- Email: [email protected]
Severe Hemoptysis of Pulmonary Arterial Origin
Original Research CHEST IMAGING
Severe Hemoptysis of Pulmonary Arterial Origin* Signs and Role of Multidetector Row CT Angiography Antoine Khalil, MD; Antoine Parrot, MD; Cosmina Nedelcu, MD; Muriel Fartoukh, MD; Claude Marsault, MD; and Marie-France Carette, MD
Background: Hemoptysis of pulmonary arterial origin is a diagnostic challenge in patients admitted to a respiratory ICU (RICU) for treatment of hemoptysis. Its early accurate recognition and treatment reduce morbidity and prevent mortality. Multidetector row CT angiography (MDCTA) is an accurate method for imaging the systemic vascular network. Our aim was to assess the MDCTA signs and role in managing hemoptysis of pulmonary arterial origin. Methods: We performed a retrospective clinical and radiologic analysis of all consecutive patients who were referred for severe hemoptysis to our RICU and were treated by endovascular means between January 2004 and December 2006. We reviewed all of those cases with hemoptysis of pulmonary arterial origin. Results: Of 272 patients who were referred for severe hemoptysis to the RICU, 189 patients were treated by endovascular means. Thirteen patients (nine men, four women; mean age, 45 years) had hemoptysis of pulmonary arterial origin. Signs of pulmonary arterial hemoptysis seen on MDCTA were of the following three types: pseudoaneurysm (n ⴝ 5); aneurysm of the pulmonary artery (n ⴝ 3); or the presence of a pulmonary artery in the inner wall of a cavity (n ⴝ 5). Hypertrophy of the bronchial arteries seen on MDCTA associated with any of these signs predicted the necessity to treat both the bronchial and pulmonary arteries. Pulmonary artery vasoocclusion was performed as a first treatment in eight patients with such an association (n ⴝ 1) or without such an association (n ⴝ 7) along with bronchial artery embolization. The remaining five patients were treated with systemic artery embolization, followed by surgery (n ⴝ 1), pulmonary artery vasoocclusion (n ⴝ 3), and death from massive hemoptysis (n ⴝ 1). Conclusions: MDCTA performed prior to endovascular treatment allows the correct identification and early appropriate management of severe hemoptysis of pulmonary arterial origin. (CHEST 2008; 133:212–219) Key words: Behc¸et disease; CT scan; hemoptysis; pulmonary artery; tuberculosis Abbreviations: BAE ⫽ bronchial artery embolization; MDCTA ⫽ multidetector CT angiography; NBSA ⫽ nonbronchial systemic artery; PAA ⫽ pulmonary artery aneurysm; PAPA ⫽ pulmonary artery pseudoaneurysm; RICU ⫽ respiratory ICU
hemoptysis is a life-threatening condition M assive that is associated with a mortality rate exceeding 50% in the absence of adequate treatment.1–3 Hemoptysis usually involves bleeding from the bronchial arteries or, less frequently, from nonbronchial systemic arteries (NBSAs). Hemoptysis of pulmonary arterial origin is rare, estimated at ⬍ 10% of hemoptysis cases. Unfortunately, clinical presentation cannot easily distinguish pulmonary and sys212
temic mechanisms, which also can be associated with hemoptysis. The pulmonary etiologies of hemoptysis are numerous. Before the era of multidetector row CT angiography (MDCTA), patient management involved fiberoptic bronchoscopy, bronchial artery embolization (BAE), and, in some cases, surgery.4 Hemoptysis of pulmonary arterial origin was usually suspected when BAE and NBSA embolization were unable to Original Research
Table 1—Clinical Characteristics, MDCTA Findings, and Treatment of Patients With Hemoptysis of Pulmonary Arterial Origin* Patient No./Age, yr/Sex/ Hemoptysis Cumulative Volume, mL 1/80/F/600 2/22/F/200
3/29/M/⬎1000 4/45/M/200
5/27/M/300 6/42/M/600
7/24/M/100
8/82/F/150
9/72/M/100
10/52/M/⬎1000 11/61/M/300
12/49/F/500 13/36/M/100
MDCTA of Pulmonary Arteries PAPA BA of normal size PAPA BA of normal size PAPA BA of normal size PAPA BA and NBSA hypertrophy PAA, contained air bubble BA of normal size Multiple PAAs, RML PAA contained air bubble and was surrounded by an alveolar consolidation BA hypertrophy Multiple PAAs; two of them were surrounded by ground-glass opacities BA hypertrophy Presence of a PA in the inner wall of the cavity BA hypertrophy Presence of a PA in the inner wall of the abscess BA of normal size PAPA in lung abscess BA hypertrophy Subsegmental left upper PA in inner wall cavity BA hypertrophy Left lower PA in inner wall cavity BA hypertrophy Right lower PA in the inner wall cavity BA of normal size
Treatment and Evolution Pulmonary angiography and superselective vasoocclusion Pulmonary angiography and selective vasoocclusion without pseudoaneurysm identification Pulmonary angiography and superselective vasoocclusion BAE relapse hemoptysis, pulmonary angiography, and selective vasoocclusion after opacification of PAPA Pulmonary angiography and superselective PA vasoocclusion Pulmonary angiography and selective vasoocclusion of RML PAA
Definitive Diagnosis Active tuberculosis Active tuberculosis
Active tuberculosis Active tuberculosis
Behçet disease Hughes-Stovin syndrome
BA vasooccluded, and two PAAs of the right lower lobe vasooccluded in the same session
Behçet disease
Subsegmental right upper lobe pulmonary artery vasoocclusion after contrast media extravasation into cavity Hemoptysis relapses 1 wk after BAE Pulmonary angiography and superselective vasoocclusion after contrast media extravasation into cavity BAE, pulmonary angiography, and superselective vasoocclusion BAE, hemoptysis relapses 1 wk later; new MDCTA revealed PAPA, which was vasooccluded BAE, massive hemoptysis relapse
Aspergilloma
BAE without major hypervascularization, patient sent to surgery
Invasive aspergillosis
Lung abscess Bronchial carcinoma excavation Bronchial carcinoma excavation Hodgkin lymphoma excavation
*BA ⫽ bronchial artery; PA ⫽ pulmonary artery; RML ⫽ right middle lobe; F ⫽ female; M ⫽ male.
control hemoptysis.5 More recently, MDCTA of the bronchial arteries and NBSAs have emerged as potential diagnostic tools for evaluating systemic *From the Radiology Department (Drs. Khalil, Nedelcu, Marsault, and Carette) and the Respiratory Intensive Care Unit (Drs. Parrot and Fartoukh), Assistance Publique-Hoˆpitaux de Paris, Tenon Hospital, Paris, France. The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Manuscript received May 16, 2007; revision accepted October 8, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Antoine Khalil, MD, Radiology Department, AP-HP Tenon Hospital, 4 Rue de la Chine, 75020 Paris, France; e-mail: [email protected] DOI: 10.1378/chest.07-1159 www.chestjournal.org
circulation in patients with hemoptysis.6,7 However, a large-scale analysis of the role of MDCTA and its influence in exploring and managing hemoptysis of pulmonary arterial origin has not yet been reported. The aim of this study was to assess the thoracic MDCTA signs and role in patients with severe hemoptysis of pulmonary arterial origin in patients admitted to the ICU for BAE.
