1
Indications & Patient Selection
Mechanism
- Application of extreme cold (−40°C to −100°C) to peripheral nerve → vasa vasorum injury, endoneural edema, microtubule dissolution → cessation of axonal transport
- Myelinated A-delta and C fibers (pain) destroyed at −20°C; motor fibers require −30°C to −40°C
- Perineurium (structural scaffold) is preserved → nerve regeneration occurs at 1–2 mm/day; effect lasts 3–6 months for benign conditions
- Longer ablation times (≥ 8–10 min × 2 cycles) required for complete Wallerian degeneration; partial ablation may cause allodynia or incomplete relief
Indications
- Oncologic pain — direct nerve involvement: lung cancer (intercostal), pancreatic cancer (celiac plexus), pelvic tumors (sciatic, pudendal, lumbosacral trunk), rib metastases
- Oncologic pain — perilesional (upstream block): intercostal cryoneurolysis for chest wall tumors not directly invading the nerve
- Non-oncologic: post-thoracotomy pain, intercostal neuralgia, occipital neuralgia, greater auricular neuralgia, Morton's neuroma
- MSK: genicular nerves (knee OA), lateral femoral cutaneous nerve / LFCN (meralgia paresthetica)
- Preferred over RFA when near major motor nerves (sciatic, pudendal) — no conduction injury risk; ice ball is visible on CT/US in real time
Contraindications
- Systemic cold intolerance (absolute): cryoglobulinemia, Raynaud's phenomenon, cold agglutinin disease
- Skin overlying target: ice ball must not contact skin (frost injury); hydrodissection required if probe is superficial
- Relative: uncorrected coagulopathy (deep target); active local infection; patient unable to cooperate with positioning during active freeze (cannot pull probe during ice ball formation)
Patient Workup
- Diagnostic nerve block must precede ablation — patient who does not respond to local anesthetic nerve block is unlikely to benefit from cryoneurolysis
- Confirm target anatomy on CT or MRI; review dermatome map; consider neurostimulation if available for accurate identification
- For mixed or motor nerves (sciatic, lumbosacral trunk): counsel patient about expected temporary weakness; arrange physical/occupational therapy and AFO brace pre-procedure for sciatic/peroneal targets
- Manage expectations: acute post-procedure pain exacerbation common for 12–24h; early inflammatory reaction may temporarily worsen pain before improvement
2
Pre-Procedure Checklist
Cryoprobe system. IceCure, Endocare, or Galil (IsoTherm). 17G (1.47 mm) probe for most peripheral nerves; larger 2.4 mm probe for celiac plexus. "IceSeed" 17G for precise nerve targeting. Confirm gas supply (argon for freeze, helium for thaw on some systems).
Guidance modality. CT for deep nerves (celiac plexus, sciatic, intercostal, pudendal, lumbosacral trunk). Ultrasound for superficial nerves (occipital, genicular, LFCN). Ice ball visible on CT as hypodense sphere; on US as hyperechoic structure with posterior shadowing.
Hydrodissection setup. 22G coaxial needle + sterile saline (5–20 mL) to create safety buffer between probe/ice ball and critical structures (pleura for intercostal; skin if probe is superficial; periosteum for LFCN). For skin protection: room air (100 cc) via 19G needle into subcutaneous tissue is an alternative.
Freeze-thaw cycle protocol. Standard: 2 cycles. Cycle 1 — freeze to −40°C for 8–10 min → passive thaw 3–5 min. Cycle 2 — refreeze to −40°C for 8–10 min → passive thaw. Do NOT use active (warm) thaw — causes microvascular tearing. Do NOT skip second cycle — single freeze gives incomplete demyelination.
Anesthesia. Local anesthesia only for most cases (cryo itself provides intraoperative analgesia once ice ball forms). Mild IV sedation (fentanyl + midazolam) optional for probe placement discomfort. General anesthesia for complex nerve targets (lumbosacral trunk, sciatic) — confirm NPO status. Note: minimize sedation if neurostimulation used for targeting (requires patient cooperation).
