Search for the Complication
http://www.100md.com
《新英格兰医药杂志》
In this Journal feature, information about a real patient is presented in stages (boldface type) to an expert clinician, who responds to the information, sharing his or her reasoning with the reader (regular type). The authors' commentary follows.
A 58-year-old woman was hospitalized for the evaluation of prolonged fever and hemoptysis. She reported an eight-month history of intermittent fevers, a productive cough, shortness of breath, and hemoptysis, which had been evaluated at another hospital. Computed tomographic (CT) scans of the chest revealed peripheral infiltrates in the upper lobe of the left lung and the lingula and a calcified left hilar opacity with additional, small mediastinal lymph nodes. Bronchoscopy demonstrated hyperemic bronchi. A transbronchial biopsy showed fibrinous material in some alveolar spaces, with a few scattered neutrophils, and capillary congestion in portions of the septa. Culture of a bronchial-lavage specimen was positive for a nontuberculous, slow-growing mycobacterium. Staining and culturing were negative for Mycobacterium tuberculosis. A course of amoxicillin–clavulanic acid and a subsequent course of cefuroxime resulted in no apparent benefit. A tuberculin skin test with purified protein derivative (5 TU) was positive, with an induration of 30 mm.
The combination of prolonged fever, unilateral pulmonary infiltrates, hemoptysis, calcified hilar lymph nodes, and a positive tuberculin skin test is highly suggestive of active tuberculosis. Although staining and culturing of a bronchoalveolar-lavage specimen were reportedly negative for M. tuberculosis, such results do not effectively rule out this diagnosis; smears of bronchoalveolar fluid are reportedly negative in almost half the cases, and cultures are negative in a quarter of the cases. It would be prudent to have three sputum specimens reexamined for M. tuberculosis despite the negative bronchoscopic findings. Respiratory precautions are needed until active tuberculosis is ruled out. In the absence of underlying immunosuppression or appreciable prior chronic lung disease, the slow-growing mycobacterium on a single culture may represent a clinically insignificant contaminant, but further evaluation is needed.
Ten days before the current admission, the patient consulted another pulmonologist, who prescribed methylprednisolone (40 mg per day) and isoniazid (300 mg per day). At presentation to our medical center, she reported increased cough, hemoptysis, pleuritic chest pain, and fever. The patient had a history of hypertension, chronic atrial fibrillation, inflammatory bowel disease, and hypothyroidism that was attributed to the use of amiodarone. She had undergone ablation therapy for atrial fibrillation one year earlier, after which the amiodarone had been stopped. She did not smoke or drink. Her medications on admission were 5-aminosalicylic acid, pravastatin, and levothyroxine.
Corticosteroid treatment could be beneficial for inflammatory lung disease, but without a specific diagnosis, especially if active tuberculosis is present, the use of this therapy arouses concern. The patient's clinical condition worsened after a few days of treatment with isoniazid and corticosteroids. Although fever is a well-known side effect of isoniazid therapy, the use of corticosteroids usually prevents it. I am concerned that the worsening clinical manifestations are related to her underlying disease. Amiodarone is a well-recognized cause of pulmonary complications, which may be manifested by organizing pneumonia. However, such complications usually develop during active treatment or soon after its cessation, and thus, they are unlikely to explain the current presentation.
Bronchiolitis obliterans is an infrequent but well-known manifestation of inflammatory bowel disease. The clinical presentation often mimics that of community-acquired pneumonia, but no improvement results with antibiotic therapy. Typically, the lung opacities are diffuse and bilateral and pleural effusion is infrequent. In rare cases, 5-aminosalicylic acid has been linked to adverse pulmonary effects, including diffuse or basilar infiltrates and bronchiolitis obliterans. I would discontinue this drug, pending further evaluation. Recurrent pulmonary thromboembolism must also be considered.
On examination, the patient appeared well nourished and in no distress. Her temperature was 39°C, and she had a pulse of 80 beats per minute, a blood pressure of 119/74 mm Hg, a respiratory rate of 18 breaths per minute, and an oxygen saturation of hemoglobin of 97 percent while she was breathing room air. Auscultation of the chest revealed no abnormalities except for scattered rales in the lower lobe of the left lung. Examination of the heart revealed no abnormalities. There was no organomegaly or abdominal tenderness. The hemoglobin level was 12.4 g per deciliter, and the white-cell count was 18,440 per cubic millimeter, with 76 percent neutrophils, 15 percent lymphocytes, 5 percent monocytes, 1.5 percent eosinophils, and 2.5 percent basophils; the platelet count was 470,000 per cubic millimeter. The erythrocyte sedimentation rate was 60 mm per hour. The C-reactive protein level was 10 mg per deciliter (normal range, 0 to 0.5). Levels of hepatic enzymes and renal function were within normal limits. A radiograph of the chest showed alveolar infiltrates in the left lateral hemithorax.
The fever and the laboratory results suggest the presence of an inflammatory reaction. The marked elevation in the white-cell count may be due to the administration of corticosteroids, but a superimposed bacterial infection should be considered. Blood and sputum should be obtained for culture, and empirical antibiotic treatment is a reasonable option. Noninfectious inflammatory lung diseases should also be considered, especially systemic lupus erythematosus, Wegener's granulomatosis, and Goodpasture's syndrome. Tests for the presence of antinuclear antibodies, antineutrophil cytoplasmic antibodies, and antiglomerular basement-membrane antibodies, as well as bronchoscopic examination, should be repeated.
