|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Respiratory Medicine, Department of Medicine (Drs. Faughnan and Hyland), and the Department of Medical Imaging (Dr. Pugash), St. Michaels Hospital, University of Toronto, Toronto, Ontario, Canada; the Department of Diagnostic and Interventional Radiology (Drs. Lui and White), Yale University School of Medicine, New Haven, CT; the Division of Pulmonary Medicine (Dr. Wirth), Maine Medical Center, Portland, ME; and Department of Medicine, Sunnybrook Health Sciences Centre (Dr. Redelmeier), University of Toronto, Toronto, Canada.
Correspondence to: Robert H. Hyland, MD, FCCP, St. Michaels Hospital, 30 Bond St, Toronto, Canada, M5B1W8; e-mail: hylandb{at}smh.toronto.on.ca
| Abstract |
|---|
|
|
|---|
Design: Retrospective chart review of all patients (n = 16) with diffuse pulmonary AVMs seen at Yale New Haven Hospital, Johns Hopkins Hospital, and St. Michaels Hospital. Up-to-date follow-up information was obtained in all living patients.
Results: All patients were severely hypoxic. Neurologic complications (stroke or brain abscess) had occurred in 70% of patients by the time of diagnosis. During the follow-up period (mean, 6 years), three patients died and two others developed new neurologic complications. One of the deaths occurred perioperatively during lung transplantation. All patients underwent transcatheter embolotherapy of any large pulmonary AVMs. A selected group underwent pulmonary flow redistribution, a novel technique. Oxygenation did not improve significantly with embolotherapy of the larger AVMs, but there was a small significant improvement in those patients who underwent pulmonary flow redistribution. The majority (85%) of the living patients are currently working or studying full-time.
Conclusions: Patients with diffuse pulmonary AVMs are at increased risk of neurologic complications. Transcatheter embolotherapy does not significantly improve the profound hypoxia, but it may reduce the risk of neurologic complications. Antibiotic prophylaxis is recommended for bacteremic procedures to prevent brain abscess. These patients can live for many years and lead productive lives. We do not recommend lung transplantation because survival with disease is difficult to predict and we have observed a perioperative transplant death.
Key Words: arteriovenous malformation brain abscess hereditary hemorrhagic telangiectasia pulmonary telangiectasia
| Introduction |
|---|
|
|
|---|
3 mm.1
The topic of
pulmonary AVMs has recently been extensively reviewed.4 Some patients with pulmonary AVMs have a more severe and diffuse pattern of disease. Little is known about the clinical characteristics and course of these patients, let alone how they should be managed, because only case reports (a total of 30 patients) appear in the literature,1 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 once cases of hepatopulmonary syndrome are excluded. These patients generally present with exercise intolerance and profound cyanosis and appear to have a poor prognosis. Several (9/30) of the cases were reported to be secondary to hereditary hemorrhagic telangiectasia (HHT). Two cases were related to the polysplenia syndrome. Fatal outcomes were reported in 5 of 30 cases, although in general, the follow-up was brief. Little is known about the risk of hemorrhagic and neurologic complications in these patients. Since occlusion or resection of all AVMs in these patients is not realistic, most of the reported patients have gone untreated, and the prognosis is believed to be poor. Some researchers have suggested that lung transplantation should be considered.12
We reviewed 16 patients with diffuse pulmonary AVMs to assess the natural history in this group, particularly with regard to the risk of neurologic complications. We also wanted to assess whether appropriate management, with the aim of preventing neurologic complications and improving hypoxemia and exercise intolerance, would improve the prognosis.
| Materials and Methods |
|---|
|
|
|---|
|
|
All patients were recommended to obtain antibiotic prophylaxis for dental work and other bacteremic procedures. TCE of large pulmonary AVMs (Fig 3 ) was performed with the goal of preventing paradoxical embolus and stroke or TIA, using the technique previously described.29 This sometimes required more than one embolization session. Pulmonary flow redistribution (PFR) was developed as an approach to improve hypoxia, as previously described.30 The patients first underwent a temporary occlusion of lobar arteries (Fig 4 ) of the most affected lobes (usually both lower lobes) to determine if there was any improvement in oxygenation. In those patients whose PaO2 increased by at least 10 mm Hg, a permanent lobar artery occlusion (Fig 5 ) was performed.
|
|
|
| Results |
|---|
|
|
|---|
|
|
|
Most of the patients (10/13) reported no persistent improvement in dyspnea with either TCE or PFR, when asked retrospectively. The current average dyspnea score in the 13 living patients is 0.8 (Table 4 ). Eleven of the 13 were working or studying full-time at follow-up. Though some of the patients have desk jobs, one was doing manual labor, one is a paramedic, and one is a nurse.
|
Three deaths occurred. The 2-year survival rate was 91% (10/11). Patient 4 died of massive hemoptysis 5 years after she was first assessed. She had not undergone any TCE because she had only small diffuse AVMs. She was the first to undergo temporary lobar occlusion; although she had a rise in PaO2 of 10 mm Hg, at the time this was not judged to be significant enough to warrant PFR. Patient 6 underwent lung transplantation at another center. His intraoperative course was complicated by significant blood loss (which necessitated cardiopulmonary bypass), hypotension, shock, and myocardial ischemia. Despite postoperative extracorporeal membrane oxygenation, he arrested several times and eventually died of multiorgan failure. Patient 14 died at the age of 3 years of congestive heart failure. She had no AVMs large enough for TCE, and her PaO2 did not improve significantly with temporary lobar occlusions.
