Im Jahr 2013 publizierte die European Federation of Societies for Ultrasound in Medicine
and Biology (EFSUMB) in dieser Zeitschrift ihre „Guidelines and Recommendations“ zu
den physikalischen Grundlagen und Methoden (Teil 1) und zur klinischen Anwendung (Teil
2) der Ultraschallelastografie (US-E) [1]
[2]. Im klinischen Teil dieser Empfehlungen liegen die Schwerpunkte auf der US-E der
Leber, der Mamma, der Schilddrüse, der Prostata, dem Gastrointestinaltrakt und an
der endoskopischen Sonoelastografie, was sich auch durch zahlreiche Publikationen
zu diesem Themen in UiM-EJU und anderen Journalen in den letzten Jahren spiegelt [3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]. Unter „zukünftige Entwicklungen“ listen die Autoren des klinischen Teils der EFSUMB-Guidelines
aber US-E-Anwendungen auf, die derzeit aufgrund der Datenlage im klinischen Stellenwert
nicht ausreichend beurteilt werden können [2]. Einige dieser „exotischen“ Anwendungen finden Sie auch in diesem Heft [19]
[20]
[21] und betreffen US-E-Evaluierungen von arteriosklerotischen Plaque der A. carotis
[19], von pulmonalen Rundherden [20] und der Steifigkeit der Milz [21]. Dabei ist die Machbarkeitsstudie einer Deutschen Arbeitsgruppe zur US-E von pulmonalen
Rundherden (RH) nach bestem Wissen die erste Anwendung dieser US-Technik in dieser
Fragestellung [20] – wenn auch nicht die erste Studie, die sich mit der bildgebenden Darstellung bzw.
Messung der „Elastizität“ der Lunge mit US, CT oder MRT auseinandersetzt [22]
[23]
[24].
Ad Machbarkeitsstudie: Alles was denkbar ist, ist auch machbar (Sokrates).
In dieser Studie [20] wurden CT-Bilddaten von 8 Patientinnen mit 18 pulmonalen Sekundärblastomen herangezogen,
um in CT-Kenntnis der genauen Lokalisation und Größe dieser subpleuralen, aber nicht
an die Pleura reichenden RH diese mit Real-time-US und US-E (7,5 mHz Schallkopf, eSie
Touch Elasticity Imaging, Accuson Antares, Siemens, Erlangen, Deutschland, in der
Arbeit als real-time Elastografie [RTE] bezeichnet) aufzusuchen.
Die Ergebnisse sind überraschend: In der Real-time-Sonografie war, wie zu erwarten,
keiner der 18 pulmonalen Rundherde > 1 cm sonografisch fassbar, da keiner der Rundherde
einen direkten Kontakt zur Pleura aufwies. Mehr als überraschend konnten aber in der
US-E alle 18 Rundherde dargestellt werden, die im Real-time-US nicht zu detektieren
waren. Die Größe der Läsionen und der Abstand der Läsionen zur Pleura zeigten eine
gute Korrelation zwischen US-E und CT. Die Abb. 3 und 4 dieser Studie [20] zeigen eine einschmelzende, zentral lufthaltige Metastase im linken Unterlappen
in der CT und das entsprechende Korrelat in Real-time-US und US-E. Die Bildlegende
lautet: „Die Läsion hat einen soliden Randsaum, welcher sich in der RTE als unelastischer,
roter Kranz darstellt. Das Zentrum der Läsion erscheint in der RTE elastisch.“ In
der Diskussion prognostizieren die Autoren der RTE von RH der Lunge und der RTE der
Lunge insgesamt eine bunte, prosperierende klinische Zukunft, trotz mancher Limitationen
[20].
Der Artikel wirft Fragen auf:
-
Wie ist es physikalisch-technisch überhaupt möglich, dass im Real-time-US nicht erkennbare,
von belüftetem Lungengewebe umgebende solide RH mit US-E darstellbar sind?
Die Autoren machen sich nur wenige spekulative Gedanken, welche physikalischen Grundlagen
ihre doch überraschende Darstellung intrapulmonaler, von ventilierter Lunge umgebene
Läsionen mit einer Ultraschalltechnik – Schallwellen! – möglich machen. Die unterschiedliche
Schallgeschwindigkeit und die unterschiedliche akustische Impedanz zwischen solidem
Gewebe und Gas in den Alveolen der Lunge ermöglichen leider nicht die Real-time-US-Darstellung
intrapulmonaler Läsionen, die nicht die Pleura tangieren oder zumindest durch solide
Gewebebrücken partiell mit der Pleura verbunden sind. Das ist herrschende Lehrmeinung.
