Image Gently/Image Wisely and the Analysis of the Pediatric Lumenal GI Tract - HD

Introduction and Lecture Topic

I am Harris Cohen.

I'm chairman of radiology at University of Tennessee Health Science Center in Memphis, and I am radiologist and chief at Le Bonheur Children's Hospital.

I'm going to be giving a lecture on Image Gently Image Wisely concepts with regard to the evaluation of the pediatric luminal gastrointestinal tract, essentially ultrasound and how it reduces radiation exposure and analysis of luminal pediatric gastrointestinal tract.

Radiation Concerns and Theoretical Models

Radiation concerns have increased over the last decade with greater use of computed tomography in diagnosis.

There are radiation effect risks that are well known, but there are two theoretical models.

One is the deterministic model, which says that there is no risk at exposures less than a given threshold, but there still could be problems.

The stochastic theory of radiation effects is that radiation effect is linear, that there's a linear dose response relationship with no safe exposure.

On the basis of articles written, there are some people that believe that each CT provides an increased risk of excess cancer in the one to 1000 to one in 2000 range.

Pierce and Preston in 2000, gotten from Slovis notes on radiation safety says there is no need to even consider a threshold when we have direct evidence of carcinogenesis at doses of less than 10 millisieverts.

On the basis of their theory at least, we should worry about any radiation given to any patient.

The question is whether this is true or not, and the consideration is always more important for the more vulnerable groups, which include fetuses, children and young adult females of childbearing age.

Effective Doses of Various Exams

As an example, this is a table from Ricke Bone's Pediatric Imaging Essentials that shows the effective dose of various exams performed on a theoretical five-year-old.

And you can see that the millisievert effective dose for an ankle exam of three views is 0.0015 millisieverts, which they say is equivalent to 0.07 chest x-rays, that a chest x-ray two view, which is 0.02 milli, is obviously the equivalent of a single chest x-ray.

If we then look at this table, one notes that an abdominal CT has an effective dose of five to 15 millisieverts equivalent to 250 to 750 chest x-rays, a head CT, the equivalent of a hundred to 200 chest x-rays and a chest CT, the equivalent of 150 to 400 chest x-rays.

On the basis of this, again, CT imaging which has increased markedly with improvement in machinery, must be considered on a case by case basis to avoid the significant exposures that children may have, consensus statements on radiation risks suggest that it is reasonable to act on the assumption that low level radiation may have a small risk, at least a small risk of causing cancer.

Consensus Statements and Campaigns

The medical community, according to Brody et al in an article in Pediatrics in 2007, should seek ways to decrease radiation exposure by using radiation doses as low as reasonably achievable the ALARA principle.

And by performing these studies only when necessary, Image Gently Image Wisely.

Choosing wisely sound judgment.

There were multiple names for various campaigns, probably first begun by Image Gently to increase and underline warnings and suggest solutions to control the problem of radiation costs of clinical workups.

For CTs, one can lower mAs one can do iterative reconstructions, one can tailor exams and most importantly, one can limit unneeded exams.

One solution when possible is to image soundly or in other words, use ultrasound and use ultrasound first ultrasound where it can triage it should, where it can make final diagnosis, it should.

Objective of the Talk

So the objective of this talk is to note ultrasound's role in radiation safe imaging workups for symptoms requiring the evaluation of the pediatric luminal gastrointestinal tract.

And I picked two scenarios, the vomiting infant and the child and adolescent with right lower quadrant and pelvic pain.

Case Example: Midgut Volvulus in 1982

In 1982, I came across this case in which we were ordered to examine a neonate for abdominal distension hepatomegaly and a dropping hematocrit with the concern of the clinician for possible liver injury due to a forceps delivery.

These were our images, the liver looked normal and we saw this echogenic material within dilated loops of bowel.

Here's another image, and the question is, what do you see?

