You are on page 1of 18


Imaging in Chronic Thromboembolic Pulmonary

Rahul D. Renapurkar, MD,*w Sankaran Shrikanthan, MD,z
Gustavo A. Heresi, MD,y Charles T. Lau, MD, MBA,8 and
Deepa Gopalan, MRCP, FRCRz

(RHC), the natural history of this disease is variable and

Abstract: Chronic thromboembolic pulmonary hypertension depends on its underlying cause.1
(CTEPH) is one of the potentially curable causes of pulmonary Chronic thromboembolic pulmonary hypertension
hypertension and is denitively treated with pulmonary throm- (CTEPH), one of the increasingly recognized causes of PH,
boendartectomy. CTEPH can be overlooked, as its symptoms are
nonspecic and can be mimicked by a wide range of diseases that
follows a single or recurrent event of acute pulmonary
can cause pulmonary hypertension. Early diagnosis of CTEPH and embolism (PE). The exact incidence of this disease is uncer-
prompt evaluation for surgical candidacy are paramount factors in tain, although various studies have suggested that it may
determining future outcomes. Imaging plays a central role in the occur after approximately 0.57% to 3.8% of acute pulmonary
diagnosis of CTEPH and patient selection for pulmonary throm- embolic events and in up to 10% of patients with recurrent
boendartectomy and balloon pulmonary angioplasty. Currently, PE.25 Often, patients present with nonspecic symptoms such
various imaging tools are used in concert, with techniques such as as exertional dyspnea secondary to dead space ventilation and
computed tomography (CT) and conventional pulmonary angiog- increased cardiac demand.6 With improved understanding of
raphy providing detailed structural information, tests such as the disease, there is increasing recognition of a subset of
ventilation-perfusion (V/Q) scanning providing functional data,
and magnetic resonance imaging providing a combination of
patients with chronic thromboembolic disease (CTED) who
morphologic and functional information. Emerging techniques have signs of vascular obstruction and functional impairment
such as dual-energy CT and single photon emission computed but with normal resting mPAP.7 The survival rate without
tomography-CT V/Q scanning promise to provide both anatomic intervention in CTEPH is poor and fairly similar to rates seen
and functional information in a single test and may change the way with other forms of PH.7 Early diagnosis and treatment are
we image these patients in the near future. In this review, we discuss critical in preventing secondary distal arteriolar vasculopathy.
the roles of various imaging techniques and discuss their merits, As the clinical presentation is fairly nonspecic, the diagnosis
limitations, and relative strengths in depicting the structural and of CTEPH requires a high degree of clinical suspicion, par-
functional changes of CTEPH. We also explore newer imaging ticularly early in its course. The greatest diculty lies in dis-
techniques and the potential value they may oer.
tinguishing CTEPH from idiopathic pulmonary arterial
Key Words: chronic thromboembolic pulmonary hypertension, hypertension (IPAH). Although IPAH is predominantly a
computed tomography, ventilation-perfusion scan, magnetic reso- small vessel disease, it can be complicated by in situ throm-
nance imaging bosis resulting in larger central thrombi. In such cases, the
concomitant presence of large vessel and microvascular dis-
(J Thorac Imaging 2017;32:7188) ease in both entities poses a diagnostic challenge.8
Medical options for patients with CTEPH include
anticoagulants, vasodilators, and remodeling agents tar-

P ulmonary hypertension (PH) is a common medical

problem characterized by elevated pulmonary arterial
pressures, which can lead to right ventricular (RV) failure
geted to decrease the pulmonary vascular resistance (PVR).
Surgery, however, remains the denitive treatment in these
patients; pulmonary thromboendarterectomy (PTE) is the
(cor pulmonale). Dened by mean pulmonary artery pressure procedure of choice, with favorable short-term and long-
(mPAP)Z25 mm Hg at rest on right heart catheterization term outcomes.912 Also, there is increasing evidence to
support PTE for patients with symptomatic CTED.13 Two
of the critical factors in identifying good candidates for
surgical intervention are accessibility of the thrombi and
From the *Section of Thoracic Imaging; wCardiovascular Imaging severity of hemodynamic and ventilatory impairment.
Laboratory; zSection of Nuclear Medicine, Imaging Institute;
ySection of Pulmonary and Critical Care Medicine, Cleveland
Recently, minimally invasive techniques such as balloon
Clinic, Cleveland, OH; 8Department of Radiology, VA Palo Alto pulmonary angioplasty (BPA) have been tried with
Health Care System, Palo Alto, CA; and zDepartment of Radiol- increasing success in patients with inoperable CTEPH.14
ogy, Imperial College Hospitals, London, UK. Imaging plays 2 critical roles in patients with CTEPH:
Gustavo Heresi: Bayer Healthcare: speakers bureau, advisory board
and consultant. The remaining authors declare no conicts of
(1) diagnosis and (2) appropriate patient selection for inter-
interest. ventional or surgical therapies. In this review, we evaluate the
Correspondence to: Rahul D. Renapurkar, MD, Section of Thoracic role of various imaging modalities in the diagnosis and pre-
Imaging, L10, Imaging Institute, Cleveland Clinic, Cleveland, OH operative evaluation of CTEPH and also analyze their
44195 (e-mail:
Supplemental Digital Content is available for this article. Direct URL
strengths and limitations.
citations appear in the printed text and are provided in the HTML
and PDF versions of this article on the journals Website, CLASSIFICATION OF PH
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. The current classication presented during the Fifth
DOI: 10.1097/RTI.0000000000000256 World Symposium held in 2013 in Nice, France, divides PH

J Thorac Imaging  Volume 32, Number 2, March 2017 | 71

Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

TABLE 1. Updated Clinical Classification of Pulmonary

Hypertension (Nice, 2013)
1. Pulmonary arterial hypertension
1.1 Idiopathic PAH
1.2 Heritable PAH
1.2.1 BMPR2
1.2.2 ALK-1, ENG, SMAD9, CAV1, KCNK3
1.2.3 Unknown 1.3 Drug and toxin induced
1.4 Associated with:
1.4.1 Connective tissue disease
1.4.2 HIV infection
1.4.3 Portal hypertension
1.4.4 Congenital heart diseases
1.4.5 Schistosomiasis
1 . Pulmonary venoocclusive disease and/or pulmonary capillary
100 . Persistent pulmonary hypertension of the newborn (PPHN)
2. Pulmonary hypertension due to left heart disease
2.1 Left ventricular systolic dysfunction
2.2 Left ventricular diastolic dysfunction
2.3 Valvular disease
2.4 Congenital/acquired left heart inow/outow tract
obstruction and congenital cardiomyopathies
3. Pulmonary hypertension due to lung diseases and/or hypoxia
3.1 Chronic obstructive pulmonary disease FIGURE 1. Diagram shows the possible results of altered reso-
3.2 Interstitial lung disease lution of the thrombus. Reprinted with permission from the
3.3 Other pulmonary diseases with mixed restrictive and Cleveland Clinic Education Institute.
obstructive pattern
3.4 Sleep-disordered breathing
3.5 Alveolar hypoventilation disorders vascular remodeling also contribute to the disease (Fig. 1).
3.6 Chronic exposure to high altitude Several factors such as infection, immune phenomena,
3.7 Developmental lung diseases inammation, circulating and vascular-resident progenitor
4. Chronic thromboembolic pulmonary hypertension (CTEPH)
cells, thyroid hormone replacement, or malignancy19 may
5. Pulmonary hypertension with unclear multifactorial mechanisms
5.1 Hematologic disorders: chronic hemolytic anemia, lead to disturbed vascular remodeling. Endothelial dysfunc-
myeloproliferative disorders, splenectomy tion and endothelial-mesenchymal transition may also play a
5.2 Systemic disorders: sarcoidosis, pulmonary histiocytosis, part.19 The end result is disturbed resolution of the throm-
lymphangioleiomyomatosis bus, with partial recanalization or formation of cords or
5.3 Metabolic disorders:glycogen storage disease, Gaucher webs in the vessel lumen, which may partially restore the
disease, thyroid disorders ow20 (Fig. 1). The organized clot can contract, causing
5.4 Others: tumoral obstruction, brosing mediastinitis, chronic shrinkage, obstruction, and atrophy of the vessel itself. The
renal failure, segmental PH mechanisms resulting in subsequent development of PH are
multifactorial. In some studies, the nding of increased
baseline PAP at the time of index PE event was associated
into 5 groups on the basis of the cause (Table 1).15 CTEPH with higher risk for subsequent development of PH, sug-
is classied as group 4 disorder. Group 1 disorders include gesting that the severity of the acute PE event might be a
PH caused by insult at the precapillary level, denoted as predictor of subsequent CTEPH.21 The development of
pulmonary arterial hypertension (PAH). Although pulmo- small vessel disease (pulmonary arteriopathy) also contrib-
nary venoocclusive disease (PVOD) and pulmonary capil- utes to the evolution of PH. It is unclear whether the degree
lary hemangiomatosis (PCH) are postcapillary diseases, of pulmonary arteriopathy is related to the duration of PH
these are included in group 1, as the histologic changes and before the diagnosis of CTEPH. Several medical conditions
clinical presentation of these conditions are similar to those have been associated with the development of small vessel
of other group 1 conditions. Conditions related to left heart disease, such as ventriculo-atrial shunts for the treatment of
disease are the most common cause of PH and are included hydrocephalus, splenectomy, inammatory bowel disease,
in group 2. low-grade malignancy, and thyroid replacement therapy.19

PATHOPHYSIOLOGY Understanding the Concept of Small Vessel

The natural history of an acute pulmonary embolic Versus Large Vessel Disease
event is near-total resolution with no or minimal residual The exact denition of small vessels in pulmonary
hemodynamic consequences. It is still unclear why certain vasculature is poorly understood. Histologically, there is a
patients develop CTEPH after acute PE.16 Several gradual transition from elastic to muscular arteries to
hypotheses have been proposed, including the presence of arterioles. The point on computed tomography (CT)
an underlying hypercoagulable state in patients who imaging at which a vessel becomes categorized as a small
develop CTEPH.17 The antiphospholipid antibody syn- vessel is partly limited by the spatial resolution of CT; a
drome is the most commonly encountered hypercoagulable pulmonary arteriole measures approximately 100 mm and is
state, occurring in up to 20% of patients with CTEPH.18 not directly visualized on standard thin-section CT
Additional factors such as thrombophilia and disturbed images.20 For clinically classifying the disorders of PH into

72 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
J Thorac Imaging  Volume 32, Number 2, March 2017 Imaging in CTEPH

small vessel and large vessel group disorders with PH with

relatively normal perfusion on imaging tests such as
ventilation-perfusion (V/Q) scanning or pulmonary angio-
gram are grouped into small vessel diseases (such as IPAH),
whereas patients with PH and at least Z1 segmental perfu-
sion defects are classied as having large vessel disorders
(such as CTEPH). This is undoubtedly an oversimplication,
and often patients show a component of both small vessel
and large vessel changes.


