GoHealthcare Knowledge Center | MSK Specialty Procedure Library™

Developed by GoHealthcare Practice Solutions

Cervical Interlaminar Epidural Steroid Injection

Prior Authorization, Medical Necessity, Clinical Documentation, Payer Policy, and Coding Fundamentals

This guide is part of the GoHealthcare MSK Specialty Procedure Library™, the operational and educational reference of the GoHealthcare Practice Solutions Knowledge Center. It bridges clinical evidence with real-world payer requirements and operational execution for cervical interlaminar epidural steroid injection (CILESI), reported under CPT® 62321 when performed with fluoroscopic or computed tomography guidance.

Explore the Procedure GuideContact GoHealthcare

Important: This document is an operational, documentation, coverage, and coding reference. It is not a procedural technique manual and does not replace physician procedural training, specialty society procedural manuals, clinical judgment, payer policy review, legal advice, or official CMS guidance. Coverage policy is perishable. Verify the live policy, coding requirements, authorization criteria, and place-of-service rules before every submission and service.

This guide is written for physicians, advanced practice providers, practice administrators, prior authorization specialists, utilization management staff, clinical nurses, clinical documentation specialists, revenue cycle leaders, ambulatory surgery centers, hospital outpatient departments, case managers, workers' compensation professionals, attorneys, and healthcare executives.

Standard	GoHealthcare Clinical Procedure Guide Standard v1.0
Library	GoHealthcare MSK Specialty Procedure Library™ — Batch B02, Epidural Steroid Injections
Version	Version 1.1 — July 2026
Primary CPT®	62321 — interlaminar epidural, cervical or thoracic, with imaging guidance
Developed by	Pinky Maniri, Founder and Chief Executive Officer, GoHealthcare Practice Solutions

98%Company-reported prior authorization approval rate

50 StatesNational prior authorization support for specialty practices

MSK FocusPain management, orthopedic, spine, neurosurgery, and ASC workflows

GoHealthcare performance statistics are company-reported operational metrics and are not a guarantee of approval, reimbursement, turnaround time, or payer outcome. Results depend on payer rules, clinical documentation, eligibility, benefit design, and case-specific medical necessity.

Procedure Guide Navigation

Explore the Complete Guide

Use the grouped links below to move directly to each approved section.

Clinical Foundations

1. Overview
2. Evidence at a Glance
3. Relevant Anatomy (High-Level Overview Only)
4. Clinical Indications
5. Patient Selection
6. Contraindications

Coverage and Authorization

7. Conservative Treatment Requirements
8. Imaging Requirements
9. Medical Necessity
10. Medicare Coverage
11. Commercial Payer Comparison
12. Prior Authorization Workflow, Denials, and Appeals

Coding, Documentation, and Operations

13. Coding and Billing Overview
14. Clinical Documentation Checklist
15. Procedure Comparison
16. GoHealthcare Clinical Insights
17. GoHealthcare Prior Authorization Insight
18. GoHealthcare Leadership Perspective

Optimization, FAQs, and Outlook

19. GoHealthcare Case Study
20. GoHealthcare Best Practices
21. Common Mistakes
22. Pearls and Pitfalls
23. Frequently Asked Questions
24. Key Takeaways, Future Outlook, and References

01

Clinical Foundations

Overview

Cervical interlaminar epidural steroid injection is a fluoroscopically or CT-guided procedure in which contrast is injected to confirm epidural placement, followed by a corticosteroid, with or without local anesthetic, delivered into the posterior cervical epidural space through a midline or paramedian interlaminar approach. Its purpose is to reduce inflammation around an irritated or compressed cervical nerve root in order to relieve radicular pain, restore function, and support participation in active rehabilitation. It is a component of a comprehensive care plan, not a standalone cure and not a substitute for rehabilitation or surgical evaluation when either is indicated.

From an operations standpoint, CILESI sits in an unusually tight regulatory corridor. Medicare's epidural steroid injection Local Coverage Determinations, which are aligned across the Medicare Administrative Contractors, permit a maximum of one level per session for interlaminar injections, prohibit bilateral interlaminar injections, prohibit treating more than one spinal region per session, and cap the cervical/thoracic region at four sessions in a rolling twelve months. Commercial payers and the utilization management organizations that administer their musculoskeletal programs impose substantially the same architecture with meaningful differences in the details, particularly around how long advanced imaging remains valid and how much relief and for how long is required before a repeat injection is authorized.

The commercially decisive fact about this procedure is that its denials are overwhelmingly administrative rather than clinical. In GoHealthcare's experience supporting musculoskeletal specialty practices, the great majority of CILESI denials are not payers disagreeing that the patient needs the injection. They are payers unable to locate, inside the submitted record, four discrete data elements: a baseline pain and function score on a named scale, a dated and specific conservative care history, an advanced imaging report that is both current under that payer's clock and concordant with the documented examination, and, for repeats, a percentage of relief with a duration

measured on the same scale used at baseline. Practices that engineer those four elements into the encounter rather than reconstructing them at appeal convert most of their denial volume into first-pass approvals.

The second decisive fact is regional. For epidural purposes, cervical and thoracic are a single anatomic spinal region under both Medicare and Carelon. A patient's thoracic interlaminar injection in March and cervical interlaminar injection in July draw from the same annual pool of four sessions. Practices that track utilization by level rather than by region discover the problem only when the fourth or fifth request is denied and the twelve-month clock has months left to run.

WHAT THIS GUIDE ANSWERS

1. What is this procedure clinically, and which patients are appropriate for it?
2. How does a practice consistently obtain insurance approval while minimizing denials and avoiding unnecessary peer-to-peer reviews?

Every section below is written to serve one or both of those questions. Sections 1 through 9 establish clinical appropriateness and the documentation that demonstrates it. Sections 10 through 12 establish coverage and the authorization pathway. Sections 13 through 22 translate both into daily operational execution.

Back to procedure guide navigation ↑

02

Clinical Foundations

Evidence at a Glance

This one-page executive summary is the orientation table for the entire guide. Each row is expanded and sourced in the sections that follow. It is written to be lifted into a payer packet cover sheet, a new hire onboarding binder, or a utilization management huddle.

Domain	Summary for Cervical Interlaminar ESI (CPT® 62321)
Medicare NCD	None. There is no National Coverage Determination governing epidural steroid injections. Coverage is determined entirely at the Medicare Administrative Contractor level through aligned Local Coverage Determinations.
Medicare LCD	A family of substantively aligned MAC LCDs titled Epidural Steroid Injections for Pain Management, including L39240 (Noridian, revision effective 04/09/2026), L39054, L39242, L39036, L38994, and L33906. Covered for cervical radiculopathy or radicular pain from disc herniation, osteophyte or osteophyte complex, severe degenerative disc disease producing foraminal or central stenosis; post-laminectomy syndrome; and acute herpes zoster associated pain. Not covered for axial neck pain, cervicogenic headache, or non-specific pain.
Commercial coverage	Broadly covered for cervical radicular pain and radiculopathy with concordant advanced imaging and documented failure of conservative care. Uniformly excluded for isolated axial neck pain. Administered for most national plans by a utilization management organization rather than the plan itself: eviCore (Cigna, some Aetna and Blue plans), Carelon (Elevance/Anthem and many Blue plans), Cohere, Evolent, or TurningPoint depending on plan and market.
Evidence level	ASIPP 2021 comprehensive evidence-based guidelines: Level I evidence for cervical interlaminar epidural injections in disc herniation with a strong recommendation for long-term effectiveness; Level II for cervical central spinal stenosis and for cervical post-surgery syndrome, each with a moderate to strong recommendation. AAN 2025 systematic review summary: epidural steroids probably provide short-term pain reduction and probably reduce short-term disability in cervical radiculopathy (moderate confidence), superseding the 2007 AAN position that cervical evidence was inadequate for any recommendation (Level U).
Image guidance	Required. Fluoroscopy or CT with contrast to confirm epidural placement. Medicare permits ultrasound without contrast only for documented low molecular weight nonionic contrast allergy or pregnancy. eviCore and UnitedHealthcare consider ultrasound-guided ESI not medically necessary or unproven without exception. Non-image-guided injection (CPT® 62320) is not reasonable and necessary.
Prior authorization	Required by essentially all commercial plans and Medicare Advantage plans. Not required by traditional fee-for-service Medicare, which enforces the same criteria retrospectively through post-payment data analysis and medical review audit rather than prospectively.
Conservative therapy	At least four weeks of documented, failed non-invasive conservative care is the common denominator across Medicare, eviCore, UnitedHealthcare, and Aetna. Carelon requires physical therapy plus at least one complementary modality and permits a reduction to two weeks where there is clear objective evidence of radiculopathy. Medicare additionally requires at least four weeks of pain duration and waives the four-week wait for acute herpes zoster refractory to conservative care.
Diagnostic block	Not applicable. A diagnostic selective nerve root block is a distinct transforaminal procedure with its own criteria and is not a prerequisite to interlaminar ESI. No diagnostic block is required before CILESI by any payer reviewed.
Key documentation	Baseline pain score and functional assessment on a named validated scale, repeated on the same scale at every follow-up; dated conservative care history with modality, duration, and outcome; MRI or CT report concordant with the examination and current under the payer's clock; a level-specific and side-specific medical necessity statement; documentation of active rehabilitation participation; and, for repeats, percentage relief with duration on the same scale.
Typical approval	Cervical radicular pain in a named dermatome, concordant MRI within the payer's window showing nerve root compression or foraminal stenosis at the level to be injected, four or more weeks of documented failed conservative care, baseline scores present, single level, single session, fluoroscopy with contrast, and an active rehabilitation plan. Repeat approval turns almost entirely on documented percentage relief and its duration.

THE FOUR-ELEMENT RULE

If a CILESI submission contains a named-scale baseline score, a dated conservative care history, a concordant and in-window imaging report, and — for repeats — a percentage of relief with a duration on the same scale, it will satisfy the great majority of payer criteria in the United States for this procedure. Nearly everything in Sections 10 through 22 of this guide is an elaboration of how to make those four elements appear reliably and without heroics.

Back to procedure guide navigation ↑

03

Clinical Foundations

Relevant Anatomy (High-Level Overview Only)

Anatomy is presented here only to the extent required to understand patient selection, medical necessity, coverage logic, and documentation. This section is deliberately not a procedural or technique reference.

The cervical epidural space lies outside the dural membrane within the spinal canal. It runs the length of the spine and contains fatty tissue, blood vessels, and the exiting nerve roots. In the cervical spine, eight pairs of nerve roots (C1 through C8) exit through the neuroforamina. Because each cervical root exits above its correspondingly numbered vertebra through C7, and C8 exits between C7 and T1, a given disc level maps to a specific root: the C5-C6 disc affects the C6 root, C6-C7 affects C7, and C7-T1 affects C8. This mapping is why payers require the injected level, the compressed root, and the symptomatic dermatome to be documented and to agree with one another. When they do not agree, the reviewer cannot establish concordance and the request is denied.

Cervical nerve roots can become a significant and disabling source of radicular pain through several mechanisms recognized in the Medicare LCD: compression from herniation of the nucleus pulposus, degenerative changes involving the spinal ligaments, discs, and zygapophyseal joints, intraspinal synovial cysts, osteophytes and osteophyte complexes, and mechanical derangement such as spondylolisthesis. Inflammation and chemical irritation of the root may produce radicular symptoms even where static imaging is unimpressive, a point the Medicare Contractor Advisory Committee subject matter experts raised directly when they expressed only low confidence that radicular pain must be concordant with a radiologist's finding of compression, noting that imaging is a static picture of a dynamic process and that non-anatomic sources such as chemical radiculitis are invisible to it. That expert reservation did not change the operational reality: commercial criteria still demand concordant imaging, and submissions without it fail.

Two anatomic facts drive the safety profile and therefore the coverage architecture. First, the epidural space narrows as it ascends, so the margin between needle and cord is smallest in the upper and mid-cervical spine; the evidence reviewed in the Medicare LCDs and in the FDA Safe Use Initiative consensus work notes that the interlaminar technique carries a risk of direct spinal cord trauma and that many authors and recommendations conclude cervical injections should be restricted to the lower cervical levels. Second, the radicular arteries that perfuse the cord course adjacent to the targeted root within the neuroforamen, which is the anatomic basis for the far higher catastrophic complication rate of cervical transforaminal injection compared with the interlaminar approach. Published safety analyses cited in the LCDs report that cervical and thoracic transforaminal epidurals constitute roughly 2.4 percent of all epidural injections and fewer than 5 percent of all transforaminal injections, yet contribute more than 99 percent of complications related to intra-arterial injection of particulate steroid. This asymmetry is the single most important reason cervical interlaminar ESI is the dominant cervical epidural approach in the United States and is the reason its evidence base and payer acceptance are stronger than that of cervical TFESI.

ANATOMY THAT MATTERS TO AN AUTHORIZATION REVIEWER

A reviewer reads anatomy in exactly one way: does the level on the request match the root on the imaging match the dermatome in the exam? Three agreeing data points approve. Two agreeing and one silent generates a peer-to-peer. Two agreeing and one contradicting denies.

The most common failure is silence, not contradiction. The MRI says C6-C7 foraminal stenosis, the exam says 'numbness in the arm,' and the request says C7-T1. Nothing there is false; nothing there is concordant either.

Back to procedure guide navigation ↑

04

Clinical Foundations

Clinical Indications

The covered indication set for CILESI is narrow, well defined, and remarkably consistent across Medicare and commercial payers. Understanding the boundary is more operationally valuable than understanding the center, because nearly all indication-based denials occur at the edges.

4.1 Covered indications

Indication	Coverage posture and what must be documented
Cervical radiculopathy or radicular pain due to disc herniation	The strongest indication across every payer reviewed. ASIPP 2021 assigns Level I evidence with a strong recommendation for long-term effectiveness of cervical interlaminar epidural injections in disc herniation. Document the named root, the concordant dermatomal distribution, the examination findings, and the MRI or CT correlation at the level to be injected.
Cervical radiculopathy or radicular pain due to osteophyte, osteophyte complex, or severe degenerative disc disease producing foraminal or central stenosis	Explicitly covered by the Medicare LCD family. ASIPP assigns Level II evidence with a moderate to strong recommendation for cervical interlaminar injections in central spinal stenosis. Carelon accepts spinal stenosis (central, lateral recess, foraminal, or extraforaminal) as a qualifying imaging finding. Note that for a fixed stenotic lesion not due to disc herniation, Carelon does not require repeat imaging for a repeat request — a meaningful and underused operational concession.
Post-laminectomy syndrome (cervical post-surgery syndrome)	Covered under the Medicare LCD. ASIPP 2021 assigns Level II evidence with a moderate to strong recommendation for cervical interlaminar epidural injections in post-surgery syndrome. Document the prior surgery, the date, the recurrent or persistent radicular pattern, and post-operative imaging. Expect closer review; this is a common peer-to-peer trigger.
Acute herpes zoster associated pain	Covered under the Medicare LCD, which uniquely waives the four-week conservative care wait where acute zoster pain is refractory to conservative management. This is a genuine Medicare-versus-commercial divergence. eviCore expressly lists ESI for post-herpetic neuralgia as not medically necessary. Do not assume a Medicare-covered zoster indication will clear a commercial utilization management review.
Cervicobrachialgia	Neck pain radiating to the arm from cervical nerve root compression. The Medicare subject matter expert panel voted 4 out of 5 that evidence supports ESI for cervicobrachialgia, noting it is caused by disc herniation and spinal stenosis, where ESIs are beneficial. Operationally, code and document it as the underlying radiculopathy, not as a standalone symptom label.

4.2 Non-covered and investigational indications

The following are the boundary conditions. Submitting for any of these, however well-supported by specialty society literature, produces a denial and consumes staff time that would be better spent elsewhere.

• Axial neck pain without radiculopathy. Excluded by the Medicare LCD as investigational. Carelon lists isolated axial neck pain among its not-medically-necessary indications. eviCore lists axial spinal pain, defined as the absence of radiculopathy, myelopathy, or myeloradiculopathy, as not medically necessary. UnitedHealthcare's coverage rationale is written entirely around Radicular Back Pain and states that there is insufficient evidence for ESI in back pain absent radicular symptoms. This is the single most consequential gap between specialty society evidence and payer policy in cervical epidural practice — see the Prior Authorization Insight in Section 17.
• Cervicogenic headache. Listed as investigational and not medically reasonable and necessary in the Medicare LCD, notwithstanding that the LCD's own evidence review acknowledges many interventional pain specialists consider epidural injection a reasonable option after conservative failure and that the expert panel voted 3 of 5 that evidence supports it. The policy language governs, not the evidence review.
• Post-herpetic neuralgia. Not medically necessary per eviCore. The Medicare position is more nuanced — the covered indication is acute herpes zoster associated pain, while the LCD's expert panel voted only 2 of 5 confidence for PHN — and the billing and coding article's covered diagnosis list does include B02.23 postherpetic polyneuropathy. Treat this as a payer-specific determination and verify before scheduling.
• Complex regional pain syndrome, widespread diffuse pain, and neuropathy from other causes. Investigational and not medically reasonable and necessary under the Medicare LCD.
• Injections performed with biologics. There are no FDA-approved biologicals for injection into the epidural space. Amniotic and placenta-derived injectates, platelet-rich plasma, stem cells, and vitamins fall in this category. Inclusion of any of these in the injectate may result in denial of the entire claim, not merely the biologic component.
• Non-image-guided injection. Not reasonable and necessary except for documented contraindication to contrast media.
• Moderate or deep sedation, general anesthesia, and monitored anesthesia care. Usually unnecessary or rarely indicated, therefore not routinely reimbursable and frequently denied. Even for needle phobia and anxiety, the LCD states oral anxiolytics typically suffice.

Back to procedure guide navigation ↑

05

Clinical Foundations

Patient Selection

Patient selection for CILESI is the point at which clinical judgment and payer criteria either align or diverge, and it is far cheaper to discover divergence in the clinic than at the authorization desk. The selection framework below is constructed so that a patient who passes it will, in the ordinary case, also clear utilization management.

