Abstract
Increased incidence of liver diseases, the development of liver surgery and other invasive methods for managing portal hypertension, plus an increasing number of liver transplant procedures pose more and more new challenges for ultrasonography. Ultrasonography, being an effective and clinically verified modality, has been used for several decades for diagnosing diseases of the liver, its vessels and portal hypertension. It is used for both initial and specialist diagnosis (performed in reference centers). The diagnostic value of ultrasonography largely depends on the knowledge of anatomy, physiology, pathophysiology and clinical aspects as well as on the mastering of the scanning technique. In the hands of an experienced physician, it is an accurate and highly effective diagnostic tool; it is of little use otherwise. The paper presents elements of anatomy, physiology and pathophysiology which make the portal system exceptional and the knowledge of which is crucial and indispensable for a correct examination and, above all, for the correct interpretation of results. The authors also present requirements regarding the equipment. Moreover, various technical aspects of the examination are presented and the normal morphological picture and hemodynamic parameters of healthy individuals are described. The authors discuss the most common clinical situations and rare cases during ultrasound examinations. The paper is based on the experience of the author who works in the largest center of liver diseases in Poland, and on the current literature.
Keywords: portal vein, hepatic veins, portal hypertension, Doppler ultrasound, liver transplantation
Abstract
Wzrost zachorowań na choroby wątroby oraz rozwój chirurgii wątroby i innych inwazyjnych metod leczenia nadciśnienia wrotnego, a także zwiększająca się liczba zabiegów transplantacji wątroby wyznaczają wciąż nowe wyzwania dla ultrasonografii. Ultrasonografia, jako skuteczna i sprawdzona klinicznie metoda, stosowana jest od kilku dekad w diagnostyce chorób wątroby, jej naczyń i nadciśnienia wrotnego. Wykorzystywana jest zarówno na poziomie diagnostyki wstępnej, jak i specjalistycznej – wykonywanej w ośrodkach referencyjnych. Wartość diagnostyczna ultrasonografii w dużym stopniu zależy od znajomości anatomii, fizjologii, patofizjologii i aspektów klinicznych, a także stopnia opanowania techniki badania. W rękach doświadczonego lekarza metoda ta jest precyzyjnym i bardzo skutecznym narzędziem diagnostycznym, w przeciwnym razie jest bezużyteczna. W opracowaniu omówiono podstawowe elementy anatomii, fizjologii i patofizjologii, które stanowią o wyjątkowości układu wrotnego, a których znajomość jest kluczowa i niezbędna dla prawidłowego wykonania badania, a przede wszystkim właściwej interpretacji wyników. Przedstawiono wymagania dotyczące zaawansowania wykorzystywanej aparatury. Omówiono różne techniczne aspekty badania oraz prawidłowy obraz morfologiczny i parametry hemodynamiczne u osób zdrowych, a także najczęstsze sytuacje kliniczne i związane z nimi odchylenia od norm w wykonywanych badaniach. Prezentowana praca oparta jest na kilkunastoletnim doświadczeniu autora pracującego w największym ośrodku chirurgii wątroby w Polsce oraz na podstawie aktualnego piśmiennictwa.
Introduction
Ultrasonography, being an effective and clinically verified modality, has been used for several decades for diagnosing diseases of the liver, its vessels and portal hypertension. It is used for both initial and specialist diagnosis (performed in reference centers). Increased incidence of chronic liver diseases, the development of liver surgery and invasive methods for managing portal hypertension, as well as an increasing number of liver transplant procedures pose more and more new challenges for ultrasonography and for physicians who use this method.
The diagnostic value of ultrasonography largely depends on the knowledge of anatomy, physiology, pathophysiology and clinical aspects of pathologies, and on the mastering of the scanning technique. In the hands of an experienced physician, it is an accurate and highly effective diagnostic tool; it is of little use otherwise.
The liver receives double blood supply: 75% from the portal vein and 25% from the hepatic artery. The portal vein is approximately 30–90 mm long and 6–12 mm wide(1). It delivers blood from unpaired abdominal organs to the liver. The vein is formed by the confluence of the splenic and superior mesenteric veins. It originates behind the head of the pancreas. The vein then runs towards the hepatic hilum in the posterior part of the hepatoduodenal ligament, medially to the bile ducts and laterally to the hepatic vein proper. Together with the accompanying hepatic artery, it divides into lobar and segmental vessels, the terminal branches of which join to create a capillary network. Portal and arterial blood mix at the level of the hepatic sinusoids and then run to the central hepatic veins and terminal hepatic venules which gradually join into larger veins and communicate with the inferior vena cava as the hepatic veins (right, middle and left). The system of the hepatic veins and portal vein branches is the basis for Couinaud's division of the liver into segments.