The initial goal of diagnostic evaluation is con-
rmation of the presence of PH. Doppler echocardiography
is the best initial screening test, but RHC is absolutely
necessary to conrm the presence of PH. Transthoracic
echocardiography allows calculation of right ventricular
FIGURE 2. Continuous wave Doppler of tricuspid regurgitation
systolic pressure by measuring the peak tricuspid regur- jet demonstrates a maximum velocity of 4.6 m/s, which corre-
gitation velocity and using the modied Bernoulli equa- sponds to a RV/RA pressure differential of 85 mm Hg.
tion21 (Fig. 2, Cine clip 1, Supplemental Digital Content 1, However, the pulmonary CTA in the diagnosis of CTEPH.27 Nevertheless, V/Q
artery systolic pressure (PASP) measured by transthoracic scanning oers a reliable and convenient way to exclude
echocardiography may dier from RHC measurements.22 CTEPH from consideration, as a low-probability V/Q scan
virtually excludes this diagnosis, and thus it remains the
Chest Radiography preferred initial screening imaging test.
Chest radiography is often performed as a part of A typical V/Q abnormality in CTEPH includes Z1
baseline evaluation and may be normal in the early course segmental or larger mismatched perfusion defects (Fig. 4).
of the disease. In advanced cases, chest radiography can This is in contrast to disorders of small vessels such as
depict ndings related to PH with dilated right and left IPAH or PVOD, in which perfusion may be normal or may
pulmonary arteries (PAs) (Fig. 3). RV enlargement in demonstrate a mottled appearance characterized by sub-
response to chronic PVR elevation may manifest radio- segmental defects.28,29 It should be noted that mismatched
graphically as enlargement of the cardiac silhouette and segmental defects have also been described with PVOD,
obliteration of the retrosternal clear space on lateral chest although they are a rare occurrence.30
radiography.23 Right atrial enlargement leading to prom- Although V/Q scanning oers a quick and easy way to
inence of the right heart border on frontal chest radiography exclude CTED as a cause of PH, this imaging technique has
may also be observed.23 Focal areas of lung parenchymal important limitations. Several other nonembolic conditions
oligemia or avascularity can be seen in regions of compro- can cause large mismatched segmental perfusion defects,
mised blood ow secondary to thromboembolic disease.23 such as the following:
Unilateral or bilateral pleural thickening and parenchymal  Intrinsic PA tumors such as sarcoma (Fig. 5),
scarring, which are likely a sequelae of previous acute  Fibrosing mediastinitis,
embolic episodes, may also be observed (Fig. 3). In a study  Fibrosis caused by radiation therapy to the mediastinal
on 36 patients, characteristic ndings of focal areas of and hilar regions,
avascularity or enlarged right descending PA (diameter >  Vasculitis.
20 mm) together with pleuritic abnormalities were found to
be helpful in dierentiating CTEPH from IPAH.24
Consequently, it is imperative to obtain additional
cross-sectional imaging such as CT to evaluate the lung
EVALUATION OF SUSPECTED PH parenchyma and mediastinal structures and to exclude other
Once the diagnosis of PH has been established, the pathologies. Second, with CTED, it is not uncommon to nd
next crucial step is to determine the cause of the PH. areas of matched V/Q defects, which can somewhat negate
CTEPH should always be considered even in patients the benets of this study. In our experience, areas of matched
without a history of venous thromboembolism or risk fac- V/Q defects likely reect the compensatory response of the
tors for thromboembolism, as their clinical presentation lung to chronic hypoperfusion leading to eventual reduction
can be clinically indistinguishable from those with PAH.25 in ventilation in those areas. Third, the degree of vascular
obstruction and hemodynamic compromise is often under-
V/Q Scintigraphy quantied by V/Q scintigraphy when compared with con-
Although CT imaging has increasingly supplanted V/ ventional angiography or at surgery.31 A possible explanation
Q scintigraphy for the diagnosis of acute pulmonary for this drawback may be the development of channels
thromboembolism, a V/Q scan is used at many centers as through the centrally obstructing lesions or partial recanali-
the initial imaging test for establishing the diagnosis of zation or organization of thrombi after an acute event, which
CTEPH. In a retrospective survey comparing V/Q scintig- permits the radiotracer to reach the periphery of the lungs.
raphy with CT angiography (CTA), the V/Q scan was The planar nature of conventional V/Q scans unmasks
found to be superior to CTA with a sensitivity of 97.4% limitations inherent to 2-dimensional (2D) techniques.
and a specicity of 90% to 95% (vs. a sensitivity of 51% Lung segment overlap and shine-through may result in
and a specicity of 99% with CTA).26 However, a more errors in the localization and quantication of perfusion
recent study found comparable sensitivities for V/Q and defects. As a result, 3D techniques, such as single photon

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. | 73
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

although the anatomic information obtained by CT can be

used to exclude obvious nonembolic causes of perfusion
defects, such as neoplasm, thereby improving the specicity
of the study. Several studies have shown improved specif-
icity and overall accuracy with SPECT-CT V/Q scanning
compared with conventional and SPECT V/Q scan-
ning.35,36 Another benet of this technique is its accurate
mapping of the perfusion defect to the respective segments
of the lung using cross-sectional imaging data (Fig. 7).37
However, the inclusion of CT does increase the radiation
dose delivered to a patient relative to SPECT V/Q and V/Q

CT is a frontline technique for the assessment of sus-
pected acute thromboembolic disease. Excellent spatial and
contrast resolution, rapid scan times, and detailed evalua-
tion of the lung parenchyma are some of the advantages
that CTA oers. For routine clinical purposes, a standard
nonelectrocardiogram (ECG)-gated CTA is usually su-
cient, although some studies have shown the added value of
ECG gating and of newer protocols such as high pitch
CT ndings in patients with CTEPH may vary
according to the severity of the disease, the amount of
vascular obstruction, and the degree of PH.39 These imag-
ing ndings can be divided into those related to CTED and
those related to PH, each of which can be subdivided into
vascular signs and lung parenchymal signs.

CT Findings Related to CTEDVascular Signs

Because of its superior spatial and contrast resolution,
CTA can directly depict the extent and burden of throm-
botic disease as peripherally as the subsegmental PA
branches. As thrombi of dierent ages may exhibit dierent
features on CT40 (Table 2), CTA may also distinguish acute
thromboembolic events superimposed on a background of
chronic disease.
In PAs, complete thrombotic occlusion may manifest
on CTA as abrupt luminal narrowing and cuto with no
distal opacication. This has been classically described as
the pouch sign on conventional angiography, with con-
trast forming a convex appearance at the distal occluded
vessel.41 Partially occlusive lling defects may be seen on
CTA as webs or bands lying in the periphery of the ves-
sel,4143 reecting partially resolved thrombi (Fig. 8A). In
advanced cases, the thrombi may manifest as eccentric
FIGURE 3. Frontal (A) and lateral (B) chest radiographs of a 40- irregular thickening of the vessel wall with reduced luminal
year-old man with CTEPH show band-like areas of scarring, which caliber (Figs. 8B, C). Organizing thrombi can contract,
are sequelae of prior infarctions (arrows). Also noted is the resulting in small and focal stenotic segments, followed by
enlarged right main and interlobar pulmonary artery (thick areas of poststenotic dilation4143 (Fig. 8D). In CTED,
arrows). Pleural thickening is noted (broken arrow), which is a residual thrombus may occasionally calcify.
common feature in patients with CTEPH. There is relatively CTA also allows the dierentiation of CTED from
decreased vascularity in the left lung compared with the right other etiologies and helps establish the diagnosis of
(thin broken arrows). CTEPH. With CTA, acute and chronic thrombotic occlu-
sion can be dierentiated by the size of the occluded vessel;
emission (SPECT) V/Q, have been advocated as a way to typically, the vessel with acute thrombus is expanded,
improve the accuracy of perfusion defect localization and whereas in CTE the aected vessel is attenuated (Table 2).
quantication. SPECT V/Q aords a signicant advantage Occasionally, patients with massively enlarged PAs as in
over conventional V/Q scans, oering improved sensitivity Eisenmenger syndrome can develop in situ thrombosis.
and specicity and a lower incidence of nondiagnostic Typical in situ thrombi adhere to the walls of the central
scans.3234 enlarged PAs and do not oat inside the lumen.44 Unlike
With the availability of SPECT-CT scanners, SPECT- CTEPH, the signs of clot retraction, reduction in the caliber
CT V/Q imaging is now possible as well (Fig. 6). A non- of aected vessels, and secondary lung parenchymal nd-
contrast CT is used primarily for attenuation correction, ings of oligemia and infarcts are not seen with in situ