Daher sind andere Arbeitsgruppen auch auf die Messung der Oberflächengeschwindigkeit
der Lunge zur Abschätzung der Elastizität ausgewichen [22]. Gelten für die hier angewendete Elastografie andere physikalische Gesetze? Recht
salopp wird auch eine zentral lufthaltige, zerfallende Metastase beschrieben, und
das „gasgefüllte Zentrum der Läsion“ als „elastisch“ angeführt. Aber Gase haben doch
weder Form- oder Volumenelastizität! Was bedeutet das RTE-Bild, wie sollen wir die
Farbkodierung der RTE in einem gasgefüllten Hohlraum interpretieren? Viele offene
Fragen!
Diese Machbarkeitsstudie erinnert an Real-time-US und Dopplererfahrungen in den 80er-
und 90er-Jahren des vorigen Jahrhunderts mit dem Ureter-Jet in der Harnblase, der
heute ein etablierter Parameter zur Beurteilung der Physiologie und Pathologie des
Harntransportes darstellt [25]
[26]
[27]
[28]. Lange blieb unklar und wurde heftig diskutiert, wie dieses Phänomen eigentlich
zustande kommen kann, da der physiologische Urin ja keine zellulären Bestandteile
(wie das in Dopplertechniken darstellbare „fließende“ Blut) enthält, die für dieses
Phänomen verantwortlich gemacht werden konnten. Theorien erstreckten sich über abgeschilferte
Zellbestandteile, Mikrobläschen und Dichteunterschiede des Harns, aber sehr wahrscheinlich
sind Turbulenzen in einer teilchenfreien, strömenden Flüssigkeit die Grundlage dieser
Schallreflexion [29].
Diese Machbarkeitsstudie [20] sollte daher dazu führen, diese In-vivo-Ergebnisse durch fundierte In-vitro-Untersuchungen
zu erhärten und zu klären, welche physikalischen Ursachen die Darstellung von luftgefüllten
Alveolen umgebenen Läsionen ermöglichen.
-
Elastografie der Lunge oder Elastografie von Rundherden und Raumforderungen?
Die Lunge enthält zahlreiche elastische Fasern und elastisches Bindegewebe, die in
der Wiener Küche beim traditionellen Gericht eines „Kalbsbeuschels“ die Grundlage
des Essgenusses darstellen [30]. Die Kalbslunge wird vor dem Kochen „pariert“, darunter versteht man die weitgehende
Entfernung von Bronchien, großen Gefäßen und festem Bindegewebe, was beim Essen stören
würde. Der köstliche Rest ist überwiegend elastisches Bindegewebe.
Alle Pneumologen wissen viel besser als der diesen Artikel schreibende Radiologe,
dass die Elastizität der Lunge, die Compliance, eine wichtige Größe der Physiologie
und Pathophysiologie der Beatmung darstellt und Störungen der Lungenelastizität letztendlich
zu einer restriktiven Ventilationsstörung führen. Würde es daher nicht Sinn machen,
die elastischen Eigenschaften der Lunge, die gerade bei diffusen interstitiellen Prozessen
gestört sind, durch eine elastografische Methode zu untersuchen, was auch Ziele anderer
Studie sind [22]
[23]
[24]? Wäre das nicht interessanter und relevanter als die Detektion pulmonaler Rundherde
mit US-E?
Die CT ist heute die etablierte Methode der Wahl zur Erfassung intrapulmonaler Rundherde
bei neoplastischen und nicht neoplastischen Erkrankungen. Dies betrifft derzeit im
klinischen Alltag viele onkologischen Fragestellungen, welche die umfassende Beurteilung
der gesamten Lunge in Diagnose und Staging erfordern. Die Limitierung auf gut zugängliche
Läsionen mit US macht hier wenig Sinn. Intrapulmonale Raumforderungen wie primäre
oder sekundäre Tumore, aber auch benigne fokale Läsionen sind „harte“, intraoperativ
gut tastbare Läsionen in der kollabierten „elastischen“ Lunge. Solange aber mit perkutanem
US und US-E nicht die Lunge gesamthaft untersucht werden kann, beschränken sich Indikation
zur Sonografie der Lunge auf etablierte diagnostische und interventionelle Fragestellungen
[31]
[32]
[33]
[34]
[35]
[36].
In 2013 the European Federation of Societies for Ultrasound in Medicine and Biology
(EFSUMB) published its “Guidelines and Recommendations” regarding the basic principles
and technology (part 1) and the clinical applications (part 2) of ultrasound elastography
(US-E) [1]
[2] in this journal. The clinical portion of these recommendations focuses on US-E of
the liver, breast, thyroid, prostate, gastrointestinal tract and on endoscopic sonoelastography
as reflected by numerous publications regarding these topics in UiM-EJU and other
journals in recent years [3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]. However, under “future developments” the authors of the clinical portion of the
EFSUMB Guidelines list US-E applications that currently cannot be sufficiently evaluated
with respect to clinical value on the basis of the available data [2]. Some of these “exotic” applications are included in this issue [19]
[20]
[21] and relate to US-E evaluations of arteriosclerotic plaque of the carotid artery
[19], pulmonary nodules [20], and spleen stiffness [21]. To the best of knowledge, the feasibility study by a German workgroup regarding
US-E of pulmonary nodules is the first use of this US technique for this purpose [20] even if it is not the first study addressing the imaging or measurement of pulmonary
“elasticity” with US, CT, or MRI [22]
[23]
[24].
