- Email: [email protected]
Surgical management of broncholithiasis
J
THoRAc CARDIOVASC SURG
90:842-848, 1985
Surgical management of broncholithiasis Fifty-two patients (31 men and 21 women) were treated for complications of broncholitbiasis between 1969 and 1984. Mean age was 50.8 years (range 26 to 74 years). IndicatioM for operation included symptolll'l in 49 patients and an abnormal chest x-ray film in three. Broncholithectomy was initially attempted by thoracotomy in 40 patients and by bronchoscopy in 12. In the thoracotomy group broncholithectomy was successful in aU patients, 32 of whom underwent pulmonary resection. Significant postoperative complicatioM occurred in five patients (12.8 %). There was one postoperative death (2.5 %). In the bronchoscopy group broncholithectomy was successful in eight patients (67% ~ significant complicatiOM occurred in two and there were no early deaths, Subsequent thoracotomy was done in three of the four patients in whom bronchoscopic removal was UMuccessfui. FoUow-upaveraged 76.5 montm (range 6 to 183 ~ntm). The 15 year survival rate (Kaplan-Meier) for aU patients was 75.1 % and did not differ from a control group of patients. No patient in the thoracotomy group bad recurrent complicatioM of broncholitbiasis. In contrast, complications recurred in three of the eight patients (37.5 %) successfuUy treated by bronchoscopy. We conclude that broncholithectomy via thoracotomy is the preferred treatment, as the risks are low and the long-term results are excellent,
Victor F. Trastek, M.D. (by invitation), Peter C. Pairolero, M.D., Eric L. Ceithaml, M.D. (by invitation), Jeffrey M. Piehler, M.D. (by invitation), W. Spencer Payne, M.D., and Philip E. Bernatz, M.D., Rochester, Minn.
Broncholithiasis is an uncommon problem with lifethreatening complications, jhe management of which can be formidable.' The purpose of this study was to review our current experience with removal of symptomatic broncholiths so as to determine the relative merits of surgical versus endoscopic management. We also wished to determine if there was any evidence associating broncholithiasis with lung cancer.
Patients and methods Between Jan. 1, 1969, and Jan. 1, 1984, 52 patients were treated for complications of broncholithiasis. Although broncholithiasis implies that at least a portion of a calcified lymph node has eroded into the lumen of the bronchus.v' we view this as an oversimplification and believe that the disease is a dynamic process with a spectrum of presentations. 1 Erosion of the broncholith
From the Section of Thoracic and Cardiovascular Surgery, Department of Surgery, Mayo Clinic and Mayo Foundation, Rochester, Minn. Read at the Sixty-fifth Annual Meeting of The American Association for Thoracic Surgery, New Orleans, Louisiana, April 29-May I, 1985. Address for reprints: Peter C. Pairolero, M.D., 200 First St., S.W., Rochester, Minn. 55905.
842
Table I. Presenting symptoms in 52 patients with broncholithiasis Symptom
No.
Percent
Chronic cough Purulent sputum Hemoptysis Fever/chills Chest pain Expectoration of stones Wheezing Dyspnea Fatigue Sinusitis Asymptomatic
47 37 30 18
90.4 71.2 57.7 34.6 21.2 19.2 7.7 5.8 3.9 3.9 5.8
II
10 4 3 2 2 3
into the lumen is only one of the possible complications. We also include under the heading of broncholithiasis the distortion or partial obstruction of the tracheobronchial tree caused by a calcified peribronchial lymph node. Finally, these patients may have a pulmonary mass suggestive of malignancy. In this study, patients were considered to have broncholithiasis if calcific hilar or carinal lymph nodes were present roentgenographically, if on bronchoscopic examination there was evidence of peribronchial disease (bronchial compression, stenosis, or distortion), and if there was no other
Volume 90
Broncholithiasis 8 4 3
Number 6 December, 1985
Fig. 1. A 39-year-old man with an 18 month history of chronic cough and recurrent fever and chills. A, Posteroanterior chest roentgenogram demonstrating right hilar mass. B, Esophagogram demonstrating small traction diverticulum. C. Computed tomogram demonstrating peribronchial calcified mass.
pulmonary disease to explain these changes.' Asymptomatic patients with an indeterminate roentgenographic pulmonary mass which at thoracotomy was found to be caused by calcified lymph nodes were also considered to have broncholithiasis. During this time interval, there were 52 patients who met these criteria, and these 52 patients form the basis for this report. Survival probabilities were calculated by the KaplanMeier" method. Operative deaths and deaths owing to any cause were included in the survival statistics. Operative mortality included all deaths occurring within the first 30 postoperative days. Expected survival curves were used as controls and were based on death from all
causes from the West North Central United States 1970 Life Table data and were matched for age and sex.
Clinical features There were 31 men and 21 women. Ages ranged from 26 to 74 years (mean 50.8 years). Forty-nine patients were symptomatic (Table I). Forty-seven patients had a cough. Thirty-seven of these patients had a cough productive of purulent sputum and four had severe coughing spells. Eighteen patients had clinical evidence of acute obstructive pneumonitis. Hemoptysis of more than 4 ounces occurred in five patients. Seven of the 10 patients who coughed up stones did so on multiple
8 44
The Journal of Thoracic and Cardiovascular
Trastek et al.
Surgery
Table D. Chest roentgenographic findings in 52 patients with broncholithiasis
Peribronchial calcification Infiltrate Parenchymal mass Atelectasis Tracheobronchial distortion Hilar mass Other
No.
Percent
34 29 22 12 2 2 4
65.4 55.8 42.3 23.1 3.9 3.9 7.7
m.
Table Tomographic findings in 29 patients with broncholithiasis Hilar calcification Tracheobronchial distortion Parenchymal mass Infiltrate Other
No.