74 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
J Thorac Imaging  Volume 32, Number 2, March 2017 Imaging in CTEPH

FIGURE 4. Planar V/Q scans in a patient with CTEPH. Images on the left panel are ventilation images, whereas perfusion images are
shown on the right. Multiple mismatched perfusion defects are noted (arrow), a typical finding in CTEPH.

thrombosis.44 Dierentiation of in situ thrombosis from In the setting of CTEPH, the regions of lower lung paren-
CTEPH has clinical implications as anticoagulation would chyma attenuation usually represent hypoperfused lung
increase the risk of bleeding in patients with in situ tissue and are associated with attenuated vascular mark-
thrombosis. ings51 (Fig. 9). In some situations, air trapping may also
Enlargement of the bronchial arteries is a nonspecic contribute to lower lung parenchymal attenuation in
response to chronically occluded PAs.45,46 The degree of patients with CTEPH, most likely because of secondary
systemic collaterization including the pleural and inter- impairment of the small airways.52 Regions of higher lung
costal arteries is related to the degree of obstruction and is parenchymal attenuation in CTEPH usually represent
more common with centrally located thromboembolic dis- hyperperfused lung tissue resulting from redistribution of
ease. In severe cases of CTEPH, enlarged bronchial arteries pulmonary blood ow. Augmented perfusion from collat-
may contribute to up to 30% of the pulmonary blood eral vessels may also contribute to focal areas of hyper-
ow.47 Interestingly, 1 study found a higher incidence of attenuation. Dierentiation from small airway disorders
bronchial artery hypervascularization in patients with can be aided by assessing the size of vessels within the low
CTEPH (73%) compared with that in patients with IPAH attenuation zones (attenuated with small vessel and normal
(14%).46 Identication of the collaterals, particularly the with small airway disorders) and by expiratory imaging,
bronchial arteries, is important, as these patients have an which can highlight air trapping in areas of low attenuation
increased incidence of hemoptysis.48 Presence of dilated with small airway disorders. Minimum-intensity projection
bronchial arteries has prognostic signicance and is asso- CT technique improves the detection of the mosaic
ciated with lower mortality after PEA, possibly related to attenuation pattern, whereas maximum-intensity projection
distal pulmonary microvasculature sparing.49 Thin-section images highlight variations in vessel caliber.
maximum-intensity projection coronal images are fairly Cylindrical bronchiectasis is occasionally observed
eective in depicting bronchial arteries. However, bronchial within segmental and subsegmental bronchi in the setting of
hypervascularization is not specic to CTEPH; these can be CTED, adjacent to severely stenosed or thrombosed PAs.53
encountered in other conditions leading to chronic hypoxia, Although the exact cause is unknown, hypoxic bronchodi-
such as interstitial brotic lung disease, bronchiectasis, and lation has been suggested as a possible mechanism.54
chronic infection. Patients with CTE typically show changes related to
old infarcts, although acute infarcts secondary to super-
CT Findings Related to CTEDLung Parenchymal imposed acute thromboembolic disease are not uncommon.
Signs On CT, in its most acute form, the infarct appears as an
Mosaic attenuation is a term used to describe a mul- area of consolidation and/or ground-glass opacity, in
tilobular patchwork of varying lung parenchymal attenu- keeping with an area of pulmonary hemorrhage. As the
ation50 that is often observed in patients with CTEPH. The infarction evolves, it acquires the typical appearance of an
imaging pattern of mosaic attenuation is by and large infarct, seen as a peripheral wedge-shaped, pleura-based
nonspecic and can be seen in several pathologies, such as density. Other characteristic ndings include internal air
small airway disease, small vessel disease, and primary lung lucencies, truncated apex (cuto tip), and a thickened vessel
parenchymal disease such as hypersensitivity pneumonitis. leading to the apex of the infarct, the vascular sign.55 Over

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. | 75
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

FIGURE 5. A 54-year-old-man with false-positive V/Q scans. A, Planar V/Q scans show severely decreased perfusion to the right lung. B,
Axial CTA image shows expansile filling defect in the right pulmonary artery (arrow), which is totally occluded. C, Coronal positron
emission tomography/CT fused image shows 18F-Fluorodeoxyglucose uptake in the mass-like filling defect (arrow), which was proven to
be a pulmonary artery sarcoma.

time, the infarct contracts and resolves, leaving residual CT Findings Related to PHVascular Signs
scars or bands. There may be an associated pleural reac- Increasing PVR can act in concert with other factors
tion, manifested as pleural thickening or uid.56 Nodules and lead to the development of PH, which is reected on CT
and cavitary opacities are not uncommon. as enlargement of central PAs. PA diameters over 29 mm are

76 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
J Thorac Imaging  Volume 32, Number 2, March 2017 Imaging in CTEPH

FIGURE 6. Abnormal representative SPECT and fused SPECT-CT images of a patient with CTEPH. Top panel images show fused
perfusion images, whereas bottom panel images show ventilation. Areas of hypoperfusion and decreased ventilation are seen as dark
defects. Mismatched perfusion defects are noted, predominantly in the upper lobes bilaterally (arrows). Note that ventilation images
show tracer retention in the central airways, which is a limitation of aerosol imaging.

often used as a threshold for PH. The measurement is per- of mortality in patients with bronchiectasis.59 A ratio of distal
formed at the plane of the pulmonary bifurcation orthogonal main PA to aortic diameter of over 1 is also suggestive of PH,
to the vessel course57,58 (Fig. 10). In a recent study, right and particularly in younger patients.58 ECG-gated examinations
left PAs over 18 mm in size were found to be best predictors have been shown to improve the evaluation of functional

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. | 77
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

data set and calculation of tortuosity of the segmented lung

vessels were found to be helpful in noninvasive assessment
of severity of PH.62 Another study showed good correlation
of fractional dimension with elevated PAP.63 A similar
study on 18 patients showed that patients with CTEPH
exhibited greater pruning of the distal vasculature, greater
dilation of proximal arteries, and increased tortuosity in the
pulmonary arterial tree.64 These measurements correlated
with invasive metrics of pulmonary hemodynamics, sug-
gesting that these may be used to assess disease severity. It
is still unclear whether these morphologic changes can help
distinguish CTEPH from other groups of PH. One nding
that may help is that CTEPH was not associated with
dilation of proximal veins or increased tortuosity in the
venous system. Also, the location of the pruned vasculature
might provide clues; upper lobepredominant pruning
would be expected in smoking-related PH.65
Development of right heart dysfunction is a natural
endpoint of PH and is a main cause of death in these
patients. Chronically elevated pulmonary pressures increase
the workload on the heart, resulting in RV hypertrophy and
enlargement (Fig. 10). Assessment of RV is suboptimal on
nonECG-gated examinations but is improved on ECG-
gated studies.60 RV hypertrophy can be reliably diagnosed
when the free wall of the RV measures over 4 mm66
(Fig. 10). RV enlargement can be reliably predicted when
the ratio of the diameter of the RV to that of the left
ventricle (LV) is over 1 (Fig. 9). Increased septal angle on
ECG-gated studies is seen with elevated PAP and correlates
with PVR in patients with CTEPH67 (Fig. 10). Mild peri-
cardial reaction in the form of eusion or thickening may
also be noted.68

CT Findings Related to PHLung Parenchymal Signs

Analysis of lung parenchyma allows identication of
disorders of groups 3 and 5 PH, such as emphysema, inter-
stitial lung disease (ILD), sarcoidosis, or Langerhans cell
histiocytosis.69,70 Recognition of the CT ndings related to
FIGURE 7. Representative SPECT-CT V/Q images of a patient
these diseases allows exclusion of CTEPH as the cause of PH.
with severe left superior vein stenosis with severely decreased As described above, mosaic attenuation of the lung paren-
perfusion to the left upper lobe (B), which substantially improved chyma can be seen with any cause of PH and can be related to
following stenting (A). Fusion of CT and SPECT data improves chronic thrombotic disease or distal vasculopathy. However,
localization and mapping of the perfusion defects, which helps in it is more frequently seen in CTEPH compared with other
assessment of treatment response. etiologies of PH.51 Centrilobular lung nodules may also be
observed in patients with PH, which are believed to be related
parameters such as PA distensibility and RV function in these to repeated episodes of pulmonary hemorrhage leading to
patients.60,61 macrophage ingestion of red blood cells and formation of
In long-standing PH, mural calcications can develop. cholesterol granulomas71 (Fig. 11). Although characteristi-
Chronically elevated PAP can lead to secondary morpho- cally seen with IPAH,72 these can also be seen with long-
logic changes in the pulmonary vascular tree, such as vessel standing left to right heart shunts as well as ILDs, including
tortuosity and arterial pruning. In a pilot study on 24 subacute hypersensitivity pneumonitis and smoking-related
patients, automated extraction of lung vessels from the CT respiratory bronchiolitis.70 In PVOD, ground-glass opacities

TABLE 2. Acute Versus CTED on CT

Findings Acute Disease Chronic Disease
Total occlusion Vessel expanded secondary to Vessel distal to the obstruction markedly attenuated
pulsatile ow
Partially occlusive Typically form acute angles with Typically form obtuse angles with the vessel lumen; several areas of webs and
lling defects the vessel lumen bands with focal stenotic segments
Systemic ndings RV heart strain, better seen with If PH has developed, RV dilation and hypertrophy; dilation of the central
ECG-gated examinations pulmonary arteries; collateral vessels, including bronchial artery dilation
Lung parenchymal Area of hemorrhage or infarction Mosaic attenuation of the lung parenchyma; brotic bands, which could be
ndings resolved infarcts

