INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Broncholiths are calcified peribronchial lymph nodes that encroach upon adjacent airways and cause clinical and roentgenographic abnormalities (1). Most broncholiths are sequelae of fungal or mycobacterial granulomatous lymphadenitis (7, 8), with silicosis also having been described as a rare etiology (9, 10). Even though calcified hilar and mediastinal lymph nodes are common findings on chest roentgenographs, symptomatic broncholithiasis is uncommon. Gradual tracheobronchial impingement occurs from an interaction of the fibrocalcific changes with the repetitive visceral motions of respiration, circulation, and deglutition (7, 11). Symptoms develop when the calcified lymph node impinges on or erodes into the airway lumen. Bronchial distortion, irritation, and erosion by broncholiths can cause chronic cough, hemoptysis, stone expectoration (lithoptysis), recurrent pneumonia, and fistulas between the bronchi and adjacent mediastinal structures (3, 12). Broncholithiasis as a diagnostic possibility is rarely considered in patients with calcified granulomas on chest roentgenographs or chronic cough. The history of lithoptysis is rarely forthcoming unless the patient is specifically queried about it.

Bronchoscopy is an important component in the diagnostic evaluation of broncholithiasis. Indeed, bronchoscopy is often the only test to document the diagnosis of broncholithiasis. However, the application of bronchoscopy for the treatment of broncholithiasis has been somewhat controversial. Success rates as high as 87% for bronchoscopic removal of visible broncholiths, without life-threatening complications, have been reported (12). Results such as these have prompted some authors to conclude that bronchoscopic removal of broncholiths should be attempted first whenever possible (12, 16). It should be noted, however, that most of these recommendations have been based on small numbers of patients who underwent bronchoscopic extraction of broncholiths. Other clinicians have expressed their concerns about the potential for significant hemorrhage, bronchial tearing, or fistula formation when attempting to extract broncholiths bronchoscopically. To avoid such complications, it has been recommended, on the basis of limited experience, that bronchoscopic broncholithectomy be completely avoided (21) or limited to patients whose comorbidities preclude surgical intervention (15).

The purpose of this study was to examine the safety and efficacy of attempts at bronchoscopic treatment of broncholithiasis over a 40-yr period at Mayo Clinic, in an effort to clarify the therapeutic role of bronchoscopy in broncholithiasis.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Through the Mayo Clinic's computerized record system, we retrospectively identified all patients with a diagnosis of broncholithiasis who also underwent bronchoscopy during the period from 1954 to 1994. Bronchoscopies were performed to evaluate pulmonary symptoms and/or roentgenographic abnormalities. Some authors have broadened the definition of broncholithiasis to include roentgenographic or bronchoscopic evidence of bronchial distortion by calcified peritracheobronchial lymph nodes, without frank intraluminal erosion (13, 15). Since the focus of this study was bronchoscopic broncholithectomy, we limited the definition of broncholith to bronchoscopically visible, free or partly eroded calcified material in the lumen of an airway. Broncholiths that were exposed after bronchoscopic removal of overlying granulation tissue were also included. From review of the patients' medical records, we compiled information about demographics, presenting symptoms and signs, imaging studies, type of bronchoscope used, bronchoscopy findings, outcome and complications of bronchoscopic removal attempts, and nonbronchoscopic therapies for broncholiths. The Mayo Institutional Review Board approved the study.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We identified 95 patients (46 women and 49 men) with a mean age of 56.8 yr (range: 17 to 91 yr) with 127 bronchoscopically visible broncholiths. Forty-eight patients (50.5%) underwent an attempt at bronchoscopic broncholith removal during 61 bronchoscopy sessions. One patient required four sessions over a 14-mo period to remove recurrent calcified material in the airways. Table 1 provides demographic and clinical information for those patients who did and for those who did not undergo attempts at bronchoscopic broncholithectomy. There was a higher frequency of hemoptysis (66% versus 38%) in the group that did not undergo an attempt at bronchoscopic broncholith removal. The groups appeared otherwise equivalent. Cough and hemoptysis were the most common clinical features in both groups; lithoptysis was uncommon.

Table 2 shows the distribution of the broncholiths encountered. As in previous series (4, 11), broncholiths were more often found in the right bronchial tree. No particular tracheobronchial location appeared to preclude an attempt at bronchoscopic broncholith removal. One patient had 10 free broncholiths extracted from his right lower-lobe bronchi, which explains the preponderance of removal attempts from this region.

