CLINICAL STUDY

Results of Bronchial Artery Embolization for theTreatment of Hemoptysis Caused by Neoplasm

Ignasi Garcia-Olivé, MD, Jose Sanz-Santos, MD, Carmen Centeno, MD,Felipe Andreo, MD, Aida Muñoz-Ferrer, MD, Pere Serra, MD,

Jaume Sampere, MD, Josep Maria Michavila, MD, Jordi Muchart, MD, andJuan Ruiz Manzano, MD

ABSTRACT

Purpose: To describe experience with bronchial artery embolization (BAE) in a cohort of patients with cancer.

Materials and Methods: All consecutive patients with cancer and at least one episode of hemoptysis that required BAE duringa 14-year period were included in this observational retrospective review. The endpoints of the study were immediate success,recurrence of hemoptysis, mortality resulting from hemoptysis, and all-cause mortality.

Results: Immediate control of bleeding was achieved in 31 of 40 patients (77.5%). Recurrence requiring BAE occurred in eightpatients (20%). Cumulative hemoptysis control rate was 0.90 (95% confidence interval [CI], 0.80–1.0) at 1 month and 0.65 (95%CI, 0.44–0.86) at 6 months. Probability of survival was 0.75 (95% CI, 0.62–0.88) at 1 month, 0.42 (95% CI, 0.27–0.57) at 6months, 0.36 (95% CI, 0.21–0.51) at 12 months, and 0.08 (95% CI, 0.0–0.18) at 3 years.

Conclusions: BAE is an effective and safe technique in the treatment of hemoptysis in patients with cancer. Nevertheless, mortalityresulting from hemoptysis and recurrence rate are high among these patients secondary to progression of the underlying disease.

ABBREVIATIONS

BAE = bronchial artery embolization, CI = confidence interval, PVA = polyvinyl alcohol

Since its first description by Remy et al (1), bronchialartery embolization (BAE) has become an establishedprocedure in the management of massive and recurrenthemoptysis. Its efficacy, safety, and utility have beenestablished (2–12). Although the most common indica-tion for BAE is hemoptysis resulting from benign lungdiseases such as bronchiectasis, cystic fibrosis, or tu-berculosis, embolization may also be effective in con-trolling hemoptysis in patients with cancer. Although

& SIR, 2014

J Vasc Interv Radiol 2014; 25:221–228

http://dx.doi.org/10.1016/j.jvir.2013.09.017

None of the authors have identified a conflict of interest.

From Respiratory Service (I.G.-O., J.S.-S., C.C., F.A., A.M.-F., P.S., J.R.M.) andDepartment of Radiology (J.S., J.M.M., J.M.), Hospital Universitari GermansTrias i Pujol, Carretera del Canyet sn, 08916, Badalona, Barcelona; FundacióInstitut d’Investigació en Ciències de la Salut Germans Trias i Pujol (I.G.-O.,J.S.-S., F.A., J.R.M.), Badalona, Barcelona; CibeRes–Ciber de EnfermedadesRespiratorias (I.G.-O., F.A., J.R.M.), Bunyola, Bunyola, Mallorca; and Depart-ment of Medicine (J.R.M.), Universitat Autònoma de Barcelona, Bellaterra,Spain. Received May 6, 2013; final revision received September 26, 2013;accepted September 27, 2013. Address correspondence to I.G.-O.; E-mail:ignasi.g.olive@gmail.com

BAE is routinely performed in patients with hemoptysissecondary to neoplasm, several studies have reportedthat BAE in patients with lung cancer is associated witha high failure rate (4,10) and increased mortality(13). Nevertheless, the few studies that have focusedexclusively on patients with hemoptysis secondary tomalignancy have reported better results (14–16). Theaim of this study was to report our experience with BAEin a consecutive group of patients with cancer andhemoptysis that required BAE and to evaluate retro-spectively the efficacy and safety of the technique.