78 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
J Thorac Imaging  Volume 32, Number 2, March 2017 Imaging in CTEPH

FIGURE 8. CTA appearances of chronic thrombi in different patients with CTEPH: web (arrow in A), vessel atrophy with poor contrast
opacification (arrow in B), eccentric thrombus (arrow) and calcified mural thickening (yellow arrowhead) (C), and stenosis with
poststenotic dilation (arrow in D).

can be seen in diuse, geographic, mosaic, perihilar, patchy,

or centrilobular patterns.73 The nodules seen with PCH are
larger than the true centrilobular nodules, and these ground-
glass nodules are larger than those seen with PVOD.74 In
PVOD, additional discrimators over PCH and idiopathic
PAH are the presence of interlobular septal thickening and
pleural eusions.70,75 A rare cause of centrilobular nodules in
the setting of PH includes intravenous injection of ller sub-
stances such as talc, which induces a foreign body reaction
around the arterioles.76

Dual-Energy CT (DECT)
DECT uses attenuation dierences at various energy
levels to dierentiate between tissue materials. X-ray
attenuation is dependent on Compton scatter and photo-
electric absorption, which vary with photon energies and
material compositions. The probability of photoelectric
eect increases with increasing atomic number and
decreases with increasing photon energy.77 For materials
with higher atomic numbers, such as iodine, the photo-
FIGURE 9. Coronal CT image in lung window settings shows electric eect markedly decreases with increasing photon
mosaic attenuation of the lung parenchyma in a patient with energy, resulting in a rapid decrease in Hounseld unit
CTEPH. value. Thus, by using low and high voltages, relative

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. | 79
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

FIGURE 10. CT evaluation of pulmonary hypertension and RV parameters: A, Axial CTA image shows measurement of the main
pulmonary artery (MPA) just proximal to the bifurcation. The MPA measures 39 mm, consistent with pulmonary hypertension. B, ECG-
gated axial CTA image shows dilated RV (5.7 cm) with RV hypertrophy (arrowhead) in a patient with CTEPH. C, Axial CTA image shows
calculation of the septal angle from a line drawn between the midpoint of the sternum to the spinous process of a vertebral body and a
line drawn through the interventricular septum. Septal angle >67 degrees is a specific indicator of increased PVR.

dierences in the absorption characteristics of materials this factor. As on V/Q and magnetic resonance imaging
such as iodine and xenon can be exploited to characterize (MRI) scans, acute perfusion defects on DECT are seen as
their content within the tissue. The iodine within a voxel is wedge-shaped areas of decreased attenuation and corre-
quantied, and iodine perfusion maps are generated. It sponding decreased iodine content (Fig. 12). Some studies
should be noted that these are not true perfusion images, as have shown excellent correlation between the perfusion maps
they measure iodine content at a single time point. How- obtained using DECT and those obtained using V/Q scans
ever, these images do serve as an eective surrogate for the and SPECT-CT examinations.79,80 Limited but encouraging
assessment of perfusion.78 Automated quantication of data correlating regional perfusion maps with hemodynamic
perfused blood volume (PBV) images can be done, which parameters and lung parenchymal ndings have been
allows objective comparison of perfusion abnormalities in reported. In 1 study, DECT allowed the identication of
dierent lung segments. areas of mismatching, using a partition value of 20 Houns-
Hemodynamic impairment is a critical factor in deter- eld unit to dierentiate areas of residual perfusion distal to
mining whether a patient is eligible for surgical intervention, proximally occluded segments.81 These areas were postulated
and DECT oers the potential for one-stop assessment of to be perfused by collaterals (such as bronchial arteries); the
presence of such segments could be predictive of greater
postoperative success than matched defects. In the same
study, no signicant correlation was found between PBV
maps and RHC-derived mPAP and PVR.81 However, in
another study in 25 patients with CTEPH, automated PBV
values inversely correlated with PASP and mPAP.82 There
was also a trend of PBV values to correlate inversely with
PVR.82 Similarly, a recent study showed that automated
lung PBV scoring can serve as a noninvasive estimator of
clinical CTEPH severity, especially in comparison with the
mPAP and PVR.83 Another study on 391 patients demon-
strated that PH patients demonstrate increased main PA
enhancement with a reciprocal reduction and greater varia-
tion in parenchymal enhancement; a DECT ratio of central
to parenchymal enhancement correlated with PVR.84
Another novel application of DECT is the ability to
dierentiate between acute and CTED and assess the
amount of bronchial artery collaterization (Fig. 13). As
FIGURE 11. Axial lung window CT image shows multifocal noted earlier, some studies have shown that the presence of
ground-glass and centrilobular nodules (arrows) in a patient with bronchial artery collaterization is associated with better
IPAH. postoperative success.42,66 Using a 2-phase scanning

80 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
J Thorac Imaging  Volume 32, Number 2, March 2017 Imaging in CTEPH

protocol, in a study on 114 patients, it was shown that Although DECT oers considerable promise in
chronic PE segments showed more enhancement on delayed CTEPH, PBV maps require careful interpretation. Pseu-
phases than did acute PE segments, suggesting more col- dodefects can be seen because of artifacts such as beam-
lateral supply to these regions.85 hardening artifacts or motion artifacts near the heart or the
DECT can also help in the assessment of the various diaphragm. These artifacts need to be avoided or recog-
causes of mosaic attenuation and provide results that point nized before a diagnosis is reached.87 Typical beam-
to a specic vascular pathology.86 For example, in areas of hardening defects aect the medial right lung (predom-
ground-glass opacities, by demonstrating increased or no inantly in the right upper lobe) due to contrast in SVC.
perfusion, it can aid in dierentiating a vascular process As noted above, PBV maps always require interpretation
from primary lung inltration. DECT also aids in the dif- in relation to morphologic pulmonary reconstructions in
ferential diagnosis of various groups of PH.87 Typically, order to exclude underlying lung process such as
group 1 diseases such as IPAH demonstrate mottled per- emphysema.
fusion as compared with the segmental defects with
CTEPH. Perfusion abnormalities may also be caused by
parenchymal destruction as in ILD, diuse small airway
disease, or pulmonary emphysema. Correlating the PBV MRI and MR Angiography (MRA)
maps with lung reconstructions can show that these defects MRI and MRA play a growing role in the evaluation
match the areas of parenchymal destruction and thus help of pulmonary thrombotic disease. Some tools, such as cine
dierentiate group 3 PH disorders. Finally, DECT can be imaging and phase-contrast velocity-encoded MRI (PC-
used to assess V/Q. Use of inert agents such as xenon has MRI), provide valuable functional information and have a
been applied to map distribution in the lung parenchyma denite role in the follow-up of patients.90 However, tech-
and thereby generate information on ventilation.88 Initial nical demands and suboptimal evaluation of lung paren-
study results have been encouraging, suggesting that a true chyma currently limit the use of MRI as a single imaging
DECT V/Q study may be possible.89 test in CTEPH.

FIGURE 12. DECT images of a 51-year-old man with CTEPH with angiographic correlation. A, Coronal perfusion iodine map shows
wedge-shaped defects in the right upper and both lower lobes (arrowheads) corresponding to stenotic right upper lobe segmental
branch (arrow). B, Corresponding blood vessel map shows patchy distribution of contrast-enhanced arteries, consistent with chronic
thrombotic disease (greenish blue arteries: iodine containing arteries; red arteries: non-iodine-containing arteries). C, Parenchymal
phase image from a conventional right pulmonary angiogram shows excellent correlation with DECT images, showing the stenotic right
upper lobe segmental branch (arrow) and oligemia in the right upper and lower lobes (arrowheads).

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. | 81
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

FIGURE 13. Two-phase DECT in a patient with CTEPH. A, Axial CT image shows severely attenuated left lower lobe pulmonary artery
(arrow). B, Coronal PBV image shows large perfusion defect in the left lower lobe; on the delayed phase done 70 seconds later (C), the
perfusion to the left lower lobe has improved, which may suggest significant collateral supply to this region.

MR Findings Related to Pulmonary Thromboembolic interobserver and intraobserver variability.96 PC-MRI and
DiseaseAnatomy cine steady state free precession imaging can also be com-
Imaging features of pulmonary thromboembolic dis- bined to quantify shunt fraction (Qp/Qs), which indirectly
ease on contrast-enhanced MRA are similar to those seen reects the degree of bronchopulmonary shunting in these
on CTA, including irregular eccentric lling defects within patients.42
the PAs, intraluminal webs and bands, and areas of sten- Recent research has shown good correlation between
oses and occlusion91 (Fig. 14, Cine clip 2, Supplemental MRI-derived mPAP and PVR measurements and angio-
Digital Content 2, A lack graphic measurements.97,98 MRI can highlight indirect signs
of signal intensity in the normal lung limits evaluation of of elevated PAP, including delayed enhancement at the
the peripheral vasculature, as the low signal intensity of the septal insertion points of the myocardium and systolic
occluded vessel is indistinguishable from that of the sur- bowing of the interventricular septum toward the LV, which
rounding lung.92 Typically, branches down to the segmental is explained by the mechanical dyssynchrony resulting from
levels can be reliably assessed with MRI.93 Because of its RV overload and elongation and LV underlling.99,100
higher spatial resolution, better contrast-noise ratio, and Functional indices such as PVR, mPAP, and RV function
faster imaging times (enabling better breath-holding), CT is have important prognostic implications and can be used to
better suited for subsegmental vasculature assessment than assess functional improvement after pulmonary thromo-
MRI and remains the diagnostic test of choice for the bendarterectomy.91,101,102 Emerging techniques such as 4D
evaluation of morphologic abnormalities.94 However, as velocity ow mapping may provide additional insights into
most surgically accessible thrombi are limited to the central ow patterns in pulmonary circulation.103
and segmental branches, the ultimate signicance of sub- In theory, MRI seems to be a good option for perfusion
segmental disease characterization is unknown. imaging, which could make this modality a true one-stop
test for assessing PH and CTEPH. Limited data evaluating
MR Findings Related to Pulmonary Thromboembolic the assessment of lung perfusion with MRI have shown good
DiseaseFunction correlation with perfusion scintigraphy.93,104 One of the
Cine steady state free precession imaging is an constraints involved in assessing lung perfusion is the
accepted reference standard for the evaluation of LV and dynamic nature of pulmonary microcirculation. Quantitative
RV function, with excellent accuracy and reproducibility.95 perfusion parameters, including mean transit time, time to
PC-MRI allows for estimation of cardiac output with low peak, and blood volume, can be estimated on the basis of the