Percent
22 II 6 6 4
75.9 37.9 20.7 20.7 13.8
Table IV. Bronchoscopic findings in 52 patients with broncholithiasis Tracheobronchial distortion Inflammation Visible broncholith Bleeding Purulent secretions Normal
No.
Percent
45
86.5 69.2 34.6 21.6 11.5 3.9
36 18 II 6
2
occasions. The three asymptomatic patients underwent operation because their roentgenographic fmdings could not be differentiated from malignant lesions. Diagnostic studies. Posteroanterior chest. roentgenograms (Fig. 1) in all patients were abnormal (Table II). Hilar or peribronchial calcification was present in 34 patients, postobstructive pneumonitis in 29, and a parenchymal mass suggestive of malignancy in 22. Tomography was performed in 29 patients (Table III); 22 had hilar calcification, and 11 had tracheobronchial distortion with adjacent calcification. Bronchography was performed in seven patients and showed abnormalities in four. Three of these patients had bronchial stenosis and one had bronchiectasis. Bronchoscopy was done in all 52 patients (Table IV), and the findings were abnormal in 50. Bronchial stenosis and distortion as well as inflammation were the most common observations. A broncholith was visible in 18 patients (Fig. 2). Active bleeding was present in 11 patients but was severe in only one. None of the 52
patients had bronchial brushings or washings that were cytologically positive for malignancy. The two patients who had normal bronchoscopic examination had an indeterminate pulmonary mass. No patient had a fistula into either the esophagus or major vessels. Treatment. All 52 patients had attempted broncholithectomy with or without pulmonary resection. This was effected at thoracotomy in 40 patients and initially attempted at bronchoscopy in 12. Indications for treatment included symptomatic obstructive pneumonitis in 31 patients, recurrent significant hemoptysis in 20, a mass suggestive of malignancy in 20, chronic severe cough in two, and bronchiectasis in one. One patient had urgent thoracotomy for severe hemoptysis. At thoracotomy, pulmonary resection along with removal of the broncholith was performed in 32 patients. Twenty-one patients had lobectomy, six had either segmentectomy or wedge resection, three had pneumonectomy, and two had bilobectomy. The other eight patients had broncholithectomy without pulmonary resection. One of these latter patients had an associated bronchoplasty and one had a sleeve resection of the right bronchus intermedius. The remaining six patients had bronchial distortion without erosion, and broncholithectomy was possible without the need for either bronchotomy or bronchoplasty. Intraoperative complications developed in five patients, with significant bleeding in three and esophageal injury in two. Pneumonectomy was necessary in one of these patients to control bleeding. This patient subsequently died 3 days later of cardiorespiratory failure. Bleeding in the other two patients was controlled without further complication. The esophageal injury in each patient was repaired primarily and both patients subsequently did well. Median hospitalization for patients surviving thoracotomy was 7 days and ranged from 5 to 37 days. There was one postoperative death (2.5%) as described earlier. Significant postoperative complications occurred in five other patients (12.8%) and included bleeding in two, and vocal cord paralysis, confirmed esophageal leak, and wound infection in one patient each. None of these postoperative complications occurred in patients who had intraoperative complications. The two patients with postoperative hemorrhage required reexploration to control bleeding. The remaining three patients responded to conservative management. Twelve patients had bronchoscopic broncholithectomy. These patients were selected for endoscopic treatment because the broncholith was visible and mobile. In none of these patients was operation contraindicated for medical reasons. Ten of these patients had obstructive pneumonitis and three had recurrent hemoptysis. Com-
Volume 90
Broncholithiasis 8 4 5
Number 6 December. 1985
Fig. 2. Operative specimen showing a broncholith (arrow) which has eroded into the lumen of the bronchus.
plete removal of the broncholith was possible in eight patients. There were two complications consisting of bronchial laceration and bleeding. Both responded to conservative therapy. There were no deaths. Three of the four patients in whom broncholithectomy was incomplete had subsequent operation. These three patients are not included in the 40 patients undergoing operation reported here (see Follow-up). Pathology. The location of the broncholith varied and often involved multiple portions of the tracheobronchial tree. The broncholith was predominantly on the right in 28 patients, on the left in 19, subcarinal in three, and bilateral in two. Chronic inflammation was present in 31 of the operative specimens, bronchiectasis in 13, lung abscess in seven, pneumonitis in six, nonspecific granulomas in three, and histoplasmosis in two. No patient had evidence of active tuberculosis or carcinoma in the resected specimen. Follow-up. Two patients were lost to follow-up evaluation. Follow-up was complete in the remaining 49 surviving patients (96%) and averaged 76.5 months (range 6 to 183 months). The actuarial 15 year survival rate for all 50 patients (including postoperative death) was 75.1%. This did not differ statistically from that of a similar group of patients matched for sex and age (Fig. 3). Thirty-seven of the 39 operative survivors were available for follow-up analysis. Follow-up ranged from 6 to 183 months with a mean of 98.8 months. There were four late deaths, none related to a complication of treatment. Cause of death was atherosclerosis in two patients and complications of a duodenal ulcer operation
100
l
80
iii >
's ~
Ul
(2)
60
'0
~
:0
40
.c
a: 0
20
0
0
5
10
15
Years
Fig. 3. Overall survival rate (death from all causes) of 50 patients with broncholithiasis as compared to expected survival matched for age, sex, and calendar year. Zero time on abscissa represents day of operation.
in one. In the fourth patient, lung cancer developed 4 years later in the same area that had contained the broncholith. After pulmonary resection, this patient died 2 years later of disseminated cancer. The remaining 33 patients are alive without pulmonary symptoms. No patient had recurrent complication from broncholithiasis after operation. In three of the eight patients who had initially successful bronchoscopic broncholithectomy, the disease recurred 6, 15, and 57 months later. One of these patients subsequently had a right middle lobectomy and has done well for 5 years. Another patient had repeat bronchoscopic broncholithectomy and has done well for
The Journal Of
8 4 6 Trastek et al.