So in this image, there is an arrow now pointing to a arrowhead configuration of a blind ending area of bowel that suggests that the bowel twisted in what is a midgut volvulus in an attempt to see if we could reduce radiation dosages for evaluating some abnormalities of the neonatal upper GI tract, we reviewed the use of fluid as a contrast agent, and you saw in that image of the mid gut volvulus a little bit of fluid that was in the GI tract that allowed us to visualize the end of the twisted loop of second portion of duodenum.

Fluid as Contrast Agent in Ultrasound

So we having seen that case, reviewed multiple patients in which we used fluid water as a contrast, agent images were similar to those obtained by CT With the only difference being that water was echolucent and echolucent water is a friend of ultrasound.

Sound travels well in water.

You can see on this image that the third and fourth portion of the duodenum has contrast within it on a CT exam, and this is similar to that in which water one can see.

Some bubbles of echogenicity are seen within the stomach, but also seen within bowel.

This is the third and fourth portion of the duodenum seen directly behind the stomach.

This would suggest normal rotation and therefore an unlikely possibility of midgut volvulus or malrotation.

Ultrasound for Other Causes of Obstruction

Ultrasound can be used for other causes of obstruction in children.

We've shown I've shown you a case of mid gut volvulus, and I'll repeat those facts again In a newborn with bilious vomiting, green vomiting, it's usually due to sepsis or obstruction.

It's a radiologic emergency since one of the possibilities is midgut volvulus, which may result in distal bowel ischemia from a twist around the blood vessels feeding the gut.

This will create a great difficulty for this child's morbidity and might also increase mortality.

The analysis for mid gut volvulus, which is usually performed with upper GI contrast material, is performed emergently, despite the fact that only 20% of bilious vomiters have midgut volvulus, 69% of them have a benign and idiopathic cause, and 10% of them actually may have a lower colon cause for bilious vomiting.

So this is just another example of some fluid seen within the second portion of the duodenum with a twisted end, which looks similar to the contrast exam that our surgeons in the early 1980s made me do to confirm that there was a midgut volvulus in this case.

When sees dilated bowel, when sees opacification of contrast peripherally, and the bowel goes to a point on this lateral film where it doesn't cross over, but goes straight down inferiorly indicating malrotation a case of mid gut volvulus discussed by ultrasound or a case of mid gut volvulus proven by ultrasound, confirmed by requested upper GI series.

Upper GI series still is the favored imaging workup, but there are times when you can help by looking with ultrasound that includes mid gut volvulus includes duodenal atresia and other abnormalities discussed in the literature.

In a world in which one seeks more ultrasound work and less fluoroscopy and CT work, triaging with ultrasound would help.

SMA/SMV Relationship as a Clue

One of the things that ultrasound helps, even without water added to the GI tract, is in noting the relationship of the superior mesenteric artery to the superior mesenteric vein.

It is a possible clue to malrotation.

The SMV traditionally is on the patient's right and the SMA traditionally on the right, the SMV is traditionally anterior to the inferior vena cava and the superior mesenteric artery is traditionally anterior to the aorta.

In Weinberger's 1992, study of greater than 200 children with non-bilious vomiting, five of five children with an SMV to the left of the SMA had bowel mal rotation and one in four children in which the SMV was anterior to the SMA had bowel mal rotation.

If one sees normal SMA SMV relationship, one can feel somewhat less concerned, but one can have patients with bowel mal rotation who appear to have normal SMV SMA relationships.

Personal Campaign: Fluid-Aided Ultrasound of the Pediatric Upper GI Tract

My personal campaign to do fluid aided ultrasound of the pediatric upper GI tract began in 1981 where I worked in a hospital that charged by the day, no matter what was done, we did 40 adult and at least two to three pediatric fluoroscopies a day.

And many of the pediatric fluoroscopic exams were merely done for analysis of vomiting versus reflux decisions on how long to fluoro the patients varied.

And in truth, one can get almost everyone under three months to reflux pause and pulse.

