Post-operative Imaging - HD

Objectives

Thank you Sandra, and good afternoon.

The objectives here are to review the postoperative appearance of joints, tendons and ligaments, and talk about complications related to the surgeries.

Now looking at normal anatomy is difficult enough, but after a surgeon has been there, things really can get quite confusing.

I'll be going through each of the joints and covering some of the more common findings of post-surgical cases, beginning with the rotator cuff in the shoulder.

Shoulder: Rotator Cuff

It's important to understand what surgery has been performed as this will predict what we're gonna see at imaging.

So if you have a partial tear, if it's articular and less than 50% or bursal, usually they simply debride the tendon.

So the tendon may appear thinner after that surgery.

With a more significant articular tear, they'll usually convert it to a full thickness tear and then reattach it.

And with full thickness repairs, there are different ways of attaching the tendon to the bone.

Most of them will use trans osseous fixation into a trough and they'll decorticate the tuberosity to allow the tendon to grow into the bone.

With direct apposition, the tendon is secured to the bone with metal, plastic or bioabsorbable suture anchors and they can be in a single, multiple or double row.

So here's an example on this illustration of single and double row.

It depends on what surgeon is performing the surgery, but this tends to be the more common scenario because it reproduces the anatomy with a broader footprint compared to a single row.

So knowing this will help us determine again what to expect after surgery.

So all the primary signs that we're looking for for rotator cuff tear beyond a defect are really, they don't really apply here.

And the secondary signs definitely do not apply here.

For example, in a native cuff that's thinned or with cortical irregularity of the footprint, that can indicate a cuff tear, but you cannot use that in this scenario.

The tendon can be thick, it can be thick, it can be hypoechoic, it can be echogenic.

You'll see suture anchors and suture within the tendon.

So it really has a variable heterogeneous appearance.

So here's some examples of postoperative and intact supraspinatus tendons.

With MR Correlation, we could see that the tendons, diffusely hypoechoic, it's a little bit thin here, the cortex is irregular.

We see suture material, but this is an intact cuff.

If we look short axis, we can appreciate the suture going to the suture anchor into the bone.

With marked bone irregularity and hypoechogenicity, you can appreciate the overall heterogeneous appearance of the cuff.

So you could see that it'd be really impossible to differentiate postoperative change from tendinosis from even subtle partial tearing.

Here's a companion case.

This patient's out a little bit further in timeframe after the surgery and here we can see what looks like more homogeneous tendon.

Although it's markedly thinned, we can see the cortical irregularity.

But again, as the MR shows, the tendon is intact in short axis.

On that case, we can see the volume loss, which would be an indirect sign in the native cuff of a cuff tear, but it doesn't apply here.

We can see that there's dipping of the deltoid down into this thin tendon and there's the suture anchor material.

So how well does ultrasound perform looking at the postoperative cuff, this study shows 94% accuracy.

I wish I could get numbers that high.

We've looked at our numbers and we're a little over 90%.

So I'm thinking 90 to 95% is our target zone.

So in other words, it can actually perform well.

But it depends on how close you want to call the diagnoses.

For me, when I'm looking at a postoperative cuff, I want to convey is there a re tear or not.

If you start going into mild tendinosis or mild partial tearing, your accuracy is gonna go way down.

But if you're trying to say re-tear or not, I think you can function at greater than 90% accuracy.

So even if you see a tendon defect and you actually are correct, and there is one, these defects may actually heal up to a year after the surgery so they can go away.

Although most tendon defects will increase in size, some of them again can heal.

What's important is that the structural integrity, which is what we're looking at, does not correlate with pain or function.

So even if you do see a defect, it may be completely asymptomatic.

So how do you diagnose a recurrent tear?

Well, we're looking for a defined tendon defect filled with either hypoechoic or echogenic fluid.

When there's complete retraction, it'll be non visualization of the tendon, but again, we're focusing on a definitive defect seen in two planes.

There's an example where we see this tendon that's retracted from the greater tuberosity.

So there's an obvious defect shown here.

Here's a companion case, another tendon in long axis showing this fluid filled defect.

Also shown here on the MRI note the bioabsorbable suture anchor as shown here on the MRI as well.

And if we look in short axis, we can appreciate the suture material just surrounded by fluid, which tells me either the tendon's pulled off of the suture or the suture's pulled out of the tendon.

