Si Rf

We do a similar technique pretty frequently in my fellowhsip. Honestly, I'm not sold on the efficacy of this particular procedure. I suspect the variability in the depth distribution of the dorsal branches in the ligament makes it very difficult to get all the innervation even with strip lesions.

There is a decent review article in ASRA/Regional anesthesia a few months ago attempting to described the innervation of the SI joint. Apparently this group also describes lesions surrounding the S1 and S2 foramina as well.

The problem I have with this conceptually is that you can't denervate the anterior joint.

Baylis Medical is marketing a cooled RF attachment for their RF system for SIJ work. It is based off a study by Yin and Dreyfuss et al from Spine 2003. It involvs nine lesions and would be billed the same as RF. They claim the cooled system created a wider area so you can "carpet bomb" the lateral dosal sensory nerves that inervate the SIJ. There is a picture on the website.

http://www.baylismedical.com/PMProdList9.html

Yin W, Willard F, Carreiro J,Dreyfuss P.Sensory stimulation-guided sacroiliac joint radiofrequency neurotomy: technique based on neuroanatomy of the dorsal sacral plexus. Spine. 2003 Oct 15;28(20):2419-25.

Interventional Medical Associates of Bellingham, P.C., 2075 Barkley Boulevard, Suite 110, Bellingham, WA 98226, USA.

STUDY DESIGN: A retrospective audit and examination of anatomic findings. OBJECTIVE: To examine the effectiveness of sensory stimulation-guided radiofrequency neurotomy for the treatment of recalcitrant sacroiliac joint pain. SUMMARY OF BACKGROUND DATA: Sacroiliac joint-mediated pain is a distinct clinical entity. The prevalence of intra-articular pain arising from the sacroiliac joint in patients with low back pain has been estimated at 15% to 30%. Unfortunately, the clinical success of current treatment methods for chronic sacroiliac pain is discouraging. Based on the anatomy of the sacral posterior primary rami and their lateral branch nerves, an anatomically based sensory stimulation-guided radiofrequency technique was developed to overcome the inherent challenge posed by the variable topography of the sacral lateral branch nerves. MATERIALS AND METHODS ANATOMIC STUDY: Meticulous dissection exposing the dorsal sacral plexus and lateral branch nerves entering the sacroiliac joint complex was performed on three cadaveric specimens. Small-gauge wires were placed adjacent to the lateral branch nerves entering the joint and over the dorsal sacrum to the dorsal sacral foramina. Fluoroscopic images were obtained correlating the location and number of these branches arising from the posterior primary rami of S1-S3 to identifiable bony landmarks. CLINICAL STUDY: A retrospective chart review was performed selecting patients who underwent sensory stimulation-guided sacral lateral branch radiofrequency neurotomy after dual analgesic sacroiliac joint deep interosseous ligament analgesic testing between February 17, 1998 and March 15, 1999. RESULTS: A total of 14 patients met inclusion criteria for this retrospective study. Success was defined as greater than 60% consistent subjective relief and greater than a 50% consistent decrease in visual integer pain score, maintained for at least 6 months after the procedure. Sixty-four percent of patients experienced a successful outcome, with 36% experiencing complete relief. Fourteen percent of patients did not achieve any improvement. No patients experienced a complication or worsening of their pain from the procedure. CONCLUSIONS: A sensory stimulation-guided approach toward the identification and subsequent radiofrequency thermocoagulation of symptomatic sacral lateral branch nerves appears to offer significant therapeutic advantages over existing therapies for the treatment of chronic sacroiliac joint complex pain.

64622 and 64623 is what we have been using.

I'm in the process of cobbling together a retrospective case series similar to what the Yin study demonstrated; we went a bit higher up to including L5-S1 in addition to S1&2 foramen.

Our results are less than thrilling. Although we are finding out a lot about the practice habits of some of our staff!

I've tried (two patients)the technique recently published in ASRA journal (around January) with bipolar lesions around each foramen from S1-4. Very tedious, and no better results.