Of the 127 broncholiths identified bronchoscopically, 71 (56%) underwent an attempt at bronchoscopic removal. Table 3 reveals the number of removal attempts and bronchoscopic route by which partly eroding and loose (free in the airway) broncholiths were subjected to removal attempts. Forty-six percent (48 of 104) of partly eroding broncholiths and 100% (23 of 23) of loose broncholiths underwent bronchoscopic extraction attempts.

The outcomes of bronchoscopic attempts to remove partly eroding broncholiths are summarized in Table 3. Equivalent percentages of partly eroding broncholiths were approached for removal with rigid and with flexible bronchoscopy. Overall, 48% (23 of 48) of partly eroding broncholiths were completely removed bronchoscopically (defined as no calcified material remaining in the bronchial lumen). Use of the rigid bronchoscope led to a higher percentage of complete extractions (67%) than did use of the flexible bronchoscope (30%). It was usually not possible to retrospectively determine the bronchoscopic technique(s) used (i.e., chipping, crushing, probing, pulling, etc.) to remove a particular broncholith. In one patient the pulsed dye laser was used to fragment an eroding broncholith that could not be dislodged with either a rigid or a flexible bronchoscope (22).

Reasons for incomplete bronchoscopic attempts to remove partially eroding broncholiths included overly firm embedding of the broncholith in eight patients; inability to adequately grasp the broncholith in five patients; concerns over bleeding in five patients; and no retrospectively discernable reason in nine patients. Two explanations applied in the cases of two patients. Bleeding in one case necessitated urgent surgical intervention (see the subsequent discussion); the other four instances did not require specific interventions. Six patients with incomplete bronchoscopic removal of partly eroding broncholiths proceeded to surgery, of whom two had broncholithectomy, two had right middle lobectomy, one had right upper lobectomy, and one had left lower lobectomy.

Table 3 shows the outcome of bronchoscopic attempts to remove loose broncholiths. All loose broncholiths (23 of 23) were completely extracted regardless of the type of bronchoscope used.

Reasons for avoidance of bronchoscopic broncholithectomy are listed in Table 4 (more than one reason could apply). Unfortunately, reason(s) for no removal attempt could not be determined in a retrospective manner for 23 patients (25 broncholiths). None of the episodes of bleeding with inspection/ biopsy required specific interventions. Eighteen of the 23 patients in whom no reason could be found for lack of attempted broncholith removal proceeded to surgery, of whom six had right middle lobectomy, three had bilobectomies, two had segmentectomies, two had broncholithectomy, and one each had pneumonectomy, right upper lobectomy, left upper lobectomy, left lower lobectomy, and exploration only.

In three patients the pediatric flexible bronchoscope was used to identify broncholiths not detected with an adult bronchoscope. An attempt was made to use the pediatric flexible bronchoscope to remove a stone in a sixth-order right upper-lobe bronchus in one patient, but the broncholith was too embedded to be movable with the pediatric scope. In the other two patients, removal attempts were stopped after bleeding was encountered with broncholith inspection.

Clinically significant complications of bronchoscopic removal attempts occurred in two patients. A 54-yr-old man developed acute dyspnea several hours after rigid bronchoscopic removal of a partly eroded broncholith in the right main bronchus. Emergent rigid bronchoscopy revealed a large broncholith loose in the distal trachea. A removal attempt with forceps through the rigid bronchoscope was complicated by fracture of the broncholith in the larynx. Laryngoscopic assistance was required for complete broncholith removal. A 62-yr-old woman developed bleeding during a rigid bronchoscopic attempt to loosen a large right middle-lobe broncholith that had resisted two earlier attempts at flexible bronchoscopic removal. A chest computed tomographic scan had not revealed a major vessel in the vicinity of the broncholith. Thoracotomy for bronchotomy and broncholithectomy was performed after an estimated 300-ml blood loss; the bleeding was slow and without hemodynamic consequences.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Broncholithiasis is a rare disorder with varying clinical presentations and a poorly defined natural history (6, 23). Options for treating it include observation, bronchoscopic broncholithectomy, and surgery. Spontaneous broncholith expectoration may occasionally lead to resolution of symptoms. Yet our experience, with the largest series to date, indicates that lithoptysis is infrequent. Other investigators have also observed the rare occurrence of lithoptysis (4, 5, 11, 14). Massive hemoptysis caused by broncholithiasis is a rare complication, with only three cases of death from broncholith-associated massive hemoptysis reported in the literature (15, 18). Most patients who present with symptomatic broncholithiasis will require an attempt at surgical or bronchoscopic extraction of a broncholith.