82 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
J Thorac Imaging  Volume 32, Number 2, March 2017 Imaging in CTEPH

RHC and Conventional Angiography

RHC remains the gold standard for dening the
severity of PH and for assessing the severity of cardiac
dysfunction. Hemodynamic assessment during RHC is
typically performed at rest, although some patients may
benet from exercise-induced measurements.109 In such
patients, exercise is postulated to uncover compensatory
mechanisms and may unmask a clinically relevant stage of
CTEPH with coexisting small vessel hypertensive
changes.110 Using waveform analysis, this technique may
help to identify a subset of patients with small vessel disease
who may not benet from surgery.111
Although considered the gold standard for the diagnosis
of CTEPH, conventional angiography has steadily decreased
in utliziation with the increasing use of cross-sectional imaging
such as CT.112 The success of BPA might change this status in
the near future and mark a resurgence of catheter angiog-
raphy. The risks associated with angiography should be con-
sidered carefully, particularly in patients with PH, as contrast
agents can potentially precipitate vasodilation and hypo-
tension. Nonetheless, conventional angiography is generally
safe and can be performed as part of RHC, when appropriate
FIGURE 14. Still MR pulmonary angiography: maximum- precautions are taken.113
intensity projection demonstrating CTEPH features such as RHC and conventional angiography are used to dene
pouch defect in the right lower lobe (thin arrow) with segmental the surgical accessibility of the thrombotic lesions, to
occlusions and a tight proximal stenosis in left lower lobe (thick quantify PVR, and to determine whether PVR is secondary
arrow) with poststenotic dilatation. to surgically treatable disease or distal arteriopathy.42,114
Biplane angiography is preferred, as concurrent orthogonal
indicator dilution theory, but this estimation involves several acquisitions improve interpretation of the pulmonary ves-
assumptions and limitations.105 One of the main limiting sels, particularly lobar and segmental branches. Direct
factors is that indicator dilution theory is applicable to endovascular visualization (angioscopy) may be performed;
intravascular agents, whereas most gadolinium agents are however, its use has declined as newer noninvasive techni-
extracellular agents; the eect of the agent diusing into the ques provide satisfactory means to assess surgical accessi-
extracellular space is not fully clear. The nonlinear function bility of thrombotic disease.115
of gadolinium agents and lack of correction for patient Five angiographic patterns have been described in the
hematocrit are additional limitations of this method.106 In literature, including pouching defects, webs or bands,
addition, some of the published studies involved time- intimal irregularities, abrupt vascular narrowing, and com-
resolved MRA with view sharing, which involves incorpo- plete vascular obstruction41 (Fig. 12C). As the thrombus can
ration of data from previous time points. Although this retract and assimilate in the vessel lumen and cause varying
technique can provide qualitative assessment of perfusion degrees of retraction and occlusion, interpretation of angio-
(and defects) (Cine clip 3, Supplemental Digital Content 3, graphic patterns can be challenging. Consequently, CT, quantitative perfusion probably provides more reliable evaluation of pulmonary
imaging is not truly achieved, as this technique requires thromboembolic diseaseparticularly in subsegmental
evaluation of the blood ow in a region over multiple time branches.94 Conventional angiography allows for excellent
points. The use of noncontrast Fourier decomposition has qualitative assessment of the pulmonary blood ow and in
also been explored as a method for assessing V/Q, as this mapping perfusion defects (Cine clip 4, Supplemental Digital
technique requires no intravenous or inhaled contrast.107,108 Content 4,

TABLE 3. Approach to Imaging of Pulmonary Hypertension Based on Updated Clinical Classification

Group Initial Test of Choice Additional Imaging Tests
1 Echocardiography RHC: to denitively diagnose the presence of PH and assess its
CT: for evaluation of connective tissue disease, portal
hypertension, PVOD/PCH and so on
MRI: for assessment of congenital heart disease
2 Echocardiography MRI and cardiac catheterization may be considered
3 CT RHC in some cases
4 V/Q as initial test followed by CT for conrmation MRI: for right heart and pulmonary artery evaluation
(possibly DECT or SPECT-CT V/Q in future) RHC and conventional pulmonary angiography: for preoperative
5 CT

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. | 83
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

TABLE 4. Comparative Assessment of Merits of Imaging Tests in Assessment of CTEPH

Imaging Tests
Factor Assessed Echocardiography CTA DECT MRI V/Q V/Q
Pulmonary vasculature
Central thrombi +++ +++ ++
Peripheral thrombi +++ +++ +
Dilated pulmonary artery + +++ +++ +++
Pulmonary microcirculation
Pulmonary perfusion + + + (needs more ++ ++ ++
Mean transit time + + ++
RV and pulmonary artery functional assessment
RV size and RV hypertrophy ++ + + + (with ECG ++ +++
RV volumes ++ + + +++
Pulmonary ow + + +++
Pulmonary compliance ++
Mean pulmonary arterial ++ ++
Tricuspid regurgitation jet and +++ +++
Tricuspid valve annular plane +++ +++
Lung parenchyma
Mosaic attenuation +++ +++ ++
Infarcts/scars +++ +++ ++ ++ +++
Ventilation (if specically + + ++ +++ +++
Bronchial arteries +++ +++ ++
+ , minimal use; + + , moderately useful; + + + , very useful.

CLINICAL DECISION MAKING IN CTEPH PASP.117 Several noninvasive tools can be used in con-
PTE is a complex surgical procedure and a true junction with clinical tools such as 6-minute walk test to
endarterectomy in which surgically accessible thrombotic assess functional recovery. Echocardiography is the easiest
material is removed. Typically, thromboembolic burden in and readily available test and can be used to assess post-
main, lobar, and segmental branches is deemed operable, therapy LV and RV remodeling. Newer techniques such as
whereas distal (subsegmental/microvascular) disease is 2D speckle tracking may allow better assessment of cardiac
usually considered inoperable. However, assessment of function after therapy.118 One of the disadvantages of echo-
surgical candidacy can often be inuenced by several other cardiography, however, is that direct visualization of the
factors such as age, conditioning, degree of hemodynamic disease burden is not possible. CTA and MRI are well suited
impairment, and preoperative PVR.116 Often, a multi- for depiction of improvement of thrombus burden, although
disciplinary discussion among surgeons, pulmonologists, they are not routinely used. MRI, although slightly inferior to
and radiologists is necessary in clinical decision making. CTA in the assessment of pulmonary vascular thrombotic
For radiologists, one of the key points to note is that if the disease, can be extremely helpful in the evaluation of func-
extent of anatomic disease correlates with the degree of tional recovery. Several MRI-based noninvasive biomarkers
increased PVR, the disease is usually deemed operable. such as PA maximum ow velocity, acceleration time/ejec-
However, if the disease burden appears mild or relatively tion time, and distensibility can be evaluated using MRI.119
normal in the context of disporoportionately elevated PVR, RV remodeling and adaptation is also better evaluated with
the possibility of small vessel arteriopathy is likely.116 MRI than with any other imaging technique.120
The roles of various imaging modalities in the evalu- A major role of imaging is in the assessment of com-
ation of common causes of PH are summarized in Table 3. plications. After PTE, some of the early complications
The roles of various imaging techniques in assessing the include reperfusion pulmonary edema and PA steal syn-
morphologic and functional changes of CTEPH are sum- drome. PA steal syndrome, seen in approximately 70% of
marized in Table 4. patients after PTE, is characterized by new areas of V/Q
mismatching and reects the redistribution of blood ow
from normally perfused lung to the newly endarterectom-
ROLE OF IMAGING IN FOLLOW-UP ized segments.121 One of the intermediate to long-term
Imaging is often used to monitor patients on medical complications of PTE is residual PH, seen in approximately
therapy and for assessment of response to therapy following one-third of patients.122 Causes include distal inoperable
PEA and BPA. Successful PTE is associated with immediate subsegmental disease and/or coexisting small vessel arte-
improvement in hemodynamics, with reduction in PVR and riopathy.122 Recurrent PH is less common and is due to a