The portal system is a low-pressure, valve-free vascular bed with diversified pressure which decreases with the direction of blood flow. Normal pressure in the splenic vein ranges from 9 to 11 mm Hg, in the superior mesenteric vein – 9–10 mm Hg, and in the portal vein and its branches – 6–8 mm Hg. In the hepatic sinusoids, mixing of portal and arterial blood results in pressure of 3–5 mm Hg. Blood pressure in the hepatic veins is 1–2 mm Hg. A physiological pressure gradient ensures spontaneous continuous hepatopetal flow.
The gold standard in the assessment of pressure in the sinusoids, which reflects blood pressure in the portal vein, is the measurement of wedged pressure in one of the hepatic veins. A gradient between wedged pressure and pressure measured with a so-called “free” catheter in the hepatic veins (hepatic venous pressure gradient, HVPG) is an important parameter. Its normal values do not exceed 10 mm Hg. Higher values indicate intrahepatic, postsinusoidal block which in 85–95% of cases is caused by liver cirrhosis(2–4).
Hepatic flow blocks are divided anatomically into prehepatic (e.g. portal or splenic vein thrombosis), intrahepatic (liver cirrhosis) and posthepatic (Budd–Chiari syndrome). The functional division, based on gradient measurement, distinguishes between intrahepatic, presinusoidal and postsinusoidal blocks. The greater the gradient, the greater the hypertension and the higher the risk of its complications, such as esophageal varices, bleeding, ascites, decompensation of liver function or hepatorenal syndrome (HRS).
Portal hypertension causes increased flow resistance in the hepatic sinusoids (due to fibrosis, scarring and formation of regenerative nodules) as well as an increase in total amount of blood flowing through the portal system (as a result of an increase in the visceral arterial inflow stimulated by vasodilators released to the bloodstream) (4). Thus the portal flow in hypertension is hyperdynamic. Increased pressure in the portal system that is higher than 15 mm Hg (gradient is greater than 10 mm Hg) is compensated by increased volume of the vascular bed (enlarged spleen, dilated venous lumina) and opening of extrahepatic anastomoses with the systemic venous system and intrahepatic ones that omit the sinusoids(5). Approximately 65–90% of collateral vessels can be visualized in ultrasonography(6).
The most common portosystemic connections with hepatofugal flow include(4–6)
gastroesophageal connections (supplied by the left gastric vein and short gastric veins in the region of the fundus and cardia of the stomach; create esophageal varices; drain to the azygos vein);
abdominal veins (dilated veins in the abdominal wall; supplied by patent umbilical vein; drain through the inferior epigastric veins to the external iliac veins) form so-called “caput medusae” sign;
rectal plexuses (upper – supplied by the inferior mesenteric vein; central and lower – drained by the pudendal veins to the internal iliac veins) create anal varices;
retroperitoneal veins (veins of Retzius) are the least significant from the clinical point of view (Fig. 1 A, B);
spleorenal connections (large, tortuous vessels between the splenic hilum, upper pole of the left kidney and the left renal vein).
Fig. 1.
Retroperitoneal venous plexuses (veins of Retzius). A. Morphological image. B. Color Doppler.
Furthermore, collateral circulation channels within the portal system with hepatopetal flow may become active. They are visualized in an ultrasound examination in e.g. regional portal hypertension, in the course of partial thrombosis.
Sappey's veins – venous plexuses in the wall of the gallbladder; drain through the cystic vein to the right branch of the portal vein in the event of portal vein thrombosis (Fig. 2);
so-called “cavernous transformation” of the portal vein – multiple tortuous collateral vessels of the hepatoduodenal ligament around the occluded portal vein (Fig. 3 A, B, C);
pancreaticoduodenal veins – drained by the left gastroomental vein to the superior mesenteric vein;
accessory portal veins – in the falciform and hepatogastric ligaments.
Fig. 2.
Veins in the gallbladder wall (veins of Sappey) – power Doppler
Fig. 3.
