The Bone Healed. The Hand Did Not: Understanding Complex Regional Pain Syndrome

What this video covers

• Why CRPS is a clinical diagnosis made with the Budapest criteria, and why no scan or blood test can confirm it
• The leading theories, from neurogenic inflammation to altered brain mapping of the limb, and why they remain proposed mechanisms, several supported mainly by animal work, not tests we can run on you
• How a sympathetic block, stellate ganglion for the arm and lumbar sympathetic for the leg, is used as a diagnostic tool and a rehabilitation window, not a cure
• What the 2023 Cochrane overview found: no high-certainty evidence for any CRPS therapy, and moderate-certainty evidence that lidocaine sympathetic blockade probably does not beat placebo for pain
• What the only five-year randomized trial of spinal cord stimulation in CRPS type I showed: no better than physical therapy alone on pain or any other outcome, with effect diminishing over time, though 95 percent of implanted patients would repeat it
• Which CRPS treatments are off-label, and why graded rehabilitation remains the backbone of care
• MEDICAL DISCLAIMER: This content is for educational purposes only and is not medical advice. It does not substitute for professional diagnosis or treatment. Always consult a licensed healthcare provider regarding your condition. Viewing this video does not establish a doctor-patient relationship.

Six weeks after a wrist fracture, the cast comes off. The bone has healed and the radiograph is unremarkable. But the hand that emerges is not a hand that healed. It is swollen and glossy, mottled red or dusky blue, and warmer than the other hand — or colder. The hair on the back of it is growing coarser and the nails faster. A bedsheet is intolerable. And in a seven-minute visit, the patient is told that the fracture is healed, so this should settle down.

It does not settle down. That picture is complex regional pain syndrome — pain grossly disproportionate to the original injury, accompanied by sensory, vasomotor, sudomotor, and trophic changes. This article explains what is thought to be happening in the limb, how those proposed mechanisms line up with the symptoms you feel, why no scan can confirm the diagnosis, and what that means for how you should be evaluated.

What is actually happening in the limb — and how much of it is still theory

Four mechanisms are proposed for complex regional pain syndrome. It matters enormously that the word is proposed.

Neurogenic inflammation and microvascular dysfunction. Small nerve fibers in the limb release neuropeptides in an exaggerated, sustained way. Plasma leaks out of capillaries into the tissue, producing edema and abnormal blood flow. This is the mechanism with the most direct correspondence to what you can see: the swelling and the color change.

Sympathetic-afferent coupling. The sympathetic nervous system — the branch that controls blood vessel tone and sweating — begins driving pain fibers directly. This has been demonstrated mainly in animal models, and where it applies in people it appears to be present in only a subset of patients. It is not a universal explanation.

Dorsal horn sensitization. The relay station in the spinal cord where limb signals are first processed becomes over-responsive, so ordinary input arrives amplified. This is largely preclinical work.

Altered cortical representation. The brain maintains a map of the body. In complex regional pain syndrome, the region of that map devoted to the affected hand or foot appears to degrade and blur into its neighbors. This is visible on group imaging studies, but its causal role is uncertain — it may be a consequence rather than a cause.

Here is the part that gets left out: not one of these four is measurable in you, as an individual, in a clinic. They are frameworks for understanding a syndrome, not tests. Treat any of them as investigational when a treatment is being justified to you on their basis.

How the proposed mechanisms line up with the symptoms you feel

The reason this condition feels so strange — and so hard to describe to a physician — is that it is not one symptom. It is four families of symptom appearing together in a single limb.

• Sensory. Hyperalgesia (a painful stimulus hurts far more than it should) and allodynia (something that should not hurt at all — a sock, a bedsheet, moving air — hurts). This is what sensitization at the spinal cord and a degraded cortical map would be expected to produce: the volume control on that limb has been turned up and the tuning has been lost.
• Vasomotor. Temperature asymmetry between the two limbs and skin color change. That is the microvascular half of the story — abnormal blood flow. It is also why the limb warms when its sympathetic supply is interrupted with local anesthetic: that warming shows the injection reached the sympathetic chain, and it is not by itself evidence that the injection treats the pain.
• Sudomotor and edema. Swelling and changed sweating. Plasma leaking into tissue is the proposed explanation for the glossy, puffy limb.
• Motor and trophic. Weakness, tremor, dystonia, and changes in hair, nail, and skin growth. The nails growing faster is not your imagination; it is a trophic change and it belongs in the record.

That combination is the diagnosis. It is also why the condition is so easy to dismiss one symptom at a time.

Why no scan confirms it — and what the Budapest criteria changed

No imaging study and no blood test confirms complex regional pain syndrome. Not an MRI, not a bone scan, not a thermogram. If a physician tells you a scan made this diagnosis, that is not supported by the literature. The condition is still widely called reflex sympathetic dystrophy, or RSD, in older charts and in most patient conversations, and the same rule applies under either name.

The diagnosis is clinical, and it is made with a structured instrument: the Budapest criteria. Four categories — sensory, vasomotor, sudomotor and edema, motor and trophic. A physician needs reported symptoms in three of the four categories, signs observable on examination in two of the four, and — the clause that gets skipped — no other diagnosis that better explains the picture.