Consent. Discuss: acute post-procedure pain exacerbation (common 12–24h — may require PCA admission); expected motor weakness for mixed/motor nerves (duration weeks to months); reversibility of effect (nerve regeneration 3–6 months); pneumothorax risk for intercostal procedures (~5–8%); frostbite if ice ball contacts skin; incomplete effect if diagnostic block was false positive; repeat treatment safe if pain recurs.
Post-procedure planning. For motor nerves: arrange physical and occupational therapy consult pre-procedure; AFO brace for foot drop risk (sciatic, peroneal, lumbosacral trunk). For intercostal: plan post-procedure CT for pneumothorax check. For oncologic patients: consider same-day PCA admission for 18–24h pain management.
3
Anatomy — Common Nerve Targets
General Anatomic Principles
- Cryoneurolysis targets peripheral nerves; deep nerve targets are reached percutaneously under imaging guidance without open surgery
- Ice ball on CT must encompass the nerve with a 5 mm margin — confirm mid-freeze on CT; reposition probe if nerve is at edge of ice ball
- Pain fibers (A-delta and C) are more cold-sensitive than motor fibers — preferred over RFA when near motor nerves because precise, temperature-titrated injury is achievable and perineurium is preserved
- Ice ball shape is predictable based on probe size; dual probe technique (two 17G probes 10–15 mm apart) merges ice balls to double ablation zone for larger targets or prior incomplete ablations
Nerve Targets by Indication
| Nerve | Indication | Approach | Guidance |
|---|---|---|---|
| Intercostal (multiple) | Post-thoracotomy pain; rib met; chest wall tumor | CT posterior — inferior rib margin, costal groove | CT |
| LFCN (lateral femoral cutaneous) | Meralgia paresthetica | Medial to ASIS; beneath inguinal ligament | US or CT |
| Celiac plexus | Pancreatic / upper abdominal cancer pain | Bilateral posterior paravertebral, antecrural, T12-L1 | CT |
| Sciatic | Pelvic/leg cancer pain; tumor-involved nerve | Subgluteal; transpedicular for lumbosacral trunk involvement | CT |
| Genicular nerves | Knee OA (superior medial, superior lateral, inferior medial) | Periosteal at femoral and tibial condyles | US or CT |
| Pudendal | Pudendal neuralgia; refractory pelvic pain | Ischial spine; posterior transgluteal | CT |
| Occipital (greater/lesser) | Occipital neuralgia | Suboccipital; along superior nuchal line | US |
| Lumbosacral trunk | Metastatic disease directly involving nerve root | Transpedicular (L4/L5) to adjacent nerve root | CT |
4
Technique
Intercostal prototype + community cards
Supplies
17G cryoprobe (IceSeed or equivalent)
Cryoprobe system (IceCure / Endocare / Galil)
CT scanner (preferred) or US probe
22G coaxial needle
Sterile saline 20 mL (hydrodissection)
1% lidocaine (local anesthesia)
Fentanyl + midazolam (optional mild sedation)
ChloraPrep
Sterile drape and gloves
Sterile dressing
Prototype — CT-Guided Intercostal Cryoneurolysis
Steps
1
CT planning scan — identify target nerve
Patient prone on CT gantry. Planning scan to identify target intercostal space(s). Target is the angle of the rib, posteriorly — where the nerve exits the costal groove (inferior rib margin, neurovascular bundle). Mark skin entry point.
2
Skin prep and local anesthesia
Sterile prep and drape. Infiltrate 1% lidocaine at skin entry down to rib periosteum. Bony contact at rib cortex confirms depth.
3
Hydrodissection — pleural safety buffer
Advance 22G coaxial needle to rib inferior margin. Inject 5–10 mL sterile saline into the space between the rib and pleura to create a protective buffer. CT confirmation of saline layer displacing pleura away from planned probe path.
4
Advance 17G cryoprobe to costal groove
Advance probe to inferior rib cortex at the costal groove. The neurovascular bundle (vein, artery, nerve — VAN — inferior to superior) runs in this groove. CT confirmation: probe tip at inferior rib cortex in the correct intercostal space, not transgressing pleura.
5
CT confirmation before freeze
Multiplanar CT reformation to confirm probe position. Verify no probe transgression of pleura. Verify saline hydrodissection buffer still present. If treating multiple intercostal spaces (e.g., T8–T11 for chest wall tumor), position all probes before initiating freeze.