Ciprofloxacin and clarithromycin were begun, and orally administered corticosteroids were continued. The 5-aminosalicylic acid was stopped. The patient's temperature gradually became normal, and the hemoptysis decreased substantially. Cultures of blood and sputum yielded no growth. Staining of three sputum specimens was negative for acid-fast bacilli. The antinuclear-antibody titer was positive at 1:160; tests for antibodies against double-stranded DNA, antibodies against glomerular basement membrane, and antineutrophil cytoplasmic antibodies were negative. On pulmonary-function testing, the forced vital capacity (FVC), the forced expiratory volume in one second (FEV1), and the FEV1:FVC ratio were normal; the FEV between 25 and 75 percent of FVC was slightly reduced (60 percent of the predicted value); the lung volumes were normal; and there was a borderline reduction in diffusing capacity. Bronchoscopy revealed slight bleeding that appeared to be of lingular origin. Gram's stain and acid-fast stains of bronchoalveolar-lavage fluid were negative for bacteria, as were cultures. The transbronchial biopsy showed interstitial changes of subacute lung injury with hemosiderin-laden alveolar macrophages.
The histologic findings are inconsistent with the presence of an infectious disease in the lung. Instead, they suggest alveolar hemorrhage, although the bronchoscopist did not report the characteristic return of continuously bloody lavage fluid. Alveolar hemorrhage generally appears on chest radiographs as bilateral alveolar infiltrates; however, infiltrates may occur unilaterally.
The laboratory results are not consistent with the presence of anti–glomerular basement membrane antibody disease or systemic small-vessel vasculitis. The negative antinuclear-antibody test does not provide support for a diagnosis of systemic lupus erythematosus in the absence of clinical or serologic evidence, or both. Results of pulmonary-function tests indicate the presence of mild small-airways dysfunction, a nonspecific finding. The results of repeated bronchoscopy essentially rule out bronchial sources of bleeding. Pulmonary embolism remains a serious concern, and further evaluation is needed.
A ventilation–perfusion lung scan (Figure 1) showed an almost total absence of perfusion of the left lung. The perfusion of the right lung appeared to be preserved, except for a defect at the superior segment of the lower lobe and a slight defect in the area of the posterobasal segment.
Figure 1. Ventilation–Perfusion Lung Scan.
Ventilation is relatively preserved (Panels A and B), but perfusion is almost totally absent in the entire left lung (Panels C and D, arrows).
Such a severe perfusion defect with preserved ventilation reflects a high probability of pulmonary emboli. The location of the lung infiltrates is compatible with areas of pulmonary infarction. The prolonged period of illness raises the possibility of chronic thromboembolic disease. However, the persistent fevers, laboratory findings of prominent inflammation, and the marked unilateral nature of the perfusion defect in the left lung raise clinical uncertainty about this diagnosis, and therefore, further evaluation is warranted.
CT angiography revealed normal main pulmonary arteries. Other findings included calcified left hilar lymph nodes, subcarinal lymphadenopathy, apparent compression of the left upper pulmonary artery by enlarged lymph nodes, patchy alveolar infiltrates that were partly pleura-based, and a small pleural effusion (Figure 2). Ultrasound-guided thoracentesis was performed. The pleural fluid was clear; the differential blood count showed 65 percent lymphocytes and 26 percent neutrophils. The level of total protein was 4.7 g per deciliter in pleural fluid and 7.1 g per deciliter in serum, and the lactate dehydrogenase level was 1308 U per liter (in the serum, 302). Transthoracic echocardiography demonstrated normal heart chambers, slight mitral and tricuspid regurgitation, and normal pulmonary-artery pressure.
Figure 2. CT Angiogram of the Chest.
In a lung window (Panel A), peripheral pulmonary infiltrates (PI) are visible in the left lung. In a mediastinal window (Panel B), a mediastinal density (MD) and a left hilar density (HD) are present, with a narrowed left upper pulmonary artery (LUPA) and pleural effusion (PE).
The high levels of lactate dehydrogenase and protein in the pleural fluid meet Light's criteria for exudative pleural effusion, but they do not point to a specific diagnosis. A hilar mass compressing the pulmonary artery could cause a perfusion defect, scintigraphically simulating pulmonary thromboembolism. Since no intraarterial defect was identified angiographically, thromboembolism is doubtful. The apparent mediastinal lymphadenopathy requires explanation. One possibility is fibrosing mediastinitis, a rare entity that may be idiopathic or occur as a result of histoplasmosis or tuberculosis. Fibrosing mediastinitis may compromise the pulmonary artery or the pulmonary venous return. In the absence of a histologic diagnosis, however, it is unclear how the patient should be treated.