Materials and Methods The clinical charts of all patients referred to the respiratory ICU (RICU) of our hospital for hemoptysis and treated by endovascular means (ie, systemic artery embolization of the bronchial artery and NBSA, and pulmonary artery vasoocclusion) during a 36-month period (January 2004 to December 2006) CHEST / 133 / 1 / JANUARY, 2008
213
Figure 1. Patient 5: a 26-year-old man with Behc¸et disease revealed by pulmonary artery thrombosis and a postpuncture left radial artery aneurysm. He was referred to the RICU for hemoptysis of a volume of 100 mL. Top left, A: sagittal 1-mm reformation MDCTA on the parenchymal window shows right lower lobe consolidation (*) related to hemoptysis. Top right, B: sagittal 15-mm-thick slab maximum-intensity projection MDCTA shows two partially thrombosed (arrows) right lower lobe PAAs in area of lung bleeding. Bottom left, C: coronal 20-mm-thick slab maximum-intensity projection MDCTA shows enlargement of the right bronchointercostal trunk (arrow) and its bronchial artery (arrowheads). Bottom right, D: sagittal 5-mm-thick slab maximum-intensity projection MDCTA 3 days after endovascular treatment shows coils into both PAAs (arrows).
were collected. Each patient who was admitted for treatment of hemoptysis in our tertiary university hospital and referral center for hemoptysis underwent chest radiography, fiberoptic bronchoscopy, and MDCTA. The charts of patients with hemoptysis of pulmonary arterial origin were obtained for retrospective review. Hemoptysis of pulmonary arterial origin was assessed based on the following data: MDCTA; pulmonary angiography; histology (when available); and outcome. The cause of hemoptysis was determined from the patient’s history; findings of the physical examination, chest radiography, fiberoptic bronchos214
copy, MDCTA, microbiology, and histology (when available); and outcome. MDCTA was performed for each patient admitted to the RICU for hemoptysis in our institution as a pretreatment examination during the first hours of hospitalization. Because this was a retrospective review, neither institutional board approval nor informed consent were required. MDCTA evaluation and interpretation of the systemic (bronchial and nonbronchial) and pulmonary vascularization were performed using a 16-slice MDCT scanner, as previously detailed.8 According to the MDCTA findings, the site of bleeding Original Research
Figure 2. Patient 1: an 80-year-old woman with recent active tuberculosis treated for 1 month. She was referred to the RICU for massive hemoptysis (⬎ 600 mL at one time). Axial (top left, A) and sagittal (top right, B) 5-mm-thick slab maximum-intensity projection shows the false aneurysm (arrow) and the segment pulmonary artery (arrowhead) with irregular limits. Bottom left, C: superselective right anteroapical segmental pulmonary artery angiogram shows the PAPAs (arrow). However, the global right pulmonary artery and right upper lobe angiograms were normal, without visualization of the pseudoaneurysm. Bottom right, D: superselective angiogram using a microcatheter (arrowheads) of a subsegmental pulmonary artery shows the occlusion of the pseudoaneurysm by two microcoils (arrow). The patient’s condition improved rapidly, and no further bleeding was detected (1-year follow-up).
was assessed as the presence of lung parenchyma consolidation and/or ground-glass opacities. Signs of hemoptysis of pulmonary arterial origin were assessed as pulmonary artery aneurysm (PAA), pulmonary artery pseudoaneurysm (PAPA), air bubble within PAA, or pulmonary artery in the inner aspect of the wall of the cavity in the same area as the site of bleeding. The MDCTA criteria for PAA were focal pulmonary artery dilatation without lung or tumor necrosis, whereas the criteria for PAPA were focal pulmonary artery dilatation within the lung or tumor necrosis. The definitive attribution of the pulmonary artery as the cause of the hemoptysis in patients with pulmonary artery in the inner aspect of the wall cavity was the extravasation of the contrast media during pulmonary artery standard angiography, the recurrence of massive hemoptysis despite optimal embolization of systemic arteries, and/or findings of the pathologic examination. Bronchial arteries and NBSAs were www.chestjournal.org
also evaluated; the bronchial artery diameter was coded as enlarged when it was ⬎ 1.5 mm.6 The pulmonary artery vasoocclusion was achieved in 11 patients by steel coils and in 2 patients by ethylene vinyl alcohol copolymer (Onyx, Embolyx; Micro Therapeutics; Irvine, CA).