5.1 The appropriate candidate

• Radicular symptoms in a level-specific referral pattern. Pain, dysesthesia, or paresthesia reported in the distribution of an involved named cervical root — not diffuse neck and shoulder pain, and not 'arm pain' without a pattern.
• Significant functional limitation. Medicare requires that the radicular pain be severe enough to greatly impact quality of life or function. Carelon operationalizes this with a numeric threshold: pain at least 3 out of 10 in intensity associated with inability to perform at least two activities of daily living or instrumental activities of daily living. eviCore requires significant functional limitations producing diminished quality of life and impaired age-appropriate activities of daily living.
• Symptom duration of at least four weeks, per the Medicare LCD, with the acute zoster exception.
• Objective corroboration. Either detailed neurologic examination findings concordant with root compression — loss of strength in a named muscle or myotome, altered sensation to light touch, pressure, pin prick, or temperature in the sensory distribution, or diminished, absent, or asymmetric reflexes — documented within the prior three months, or a concordant advanced imaging or electrodiagnostic study. eviCore accepts either; the safest submission contains both.
• Failed conservative care of at least four weeks (see Section 7).
• Participation in active rehabilitation. Medicare states patients should be part of an active rehabilitation program, home exercise program, or functional restoration program. eviCore requires participation in a comprehensive pain management program that includes physical therapy, patient education, psychosocial support, and oral medications — all four elements.
• A realistic therapeutic goal. The evidence supports short-term to intermediate-term relief that creates a window for rehabilitation. It does not support a predetermined series, and it does not reliably support long-term relief or surgery avoidance. Setting that expectation at consent prevents both patient disappointment and the pressure to over-inject.

5.2 The inappropriate candidate

• Isolated axial neck pain without a radicular pattern.
• Progressive neurologic deficit, myelopathy, or any red flag requiring surgical rather than interventional evaluation. Carelon excludes moderate-to-severe myelopathy on clinical exam and myelopathy associated with intramedullary cord signal change on T1- or T2-weighted MRI. UnitedHealthcare lists rapidly progressing neurological deficit as a contraindication.
• A patient whose imaging and examination cannot be reconciled to a single level and side. This patient may still need care, but CILESI is not the next step and the request will not authorize.
• A patient who has already exhausted the four-session annual allowance in the cervical/thoracic region.
• A patient with a fixed expectation of cure, or who is seeking a scheduled series.

5.3 Special populations

Population	Selection and documentation considerations
Diabetes	Epidural corticosteroid produces dose-dependent hyperglycemia that may persist for two to six days after injection. Carelon lists uncontrolled diabetes among its contraindications. Document glycemic control, the patient's counseling on expected transient elevation, and the monitoring plan.
Postmenopausal women, osteoporosis, and osteopenia	The evidence reviewed in the Medicare LCD includes a large retrospective cohort reporting a roughly 21 percent increase in fracture risk per injection, and WIP Benelux guidance recommending corticosteroid exposure be kept to a minimum in high-risk individuals. Document the risk-benefit discussion. This is also why cumulative steroid exposure documentation matters (Section 9.3).
Patients on anticoagulant or antiplatelet therapy	Genuinely unsettled. ASRA's 2018 systematic review found interlaminar procedures carried greater hemorrhagic risk regardless of whether anticoagulants were ceased or continued, while ASIPP found good evidence of thromboembolic risk when antithrombotic therapy is discontinued. A 2012 survey found bleeding complications were similar whether or not anticoagulation was held, but thromboembolic events were seventeen times more frequent when it was held. The Medicare expert panel registered only low confidence (2 of 5) that evidence supports continuation. Document the shared decision, the consultation with the prescriber, and the rationale — not a protocol.
Pregnancy	Medicare permits ultrasound guidance without contrast where the patient is pregnant. Note this is a Medicare accommodation; eviCore and UnitedHealthcare do not extend the same latitude for ESI. Verify before scheduling.
Contrast allergy	Medicare permits ultrasound guidance without contrast for documented allergy to low molecular weight nonionic contrast. The allergy must be documented in the record, not asserted on the authorization form.
Immunocompromised patients and vaccination timing	Dose-dependent immune suppression after ESI is reported. A retrospective study found corticosteroid reduced vaccine effectiveness with increased influenza risk after major joint corticosteroid injection (relative risk 1.52). Coordinate timing; document the coordination.

Back to procedure guide navigation ↑

06

Clinical Foundations

Contraindications

Contraindications are listed here because they are coverage criteria as well as safety criteria. Payers screen for them, and a record that does not affirmatively address the relevant ones invites a peer-to-peer. The list below consolidates the Medicare LCD, the Carelon guideline, and the UnitedHealthcare policy.

6.1 Conditions requiring evaluation before any consideration of CILESI

• New onset of neck pain in the setting of established malignancy, or where malignancy is suspected on clinical presentation. The Medicare LCD independently lists suspicion or major risk factors for cancer as a contraindication.
• New onset of neck pain in a patient with risk factors for spinal infection or osteoporotic fracture.
• Comorbid conditions associated with increased bleeding risk from coagulopathy or anticoagulant therapy.
• Neck pain in the setting of acute trauma.

6.2 Known or suspected conditions that contraindicate the procedure

• Suspected or active localized spinal infection; significant systemic infection; local infection at the injection site.
• Compressive lesions of the spinal cord, conus medullaris, or cauda equina.
• Cauda equina syndrome; conus medullaris syndrome.
• Epidural hematoma, epidural abscess, epidural mass, subarachnoid hemorrhage, spinal cord ischemia.
• Prior imaging demonstrating inadequate epidural space for needle placement at the target level. In the cervical spine this is not a theoretical exclusion — it is a routine and appropriate reason to select a different level or a different procedure entirely.
• Spinal fracture occurring less than six weeks before the proposed injection.
• Demyelinating disease or other central nervous system processes predisposing to transverse myelitis.
• Uncontrolled diabetes (Carelon).
• Spinal neoplasm; rapidly progressing neurological deficit (UnitedHealthcare).

A DOCUMENTATION POINT THAT PREVENTS PEER-TO-PEERS

Most records document contraindications by omission — if the note does not mention infection, presumably there is none. Reviewers do not read omission as clearance.

One affirmative sentence in the procedure planning note — 'No fever, no constitutional symptoms, no history of malignancy, no recent trauma, no coagulopathy; anticoagulation reviewed with prescribing cardiologist on [date]; MRI demonstrates adequate epidural space at the planned level' — closes every screening question a reviewer has and is one of the highest-yield sentences a practice can standardize.

Back to procedure guide navigation ↑

07

Coverage and Authorization

Conservative Treatment Requirements

Conservative care is the most frequently denied element of a CILESI submission and, paradoxically, the element practices are most confident they have documented. The gap arises because clinicians document conservative care as a conclusion and payers require it as a record. 'Failed conservative management' is a conclusion. 'Physical therapy at Northside Rehabilitation, twelve visits from 02/03/2026 to 03/19/2026, discharged with no improvement in radicular symptoms; meloxicam 15 mg daily from 01/28/2026 to present without benefit; gabapentin titrated to 900 mg daily, discontinued 03/02/2026 for sedation' is a record.

7.1 What each payer requires

Payer / UM organization	Conservative therapy requirement
Medicare (LCD family)	Pain duration of at least four weeks, and either inability to tolerate non-invasive conservative care or documented failure to respond to four weeks of non-invasive conservative care. Conservative therapy is defined as an appropriate combination of medication (for example NSAIDs or analgesics) in addition to physical therapy, spinal manipulation therapy, cognitive behavioral therapy, home exercise program, or other interventions based on the individual's presentation, physical findings, and imaging. ESIs should be performed in conjunction with conservative treatments, and the patient should be part of an active rehabilitation program, home exercise program, or functional restoration program. Acute herpes zoster refractory to conservative management does not require the four-week wait.
eviCore (Cigna CMM-200, effective 08/04/2026)	Failure to respond to at least four weeks of conservative treatment — for example exercise, physical therapy, chiropractic care, or medications including NSAIDs or analgesics. Separately and additionally, the individual must be participating in a comprehensive pain management program that includes all of: physical therapy, patient education, psychosocial support, and oral medications. The 'and' is load-bearing — four weeks of failed PT alone does not satisfy the second requirement.
Carelon (Interventional Pain Management, effective 06/14/2026)	Conservative management must combine strategies to reduce inflammation, alleviate pain, and correct underlying dysfunction, including physical therapy AND at least one complementary conservative treatment strategy. The PT requirement is satisfied by PT from a qualified provider, or by a supervised home program with a tailored plan, initial active instruction by an MD, DO, or PT with patient redemonstration, and documented compliance or clinician attestation. Complementary strategies include anti-inflammatory medications and analgesics, adjunctive medications such as nerve membrane stabilizers or muscle relaxants, or alternative therapies including acupuncture, chiropractic manipulation, massage, activity modification, or a trial of rest. Failure requires completion of a full course, worsening or no significant improvement on clinical reevaluation, and consideration of more invasive therapy. Standard duration is four weeks, reducible to two weeks where there is clear evidence of radiculopathy, defined as pain along a specific nerve root with corresponding dermatomal sensory changes, myotomal weakness, and/or reflex abnormality.
UnitedHealthcare (2026T0616N, effective 06/01/2026)	Radicular Back Pain unresponsive to at least four weeks of conservative treatment, satisfied by any one of: pharmacotherapy such as NSAIDs or acetaminophen; activity modification including restriction of heavy lifting, bending, and spinal torsion; or physical therapy or home exercise. The disjunctive structure makes this the least demanding conservative care standard among the major national policies — a fact worth knowing when triaging which submissions need the most build-out.
Aetna (CPB 0016)	Consistent with the general four-week failed conservative care framework. Aetna's musculoskeletal program materials emphasize documented conservative care and, for repeats, documented relief with functional gain and medication reduction.

THE CARELON TWO-WEEK PATHWAY IS UNDERUSED

Carelon permits epidural steroid injection after two weeks rather than four when there is clear objective evidence of radiculopathy — pain along a specific nerve root with corresponding dermatomal sensory changes, myotomal weakness, and/or reflex abnormality.

Most practices never invoke it, because the examination note says 'positive Spurling, decreased sensation' rather than naming the root, the dermatome, the myotome, and the reflex. The pathway is unlocked by examination specificity, not by advocacy. For an Anthem or Blue plan patient in acute radicular distress with a clean neurologic exam, this halves time to treatment — and it costs nothing but one better-written physical examination.

7.2 What to capture, and where

Conservative care documentation fails at the source, not at the authorization desk. The authorization specialist cannot manufacture a PT discharge summary that does not exist. The following belong in the referring or treating clinician's note, not in a prior authorization worksheet:

• Modality — named specifically. Not 'therapy.' Physical therapy, chiropractic manipulation, home exercise program, acupuncture, activity modification.
• Provider and location — where it was done and by whom.
• Dates — start and end, or start and 'ongoing.' A duration without dates does not establish the four-week clock.
• Dose and duration for medications — drug, dose, dates, and why it was stopped. 'Failed NSAIDs' is not a medication history.
• Outcome — and specifically, outcome on the radicular symptoms, which is what is being treated. 'Neck mobility improved but radicular arm pain unchanged' is a far stronger record than 'no better,' because it demonstrates the clinician distinguished the two.
• Compliance — for home programs, Carelon accepts documented compliance or clinician attestation on follow-up. One attestation sentence satisfies a criterion that otherwise fails.
• Contraindication or intolerance where applicable — Medicare covers inability to tolerate conservative care as an alternative to failure of it. If the patient could not take NSAIDs because of chronic kidney disease and could not tolerate PT because of pain severity, document that; it satisfies the criterion on a different branch.

Back to procedure guide navigation ↑

08

Coverage and Authorization

Imaging Requirements

Imaging drives CILESI authorization in two distinct ways that are frequently confused: advanced diagnostic imaging establishes medical necessity before the procedure, and procedural image guidance is a condition of coverage during the procedure. Both are mandatory. They fail for different reasons and are appealed differently.

8.1 Advanced diagnostic imaging (pre-procedure)

For cervical interlaminar ESI, advanced imaging is effectively non-negotiable. This is a point of genuine divergence from the lumbar spine, where both Carelon and the LCD framework permit more latitude.

Payer	Advanced imaging requirement for cervical ESI	Currency window
Medicare	History, physical examination, and concordant radiological image-based diagnostic testing supporting the covered indication. Any prior imaging should be reviewed for anatomy before the interventional procedure.	Not specified numerically. Concordance and review are the operative requirements.
eviCore (Cigna CMM-200)	Advanced diagnostic imaging is expressly required for cervical/thoracic interlaminar and transforaminal epidural steroid injections — for initial radiculitis or radicular pain requests, for radiculopathy requests, and again for repeat requests.	Within 24 months for each of the three.
Carelon	Evidence of either nerve root compression secondary to herniated disc, or spinal stenosis (central, lateral recess, foraminal, extraforaminal), seen on MRI or CT and correlating with clinical findings. The narrow no-imaging exception applies only to caudal and interlaminar approaches in the lumbar spine — it does not reach the cervical spine.	Within 18 months for herniated disc. For a fixed lesion at the target level not due to herniation, no repeat imaging is required for a repeat request.
UnitedHealthcar e	Evidence of structural and/or functional nerve root involvement by imaging or electrodiagnostic studies. The disjunction is meaningful: EMG/NCV can substitute.	Not specified numerically.
TWO IMAGING CLOCKS, NOT ONE eviCore's window is 24 months. Carelon's is 18 months for herniated disc. A single practice serving both a Cigna and an Anthem population is running two different clocks on the same shelf of MRI reports — and the eighteen-month one expires first. The operational fix is a single field in the pre-authorization worklist: imaging date, with an automated flag at 17 months. Not at twenty-four. Build to the tightest clock in the payer mix and the loosest one takes care of itself. The corresponding concession is equally underused: for a fixed stenotic lesion not due to disc herniation, Carelon requires no repeat imaging at all for a repeat injection. Practices routinely order — and patients routinely wait for — an MRI that the policy does not ask for.

8.2 Procedural image guidance (intra-procedure)

This is a condition of coverage, not a technique preference, and it is the cleanest bright line in the entire policy set.

• Medicare: ESIs must be performed under CT or fluoroscopic image guidance with contrast. Injections performed without image guidance, or by ultrasound, are not reasonable and necessary except for documented contraindication to contrast media such as allergy or pregnancy, in which case ultrasound guidance without contrast may be considered. An initial injection of contrast is required to confirm epidural placement absent a contraindication.
• eviCore: ESI should be performed with fluoroscopic or CT guidance and injection of contrast, excepting an emergent situation where guidance or contrast is contraindicated, such as pregnancy. ESI performed with ultrasound guidance is not medically necessary.
• Carelon: Injections must be performed under fluoroscopy or CT guidance with contrast to confirm needle placement, unless CT or fluoroscopy cannot be performed due to contraindications.
• UnitedHealthcare: The injection must be performed under fluoroscopic or CT guidance. Ultrasound guidance for ESI is unproven and not medically necessary due to insufficient evidence of efficacy — a position UHC reaffirmed in its June 2026 evidence update notwithstanding recent meta-analytic data showing lower intravascular injection rates with ultrasound.
• Documentation: Films adequately documenting final needle position and contrast flow — a minimum of two views — should be retained and made available on request. This is an explicit Medicare documentation requirement and one of the most commonly missed items in a post-payment audit response.
• Coding consequence: Image guidance is bundled into CPT® 62321. It is not separately reportable. Reporting fluoroscopic guidance separately with 62321 is a coding error, not an optimization.

THE ULTRASOUND TRAP

The evidence on ultrasound guidance is moving. UnitedHealthcare's own June 2026 policy narrative summarizes a 2025 systematic review finding significantly lower intravascular injection rates and shorter procedure times with ultrasound guidance for cervical transforaminal injection, and a 2023 meta-analysis finding decreased vascular puncture risk under ultrasound.

UnitedHealthcare read that evidence and left the policy unchanged: ultrasound remains unproven and not medically necessary.

The operational lesson is broader than ultrasound. Favorable evidence in a payer's own policy narrative is not favorable policy. Practices sometimes cite the evidence section of a policy back to the payer in an appeal. Reviewers apply the coverage rationale, not the literature review that precedes it. Cite the coverage rationale.

Back to procedure guide navigation ↑

09

Coverage and Authorization

Medical Necessity

Medical necessity for CILESI is not established by clinical conviction. It is established by a reviewable record containing specific, named, dated elements. This section addresses what demonstrates medical necessity for insurance approval, which is a narrower and more mechanical question than whether the patient clinically needs the injection.

9.1 The Medicare three-part test

The Medicare LCD family conditions coverage on three requirements, all of which must be met:

1. Concordant diagnosis. History, physical examination, and concordant radiological image-based diagnostic testing supporting one of: cervical radiculopathy or radicular pain due to disc herniation, osteophyte or osteophyte complex, or severe degenerative disc disease producing foraminal or central spinal stenosis; or post-laminectomy syndrome; or acute herpes zoster associated pain.
2. Severity with objective measurement. The radicular pain must be severe enough to greatly impact quality of life or function. An objective pain scale or functional assessment must be performed at baseline, prior to interventions, and the same scale must be repeated at each follow-up. The LCD names acceptable scales without limitation: verbal rating scales, the Numerical Rating Scale, and the Visual Analog Scale for pain; and the Pain Disability Assessment Scale, Oswestry Disability Index, Oswestry Low Back Pain Disability Questionnaire, Quebec Back Pain Disability Scale, Roland Morris Pain Scale, Back Pain Functional Scale, and PROMIS profile domains for function.
3. Duration and conservative care. Pain duration of at least four weeks and either inability to tolerate non-invasive conservative care or documented failure to respond to four weeks of it, with the acute zoster exception.

"THE SAME SCALE" IS THE MOST EXPENSIVE PHRASE IN THIS POLICY

The LCD does not merely require a scale. It requires the same scale at baseline and at every follow-up, and it requires the scale used to be documented in the medical record.

This is where practices lose repeat authorizations they clinically deserve. Baseline is captured as VAS in the consultation. Follow-up is captured as NRS in the post-procedure call, because a different staff member used a different template. The relief is real, the patient is better, and the reviewer cannot compute a percentage from two different instruments.

Pick one pain scale and one function scale for the entire practice. Put them in the intake template, the procedure note template, and the follow-up call script. Do not permit substitution. This single standardization step resolves more repeat-injection denials than any appeal letter ever written.

9.2 Commercial medical necessity: the concordance triangle

Commercial and UM criteria converge on a structure worth naming explicitly because it is how reviewers actually read. Three points must agree:

Vertex	What the reviewer looks for	How it fails
Symptom	Pain, dysesthesia, or paresthesia in a level-specific referral pattern of an involved named root, with significant functional limitation.	'Neck and arm pain, 7/10.' No root, no dermatome, no named function lost.