On open biopsy of the left lung, 300 ml of clear pleural fluid was evacuated. The visceral pleura of the left upper lobe was adherent to the chest wall, and several lesions were seen throughout the left lung, but the surgeon found no mediastinal lymphadenopathy. Wedge resection at the level of the lingula with pleural excision was performed. Histopathological study disclosed hemorrhagic infarcts, arteries with nonconcentric intimal thickening that was highly suggestive of the presence of thrombotic arteriopathy, and fibrotic pleural foci with intimal thickening of the venules (Figure 3). Stains for acid-fast bacilli were negative.
Figure 3. Biopsy Specimen of the Lung (Hematoxylin and Eosin).
In Panel A, hemorrhagic infarcts (HI) and a thrombotic artery (A) are present. In Panel B, there is intimal thickening of a venule (V).
The pathological findings are consistent with the presence of arteriopathy, venulopathy, and secondary hemorrhagic infarcts; nothing suggests necrotizing vasculitis. The rare entity of pulmonary veno-occlusive disease should be considered. Its pathological hallmark is extensive occlusion of the pulmonary veins by fibrous tissue, sometimes with arterialized media with increased elastic fibers. In about 50 percent of the cases, the pulmonary arterioles have medial hypertrophy.
Further laboratory testing revealed normal levels of proteins C and S, antithrombin III, and factors VIII, IX, and XI. Testing for antiphospholipid antibodies, lupus anticoagulant, activated protein C resistance, and mutations of factor V Leiden, methyltetrahydrofolate reductase, and prothrombin G20210A was negative.
Hypercoagulability may contribute to the pathogenesis of pulmonary veno-occlusive disease, and thus, it is prudent to rule out thrombophilic disorders, as was done in this case. Despite the negative evaluation for a coagulation disorder, anticoagulant treatment is reasonable to attempt to inhibit a thrombotic tendency. To establish the diagnosis of veno-occlusive disease, we still need to rule out any secondary cause of obstruction of pulmonary venous outflow.
Anticoagulation treatment was started. The pleuritic chest pain and hemoptysis resolved. Transesophageal echocardiography demonstrated complete occlusion of the two left pulmonary veins and absence of blood flow in the left pulmonary artery (Figure 4). Pulmonary angiography revealed nearly complete occlusion of the pulmonary veins on the left side of the chest. Some collateral veins and a very small pulmonary vein draining a small volume of blood from the lower lobe of the left lung were visualized (Figure 5). Pulmonary-vein stenosis was recognized to be a potential complication of the patient's prior ablation procedure for atrial fibrillation. She was referred to a facility experienced in the use of the transcatheter technique for pulmonary-vein dilatation. She underwent successful recannulation of the completely occluded left lower pulmonary vein, followed by placement of a stent. Treatment of the left upper pulmonary vein was deferred because the occlusion was not thought to be readily amenable to stenting. Follow-up CT scanning showed improvement in the patchy parenchymal and ground-glass opacities within the left lower lobe of the lung. A quantitative lung-perfusion scan demonstrated some improvement in flow (14.1 percent from the left lung, as compared with 6.2 percent before the intervention). Cultures of specimens of sputum, bronchoalveolar fluid, and lung tissue subsequently proved negative for mycobacteria.
Figure 4. Multiplane Transesophageal Echocardiogram Showing Thrombotic Occlusion of a Pulmonary Vein.
LAA denotes left atrial appendage, LA left atrium, PV pulmonary vein, and TH thrombus.
Figure 5. Pulmonary Angiogram Showing Nearly Complete Occlusion of Pulmonary-Vein Flow.
The flow is limited to a very narrow pulmonary vein. LA denotes left atrium, LPA left pulmonary artery, and PV left lower pulmonary vein.
At follow-up examination two months later, the patient had no cough or fever and reported decreased shortness of breath. A repeated CT scan demonstrated a widely patent stent in the left lower pulmonary vein (Figure 6).
Figure 6. CT Scan Showing the Opened Left Lower Pulmonary Vein with the Inserted Stent.
Commentary
A diagnosis made after much testing sometimes raises the question of why the correct diagnosis was not made earlier. In some cases, diseases are overlooked because they are rare, mimic other diseases, or present atypically. The diagnosis of a relatively new clinical syndrome is especially challenging.
The case under discussion reflects just that scenario — pulmonary disease developing as a complication of an ablation procedure for atrial fibrillation. Ablation of the pulmonary veins was introduced in 1998, after the observation that the pulmonary veins are an important source of ectopic beats initiating atrial fibrillation and that these foci respond to radiofrequency ablation.1
With increasing use of the ablation procedure, its complications have become better understood. In a large international survey of physicians whose patients had undergone catheter ablation of atrial fibrillation, marked pulmonary-vein stenosis occurred in 1.3 percent and interventional treatment was required in about half the patients.2 In a retrospective follow-up study of 335 patients undergoing pulmonary-vein electrical isolation with the use of radiofrequency catheter ablation, severe pulmonary-vein stenosis was detected in 18 patients (5 percent), 8 of whom were asymptomatic; symptoms included shortness of breath (44 percent), cough (39 percent), hemoptysis (28 percent), and pleuritic pain (22 percent).3 In a case series of 23 patients with pulmonary-vein stenosis, 3 had influenza-like symptoms and low-grade fever and were treated with antibiotics, without benefit.4 The patient in this case also had these symptoms. In the case series, radiographic abnormalities were observed in 50 percent of patients with severe stenosis of the pulmonary veins; parenchymal consolidation was seen most often, followed by left pleural effusion.3 Pulmonary-vein stenosis was not initially considered in any of the patients in the series, and misdiagnosis and unwarranted diagnostic and therapeutic procedures were common.3
The evaluation of any clinical problem should include consideration of the patient's history and whether it might relate to the current condition. In this case, the connection with the patient's history was missed because this rare potential complication of ablation was unfamiliar to the clinicians caring for the patient and because there was a prolonged interval between the procedure and the patient's presentation. Had the specialist who performed the cardiac procedure been consulted, the correct diagnosis would presumably have been made sooner.