84 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
J Thorac Imaging  Volume 32, Number 2, March 2017 Imaging in CTEPH

new thromboembolic event after successful PTE.116 CTA 5. Berghaus TM, Barac M, von Scheidt W, et al. Echocardio-
and MRI are promising tools for assessing residual/recur- graphic evaluation for pulmonary hypertension after
rent PH, with both having their respective strengths. DECT recurrent pulmonary embolism. Thromb Res. 2011;128:
with its ability to provide perfusion information along with e144e147.
6. Kapitan KS, Buchbinder M, Wagner PD, et al. Mechanisms
excellent depiction of thrombotic disease might evolve as a of hypoxemia in chronic thromboembolic pulmonary hyper-
frontline test for evaluation of complications. tension. Am Rev Respir Dis. 1989;139:11491154.
7. Held M, Kolb P, Grun M, et al. Functional characterization
of patients with chronic thromboembolic disease. Respiration.
FUTURE DIRECTIONS 2016;91:503509.
The role of imaging in the diagnosis and management 8. Berger G, Azzam ZS, Hardak E, et al. Idiopathic pulmonary
of CTEPH continues to expand rapidly. With the success of arterial hypertension or chronic thromboembolic pulmonary
newer minimally invasive therapies such as BPA, a variety hypertension: can we be certain? Isr Med Assoc J. 2011;13:
of therapy options are available. One of the holy grails in 106110.
the imaging of CTEPH remains the identication of 9. Riedel M, Stanek V, Widimsky J, et al. Long-term follow-up
microvascular disease. So far, there is no gold standard for of patients with pulmonary thromboembolism: late prognosis
and evolution of hemodynamic and respiratory data. Chest.
direct identication of microvascular disease. Angiography-
based partitioning of PVR by a PA occlusion technique 10. Fedullo P, Kerr KM, Kim NH, et al. Chronic thromboem-
may allow the identication of patients with small vessel bolic pulmonary hypertension. Am J Respir Crit Care Med.
disease but is invasive and technically demanding.111 A 2011;183:16051613.
handful of studies have shown the potential role of echo- 11. Corsico AG, DArmini AM, Cerveri I, et al. Long-term
cardiography and MRI in identifying these patients. Using outcome after pulmonary endarterectomy. Am J Respir Crit
Pulsed Doppler and PC-MRI, the PA systolic prole is Care Med. 2008;178:419424.
mapped and assessed for the presence of systolic notching. 12. Kim NH, Delcroix M, Jenkins DP, et al. Chronic throm-
The timing of the systolic notch is used as a predictor of the boembolic pulmonary hypertension. J Am Coll Cardiol. 2013;
site of obstruction, with a late systolic notch indicating the 62(suppl):D92D99.
13. Taboada D, Pepke-Zaba J, Jenkins DP, et al. Outcome of
presence of microvascular disease.123 Although such pulmonary endarterectomy in symptomatic chronic throm-
advances are promising, larger studies are needed to fully boembolic disease. Eur Respir J. 2014;44:16351645.
assess the benet of these noninvasive markers of micro- 14. Ogawa A, Matsubara H. Balloon pulmonary angioplasty: a
vascular disease. treatment option for inoperable patients with chronic
thromboembolic pulmonary hypertension. Front Cardiovasc
Med. 2015;2:4.
CONCLUSIONS 15. Simonneau G, Gatzoulis MA, Adatia I, et al. Updated clinical
Imaging plays an important role in the diagnosis of classification of pulmonary hypertension. J Am Coll Cardiol.
CTEPH, its preoperative evaluation, and in the assessment 2013;62(suppl):D34D41.
of a patients response to therapy. Although V/Q scanning 16. Auger WR, Fedullo PF. Chronic thromboembolic pulmonary
continues to be favored as the initial screening test of hypertension. Semin Respir Crit Care Med. 2009;30:471483.
choice, CT has emerged as the denitive imaging test of 17. Kyrle PA, Minar E, Hirschl M, et al. High plasma levels of
factor VIII and the risk of recurrent venous thromboembo-
choice in depicting the structural and vascular abnormal- lism. N Engl J Med. 2000;343:457462.
ities in CTEPH. MRI plays a complementary role, pro- 18. Wolf M, Boyer-Neumann C, Parent F, et al. Thrombotic risk
viding crucial functional and physiological information that factors in pulmonary hypertension. Eur Respir J. 2000;15:
carries prognostic value. The emergence of new methods 395399.
such as DECT, SPECT V/Q, SPECT-CT V/Q, and newer 19. Lang IM, Pesavento R, Bonderman D, et al. Risk factors and
MRI techniques heralds an exciting and promising shift in basic mechanisms of chronic thromboembolic pulmonary
imaging paradigms that may improve clinical decision hypertension: a current understanding. Eur Respir J. 2013;41:
making and ultimately lead to more favorable patient 462468.
outcomes. 20. Hansell DM. Small-vessel diseases of the lung: CT-pathologic
correlates. Radiology. 2002;225:639653.
21. Raisinghani A, Ben-Yehuda O. Echocardiography in chronic
REFERENCES thromboembolic pulmonary hypertension. Semin Thorac
1. Galie N, Hoeper MM, Humbert M, et al. Guidelines for the Cardiovasc Surg. 2006;18:230235.
diagnosis and treatment of pulmonary hypertension: the Task 22. Finkelhor RS, Lewis SA, Pillai D. Limitations and strengths
Force for the Diagnosis and Treatment of Pulmonary of doppler/echo pulmonary artery systolic pressure-right
Hypertension of the European Society of Cardiology (ESC) heart catheterization correlations: a systematic literature
and the European Respiratory Society (ERS), endorsed by thr review. Echocardiography. 2015;32:1018.
International Society of Heart and Lung Transplantation 23. Woodruff WW III, Hoeck BE, Chitwood WR Jr, et al.
(ISHLT). Eur Heart J. 2009;30:24932537. Radiographic findings in pulmonary hypertension from
2. Miniati M, Monti S, Bottai M, et al. Survival and restoration unresolved embolism. AJR Am J Roentgenol. 1985;144:
of pulmonary perfusion in a long-term follow-up of patients 681686.
after acute pulmonary embolism. Medicine (Baltimore). 24. Satoh T, Kyotani S, Okano Y, et al. Descriptive patterns of
2006;85:253262. severe chronic pulmonary hypertension by chest radiography.
3. Klok FA, van Kralingen KW, van Dijk AP, et al. Prospective Respiratory medicine. 2005;99:329336.
cardiopulmonary screening program to detect chronic throm- 25. Galie` N, Hoeper MM, Humbert M, et al. Guidelines for the
boembolic pulmonary hypertension in patients after acute diagnosis and treatment of pulmonary hypertension: the task
pulmonary embolism. Haematologica. 2010;95:970975. force for the diagnosis and treatment of pulmonary hyper-
4. Pengo V, Lensing AW, Prins MH, et al. Thromboembolic tension of the European Society of Cardiology (ESC) and the
Pulmonary Hypertension Study Group. Incidence of chronic European Respiratory Society (ERS), endorsed by the
thromboembolic pulmonary hypertension after pulmonary International Society of Heart and Lung Transplantation
embolism. N Engl J Med. 2004;350:22572264. (ISHLT). Eur Heart J. 2009;30:24932537.

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. | 85
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

