see also: Salivary ultrasound standardized diagnostic approach and report; Salivary stone imaging correlates; Ultrasound CT Sialendoscopy and Gross Appearance of Submandibular Stone

Salivary Swelling; Sialolithiasis; Sialograms and Sialography; Ultrasound Guided FNA Fine Needle Aspiration Biopsy; Parotid Duct Stricture Dilation with Salivary Balloon and Ultrasound Guidance

Botox injection to salivary glands for hypersalivation; Submandibular sialogram ductal stricture; Parotid sialogram, ultrasound, and CT for Sjogrens syndrome; Ultrasound appearance of facial venolymphatic malformation with phlebolith

Lecture Ultrasound Assisted Surgical Techniques (Hoffman) Mt Sinai May 5 2018

Thyroid Ultrasound

I. Background

A. Terminology

Echogenicity of tissue: capacity to reflect or transmit US waves in the context of surrounding tissue. Where there is an interface of structures with different echogenicities, a visible difference in contrast appears.

Hyperechoic: white  //  Hypoechoic: grey  //  Anechoic: (black)

Examples: Bone has a bright hyperechoic rim and black or anechoic center because ultrasound casts and acoustic shadow beyond it.

Lymph nodes are hypoechoic or anechoic; muscles are hypoechoic with striate structure; fat is usually anechoic

Echogenecity = even (fine echotexture is normal for parotid/smg) or coarse (uneven echotexture seen after I131 therapy in ~ 50% of parotid glands)

Heterogeneity of Echo Signal = Heterogeneity defined for Sjogrens syndrome by: "hypoechoic areas, lines or spots or hyoechoic areas surrounded by hyperechoic lines and/or spots resembling a reticular or honeycomb image" (Saied 2013). Heterogeneity refers to the presence of dissimilar components as determined by difference in ultrasound signals - discriminated from homogeneity or homogeneous- meaning structures with similar ultrasound characteristics

color Doppler: flow toward the probe = red; flow away from the probe = blue (mnemonic: BART = Blue Away, Red Toward)

Scanning planes:

Transverse = Perpendicular to the long axis of the patient (termed axial for CT imaging)

Longitudinal = Parallel to the long axis of the patient (termed sagittal for CT imaging)

Other: Oblique, Coronal

Angle of incidence: the angle at which US waves encounter the surface of a structure - best images with angle of incidence = 90 degrees (perpendicular)

Anisotropy: a tissue property responsible for US reflection even with mild changes in angel of incidence. 'now you see it, now you don't' phenomenon

different tissues have differing degrees of anisotropy

requires adjustments to optimize visualization via US probe manuevers of pressure, tilt, and rotation to optimize the angle of incidence to get the best reflection

High frequency probes (10-15 MHz) are best for US imaging of superficial structures (2-4 cm)

Mid frequency probes (5-10 MHz) preferred for slightly deeper structures (5-6 cm)

Low frequency probes (2-5 MHz) preferred for deeper still (i.e. 10 cm)

Small footprint probe: narrow probe -  advantage for in-plane needle placement (allows operator to place needle entry closer to target) - disadvantage relative to 

Larger footprint probe provides a wider picture and better lateral resolution

Gain: most US machines have an 'auto-gain knob - some have gain modes designed for structures ('nerve'/'angio'/'general') but usually not needed if appropriate probe and depth selected

Depth: reasonable to set a greater depth reading initially to get the 'big picture' and the decrease the depth when the targeted structure is found - usually 1 cm deeper that the target to be injected

Probe Manipulation: mnemonic PART (Pressure, Alignment, Rotation and Tilt)

Pressure: affects echogenicity of the tissue (also shortens distance to structure of interest)

Alignment: 'sliding' - find the structure of interest and position it optimally on the screen for needle advancement

in the middle of screen for 'out-of-plane' approach

out-of-plane with long axis of needle directed across the scanning plane (out-of-plane needling commoner catheter placement)

out-of-plane needle advancement often warrants dynamic tilting or sliding of the transducer when advancing the needle to help track the tip of the needle - visualizing the tip of the needle is challenging but essential

