Cover of Evidence reviews for interventions for shoulder pain after stroke

Evidence reviews for interventions for shoulder pain after stroke

Stroke rehabilitation in adults

Evidence review O

NICE Guideline, No. 236

London: National Institute for Health and Care Excellence (NICE); .
ISBN-13: 978-1-4731-5464-3
Copyright © NICE 2023.

1. Managing post-stroke shoulder pain

1.1. Review question

In people with shoulder pain after stroke, what is the clinical and cost effectiveness of transcutaneous electrical nerve stimulation, acupuncture, functional electrical stimulation and intra-articular steroid injection in reducing pain?

1.1.1. Introduction

Shoulder pain is very common after a stroke, in particular among individuals with arm weakness. This pain can be disabling and can prevent or interrupt rehabilitation programmes. While there is extensive literature on the management of shoulder pain in the healthy adult population, there is little research and clinical guidance for the management of post-stroke shoulder pain. Shoulder pain in this clinical cohort is complex and multifactorial in aetiology, and there has been an increase in treatment options such as electrical stimulation becoming available over the past few years. Despite this, a lack of national clinical standards means that current clinical practice tends to be more reactive rather than proactive, and clinicians may be uncertain which physical or pharmacological intervention may be the most appropriate for their patient.

1.1.2. Summary of the protocol

Table 1. PICO characteristics of review question.

Table 1

PICO characteristics of review question.

For full details see the review protocol in Appendix A.

1.1.3. Methods and process

This evidence review was developed using the methods and process described in Developing NICE guidelines: the manual. Methods specific to this review question are described in the review protocol in Appendix A and the methods document.

Declarations of interest were recorded according to NICE’s conflicts of interest policy.

1.1.4. Effectiveness evidence

1.1.4.1. Included studies

Twenty eight randomised controlled trial studies (32 papers) were included in the review;2–9, 14–17, 21–24, 26, 27, 33, 34, 36, 37, 39–41, 45, 46, 48–52 these are summarised in Table 2 below. Evidence from these studies is summarised in the clinical evidence summary below (Table 3).

The following interventions were compared:

  • Transcutaneous electrical nerve stimulation (TENS) compared to:
    • Neuromuscular electrical stimulation (NMES)6, 52
    • Nerve blocks (local anaesthetic)9
    • Usual care or no treatment52
  • Functional electrical stimulation (FES) compared to:
    • Usual care or no treatment21
  • Neuromuscular electrical stimulation (NMES) compared to:
    • Transcutaneous electrical nerve stimulation (TENS)6, 52
    • Devices – slings5
    • Placebo/sham7, 23
    • Usual care or no treatment40, 45, 52
  • Devices – tape compared to:
  • Devices – slings compared to:
    • Neuromuscular electrical stimulation (NMES)5
    • Usual care or no treatment27, 41
  • Devices – braces compared to:
    • Usual care or no treatment14
  • Acupuncture/dry needling compared to:
    • Placebo/sham24
    • Usual care or no treatment8, 26, 50, 51
  • Electroacupuncture compared to:
    • Placebo/sham37
  • Intra-articular medicine injections – Corticosteroids compared to:
  • Nerve blocks (local anaesthetic) compared to:
    • Transcutaneous electrical nerve stimulation (TENS)9
    • Placebo/sham2, 39

No relevant clinical studies including the following interventions were identified:

  • Devices – supports and other devices
  • Intra-articular medicine injections – saline
  • Injections into other sites (for example: bursae) – corticosteroids and saline

Population and concomitant therapy factors

The populations included in the review were somewhat similar. There was a mixture of studies investigating the use of interventions in different time periods after stroke, mostly including people in the subacute or chronic time periods. The majority of studies excluded people with previous shoulder pathology, while others did not state whether they were excluded. No study reported specifically including people with previous shoulder pathology.

Concomitant therapy use varied between studies. In the majority of cases, physiotherapy including exercise with or without manual therapy was available with the therapy being of varied intensity. In some cases, occupational therapy and speech and language therapy were provided as required. In others, additional pharmacological therapy, including paracetamol, non-steroidal anti-inflammatory drugs and opioids for pain relief and occasionally antispasticity medication, such as tizanidine were available.

Inconsistency

The majority of outcomes included evidence from one study only. Where outcomes included multiple studies, some showed inconsistency that could not be resolved by sensitivity analysis or subgroup analysis. In the majority of cases, there were less than four studies, which meant that valid subgroups could not be formed.

Background rate of oral drug use

When investigating the studies, the possibility of study enrichment through inclusion criteria specifying previous oral medication use was considered. Most studies did not report specific response criteria, while the others that discussed this possibility did not specifically include or exclude people based on this. Instead, they provided the opportunity to use oral pain relief medication to all participants. In some studies, this appeared to be provided to all participants, while in others only some of the participants received therapy. A series of sensitivity analyses were conducted investigating this and did not find that considering this resolved heterogeneity.

