Neurosurgery Coding Alert

Co-surgeon status is denoted by attaching modifier -62 (two surgeons) to the procedure, which CPT 2000 defines as when two surgeons work together as primary surgeons performing distinct part(s) of a single reportable procedure, each surgeon should report his/her distinct operative work by adding modifier -62 to the single definitive procedure code. Each surgeon should report the co-surgery once using the same procedure code.

Co-surgery commonly occurs when a single procedure requires the talents of two different specialties or sub-specialties. In this case study, the neurosurgeon using the retroperitoneal approach to perform disc surgery requires the services of a vascular surgeon to access the surgical site.

Case Description

The patient is a 55-year-old male with a diagnosis of a fracture of vertebral column without mention of spinal cord injury; lumbar, closed (805.4) and pathologic fracture of vertebrae (733.13).

Coding Notebook

The operative procedures begin with a description of the work performed by the vascular surgeon and the neurosurgeon acting as a co-surgeon. Code 63090 (vertebral corpectomy [vertebral body resection], partial or complete, transperitoneal or retroperitoneal approach with decompression of spinal cord, cauda equina or nerve root(s), lower thoracic, lumbar, or sacral; single segment) would be billed with a -62 modifier (two surgeons) by both surgeons.

Once the vascular surgeon accessed the site, however, the neurosurgeon, performed additional procedures. These include 22558 (arthrodesis, anterior interbody technique, including minimal diskectomy to prepare interspace [other than for decompression]; lumbar) and 22585 ( each additional interspace). Because the fusion extended over two interspaces, 22845 (anterior instrumentation; two to three vertebral segments) was billed for the instrumentation inserted to stabilize the spine and 20931 (allograft for spine surgery only; structural) was billed for the bone graft that was used as an additional strut for the spine. For these procedures the vascular surgeon was no longer a co-surgeon; he or she assisted during these services and should bill using modifier -80 (assistant surgeon). The neurosurgeon would bill these procedures without a modifier.

Be Careful Not to Confuse Modifier Use

Both surgeons would submit separate claims for 63090 with modifier -62 attached. In addition, both surgeons should submit separate operative reports describing the procedure. The same diagnosis codes should be used and the documentation of both surgeons needs to state that they were co-surgeons for the procedure, says Barbara Cobuzzi, MBA, CPC, CHBME, president of Cash Flow Solutions, a coding and reimbursement consulting firm in Lakewood, N.J.

According to CPT guidelines regarding modifier -62 use, it is acceptable to bill as a co-surgeon on one procedure and as an assistant surgeon on another. As a result, some private carriers may accept such claims. Many Medicare carriers, however, may not allow payment for a co-surgeon and an assistant during the same operative session. In other words, the surgeon may bill as a co-surgeon, or an assistant, but not both.

Co-surgeons are reimbursed at a substantially higher level than assistant surgeons. Billing as an assistant when co-surgery was warranted can be a costly error. Medicare pays for co-surgery by multiplying the fee for the procedure by 1.25 and splitting that between the surgeons. If a procedure normally is reimbursed at $1,000, Medicare will pay $1,250 for the procedure when co-surgeons perform it and pay each surgeon half. When billing for co-surgery, neurosurgeons should bill at 62.5 percent the regular fee. The explanation of benefits needs to be carefully checked to ensure that the proper fee was paid for the co-surgery, and an appeal may be required to receive the correct payment.

Assistant surgeons receive much less. Medicare will reimburse the assistant surgeon at 16 percent of the procedure fee, while private carriers pay better from 25 percent to 50 percent Cobuzzi says.

Some procedures do not allow for payment of co-surgeons and assistant surgeons. The national Physician Fee Schedule Relative Value Guide, a listing of all CPT and HCPCS procedures, outlines the services that can be billed with modifier -62 and modifier -80. Some private payers do not publish the lists. Neurosurgeons should check with carriers beforehand or be prepared to launch an appeal to get paid, Cobuzzi says.