Results During the 36-month period, 272 patients were referred to the RICU for hemoptysis; 189 of them (151 men and 48 women; mean age, 57.1 years of age) were treated by endovascular means. Thirteen patients (6.9%; 9 men and 4 women; mean age, 45 years of age) had hemoptysis of pulmonary arterial origin. CHEST / 133 / 1 / JANUARY, 2008
215
Figure 3. Patient 11: a 64-year-old man with massive hemoptysis. Chest radiography examination (not shown) revealed a hilar necrotic mass with disseminated bronchiectasis. Bronchoscopy (not shown) did not locate the bleeding site but showed endobronchial infiltration of the left upper lobe 216
Original Research
Clinical data and appearance on the MDCTA are detailed in Table 1. The diseases underlying hemoptysis of pulmonary arterial origin were active tuberculosis (n ⫽ 4), necrotic bronchial carcinoma (n ⫽ 2), Behc¸et disease (n ⫽ 2), Hughes-Stovin syndrome (n ⫽ 1), mycetoma (n ⫽ 1), invasive aspergillosis (n ⫽ 1), necrotic Hodgkin lymphoma (n ⫽ 1), and lung parenchyma abscess (n ⫽ 1). Signs of hemoptysis of pulmonary arterial origin, as determined by MDCTA, were PAPA (n ⫽ 5), pulmonary artery in the inner wall cavity (n ⫽ 5), multiple PAA associated with pulmonary artery occlusion due to systemic disease (n ⫽ 2), and isolated PAA in the patient with Behc¸et disease (n ⫽ 1). Two of the three patients with PAA had an air bubble within the aneurysm sac. The hypertrophy of bronchial arteries was apparent by MDCTA in 7 of 13 patients, as follows: 1 patient with PAA (Fig 1); 3 patients with PAPA; and 3 patients with the pulmonary artery in the inner wall of the cavity. In patients with PAA or PAPA, pulmonary artery vasoocclusion was performed in six patients with BAE (n ⫽ 1) or without BAE (n ⫽ 5). Two patients were treated by BAE because of the misinterpretation of MDCTA images (one patient in our institute, and one patient who underwent BAE in another hospital before being referred to our hospital). In these latter cases, hemoptysis relapsed 1 week later in one patient and continued in the second patient; the reinterpretation of the MDCTA images showed clearly that both were PAPAs, and they were adequately treated by pulmonary artery vasoocclusion. Of the 18 patients with active tuberculosis, 4 patients had PAPA and 14 patients had hemoptysis from a bronchial arterial origin. The connection between pseudoaneurysm and the pulmonary artery tree was clearly identified in one patient and less clearly in three patients. The latter finding predicted difficulties in catheterizing the pulmonary artery feeding the pseudoaneurysm (Fig 2) despite visualization on MDCTA. In patients with systemic disease, the connection between the aneurysm sac and the pulmonary artery tree was clearly identified by MDCTA, leading to easy pulmonary artery catheterization.
In patients with a pulmonary artery in the inner wall of the cavity, pulmonary artery vasoocclusion was performed in three patients; one procedure was performed 1 week after BAE and hemoptysis recurrence (patient 11) [Table 1, Fig 3] and one procedure was followed by BAE for hemoptysis recurrence 1 week later (patient 8). We have observed contrast media extravasation into the cavity in two patients (patients 8 and 9). Patients 12 and 13, who had a pulmonary artery in the inner wall of the cavity, were treated by BAE and experienced a recurrence of massive hemoptysis. Patient 12 had hemoptysis with a volume of ⬎ 1.5 L, which led to cardiac arrest and suffocation, despite mechanical ventilation. Patient 13 had a massive hemoptysis and experienced a cardiac arrest 1 h before undergoing surgery; he was resuscitated but, despite resection of the right lower lobe, he died 1 week later due to multisystemic failure. Isolated treatment by pulmonary artery vasoocclusion was effective in five patients (all patients had a bronchial artery of normal size). The other three patients underwent pulmonary artery vasoocclusion as a first treatment, which was performed simultaneously with BAE in one patient and completed by BAE 1 week or 6 months later for hemoptysis relapse in two patients. All three patients had bronchial artery enlargement. In four patients, MDCTA identified bronchial artery enlargement and signs of hemoptysis of pulmonary arterial origin, and these patients were treated by BAE as a first treatment. Hemoptysis relapse was observed in all four patients within 1 week. Hemoptysis was stopped in three patients by pulmonary artery vasoocclusion, and it was fatal in the last patient.
Discussion Hemoptysis requires prompt management, and the identification of the site and the cause of the bleeding is essential for proper treatment. Emergency surgery in a patient with active bleeding is associated with high morbidity and mortality,9,10 so interventional radiology is recommended.11 The primary intervention to treat patients with hemoptysis is the embolization of systemic arteries (bronchial artery and/or NBSA). Until
Figure 3. (Continued) bronchus. Top left, A: coronal 1-mm-thick slab on the parenchymal window shows the communication between the tumor cavity and the left upper lobe, and the diffuse bronchiectasis. Top right, B: coronal 3-mm-thick slab maximum-intensity projection slices of MDCTA show a subsegmental pulmonary artery in the inner wall of the cavity (arrow) and enlargement of the bronchial arteries (arrowhead). The patient was treated by BAE. One week later, he was readmitted to the RICU for hemoptysis recurrence (400 mL). A new MDCTA depicted PAPA. Center left, C: oblique coronal 3-mm-thick slab maximum-intensity projection image clearly shows the PAPA (arrow) and microcoil (arrowhead) from a previous treatment session. Center right, D: anterior view of the volume-rendering technique of the pulmonary artery tree shows clearly the architecture and the three-dimensional appearance of the PAPA (arrow). Bottom left, E: subsegmental pulmonary artery angiogram confirms the false aneurysm (arrow). This subsegmental artery was occluded by coils. Hemoptysis did not recur during the 2 years of follow-up. Reproduced with permission of Khalil et al.8 www.chestjournal.org
CHEST / 133 / 1 / JANUARY, 2008
217