“Cavernous transformation” of the PV. A. Morphological image. B. Power Doppler. C. Spectral recording of flow
Equipment
In many cases, the evaluation of portal system and hepatic veins is a complex problem, and usually concerns so-called difficult patients. The equipment requirements are therefore high. The difficulty in examining patients with chronic liver diseases is a consequence of disorders in the liver size and shape as well as echogenicity of its parenchyma and echostructure, which can considerably blur the image of anatomic structures, including vessels. Moreover, ascites and intestinal flatulence make the vessels poorly visible and poorly accessible at the angle of insonation appropriate for Doppler assessment. Blood flow velocities in the portal system, particularly in pathological conditions, are relatively low and frequently recorded from considerable depths among numerous artefacts. At the same time, flow velocities are high when vessels are narrowed, in the arterioportal and arteriovenous shunts as well as in patients after transjugular intrahepatic portosystemic shunt (TIPS). This is a considerable challenge for the Doppler mode.
Since flow assessment in the portal system and hepatic veins is conducted with the patient breathing normally, the scanner should be equipped with the triplex Doppler mode (but the duplex Doppler mode can also be used for assessment of frozen B-mode images).
Therefore, premium-class equipment should be used ensuring good B-mode imaging, tissue harmonic imaging (THI), sensitive and full-range spectral Doppler, color Doppler and power Doppler.
The scanner should be outfitted with a conventional broadband 2–5 MHz convex probe for evaluation of the abdominal structures as well as a typical linear probe (5–10 MHz) for evaluation of the anterior aspect of the liver in search for cirrhotic changes. Furthermore, specialist centers must be equipped with a scanner able to operate with a low mechanical index (MI lower than 0.1), with second harmonic imaging by reverse impulse phase to enable contrast-enhanced imaging (CEUS) with the use of second-generation contrast agents. This is needed for the assessment of the liver parenchyma, its focal lesions and patency of the vessels in the event of doubtful results of the Doppler examination. This does not mean, however, that the portal vein and hepatic vessels cannot be examined with less sophisticated scanners. Examinations at the level of initial diagnosis, of follow-up scans to check-up the portal system in patients with chronic liver diseases with no advanced cirrhosis or portal hypertension, can be conducted with the use of medium-class scanners.
Preparation for examination
The examination should be conducted with the patient fasting (the last meal ingested 6–8 hours before the examination). It is preferable to conduct the examination in the morning. At least 30 minutes before the test, the patient should refrain from physical exertion. To reduce the amount of gas in the intestine, oral agents to decrease surface tension can be used on the day before the scan(2).
Scanning technique
Visualization of the hepatic vessels and the entire portal system in the B-mode is a part of abdominal examination, particularly the examination of the epigastric region, that involves the assessment of the liver, pancreas and spleen, with the use of these organs as “acoustic windows” enabling the evaluation of their surroundings and vessels running in the parenchyma and in the vicinity of these organs.
The examination is conducted via the subcostal and intercostal access and from below the xiphoid process. The right subcostal access enables assessment of the larger part of the liver, hepatic veins and their confluence as well as the inferior vena cava and the portal vein. A similar, or sometimes even better insight into these structures is provided by the right lateral access, through the intercostal spaces. It provides a good angle of insonation for the Doppler assessment of the right and middle hepatic veins, inferior vena cava and portal vein and its right branch. The access from below the xiphoid process and the right subcostal access at the right pararectus line guarantee a good insight into the left and middle hepatic veins, their confluence and, above all, into the left branch of the portal vein; the angle of insonation can be even 0–10°.
In order to visualize and assess the morphology of the liver, spleen and pancreas, perpendicular, transverse, longitudinal and oblique planes are used. The patient must be examined in a systematized way in the supine position with the arms resting along the body or above the head, and in the left and right lateral decubitus position. This reduces the risk of overseeing pathologies, enables better visualization of the structures examined due to movement and elimination of obstacles, such as intestinal loops. Moreover, it makes it easier to distinguish artefacts.
If possible, vascular structures should be visualized on their entire course along the long axis and transversely. The assessment should begin at their origin an continue up to their branching or confluence; branches should be identified.
In order to make the assessment of the vascular walls and lumina reliable and to enable accurate width measurements, it is best to use the angle of insonation of 90° in the B-mode without using the color Doppler. In the Doppler assessment, i.e. to evaluate the patency of the vessels as well as the direction and velocity of blood flow, it is necessary to find such access sites to obtain the angle between the vessel axis and the direction of insonation below 60°, or even below 40°. The lower the access angle, the stronger and more unequivocal Doppler signal in terms of direction and velocity of flow(7).
The examination should be conducted in different respiratory phases, depending on the needs. Frequently, deep inspiration is needed to visualize certain regions. However, when portal vessels and hepatic veins are assessed, the measurements should be taken during normal breathing or while holding normal breath. Holding a deep inspiration may increase the width of the portal vessels and change their blood flow thus dampening the physiological phasicity in the hepatic veins.