In this patient, the lung-perfusion scan, initially interpreted as showing a high probability of pulmonary emboli, served as the clue to the vascular disease underlying the parenchymal infiltrative disorder. In a series of 11 patients evaluated for suspected pulmonary-vein stenosis complicating catheter ablation for atrial fibrillation, 6 had more than 50 percent narrowing on direct pulmonary venography, and all 6 had perfusion defects in the affected pulmonary lobe.5 The quantitative pulmonary ventilation–perfusion scans have been reported to correlate well with the severity of pulmonary-vein stenosis documented by spiral CT, and they may provide information regarding the functional value of pulmonary-vein stenosis.6 In this case, the grossly abnormal perfusion scan and the patient's symptoms were consistent with the presence of severe disease, which was documented on further testing. Transesophageal echocardiography, as was performed in this case, and magnetic resonance imaging, which allows good visualization of the pulmonary veins and can be used to estimate blood-flow velocity, can also be used to suggest the diagnosis.7 Balloon angioplasty has been recommended as a definitive diagnostic method; in some cases, veins that appeared severely stenosed or occluded on spiral CT have been found to be patent on balloon angioplasty.8 In case series, transcatheter dilatation relieved obstruction and improved flow, but longer follow-up and larger series are needed to predict more accurately the long-term success of this procedure and the factors predisposing to its success.3,8
The recognition that pulmonary-vein stenosis may complicate ablation has led to changes in the procedure to minimize this risk. Ostial ablation appears to be associated with a markedly lower risk of stenosis than distal pulmonary-vein ablation,4,9 as was performed in the case under discussion. The risk of this complication also appears to be reduced by the use of lower temperatures and energy for ablation, the use of intracardiac echocardiography or various mapping systems, and increased operator experience.6 These approaches should make pulmonary-vein stenosis even less common in the future.
This case highlights the need for clinicians to consider a patient's history and treatment carefully when evaluating a new problem. Greater awareness of this uncommon complication of ablation therapy is needed among the generalist physicians who follow these patients on a long-term basis.
No potential conflict of interest relevant to this article was reported.
We are indebted to Drs. Yossi Manisterski, Michal Cohen, Itshak Shechtman, Ofer Biniaminov, and Hanna Bernstein for their assistance with the evaluation and treatment of the patient.
Source Information
From the Departments of Medicine D (F.S., R.T.-K.), Cardiology (R.H.), and Pulmonary Medicine (M.R.K.), Rabin Medical Center, Beilinson Campus, Petah Tiqwa; and Sackler School of Medicine, Tel Aviv University, Tel Aviv (F.S., R.H., R.T.-K., M.R.K.) — both in Israel.
Address reprint requests to Dr. Salamon at the Department of Medicine D, Rabin Medical Center, Beilinson Campus, Petah Tiqwa 49100, Israel, or at salamonf@post.tau.ac.il.
References
Haissaguerre M, Jais P, Shah DC, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998;339:659-666.
Cappato R, Calkins H, Chen S-A, et al. Worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circulation 2005;111:1100-1105.
Saad EB, Marrouche NF, Saad CP, et al. Pulmonary vein stenosis after catheter ablation of atrial fibrillation: emergence of a new clinical syndrome. Ann Intern Med 2003;138:634-638.
Packer DL, Keelan P, Munger TM, et al. Clinical presentation, investigation, and management of pulmonary vein stenosis complicating ablation for atrial fibrillation. Circulation 2005;111:546-554.
Nanthakumar K, Mountz JM, Plumb VJ, Epstein AE, Kay GN. Functional assessment of pulmonary vein stenosis using radionuclide ventilation/perfusion imaging. Chest 2004;126:645-651.
Saad EB, Rossillo A, Saad CP, et al. Pulmonary vein stenosis after radiofrequency ablation of atrial fibrillation: functional characterization, evolution, and influence of the ablation strategy. Circulation 2003;108:3102-3107.
Yang M, Akbari H, Reddy GP, Higgins CB. Identification of pulmonary vein stenosis after radiofrequency ablation for atrial fibrillation using MRI. J Comput Assist Tomogr 2001;25:34-35.
Qureshi AM, Prieto LR, Latson LA, et al. Transcatheter angioplasty for acquired pulmonary vein stenosis after radiofrequency ablation. Circulation 2003;108:1336-1342.