Are people using a double diagnostic comparitive block technique ala Cervical and Lumbar RFA?

correct coding would be neurodestruction of peripheral branch --- problem is that most insurances will only pay for one branch at a time - and therefore you will likely get paid only once for you "carpet bomb" 9 lesions ---

i have done a few and have not been impressed with results (anecdotal)

and 64623/3 is not correct.

I've not tried it, but have sent refractory SI cases to others for it. I have one guy in town who does it. Seems to be about 1/2 cases get any longer relief than from the SIJI itself, making me wonder if it's just addition of steroid after the procedure that helps.

Stryker rep has talked to me about doing bipolar for it. Until there's good evidence for it and a good code, I'm waiting.

Tenesma said:

correct coding would be neurodestruction of peripheral branch --- problem is that most insurances will only pay for one branch at a time - and therefore you will likely get paid only once for you "carpet bomb" 9 lesions ---

I have done a few and have not been impressed with results (anecdotal)

and 64623/3 is not correct.

Click to expand...

I called Baylis - they recommend either

64622 for L5 Destruction by a neurolytic agent, paravertebral facet joint nerve, lumbar or sacral, single level

and

64623 x3: Lumbar or sacral each additional level (list separately in addition to code for primary procedure)

or

64622 for L5 Destruction by a neurolytic agent, paravertebral facet joint nerve, lumbar or sacral, single level

and

64640 for S1: Destruction by a neurolytic agent - other peripheral nerve or branch

plus

64640-51 for S2 and S3

the 64622 can only be used for dorsal ramus of L5 --- all the other lesions are technically peripheral branches

i wouldn't rely on manufacturers for coding because they will do everything to make sure you do the procedure and use their equipment

i can tell you from my experience i have done these procedures, have tried multiple different ways for coding - and I have been either denied or paid a ridiculously low amount....

Tenesma said:

the 64622 can only be used for dorsal ramus of L5 --- all the other lesions are technically peripheral branches

i wouldn't rely on manufacturers for coding because they will do everything to make sure you do the procedure and use their equipment

i can tell you from my experience i have done these procedures, have tried multiple different ways for coding - and I have been either denied or paid a ridiculously low amount....

Click to expand...

Tenesma is right on (as usual!)

The definition of 64622 is "destruction by neurolytic agent, paravertebral facet joint nerve"

SI nerves are NOT facet joint nerves, so you can't use that code. You can try the peripheral code (64640) but they probably won't pay you for the number of nerves you are doing. I would try the dreaded 64999 and in line 19 say "comparable to 64622 and 64623 x 3" Then price the same as you would for said 64622 and 64623. If you have an ace in your billing department, this can be made to work for you.

This is of course the great craps shoot- sometimes you win with 64999, sometimes you lose. But hey, you won't go to Medicare jail at least.

Ligament said:

Tenesma is right on (as usual!)

The definition of 64622 is "destruction by neurolytic agent, paravertebral facet joint nerve"

SI nerves are NOT facet joint nerves, so you can't use that code.

Click to expand...

You only quoted the portion of the definition that suites your purposes - the full definition actually is: "Destruction by a neurolytic agent, paravertebral facet joint nerve, lumbar or sacral"

I readily acknowledge that there are no facet joints in the sacrum, but by implication of the above definition, I believe you could make the good faith argument that CMS, by making reference to sacral facet joint nerves, acknowledges that they exist, regardless of whether they have a joint to enervate or not.

well i think the issue is not so much that it is lumbar or sacral (obviously they don't know too much of what they are talking about) but rather that the SI joint is not a paravertebral facet joint....

kind of like coding a cervical medial branch block as a cervical TFESI just because it says cervical in the CPT definition...

by the way, i did try coding it that way when i first started doing those and was quickly slapped on the hand --- so the insurances do know a bit about anatomy...

ampaphb said:

You only quoted the portion of the definition that suites your purposes- the full definition actually is: "Destruction by a neurolytic agent, paravertebral facet joint nerve, lumbar or sacral"

I readily acknowledge that there are no facet joints in the sacrum, but by implication of the above definition, I believe you could make the good faith argument that CMS, by making reference to sacral facet joint nerves, acknowledges that they exist, regardless of whether they have a joint to enervate or not.

Click to expand...