10 years. The last patient had early recurrence of hemoptysis. Further treatment was refused, and this patient died of massive hemoptysis 39 months later. A second late death also occurred during follow-up. In this patient a squamous cell carcinoma developed in the same lobe from which the broncholith had been removed 38 months previously. Pulmonary resection was refused and the patient died 43 months later. The remaining four patients have done well without complication or recurrence at 21, 70, 129, and 132 months. Four patients had unsuccessful endoscopic attempts to remove the broncholith. Three of these patients subsequently had thoracotomy, and all three are currently free of symptoms 9, 73, and 78 months after their operation. The fourth patient refused further treatment. Bronchogenic carcinoma, however, did develop in the contralateral lung 10 years later. This patient is currently alive 4 months after pulmonary resection.
Discussion The pathogenesis of broncholithiasis is thought to be related to late tissue response to healing granulomatous pulmonary infections, most commonly histoplasmosis or tuberculosis. 1-3,5-7 Following spread to central lymph nodes after initial involvement of lung parenchyma, calcium salts may precipitate in the node during the healing process. Eventually, constant motion created by respiration and beating of the heart may cause the area of calcification to migrate into adjacent structures, most commonly the tracheobronchial tree but also the esophagus and pulmonary vessels. Compression or distortion of these structures is followed by erosion into the hollow viscus with bleeding, pulmonary obstruction, or, if the offending calcification breaks free, expectoration of calcareous material.' The exact incidence of broncholithiasis is unknown but is related to the incidence of specific granulomatous lung infection. The number of patients today, however, remains low, with approximately three or four being treated annually at our institution. Men predominated in our series, although this has not been the experience of others.': 2,5 The disease is most common in patients in their sixth decade. I, 2, 5 The diagnosis of broncholithiasis may be difficult. Expectoration of stones and visualization of stones through a bronchoscope are evidence of this disease process. Hilar and mediastinal calcification on chest roentgenograms (including tomography) is suggestive of broncholithiasis, However, neoplasm must always be suspected, and biopsy with bronchial brushing for cytologic examination should always be performed during bronchoscopy. Most commonly, however, the diag-
Thoracic and Cardiovascular Surgery
nosis of broncholithiasis is based upon symptoms (cough, hemoptysis, and expectoration of stones) and roentgenographic and bronchoscopic fmdings. Esophagobronchial fistula is an uncommon fmding in broncholithiasis. In our current series, no patient had this complication. Similarly, Dixon and associates' reported in 1984 on 19 patients with complications of broncholithiasis, none of whom had a fistula. In a previous report from our institution by Arrigoni, Bernatz, and Donoghue,' only two of 253 patients with broncholithiasis had an esophagobronchial fistula. In contrast, Faber and colleagues' reported on 43 patients in 1975, five of whom had fistulas. The frequent use of preoperative bronchography and esophageal contrast studies in their patients may have detected an increased incidence of this particular complication. Indications for treatment generally include intractable cough, recurrent hemoptysis, chronic suppuration resulting from bronchial stenosis, secondary bronchoesophageal fistula, or uncertainty about the diagnosis, and not the mere presence of calcified lymph nodes alone. The treatment of broncholithiasis depends upon the severity of the presenting complication and the medical condition of the patient. Goals of treatment should be removal of all offending calcification and irreversibly damaged bronchial or lung parenchyma along with conservation of as much normal lung tissue as possible.':? Since tracheobronchial stenosis is frequently associated with postobstructive pneumonitis, lung abscess,or bronchiectasis, pulmonary resection rather than isolated broncholithectomy is often indicated. Lymph node calcification in this disease is frequently associated with an intense inflammatory response, which makes surgical dissection difficult and increases significantly the likelihood of intraoperative complications, specifically, hemorrhage and esophageal damage. Preoperative preparation by having large intravenous lines in place and extra blood available is recommended. Meticulous dissection of the major vascular structures cannot be stressed enough. Early mobilization of the central pulmonary artery is important so that bleeding, if it occurs, can be readily controlled. Proximity of calcified lymph nodes to the esophagus may also result in esophageal changes, such as traction diverticulum (Fig. 1), most of which can be corrected by removal of the calcified lymph nodes. However, ('.are must be taken to avoid iatrogenic esophageal injury, such as transection of these diverticula. Placing an intraluminal dilator by mouth can help locate the esophagus. If there is any injury to the esohagus, direct repair with autogenous tissue coverage should be performed. If esophageal damage is suspected postopera-
Volume 90 Number 6 December, 1985
tively, early esophagograms can be of value in diagnosis of an esohageal leak. Endoscopic broncholithectomy is not new. I-3• 5• 8 Arrigoni, Bematz, and Donoghue' reported on 40 patients who underwent this procedure. Faber and associates' likewise reported on two patients with endoscopic removal, one of whom developed a tracheoesophageal fistula. Dixon and co-workers' reported on five patients who had attempted bronchoscopic removal, with success in four. In our current series, bronchoscopic broncholithectomy was done in patients in whom the broncholith was visible and appeared mobile and easily removable. Two of 12 patients had significant complications. Although this is a reasonable form of treatment provided that risks related to bleeding are accepted and that emergent surgical intervention is available if needed, this procedure should probably be reserved for patients who are in poor medical condition. Mortality and morbidity from surgical treatment remain low and acceptable. Although broncholithectomy alone, which is theoretically ideal, can be done at thoracotomy in selected patients, pulmonary resection is frequently required to remove