MATERIALS AND METHODS

Patient PopulationThis observational retrospective study comprised patientspresenting with life-threatening or persisting hemoptysiswho underwent BAE at Hospital Universitari GermansTrias i Pujol, a tertiary-care hospital in Badalona, Spain.Life-threatening hemoptysis was defined as bleedingamounting to 200 mL during 24 hours, 100 mL dailyduring at least 3 days, or minor bleeding with hemody-namic instability. All consecutive patients with at least

Garcia-Olivé et al ’ JVIR222 ’ BAE in Patients with Cancer

one episode of hemoptysis that required BAE during a14-year period (January 1999 to December 2012) with ahistologically confirmed diagnosis of either primary lungcancer or extrapulmonary cancer with pulmonary meta-stases were included. The clinical records were reviewed,and the following data were collected for analysis: age,sex, diagnosis, BAE findings, and time period betweenembolizations in case of recurrence. Written informedconsent was obtained from patients before embolization.The research protocol was approved by the regional ethicscommittee (Ethics Committee for Clinical Research of theHospital Universitari Germans Trias i Pujol).

BAEAccording to our institution protocol, platelet count,international normalized ratio, and activated partialthromboplastin time were reviewed before angiography,and any coagulation disorders were corrected before theprocedure. No patient was coagulopathic at the time ofangiography. No other therapeutic measures wereattempted before BAE. All procedures were performedby one of two interventional radiologists with 4 10 yearsof experience. The choice of catheter, microcatheter, andembolization materials was made at the discretion of theoperating physician. BAE was performed using theSeldinger technique via the femoral approach except inone patient in whom BAE was performed via a brachialapproach owing to extreme tortuosity of the supraaortictrunks. A descending thoracic aortogram was obtained atthe beginning of the procedure, followed by selectivecatheterization of bronchial arteries using various shaped4-F (Cobra 2, Radiofocus Glidecath; Terumo, Somerset,New Jersey) and 5-F angiographic catheters (Pigtail,Imager II; Boston Scientific, Natick, Massachusetts;Amplatz Left 2 Performa; Merit Medical, South Jordan,Utah). If no abnormal bronchial arteries were identified,nonbronchial systemic arteries were selectively catheter-ized, depending on the known site of pulmonary disease(assessed by x-ray, computed tomography (CT) scan, ordiagnostic bronchoscopy). Target vessels were superse-lectively catheterized using a 3-F coaxial microcathetersystem (Progreat; Terumo) if stable cannulation couldnot be achieved or if more distal cannulation of the vesselwas required to avoid important side branches, such asthe anterior spinal artery. The embolic material used wasnonspherical polyvinyl alcohol (PVA) particles (Contour;Boston Scientific) (years 1999–2008) or gelatin cross-linked acryl microspheres (Bead Block; Biocompatibles,Farnham, Surrey, United Kingdom) (years 2009–2012)for proximal vessels and metal coils (VortX 18 Diamond;Boston Scientific) for proximal and larger arteries (espe-cially when pseudoaneurysms in the bronchial arteries orarteriovenous malformations were present, whichoccurred in two patients). Embolization was performedof all abnormal vessels supplying the area of interest iftechnically possible. If there was no detectable underlying

pulmonary abnormality, embolization was performed ofall bronchial arteries on the side of the tumor that couldbe catheterized. Embolization was performed to completestasis. Contraindications to embolization were visual-ization of the anterior spinal branch (if it was not possibleto advance the catheter beyond this point) and catheterinstability. Pulmonary angiography was not routinelyperformed.No embolizations were performed by using coils alone.

Of the 40 embolization procedures, 23 (57.5%) wereperformed by using PVA particles alone. The size ofPVA particles ranged from 300–500 μm (n = 12) and 500–700 μm (n = 11). Of the 40 embolization procedures, 16(40%) were performed by using microspheres alone. Thesize of the microspheres ranged from 300–500 μm (n = 10)and 500–700 μm (n = 4). In two procedures, a combina-tion of microspheres 300–500 μm and 500–700 μm wasused. In one procedure, a combination of microspheres300–500 μm and PVA particles 300–500 μm was used.

Data AnalysisThe endpoints of this study were immediate success of theprocedure, recurrence of hemoptysis, mortality secondaryto hemoptysis, and all-cause mortality. Immediate successwas defined as cessation of hemoptysis during the sameadmission, whereas failure was indicated by continued orrecurrent hemoptysis during the same admission. Recur-rence was defined as an episode of life-threateninghemoptysis or persisting hemoptysis requiring BAE in apatient with a previous BAE performed. Indications forBAE in recurring hemoptysis were the same as for theinitial bleeding event (bleeding of 200 mL during 24 hours,100 mL daily during at least 3 days, or minor bleedingwith hemodynamic instability). Patient status at the end ofthe follow-up period was thoroughly investigated, includ-ing a review of hospital and outpatient medical records,computerized databases, and telephone contact with thepatient or family members or primary care physician whennecessary. When a patient died, the date of death wasobtained from hospital medical records, if he or she died inthe hospital, or from official death certificates.