recently, the suggested treatment for hemoptysis of pulmonary arterial origin was the generally ineffective embolization of systemic arteries, and the condition was commonly associated with destructive lung disease.12,13 In some cases, a PAPA can be identified during systemic arterial angiography (bronchial artery and/or NBSA angiograms) resulting from flow reversal within the pulmonary artery branches secondary to the systemic artery-to-pulmonary artery shunting or by conventional CT scans.13,14 To our knowledge, the use of MDCTA for the diagnosis of a condition of pulmonary arterial origin has received little attention in the literature, with only a few case reports.15,16 In our series, the signs of pulmonary arterial hemoptysis determined by MDCTA were PAA or PAPA (n ⫽ 8) and pulmonary artery within the inner aspect of a cavity wall (n ⫽ 5). In fact, the MDCTA results depended on the underlying disease. The potential causes of pulmonary arterial hemoptysis are numerous and include all pathologic processes, such as pulmonary necrosis (eg, active tuberculosis, pulmonary abscess, aspergillosis, and necrotic cavitary from lung carcinoma), vasculitis (eg, Behc¸et disease or HughesStovin syndrome), trauma from a by Swan-Ganz catheter, and pulmonary arteriovenous malformation. The pulmonary embolism is not a cause of pulmonary arterial hemoptysis; in fact, the pulmonary embolism provokes a lung ischemia leading to systemic hypervascularization by bronchial arteries. In this case, the direct hemoptysis source is the bronchial artery hypervascularization and not the pulmonary artery injury. The hemoptysis source in the patient with pulmonary emboli is the bronchial arteries as a consequence of the lung ischemia, the bleeding source. In our patients, MDCTA suggested diagnoses of vasculitis (n ⫽ 2), tuberculosis (n ⫽ 3), Aspergillus infections (n ⫽ 2), lung cancer (n ⫽ 1), and Hodgkin lymphoma (n ⫽ 1). In the remaining four patients, the etiologic diagnosis was known before the onset of hemoptysis. The most common forms of vasculitis associated with PAA and hemoptysis are Behc¸et disease and HughesStovin syndrome. MDCTA in such patients identifies clearly the PAA causing hemoptysis by detecting air bubbles in the aneurysm sac (ie, communication between airways and aneurysm sac) and/or lung parenchyma abnormalities related to the hemoptysis as lung parenchyma consolidation and/or ground-glass attenuation surrounding the aneurysm. In such patients, the presence of bronchial artery enlargement is usually observed because they have a long history of pulmonary artery occlusion (pulmonary embolism recurrence or pulmonary artery vasculitis with occlusion). In these patients, the hemoptysis mechanism could be the PAA itself (patients 5 and 6) or the association of PAA and systemic hypervascularization secondary to the chronic 218
pulmonary artery occlusion (patient 7). The treatment in these two situations is different. In the first case, the pulmonary artery is treated by vasoocclusion alone, whereas BAE must be considered in the second case. In patients with tuberculosis, lung abscess, lung tumor, or aspergillosis, the mechanism of hemoptysis of pulmonary arterial origin is the erosion of the pulmonary artery and the formation of a pseudoaneurysm. According to MDCTA findings, the pathophysiology is revealed by the visualization of the pulmonary artery in the inner aspect of the wall cavity followed by the formation of PAPA (Fig 3), as in patient 11. This natural history suggests the possibility of a pulmonary arterial origin of the hemoptysis when MDCTA shows a pulmonary artery in the inner aspect of the wall cavity without finding a PAPA, as occurred in four patients in this series. Contrast media extravasation was observed during vasoocclusion in two patients; pathologic examination confirmed this situation in one patient, and the last patient died from massive hemoptysis. An imperfect MDCTA display of the connection between a PAPA and the pulmonary artery tree predicts the difficulty of endovascular treatment. This situation occurred in three patients with acute tuberculosis. In these cases, the PAPA was identified only by a superselective angiogram of the suspected subsegmental pulmonary artery despite a lack of visualization of the PAPA on the global or lobar angiogram. These difficulties have been highlighted by Sbano et al.13 Conversely, a good display on the MDCTA of this connection15 between the PAPA or PAA and the pulmonary artery tree predicts an easy endovascular treatment; this was the situation in all patients with PAA and in two patients with PAPA. In our institution, MDCTA is used routinely during the first hours after admittance to manage patients who have been referred to the RICU for treatment of hemoptysis. The incidence of hemoptysis of pulmonary arterial origin was 6.9% in this series, which agrees with the results of other studies.13,14 In our series, MDCTA findings influenced disease management and spared patients from undergoing systematic bronchial arteriography. While patients were referred to our center for this procedure, 8 of 13 patients (61.5%) were treated for pulmonary artery involvement during the first session. Furthermore, the presence of bronchial artery enlargement on MDCTA indicated the involvement of systemic arteries in the bleeding mechanism, leading to systemic artery embolization to avoid hemoptysis relapse. In conclusion, thoracic MDCTA should be considered prior to endovascular treatment because it allows for the correct identification and early management of patients with hemoptysis of pulmonary arterial origin, and for the assessment of the eventual Original Research
involvement of systemic arteries with the bleeding pathophysiology. Furthermore, MDCTA guides endovascular treatment by showing the precise bleeding site for accurate superselective vasoocclusion.
9
10
References 1 Crocco JA, Rooney JJ, Fankushen DS, et al. Massive hemoptysis. Arch Intern Med 1968; 121:495– 498 2 Garzon AA, Cerruti M, Gourin A, et al. Pulmonary resection for massive hemoptysis. Surgery 1970; 67:633– 638 3 Garzon AA, Gourin A. Surgical management of massive hemoptysis: a ten-year experience. Ann Surg 1978; 187:267–271 4 Haponik EF, Fein A, Chin R. Managing life-threatening hemoptysis: has anything really changed? Chest 2000; 118: 1431–1435 5 Sanyika C, Corr P, Royston D, et al. Pulmonary angiography and embolization for severe hemoptysis due to cavitary pulmonary tuberculosis. Cardiovasc Intervent Radiol 1999; 22:457– 460 6 Remy-Jardin M, Bouaziz N, Dumont P, et al. Bronchial and nonbronchial systemic arteries at multi-detector row CT angiography: comparison with conventional angiography. Radiology 2004; 233:741–749 7 Chung MJ, Lee JH, Lee KS, et al. Bronchial and nonbronchial systemic arteries in patients with hemoptysis: depiction on MDCT angiography. AJR Am J Roentgenol 2006; 186: 649 – 655 8 Khalil A, Fartoukh M, Tassart M, et al. Role of MDCT in
www.chestjournal.org
11
12
13 14 15
16
identification of the bleeding site and the vessels causing hemoptysis. AJR Am J Roentgenol 2007; 188:W117–W125 McCollun WB, Mattox KL, Guinn GA, et al. Immediate operative treatment for massive hemoptysis. Chest 1975; 67:152–155 Conlan AA, Hurwitz SS, Krige L, et al. Massive hemoptysis: review of 123 cases. J Thorac Cardiovasc Surg 1983; 85:120 – 124 Fartoukh M, Khalil A, Louis L, et al. An integrated approach to diagnosis and management of severe haemoptysis in patients admitted to the intensive care unit: a case series from a referral centre. Respir Res 2007; 8:11–20 Remy-Jardin M, Wattinne L, Remy J. Transcatheter occlusion of pulmonary arterial circulation and collateral supply: failures, incidents, and complications. Radiology 1991; 180: 699 –705 Sbano H, Mitchell AW, Ind PW, et al. Peripheral pulmonary artery pseudoaneurysms and massive hemoptysis. AJR Am J Roentgenol 2005; 184:1253–1259 Remy J, Lemaitre L, Lafitte JJ, et al. Massive hemoptysis of pulmonary arterial origin: diagnosis and treatment. AJR Am J Roentgenol 1984; 143:963–969 Picard C, Parrot A, Boussaud V, et al. Massive hemoptysis due to Rasmussen aneurysm: detection with helicoidal CT angiography and successful steel coil embolization. Intensive Care Med 2003; 29:1837–1839 Kierse R, Jensen U, Helmberger H, et al. Value of multislice CT in the diagnosis of pulmonary artery pseudoaneurysm from Swan-Ganz catheter placement. J Vasc Interv Radiol 2004; 15:1133–1137
CHEST / 133 / 1 / JANUARY, 2008
219
Severe Hemoptysis of Pulmonary Arterial Origin* Signs and Role of Multidetector Row CT Angiography Antoine Khalil, MD; Antoine Parrot, MD; Cosmina Nedelcu, MD; Muriel Fartoukh, MD; Claude Marsault, MD; and Marie-France Carette, MD
Background: Hemoptysis of pulmonary arterial origin is a diagnostic challenge in patients admitted to a respiratory ICU (RICU) for treatment of hemoptysis. Its early accurate recognition and treatment reduce morbidity and prevent mortality. Multidetector row CT angiography (MDCTA) is an accurate method for imaging the systemic vascular network. Our aim was to assess the MDCTA signs and role in managing hemoptysis of pulmonary arterial origin. Methods: We performed a retrospective clinical and radiologic analysis of all consecutive patients who were referred for severe hemoptysis to our RICU and were treated by endovascular means between January 2004 and December 2006. We reviewed all of those cases with hemoptysis of pulmonary arterial origin. Results: Of 272 patients who were referred for severe hemoptysis to the RICU, 189 patients were treated by endovascular means. Thirteen patients (nine men, four women; mean age, 45 years) had hemoptysis of pulmonary arterial origin. Signs of pulmonary arterial hemoptysis seen on MDCTA were of the following three types: pseudoaneurysm (n ⴝ 5); aneurysm of the pulmonary artery (n ⴝ 3); or the presence of a pulmonary artery in the inner wall of a cavity (n ⴝ 5). Hypertrophy of the bronchial arteries seen on MDCTA associated with any of these signs predicted the necessity to treat both the bronchial and pulmonary arteries. Pulmonary artery vasoocclusion was performed as a first treatment in eight patients with such an association (n ⴝ 1) or without such an association (n ⴝ 7) along with bronchial artery embolization. The remaining five patients were treated with systemic artery embolization, followed by surgery (n ⴝ 1), pulmonary artery vasoocclusion (n ⴝ 3), and death from massive hemoptysis (n ⴝ 1). Conclusions: MDCTA performed prior to endovascular treatment allows the correct identification and early appropriate management of severe hemoptysis of pulmonary arterial origin. (CHEST 2008; 133:212–219) Key words: Behc¸et disease; CT scan; hemoptysis; pulmonary artery; tuberculosis Abbreviations: BAE ⫽ bronchial artery embolization; MDCTA ⫽ multidetector CT angiography; NBSA ⫽ nonbronchial systemic artery; PAA ⫽ pulmonary artery aneurysm; PAPA ⫽ pulmonary artery pseudoaneurysm; RICU ⫽ respiratory ICU
hemoptysis is a life-threatening condition M assive that is associated with a mortality rate exceeding 50% in the absence of adequate treatment.1–3 Hemoptysis usually involves bleeding from the bronchial arteries or, less frequently, from nonbronchial systemic arteries (NBSAs). Hemoptysis of pulmonary arterial origin is rare, estimated at ⬍ 10% of hemoptysis cases. Unfortunately, clinical presentation cannot easily distinguish pulmonary and sys212
temic mechanisms, which also can be associated with hemoptysis. The pulmonary etiologies of hemoptysis are numerous. Before the era of multidetector row CT angiography (MDCTA), patient management involved fiberoptic bronchoscopy, bronchial artery embolization (BAE), and, in some cases, surgery.4 Hemoptysis of pulmonary arterial origin was usually suspected when BAE and NBSA embolization were unable to Original Research
Table 1—Clinical Characteristics, MDCTA Findings, and Treatment of Patients With Hemoptysis of Pulmonary Arterial Origin* Patient No./Age, yr/Sex/ Hemoptysis Cumulative Volume, mL 1/80/F/600 2/22/F/200
3/29/M/⬎1000 4/45/M/200
5/27/M/300 6/42/M/600
7/24/M/100
8/82/F/150
9/72/M/100
10/52/M/⬎1000 11/61/M/300
12/49/F/500 13/36/M/100
MDCTA of Pulmonary Arteries PAPA BA of normal size PAPA BA of normal size PAPA BA of normal size PAPA BA and NBSA hypertrophy PAA, contained air bubble BA of normal size Multiple PAAs, RML PAA contained air bubble and was surrounded by an alveolar consolidation BA hypertrophy Multiple PAAs; two of them were surrounded by ground-glass opacities BA hypertrophy Presence of a PA in the inner wall of the cavity BA hypertrophy Presence of a PA in the inner wall of the abscess BA of normal size PAPA in lung abscess BA hypertrophy Subsegmental left upper PA in inner wall cavity BA hypertrophy Left lower PA in inner wall cavity BA hypertrophy Right lower PA in the inner wall cavity BA of normal size
Treatment and Evolution Pulmonary angiography and superselective vasoocclusion Pulmonary angiography and selective vasoocclusion without pseudoaneurysm identification Pulmonary angiography and superselective vasoocclusion BAE relapse hemoptysis, pulmonary angiography, and selective vasoocclusion after opacification of PAPA Pulmonary angiography and superselective PA vasoocclusion Pulmonary angiography and selective vasoocclusion of RML PAA
Definitive Diagnosis Active tuberculosis Active tuberculosis
Active tuberculosis Active tuberculosis
Behçet disease Hughes-Stovin syndrome
BA vasooccluded, and two PAAs of the right lower lobe vasooccluded in the same session
Behçet disease
Subsegmental right upper lobe pulmonary artery vasoocclusion after contrast media extravasation into cavity Hemoptysis relapses 1 wk after BAE Pulmonary angiography and superselective vasoocclusion after contrast media extravasation into cavity BAE, pulmonary angiography, and superselective vasoocclusion BAE, hemoptysis relapses 1 wk later; new MDCTA revealed PAPA, which was vasooccluded BAE, massive hemoptysis relapse
Aspergilloma
BAE without major hypervascularization, patient sent to surgery
Invasive aspergillosis
Lung abscess Bronchial carcinoma excavation Bronchial carcinoma excavation Hodgkin lymphoma excavation
*BA ⫽ bronchial artery; PA ⫽ pulmonary artery; RML ⫽ right middle lobe; F ⫽ female; M ⫽ male.
control hemoptysis.5 More recently, MDCTA of the bronchial arteries and NBSAs have emerged as potential diagnostic tools for evaluating systemic *From the Radiology Department (Drs. Khalil, Nedelcu, Marsault, and Carette) and the Respiratory Intensive Care Unit (Drs. Parrot and Fartoukh), Assistance Publique-Hoˆpitaux de Paris, Tenon Hospital, Paris, France. The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Manuscript received May 16, 2007; revision accepted October 8, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Antoine Khalil, MD, Radiology Department, AP-HP Tenon Hospital, 4 Rue de la Chine, 75020 Paris, France; e-mail: [email protected] DOI: 10.1378/chest.07-1159 www.chestjournal.org
circulation in patients with hemoptysis.6,7 However, a large-scale analysis of the role of MDCTA and its influence in exploring and managing hemoptysis of pulmonary arterial origin has not yet been reported. The aim of this study was to assess the thoracic MDCTA signs and role in patients with severe hemoptysis of pulmonary arterial origin in patients admitted to the ICU for BAE.
Materials and Methods The clinical charts of all patients referred to the respiratory ICU (RICU) of our hospital for hemoptysis and treated by endovascular means (ie, systemic artery embolization of the bronchial artery and NBSA, and pulmonary artery vasoocclusion) during a 36-month period (January 2004 to December 2006) CHEST / 133 / 1 / JANUARY, 2008
213
Figure 1. Patient 5: a 26-year-old man with Behc¸et disease revealed by pulmonary artery thrombosis and a postpuncture left radial artery aneurysm. He was referred to the RICU for hemoptysis of a volume of 100 mL. Top left, A: sagittal 1-mm reformation MDCTA on the parenchymal window shows right lower lobe consolidation (*) related to hemoptysis. Top right, B: sagittal 15-mm-thick slab maximum-intensity projection MDCTA shows two partially thrombosed (arrows) right lower lobe PAAs in area of lung bleeding. Bottom left, C: coronal 20-mm-thick slab maximum-intensity projection MDCTA shows enlargement of the right bronchointercostal trunk (arrow) and its bronchial artery (arrowheads). Bottom right, D: sagittal 5-mm-thick slab maximum-intensity projection MDCTA 3 days after endovascular treatment shows coils into both PAAs (arrows).
were collected. Each patient who was admitted for treatment of hemoptysis in our tertiary university hospital and referral center for hemoptysis underwent chest radiography, fiberoptic bronchoscopy, and MDCTA. The charts of patients with hemoptysis of pulmonary arterial origin were obtained for retrospective review. Hemoptysis of pulmonary arterial origin was assessed based on the following data: MDCTA; pulmonary angiography; histology (when available); and outcome. The cause of hemoptysis was determined from the patient’s history; findings of the physical examination, chest radiography, fiberoptic bronchos214
copy, MDCTA, microbiology, and histology (when available); and outcome. MDCTA was performed for each patient admitted to the RICU for hemoptysis in our institution as a pretreatment examination during the first hours of hospitalization. Because this was a retrospective review, neither institutional board approval nor informed consent were required. MDCTA evaluation and interpretation of the systemic (bronchial and nonbronchial) and pulmonary vascularization were performed using a 16-slice MDCT scanner, as previously detailed.8 According to the MDCTA findings, the site of bleeding Original Research
Figure 2. Patient 1: an 80-year-old woman with recent active tuberculosis treated for 1 month. She was referred to the RICU for massive hemoptysis (⬎ 600 mL at one time). Axial (top left, A) and sagittal (top right, B) 5-mm-thick slab maximum-intensity projection shows the false aneurysm (arrow) and the segment pulmonary artery (arrowhead) with irregular limits. Bottom left, C: superselective right anteroapical segmental pulmonary artery angiogram shows the PAPAs (arrow). However, the global right pulmonary artery and right upper lobe angiograms were normal, without visualization of the pseudoaneurysm. Bottom right, D: superselective angiogram using a microcatheter (arrowheads) of a subsegmental pulmonary artery shows the occlusion of the pseudoaneurysm by two microcoils (arrow). The patient’s condition improved rapidly, and no further bleeding was detected (1-year follow-up).
was assessed as the presence of lung parenchyma consolidation and/or ground-glass opacities. Signs of hemoptysis of pulmonary arterial origin were assessed as pulmonary artery aneurysm (PAA), pulmonary artery pseudoaneurysm (PAPA), air bubble within PAA, or pulmonary artery in the inner aspect of the wall of the cavity in the same area as the site of bleeding. The MDCTA criteria for PAA were focal pulmonary artery dilatation without lung or tumor necrosis, whereas the criteria for PAPA were focal pulmonary artery dilatation within the lung or tumor necrosis. The definitive attribution of the pulmonary artery as the cause of the hemoptysis in patients with pulmonary artery in the inner aspect of the wall cavity was the extravasation of the contrast media during pulmonary artery standard angiography, the recurrence of massive hemoptysis despite optimal embolization of systemic arteries, and/or findings of the pathologic examination. Bronchial arteries and NBSAs were www.chestjournal.org
also evaluated; the bronchial artery diameter was coded as enlarged when it was ⬎ 1.5 mm.6 The pulmonary artery vasoocclusion was achieved in 11 patients by steel coils and in 2 patients by ethylene vinyl alcohol copolymer (Onyx, Embolyx; Micro Therapeutics; Irvine, CA).