Either way, this is an indirect finding to say that this truly is a defect indicating a recurrent tear.

So the question exists, what does the tendon look like at certain time points after surgery?

How can we focus this exam to say, is there a retear or not?

And how's it changed over time and when is it quote normal appearing?

So a study we completed, we looked at 25 shoulders where I imaged the shoulder at day two, week two, one month, two month, three, month six, nine, and 12 months, one and two, three year that the study took forever, as you can sense.

But I learned a lot because I was able to track the defects over time to see what happened to them.

And here's one such case where I would've called this a retear.

It looks pretty much like a defect, but over time, that completely healed.

And then we could see by nine months to a year it starts to look more normal.

And that was the trend that we saw in this study.

In that most recurrent tears occurred within three months after surgery and it occurred at the bone attachment.

The tendons started to look normal, whatever that means, more fibrillar and echogenic by six to nine months.

So in the first six months after surgery, the tendon's more heterogeneous than ever and it really is risky to call subtle things in that early six to nine month period when you see a defect.

If they're equivocal, they can heal or they can get larger.

So if you're not sure, you should get a followup examination.

So that's the key.

When my surgeon says, is there a retear, the surgeon wants to know, without a doubt, do I need to take the patient back in?

And if I'm questioning a retear, I'll say I'm not sure I'll be honest and get a follow up test, be follow up ultrasound because either it will heal or it will get larger.

So even if the tendon is completely intact, they may have continued symptoms.

Patients with recurrent tears may be asymptomatic and large, although large recurrent tears tend to be more symptomatic.

Again, even though we're looking at the structural integrity of the tendon, it may not correlate with the patient.

Biceps Tendon

Still in the topic of postoperative shoulder, a few slides on the biceps tendon.

When someone has biceps longhead tendon pain and pathology, the surgeon will often perform a tenotomy or release the tendon at the labrum to treat the symptom.

And then after that the patient or the surgeon may reattach the tendon to the bicipital groove called a tenodesis, or they may not and it'll just basically let it retract distally.

So when you're looking at the bicipital groove and it looks empty, it could be a tendon tear, but realize that the surgeon may have performed a tenotomy.

But here's a case where they performed a tenodesis with the tenotomy and they tack down the tendon back in the bicipital groove.

We can see it intact as we move proximal then there's nothing more proximal.

It's important when you see this not to misinterpret this as an intraarticular body, the fact that there's no tendon proximal tells me that this is post-surgical material.

Here's an example of a failed tenodesis where the tendon is retracted with that suture anchor.

Here's the MRI, which showed that the suture anchor has been pulled away from its normal attachment, more proximally in the bicipital groove.

Shoulder Arthroplasty

And then the last bit here on the shoulder talking about shoulder arthroplasty.

Ultrasound is an ideal method to look at the rotator cuff after a shoulder replacement and that's because the artifact from the metal does not obscure the overlying tendon.

But what is important is to understand which arthroplasty is performed.

This is critical when approaching the shoulder with ultrasound.

Now if you look at these various arthroplasties, I try to separate into two general groups.

The group where the tuberosity is intact and the group where it's resected.

To me, that's the first wave of approaching this because if the greater tuberosity is resected, you really can't perform an ultrasound because the rotator cuff is torn already and the bone landmarks are missing.

So this is called a reverse total shoulder arthroplasty where the ball is on the glenoid side where this is the more conventional variety where the ball is on the glenoid on the humeral side.

So if it says to evaluate the rotator cuff after an arthroplasty, if I see this x-ray, I'm gonna think, well the rotator cuff is already torn, that's why they chose the reverse total arthroplasty.

I'm not gonna even be able to see it because it's gonna be retracted.

I wanna approach this case of simply looking for a fluid collection or a in a generic post-surgical complication.

Now if you look at the, these on the left side, again, the greater tuberosities are still intact.

So you would perform your ultrasound exam like a native shoulder exactly the same.

This is a total shoulder arthroplasty, meaning both sides of the joint are replaced.

These are hemi arthroplasties where only one side of the joint is replaced.

But the key is the greater tuberosities are the same, they're intact.

So here's an example of a conventional total shoulder arthroplasty.

Here are the bone landmarks of the greater tuberosity and the acromion.

And this is the humeral head.

So this looks exactly like a native shoulder other than you have reverberation artifact because of the metal head that's been replaced.