A current Image Gently recommendation for fluoroscopy was partially not possible at the time because there was no such thing as pulse fluoroscopy.

Whereas pulse fluoroscopy helps decrease radiation exposure.

Ultrasound would cause much less exposure, it would cause no exposure.

Ultrasound for Gastroesophageal Reflux

So ultrasound has been used in the workup for gastroesophageal reflux according to Ricko Bonin, according to Cohen, me, ultrasound shows more reflux episodes in the upper GI series.

RIC Bone in 92 noted a hundred percent sensitivity and 87.5% specificity.

While we noted 48 true positives, six true negatives and one perhaps false negative.

These are longitudinal and transverse images of the stomach with fluid in it that was placed.

The NG tube, the NG tube was removed because an NG tube crossing the GE junction would create reflux so the NG tube was removed, so there would be no creation of false reflux and the fluid-filled stomach was then studied.

In this longitudinal view, one can see Bubbles coming above the diaphragm.

This is superior and this is inferior.

And in the transverse view, one can see a stomach and through the esophageal hiatus the distal esophagus filled with fluid above the diaphragm frame.

Ultrasound does not show the entire esophagus as would a fluoroscopy through the chest, but McCauley in assessing the upper GI series on fluoroscopy said that gastroesophageal grading could be determined in part on the basis of the amount of dilation of the distal esophagus.

The greater the distal esophageal dilation, the greater the reflux.

And this is an example of fluid in someone's stomach that is now seen in the esophagus in a patient who had esophageal reflux.

This is a video in which an individual, you don't see the stomach well here, but I'm just showing the dilated esophagus in an individual we diagnosed as having reflux.

Projectile Vomiting and Hypertrophic Pyloric Stenosis

One of the key areas in vomiting that is definitively gone to ultrasound for diagnosis is the concept of projectile vomiting.

And we discuss a differential diagnosis of projectile vomiting in six weeks as being possible.

Viral gastroenteritis, gastroesophageal reflux, hypertrophic pyloric stenosis or pyloric spasm.

These can the evaluated via ultrasound, although viral gastroenteritis would be a diagnosis of exclusion.

Hypertrophic pyloric stenosis is the idiopathic thickening of circular muscle.

It occurs more commonly in males than females.

It's more common in firstborns in those with a positive family history, particularly if the positive family history is on the maternal side.

Classically, a normal infant Develops projectile vomiting sometime around four to six weeks of age, and that is a typical patient with hypertrophic pyloric stenosis.

One can develop it at birth, one can develop it in a couple of weeks, but typically it first arises at four to six weeks of age.

Findings of Hypertrophic Pyloric Stenosis on Plain Film and GI

These are plain film and GI findings of hypertrophic pyloric stenosis.

One sees the caterpillar sign in which peristalsing stomach at an obstructed point distal to It allows one to see the stomach as if it's a caterpillar crossing from the north east to the southwest.

A mass impression from a thickened pyloric muscle can be seen on the gastric antrum in the shoulder sign, and the obstruction prevents contrast from entering the pylorus.

Here, it's just seen in the proximal pylorus and sometimes contrast is seen throughout an elongated pylorus as in this case, and that would be known as the string sign.

There's also a double track sign on fluoroscopy in which pressure on the pyloric channel from 360 degrees surrounding it can make it look like there are two channels of contrast rather than one channel.

Ultrasound Findings of Hypertrophic Pyloric Stenosis

Ultrasound, however, can show many of these signs and additional signs Ultrasound in the seven week firstborn male who had projectile vomiting for four weeks.

So this patient had this problem for three weeks and was not yet diagnosed.

One sees this Bagel or donut in which there's thickened pyloric or pyloric mucosa or mucus and pyloric mucosa, and there's prominent thick muscular wall to the pylorus, which in a normal individual you wouldn't see at all three millimeters is considered positive, and this is beyond three millimeters in thickness and one can see the thickness of the various areas I measured.

