batten JC: Pleurectomy for spontaneous pneumothorax in cystic fibrosis. Thorax 25:165-171, 1970