26. Tunariu N, Gibbs SJ, Win Z, et al. Ventilation-perfusion hypertension: evaluation with helical CT and MR imaging.
scintigraphy is more sensitive than multidetector CTPA in AJR Am J Roentgenol. 2002;179:12091215.
detecting chronic thromboembolic pulmonary disease as a 46. Remy-Jardin M, Duhamel A, Deken V, et al. Systemic collateral
treatable cause of pulmonary hypertension. J Nucl Med. 2007; supply in patients with chronic thromboembolic and primary
48:680684. pulmonary hypertension: assessment with multi-detector row
27. He J, Fang W, Lv B, et al. Diagnosis of chronic helical CT angiography. Radiology. 2005;235:274281.
thromboembolic pulmonary hypertension: comparison of 47. Endrys J, Hayat N, Cherian G. Comparison of bronchopul-
ventilation/perfusion scanning and multidetector computed monary collaterals and collateral blood flow in patients with
tomography pulmonary angiography with pulmonary angiog- chronic thromboembolic and primary pulmonary hyper-
raphy. Nucl Med Commun. 2012;33:459463. tension. Heart. 1997;78:171176.
28. Fishman AJ, Moser KM, Fedullo PF. Perfusion scans vs 48. Thomas CS, Endrys J, Abul A, et al. Late massive
pulmonary angiography in evaluation of suspected primary haemoptyses from bronchopulmonary collaterals in infarcted
pulmonary hypertension. Chest. 1983;84:679683. segments following pulmonary embolism. Eur Respir J. 1999;
29. Seferian A, Helal B, Ja s X, et al. Ventilation/perfusion lung 13:463464.
scan in pulmonary veno-occlusive disease. Eur Respir J. 2012; 49. Kauczor HU, Schwickert HC, Mayer E, et al. Spiral CT of
40:7583. bronchial arteries in chronic thromboembolism. J ComputAs-
30. Bailey CL, Channick RN, Auger WR, et al. High sist Tomogr. 1994;18:855861.
probability perfusion lung scans in pulmonary venoocclusive 50. Hansell DM, Bankier AA, MacMahon H, et al. Fleischner
disease. Am J Respir Crit Care Med. 2000;162:19741978. Society: glossary of terms for thoracic imaging. Radiology.
31. Ryan KL, Fedullo PF, Davis GB, et al. Perfusion scan 2008;246:697722.
findings understate the severity of angiographic and hemody- 51. Sherrick AD, Swensen SJ, Hartman TE. Mosaic pattern of
namic compromise in chronic thromboembolic pulmonary lung attenuation on CT scans: frequency among patients with
hypertension. Chest. 1988;93:11801185. pulmonary artery hypertension of different causes. AJR Am J
32. Collart JP, Roelants V, Vanpee D, et al. Is a lung perfusion Roentgenol. 1997;169:7982.
scan obtained by using single photon emission computed 52. Arakawa H, Stern EJ, Nakamoto T, et al. Chronic pulmonary
tomography able to improve the radionuclide diagnosis of thromboembolism. Air trapping on computed tomography
pulmonary embolism? Nucl Med Commun. 2002;23: and correlation with pulmonary function tests. J Comput
11071113. Assist Tomogr. 2003;27:735742.
33. Reinartz P, Wildberger JE, Schaefer W, et al. Tomographic 53. Remy-Jardin M, Remy J, Louvegny S, et al. Airway changes
imaging in the diagnosis of pulmonary embolism: a compar- in chronic pulmonary embolism: CT findings in 33 patients.
ison between V/Q lung scintigraphy in SPECT technique and Radiology. 1997;203:355360.
multislice spiral CT. J Nucl Med. 2004;45:15011508. 54. Wetzel RC, Herold CJ, Zerhouni EA. Hypoxic bronchodila-
34. Bajc M, Olsson B, Palmer J, et al. Ventilation/perfusion tion. J Appl Physiol. 1992;73:12021206.
SPECT for diagnostics of pulmonary embolism in clinical 55. He H, Stein MW, Zalta B, et al. Pulmonary infarction:
practice. J Intern Med. 2008;264:379387. spectrum of findings on multidetector helical CT. J Thorac
35. Gutte H, Mortensen J, Jensen CV, et al. Detection of Imaging. 2006;21:17.
pulmonary embolism with combined ventilation-perfusion 56. Bergin CJ, Rios G, King MA, et al. Accuracy of high-
SPECT and low-dose CT: head-to-head comparison with resolution CT in identifying chronic pulmonary thromboem-
multidetector CT angiography. J Nucl Med. 2009;50: bolic disease. AJR Am J Roentgenol. 1996;166:13711377.
19871992. 57. Tan RT, Kuzo R, Goodman LR, et al. Utility of CT scan
36. Herald P, Roach P, Schembri GP. Does the addition of low evaluation for predicting pulmonary hypertension in patients
dose CT improve diagnostic accuracy of V/Q SPECT with parenchymal lung disease. Medical College of Wisconsin
scintigraphy? [abstract]. J Nucl Med. 2008;suppl:91P. Lung Transplant Group. Chest. 1998;113:12501256.
37. Gradinscak DJ, Roach P, Schembri GP, et al. Can CT 58. Ng CS, Wells AU, Padley SPA. CT sign of chronic
coregistration improve the accuracy of segmental localisation pulmonary arterial hypertension: the ratio of main pulmonary
on V/Q SPECT? [abstract]. Eur J Nucl Med Mol Imaging. artery to aortic diameter. J Thorac Imaging. 1999;14:
2009;suppl:S463. 270278.
38. Bolen MA, Renapurkar RD, Popovic ZB, et al. High-pitch 59. Devaraj A, Wells AU, Meister MG, et al. Pulmonary
ECG-synchronized pulmonary CT angiography versus stand- hypertension in patients with bronchiectasis: prognostic
ard CT pulmonary angiography: a prospective randomized significance of CT signs. AJR Am J Roentgenol. 2011;196:
study. AJR Am J Roentgenol. 2013;201:971976. 13001304.
39. Castaner E, Gallardo X, Ballesteros E, et al. CT diagnosis of 60. Abel E, Jankowski A, Pison C, et al. Pulmonary artery and right
chronic pulmonary thromboembolism. Radiographics. 2009; ventricle assessment in pulmonary hypertension: correlation
29:3150. between functional parameters of ECG-gated CT and right-side
40. Wittram C, Kalra MK, Maher MM, et al. Acute and chronic heart catheterization. Acta Radiol. 2012;53:720727.
pulmonary emboli: angiography-CT correlation. AJR Am J 61. Kasai H, Sugiura T, Tanabe N, et al. Electrocardiogram-
Roentgenol. 2006;186(suppl 2):S421S429. gated 320-slicemultidetector computed tomography for the
41. Auger WR, Fedullo PF, Moser KM, et al. Chronic major- measurement of pulmonary arterial distensibility in chronic
vessel thromboembolic pulmonary artery obstruction: thromboembolic pulmonary hypertension. PLoS One. 2014;9:
appearance at angiography. Radiology. 1992;182:393398. e111563.
42. Bergin CJ, Sirlin CB, Hauschildt JP, et al. Chronic 62. Helmberger M, Pienn M, Urschler M, et al. Quantification of
thromboembolism: diagnosis with helical CT and MR tortuosity and fractal dimension of the lung vessels in
imaging with angiographic and surgical correlation. Radiol- pulmonary hypertension patients. PLoS One. 2014;9:e87515.
ogy. 1997;204:695702. 63. Haitao S, Ning L, Lijun G, et al. Fractal dimension analysis
43. Korn D, Gore I, Blenke A, et al. Pulmonary arterial bands of MDCT images for quantifying the morphological changes
and webs: an unrecognized manifestation of organized of the pulmonary artery tree in patients with pulmonary
pulmonary emboli. Am J Pathol. 1962;40:129151. hypertension. Korean J Radiol. 2011;12:289296.
44. Attina` D, Niro F, Garzillo G, et al. Thoracic manifestation of 64. Rahaghi FN, Ross JC, Agarwal M, et al. Pulmonary vascular
Eisenmengers syndrome in adult patients: a MDCT review. morphology as an imaging biomarker in chronic thromboem-
Lung. 2015;193:173181. bolic pulmonary hypertension. Pulm Circ. 2016;6:7081.
45. Ley S, Kreitner KF, Morgenstern I, et al. Bronchopulmonary 65. Estepar RS, Kinney GL, Black-Shinn JL, et al. COPDGe-
shunts in patients with chronic thromboembolic pulmonary neStudy. Computed tomographic measures of pulmonary

86 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
J Thorac Imaging  Volume 32, Number 2, March 2017 Imaging in CTEPH