'hydrolocation' - small volume of saline or destrose can be injected to serve as needle tip locator

somewhat on the opposite side of the screen for 'in-plane' approach

in-plane approach: needle can be seen on the US monitor in the long-axis view (in-plane needling common for single injections)

larger needles designed for enhanced echogenicity enhances visualization - also if more favorable angle of incidence is employed

Rotation: achieve several goals:

1. maintain true axial view of a structure as you are sliding to keep its long axis parallel to the probe

2. can change view back and forth from long axis to short axis

Tilt: confusion can arise from the fact that when the probe is tilted in one direction, the US plane is sweeping to the opposite direction

Acoustic enhancement: artifact of increased echogenecity in region distal to a fluid collection

Acoustic shadow: anechoic area behind bony structures

Attenuation: Progressive reduction of amplitude and intensity of a transmitted signal as it moves through a medium

Higher frequency waves attenuate closer to the source address by adjusting Time-gain compensation for attenuation (Boosts signal to compensate)

Time-gain compensation for attenuation: signal gain is increased as time passes from the emitted wave pulse. This correction makes equally echogenic tissues look the same even if they are located in different depths. If the ultrasound image was formed directly by the raw returned echoes, image would appear lighter in superficial layers and darker in deep layers but TGC is used to overcome this artifact.

Artifacts: do not correspond to real structure

Artifact lines: Comet-tail artifacts: vertical hyperchoic lines (reverberation occurring in a small space) streak of white running away (deep) often seen in colloidal cysts. Not calcium generally considered benign different from calcification which won't have this artifact

acoustic shadow

Tangential artifact. Echo-free zone

Acoustic enhancement (enhancement artifact) area behind cyst brighter because waves pass thru cyst readily)

reverberation artifact: repetitive cycle losing energy each time i.e. trachea

Mirror artifact - internal reflection

Doppler: technique for imagine moving particles by the detection of a change in the frequency of the reflected ultrasound energy

Modes: A-mode not used = amplitude mode = rejected energy is shown as peaks of different size

B-mode is used = brightness mode = reflected energy is displayed as areas of different brightness

M-mode = reflected energy is shown as areas of brightness traced from left to right on screen with time on x-axis (adjust to B-mode)

Axial resolution; distinguish structures in line with one another (depth)

Lateral resolution: distinguish adjacent structures

Contrast agents: (see Oh 2019) novel device to create 'saline-air mixture' as contast medium for salivary ultrasound

Terminology (adapted from multiple references)	Comments
Echogenicity (brightness)
Normal echogenicity (equivalent to thyroid; greater than adjacent muscle)	Echogenicity can be compared to either normal thyroid gland (identical) or adjacent muscles (should be more echoic than adjacent muscles)
Abnormal echogenicity	hypoechogenicity is characterized by numerous hypo/anechoic areas
	hyperechogeneicity can result from fatty infiltration/fibrosis with hyperechoic bands
	in older patients involution and fibrosis may result in atrophy and hyperechogenicity
	Echogenicity also influenced by amount of intraglandular fatty tissue (increased echogenicity with fat infiltration)
Homogeneity (consistency)
Normal homogeneity similar to thyroid gland
Abnormal homogeneity or abnormal echostructure (heterogeneity)	Sjogrens characterized by heterogeneity or abnormal homogeneity defined as numerous hypo/anechoic areas and numerous hyperechoic bands
	Heterogeneity may not be obvious in patients with few or irregularly distrubuted cyst-like or hypo/anechoic areas
Hypo/anechoic areas
Defined as areas generating few or no echoes	Full definition: small areas generating few or no echoes located anywhere within the gland, not compressible by the probe
Fibrous Stranding
Hyperechoic bands identified	As oer Quéré et al  2024 correspond to fascia between lobules in parotid - clinical significance uncertain but numerous hyperechoic bands may indicate presence of fat deposits or fibrous tissue

