INTRODUCTION

TOP ABSTRACT INTRODUCTION CASE REPORT DISCUSSION REFERENCES

Ventilation/perfusion (/) lung scanning is an integral part of the evaluation of a patient with pulmonary hypertension (1). Multiple segmental perfusion defects usually indicate proximal thromboemboli as the cause of pulmonary hypertension (2). Rarely, a "high-probability" / scan is caused by other processes obstructing the pulmonary arterial tree, such as fibrosing mediastinitis, pulmonary artery sarcoma, and vasculitis (3). In all of these processes, pulmonary arteriography and chest computed tomography (CT) will confirm the presence of proximal obstructing lesions.

In this paper we describe three patients in whom marked discrepancies were noted between findings on / scans and those in pulmonary arteriograms. Two of three patients had biopsy-proven pulmonary venoocclusive disease (PVOD) as the cause of pulmonary hypertension; the other patient had PVOD diagnosed by pulmonary venography. These cases emphasize the importance of considering alternative causes for a high-probability / scan when pulmonary angiography fails to disclose evidence of pulmonary thromboembolism.

	CASE REPORT

TOP ABSTRACT INTRODUCTION CASE REPORT DISCUSSION REFERENCES

Patient 1

A 47-yr-old male had been well until 1 yr before referral, when he presented with rapidly worsening dyspnea. As part of his initial evaluation, he underwent a / scan that revealed multiple mismatched segmental perfusion defects. An open-lung biopsy was performed because of progressive pulmonary infiltrates, and demonstrated pulmonary arteriopathy with tortuous, dilated pulmonary veins. Upon referral, the patient underwent a repeat / scan, which confirmed multiple mismatched segmental perfusion defects involving the right upper lobe, right lower lobe, lingula, and part of the left lower lobe (Figure 1). Cardiac catheterization disclosed a right atrial pressure (Pra) of 6 mm Hg and a pulmonary artery pressure (Ppa) of 85/35 mm Hg (mean: 60 mm Hg). The pulmonary capillary wedge pressure (Ppcw) was 8 mm Hg and 12 mm Hg in the right and left pulmonary arteries, respectively. The cardiac output (CO) was 5.0 L/min, and the pulmonary vascular resistance (Rpv) was 832 dynes · s/cm⁵. Pulmonary angiography, done according to standard technique with separate injections of 55 to 60 ml of contrast medium into each main pulmonary artery, showed smoothly tapering major branches, with peripheral pruning bilaterally (Figures 2A and 2B). In the late phase of the pulmonary arteriogram, the pulmonary veins were extremely narrowed throughout their course. A spiral CT revealed increased interstitial markings, a small left pleural effusion, thickening of the interlobar septa, and attenuated pulmonary veins bilaterally. The patient subsequently underwent heart-lung transplantation. Histologic examination of explanted tissue was diagnostic for PVOD.

View larger version (81K): [in this window] [in a new window]   Figure 1.   Perfusion lung scan for Case 1, showing segmental perfusion defects involving the right upper lobe, right middle lobe, right lower lobe, lingula, and part of the left lower lobe. Ventilation scan (not shown) was normal.

View larger version (85K): [in this window] [in a new window]   View larger version (92K): [in this window] [in a new window]   Figure 2.   (A and B) Pulmonary angiogram for Case 1, showing smoothly tapering major pulmonary branches with peripheral pruning on both sides. There is no evidence of thromboembolic disease.

Patient 2

A 34-yr-old man with a history of mild systemic hypertension was in good health until 4 mo before admission, when he developed dyspnea on exertion of insidious onset. The dyspnea progressed rapidly and the patient was admitted to a local hospital, where he was found to have evidence of pulmonary hypertension on echocardiography and segmental mismatched perfusion defects on / scanning. Over the following 2 mo the patient's condition deteriorated rapidly and he was referred to this center for evaluation for pulmonary thromboendarterectomy. A repeat / scan was remarkable for multiple segmental and subsegmental mismatched perfusion defects involving both lungs (Figure 3). A chest CT showed a mosaic perfusion pattern, eccentric pleural thickening, prominent hilar and mediastinal lymphadenopathy, and marked enlargement of the central pulmonary arteries. Cardiac catheterization showed a Pra of 4 mm Hg, Ppa of 90/34 mm Hg (mean: 57 mm Hg), Ppcw of 5 mm Hg, cardiac index (CI) of 3.2 L/min/m², and Rpv of 1,291 dynes · s/cm⁵. No intracardiac shunt or anomalous pulmonary venous return was detected. Pulmonary angiography showed enlarged central pulmonary arteries and pruning of the vessels distally (Figures 4A and 4B). No thromboemboli were detected on angiography or angioscopy. Inhaled NO at 20 ppm was given for 10 min, and repeat hemodynamics showed a Pra of 0 mm Hg, Ppa of 57/24 mm Hg (mean: 38 mm Hg), Ppcw of 7 mm Hg, CI of 2.7 L/min/m², and Rpv of 917 dynes · s/cm⁵. Over the following several days, the patient became increasingly hypoxemic and ultimately expired from cardiac failure. The results of autopsy confirmed the diagnosis of PVOD.