Feasibility study: Whatever is conceivable is possible (Socrates).
CT data from 8 patients with 18 pulmonary metastases were included in this study [20] to use real-time US and US-E (7.5 mHz transducer, eSie Touch Elasticity Imaging,
Accuson Antares, Siemens, Erlangen, Germany, referred to as real-time elastography
(RTE) in the study) to locate these lesions with CT knowledge of the exact location
and size of these subpleural nodules that do not extend to the pleura.
The results are surprising: As expected, none of the 18 pulmonary nodules > 1 cm could
be visualized with real-time ultrasound since none of the nodules was in direct contact
with the pleura. However, it is more than surprising that all 18 nodules that could
not be detected with real-time US could be visualized with US-E. There was a good
correlation between US-E and CT with regard to the size of the lesions and the distance
of the lesions from the pleura. Figures 3 and 4 of this study [20] show a liquefactive metastasis with air in the center located in the left inferior
pulmonary lobe in CT and the corresponding correlate in real-time US and US-E. The
caption reads as follows: “The lesion showed a spreaching solid border which is reproducible
as an inelastic, red margin in RTE. The center of the lesion appeared elastic in RTE.”
In the discussion the authors predict a bright future for RTE of pulmonary nodules
and RTE of the lung despite certain limitations [20].
The article raises some questions:
-
How is it physically-technically even possible that solid nodules that are surrounded
by aerated lung tissue and are not visible with real-time US can be visualized with
US-E?
The authors only speculate briefly as to which physical basic principles allow the
surprising visualization of intrapulmonary lesions surrounded by ventilated lung tissue
using ultrasound technology (i. e., sound waves!). The different velocities of sound
and different acoustic impedances between solid tissue and gas in pulmonary alveoli
unfortunately do not allow real-time US visualization of intrapulmonary lesions that
are not tangent to the pleura or are at least partially connected to the pleura via
solid tissue bridges. That is the prevailing scientific opinion. Therefore, other
workgroups have decided to measure the surface wave speed of the lung in order to
estimate elasticity [22]. Do other laws of physics apply to the elastography used here? The article rather
sloppily describes a liquefactive metastasis with air in the center and then refers
to the “gas-filled center of the lesion” as “elastic”. Gases have neither form nor
volume elasticity! What does the RTE image mean? How is the RTE color coding in a
gas-filled cavity to be interpreted? So many unanswered questions!
This feasibility study is reminiscent of the use of real-time US and Doppler in the
1980s and 1990s to assess the ureteral jet in the bladder, which is now an established
parameter for assessing the physiology and pathology of urine transport [25]
[26]
[27]
[28]. For a long time the mechanism of this phenomenon was unclear and intensely debated
since physiological urine does not contain any cellular components (like in the case
of “flowing” blood as can be visualized with Doppler techniques) that could explain
the phenomenon. Theories included peeled-off cell components, microbubbles, and urine
density differences. However, turbulences in a particle-free, flowing liquid are very
probably the reason for the sound reflection [29].
This feasibility study [20] should therefore enhance in-vitro studies to clarify the physical reasons behind
the visualization of lesions surrounded by air-filled alveoli.
-
Elastography of the lung or elastography of nodules and masses?
The lung contains numerous elastic fibers and elastic connective tissue and this is
what makes the traditional Viennese calf lung dish such a culinary delight [30]. Prior to cooking, the calf lung is trimmed, which means that the difficult-to-eat
bronchi, large vessels, and solid connective tissue are removed. What remains is
the delicious elastic connective tissue.
All pulmonologists know far better than the radiologist writing this article that
the elasticity of the lung, the compliance, is an important parameter of the physiology
and pathophysiology of ventilation and that lung elasticity defects can ultimately
result in a restrictive ventilatory disorder. Therefore, wouldn’t it make sense to
use an elastographic method to examine the elastic properties of the lung that are
impaired in the case of diffuse interstitial processes in particular as is the objective
of other studies [22]
[23]
[24]? Wouldn’t that be more interesting and more relevant than the detection of pulmonary
nodules with US-E?
CT is currently the established method of choice for detecting intrapulmonary nodules
in neoplastic and non-neoplastic diseases. This currently affects many oncological
issues in the clinical routine that require comprehensive evaluation of the entire
lung for diagnosis and staging. A limitation to lesions that are easily accessible
with US does not make much sense here. Intrapulmonary masses such as primary or secondary
tumors as well as benign focal lesions are “hard” lesions that are easily intraoperatively
palpable in the collapsed “elastic” lung. As long as the lung cannot be examined in
its entirety with percutaneous US and US-E, the indication for ultrasound of the lung
will be limited to established diagnostic and interventional issues [31]
[32]
[33]
[34]
[35]
[36].