This is just a drawing showing how the thickened muscle of the pylorus allows it to appear elongated and causes relative obstruction.

And this is an older image of a hypertrophic pyloric stenosis highlighting again on a longitudinal plain view through the pylorus, the thickening of the muscle one can see fluid with some air bubbles in it in the stomach itself, which allows us to see where the beginning of the pylorus is as time has gone on.

And I think a pyloric length of 18 millimeters or greater is definitively positive.

You'll see some lower numbers in the literature.

This is just an example.

When we did this original film, the pediatric surgeon didn't believe us, so I had to use a leaded glove and push the contrast into the Pyloric channel, which was extremely elongated.

We don't do this anymore.

This was something that we were forced to do in the early 1980s as we brought ultrasound of the Pylorus to New York, which hadn't been using it until then.

This is an example of a modern high frequency linear array transducer showing elongated pyloric channel and thickened pyloric muscle.

I often see these vertical echogenicity on the imaging using the better transducers.

If one had trouble knowing where the pylorus was, one would simply find the gallbladder, and the gallbladder usually sits right near the duodenal bulb.

So just like there were findings of hypertrophic pyloric stenosis on fluoroscopy, we see the equivalent using ultrasound.

There's a double track sign in which echogenic parallel lines filled with water in a fluid aided exam denotes the mucosal complex of the pylorus.

There is the shoulder sign in which hypertrophic muscle has an impression on the gastric wall.

There's the antral nipple sign or what we've called mucosal heaping sign in which some of the mucosa of the pylorus protrudes approximately into the antrum, perhaps on the basis of redundant pyloric mucosa or long term pressure on the pylorus going forward.

That allows in a way a herniation of mucosa posteriorly.

I'll show you an example of that.

And then there's the two bright liver and stomach wall sign, a relatively newly determined finding in hypertrophic pyloric stenosis.

So this is an example of the double track sign.

The ultrasound beam crosses a pyloric channel that no longer is a simple circle but has impression on it from the thickened muscle.

And this allows the single channel to look like it has several channels, and this is an ultrasound example of two channels seen within the pyloric mucosa.

These are examples of mucosal heaping, the redundant pyloric mucosa described by Hernan Schulman as the antral nipple sign.

And the arrows point to the small amount of mucosa now seen approximately within the fluid aided stomach, the fluid, the stomach that was filled by fluid, because of fluid aided ultrasound, one can see mass impression from the thickened pyloric muscle.

And again, we see the little bit of redundant or retro position pyloric mucosa.

This is an example of the overly bright liver and stomach wall in which we're looking at the liver and we see very, very bright echogenicity.

If one sees peripheral bright echogenicity like this.

One of the concerns is that we have portal venous air in a view of the proximal portal vein when actually sees echogenic air within it, which on real time we were able to see move.

And on this motion you can see these little bright spots that the arrows are pointing at, which show motion at the same time.

The arrow points the bright echogenicity in the stomach wall.

And this individual had both the bright echoes in the liver and the bright echogenicity of the stomach wall itself, we knew had hypertrophic pyloric stenosis, which we proved.

Simulators of Hypertrophic Pyloric Stenosis

All that we see is not necessarily hypertrophic pyloric stenosis.

So this is a six week old who was vomiting with the clinician thinking there might be pyloric stenosis, but what we saw was a dilated pyloric channel with fluid in it, a duodenal bulb and it dilated as well.

And this was because there was an obstruction just beyond the bulb.

The dilated proximal channel was an indicator of stenosis in the proximal duodenum, which was proven in surgery.

An upper GI series which was performed to confirm the finding for a surgeon who was a doubting Thomas, shows this relative narrowing beyond the duodenal bulb.

This was a vomiting newborn.

On initial viewing of this, someone thought that the stomach was seen normal position, but in reality this was colon.

The stomach air was seen on the side of the abdomen and ultrasound placed into the stomach showed midline where the vertebral body was to have half the stomach on one side anterior to the diaphragm, and half the stomach on the other side posterior to the echogenic diaphragm.