But we can see the cuff normally attaching on the greater tuberosity.

Also seen in short axis.

Again, just like in the native shoulder, here's a case of a tear of the rotator cuff after arthroplasty, we can see the high riding humeral head, which is an indirect sign similar to a native shoulder and the cuff is completely absent.

So as a side point about shoulder replacements, the one area where I continue to struggle is the subscapularis because they go through the subscapularis to do the replacement and then they suture it back up.

So approaching the subscapularis would be like approaching the postoperative supraspinatus that I started this lecture with where we'll always look heterogeneous.

I'm looking for an obvious defect.

Other Shoulder Complications

So as we lead the shoulder, I just wanna show you an example of a retained sponge, which is called a gossypiboma.

And this is showing with MRI where basically we see this area of the sponge with this ring enhancement and with ultrasound, this can be very confusing because remember the echoes on ultrasound are interfaces and if you put the ultrasound transducer here, you would see a bunch of echogenic areas and artifacts.

So that can be very problematic.

Now this is a case of a transection neuroma.

This is after an amputation and what will happen here is the nerve will form a neuroma and it can form a mass like area.

And if we're not aware of this phenomenon, which is an expected finding after a nerve transection, this could be called a mass or something.

And this was done for a sarcoma.

And that's why when looking at this before calling this sarcoma recurrence, realize that the nerve is going into this.

Of course when you push on this, it'll have phantom pain.

So just realize what happens after an amputation and transection of a nerve that a neuroma is a normal or expected response.

Elbow

Biceps Tendon Repair

Okay, moving on to the elbow first biceps tendon repair.

Now this appearance is similar to any other tendon repair.

I could show you an achilles or any other and it is appearance of a thickened hypoechoic tendon.

So at first glance it looks like tendinosis, but then you see this suture material weaving within and out of the fibers along the long axis of the tendon.

So this is the typical appearance that we see of a repaired and intact biceps brachii tendon at the elbow.

And if we're looking for a retear, we look for a defect or retraction.

Ulnar Nerve

And on the elbow topic also realize that when someone has ulnar nerve dislocation like this, the surgeon can come in and then basically, move this nerve so it doesn't keep flopping back and forth with flexion and extension.

So this is what they would be treating would be the ulnar nerve dislocation.

And the surgeon has two options.

The nerve is normally back here and so they'll either more commonly move the nerve or transpose it under the subcutaneous fat or less common underneath the pronator teres.

So you can do submuscular or subcutaneous.

And when they transpose the nerve, it often goes back to normal, but many times it still remains hypoechoic and you have to really correlate with symptoms.

Push on that nerve segment that's abnormal to see if that nerve is still irritated, perhaps with scar tissue.

Wrist

Carpal Tunnel Syndrome

Okay, moving to the wrist first, carpal tunnel syndrome.

We know the criteria for nerve entrapment by ultrasound and that would be enlargement and hypoechogenicity or edema of the nerve.

The term neuritis is actually a misnomer because it's actually edema followed by demyelination and axonal death.

But nonetheless the finding is hypoechoic enlargement as the nerve then goes into the carpal canal.

So when a surgeon is treating this, they'll do a carpal tunnel release.

Now we are unable to determine on the ultrasound if the release was a complete or incomplete release 'cause it always looks the same, meaning the retinaculum will look thickened and will look tented and the nerve will be displaced somewhat in a palmar direction.

So we're unable to say if the release is complete or not.

What is interesting is after an effective release, we expect the nerve to return back to normal size and echogenicity, but it isn't always the case.

The median nerve size does not always correlate with success.

This is one where the nerve was still large and hypoechoic and had residual symptoms, but this person could be completely asymptomatic.

So how I approach a postoperative carpal tunnel release is I'm looking for nerve transection, neuroma suture, granuloma fluid collection.

I could really say something about the nerve, but I know that it really will depend on the patient's clinical symptoms to determine the significance of that finding.

Here's a companion case where we see, again this markedly thickened retinaculum.

Was it an incomplete release? I have no idea.

And here we can see this nerve is hypoechoic enlarged, which could correlate with recurrent symptoms, but again, would have to check with the patient in electrodiagnostic studies to confirm that.

Wartenberg Syndrome

And then Wartenberg syndrome is a condition where you have an entrapment or injury to the superficial branch of the radial nerve that goes over the first wrist compartment.