That discipline has been measured. Harden and colleagues, publishing in Pain in 2010, tested these criteria in 113 patients with CRPS type I against 47 patients with other neuropathic pain conditions.¹ The older International Association for the Study of Pain criteria had a sensitivity of 1.00 at a specificity of 0.41 — meaning they caught essentially everyone who had the condition, but also captured 59 percent of the people who did not. The Budapest clinical criteria hold sensitivity at 0.99 and raise specificity to 0.68. The stricter research version of the criteria reaches 0.79 specificity, at lower sensitivity.

That difference — 0.41 to 0.68 — is the difference between a label and a diagnosis. Note also what the study population was: these criteria were validated in CRPS type I. Most trials in this condition enrolled type I only.

What standard care tends to miss

Complex regional pain syndrome is mishandled in both directions at once, and both errors cost the same currency: time.

Some patients wait a year to hear the words. Their limb pain is attributed to anxiety or deconditioning, or each symptom is addressed separately — the swelling to one clinician, the burning to another — so the pattern is never assembled. Meanwhile the limb is being used less every week.

Other patients are labeled in ten minutes because a limb looked red, and are then treated indefinitely for a disease they may not have. This is where the skipped clause matters. Before this label is attached, a physician should exclude infection, deep vein thrombosis, cellulitis, occult fracture, and nonunion. A compressive neuropathy, a radiculopathy, or a discrete peripheral nerve injury should be identified or ruled out by examination and, where indicated, by electrodiagnostic study. Inflammatory arthritis and arterial insufficiency should be considered.

Those tests do not confirm the condition. They subtract the alternatives. That is a different job, and it is the one that gets skipped when a diagnosis is made from across the room.

The third thing standard care misses is the clock. Escalating medication is frequently the only thing on offer, and no medication retrains a limb. The intervention with the strongest claim on your outcome is graded rehabilitation of that limb, started early and pushed consistently — graded motor imagery, mirror therapy, desensitization, and progressive loading. Be exact about what that sentence is and is not. It is a statement about where the priority belongs, because rehabilitation is the only part of the plan aimed directly at restoring use of the limb. It is not a claim of proven benefit: the 2023 Cochrane overview of systematic reviews found no high-certainty evidence for the effectiveness of any therapy in this condition, and rehabilitation is not exempt from that finding.² What is not in doubt is the cost of delay. Every month spent in diagnostic limbo, or on an injection schedule that was never designed to end, is a month that limb spends learning not to move.

What this means for evaluation and treatment

The framework used at the Padda Institute is straightforward to state: establish the diagnosis properly, then use interventions to open a window that rehabilitation fills.

Sympathetic blocks as a diagnostic probe. The candidate for one is a patient who meets the Budapest criteria and remains limited despite active rehabilitation — not a patient who happens to have been offered an injection. If the affected limb is an arm, the sympathetic supply is interrupted at the cervicothoracic (stellate) ganglion, at the C6 or C7 level, under high-resolution ultrasound, because ultrasound shows the carotid, the inferior thyroid artery, the vertebral artery, and the esophagus that fluoroscopy cannot. If the limb is a leg, the target is the lumbar sympathetic chain at the second through fourth lumbar vertebral bodies, under live fluoroscopy with contrast. The step-by-step technique, what the appointment involves, and recovery are covered on our complex regional pain syndrome treatment page; this page is about whether the diagnosis is right before any of that is scheduled.

These are real procedures with real risks, and you should hear them before a consent form. In the neck: Horner syndrome, hoarseness, difficulty swallowing, hematoma, seizure from inadvertent vertebral artery injection, pneumothorax. In the back: low blood pressure, transient leg weakness or numbness, genitofemoral neuralgia, retroperitoneal injury, bleeding, infection. Blood thinners are reviewed before anything is scheduled.

What the evidence says about whether these blocks relieve pain. The 2023 Cochrane overview of systematic reviews by Ferraro and colleagues found no high-certainty evidence for the effectiveness of any therapy in this condition, and moderate-certainty evidence that lidocaine sympathetic blockade probably does not reduce pain intensity compared with placebo.² That finding is not contradicted here. In this practice, a sympathetic block is used as a diagnostic probe and as a window — a period in which a limb that could not tolerate touch can tolerate work. In our clinical experience that window, when it opens at all, runs from hours to a few weeks; that is a practice observation, not a trial result, and individual results vary. If it is not spent on graded motor imagery, mirror therapy, desensitization, and progressive loading, it was wasted.

Hardware deserves the same scrutiny. Kemler and colleagues, in the Journal of Neurosurgery in 2008, reported the five-year follow-up of the only randomized trial of spinal cord stimulation in CRPS type I.³ That trial enrolled a narrow population — CRPS type I of the hand or foot, present for six months or longer, severe, and refractory to conventional treatment — so what it found applies to that group and no wider. At five years, stimulation plus physical therapy produced results similar to physical therapy alone on pain relief and every other measured variable, and the effect diminished over time. The trial was open-label with no sham comparison, and 10 of the 54 randomized patients were excluded from that final analysis. A non-randomized subgroup of implanted patients did report better global perceived effect (p=0.02), and 95 percent said they would repeat the treatment — but their pain relief did not reach statistical significance (p=0.06). That is the honest shape of the evidence. If you are being walked toward an implant without hearing it, get a second opinion. Hardware risks include lead migration, infection, dural puncture and spinal headache, device failure, revision surgery, and rarely epidural hematoma or abscess with lasting nerve injury — plus permanent restrictions on future MRI.