Vertex	What the reviewer looks for	How it fails
Sign	Detailed neurologic examination within the prior three months (eviCore) showing findings concordant with compression of the named root: named muscle or myotomal weakness, altered sensation in the sensory distribution, or diminished, absent, or asymmetric reflexes.	'Neuro intact' or 'strength 5/5 throughout.' A negative exam does not defeat coverage on its own — eviCore accepts imaging as an alternative — but it removes a vertex and shifts all weight to imaging.
Image	Concordant radiologist's interpretation of MRI or CT demonstrating compression of the involved named root or foraminal stenosis at the concordant level, within the payer's currency window.	Imaging that is out of window; imaging read at a level other than the one requested; or multi-level findings with no statement selecting the symptomatic level.

The clinician's job is to close the triangle in one sentence. The most valuable sentence in a CILESI record is a concordance statement that names all three vertices together: 'Right C7 radicular pain in a C7 dermatomal distribution with 4/5 triceps weakness and diminished triceps reflex, concordant with the 03/14/2026 cervical MRI demonstrating right C6-C7 foraminal stenosis with C7 nerve root compression; right C6-C7 interlaminar epidural steroid injection is planned.' That sentence does more work than three pages of narrative, and it is the difference between an auto-approval and a clinical review queue.

9.3 Utilization and dosing parameters as coverage criteria

Several parameters that appear clinical are in fact coverage conditions, and are enforced as such. They are presented here as documentation and coverage requirements, not as clinical dosing guidance, which is outside the scope of this library.

• One level per session. Caudal and interlaminar ESIs involving a maximum of one level are medically reasonable and necessary. CPT® 62321 may be reported for one level per session only.
• Not bilateral. It is not medically reasonable and necessary to perform caudal or interlaminar ESIs bilaterally. CPT® 62321 is not a bilateral procedure and modifier 50 does not apply to it.
• One region per session. It is not medically reasonable and necessary for more than one spinal region to be injected in the same session.
• Not with other blocks. Multiple blocks — ESI, sympathetic blocks, facet blocks, trigger point injections — in the same session as an ESI are not medically reasonable and necessary, with the narrow exception of a facet synovial cyst procedure performed with an ESI in the same session. eviCore and Carelon each carve a comparable exception, framed as a transforaminal injection combined with intra-articular facet synovial cyst aspiration.
• Four sessions per region per rolling twelve months. Applies regardless of the number of levels involved. Enforced by Medicare, Carelon, UnitedHealthcare, Aetna, and eviCore alike.
• eviCore's additional episode limit. Up to three sessions per episode of pain per region in six months, not to exceed four per region in a rolling twelve months. The three-per-episode limit is a second, tighter constraint that sits inside the annual four and is frequently overlooked.
• No predetermined series. It is not medically reasonable and necessary to prescribe a predetermined series of ESIs. eviCore states there is insufficient evidence to support scheduling a 'series of three' in either a diagnostic or therapeutic approach and will not preservice authorize a series; each date of service requires its own request.
• Steroid dose documentation. Medicare states steroid dosing should be the lowest effective amount and publishes recommended per-session ceilings. Carelon goes further: after three injections in the same region, the primary care provider be notified of prolonged repeat steroid use (effective 06/14/2026, Carelon removed its prior fixed cumulative-dose limits in favor of this notification requirement). This is a documentation obligation most practices do not track, and it surfaces as a denial precisely when the patient is doing well enough to warrant a fourth injection.
• Twelve-month horizon. Medicare states it generally would not be reasonable and necessary for ESI treatment to extend beyond twelve months, and that frequent continuation beyond twelve months may trigger a focused medical review. Continuation beyond twelve months requires severity causing significant functional or vocational disability; at least 50 percent sustained improvement in pain and/or 50 percent objective functional improvement on the same scale; a documented rationale such as high surgical risk, patient preference against surgery, or recurrence of pain in the same location relieved by ESI for at least three months; and notification of the primary care provider regarding continuation of procedures and prolonged repeat steroid use. Carelon adopted a comparable PCP notification expectation in its 2026 revision. Almost no practice does this, and it is a stated criterion.

9.4 Provider qualifications as a coverage criterion

The Medicare Program Integrity Manual provides that services are considered medically reasonable and necessary only if performed by appropriately trained providers, and the LCD family makes this explicit for epidural procedures. Practitioners performing epidural procedures for chronic pain must be appropriately trained and credentialed by a formal residency or fellowship program, or certified by an accredited nationally recognized organization, or trained through an accredited post-graduate course or professional training program whose core curriculum includes performance and management of these procedures. Credentialing or privileges are required in both inpatient and outpatient settings. Where imaging is used, state licensure or certification to use and interpret the imaging modality is required. The LCD is explicit that these procedures do not constitute anesthesia services and that the evaluation, methods, and techniques specified are not considered routine for surgical or perioperative anesthesia. Operationally: keep credentialing files current and audit-ready. This is a coverage criterion, not an HR formality, and it is examined during post-payment review.

Back to procedure guide navigation ↑

10

Coverage and Authorization

Medicare Coverage

10.1 There is no NCD

This is the first thing to know and the thing most often gotten wrong. There is no National Coverage Determination for epidural steroid injections. Coverage is determined entirely at the Medicare Administrative Contractor level. The LCDs supplement but do not replace, modify, or supersede existing applicable NCDs or payment policy rules for epidural procedures. The statutory anchor is Title XVIII of the Social Security Act, section 1862(a)(1)(A), under which no Medicare payment may be made for items or services that are not reasonable and necessary for the diagnosis or treatment of illness or injury. The reasonable-and-necessary provision in an LCD is addressed in CMS Internet-Only Manual Publication 100-08, Medicare Program Integrity Manual, Chapter 13, Section 13.5.4.

The practical consequence for a multi-state MSO or a practice near a jurisdictional border: the governing LCD follows the place of service, not the practice's headquarters. The LCDs are substantively aligned, which is a deliberate multi-jurisdictional design, but the document numbers, revision dates, and associated billing and coding articles differ. Cite the correct one.

10.2 The aligned LCD family

LCD	Contractor	Note
L39240	Noridian Healthcare Solutions (Jurisdictions E and F)	Original effective 06/19/2022. Revision effective 04/09/2026. The 2026 revision made editorial changes clarifying needle placement for TFESIs, the conditions in which contrast is contraindicated, and definitions.
L39054	Multi-jurisdictional (see MCD for current contractor assignment)	Paired with Billing and Coding Article A58777.
L39242	Multi-jurisdictional	Paired with Billing and Coding Article A58995; note that versions of this LCD have been retired and superseded — confirm current status in the Medicare Coverage Database.
L39036 / L38994 / L33906 / L36920	Additional aligned MAC policies	Substantively aligned coverage architecture. Verify the LCD applicable to your jurisdiction.

The LCD family was informed by a multi-jurisdictional Contractor Advisory Committee meeting of subject matter experts convened on February 11, 2021, comprising experts in pain management, anesthesiology, physical medicine and rehabilitation, neuroradiology, internal medicine, and nurse anesthesia, with academic and community representation. The transcript, voting results, and audio are published on each MAC's website. Reading the panel's votes is one of the highest-yield hours a utilization management leader can spend, because the votes reveal where the policy is firm and where it is a compromise — and the compromises are where appeals succeed.

10.3 Covered indications, limitations, and documentation

Coverage requires all three elements of the test in Section 9.1, plus:

• Image guidance with contrast (Section 8.2).
• One level per session for interlaminar; not bilateral; one region per session.
• A maximum of four ESI sessions per spinal region per rolling twelve-month period, regardless of levels.
• Repeat coverage requires documented minimum consistent 50 percent improvement in pain for at least three months, or at least 50 percent consistent improvement in the ability to perform previously painful movements and activities of daily living compared with baseline on the same scale. If the patient fails to respond well to the initial ESI, a repeat may be performed after fourteen days using a different approach, level, and/or medication if appropriate, with the rationale and medical necessity documented.
• ESIs performed in conjunction with conservative treatment, with the patient participating in an active rehabilitation, home exercise, or functional restoration program.
• No biologicals or other non-FDA-designated substances in the injectate.
• No moderate or deep sedation, general anesthesia, or MAC absent unequivocal, clearly documented necessity in an exceptional and unique case.

10.4 Medicare documentation requirements (Billing and Coding Article)

The billing and coding article that accompanies the LCD carries the documentation obligations that are tested in post-payment review. These are not aspirational; they are the audit checklist.

• All documentation maintained in the patient's medical record and available to the contractor on request.
• Every page legible, with patient identification including complete name and dates of service, and the legible signature of the physician or non-physician practitioner responsible for and providing the care.
• The submitted medical record must support the ICD-10-CM code selected, and the CPT®/HCPCS code reported must describe the service performed.
• Films documenting final needle position and contrast flow, a minimum of two views, retained and available on request.
• The performing provider's assessment as it relates to the patient's complaint for that visit; relevant medical history; results of pertinent tests and procedures; a signed and dated office visit record and operative report; and documentation supporting medical necessity of the procedure.

FEE-FOR-SERVICE MEDICARE DOES NOT PRE-AUTHORIZE. IT AUDITS.

Traditional Medicare requires no prior authorization for CILESI. Practices sometimes read that as a lighter compliance burden. It is the opposite.

The LCD states plainly that compliance with its provisions may be monitored and addressed through post-payment data analysis and subsequent medical review audits, that frequent continuation of ESIs beyond twelve months may trigger a focused medical review, and that aberrant use of the KX modifier may trigger focused medical review.

A commercial denial costs a resubmission. A Medicare post-payment finding costs an extrapolated overpayment across a sampled universe, plus the cost of the review itself. Practices should hold their Medicare documentation to a higher standard than their commercial documentation, not a lower one. Most do the reverse, because commercial payers ask and Medicare does not.

Back to procedure guide navigation ↑

11

Coverage and Authorization

Commercial Payer Comparison

The controlling document for a commercial CILESI request is usually not the health plan's own medical policy. It is the clinical guideline of the utilization management organization delegated to administer the plan's musculoskeletal or interventional pain program — eviCore, Carelon, Cohere, Evolent, or TurningPoint — and delegation varies by plan, product, line of business, and market. Confirm the delegated vendor and product line before building the packet, because the criteria differ and the submission channel differs with them.

Below is a synthesis of the criteria in effect as of July 2026. It is a practical comparison, not a reproduction of any payer manual; each source is cited in full in Section 24 and should be read in the original before it is relied upon.

11.1 Initial cervical interlaminar ESI — criteria comparison

Criterion	Medicare LCD	eviCore / Cigna	Carelon	UnitedHealthcare
Indication	Cervical radiculopathy or radicular pain from HNP, osteophyte, or severe DDD with stenosis; post-laminectomy syndrome; acute zoster pain	Presumed radiculitis or radicular pain; presumed radiculopathy	Significant radicular pain or cervical radiculopathy with functional impairment	Radicular Back Pain by history and exam
Conservative care	4 weeks failed or intolerant; pain duration 4+ weeks	4 weeks failed AND participation in a comprehensive pain program (PT + education + psychosocial + oral meds)	PT + 1 complementary modality; 4 weeks, or 2 weeks with clear radiculopathy	4 weeks unresponsive; any one of pharmacotherapy, activity modification, or PT/home exercise
Advanced imaging	Concordant image-based testing; review prior imaging	Required for cervical/thoracic; within 24 months	MRI/CT nerve root compression or stenosis correlating with clinical findings; 18 months for HNP	Imaging or electrodiagnostic evidence of nerve root involvement
Objective severity	Pain greatly impacting QOL or function; baseline scale required	Significant functional limitations; detailed neuro exam within 3 months or imaging/EDX within 24 months	Pain 3/10 or more and inability to perform 2+ ADLs/IADLs	Not numerically specified
Image guidance	Fluoro or CT with contrast; US only for contrast allergy or pregnancy	Fluoro or CT with contrast; US not medically necessary	Fluoro or CT with contrast	Fluoro or CT; US unproven / not medically necessary
Levels per session	1 (interlaminar); not bilateral	1 spinal level for ILESI; not with a TFESI	1 injection per session for interlaminar; not with a TFESI	Not separately specified
Regions per session	1	1 invasive modality per date of service	1	Not separately specified
Annual limit	4 sessions per region per rolling 12 months	3 per episode per region in 6 months, not to exceed 4 per region per rolling 12 months	4 therapeutic sessions per region per rolling 12 months, any approach, any number of levels	4 sessions per region per year (year runs from the first approved injection)
Region definition	Cervical/thoracic = ONE region (62321, 64479, 64480)	Cervical, thoracic, lumbar treated as regions for frequency	Cervical/thoracic = ONE region for epidurals	Cervical, thoracic, or lumbosacral
Prior authorization	Not required (post-payment review instead)	Required; each date of service requires its own request; no series authorized	Required; peer-to-peer likely where elements are not established	Required

11.2 Repeat cervical interlaminar ESI — criteria comparison

This is where the money is. Initial requests are comparatively easy; repeat requests are where practices bleed. The criteria diverge more here than anywhere else, and the divergence is subtle enough that staff routinely apply the wrong payer's threshold.

Payer	Minimum interval	Response threshold for a repeat	Other
Medicare LCD	14 days after a failed initial injection (different approach, level, and/or medication, with documented rationale)	Consistent 50% or greater improvement in pain for at least 3 months, or 50% or greater consistent improvement in previously painful movements and ADLs vs. baseline on the same scale	Beyond 12 months requires disability, sustained 50% relief or objective function gain, documented rationale, and PCP notification
eviCore / Cigna	At least 14 days since the prior ESI	50% or greater relief of radicular pain for 2 or more weeks, and either increased function/physical activity (e.g., return to work) or reduced use of pain medication and/or additional services such as PT or chiropractic	Advanced imaging within 24 months required again for the cervical/thoracic repeat
Carelon	Not less than the expected duration of relief; a one-time second injection no sooner than 2 weeks after a failed initial	The prior injection produced at least 50% pain reduction AND functional improvement of at least 3 months' duration, documented in a follow-up evaluation	Imaging must show compression or stenosis at the requested level and laterality; no repeat imaging needed for a fixed non-HNP lesion; conservative management must continue between injections
UnitedHealthcar e	Not specified numerically	Pain returned and/or function deteriorated, and either: prior injection gave 50% or less relief for under 3 months with reassessment of the patient, injection site, and technique; or prior injection gave 50% or more relief for 3 or more months	Relief must be measured by validated measurement tools
Aetna (CPB 0016)	Not less than the stated interval in policy	At least 50% pain relief with documented increase in functional abilities and documented reduction in pain medication and/or additional services	Up to 3 interlaminar ESIs per region per episode of pain in 6 months; up to 4 per region per rolling 12 months
THE SINGLE MOST VALUABLE DIVERGENCE IN THIS GUIDE eviCore's repeat threshold is 50% relief for two or more weeks. Carelon's and Medicare's is 50% relief for three or more months. That is not a nuance. It is a six-fold difference in the duration a patient must sustain relief before a second injection is authorized — and the two criteria live side by side in the same practice, reviewed by the same coordinator, using the same follow-up call script. A patient with eight weeks of 60 percent relief is an approval at eviCore and a denial at Carelon. A coordinator who has internalized 'you need three months' will not even submit the eviCore case. A coordinator who has internalized 'you need two weeks' will submit the Carelon case and lose it, then escalate to peer-to-peer and lose it again. UnitedHealthcare is different again, and in an underappreciated way: UHC authorizes a subsequent ESI even where the prior injection gave 50 percent or less relief for under three months — provided there has been a documented reassessment of the individual and the injection site and technique. That is a coverage pathway for a partial responder that neither Medicare nor Carelon offers, and it is unlocked by a single documented reassessment. Operational action: build a payer-specific repeat-eligibility rule into the follow-up call script, not into the coordinator's memory. The call script should branch on payer.

11.3 Other government programs

• Medicare Advantage. MA organizations must follow applicable NCDs and LCDs. Because there is no ESI NCD, the LCD applicable to the place of service governs. Where no NCD, LCD, or other Medicare coverage guidance exists, CMS permits an MA organization to create its own coverage determination using objective evidence-based rationale relying on authoritative evidence (Medicare IOM Pub. 100-16, Ch. 4, §90.5). Practical consequence: an MA plan generally may not apply a commercial criterion that is more restrictive than the LCD for this procedure. When an MA denial cites a commercial policy or a vendor guideline that conflicts with the LCD, say so explicitly in the appeal and cite the LCD by number and revision date. This is one of the most reliably successful MA appeal arguments in interventional pain.
• Medicaid and state Medicaid programs. Criteria vary substantially by state and by managed care organization. Many states adopt the Medicare LCD framework; others impose tighter limits or require additional documentation. Verify the state plan and the specific MCO.
• TRICARE. Coverage generally follows the reliable evidence standard; verify current policy and regional contractor requirements before scheduling.
• Veterans Affairs Community Care. Authorization is issued through the referral, and the authorization scope — number of visits, date range, and procedures — controls. Confirm the referral covers both the evaluation and the procedure before scheduling; a referral for consultation only will not pay for the injection.
• Workers' compensation. Governed by state-specific treatment guidelines and utilization review rules that frequently diverge from group health criteria on both frequency and conservative care duration. Many states have adopted formulary and treatment guideline frameworks that treat cervical ESI more restrictively than commercial policy. Verify the state's adopted guideline, not the group health criteria, and calendar the state's UR response deadlines — which are usually shorter and more consequential than commercial ones.