irreversibly damaged parenchyma. I. 2 Coupled with this low surgical risk is the excellent prognosis such treatment provides. In our experience there was no decrease in expected survival after treatment. Similarly, the quality of life is improved. Of the 33 patients treated surgically who were still alive at the end of our study, all were free of pulmonary symptoms. No recurrences were documented for the thoracotomy group during follow-up. It appears, therefore, that surgical treatment is both safe and definitive and still should be considered the treatment of choice for complications of broncholithiasis. The question of whether broncholithiasis is associated with carcinoma of the lung remains unanswered. Arrigoni, Bematz, and Donoghue' reported on three patients who presented with both lung cancer and broncholithiasis.1 Lung cancer developed during follow-up in three of their other patients. None of our patients initially had cancer of the lung, although the disease developed during follow-up in three of them (6.1%). In two of these patients the malignant tumor developed in the same area that had contained the broncholith. It remains impossible, however, to determine if there is a causeand-effect relationship between broncholithiasis and cancer. Nonetheless, follow-up on a regular basis seems appropriate because of the high incidence of associated lung cancer. In summary, broncholithiasis is a rare but troublesome disease that can cause life-threatening complications. After evaluation with chest roentgenography and
Broncholithiasis 8 4 7
bronchoscopy, the patient should undergo thoracotomy for local resection of the offending broncholith and conservative pulmonary resection of irreversible damaged bronchi and parenchyma. The risk of surgical treatment is low and benefit over time is excellent. Endoscopic removal of mobile broncholiths may be performed in selected patients in poor medical condition. Regardless of the method of treatment, follow-up on a regular basis is recommended. REFERENCES
2
3 4
5
6 7 8
Arrigoni MG, Bernatz PE, Donoghue FE: Broncholithiasis. J THoRAc CARDIOVASC SURG 62:231-237, 1971 Faber LP, Jensik RJ, Chawla SK, Kittle CF: The surgical implication of broncholithiasis. J THORAC CARDIOVASC SURG 70:779-789, 1975 Schmidt HW, Clagett or, McDonald JR: Broncholithiasis. J THoRAc SURG 19:226-243, 1950 Kaplan EL, Meier P: Nonparametric estimation from incomplete observation. J Am Stat Assoc 53:457-481, 1958 Dixon GF, Donnerberg RL, Schonfeld SA, Whitcomb ME: Clinical commentary. Advances in diagnosis and treatment of broncholithiasis. Am Rev Respir Dis 129:1028-1030, 1984 Kelley WA: Broncholithiasis. Current concepts of an ancient disease. Postgrad Moo 66:81-90, 1979 Weed LA, Andersen HA: Etiology of broncholithiasis. Chest 37:270-277, 1960 Moersch HJ, Schmidt HW: Scientific papers of The American Bronchoesophagological Association. XXXIX Broncholithiasis. Ann Otol Rhinol Laryngol 68:548-563, 1959
Discussion DR. L. PENFIELD FABER Chicago. 1/1.
We presented our experience with thoracotomy in 33 patients who had broncholithiasis at the Association meeting in 1975. Since that time, we have accumulated an additional 16 patients and have now done a thoracotomy for this problem in a total of 49 patients. The symptoms and diagnostic studies of our patients are similar to those reported by the authors. Our indications for operation include bronchoesophageal fistula, persistent hemoptysis, recurrent pneumonitis, and suspected carcinoma. A broncholith that erodes through the bronchial wall creates granulation tissue with irregular friable mucosa that has the appearance of carcinoma when viewed through the bronchoscope. Normal findings on biopsy and brushing coupled with calcification seen on the chest x-ray film in the area of the bronchial disease should always raise the question of a broncholith. In contradistinction to the authors' experience, we have had seven patients with a fistula between the esophagus and the bronchus. The mechanism for the development of the fistula is that a traction diverticulum of the esophagus develops from
8 4 8 Trastek et al.
the fibrosis and inflammation of the broncholith, with subsequent connection to the adjacent bronchus or lung parenchyma. I believe that one of the most important slides the authors showed was the one describing complications. Operations for broncholithiasis can be very difficult and the surgeon must be prepared for problems when performing a resection for this disease. It is important to gain proximal control of the pulmonary artery before dissection of its branches, as the vascular planes are frequently obliterated by the longstanding fibrosis. The authors noted three instances of significant intraoperative bleeding, and one patient required a pneumonectomy because of it. Two patients had esophageal injury and two additional patients required reoperation because of postoperative bleeding. These complications are not unique and their occurrence emphasizes the technical difficulty of this procedure. We recommend the use of a pleural flap for bronchoplasty or for interposition of tissue in the repair of an esophagobronchial fistula. Conservation of pulmonary tissues
The Journalof Thoracic and Cardiovascular Surgery
is an important aspect of operations for broncholithiasisand a bronchoplasty can frequently be utilized. Surgical technique is important when dealing with this problem. I would like to ask the authors if they can give us any hints regarding technical aids that can help in operations for broncholithiasis. DR. TRASTEK (Closing) I would like to thank Dr. Faber for his discussion and interest in our paper. We agree with his indications and technical considerations wholeheartedly, particularly the proximal control of the pulmonary artery and the use of pleural flaps when indicated. As he pointed out, the risk of bleeding is definitely present and anything that can be done to prepare for this ahead of time is certainly warranted in this type of procedure. The only helpful hints I can offer are to use careful surgical technique and be prepared for problems, as I have just mentioned.
Notice of correction
In the March, 1985, issue of the JOURNAL, in the article by Cosgrove and associates entitled, "In Vivo Hemodynamic Comparison of Porcine and Pericardial Valves," reference 18 (page 366) is incorrect. The correct reference is as follows: Tandon AP, Smith DR, Mary DAS, Ionescu MI: Sequential hemodynamic studies in patients having aortic valve replacement with the Ionescu-Shiley pericardial xenograft. Ann Thorac Surg 24:149-155, 1977.