I absolutely agree with your point, but to be clear, MY purpose is to get PAID AND KEEP the $$$$$, not hand it back after an audit. If using 64622/23 for SIJ RF allowed me to do this, heck yeah I would! It pays very well! But because it pays well, the insurance company uses my argument instead of yours... it suits their purposes, not mine!

Also, remember that the AMA has an exclusive contract (cough fascist monopoly coughcough) to make up these CPT codes *for CMS*, so it is the AMA that doesn't know what they are talking about. Go figure...

SIJ PAIN is most accurately diagnosed with a flouro guided diagnostic anesthetic injection using time concordant relief, in the presence of no additional analgesic medications and with perfect arthrogram recorded.

In these cases, I have had good results with providing the patient relief with bipolar RF strip lesions medial to the joint, starting inferiorly and going up to the dorsal ramus L5 as the final location. Patients usually get 70-90 percent relief with the RF if they had 100% relief with the diagnostic block. Most common profile is that of a rheumatologic patient with synovitis, and most of these patients also have a facet mediated component. Thus, steroids anywhere will always be glorious SHORT LIVED relief. So I find RF to be a great option for them as opposed to chronic polypharmacy.

I am now using the cooled RF probe by Bayliss and will have more results to share in 90 days. It is definately easier and quicker using the cooled RF, and we will see how it compares to standard bipolar RF of the SIJ.

THE PROFESSOR said:

SIJ PAIN is most accurately diagnosed with a flouro guided diagnostic anesthetic injection using time concordant relief, in the presence of no additional analgesic medications and with perfect arthrogram recorded.

In these cases, I have had good results with providing the patient relief with bipolar RF strip lesions medial to the joint, starting inferiorly and going up to the dorsal ramus L5 as the final location. Patients usually get 70-90 percent relief with the RF if they had 100% relief with the diagnostic block. Most common profile is that of a rheumatologic patient with synovitis, and most of these patients also have a facet mediated component. Thus, steroids anywhere will always be glorious SHORT LIVED relief. So I find RF to be a great option for them as opposed to chronic polypharmacy.

I am now using the cooled RF probe by Bayliss and will have more results to share in 90 days. It is definately easier and quicker using the cooled RF, and we will see how it compares to standard bipolar RF of the SIJ.

Click to expand...

What are you billing your strip lesion as? What anatomic structure do you believe you are cooking?

I would look at 64640 or 64999 for the destruction depending on payor preference. In my experience, payors consider SI joint RF investigational and will not cover the procedure.

I would have your pre-auth person contact the patient's insurance to find out if it is a covered benefit before proceeding.

PAINBILLER said:

I would look at 64640 or 64999 for the destruction depending on payor preference. In my experience, payors consider SI joint RF investigational and will not cover the procedure.

I would have your pre-auth person contact the patient's insurance to find out if it is a covered benefit before proceeding.

Click to expand...

Painbiller, you clearly get my point - there is no identifiable anatomic structure you are legitimately targeting, so I am not even sure you could justify 64640. You are left with 64999 as your only legitimate code, which, as we all know is unlikely to be covered.

C Fiber said:

64622 and 64623 is what we have been using.

Click to expand...

this is not correct coding

Sorry to bump such an old thread. Any new thoughts on billing or efficacy of the procedure? Are insurers paying now in 2012?

We were billing as 64640 x 4 with no problems but my billers just told me that starting next month this would not get reimbursed with regards to the SI joint. Has anyone else heard this??

giddyup said:

We were billing as 64640 x 4 with no problems but my billers just told me that starting next month this would not get reimbursed with regards to the SI joint. Has anyone else heard this??

Click to expand...

Havent heard of it not getting pain but you are coding it correctly...

Are you also lesioning L4 and L5?

lobelsteve said:

The guys who taught you the technique and/or published it are probably posting on this forum and reading your message. We are currently discussing how to get paid for this procedure by linking a CPT code to the ICD9 720.2 and not getting acused of fraud.

