| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Division of Pulmonary and Critical Care Medicine, Stanford University School
of Medicine, 300 Pasteur Dr., Stanford, CA 94305.
2
Division of Respiratory and Critical Care Medicine, Santa Clara Valley Medical
Center, 751 S. Bascom Ave., San Jose, CA 95128, and Stanford University School
of Medicine, Stanford, CA 94305.
3
Department of Radiology, Alta Bates Hospital, Ashby Campus, 2450 Ashby Ave.,
Berkeley, CA 94705, and Stanford University School of Medicine. Stanford, CA
94305.
Received September 18, 2000;
accepted after revision April 19, 2001.
Address correspondence to C. M. Kirsch at San Jose address.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We retrospectively reviewed the cases of all patients with hemoptysis who had presented at either of two local hospitals, one county hospital and one community hospital, between 1988 and 2000 and who had undergone fiberoptic bronchoscopy before bronchial arteriography. All data were abstracted using a standardized coding form, and radiographs were independently reviewed by two of the authors.
RESULTS. Twenty-nine patients meeting the inclusion criteria were identified; one patient was excluded because of missing radiographs. The remaining 28 patients consisted of 19 men and nine women, with an average age of 54.6 years (age range, 16-91 years). The clinically determined diagnoses of their symptoms were tuberculous bronchiectasis (n = 14; 50.0%); bronchogenic carcinoma (n = 4; 14.3%); active tuberculosis (n = 2; 7.1%); nontuberculous bronchiectasis (n = 2; 7.1%); active coccidioidomycosis, pancreaticobronchial fistula, arteriovenous malformation, and tetralogy of fallot (n =1 each; 3.6% each); and unknown cause (n = 2; 7.1%). The bleeding site determined through bronchoscopy was consistent with that determined through radiographs in 23 patients (82.1%); all had either unilateral disease (n = 15), bilateral disease with unilateral cavities (n = 5), or a preponderance of disease on one side (n = 3). Bronchoscopy was an essential tool in determining the bleeding site in only three patients (10.7%), all of whom had bronchiectasis without localizing features visible on chest radiographs. In the remaining two patients (7.1%), bronchoscopic findings were indeterminate, but radiographs were helpful.
CONCLUSION. Fiberoptic bronchoscopy before bronchial artery embolization is unnecessary in patients with hemoptysis of known causation if the site of bleeding can be determined from radiographs and no bronchoscopic airways management is needed.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
If the bleeding lung can be confidently identified before the bronchial artery embolization procedure, therapeutic embolization can be performed in a focused and efficient manner. Although identification of the lobe in which bleeding originates is important in patients who are undergoing lobectomy for hemoptysis, lateralization of the bleeding source is usually all that is required for guidance of bronchial artery embolization. Many experts advocate the use of fiberoptic bronchoscopy as the primary method of localizing the site of bleeding in massive hemoptysis [8, 9], but some studies have placed less importance on this modality. Researchers have noted that if a focal infiltrate or region of most severe disease is visible on the chest radiograph, its presence is a fairly accurate indication that this area is the site of bleeding [10, 11]. In addition, flexible bronchoscopy often has limited usefulness if the patient's airways are filled with a large quantity of blood. Performing bronchoscopy in patients with massive hemoptysis who are already scheduled for bronchial artery embolization may not be as necessary or useful as previously thought. The risks of bronchoscopy, including possible airway compromise from sedation, delay in definitive treatment, and cost, must all be weighed against the clinical benefits of performing fiberoptic bronchoscopy before bronchial artery embolization for hemoptysis.
In a study by Ramakantan et al. [5], 140 patients with massive hemoptysis underwent bronchial artery embolization between 1988 and 1994. For logistic reasons, bronchoscopy was not performed before embolization; the decision to perform angiographic intervention was based solely on chest radiograph reviews. The chest radiographs of all patients showed abnormalities (unilateral lesions, n = 106; bilateral lesions, n = 34). Embolization was carried out on the side with the greater extent of disease and on both sides if there were no findings to indicate which side had the greater extent. The side with cavities was always embolized regardless of the size of the lesion. Effective control of hemorrhage was achieved in 131 patients (94%). These findings suggest that, in most cases, fiberoptic bronchoscopy is not essential for patients undergoing bronchial artery embolization as treatment for hemoptysis. The purpose of our study was to specifically investigate the clinical benefits of performing fiberoptic bronchoscopy before bronchial artery embolization in patients with hemoptysis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Chest radiographs and CT scans were reviewed independently by four of the investigators who were unaware of the bronchoscopic findings. The likely radiographic location of bleeding site was identified by consensus. Endoscopic findings, including the bleeding site as determined by bronchoscopy, were recorded from bronchoscopy reports.