THoRAc CARDIOVASC SURG
90:842-848, 1985
Surgical management of broncholithiasis Fifty-two patients (31 men and 21 women) were treated for complications of broncholitbiasis between 1969 and 1984. Mean age was 50.8 years (range 26 to 74 years). IndicatioM for operation included symptolll'l in 49 patients and an abnormal chest x-ray film in three. Broncholithectomy was initially attempted by thoracotomy in 40 patients and by bronchoscopy in 12. In the thoracotomy group broncholithectomy was successful in aU patients, 32 of whom underwent pulmonary resection. Significant postoperative complicatioM occurred in five patients (12.8 %). There was one postoperative death (2.5 %). In the bronchoscopy group broncholithectomy was successful in eight patients (67% ~ significant complicatiOM occurred in two and there were no early deaths, Subsequent thoracotomy was done in three of the four patients in whom bronchoscopic removal was UMuccessfui. FoUow-upaveraged 76.5 montm (range 6 to 183 ~ntm). The 15 year survival rate (Kaplan-Meier) for aU patients was 75.1 % and did not differ from a control group of patients. No patient in the thoracotomy group bad recurrent complicatioM of broncholitbiasis. In contrast, complications recurred in three of the eight patients (37.5 %) successfuUy treated by bronchoscopy. We conclude that broncholithectomy via thoracotomy is the preferred treatment, as the risks are low and the long-term results are excellent,
Victor F. Trastek, M.D. (by invitation), Peter C. Pairolero, M.D., Eric L. Ceithaml, M.D. (by invitation), Jeffrey M. Piehler, M.D. (by invitation), W. Spencer Payne, M.D., and Philip E. Bernatz, M.D., Rochester, Minn.
Broncholithiasis is an uncommon problem with lifethreatening complications, jhe management of which can be formidable.' The purpose of this study was to review our current experience with removal of symptomatic broncholiths so as to determine the relative merits of surgical versus endoscopic management. We also wished to determine if there was any evidence associating broncholithiasis with lung cancer.
Patients and methods Between Jan. 1, 1969, and Jan. 1, 1984, 52 patients were treated for complications of broncholithiasis. Although broncholithiasis implies that at least a portion of a calcified lymph node has eroded into the lumen of the bronchus.v' we view this as an oversimplification and believe that the disease is a dynamic process with a spectrum of presentations. 1 Erosion of the broncholith
From the Section of Thoracic and Cardiovascular Surgery, Department of Surgery, Mayo Clinic and Mayo Foundation, Rochester, Minn. Read at the Sixty-fifth Annual Meeting of The American Association for Thoracic Surgery, New Orleans, Louisiana, April 29-May I, 1985. Address for reprints: Peter C. Pairolero, M.D., 200 First St., S.W., Rochester, Minn. 55905.
842
Table I. Presenting symptoms in 52 patients with broncholithiasis Symptom
No.
Percent
Chronic cough Purulent sputum Hemoptysis Fever/chills Chest pain Expectoration of stones Wheezing Dyspnea Fatigue Sinusitis Asymptomatic
47 37 30 18
90.4 71.2 57.7 34.6 21.2 19.2 7.7 5.8 3.9 3.9 5.8
II
10 4 3 2 2 3
into the lumen is only one of the possible complications. We also include under the heading of broncholithiasis the distortion or partial obstruction of the tracheobronchial tree caused by a calcified peribronchial lymph node. Finally, these patients may have a pulmonary mass suggestive of malignancy. In this study, patients were considered to have broncholithiasis if calcific hilar or carinal lymph nodes were present roentgenographically, if on bronchoscopic examination there was evidence of peribronchial disease (bronchial compression, stenosis, or distortion), and if there was no other
Volume 90
Broncholithiasis 8 4 3
Number 6 December, 1985
Fig. 1. A 39-year-old man with an 18 month history of chronic cough and recurrent fever and chills. A, Posteroanterior chest roentgenogram demonstrating right hilar mass. B, Esophagogram demonstrating small traction diverticulum. C. Computed tomogram demonstrating peribronchial calcified mass.
pulmonary disease to explain these changes.' Asymptomatic patients with an indeterminate roentgenographic pulmonary mass which at thoracotomy was found to be caused by calcified lymph nodes were also considered to have broncholithiasis. During this time interval, there were 52 patients who met these criteria, and these 52 patients form the basis for this report. Survival probabilities were calculated by the KaplanMeier" method. Operative deaths and deaths owing to any cause were included in the survival statistics. Operative mortality included all deaths occurring within the first 30 postoperative days. Expected survival curves were used as controls and were based on death from all
causes from the West North Central United States 1970 Life Table data and were matched for age and sex.
Clinical features There were 31 men and 21 women. Ages ranged from 26 to 74 years (mean 50.8 years). Forty-nine patients were symptomatic (Table I). Forty-seven patients had a cough. Thirty-seven of these patients had a cough productive of purulent sputum and four had severe coughing spells. Eighteen patients had clinical evidence of acute obstructive pneumonitis. Hemoptysis of more than 4 ounces occurred in five patients. Seven of the 10 patients who coughed up stones did so on multiple
8 44
The Journal of Thoracic and Cardiovascular
Trastek et al.
Surgery
Table D. Chest roentgenographic findings in 52 patients with broncholithiasis
Peribronchial calcification Infiltrate Parenchymal mass Atelectasis Tracheobronchial distortion Hilar mass Other
No.
Percent
34 29 22 12 2 2 4
65.4 55.8 42.3 23.1 3.9 3.9 7.7
m.
Table Tomographic findings in 29 patients with broncholithiasis Hilar calcification Tracheobronchial distortion Parenchymal mass Infiltrate Other
No.