Click to expand...

hehehe

This is one of my old threads on the physicians only forum that may shed light on things. BTW, I'm still coding it this way, I do about 2 SIJ RF per month, still getting paid........

http://forums.studentdoctor.net/showthread.php?t=870287

Humana medicare replacement just told me that RF for the SI joint was not proven and uncovered (experimental) . Weird since the reason she has SI pain is because they paid for a 100K lumbar fusion that didn't work. I told her that I would see her back after she got off Humana

bedrock said:

This is one of my old threads on the physicians only forum that may shed light on things. BTW, I'm still coding it this way, I do about 2 SIJ RF per month, still getting paid........

http://forums.studentdoctor.net/showthread.php?t=870287

Click to expand...

The question has never been will you get paid. The appropriate issue, that should keep you up nights, is will you survive a Medicare audit?

ampaphb said:

The question has never been will you get paid. The appropriate issue, that should keep you up nights, is will you survive a Medicare audit?

Click to expand...

why wouldnt he?

If he's seeing the patient and writing down what he is doing and actuallydoing it? seems reaonsable?

Also I'm assuming they are prior authorizing it, so why wouldnt it be ok?

Does medicare specifically forbid RF of the sacral nerves?

Seems there are a lot of nerves you could or couldn't RF in the body and I'm not sure medicare has some policy about each branch.......

lobelsteve said:

Medicare does not require PA or precert. But if you break their rules.....

Click to expand...

I thought they did on 'bigger' procedures like RFA, SCS, Vplasty?

My scheduler has always told me that we need to get the RFA (or SCS,or whatever bigger procedure) approved before we schedule them? I'm at a hospital and they usually mk sure things are approved first so that the patient doesnt get shafted. The private group across the street does procedures,etc or whatever on patients and then have had the patient stuck with a bill (urban legend at my place atleast).

Correct me if this isnt true???? I'll have to have a 'discussion' with my scheduler if it isnt.

PinchandBurn said:

I thought they did on 'bigger' procedures like RFA, SCS, Vplasty?

My scheduler has always told me that we need to get the RFA (or SCS,or whatever bigger procedure) approved before we schedule them? I'm at a hospital and they usually mk sure things are approved first so that the patient doesnt get shafted. The private group across the street does procedures,etc or whatever on patients and then have had the patient stuck with a bill (urban legend at my place atleast).

Correct me if this isnt true???? I'll have to have a 'discussion' with my scheduler if it isnt.

Click to expand...

Medicare does not require prior authorizations for anything even though it may be changing for some orthopedic and interventional cardiology procedures.


Also in most private insurance plans, if you do something that is not approved you are not typically allowed to flip it to the patient. It will say that on the EOB. I read 100% of mine which is why I know this.

Medicare doesn't forbid anything, however, there are many things they simply don't cover, like intradiscal thermal procedures.

Bedrock was advocating using 64450 for lateral branch RF. The code is defined as " Injection for nerve block. Injection, anesthetic agent; other peripheral nerve or branch." While not fraudulent, this is clearly not the primary procedure.

The appropriate codes to use are 64622 and 64623 (Radiofrequency (RF) Ablation or Destruction of Lumbar-Sacral Nerves)

ampaphb said:

Medicare doesn't forbid anything, however, there are many things they simply don't cover, like intradiscal thermal procedures.

Bedrock was advocating using 64450 for lateral branch RF. The code is defined as " Injection for nerve block. Injection, anesthetic agent; other peripheral nerve or branch." While not fraudulent, this is clearly not the primary procedure.

The appropriate codes to use are 64622 and 64623 (Radiofrequency (RF) Ablation or Destruction of Lumbar-Sacral Nerves)

Click to expand...

64622 and 64623 are not appropriate for two reasons....

First, these codes were deleted at the start of 2012. If you use them they will be denies The appropriate codes are 64635 and 64636. Please note that these are now billed per joint and not per nerve.

Second, these codes are for denervation of paravertebral facet joint lumbar/sacral. To be fair that doesnt really accurately describe the above.

ampaphb said:

Medicare doesn't forbid anything, however, there are many things they simply don't cover, like intradiscal thermal procedures.

Bedrock was advocating using 64450 for lateral branch RF. The code is defined as " Injection for nerve block. Injection, anesthetic agent; other peripheral nerve or branch." While not fraudulent, this is clearly not the primary procedure.