Arentz T, Jander N, von Rosenthal J, et al. Incidence of pulmonary vein stenosis 2 years after radiofrequency catheter ablation of refractory atrial fibrillation. Eur Heart J 2003;24:963-969.(Francis Salamon, M.D., Ra)
A 58-year-old woman was hospitalized for the evaluation of prolonged fever and hemoptysis. She reported an eight-month history of intermittent fevers, a productive cough, shortness of breath, and hemoptysis, which had been evaluated at another hospital. Computed tomographic (CT) scans of the chest revealed peripheral infiltrates in the upper lobe of the left lung and the lingula and a calcified left hilar opacity with additional, small mediastinal lymph nodes. Bronchoscopy demonstrated hyperemic bronchi. A transbronchial biopsy showed fibrinous material in some alveolar spaces, with a few scattered neutrophils, and capillary congestion in portions of the septa. Culture of a bronchial-lavage specimen was positive for a nontuberculous, slow-growing mycobacterium. Staining and culturing were negative for Mycobacterium tuberculosis. A course of amoxicillin–clavulanic acid and a subsequent course of cefuroxime resulted in no apparent benefit. A tuberculin skin test with purified protein derivative (5 TU) was positive, with an induration of 30 mm.
The combination of prolonged fever, unilateral pulmonary infiltrates, hemoptysis, calcified hilar lymph nodes, and a positive tuberculin skin test is highly suggestive of active tuberculosis. Although staining and culturing of a bronchoalveolar-lavage specimen were reportedly negative for M. tuberculosis, such results do not effectively rule out this diagnosis; smears of bronchoalveolar fluid are reportedly negative in almost half the cases, and cultures are negative in a quarter of the cases. It would be prudent to have three sputum specimens reexamined for M. tuberculosis despite the negative bronchoscopic findings. Respiratory precautions are needed until active tuberculosis is ruled out. In the absence of underlying immunosuppression or appreciable prior chronic lung disease, the slow-growing mycobacterium on a single culture may represent a clinically insignificant contaminant, but further evaluation is needed.
Ten days before the current admission, the patient consulted another pulmonologist, who prescribed methylprednisolone (40 mg per day) and isoniazid (300 mg per day). At presentation to our medical center, she reported increased cough, hemoptysis, pleuritic chest pain, and fever. The patient had a history of hypertension, chronic atrial fibrillation, inflammatory bowel disease, and hypothyroidism that was attributed to the use of amiodarone. She had undergone ablation therapy for atrial fibrillation one year earlier, after which the amiodarone had been stopped. She did not smoke or drink. Her medications on admission were 5-aminosalicylic acid, pravastatin, and levothyroxine.
Corticosteroid treatment could be beneficial for inflammatory lung disease, but without a specific diagnosis, especially if active tuberculosis is present, the use of this therapy arouses concern. The patient's clinical condition worsened after a few days of treatment with isoniazid and corticosteroids. Although fever is a well-known side effect of isoniazid therapy, the use of corticosteroids usually prevents it. I am concerned that the worsening clinical manifestations are related to her underlying disease. Amiodarone is a well-recognized cause of pulmonary complications, which may be manifested by organizing pneumonia. However, such complications usually develop during active treatment or soon after its cessation, and thus, they are unlikely to explain the current presentation.
Bronchiolitis obliterans is an infrequent but well-known manifestation of inflammatory bowel disease. The clinical presentation often mimics that of community-acquired pneumonia, but no improvement results with antibiotic therapy. Typically, the lung opacities are diffuse and bilateral and pleural effusion is infrequent. In rare cases, 5-aminosalicylic acid has been linked to adverse pulmonary effects, including diffuse or basilar infiltrates and bronchiolitis obliterans. I would discontinue this drug, pending further evaluation. Recurrent pulmonary thromboembolism must also be considered.
On examination, the patient appeared well nourished and in no distress. Her temperature was 39°C, and she had a pulse of 80 beats per minute, a blood pressure of 119/74 mm Hg, a respiratory rate of 18 breaths per minute, and an oxygen saturation of hemoglobin of 97 percent while she was breathing room air. Auscultation of the chest revealed no abnormalities except for scattered rales in the lower lobe of the left lung. Examination of the heart revealed no abnormalities. There was no organomegaly or abdominal tenderness. The hemoglobin level was 12.4 g per deciliter, and the white-cell count was 18,440 per cubic millimeter, with 76 percent neutrophils, 15 percent lymphocytes, 5 percent monocytes, 1.5 percent eosinophils, and 2.5 percent basophils; the platelet count was 470,000 per cubic millimeter. The erythrocyte sedimentation rate was 60 mm per hour. The C-reactive protein level was 10 mg per deciliter (normal range, 0 to 0.5). Levels of hepatic enzymes and renal function were within normal limits. A radiograph of the chest showed alveolar infiltrates in the left lateral hemithorax.
The fever and the laboratory results suggest the presence of an inflammatory reaction. The marked elevation in the white-cell count may be due to the administration of corticosteroids, but a superimposed bacterial infection should be considered. Blood and sputum should be obtained for culture, and empirical antibiotic treatment is a reasonable option. Noninfectious inflammatory lung diseases should also be considered, especially systemic lupus erythematosus, Wegener's granulomatosis, and Goodpasture's syndrome. Tests for the presence of antinuclear antibodies, antineutrophil cytoplasmic antibodies, and antiglomerular basement-membrane antibodies, as well as bronchoscopic examination, should be repeated.
Ciprofloxacin and clarithromycin were begun, and orally administered corticosteroids were continued. The 5-aminosalicylic acid was stopped. The patient's temperature gradually became normal, and the hemoptysis decreased substantially. Cultures of blood and sputum yielded no growth. Staining of three sputum specimens was negative for acid-fast bacilli. The antinuclear-antibody titer was positive at 1:160; tests for antibodies against double-stranded DNA, antibodies against glomerular basement membrane, and antineutrophil cytoplasmic antibodies were negative. On pulmonary-function testing, the forced vital capacity (FVC), the forced expiratory volume in one second (FEV1), and the FEV1:FVC ratio were normal; the FEV between 25 and 75 percent of FVC was slightly reduced (60 percent of the predicted value); the lung volumes were normal; and there was a borderline reduction in diffusing capacity. Bronchoscopy revealed slight bleeding that appeared to be of lingular origin. Gram's stain and acid-fast stains of bronchoalveolar-lavage fluid were negative for bacteria, as were cultures. The transbronchial biopsy showed interstitial changes of subacute lung injury with hemosiderin-laden alveolar macrophages.
The histologic findings are inconsistent with the presence of an infectious disease in the lung. Instead, they suggest alveolar hemorrhage, although the bronchoscopist did not report the characteristic return of continuously bloody lavage fluid. Alveolar hemorrhage generally appears on chest radiographs as bilateral alveolar infiltrates; however, infiltrates may occur unilaterally.
The laboratory results are not consistent with the presence of anti–glomerular basement membrane antibody disease or systemic small-vessel vasculitis. The negative antinuclear-antibody test does not provide support for a diagnosis of systemic lupus erythematosus in the absence of clinical or serologic evidence, or both. Results of pulmonary-function tests indicate the presence of mild small-airways dysfunction, a nonspecific finding. The results of repeated bronchoscopy essentially rule out bronchial sources of bleeding. Pulmonary embolism remains a serious concern, and further evaluation is needed.
A ventilation–perfusion lung scan (Figure 1) showed an almost total absence of perfusion of the left lung. The perfusion of the right lung appeared to be preserved, except for a defect at the superior segment of the lower lobe and a slight defect in the area of the posterobasal segment.
Figure 1. Ventilation–Perfusion Lung Scan.
Ventilation is relatively preserved (Panels A and B), but perfusion is almost totally absent in the entire left lung (Panels C and D, arrows).
Such a severe perfusion defect with preserved ventilation reflects a high probability of pulmonary emboli. The location of the lung infiltrates is compatible with areas of pulmonary infarction. The prolonged period of illness raises the possibility of chronic thromboembolic disease. However, the persistent fevers, laboratory findings of prominent inflammation, and the marked unilateral nature of the perfusion defect in the left lung raise clinical uncertainty about this diagnosis, and therefore, further evaluation is warranted.
CT angiography revealed normal main pulmonary arteries. Other findings included calcified left hilar lymph nodes, subcarinal lymphadenopathy, apparent compression of the left upper pulmonary artery by enlarged lymph nodes, patchy alveolar infiltrates that were partly pleura-based, and a small pleural effusion (Figure 2). Ultrasound-guided thoracentesis was performed. The pleural fluid was clear; the differential blood count showed 65 percent lymphocytes and 26 percent neutrophils. The level of total protein was 4.7 g per deciliter in pleural fluid and 7.1 g per deciliter in serum, and the lactate dehydrogenase level was 1308 U per liter (in the serum, 302). Transthoracic echocardiography demonstrated normal heart chambers, slight mitral and tricuspid regurgitation, and normal pulmonary-artery pressure.
Figure 2. CT Angiogram of the Chest.
In a lung window (Panel A), peripheral pulmonary infiltrates (PI) are visible in the left lung. In a mediastinal window (Panel B), a mediastinal density (MD) and a left hilar density (HD) are present, with a narrowed left upper pulmonary artery (LUPA) and pleural effusion (PE).
The high levels of lactate dehydrogenase and protein in the pleural fluid meet Light's criteria for exudative pleural effusion, but they do not point to a specific diagnosis. A hilar mass compressing the pulmonary artery could cause a perfusion defect, scintigraphically simulating pulmonary thromboembolism. Since no intraarterial defect was identified angiographically, thromboembolism is doubtful. The apparent mediastinal lymphadenopathy requires explanation. One possibility is fibrosing mediastinitis, a rare entity that may be idiopathic or occur as a result of histoplasmosis or tuberculosis. Fibrosing mediastinitis may compromise the pulmonary artery or the pulmonary venous return. In the absence of a histologic diagnosis, however, it is unclear how the patient should be treated.
On open biopsy of the left lung, 300 ml of clear pleural fluid was evacuated. The visceral pleura of the left upper lobe was adherent to the chest wall, and several lesions were seen throughout the left lung, but the surgeon found no mediastinal lymphadenopathy. Wedge resection at the level of the lingula with pleural excision was performed. Histopathological study disclosed hemorrhagic infarcts, arteries with nonconcentric intimal thickening that was highly suggestive of the presence of thrombotic arteriopathy, and fibrotic pleural foci with intimal thickening of the venules (Figure 3). Stains for acid-fast bacilli were negative.
Figure 3. Biopsy Specimen of the Lung (Hematoxylin and Eosin).
In Panel A, hemorrhagic infarcts (HI) and a thrombotic artery (A) are present. In Panel B, there is intimal thickening of a venule (V).
The pathological findings are consistent with the presence of arteriopathy, venulopathy, and secondary hemorrhagic infarcts; nothing suggests necrotizing vasculitis. The rare entity of pulmonary veno-occlusive disease should be considered. Its pathological hallmark is extensive occlusion of the pulmonary veins by fibrous tissue, sometimes with arterialized media with increased elastic fibers. In about 50 percent of the cases, the pulmonary arterioles have medial hypertrophy.
Further laboratory testing revealed normal levels of proteins C and S, antithrombin III, and factors VIII, IX, and XI. Testing for antiphospholipid antibodies, lupus anticoagulant, activated protein C resistance, and mutations of factor V Leiden, methyltetrahydrofolate reductase, and prothrombin G20210A was negative.
Hypercoagulability may contribute to the pathogenesis of pulmonary veno-occlusive disease, and thus, it is prudent to rule out thrombophilic disorders, as was done in this case. Despite the negative evaluation for a coagulation disorder, anticoagulant treatment is reasonable to attempt to inhibit a thrombotic tendency. To establish the diagnosis of veno-occlusive disease, we still need to rule out any secondary cause of obstruction of pulmonary venous outflow.
Anticoagulation treatment was started. The pleuritic chest pain and hemoptysis resolved. Transesophageal echocardiography demonstrated complete occlusion of the two left pulmonary veins and absence of blood flow in the left pulmonary artery (Figure 4). Pulmonary angiography revealed nearly complete occlusion of the pulmonary veins on the left side of the chest. Some collateral veins and a very small pulmonary vein draining a small volume of blood from the lower lobe of the left lung were visualized (Figure 5). Pulmonary-vein stenosis was recognized to be a potential complication of the patient's prior ablation procedure for atrial fibrillation. She was referred to a facility experienced in the use of the transcatheter technique for pulmonary-vein dilatation. She underwent successful recannulation of the completely occluded left lower pulmonary vein, followed by placement of a stent. Treatment of the left upper pulmonary vein was deferred because the occlusion was not thought to be readily amenable to stenting. Follow-up CT scanning showed improvement in the patchy parenchymal and ground-glass opacities within the left lower lobe of the lung. A quantitative lung-perfusion scan demonstrated some improvement in flow (14.1 percent from the left lung, as compared with 6.2 percent before the intervention). Cultures of specimens of sputum, bronchoalveolar fluid, and lung tissue subsequently proved negative for mycobacteria.
Figure 4. Multiplane Transesophageal Echocardiogram Showing Thrombotic Occlusion of a Pulmonary Vein.
LAA denotes left atrial appendage, LA left atrium, PV pulmonary vein, and TH thrombus.
Figure 5. Pulmonary Angiogram Showing Nearly Complete Occlusion of Pulmonary-Vein Flow.
The flow is limited to a very narrow pulmonary vein. LA denotes left atrium, LPA left pulmonary artery, and PV left lower pulmonary vein.
At follow-up examination two months later, the patient had no cough or fever and reported decreased shortness of breath. A repeated CT scan demonstrated a widely patent stent in the left lower pulmonary vein (Figure 6).
Figure 6. CT Scan Showing the Opened Left Lower Pulmonary Vein with the Inserted Stent.
Commentary
A diagnosis made after much testing sometimes raises the question of why the correct diagnosis was not made earlier. In some cases, diseases are overlooked because they are rare, mimic other diseases, or present atypically. The diagnosis of a relatively new clinical syndrome is especially challenging.
The case under discussion reflects just that scenario — pulmonary disease developing as a complication of an ablation procedure for atrial fibrillation. Ablation of the pulmonary veins was introduced in 1998, after the observation that the pulmonary veins are an important source of ectopic beats initiating atrial fibrillation and that these foci respond to radiofrequency ablation.1
With increasing use of the ablation procedure, its complications have become better understood. In a large international survey of physicians whose patients had undergone catheter ablation of atrial fibrillation, marked pulmonary-vein stenosis occurred in 1.3 percent and interventional treatment was required in about half the patients.2 In a retrospective follow-up study of 335 patients undergoing pulmonary-vein electrical isolation with the use of radiofrequency catheter ablation, severe pulmonary-vein stenosis was detected in 18 patients (5 percent), 8 of whom were asymptomatic; symptoms included shortness of breath (44 percent), cough (39 percent), hemoptysis (28 percent), and pleuritic pain (22 percent).3 In a case series of 23 patients with pulmonary-vein stenosis, 3 had influenza-like symptoms and low-grade fever and were treated with antibiotics, without benefit.4 The patient in this case also had these symptoms. In the case series, radiographic abnormalities were observed in 50 percent of patients with severe stenosis of the pulmonary veins; parenchymal consolidation was seen most often, followed by left pleural effusion.3 Pulmonary-vein stenosis was not initially considered in any of the patients in the series, and misdiagnosis and unwarranted diagnostic and therapeutic procedures were common.3
The evaluation of any clinical problem should include consideration of the patient's history and whether it might relate to the current condition. In this case, the connection with the patient's history was missed because this rare potential complication of ablation was unfamiliar to the clinicians caring for the patient and because there was a prolonged interval between the procedure and the patient's presentation. Had the specialist who performed the cardiac procedure been consulted, the correct diagnosis would presumably have been made sooner.
In this patient, the lung-perfusion scan, initially interpreted as showing a high probability of pulmonary emboli, served as the clue to the vascular disease underlying the parenchymal infiltrative disorder. In a series of 11 patients evaluated for suspected pulmonary-vein stenosis complicating catheter ablation for atrial fibrillation, 6 had more than 50 percent narrowing on direct pulmonary venography, and all 6 had perfusion defects in the affected pulmonary lobe.5 The quantitative pulmonary ventilation–perfusion scans have been reported to correlate well with the severity of pulmonary-vein stenosis documented by spiral CT, and they may provide information regarding the functional value of pulmonary-vein stenosis.6 In this case, the grossly abnormal perfusion scan and the patient's symptoms were consistent with the presence of severe disease, which was documented on further testing. Transesophageal echocardiography, as was performed in this case, and magnetic resonance imaging, which allows good visualization of the pulmonary veins and can be used to estimate blood-flow velocity, can also be used to suggest the diagnosis.7 Balloon angioplasty has been recommended as a definitive diagnostic method; in some cases, veins that appeared severely stenosed or occluded on spiral CT have been found to be patent on balloon angioplasty.8 In case series, transcatheter dilatation relieved obstruction and improved flow, but longer follow-up and larger series are needed to predict more accurately the long-term success of this procedure and the factors predisposing to its success.3,8
The recognition that pulmonary-vein stenosis may complicate ablation has led to changes in the procedure to minimize this risk. Ostial ablation appears to be associated with a markedly lower risk of stenosis than distal pulmonary-vein ablation,4,9 as was performed in the case under discussion. The risk of this complication also appears to be reduced by the use of lower temperatures and energy for ablation, the use of intracardiac echocardiography or various mapping systems, and increased operator experience.6 These approaches should make pulmonary-vein stenosis even less common in the future.
This case highlights the need for clinicians to consider a patient's history and treatment carefully when evaluating a new problem. Greater awareness of this uncommon complication of ablation therapy is needed among the generalist physicians who follow these patients on a long-term basis.
No potential conflict of interest relevant to this article was reported.
We are indebted to Drs. Yossi Manisterski, Michal Cohen, Itshak Shechtman, Ofer Biniaminov, and Hanna Bernstein for their assistance with the evaluation and treatment of the patient.
Source Information
From the Departments of Medicine D (F.S., R.T.-K.), Cardiology (R.H.), and Pulmonary Medicine (M.R.K.), Rabin Medical Center, Beilinson Campus, Petah Tiqwa; and Sackler School of Medicine, Tel Aviv University, Tel Aviv (F.S., R.H., R.T.-K., M.R.K.) — both in Israel.
Address reprint requests to Dr. Salamon at the Department of Medicine D, Rabin Medical Center, Beilinson Campus, Petah Tiqwa 49100, Israel, or at salamonf@post.tau.ac.il.
References
Haissaguerre M, Jais P, Shah DC, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998;339:659-666.
Cappato R, Calkins H, Chen S-A, et al. Worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circulation 2005;111:1100-1105.
Saad EB, Marrouche NF, Saad CP, et al. Pulmonary vein stenosis after catheter ablation of atrial fibrillation: emergence of a new clinical syndrome. Ann Intern Med 2003;138:634-638.
Packer DL, Keelan P, Munger TM, et al. Clinical presentation, investigation, and management of pulmonary vein stenosis complicating ablation for atrial fibrillation. Circulation 2005;111:546-554.
Nanthakumar K, Mountz JM, Plumb VJ, Epstein AE, Kay GN. Functional assessment of pulmonary vein stenosis using radionuclide ventilation/perfusion imaging. Chest 2004;126:645-651.
Saad EB, Rossillo A, Saad CP, et al. Pulmonary vein stenosis after radiofrequency ablation of atrial fibrillation: functional characterization, evolution, and influence of the ablation strategy. Circulation 2003;108:3102-3107.
Yang M, Akbari H, Reddy GP, Higgins CB. Identification of pulmonary vein stenosis after radiofrequency ablation for atrial fibrillation using MRI. J Comput Assist Tomogr 2001;25:34-35.
Qureshi AM, Prieto LR, Latson LA, et al. Transcatheter angioplasty for acquired pulmonary vein stenosis after radiofrequency ablation. Circulation 2003;108:1336-1342.
Arentz T, Jander N, von Rosenthal J, et al. Incidence of pulmonary vein stenosis 2 years after radiofrequency catheter ablation of refractory atrial fibrillation. Eur Heart J 2003;24:963-969.(Francis Salamon, M.D., Ra)