Percent
22 II 6 6 4
75.9 37.9 20.7 20.7 13.8
Table IV. Bronchoscopic findings in 52 patients with broncholithiasis Tracheobronchial distortion Inflammation Visible broncholith Bleeding Purulent secretions Normal
No.
Percent
45
86.5 69.2 34.6 21.6 11.5 3.9
36 18 II 6
2
occasions. The three asymptomatic patients underwent operation because their roentgenographic fmdings could not be differentiated from malignant lesions. Diagnostic studies. Posteroanterior chest. roentgenograms (Fig. 1) in all patients were abnormal (Table II). Hilar or peribronchial calcification was present in 34 patients, postobstructive pneumonitis in 29, and a parenchymal mass suggestive of malignancy in 22. Tomography was performed in 29 patients (Table III); 22 had hilar calcification, and 11 had tracheobronchial distortion with adjacent calcification. Bronchography was performed in seven patients and showed abnormalities in four. Three of these patients had bronchial stenosis and one had bronchiectasis. Bronchoscopy was done in all 52 patients (Table IV), and the findings were abnormal in 50. Bronchial stenosis and distortion as well as inflammation were the most common observations. A broncholith was visible in 18 patients (Fig. 2). Active bleeding was present in 11 patients but was severe in only one. None of the 52
patients had bronchial brushings or washings that were cytologically positive for malignancy. The two patients who had normal bronchoscopic examination had an indeterminate pulmonary mass. No patient had a fistula into either the esophagus or major vessels. Treatment. All 52 patients had attempted broncholithectomy with or without pulmonary resection. This was effected at thoracotomy in 40 patients and initially attempted at bronchoscopy in 12. Indications for treatment included symptomatic obstructive pneumonitis in 31 patients, recurrent significant hemoptysis in 20, a mass suggestive of malignancy in 20, chronic severe cough in two, and bronchiectasis in one. One patient had urgent thoracotomy for severe hemoptysis. At thoracotomy, pulmonary resection along with removal of the broncholith was performed in 32 patients. Twenty-one patients had lobectomy, six had either segmentectomy or wedge resection, three had pneumonectomy, and two had bilobectomy. The other eight patients had broncholithectomy without pulmonary resection. One of these latter patients had an associated bronchoplasty and one had a sleeve resection of the right bronchus intermedius. The remaining six patients had bronchial distortion without erosion, and broncholithectomy was possible without the need for either bronchotomy or bronchoplasty. Intraoperative complications developed in five patients, with significant bleeding in three and esophageal injury in two. Pneumonectomy was necessary in one of these patients to control bleeding. This patient subsequently died 3 days later of cardiorespiratory failure. Bleeding in the other two patients was controlled without further complication. The esophageal injury in each patient was repaired primarily and both patients subsequently did well. Median hospitalization for patients surviving thoracotomy was 7 days and ranged from 5 to 37 days. There was one postoperative death (2.5%) as described earlier. Significant postoperative complications occurred in five other patients (12.8%) and included bleeding in two, and vocal cord paralysis, confirmed esophageal leak, and wound infection in one patient each. None of these postoperative complications occurred in patients who had intraoperative complications. The two patients with postoperative hemorrhage required reexploration to control bleeding. The remaining three patients responded to conservative management. Twelve patients had bronchoscopic broncholithectomy. These patients were selected for endoscopic treatment because the broncholith was visible and mobile. In none of these patients was operation contraindicated for medical reasons. Ten of these patients had obstructive pneumonitis and three had recurrent hemoptysis. Com-
Volume 90
Broncholithiasis 8 4 5
Number 6 December. 1985
Fig. 2. Operative specimen showing a broncholith (arrow) which has eroded into the lumen of the bronchus.
plete removal of the broncholith was possible in eight patients. There were two complications consisting of bronchial laceration and bleeding. Both responded to conservative therapy. There were no deaths. Three of the four patients in whom broncholithectomy was incomplete had subsequent operation. These three patients are not included in the 40 patients undergoing operation reported here (see Follow-up). Pathology. The location of the broncholith varied and often involved multiple portions of the tracheobronchial tree. The broncholith was predominantly on the right in 28 patients, on the left in 19, subcarinal in three, and bilateral in two. Chronic inflammation was present in 31 of the operative specimens, bronchiectasis in 13, lung abscess in seven, pneumonitis in six, nonspecific granulomas in three, and histoplasmosis in two. No patient had evidence of active tuberculosis or carcinoma in the resected specimen. Follow-up. Two patients were lost to follow-up evaluation. Follow-up was complete in the remaining 49 surviving patients (96%) and averaged 76.5 months (range 6 to 183 months). The actuarial 15 year survival rate for all 50 patients (including postoperative death) was 75.1%. This did not differ statistically from that of a similar group of patients matched for sex and age (Fig. 3). Thirty-seven of the 39 operative survivors were available for follow-up analysis. Follow-up ranged from 6 to 183 months with a mean of 98.8 months. There were four late deaths, none related to a complication of treatment. Cause of death was atherosclerosis in two patients and complications of a duodenal ulcer operation
100
l
80
iii >
's ~
Ul
(2)
60
'0
~
:0
40
.c
a: 0
20
0
0
5
10
15
Years
Fig. 3. Overall survival rate (death from all causes) of 50 patients with broncholithiasis as compared to expected survival matched for age, sex, and calendar year. Zero time on abscissa represents day of operation.
in one. In the fourth patient, lung cancer developed 4 years later in the same area that had contained the broncholith. After pulmonary resection, this patient died 2 years later of disseminated cancer. The remaining 33 patients are alive without pulmonary symptoms. No patient had recurrent complication from broncholithiasis after operation. In three of the eight patients who had initially successful bronchoscopic broncholithectomy, the disease recurred 6, 15, and 57 months later. One of these patients subsequently had a right middle lobectomy and has done well for 5 years. Another patient had repeat bronchoscopic broncholithectomy and has done well for
The Journal Of
8 4 6 Trastek et al.