Back to procedure guide navigation ↑

12

Coverage and Authorization

Prior Authorization Workflow, Denials, and Appeals

12.1 The complete workflow, patient evaluation through scheduling

Stage	Owner	Actions and controls
1. Evaluation & benefit check	Front office / clinical	Verify eligibility. Identify the delegated UM vendor and product line, not just the plan name. Capture the payer's imaging window and repeat threshold into the case record at this moment, because everything downstream branches on them. Confirm remaining sessions in the cervical/thoracic region in the rolling 12 months — including any thoracic injections.
2. Clinical documentation	Physician / APP	Capture the concordance statement (Section 9.2). Capture baseline pain and function on the practice's single standard scales. Document conservative care as a record, not a conclusion. Document the contraindication screen affirmatively. Document rehabilitation participation. State the planned level and side.
3. Imaging review	Physician / APP	Confirm the MRI or CT report is in window for this payer and that the radiologist's interpretation names the level and side you intend to inject. If the report is multi-level, write the sentence that selects the symptomatic level. Confirm adequate epidural space at the planned level.
4. Packet assembly	PA specialist	Assemble against a payer-specific checklist, not a generic one. Order the packet the way the reviewer reads it: concordance statement first, then imaging report, then conservative care, then scores. Do not send the entire chart. A 90-page dump buries the four elements and materially raises the odds of a clinical review.
5. Internal QA	PA lead / clinical reviewer	A second set of eyes against the four-element rule before submission. This is the single highest-return control in the whole workflow — it is far cheaper to fix a gap pre-submission than to appeal it.
6. Submission	PA specialist	Submit through the correct channel — vendor portal, NaviNet or the plan's provider portal, or the plan's own intake. Record the reference number, the submitting user, the date and time, and the criteria version cited. Each date of service requires its own request; series are not authorized.
7. Determination tracking	PA specialist	Calendar the payer's decision deadline. Track pended requests daily. A pended request is not a denial, and it is usually a request for one missing element — respond within the pend window rather than letting it convert to a denial and an appeal.
8. Peer-to-peer	Physician + PA lead	Prepare the physician with the specific criterion at issue and the specific record location that satisfies it. Never send a physician into a P2P to argue clinical merit in the abstract; send them to close a named criterion. (Section 12.4.)
9. Scheduling	Scheduling	Schedule only against an active authorization, with the authorized level, side, CPT code, place of service, and date range. Confirm the authorized level matches the planned level. Any intra-procedure change of level requires a new authorization, not a modifier.
10. Post-procedure & follow-up	Clinical / PA	Capture percentage relief and duration on the same scale as baseline at the payer's relevant interval. Retain the two-view films with contrast flow. Populate the next request while the data is fresh — not ninety days later from memory.

12.2 Common prior authorization denials

The denial reasons below are presented in rough descending order of frequency in GoHealthcare's experience across musculoskeletal specialty practices. Note how few of them are clinical disagreements.

Denial reason	What actually happened	Prevention
Conservative care not documented / insufficient	The record says 'failed conservative therapy' with no modality, dates, or outcome; or PT was done at an outside facility and the records were never obtained.	Structured conservative care field in the note template. Obtain outside PT records at referral intake, not at PA.

Denial reason	What actually happened	Prevention
No baseline pain or functional assessment	A score exists but not on a named scale, or exists in a nursing flowsheet the packet did not include.	One standard pain scale and one standard function scale. Pull them into the packet automatically.
Imaging not current / not concordant	The MRI is 20 months old and the payer is Carelon (18-month window). Or the MRI names C5-C6 and the request names C6-C7.	Flag imaging at 17 months. Require the concordance statement before the request leaves the practice.
Axial pain, not radicular	The note documents neck pain and reduced range of motion. Radicular symptoms are present but were never described in a dermatomal pattern.	The concordance statement. This denial is almost always a documentation failure, not a selection failure.
Repeat: response to prior injection not documented	The patient reported improvement at the follow-up call; nobody recorded a percentage or a duration.	Payer-branched follow-up call script capturing percentage and duration on the baseline scale.
Frequency limit exceeded	The fifth session in twelve months — or the fourth, where a thoracic injection earlier in the year already consumed one of the four in the shared cervical/thoracic region.	Track utilization by region, not by level. Surface the count at scheduling.
Multiple procedures same session	A facet injection and an ESI were scheduled the same day for patient convenience.	Hard scheduling rule: one invasive modality per date of service, absent the synovial cyst exception.
Sedation / MAC denied	Anesthesia was billed with the ESI without documented exceptional necessity.	Default to local. Where sedation is genuinely required, document the unique, exceptional circumstances at the time — not on appeal.
Ultrasound guidance	US was used for a patient with a contrast allergy at a commercial payer that has no US exception.	Know the payer's exception structure. Medicare's contrast-allergy accommodation is not universal.
Non-covered indication	Cervicogenic headache, PHN at a payer that excludes it, or CRPS.	Screen indication against the payer's exclusion list before the visit, not after.
No authorization on file / wrong level authorized	The authorization was obtained for C7-T1; the physician injected C6-C7.	Confirm level and side at time-out against the authorization. A level change requires a new authorization.

12.3 Appeal strategies

The purpose of an appeal is not to persuade. It is to close a named criterion with a named document at a named page. Appeals written as clinical advocacy lose; appeals written as criterion mapping win. The structure below is what GoHealthcare uses and teaches.

1. Read the denial letter for the criterion, not the conclusion. Every denial cites a policy and a criterion. The criterion is the only thing that matters. If the letter is vague, request the specific criterion and the reviewer's credentials in writing — that request alone sometimes resolves the case.
2. Open with a criterion-mapping table. Two columns: the payer's criterion, verbatim by reference and citation; and where in the enclosed record it is satisfied, by document name, date, and page. A reviewer who can verify every criterion in ninety seconds overturns. A reviewer who must hunt does not.
3. Cite the payer's own coverage rationale, by version and effective date. Not the evidence review, not the literature. 'Per Cigna Coverage Policy CMM-200, effective 08/04/2026, Repeat Interlaminar, Caudal, or Transforaminal Epidural Steroid Injection' is the language that moves a file.
4. Fix the gap, then say you fixed it. If the denial was for absent conservative care documentation and the care in fact occurred, obtain the outside records and enclose them. Do not argue that the care occurred. Show it.
5. For Medicare Advantage, argue the LCD. Where an MA denial applies a criterion more restrictive than the applicable LCD, cite the LCD by number and revision effective date and state the conflict expressly. This is the highest-yield MA argument in this procedure family.
6. For cervical evidence challenges, cite the 2025 AAN update. Several payer policies still carry the 2007 AAN position that data on ESI for cervical radicular pain were inadequate to make any recommendation (Level U) — Aetna's CPB 0016 quotes it. The AAN Guidelines Subcommittee's 2025 systematic review summary (Armon et al., Neurology 2025;104(5):e213361) concluded that ESIs probably provide short-term pain reduction and probably reduce short-term disability in cervical radiculopathy, at moderate confidence. Where a denial rests on the 2007 Level U statement, the 2025 update is directly responsive and is the single strongest evidence citation available for cervical ESI. Pair it with ASIPP 2021 Level I for cervical interlaminar injection in disc herniation.
7. Escalate on procedure, not just substance. Track and assert the payer's own timeliness obligations, the reviewer specialty-match requirement where state law provides one, and external review rights. Many denials are resolved on procedural grounds without ever reaching the clinical merits.
8. Close the loop upstream. Every overturned appeal is a defect report on the practice's documentation template. If the same criterion is being appealed twice a month, the template is wrong. Appeals volume is a symptom; template design is the disease.

12.4 Avoiding unnecessary peer-to-peer reviews

Carelon's guideline is unusually candid on this point: where the elements of clinical appropriateness are not established with respect to a given request, the determination will most likely require a peer-to-peer conversation to understand the individual and unique facts. That sentence is the whole strategy. A peer-to-peer is not an adverse event; it is a signal that the packet did not establish an element. Practices that treat peer-to-peer volume as a documentation metric rather than a physician-time cost eliminate most of it within two quarters.

• Track the criterion cited in every P2P. If 40 percent of your CILESI peer-to-peers turn on conservative care documentation, you do not have a peer-to-peer problem. You have a note template problem.
• Never let a coordinator schedule a P2P without identifying the criterion at issue. The physician should walk in knowing exactly which sentence in which policy is unsatisfied and exactly where in the record it is satisfied.
• Carelon flags simultaneous ordering as a P2P trigger — requests for multiple diagnostic or therapeutic interventions at the same time will often require a P2P because appropriateness of additional intervention depends on the outcome of the initial one. Do not batch requests.
• Protect physician time deliberately. A physician hour spent on an avoidable P2P is an hour not spent in clinic. Quantify it and report it monthly. It is the fastest way to get clinical buy-in for documentation standardization.

Back to procedure guide navigation ↑

13

Coding, Documentation, and Operations

Coding and Billing Overview

This section is a concise operational overview. Detailed reimbursement methodology, fee schedules, payment rates, NCCI edit tables, MUE values, and extended revenue cycle analysis are outside the scope of the Procedure Library and belong to the GoHealthcare Revenue Cycle Knowledge Center.

13.1 Applicable CPT® codes

CPT®	Descriptor (abbreviated)	Use for cervical interlaminar ESI
62321	Injection(s) of diagnostic or therapeutic substance(s), not including neurolytic substances, including needle or catheter placement, interlaminar epidural or subarachnoid, cervical or thoracic; with imaging guidance (fluoroscopy or CT)	The primary code. This is the code for a cervical interlaminar ESI performed to the standard of care. Imaging guidance is included and is not separately reportable. One level per session. Not a bilateral procedure.
62320	Same, without imaging guidance	Avoid. Non-image-guided ESI is not reasonable and necessary except for documented contraindication to contrast media. Reporting 62320 is, in effect, reporting a non-covered service. Note that Carelon requires preauthorization for notification purposes only when 62320 is used for post-procedural pain with G89.11, G89.12, or G89.18.
62324 / 62325	Interlaminar epidural or subarachnoid, cervical or thoracic, including indwelling catheter placement, continuous infusion or intermittent bolus; without (62324) or with (62325) imaging guidance	Not a single-shot ESI. Reserve for catheter-based continuous or intermittent bolus delivery. Reporting a catheter code for a single injection is a coding error.
64479 / +64480	Transforaminal epidural, with imaging guidance; cervical or thoracic, single level (64479) / each additional level (+64480)	Cervical transforaminal — a different procedure, not an alternative code for the same one. Included here because 62321, 64479, and 64480 together define the cervical/thoracic region for Medicare frequency counting.
THE REGION RULE, STATED PLAINLY Per the Medicare billing and coding article, an anatomic spinal region for epidurals is defined as cervical/thoracic (CPT® 62321, 64479, 64480) or lumbar/sacral (CPT® 62323, 64483, 64484). Carelon uses the same definition. Therefore: a thoracic interlaminar ESI and a cervical interlaminar ESI draw from the same pool of four sessions per rolling twelve months. So does a cervical transforaminal injection. So does a thoracic transforaminal injection. All four codes count against one allowance. Practices that count 'four cervical injections per year' rather than 'four cervical/thoracic sessions per rolling twelve months' will be surprised, and the surprise arrives as a denial for a patient who is already scheduled.

13.2 HCPCS codes (when applicable)

Where the practice supplies the injectate in a physician office setting, the corticosteroid may be separately reportable with the applicable HCPCS Level II J-code — for example J1030 or J1040 (methylprednisolone acetate), J3301 (triamcinolone acetonide, not otherwise specified), J0702 (betamethasone acetate and betamethasone sodium phosphate), or J1100 (dexamethasone sodium phosphate) — reported with the correct number of units and, where applicable, the JZ or JW modifier for discarded drug reporting. In the ASC and hospital outpatient setting, the drug is generally packaged into the facility payment and is not separately reportable. Contrast media is generally not separately reportable with 62321. Verify the specific payer and setting; this is one of the most variable areas in the whole workflow and one of the easiest to get wrong at scale.

13.3 Common ICD-10-CM diagnosis codes

The following codes appear on the Medicare billing and coding article's list of ICD-10-CM codes that support medical necessity for CPT® 62321 and are the codes most relevant to cervical interlaminar ESI. It is the provider's responsibility to select codes to the highest level of specificity from the code book applicable to the year the service is rendered.

ICD-10-CM	Description	Operational note
M50.121	Cervical disc disorder at C4-C5 level with radiculopathy	Level-specific. Use where the MRI and exam both support C5 root involvement.
M50.122	Cervical disc disorder at C5-C6 level with radiculopathy	Level-specific. C6 root.
M50.123	Cervical disc disorder at C6-C7 level with radiculopathy	Level-specific. C7 root. The most common cervical ESI diagnosis.
M50.13	Cervical disc disorder with radiculopathy, cervicothoracic region	C7-T1 / C8 root.
M54.12	Radiculopathy, cervical region	Acceptable, but less specific. Prefer a level-specific M50.12x where the record supports it — specificity is what demonstrates concordance.
M54.13	Radiculopathy, cervicothoracic region	Cervicothoracic junction.
M47.22	Other spondylosis with radiculopathy, cervical region	Degenerative/osteophytic etiology.
M47.23	Other spondylosis with radiculopathy, cervicothoracic region	Degenerative etiology.
M96.1	Postlaminectomy syndrome, not elsewhere classified	Post-surgery syndrome. Expect closer review; document prior surgery and date.
M99.21 / M99.31 / M99.41 / M99.51	Subluxation / osseous / connective tissue / intervertebral disc stenosis of neural canal, cervical region	Central canal stenosis family.
M99.61 / M99.71	Osseous and subluxation / connective tissue and disc stenosis of intervertebral foramina, cervical region	Foraminal stenosis family.
B02.23 / B02.29 / B02.8 / B02.9	Postherpetic polyneuropathy / other postherpetic nervous system involvement / zoster with other complications / zoster without complications	Zoster family. Medicare's covered indication is acute zoster-associated pain; several commercial payers exclude post-herpetic neuralgia outright. Verify by payer.
G89.3	Neoplasm related pain (acute) (chronic)	On the Medicare covered list. Note the LCD separately contraindicates ESI where there is suspicion or major risk factors for cancer — these are not in conflict, but the record must make clear which situation applies.

13.4 Applicable modifiers

Modifier	Applicability to CPT® 62321
50 — Bilateral	Does not apply. It is not medically reasonable and necessary to perform interlaminar ESIs bilaterally; the Medicare billing and coding article states expressly that 62321 and 62323 are not bilateral procedures. Appending modifier 50 to 62321 is an error. (Modifier 50 is used for bilateral transforaminal injections, 64479-64484 — do not carry that habit across to the interlaminar code.)
RT / LT	Not applicable to a midline or paramedian interlaminar injection, which is not a laterality-coded procedure. (RT/LT reporting on separate lines is an ASC facility convention for bilateral transforaminal procedures; physicians continue to use modifier 50.)

Modifier	Applicability to CPT® 62321
KX — Requirements specified in the medical policy have been met	Reserved for a diagnostic selective nerve root block, which is coded identically to an epidural injection; the KX modifier distinguishes the DSNRB from an ESI. It is not a general-purpose attestation to be appended to 62321. Aberrant use of the KX modifier may trigger focused medical review. This is one of the most consequential and least understood modifier rules in interventional pain.
59 / X{EPSU}	Rarely appropriate. Multiple blocks in the same session as an ESI are not medically reasonable and necessary, with the narrow synovial cyst exception. A distinct procedural service modifier does not convert a non-covered same-session combination into a covered one.
GA / GX / GY / GZ	Where a service is known to be non-covered, bill it as non-covered with the appropriate modifier and an executed ABN where applicable. The LCD states it is not appropriate to bill Medicare for non-covered services as if they were covered.
26 / TC	Not applicable. Imaging guidance is bundled into 62321 and is not separately reportable, so there is no professional/technical split to report.
JW / JZ	Discarded drug reporting where the injectate is separately reportable in the office setting.

13.5 Place of service and ASC considerations

• Common places of service: 11 (office), 24 (ambulatory surgical center), 22 (on-campus hospital outpatient), 19 (off-campus hospital outpatient).
• Site-of-service policies matter. UnitedHealthcare maintains a separate Office Based Procedures — Site of Service policy referenced from its ESI policy. A procedure that is medically necessary may still be denied at the requested site of service. Authorize the procedure and the site.
• ASC facility reporting. For bilateral transforaminal procedures, physicians continue to use modifier 50 while the ASC facility reports the procedure on two separate lines with one unit each and RT and LT appended. This does not apply to 62321, which is not bilateral, but the distinction is a frequent source of ASC coding error in mixed epidural schedules.
• Anesthesia. Do not assume anesthesia is billable. Moderate or deep sedation, general anesthesia, and MAC are usually unnecessary or rarely indicated for these procedures and are not routinely reimbursable; individual consideration may be available on appeal where medical necessity is unequivocal and clearly documented.
• Post-operative pain management exception. When epidural injections are used for post-operative pain management, the diagnosis code restrictions in the Medicare article do not apply. This is a distinct scenario from therapeutic CILESI and should not be conflated with it.

13.6 Common coding and billing pitfalls

• Reporting fluoroscopic guidance separately with 62321. It is bundled.
• Appending modifier 50 to 62321. It is not a bilateral procedure.
• Using the KX modifier as a general medical-necessity attestation rather than as the DSNRB distinguisher. This can trigger focused medical review.
• Reporting 62320 (no imaging) when fluoroscopy was in fact used — a template default error that converts a covered service into a non-covered one.
• Reporting a catheter code (62324/62325) for a single-shot injection.
• Reporting more than one unit of 62321 for a single session. One level per session.
• Billing a cervical and a thoracic injection on the same date. One region per session — and they are the same region.
• Selecting M54.12 (radiculopathy, cervical region) when the record supports level-specific M50.123. Specificity is what demonstrates concordance.
• Using epidural CPT® codes to bill chemotherapy or complex drug administration — expressly addressed and prohibited in the Medicare billing and coding article.
• Including a biologic in the injectate. This may result in denial of the entire claim, not just the biologic.

Back to procedure guide navigation ↑

14

Coding, Documentation, and Operations

Clinical Documentation Checklist

Print this page. It is designed to be used as a pre-submission gate, not as a retrospective audit tool. Every unchecked box is a denial the practice has not yet received.