Statistical AnalysisThe statistical package SPSS, version 19.0 (SPSS, Inc,Chicago, Illinois), was used for statistical analysis. Alldata were tabulated as mean and standard deviation inthe case of quantitative variables and as absolutenumbers and percentages in the case of qualitativevariables. Survival curves for the groups were con-structed according to the Kaplan-Meier method. Sur-vival curves were compared using the log-rank test.

RESULTS

During the study period, 40 patients with life-threaten-ing hemoptysis secondary to neoplastic disease were

Table 2 . Histology and Staging of 34 Patients with Primary

Lung Cancer

Histology

Squamous cell carcinoma 12 (35.3%)

Adenocarcinoma 10 (29.4%)

Non–small cell lung cancer 4 (11.8%)

Small cell carcinoma 2 (5.9%)

Poorly differentiated carcinoma 2 (5.9%)

Large cell carcinoma 1 (2.9)

Volume 25 ’ Number 2 ’ February ’ 2014 223

included. Characteristics of the patients are summarizedin Table 1. Histology and staging of patients withprimary lung cancer are summarized in Table 2.Results of complementary examinations are presentedin Table 3. All patients underwent embolization. Nomajor complications, according to the Society of Inter-ventional Radiology (SIR) definitions, occurred (17).There were no complications such as spinal cord ische-mia, transient pleurisy, or groin hematoma.

Unknown 3 (8.8%)

Staging

IA 1 (2.9%)

IIA 4 (11.8%)

IIB 2 (5.9%)

IIIA 5 (14.7%)

IIIB 6 (17.6%)

IV 13 (38.2%)

Immediate Control of HemoptysisImmediate arrest of hemoptysis was achieved in 31(77.5%) of 40 patients who underwent BAE. BAE failedto control bleeding in nine (22.5%) patients. Seven ofthese patients died of hemoptysis within 4 weeks of theprocedure, and two underwent successful repeat embo-lization.

Table 1 . Characteristics of 40 Patients Included in the Study

Age (y)

Mean (SD) 62.8 (11.7)

Sex

Male 34 (85%)

Female 6 (15%)

Smoker

Never 9 (22.5%)

Former 22 (55.0%)

Current 9 (22.5%)

Hemoptysis volume (mL)

Mean (SD) 241 (162)

Primary cancer

Lung 34 (85%)

Breast 2 (5.0%)

Colon 1 (2.5%)

Choriocarcinoma 1 (2.5%)

Sarcoma 1 (2.5%)

Hypernephroma 1 (2.5%)

Previous radiotherapy

No 27 (67.5%)

Yes 13 (32.5%)

Radiotherapy dose (Gy)

Mean (SD) 49.1 (18.4)

Recurrence

No 32 (80.0%)

Yes 8 (20.0%)

1 recurrence 6 (15.0%)

2 recurrences 2 (5.0%)

Mortality from hemoptysis

No 28 (70.0%)

Yes 12 (30.0%)

All-cause mortality

No 6 (15%)

Yes 34 (85%)

Values are number (percentage) unless otherwise indicated.

SD ¼ standard deviation.

Unknown 3 (8.8%)

Values are number (percentage).

Table 3 . Complementary Examinations

Imaging

Chest x-ray 10 (25%)

CT scan 24 (60.0%)

None 6 (15.0%)

Imaging findings

Normal 2 (5.0%)

Pulmonary nodule 4 (10.0%)

Multiple pulmonary nodules 4 (10.0%)

Pulmonary mass 23 (57.5%)

Ground glass 1 (2.5%)

None 6 (15.0%)

Great intrathoracic vessels affected CT

No 20 (83.3%)

Yes 4 (16.7%)

Bronchoscopy

No 20 (50.0%)

Yes 20 (50.0%)

Bronchoscopy findings

Blood 10 (50.0%)

Tumor 10 (50.0%)

Values are number (percentage).

Recurrence and Angiographic FindingsCumulative survival free of recurrence is shown inFigure 1. Probability of surviving free of recurrenceat 1 month was 0.90 (95% confidence interval [CI],0.80–1.0). Probability of surviving free of recurrence at 6months was 0.65 (95% CI, 0.44–0.86). Recurringhemoptysis that required further BAE occurred ineight patients. Characteristics of patients with hemo-ptysis recurrence after BAE and characteristics of theprocedures are presented in Table 4. Two of these eightpatients required more than one procedure to achievecontrol of the bleeding. Recurrence occurred within 1

Figure 1. Cumulative hemoptysis control rate.