6
Freeze cycle 1 — activate cryoprobe
Activate cryo gas flow. Monitor ice ball formation on CT at 5 min — ice ball appears as hypodense sphere. Confirm ice ball encompasses nerve with ≥5 mm margin. Ensure ice ball does not contact skin surface (frost injury) or pleural surface (pneumothorax). Continue freeze to 8–10 min total.
7
Passive thaw — cycle 1
Stop cryo gas. Allow passive thaw only for 3–5 min. CT shows ice ball receding. Do NOT pull probe during active ice ball phase — vascular and tissue tearing will occur. Wait until probe is fully free of ice on CT before any repositioning.
8
Freeze cycle 2
Second freeze cycle to −40°C for 8–10 min. The second cycle on "sensitized" tissue extends the ablation zone and ensures complete demyelination. Monitor ice ball again on CT — ensure margins are adequate.
9
Passive thaw — cycle 2
Final passive thaw until probe is freely movable. Do not pull probe while ice ball is still visible on CT.
10
Probe withdrawal + post-procedure CT
Withdraw probe after full thaw. Immediate post-procedure CT of the treated region. For intercostal target: evaluate for pneumothorax (lung windows). Small pneumothorax (<15%, asymptomatic) — observe and recheck CT in 1–2h. Significant or symptomatic pneumothorax → chest tube placement.
Sign in to view and create community cards
5
Imaging Guidance Principles
CT Guidance
- Ice ball appearance: hypodense sphere on CT; highly conspicuous and monitorable in real time during freeze
- Obtain CT scan at 5 min during freeze to verify ice ball encompasses nerve; repeat at 8–10 min to confirm full extent before ending cycle
- Ice ball must not contact skin (frost injury) or pleura (pneumothorax risk for intercostal targets)
- Preferred for deep nerves: celiac plexus, sciatic, pudendal, intercostal, lumbosacral trunk
- Multiplanar reformats essential for accurate depth and proximity assessment
Ultrasound Guidance
- Ice ball appearance: hyperechoic structure with posterior acoustic shadowing — distinct and easily identified in real time
- Preferred for superficial nerves: occipital, genicular, LFCN, greater auricular
- Allows dynamic visualization without radiation; suitable for repeat procedures
- Limitation: ice ball posterior shadowing obscures structures deep to ice ball — plan probe trajectory to avoid critical structures in the acoustic shadow
Hydrodissection
- Inject sterile saline (or room air for subcutaneous protection) to displace critical structures away from planned ice ball path
- Intercostal nerves: 5–10 mL saline between rib and pleura → pleural safety buffer
- Skin protection: 10–20 mL saline (or 50–100 mL room air) subcutaneously between probe and skin; prevents frost injury if probe is superficial
- LFCN: saline to protect ASIS periosteum from ice ball
- Deep pelvic nerves: saline to displace bowel or vessels from ice ball margin
- Confirm displacement on CT or US before initiating freeze
Dual Probe Technique
- Two 17G probes placed 10–15 mm apart — ice balls merge to create a larger confluent ablation zone
- Doubles effective ablation volume; use for larger nerves (sciatic, celiac plexus bilaterally), prior incomplete ablation, or nerve targets with large cross-sectional area
- Both probes activated simultaneously; both ice balls monitored on CT mid-freeze to confirm confluence
6
Troubleshooting
Problem
Ice ball not encompassing nerve
Likely cause: Probe too far from nerve; nerve displacement by adjacent structures; probe tip not reaching target depth.
Next step: Allow passive thaw. Reposition probe 1–2 mm closer to nerve on CT. Consider dual probe technique — place second 17G probe 10–15 mm from first to merge ice balls and increase ablation zone. Confirm on CT mid-freeze of repositioned probe.
Problem
Ice ball approaching or contacting skin
Likely cause: Probe too superficial; inadequate initial hydrodissection; subcutaneous tissue thinner than anticipated.
Next step: Pause freeze and monitor. Retract probe slightly if ice ball not yet contacting skin. Inject 10–20 mL saline or 50–100 mL room air (via 19G needle) into subcutaneous tissue between probe and skin to create insulating buffer. Resume freeze once adequate protection confirmed on CT/US.
