ECMO Coding

As you know, there are now three ICD-10-PCS codes to capture ECMO based on the method of cannulation.  ECMO with central cannulation continues to be grouped to a surgical MS-DRG.  Whereas, VA-ECMO and VV-ECMO are captured as peripheral cannulation and no longer a surgical MS-DRG.   This coding change has great  implication financially for hospitals as the peripheral cannulation reimbursement has dramatically been reduced for these critically ill patients. 

My facility has had some discussion and some feel that the VA ECMO should maybe be coded to the heart assist device procedure code, since a Centrimag or Pedimag is used and this is used for heart assistance. 

So, I reaching out for some assistance from some CDI experts on what they think of this discussion and the general censuses for capturing the heart assist device procedure code with VA ECMO.  Is your facility having any discussions regarding these new ECMO coding and taking any steps to help migrate the decrease reimbursement? 

I know that the ECMO Society, along with the Cardiothoracic Physician Society has taken this issue to CMS and requesting changes, but in the meantime, what do we do? 

Thanks!  Karen

Karen Bridgeman

Medical University of South Carolina

bridgema@musc.edu

                                                                                                                                                                              

Comments

  • Hi Karen,

    We use ECMO quite a bit and yes this change is having substantial financial implications for hospitals.

    In the scenario you provided, I would code the ECMO VA PCS code and the 5A02216 Assistance with Cardiac Output using other cardiac pump. Adding the additional assistance code for the Centrimag should take you to DRG 221 depending on the principal diagnosis.

    I do not believe you can use the additional insertion code for the Centrimag because it is not directly cannulated in the heart. But I would love to hear other's opinions.

    I hope this helps.

    Joan Oliver

    joan.oliver@duke.edu

Sign In or Register to comment.