The appropriate codes to use are 64622 and 64623 (Radiofrequency (RF) Ablation or Destruction of Lumbar-Sacral Nerves)

Click to expand...

I think what he meant to say was 64640 which is "destruction of peripheral nerves'. This is more appropriate than the 64635 and 64636 as you are not burning a joint. It's the nerves. I think if you meticulously burn each o f the s1,2 and 3 nerves correctly 64640 is veyr appropriate and in fact requires a lot of work/time.

PinchandBurn said:

I think what he meant to say was 64640 which is "destruction of peripheral nerves'. This is more appropriate than the 64635 and 64636 as you are not burning a joint. It's the nerves. I think if you meticulously burn each o f the s1,2 and 3 nerves correctly 64640 is veyr appropriate and in fact requires a lot of work/time.

Click to expand...

I totally agree with you as I said in post #26

Im still hearing rumblings about this not getting reimbursed... Apparently our Neurotherm rep told one of the other docs in my group SI RF with 64640 was not getting reimbursed after August 23rd. Anyone else get wind of this?

giddyup said:

Im still hearing rumblings about this not getting reimbursed... Apparently our Neurotherm rep told one of the other docs in my group SI RF with 64640 was not getting reimbursed after August 23rd. Anyone else get wind of this?

Click to expand...

Sounds like this is an issue with a local insurer. Aug 23rd sound like an arbitrary day. Medicare generally makes changes on Jan 1. Are you sure that every single one of your local insurance companies is going to stop paying for 64640 on the same day in August?

I haven't heard of 64640 not getting paid in the northeast recently, (or that payment would stop last week).

Does anyone have an update on this? I just head from our billing company that medicare is refusing to pay anything on 64640. Also a BCBS of Florida (but not Alabama.) Why do we even play this game anymore? How is everyone getting around this? I actually get pretty good results with the SI RFs.

mille125 said:

Second, these codes are for denervation of paravertebral facet joint lumbar/sacral. To be fair that doesnt really accurately describe the above.

Click to expand...

What is a sacral facet joint if not the sacroiliac?

We simply need a new code for SI RF. It's at least as much work as a bilateral two level lumbar RF.

We do mostly the neurotherm simplicity III (single probe, 3 electrodes) and the Bayliss cooled system in my fellowship. Seems like we have good results with both. Simplicity is much quicker. 5 min burn time and pretty easy placement.

This is how RF is coded, as I understand it.

64635: Destruction by neurolytic agent, paravertebral facet joint nerve; (Fluoroscopy or CT); lumbar or
sacral, single level
64640: Destruction by neurolytic agent; other peripheral nerve or branch
RF denervation in the sacroiliac region is commonly done at L5, S1, S2, and S3 levels. Physicians
who currently perform RF denervation procedure in the sacroiliac region commonly use the following
approach in coding:
RF lesion at L5: 64635
RF Lesions at S1: 64640-59
RF lesions at S2: 64640-59
RF lesions at S3: 64640-59
Note: For bilateral procedures, use Modifier-50
According to the AMA, as published in the CPT Assistant, December 2009:
"To differentiate between the work when performing sacral nerve destruction of S1, S2, S3, and
S4, each individually separate peripheral nerve root neurolytic block is reported as destruction of a
peripheral nerve, using code 64640, Destruction by neurolytic agent; other peripheral nerve or branch.
In this instance, code 64640 is reported four times. It is suggested that Modifier 59, Distinct Procedural
Service, be appended as well."

tugbug said:

We do mostly the neurotherm simplicity III (single probe, 3 electrodes) and the Bayliss cooled system in my fellowship. Seems like we have good results with both. Simplicity is much quicker. 5 min burn time and pretty easy placement.

This is how RF is coded, as I understand it.