Subsequently, blood flow is assessed with the use of the color Doppler. Based on the color-coded Doppler signal, or the lack of it, it is possible to determine whether the vessel is patent or not and what the direction of flow is, as well as if blood flows in the entire lumen or whether there are signal defects suggesting, e.g. perimural thrombi. If there are doubts concerning patency, particularly resulting from technical problems, the color Doppler mode should be turned off and the spectral Doppler, which is more sensitive, should be used for flow determination. The Doppler waveform enables the assessment of the character of flow (turbulent, laminar, venous, arterial, low- or highresistance, arterialized venous, venous with pulsations or oscillations, or without them) as well as the measurements of the maximum, minimum and mean velocities, resistivity indices, pulsation indices and other quantitative parameters.
The color velocity scale should be set based on current assessment at 1/3–1/2 of the maximum velocity, with gain at approximately 75%. In practice, initial color scale for portal flows should be set at -15 cm/s to +15 cm/s. If flows are very slow, it should be lowered to 10 cm/s or lower.
The spectral Doppler reading should occupy 1/2 or 2/3 of the scale. This enables a thorough analysis of the waveform and improves measurements of quantitative parameters. One should also remember about the wall filter value; it needs to be reduced from 100 Hz to 50 Hz if flow velocities are low. The Doppler gate should be set at 50% of the vessel's width in its central part. One should register the reading from at least three cardiac cycles or for 2–3 seconds. The flow velocity in the portal vein is calculated from the manual or automatic outline of the Doppler spectrum. It is a mean value of maximum velocities registered at a given time. This is so-called TAM velocity – time averaged max velocity (1, 2, 6, 7).
Normal presentation of vessels in Doppler Ultrasonography
Portal vein (PV):
the width, measured during normal breathing 2 cm above the branching at the site where it crosses the hepatic artery, varies: 6–13 mm, usually 9–11 mm, and when the patient inhales deeply, it may increase to 16 mm (it also increases after a meal and depends on the body surface);
continuous hepatopetal, monophasic flow with slight oscillations of the waveform that depend on the cardiac cycle and respiratory phases (Fig. 4);
maximum velocity: 15–30 cm/s (increases after a meal from 50% to 100%, declines after assuming a vertical position and upon physical exertion)(8);
mean velocity (Vmean): 12–20 cm/s, TAM >20 cm/s.
Fig. 4.
Normal flow in the PV with slight waveform oscillations
Superior mesenteric vein (SMV):
diameter: 4–13 mm, measured 2 cm above the confluence;
maximum velocity: 8–25 cm/s;
mean velocity: 12–18 cm/s.
Splenic vein (SV):
diameter: 5–10 mm, measured 2 cm above the confluence;
maximum velocity: 9–30 cm/s;
mean velocity: 12–16 cm/s.
In normal conditions, the diameters of SMV and SV increase upon a deep inspiration by 50–100%(1).
Hepatic veins (HVs):
diameter ≤1 cm;
velocity: 16–40 cm/s;
triphasic flow associated with the cardiac cycle – the changes in pressures in the right atrium, so-called type I (it is assessed upon holding normal breath; deep inspiration dampens pulsations – flattens the waveform) (Fig. 5).
Fig. 5.
Triphasic flow in the HV – type I
Inferior vena cava (IVC):
the diameter does not usually exceed 2 cm and depends on the respiratory phase;
tri- of tetraphasic flow;
tends to collapse on expiration and dilate on inspiration.
Hepatic artery (HA) (Fig. 6):
diameter: 3.5–4.5 mm;
systolic velocity: 40–80 cm/s;
diastolic velocity: 15–40 cm/s;
resistivity index (RI): 0.59–0.8,
pulsatility index (PI): 1.0–1.5.
Fig. 6.
Normal flow in the HA
The hemodynamic normal ranges are broad. Frequently, normal values reported by one author overlap with values describes as pathological by other authors. It should be in mind that the results of morphological and hemodynamic parameters are imperfect and may depend on technical conditions and possibilities, experience of the examiner and equipment. These values are then approximate. This is confirmed by the author's own material. There are still no randomized studies with appropriate number of patients(2, 3, 5–7).