See also the study selection flow chart in Appendix C, study evidence tables in Appendix D, forest plots in Appendix E, and GRADE tables in Appendix F.

1.1.4.2. Excluded studies

Two Cochrane reviews, Ada 20051 and Price 200035 were identified but were not included in the review. The reasons included reviewing a different population and not investigating the effect of the intervention on pain1 and including people where it was not explicitly stated they had shoulder pain and not including all of the comparisons stated in the protocol35. In these cases, the citation list was checked and all relevant studies were included in the review.

See the excluded studies list in Appendix J.

1.1.5. Summary of studies included in the effectiveness evidence

Table 2. Summary of studies included in the evidence review.

Table 2

Summary of studies included in the evidence review.

See Appendix D for full evidence tables.

1.1.5.1. Summary matrices
Table 3. Summary matrix of the protocol interventions compared to placebo/sham.

Table 3

Summary matrix of the protocol interventions compared to placebo/sham.

Table 4. Summary matrix of the protocol interventions compared to usual care or no treatment.

Table 4

Summary matrix of the protocol interventions compared to usual care or no treatment.

Table 5. Summary matrix of the protocol interventions compared to each other.

Table 5

Summary matrix of the protocol interventions compared to each other.

1.1.6. Summary of the effectiveness evidence

1.1.6.1. Transcutaneous electrical nerve stimulation (TENS) compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment
Table 6. Clinical evidence summary: transcutaneous electrical nerve stimulation (TENS) compared to neuromuscular electrical stimulation (NMES).

Table 6

Clinical evidence summary: transcutaneous electrical nerve stimulation (TENS) compared to neuromuscular electrical stimulation (NMES).

Table 7. Clinical evidence summary: transcutaneous electrical nerve stimulation (TENS) compared to usual care or no treatment.

Table 7

Clinical evidence summary: transcutaneous electrical nerve stimulation (TENS) compared to usual care or no treatment.

1.1.6.2. Functional electrical stimulation (FES) compared to usual care or no treatment
Table 8. Clinical evidence summary: functional electrical stimulation (FES) compared to usual care or no treatment.

Table 8

Clinical evidence summary: functional electrical stimulation (FES) compared to usual care or no treatment.

1.1.6.3. Neuromuscular electrical stimulation (NMES) compared to placebo/sham and usual care or no treatment
Table 9. Clinical evidence summary: neuromuscular electrical stimulation (NMES) compared to placebo/sham.

Table 9

Clinical evidence summary: neuromuscular electrical stimulation (NMES) compared to placebo/sham.

Table 10. Clinical evidence summary: neuromuscular electrical stimulation (NMES) compared to usual care or no treatment.

Table 10

Clinical evidence summary: neuromuscular electrical stimulation (NMES) compared to usual care or no treatment.

1.1.6.4. Devices - tape compared to placebo/sham and usual care or no treatment
Table 11. Clinical evidence summary: devices – tape compared to placebo/sham.

Table 11

Clinical evidence summary: devices – tape compared to placebo/sham.

Table 12. Clinical evidence summary: devices – tape compared to usual care or no treatment.

Table 12

Clinical evidence summary: devices – tape compared to usual care or no treatment.

1.1.6.5. Devices - slings compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment
Table 13. Clinical evidence summary: devices – slings compared to neuromuscular electrical stimulation (NMES).

Table 13

Clinical evidence summary: devices – slings compared to neuromuscular electrical stimulation (NMES).

Table 14. Clinical evidence summary: devices – slings compared to usual care or no treatment.

Table 14

Clinical evidence summary: devices – slings compared to usual care or no treatment.

1.1.6.6. Devices - braces compared to usual care or no treatment
Table 15. Clinical evidence summary: devices – braces compared to usual care or no treatment.

Table 15

Clinical evidence summary: devices – braces compared to usual care or no treatment.

1.1.6.7. Acupuncture/dry needling compared to placebo/sham and usual care or no treatment
Table 16. Clinical evidence summary: acupuncture/dry needling compared to placebo/sham.

Table 16

Clinical evidence summary: acupuncture/dry needling compared to placebo/sham.

Table 17. Clinical evidence summary: acupuncture/dry needling compared to usual care or no treatment.

Table 17

Clinical evidence summary: acupuncture/dry needling compared to usual care or no treatment.

1.1.6.8. Electroacupuncture compared to placebo/sham
Table 18. Clinical evidence summary: electroacupuncture compared to placebo/sham.

Table 18

Clinical evidence summary: electroacupuncture compared to placebo/sham.

1.1.6.9. Intra-articular medicine injections - corticosteroids compared to placebo/sham
Table 19. Clinical evidence summary: intra-articular medicine injections – corticosteroids compared to placebo/sham.

Table 19

Clinical evidence summary: intra-articular medicine injections – corticosteroids compared to placebo/sham.

1.1.6.10. Nerve blocks (local anaesthetic) compared to transcutaneous electrical nerve stimulation (TENS) and placebo/sham
Table 20. Clinical evidence summary: nerve blocks (local anaesthetic) compared to transcutaneous electrical nerve stimulation (TENS).

Table 20

Clinical evidence summary: nerve blocks (local anaesthetic) compared to transcutaneous electrical nerve stimulation (TENS).

Table 21. Clinical evidence summary: nerve blocks (local anaesthetic) compared to placebo/sham.

Table 21

Clinical evidence summary: nerve blocks (local anaesthetic) compared to placebo/sham.

See Appendix F for full GRADE tables.

1.1.7. Economic evidence

1.1.7.1. Included studies

No health economic studies were included in this review.

1.1.7.2. Excluded studies

No relevant health economic studies were excluded due to assessment of limited applicability or methodological limitations.

See also the health economic study selection flow chart in Appendix G.

1.1.8. Summary of included economic evidence

No health economic studies were included.

1.1.9. Economic model

This area was not prioritised for new cost-effectiveness analysis.

1.1.10. Unit costs

The tables below include unit costs relevant to the interventions being considered in this review. Table 22 presents staff costs related to people who may delivering interventions to reduce shoulder pain.

Electrotherapies (FES, NMES, TENS)

The cost of electrotherapies relates primarily to the staff time to administer it and will depend on how long sessions are and how often they are given, and duration of treatment. There are also equipment costs.

NMES was the most frequently evaluated of out the electrotherapy interventions (7 studies included in clinical review). The interventions varied between studies in terms of frequency and duration, with sessions ranging from 1–6-hours and were delivered between 3–7 days per week for 3–8 weeks. The included evidence for TENS reported sessions lasting 45–60 minutes, 3–7 days per week for 4–8 weeks. TENS can be delivered at home then returned for use by other patients which could lower resource use. The one study (Karaahmet 201921) that assessed functional electrical stimulation (FES) was structured around 30-minute sessions of FES-cycling, delivered 5 times a week over 4 weeks (20 sessions total).

Table 22. Unit costs of health care professionals who may be involved in delivering interventions to reduce shoulder pain.

Table 22

Unit costs of health care professionals who may be involved in delivering interventions to reduce shoulder pain.

Table 23 shows some the equipment costs related to TENS. The cost of a TENS machine varies (approximately £18-£50) depending on the type as a few are recorded in the NHS supply chain catalogue.32 Costs for NMES and FES machines were not listed.

Previous economic evaluations of electrotherapy (TENS, NMES, FES) for treating other types of pain have not included the costs of equipment used by physiotherapists in the analysis as the per-use costs were expected to be small (MacPherson 2017,25 Woods 201747).

Table 23. Equipment costs transcutaneous electrical nerve stimulation (TENS).

Table 23

Equipment costs transcutaneous electrical nerve stimulation (TENS).

A 2010 NHS Purchasing and Supply Agency report38 on FES for drop foot of central neurological origin included an initial assessment appointment costing £140. This analysis also included a clinic model in which the costs of the FES device are incorporated in the ongoing clinical charges. Each ongoing clinical appointment was estimated at £300. FES can also be delivered at home; however, availability varies across current practice.

Acupuncture and electroacupuncture

The cost of acupuncture relates primarily to the staff time to administer it and will depend on how long sessions are and how often they are given, and duration of treatment.

In the clinical review, the frequency and duration for delivering acupuncture and electroacupuncture varied across studies. Acupuncture ranged from being delivered once with a 1-week follow-up to once daily for one month continuously. Sessions typically lasted 30 minutes.

Equipment costs for acupuncture relate to the needles used. A previous economic model developed for the Chronic Pain NICE guideline (NG193)28 used a cost per needle of £0.06. A large acupuncture individual patient meta-analysis in chronic pain reported the number of needles across studies, and the most frequent range was between 10 and 14.42

An outpatient procedure for acupuncture for pain management is £141 (2019/2020 NHS reference costs31). Costs in the community setting may be lower.

One study included in the clinical review (Sui 202137) provided acupuncture followed by 30 minutes of electroacupuncture delivered once a day, five days a week for two weeks. Example electroacupuncture equipment costs shown in Table 24 were taken from the analysis conducted as part of the osteoarthritis guideline update29. These devices were the ES-160 (included as it was used in two of the four clinical studies in the osteoarthritis review of electroacupuncture) and AS-super 4, which is a popular alternative in clinical practice. The analysis assumed that both devices have a lifespan of 5 years. Other costs associated with electrotherapy include batteries, needles, disinfectant swabs, and surgeons’ gloves. The last electroacupuncture device included was the HANS-200A instrument, which was used in Sui 2021,37 however this would not be as frequently used in an NHS clinical setting.