10 years. The last patient had early recurrence of hemoptysis. Further treatment was refused, and this patient died of massive hemoptysis 39 months later. A second late death also occurred during follow-up. In this patient a squamous cell carcinoma developed in the same lobe from which the broncholith had been removed 38 months previously. Pulmonary resection was refused and the patient died 43 months later. The remaining four patients have done well without complication or recurrence at 21, 70, 129, and 132 months. Four patients had unsuccessful endoscopic attempts to remove the broncholith. Three of these patients subsequently had thoracotomy, and all three are currently free of symptoms 9, 73, and 78 months after their operation. The fourth patient refused further treatment. Bronchogenic carcinoma, however, did develop in the contralateral lung 10 years later. This patient is currently alive 4 months after pulmonary resection.
Discussion The pathogenesis of broncholithiasis is thought to be related to late tissue response to healing granulomatous pulmonary infections, most commonly histoplasmosis or tuberculosis. 1-3,5-7 Following spread to central lymph nodes after initial involvement of lung parenchyma, calcium salts may precipitate in the node during the healing process. Eventually, constant motion created by respiration and beating of the heart may cause the area of calcification to migrate into adjacent structures, most commonly the tracheobronchial tree but also the esophagus and pulmonary vessels. Compression or distortion of these structures is followed by erosion into the hollow viscus with bleeding, pulmonary obstruction, or, if the offending calcification breaks free, expectoration of calcareous material.' The exact incidence of broncholithiasis is unknown but is related to the incidence of specific granulomatous lung infection. The number of patients today, however, remains low, with approximately three or four being treated annually at our institution. Men predominated in our series, although this has not been the experience of others.': 2,5 The disease is most common in patients in their sixth decade. I, 2, 5 The diagnosis of broncholithiasis may be difficult. Expectoration of stones and visualization of stones through a bronchoscope are evidence of this disease process. Hilar and mediastinal calcification on chest roentgenograms (including tomography) is suggestive of broncholithiasis, However, neoplasm must always be suspected, and biopsy with bronchial brushing for cytologic examination should always be performed during bronchoscopy. Most commonly, however, the diag-
Thoracic and Cardiovascular Surgery
nosis of broncholithiasis is based upon symptoms (cough, hemoptysis, and expectoration of stones) and roentgenographic and bronchoscopic fmdings. Esophagobronchial fistula is an uncommon fmding in broncholithiasis. In our current series, no patient had this complication. Similarly, Dixon and associates' reported in 1984 on 19 patients with complications of broncholithiasis, none of whom had a fistula. In a previous report from our institution by Arrigoni, Bernatz, and Donoghue,' only two of 253 patients with broncholithiasis had an esophagobronchial fistula. In contrast, Faber and colleagues' reported on 43 patients in 1975, five of whom had fistulas. The frequent use of preoperative bronchography and esophageal contrast studies in their patients may have detected an increased incidence of this particular complication. Indications for treatment generally include intractable cough, recurrent hemoptysis, chronic suppuration resulting from bronchial stenosis, secondary bronchoesophageal fistula, or uncertainty about the diagnosis, and not the mere presence of calcified lymph nodes alone. The treatment of broncholithiasis depends upon the severity of the presenting complication and the medical condition of the patient. Goals of treatment should be removal of all offending calcification and irreversibly damaged bronchial or lung parenchyma along with conservation of as much normal lung tissue as possible.':? Since tracheobronchial stenosis is frequently associated with postobstructive pneumonitis, lung abscess,or bronchiectasis, pulmonary resection rather than isolated broncholithectomy is often indicated. Lymph node calcification in this disease is frequently associated with an intense inflammatory response, which makes surgical dissection difficult and increases significantly the likelihood of intraoperative complications, specifically, hemorrhage and esophageal damage. Preoperative preparation by having large intravenous lines in place and extra blood available is recommended. Meticulous dissection of the major vascular structures cannot be stressed enough. Early mobilization of the central pulmonary artery is important so that bleeding, if it occurs, can be readily controlled. Proximity of calcified lymph nodes to the esophagus may also result in esophageal changes, such as traction diverticulum (Fig. 1), most of which can be corrected by removal of the calcified lymph nodes. However, ('.are must be taken to avoid iatrogenic esophageal injury, such as transection of these diverticula. Placing an intraluminal dilator by mouth can help locate the esophagus. If there is any injury to the esohagus, direct repair with autogenous tissue coverage should be performed. If esophageal damage is suspected postopera-
Volume 90 Number 6 December, 1985
tively, early esophagograms can be of value in diagnosis of an esohageal leak. Endoscopic broncholithectomy is not new. I-3• 5• 8 Arrigoni, Bematz, and Donoghue' reported on 40 patients who underwent this procedure. Faber and associates' likewise reported on two patients with endoscopic removal, one of whom developed a tracheoesophageal fistula. Dixon and co-workers' reported on five patients who had attempted bronchoscopic removal, with success in four. In our current series, bronchoscopic broncholithectomy was done in patients in whom the broncholith was visible and appeared mobile and easily removable. Two of 12 patients had significant complications. Although this is a reasonable form of treatment provided that risks related to bleeding are accepted and that emergent surgical intervention is available if needed, this procedure should probably be reserved for patients who are in poor medical condition. Mortality and morbidity from surgical treatment remain low and acceptable. Although broncholithectomy alone, which is theoretically ideal, can be done at thoracotomy in selected patients, pulmonary resection is frequently required to remove