Medical Necessity, Documentation, Coding, and Billing for Spinal Cord Stimulation

May 1, 2019, 16:39 PM by Yeshvant Navalgund, MD

Background

Chronic back pain affects a large portion of the global population, costing billions in direct and indirect medical costs and disrupting the lives of millions of people. In the United States alone, an estimated 7.9 million adults experience chronic back pain.^[1]

Spinal cord stimulation (SCS) has become an important tool in the management of otherwise intractable pain and can be a life-changing therapy for many patients.^[2] SCS targets the dorsal columns of the spinal cord for relief of neuropathic pain. Electrical stimulation of the spinal cord is approved by the Food and Drug Administration for chronic painful disorders of the trunk and extremities such as failed back surgery syndrome and complex regional pain syndrome types I and II.^[3] Currently, the annual worldwide SCS system implantation rate is between 35,000 and 50,000 units.^[4]

The efficacy, safety, and cost-effectiveness of traditional SCS for chronic pain conditions are well-established with level 1 and level 2 evidence.^[5] In the treatment algorithm, SCS performed early in the course of patients’ chronic pain processes is associated with better outcomes than SCS performed late in the disease. Research showed that success was inversely proportional to time between initial pain diagnosis and implantation, and the Neuromodulation Appropriateness Consensus Committee (NACC) recommended that SCS be considered and trialed within the first 2 years of chronic pain.^[2],[6]

Cost

Substantial costs associated with the SCS system arise at the time of surgical implantation as well as at the time of revision (for reasons such as implantable pulse generator battery depletion, lead replacement, device malfunction, and infection).

Several cost-effectiveness analyses, spanning several different countries and cost/reimbursement patterns, have repeatedly demonstrated, at even relatively short follow-up periods (eg, 24 months or less), that patient-reported health outcomes are sufficiently improved such that widely accepted standards of “willingness to pay” are met. The break-even point at which savings associated with SCS are observed in therapy responders has been shown to be 2.1 to 2.5 years.^[7]

Most data covering costs of SCS argue in favor of its cost-effectiveness for chronic neuropathic pain, especially in comparison to reoperation and medical management. A review of cost-effectiveness data implied that the largest reductions in health care expenditure come not only with consideration of SCS but also including it earlier as part of a comprehensive treatment paradigm.^[7]

Trialing

NACC recommended a multiday SCS trial for the treatment of pain to assess the therapy before committing to permanent implantation of an expensive and potentially more invasive device. Trialing is typically done with a pulse generator (current procedural terminology [CPT] code 63685) and two percutaneous leads (code 63650) or one paddle lead (code 63655).

The opportunity for sizable profit resulted in a large number of unnecessary trials, evidenced by the fact that only 30% of patients progressed to permanent SCS implant. Unnecessary trials have decreased significantly because the exorbitant profits no longer exist.

Clinician assessment of the trial outcome includes evaluations of pain relief, improvement in patient function, associated treatment (especially medication) use, and any complications of therapy. From the patient’s perspective, assessment includes acceptance and satisfaction with the outcomes of the treatment.^[2] NACC recommended that a successful trial be defined as the patient experiencing and recording at least 50% pain relief during the trial.

As of 2014, the Healthcare Common Procedure Coding System (HCPCS) code L8680 is no longer separately billable for Medicare (the payment for electrodes was incorporated in CPT code 63650). The change simplified the reimbursement process for trials but also had a significant impact on the practice of trialing. Previously, a physician could purchase the trial devices for about $1,000 and receive approximately $6,000 reimbursement for the trial. The opportunity for sizable profit resulted in a large number of unnecessary trials, evidenced by the fact that only 30% of patients progressed to permanent SCS implant. Unnecessary trials have decreased significantly because the exorbitant profits no longer exist. Current reimbursement rates are approximately $1,800 per trial, which is also intended to cover the cost of the trial devices.