Assessment of pathological lesions
Portal hypertension in a US examination(2, 3, 5–8)
PV diameter: >13 mm, sensitivity: <50%, specificity: 90–100%;
SV and SMV: >11 mm, sensitivity: 72%, specificity: 100%
an increase in PV, SMV and SV diameter by <20–40% or less on deep inspiration (the parameter is more sensitive than PV width measurements, sensitivity: 79.7%, specificity: 100%);
splenomegaly (size along the long axis >12 cm, area: >45 cm2, sensitivity: 93%, specificity: 36%);
macroscopic features of cirrhosis (nodular liver surface, coarse-grained parenchymal echotexture, smaller right lobe, enlarged caudate lobe and left lateral segments, ascites);
hepatofugal flow in the PV, SMV and SV, specificity 100%; rarely observed;
no undulation of the Doppler waveform in the PV (Fig. 7);
slower or oscillating flow or no flow;
presence of portosystemic collateral vessels, sensitivity: 83%, specificity: 100%, dilated umbilical vein >3 mm, dilated left gastric vein >5 mm and in considerable hypertension – above 7 mm;
venous “aneurysms” in the PV and SV (Fig. 8 A, B);
loss of Doppler waveform phasicity in the HVs (biphasic waveform, so-called type II, observed in approximately 31% of patients with cirrhosis or monophasic, type III, waveform observed in 18% of cirrhotic patients) (Fig. 9 A, B);
RI in the renal artery >0.7, measured intrarenally (may indicate hepatorenal syndrome);
RI in the splenic artery >0.63, measured in the parenchyma;
dilated and tortuous hepatic artery due to an increase in arterial flow in hypertension and cirrhosis.
Fig. 7.
Hepatopetal flow in the PV, with nor oscillations, in portal hypertension
Fig. 8.
“Venous aneurysm.” A. Morphological image. B. Color Doppler
Fig. 9.
A. Biphasic flow in the HV – type II. B. Monophasic flow in the HV – type III
Hyperkinetic portal hypertension(9)
a particular form of portal hypertension caused by increased blood volume in the portal system in the course of arterioportal fistula (APF);
APF can be intrahepatic or extrahepatic;
PV flow is always hepatopetal with decreased velocity in intrahepatic fistulae and increased velocity in extrahepatic ones;
in extrahepatic APF, blood flow in both PV branches is hepatopetal;
in intrahepatic APF, blood flow in the PV branch that drains the fistula is always hepatofugal, and in the other branch, it is hepatopetal;
intrahepatic fistulae are focal fluid areas with the dilated draining PV branch and dilated supplying artery with rapid, turbulent and low-resistance flow;
blood flow in the venous bed is pulsatile, low-resistance or biphasic, consistent with the pulse (so-called arterialization);
in CEUS, the signal of flow in the portal branch draining the intrahepatic fistula and in the PV appears in the early arterial phase (7–10 s).
Portal hypertension in the course of cardiac failure (cardiogenic transsinusoidal shunting, CTS)(9)
the PV waveform – pulsatile, hump-shaped and susceptible to pressure changes in the right atrium (in this mechanism, portal pressure raises periodically);
hepatopetal flow with inverted, periodical hepatofugal waves or without them (Fig. 10 A);
dilated HVs and IVC, also with increased pulsation (Fig. 10 B, C).
Fig. 10.
A. Pulsatile, hump-shaped flow waveforms in the PV – CTS. B. Dilation of the HVs – CTS. C. Deeper pulsations in the hepatic vein with inverted S wave – CTS
Portal vein obstruction in a US examination
Portal vein thrombosis (PVT) or neoplastic invasion are the most common causes. The sensitivity of the color Doppler in detecting thrombosis is 93% and specificity – 99%(3).
- acute thrombosis:
- hypoechoic or anechoic clots in the venous lumen;
- blurred wall;
- PV dilation;
- absence of flow (Fig. 11 A);
- slowed or accelerated flow at the stenosed site or inverted flow (in partial thrombosis) (Fig. 11 B). It is important to adjust Doppler settings for low flows and to the angle <60°; in the event of doubts, the examination can be repeated after a meal, or a CEUS should be performed;
- chronic thrombosis:
- the PV is often narrow and invisible, hyperechoic and fibrotic;
- so-called cavernous transformation – numerous tortuous slight collateral vessels with hepatopetal flow running in the hepatoduodenal ligament; they surround the occluded and fibrotic PV; it develops after 6–20 days following PVT and, together with arterial hyperperfusion, ensures good blood supply to the liver(6);
- neoplastic invasion:
- the PV dilated to 20–25 mm usually suggests that the lumina are filled with neoplastic masses;
- arterial vessels can develop within the tissue masses growing into the vascular lumina (Fig. 11 C).
Fig. 11.
