Not a month goes by without the ACR’s economics staff being contacted about establishing new Current Procedural Terminology (CPT®) codes. Some requests are for new ways of doing something (such as ultra-low-dose lung CT or abbreviated breast MRI), while others are for screening codes (such as screening breast MRI or screening liver MRI). Some requests are for more specific forms of existing codes (prostate MRI rather than pelvic MRI), and some requests are for exams that may not currently have great options for coding (such as whole-body CT or MRI). Sometimes, codes are desired for unique clinical circumstances.

What most radiologists are unaware of is that everything we do in creating new codes has a risk/reward ratio. A decision to bring forth new codes must be carefully considered in terms of current coding and reimbursement. The purpose of this two-part series on CPT codes is to detail some of the many considerations that your ACR CPT and Relative Value Scale Update Committee (RUC) volunteers and staff undergo when contemplating new codes.

First, a short discussion of current CPT codes is in order, especially for those members who are new to this topic. Each CPT code descriptor (that you will find in the CPT book or online version) delineates the service that is performed. It is required that the service you are coding for matches the service described in CPT. That said, there are a few important points regarding coding.

For CT and radiographs, there is no definition of “dose.” Folks often want codes for “low-dose” or “ultra-low-dose,” but those are not defined, and it would then be necessary to define “high-dose” (and who would want to get a high-dose CT?). So, for example, whatever dose you use (it should be as low as possible, obviously) for a CT of the abdomen and pelvis without contrast, the procedure is coded as 74146. The only exception is CPT code 71271: Computed tomography, thorax, low-dose for lung cancer screening, without contrast material(s). This particular code had the term “low-dose” placed into the descriptor over our objections, to parallel the format of an existing temporary code (also known as G-code) that CMS insisted we follow — even though we argued that we always use the lowest dose possible.

Similarly, there are no separate MRI codes depending on how many sequences you perform. There are no separate codes to differentiate how many post-contrast sequences you perform (just a portal venous phase versus arterial, portal venous, and three sets of delayed sequences coded the same). Finally, there are no separate codes to capture how advanced your equipment is (0.3T versus 3T; 4-slice versus 246-slice CT; big-box US versus laptop US). Many of us, myself included, think there should be a quality differentiator and different codes that would recognize some of these differences, but as of now that is not possible — and trying to create those differentiating factors would require a complete revamp of existing CPT codes, with the associated risks outlined in part two of this series in the June issue.

So, what are the CPT rules that we play by? First of all, you can’t “fragment” an existing CPT code without also updating the existing code. So, if a code already covers something we do and you want a new code, you will have to redefine the old code. If we wanted to create a new low-dose abdomen CT, we would have to define parameters of what constitutes low-dose and redefine the current codes as higher than that dose. Similarly, if we wanted a code for CT of just the thyroid or liver, we would have to differentiate those from CT neck or CT abdomen, which already include those body parts. If we wanted a new abbreviated MR code, we would first need to define specifically what “full” and “abbreviated” mean (and the literature would have to be consistent).

As you can see, this process is a little more complex that it might seem. Assuming we were able to define a new service as something different than what is covered by existing codes (and redefine the old codes) and we were successful in creating the new desired code(s), what’s next? In the June issue of the Bulletin, we’ll outline the subsequent steps in the lifecycle of a new code.