Garcia-Olivé et al ’ JVIR224 ’ BAE in Patients with Cancer

month after the first episode in six cases (75%). Fourof the patients who underwent repeat embolizationprocedures had previous radiotherapy.

Mortality Secondary to HemoptysisDeath as a result of hemoptysis occurred in 12 (30%)patients. Eight (66.7%) patients died within the firstmonth after the episode. Three of four (75%) patientswith invasion of thoracic great vessels as visualized byCT scan died as a result of massive hemoptysis despiteBAE (range, 2–21 days). Probability of not dying as aresult of hemoptysis at 1 month and at 6 months was0.77 (95% CI, 0.64–0.90) and 0.68 (95% CI, 0.52–0.85).No patient presented with fatal hemoptysis after 8months of follow-up.

All-Cause MortalityThere were 28 (70%) patients who died as a result ofetiologies other than hemoptysis. There were 12 deathsattributed to progression of neoplastic disease, 4 attrib-uted to pneumonia, 3 attributed to cerebral hemorrhage,and 3 attributed to myocardial infarction. The cause ofdeath was unavailable for six (15%) patients. Cumulativesurvival is shown in Figure 2. Probability of survivingwas 0.75 (95% CI, 0.62–0.88) at 1 month, 0.42 (95% CI,0.27–0.57) at 6 months, 0.36 (95% CI, 0.21–0.51) at 12months, and 0.08 (95% CI, 0.0–0.18) at 3 years.Survival curve for patients who had previously

received radiation therapy was not statistically significant

compared with patients with no previous history ofradiotherapy (P ¼ .67). For the 34 patients with primarylung cancer, the mean overall survival was 14.2 months(95% CI, 7.41–21.03); the mean overall survival for the 6patients with extrapulmonary primary cancer was 10.1months (95% CI, 3.25–17.01). Difference in survivalbetween the two groups was not statistically significant(P ¼ .83).

Imaging FindingsThe vessels interrogated at diagnostic angiographyincluded the bronchial arteries (n ¼ 48), various inter-costal arteries (n ¼ 37), internal mammary arteries (n ¼4), and lateral thoracic artery (n ¼ 1). Angiographicfindings (Figs 3, 4) included active extravasation (17.5%;5 of 40 cases), tumor blush (90%; 36 of 40 cases),bronchial artery enlargement (62.5%; 25 of 40 cases),bronchial artery tortuosity (40%; 16 of 40 cases), andarteriovenous fistula (2.5%, 1 of 40 cases). Three (7.5%)of the cases manifested with a single abnormality: en-larged bronchial artery (n = 2) and arteriovenous fistula(n = 1). There were no differences in immediate successof BAE or mortality according to angiographic findingsor arteries involved.

DISCUSSION

Our data demonstrate that BAE is an effective and safetechnique in the treatment of hemoptysis in patients with

Table

4.CasesofHemoptysis

RecurrenceafterBAE

CaseNo.

Recurrenceafter

BAE(m

o)

Diagnosis

Staging

(LungCancer)

PreviousRT

MaterialUsedatBAE

Death

Caused

byHemoptysis

EmbolizedArteries

FirstBAE

SecondBAE

10.9

Sarcoma

No

PVA

No

BBT,RICBT

LBA

21.4

Lung(squamous)

IIA

No

PVA

No

LBA

LBA

35.6

Lung(N

SCLC)

Unknown

Yes

PVA

No

RICBT,LBA

BBT

41.4

Breast

No

PVA

No

LBA

BBT

57.7

Lung(N

SCLC)

IVYes

Microspheres

Yes

LBA

RBA

61.1

Lung(squamous)

IIA

Yes

Microspheresandcoils

Yes

RICBT,BBT

RLTA

75.4

Choriocarcinoma

No

Coils

Yes

AVF*

BBT

81.0

Lung(squamous)

IIIB

Yes

Microspheresandcoils

Yes

BBT

None†

AVF¼

arteriovenousfistula,BAE¼

bronchialartery

embolization,BBT¼

bibronchialtrunk,LBA¼

leftbronchialartery,NSCLC¼

non–smallcelllungcancer,PVA¼

polyvinylalcohol,

RBA

¼rightbronchialartery,RICBT¼

rightintercostobronchialtrunk,RLTA

¼rightlateralthoracic

artery,RT¼

radiotherapy.

nFedbysegmentalartery.

