CRPS after a stroke (shoulder-hand syndrome)

Complex regional pain syndrome can develop in the weakened arm after a stroke, where it was long called shoulder-hand syndrome. Two prospective studies report very different rates: 48.8 percent of 82 patients within 28 weeks in Turkey (Kocabas, 2007) and 15.4 percent of 195 rehabilitation patients at three months in Japan (Katsura, 2024). Weakness, limb disuse and inactivity predicted it.

A painted pair of crutches tied with an orange ribbon among poppies.

What is shoulder-hand syndrome?

Shoulder-hand syndrome is an older name for complex regional pain syndrome in the arm, used most often when it follows a stroke. The Royal College of Physicians' 2018 guideline lists shoulder-hand syndrome among the earlier names for CRPS, beside reflex sympathetic dystrophy, causalgia, algodystrophy and Sudeck's atrophy.

After a stroke, the condition appears in the arm on the weakened side of the body, which doctors call the hemiplegic arm. Recent studies use the terms post-stroke CRPS and CRPS type I in place of shoulder-hand syndrome. A person may see any of the three names in a medical record, and they refer to the same condition.

How often does CRPS develop after a stroke?

Two prospective studies give rates of 15.4 percent and 48.8 percent, and both studied patients who were already receiving stroke rehabilitation. Neither figure describes everyone who has a stroke. The table shows who was studied and when the patients were assessed.

Prospective studies of CRPS after stroke, with the population and follow-up for each rate
StudyWho was studiedRate of CRPSFollow-up and criteria
Kocabas and colleagues, International Journal of Rehabilitation Research, 2007, Turkey82 patients with weakness on one side of the body after a strokeCRPS type I of the arm in 48.8 percentAssessed at 2, 6, 14 and 28 weeks after the stroke; diagnostic criteria not stated in the abstract
Katsura and colleagues, Scientific Reports, 2024, Japan195 stroke patients admitted to Kishiwada Rehabilitation HospitalPost-stroke CRPS in 15.4 percentAssessed at admission and 3 months after admission; diagnostic criteria not stated in the abstract

Why do the two stroke studies report such different rates?

The Kocabas and Katsura studies differ in their patients, their follow-up and possibly their criteria. Kocabas followed patients for 28 weeks from the stroke, and Katsura counted cases three months after admission to a rehabilitation hospital. A longer follow-up can find more cases. The severity of paralysis in each group also matters, because measures of weakness predicted CRPS in both studies.

Diagnostic criteria can change a CRPS rate several times over. In a 2012 Dutch study of 596 fracture patients (Beerthuizen and colleagues), the same patients produced a rate of 7.0 percent under the Harden and Bruehl criteria and 48.5 percent under the older International Association for the Study of Pain criteria. Neither stroke abstract states which criteria were applied, so the two rates cannot be compared directly.

Who is at higher risk of CRPS after a stroke?

Stroke patients with more severe arm weakness and shoulder problems were at higher risk in the Kocabas study. In those 82 patients, CRPS type I was significantly correlated with partial dislocation of the shoulder joint (subluxation), with spasticity, with lower arm strength scores and with a lower stage of motor recovery. The correlations were modest, ranging from 0.26 to 0.39 in size.

The authors summarized the finding as a relationship between CRPS type I and "subluxation, loss of range of motion, spasticity of shoulder muscles and muscle strength." A correlation shows that two things occur together and does not show that one causes the other.

How much does inactivity matter after a stroke?

Inactivity was the strongest dividing line in the 2024 Katsura study. The researchers measured arm function, leg function, balance and daily inactive time in 195 patients at admission, then used a decision tree to find which measures predicted CRPS three months later. Inactive time, arm motor score and leg motor score were the three predictors.

Among patients with a short inactive time, 1.5 percent developed post-stroke CRPS. Among patients with a long inactive time plus severe disuse of both the arm and the leg, 84.6 percent developed it. The authors concluded that the incidence of post-stroke CRPS "may increase with lower-limb disuse and physical inactivity, in addition to upper-limb disuse."

When does CRPS appear after a stroke?

CRPS after a stroke appears over the weeks and months of recovery, not at the moment of the stroke. Kocabas and colleagues examined patients at 2, 6, 14 and 28 weeks and reported the incidence across the first 28 weeks. Katsura and colleagues counted cases at three months after admission to rehabilitation.

For CRPS after an injury, the Royal College of Physicians' 2018 guideline states that onset "for the majority occurs within 1 month of the trauma or immobilisation of the limb." A stroke is not a limb injury, and the published stroke abstracts do not report the average week of onset. Pain, swelling, or color and temperature change in the weak hand during rehabilitation are the features the diagnostic criteria describe.

Is CRPS after a stroke diagnosed the same way as CRPS after an injury?

Yes. Doctors apply the same Budapest criteria, which require continuing disproportionate pain, symptoms in three of four categories, signs at the examination in two of four, and no other diagnosis that better explains the findings. The criteria do not require a limb injury.

The last rule takes care after a stroke, because the weak arm can be painful for other reasons, such as shoulder subluxation and spasticity, which Kocabas and colleagues measured as separate findings. The Royal College of Physicians' guideline lists central nervous system lesions and nerve damage among the conditions a doctor weighs before diagnosing CRPS. The page on how CRPS is diagnosed explains each rule.

Does depression cause CRPS after a stroke?