B. Physics

1. Ultrasound = non audible sound energy

a. Audible sound energy = 2-20kHz

b. Ultrasound for medical application = 1,000x greater than audible energy (2 to 20MHz)

2. Reflection of sound waves creates image based on changes of the sound waves as they traverse tissue including: attenuation, deflection, scattering, refraction, and reflection.

a. Reflection depends on tissue impedance = measure of resistance to propagation of sound waves from one medium to another

b. At the boundary between tissue structures with difference in impedance (impedance mismatch) an ultrasound pulse is partially reflected

c. The impedance difference between air and tissue is greatest - and need be removed by using a coupling gel to remove air between the transducer and the skin

d. Signal processing of the reflected acoustic signal is translated into a gray-scale (B-mode) image on the the screen

3. The ultrasound pulse is generated by piezoelectric crystals in the transducer probe

a. An alternative current applied to the crystals causes them to vibrate to convert electrical energy into mechanical energy to emit a pulse

b. After the pulse is generated, the system awaits the return of the reflected signals - picked up by the same piezoelectric crystals in the transducer.

c. This returning energy provides information based on analysis of the amplitude and timing of the signal that is translated into an image.

C. Clinical Use

U/S guided needle placement hints:

initially take the operator's attention off the screen and focus on attaining correct needle alignment with the US probe

only after some advancement of the needle through the skin has occurred should the probe operator shift attention back to screen

then, if the needle is visualized on the screen, further attention to the probe position and needle can be done based on feedback from the US screen

if the tip of the needle is lost from view on the screen, then look back at the bore and find the best possible way to realign the needle with the US plane

remember the US beam is thin (1 mm = the width of a credit card) so subtle movements will bring the needle in and out of view.    

to prepare the transducer to avoid exposure to body fluids and alcholol (used as asepsis for the skin) it is reasonable to use cling wrap as a protective barrier for the probe

a generous amount of US gel should be placed on the end of the transducer before placement of the protective wrap (Smith 2010)

Size Measurements

Parotid gland has a more complex anatomy that makes measurement of size difficult

1. As per Carona et al (2015) - AP = anterior-posterior diameter = the maximum distance in the transverse axis // ML = medio-lateral dimension maximum width/thickness of the gland in the coronal axis.

a. AP and ML diameters of the parotid gland were measured on the same transverse view obtained at the level of the angle of the mandible

a. patient in sitting position with neck hyperextended and head slightly turned ot the side opposite the gland being examined

Submandibular gland is more amenable to analyzing size in three dimensions (depth, width, length) by transverse and longitudinal placement of the probe

1. As per Carona et al (2015) 
   1. AP dimension measured on longitudinal view parallel to the horizontal ramus of the  mandible
   2. ML measured on a perpendicular view obtained at the half point of the AP dimension.
2. As per Marteau et al (2019) "gland size can be appreciated by surface-area measuring - with the normal paortid surface ~3-4 cm² and the normal submandibular gland surface area ~ 1-2 cm (referring to 240
   1. They identify that although semi-quantitative scores have been developed for salivary US changes with Sjogrens - 'gland size is seldom considered'
   2. Through a crude analysis of surface measurements comparing US with palpation they proposed 'hypertrophy' characterize surface US measures of > 5 cm² for the parotid and >3 cm² for the submandibular glands
   3. They identified that volumetric evaluation of the parotid glands by US is limited by morphologic features
   4. They conclude that US as a "complementary exam' to palpation 'might be helpful' to evaluate for hypertrophy 

Lymph node evaluation

Normal lymph nodes: "No single ultrasonographic feature defines a normal from an abnormal cervical lymph node (Chan 2007)

Cancer more likely with the following (Chan 2007): "combination of characteristics such as increased size, round shape, absence of an echogenic hilus, intranodal necrosis and peripheral or displaced vascularity make malignancy more likely"

Abnormal lymph nodes (use of size criteria questionable - Van Den Brekel (1998) - level I nodes: minimal axial diameter or short axis 4-6 mm; level II nodes 6-8 mm; level III-V nodes 4-7 mm;

As per Chan (2007) "best approach is not defining absolute size criteria"

Peripheral vascularity

Shape 

1. Short axis-long axis ratio (S:L) - S/L ≥0.5 abnormal
2. Nodal shape transformation from oval to round is 'most likely because of the malignant infiltration changing the architecture of the lymph node' (Chan 2007)
3. Submandibular and parotid nodes can aslo be round and normal.