Several retrospective studies have described surgical management of broncholithiasis (5, 11, 15, 16). Using an expanded definition that included cases of airway distortion by calcified peribronchial lymph nodes without frank intraluminal erosion by a node, Arrigoni and colleagues summarized the Mayo Clinic surgical experience with broncholithiasis through 1968 (13). Sixty-eight of the 253 patients underwent surgery. Two patients (3%) died in the immediate postoperative period and another 12 (18%) developed postoperative complications. Using the same expanded broncholithiasis definition, Trastek and coworkers (15) updated the Mayo Clinic surgical experience with broncholithiasis from 1969 through 1983. Fifty-two patients with symptomatic broncholithiasis were identified. Broncholithectomy by thoracotomy was attempted in 40 patients. Intraoperative complications occurred in five patients (13%), with one (2.5%) of these patients dying 3 d later. Postoperative complications occurred in another five (13%) patients. Other surgical series of broncholithiasis include that of Groves and Effler (5), who reviewed 27 cases of broncholithiasis, of which 20 required surgical intervention. There were two (7%) postoperative deaths and two empyemas. Faber and associates (11) described their experience with thoracotomy for symptomatic broncholithiasis in 33 patients. Only two patients (6%) developed postoperative morbidities, and there were no deaths. Cole and colleagues (16) reported the results of thoracotomy for 25 patients with broncholithiasis, none of whom died perioperatively.

A review of these reports reveals several common themes. First, usual indications for surgery for broncholithiasis include chronic pulmonary suppurative disease (bronchiectasis), massive hemoptysis, bronchoesophageal fistulas, and uncertainty about the diagnosis (11). Second, the mediastinal and hilar fibrocalcific reaction accompanying broncholithiasis can alter tissue planes, obscure anatomic landmarks, and increase blood vessel fragility in the operative field, making surgical dissection more difficult and increasing the risk of complications (11, 13, 15). Pneumonectomy was required in one patient to control intraoperative bleeding in the series examined by Trastek and coworkers (15), and two patient sustained intraoperative esophageal injury. However, perioperative morbidity appeared to be within acceptable limits. Third, most surgeries for broncholithiasis involve pulmonary resection, which may be due at least in part to the surrounding intense inflammatory changes. Eighty to 95% of patients have required segmentectomy or more extensive resection as part of the surgical therapy for broncholithiasis (5, 11, 13, 15, 16). Fourth, the long-term results of surgery are usually excellent. Follow-up was available for 59 surgical patients in the series reported by Arrigoni and colleagues (13), with 50 patients remaining completely asymptomatic. In the series reported by Trastek and coworkers (15), no surgical patient had recurrent problems with broncholithiasis, and 15-yr survival for the surgical cohort was equivalent to that of a matched control group. Among the 13 surgical patients followed by Cole and associates (16), all were reported as being well from 3 to 30 yr postoperatively.

All of these surgical series also reported concurrent efforts at bronchoscopic broncholithectomy, with the authors drawing different conclusions. Among the 63 patients studied by Arrigoni and colleagues (13), broncholiths were removed bronchoscopically from 40 patients (63%) whose bronchoscopies revealed visible broncholiths. Complications were not reported, and it is not possible to discern how the bronchoscopic broncholithectomy patients fared as compared with the surgical patients. Nonetheless, the authors concluded that bronchoscopic extraction of a visualized broncholith was "reasonable" as long as irreversible distal bronchial and parenchymal damage had not occurred (13). On the basis of their successful bronchoscopic removal of intraluminal broncholiths from eight patients without severe bleeding, Cole and associates (16) likewise concluded that bronchoscopic broncholithectomy was a "useful adjunct" and should be judiciously attempted before complications of broncholithiasis occur. Among the patients studied by Trastek and coworkers (15), complete bronchoscopic broncholith removal was achieved in eight of the 12 patients (66.7%) who underwent bronchoscopic extraction. There were two minor complications that responded to conservative maneuvers. Broncholithiasis recurred in three of these eight patients, for which one underwent repeat bronchoscopic broncholithectomy, one required right middle lobectomy, and one refused further intervention and died of massive hemoptysis. Three of the four patients with unsuccessful bronchoscopic removal attempts went to surgery. Trastek and coworkers (15) concluded that bronchoscopic broncholithectomy "should probably be reserved for patients who are in poor medical condition." Only two of the 33 patients studied by Faber and colleagues (11) underwent bronchoscopic stone removal, and one attempt was complicated by development of a bronchoesophageal fistula. The authors concluded that bronchoscopic broncholith removal was indicated only if the broncholith was "loose and mobile" and extraction did "not require extensive manipulation" (11).