A. Acute complete PVT. B. Partial PVT – power Doppler. C. Neoplastic “thrombosis” – cancerous masses in the lumen of the left branch of the PV; arterial vasculature with hepatofugal flow
Budd–Chiari syndrome in a US examination(3, 5–7)
This rarely observed compromised venous conf luence is manifested with abdominal pain and massive, treatment-resistant ascites. As a primary syndrome, it usually develops due to thrombosis; as a secondary syndrome – due to neoplastic invasion or external compression. It may be seen only in the hepatic veins or in the inferior vena cava, or in both of these vessels.
Acute Budd–Chiari syndrome:
the HVs are dilated, partially or completely filled by hypoechoic clots;
the HVs can be invisible with no signs of flow;
the liver is enlarged, edematous; the parenchyma is hypoechoic or heterogeneous due to edema, hemorrhagic lesions and perfusion disorders;
segment 1 is not enlarged or is slightly enlarged;
PV flow is slower and continuous with no Doppler waveform oscillation; hepatofugal flow is possible;
Chronic Budd–Chiari syndrome:
the clinical presentation depends on the severity of hepatic parenchyma damage;
considerable compensatory hypertrophy of segment 1 with its veins (invisible in normal conditions) dilated to 3 mm;
liver deformity, atrophy of the lateral aspects of both lobes, hypertrophy of the central segments, i.e. 8 and 4a, fragmentary pseudo-nodular hypertrophy of the parenchyma, increased echogenicity of the parenchyma in the fibrotic portions of the liver;
the hepatic veins are invisible, occluded or partially patent with stenoses;
inverted flow with a monophasic waveform in the fragments of patent hepatic veins with the redistribution of blood to patent accessory veins by intrahepatic collateral vessels;
PV flow can be slower, inverted or absent – thrombosis;
Budd-Chiari syndrome leads to secondary portal hypertension.
Veno-occlusive disease (VOD) in patients after radiotherapy, chemotherapy, bone marrow transplantation and toxic injuries gives similar clinical symptoms to Budd-Chiari syndrome. By contrast with Budd-Chiari syndrome, hepatic veins are patent, but PV flow can be slower, inverted or oscillating. PV thrombosis is possible.
Ultrasound assessment of TIPS (transjugular intrahepatic portosystemic shunt)(2, 5–7)
TIPS is conducted by creating an anastomosis between the hepatic vein and branches of the portal vein with the use of expandable metal stents (8–12 mm wide). It is usually conducted to connect the right hepatic vein with the right branch of the portal vein. In the liver parenchyma, the ultrasound examination reveals a hyperechoic tubular structure that is reticular or solid (in coated stents) with a gentle arching course.
Normal hemodynamic parameters:
the flow in the stent is slightly pulsatile, turbulent and rapid with the velocity of at least 50–60 cm/s (usually 90–120 cm/s). The velocities of 100–200 cm/s can also be observed (Fig. 12 A);
the flow velocities measured at the portal end, in the stent and at the venous end are similar;
PV and SV flow is hepatopetal with the velocity of 35–45 cm/s;
hepatofugal flow can be observed in the left branch of the portal vein.
Fig. 12.
A. TIPS – normal flow. B. TIPS – occlusion, the absence of flow in the color Doppler
Abnormal hemodynamic parameters:
velocity >220 cm/s, highly turbulent;
velocity changes within the stent >100 cm/s;
velocity drops in two consecutive examinations by >50 cm/s;
visible stenosis;
stent flow velocity <50 cm/s;
continuous, pulsation-free flow;
decrease in the velocity in the PV compared to the value obtained prior to TIPS;
development of hepatofugal or oscillating flow in the PV and SV;
absence of flow (Fig. 12 B);
recurrence of esophageal varices and development of ascites.
The aim of a US examination is to qualify patients for? TIPS, assess the position, patency and flow parameters after the procedure and monitor the patient over time in order to identify stenosis before recurrence of portal hypertension symptoms.
Application of a US examination for diagnosing esophageal varices (EV)(10)
It is attempted to find less invasive methods for diagnosing varices and variceal bleeding than HVPG and endoscopy. Despite initial promising reports, none of the ultrasound parameters have become commonly acknowledged in a non-invasive algorithm to monitor EV.
Nevertheless, varices are more likely to develop if:
the diameter of the portal vein is >13 mm;
the resistivity index in the renal artery is >0.7.
The risk of variceal bleeding is greater when the following are observed:
splenomegaly >15 cm;
congestion index (CI) >0.154 cm/s (ratio of the PV crosssectional area to mean velocity – upper limit: 0.07 cm/s); sensitivity 70% and specificity 64.9%;
PHI >2.08 (Portal Hypertensive Index = HARI × 0.69 × SARI × 0.87/PVVmean);
presence of new collateral vessels in the subsequent examination.