View larger version (87K): [in this window] [in a new window]   Figure 3.   Perfusion lung scan for Case 2, showing multiple mismatched segmental and subsegmental perfusion defects involving both lungs. Ventilation scan (not shown) was normal.

View larger version (101K): [in this window] [in a new window]   View larger version (94K): [in this window] [in a new window]   Figure 4.   (A and B) Pulmonary angiogram for Case 2, showing enlarged central pulmonary arteries and pruning of vessels distally. There is no evidence of thromboembolic disease.

Patient 3

A 38-yr-old woman with a history of hypothyroidism was well until 5 yr before presentation to our center, when she developed dyspnea during a pregnancy and was treated for a presumptive pulmonary embolism. Three years later she noted worsening exertional dyspnea. Further evaluation at that time included a cardiac catheterization showing moderate pulmonary hypertension and pulmonary angiography showing no evidence of proximal thromboemboli. She had an acute pulmonary vasodilator response to 20 ppm of inhaled NO and subsequently began treatment with diltiazem. Over the next several months her dyspnea and hypoxemia worsened, and she was referred to our center for evaluation for possible pulmonary thromboendarterectomy. A / scan was remarkable for multiple segmental and subsegmental mismatched perfusion defects, most prominent in the right upper lobe and bilateral bases (Figure 5). Cardiac catheterization showed a Pra of 2 mm Hg, Ppa of 47/20 mm Hg (mean: 26 mm Hg), Ppcw of 7 mm Hg, CO of 5.34 L/min, and Rpv of 284 dynes · s/cm⁵. With exercise, the Ppa increased to 80/33 mm Hg (mean: 50 mm Hg). The Ppcw was 10 mm Hg, the Rpv was 334 dynes · s/cm⁵, and the CO was 9.58 L/min. Pulmonary angiography revealed significant peripheral pruning of the left-sided vessels and a minor irregularity at the takeoff of the right middle-lobe vessel, but no evidence of pulmonary thromboemboli (Figures 6A and 6B). Pulmonary angioscopy confirmed the absence of thromboembolic disease. A chest CT showed a mosaic pattern of lung attenuation, enlarged central pulmonary arteries, areas of ground-glass opacity, and mediastinal adenopathy. On the basis of these results the patient was given a presumptive diagnosis of PVOD. Chronic inhaled NO therapy was begun, and the patient is doing well 2 yr after discharge. A pulmonary venogram performed during this 2-yr period showed several sites of stenosis in the pulmonary veins. Balloon dilatation of several of these stenoses reduced the patient's dyspnea.

View larger version (80K): [in this window] [in a new window]   Figure 5.   Perfusion scan for Case 3, showing multiple segmental and subsegmental mismatched perfusion defects, most prominent in the right upper lobe and bilateral bases. Ventilation scan (not shown) was normal.

View larger version (81K): [in this window] [in a new window]   View larger version (76K): [in this window] [in a new window]   Figure 6.   (A and B) Pulmonary angiogramy for Case 3, showing significant peripheral pruning of the left-sided vessels and a minor irregularity at the takeoff of the right middle lobe vessel, but no evidence of pulmonary thromboembolic disease.

	DISCUSSION

TOP ABSTRACT INTRODUCTION CASE REPORT DISCUSSION REFERENCES

In this report we describe three patients with pulmonary hypertension and high-probability / scans resulting from PVOD. PVOD is a rare but increasingly recognized cause of pulmonary hypertension (6). The prognosis in PVOD is usually poor, with severe, progressive pulmonary hypertension and right ventricular failure usually occurring within 2 yr of diagnosis (7). Options for medical therapy other than lung transplantation have been disappointing. Intravenous epoprostenol, which has clear efficacy in primary pulmonary hypertension (PPH), has been reported to cause acute pulmonary edema and even death in patients with PVOD. Oral nifedipine, high-dose prostacyclin, inhaled NO, and aerosolized iloprost have been described as having beneficial hemodynamic effects in some patients (8). Distinguishing PVOD from both PPH and chronic thromboembolic disease, a surgically curable form of pulmonary hypertension (11, 12), is therefore critical.