Problem
Pneumothorax on post-procedure CT (intercostal target)
Likely cause: Ice ball contacted parietal or visceral pleura; probe placement inadvertently transgressed pleural space; inadequate hydrodissection buffer.
Next step: Assess size and symptoms. Small pneumothorax (<15%, asymptomatic) — observe; serial CT in 1–2h; most resolve spontaneously. Moderate-to-large (>15%) or symptomatic (dyspnea, O2 desaturation) → chest tube placement. Have chest tube kit available for all intercostal procedures.
Problem
Patient movement during active freeze
Likely cause: Inadequate sedation; pain from probe placement; patient anxiety.
Next step: Do NOT pull probe during active ice ball phase — adherent frozen tissue will tear. Pause freeze. Administer additional analgesia/sedation. Wait until ice ball fully thaws and probe is freely movable before any repositioning. Resume procedure with improved sedation.
Problem
Inadequate or absent pain relief after procedure
Likely cause: Diagnostic nerve block was a false positive; incorrect nerve targeted; partial ablation from single cycle or inadequate freeze duration; early post-procedure inflammatory pain masking effect.
Next step: Distinguish early inflammatory pain (first 24–48h) from true failure. Reassess at 2–4 weeks. If persistent failure: review imaging to confirm probe position relative to nerve; consider if correct nerve was targeted; repeat diagnostic block at adjacent level. Repeat cryoneurolysis is safe if failure was technical.
7
Complications
Periprocedural
- Acute post-procedure pain exacerbation — common; occurs in distribution of treated nerve within 24h; likely due to acute nerve injury and central sensitization; manage with PCA; resolves within 18–24h; warn all patients before procedure
- Frostbite / skin injury — if ice ball contacts skin surface; prevent with hydrodissection (saline or air); treat with warming and wound care if occurs
- Pneumothorax (intercostal target) — most common serious complication; ~5–8% incidence; most are small and self-limiting; have chest tube kit available; see troubleshooting
- Hemorrhage at probe site — minor and self-limited; probe tamponades tract during active freeze; occurs after removal; direct pressure; rare significant bleeding
Delayed
- Motor weakness — expected and predictable for mixed or motor nerve targets (sciatic, lumbosacral trunk, peroneal); onset hours after procedure; recovery at 1–2 mm/day nerve regeneration rate; AFO brace for foot drop; physical therapy essential
- Incomplete effect / early nerve regeneration — effect duration 3–6 months for benign conditions; in oncologic patients, nerve regeneration may be slower due to disease progression; repeat treatment safe and recommended
- Neuroma formation — rare compared to RFA or chemical neurolysis; perineurium preservation with cryoneurolysis reduces neuroma risk; if suspected, MR neurography to evaluate
- Infection at probe site — rare; standard sterile technique; treat with antibiotics if erythema/cellulitis develops
- CSF leak — rare; reported with cryoneurolysis of sacral/coccygeal nerves; monitor for positional headache post-procedure; manage conservatively (blood patch if persistent)
8
Critical Pearls
✶
Two freeze-thaw cycles are required — do not skip the second. A single freeze gives incomplete demyelination and risks partial ablation, which may cause allodynia or only short-term relief. The second cycle on "sensitized" tissue extends the ablation zone and ensures Wallerian degeneration. Both cycles must reach target temperature for the full 8–10 minutes.
✶
Passive thaw only — never active (warm) thaw. Active warming during an existing ice ball causes microvascular tearing within the nerve and surrounding tissue. Allow the ice ball to thaw passively (3–5 min). Never pull the probe while the ice ball is still present — it is adherent to frozen tissue and will cause injury on extraction.
✶
Ice ball must include the nerve with a 5 mm margin — confirm on CT mid-freeze. Obtain CT at 5 min into the freeze cycle. If the nerve is at the margin of or outside the ice ball, pause, thaw, reposition the probe 1–2 mm closer, and repeat. Do not assume adequate coverage without imaging confirmation.