†Pulm

onary

artery

involved.

Volume 25 ’ Number 2 ’ February ’ 2014 225

cancer. Immediate control of bleeding was achieved inmost patients. Recurrence requiring BAE occurred in20% of patients.Hemoptysis in patients with cancer generally occurs

secondary to local necrosis and inflammation of bloodvessels within the tumor bed, rather than direct tumorinvasion of blood vessels (18). In these patients, thetumors feeding bronchial arteries are extended andenlarged (19). BAE has become a common first-linetherapy for life-threatening hemoptysis. Although mostprevious studies have focused on patients with hemopt-ysis secondary to benign lung diseases, such as bron-chiectasis, tuberculosis, and aspergilloma, some of theseseries included patients with lung cancer in their analysis(4,13). These studies showed that BAE in patients withlung cancer was associated with a high failure rate (4)and increased mortality (18), even though angiographicfindings in tumor-related bleeding were not significantlydifferent from findings in benign causes of hemoptysis(15,16). However, studies exclusively in patients withhemoptysis secondary to malignancy showed betterresults (14–16).Witt et al (14) described their experience with BAE

using platinum coils in patients with tumorous pulmonarybleeding. They prospectively studied 30 patients with lungcancer who underwent BAE and compared their resultswith 15 patients with moderate hemoptysis secondaryto lung cancer who had been treated conventionally.These authors found that the mean survival time wassignificantly lower in the control group than in the BAEgroup. Park et al (15) included in their analysis 19 patientswith a diagnosis of primary lung cancer who underwentBAE. They concluded that BAE was an effective tech-nique in the treatment of hemoptysis in patients with lungcancer. Wang et al (16) studied 30 patients with a diag-nosis of either lung cancer or pulmonary metastases fromextrathoracic disease. They concluded that BAE was safeand effective, especially in patients with cancer in whichbleeding was secondary to a benign cause and not relatedto the tumor.To the best of our knowledge, our series is among the

largest analyzing the results of BAE in patients withhemoptysis resulting solely from malignancy and showsresults similar to previous studies. Immediate control ofbleeding was achieved 77.5% of patients. Althoughimmediate success is defined differently in each article,it ranges from 79%–100% (15,16). Recurrence occurredin 20% of patients in our series, whereas in previousstudies it ranged from 26%–53% (14,16). In our series,the mean overall survival for the 34 patients withprimary lung cancer was 14.2 months. Wang et al (16)described a median overall survival time of 5.5 months.In the study by Witt et al (14), the mean survival time ofpatients with BAE treatment was 139 days. In our series,survival rate was 0.75 at 1 month, 0.42 at 6 months, 0.36at 12 months, and 0.08 at 3 years. In the series by Wanget al, survival rate was 0.70 at 1 month, 0.25 at

Figure 2. Cumulative survival (all cause mortality).

Figure 3. Selective bibronchial trunk arteriography (arrow) of a

79-year-old man with a pulmonary mass in the left lower lobe

and hemoptysis. Embolization with microspheres was success-

fully performed. The patient died 13 days after the procedure

because of acute respiratory failure and pneumonia.

Garcia-Olivé et al ’ JVIR226 ’ BAE in Patients with Cancer

12 months, and 0.087 at 3 years. In the series by Wittet al, survival rate was 0.71 at 15 days and 0.28 at 6months. Our study, with a large number of patients anda long follow-up period, shows that BAE is useful in themanagement of hemoptysis related to malignancy. It isalso the first study that differentiates mortality secon-dary to hemoptysis from all-cause mortality.Wang et al (16) analyzed the effect of previous

radiotherapy on the outcome of the technique. Theyfound that the survival of patients who had previouslyreceived radiation therapy was not statistically signi-ficant compared with patients with no prior history ofradiation therapy. In our group, we did not find differ-ences in survival between patients who had receivedradiotherapy and patients who had not. Also, Wang et al(16) compared the results of BAE performed forhemoptysis related to tumor and not related to tumor(16). In our series all patients had tumor-related hemo-ptysis, so we did not perform this analysis.Apart from BAE, other interventional procedures, such