vascular morphology in smokers and their clinical implica- parenchymal enhancement in suspected pulmonary hyper-
tions. Am J Respir Crit Care Med. 2013;188:231239. tension. J Thorac Imaging. 2014;29:98106.
66. Remy-Jardin M, Remy J, Mayo JR, et al. Pulmonary 85. Hong YJ, Kim JY, Choe KO, et al. Different perfusion
hypertension. In: Remy-Jardin M, Remy J, Mayo JR, pattern between acute and chronic pulmonary thromboemb-
Muller NL, eds. CT Angiography of the Chest. Philadelphia, olism: evaluation with two-phase dual-energy perfusion CT.
PA: Lippincott, Williams & Wilkins; 2001:6781. AJR Am J Roentgenol. 2013;200:812817.
67. Liu M, Ma ZH, Guo XJ, et al. A septal angle measured on 86. Pontana F, Remy-Jardin M, Duhamel A, et al. Lung
computed tomographic pulmonary angiography can non- perfusion with dual-energy multi-detector row CT: can it
invasively estimate pulmonary vascular resistance in patients help recognize ground glass opacities of vascular origin? Acad
with chronic thromboembolic pulmonary hypertension. Radiol. 2010;17:587594.
J Thorac Imaging. 2012;27:325330. 87. Hachulla AL, Lador F, Soccal PM, et al. Dual-energy
68. Baque-Juston MC, Wells AU, Hansell DM. Pericardial thicken- computed tomographic imaging of pulmonary hypertension.
ing or effusion in patients with pulmonary artery hypertension: a Swiss Med Wkly. 2016;146:w14328.
CT study. AJR Am JRoentgenol. 1999;172:361364. 88. Thieme SF, Hoegl S, Nikolaou K, et al. Pulmonary
69. Ryu JH, Krowka MJ, Pellikka PA, et al. Pulmonary ventilation and perfusion imaging with dual-energy CT. Eur
hypertension in patients with interstitial lung diseases. Mayo Radiol. 2010;20:28822889.
Clin Proc. 2007;82:342350. 89. Zhang LJ, Zhou CS, Schoepf UJ, et al. Dual-energy CT lung
70. Lewis G, Hoey ET, Reynolds JH, et al. Multi-detector CT ventilation/perfusion imaging for diagnosing pulmonary
assessment in pulmonary hypertension: techniques, systematic embolism. Eur Radiol. 2013;23:26662675.
approach to interpretation and key findings. Quant Imaging 90. Kreitner KF, Kunz RP, Ley S, et al. Chronic thromboembolic
Med Surg. 2015;5:423432. pulmonary hypertensionassessment by magnetic resonance
71. Horton MR, Tuder RM. Primary pulmonary arterial hyper- imaging. Eur Radiol. 2007;17:1121.
tension presenting as diffuse micronodules on CT. Crit Rev 91. Kreitner KF, Ley S, Kauczor HU, et al. Chronic throm-
Computed Tomogr. 2004;45:335341. boembolic pulmonary hypertension: pre- and postoperative
72. Kacprzak A, Szturmowicz M, Burakowska B, et al. Abnor- assessment with breath-hold MR imaging techniques. Radi-
malities in high-resolution computed tomography of the lungs ology. 2004;232:535543.
in patients with idiopathic pulmonary arterial hypertension 92. Revel MP, Sanchez O, Couchon S, et al. Diagnostic accuracy
correlation with hemodynamic parameters and prognostic of magnetic resonance imaging for an acute pulmonary
significance. Pneumonol Alergol Pol. 2009;77:2330. embolism: results of the IRM-EP study. J Thromb Haemost.
73. Frazier AA, Franks TJ, Mohammed TL, et al. From the 2012;10:743750.
Archives of the AFIP: pulmonary veno-occlusive disease and 93. Nikolaou K, Schoenberg SO, Attenberger U, et al. Pulmo-
pulmonary capillary hemangiomatosis. Radiographics. 2007; nary arterial hypertension: diagnosis with fast perfusion MR
27:867882. imaging and high-spatial-resolution MR angiography: prelim-
74. Miura A, Akagi S, Nakamura K, et al. Different sizes of inary experience. Radiology. 2005;236:694703.
centrilobular ground-glass opacities in chest high-resolution 94. Ley S, Ley-Zaporozhan J, Pitton MB, et al. Diagnostic
computed tomography of patients with pulmonary veno- performance of state-of-the-art imaging techniques for
occlusive disease and patients with pulmonary capillary morphological assessment of vascular abnormalities in
hemangiomatosis. Cardiovasc Pathol. 2013;22:287293. patients with chronic thromboembolic pulmonary hyper-
75. Grosse C, Grosse ACT. CT findings in diseases associated tension (CTEPH). Eur Radiol. 2012;22:607616.
with pulmonary hypertension: a current review. Radio- 95. Alfakih K, Plein S, Bloomer T, et al. Comparison of right
graphics. 2010;30:17531777. ventricular volume measurements between axial and short
76. Nguyen VT, Chan ES, Chou SH, et al. Pulmonary effects of axis orientation using steady-state free precession magnetic
i.v. injection of crushed oral tablets: excipient lung disease. resonance imaging. J Magn Reson Imaging. 2003;18:2532.
AJR Am J Roentgenol. 2014;203:W506W515. 96. Gatehouse PD, Keegan J, Crowe LA, et al. Applications of
77. Kang MJ, Park CM, Lee CH, et al. Dual-energy CT: clinical phase-contrast flow and velocity imaging in cardiovascular
applications in various pulmonary diseases. Radiographics. MRI. Eur Radiol. 2005;15:21722184.
2010;30:685698. 97. Swift AJ, Rajaram S, Hurdman J, et al. Noninvasive
78. Hagspiel KD, Flors L, Housseini AM, et al. Pulmonary blood estimation of PA pressure, flow, and resistance with CMR
volume imaging with dual-energy computed tomography: imaging: derivation and prospective validation study from the
spectrum of findings. Clin Radiol. 2012;67:6977. ASPIRE registry. JACC Cardiovasc Imaging. 2013;6:
79. Dournes G, Verdier D, Montaudon M, et al. Dual-energy CT 10361047.
perfusion and angiography in chronic thromboembolic pulmo- 98. Garcia-Alvarez A, Fernandez-Friera L, Mirelis JG, et al.
nary hypertension: diagnostic accuracy and concordance with Non-invasive estimation of pulmonary vascular resistance
radionuclide scintigraphy. Eur Radiol. 2014;24:4251. with cardiac magnetic resonance. Eur Heart J. 2011;32:
80. Thieme SF, Graute V, Nikolaou K, et al. Dual energy CT 24382445.
lung perfusion imagingcorrelation with SPECT/CT. Eur J 99. Sanz J, Dellegrottaglie S, Kariisa M, et al. Prevalence and
Radiol. 2012;81:360365. correlates of septal delayed contrast enhancement in patients
81. Hoey ET, Mirsadraee S, Pepke-Zaba J, et al. Dual-energy CT with pulmonary hypertension. Am J Cardiol. 2007;100:731735.
angiography for assessment of regional pulmonary perfusion 100. Marcus JT, Gan CT, Zwanenburg JJ, et al. Interventricular
in patients with chronic thromboembolic pulmonary hyper- mechanical asynchrony in pulmonary arterial hypertension:-
tension: initial experience. AJR Am J Roentgenol. 2011;196: left-to-right delay in peak shortening is related to right
524532. ventricular overload and left ventricular underfilling. J Am
82. Meinel FG, Graef A, Thierfelder KM, et al. Automated Coll Cardiol. 2008;51:750757.
quantification of pulmonary perfused blood volume by dual- 101. Schoenfeld C, Cebotari S, Hinrichs J, et al. MR imaging-
energy CTPA in chronic thromboembolic pulmonary hyper- derived regional pulmonary parenchymal perfusion and
tension. Rofo. 2014;186:151156. cardiac function for monitoring patients with chronic
83. Takagi H, Ota H, Sugimura K, et al. Dual-energy CT to thromboembolic pulmonary hypertension before and after
estimate clinical severity of chronic thromboembolic pulmo- pulmonary endarterectomy. Radiology. 2016;279:925934.
nary hypertension: comparison with invasive right heart 102. Swift AJ, Wild JM, Nagle SK, et al. Quantitative magnetic
catheterization. Eur J Radiol. 2016;85:15741580. resonance imaging of pulmonary hypertension: a practical
84. Ameli-Renani S, Ramsay L, Bacon JL, et al. Dual-energy approach to the current state of the art. J Thorac Imaging.
computed tomography in the assessment of vascular and 2014;29:6879.

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. | 87
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
Renapurkar et al J Thorac Imaging  Volume 32, Number 2, March 2017

103. Bachler P, Pinochet N, Sotelo J, et al. Assessment of normal major-vessel thromboembolic pulmonary hypertension. Cir-
flow patterns in the pulmonary circulation by using 4D culation. 1996;94:24852491.
magnetic resonance velocity mapping. Magn Reson Imaging. 114. Nicod P, Peterson K, Levine M, et al. Pulmonary angiog-
2013;31:178188. raphy in severe chronic pulmonary hypertension. Ann Intern
104. Rajaram S, Swift AJ, Telfer A, et al. 3D contrast-enhanced Med. 1987;107:565568.
lung perfusion MRI is an effective screening tool for chronic 115. Shure D, Gregoratos G, Moser KM. Fiberoptic angioscopy:
thromboembolic pulmonary hypertension: results from the role in the diagnosis of chronic pulmonary arterial obstruc-
ASPIRE Registry. Thorax. 2013;68:677678. tion. Ann Intern Med. 1985;103(part 1):844850.
105. Ohno Y, Hatabu H, Murase K, et al. Primary pulmonary 116. Jenkins D. Pulmonary endarterectomy: the potentially
hypertension: 3D dynamic perfusion MRI for quantitative curative treatment for patients with chronic thromboembolic
analysis of regional pulmonary perfusion. AJR Am J Roent- pulmonary hypertension. Eur Respir Rev. 2015;24:263271.
genol. 2007;188:4856. 117. Moser KM, Daily PO, Peterson K, et al. Thromboendarter-
106. Schiebler ML, Bhalla S, Runo J, et al. Magnetic resonance ectomy for chronic, major-vessel thromboembolic pulmonary
and computed tomography imaging of the structural and hypertension. Immediate and long-term results in 42 patients.
functional changes of pulmonary arterial hypertension. Ann Intern Med. 1987;107:560565.
J Thorac Imaging. 2013;28:178193. 118. Sunbul M, Kivrak T, Durmus E, et al. Evaluation of right
107. Lederlin M, Bauman G, Eichinger M, et al. Functional MRI and left heart mechanics in patients with chronic throm-
using Fourier decomposition of lung signal: reproducibility of boembolic pulmonary hypertension before and after pulmo-
ventilation- and perfusion-weighted imaging in healthy nary thromboendarterectomy. Int J Cardiovasc Imaging.
volunteers. Eur J Radiol. 2013;82:10151022. 2015;31:11591167.
108. Bauman G, Scholz A, Rivoire J, et al. Lung ventilation- and 119. Rolf A, Rixe J, Kim WK, et al. Pulmonary vascular
perfusion-weighted Fourier decomposition magnetic reso- remodeling before and after pulmonary endarterectomy in
nance imaging: in vivo validation with hyperpolarized 3He patients with chronic thromboembolic pulmonary hyper-
and dynamic contrast-enhanced MRI. Magn Res Med. 2013; tension: a cardiac magnetic resonance study. Int J Cardiovasc
69:229237. Imaging. 2015;31:613619.
109. Kovacs G, Berghold A, Scheidl S, et al. Pulmonary arterial 120. Rolf A, Rixe J, Kim WK, et al. Right ventricular adaptation
pressure during rest and exercise in healthy subjects: a to pulmonary pressure load in patients with chronic
systematic review. Eur Respir J. 2009;34:888894. thromboembolic pulmonary hypertension before and after
110. Auger WR, Kerr KM, Kim NH, et al. Evaluation of patients successful pulmonary endarterectomya cardiovascular
with chronic thromboembolic pulmonary hypertension for magnetic resonance study. J Cardiovasc Magn Reson. 2014;16:
pulmonary endarterectomy. Pulm Circ. 2012;2:155162. 96.
111. Kim NH, Fesler P, Channick RN, et al. Preoperative 121. Olman MA, Auger WR, Fedullo PF, et al. Pulmonary
partitioning of pulmonary vascular resistance correlates with vascular steal in chronic thromboembolic pulmonary hyper-
early outcome after thromboendarterectomy for chronic tension. Chest. 1990;98:14301434.
thromboembolic pulmonary hypertension. Circulation. 2004; 122. Jamieson SW, Kapelanski DP, Sakakibara N, et al. Pulmo-
109:1822. nary endarterectomy: experience and lessons learned in 1,500
112. Coulden R. State-of-the-art imaging techniques in chronic cases. Ann Thorac Surg. 2003;76:14571462.
thromboembolic pulmonary hypertension. Proc Am Thorac 123. Hardziyenka M, Reesink HJ, Bouma BJ, et al. A novel
Soc. 2006;3:577583. echocardiographic predictor of in-hospital mortality and mid-
113. Pitton MB, Duber C, Mayer E, et al. Hemodynamic effects of term haemodynamic improvement after pulmonary endarter-
nonionic contrast bolus injection and oxygen inhalation ectomy for chronic thromboembolic pulmonary hypertension.
during pulmonary angiography in patients with chronic Eur Heart J. 2007;28:842849.

88 | Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.