This is something we have to keep an eye on when we're injecting for De Quervain's.

But why I'm showing this case, this person had a radius fracture and had ORIF or fixation and what has happened is there's scar tissue related to the screw and the superficial branch of the radial nerve is going into the scar tissue.

You can see the nerve going directly into it.

So just be aware that when patients develop new nerve symptoms after surgery, we've gotta track these individual nerve trunks and bundles to see how this relates to the surgery.

De Quervain's Tenosynovitis

Now here's a case of De Quervain's tenosynovitis where we see a thickened sheath and hyperemia.

So this is a typical but more severe example given the degree of hyperemia.

Remember the primary finding of De Quervain's is actually the thickened sheath more so than the tendon problem 'cause it's a stenosis problem.

You can see the tendon deviating underneath this markedly thickened retinaculum or sheath.

So one way to treat this is to release the retinaculum.

This is a patient who had snapping that occurred with pronation and supination.

And what we found is that the abductor pollicis brevis was actually snapping over the edge of the radius due to this release.

So ultrasound is effective after these surgeries when there's popping or snapping because you can move the patient in certain positions to see what is happening.

Hip Arthroplasty

Okay, moving on to the hip first, looking at arthroplasty.

So the bone landmarks of the hip are of course replaced with the hardware of a hip arthroplasty.

Here we see the acetabular femoral head and neck components, no reverberation artifact deep to these metal components.

They're shadowing deep to the femur and the acetabulum.

So you can see the artifact, which is deep, which is nice 'cause it doesn't obscure the area over the top.

Now we have completed a study looking at this hypoechoic layer over the top of the arthroplasty.

We try to determine could we predict if this was fluid or not and we could not.

So what came from the study is that a hypoechoic halo around the hardware could be post-surgical change or fluid.

And we really can't tell when it's that thin.

So if the clinicians worry about infection, I'm unable given this finding, I'm unable to say there's no fluid.

What I would say is that this minimal hypoechoic layer could be post-surgical.

The lack that is not significantly distended, argues against infection.

But of course I cannot exclude infection and you'd have to put your needle into the joint.

When the pseudocapsule starts to form and distended even more, then there's a higher likelihood of infection.

If you're into numbers, this was a study that came up with 3.2 millimeters.

See, I wouldn't rely on that personally.

I would look at this and say, well this isn't a thin halo, this is more than that.

If you're worried about infection, we have to put a needle in.

I don't care if it's 3.1 or 3.3 millimeters.

Now when the fluid becomes really marked and extending outside of the joint, this is highly suspicious for infection, which this truly was infected.

Now in that note, when you are approaching a postoperative hip or any joint, you really have to look at the radiograph to look at the extent of where the hardware is located.

For example, this is an endoprosthesis that's placed for a sarcoma resection.

Now if you just focused here, we'd be missing the abscess, which is located right here.

In fact, we aspirated this joint and the patient pointed down here and said, doc, I hurt down here.

We put the probe down here and saw the abscess.

There's the femur with the shadowing, there's the endoprosthesis with the reverberation.

Now I've seen abscesses down here at the stem.

So if you weren't aware of how long the stem is, we could totally miss screening this area for fluid.

So look at the x-ray, see the entire extent of where the surgeon has been and where the hardware's located and cover all that ground.

Pseudotumor Formation

And as we leave the hip, we're aware of what's called pseudotumor formation, which occurs with metal on metal arthroplasties where it's, there's a metal debris that has a reaction that produces this thickened hypoechoic tissue coming out of the joint, going around the greater trochanter.

And it looks like a tumor, hence the term pseudotumor formation.

But this is really a reaction to the metal on metal debris from the arthroplasty.

Now ultrasound has been shown to be effective in looking for this in the orthopedic literature.

However, I really don't personally rely on it and that's because I've seen a number of cases where there's been only minimal debris or reaction at the joint.

But there was much extent going into the pelvis.

So you could screen with ultrasound, but I wouldn't stop with a negative ultrasound 'cause it could be actually going into the pelvis and not seen well with ultrasound.

Incisional Abscess

And when you're looking at a postoperative hip or any other joint, I'm also focusing wherever the suture line is because if you're gonna have an incisional abscess, of course that's where you would see that.

And then if you see bright echoes within the fluid collection with these dirty shadowing or ring down artifacts, that's typical for gas.

