| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Full paper |
Istanbul Medical Faculty, Department of Radiology, Capa 34390, Sehremini, Istanbul, Turkey
Correspondence: Arzu Poyanli, Assistant Professor, IU Istanbul Tip Fakultesi Radiodiagnostik ABD, Department of Radiology, Capa 34390, Turkey. E-mail: aopoyanli{at}hotmail.com
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Angiographic detection and endovascular treatment of the bleeding source is reported to be an effective and safe palliative treatment alternative [1, 5]. Recent successful results have broadened the indications for endovascular treatment to include moderate (more than or equal to three episodes of 100 ml of bleeding per day within 1 week) and even mild (chronic or slowly increasing) haemoptysis cases [6].
In this study, we aimed to evaluate our experience with 140 patients with massive and moderate haemoptysis treated by bronchial artery embolisation.
| Methods and materials |
|---|
|
|
|---|
Haemoptysis was caused by tuberculosis in 136 (97.1%) patients and lung cancer in four (2.9%). Haemoptysis was massive in all patients with lung cancer. All patients diagnosed with tuberculosis were in the inactive stage of the disease.
Bronchoscopy findings were taken into consideration whenever they were available. We did not insist on this procedure in order not to delay the treatment. Bilateral bronchial angiography was performed in every case.
Angiography was performed via femoral access in all patients. Embolisation was performed with high brachial puncture in four cases because the feeders originating from the subclavian artery could not be catheterized via femoral access. A 5F vascular sheath was used in all cases to facilitate catheter changes. 4F cobra or Simmons-1 catheters were used for selective catheterization. A microcatheter was used when necessary to ensure safe catheterization.
Assessment included selective bronchial angiography in every case. Additional angiography of the intercostal, subclavian or inferior phrenic arteries was performed depending on the location of the disease in the lung (Figure 1
). Further flush angiography of the aortic arch and thoracic aorta was performed if the source was not identified. Pulmonary angiography was not routinely performed.
|
After selective catheterization of the bleeding source in the bronchial artery or non-bronchial systemic vessels, embolisation was performed under fluoroscopic control using 250350 µm and 350500 µm sized polyvinyl alcohol (PVA) particles (Contour; Boston Scientific, Natick, MA) suspended in contrast medium. 350500 µm PVA particles were used in patients with bronchopulmonary shunts and large bronchial arteries. Embolisation was terminated when antegrade flow was observed to stagnate or terminate.
Two 5 mm/10 mm and 3 mm/4 mm sized coils (Cook, Bjaeverskov, Denmark) were used to embolise the bleeding source originating from the main trunk supplying the apical and posterior branches of the right superior lobe artery in one case.
In patients with recurrent bleeding, additional bronchial artery, aberrant bronchial artery and abnormal non-bronchial systemic and pulmonary supply were investigated with angiography.
| Results |
|---|
|
|
|---|
Angiography findings are summarized in Table 1
and the distribution and number of embolised arteries in Table 2
.
|
|
The immediate success rate after bronchial artery embolisation, defined as no recurrence of bleeding within 24 h, was 98.5% (138 of 140 patients). Additional left bronchial artery supply, not detected in the first session, was detected and embolised during repeat angiography in two patients who had 150 ml of bleeding during the first 24 h. There was no recurrence at 3 months and 6 months follow up.
Control of haemoptysis at 1 month was observed in 126 patients (90%). 86 of these patients had massive and 40 had moderate haemoptysis and there was no recurrence during a mean follow up period of 33.6 months (190 months).
Severe haemoptysis developed in 11 prior massive and three prior moderate haemoptysis cases during an average time of 3.7 months (17 months) following the last intervention. There was systemic collateral supply from the costocervical trunk and abrupt termination of the segmentary pulmonary arterial branch in 1 case, and bronchial artery and systemic collateral development (not present in the previous session) in 11 cases. The bleeding sources or sources in these patients were embolised following selective catheterisation. Two sessions of endovascular treatment were performed in ten and three sessions were performed in two patients with recurrent haemoptysis. All of the recurrences were moderate. No haemoptysis was reported after the last embolisation during a mean follow up period of 46.6 months (1686 months). Two of the cases with malignancy died during abundant bleeding following an asymptomatic 1 month period before any intervention could be performed.