irreversibly damaged parenchyma. I. 2 Coupled with this low surgical risk is the excellent prognosis such treatment provides. In our experience there was no decrease in expected survival after treatment. Similarly, the quality of life is improved. Of the 33 patients treated surgically who were still alive at the end of our study, all were free of pulmonary symptoms. No recurrences were documented for the thoracotomy group during follow-up. It appears, therefore, that surgical treatment is both safe and definitive and still should be considered the treatment of choice for complications of broncholithiasis. The question of whether broncholithiasis is associated with carcinoma of the lung remains unanswered. Arrigoni, Bematz, and Donoghue' reported on three patients who presented with both lung cancer and broncholithiasis.1 Lung cancer developed during follow-up in three of their other patients. None of our patients initially had cancer of the lung, although the disease developed during follow-up in three of them (6.1%). In two of these patients the malignant tumor developed in the same area that had contained the broncholith. It remains impossible, however, to determine if there is a causeand-effect relationship between broncholithiasis and cancer. Nonetheless, follow-up on a regular basis seems appropriate because of the high incidence of associated lung cancer. In summary, broncholithiasis is a rare but troublesome disease that can cause life-threatening complications. After evaluation with chest roentgenography and
Broncholithiasis 8 4 7
bronchoscopy, the patient should undergo thoracotomy for local resection of the offending broncholith and conservative pulmonary resection of irreversible damaged bronchi and parenchyma. The risk of surgical treatment is low and benefit over time is excellent. Endoscopic removal of mobile broncholiths may be performed in selected patients in poor medical condition. Regardless of the method of treatment, follow-up on a regular basis is recommended. REFERENCES
2
3 4
5
6 7 8
Arrigoni MG, Bernatz PE, Donoghue FE: Broncholithiasis. J THoRAc CARDIOVASC SURG 62:231-237, 1971 Faber LP, Jensik RJ, Chawla SK, Kittle CF: The surgical implication of broncholithiasis. J THORAC CARDIOVASC SURG 70:779-789, 1975 Schmidt HW, Clagett or, McDonald JR: Broncholithiasis. J THoRAc SURG 19:226-243, 1950 Kaplan EL, Meier P: Nonparametric estimation from incomplete observation. J Am Stat Assoc 53:457-481, 1958 Dixon GF, Donnerberg RL, Schonfeld SA, Whitcomb ME: Clinical commentary. Advances in diagnosis and treatment of broncholithiasis. Am Rev Respir Dis 129:1028-1030, 1984 Kelley WA: Broncholithiasis. Current concepts of an ancient disease. Postgrad Moo 66:81-90, 1979 Weed LA, Andersen HA: Etiology of broncholithiasis. Chest 37:270-277, 1960 Moersch HJ, Schmidt HW: Scientific papers of The American Bronchoesophagological Association. XXXIX Broncholithiasis. Ann Otol Rhinol Laryngol 68:548-563, 1959
Discussion DR. L. PENFIELD FABER Chicago. 1/1.
We presented our experience with thoracotomy in 33 patients who had broncholithiasis at the Association meeting in 1975. Since that time, we have accumulated an additional 16 patients and have now done a thoracotomy for this problem in a total of 49 patients. The symptoms and diagnostic studies of our patients are similar to those reported by the authors. Our indications for operation include bronchoesophageal fistula, persistent hemoptysis, recurrent pneumonitis, and suspected carcinoma. A broncholith that erodes through the bronchial wall creates granulation tissue with irregular friable mucosa that has the appearance of carcinoma when viewed through the bronchoscope. Normal findings on biopsy and brushing coupled with calcification seen on the chest x-ray film in the area of the bronchial disease should always raise the question of a broncholith. In contradistinction to the authors' experience, we have had seven patients with a fistula between the esophagus and the bronchus. The mechanism for the development of the fistula is that a traction diverticulum of the esophagus develops from
8 4 8 Trastek et al.
the fibrosis and inflammation of the broncholith, with subsequent connection to the adjacent bronchus or lung parenchyma. I believe that one of the most important slides the authors showed was the one describing complications. Operations for broncholithiasis can be very difficult and the surgeon must be prepared for problems when performing a resection for this disease. It is important to gain proximal control of the pulmonary artery before dissection of its branches, as the vascular planes are frequently obliterated by the longstanding fibrosis. The authors noted three instances of significant intraoperative bleeding, and one patient required a pneumonectomy because of it. Two patients had esophageal injury and two additional patients required reoperation because of postoperative bleeding. These complications are not unique and their occurrence emphasizes the technical difficulty of this procedure. We recommend the use of a pleural flap for bronchoplasty or for interposition of tissue in the repair of an esophagobronchial fistula. Conservation of pulmonary tissues
The Journalof Thoracic and Cardiovascular Surgery
is an important aspect of operations for broncholithiasisand a bronchoplasty can frequently be utilized. Surgical technique is important when dealing with this problem. I would like to ask the authors if they can give us any hints regarding technical aids that can help in operations for broncholithiasis. DR. TRASTEK (Closing) I would like to thank Dr. Faber for his discussion and interest in our paper. We agree with his indications and technical considerations wholeheartedly, particularly the proximal control of the pulmonary artery and the use of pleural flaps when indicated. As he pointed out, the risk of bleeding is definitely present and anything that can be done to prepare for this ahead of time is certainly warranted in this type of procedure. The only helpful hints I can offer are to use careful surgical technique and be prepared for problems, as I have just mentioned.
Notice of correction
In the March, 1985, issue of the JOURNAL, in the article by Cosgrove and associates entitled, "In Vivo Hemodynamic Comparison of Porcine and Pericardial Valves," reference 18 (page 366) is incorrect. The correct reference is as follows: Tandon AP, Smith DR, Mary DAS, Ionescu MI: Sequential hemodynamic studies in patients having aortic valve replacement with the Ionescu-Shiley pericardial xenograft. Ann Thorac Surg 24:149-155, 1977.