Four of the women successfully carried pregnancies to term. Two of these women (patients 7 and 11) delivered without complication before diffuse pulmonary AVMs were diagnosed; therefore, we do not know what their PaO2 measurements were in retrospect. Patient 16 had four miscarriages before unilateral but diffuse pulmonary AVMs were diagnosed. Her PaO2 rose to normal levels after PFR. She became pregnant soon after PFR and delivered a normal child at term with no complications. Patient 9 became pregnant, with a PaO2 ranging from 50 to 60 mm Hg. She became increasingly dyspneic at 32 weeks of pregnancy and had significant peripheral edema. Investigation did not reveal any evidence of congestive heart failure or worsening hypoxia. Her worsening dyspnea was thought to be due to chest wall restriction late in pregnancy. Also, the fetus was noted to have intrauterine growth retardation; therefore, a cesarean section was performed at 33 weeks. The neonate was small for its gestational age but was otherwise well.
| Discussion |
|---|
|
|
|---|
We have reviewed our 16 patients with diffuse pulmonary AVMs, describing their clinical characteristics and course with 101 patient-years of follow-up, The majority of our patients with diffuse pulmonary AVMs have HHT, similar to patients with discrete AVMs. The majority had a neurologic complication from pulmonary AVMs by the time of diagnosis, unlike patients with single or multiple discrete pulmonary AVMs. Approximately 40% of patients with pulmonary AVMs will have had a neurologic complication.1 It is not surprising that the prevalence of neurologic complications is higher in this group, since they may have hundreds of AVMs. Strokes and TIAs are likely caused by paradoxical embolization through the AVM. The pathophysiology of brain abscess is probably different. Abscess formation may result from seeding of bacteria onto areas of encephalomalacia.31 The encephalomalacia may be caused in turn by previous microemboli, hypoxia, or sludging with polycythemia. Whatever the cause, these patients have a higher prevalence of brain abscess (38%) than reported in the literature (9%) for discrete pulmonary AVMs.1 As in other series, half of the patients who had a brain abscess had visited the dentist recently. Since both stroke and brain abscess are more frequent in these patients, transcatheter occlusion of the AVMs and antibiotic prophylaxis for bacteremic procedures are justified.
TCE of the large AVMs was performed in all patients. There was no statistically significant improvement in oxygenation after TCE of the large AVMs, which is not surprising since the many tiny diffuse AVMs were unaffected. PFR, a novel approach, was attempted to significantly improve oxygenation. This involved occlusion of lobar arteries to the most severely affected lobes (often the lower lobes), so that pulmonary flow would be redistributed to less-affected areas. Immediately after PFR, there was a small but statistically significant rise in PaO2. Of note, these patients had first undergone TCE of the larger AVMs, with no significant change in PaO2; therefore, the rise in PaO2 was not secondary to the occlusion of the larger AVMs. The fact that the rise in PaO2 is small reflects the diffuse involvement in this disease. In the two patients who had diffuse disease only unilaterally, the rise in PaO2 was much greater, indicating that redistribution is technically possible. However, in the remainder of the patients, the disease is too diffuse for redistribution of blood flow to significantly alter the shunt. Although we thought a small improvement in PaO2 would significantly improve patients dyspnea, most of the patients did not report any improvement with PFR.
The patients had few serious neurologic events once they were treated. We propose that this is because the larger AVMs had been occluded, thereby reducing the risk of significant cerebral embolization. Despite 101 patient-years of follow-up, only one new brain abscess occurred, which would suggest that the combination of antibiotic prophylaxis for bacteremic procedures and occlusion of the larger AVMs is effective in preventing brain abscess. Since brain abscess has a mortality of up to 24%,32 33 34 35 this is an important event to prevent.