Results During the 36-month period, 272 patients were referred to the RICU for hemoptysis; 189 of them (151 men and 48 women; mean age, 57.1 years of age) were treated by endovascular means. Thirteen patients (6.9%; 9 men and 4 women; mean age, 45 years of age) had hemoptysis of pulmonary arterial origin. CHEST / 133 / 1 / JANUARY, 2008
215
Figure 3. Patient 11: a 64-year-old man with massive hemoptysis. Chest radiography examination (not shown) revealed a hilar necrotic mass with disseminated bronchiectasis. Bronchoscopy (not shown) did not locate the bleeding site but showed endobronchial infiltration of the left upper lobe 216
Original Research
Clinical data and appearance on the MDCTA are detailed in Table 1. The diseases underlying hemoptysis of pulmonary arterial origin were active tuberculosis (n ⫽ 4), necrotic bronchial carcinoma (n ⫽ 2), Behc¸et disease (n ⫽ 2), Hughes-Stovin syndrome (n ⫽ 1), mycetoma (n ⫽ 1), invasive aspergillosis (n ⫽ 1), necrotic Hodgkin lymphoma (n ⫽ 1), and lung parenchyma abscess (n ⫽ 1). Signs of hemoptysis of pulmonary arterial origin, as determined by MDCTA, were PAPA (n ⫽ 5), pulmonary artery in the inner wall cavity (n ⫽ 5), multiple PAA associated with pulmonary artery occlusion due to systemic disease (n ⫽ 2), and isolated PAA in the patient with Behc¸et disease (n ⫽ 1). Two of the three patients with PAA had an air bubble within the aneurysm sac. The hypertrophy of bronchial arteries was apparent by MDCTA in 7 of 13 patients, as follows: 1 patient with PAA (Fig 1); 3 patients with PAPA; and 3 patients with the pulmonary artery in the inner wall of the cavity. In patients with PAA or PAPA, pulmonary artery vasoocclusion was performed in six patients with BAE (n ⫽ 1) or without BAE (n ⫽ 5). Two patients were treated by BAE because of the misinterpretation of MDCTA images (one patient in our institute, and one patient who underwent BAE in another hospital before being referred to our hospital). In these latter cases, hemoptysis relapsed 1 week later in one patient and continued in the second patient; the reinterpretation of the MDCTA images showed clearly that both were PAPAs, and they were adequately treated by pulmonary artery vasoocclusion. Of the 18 patients with active tuberculosis, 4 patients had PAPA and 14 patients had hemoptysis from a bronchial arterial origin. The connection between pseudoaneurysm and the pulmonary artery tree was clearly identified in one patient and less clearly in three patients. The latter finding predicted difficulties in catheterizing the pulmonary artery feeding the pseudoaneurysm (Fig 2) despite visualization on MDCTA. In patients with systemic disease, the connection between the aneurysm sac and the pulmonary artery tree was clearly identified by MDCTA, leading to easy pulmonary artery catheterization.
In patients with a pulmonary artery in the inner wall of the cavity, pulmonary artery vasoocclusion was performed in three patients; one procedure was performed 1 week after BAE and hemoptysis recurrence (patient 11) [Table 1, Fig 3] and one procedure was followed by BAE for hemoptysis recurrence 1 week later (patient 8). We have observed contrast media extravasation into the cavity in two patients (patients 8 and 9). Patients 12 and 13, who had a pulmonary artery in the inner wall of the cavity, were treated by BAE and experienced a recurrence of massive hemoptysis. Patient 12 had hemoptysis with a volume of ⬎ 1.5 L, which led to cardiac arrest and suffocation, despite mechanical ventilation. Patient 13 had a massive hemoptysis and experienced a cardiac arrest 1 h before undergoing surgery; he was resuscitated but, despite resection of the right lower lobe, he died 1 week later due to multisystemic failure. Isolated treatment by pulmonary artery vasoocclusion was effective in five patients (all patients had a bronchial artery of normal size). The other three patients underwent pulmonary artery vasoocclusion as a first treatment, which was performed simultaneously with BAE in one patient and completed by BAE 1 week or 6 months later for hemoptysis relapse in two patients. All three patients had bronchial artery enlargement. In four patients, MDCTA identified bronchial artery enlargement and signs of hemoptysis of pulmonary arterial origin, and these patients were treated by BAE as a first treatment. Hemoptysis relapse was observed in all four patients within 1 week. Hemoptysis was stopped in three patients by pulmonary artery vasoocclusion, and it was fatal in the last patient.
Discussion Hemoptysis requires prompt management, and the identification of the site and the cause of the bleeding is essential for proper treatment. Emergency surgery in a patient with active bleeding is associated with high morbidity and mortality,9,10 so interventional radiology is recommended.11 The primary intervention to treat patients with hemoptysis is the embolization of systemic arteries (bronchial artery and/or NBSA). Until
Figure 3. (Continued) bronchus. Top left, A: coronal 1-mm-thick slab on the parenchymal window shows the communication between the tumor cavity and the left upper lobe, and the diffuse bronchiectasis. Top right, B: coronal 3-mm-thick slab maximum-intensity projection slices of MDCTA show a subsegmental pulmonary artery in the inner wall of the cavity (arrow) and enlargement of the bronchial arteries (arrowhead). The patient was treated by BAE. One week later, he was readmitted to the RICU for hemoptysis recurrence (400 mL). A new MDCTA depicted PAPA. Center left, C: oblique coronal 3-mm-thick slab maximum-intensity projection image clearly shows the PAPA (arrow) and microcoil (arrowhead) from a previous treatment session. Center right, D: anterior view of the volume-rendering technique of the pulmonary artery tree shows clearly the architecture and the three-dimensional appearance of the PAPA (arrow). Bottom left, E: subsegmental pulmonary artery angiogram confirms the false aneurysm (arrow). This subsegmental artery was occluded by coils. Hemoptysis did not recur during the 2 years of follow-up. Reproduced with permission of Khalil et al.8 www.chestjournal.org
CHEST / 133 / 1 / JANUARY, 2008
217
recently, the suggested treatment for hemoptysis of pulmonary arterial origin was the generally ineffective embolization of systemic arteries, and the condition was commonly associated with destructive lung disease.12,13 In some cases, a PAPA can be identified during systemic arterial angiography (bronchial artery and/or NBSA angiograms) resulting from flow reversal within the pulmonary artery branches secondary to the systemic artery-to-pulmonary artery shunting or by conventional CT scans.13,14 To our knowledge, the use of MDCTA for the diagnosis of a condition of pulmonary arterial origin has received little attention in the literature, with only a few case reports.15,16 In our series, the signs of pulmonary arterial hemoptysis determined by MDCTA were PAA or PAPA (n ⫽ 8) and pulmonary artery within the inner aspect of a cavity wall (n ⫽ 5). In fact, the MDCTA results depended on the underlying disease. The potential causes of pulmonary arterial hemoptysis are numerous and include all pathologic processes, such as pulmonary necrosis (eg, active tuberculosis, pulmonary abscess, aspergillosis, and necrotic cavitary from lung carcinoma), vasculitis (eg, Behc¸et disease or HughesStovin syndrome), trauma from a by Swan-Ganz catheter, and pulmonary arteriovenous malformation. The pulmonary embolism is not a cause of pulmonary arterial hemoptysis; in fact, the pulmonary embolism provokes a lung ischemia leading to systemic hypervascularization by bronchial arteries. In this case, the direct hemoptysis source is the bronchial artery hypervascularization and not the pulmonary artery injury. The hemoptysis source in the patient with pulmonary emboli is the bronchial arteries as a consequence of the lung ischemia, the bleeding source. In our patients, MDCTA suggested diagnoses of vasculitis (n ⫽ 2), tuberculosis (n ⫽ 3), Aspergillus infections (n ⫽ 2), lung cancer (n ⫽ 1), and Hodgkin lymphoma (n ⫽ 1). In the remaining four patients, the etiologic diagnosis was known before the onset of hemoptysis. The most common forms of vasculitis associated with PAA and hemoptysis are Behc¸et disease and HughesStovin syndrome. MDCTA in such patients identifies clearly the PAA causing hemoptysis by detecting air bubbles in the aneurysm sac (ie, communication between airways and aneurysm sac) and/or lung parenchyma abnormalities related to the hemoptysis as lung parenchyma consolidation and/or ground-glass attenuation surrounding the aneurysm. In such patients, the presence of bronchial artery enlargement is usually observed because they have a long history of pulmonary artery occlusion (pulmonary embolism recurrence or pulmonary artery vasculitis with occlusion). In these patients, the hemoptysis mechanism could be the PAA itself (patients 5 and 6) or the association of PAA and systemic hypervascularization secondary to the chronic 218
pulmonary artery occlusion (patient 7). The treatment in these two situations is different. In the first case, the pulmonary artery is treated by vasoocclusion alone, whereas BAE must be considered in the second case. In patients with tuberculosis, lung abscess, lung tumor, or aspergillosis, the mechanism of hemoptysis of pulmonary arterial origin is the erosion of the pulmonary artery and the formation of a pseudoaneurysm. According to MDCTA findings, the pathophysiology is revealed by the visualization of the pulmonary artery in the inner aspect of the wall cavity followed by the formation of PAPA (Fig 3), as in patient 11. This natural history suggests the possibility of a pulmonary arterial origin of the hemoptysis when MDCTA shows a pulmonary artery in the inner aspect of the wall cavity without finding a PAPA, as occurred in four patients in this series. Contrast media extravasation was observed during vasoocclusion in two patients; pathologic examination confirmed this situation in one patient, and the last patient died from massive hemoptysis. An imperfect MDCTA display of the connection between a PAPA and the pulmonary artery tree predicts the difficulty of endovascular treatment. This situation occurred in three patients with acute tuberculosis. In these cases, the PAPA was identified only by a superselective angiogram of the suspected subsegmental pulmonary artery despite a lack of visualization of the PAPA on the global or lobar angiogram. These difficulties have been highlighted by Sbano et al.13 Conversely, a good display on the MDCTA of this connection15 between the PAPA or PAA and the pulmonary artery tree predicts an easy endovascular treatment; this was the situation in all patients with PAA and in two patients with PAPA. In our institution, MDCTA is used routinely during the first hours after admittance to manage patients who have been referred to the RICU for treatment of hemoptysis. The incidence of hemoptysis of pulmonary arterial origin was 6.9% in this series, which agrees with the results of other studies.13,14 In our series, MDCTA findings influenced disease management and spared patients from undergoing systematic bronchial arteriography. While patients were referred to our center for this procedure, 8 of 13 patients (61.5%) were treated for pulmonary artery involvement during the first session. Furthermore, the presence of bronchial artery enlargement on MDCTA indicated the involvement of systemic arteries in the bleeding mechanism, leading to systemic artery embolization to avoid hemoptysis relapse. In conclusion, thoracic MDCTA should be considered prior to endovascular treatment because it allows for the correct identification and early management of patients with hemoptysis of pulmonary arterial origin, and for the assessment of the eventual Original Research
involvement of systemic arteries with the bleeding pathophysiology. Furthermore, MDCTA guides endovascular treatment by showing the precise bleeding site for accurate superselective vasoocclusion.
9
10
References 1 Crocco JA, Rooney JJ, Fankushen DS, et al. Massive hemoptysis. Arch Intern Med 1968; 121:495– 498 2 Garzon AA, Cerruti M, Gourin A, et al. Pulmonary resection for massive hemoptysis. Surgery 1970; 67:633– 638 3 Garzon AA, Gourin A. Surgical management of massive hemoptysis: a ten-year experience. Ann Surg 1978; 187:267–271 4 Haponik EF, Fein A, Chin R. Managing life-threatening hemoptysis: has anything really changed? Chest 2000; 118: 1431–1435 5 Sanyika C, Corr P, Royston D, et al. Pulmonary angiography and embolization for severe hemoptysis due to cavitary pulmonary tuberculosis. Cardiovasc Intervent Radiol 1999; 22:457– 460 6 Remy-Jardin M, Bouaziz N, Dumont P, et al. Bronchial and nonbronchial systemic arteries at multi-detector row CT angiography: comparison with conventional angiography. Radiology 2004; 233:741–749 7 Chung MJ, Lee JH, Lee KS, et al. Bronchial and nonbronchial systemic arteries in patients with hemoptysis: depiction on MDCT angiography. AJR Am J Roentgenol 2006; 186: 649 – 655 8 Khalil A, Fartoukh M, Tassart M, et al. Role of MDCT in
www.chestjournal.org
11
12
13 14 15
16
identification of the bleeding site and the vessels causing hemoptysis. AJR Am J Roentgenol 2007; 188:W117–W125 McCollun WB, Mattox KL, Guinn GA, et al. Immediate operative treatment for massive hemoptysis. Chest 1975; 67:152–155 Conlan AA, Hurwitz SS, Krige L, et al. Massive hemoptysis: review of 123 cases. J Thorac Cardiovasc Surg 1983; 85:120 – 124 Fartoukh M, Khalil A, Louis L, et al. An integrated approach to diagnosis and management of severe haemoptysis in patients admitted to the intensive care unit: a case series from a referral centre. Respir Res 2007; 8:11–20 Remy-Jardin M, Wattinne L, Remy J. Transcatheter occlusion of pulmonary arterial circulation and collateral supply: failures, incidents, and complications. Radiology 1991; 180: 699 –705 Sbano H, Mitchell AW, Ind PW, et al. Peripheral pulmonary artery pseudoaneurysms and massive hemoptysis. AJR Am J Roentgenol 2005; 184:1253–1259 Remy J, Lemaitre L, Lafitte JJ, et al. Massive hemoptysis of pulmonary arterial origin: diagnosis and treatment. AJR Am J Roentgenol 1984; 143:963–969 Picard C, Parrot A, Boussaud V, et al. Massive hemoptysis due to Rasmussen aneurysm: detection with helicoidal CT angiography and successful steel coil embolization. Intensive Care Med 2003; 29:1837–1839 Kierse R, Jensen U, Helmberger H, et al. Value of multislice CT in the diagnosis of pulmonary artery pseudoaneurysm from Swan-Ganz catheter placement. J Vasc Interv Radiol 2004; 15:1133–1137
CHEST / 133 / 1 / JANUARY, 2008
219