The evidence does not show that. Kocabas and colleagues found a statistically significant but weak correlation (0.29) between depression scores and CRPS type I in 82 stroke patients. A correlation does not show which came first, the low mood or the painful arm.

In the general CRPS population, a 2008 Dutch case-control study of 186 patients and 697 controls reported that "psychological factors were not associated with CRPS onset" (de Mos and colleagues, Pain). Katsura's 2024 analysis of stroke patients identified physical measures, inactivity and limb disuse, as the predictors.

Can CRPS after a stroke be prevented?

No trial reviewed for this page proves a way to prevent CRPS after a stroke. Both sets of researchers suggested prevention strategies based on what they observed. Kocabas and colleagues recommended exercises to increase the range of motion of the shoulder joint, strengthening of the shoulder muscles and reduction of spasticity. Katsura and colleagues wrote that increasing physical activity and addressing arm and leg paralysis "may reduce post-stroke CRPS."

Those are suggestions drawn from observational studies, which can show an association and cannot show that changing the factor changes the outcome. Decisions about a rehabilitation program belong to the patient and the treating team.

What do the medical facts mean for a California claim after a stroke?

The studies tie CRPS after a stroke to the effects of the stroke itself: weakness, limb disuse and inactivity. They do not suggest that the condition signals a mistake in care. California's jury instruction CACI No. 505 states that a medical practitioner "is not necessarily negligent just because" the practitioner's efforts are unsuccessful. Whether care met the standard is decided on the facts of the case, usually with testimony from medical expert witnesses (CACI No. 501).

Where a claim does exist, the deadlines are short. A medical negligence claim must be filed within three years of the injury or one year of discovery, whichever comes first, under Code of Civil Procedure section 340.5, with 90 days' notice under section 364. Noneconomic damages are capped at $470,000 for an injury in 2026 (Civil Code section 3333.2). The page on medical negligence claims for CRPS explains the law.

Common questions

Yes. The Royal College of Physicians' 2018 guideline lists shoulder-hand syndrome among the earlier names for complex regional pain syndrome. The term is still used for CRPS that develops in the weakened arm after a stroke. Research papers now call it post-stroke CRPS or CRPS type I. CRPS type I was formerly called reflex sympathetic dystrophy (RSD); type II was formerly called causalgia.

The published studies concern the arm. Kocabas and colleagues (2007) studied CRPS type I in the upper limb of 82 stroke patients. Katsura and colleagues (2024) found that disuse of the leg, measured by a leg motor score, helped predict post-stroke CRPS, along with arm disuse and inactivity. Neither abstract reports a rate of CRPS in the leg after a stroke.

No. Shoulder pain after a stroke has several causes, including partial dislocation of the shoulder joint and muscle spasticity, both of which Kocabas and colleagues measured as separate findings. CRPS is diagnosed only when the Budapest criteria are met, which requires signs such as swelling, color or temperature change, or sweating changes at the examination, and no other diagnosis that better explains the findings.

The populations are different. The stroke studies followed patients in rehabilitation programs, many with severe paralysis, and severe disuse was the strongest predictor of CRPS. Fracture studies follow people who are otherwise mobile. The stroke abstracts also do not state their diagnostic criteria, and looser criteria produce higher rates: one 2012 fracture study found 7.0 percent under stricter criteria and 48.5 percent under older ones in the same 596 patients.

The studies do not show that. Both studies followed patients who were receiving stroke rehabilitation, and CRPS developed in 15.4 percent of the Japanese group and 48.8 percent of the Turkish group. The strongest predictors were the severity of weakness and limb disuse, which are effects of the stroke. The authors suggested that more activity may reduce the risk, and no trial reviewed for this page has tested that suggestion.

Sources

Every source below was opened and read for this page. The date is when it was last checked.

  1. Complex regional pain syndrome in stroke patientsStudy · International Journal of Rehabilitation Research (Kocabas et al.), 2007 · checked
  2. A decision tree algorithm to identify predictors of post-stroke complex regional pain syndromeStudy · Scientific Reports (Katsura et al.), 2024 · checked
  3. Demographic and medical parameters in the development of complex regional pain syndrome type 1 (CRPS1): prospective study on 596 patients with a fractureStudy · Pain (Beerthuizen et al.), 2012 · checked
  4. Medical history and the onset of complex regional pain syndrome (CRPS)Study · Pain (de Mos et al.), 2008 · checked
  5. Complex regional pain syndrome in adults: UK guidelines for diagnosis, referral and management in primary and secondary care, 2nd editionClinical guideline · Royal College of Physicians, 2018 · checked
  6. Validation of proposed diagnostic criteria (the "Budapest Criteria") for Complex Regional Pain SyndromeStudy · Pain (Harden et al.), 2010 · checked
  7. Judicial Council of California Civil Jury Instructions (CACI), 2026 edition: Nos. 501 and 505Jury instruction · Judicial Council of California, 2026 · checked
  8. Code of Civil Procedure section 340.5Statute · California Legislative Information · checked
  9. Code of Civil Procedure section 364Statute · California Legislative Information · checked
  10. Civil Code section 3333.2Statute · California Legislative Information · checked

Ask a lawyer about your CRPS case

Free case review. No fee unless we recover. A person from Wise Law reads every message.

What happened before the CRPS started? (optional)
Has a doctor diagnosed CRPS? (optional)
Or call (424) 235-7879

Submitting this form does not create an attorney-client relationship. An attorney-client relationship is formed only by a signed written fee agreement.

Text usCall