Billing Codes

Diagnosis codes document the indication for the procedure. Pain codes from the G89 series are used as the principal diagnosis when the encounter is for pain control or pain management, rather than for management of the underlying condition. Neurostimulation therapy is directed at managing chronic, intractable pain rather than treating the underlying disorder. When a patient is admitted for insertion of a neurostimulator for pain control, the pain code is sequenced as the principal diagnosis.^[8] Additional codes may then be assigned to identify the underlying cause and to give more detail about the nature and location of the pain.

Tables 1 and 2 provide HCPCS II device codes for non-Medicare and Medicare billing, respectively. Table 3 provides CPT codes for physician payment.

Beyond reimbursement for the trial and permanent devices, programming codes are applicable for office visits (see Table 3). Programming codes cannot be used for the initial implant procedure and should be used only when the physician or a direct employee is performing the programming.

Table 1: Healthcare Common Procedure Coding System (HCPCS) II device codes^a (non-Medicare)^b

These codes are used by the entity that purchased and supplied the medical device, DME, drug, or supply to the patient. For implantable devices, that is generally the facility. It may also be the physician, most commonly for trial leads placed in the office. For specific Medicare hospital outpatient instructions for medical devices, see the Device C-codes (Medicare) below.
Lead^c	L8680	Implantable neurostimulator electrode, each
Pulse Generator^d	L8679	Implantable neurostimulator generator, any type
	L8686	Implantable neurostimulator pulse generator, single array, nonrechargeable, includes extension
	L8687	Implantable neurostimulator pulse generator, dual array, rechargeable, includes extension
	L8688	Implantable neurostimulator pulse generator, dual array, nonrechargeable, includes extension
External Recharger	L8689	External recharging system for battery (internal) for use with implantable neurostimulator, replacement only
Patient Programmer	L8681	Patient programmer (external) for use with implantable programmable neurostimulator pulse generator, replacement only

Abbreviation: DME, durable medical equipment.
a. Healthcare Common Procedure Coding System (HCPCS) Level II codes are maintained by the Centers for Medicare and Medicaid Services. http://www.cms.gov/Medicare/Coding/MedHCPCSGenInfo/index.html. Accessed November 21, 2017.
b. This table is reprinted with the permission of Medtronic, Inc.©
c. Physicians should not submit code L8680 to Medicare for leads placed in the office. This code is not separately billable to Medicare because the cost of the lead is already valued in the CPT procedure code. Centers for Medicare and Medicaid Services. MLN Matters Number MM8645. http://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/MLNMattersArticles/downloads/MM8645.pdf. Accessed November 21, 2017. Code L8680 remains available for use with non-Medicare payers, although physicians should check with the payer for specific coding and billing instructions. Likewise, hospitals and ambulatory surgical centers (ASCs) may be able to submit L8680 for non-Medicare payers but should check with the payer for instructions.
d. Effective January 2014, generator codes L8685–L8688 are not recognized by Medicare. Specifically, for billing Medicare, code L8679 is available for physician use, while hospitals typically use C-codes and ASCs generally do not submit HCPCS II codes for devices. For non-Medicare payers, codes L8685–L8688 remain available. However, all providers should check with the payer for specific coding and billing instructions.

Table 2: Device C-codes^a (Medicare)^b

Medicare provides C-codes for hospital use in billing Medicare for medical devices in the outpatient setting. Although other payers may also accept C-codes, regular HCPCS II device codes are generally used for billing non-Medicare payers. Unlike regular HCPCS II device codes, the extension is separately codable using C-codes. ASCs, however, usually should not assign or report HCPCS II device codes for devices on claims sent to Medicare. Medicare generally does not make a separate payment for devices in the ASC. Instead, payment is “packaged” into the payment for the ASC procedure. ASCs are specifically instructed not to bill HCPCS II device codes to Medicare for devices that are packaged.^c
Pulse Generator (nonrechargeable)	C1767	Generator, neurostimulator (implantable) nonrechargeable
Pulse Generator (rechargeable)^d	C1820	Generator, neurostimulator (implantable), with rechargeable battery and charging system
Extension	C1883	Adaptor/extension, pacing lead or neurostimulator lead (implantable)
Leads	C1778	Lead, neurostimulator (implantable)
Leads	C1897	Lead, neurostimulator, test kit (implantable)
Patient Programmer	C1787	Patient programmer, neurostimulator