Concerns about major bleeding prompted Brantigan to conclude that attempts at bronchoscopic broncholithectomy should be completely avoided (21). However, cumulative experience indicates that massive hemorrhage complicating bronchoscopic broncholith removal is rare. None of the large series studies of broncholithiasis have described this complication. In our series, one patient (2% of the total undergoing bronchoscopic broncholithectomy) experienced an estimated 300 ml blood loss after an extraction attempt. Although the patient was never hemodynamically unstable, the decision was made to proceed directly to surgery, since two previous bronchoscopic attempts to dislodge the broncholith had also been unsuccessful.

The limitations of retrospective reviews such as the present one in determining the risk of complication in bronchoscopic broncholithectomy must be recognized. Although ours is the first review to list reasons for the lack of attempts at bronchoscopic extractions and why attempts were stopped, it was admittedly difficult in many cases to retrospectively decipher how the bronchoscopist decided whether to attempt bronchoscopic removal, how vigorous the extraction efforts were, and why attempts were aborted. The most frequently discernible reason for avoiding bronchoscopic removal was bleeding with initial stone manipulation. Interestingly, the frequency of hemoptysis was greater in the group of patients that did not undergo attempts at bronchoscopic broncholith extraction. How this symptom influenced decisions during bronchoscopy is unclear. Igoe and associates (24) urged caution in bronchoscopically approaching broncholiths in patients presenting with hemoptysis. They reported a case of a patient with hemoptysis whose bronchoscopically visible broncholith was later found at thoracotomy to have also eroded into a branch of the pulmonary artery. It should also be recognized that not all instances of hemoptysis are caused by airway-vascular communication (fistula) caused by a broncholith. The granulation tissue that develops around the broncholith often tends to bleed upon bronchoscopic instrumentation. However, it seems prudent whenever possible to define the relation of a broncholith to adjacent vascular and mediastinal structures before attempting bronchoscopic extraction. In our review it was not possible to retrospectively measure the influence of plain and computed tomographic results on the bronchoscopists.

Comprehensive bronchoscopic techniques enhance the outcome of attempted broncholith removal. In our series, the rigid bronchoscope resulted in a greater frequency of complete broncholith removals than did exclusive use of the flexible bronchoscope. This statement is made with the acknowledgment that many attempts at rigid bronchoscopy for broncholith removal were made before the availability of flexible bronchoscopy. The utility of rigid bronchoscopy is also pointed out by Dixon and coworkers (14), who reported successful rigid bronchoscopic broncholithectomy in three of four patients, but success in only one patient with the flexible bronchoscope. Extraction efforts in seven of our patients incorporated both the rigid and flexible bronchoscopes. Miks and colleagues (25) used an Nd:YAG laser to fragment a mobile broncholith that was too hard to be broken with the biopsy forceps and too large to be pulled through the upper airway. The Nd:YAG laser has also been used to remove partly obstructing granulation tissue associated with calcified peribronchial lymph nodes (26). We have reported the use of combined electrohydraulic and pulsed-dye laser broncholithotripsy to facilitate removal of a large broncholith in a patient whose comorbidities prevented surgical intervention (22). Although limited in its extraction capabilities, the pediatric flexible bronchoscope may enhance the ability to diagnose broncholithiasis (27). In our series, three broncholiths that would otherwise have escaped detection were identified with the pediatric bronchoscope.

We conclude that bronchoscopy is a safe and effective therapeutic option for the management of loose broncholiths. In our series, 23 of 23 (100%) free broncholiths were completely removed bronchoscopically without complication. For partly eroding broncholiths, bronchoscopy should be considered a therapeutic option with an acceptably low complication rate in carefully selected patients. In our series, 23 of 48 (48%) partly eroding broncholiths were completely removed bronchoscopically, with only two patients (4% of the bronchoscopic broncholithectomy group) experiencing clinically significant complications. The fibrocalcific reaction accompanying broncholithiasis may theoretically increase the risk of hemorrhage, bronchial tearing, and fistula formation with attempts at bronchoscopic broncholithectomy, but these complications appear to have been uncommon in this and other series when bronchoscopic intervention was prudently applied. As compared with the morbidity and mortality of surgical intervention, bronchoscopic management, as judged from our retrospective review, appears favorable in patients with loose or partly eroded broncholiths. Attempts at bronchoscopic extraction of a broncholith should ideally be conducted in a setting with capabilities for rigid and flexible bronchoscopy and immediate thoracic surgical support, and after the relation of the broncholith to adjacent vascular structures has been studied tomographically.