And here you can barely see the gas, but we see it better in here.

So that of course would make it highly suspicious for infection in this case and warrant an aspiration.

Nerve Transection and Neuromas

Now getting back to nerve transection, I showed a case in the shoulder after the amputation where there's that large neuroma that almost simulated a sarcoma recurrence.

Again, I wanna reiterate the fact that when you have an amputation, all those nerves that are transected, they will form neuromas.

Now this is the value of ultrasound after a knee amputation because all those nerve trunks are gonna have these neuromas.

And if I just look with MRI, I'll see the neuromas unless they're really small, but I won't know which one is symptomatic.

So with ultrasound, what I do is I'll scan each of the individual nerves, find the neuromas and push on them.

And with the transducer that is, and I'll ask the patient, is this the one producing your symptoms?

And that's how we can decide which one we need to address perhaps with an injection.

So here's an example of a sciatic nerve neuroma.

The nerve splay is going directly into this.

This one is a small one. Here it is on MRI as well.

And when these nerves are transected, they're almost like a rubber band where they recoil and they have this in and out of plane appearance.

So the key is with ultrasound is to find these neuromas and see which one is truly symptomatic.

Ankle: Tendon and Ligament Repairs

Okay? Then finally, tendon injury and ligament repair in the ankle.

Once again, if someone has symptoms, we need to look at the radiograph to see where the hardware is located because this would be the culprit.

Then we can see these metal fragments that are within the area, the ankle.

And you can see this tibialis posterior tendon is abnormal.

There's tendinosis, there's clefts in here, there's metal debris even within the tendon.

I'm not quite sure what happened here.

So you have to be very careful how you dictate the impression of this case.

You don't wanna imply or pass blame.

And you know, I wasn't at the surgery.

I'm not sure what happened.

I'm sure the surgeon did the best that they could.

But nonetheless you would say that there's metal debris in and about the tendon with partial tearing and they can figure out why it's there, although there's a clue.

Here's a case where a patient had foot pain after plate fixation of the proximal phalanx and we look at the radiograph and say, okay, we can see there the screw tips are here.

In fact, these screw tips are actually going into the flexor hallucis longus causing the patient's symptoms.

There's a little bit of a halo here of some irritation or fluid as well.

And of course we push on this with the transducer and the patient says that that's the pain that I'm having.

This person had a pain at the top of the foot and what we see is the screw head is protruding and there's fluid around it.

And you can see when the patient's moving that it's basically going over the top of this screw head.

It's deviated and the patient had symptoms with this movement and I could push with the transducer and say, yes, this is what's causing your symptoms.

Now to finish with ligament tears, here are just some examples of anterior talofibular ligament tears, the most common ligament that's torn in the lateral ankle.

Basically hematoma is formed here.

There's a bone fragment as well where we see where we should normally see the ligament.

It's replaced with the hematoma.

So there are a whole bunch of surgeries of how to repair this and one of them, which I'm not getting into, is where they take the various peroneus or fibular tendons and reroute them.

But where I'm showing this is, this is a modified Brostrom procedure where basically they take the retinaculum and they tack it down into the ligament and try to reconstruct it.

So if you're not aware of what's going on, we'd look at this and say, this ligament looks funny, there's suture material.

This is what we'll see with this procedure where they try to do a primary reconstruction.

It's still a commonly performed procedure.

Take Home Points

So the take home points of what I tried to cover with postoperative imaging, I first focused on the rotator cuff.

Now remember that most tears occur within three months.

It doesn't look quote normal till six or nine months.

You can't use the secondary findings.

And I basically thinking is there a definitive retear or not?

My dictation is such that I'll say postoperative changes, no retear or there is a retear.

And if you question something, get a follow up because they'll either go away or they will get larger and make your life easier.

I went on to talk about hip arthroplasty and basically that thin halo can is equivocal of what that means.

And to be honest, regardless of what you see, if you worry about infection, you'd have to put a needle into the joint.

But of course, the more fluid there is, the more likely it's infected nerve injury.

Basically a neuroma is an expected response and ultrasound can determine which neuroma is symptomatic.

Then lastly, a primary repair of tendon or ligament.

The tendon will be hypoechoic.

It'll be thickened, you'll see suture material in it.

And basically again, I'm looking for an obvious unequivocal defect in two planes before calling a tear with confidence.

Thank you very much for your attention.