✶
Cryoneurolysis is reversible — preferred near motor nerves. Unlike RFA, cryoneurolysis preserves the perineurium, enabling nerve regeneration. Motor weakness after sciatic, pudendal, or lumbosacral trunk cryoneurolysis is predictable and temporary. Counsel patients and arrange PT/AFO before the procedure, not after. Nerve regeneration proceeds at 1–2 mm/day.
✶
Dual probe technique doubles ablation zone for larger nerves. Place two 17G probes 10–15 mm apart; ice balls merge during freeze to create a confluent, larger ablation zone. Use for sciatic nerve, bilateral celiac plexus targets, prior incomplete ablation, or any nerve with cross-sectional area larger than a single ice ball can cover.
✶
Diagnostic nerve block is mandatory before ablation. A patient who does not respond to local anesthetic block at the target nerve will not respond to cryoneurolysis. A positive diagnostic block (≥50% relief) confirms the nerve is the pain generator and predicts procedural success. Do not skip this step — particularly for atypical or multifactorial pain presentations.
✶
Oncologic patients benefit most from durability — repeat freely. In cancer-related pain, nerve regeneration may be slow or incomplete due to disease progression or radiation effects, extending the duration of relief. Repeat cryoneurolysis is safe at the same site. Patients with direct nerve involvement by tumor (as with lumbosacral trunk or intercostal invasion) may obtain sustained palliation lasting months.
9
References
Citations
- Prologo JD, Ray CE Jr., eds. Advanced Pain Management in Interventional Radiology: A Case-Based Approach. Thieme; 2024. Ch. 12–13 (Prologo JD, Zabala ZE; Moussa AM, Santos E, Camacho JC).
- Prologo JD, et al. Percutaneous CT-guided cryoablation for the treatment of refractory pudendal neuralgia. J Vasc Interv Radiol. 2015;26(9):1357–1362.
- Prologo JD, Johnson C, Hawkins CM, et al. Natural history of mixed and motor nerve cryoablation in humans — a cohort analysis. J Vasc Interv Radiol. 2020;31(6):912–916.e1.
- Prologo JD, Gilliland CA, Miller M, et al. Percutaneous image-guided cryoablation for the treatment of phantom limb pain in amputees: a pilot study. J Vasc Interv Radiol. 2017;28(1):24–34.e4.
- Yoon JHE, Grechushkin V, Chaudhry A, Bhattacharji P, Durkin B, Moore W. Cryoneurolysis in patients with refractory chronic peripheral neuropathic pain. J Vasc Interv Radiol. 2016;27(2):239–243.
- Bittman RW, Peters GL, Newsome JM, et al. Percutaneous image-guided cryoneurolysis. AJR Am J Roentgenol. 2018;210(2):454–465.
- Ilfeld BM, Preciado J, Trescot AM. Novel cryoneurolysis device for the treatment of sensory and motor peripheral nerves. Expert Rev Med Devices. 2016;13(8):713–725.
- Trescot AM. Cryoanalgesia in interventional pain management. Pain Physician. 2003;6(3):345–360.
- Lloyd JW, Barnard JDW, Glynn CJ. Cryoanalgesia: a new approach to pain relief. Lancet. 1976;2(7992):932–934.
- Erinjeri JP, Clark TWI. Cryoablation: mechanism of action and devices. J Vasc Interv Radiol. 2010;21(8 Suppl):S187–S191.
- Radnovich R, Scott D, Patel AT, et al. Cryoneurolysis to treat the pain and symptoms of knee osteoarthritis: a multicenter, randomized, double-blind, sham-controlled trial. Osteoarthritis Cartilage. 2017;25(8):1247–1256.
9
References & Resources
Primary sources · Key data · Related procedures
Key Guidelines
- SIR Standards of Practice for Thermal/Cryoablation
- ASRA Guidelines for Cryotherapy in Pain Management
Primary References
- Prologo JD et al. Percutaneous image-guided cryoablation for the treatment of pain in cancer patients. J Vasc Interv Radiol. 2015;26(7):1007-1013.
- Prologo JD, Ray CE Jr., eds. Advanced Pain Management in Interventional Radiology. Thieme; 2024. Cryoneurolysis chapters.
- Callstrom MR, Kurup AN. Percutaneous ablation for musculoskeletal and soft tissue neoplasms. Semin Intervent Radiol. 2010;27(3):285-295.