Patient identification and eligibility

• Eligibility verified; plan, product line, and delegated UM vendor identified (eviCore, Carelon, Cohere, Evolent, TurningPoint, or plan-direct)
• This payer's imaging currency window recorded in the case (18 months? 24 months? unspecified?)
• This payer's repeat response threshold recorded in the case (2 weeks? 3 months? partial-responder pathway?)
• Sessions used in the cervical/thoracic region in the rolling 12 months confirmed — including any thoracic injections and any cervical or thoracic transforaminal injections

Clinical history

• Symptom onset date and duration of at least 4 weeks documented
• Radicular pain described in a named, level-specific dermatomal distribution
• Laterality documented
• Functional impairment named specifically — which ADLs or IADLs, and how
• Prior cervical spine surgery documented with date, if applicable
• Red flag screen documented affirmatively (malignancy, infection, trauma, progressive deficit, myelopathy)

Objective measurement

• Baseline pain score on a named scale, with the scale named in the record (NRS, VAS, or verbal rating scale)
• Baseline functional assessment on a named scale (Neck Disability Index, PROMIS domain, PDAS, or equivalent)
• The practice's standard scales were used — the same instruments that will be used at every follow-up

Physical examination

• Detailed neurologic examination performed within the prior 3 months (eviCore requirement)
• At least one objective finding concordant with the named root documented: named muscle or myotomal weakness, altered sensation in the sensory distribution, or diminished / absent / asymmetric reflex
• Provocative testing documented
• Findings tied explicitly to the named root, not to 'the arm'

Conservative treatment

• At least 4 weeks of conservative care documented (or 2 weeks with clear objective radiculopathy, Carelon only)
• Each modality named, with provider, dates, and outcome
• Medications listed with drug, dose, dates, and reason for discontinuation
• Outcome stated as to the radicular symptoms specifically
• For eviCore: participation in a comprehensive pain management program documented — PT AND patient education AND psychosocial support AND oral medications
• For Carelon: PT AND at least one complementary modality documented; home program compliance documented or attested
• Intolerance or contraindication to conservative care documented where failure is not the basis
• Ongoing active rehabilitation, home exercise, or functional restoration participation documented

Imaging

• MRI or CT report obtained, not just the impression line
• Imaging date within this payer's window
• Radiologist's interpretation names the level and the side to be injected
• Where imaging is multi-level: a sentence selecting the symptomatic level is in the record
• Adequate epidural space at the planned level confirmed

Medical necessity

• Concordance statement present, naming symptom, sign, image, level, and side in one sentence
• Planned level, side, and approach stated
• Rationale for interlaminar rather than transforaminal or caudal, where relevant
• Statement that the injection is part of a comprehensive care plan, not a standalone treatment
• No predetermined series requested

Repeat injections (in addition to all of the above)

• Percentage of relief from the prior injection documented
• Duration of that relief documented
• Both measured on the same scale as baseline
• Functional improvement documented (increased activity, return to work, reduced medication, or reduced use of PT/chiropractic)
• Conservative management between injections documented (Carelon requirement)
• For UnitedHealthcare partial responders: reassessment of the patient, the injection site, and the technique documented
• Interval since prior injection meets the payer's minimum
• Session count within the region limit confirmed
• For repeated injections under Carelon: primary care provider notified of prolonged repeat steroid use (fixed cumulative-dose limits removed 06/14/2026)
• For continuation beyond 12 months under Medicare: disability, sustained response, documented rationale, and primary care provider notification all present

Procedure record and retention

• Signed, dated operative report describing the level, side, approach, guidance modality, and contrast use
• Films retained: minimum of 2 views documenting final needle position and contrast flow
• Legible signature of the performing provider
• Provider credentialing current and on file
• Authorization on file matches the level, side, CPT code, place of service, and date actually performed

Back to procedure guide navigation ↑

15

Coding, Documentation, and Operations

Procedure Comparison

The three epidural approaches are not interchangeable, and in the cervical spine the choice among them is driven far more by safety than by efficacy. Understanding why cervical interlaminar dominates clarifies both the clinical reasoning and the payer posture.

Attribute	Cervical interlaminar (62321)	Cervical transforaminal (64479 / +64480)	Caudal (62323)
Access	Posterior, between the laminae, midline or paramedian	Through the neuroforamen, adjacent to the exiting root and dorsal root ganglion	Through the sacral hiatus into the sacral canal
Applicability to the cervical spine	Yes — the dominant cervical approach	Yes, but see safety	No. Caudal access reaches lumbosacral levels only; eviCore notes a CESI involves symptomatic levels below L4-L5
Delivery	Broader, less selective spread; well suited to central or multi-root involvement	More selective, targeted to a single named root	Non-selective; volume-dependent
Levels per session	1	Up to 2 levels in one region (Medicare, eviCore, Carelon); most conditions do not require 2	1
Bilateral permitted	No	Yes, when clinically indicated (modifier 50 for the physician)	No
Evidence (ASIPP 2021)	Level I, strong recommendation, for cervical disc herniation; Level II, moderate to strong, for cervical central stenosis and cervical post-surgery syndrome	No randomized trials assessing cervical TFESI effectiveness were available in the systematic reviews underlying the LCD; a 2020 GRADE systematic review found roughly half of patients achieved 50% or greater pain reduction at short and intermediate term, on very low quality evidence	Level I for lumbar disc herniation; not applicable cervically
Safety	Principal risk is direct trauma to the spinal cord; the epidural space narrows as it ascends, and evidence reviewed in the LCDs supports restricting cervical injections to the lower cervical levels	Cervical and thoracic transforaminal epidurals are roughly 2.4% of all epidural injections and under 5% of all transforaminal injections, yet contribute over 99% of complications related to intra-arterial injection of particulate steroid. A 2015 safety review concluded cervical TFESIs must not be performed until appropriate evidence develops and safe steroid preparations are available	Lowest-risk epidural approach; not cervically applicable
Payer posture	Broadly covered for cervical radicular pain with concordant imaging	Covered, but reviewed more closely; particulate steroid concerns are explicit in the FDA Safe Use Initiative consensus that Carelon requires providers to follow	Not applicable cervically

WHY THIS COMPARISON MATTERS OPERATIONALLY

The cervical safety asymmetry is not a footnote — it is the reason the interlaminar approach owns the cervical spine while the transforaminal approach owns the lumbar spine.

That has a direct authorization consequence: patterns that are routine and defensible lumbar practice do not transfer cervically. Two-level transforaminal injections, bilateral transforaminal injections, and modifier 50 are all normal lumbar behavior. Carried into a cervical schedule, they produce denials, peer-to-peers, and — in the case of modifier 50 on 62321 — a coding error.

Practices that do high lumbar volume and occasional cervical volume are the most exposed, because the muscle memory is lumbar.

Back to procedure guide navigation ↑

16

Coding, Documentation, and Operations

GoHealthcare Clinical Insights

The recommendations below reflect GoHealthcare's operational experience supporting musculoskeletal specialty practices, ambulatory surgery centers, and hospital outpatient departments. They are written for the people who have to make this work on a Tuesday afternoon.

Documentation quality

• Write one sentence, not three pages. The concordance statement in Section 9.2 outperforms narrative volume. Reviewers work a queue; the packet that lets them verify every criterion quickly is the packet that approves quickly.
• Document conservative care at the source. The single most common structural failure in this procedure is that the conservative care happened somewhere else and the records were never obtained. Build record retrieval into referral intake — not into prior authorization, where it becomes an urgent request against a scheduled procedure date.
• Say what the conservative care did to the radicular pain. Payers are authorizing treatment of a root, not of a neck. 'Range of motion improved; radicular arm pain unchanged' is a stronger record than 'no improvement,' because it shows the clinician tracked the right thing.
• Screen contraindications affirmatively. One standardized sentence closes an entire category of reviewer questions (Section 6).

Workflow improvement

• Move the payer's rules to the front of the workflow. Imaging window and repeat threshold should be captured at eligibility verification, before the visit — not discovered at submission. A 20-month-old MRI is a solvable problem in week one and a cancelled procedure in week six.
• Build the pre-submission QA gate. One reviewer, one checklist, before submission. This control returns more than any other single intervention in this procedure family, because the marginal cost of fixing a gap pre-submission is minutes and the marginal cost of fixing it post-denial is weeks plus physician time.
• Standardize the scales and forbid substitution. One pain scale, one function scale, in every template and script (Section 9.1).
• Track utilization by region, not by level. Surface the cervical/thoracic session count on the scheduling screen.
• Set the imaging flag at 17 months. Build to the tightest clock in your payer mix.

Prior authorization success

• Confirm the delegated vendor first. The plan name on the card does not tell you whose criteria will be applied. eviCore, Carelon, Cohere, Evolent, and TurningPoint apply different rules through different channels.
• Submit one date of service at a time. No payer preauthorizes a series, and Carelon flags simultaneous ordering as a peer-to-peer trigger.
• Do not send the whole chart. A large, unordered packet increases the probability of clinical review, because the reviewer cannot find the elements.
• Work pends the day they arrive. A pend is a request for one missing element with a deadline attached. Left alone, it becomes a denial and an appeal.

Denial prevention

• Instrument the follow-up call. The repeat-injection denial is created ninety days before it arrives, at the follow-up call where nobody wrote down a percentage and a duration. Fix the script and the denial disappears.
• Branch the script by payer. The eviCore two-week threshold and the Carelon three-month threshold cannot both live in a coordinator's memory. Put the branch in the tool.
• Audit your own denials monthly by criterion, not by payer. Payer-level denial reporting tells you where the pain is. Criterion-level reporting tells you what to fix.

Communication between providers and authorization teams

• Give the physician the criterion, not the case. Before any peer-to-peer, the coordinator's job is to identify the single unsatisfied criterion and the exact record location that satisfies it.
• Report avoidable physician time monthly. Nothing drives clinical engagement with documentation standards faster than a number showing how many hours of clinic time were spent on peer-to-peers that a better sentence would have prevented.
• Close the loop from appeals to templates. Every overturned appeal is a defect report. If the same criterion appears twice in a month, change the template — do not write a better appeal letter.

Back to procedure guide navigation ↑

17

Coding, Documentation, and Operations

GoHealthcare Prior Authorization Insight

This section is exclusive to GoHealthcare. It contains the operational knowledge that is not available in payer policies or clinical guidelines — the things a national musculoskeletal specialty management services organization learns from working thousands of these requests, and that no single policy document will tell a practice.

17.1 The axial pain gap: where the evidence and the policy openly disagree

This is the most consequential thing in this guide that a practice will not learn from reading a payer policy, because it requires reading two literatures against each other.

ASIPP's 2021 comprehensive evidence-based guidelines assign Level II evidence with a moderate to strong recommendation for fluoroscopically guided cervical interlaminar epidural injection in axial discogenic pain without facet joint pain, based on a relevant high-quality randomized controlled trial, with or without steroids. Manchikanti and colleagues published two-year follow-up data on fluoroscopic cervical epidural injections in chronic axial or discogenic neck pain. The evidence for this indication is not absent. It is real, it is published in the peer-reviewed literature, and it is graded.

Every payer reviewed in this guide excludes it. Medicare's LCD lists axial spine pain as investigational and therefore not medically reasonable and necessary. Carelon lists isolated axial neck pain among its not-medically-necessary indications. eviCore defines axial spinal pain as the absence of radiculopathy, myelopathy, or myeloradiculopathy and calls it not medically necessary. UnitedHealthcare's entire coverage rationale is built around Radicular Back Pain and states there is insufficient evidence for ESI in back pain absent radicular symptoms.

WHAT PRACTICES DO WRONG HERE, AND WHAT TO DO INSTEAD

The mistake: a physician who knows the ASIPP evidence submits an axial-pain request and cites ASIPP in the clinical rationale. It is denied. The physician, correctly believing the evidence supports the treatment, escalates to peer-to-peer and cites ASIPP again. It is denied again. Two hours of physician time are gone and the patient is no closer to treatment.

Why it fails: the reviewer is not empowered to weigh evidence. The reviewer applies the coverage rationale. An evidence argument against an explicit exclusion cannot succeed at utilization review, no matter how good the evidence is.

What to do instead:

1. Screen for it before the visit. If the presentation is axial without a radicular component, CILESI is not the next covered step at any payer in this guide. Route the patient to a different pathway — facet evaluation, rehabilitation, or another modality — rather than to a denial.
2. Do not manufacture radiculopathy. This must be said plainly. The pressure to document a dermatomal pattern that is not there is real, and it is the beginning of a compliance problem, not a solution to an authorization problem. If the radicular component is genuinely present, document it properly. If it is not, do not.
3. Where a genuine radicular component coexists with axial pain — which is common — document the radicular component as the indication, with the concordance triangle intact, and describe the axial pain as an associated symptom rather than as the target. That is not gamesmanship; it is accurate characterization of what is being treated and why.
4. Advocate through policy channels, not through individual appeals. The place to move an exclusion is the LCD reconsideration process and the payer's policy comment cycle — not a Tuesday peer-to-peer.

17.2 Where the LCD's own experts disagreed with the LCD

The February 2021 multi-jurisdictional Contractor Advisory Committee is a public record, and its voting results are one of the most useful and least read documents in interventional pain. The panel's confidence votes reveal where the LCD is settled and where it is a policy choice made over expert uncertainty:

• The panel voted only 2 of 5 confidence that radicular pain should be concordant with a radiologist's interpretation showing compression of the involved root — noting imaging is a static image of a dynamic process and that chemical radiculitis is invisible to it. The concordant imaging requirement nonetheless survived into the policy.
• The panel voted 3 of 5 that evidence supports periods of conservative management before ESI, observing that the time interval relates more to natural history than to evidence.
• The panel voted only 2 of 5 confidence in evidence supporting documentation of failure of two medication classes before ESI.
• On the number of ESIs that is safe or harmful per year, 7 of 11 experts voted that the literature provides no answer, and there was no consensus, with a suggested range of three to four in twelve months. The four-per-region-per-year limit is therefore a policy construct informed by expert opinion, not an evidence-derived threshold.
• The panel voted 4 of 5 confidence for cervical ESIs generally, and 4 of 5 for cervicobrachialgia.

Why this matters operationally. It does not change what to submit. It changes how to think about appeals and about clinical conversations. Where a criterion is a policy construct built over expert uncertainty — the four-session cap is the clearest example — an individual appeal will not move it, and staff should not burn hours trying. Where a criterion is evidence-anchored, it is worth meeting precisely. Knowing which is which is the difference between a team that appeals strategically and a team that appeals reflexively.

17.3 Frequent payer expectations that practices systematically miss

• The PCP notification requirement. Medicare requires that the primary care provider be notified regarding continuation of procedures and prolonged repeat steroid use beyond twelve months. Carelon adopted a comparable expectation in 2026. In GoHealthcare's experience, almost no practice does this, and almost no practice knows it is a stated criterion. It is a single letter or a single message from the EHR. Build it into the twelve-month review.
• Prolonged steroid use notification (Carelon). Effective 06/14/2026, Carelon removed its fixed cumulative-dose limits and instead requires the ordering provider to notify the primary care provider of prolonged repeat steroid use. Documenting cumulative steroid dose remains sound practice, but the operative trigger is now the PCP notification, not a numeric ceiling. Add a PCP-notification field to the procedure note for repeat injections.
• The 'same scale' requirement. Enforced literally. See Section 9.1.
• eviCore's comprehensive pain management program. Four elements, all required: physical therapy, patient education, psychosocial support, and oral medications. Practices document the PT and the medications and are surprised by the denial. Patient education and psychosocial support are usually happening — they are just not being written down.
• The two-view films with contrast flow. An explicit Medicare documentation requirement and a common gap in post-payment audit response. Retention policy should be explicit and tested.
• Carelon's home program attestation. Carelon accepts clinician attestation of compliance on follow-up evaluation. One sentence satisfies a criterion that otherwise requires documentation the practice does not have.

17.4 Internal quality assurance recommendations

• Pre-submission QA gate against the four-element rule, staffed by one accountable reviewer.
• Monthly criterion-level denial analysis. Report by criterion, not by payer. The output should be template changes, not appeal letters.
• Quarterly policy-currency review. Every payer in this guide revised its ESI policy within the eighteen months preceding publication: the Noridian LCD effective 04/09/2026, eviCore CMM-200 effective 08/04/2026, UnitedHealthcare 2026T0616N effective 06/01/2026, and Carelon effective 06/14/2026. A practice operating on a policy binder assembled a year ago is operating on a fiction.
• Region-based utilization dashboard surfacing cervical/thoracic session counts at scheduling.
• Peer-to-peer log capturing the criterion at issue and the outcome. This is your documentation defect list.
• Credentialing audit aligned to the LCD's provider qualification criterion, refreshed annually and held audit-ready.
• Scale-standardization audit. Sample ten charts a month. Confirm baseline and follow-up used the same instrument. This one audit protects the entire repeat-injection revenue stream.

Back to procedure guide navigation ↑

18

Coding, Documentation, and Operations

GoHealthcare Leadership Perspective

Cervical interlaminar epidural steroid injection is a useful lens on musculoskeletal specialty operations generally, because it is a procedure where the clinical decision is comparatively straightforward and almost all of the organizational difficulty lives downstream of it. An executive reading this guide should take away four structural observations.

Documentation excellence is a design problem, not a compliance problem

Practices routinely respond to documentation-driven denials by exhorting physicians to document better. This does not work, and it is not a physician failing. The elements payers require — a named-scale baseline, a dated modality-specific conservative history, a concordance statement — are not elements that a well-trained clinician naturally produces, because they are not elements that clinical reasoning requires. A physician who has established that a patient has C7 radiculopathy from a C6-C7 herniation and needs an injection has completed the clinical task. Everything the payer additionally wants is an artifact of the review process, not of the diagnosis. The fix is therefore in template design, not in physician behavior. Organizations that engineer the required elements into the structure of the note — so that producing them is the path of least resistance rather than an additional task — solve this permanently. Organizations that rely on reminders solve it for about six weeks.

Prior authorization is a data-capture problem masquerading as a negotiation

The intuition that prior authorization is adversarial — that approval is won by advocacy — is durable and mostly wrong for this procedure. The overwhelming majority of CILESI denials are not disagreements. They are the payer's inability to locate four data elements in the submitted record. This reframing has real budgetary consequence: it argues for investment in intake, templates, records retrieval, and pre-submission QA, and against investment in appeal-writing capacity and physician peer-to-peer time. An organization whose prior authorization spend is concentrated downstream of submission is spending on the symptom.

Medical necessity and revenue cycle are the same conversation, held twice

For CILESI under commercial coverage, the medical necessity conversation happens prospectively at utilization review. Under traditional fee-for-service Medicare, the identical conversation happens retrospectively at post-payment medical review — but with extrapolation risk attached and without the opportunity to fix the record. Practices consistently underinvest in the Medicare version precisely because nobody asked. The absence of prior authorization is not the absence of criteria; it is the deferral of the audit. Executives should insist that Medicare documentation be held to the same or a higher standard than commercial documentation, and should treat the LCD's stated triggers for focused medical review — continuation beyond twelve months, aberrant KX modifier use — as active operational risk indicators, not as background noise.

Policy is perishable, and most organizations treat it as durable

Within the eighteen months before this guide was published, every major policy governing this single procedure was revised: the Noridian LCD (revision effective April 16, 2026), the eviCore/Cigna CMM-200 guideline (effective February 25, 2026), the UnitedHealthcare medical policy (2026T0616N, effective June 1, 2026), and the Carelon interventional pain guideline (effective June 14, 2026, archiving the prior version). The 2026 Carelon revision alone added a primary care notification expectation and restructured the repeat-injection language. A payer policy binder is not a reference; it is a perishable asset with a maintenance cost. Organizations that do not own that maintenance cost explicitly — with a named owner and a quarterly cycle — are running on assumptions that were

true last year.

On AI-enabled workflow

There is a real and specific opportunity here, and it is narrower than the market rhetoric suggests. The tasks in this workflow that are genuinely amenable to automation are the deterministic ones: flagging an imaging date against a payer-specific window; counting sessions within a rolling twelve-month window against a region definition; branching a follow-up script on payer; checking a packet for the presence of four named elements before submission; detecting that a note used VAS at baseline and NRS at follow-up. These are rule applications, and they are exactly where practices currently lose money. They do not require clinical judgment, and they should not be given clinical judgment.

The tasks that are not amenable are the ones organizations are most tempted to automate: generating the concordance statement, drafting clinical rationale, and composing appeal narratives. A generated concordance statement is an assertion about a patient that no clinician has verified, and the fact that it will usually be right is not a governance answer. GoHealthcare's position is that AI in this workflow should be deployed against rules and completeness, with a human clinician authoring every clinical assertion, and with an audit trail identifying which elements were machine-checked and which were clinician-authored. Automate the checking. Do not automate the claiming. That line is where AI governance in prior authorization is actually decided, and it is worth an organization's leadership drawing it deliberately rather than discovering it in an audit.

Back to procedure guide navigation ↑

19

Optimization, FAQs, and Outlook

GoHealthcare Case Study

This is an original educational case study. It does not describe any actual patient or organization. All names, dates, and details are constructed for instructional purposes.

Clinical scenario

A 58-year-old right-hand-dominant commercial electrician presents to a multi-site interventional pain practice with six weeks of right-sided neck pain radiating into the posterior arm and into the middle finger, with associated numbness. He reports difficulty gripping tools overhead and has stopped working. He has a commercial plan administered by a national carrier whose interventional pain program is delegated to a utilization management vendor. Examination documents reduced cervical range of motion, a positive Spurling maneuver on the right, 4/5 triceps strength, diminished right triceps reflex, and decreased sensation over the middle finger. Cervical MRI performed twenty-two months earlier for a prior episode demonstrates a right paracentral C6-C7 disc protrusion with foraminal narrowing and C7 nerve root contact. He has completed eight sessions of physical therapy over five weeks at an outside facility, with improvement in neck stiffness but no change in arm symptoms, and has taken naproxen without benefit. The physician plans a right C6-C7 interlaminar epidural steroid injection.

Documentation challenges

• The examination is excellent but the note does not name the root. It describes findings; it does not connect them. There is no concordance statement.
• Baseline pain was recorded as '7/10' in the nursing intake flowsheet with no scale named. No functional assessment instrument was administered at all.
• The physical therapy occurred at an outside facility. The clinic note says 'completed PT, no improvement.' The PT records are not in the chart.
• The naproxen is listed on the medication list. There is no start date, no dose, and no documented outcome.
• The MRI is twenty-two months old.
• There is no documentation of ongoing rehabilitation participation, patient education, or psychosocial support.

Prior authorization barriers

The coordinator submits with the clinic note and the MRI report. The request pends, then denies. The denial cites two criteria: conservative treatment not adequately documented, and no baseline measurement on a validated scale. Notably, the payer did not dispute that the patient has C7 radiculopathy. The clinical judgment was never in question. The physician, reading the denial as a clinical challenge, requests a peer-to-peer.

Payer considerations

The reviewing organization's criteria required four weeks of failed conservative treatment and participation in a comprehensive pain management program comprising physical therapy, patient education, psychosocial support, and oral medications. Its imaging window for cervical interlaminar ESI was twenty-four months — the MRI was in window, with two months to spare, and nobody had checked. Had this patient carried an Anthem plan reviewed under the Carelon guideline, the eighteen-month window would have expired four months earlier and a new MRI would have been genuinely required. The practice did not know which clock it was running.

Operational workflow

The GoHealthcare team stopped the peer-to-peer before it was scheduled and worked the denial as a documentation gap rather than a clinical dispute:

1. Requested and obtained the outside physical therapy evaluation and discharge summary, which documented eight visits across five weeks with dates, a home exercise program with documented compliance, and a discharge note explicitly recording persistent right upper extremity radicular symptoms despite improvement in cervical mobility.
2. Returned to the physician for a brief addendum containing: the concordance statement naming the C7 root, the C7 dermatome, the triceps findings, and the C6-C7 MRI correlation with the imaging date; the naproxen dose, start date, and outcome; and an attestation of ongoing home exercise participation, patient education provided at the initial visit, and the psychosocial screen already completed at intake but never abstracted into the note.
3. Administered the practice's standard function instrument and documented a baseline pain score on the practice's standard named pain scale, with both scales explicitly named in the record.
4. Confirmed the imaging date against the payer's twenty-four-month window and documented the confirmation in the case record.
5. Confirmed zero prior injections in the cervical/thoracic region in the rolling twelve months.
6. Reassembled the packet in reviewer order: concordance statement, imaging report, conservative care record with dates, baseline scores. Total packet length, eleven pages — down from the fifty-three originally submitted.

Resolution strategy

The team did not appeal. It resubmitted with the corrected record, on the theory that a clean first-level submission moves faster than an appeal of a denial that was, on the record as submitted, correct. The peer-to-peer was cancelled. The request was approved without clinical review. The physician spent zero additional minutes on the payer.

Outcome

• Authorization obtained; procedure performed at the authorized level and side.
• At the payer's follow-up interval, the coordinator captured percentage relief and duration on the same named scales used at baseline, using the payer-branched follow-up script. The repeat authorization, when it became necessary, approved on first submission.
• The practice reviewed the case at its monthly criterion-level denial meeting and made three template changes: a required concordance statement field in the procedure planning note, a structured conservative care block requiring modality, provider, dates, and outcome, and a hard-stop requiring the standard pain and function scales at the initial visit.
• Outside records retrieval was moved from prior authorization to referral intake.

Lessons learned

1. The denial was correct on the record submitted, and the treatment was correct on the clinical facts. Both things were true at once. That is the ordinary condition of a CILESI denial, and recognizing it is what prevents a practice from burning physician time arguing merit.
2. The peer-to-peer would have failed, because the physician would have been arguing that the patient had radiculopathy — which the payer never disputed — against criteria about documentation, which the physician could not satisfy verbally.
3. Resubmission beat appeal. When the gap is a missing document rather than a disputed judgment, supply the document.
4. The imaging window was luck. The MRI happened to be in window for this payer. Under a different payer's clock it would not have been, and the practice would have discovered that at week six with a patient already scheduled. Luck is not an imaging policy.
5. Everything the payer wanted already existed. The PT happened. The education happened. The psychosocial screen happened. The naproxen trial happened. None of it was written down in a form a reviewer could verify. This is the characteristic shape of a CILESI denial, and it is a template problem with a template solution.

Back to procedure guide navigation ↑

20

Optimization, FAQs, and Outlook

GoHealthcare Best Practices

Domain	Practice	Why it works
Scales	Adopt one pain scale and one function scale practice-wide. Embed in intake, procedure note, and follow-up script. Forbid substitution.	Medicare requires the same scale at baseline and follow-up. Scale drift silently destroys repeat authorizations.
Concordance	Require a one-sentence concordance statement naming symptom, sign, image, level, and side before any request leaves the practice.	It is the sentence a reviewer needs and the sentence clinical reasoning does not naturally produce.
Imaging	Flag imaging at 17 months, not 24. Record the payer's window in the case at eligibility verification.	Build to the tightest clock in the payer mix (Carelon, 18 months for HNP) and the looser ones take care of themselves.
Records	Move outside records retrieval from prior authorization to referral intake.	The conservative care happened elsewhere. Retrieval at PA is an urgent request against a scheduled date.
Utilization	Track sessions by region (cervical/thoracic as one) on a rolling 12-month window. Surface the count at scheduling.	62321, 64479, and 64480 all draw from one allowance. Level-based counting fails silently.
QA gate	One accountable reviewer, one checklist, before every submission.	Highest-return control in the workflow. Minutes pre-submission versus weeks post-denial.
Follow-up	Payer-branched follow-up script capturing percentage AND duration on the baseline scale.	eviCore's threshold is 2 weeks; Carelon's and Medicare's is 3 months; UHC has a partial-responder pathway. One script cannot serve all three from memory.
Peer-to-peer	No P2P is scheduled until the coordinator has identified the criterion at issue and the record location that satisfies it.	A P2P is a signal that an element was not established. Send the physician to close a criterion, not to argue merit.
Denial analysis	Report monthly by criterion, not by payer. Output = template changes.	Payer-level reporting locates pain. Criterion-level reporting locates cause.
Policy currency	Named owner; quarterly review of every governing policy with version and effective date recorded.	Every major policy for this procedure revised within 18 months of publication.
12-month review	At month 12, run the Medicare continuation checklist — including PCP notification.	A stated criterion almost nobody meets, and a stated trigger for focused medical review.
Steroid tracking	Cumulative steroid dose field in the procedure note, running by region.	Carelon requires primary care provider notification of prolonged repeat steroid use; fixed cumulative-dose limits were removed effective 06/14/2026.
Packet design	Assemble in reviewer order. Cap length. Never send the full chart.	Findability drives auto-approval. Volume drives clinical review.
Credentialing	Annual audit against the LCD provider qualification criterion; hold audit-ready.	Provider qualification is a coverage criterion, examined in post-payment review.
AI governance	Automate rules and completeness checking. Never automate clinical assertions.	A generated concordance statement is an unverified claim about a patient.

Back to procedure guide navigation ↑

21

Optimization, FAQs, and Outlook

Common Mistakes

Clinical and selection

• Submitting for axial neck pain because the ASIPP evidence supports it. Every payer in this guide excludes it (Section 17.1).
• Treating the request as an argument about whether the patient has radiculopathy. The payer almost never disputes that.
• Missing myelopathy or a progressive deficit that requires surgical rather than interventional referral.
• Proposing a predetermined series. No payer authorizes one.

Documentation

• 'Failed conservative therapy' with no modality, provider, dates, or outcome.
• A pain score with no named scale.
• Scale drift — VAS at baseline, NRS at follow-up.
• No functional assessment instrument at all, only a pain number.
• An examination that describes findings without naming the root.
• Documenting outcome as to neck pain when the covered indication is radicular pain.
• Omitting the contraindication screen and assuming silence reads as clearance.
• No documentation of rehabilitation participation — a stated Medicare expectation and half of eviCore's comprehensive program requirement.

Imaging

• Running one imaging clock across a mixed payer panel.
• Sending the impression line instead of the full report.
• Submitting multi-level imaging with no sentence selecting the symptomatic level.
• Requesting a level the radiologist did not read.
• Ordering a repeat MRI Carelon does not require for a repeat injection on a fixed non-HNP lesion.
• Using ultrasound guidance for a contrast-allergic patient at a payer with no ultrasound exception.

Coding and billing

• Modifier 50 on 62321. It is not a bilateral procedure.
• KX on 62321 as a general medical-necessity attestation. It is the DSNRB distinguisher, and aberrant use may trigger focused medical review.
• Billing fluoroscopy separately. It is bundled.
• 62320 by template default when fluoroscopy was used.
• A catheter code for a single-shot injection.
• Cervical and thoracic injections on the same date — one region per session, and they are the same region.
• M54.12 where the record supports M50.123.

Workflow and organization

• Discovering the payer's rules at submission instead of at eligibility verification.
• Counting injections by level instead of by region.
• Letting pends age into denials.
• Sending fifty-page packets.
• Batching multiple requests — a stated Carelon peer-to-peer trigger.
• Scheduling before the authorization is active, or scheduling a level the authorization does not cover.
• Responding to denials with better appeal letters instead of better templates.
• Holding Medicare documentation to a lower standard than commercial documentation because nobody asked for it prospectively.

Back to procedure guide navigation ↑

22

Optimization, FAQs, and Outlook

Pearls and Pitfalls

Pearls

• The four-element rule. Named-scale baseline, dated conservative history, concordant in-window imaging, and — for repeats — percentage relief with duration on the same scale. Nearly every criterion in the United States for this procedure is satisfied by those four.
• One sentence beats three pages. The concordance statement is the highest-leverage sentence in the record.
• Carelon's two-week pathway exists and is unlocked by examination specificity, not advocacy — pain along a named root with dermatomal sensory change, myotomal weakness, and/or reflex abnormality.
• Carelon requires no repeat imaging for a fixed non-HNP lesion. Stop ordering the MRI the policy does not ask for.
• UnitedHealthcare has a partial-responder pathway. Less than 50 percent relief for under three months still authorizes a subsequent injection with a documented reassessment of the patient, the injection site, and the technique. Neither Medicare nor Carelon offers this.
• UnitedHealthcare's conservative care standard is disjunctive — any one of pharmacotherapy, activity modification, or PT. Triage accordingly.
• Medicare covers intolerance as well as failure of conservative care. A second branch to the same criterion.
• Medicare Advantage generally cannot be more restrictive than the LCD for a procedure with no NCD. Cite the LCD by number and revision date.
• AAN 2025 supersedes AAN 2007 for cervical. Where a denial rests on the 2007 Level U statement, Armon et al., Neurology 2025;104(5):e213361 is directly responsive.
• Resubmit rather than appeal when the gap is a missing document rather than a disputed judgment.

Pitfalls

• Lumbar muscle memory. Two levels, bilateral, modifier 50 — all normal lumbar behavior, all wrong cervically.
• The region trap. Cervical and thoracic are one region. A thoracic injection in March costs a cervical injection in July.
• The scale trap. Different instruments at baseline and follow-up make relief uncomputable.
• The evidence trap. Citing a payer's own literature review back to it. Reviewers apply the coverage rationale, not the evidence section (the ultrasound example in Section 8.2 is the cleanest illustration).
• The zoster trap. Medicare covers acute zoster-associated pain and waives the four-week wait; eviCore expressly excludes post-herpetic neuralgia. A Medicare-covered indication is not a commercial one.
• The eviCore episode trap. Three per episode per region in six months sits inside the annual four and is a separate, tighter limit.
• The KX trap. A modifier that looks like a medical-necessity attestation, is not one, and whose aberrant use may trigger focused medical review.
• The sedation trap. Usually unnecessary, rarely indicated, routinely denied, and the justification must be contemporaneous rather than retrospective.
• The biologic trap. A biologic in the injectate may deny the entire claim.
• The audit trap. No prior authorization under fee-for-service Medicare does not mean no criteria. It means a deferred, extrapolated, and unfixable version of the same review.
• The stale binder trap. Every governing policy for this procedure revised within eighteen months of publication.

Back to procedure guide navigation ↑

23

Optimization, FAQs, and Outlook

Frequently Asked Questions

1. Is there a National Coverage Determination for cervical epidural steroid injection?

No. There is no NCD for epidural steroid injections. Coverage under traditional Medicare is determined entirely by the Medicare Administrative Contractor's Local Coverage Determination applicable to the place of service, within the aligned LCD family titled Epidural Steroid Injections for Pain Management.

2. Does traditional fee-for-service Medicare require prior authorization for CPT® 62321?

No. But the LCD states expressly that compliance may be monitored and addressed through post-payment data analysis and subsequent medical review audits. The criteria are identical; only the timing of enforcement differs — and the retrospective version carries extrapolation risk and no opportunity to correct the record.

3. How many cervical interlaminar ESIs can a patient have in a year?

A maximum of four ESI sessions per spinal region per rolling twelve-month period, regardless of the number of levels involved. This limit is common to Medicare, Carelon, UnitedHealthcare, and Aetna. eviCore applies the same annual cap plus a tighter internal limit of three sessions per episode of pain per region in six months. UnitedHealthcare defines the year as running from the date of service of the first approved injection.

4. Do cervical and thoracic injections count separately toward that limit?

No, and this is the most commonly missed rule in the entire procedure. The Medicare billing and coding article defines an anatomic spinal region for epidurals as cervical/thoracic (CPT® 62321, 64479, 64480) or lumbar/sacral (62323, 64483, 64484). Carelon uses the same definition. A thoracic interlaminar injection, a cervical interlaminar injection, and a cervical or thoracic transforaminal injection all draw from the same pool of four.

5. Can a cervical interlaminar ESI be performed bilaterally?

No. It is not medically reasonable and necessary to perform caudal or interlaminar ESIs bilaterally, and the Medicare billing and coding article states that 62321 and 62323 are not bilateral procedures. Modifier 50 does not apply to 62321. Modifier 50 is appropriate for bilateral transforaminal injections (64479-64484), which is exactly why the error is so common in mixed epidural schedules.

6. How many levels can be injected in one session?

One, for the interlaminar approach. Caudal and interlaminar ESIs involving a maximum of one level are medically reasonable and necessary. Transforaminal injections may involve up to two levels in one region, though the LCD notes most conditions would not ordinarily require two.

7. Can two spinal regions be injected on the same date?

No. It is not medically reasonable and necessary for more than one spinal region to be injected in the same session, and the same rule appears across eviCore and Carelon. Note again that cervical and thoracic are one region, so a same-day cervical and thoracic injection fails twice over.

8. Is fluoroscopy or CT guidance actually required, or is it just recommended?

Required, as a condition of coverage. ESIs must be performed under CT or fluoroscopic guidance with contrast. Injections performed without image guidance, or by ultrasound, are not reasonable and necessary except for a documented contraindication to contrast media. This is a bright line, not a preference.

9. When is ultrasound guidance acceptable?

Under Medicare, ultrasound without contrast may be considered where there is a documented allergy to low molecular weight nonionic contrast, or where the patient is pregnant. That accommodation is not universal. eviCore considers ultrasound-guided ESI not medically necessary; UnitedHealthcare considers it unproven and not medically necessary, a position it reaffirmed in June 2026 notwithstanding recent meta-analytic data showing lower intravascular injection rates with ultrasound.

10. Can fluoroscopic guidance be billed separately with 62321?

No. Imaging guidance is included in 62321 and is not separately reportable. This is a coding error, not an optimization.