We used the combination of conventional chest radiographs, chest CT scans (when available), fiberoptic bronchoscopic studies, bronchial arteriograms, and clinical results of bronchial artery embolization to determine the pulmonary site of hemorrhage.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The underlying causes of hemoptysis in the 28 patients are shown in Table 1. All 28 patients had chest radiographs obtained before fiberoptic bronchoscopic procedure. A chest CT scan was obtained in 13 patients before endoscopy during the initial episode of hemoptysis. However, the CT scan was useful in only two patients, both of whom had small lateralizing infiltrates that were visible only on CT scans, not on conventional radiographs. For the other 11 patients, the chest CT scan did not reveal any additional useful clinical information despite its ability to show parenchymal disease in more detail than the chest radiograph. Chest CT scanning was not performed in 15 patients because the clinicians did not believe the study would provide additional useful clinical information. The investigators agreed consistently and universally in their separate and unbiased interpretations of the radiographic studies of all 28 patients.
|
During the initial episode of hemoptysis, the bleeding lung, as determined through bronchoscopy, was the same as the one determined through radiographic studies (either chest radiograph, chest CT scan, or both) in 23 (82.1%) of the 28 patients (Fig. 1). These 23 patients had either unilateral disease (n =15, Fig. 2), bilateral disease with unilateral cavities (n = 5, Fig. 3), or a preponderance of disease on one side (n = 3, Fig. 4). The bleeding site was accurately determined through radiographic studies (either chest radiograph, chest CT scan, or both) in 25 (89.3%) of the 28 patients. Of these 25 patients, the source of pulmonary hemorrhage could be identified on chest radiographs in 23 individuals. In only two patients were chest CT scans required to locate the site of bleeding (Fig. 5A,5B). Fiberoptic bronchoscopy, on the other hand, accurately identified the bleeding source in 26 patients (92.9%). In 19 of them, the bleeding site was localized to a specific lobe. For the other seven patients, fiberoptic bronchoscopy identified only the bleeding lung, without providing a more precise location. In 23 (88.5%) of the 26 patients, however, the bleeding site was readily discernible from radiographic studies. No contradictions concerning the location of pulmonary hemorrhage were found between radiographic studies and fiberoptic bronchoscopy. However, three patients had inconclusive radiographic studies, and two patients had inconclusive fiberoptic bronchoscopic examinations.
|
|
|
|
|
|
Bronchoscopy was essential for determining the bleeding site in only three patients (10.7%), all of whom had bronchiectasis without localizing features on chest film. None of these three patients had chest CT scans to help localize the source of hemoptysis. Bronchoscopic findings were normal in two patients (7.1%) and, therefore, were not useful in determining the bleeding site. In these two patients, however, localizing features on radiographic studies were used to determine the source of hemoptysis (Fig. 1). No other diagnoses were made from fiberoptic bronchoscopic findings in any of the 28 patients. On the basis on these results, we conclude that fiberoptic bronchoscopy provided additional helpful information in only three patients (10.7%).
Among the four patients with recurrent hemoptysis requiring repeated embolization, fiberoptic bronchoscopy performed for subsequent episodes of hemoptysis did not reveal additional clinical information. All four patients had recurrence from the same site. In two of these patients, fiberoptic bronchoscopy was not performed during one or more of their subsequent episodes. The decision to perform embolization in these incidences was based on radiographic findings alone. Neither patient's chest radiographs showed a significant change from those previously obtained. Bronchial artery embolization achieved immediate hemostasis in all the bleeding recurrences for both patients.