Chest radiographic findings in PVOD include enlargement of the right side of the heart, dilated central pulmonary arteries, pleural effusions, and signs of pulmonary edema with bilateral increased interstitial markings and Kerley B lines (13, 14). Mediastinal adenopathy caused by vascular congestion may also be seen (15). Swensen and colleagues described common chest CT findings in eight patients with PVOD, which included smooth interlobular septal thickening, diffuse multifocal regions of ground-glass opacity, pleural effusions, a dilated main pulmonary artery and central pulmonary arteries, pulmonary veins of normal caliber, and a mosaic pattern of lung attenuation (16). Pulmonary angiography has been described as showing dilated patent pulmonary arteries, a prolonged circulation time, and a normal left atrium and pulmonary veins (13).

/ lung scanning plays a pivotal role in the diagnostic approach to patients with pulmonary hypertension. In PPH (and other forms of small-vessel pulmonary hypertension), / scans are characteristically normal or may demonstrate a mottled appearance (17). In chronic thromboembolic pulmonary hypertension, segmental or larger perfusion defects are invariably present. In addition to thromboembolism, processes including pulmonary artery sarcoma, vasculitis, and extrinsic compression by mediastinal adenopathy or fibrosis will cause obstruction of segmental or larger arteries and lead to a high-probability / scan. Pulmonary angiography in these cases will confirm occlusion or extrinisic compression of the pulmonary arterial branches.

In PVOD, / scans are frequently reported to be normal (13). Other reported findings have included diffuse, patchy distribution of tracer material without clear segmental or subsegmental defects, and unilateral absence of perfusion due to severe asymmetric involvement of the major pulmonary veins (17). To our knowledge, this is the first report of multiple segmental mismatched perfusion defects with negative pulmonary angiograms in patients with PVOD.

In the third case, in which lung biopsy was not performed, the diagnosis of PVOD was initially suggested on clinical grounds. This patient's chest CT had several findings reported to be consistent with the diagnosis of PVOD, including a mosaic pattern of lung attenuation, enlarged central pulmonary arteries, areas of smooth interlobular septal thickening, and mediastinal adenopathy (16, 20). In addition, the pulmonary venogram demonstrated focal venous obstructions, a finding highly suggestive of the diagnosis of PVOD. The patient in this case has been treated with chronic inhaled NO for 2 yr, and has continued to do well.

Other potential causes of pulmonary venous obstruction that could result in the abnormalities observed in the / scan in the third case include extrinsic compression of the pulmonary veins or pulmonary capillary hemangiomatosis (PCH) (21). Although PCH could not be definitively ruled out in this case without histologic examination of lung tissue, extrinsic compression of the pulmonary veins was excluded by chest CT.

One hypothesis that could explain the discordance between the findings in the / scan and pulmonary angiogram in PVOD relates to the increased downstream resistance caused by the narrowing and obliteration of pulmonary veins and venules. Since the distribution of particles in the lungs is proportional to regional pulmonary blood flow, high downstream resistance could reduce tracer deposition in the precapillary arterioles upstream of the venous occlusion, resulting in a mismatched perfusion defect in the / scan. That these areas fail to demonstrate corresponding defects in the pulmonary angiogram could be explained by the higher pressure used in injecting contrast medium into the pulmonary arteries, which would overcome the increased downstream resistance.

Why some cases of PVOD show this / scan abnormality while others do not is unclear. Perhaps disease resulting in more extensive involvement of larger veins leads to segmental defects, whereas diffuse involvement of smaller venules produces the more frequently described / scan abnormality of diffuse patchy distribution of tracer material. This hypothesis is supported by the findings on pulmonary venography described in the third case, with extensive involvement of larger veins.

We found one report of unilateral absence of perfusion associated with stenosis involving the main pulmonary veins in PVOD (19). In contrast to our patients, this patient did not undergo pulmonary angiography, and pulmonary artery embolic disease could therefore not be excluded.

The findings in this report have important clinical implications. Systemic vasodilators such as prostacyclin have been reported to cause pulmonary edema and even death in patients with PVOD (22, 23). Therefore, a high index of suspicion for PVOD should be present when a discrepancy exists between the / scan and pulmonary angiogram.

In conclusion, a high-probability / scan in a patient with pulmonary hypertension does not always indicate a proximal pulmonary arterial process; when coupled with an angiogram showing no arterial obstruction, this finding is probably due to a focal "downstream" process such as PVOD.