as bronchoscopic techniques or palliative thoracic radio-therapy, can also be used. Several bronchoscopic treat-ments have been described for the treatment of life-threatening hemoptysis in patients with lung cancer.Endoscopic methods include cold saline lavage, instilla-tion of topical vasoconstrictive agents or fibrinogen-thrombin, balloon or stent tamponade, endobronchialairway blockade, laser therapy, argon plasma coagula-tion, and electrocautery (20). In our series, 10 patients had

endobronchial lesions. Only 25% of our patients wouldhave benefited from endoscopic treatment of hemoptysis.None of these bronchoscopic techniques were attempted inour patients. No patient received palliative thoracicradiotherapy in our series.Causes for recurrent bleeding include progression of

underlying disease; recanalization or revascularization;incomplete embolization; and emergence of other blood

Figure 4. (a, b) Selective catheterization of an intercostobronchial trunk with hypervascularity in the intercostal artery (a and b, arrow)

and right bronchial artery (a, arrowhead) in a 53-year-old man with a right pulmonary mass and hemoptysis. Embolization with

microspheres and one coil was successfully performed. The patient was alive 10 months after the procedure.

Volume 25 ’ Number 2 ’ February ’ 2014 227

supply to the affected area, such as from systemiccollaterals (21). Early rebleeding has been related toincomplete embolization, whereas late rebleeding isgenerally due to disease progression (2,4,7,9). Garcia-Olivé et al (22) reported more recently that recurringhemoptysis secondary to recanalization was relatedto time to recurrence but not to use of coils, numberof arteries that received embolization, or neoplasticetiology. More specifically, we found that the arterythat received embolization in the first 80 weeks afterthe first procedure would be a different one in mostcases (suggesting incomplete embolization), whereas theartery that received embolization after 80 weeks wouldbe the same as in the previous procedure in most cases(suggesting recanalization) (22). Our series includedhemoptysis secondary to many conditions, includingbenign and malignant conditions. In our study, theartery that received embolization in the secondprocedure was a different one in seven of the eightpatients with recurring hemoptysis, which may suggestincomplete embolization or emergence of other supplyto the affected area. The small size of the sample didnot allow us to compare the results between differentagents.Multidetector CT has a role in the identification of the

bronchial and nonbronchial arteries causing hemoptysisand improves the results of BAE (23–25). In our series, 6(15%) patients had no imaging before BAE, and only60% had a CT scan performed before the procedure. Allsix patients with no imaging were included in the studybefore 2004. Since then, our institution protocol for themanagement of life-threatening hemoptysis has changed,and it is now mandatory to perform a CT scan beforeBAE except when there is hemodynamic instability orextremely rapid bleeding.We have not seen any severe treatment-related com-

plications. Likewise, Witt et al (14) did not report anycomplications. However, Park et al (15) described twominor complications (one episode of transient chest pain

and one puncture site hematoma), and Wang et al (16)described two minor complications (two cases of vesseldissection) and one spinal cord infarction.There are several limitations to this study. First, the

sample size is small. Second, this is a retrospective study,with embolization procedures performed by different op-erators using different tools and embolic agents. Finally,there is no control group with which compare the results.In conclusion, BAE is an effective and safe technique

in the treatment of hemoptysis in patients with oncologicdisease. Nevertheless, mortality secondary to hemoptysisand recurrence are high among these patients, most likelyresulting from progression of the underlying disease.

REFERENCES

1. Remy J, Voisin C, Ribet M, et al. Treatment, by embolization, of severeor repeated hemoptysis associated with systemic hypervascularization.Nouv Presse Med 1973; 2:2060 [in French].

2. Rabkin JE, Astafjev VI, Gothman LN, Grigorjev Y. Transcatheter embo-lization in the management of pulmonary haemorrhage. Radiology 1987;163:361–365.

3. Katoh O, Kishikawa T, Yamada H, Matsumoto S, Kudo S. Recurrentbleeding alter arterial embolization in patients with hemoptysis. Chest1990; 97:541–546.

4. Hayakawa K, Tanaka F, Torizuka T, et al. Bronchial artery embolizationfor hemoptysis: immediate and long-term results. Cardiovasc InterventRadiol 1992; 15:154–159.

5. Ramakantan R, Bandekar VG, Gandhi MS, Aulakh BG, DeshmukhHL. Massive hemoptysis due to pulmonary tuberculosis: control withbronchial artery embolization. Radiology 1996; 200:691–694.