Mild chest pain developed during and after the process in 67 (47.8%) cases. It resolved spontaneously in 27 days.
There were no major procedure related complications.
| Discussion |
|---|
|
|
|---|
It is reported that bronchoscopy helps to detect the bleeding site in a lung or lobe in patients with diffuse pulmonary disease [9]. However, in patients with severe haemoptysis, the airways are filled with large volumes of blood restricting the use of bronchoscopy and similarly rendering endobronchial treatment ineffective [7]. Hsiao et al reported that bronchoscopy was not a prerequisite in the treatment process considering the risk of airway compromise from sedation, delay in definitive treatment, hypoxaemia and high cost [10]. In this study, bronchoscopy findings were taken into consideration whenever they were available. Not having performed a bronchoscopy did not affect our endovascular treatment progress. Bronchoscopy findings have not altered the course of angiography and endovascular treatment in any of our patients. In our experience, there was no need for emergency bronchoscopy during active severe bleeding.
CT has diagnostic superiority over bronchoscopy and chest radiograph for demonstrating underlying pathology and the site of bleeding in haemoptysis, especially in bronchiectasis, bronchogenic carcinoma and aspergilloma cases [8, 11]. Vascular pathologies such as arteriovenous malformation or aneurysm, which are rare causes of haemoptysis, are also depicted very clearly in contrast enhanced CT examinations [12]. With recent developments in multidetector CT technique it is now possible to scan the whole thorax with very thin slices (1.25 mm) in a very short time (1215 s) [13, 14]. Both the lesion causing haemoptysis and the bronchial and non-bronchial systemic feeding arteries are detected during the same study using 80100 ml of contrast medium. In angiography controlled studies, 8687% of the pathologic vessels detected by angiography are reportedly demonstrated with CT angiography (CTA) as well [13, 14]. The road map thus prepared for the interventional radiologist by CTA is thought to shorten the examination time in critical patients.
The source of severe haemoptysis is the bronchial circulation in 90% of cases [15]. Significant bleeding from the pulmonary circulation is rare (5%) and is usually secondary to erosion of a pulmonary arterial branch by chronic inflammation. It is recommended to investigate the pulmonary circulation for a bleeding source following systemic arterial embolisation, in patients with recurring haemoptysis due to fibrocavitary tuberculosis [5]. In a minority of cases (approximately 5%) severe haemoptysis may originate from the aorta (e.g. aortobronchial fistula, ruptured aortic aneurysm) or non-bronchial systemic pulmonary arteries [16, 17]. Intercostal arteries, thoracic arteries originating from axillary and subclavian arteries in the upper and inferior phrenic artery in the lower zone disease should be investigated during the initial session, particularly in cases with chronic infection and pleural disease because non-bronchial systemic artery supply is the most important cause of early recurrent haemoptysis after successful bronchial artery embolisation [5]. Since history of chronic infection was present in the majority of patients in this study, systemic vascular supply was investigated in all sessions.
Chun et al reported that they found and embolised 29 additional arteries (14.5%) which were identified on post-embolisation thoracic aortography. They stressed that inferior phrenic artery and intercostal arteries could be overlooked when thoracic aortography was not performed. It is known that the bronchial arteries do not have a constant origin and that they originate from the inferior aspect of the aortic arch in up to 15% of the population [19]. Regarding these findings, we injected the aortic arch and thoracic aorta after searching for bronchial artery and systemic vascular supply.
It has been suggested that the artery of Adamkiewicz will arise from the right intercostobronchial artery in up to 10% of the population, but it is generally believed that the true incidence is considerably lower [20]. Embolisation distal to the origin of artery of Adamkiewicz is the ideal treatment method in these cases, especially with the modern angiography equipment making distal navigation possible. However, in some cases, a thin spinal branch may appear after initiation of embolisation resistance is increased in the vascular bed. In animal studies, particles larger than 200250 µm were found to be too large to reach the spinal arteries and were considered safe when spinal arteries were not visualized [21]. In another experimental study, bronchopulmonary anastomoses in human lung were found to be 325 µm in diameter; therefore, the use of smaller particles could cause pulmonary infarction [22]. In this study, 250350 µm and 350500 µm sized PVA particles were used. 350500 µm PVA particles were preferred in patients with bronchopulmonary shunts and large bronchial arteries. No spinal arteries were observed in selective injections prior to or during embolisation in any of the cases, and there were no neurological complications. Furthermore, no pulmonary infarction developed in any of the cases with bronchopulmonary shunts.