It seems that monitoring of the aforementioned parameters is more important than the interpretation of individual measurements.
Ultrasound assessment of hepatic vessels after orthotropic liver transplantation (OLTx)(5)
One should assess the patency of the artery, portal vein and hepatic veins as well as functioning of the vascular anastomoses. The most important parameter is arterial patency.
HA (normal values):
low-resistance flow, RI 0.5–0.8;
acceleration time (AT) <80 ms.
HAS (hepatic artery stenosis):
an increase in velocity, PSV >200 cm/s and turbulence at the site of stenosis (Fig. 13 A);
RI <0.5 – distally to the site of stenosis (Fig. 13 B);
AT >80 ms – distally to the site of stenosis (Fig. 13 C).
Fig. 13.
A. HAS – Doppler waveform in the stenosed area (PSV 226 cm/s). B. HAS – decreased RI (0.37) – distally to the stenosed area. C. HAS – prolonged acceleration time (104 ms) – distally to the stenosed area
HAT (hepatic artery thrombosis):
absence of flow in the visible artery;
absence of the flow signal in the region of the hilum and intrahepatically.
HAP (hepatic artery pseudoaneurysm):
it not always develops at the site of an anastomosis (it may develop due to fungal infection of percutaneous intervention);
rupture or fistula to the bile ducts and portal vein are possible.
PV (normal value):
hepatopetal turbulent flow with the mean velocity of 40 cm/s.
PVS (portal vein stenosis) (Fig. 14 A, B):
severe turbulence and increased flow velocity to over 100 cm/s, or its 3–4-fold increase.
Fig. 14.
A. PVS – morphological image. B. PVS – Doppler examination; turbulence and increased velocity to 125 cm/s
PVT:
absence of flow, thrombotic masses in the vascular lumen;
in partial thrombosis, flow is observed in a part of the lumen.
Disturbance of venous confluence at the level of anastomosis: stenosis or twisting of the IVC:
severe turbulence and 3–4-fold increase of flow velocity (Fig. 15);
flow slower than 15 cm/s in the HVs, absence of the Doppler waveform phasicity.
Fig. 15.
Stenosis of IVS anastomosis
Test description
The assessment of the portal system and hepatic vessels is an indispensable part of an abdominal ultrasound examination.
All parenchymal abdominal organs and the bile ducts with the gallbladder must be described. If ascites is present, its grade must be specified. In the case of the liver, one must describe its size, outlines and echostructure (to assess possible parenchymal damage, cirrhosis), possible focal lesions with their differentiation and their relationship with vascular structures (hepatocellular carcinoma may cause infiltration and thrombosis of the portal vessels or, more rarely, hepatic veins).
The spleen and pancreas should be described in a similar fashion (inflammatory tumors may cause infiltration and thrombosis of the SV and SMV and consequently, lead to regional portal hypertension). Subsequently, hepatic veins should be assessed: the right, middle and left one as well as their confluence, possible accessory veins, their width, patency and character of flow – phasicity. The assessment of the inferior vena cava is similar.
Subsequently, the portal vein with its lobar, sectorial and segmental branches must be assessed in terms of: their width, lumina, patency, direction of blood flow, character of the Doppler waveform and the presence or absence of physiological changes in flow and width of the vessels depending on the respiratory and cardiac cycles. A similar approach is conducted for the splenic, superior and inferior mesenteric veins if it is accessible for evaluation.
Next, one should search for collateral vessels or portosystemic vessels, which in physiological conditions are not visible in ultrasonography. In cases of portal hypertension, the most commonly assessed vessels are: venous plexuses in the splenic hilum and at the lesser curvature of the stomach, dilated left gastric vein or dilated patent umbilical vein. Their evaluation involves their width, patency and, if possible, the direction and velocity of blood flow (the tortuous course of the veins makes the examination difficult).
It is necessary to assess arteries (hepatic artery, splenic artery and superior mesenteric artery), the character of Doppler waveform, flow velocity and resistivity indices: RI and PI. They can help assess the portal system, but they are also basic and essential parameters in diagnosing arterioportal fistulae and compensatory arterial hyperperfusion in the liver when portal perfusion is reduced due to liver cirrhosis or thrombosis of the portal vein and/or its lobar branches.
Documentation
The examination can be documented either electronically as video recordings or in single “frozen” images (black and white or in color). The latter are more common and sufficient.
The scope of documentation depends on the anatomic situation, causes of portal hypertension and progression of lesions.
The following must be documented:
signs of cirrhotic transformation of the liver, possible liver tumors, size of the spleen and ascites;
image of the hepatic veins, together with their width measurements, and of the Doppler waveform;
PV, SV and SMV width at normal respiration or when holding breath, and at deep respiration;
recording of the Doppler waveform with mean velocities in the PV, SV and SMV and possibly also in the selected collateral vessels;
2D image of thrombi in the veins of the portal system supplemented with color Doppler or power Doppler images;
recording of the Doppler waveform from the HA, SA and intrarenal arteries (if cirrhosis and hepatorenal syndrome are suspected);
color and spectral Doppler images of morphology in stents (patients after TIPS);
recording of flows of absence of flow in the PV, HA and HVs in patients with liver transplants.
Summary
Doppler ultrasonography is a commonly used, wellestablished and non-invasive method to assess abdominal organs and the portal system. It is the first choice among imaging modalities for evaluating the signs of portal hypertension and progression of anatomic changes in the course of this disease. It does not enable assessment of pressures in the portal system or their gradient and thus the grade of portal hypertension cannot be determined. Moreover, ultrasonography provides no information about the progression of liver fibrosis (this can be done with elastography) that corresponds to clinical advancement of cirrhosis and grade of portal hypertension and also with the risk of bleeding from esophageal varices. Endoscopy plays a major role in such cases.
In the hands of an experienced physician, Doppler ultrasonography is a modality of high clinical efficacy, which is abundantly discussed in the literature from the 1980s and 1990s as well as in contemporary reports. It is an effective tool in detecting the signs of cirrhosis and neoplastic lesions of the liver as well as in the evaluation of the spleen and ascites. It is valuable in morphological presentation of the portal vessels and hepatic veins as well as in evaluating their patency, direction of blood flow and basic hemodynamic parameters. It is used for initial diagnosis, monitoring of diagnosed pathologies, qualifying patients for TIPS and OLTx and monitoring of clinical effects of the treatment.
Conflict of interest
The authors do not report any financial or personal links with other persons or organizations, which might negatively affect the content of this publication and claim authorship rights to this publication.
References
- 1.Hennerici M, Neuerburg-Heusler D. Vascular Diagnosis with Ultrasound. Stuttgart & New York: Thieme; 1998. [Google Scholar]
- 2.Berzigotti A, Piscaglia F, the EFSUMB Education and Professional Standard Committee . EFSUMB guidelines. 2013. Clinical recommendation for the performance and reporting of ultrasound examination for portal hypertension. [DOI] [PubMed] [Google Scholar]
- 3.Middleton WD, Robinson KA. Ultrasound assessment of hepatic vasculature. In: Pellerito JS, Polak JF, editors. Introduction to vascular ultrasonography. Philadelphia: Elsevier Saunders; 2012. pp. 495–515. [Google Scholar]
- 4.Krawczyk M, editor. Warszawa: Wydawnictwo Lekarskie PZWL; 2013. Chirurgia dróg żółciowych i wątroby. [Google Scholar]
- 5.Ahuja AT, editor. Diagnostic imaging ultrasound. Amirsys; 2007. [Google Scholar]
- 6.Harvey CJ, Lim AKP. Vascular disorders of the liver. In: Allan PL, Baxter GM, Weston MJ, editors. Clinical Ultrasound. Vol. 1. Elsevier; 2011. [Google Scholar]
- 7.Małek G, Elwetowski M. Wątroba, układ wrotny. In: Małek G, editor. Ultrasonografia Dopplerowska. Zastosowanie kliniczne. Warszawa: Medipage; 2003. [Google Scholar]
- 8.Patriquin H, Lafortune M, Burns PN, Dauzat M. Duplex Doppler examination in portal hypertension: technique and anatomy. AJR Am J Roentgenol. 1987;149:71–76. doi: 10.2214/ajr.149.1.71. [DOI] [PubMed] [Google Scholar]
- 9.Meng Q, Lv L, Yang B, Fu N, Lu G. Fluctuating portal velocity tracing with rhytmicity: ultrasonic differential diagnosis and clinical significance. Radiol Oncol. 2012;46:198–206. doi: 10.2478/v10019-012-0028-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bintintan A, Chira RI, Mircea PA. Non-invasive ultrasound-based diagnosis and staging of esophageal varices in liver cirrhosis. A systematic review of the literature published in the third millenium. Med Ultrason. 2013;15:116–124. doi: 10.11152/mu.2013.2066.152.ab1ric2. [DOI] [PubMed] [Google Scholar]















