Ask Dr. Z

I hope you can give some insight into this procedure. Basically the physician performed open revision with thrombectomy (36832), then performed fistulogram (36147), followed by percutaneous venous angioplasy (35476 and 75978-26), and then decided to ligate the entire fistula (37607) and create a whole new graft (36830).  Based on the below documentation, would you bill all those codes? Or should only the open procedure be coded as per NCCI Chapter 5, Section D, #9? Any assistance will be appreciated!

A linear incision was made in the fistula at the arterial anastomosis. I noted immediately that the thrombus was well organized and adherent to the fistula walls. It required mechanical removal. I carefully inspected the area of the arterial anastomosis, removing the fibrin plug. I passed a #3 Fogarty catheter distally in the brachial artery and retrieved no additional thrombus. I sounded the proximal brachial artery with the right angle, and there was no evidence of a stricture at the arterial anastomosis. I removed as much thrombus from the body of the fistula as allowed by the arterial cuff, which had been placed proximally. In order to control the arterial inflow and to avoid stricturing of the fistula, I acquired a bovine patch and partially closed the fistulotomy with the bovine patch and 6-0 Prolene suture. This allowed for application of an atraumatic clamp at the arterial anastomosis and removal of the proximal arterial tourniquet. I evacuated the clot from the remaining portion of the fistula body by vigorous manipulation beginning at the axilla. I removed a relatively small amount of clot. I did retrieve venous backbleeding. Heparinized saline was instilled, and an atraumatic clamp was placed on the body of the graft. The patch angioplasty was completed. There was a pulse within the graft with removal of the arterial tourniquet. This was not accompanied by a thrill though there was a continuous Doppler signal. I cannulated the patch with a 21 gauge micropuncture needle. I advanced the 0.018 guidewire under fluoroscopy. The needle was exchanged for a 5 French transitional dilator. I removed the inner stiffener and 0.018 guidewire, and through the transitional dilator, I performed a fistulogram. Although there was continuous flow in the fistula the fistula was noted to be quite sclerotic. This did not appear to be thrombus. A retrograde filling of the brachial artery revealed the arterial anastomosis to be widely patent. I attempted to pass a short 0.035 guidewire through the transitional dilator, but it would not negotiate the fistula. I acquired a 0.035 Glidewire, and with some manipulation the Glidewire traversed the fistula and was placed in the superior vena cava. I removed the 5 French dilator and advanced a 6 French short sheath. I advanced a 5 French Kumpe catheter over the Glidewire and exchanged the Glidewire for a 0.035 Rosen wire. I repeated the fistulogram documenting the fairly extensive sclerotic changes within the fistula. Again, these did not appear to be thrombus. I acquired a 5 French and subsequently a 6 French x 4 centimeter balloon catheter and proceeded to dilate the entire fistula from the end of the sheath to the basilic vein junction with the brachial vein. There was no evidence of a central stenosis. The balloons were inflated to pressures of 14 millimeters of mercury. Following the balloon angioplasty, I repeated the fistulogram. While there was some improvement in the luminal diameter of the fistula, it remained quite ratty and there was sluggish flow. I did not feel that further efforts at maintaining the fistula would be productive. I ligated the fistula just beyond the arterial anastomosis. I proceeded with an AV graft insertion. A short incision was made in the axilla, and I identified a 12 millimeter brachial vein. I carefully dissected between the nerve trunks and identified a 6-7 millimeter axillary artery. The artery lies medial and deep to the vein. A counterincision was made on the upper arm to allow for tunneling in a loop configuration. The patient was given an additional 1000 units of heparin. I carefully exposed the artery, placing no tension on the nerve trunks. An end-to-side arterial anastomosis was completed with 5-0 Prolene suture. Two of the three large nerve trunks lie medial to the graft and one lies lateral. Upon completion of the anastomosis, there was no anastomotic bleeding. The bovine graft was then withdrawn through the subcutaneous tunnel in two movements. It was allowed to lie in a gentle loop configuration. A partial occlusion clamp was placed on the axillobrachial vein, and an end-to-side anastomosis was completed between the bovine graft and the vein with a 5-0 Prolene suture. Whereas the arterial anastomosis is 5-6 millimeters in length, the venous anastomosis is 8-10 millimeters in length. Prior to completing the anastomosis, the vessels were vented and were flushed with heparinized saline. There was minimal anastomotic oozing. This was readily controlled with Fibrillar. Once hemostasis was confirmed, the three operative wounds were closed with two layers of absorbable suture.