And this was an unusual case in which the person just had a large hiatal hernia naturally with a non normally rotated and improperly rotated bowel.

This was a hiatal hernia With abnormal bowel rotation.

A caveat in diagnosing hypertrophic pyloric stenosis is that there's an entity called pyloric spasm, which we've written about that can create at least for a portion of the study findings that simulate hypertrophic pyloric stenosis.

This includes the elongated length of the pylorus, the thickened pyloric muscle, the double track sign.

The main difference between hypertrophic pyloric stenosis findings are that they are unchangeable while the Pyloric spasm findings something that is not treated surgically.

Those findings change during the study.

So if during a study I have abnormal measurements and then they become normal or relatively normal, I know I'm dealing with Pyloric spasm occasionally.

One of the considerations in the very young is, is it simulating pyloric spasm that will develop into HPS In a large number of patients that we've prevented from going to the OR with a diagnosis of pyloric spasm, we've never had a patient develop hypertrophic pyloric stenosis, but I normally give a caveat to my clinicians that if the child's very young, we have to worry about that and they should follow the patient clinically closely.

So this is an example of an individual who has an elongated thickened pylorus on one view several seconds later when sees fluid in the proximal pyloric area, and then several seconds later when sees fluid in a completely open area.

We looked at pyloric length measurements obtained in patients with pyloric spasm and we noted they could simulate hypertrophic pyloric stenosis.

So at one time people thought that pyloric spasm measurements would be far less than hypertrophic pyloric stenosis measurements.

And again, I use 18 millimeters as abnormal.

There are some people that use 14 millimeters is abnormal.

You can see that everywhere is a pyloric spasm case and we can see some of them with very, very elongated pylorus and some with borderline elongated pylorus and many 14 millimeters is abnormal, having lengths that would be considered hypertrophic pyloric stenosis.

This slide is somewhat problematic in that this arrow refers to this point in analyzing ultrasound for pyloric spasm.

Watch how the Pylorus handles a fluid load note if the wall thickness or length measurements that appear abnormal or ever normal.

If the antrum pylorus region is changeable, it's pyloric spasm.

Follow the patient clinically.

However, patience is a virtue in these ultrasound exams.

Diagnosis of Appendicitis

The second part, diagnosis of appendicitis, and this is because of time limitations gonna be very short and sweet At times.

The diagnosis of appendicitis is clinically confusing.

One can look at Copes the acute abdomen and see the many confusing things that can simulate appendicitis including right lower lobe pneumonia, left lower lobe pneumonia.

Classically, the patient has rebound point pain that occurs after original periumbilical pain.

The patients however, may not be febrile.

The patients may have different pain complaints related to the different positions of the appendix, including right upper quadrant, pelvis or retrocecal position.

Some personal history on the diagnosis of appendicitis by ultrasound from my point of view until 1986, we were only able to see collections in the pelvis from perforated appendicitis or use ultrasound to diagnose a gynecologic simulator of appendicitis.

We didn't have the ability to see the appendix well and until TS work in the mid to late 1980s out of the Netherlands.

One didn't know to use high frequency linear array transducers and attempt to compress the appendix.

An appendix that doesn't compress that has a blind ending structure greater than six millimeters was an indicator of a non perforated appendicitis.

Since that time, we've also used color flow imaging to note the inflamed wall and have seen bright echogenicity in the periappendiceal soft tissues.

The equivalent of the increased markings seen in the periappendiceal fat on CT.

His work changed our ability to read appendicitis and with that change, the ability to decrease radiation exposure in patients.

Ultrasound helps diagnose simulators of appendicitis, whether GU related, whether GYN related or whether GI related.

Technically I prefer doing ultrasound on children with focal pain because I think that is where ultrasound's most helpful.