64635: Destruction by neurolytic agent, paravertebral facet joint nerve; (Fluoroscopy or CT); lumbar or
sacral, single level
64640: Destruction by neurolytic agent; other peripheral nerve or branch
RF denervation in the sacroiliac region is commonly done at L5, S1, S2, and S3 levels. Physicians
who currently perform RF denervation procedure in the sacroiliac region commonly use the following
approach in coding:
RF lesion at L5: 64635
RF Lesions at S1: 64640-59
RF lesions at S2: 64640-59
RF lesions at S3: 64640-59
Note: For bilateral procedures, use Modifier-50
According to the AMA, as published in the CPT Assistant, December 2009:
“To differentiate between the work when performing sacral nerve destruction of S1, S2, S3, and
S4, each individually separate peripheral nerve root neurolytic block is reported as destruction of a
peripheral nerve, using code 64640, Destruction by neurolytic agent; other peripheral nerve or branch.
In this instance, code 64640 is reported four times. It is suggested that Modifier 59, Distinct Procedural
Service, be appended as well.”

Click to expand...

Any follow-up on coding for SIJ RF. Is this correct? I have not done it in a while, because I did not think it was being covered.

I am considering asking for a new Neurotherm RF machine to be able to do the Simplicity lesioning. I've heard so many different answers on how to bill for SIJ RF- I am confused. Is it worth shelling out the extra bucks for a new machine with SImplicity Probe capability, or stick with an older machine. For MBRF, I do one lesion at a time, and can do a 3 level within 20-30min, so having the capability to do multiple MB lesions for facets at one time is not an important feature for me. However, the Simplicity probe is attractive and would certainly cut down the time with SIJ RF, so long as I can bill for it correctly and get paid. Volume is 6-8 RF's/wk

What is the latest verdict on SIJ RF coding and payment?
WHAT IS THE CORRECT WAY TO CODE SIJ RF?

Thanks!

wscott said:

Any follow-up on coding for SIJ RF. Is this correct? I have not done it in a while, because I did not think it was being covered.

I am considering asking for a new Neurotherm RF machine to be able to do the Simplicity lesioning. I've heard so many different answers on how to bill for SIJ RF- I am confused. Is it worth shelling out the extra bucks for a new machine with SImplicity Probe capability, or stick with an older machine. For MBRF, I do one lesion at a time, and can do a 3 level within 20-30min, so having the capability to do multiple MB lesions for facets at one time is not an important feature for me. However, the Simplicity probe is attractive and would certainly cut down the time with SIJ RF, so long as I can bill for it correctly and get paid. Volume is 6-8 RF's/wk

What is the latest verdict on SIJ RF coding and payment?
WHAT IS THE CORRECT WAY TO CODE SIJ RF?

Thanks!

Click to expand...

If you do L4 and L5 medial branches, you can get one facet level. For S1-3, the best you're going to do is peripheral nerve RF codes, which pay poorly relative to the cost of a Simplicity probe. Unless you have awesome payors, or get facility fees, it's not remotely worth it for the time you invest in the procedure. I think of SI RF as a charity case.

powermd said:

If you do L4 and L5 medial branches, you can get one facet level. For S1-3, the best you're going to do is peripheral nerve RF codes, which pay poorly relative to the cost of a Simplicity probe. Unless you have awesome payors, or get facility fees, it's not remotely worth it for the time you invest in the procedure. I think of SI RF as a charity case.

Click to expand...

I disagree regarding facet codes. I bill a single level lumbar facet RF, just for doing L5. Never had a problem with billing that, despite burning just L5 and not L4. I have billed approx 40 SIJ RF cases that way.
It's not my fault that CMS hasn't come up with a code for this situation.

To summarize, I bill
64635 X 1 for L5 RF
64640 X 3 for S1-S3 RF
77003 X 1 for fluoro guidance during S1-S3 RF portion of procedure
(59 modifiers are a good idea on the 64640 and 77003 codes to clarify you're billing fluoro just for the S1-S3 portion of the procedure)


I do agree that it's not worth it relative to the cost of a Simplicity probe.

I used to do a lot of simplicity cases at the local ASC, where they paid for the probes.
I now do 99% of my cases in the office and do SIJ RF with bipolar burns. Results are similar to simplicity burns. Only need to pay for 4 regular RF probes that way. Makes the cases much more financially viable. Not a real money maker, but I can do the case in 30 minutes and don't lose money on it either.

SIJ RF can be a life changer for some patients, I wouldn't want to deny it to them.