Nodal Border

1. Sharp nodal border can be seen with normal lymph nodes - but unsharp borders are commonly thought to be a finding suggestive of normal lymph nodes
2. Increased sharpness of nodal border thought to be due to increased acoustic impedance difference resulting rom tumor infiltration and also from loss of intranodal fat
3. Malignant nodes with unsharp borders can represent extracapsular spread of tumor

Echogenic Hilus

1. Represents collecting sinuses of a normal lymph node
2. Hilar vascularity should be evaluated with color doppler
3. Older patiens have more intranodal fat and therefore have more prominent hilus

Echogenicity (overall)

1. Metastatic papillary thyroid nodes then to be hyperechoic (correlate with psammoma bodies)
2. Most malignant nodes, tuberculous nodes, and normal nodes tend to be hypoechoic compared to adjacent muscle

Vascular Pattern

1. Normal and reactive lymph nodes are generally either avascular or have vasculature confined to the hilum
2. Metastatic lymph nodes have characteristically more peripheral vascularity

Absence of hilus

Presence of nodal microcalcifications

Cystic changes

Lymph node classification per Lo et al (2016) "The Effect of Radiotherapy on Ultrasound-Guided Fine Needle Aspiration Biopsy and the Ultrasound Characteristics of Neck Lymph Nodes in Oral Cancer Pts After Treatment	Classifications	Classifications Features found to correlate with recurrent nodal disease following XRT (in bold)	Classifications
Echo structure	Predominantly solid	Mixed cyst type
Short-axis diameter Long-axis diameter	S/L ratio	Increased short and long axis diameter
Nodal margin	Regular	Irregular
Echogenicity (compared to adjacent muscle)	Hypoechoic	Isoechoic	Hyperechoic
Echogenic hilum	Present	Absent
Calcification	Present	Absent
Internal echo	Homogeneous	Heterogeneous
Vascular pattern (Doppler - set for high sensitivity with low wall filter (detection of vessels with low flow)	Avascular or Hilar type	Mixed / Spotted	Peripheral type

II. Guidelines for performing and reporting Salivary Gland Ultrasonography

A. AIUM Practice Guidelines for evaluation (American Institute of Ultrasound in Medicine 2014 - (note updated information published 2019 accessed 1-26-2023: https://onlinelibrary.wiley.com/doi/10.1002/jum.14830)

1. Examination done with patient sitting or supine with neck in extension
2. A systematic exam should be done according to examiners preference in a standardized and thorough fashion
3. For a right-handed examiner, the console should be located next to the patient's right shoulder
4. Systematically examine salivary glands in transverse, anteroposterior, and longitudinal planes
5. Note parenchymal echogenicity and presence of focal abnormalities comparing each gland with contralateral
6. Normal echogenecity of major salivary glands
   1. Typically homogeneous with variation from bright (hyper echoic) to 'slightly' hyperechoic compared to adjacent musculature
   2. Variation in salivary echogenicity may depend on amount of intraglandular fat (fat appears bright on sonography and suppresses transmission to deeper portions of gland)
   3. In some the gland may be so fatty as to limit diagnostic value of sonography

B. AIUM Indications for salivary gland ultrasound evaluation (2014) - "include, but not limited to the following":

1. Diffuse enlargement and tenderness
2. Suspected abscess
3. Recurrent swelling
4. Swelling with eating
5. Discrete solitary mass
6. Multiple masses
7. Anterior floor of mouth mass