Neurologic complications, rather than pulmonary complications, are the rule in this patient population. None of these patients developed spontaneous hemothorax. Two patients had minor hemoptysis; one of the two was shown to have diffuse tracheobronchial telangiectases, which have been previously described in HHT. One patient had massive hemoptysis due to bronchial artery hypertrophy, and one patient developed rib notching after PFR. We do not know whether bronchial artery hypertrophy is part of the disease process or whether it developed subsequent to PFR. Reports in the literature suggest that bronchial artery hypertrophy can occur near untreated pulmonary AVMs36 or after TCE of pulmonary AVMs.37 None of the patients who were successfully treated for focal AVMs have ever developed late hemoptysis following TCE.38 We were surprised by how well these patients function in their daily lives. Few complain of more than mild dyspnea, and the unemployment rate is less than that of the general population. Adaptation to chronic hypoxia is not very well understood. The majority of these patients have polycythemia, which likely helps them to compensate for hypoxia. Those who are not polycythemic have anemia secondary to chronic epistaxis. Pulmonary artery pressures are normal or even reduced in these patients, presumably because of the decreased pulmonary vascular resistance secondary to the diffuse AVMs. Some might expect these patients to develop pulmonary hypertension over time, secondary to the chronic hypoxia; however, these patients do not have alveolar hypoxia. Since hypoxic vasoconstriction occurs secondary to alveolar hypoxia rather than hypoxemia, it is not surprising that these patients do not develop pulmonary hypertension.
Pregnancy in the hypoxic patient is possible, but the rate of fetal loss is increased. This has been best documented in patients with cyanotic heart disease, where only 59% of pregnancies, in women who have not had corrective surgery, have normal fetal outcomes.39 Two of our patients had a total of five normal pregnancies, years before pulmonary AVMs were diagnosed. In the one cyanotic patient who became pregnant during the follow-up period, intrauterine growth retardation developed, which was likely secondary to fetal hypoxia. As patient 16 exemplifies, hypoxia can also cause infertility. She had multiple miscarriages until her hypoxia was corrected, and then she was able to conceive and have a normal pregnancy.
When confronted with a patient with diffuse pulmonary AVMs, a clinician might incorrectly conclude that such a patient has a poor prognosis and should be referred for lung transplantation. However, the 2-year survival rate in this group (91%) is better than that seen after lung transplantation (63%)40 ; therefore, referral for lung transplantation does not seem justified. In fact, one of the three deaths in our group occurred after lung transplantation. Survival is likely underestimated in our series because these patients have been referred to tertiary care centers and therefore are likely to be the more severe and symptomatic cases. We have not found any other reports in the literature of lung transplantation for this disease. There may be a time for lung transplantation in these patients, but we have not yet found a marker of disease progression to enable us to determine the most appropriate timing.
| Conclusion |
|---|
|
|
|---|
| Footnotes |
|---|
This work was supported by a fellowship (for Dr. Faughnan) from the Canadian Lung Association/Glaxo/Medical Research Council of Canada, as well as funding from the Nelson Arthur Hyland Foundation. Dr. Robert I. White, Jr. is supported in part by a grant from the Josephine Laurence Hopkins Foundation.
Received for publication December 10, 1998. Accepted for publication July 15, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Pierucci, J. Murphy, K. J. Henderson, D. A. Chyun, and R. I. White Jr New Definition and Natural History of Patients With Diffuse Pulmonary Arteriovenous Malformations: Twenty-Seven-Year Experience Chest, March 1, 2008; 133(3): 653 - 661. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Mei-Zahav, M. Letarte, M. E. Faughnan, S. A. Abdalla, U. Cymerman, and I. B. MacLusky Symptomatic Children With Hereditary Hemorrhagic Telangiectasia: A Pediatric Center Experience Arch Pediatr Adolesc Med, June 1, 2006; 160(6): 596 - 601. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Maimon, R. P. Chan, M. D. Cusimano, and M. E. Faughnan A 50-Year-Old Man With Clubbing Has Seizures After a Dental Procedure Chest, November 1, 2005; 128(5): 3634 - 3638. [Full Text] [PDF] |
||||
![]() |
S Koppen, C R W Korver, M Dalinghaus, and C J J Westermann Neonatal pulmonary arteriovenous malformation in hereditary haemorrhagic telangiectasia Arch. Dis. Child. Fetal Neonatal Ed., November 1, 2002; 87(3): F226 - 227. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Engelke, C. Schaefer-Prokop, E. Schirg, J. Freihorst, S. Grubnic, and M. Prokop High-Resolution CT and CT Angiography of Peripheral Pulmonary Vascular Disorders RadioGraphics, July 1, 2002; 22(4): 739 - 764. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-P. Dinkel and J. Triller Pulmonary Arteriovenous Malformations: Embolotherapy with Superselective Coaxial Catheter Placement and Filling of Venous Sac with Guglielmi Detachable Coils Radiology, June 1, 2002; 223(3): 709 - 714. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. J. Grau, S. O. Schoenberg, C. Lichy, F. Buggle, M. Bock, and W. Hacke Lack of Evidence for Pulmonary Venous Thrombosis in Cryptogenic Stroke: A Magnetic Resonance Angiography Study Stroke, May 1, 2002; 33(5): 1416 - 1419. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Moussouttas, P. Fayad, M. Rosenblatt, M. Hashimoto, J. Pollak, K. Henderson, T. Y.- Z. Ma, and R. I. White Pulmonary arteriovenous malformations: Cerebral ischemia and neurologic manifestations Neurology, October 10, 2000; 55(7): 959 - 964. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |