Do Patients With Primary Pulmonary Hypertension Develop Extensive Central Thrombi?
Background Distinguishing chronic major vessel thromboembolic pulmonary hypertension from primary pulmonary hypertension is critical because the treatment options differ markedly. Surgical thromboendarterectomy is potentially curative in the former condition, whereas oxygen, vasodilators, perhaps anticoagulation, and lung transplantation are the options for the latter. The development of large thrombi in the main, right, or left pulmonary arteries has not been previously described in patients with primary pulmonary hypertension.
Methods and Results Three pulmonary hypertensive patients with massive thrombi in the central pulmonary arteries are described. The data indicate that the large central thrombi in these three patients were not hemodynamically significant. In none did perfusion lung scans demonstrate segmental or larger defects.
Conclusions Large central thrombi can develop in patients with primary pulmonary hypertension. Perfusion lung scans that do not demonstrate segmental or larger defects should alert physicians to this possibility. Chest computed tomography and other studies identifying such thrombi are not adequate in distinguishing such a development from operable chronic major vessel thromboembolic hypertension. Careful review of lobar and segmental artery findings and the pulmonary angiogram, angioscopy, and cardiac catheterization data demonstrating the hemodynamic significance (or lack thereof) of these thrombi are essential in making this important distinction. Furthermore, these observations may constitute an additional indication for anticoagulant therapy in primary pulmonary hypertension.
Differentiating between primary pulmonary hypertension (PPH) and chronic major vessel thromboembolic pulmonary hypertension (CTEPH) has considerable importance because of the very different therapeutic approaches to these two disorders. In PPH, therapy involves vasodilator drugs, oxygen, consideration of anticoagulant therapy, and lung transplantation.1 2 CTEPH is potentially curable by surgical thromboendarterectomy.3 4
Extensive experience with both entities has indicated that the perfusion lung scan is a key test in differential diagnosis.1 4 5 6 In CTEPH, the scan has consistently shown one or more segmental or lobar-sized defects. In PPH, the scan is either normal or shows mottling: that is, subsegmental areas of reduced to absent perfusion. This scan differential has been so consistent that our centers do not perform pulmonary angiography in patients with normal or mottled lung scans, a position also adopted by the National Heart, Lung, and Blood Institute Primary Pulmonary Hypertension Registry.1 6 Furthermore, at the University of California, San Diego (USCD), anticoagulant therapy is not routinely given to patients with primary pulmonary hypertension unless right ventricular failure, immobility, or other embolic risk factors develop, while others recommend this therapy for all patients with PPH.7
Recently, we encountered three patients whose case histories raised issues about the potential fallibility of the perfusion lung scan in making the differential diagnoses and the decisions regarding anticoagulant therapy. We felt that this report might be useful to physicians attempting to differentiate between PPH complicated by thrombosis of the central pulmonary arteries and CTEPH and deciding on appropriate therapy.
A 75-year-old white female first came to UCSD Medical Center for consultation regarding two problems: severe pulmonary hypertension and a mass in the right kidney. History disclosed that the patient had first developed dyspnea on exertion and easy fatigue in mid-1981. Interval history was sketchy. In mid-1988, dyspnea increased slightly. Pulmonary spirometry disclosed mild restriction and a low diffusing capacity for carbon dioxide (Dco). The chest x-ray film disclosed a large heart, large central pulmonary arteries, and possible scattered pulmonary infiltrates. A resting arterial blood gas analysis disclosed a Pao2 in the low 40s (mm Hg). Probable diffuse interstitial pneumonitis was diagnosed, and the patient was treated with oral corticosteroids without improvement.
In November 1989, another physician obtained a cardiac echocardiogram disclosing an enlarged right atrium and ventricle; estimated pulmonary artery systolic pressure was 80 mm Hg, and left ventricular size and function were normal, as was a perfusion lung scan. A chest computed tomogram, with and without contrast, showed no interstitial fibrosis but confirmed bilateral pulmonary artery, right atrial, and ventricular enlargement. Inadvertent extension of the chest computed tomogram to the abdomen disclosed a mass in the right kidney. An aortogram revealed a pattern compatible with hypernephroma in the right kidney. Right heart catheterization disclosed a pulmonary artery pressure of 100/50 mm Hg (mean, 64 mm Hg), wedge pressure of 5 mm Hg, and cardiac output above 4 L/min. A balloon-occlusion pulmonary angiogram was reported as normal.
The patient was referred to UCSD for consultation in August 1990. She was well developed and not dyspneic at rest. Blood pressure was 150/70 mm Hg and pulse was 80. The jugular venous pulse was flat. There was no adenopathy. The diaphragms descended well and the lungs were clear. Cardiac exam revealed a right ventricular tap, palpable P2, fixed splitting of S2, and a 1/6 tricuspid regurgitant murmur. Examination of the abdomen and lower extremities was normal. A routine chest x-ray film in June 1984 was normal. The next, in February 1989, disclosed right ventricular cardiomegaly, very large central pulmonary arteries, and clear lung fields. This appearance persisted through August 1990.
Spirometry disclosed mild restriction and normal flow rates. The diffusing capacity for carbon monoxide was 80% of predicted. An arterial blood gas analysis (rest, room air) disclosed a Pao2 of 46 mm Hg, a Paco2 of 34 mm Hg, and a pH of 7.42. Echocardiogram revealed an enlarged right atrium and ventricle and an estimated pulmonary artery systolic pressure of 90 mm Hg. Review of the previous computed tomographic scan with contrast disclosed no parenchymal disease and dilated pulmonary arteries. A repeat perfusion lung scan disclosed no abnormalities other than an “imprint” of the markedly enlarged left pulmonary artery. Impedance plethysmography was negative.
At left nephrectomy in September 1990, a well-encapsulated hypernephroma was found. Abdominal node biopsies were negative. Postoperatively the patient was begun on oral diltiazem. She was followed up clinically and by echocardiography at intervals of 6 to 9 months and remained stable. In April 1993, another chest computed tomogram with and without contrast disclosed intraluminal masses in the right and left main pulmonary arteries. She was begun on Coumadin. A repeat chest computed tomogram in June 1993 was unchanged (Fig 1⇓).
On reevaluation at UCSD in July 1993, her physical exam was unchanged. Chest x-ray film, perfusion lung scan, and impedance plethysmography and echocardiogram showed no change from previous studies. Right heart catheterization disclosed right atrial mean pressure of 11 mm Hg, a right ventricular pressure of 95/15 mm Hg, pulmonary artery pressure of 100/49 mm Hg (mean, 69 mm Hg), wedge pressure of 13 mm Hg, and cardiac output of 4.05 L/min: all measurements were virtually identical to those made in November 1989. Furthermore, the pressures in the central main pulmonary artery were identical to those in the right and left descending arteries. Pulmonary angiography did not define the defects in the main pulmonary arteries, but there were small defects in the proximal right and left descending pulmonary arteries. The lobar and segmental arteries were normal. The distal pulmonary arterial tree was severely “pruned.”
Pulmonary angioscopy was performed. The main right and left pulmonary arteries were too capacious to permit adequate visualization. However, all lobar and segmental branches were visualized and disclosed no acute or chronic thrombi.
The patient was discharged and given Coumadin and diltiazem therapy. Her clinical deterioration slowly progressed with increasing right ventricular failure. She died in October 1993. Permission for autopsy was not granted.
A 51-year-old man came to the UCSD Medical Center for evaluation in July 1993 for potential pulmonary thromboendarterectomy. He had first experienced dyspnea on exertion in early 1986 and had undergone a right heart catheterization that disclosed pulmonary hypertension. No therapy was instituted.
His dyspnea on exertion slowly increased, and in early 1993 he developed exertional near-syncope and peripheral edema. In March 1993, on admission to another medical center, chest x-ray film disclosed massive cardiomegaly and large central pulmonary arteries. Spirometry disclosed a moderate restrictive defect and a Dco 46% of predicted. An arterial blood gas analysis disclosed a Pao2 of 67 mm Hg, a Paco2 of 37 mm Hg, and a pH of 7.41. A perfusion lung scan was reported as showing gray zones in the right upper and lower lobes but no total defects. Ventilation scan was normal. A transthoracic echocardiogram disclosed severe dilatation of the right atrium and ventricle, paradoxical septal motion, normal left ventricular function, and tricuspid regurgitation with a Doppler-estimated pulmonary artery systolic pressure of 95 mm Hg. A transesophageal echocardiogram confirmed these findings and demonstrated a mass in the main pulmonary artery extending distally to the right. Bubble study was negative.
On referral to UCSD, he was in New York Heart Association class 4. He had no history of venous thrombosis or pulmonary embolism. Blood pressure was 110/70 mm Hg, pulse was 92, and respirations were 16 per minute. The jugular venous pulse was markedly elevated and pulsatile. The lungs were clear. Cardiac exam disclosed a right ventricular heave, palpable P2, fixed splitting of S2, marked accentuation of P2, and a 3/6 tricuspid regurgitant murmur. No murmur was heard over the pulmonary arteries or the lung fields. The liver was 14 cm and pulsatile. There was 2+ peripheral edema.
A complete blood count and a chemistry battery were normal. Chest x-ray film revealed massive cardiomegaly, large central pulmonary arteries, and clear lung fields. An ECG revealed right axis deviation, right ventricular hypertrophy, and strain. Impedance plethysmography was bilaterally positive. Ultrasound examination disclosed no popliteal or femoral vein thrombus. Spirometry disclosed a restrictive defect, and a Dco was 51% of predicted.
Perfusion and ventilation lung scans were within normal limits except for a defect due to the very large left pulmonary artery (Fig 2⇓). Arterial blood gas analyses were normal at rest and remained so during mild exercise. A chest computed tomographic scan disclosed bilaterally enlarged central pulmonary arteries and a mass (8×3×4 cm3) in the right pulmonary artery. An echocardiogram confirmed the prior findings, including a mass starting just to the right of the pulmonary valve and extending into the right main pulmonary artery.
At right heart catheterization, right atrial mean pressure was 20 mm Hg, right ventricular pressure was 103/23 mm Hg, pulmonary artery pressure was 105/52 mm Hg (mean, 68 mm Hg), wedge pressure was 6 mm Hg, and cardiac output was 2.57 L/min. Calculated pulmonary vascular resistance (PVR) was 1898 dynes · sec · cm−5. There was no difficulty in passing the catheter into the right or left descending pulmonary artery, and there was no gradient from the main pulmonary artery into either descending pulmonary artery.
Pulmonary angiography disclosed a mass in the right main pulmonary artery extending into the right descending pulmonary artery and a possible defect in the distal left descending pulmonary artery. There was a sparsity of small vessels (“pruning”) bilaterally (Fig 3⇓). On pulmonary angioscopy, thrombus was visible in the right main pulmonary artery that extended into the right upper lobe, partially occluding it, and into the right descending pulmonary artery. However, the right middle and lower lobe orifices were minimally narrowed. The left upper lobe and lingular arteries were normal. Thrombus slightly narrowed but did not occlude the left descending artery.
Intensive discussions with the patient and his family were held in which a very high risk for thromboendarterectomy was posed and an equivocal extent of improvement was stressed. Lung transplantation was offered. The patient requested thromboendarterectomy.
At surgery, substantial thrombotic material was removed from the right and left main pulmonary arteries with lesser amounts from lobar branches on the right (Fig 4⇓). Pathological examination confirmed that this was organized thrombus. At the conclusion of the procedure, attempts to remove the patient from cardiopulmonary bypass led to right ventricular dilatation and an inadequate cardiac output. After multiple efforts to resuscitate him, the patient died. Permission for postmortem examination was not granted.
In 1982, after her second uneventful full-term pregnancy, a then 28-year-old patient noted onset of mild dyspnea on exertion. She had used oral contraceptives from age 16 to age 24 and had a history of gout and migraine headaches. She underwent a bilateral tubal ligation in 1983. There was no history of venous thrombosis. In 1984 she had a syncopal episode after running upstairs. In 1986 she developed severe dyspnea while vacationing at high altitude, but this subsided on return to sea level.
In April 1987, easy fatigue and increasing dyspnea led to an evaluation. Physical exam disclosed normal vital signs. The lungs were clear. Cardiac exam revealed a right ventricular lift, accentuated P2, and no S3 or S4. Chest x-ray film disclosed modest cardiomegaly and large pulmonary artery shadows, and cardiac echocardiogram revealed an enlarged right ventricle, a flattened interventricular septum, and mild tricuspid regurgitation.
A perfusion scan disclosed decreased perfusion to the left upper lobe and the anterior portion of the left lower lobe but no total perfusion defects. Ventilation scan was normal. Right heart catheterization disclosed right atrial mean pressure of 3 mm Hg; right ventricular pressure was 110/10 mm Hg, pulmonary artery pressure was 105/41 mm Hg (mean, 62 mm Hg), wedge pressure was 8 mm Hg, cardiac output was 3.75 L/min, and pulmonary vascular resistance was 1072 dynes · sec · cm−5. No shunt was demonstrated. An arterial blood gas analysis disclosed a Pao2 of 90 mm Hg, a Paco2 of 35 mm Hg, and a pH of 7.43. The patient declined pulmonary angiography.
During 1992 and 1993, the patient noted progressive dyspnea on exertion, fatigue, and the onset of peripheral edema. In early 1993 she accepted placement in a lung transplant program and treatment with Coumadin was initiated. Her condition continued to deteriorate, and she was admitted in late December 1993. Physical examination disclosed her condition as near-moribund and severely dyspneic; her blood pressure was 70/20 mm Hg, her heart rate was 114 beats per minute, and the jugular venous pulse was distended to the angle of the jaw with large V waves. Cardiac exam revealed a markedly accentuated P2, a right ventricular thrust, and a 5/6 tricuspid regurgitant murmur. Despite intensive supportive therapy, the patient’s condition continued to deteriorate, and she died on January 5, 1994.
At autopsy, the major gross findings included a markedly hypertrophied and dilated right ventricle, a dilated right atrium, and a patent foreman ovale. The right and left pulmonary arteries were markedly dilated. The left pulmonary artery contained a large mural thrombus, 4.0 cm at its greatest diameter, that began at the origin of the left main pulmonary artery and extended approximately 9 cm to the orifice of the left upper lobe branch (Fig 5⇓). There was a channel around this thrombus that fed all branches except the apical posterior branch of the left upper lobe, which was totally occluded. One distal lingular artery (1.3 mm in diameter) contained a recanalized thrombus. All other elastic arteries were patent. There were fatty streaks and yellowish plaques on the intimal surfaces of the central pulmonary arteries. The lungs were grossly normal. The inferior vena cava was dilated but otherwise unremarkable.
Microscopic examination disclosed widespread, marked medial hypertrophy and concentric intimal hyperplasia of the small pulmonary arteries and arterioles. Scattered small pulmonary arteries demonstrated eccentric intimal hyperplasia. Rare plexiform (Fig 6⇓) and numerous dilatation lesions were present.
These three case histories raise several questions regarding the differential diagnosis and management of PPH and CTEPH. First, how reliable is the perfusion lung scan in distinguishing between these two entities? Second, what are the roles of such techniques as pulmonary angiography, chest computed tomography, and echocardiography in making this differentiation? Third, what is the role of anticoagulant therapy in patients with known PPH? Fourth, what are the mechanisms by which patients with both these forms of pulmonary hypertension develop hypertensive changes in the small resistance vessels? And fifth, how can one identify those patients with central thrombi who may benefit from surgical thromboendarterectomy? While all of these questions are raised by these case histories, only the first two can be addressed by the data provided.
The value of the perfusion scan in differentiating PPH from CTEPH was reaffirmed in all three patients. Although two zones of decreased perfusion were seen in patient 3 some 6 years before her death, none of the patients had perfusion scans with a segmental or larger defect. Therefore, at this time, the concept remains intact that a perfusion scan that discloses no segmental or larger defects is diagnostic of PPH without the need for confirmatory pulmonary angiography.1 5 In patient 1, her long history, a previous computed tomographic scan demonstrating no central thrombi, the angiogram demonstrating characteristic pruning, and the angioscopic findings of normal lobar and segmental arteries all suggest that her central thrombi developed years after the development of PPH. In patient 2, the angioscopy did not clearly define sufficient obstruction at the lobar or segmental arteries to explain the degree of pulmonary hypertension, a concern confirmed at surgery. In patient 3, autopsy did not disclose obstruction sufficient to explain her pulmonary hypertension.
These patients also indicate that the diagnostic techniques available to define operable CTEPH must be used with caution. Clearly, the visualization of large defects in the right, left, or both pulmonary arteries by chest computed tomographic scans, with or without contrast, does not ensure a successful outcome of pulmonary thromboendarterectomy. Thromboendarterectomy was clearly contraindicated in patient 1, whose severe pulmonary hypertension long antedated the central artery abnormalities; it was also contraindicated in this patient by the absence of a pressure gradient at catheterization and the normal angioscopic findings. In patient 2, the striking computed tomographic and angiographic findingsas well as the outcome after removal of the central thrombus again indicate that computed tomographic defects in the main pulmonary arteries alone cannot be used to make a surgical decision. In patient 3, had studies defined the large left main pulmonary artery thrombus, they again would not have provided an adequate indication for thromboendarterectomy. The same caution must be applied to transesophageal echocardiographic or magnetic resonance imaging studies that reveal main pulmonary artery thrombi but do not define the status of lobar or segmental arteries.
These case histories also raise the issue of the role of anticoagulant therapy in PPH. Had in situ thrombosis developed in the course of PPH, or did these patients suffer pulmonary embolism? The weight of evidence favors the former conclusion. None had a known episode of venous thrombosis. Only one, patient 3, had received anticoagulant therapy prior to the discovery of these central thrombi, and it is not clear whether the thrombi were present before or after such therapy was instituted. It is not apparent, therefore, whether anticoagulant therapy might have prevented the central thrombi or, more importantly, whether such prevention would have altered outcomes. The absence of a gradient from the main pulmonary artery to each descending pulmonary artery in two patients suggests that these large thrombi were not hemodynamically significant. However, should such central thrombi be detected in a patient with PPH, they are a clear indication for anticoagulant therapy if this has not previously been instituted.
In patient 2 it was apparent from the lack of hemodynamic improvement that the removal of these thrombi was not of value. That was the source of our concern in submitting him to surgery: ie, that the basis of his pulmonary hypertension lay in the small (resistance) arteries. This supposition remains unproven in the absence of an autopsy. In patient 3, the autopsy findings indicated that the central thrombus was an incidental problem. Therefore, we propose that the thrombi in all three patients developed in situ, engrafted upon intimal injury induced by long-standing pulmonary hypertension. Furthermore, given the known distribution of pulmonary emboli (favoring both lower lobes), the limitation of emboli to the main central pulmonary arteries would be most unusual. With respect to hypertensive changes in the resistance arteries, patients with CTEPH demonstrate the same lesions as those seen in PPH.8 The mechanisms responsible remain unknown. However, despite changes in the resistance arteries, there is a successful outcome after thromboendarterectomy in more than 90% of patients in whom lobar and segmental obstructions are relieved, despite lung biopsies demonstrating these lesions. The explanation of this apparent paradox must lie in the extent and severity of these pulmonary hypertensive lesions.
The observations made in these and other patients lead to the final question: how can one reliably identify patients who will benefit from thromboendarterectomy? Unfortunately, at this juncture, as demonstrated in the case of patient 2, definitive answers are not available for every patient. Lung biopsy cannot be relied upon, even if this were practical. The history and previous studies, as in patient 1, may provide the answer for some. Angiographic demonstration of diffuse pruning may be instructive, but the presence of more proximal vessel obstructions may make interpretation of pruning difficult, as in patient 2. The ultimate role of angioscopy remains to be defined as further experience is gained. Clearly, then, at this time there is no fail-safe method for identifying patients, like patient 2, who will not benefit from surgical intervention. It is hoped that future investigations will provide the answer. In the meantime, it is clear that demonstration of large thrombi in the main, right, or left pulmonary artery alone (by any technique) is not adequate. Accessible obstructions at the lobar or segmental level sufficient to explain the extent of pulmonary hypertension must be identified.
This study was supported in part by National Institutes of Health grants HL-07022 and HL-23584.
- Received May 24, 1994.
- Accepted August 19, 1994.
- Copyright © 1995 by American Heart Association
Rich S, Dantzker DR, Ayres SM, et al. Primary pulmonary hypertension: a national prospective study. Ann Intern Med. 1987;107:216-223.
Moser KM, Fedullo PF, Auger WR, Jamieson SW. Chronic thromboembolic pulmonary hypertension: clinical picture and surgical treatment. Eur Respir J. 1992;5:334-343.
Moser KM, Auger WR, Fedullo PF. Chronic major vessel thromboembolic pulmonary hypertension. Circulation. 1990;81:1735-1743.
D’Alonzo GE, Barst, Ayres SM, et al. Survival in patients with primary pulmonary hypertension. Ann Intern Med. 1991;115:343-349.
Rich S, Kaufman E, Levy S. The effect of high doses of calcium channel blockers on survival in primary pulmonary hypertension. N Engl J Med. 1992;317:76-81.