Abbreviations: ASCs, ambulatory surgical centers; HCPCS, Healthcare Common Procedure Coding System.
a. Device C-codes are HCPCS Level II codes and also maintained by the Centers for Medicare and Medicaid Services. Healthcare Common Procedure Coding System. http://www.cms.gov/Medicare/Coding/HCPCSReleaseCodeSets/Alpha-Numeric-HCPCS.html. Accessed November 21, 2017.
b. This table is reprinted with the permission of Medtronic, Inc.©
c. ASCs should report all charges incurred. However, only charges for nonpackaged items should be billed as separate line items. For example, the ASC should report its charge for the generator. However, because the generator is a packaged item, the charge should not be reported on its own line. Instead, the ASC should bill a single line for the implantation procedure with a single total charge, including not only the charge associated with the operating room but also the charges for the generator device and all other packaged items. Because of a Medicare requirement to pay the lesser of the ASC rate or the line-item charge, breaking these packaged charges out onto their own lines can result in incorrect payment to the ASC. Centers for Medicare and Medicaid Services. Medicare Claims Processing Manual, Chapter 14—Ambulatory Surgical Centers, Section 40. http://www.cms.gov/Regulations-and-Guidance/Guidance/Manuals/downloads/clm104c14.pdf. Accessed November 21, 2017. See also MLN Matters SE0742 pp. 9–10: Centers for Medicare and Medicaid Services. MLN Matters Number SE0742 Revised. http://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/MLNMattersArticles/downloads/SE0742.pdf. Accessed November 21, 2017.
d. HCPCS C-code C1822, Generator, neurostimulator (implantable), high frequency, with rechargeable battery and charging system, is also used for certain types of spinal neurostimulators. However, this code does not represent Medtronic spinal cord stimulation generators.

Table 3: Current procedural terminology (CPT) procedure codes^a

Physicians use CPT codes for all services. Under Medicare’s Resource-Based Relative Value Scale (RBRVS) methodology for physician payment, each CPT code is assigned a point value, known as the relative value unit (RVU), which is then converted to a flat payment amount.
Procedure	CPT code and description^b	Medical RVUs^c		Medicare National Average^d
		For physician services provided in:^e
		Physician Office	Facility	Physician Office	Facility
Screening Test^g,h,i	63650 Percutaneous implantation of neurostimulator electrode array, epidural^j,k	37.59	11.83	$1,353	$426
Lead Implantation^g,h,i	63650 Percutaneous implantation of neurostimulator electrode array, epidural^j,k	37.59	11.83	$1,353	$426
Lead Implantation^g,h,i	63655 Laminectomy for implantation of neurostimulator electrodes, plate/paddle, epidural	N/A	24.07	N/A	$867
Generator Implantation or Replacement^h,l	63685 Insertion or replacement of spinal neurostimulator pulse generator or receiver, direct or inductive coupling	N/A	10.47	N/A	$377
Removal of Leads^h,m,n,o	63661 Removal of spinal neurostimulator electrode percutaneous array(s), including fluoroscopy, when performed	16.73	9.33	$602	$336
Removal of Leads^h,m,n,o	63662 Removal of spinal neurostimulator electrode plate/paddle(s) placed via laminotomy or laminectomy, including fluoroscopy, when performed	N/A	24.33	N/A	$876
Revision or Replacement of Leads^h,n,o	63663 Revision including replacement, when performed, of spinal neurostimulator electrode percutaneous array(s), including fluoroscopy when performed	22.49	12.98	$810	$467
Revision or Replacement of Leads^h,n,o	63664 Revision including replacement, when performed, of spinal neurostimulator electrode plate/paddle(s) placed via laminotomy or laminectomy, including fluoroscopy when performed	N/A	25.33	N/A	$912
Revision or Removal of Generator^h,l	63688 Revision or removal of implanted spinal neurostimulator pulse generator or receiver	N/A	10.76	N/A	$387
Analysis/Programming Note: In the office, analysis and programming may be furnished by a physician, practitioner with an “incident to” benefit, or auxiliary personnel under the direct supervision of the physician (or other practitioner), with or without support from a manufacturer’s representative. The patient or payer should not be billed for services rendered by the manufacturer’s representative. Contact your local contractor or payer for interpretation of applicable policies.	95970 Electronic analysis of implanted neurostimulator pulse generator system (eg, rate, pulse amplitude, pulse duration, configuration of wave form, battery status, electrode selectability, output modulation, cycling, impedance, and patient compliance measurements); simple or complex brain, spinal cord, or peripheral (ie, cranial nerve, peripheral nerve, sacral nerve, neuromuscular) neurostimulator pulse generator/transmitter, without reprogramming	1.97	0.69	$71	$25
	95971 Electronic analysis of implanted neurostimulator pulse generator system (eg, rate, pulse amplitude, pulse duration, configuration of wave form, battery status, electrode selectability, output modulation, cycling, impedance, and patient compliance measurements); simple spinal cord or peripheral (ie, peripheral nerve, sacral nerve, neuromuscular) neurostimulator pulse generator/transmitter, with intraoperative or subsequent programming^p	1.45	1.17	$52	$42
	95972 Electronic analysis of implanted neurostimulator pulse generator system (eg, rate, pulse amplitude, pulse duration, configuration of wave form, battery status, electrode selectability, output modulation, cycling, impedance, and patient compliance measurements); complex spinal cord or peripheral (ie, peripheral nerve, sacral nerve, neuromuscular) (except cranial nerve) neurostimulator pulse generator/transmitter, with intraoperative or subsequent programming^p	1.67	1.19	$60	$43

a. This table is reprinted with the permission of Medtronic, Inc.©
b. CPT copyright 2017 American Medical Association. All rights reserved. CPT® is a registered trademark of the American Medical Association (AMA). Applicable FARS/DFARS. Restrictions Apply to Government Use. Fee schedules, relative value units, conversion factors, and/or related components are not assigned by the AMA, are not part of CPT, and the AMA is not recommending their use. The AMA does not directly or indirectly practice medicine or dispense medical services. The AMA assumes no liability for data contained or not contained herein.
c. Centers for Medicare & Medicaid Services. Medicare Program; Payment Policies Under the Physician Fee Schedule and Other Revisions to Part B for CY 2018 Final Rule; 82 Fed. Reg. 52976-53371. https://www.gpo.gov/fdsys/pkg/FR-2017-11-15/pdf/2017-23953.pdf Published November 15, 2017. Accessed November 21, 2017. The total RVU as shown here is the sum of three components: physician work RVU, practice expense RVU, and malpractice RVU.
d. Medicare national average payment is determined by multiplying the sum of the three RVUs by the conversion factor. The conversion factor for CY 2018 is $35.9996 per 82 Fed. Reg. 53344. https://www.gpo.gov/fdsys/pkg/FR-2017-11-15/pdf/2017-23953. Published November 15, 2017. Accessed November 21, 2017. See also the January 2018 release of the PFS Relative Value File RVU18A at http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeeSched/PFS-Relative-Value-Files.html. Released November 15, 2017. Accessed November 21, 2017. Final payment to the physician is adjusted by the Geographic Practice Cost Indices (GPCI). Also note that any applicable coinsurance, deductible, and other amounts that are patient obligations are included in the payment amount shown.
e. The RVUs shown are for the physician’s services, and payment is made to the physician. However, there are different RVUs and payments depending on the setting in which the physician rendered the service. “Facility” includes physician services rendered in hospitals, ambulatory surgical centers, and skilled nursing facilities. Physician RVUs and payments are generally lower in the “Facility” setting because the facility is incurring the cost of some of the supplies and other materials. Physician RVUs and payments are generally higher in the “Physician Office” setting because the physician incurs all costs there.
f. “N/A” shown in Physician Office setting indicates that Medicare has not developed RVUs in the office setting because the service is typically performed in a facility (eg, in a hospital). However, if the local contractor determines that it will cover the service in the office, then it is paid using the facility RVUs at the facility rate. Centers for Medicare & Medicaid Services. Details for Title: CMS-1676-F. CY 2018 PFS Final Rule Addenda. Addendum A: Explanation of Addendum B and C. https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeeSched/PFS-Federal-Regulation-Notices-Items/CMS-1676-F.html. Released November 6, 2017. Accessed November 21, 2017.
g. As defined and as published by the AMA (CPT Assistant, June 1998, p. 4), these codes represent a single lead. When more than one lead is placed, each is coded separately. However, Medicare does not permit the use of bilateral modifier –50 or –LT/ –RT on these codes. Some payers recognize that each code represents a distinct lead when modifier –51 or modifier –59 is appended to the additional codes. Note that Medicare’s Medically Unlikely Edits allow 2 units for code 63650 on the same date of service but only 1 unit for code 63655. Denials for units in excess of the MUE values may be appealed.