11. What is the difference between 62320 and 62321?

62320 is the interlaminar cervical or thoracic epidural injection without imaging guidance; 62321 is the same procedure with imaging guidance (fluoroscopy or CT). Because image guidance is a condition of coverage, 62321 is the code for a cervical ESI performed to the standard of care. Reporting 62320 by template default when fluoroscopy was in fact used converts a covered service into a non-covered one.

12. Is a diagnostic block required before a cervical interlaminar ESI?

No. A diagnostic selective nerve root block is a distinct transforaminal procedure with its own criteria, used to identify a symptomatic level or to assist surgical planning. No payer reviewed requires one before an interlaminar ESI.

13. How much conservative care is required, and for how long?

Four weeks is the common denominator. Medicare requires four weeks of pain duration and either four weeks of failed non-invasive conservative care or documented inability to tolerate it. eviCore requires four weeks of failed conservative treatment and participation in a comprehensive pain management program including physical therapy, patient education, psychosocial support, and oral medications. Carelon requires physical therapy plus at least one complementary modality for four weeks, reducible to two weeks with clear objective radiculopathy. UnitedHealthcare requires four weeks unresponsive to any one of pharmacotherapy, activity modification, or physical therapy/home exercise.

14. How current must the MRI be?

It depends entirely on the payer, and this is a live operational risk. eviCore requires advanced imaging within 24 months for cervical/thoracic interlaminar and transforaminal ESI — for initial requests and again for repeats. Carelon requires imaging within 18 months for herniated disc, and requires no repeat imaging at all for a fixed non-HNP lesion at the target level on a repeat request. Medicare specifies concordance and review rather than a numeric window. Flag at 17 months and the tightest clock is covered.

15. Can EMG substitute for MRI?

For UnitedHealthcare, yes — the criterion is evidence of structural and/or functional nerve root involvement by imaging or electrodiagnostic studies. eviCore accepts electrodiagnostic studies diagnostic of nerve root compression within the prior 24 months as an alternative to a concordant imaging interpretation for the radiculopathy definition — but note that eviCore separately and independently requires advanced diagnostic imaging within 24 months for cervical and thoracic epidurals. For cervical ESI at eviCore, EMG does not substitute for the imaging requirement.

16. What relief is required before a repeat injection?

This is the sharpest divergence in the guide. Medicare and Carelon: at least 50 percent relief sustained for at least three months (Medicare permits 50 percent functional improvement on the same scale as an alternative; Carelon requires both pain reduction and functional improvement). eviCore: 50 percent or greater relief of radicular pain for two or more weeks, plus either increased function or reduced medication/services, with at least 14 days since the prior injection. UnitedHealthcare: either 50 percent or more relief for three or more months, or 50 percent or less relief for under three months with a documented reassessment of the patient, the injection site, and the technique.

17. Our patient had 60 percent relief for eight weeks. Is that a repeat?

It depends on the payer, and that is the whole point. At eviCore, yes — comfortably. At Carelon or under Medicare, no — eight weeks is short of three months. At UnitedHealthcare, potentially yes via the partial-responder pathway with a documented reassessment. The same clinical fact produces three different answers. This is why the follow-up script must branch by payer.

18. What if the first injection did not work at all?

Under Medicare, if the patient fails to respond well to the initial ESI, a repeat may be performed after 14 days using a different approach, level, and/or medication if appropriate, with the rationale and medical necessity documented. Carelon permits a one-time second injection no sooner than two weeks after a failed initial, also using a different approach, level, or steroid, with documented rationale. This is a distinct pathway from the repeat-for-success pathway and is frequently overlooked.

19. Can we schedule a series of three?

No. It is not medically reasonable and necessary to prescribe a predetermined series of ESIs. eviCore states there is insufficient evidence to support a 'series of three' in either a diagnostic or therapeutic approach and will not preservice authorize a series; each date of service requires a separate request, with the decision driven by the response to the prior injection.

20. Is sedation or monitored anesthesia care covered?

Usually not. Moderate or deep sedation, general anesthesia, and MAC are usually unnecessary or rarely indicated for these procedures and are therefore not considered medically reasonable and necessary. The LCD notes that even for needle phobia and anxiety, oral anxiolytics typically suffice. Individual consideration may be available on appeal where necessity is unequivocal and clearly documented — but the documentation must be contemporaneous.

21. Can we combine an ESI with a facet injection on the same day for patient convenience?

No. Multiple blocks — ESI, sympathetic blocks, facet blocks, trigger point injections — in the same session as an ESI are not medically reasonable and necessary. The one recognized exception across Medicare, eviCore, and Carelon involves a facet joint synovial cyst: a transforaminal injection combined with intra-articular facet synovial cyst aspiration and steroid injection may be performed together.

22. Is cervical ESI covered for axial neck pain?

No, at any payer in this guide, notwithstanding that ASIPP 2021 assigns Level II evidence with a moderate to strong recommendation for cervical interlaminar injection in axial discogenic pain without facet joint pain. Medicare lists axial spine pain as investigational; Carelon lists isolated axial neck pain as not medically necessary; eviCore lists axial spinal pain as not medically necessary; UnitedHealthcare's rationale covers Radicular Back Pain only. See Section 17.1 for how to handle this.

23. Is cervical ESI covered for cervicogenic headache?

Not under the Medicare LCD, which lists cervicogenic headaches as investigational and therefore not medically reasonable and necessary. The LCD's own evidence review acknowledges that many interventional pain specialists consider epidural injection a reasonable option for this indication after conservative failure, and the expert panel voted 3 of 5 that evidence supports it. The coverage language governs.

24. Can a Medicare Advantage plan apply criteria stricter than the LCD?

Generally no, for this procedure. MA organizations must follow applicable NCDs and LCDs. Because there is no ESI NCD, the LCD applicable to the place of service governs. CMS permits an MA organization to create its own coverage determination only in the absence of an NCD, LCD, or other Medicare coverage guidance (Medicare IOM Pub. 100-16, Ch. 4, §90.5). Where an MA denial applies a commercial or vendor criterion that conflicts with the LCD, cite the LCD by number and revision effective date and state the conflict expressly. This is among the most reliable MA appeal arguments in interventional pain.

25. What documentation is tested in a Medicare post-payment audit?

All documentation must be in the medical record and available on request; every page legible with patient identification and dates of service and the legible signature of the responsible practitioner; the record must support the ICD-10-CM code and the CPT®/HCPCS code reported; and films documenting final needle position and contrast flow — a minimum of two views — must be retained and available on request. The record should also contain the performing provider's assessment, relevant history, pertinent test results, a signed and dated office visit record and operative report, and documentation supporting medical necessity.

26. What triggers a focused medical review?

The LCD names two triggers explicitly: frequent continuation of epidural steroid injections beyond twelve months, and aberrant use of the KX modifier. Both are avoidable and both should be monitored as active operational risk indicators.

27. What does the twelve-month rule actually require?

Medicare states it generally would not be reasonable and necessary for ESI treatment to extend beyond twelve months. Continuation beyond twelve months requires all of: pain severe enough to cause a significant degree of functional or vocational disability; at least 50 percent sustained improvement in pain and/or 50 percent objective functional improvement on the same scale used at baseline; a documented rationale, such as high surgical risk, the patient not desiring surgery, or recurrence of pain in the same location relieved by ESI for at least three months; and notification of the primary care provider regarding continuation of procedures and prolonged repeat steroid use. The PCP notification element is almost universally missed.

28. Are biologics ever acceptable in the injectate?

No. There are no FDA-approved biologicals for injection into the epidural space. Amniotic and placenta-derived injectates, platelet-rich plasma, stem cells, and vitamins fall in this category, and their inclusion may result in denial of the entire claim, not merely the biologic component. eviCore and Carelon carry parallel exclusions.

29. Why is the interlaminar approach preferred cervically when transforaminal dominates lumbar practice?

Safety. The radicular arteries that perfuse the spinal cord course adjacent to the targeted root within the neuroforamen. Published safety analyses cited in the Medicare LCDs report that cervical and thoracic transforaminal epidurals constitute roughly 2.4 percent of all epidural injections and fewer than 5 percent of all transforaminal injections, yet contribute more than 99 percent of complications related to intra-arterial injection of particulate steroid. A 2015 safety review concluded cervical TFESIs must not be performed until appropriate evidence develops and safe steroid preparations are available. This asymmetry does not exist in the lumbar spine.

30. Our denial cites the 2007 AAN statement that cervical evidence is inadequate. Is that still current?

No, and this is the single most useful evidence citation in cervical ESI appeals. The AAN Guidelines Subcommittee published an updated systematic review summary in 2025 (Armon et al., Neurology 2025;104(5):e213361) concluding that epidural steroids probably provide short-term pain reduction and probably reduce short-term disability in cervical radiculopathy, at moderate confidence — superseding the 2007 Level U position. Several payer policies, including Aetna's CPB 0016, still quote the 2007 statement. Where a denial rests on it, the 2025 update is directly responsive. Pair it with ASIPP 2021 Level I for cervical interlaminar injection in disc herniation.

Back to procedure guide navigation ↑

24

Optimization, FAQs, and Outlook

Key Takeaways, Future Outlook, and References

1. There is no NCD. Traditional Medicare coverage is governed entirely by the aligned MAC LCD family. The LCD follows the place of service, not the practice's headquarters.
2. Fee-for-service Medicare does not pre-authorize; it audits. Same criteria, deferred enforcement, extrapolation risk, and no chance to fix the record. Hold Medicare documentation to the higher standard, not the lower one.
3. Cervical and thoracic are one region. 62321, 64479, and 64480 all draw from four sessions per rolling twelve months. Track by region, not by level.
4. 62321 is not bilateral, one level per session, one region per session. Modifier 50 does not apply. Lumbar habits do not transfer.
5. Image guidance with contrast is a condition of coverage, and it is bundled into 62321. Ultrasound is accommodated by Medicare for contrast allergy or pregnancy and by essentially nobody else.
6. The four-element rule carries the procedure. Named-scale baseline, dated conservative history, concordant in-window imaging, and percentage relief with duration on the same scale for repeats.
7. 'The same scale' is enforced literally. Standardize one pain scale and one function scale practice-wide and forbid substitution. This single step protects the repeat-injection revenue stream.
8. Repeat thresholds diverge sharply. eviCore: 50 percent for two weeks. Medicare and Carelon: 50 percent for three months. UnitedHealthcare: a partial-responder pathway available with a documented reassessment. Branch the follow-up script by payer.
9. Imaging clocks diverge too. eviCore 24 months, Carelon 18 months for HNP, no repeat imaging for a fixed non-HNP lesion. Flag at 17 months.
10. Axial neck pain is excluded by every payer, notwithstanding ASIPP Level II evidence. Screen for it before the visit; do not manufacture radiculopathy; do not spend physician time appealing it.
11. Most denials are documentation failures, not clinical disagreements. Invest upstream in templates, intake, records retrieval, and pre-submission QA — not downstream in appeal capacity.
12. A peer-to-peer is a signal that an element was not established. Track the criterion at issue and fix the template.
13. Policy is perishable. Every governing policy for this procedure revised within eighteen months of publication. Name an owner and run a quarterly cycle.
14. Two stated criteria almost nobody meets: primary care provider notification at twelve months, and primary care provider notification of prolonged repeat steroid use (Carelon; fixed cumulative-dose limits removed 06/14/2026).
15. AAN 2025 supersedes AAN 2007 for cervical. It is the strongest available evidence citation when a denial rests on the old Level U position.

Future Outlook

• The cervical evidence base is strengthening. The AAN's 2025 update — moving from 'evidence inadequate for any recommendation' in 2007 to 'probably provides short-term pain reduction and probably reduces short-term disability, moderate confidence' — is the most significant shift in cervical ESI evidence in nearly two

decades. Expect it to surface in payer policy narratives over the coming revision cycles. Expect it to move coverage language slowly, if at all, because the covered indication set is already radicular.

• Countervailing evidence is also strengthening. The 2025 BMJ network meta-analysis of interventional procedures for chronic non-cancer spine pain (Wang et al.) found that no commonly performed interventional procedure demonstrated convincing evidence of important pain relief or functional improvement for axial or radicular chronic spine pain, with moderate certainty of little to no effect in many instances. UnitedHealthcare has already incorporated it. Practices should expect the criteria to tighten, not loosen, and should not build operating models on the assumption that utilization limits will relax.
• Ultrasound guidance is the live technical question. Recent meta-analytic data show materially lower intravascular injection rates and shorter procedure times with ultrasound. Payer policy has not moved and, in UnitedHealthcare's case, expressly declined to move in June 2026. Watch for divergence between the safety literature and the coverage position; this is the most likely place for a policy shift in the next two to three years, and it will likely arrive first as a narrow exception rather than as a general acceptance.
• Utilization management consolidation continues. The practical determinant of a commercial authorization is increasingly the delegated vendor's guideline rather than the plan's own policy. Practices should organize their payer intelligence around vendors, not around plan names.
• Documentation-derived automation is the near-term operational frontier. The deterministic checks in this workflow — imaging windows, region-based session counts, scale consistency, packet completeness — are precisely where practices lose money and precisely where rule-based automation performs well without touching clinical judgment. GoHealthcare's position remains that AI should be deployed against rules and completeness, with a clinician authoring every clinical assertion.
• Steroid exposure is an emerging documentation frontier. Carelon's 2026 shift from a fixed cumulative-dose cap to a prolonged-use PCP notification and Medicare's PCP notification requirement both point in the same direction: payers are beginning to manage cumulative corticosteroid burden, not just procedure counts. Expect this to expand.

References

All web addresses were verified as of July 2026. Coverage policies revise on independent cycles; verify the live document before relying on any criterion summarized in this guide.

Centers for Medicare & Medicaid Services

• Centers for Medicare & Medicaid Services. Local Coverage Determination: Epidural Steroid Injections for Pain Management (L39240). Noridian Healthcare Solutions, Jurisdictions E and F. Original effective 06/19/2022; revision effective 04/09/2026. Website: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=39240
• Centers for Medicare & Medicaid Services. Local Coverage Determination: Epidural Steroid Injections for Pain Management (L39054). Website: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?LCDId=39054
• Centers for Medicare & Medicaid Services. Local Coverage Determination: Epidural Steroid Injections for Pain Management (L39242). Website: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=39242
• Centers for Medicare & Medicaid Services. Local Coverage Determination: Epidural Steroid Injections for Pain Management (L39036). Website: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=39036
• Centers for Medicare & Medicaid Services. Local Coverage Determination: Epidural Steroid Injections for Pain Management (L38994). Website: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=38994
• Centers for Medicare & Medicaid Services. Local Coverage Determination: Epidural Steroid Injections for Pain Management (L33906). Website: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=33906
• Centers for Medicare & Medicaid Services. Local Coverage Determination: Epidural Steroid Injections for Pain Management (L36920). Website: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?LCDId=36920
• Centers for Medicare & Medicaid Services. Billing and Coding: Epidural Steroid Injections for Pain Management (A58777). Original effective 12/05/2021; revision effective 11/30/2023. Website: https://www.cms.gov/medicare-coverage-database/view/article.aspx?articleid=58777
• Centers for Medicare & Medicaid Services. Billing and Coding: Epidural Steroid Injections for Pain Management (A58995). Website: https://www.cms.gov/medicare-coverage-database/view/article.aspx?articleId=58995
• Centers for Medicare & Medicaid Services. Response to Comments: Epidural Steroid Procedures Injections for Pain Management (A59079). Website: https://www.cms.gov/medicare-coverage-database/view/article.aspx?articleId=59079
• Centers for Medicare & Medicaid Services. Medicare Program Integrity Manual (IOM Pub. 100-08), Chapter 13, Section 13.5.4, Reasonable and Necessary Provision in an LCD. Website: https://www.cms.gov/regulations-and-guidance/guidance/manuals/internet-only-manuals-ioms-items/cms019033
• Centers for Medicare & Medicaid Services. Medicare Benefit Policy Manual (IOM Pub. 100-02), Chapter 15, Section 50.2 (Self-Administered Drug determinations) and Chapter 16, Section 180. Website: https://www.cms.gov/Regulations-and-Guidance/Guidance/Manuals/Downloads/bp102c15.pdf
• Centers for Medicare & Medicaid Services. Medicare Coverage Database (search portal for the LCD applicable to your jurisdiction). Website: https://www.cms.gov/medicare-coverage-database
• Centers for Medicare & Medicaid Services. Medicare Learning Network (MLN) Provider Education. Website: https://www.cms.gov/training-education/medicare-learning-network
• Social Security Act, Title XVIII, Section 1862(a)(1)(A) (reasonable and necessary). Website: https://www.ssa.gov/OP_Home/ssact/title18/1862.htm

Utilization management organizations

• eviCore by EVERNORTH. Cigna Medical Coverage Policies — Musculoskeletal: Epidural Steroid Injections (CMM-200). Comprehensive Musculoskeletal Management Guidelines V1.0.2026. Effective 08/04/2026. Website: https://www.evicore.com/provider/clinical-guidelines
• Carelon Medical Benefits Management, Inc. Clinical Appropriateness Guidelines — Musculoskeletal: Appropriate Use Criteria: Interventional Pain Management. Doc ID MSK01-0626.1. Effective 06/14/2026; last review 10/09/2025. Website: https://guidelines.carelonmedicalbenefitsmanagement.com/interventional-pain-management-2026-06-14/
• Carelon Medical Benefits Management, Inc. Clinical Guidelines and Pathways (current and archived guideline library). Website: https://guidelines.carelonmedicalbenefitsmanagement.com
• eviCore by EVERNORTH. Musculoskeletal Program Frequently Asked Questions. Website: https://www.evicore.com/resources
• Cohere Health. Clinical Guidelines and Provider Resources. Website: https://coherehealth.com
• Evolent (formerly National Imaging Associates). Clinical Guidelines. Website: https://www.evolent.com
• TurningPoint Healthcare Solutions. Clinical Guidelines and Provider Resources. Website: https://www.turningpoint-healthcare.com
• NantHealth / NaviNet. NaviNet Open provider portal (authorization workflow and submission). Website: https://navinet.navimedix.com

Commercial payers

• UnitedHealthcare. Epidural Steroid Injections for Spinal Pain. Commercial and Individual Exchange Medical Policy 2026T0616N. Effective 06/01/2026. Website: https://www.uhcprovider.com/content/dam/provider/docs/public/policies/comm-medical-drug/epidural-steroid-injections-spinal-pain.pdf
• UnitedHealthcare. Medical and Drug Policies and Coverage Determination Guidelines. Website: https://www.uhcprovider.com/en/policies-protocols.html
• Aetna. Clinical Policy Bulletin 0016: Back Pain — Invasive Procedures. Website: https://www.aetna.com/cpb/medical/data/1_99/0016.html
• Aetna. Clinical Policy Bulletin 0722: Transforaminal Epidural Injections. Website: https://www.aetna.com/cpb/medical/data/700_799/0722.html
• Cigna Healthcare. Medical Coverage Policies. Website: https://www.cigna.com/health-care-providers/coverage-and-claims/policies
• Elevance Health (Anthem). Medical Policies and Clinical Utilization Management Guidelines. Website: https://www.anthem.com/provider/policies/
• Humana. Medical and Pharmacy Coverage Policies. Website: https://www.humana.com/provider/medical-resources/clinical/coverage-policies
• Blue Cross Blue Shield Association. Evidence Street and member plan medical policy libraries (policy varies by plan and market). Website: https://www.bcbs.com
• Molina Healthcare. Medical Coverage Policies. Website: https://www.molinahealthcare.com/providers/common/medicaid/medical-policies.aspx
• Centene / WellCare. Clinical Policies. Website: https://www.centene.com/providers/clinical-policy.html
• Kaiser Permanente. Clinical Review Criteria. Website: https://healthy.kaiserpermanente.org

Professional organizations and clinical guidelines

• Manchikanti L, Knezevic NN, Navani A, et al. Epidural Interventions in the Management of Chronic Spinal Pain: American Society of Interventional Pain Physicians (ASIPP) Comprehensive Evidence-Based Guidelines. Pain Physician. 2021;24(S1):S27-S208. Website: https://www.painphysicianjournal.com
• Armon C, Narayanaswami P, Potrebic S, et al. Epidural steroids for cervical and lumbar radicular pain and spinal stenosis systematic review summary: report of the AAN Guidelines Subcommittee. Neurology. 2025 Mar;104(5):e213361.