Currently, PVA particles are the most commonly used agents for bronchial artery embolisation worldwide. PVA particles are biocompatible and non-biodegradable and are considered to be a permanent embolic agent. However, unpredictable proximal vessel occlusion and catheter blockage caused by clumping or aggregation of irregular-shaped PVA particles have been reported [23]. Recently, the interest in spherical agents has grown, to overcome the drawbacks of PVA. Trisacryl gelatine microsphere (Embosphere; Biosphere, Rockland, MA; Contour SE; Boston Scientific, Boston, MA) is a new embolic agent that is increasingly used for uterine fibroid embolisation. In vivo and in vitro studies have shown that these particles clearly have better sizing and penetration characteristics than PVA [24, 25]. Vinaya et al reported a myocardial infarction followed by stroke in a patient with haemoptysis due to spontaneous passage of particles through the bronchopulmonary shunts with 500 µm sized embospheres [26]. It is obvious that further clinical and experimental studies investigating the effectiveness and safety of bronchial artery embolisation with these particles are needed.
Acute control of haemoptysis with endovascular treatment can be achieved in 7398% of the patients with massive haemoptysis [2729]. However, haemoptysis can recur with a frequency of up to 20%, especially in cases with chronic tuberculosis, aspergilloma and neoplasia [3032]. Recurrent bleeding during the first 6 months after the embolisation procedure is usually due to undetected bronchial and systemic collaterals caused by diffuse pulmonary disease [5]. Late recurrent bleeding (after 6 months) usually develops due to disease progression. In this study, haemoptysis recurred in 17 months (mean 3.7 months) after the first intervention in 14 cases (10%). Haemoptysis following re-embolisation was encountered in two cases after 1 month and 6 months and a repeat endovascular treatment was performed. No haemoptysis was reported in the mean 46.6 months (1686 months) follow-up period after the last embolisation until today. Two of the cases with malignancy died because of abundant bleeding after an asymptomatic 1 month period.
The successful results achieved with endovascular treatment in the control of massive haemoptysis have also brought up the idea of embolisation of moderate and chronic recurrent, mild haemoptysis compromising the life quality of the patient. Barben et al reported the results of bronchial artery embolisation in 46 bleeding episodes in 20 patients with cystic fibrosis. There were 22 massive, 10 moderate and 14 chronic recurrent bleeding episodes [33]. They observed that immediate success was achieved in 95% of the cases, whereas recurrent haemoptysis occurred in 55% of the cases after a mean period of 4 months. Yu-Tang et al reported the success rate as 81.5% in their study group consisting of 103 patients embolised for moderate to massive haemoptysis [34]. They stressed that in 15.5% of cases re-embolisation would be necessary and that endovascular treatment should be repeated whenever haemoptysis could be controlled by this method. In this study, moderate haemoptysis comprised 30.8% of the cases and successful results were achieved with endovascular treatment in these patients similar to those with massive haemoptysis (haemoptysis control at 1 month; 93% and 88.6% control with repeat session(s); 7% and 9.2% respectively, in moderate and massive haemoptysis cases).
In summary, effective palliation with endovascular treatment is achieved in the majority of moderate and massive haemoptysis cases after a single intervention. If necessary, the procedure may be repeated as long as the haemoptysis can still be controlled by this method. Endovascular treatment performed by an experienced staff with sufficient technical back-up is an effective and safe choice in the control of primary and recurrent haemoptysis. We believe that bronchial artery embolisation should always be considered for treatment of moderate to massive hemoptysis.
Received for publication June 14, 2006. Revision received July 18, 2006. Accepted for publication August 15, 2006.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. M. T. Lau, V. Yozghatlian, C. Kosky, C. Moriarty, R. Dentice, R. Waugh, P. J. Torzillo, and P. T. Bye Recombinant Activated Factor VII for Massive Hemoptysis in Patients With Cystic Fibrosis Chest, July 1, 2009; 136(1): 277 - 281. [Abstract] [Full Text] [PDF] |
||||
![]() |
BJR review of the year -- 2007 Br. J. Radiol., April 1, 2008; 81(964): 265 - 269. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |