0
Original Research: PULMONARY HYPERTENSION |

Is Pulmonary Arterial Hypertension in Neurofibromatosis Type 1 Secondary to a Plexogenic Arteriopathy?* FREE TO VIEW

Douglas R. Stewart, MD; Joy D. Cogan, PhD; Mordechai R. Kramer, MD, FCCP; Wallace T. Miller, Jr, MD; Leif E. Christiansen, DO; Michael W. Pauciulo, MBA; Ludwine M. Messiaen, PhD; George S. Tu, MD, FCCP; William H. Thompson, MD, FCCP; Reed E. Pyeritz, MD, PhD; Jay H. Ryu, MD, FCCP; William C. Nichols, PhD; Makoto Kodama, MD; Barbara O. Meyrick, PhD; David J. Ross, MD
Author and Funding Information

*From the National Human Genome Research Institute (Dr. Stewart), National Institutes of Health, Bethesda, MD; the Department of Pediatrics (Dr. Cogan), Vanderbilt University Medical Center, Nashville, TN; Institute of Pulmonary Medicine (Dr. Kramer), Rabin Medical Center, Petach Tikva, Israel; the Departments of Radiology (Dr. Miller) and Medicine (Dr. Pyeritz), University of Pennsylvania School of Medicine, Philadelphia, PA; St. Joseph’s Hospital (Dr. Christiansen), Reading PA; Cincinnati Children’s Hospital Medical Center (Mr. Pauciulo and Dr. Nichols), Cincinnati, OH; the Department of Genetics (Dr. Messiaen), University of Alabama at Birmingham, Birmingham, AL; Lung Center of Nevada (Dr. Tu), Las Vegas, NV; the Division of Pulmonary and Critical Care Medicine (Dr. Thompson), University of Washington, Seattle, WA; the Division of Pulmonary and Critical Care Medicine (Dr. Ryu), Mayo Clinic, Rochester, MN; the First Department of Internal Medicine (Dr. Kodama), Niigata University School of Medicine, Niigata, Japan; the Department of Pathology and Medicine (Dr. Meyrick), Vanderbilt University Medical Center, Nashville, TN; and the Department of Pulmonary and Critical Care Medicine and Hospitalists (Dr. Ross), David Geffen School of Medicine, University of California, Los Angeles, CA.

Correspondence to: Douglas R. Stewart, MD, National Human Genome Research Institute, National Institutes of Health, 49 Convent Dr, Building 49, Room 4A62, Bethesda, MD 20892; e-mail: drstewart@mail.nih.gov



Chest. 2007;132(3):798-808. doi:10.1378/chest.06-3017
Text Size: A A A
Published online

Background: Neurofibromatosis type 1 (NF1) is a common disorder of dysregulated tissue growth secondary to mutations in the tumor suppressor gene NF1. Pulmonary arterial hypertension (PAH) in patients with NF1 is hypothesized to be secondary to an underlying vasculopathy.

Methods: We describe the entity we term NF1-associated PAH (NF1-PAH) in four new patients and update the data on four previously published reports of patients with PAH and NF1. We performed genetic testing of the bone morphogenic protein receptor 2 (BMPR2) gene, which is mutated in 70% of patients with familial PAH and approximately 25% of patients with idiopathic PAH. We report, for the first time, pathologic findings in the autopsy-obtained lung of one patient with NF1-PAH.

Results: Patients with NF1-PAH have a generally poor long-term prognosis. In four patients, we observed the mosaic pattern of lung attenuation on a CT scan of the chest, a radiographic finding that can be consistent with an underlying vasculopathy. No mutations or rearrangements in the BMPR2 gene were found. We observed complex plexiform lesions in the one available autopsy specimen. Similar lesions are a hallmark of plexogenic pulmonary arteriopathy and are associated with several severe types of PAH. (Plexiform lesions should not be confused with plexiform neurofibromas, which are distinctive tumors seen in NF1.)

Conclusions: Our findings suggest that NF1 should be considered as being “associated with PAH” as outlined in the Revised Clinical Classification of Pulmonary Hypertension. Understanding the mechanism of PAH in NF1 may inform the pathogenesis of PAH, NF1-PAH itself, and other NF1-associated vasculopathies. The pulmonary vasculature should now be included among the arterial beds affected by NF1 vasculopathy.

Figures in this Article

Neurofibromatosis type 1 (NF1) is a common (1 per 3,000 individuals) monogenic disorder of dysregulated tissue growth secondary to mutations in NF1, a tumor suppressor gene that regulates the protooncogene RAS. It is inherited in an autosomal-dominant pattern.1Pulmonary arterial hypertension (PAH) in patients with NF1 has been recognized and is hypothesized to be secondary to an underlying vasculopathy.23 Vasculopathies are uncommon but well-recognized complications of NF1.45 Patients with NF1 and PAH typically receive diagnoses of primary pulmonary hypertension, which is now termed idiopathic PAH (IPAH).6

In this article, we describe four new patients with NF1 and PAH, and review and update data on four previously published reports of patients with PAH and NF1. We report, for the first time, complex plexiform lesions in the lung of one patient with NF1 and PAH. In four patients, we observed the mosaic pattern of lung attenuation (hereafter called mosaic pattern) on a CT scan of the chest, which is a radiographic finding that can be consistent with an underlying vasculopathy. We performed genetic testing of the bone morphogenic receptor protein 2 (BMPR2) gene, which is mutated in 70% of patients with familial PAH and approximately 25% of patients with IPAH.7 Our findings suggest that NF1 should be considered as “associated with PAH” (APAH), as outlined in the Revised Clinical Classification of Pulmonary Hypertension.,6 Furthermore, this association confirms the susceptibility of the pulmonary arterial bed to an NF1-associated vasculopathy. In recognition of NF1 as a distinct cause of PAH, we propose that it be termed NF1-associated PAH (NF1-PAH).

Table 1 summarizes the clinical evaluation of four previously unreported patients with NF1 and PAH. Tables 2 and 3 summarize detailed echocardiographic and cardiac catheterization data. All patients underwent an appropriate workup to rule out secondary causes of pulmonary hypertension.

Patient 1

A 72-year-old female with NF1 presented with a several-month history of progressive dyspnea, lower extremity edema, exercise intolerance, and cyanosis. She denied chest pain, cough, wheezing, hemoptysis, fever, or weight loss. She had been hospitalized recently with similar complaints; echocardiography findings at that time showed right ventricular (RV) dilatation, triscuspid regurgitation, and severely elevated pulmonary pressures (approximately 76 mm Hg). A cardiac catheterization showed noncritical coronary artery disease. A lower extremity ultrasound showed a superficial venous thrombosis in the right leg, but the findings of a ventilation-perfusion (V̇/Q̇) lung scintigraphy scan were normal. She was discharged from the hospital receiving therapy with warfarin, furosemide, and home oxygen. Despite compliance, her symptoms progressed and she was readmitted to the hospital. Her medical history was remarkable only for NF1. There was no history of hypertension, pheochromocytomas, or vasculopathy. She was a retired factory worker but denied significant toxic exposures. She denied the use of alcohol, tobacco, or illicit drugs.

On physical examination, she had cyanotic lips and was in moderate respiratory distress. Her pulse oximetry was 84% while breathing oxygen at 6 L/min with a nasal cannula, with desaturations to 70% with minimal exertion. She had a jugular venous pulsation to the angle of the jaw, a right parasternal heave, and a loud second pulmonic sound. Her lungs were clear. She had 4+ peripheral pitting edema to the thigh. She had multiple neurofibromas and café-au-lait spots, mild scoliosis, and moderate obesity.

The patient was admitted to the ICU with the diagnosis of cor pulmonale. Repeat echocardiography showed progressive RV failure. A chest CT scan showed a mosaic pattern but no evidence of interstitial lung disease or fibrosis (Fig 1 ). Right heart catheterization confirmed severe pulmonary hypertension (65 to 80 mm Hg). Pulmonary function tests (PFTs) demonstrated a restrictive process with an FVC of 1.40 L (66% predicted), an FEV1 of 1.22 L (73% predicted), and an FEV1/FVC ratio of 88 (110% predicted). The results of tests for rheumatoid factor, antinuclear antibody (ANA), and anti-Scl antibodies were all negative. The patient denied any history of diet drug use, HIV risk factors, obstructive sleep apnea (OSA) symptoms, or rheumatologic symptoms. She received a diagnosis of primary pulmonary hypertension (PPH). Trials of calcium channel blockers (CCBs), angiotensin-converting enzyme inhibitors, and IV epoprostenol were not tolerated secondary to systemic hypotension; she died shortly after hospital discharge from respiratory failure. A request for autopsy was denied.

Patient 2

A 56-year-old woman with NF1, hyperparathyroidism, and multinodular goiter presented with progressive, severe dyspnea, one-block exercise intolerance, and near syncope. She experienced intermittent anterior chest discomfort associated with exertion without pain radiation, palpitations, or pleuritic quality. She used home oxygen. She was estimated to have New York Heart Association class 3 symptoms. The patient had no history of significant interstitial lung disease or sarcoidosis, though possible upper lung zone distribution “alveolitis” consistent with NF1 was noted on serial chest CT scan imaging. There was no prior thrombophlebitis or suspected pulmonary embolus (PE). She denied any signs or symptoms of collagen vascular disease. She denied OSA symptoms. She had a 50-pack-year smoking history, but had quit 15 years earlier. There was no history of illicit substance abuse, and she denied the usage of anorexigens.

A physical examination revealed multiple neurofibromas, mild goiter, and a murmur of tricuspid regurgitation. Serologic studies included an ANA of 1:320 with a homogenous pattern, and tests for antiscleroderma antibodies were negative. Additional serologies were negative for anti-Smith antibody, anti-Ro antibody, anti-La antibody, and the antiphospholipid antibody syndrome. The results of a dilute Russell viper venom test were normal. Assessments of liver function and HIV were unremarkable. A restrictive process was observed on PFTs, as follows: total lung capacity (TLC), 3.70 L (78% predicted); FVC, 2.42 L (83% predicted); FEV1, 1.80 L (76% predicted); FEV1/FVC ratio, 91% predicted; and diffusing capacity of the lung for carbon monoxide adjusted for hemoglobin, 9.8 (41% predicted [which was disproportionately reduced for reduction in lung volumes]). A high-resolution CT (HRCT) scan of the chest demonstrated a mild ground-glass attenuation involving bilateral upper lung zones, but without definitive reticulation, interstitial fibrosis, bronchiectasis, or honeycomb formation. Small, scattered subpleural blebs were noted in the lung apices. A helical chest CT scan demonstrated no evidence of acute or chronic thromboembolic pulmonary hypertension; a mild mosaic pattern (Fig 2 ) with ground-glass opacity in the upper lung zones was present. Echocardiography demonstrated marked right atrial and RV dilatation and decreased RV contractility with normal left ventricular (LV) ejection fraction (EF) of 50%. Combined left and right heart catheterization demonstrated pulmonary arterial pressure (PAP) of 100/40 mm Hg (mean PAP, 68 mm Hg). There was no evidence of intracardiac shunt or coronary atherosclerosis. A V̇/Q̇ lung scintigraphy scan demonstrated decreased perfusion to the upper lung zones but no segmental or subsegmental defects, and was deemed to have “low probability” for PE.

The patient received a diagnosis of PPH and started receiving therapy warfarin. Her symptoms and pulmonary pressures were initially improved with therapy with continuously infused epoprostenol; however, she died after 2 years of treatment during a hospitalization for progressive respiratory failure.

Patient 3

Patient 3 was a 68-year-old man with a history of NF1 complicated by a hydrocephalus managed with a ventricular-peritoneal shunt with an 18-month history of progressive dyspnea with intermittent pedal edema. He also had atrial fibrillation, coronary artery disease, and transient ischemic attacks but no history of known lung disease, OSA symptoms, arthritis, rashes, or sun hypersensitivities. He had a 40-pack-year smoking history, but had quit smoking 25 years earlier. The patient denied environmental or animal exposures. He denied resting dyspnea, orthopnea, chest tightness, cough, hemoptysis, or weight loss. The results of tests for sedimentation rate, ANA, and rheumatoid factor were all negative. A transthoracic echocardiogram showed cor pulmonale with a PAP estimated at 90 mm Hg. A chest CT scan with CT angiogram showed some fine cystic changes but no evidence of a PE. Mild obstruction was seen on the results of PFTs. A technetium persantine study with single-photon emission CT imaging showed small fixed defects near the base of the heart as well as the inferior septal and inferolateral junction, but with no evidence of reversible ischemia and an EF of 63%. Right heart catheterization showed an initial PAP of 68/20 mm Hg and a minimal response to epoprostenol. Therapy with oral nifedipine resulted in reduced pulmonary vascular resistance. The patient’s condition was managed using therapy with nifedipine and home oxygen. Subsequent transthoracic echocardiograms showed improvement in estimated PAP to 60 mm Hg. He died in a hospice approximately 6 years after the diagnosis of PPH from presumed right heart failure. No autopsy was performed. He had a brother who also had NF1-PAH, but further clinical details are not available.

Patient 4

Patient 4 was a 33-year-old woman with history of NF1 and a 2-year history of progressive dyspnea. Her NF1 was complicated by an optic pathway glioma and hydrocephalus managed with a ventricular-atrial shunt, later revised to a ventricular-peritoneal shunt. She also had a history of a carotid artery occlusion complicated by a stroke and residual right-sided paresis. At 31 years of age, she had increased dyspnea and was found to have small peripheral PE by pulmonary angiogram; the source of the clot was attributed to her ventricular-atrial shunt. She underwent anticoagulation with aspirin only because her optic pathway glioma was a contraindication to anticoagulation with warfarin. At 32 years of age, she was evaluated for dyspnea and hypoxia. A pulmonary artery catheterization showed a PAP of 60/28 mm Hg (mean, 40 mm Hg). Helical CT scan findings were negative for PE. At 33 years of age, her dyspnea worsened, and her exercise tolerance was limited to a few feet and was associated with sharp substernal chest pain. She denied alcohol, tobacco, and illicit drug use. An HRCT scan of the chest showed enlarged central pulmonary arteries and mosaic perfusion (Fig 3 ). No emboli or mural thrombus was observed. A moderate restrictive pattern was seen in PFT results, as follows: TLC, 2.31 L (63% predicted); FVC, 0.97 (34% predicted); and FEV1, 0.96 (38% predicted). A transesophageal echocardiogram showed an estimated RV systolic pressure of 68 mm Hg with mild-moderate tricuspid regurgitation, normal LV EF, and no evidence of intracardiac thrombus or other abnormality. Ultrasound of the lower extremities showed no thrombus. Right heart catheterization showed a PAP of 85/33 mm Hg (mean, 53 mm Hg) with a decrease to 71/27 mm Hg (mean, 45 mm Hg) with epoprostenol therapy. A pulmonary angiogram showed severe “pruning” consistent with severe pulmonary hypertension and no evidence of chronic PE. Treatment with a CCB, bosentan, and sildenafil did not significantly improve her symptoms. She lived for approximately 1 year after the diagnosis of PPH before dying in a hospice from respiratory failure. No autopsy was performed.

Update of Previously Published Case Reports

We obtained updated information on four previously published reports of patients with PAH and NF1 (Table 4 ). Follow-up information on the patient reported by Porterfield et al8 was not available. Samuels et al3 described a 51-year-old man with NF1 and severe PAH diagnosed after endarterectomy secondary to extensive irregular thickening of the intima of the pulmonary artery. Their report was, until this report, the only account with pulmonary histology. We reviewed the unpublished HRCT scan of the chest and found a mosaic pattern (Fig 4 ), which is consistent with his histologically proven vasculopathy. We arranged for BMPR2 gene testing; no mutations or rearrangements were found.

Aoki et al2 reported on two Japanese women with NF1 and PPH, and proposed a causal relationship between the vasculopathy of NF1 and PAH. Shortly after publication of the case report, their patient 1 died at 20 years of age from her NF1-PAH. BMPR2 gene testing was performed, but no mutations were found; rearrangements of BMPR2 were not examined (H. Hanawa, MD; personal communication; August 24, 2006). We reviewed a chest CT scan (not an HRCT scan) obtained when the patient was 16 years old. No mosaic pattern was observed. On autopsy, café-au-lait macules and neurofibromas were noted. The patient’s RV cavity had expanded to about twice its normal size. The liver showed mild hemostasis in the middle of the central veins. We reviewed three slides prepared from samples of her lungs (Fig 5 ). The appearance was consistent with IPAH with intimal and medial thickening of the arteries that often nearly or completely occluded the lumen (Fig 5, top left, a, and top right, b). Evidence of angiogenesis was sometimes apparent in the thickened intimal layer (Fig 5, bottom left, c). Many of the smaller arteries were associated with plexiform lesions (Fig 5, bottom right, d). The adventitia seemed unaffected, and the veins were normal in appearance.

Last, in the Spanish-language literature, Garcia Hernández et al9 described a 44-year-old man in whom NF1 and PPH had been diagnosed. For the benefit of English-language readers, we provide a translated report (as published in 2002) of their patient. Follow-up with the authors revealed that the patient’s symptoms progressed, and, despite treatment with sildenafil and bosentan, he died of NF1-PAH. The authors reviewed for us the CT scan of the chest, but no mosaic pattern was observed (J.S. Román, MD, PhD; personal communication; November 21, 2005). No DNA for BMPR2 analysis or pathologic material was available.

The patient reported by Garcia Hernández et al9 was a 44-year-old man with NF1. Over a 6-week period, a dry cough and increasing dyspnea with minimal exertion developed. The patient had never smoked or taken anorexigens. On physical examination, he had New York Heart Association class 3 symptoms with a left parasternal systolic murmur of II/VI intensity. He had malleolar edema. CBC count, sedimentation rate, rheumatoid factor, complement assay, antinuclear antibody tests, and “retrovirus serology” were all normal or negative. The Exner test was lengthened at 20.5 s, but a coagulation profile and tests for anticardiolipin antibodies were normal or negative. On an echocardiogram, the estimated pulmonary artery pressure was 90 mm Hg with evidence of RV dilatation but normal LV function. Chest radiograph findings showed enlarged pulmonary arteries and branches but without infiltrates or other parenchymal changes. A chest CT scan did not show any infiltrative lesions that were consistent with parenchymal pulmonary disease. PFT results showed an FVC of 5.3 L (107% predicted), an FEV1 of 4.2 L (110% predicted), and an FEV1/FVC ratio of 79% predicted. The findings of a lower extremity ultrasound and a V̇/Q̇ scan were unremarkable. Right heart catheterization showed pulmonary artery pressures of 62 mm Hg. Therapy with IV epoprostenol resulted in a 20% decrease in systolic pulmonary artery pressure. Pulmonary artery biopsy was not performed due to technical difficulties. Primary PAH was diagnosed and was treated with oral anticoagulants and the continuous infusion of IV epoprostenol. His exercise tolerance improved.

All investigations were performed under protocols approved by the University of Pennsylvania School of Medicine, Vanderbilt University School of Medicine, Cincinnati Children’s Hospital Medical Center, and the National Human Genome Research Institute institutional review boards.

Mutations in NF1

The total coding region of the NF1 gene from patient 1 was analyzed by reverse-transcription polymerase chain reaction and in vitro transcription/translation, as previously reported.10

Mutations in BMPR2

Lymphoblastoid cell lines from patient 1, the brother of patient 3 (who also had NF1-PAH), and the patient reported by Samuels et al3 were cultured as previously described11; BMPR2 complementary DNA was generated and sequenced by the reverse-transcription polymerase chain reaction method detailed in the work of Cogan and colleagues.,1112 Patient 2 had a similar mutation screen using DNA from blood, as previously described.13 For patient 1 reported by Aoki et al,2 sequencing only of the exons of the BMPR2 gene was performed on complementary DNA produced from messenger RNA extracted from lung samples that had been obtained at autopsy.

Multiplex ligation-dependent probe amplification to assess the exon copy number was performed on samples from patients 1, 2, and 3, and the individual reported by Samuels et al3 as previously described.11 Multiplex ligation-dependent probe amplification was not performed on the autopsy material that was available from the patient in the study by Aoki et al.2

Mutations in NF1
Patient 1:

Complementary DNA sequencing of fragment 2 identified a single-nucleotide insertion in exon 13: 2034dupC. The mutation was confirmed by sequencing exon 13 in genomic DNA.

Mutations in BMPR2

All samples had a wild-type sequence only for BMPR2 without evidence of rearrangements. In patient 1, a rare (minor allele frequency, 2%), nonconservative single-nucleotide polymorphism (rs2228545; c.2324G>A; p.Ser775Asn) in exon 12 was found.

Summary of NF1-PAH Phenotype

Detailed phenotype data are available in nine patients (Table 4), and detailed echocardiographic and catheterization data are available in four patients (Tables 2, 3). In all patients, systolic PAP as measured by catheterization or estimated by echocardiography was severely elevated (≥ 60 mm Hg), sometimes dramatically so. No secondary cause of pulmonary hypertension, including thrombophilia, was identified in any of the eight patients reported on or updated here. Two patients (patients 2 and 3) had a remote history of moderate smoking. However, in both patients, symptoms developed years after smoking cessation. Furthermore, the degree of pulmonary artery pressure elevation exceeds that typically seen in patients with pulmonary hypertension associated with COPD.6,14 In the patient reported on by Porterfield et al,8 the pulmonary hypertension was ascribed to interstitial fibrosis. However, no chest CT scan was reported in that patient.

The patient in the study by Samuels et al3 improved dramatically after undergoing endarterectomy. Continuously infused epoprostenol reduced pulmonary pressures and/or improved symptoms in all but one of the seven patients to whom it was administered. However, the long-term prognosis was poor. Of the seven patients for whom vital status was known, five patients (four women and one man) died from their NF1-PAH within 3 years of presentation. One man (patient 3) was maintained on therapy with oral nifedipine and died approximately 6 years after the diagnosis of NF1-PAH. Among the four new patients whom we reported, none died suddenly. They all died either in the hospital or in a hospice from progressive respiratory failure and/or cor pulmonale.

NF1-PAH is a rare complication of a classic Mendelian disorder. We provide radiographic and pathologic evidence of an underlying pulmonary vasculopathy in patients with NF1 and infer causality with the observed PAH. Despite the rarity of NF1-PAH, the association of NF1 and PAH is informative since it implicates mutations in the tumor suppressor gene NF1 with the pathogenesis of a pulmonary vasculopathy and subsequent pulmonary hypertension. Similar insights have been gleaned from other genetic causes deemed to be APAH (eg, type Ia glycogen storage disease, Gaucher disease, and hereditary hemorrhagic telangiectasia).6

The clinical features of NF1-PAH include the following: (1) an unequivocal diagnosis of NF115; (2) often rapidly progressive dyspnea; (3) a systolic PAP of > 60 mm Hg; (4) an absence of a secondary explanation despite adequate evaluation; (5) a lack of BMPR2 mutations or rearrangements; and (6) a poor long-term prognosis. Despite the prognosis, some improvement in both pulmonary hemodynamics and symptoms can be achieved by therapy with epoprostenol, bosentan, sildenafil, and beraprost. Interestingly, one man (patient 3) survived 6 years after the diagnosis of NF1-PAH while receiving therapy with oral nifedipine and home oxygen only. This variation in the pharmacologic response to CCBs is similar to the heterogeneity in response to CCBs observed in patients with IPAH.16

We infer the presence of a pulmonary vasculopathy on the basis of the mosaic pattern of lung attenuation seen on the chest CT or HRCT scan in patients 1, 2, and 4. The mosaic pattern is defined as sharply demarcated areas of heterogeneous attenuation in the pulmonary parenchyma; it can arise from cardiac, lung, and vascular disease. In patients with PAH, the mosaic pattern is more commonly due to vascular disease rather than cardiac or lung disease.17In the patients we describe, many nonvascular causes can be ruled out by history, physical examination, or testing.1819 The mosaic pattern was clearly observed on the chest CT scan of the patient reported by Samuels et al3(Fig 4), who had a histologically proven vasculopathy. It was not seen in patient 1 in the study by Aoki et al2 (who also had a histologically proven vasculopathy); however, the only chest CT scan available for review was not an HRCT scan and was obtained at the onset of the patient’s symptoms 4 years before her death.

Little is known about the frequency, pathogenesis, and natural history of NF1-associated vasculopathies. They primarily affect the renal, cerebral, and peripheral vascular beds, and appear to contribute to the excess mortality of children and young adults with NF1.5 NF1-associated vasculopathies are heterogeneous and affect a variety of arteries of different sizes throughout the body. Proposed mechanisms to explain their pathogenesis include (1) altered vascular histogenesis, (2) altered vascular maintenance and repair, (3) somatic mutation elsewhere, and (4) environmental factors.5 Their pathologic schema is evolving. The formulation of Lie4 is the most recent; it categorizes NF1-associated vasculopathies as (1) intimal vascular smooth muscle cell proliferation in large elastic arteries, (2) intimal vascular smooth muscle cell proliferation with fibrosis and neoangiogenesis in medium-sized elastic arteries, or (3) plexiform (or angiomatoid) intimal proliferation in small arteries and arterioles.4 The observed clinical and pathologic heterogeneity of NF1-associated vasculopathies probably reflects heterogeneity of underlying pathogenetic mechanisms.

We report, for the first time, the pathology of the autopsy-obtained lungs of a patient with NF1-PAH. The lesions we observed in the small-sized and medium-sized arteries of the lung were similar to the plexiform intimal proliferation described in patients with NF1-associated vasculopathies. The plexiform arteriopathy (or lesions) we observed in the lungs were more complex than those previously reported in the pancreas and digital arteries of individuals with NF1.4 (Plexiform arteriopathy or plexiform lesions should not be confused with plexiform neurofibromas, which are distinctive congenital tumors that are seen in 25 to 40% of patients with NF1.,1)

To our knowledge, the only other description of pulmonary vascular pathology in a patient with NF1 is the patient reported by Samuels et al.3 This patient, who also had NF1-PAH, had extensive irregular thickening of the intima of the pulmonary arteries by fibrous tissue. No plexiform lesions were found in that specimen. This may be due to the limited tissue sampling that is inherent in patients undergoing pulmonary endarterectomy.3 Alternatively, the vasculopathy in that patient may be truly nonplexogenic. If so, this would be consistent with clinical and pathogenetic heterogeneity within NF1-PAH itself, a feature that has been observed in other NF1-associated vasculopathies.

Pulmonary plexiform lesions are a hallmark of plexogenic pulmonary arteriopathy and are associated with several severe types of pulmonary hypertension.2021 The pathogenesis of these lesions is comparatively better understood than the origins of NF1-associated vasculopathies. Plexogenic pulmonary arteriopathy may feature either monoclonal or polyclonal endothelial cell proliferation.22 Monoclonal plexiform lesions are observed in patients with IPAH and PAH secondary to appetite suppressant use and HIV infection. They are hypothesized to arise secondary to an acquired selective growth advantage of a cell, presumably from somatic mutation. Polyclonal plexiform lesions are observed in patients with pulmonary hypertension secondary to congenital heart disease and schistosomiasis, and reflect multicellular endothelial proliferation due to shear stress or inflammation.22 Endothelial cell proliferation, whether monoclonal or polyclonal, generates cytokines and growth factors, inflammatory mediators, vascular remodeling, and, ultimately, dysregulated angiogenesis. In this scheme, the primary event is the dysregulated, proliferating endothelial cell, which then prompts an exuberant and pathologic vascular smooth muscle cell response.7,23 In our lung specimen, the disorganized, near-perpendicular smooth muscle fibers in the media of the smaller arteries (Fig 5, top left, a) testify to this dysfunctional process.

The patients in this report with NF1-PAH lacked either secondary causes or BMPR2 mutations or rearrangements, suggesting that the pathogenesis of NF1-PAH lies elsewhere. Neurofibromin, the protein product of the NF1 gene, and an important regulator of Ras, is expressed in the endothelial cells and vascular smooth muscle cells of many species, including humans.5 Patients with NF1 have a germline mutation in one copy of their NF1 gene and are therefore left with only one functioning (“wild-type”) copy (a state called haploinsufficiency). The loss of that second copy due to deletion or mitotic recombination is called loss of heterozygosity (LOH) and results in abnormal cell growth (eg, in NF1, neurofibromas, and leukemia) secondary to Ras dysregulation.

We previously noted the multiplicity of hypotheses about the pathogenesis of NF1-associated vasculopathies. The pathogenesis of NF1-PAH may also be due to impaired vascular wound repair arising from NF1 haploinsufficiency. Indeed, NF1 haploinsufficiency alone may account for many nontumor manifestations in patients with NF1, including macrocephaly, short stature, and learning disabilities. However, given that the LOH of NF1 is a common pathogenetic mechanism in tumor formation in patients with NF1, we speculate that the pathogenesis of NF1-PAH stems from the LOH of NF1 in endothelial cells, the subsequent dysregulation of the RAS pathway, the monoclonal expansion of the endothelial cells, abnormal vascular cell proliferation, and misguided angiogenesis. (Other mechanisms such as point mutations in the NF1 gene may also cause the functional loss of the wild-type copy of NF1.)

The clonality of endothelial cells in patients with NF1-PAH and other NF1-associated vasculopathies is unknown; this may be due to the difficulty in obtaining appropriate tissue. However, in NF1-associated juvenile myelomonocytic leukemia, the LOH of NF1 results in a monoclonal expansion.2425 The severity of NF1-PAH is akin to other forms of IPAH in which monoclonal plexiform arteriopathy has been documented.7 Mast cells are implicated in the pathogenesis of neurofibromas2627 and in dysregulated angiogenesis in pulmonary allergic disorders28; their role in NF1-PAH remains to be investigated.

In summary, our observations (1) demonstrate the susceptibility of the pulmonary arterial bed to an NF1-associated vasculopathy, (2) suggest a possible mechanism of the pathogenesis of both NF1-PAH and NF1-associated vasculopathies, and (3) broaden the genetic basis of the pathophysiology of PAH. NF1 should be considered as APAH, as outlined in the Revised Clinical Classification of Pulmonary Hypertension).6 We speculate that in patients with NF1-PAH, the LOH of the tumor suppressor gene NF1 in endothelial cells leads to monoclonal endothelial cell proliferation, misguided angiogenesis, and a plexogenic pulmonary arteriopathy. Our observations and hypothesis, if correct, also suggest that pathologic variation in RAS pathway genes may influence the pathogenesis of other forms of PAH. Experimental proof of a monoclonal origin of endothelial cell proliferation and the LOH of NF1 in patients with NF1-PAH would support the “cancer-like” model of PAH pathogenesis in which the loss of tumor suppressor genes leads to pathologic vascular remodeling.,22,29

Abbreviations: ANA = antinuclear antibody; APAH = associated with pulmonary arterial hypertension; BMPR2 = bone morphogenic protein receptor 2; CCB = calcium channel blocker; EF = ejection fraction; HRCT = high-resolution CT; IPAH = idiopathic pulmonary arterial hypertension; LOH = loss of heterozygosity; LV = left ventricle, ventricular; NF1 = neurofibromatosis type 1; NF1-PAH = NF1-associated pulmonary arterial hypertension; OSA = obstructive sleep apnea; PAH = pulmonary arterial hypertension; PAP = pulmonary arterial pressure; PE = pulmonary embolus; PFT = pulmonary function test; PPH = primary pulmonary hypertension; RV = right ventricle, ventricular; TLC = total lung capacity; V̇/Q̇ = ventilation-perfusion

This work was performed at the University of Pennsylvania, the National Human Genome Research Institute, Vanderbilt University Medical Center and University of Alabama at Birmingham.

The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of commercial products or organizations imply endorsement by the US Government.

This research was supported in part by the Intramural Research Program of the National Human Genome Research Institute, National Institutes of Health, and program project grant WCN P01 HL072058.

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.

Table Graphic Jump Location
Table 1. Clinical Evaluation of Patients With NF1 and PAH
* 

LE = lower extremity; U/S = ultrasound; anti-Smith = anti-Smith antibody; SSA = anti-Ro antibody; SSB = anti-La antibody; ESR = erythrocyte sedimentation rate; Dlco = diffusing capacity of the lung for carbon monoxide; F = female; M = male.

Table Graphic Jump Location
Table 2. Summary of Echocardiographic Data of Patients With NF1 and PAH*
* 

All echocardiograms were transthoracic. In cases when quantitative information was not available, qualitative descriptors are quoted directly from report. ND = no data; ITV = interventricular.

Table Graphic Jump Location
Table 3. Summary of Cardiac Catheterization Data of Patients With NF1 and PAH*
* 

Patient 1 had separate right and left heart catheterization procedures; in patient 2 they were combined. Multiple measurements are listed as a range if available. Vasoactive medications were not infused while measurements were made. See Table 2 for abbreviation not used in the text.

Figure Jump LinkFigure 1. Mosaic pattern of lung attenuation on a chest CT scan of patient 1.Grahic Jump Location
Figure Jump LinkFigure 2. Mosaic pattern of lung attenuation on a helical chest CT scan of patient 2.Grahic Jump Location
Figure Jump LinkFigure 3. Mosaic pattern of lung attenuation on a chest HRCT scan of patient 4.Grahic Jump Location
Table Graphic Jump Location
Table 4. Summary of Clinical Features of Patients With NF1 and PAH*
* 

SNP = single-nucleotide polymorphism; ACE-I = angiotensin-converting enzyme inhibitor; atx = atelectasis; PA = pulmonary artery.

Figure Jump LinkFigure 4. Mosaic pattern of lung attenuation on a chest HRCT scan of the patient in the study by Samuels et al.3Grahic Jump Location
Figure Jump LinkFigure 5. Complex plexiform lesions in NF1-PAH. Hematoxylin and eosin-stained lung preparations (original × 160 for all panels) from autopsy samples of patient 1 in the study by Aoki et al.2 The bar represents 100 μm. Top left, a: an artery (approximately 250 μm in external diameter) with increased medial and intimal thickness. The lumen of the artery is virtually occluded, and the smooth muscle cells of the medial layer seem disorganized. Top right, b: an artery (approximately 170 μm in external diameter) showing increased medial and intimal thickness; the lumen is markedly reduced by the intimal layer. Bottom left, c: an artery (approximately 270 μm in external diameter) showing striking intimal thickening and the appearance of small capillaries in the intima (arrows). Bottom right, d: a complex plexiform lesion.Grahic Jump Location

We thank Caroline Moore, MS, for help with cell cultures, Lisa Wheeler for administrative assistance, Grisel Lopez, MD, and Reiko Horai, PhD, for translations, and Mark Bryant, DVM, for discussions regarding the pathology findings.

Friedman, JM Gutmann, DH MacCollin, Met al eds.Neurofibromatosis: phenotype, natural history and pathogenesis 3rd ed.1999 Johns Hopkins University Press. Baltimore, MD:
 
Aoki, Y, Kodama, M, Mezaki, T, et al von Recklinghausen disease complicated by pulmonary hypertension.Chest2001;119,1606-1608. [PubMed] [CrossRef]
 
Samuels, N, Berkman, N, Milgalter, E, et al Pulmonary hypertension secondary to neurofibromatosis: intimal fibrosis versus thromboembolism.Thorax1999;54,858-859. [PubMed]
 
Lie, JT Vasculopathies of neurofibromatosis type 1 (von Recklinghausen disease).Cardiovasc Pathol1998;7,97-108
 
Hamilton, SJ, Friedman, JM Insights into the pathogenesis of neurofibromatosis 1 vasculopathy.Clin Genet2000;58,341-344. [PubMed]
 
Simonneau, G, Galie, N, Rubin, LJ, et al Clinical classification of pulmonary hypertension.J Am Coll Cardiol2004;43(suppl),5S-12S
 
Zaiman, A, Fijalkowska, I, Hassoun, PM, et al One hundred years of research in the pathogenesis of pulmonary hypertension.Am J Respir Cell Mol Biol2005;33,425-431. [PubMed]
 
Porterfield, JK, Pyeritz, RE, Traill, TA Pulmonary hypertension and interstitial fibrosis in von Recklinghausen neurofibromatosis.Am J Med Genet1986;25,531-535. [PubMed]
 
Garcia Hernández, FJ, Sanchez Román, J, Ocana Medina, C, et al Pulmonary hypertension in a patient with neurofibromatosis.Med Clin (Barc)2002;118,78-79
 
Messiaen, LM, Callens, T, Mortier, G, et al Exhaustive mutation analysis of the NF1 gene allows identification of 95% of mutations and reveals a high frequency of unusual splicing defects.Hum Mutat2000;15,541-555. [PubMed]
 
Cogan, JD, Pauciulo, MW, Batchman, AP, et al High frequency of BMPR2 exonic deletions/duplications in familial pulmonary arterial hypertension.Am J Respir Crit Care Med2006;174,590-598. [PubMed]
 
Cogan, JD, Vnencak-Jones, CL, Phillips, JA, III, et al Gross BMPR2 gene rearrangements constitute a new cause for primary pulmonary hypertension.Genet Med2005;7,169-174. [PubMed]
 
Koehler, R, Grunig, E, Pauciulo, MW, et al Low frequency of BMPR2 mutations in a German cohort of patients with sporadic idiopathic pulmonary arterial hypertension.J Med Genet2004;41,e127. [PubMed]
 
Tuder, RM, Cool, CD, Yeager, M, et al The pathobiology of pulmonary hypertension: endothelium.Clin Chest Med2001;22,405-418. [PubMed]
 
National Institutes of Health.. Neurofibromatosis: conference statement; National Institutes of Health consensus development conference.Arch Neurol1988;45,575-578. [PubMed]
 
Sitbon, O, Humbert, M, Jais, X, et al Long-term response to calcium channel blockers in idiopathic pulmonary arterial hypertension.Circulation2005;111,3105-3111. [PubMed]
 
Sherrick, AD, Swensen, SJ, Hartman, TE Mosaic pattern of lung attenuation on CT scans: frequency among patients with pulmonary artery hypertension of different causes.AJR Am J Roentgenol1997;169,79-82. [PubMed]
 
Worthy, SA, Muller, NL, Hartman, TE, et al Mosaic attenuation pattern on thin-section CT scans of the lung: differentiation among infiltrative lung, airway, and vascular diseases as a cause.Radiology1997;205,465-470. [PubMed]
 
Stern, EJ, Muller, NL, Swensen, SJ, et al CT mosaic pattern of lung attenuation: etiologies and terminology.J Thorac Imaging1995;10,294-297. [PubMed]
 
Runo, JR, Loyd, JE Primary pulmonary hypertension.Lancet2003;361,1533-1544. [PubMed]
 
Pietra, GG, Capron, F, Stewart, S, et al Pathologic assessment of vasculopathies in pulmonary hypertension.J Am Coll Cardiol2004;43(suppl),25S-32S
 
Lee, SD, Shroyer, KR, Markham, NE, et al Monoclonal endothelial cell proliferation is present in primary but not secondary pulmonary hypertension.J Clin Invest1998;101,927-934. [PubMed]
 
Fishman, AP Changing concepts of the pulmonary plexiform lesion.Physiol Res2000;49,485-492. [PubMed]
 
Emanuel, PD Juvenile myelomonocytic leukemia.Curr Hematol Rep2004;3,203-209. [PubMed]
 
Shannon, KM, O’Connell, P, Martin, GA, et al Loss of the normal NF1 allele from the bone marrow of children with type 1 neurofibromatosis and malignant myeloid disorders.N Engl J Med1994;330,597-601. [PubMed]
 
Yang, FC, Ingram, DA, Chen, S, et al Neurofibromin-deficient Schwann cells secrete a potent migratory stimulus for Nf1+/- mast cells.J Clin Invest2003;112,1851-1861. [PubMed]
 
Yang, FC, Chen, S, Clegg, T, et al Nf1+/- mast cells induce neurofibroma like phenotypes through secreted TGF-beta signaling.Hum Mol Genet2006;15,2421-2437. [PubMed]
 
Puxeddu, I, Ribatti, D, Crivellato, E, et al Mast cells and eosinophils: a novel link between inflammation and angiogenesis in allergic diseases.J Allergy Clin Immunol2005;116,531-536. [PubMed]
 
Yeager, ME, Halley, GR, Golpon, HA, et al Microsatellite instability of endothelial cell growth and apoptosis genes within plexiform lesions in primary pulmonary hypertension.Circ Res2001;88,E2-E11. [PubMed]
 

Figures

Figure Jump LinkFigure 1. Mosaic pattern of lung attenuation on a chest CT scan of patient 1.Grahic Jump Location
Figure Jump LinkFigure 2. Mosaic pattern of lung attenuation on a helical chest CT scan of patient 2.Grahic Jump Location
Figure Jump LinkFigure 3. Mosaic pattern of lung attenuation on a chest HRCT scan of patient 4.Grahic Jump Location
Figure Jump LinkFigure 4. Mosaic pattern of lung attenuation on a chest HRCT scan of the patient in the study by Samuels et al.3Grahic Jump Location
Figure Jump LinkFigure 5. Complex plexiform lesions in NF1-PAH. Hematoxylin and eosin-stained lung preparations (original × 160 for all panels) from autopsy samples of patient 1 in the study by Aoki et al.2 The bar represents 100 μm. Top left, a: an artery (approximately 250 μm in external diameter) with increased medial and intimal thickness. The lumen of the artery is virtually occluded, and the smooth muscle cells of the medial layer seem disorganized. Top right, b: an artery (approximately 170 μm in external diameter) showing increased medial and intimal thickness; the lumen is markedly reduced by the intimal layer. Bottom left, c: an artery (approximately 270 μm in external diameter) showing striking intimal thickening and the appearance of small capillaries in the intima (arrows). Bottom right, d: a complex plexiform lesion.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1. Clinical Evaluation of Patients With NF1 and PAH
* 

LE = lower extremity; U/S = ultrasound; anti-Smith = anti-Smith antibody; SSA = anti-Ro antibody; SSB = anti-La antibody; ESR = erythrocyte sedimentation rate; Dlco = diffusing capacity of the lung for carbon monoxide; F = female; M = male.

Table Graphic Jump Location
Table 2. Summary of Echocardiographic Data of Patients With NF1 and PAH*
* 

All echocardiograms were transthoracic. In cases when quantitative information was not available, qualitative descriptors are quoted directly from report. ND = no data; ITV = interventricular.

Table Graphic Jump Location
Table 3. Summary of Cardiac Catheterization Data of Patients With NF1 and PAH*
* 

Patient 1 had separate right and left heart catheterization procedures; in patient 2 they were combined. Multiple measurements are listed as a range if available. Vasoactive medications were not infused while measurements were made. See Table 2 for abbreviation not used in the text.

Table Graphic Jump Location
Table 4. Summary of Clinical Features of Patients With NF1 and PAH*
* 

SNP = single-nucleotide polymorphism; ACE-I = angiotensin-converting enzyme inhibitor; atx = atelectasis; PA = pulmonary artery.

References

Friedman, JM Gutmann, DH MacCollin, Met al eds.Neurofibromatosis: phenotype, natural history and pathogenesis 3rd ed.1999 Johns Hopkins University Press. Baltimore, MD:
 
Aoki, Y, Kodama, M, Mezaki, T, et al von Recklinghausen disease complicated by pulmonary hypertension.Chest2001;119,1606-1608. [PubMed] [CrossRef]
 
Samuels, N, Berkman, N, Milgalter, E, et al Pulmonary hypertension secondary to neurofibromatosis: intimal fibrosis versus thromboembolism.Thorax1999;54,858-859. [PubMed]
 
Lie, JT Vasculopathies of neurofibromatosis type 1 (von Recklinghausen disease).Cardiovasc Pathol1998;7,97-108
 
Hamilton, SJ, Friedman, JM Insights into the pathogenesis of neurofibromatosis 1 vasculopathy.Clin Genet2000;58,341-344. [PubMed]
 
Simonneau, G, Galie, N, Rubin, LJ, et al Clinical classification of pulmonary hypertension.J Am Coll Cardiol2004;43(suppl),5S-12S
 
Zaiman, A, Fijalkowska, I, Hassoun, PM, et al One hundred years of research in the pathogenesis of pulmonary hypertension.Am J Respir Cell Mol Biol2005;33,425-431. [PubMed]
 
Porterfield, JK, Pyeritz, RE, Traill, TA Pulmonary hypertension and interstitial fibrosis in von Recklinghausen neurofibromatosis.Am J Med Genet1986;25,531-535. [PubMed]
 
Garcia Hernández, FJ, Sanchez Román, J, Ocana Medina, C, et al Pulmonary hypertension in a patient with neurofibromatosis.Med Clin (Barc)2002;118,78-79
 
Messiaen, LM, Callens, T, Mortier, G, et al Exhaustive mutation analysis of the NF1 gene allows identification of 95% of mutations and reveals a high frequency of unusual splicing defects.Hum Mutat2000;15,541-555. [PubMed]
 
Cogan, JD, Pauciulo, MW, Batchman, AP, et al High frequency of BMPR2 exonic deletions/duplications in familial pulmonary arterial hypertension.Am J Respir Crit Care Med2006;174,590-598. [PubMed]
 
Cogan, JD, Vnencak-Jones, CL, Phillips, JA, III, et al Gross BMPR2 gene rearrangements constitute a new cause for primary pulmonary hypertension.Genet Med2005;7,169-174. [PubMed]
 
Koehler, R, Grunig, E, Pauciulo, MW, et al Low frequency of BMPR2 mutations in a German cohort of patients with sporadic idiopathic pulmonary arterial hypertension.J Med Genet2004;41,e127. [PubMed]
 
Tuder, RM, Cool, CD, Yeager, M, et al The pathobiology of pulmonary hypertension: endothelium.Clin Chest Med2001;22,405-418. [PubMed]
 
National Institutes of Health.. Neurofibromatosis: conference statement; National Institutes of Health consensus development conference.Arch Neurol1988;45,575-578. [PubMed]
 
Sitbon, O, Humbert, M, Jais, X, et al Long-term response to calcium channel blockers in idiopathic pulmonary arterial hypertension.Circulation2005;111,3105-3111. [PubMed]
 
Sherrick, AD, Swensen, SJ, Hartman, TE Mosaic pattern of lung attenuation on CT scans: frequency among patients with pulmonary artery hypertension of different causes.AJR Am J Roentgenol1997;169,79-82. [PubMed]
 
Worthy, SA, Muller, NL, Hartman, TE, et al Mosaic attenuation pattern on thin-section CT scans of the lung: differentiation among infiltrative lung, airway, and vascular diseases as a cause.Radiology1997;205,465-470. [PubMed]
 
Stern, EJ, Muller, NL, Swensen, SJ, et al CT mosaic pattern of lung attenuation: etiologies and terminology.J Thorac Imaging1995;10,294-297. [PubMed]
 
Runo, JR, Loyd, JE Primary pulmonary hypertension.Lancet2003;361,1533-1544. [PubMed]
 
Pietra, GG, Capron, F, Stewart, S, et al Pathologic assessment of vasculopathies in pulmonary hypertension.J Am Coll Cardiol2004;43(suppl),25S-32S
 
Lee, SD, Shroyer, KR, Markham, NE, et al Monoclonal endothelial cell proliferation is present in primary but not secondary pulmonary hypertension.J Clin Invest1998;101,927-934. [PubMed]
 
Fishman, AP Changing concepts of the pulmonary plexiform lesion.Physiol Res2000;49,485-492. [PubMed]
 
Emanuel, PD Juvenile myelomonocytic leukemia.Curr Hematol Rep2004;3,203-209. [PubMed]
 
Shannon, KM, O’Connell, P, Martin, GA, et al Loss of the normal NF1 allele from the bone marrow of children with type 1 neurofibromatosis and malignant myeloid disorders.N Engl J Med1994;330,597-601. [PubMed]
 
Yang, FC, Ingram, DA, Chen, S, et al Neurofibromin-deficient Schwann cells secrete a potent migratory stimulus for Nf1+/- mast cells.J Clin Invest2003;112,1851-1861. [PubMed]
 
Yang, FC, Chen, S, Clegg, T, et al Nf1+/- mast cells induce neurofibroma like phenotypes through secreted TGF-beta signaling.Hum Mol Genet2006;15,2421-2437. [PubMed]
 
Puxeddu, I, Ribatti, D, Crivellato, E, et al Mast cells and eosinophils: a novel link between inflammation and angiogenesis in allergic diseases.J Allergy Clin Immunol2005;116,531-536. [PubMed]
 
Yeager, ME, Halley, GR, Golpon, HA, et al Microsatellite instability of endothelial cell growth and apoptosis genes within plexiform lesions in primary pulmonary hypertension.Circ Res2001;88,E2-E11. [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

CHEST Journal Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543