Website: https://www.neurology.org

• American Society of Interventional Pain Physicians (ASIPP). Website: https://www.asipp.org
• American Society of Pain and Neuroscience (ASPN). Website: https://aspnpain.com
• North American Spine Society. Coverage Policy Recommendations: Epidural Steroid Injections and Selective Spinal Nerve Blocks. 2020. Burr Ridge, IL. Website: https://www.spine.org
• North American Spine Society. Evidence-Based Clinical Guidelines for Multidisciplinary Spine Care: Diagnosis and Treatment of Cervical Radiculopathy from Degenerative Disorders. 2010. Burr Ridge, IL. Website: https://www.spine.org
• North American Spine Society. Choosing Wisely recommendations on image guidance for elective spinal injections. Website: https://www.choosingwisely.org
• Spine Intervention Society (formerly ISIS) / International Pain and Spine Intervention Society (IPSIS). Bogduk N, ed. Practice Guidelines for Spinal Diagnostic and Treatment Procedures, 2nd ed., and Safety Practices for Interventional Pain Procedures. 2021. Website: https://www.spineintervention.org
• American Academy of Orthopaedic Surgeons (AAOS). Website: https://www.aaos.org
• American Association of Neurological Surgeons (AANS). Website: https://www.aans.org
• Congress of Neurological Surgeons (CNS). Website: https://www.cns.org
• International Neuromodulation Society (INS). Website: https://www.neuromodulation.com
• American Society of Regional Anesthesia and Pain Medicine (ASRA). Website: https://www.asra.com
• American Medical Association. CPT® Professional Edition (current year). Website: https://www.ama-assn.org/practice-management/cpt
• American College of Occupational and Environmental Medicine. Invasive Treatments for Low Back Disorders. 2021. Website: https://acoem.org
• Rathmell JP, Benzon HT, Dreyfuss P, et al. Safeguards to prevent neurologic complications after epidural steroid injections: consensus opinions from a multidisciplinary working group and national organizations (FDA Safe Use Initiative). Anesthesiology. 2015;122(5):974-984. Website: https://pubs.asahq.org/anesthesiology/article/122/5/974/12571

Peer-reviewed literature

• Cohen SP, Hayek S, Semenov Y, et al. Epidural steroid injections, conservative treatment, or combination treatment for cervical radicular pain: a multicenter, randomized, comparative-effectiveness study. Anesthesiology. 2014;121(5):1045-1055.
• Manchikanti L, Nampiaparampil DE, Candido KD, et al. Do cervical epidural injections provide long-term relief in neck and upper extremity pain? A systematic review. Pain Physician. 2015;18(1):39-60.
• Diwan S, Manchikanti L, Benyamin RM, et al. Effectiveness of cervical epidural injections in the management of chronic neck and upper extremity pain. Pain Physician. 2012;15(4):E405-E434.
• Conger A, Cushman DM, Speckman RA, Burnham T, Teramoto M, McCormick ZL. The effectiveness of fluoroscopically guided cervical transforaminal epidural steroid injection for the treatment of radicular pain: a systematic review and meta-analysis. Pain Medicine. 2020;21(1):41-54.
• Bicket MC, Chakravarthy K, Chang D, Cohen SP. Epidural steroid injections: an updated review on recent trends in safety and complications. Pain Management. 2015;5(2):129-146.
• Wang X, Martin G, Sadeghirad B, et al. Common interventional procedures for chronic non-cancer spine pain: a systematic review and network meta-analysis of randomised trials. BMJ. 2025;388:e079971.
• Oliveira CB, Maher CG, Ferreira ML, et al. Epidural corticosteroid injections for lumbosacral radicular pain. Cochrane Database of Systematic Reviews. 2020;4(4):CD013577.
• Carette S, Leclaire R, Marcoux S, et al. Epidural corticosteroid injections for sciatica due to herniated nucleus pulposus. New England Journal of Medicine. 1997;336:1634-1640.
• Friedly JL, Comstock BA, Turner JA, et al. A randomized trial of epidural glucocorticoid injections for spinal stenosis. New England Journal of Medicine. 2014;371:11-21.
• Chou R, Hashimoto R, Friedly J, et al. Pain Management Injection Therapies for Low Back Pain. Rockville, MD: Agency for Healthcare Research and Quality; 2015.
• Verheijen EJA, Bonke CA, Amorij EMJ, et al. Epidural steroid compared to placebo injection in sciatica: a systematic review and meta-analysis. European Spine Journal. 2021;30(11):3255-3264.
• Helm S 2nd, Harmon PC, Noe C, et al. Transforaminal epidural steroid injections: a systematic review and meta-analysis of efficacy and safety. Pain Physician. 2021;24(S1):S209-S232.
• Kleimeyer J, Koltsov J, Smuck M, Wood K, Cheng I, Hu S. Cervical epidural steroid injections: incidence and determinants of subsequent surgery. The Spine Journal. 2020;20(11):1729-1736.
• Benzon HT, Huntoon MA, Rathmell JP. Improving the safety of epidural steroid injections. JAMA. 2015;313(17):1713-1714.
• Miranda AM, Marinho FD, Wegner GRM, et al. Ultrasound-guided versus conventional radioscopic-guided transforaminal epidural steroid injections for cervical radicular pain: a systematic review and meta-analysis. Anesthesia & Analgesia. 2025 (online ahead of print).

Regulatory and safety

• U.S. Food and Drug Administration. FDA Drug Safety Communication: FDA requires label changes to warn of rare but serious neurologic problems after epidural corticosteroid injections for pain. April 2014. Website: https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-requires-label-changes-warn-rare-serious-neurologic-problems-after-epidural-corticosteroid-injections-pain
• U.S. Food and Drug Administration. Center for Drug Evaluation and Research. Website: https://www.fda.gov/about-fda/fda-organization/center-drug-evaluation-and-research-cder

Reading Recommendations

For teams building depth in this procedure, GoHealthcare recommends the following in this order. The sequence matters: read the policy that governs your patients before reading the evidence that governs the policy.

1. The LCD applicable to your jurisdiction, in full — including the Summary of Evidence and Analysis of Evidence sections, which most readers skip and which explain why the criteria are what they are.
2. The billing and coding article paired with your LCD (A58777 or A58995). This is the operational document: the utilization parameters, the modifier rules, the documentation requirements, and the covered ICD-10-CM list all live here, not in the LCD.
3. The February 11, 2021 multi-jurisdictional Contractor Advisory Committee transcript and voting results, published on each MAC's website. The single most useful and least read document in interventional pain coverage (see Section 17.2).
4. The current guideline of the utilization management vendor that reviews the majority of your commercial volume — eviCore CMM-200, or the Carelon Interventional Pain Management guideline. Read the version in effect, not the one in your binder.
5. Manchikanti L, Knezevic NN, Navani A, et al. ASIPP Comprehensive Evidence-Based Guidelines. Pain Physician. 2021;24(S1):S27-S208. The reference evidence document for the field.
6. Armon C, Narayanaswami P, Potrebic S, et al. Neurology. 2025;104(5):e213361. The current AAN position, and the appeal citation that matters most for cervical.
7. Rathmell JP, Benzon HT, Dreyfuss P, et al. Anesthesiology. 2015;122(5):974-984. The FDA Safe Use Initiative consensus safeguards that Carelon expects providers to follow.
8. Wang X, Martin G, Sadeghirad B, et al. BMJ. 2025;388:e079971. Read it because your payers are reading it.

Related GoHealthcare Resources

This guide is one page of the GoHealthcare MSK Specialty Procedure Library™. The following related resources in the GoHealthcare Practice Solutions Knowledge Center extend it. Internal links are published at www.GoHealthcareLLC.com.

Batch B02 — Epidural Steroid Injections (companion guides)

• Thoracic Interlaminar Epidural Steroid Injection — shares CPT® 62321 and the cervical/thoracic region allowance with this guide
• Lumbar Interlaminar Epidural Steroid Injection
• Caudal Epidural Steroid Injection
• Cervical Transforaminal Epidural Steroid Injection — the safety and coverage counterpoint to this guide
• Thoracic Transforaminal Epidural Steroid Injection
• Lumbar Transforaminal Epidural Steroid Injection

Related procedure guides

• Cervical Medial Branch Block and Cervical Radiofrequency Ablation
• Cervical Facet Joint Injection
• Diagnostic Selective Nerve Root Block
• Spinal Cord Stimulation Trial and Implant

Knowledge Center resources

• GoHealthcare Prior Authorization Playbook — MSK Specialty
• GoHealthcare Payer Intelligence Center — eviCore, Carelon, Cohere, Evolent, and TurningPoint profiles
• GoHealthcare Medical Necessity and Clinical Documentation Excellence Standards
• GoHealthcare Revenue Cycle Knowledge Center — reimbursement methodology, NCCI edits, MUE values, and payment analysis for epidural procedures
• GoHealthcare Denials and Appeals Library — MSK Specialty
• GoHealthcare Utilization Management Operations Guide
• GoHealthcare AI Governance Framework for Prior Authorization
• GoHealthcare ASC Operations and Compliance Center

Recommended Downloads

• Cervical Interlaminar ESI Clinical Documentation Checklist (Section 14, one page, printable)
• Four-Element Rule Pre-Submission QA Card (one page, for the QA gate)
• Payer Repeat-Threshold Quick Reference (Section 11.2, laminated desk card for the follow-up call)
• Payer Imaging Currency Window Quick Reference (17-month flag card)
• Cervical/Thoracic Region Utilization Tracker (rolling 12-month, region-based)
• Concordance Statement Template and Examples
• Payer-Branched Follow-Up Call Script (captures percentage and duration on the baseline scale)
• CILESI Criterion-Mapping Appeal Template
• Twelve-Month Continuation Review Checklist (including PCP notification)
• Cumulative Steroid Dose Tracking Log (by region)
• Peer-to-Peer Preparation Worksheet (criterion and record location)
• Monthly Criterion-Level Denial Analysis Template

Visual Recommendations

The following visuals are recommended to accompany the web publication of this guide.

Visual	Type	Purpose
Cervical spine and epidural space	High-level anatomy illustration	Orientation only. Show the epidural space, laminae, exiting roots, and the narrowing of the space as it ascends. Deliberately non-procedural — no needle trajectory, no technique.
Cervical dermatome and root-level map	Anatomy illustration	Supports the concordance triangle: disc level to root to dermatome. The single most useful visual for authorization staff.
The concordance triangle	Decision diagram	Symptom / Sign / Image, with the failure modes on each vertex (Section 9.2).

Visual	Type	Purpose
The four-element rule	Callout graphic	The organizing concept of the guide.
CILESI prior authorization workflow	Process map / swimlane	Ten stages by owner (Section 12.1), with the pre-submission QA gate visually emphasized.
Payer coverage matrix — initial ESI	Comparison table	Section 11.1, rendered as a scannable web table.
Payer repeat-threshold comparison	Comparison table / bar visual	Section 11.2. Consider visualizing the 2-week versus 3-month divergence on a timeline — it is the single most memorable graphic in this guide.
Imaging currency windows	Timeline	18-month, 24-month, and the 17-month flag point.
Cervical/thoracic region allowance	Coverage matrix	Show 62321, 64479, and 64480 flowing into a single pool of four.
Approach comparison	Comparison table	Section 15, with the cervical safety asymmetry (2.4% of injections, over 99% of particulate intra-arterial complications) called out.
Documentation checklist	Printable checklist	Section 14, as a downloadable PDF.
Denial reason and prevention map	Decision tree	Section 12.2, mapping each denial reason to its upstream prevention.

Document History

Version	Date	Author	Summary of changes
1.1	July 2026	Pinky Maniri, Founder and CEO, GoHealthcare Practice Solutions	Payer-policy currency sweep (all citations re-verified against live sources). Updates: Carelon Interventional Pain Management — the prior fixed cumulative-steroid-dose limitation was removed effective 06/14/2026 and replaced by a requirement to notify the primary care provider of prolonged repeat steroid use; all references corrected accordingly (a newer Carelon revision effective 09/19/2026 was noted and did not alter the epidural criteria summarized here). eviCore/Cigna CMM-200 updated from V1.1.2025 (eff. 02/25/2026) to V1.0.2026 (eff. 08/04/2026); summarized criteria unchanged. Medicare LCD L39240 revision date corrected to 04/09/2026 (R4). UnitedHealthcare 2026T0616N (eff. 06/01/2026) confirmed current. Underlying clinical evidence, coding rules, and secondary-citation audit unchanged from the prior version.
1.0	July 2026	Pinky Maniri, Founder and CEO, GoHealthcare Practice Solutions	Initial publication. Built to the GoHealthcare Clinical Procedure Guide Standard v1.0. Payer criteria verified against: CMS LCD L39240 (revision effective 04/09/2026) and the aligned MAC LCD family; CMS Billing and Coding Article A58777 (revision effective 11/30/2023); eviCore/Cigna CMM-200 (V1.0.2026, effective 08/04/2026); Carelon Interventional Pain Management (MSK01-0626.1, effective 06/14/2026); UnitedHealthcare 2026T0616N (effective 06/01/2026); Aetna CPB 0016. Evidence current through the AAN Guidelines Subcommittee systematic review summary (Armon et al., Neurology 2025) and ASIPP 2021 comprehensive evidence-based guidelines.

Back to procedure guide navigation ↑

Related Reading

Verified GoHealthcare Resources

Prior Authorization Services

Specialty prior authorization and utilization management support across all 50 states.

https://www.gohealthcarellc.com/prior-authorization-services.html

Pain Management Prior Authorization

Procedure-specific authorization support for interventional pain practices.

https://www.gohealthcarellc.com/pain-management-prior-authorization.html

Prior Authorization and Utilization Management Strategy

GoHealthcare's strategic framework for financial performance, compliance integrity, and scalable operations.

https://www.gohealthcarellc.com/blog/the-future-of-prior-authorization-and-utilization-management

Prior Authorization Case Study

Operational case study covering clinical documentation, workflow control, and authorization execution.

https://www.gohealthcarellc.com/case-study-prior-authorization-clinical-operations.html

GoHealthcare Blog

Operational, prior authorization, revenue cycle, compliance, and MSK specialty healthcare insights.

https://www.gohealthcarellc.com/blog

Contact GoHealthcare

Request a specialty practice assessment or discuss prior authorization operations.

https://www.gohealthcarellc.com/contact-us.html

Strengthen Prior Authorization Performance

GoHealthcare Practice Solutions supports pain management, orthopedic, spine, neurosurgery, neuromodulation, ambulatory surgery center, and hospital outpatient teams with prior authorization, utilization management, medical necessity, clinical documentation, denial prevention, appeals, revenue cycle, compliance, and operational optimization.

Request a Practice AssessmentCall 1 (800) 267-8752

Developed by

Pinky Maniri

MSc, BSc, CRCR, CSAPM, CSPPM, CSBI, CSPR, CSAF, Certified in Healthcare A.I. Governance

Founder and Chief Executive Officer of GoHealthcare Practice Solutions

Connect on LinkedInVisit GoHealthcare

Educational and Operational Disclaimer

EDUCATIONAL AND OPERATIONAL REFERENCE ONLY

This document is intended for educational and operational reference purposes only.

It does not replace physician clinical judgment, payer policy review, legal advice, or official CMS guidance. It is not a procedural technique manual, a physician training guide, or a substitute for specialty society procedural manuals or clinical textbooks. It contains no procedural technique, surgical technique, needle placement instruction, medication dosing guidance, or fluoroscopic guidance instruction, and none should be inferred from it.

Coverage policies, coding guidance, and reimbursement requirements change frequently and vary by payer, plan, product line, jurisdiction, and place of service. All coverage, coding, documentation, and reimbursement requirements should always be verified with the applicable payer and current regulatory sources before any clinical, coding, or billing decision is made.

Payer criteria summarized in this guide are GoHealthcare's synthesis of publicly available policy documents as of July 2026, prepared for educational use. They are not reproductions of payer policy and are not a substitute for reading the governing policy in its original form. Where this guide and a payer policy differ, the payer policy governs.

CPT® is a registered trademark of the American Medical Association. CPT® five-digit codes, descriptions, and other data are copyright of the American Medical Association. All rights reserved. The AMA does not directly or indirectly practice medicine or dispense medical services and assumes no liability for data contained or not contained herein. All other trademarks and policy documents referenced are the property of their respective owners.

This website content is intended for healthcare organizations and professionals and does not provide patient-specific medical advice. Do not submit protected health information or other sensitive patient information through a public webpage.