h. Surgical procedures are subject to a “global period.” The global period defines other physician services that are generally considered part of the surgery package. The services are not separately coded, billed, or paid when rendered by the physician who performed the surgery. These services include preoperative visits the day before or the day of the surgery, postoperative visits related to recovery from the surgery for 10 or 90 days depending on the specific procedure, treatment of complications unless they require a return visit to the operating room, and minor postoperative services such as dressing changes and suture removal.
i. The published vignettes for codes 63650 and 63655 include fluoroscopy, and, according to guidelines published by the American Association of Neurological Surgeons (2017 AANS Guide to Coding, 2016 Edition, p. 68), its use is inherent to lead implantation and should not be coded separately. In addition, National Correct Coding Initiative (NCCI) edits prohibit coding fluoroscopy separately with 63650 and 63655. See also CPT Assistant, January 2016, p. 12.
j. The Physician Office RVUs for code 63650 are valued to include payment for the lead and other practice expenses associated with office-based lead insertion, eg, trials. HCPCS code L8680 should not be reported separately for the lead in conjunction with office-based lead insertion.
k. The AMA has published (CPT Assistant, October 2013, p. 19) that the use of an incision to admit the needle or to anchor the lead is inherent to percutaneous placement and does not alter the use of code 63650. See also 2017 AANS Guide to Coding, p. 68.
l. When an existing generator is removed and replaced by a new generator, only the generator replacement code 63685 may be assigned. NCCI edits do not allow removal of the existing generator to be coded separately. Also note that, according to NCCI policy, use of the CPT code for generator “insertion or replacement” requires placement of a new generator. When the same generator is removed and then reinserted, the “revision” code is used (NCCI Policy Manual 1/1/2018, p. VIII-8).
m. The AMA has published that the work of removing a temporary trial lead is inherent to the original percutaneous placement code 63650 and is not coded separately. Code 63661 cannot be assigned for removal of a temporary trial lead that was placed percutaneously. Further, codes 63661 and 63662 apply to surgical removal of permanent leads. Removal of a permanent lead by simple pull is not coded (CPT Assistant, August 2010, p. 8,15; April 2011, pp. 10–11, 15).
n. The AMA has published that replacement codes 63663 and 63664 are assigned when a permanent lead is replaced by another permanent lead of the same type via the same approach at the same spinal level. The work of removing the existing permanent lead is included and is not coded separately (CPT Assistant, August 2010, p. 8,15; April 2011, pp. 10–11, 15). In addition, NCCI edits do not permit removal codes 63661 and 63662 to be assigned separately with replacement codes 63663 and 63664.
o. The AMA has published that when a permanent percutaneous lead is removed and a new lead is placed via a fresh laminectomy at the same or a different spinal level, insertion code 63655 is assigned with removal code 63661 (CPT Assistant, April 2011, pp. 11, 15). NCCI edits allow this combination without use of a modifier.
p. According to CPT manual instructions, “simple” programming involves changes to three or fewer parameters and “complex” programming involves changes to four or more parameters. The parameters that qualify are rate, pulse amplitude, pulse duration, pulse frequency, eight or more electrode contacts, cycling, stimulation train duration, train spacing, number of programs, number of channels, alternating electrode polarities, dose time (stimulation parameters changing in time periods of minutes including dose lockout times), assessing more than one clinical feature. (See also CPT Assistant, July 2016, p. 7 and p. 9.)