6. Brinson GM, Noone PG, Mauro MA, et al. Bronchial artery embolizationfor the treatment of hemoptysis in patients with cystic fibrosis. Am JRespir Crit Care Med 1998; 157:1951–1958.

7. Mal H, Rullon I, Mellot F, et al. Immediate and long-term results ofbronchial artery embolization for life-threatening hemoptysis. Chest 1999;115:996–1001.

8. Osaki S, Nakanishi Y, Wataya H, et al. Prognosis of bronchial arteryembolization in the management of hemoptysis. Respiration 2000; 67:412–416.

9. Kato A, Kudo S, Matsumoto K, et al. Bronchial artery embolization forhemoptysis due to benign diseases: immediate and long-term results.Cardiovasc Intervent Radiol 2000; 23:351–357.

10. Swanson KL, Johnson M, Prakash UBS, McKusick MA, Andrews JC,Stanson AW. Bronchial artery embolization. Experience with 54patients. Chest 2002; 121:789–795.

Garcia-Olivé et al ’ JVIR228 ’ BAE in Patients with Cancer

11. Kim YG, Yoon HK, Ko GY, Lim CM, Kim WD, Koh Y. Long-term effect ofbronchial artery embolization in Korean patients with haemoptysis.Respirology 2006; 11:776–781.

12. Chun JY, Belli AM. Immediate and long-term outcomes of bronchialand non-bronchial systemic artery embolisation for the management ofhaemoptysis. Eur Radiol 2010; 20:558–565.

13. Fartoukh M, Khoshnood B, Parrot A, et al. Early prediction of in-hospitalmortality of patients with hemoptysis: an approach to defining severhemoptysis. Respiration 2012; 83:106–114.

14. Witt CH, Schmidt B, Geisler A, et al. Value of bronchial arteryembolisation with platinum coils in tumorous pulmonary bleeding. Eur JCancer 2000; 36:1949–1954.

15. Park HS, Kim YI, Kim HY, Zo JI, Lee JH, Lee JS. Bronchial artery andsystemic artery embolization in the management of primary lung cancerin patients with hemoptysis. Cardiovasc Intervent Radiol 2007; 30:638–643.

16. Wang GR, Ensor JE, Gupta S, Hicks ME, Tam AL. Bronchial arteryembolization for the management of hemoptysis in oncology patients:utility and prognostic factors. J Vasc Interv Radiol 2009; 20:722–729.

17. Angle JF, Siddiqi NH, Wallace MJ, et al. Quality improvement guide-lines for percutaneous transcatheter embolization. Society of Interven-tional Radiology standards of practice committee. J Vasc Interv Radiol2010; 21:1479–1486.

18. Winter SM, Ingbar DH. Massive hemoptysis: pathogenesis and man-agement. J Intens Care Med 1988; 3:171–188.

19. Newton TH, Preger L. Selective bronchial arteriography. Radiology1965; 84:1043–1051.

20. Gompelmann D, Eberhardt R, Herth FJF. Advanced malignant lungdisease: what the specialist can offer. Respiration 2011; 82:111–123.

21. Yoon W, Kyu Kim J, Hyun Kim Y, Woong Chung T, Keun KangH. Bronchial and nonbronchial systemic artery embolization for lifethreatening hemoptysis: a comprehensive review. Radiographics 2002;22:1395–1409.

22. Garcia-Olivé I, Sanz-Santos J, Centeno C, et al. Predictors of recanaliza-tion in patients with life-threatening hemoptysis requiring artery embo-lization. Arch Bronconeumol 2013; pii: S0300-2896(13)00188-9.

23. Yoon YC, Lee KS, Jeong YJ, Shin SW, Chung MJ, Kwon OJ. Hemopt-ysis: bronchial and nonbronchial systemic arteries at 16-detector row CT.Radiology 2005; 234:292–298.

24. Mori H, Ohno Y, Tsuge Y, et al. Use of multidetector row CT to evaluatethe need for bronchial arterial embolization in hemoptysis patients.Respiration 2010; 80:24–31.

25. Hayes D Jr, Winkler MA, Kirkby S, Capasso P, Mansour HM,Attili AK. Preprocedural planning with prospectively triggered multidetectorrow CT angiography prior to bronchial artery embolization in cystic fibrosispatients with massive hemoptysis. Lung 2012; 190:221–225.