During the 12-year period of our review, four patients did not have fiberoptic bronchoscopy performed before undergoing bronchial artery embolization for an initial occurrence of massive hemoptysis. Fiberoptic bronchoscopy was not performed in these patients for the following reasons: one patient had smear-positive active tuberculosis with unilateral cavitary lesions visible on radiographs, and the pulmonary consultant felt that the site of hemorrhage was reasonably well established without risking exposure to tuberculosis during fiberoptic bronchoscopy; one patient had active pulmonary coccidioidomycosis with a unilateral cavity, but fiberoptic bronchoscopy was deemed too risky by the pulmonary consultant, given the patient's age, poor pulmonary reserve, and refusal to be intubated; one patient had a unilateral cavitary lesion and a bruit on auscultation of the chest that the clinician believed located the bleeding vascular abnormality reliably without the necessity for fiberoptic bronchoscopy; and one patient had an unremarkable fiberoptic bronchoscopic examination that had been obtained 2 weeks before the episode of hemoptysis, and both the cause and source of hemoptysis were apparent from previous history and radiographs. We excluded these four patients from our statistical analysis. However, it is worth noting that in all four patients, the site for embolization was directed by radiographic findings and that bleeding had ceased immediately after embolotherapy in all patients.
Of these four patients, two did not have any recurrent hemorrhage and received no further treatment for hemoptysis during the follow-up period (follow-up time range, 18-24 months). One patient was lost to follow-up shortly after undergoing embolization. The fourth patient underwent elective lobectomy and did not have any further bleeding, although he later died of sepsis and multiorgan failure 3 years after surgery.
Four of our 28 patients ultimately had lobectomy for hemoptysis. Two of them required surgery because of recurrent hemoptysis despite multiple embolotherapy procedures (range, three to four procedures). In one patient, the bronchial arteriogram failed to identify an embolizable vessel. As a result, no embolization was performed, and the patient underwent lobectomy. The fourth patient was treated surgically for repeated hemoptysis, although bronchial artery embolization was performed only during his initial episode of bleeding. None of the four patients who were treated surgically had further pulmonary hemorrhage after lobectomy, although one of them died 21 months later from end-stage lung disease.
Vascular abnormalities, such as vascular hypertrophy and irregularity, neovascularity, and bronchial artery to pulmonary artery shunting, were seen on 29 of the 37 bronchial arteriograms (Fig. 6A,6B,6C). Active contrast extravasation was seen in only three of them. Of the eight patients with no noted vascular abnormalities, one underwent lobectomy, six had embolization performed on the vessels supplying the chosen areas of radiographic abnormalities, and one did not receive further therapy. The decision to embolize radiographically normal vessels in those six patients was based on the clinical supposition that these vessels supplied the bleeding site. Our retrospective review did not elucidate why the two additional patients with radiographically normal vasculature did not have bronchial artery embolization done.
|
|
|
Because of the extent of neovascularity, multiple vessels were often embolized in one procedure. A bronchial artery was the most commonly embolized vessel (n = 29), followed by internal mammary artery (n = 5), intercostal artery (n = 5), long thoracic artery (n = 2), thoracoacromial artery (n = 1), and pulmonary artery (n = 1). Both gelatin and polyvinyl alcohol particles were commonly used as embolizing agents in our series. None of our 28 patients had bilateral embolization. No complications were noted in any of the bronchial artery embolization procedures (including 37 bronchial arteriograms and 33 embolizations). The anterior spinal artery was not visualized during any of the procedures. All patients except one had cessation of active hemoptysis after bronchial artery embolization during the index hospitalization. That patient ultimately required lobectomy for hemoptysis that persisted despite three embolization procedures over the course of 1 week. None of the 28 patients in our study died within 30 days after their initial episode of hemoptysis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Historically, active tuberculosis and bronchiectasis from various diseases have been the most common causes of massive hemoptysis in the United States [8, 25]. The prevalence of these entities, however, has declined considerably since the first half of the 20th century [19]. Other conditions, including aspergilloma, bronchogenic carcinoma, lung abscess, mitral stenosis, cystic fibrosis, and trauma, are also associated with massive hemoptysis [8, 26]. Contrary to the statistics reported in many recent studies [19], tuberculous bronchiectasis is the most common underlying etiology for massive hemoptysis in our patients, probably because of our region's large immigrant population from areas in which tuberculosis is endemic. The large percentage of patients with tuberculous bronchiectasis (50.0%) in our study, however, does not appear to be a contributing factor to the high sensitivity of chest radiography in localizing the site of hemoptysis in our patients. In fact, tuberculous bronchiectasis was the underlying cause of hemoptysis in all three patients without lateralizing features on their chest radiographs (Table 1). They were the only patients for whom fiberoptic bronchoscopy proved to be helpful.
Immediate localization of the bleeding site followed by early thoracotomy and resection of the bleeding source constitutes ideal management [13, 27]. Unfortunately, in many instances, poor pulmonary reserve and other medical comorbidities decrease the patient's ability to tolerate surgical intervention. Overall, the mortality rate of patients with massive hemoptysis has been reported as ranging from 7% to 80% [9, 13, 18, 22, 26, 28], with a 30-40% operative mortality when treated surgically during the acute episode [16, 20]. Active bleeding at the time of surgery is associated with an increased mortality rate caused by asphyxiation [16, 20, 21, 29, 30]. There is a clear advantage to operating between episodes of bleeding in patients who survive massive hemoptysis [15, 31].
Nonsurgical interventions for hemoptysis may be used as an interim solution before surgery or may constitute definitive therapy in a patient who is a not a candidate for surgery. Bronchial artery embolization has gained popularity as a means of controlling massive hemoptysis in recent years, especially for patients with significant operative risk. Angiographic signs in hemoptysis include hyperplasia of the bronchial arterial trunk and branches, bronchopulmonary anastomoses, and bronchial arterial aneurysms [32]. Although extravasation of contrast medium is a frequent and specific angiographic sign in extrapulmonary bleeding, such extravasation is rarely seen in patients with hemoptysis [26, 32]. Extravasation of contrast material was observed in only three (10.7%) of our 28 patients. Therefore, it was not practical for us to use contrast extravasation as the gold standard for localization of hemoptysis. Instead, we believe that the pulmonary site of hemorrhage can be reliably determined using a combination of information from radiographic, angiographic, and endoscopic studies and correlation of this information with the clinical results of embolization.
Nonbronchial systemic collateral vessels from intercostal, axillary, subclavian, internal mammary, and phrenic arteries can also contribute to hemoptysis when they supply areas of parenchymal lung disease [8]. Embolization of these additional vessels may be required to control bleeding in selected patients. The time required to successfully cannulate and embolize the appropriate arterial vasculature may take minutes to hours, or the procedure may be ultimately unsuccessful for a number of technical reasons. Therefore, identification of the bleeding lung and the consequent ability to concentrate on the relevant vasculature will reduce operating time in a potentially critically ill patient. Despite its immediate effectiveness in achieving hemostasis, bronchial artery embolization may not be a permanent solution for some patients with massive hemoptysis. Bleeding recurs in as many as 30% of these patients as a result of incomplete embolization, revascularization, or recanalization of embolized vessels [8]. Elective resection should be considered the definitive treatment for those patients without contraindications to surgery, especially in patients who have had recurrent hemorrhage.
The safety of both bronchial arteriographic studies and therapeutic embolization has been well established [32]. Minor reactions that may occur with embolotherapy include chest pain, fever, and dysphagia, which usually resolve within a few days [26]. The most serious complication of the procedures is potential injury to the spinal cord [3, 33]. Spinal arteries arise from bronchial arteries, particularly the right bronchocostal trunk, in about 5% of patients [32, 34]. Injection of contrast agents (particularly those with high neurotoxicity) or therapeutic embolization in these patients can result in serious neurologic complications. The arteriogram, therefore, should include the thoracic spinal region, so that spinal arteries originating from the bronchial artery may be detected. Visualization of spinal arteries arising from the bronchial artery is a relative contraindication to repeated contrast injections as well as to therapeutic embolization [32]. In addition, ischemic bronchial or pulmonary injury associated with the embolization is possible. However, the ability of the bronchopulmonary circulation to withstand bronchial artery occlusion is illustrated by transplantation of the lung, which can be performed without reconstituting the bronchial arteries [32, 35]. It is only when very small microspheres virtually fill the terminal bronchial branches that bronchial injury can occur [32, 36].
Our study was designed to evaluate the clinical utility of performing fiberoptic bronchoscopy before bronchial artery embolization for massive hemoptysis of known etiology. The key question we addressed was whether fiberoptic bronchoscopy provided any additional information in helping clinicians with the management of massive hemoptysis that required embolization. Our analysis shows that fiberoptic bronchoscopy seldom provides information that cannot otherwise be derived from radiographic studies before patients undergo bronchial artery embolization. For patients in whom the source of hemorrhage was visible on both fiberoptic bronchoscopic and radiographic examinations, we found no disagreement between the findings of the two modalities. Fiberoptic bronchoscopy, in retrospect, did not provide any additional clinical information in these cases. Our findings indicate that both radiography and fiberoptic bronchoscopy are highly sensitive in localizing the bleeding site in patients with massive hemoptysis (89.3% vs 92.9%, respectively). It has been our experience that patients with massive hemoptysis, in general, do not bleed from airway lesions amendable to bronchoscopy. None of the patients in our study required therapeutic bronchoscopy to manage the airway, prevent asphyxia, or control a proximal endobronchial source of bleeding. Most of these patients were poor surgical candidates because of severe underlying lung disease. Bronchial artery embolization, therefore, is often the treatment of choice in this setting. The small difference between fiberoptic bronchoscopy and radiography in the ability to localize the bleeding source means that fiberoptic bronchoscopy has limited usefulness in patients with massive hemoptysis if the diagnosis is well established from previous history and if bronchoscopic airway management is not needed.
With the exception of two patients, chest CT scans did not provide any further clues to the source of bleeding. Although the data are limited, our observation is in agreement with that by Brinson et al. [10] in their review of cystic fibrosis patients who had bronchial artery embolization for hemoptysis. We share their view that chest CT scanning should be reserved for those patients whose symptoms and chest radiographs are not helpful. In our study, none of the three patients whose bleeding sites were localized only by bronchoscopy had CT scans of the chest. We do not know whether or not chest CT scanning would have been helpful in any of these three patients. Nevertheless, given its noninvasive nature, chest CT scanning should probably be considered before subjecting patients to fiberoptic bronchoscopy.
In conclusion, we advocate using radiographic studies to guide the approach to bronchial artery embolization in patients with a known cause of hemoptysis who do not need bronchoscopic airway management. Bronchoscopy should be reserved for those patients in whom the bleeding site cannot be located by radiographic methods (Fig. 7). This recommendation does not obviate the need for bronchoscopic evaluation (flexible or rigid bronchoscopy) in patients in whom the underlying cause of hemoptysis is not reasonably clear or in whom life-saving airway tamponade is needed. In addition, given the dire consequences of surgical intervention if performed on the wrong side, fiberoptic bronchoscopy verification of the bleeding source before surgery would be prudent.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. SIRAJUDDIN and T.-L. H. MOHAMMED A 44-year-old man with hemoptysis: A review of pertinent imaging studies and radiographic interventions Cleveland Clinic Journal of Medicine, August 1, 2008; 75(8): 601 - 607. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Savale, A. Parrot, A. Khalil, M. Antoine, J. Theodore, M.-F. Carette, C. Mayaud, and M. Fartoukh Cryptogenic Hemoptysis: From a Benign to a Life-threatening Pathologic Vascular Condition Am. J. Respir. Crit. Care Med., June 1, 2007; 175(11): 1181 - 1185. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Poyanli, B Acunas, I Rozanes, K Guven, S Yilmaz, A Salmaslioglu, E Terzibasioglu, and R Cirpin Endovascular therapy in the management of moderate and massive haemoptysis Br. J. Radiol., May 1, 2007; 80(953): 331 - 336. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A Khalil, M Soussan, G Mangiapan, M Fartoukh, A Parrot, and M-F Carette Utility of high-resolution chest CT scan in the emergency management of haemoptysis in the intensive care unit: severity, localization and aetiology Br. J. Radiol., January 1, 2007; 80(949): 21 - 25. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. F. Bruzzi, M. Remy-Jardin, D. Delhaye, A. Teisseire, C. Khalil, and J. Remy Multi-Detector Row CT of Hemoptysis RadioGraphics, January 1, 2006; 26(1): 3 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Revel, L. S. Fournier, A. S. Hennebicque, C. A. Cuenod, G. Meyer, P. Reynaud, and G. Frija Can CT Replace Bronchoscopy in the Detection of the Site and Cause of Bleeding in Patients with Large or Massive Hemoptysis? Am. J. Roentgenol., November 1, 2002; 179(5): 1217 - 1224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Yoon, J. K. Kim, Y. H. Kim, T. W. Chung, and H. K. Kang Bronchial and Nonbronchial Systemic Artery Embolization for Life-threatening Hemoptysis: A Comprehensive Review RadioGraphics, November 1, 2002; 22(6): 1395 - 1409. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