When a child can point out the area of focal pain, I can take the transducer as if I am examining spokes on a wheel going toward the center where there's the most focal pain and able to therefore determine what is the cause of the area of focal pain, whether it's lymph nodes there, that's can sometimes cause it or whether I see a tubular non-compressible blind ending structure that is a non-perforated appendicitis, right lower quadrant pelvic pain in the child or teenager can have non gynecologic considerations.

They include appendicitis, Crohn's disease, mesenteric adenitis, ureteral stones or bladder infections.

All these entities may be diagnosed by ultrasound.

Pelvic pain in the pediatric patient and right lower quadrant pain have gynecologic considerations as well and that includes hemorrhagic cyst, uterine obstruction, ovarian torsion, teratomas, and or other ovarian masses, teratomas of the most common pelvic inflammatory disease or ectopic pregnancy.

We do not have the time to talk about those.

I would reference one of my sonar world talks on pelvic pain to get images that show that this is an example of an appendicitis exam.

Ultrasound Examples of Appendicitis

This is an example of an appendicitis case.

16-year-old with right pelvic pain in 21,000 white count Who we asked for a pelvic ultrasound prior to doing an empty bladder, right lower quadrant compression ultrasound.

The individual had a normal pelvic ultrasound, but the right lower quadrant ultrasound is what you see here.

It's using a linear transducer of high frequency in the right lower quadrant.

The patient was somewhat heavy.

You can see the significant soft tissues and what we saw was a greater than six millimeter tubular non-compressible blind ending structure with somewhat thickened walls.

There was a contained calcification within it and the diagnosis was appendicitis with the echogenic area, which sometimes would show better shadowing than that, containing an appendicolith.

This is an example of that same patient with color doppler showing increased flow, at least at the tip of the appendix.

And remember that periappendiceal echogenicity may be due to inflamed fat.

This is an example of another example of non-perforated appendicitis, a different child, a more classic tubular blind ending.

Non-compressible structure greater than six millimeter, with color flow within its wall.

Some people are now using seven millimeters as the width of abnormality to prevent false positive exams.

I think that's more of a situation of What has occurred with CT exams and which originally CT exams were calling appendicitis at 10 millimeters.

That drop down is six millimeters and I think between six and seven millimeters there's somewhat of a gray zone.

This is an example of a normal appendix.

Some people find normal appendices relatively easily, other people not, but you can either go from the cecum out or if you're fortunate enough, see this not wide compressible tubular structure that is blind ending, that is the normal appendix.

GI and Non-GI Simulators

Again, there are other GI and non GI simulators that we're not gonna go through them other than the show you this one image of a 15-year-old with right pelvic pain who did not have a gynecologic abnormality, who did not have appendicitis, but had this thick walled elongated bowel loop with central echogenic mucosa highlighting how thick the bowel wall is.

And this patient had Crohn's disease and color flow showed some increased flow to the very thick muscle in this patient with Crohn's disease.

Recent Articles and Impact

With all this work done, pursuing ultrasound diagnosis within pelvis and right lower quadrant, recent articles have touted how this has helped.

And a key article in a AJR 2013 out of Children's Hospital of Montefiore by Lee Et al States that collaboration between radiologists surgery and emergency department physicians increased the use of ultrasound as the first imaging option in their institution from 33% at the beginning of the study to 90% at the end of the study.

That the use of a CT scan is the first and only diagnostic test decreased from 43% to less than 10% by the end of the study, suggesting that one can make final diagnosis via ultrasound.

In many cases, one can determine how aggressively one will look with another modality and that one can save CT for second line modality if necessary.

Summary

In summary, adhering to the ALARA principle is necessary in pediatric workups.

Ultrasound is a key tool in keeping workup radiation doses down.

Examples have been shown of its aid in assessing pediatric luminal gastrointestinal tract abnormalities.

It helps elsewhere use the transducer.

It works if you work it and don't give up no matter what the end.

And again, because of time constraints, we have not looked at ultrasound and excellent imaging of gynecologic simulators of appendicitis.