Medicare Warning

False information within neurology practices regarding how to bill this type of surgery has created problems. This misinformation has been dealt with through regulations on unbundling found in HCFAs national Correct Coding Initiative (CCI), which went into effect in July 1996. This banned the use of laparotomy (i.e., 49000 and 49010) or thoracotomy (32100) codes for exposing the spine and considers the use of these codes in such procedures as unbundling.

Medicare has labeled it as potentially fraudulent for a neurosurgeon doing the vertebral body resection to bill 63090 and the surgeon who did the approach to bill 49010 (exploration, retroperitoneal area with or without biopsy). The issue of fraud arises because 63090 includes the approach and the closure, while 49010 is an exploratory code meant for diagnosis, not for complex spinal surgery. Having one surgeon bill for the approach and another bill for the vertebral excision, which also includes the approach, is considered double dipping.

Susan Callaway-Stradley, CPC, CCS-P, an independent coding consultant and educator in North Augusta, S.C., notes that private payers are looking at HCFAs CCI and retroactively going after payments. Even a successfully paid 49010 claim could be recalled by the payer in the future if an audit reveals it has been wrongly paid.

Operative Report The patient was brought to the operating room and given general endotracheal anesthesia. Foley catheter, arterial and additional peripheral IV lines were placed. The patient was given two grams of Ancef intravenously. He was turned to the lateral decubitus position with the right side up. The hips were moderately flexed to relax tension on the psoas muscles to facilitate exposure of the anterior spine. The vascular surgeon performed the retroperitoneal exposure with mobilization of the vena cava by ligating and then severing the segmental vessels across L1-2 and L3 vertebral bodies with the neurosurgeon acting as a co-surgeon. The vascular surgeon then assisted for the decompression, fusion and internal fixation portion of the procedure. The patient was found to have evidence of retroperitoneal hematoma, presumably due to the fracture. The psoas muscle was edamatous and indurated, making retraction difficult. Across the L2 vertebral body the periosteum was thickened and there was moderate induration and scarring making identification of the segmental vessels difficult, however. The segmental vessels at L1, L2, and L3 were clamped with hemostatic clips or suture ties before being released. The vena cava was carefully protected during the procedure. Once adequate exposure of the anterior portion of the spine had been achieved, we then worked to achieve exposure posteriorly on the vertebral bodies back to the pedicles at L1, L2, and L3. Diskectomies were performed at L1-2 and L2-3. Prior to taking the segmental vessels, the levels were confirmed by taking an x-ray with needle in one of the disc spaces which proved to be L1-2. Vertebrectomy of L2 was then performed. There were some areas of soft tissue within the L2 vertebral body, particularly on the right side. Some of this had the appearance of herniated nucleus pulposus that possibly had herniated into the vertebral body at the time of the fracture. The posterior Cordis cortex was bent carefully with a high speed bur under direct visualization until most of it could be gently pulled forward off the posterior longitudinal ligament and decompressing the spinal canal, using curets. Adequate decompression was achieved in this fashion. The end-plates were scraped down to the subchondral bone at L1 and L3. The vertebrectomy site was thoroughly irrigated with antibiotic solution and a femoral strut allograft was sized and placed in the space between L1 and L2 after putting in a laminar spreader to restore normal curvature and partially correct the scoliotic deformity due to the fracture. This was supplemented with a piece of the patients 12th rib as an additional strut. This has been resected during the approach. A Synthes thoracolumbar anterior locking plate was then temporarily affixed to the spine with the 4.0 compression screws. Care was taken to put the plate as far posteriorly as possible so that it was more or less directly in the lateral aspect of the vertebral bodies to ensure that the screws would be passing in a transverse fashion rather than toward the spinal canal. Excellent purchase was obtained with all of the screws. The compression screws were used to apply compression across the graft for added stability. The wound was thoroughly irrigated again with antibiotic solution. The vascular surgeon then performed closure of the wound. A sterile dressing was applied. The patient was awakened, extubated and taken to the recovery room in stabile condition. He could flex and extend the toes and ankles bilaterally to command after awakening.