What is off-label. Ketamine, bisphosphonates, and intravenous immunoglobulin are off-label in this condition, as is corticosteroid added to these injections. On bisphosphonates specifically, the Cochrane overview reported low-certainty evidence that they may reduce pain intensity after treatment alongside moderate-certainty evidence that they probably increase adverse events.² Where metabolic health is poor, glycemic control and nutrition are addressed because tissue repair depends on it — but that is supportive care, not treatment for the syndrome. Photobiomodulation and specialized pro-resolving mediators are adjunctive and unproven in this condition, and should be labeled exactly that way or not offered.

What recovery realistically looks like. The following is a composite — a picture assembled from many patients with this condition, not one person’s chart. An adult has routine foot surgery. It heals; the pain does not follow the healing. Four months later the foot is swollen, cold, purple when it hangs down, and cannot tolerate a sock. Two clinics have offered only escalating medication. The Budapest criteria are met on examination; vascular and electrodiagnostic studies are ordered to exclude alternatives, not to prove the diagnosis. A lumbar sympathetic block is performed under fluoroscopy. The limb warms and the relief lasts about ten days. Those ten days are the entire point, and they are hard — desensitization is unpleasant before it is useful, and mirror therapy feels absurd for two weeks. A second block, weeks later, gives less than the first. A cold snap in February undoes a month of progress. Nine months on, that patient walks a mile, wears a shoe, sleeps through the night, and drives again. That patient does not run, and the foot still tells the weather. That is a real result. It is not a cure, and no one should promise you one. Individual results vary.

If your limb hurts far more than the injury explains, and it is swelling, sweating, or changing color and temperature, do not accept vagueness. Ask for the Budapest criteria to be applied to you on paper. Ask how the alternatives were excluded. Ask what any proposed block is for, and what rehabilitation is already on the calendar to fill the window it opens.

Frequently asked questions

Key takeaways

• Complex regional pain syndrome is pain grossly out of proportion to the injury, with sensory, vasomotor, sudomotor, and trophic changes in the same limb; the four proposed mechanisms are frameworks, not tests, and none can be measured in an individual patient.
• No scan or blood test confirms it. The diagnosis is clinical and made with the Budapest criteria, which raised specificity from 0.41 under the older IASP criteria to 0.68 — imaging and nerve studies exist to exclude alternatives.
• The evidence for treatment is genuinely limited: no high-certainty evidence for any therapy — rehabilitation included — moderate-certainty evidence that lidocaine sympathetic blockade probably does not beat placebo, and a five-year randomized trial in a narrow refractory type I population in which spinal cord stimulation was no better than physical therapy alone.
• Sympathetic blocks are best understood as a diagnostic probe and a rehabilitation window in patients who meet the criteria and are still limited despite active therapy; graded motor imagery, mirror therapy, desensitization, and progressive loading are the backbone of care.
• Insist on the diagnosis before the procedure, and on the rehabilitation before the hardware.

Medically reviewed by Gurpreet Singh Padda, MD — Board Certified in Anesthesiology, Pain Medicine, Interventional Pain Management, Addiction Medicine, and Obesity Medicine. Last reviewed July 2026.

This article is educational and is not a substitute for evaluation, diagnosis, or treatment by a physician. Individual results vary. Do not start, stop, or change any medication without consulting your physician. To be evaluated, request an appointment at painmd.tv, call (314) 481-5000, or text (314) 886-5902.

References

1. Ferraro MC, Cashin AG, Wand BM, Smart KM, Berryman C, Marston L, Moseley GL, McAuley JH, O’Connell NE. Interventions for treating pain and disability in adults with complex regional pain syndrome: an overview of systematic reviews. Cochrane Database Syst Rev. 2023 Jun 12;6(6):CD009416. PMID 37306570101002146518580094163. PubMed
2. Harden NR, Bruehl S, Perez RSGM, Birklein F, Marinus J, Maihofner C, Lubenow T, Buvanendran A, Mackey S, Graciosa J, Mogilevski M, Ramsden C, Chont M, Vatine JJ. Validation of proposed diagnostic criteria (the “Budapest Criteria”) for Complex Regional Pain Syndrome. Pain. 2010;150(2):268-274. PMID 20493633. PubMed
3. Kemler MA, de Vet HC, Barendse GA, van den Wildenberg FA, van Kleef M. Effect of spinal cord stimulation for chronic complex regional pain syndrome Type I: five-year final follow-up of patients in a randomized controlled trial. J Neurosurg. 2008;108(2):292-298. PMID 18240925. PubMed

Dr. Gurpreet Singh Padda, MD, MBA, MHP