C. Selected AIUM Recommendations regarding documentation (accessed 4-21-2016 http://www.aium.org/resources/guidelines/documentation.pdf)

1. There should be a permanent record of the U/S exam and its interpretation
2. Variations from normal size should be accompanied by measurements
3. Label images with patient identification, facility identification, exam date, side (R v L)
4. "Retention of the ultrasound examination should be consistent both with the clinical needs and with relevant legal and local health care facility requirements"
5. Limitations that compromise the quality of the exam should be noted (eg. high body mass index)
6. Comparison with prior relevant imaging if available
7. Final reports should be available within 24 hours or, for nonemergency cases, by the next business day

III. Ultrasound Findings

A. General Findings - the role for Ultrasound versus other imaging (CT, MRI, sialography) is still evolving as is the technology accompanying these techniques

as per Bag et al (2015) "Ultrasound is most commonly used to obtain core tissue biopsies, rather than as a means of evaluating the parotid gland's primary pathology." These investigators affirm that "Contrast-enhanced CT scan is typically used to evaluate patients with suspected inflammatory pathologies". Although many would disagree with their assigning a lesser role for the diagnostic capabilities of ultrasound, they also assert that "No consensus or evidenced-based guidelines indicating usage for these modalities to image the parotids exist currently"

B. Disease Processes

1. Radioactive Iodine Ablation (I131)

a. Of 202 patients treated with radioiodine (Kim 2015)

i. 46.5% demonstrated post-RIA changes on ultrasound (46% of parotid glands/3.5% submandibular)

ii. Findings: coarse echotexture, decreased echogenecity, lobulated margin, decreased gland size

2. Sjogren's syndrome

a. Most valuable diagnostic criterion in Sjogrens is heterogeneity of the glandular parenchyma (Saied F 2013) 

i. Heterogeneity defined by: "hypoechoic areas, lines or spots or hyoechoic areas surrounded by hyperechoic lines and/or spots resembling a reticular or honeycomb image" 

ii. DDX for salivary heterogeneity: hematoma / bacteria infection / proliferative process /viral infections/ HIV / chronic sialadenitis /sarcoidosis 

b. However, as per Cornec et al (2014): "Salivary gland US needs standardization and validaion before its use in Sjogrens Syndrome diagnosis"

c. More recent work by Jousee-Joulin et al (2017) identify reproducible abnormalities of abnormal echogenicity and homogeneity showed inter-observer reliability in assisting with the diagnosis of Sjogren's syndrome 

3. Plunging ranula and sublingual gland herniation through mylohyoid

a. Mylohyoid defect commonly seen in cases of plunging ranulas (all but one of 80 in a series by Jain et al 2014)

i. Mylohyoid defects are most commonin the anterior third of the muscle

ii. With active tongue movement the sublingula gland herniation can be seen to move through the defect in a slide-like motion

iii. Four different sonographic types to sublingual gland herniation: slide/wobble/mushroom/and and retrusion (Jain 2014)

b. A Asymptomatic herniation of thte sublingual gland through mylohyoid defects reported from 10% to 39-45% (Castelli 1969 and Nathan 1985)

c. May be thin or thick walled - generally well-defined, anechoic with no internal vascularity with ddx: (as per Rhys 2011):

i. Congenital: first branchial cleft cysts (solitary cystic structure) / lymphatic malformations (multispetated, often trans-spatial)

ii. Acquired: chronic inflammation, autoimmune disease, HIV infection (usually in presence of benign cervical lymphadenopathy - lympoepithelial cysts); Warthin's tumor

4. Sialolithiasis

a. Reported sensitivity for ultrasound detection of salivary stones ranges from 71% to 94% (Jager 2000, Gritzman 1989, and Diedrich 1987)

b. Sonographic diagnosis of calculi (Terraz 2013)

i. Presence of hyper-echoic linear, oval or round formations casting an acoustic shadow in normal or dilated salivary ducts

ii. Presence of ductal dilation caused by a hyperechoic formation without an obvious acoustic shadow

c. Terraz et al (2013) compared ultrasound with either sialography with sialendoscopy or sialography alone to establish the sensitivity/specificity/accuracy for ultrasound (7.5-12 MHz linear probes) in evaluating 82 swollen glands in 79 patients. They identified a negative predictive value of only 78% for ultrasound detection of stones.

i. The concluded that the limited sensitivity and limited negative predictive value of ultrasound does not allow reliable exclusion of small salivary gland calculi

ii. They recommend additional diagnostic investigations to detect calculi in patients with normal sonographic findings and suspected lithiasis

iii. Negative ultrasound but still suspected sialolithiasis in their hands is followed either by sialography (technique well described in article) - if the suspicion for stones is low or interventional sialendoscopy if the degree of suspicion is high

iv. In our hands (Hoffman, U of Iowa) CT imaging with contrast is a standard method of evaluation done if still warranted after in-clinic diagnostic salivary ultrasound; sialography is commonly used for diagnostic purposes and to map the parotid ductal system before interventional sialendoscopy and less commonly for the submandibular gland. When used for the submandiular gland, sialography is more commonly used for diagnostic dilemmas (see: Sialograms and Sialography)

Classification of stone location (Goncalves et al 2017)

Sialolithiasis Definition: "Hyperechoic reflexes with distal signal loss along the course of the duct" from Goncalves et al "Sonography in Diagnosis of Sialolithiasis" J Ultrasound Med 2017;36:2227-2235
	Submandibular sonographic landmarks	Parotid sonographic landmarks
Intraparenchymal stone	proximally located in parenchyma	proximally located in parenchyma
Proximal/hilar stone	1 cm proximal to 1 cm distal to the edge of the mylohyoid muscle	1 cm proximal to the posterior edge of masseter to middle of masseter muscle
Middle third	1 cm distal to the edge of the mylohyoid muscle to the sublingual gland	middle of masseter muscle to to anterior edge of masseter
Distal duct including papillary region	from the main mass of the sublingual gland to the papilla	anterior edge of masseter muscle to papilla

5. Obstructive Sialadenitis

a. The duct network of submandibular and parotid glands is not normally visible during US assessment

b. Salivary stimulation with oral ascorbic acid (vitamin C) enhances US visualization both in normal and obstructed ducts (Bozzato 2009)

c. Accuracy for use of ultrasound to identify ductal stenosis (in the absence of sialolithiasis) was shown by Larson et al (2017) had a negative predictive value of only 50% when correlated with intraoperative findings on sialendoscopy.  However, if ductal dilation (suggesting stenosis) were identified on ultrasound, then the positive predictive value for the finding of stenosis on sialendoscopy was 93%. These investigators conclude that the surgeon should feel confident that ductal stenosis will be encountered intraoperatively when duct dilation is identified on preoperative ultrasound - however, the absence of this finding on ultrasound does not preclude the identification of a treatable stenosis identified at the time of sialendosocpy

6. HIV-associated lymphoepithelial cysts of the parotid

a.  1-10% of patients with HIV infection reported to have salivary gland enlargement (Sekikawa and Hongo 2017)

b. BLEC - benign lymphoepithelial cyst - with ddx including Warthin's tumor and cystic degeneration of benign and malignant tumors

- can be an indicator of HIV infection - with cysts as reservoirs of HIV-1 p24 antigen

- characterized by bilateral parotid gland swelling and cervical lymphadenopathy; can occur as a single cyst but more typically bilateral

- treatment often dictated by cosmetic issues in that most are asymptomatic (assuming management of HIV infeciton is ongoing)

options include repeat aspiration, ART (anti-retroviral therapy), sclerotherapy, radiation, surgical removal

References

Ihnatsenka B and Boezaart AP: Ultrasound: Basic Understanding and learning the language   Int J Shoulder Surg. 2010 Jul-Sep;4(3):55-62