Recent Technologic Advancements

Technologic advancements such as novel waveforms, higher-stimulation frequencies, and new anatomical targets have vastly expanded the field of SCS, resulting in greater efficacy and broader applicability.^[5]

In 2015, the SEZNA-RCT trial demonstrated that the HF10 waveform was significantly better than traditional SCS in terms of the proportion of responders (84.5% of subjects were HF10 responders for back pain and 83.1% were HF10 responders for leg pain, compared with 43.8% of subjects for back pain and 55.5% for traditional SCS for leg pain).^[9] In the field of pain management, clinicians generally appreciate a response rate of more than 40% as better than most available therapies; the fact that a new waveform essentially doubles the response rate is extremely encouraging for the potential of this relatively novel therapy.

In 2017, the SUNBURST study showed that burst stimulation is superior over tonic stimulation and preferred by significantly more patients.^[10] Interestingly, 69% of subjects responded to tonic stimulation, burst stimulation, or both. For the individual stimulation modes, 60% of subjects were responders to burst stimulation and 51% to tonic stimulation.^[10] This is another example of the incremental effect that the development of new stimulation parameters can have on therapy response and patient satisfaction rates.

In addition to improving response rates, the advancements can help physicians to optimize stimulation settings for patients earlier in the therapy timeline, which may ensure that the therapy continues successfully beyond the cost-effectiveness break-even point. That notion further underscores the importance of ongoing development of stimulation parameters that will continue to advance the therapy, making its future brighter with even better outcomes.

References

Staats P. Trial design and endpoint evaluation in clinical studies addressing chronic back pain. Spine 2017;42:S93–S97.
Deer TR, Mekhail N, Provenzano D, et al. The appropriate use of neurostimulation of the spinal cord and peripheral nervous system for the treatment of chronic pain and ischemic diseases: the Neuromodulation Appropriateness Consensus Committee. Neuromodulation 2014;17:515–550.
Deer TR, Krames E, Mekhail N, et al. The appropriate use of neurostimulation: new and evolving neurostimulation therapies and applicable treatment for chronic pain and selected disease states. Neuromodulation 2014;17:599–615.
Provenzano DA, Amirdelfan K, Kapural L, et al. Evidence gaps in the use of spinal cord stimulation for treating chronic spine conditions. Spine 2017;42:S80–S92.
Kapural L, Peterson E, Provenzano DA, et al. Clinical evidence for spinal cord stimulation for failed back surgery syndrome (FBSS). Spine 2017;42:S61–S66.
Kumar K, Hunter G, Demeria D. Spinal cord stimulation in treatment of chronic benign pain: challenges in treatment planning and present status, a 22-year experience. Neurosurgery 2006;58:481–496.
Hoelscher C, Riley J, Wu C, et al. Cost-effectiveness data regarding spinal cord stimulation for low back pain. Spine 2017;42:S80–S92.
Centers for Disease Control and Prevention. ICD-10-CM official guidelines for coding and reporting FY 2018, I.C.6.b.1 (a). Available at: https://www.cdc.gov/nchs/data/icd/10cmguidelines_fy2018_final.pdf. Accessed March 29, 2019.
Kapural L, Yu C, Doust MW, et al. Novel 10-kHz high-frequency therapy (HF10 therapy) is superior to traditional low-frequency spinal cord stimulation for the treatment of chronic back and leg pain: the SENZA-RCT randomized controlled trial. Anesthesiology 2015;123:851–860.
Deer T, Slavin KV, Amirdelfan K, et al. Success using neuromodulation with BURST (SUNBURST) study: results from a prospective, randomized controlled trial using a novel burst waveform. Neuromodulation 2018;21:56–66.

ASRA Pain Medicine Update