Table 24. Example equipment costs for electroacupuncture.

Table 24

Example equipment costs for electroacupuncture.

Devices

Table 25 reports the costs associated with the devices reported in the clinical review. Slings and tape are relatively low-cost compared43 to the other interventions reported as the equipment costs and staff time involved in the application and correction of the devices are less resource intensive and can be incorporated into standard therapy. Taping was typically kept on for three days before being reapplied, meaning frequent visits may increase staff time compared to the sling. Shoulder braces were more expensive, with one study (Hartwig 2012)14 reporting the use of a shoulder brace (Functional orthosis Neuro-Lux (Sporlastic GmbH, Nürtingen, Germany)) which retails online for almost €233 (£212).43 Although this specific device was not reported in the NHS supply chain catalogue, it was noted by the committee to be one of the braces used in current practice. These interventions could also take place at home, with people tasked with wearing the devices all day or whenever the upper limb was unsupported.

Table 25. Example equipment costs of devices.

Table 25

Example equipment costs of devices.

Intra-articular medicine injections and nerve blocks

Resource use for intra-articular medicine injections and nerve blocks will relate to the drugs injected and the staff time to deliver the injections.

Table 26 presents unit costs for drugs used in intra-articular injections and nerve blocks. Saline injections were also included in the protocol, but no studies were found in the clinical review related to this and so costs are not shown.

Participants in all studies included in the clinical review received a single injection and were followed up from between 3 to 12 weeks post-intervention. This reflects current practice, with people receiving typically 1–2 injections. Drug costs per injection estimated based on the drugs and doses used in these studies are summarised in Table 26.

Table 26. Unit costs of intra-articular medicine injections and nerve blocks (local anaesthetics).

Table 26

Unit costs of intra-articular medicine injections and nerve blocks (local anaesthetics).

Resource use also differed for staff involvement in the injection procedure. Table 27 illustrates outpatient appointment costs associated with having an injection for pain management. All studies reported using a rehabilitation doctor to provide the injection, however, one study (Rah 201236) reported that two rehabilitation doctors (physiatrists) and a radiologist attended a 2-day training course on the study procedure, physical tests, home exercise program, and ultrasonography for diagnosis and injection procedure. The training costs and ultrasound-guided subacromial injection would incur additional costs compared to the other interventions. Shoulder pain injections can also be provided in primary care settings which would incur lower costs, however this varies depending on the clinician being comfortable with administering the injection.

Table 27. Example procedural costs of intra-articular medicine injections and nerve blocks (local anaesthetics).

Table 27

Example procedural costs of intra-articular medicine injections and nerve blocks (local anaesthetics).

1.1.11. Evidence statements

Effectiveness/Qualitative
Economic

No relevant economic evaluations were identified.

1.1.12. The committee’s discussion and interpretation of the evidence

1.1.12.1. The outcomes that matter most

The committee included the following outcomes: person/participant generic health-related quality of life, carer generic health-related quality of life, pain, physical function – upper limb, activities of daily living, stroke-specific Patient-Reported Outcome Measures and withdrawal due to adverse events. All outcomes were considered equally important for decision making and therefore have all been rated as critical.

Person/participant health-related quality of life outcomes were considered particularly important as a holistic measure of the impact on the person’s quality of living. Pain was considered important as a direct answer to the question. Withdrawal due to adverse events was used to understand the tolerability to the intervention, with the committee acknowledging that different interventions may have different adverse events. Mortality was not considered as it was deemed unlikely to be a result of the treatment and would be included in withdrawal due to adverse events. If mortality was an adverse events then this was highlighted to the committee during their deliberation.

The committee chose to investigate these outcomes at less than 6 months and more than and equal to 6 months, as they considered that there could be a difference in the short term and long term effects of the interventions.

The evidence for this question was limited, with some outcomes not being reported for every comparison. No study investigated the effects of interventions on carer generic health-related quality of life. The majority of outcomes were reported at less than 6 months, with only one outcome being reported at more than and equal to 6 months. The most widely reported outcome was pain.

1.1.12.2. The quality of the evidence

Twenty eight randomised controlled trial studies were included in the review. These reported a range of different comparisons:

The following interventions were compared:

  • Transcutaneous electrical nerve stimulation (TENS) compared to neuromuscular electrical stimulation (NMES), nerve blocks (local anaesthetic) and usual care or no treatment
  • Functional electrical stimulation (FES) compared to usual care or no treatment
  • Neuromuscular electrical stimulation (NMES) compared to transcutaneous electrical nerve stimulation (TENS), devices – slings, placebo/sham and usual care or no treatment
  • Devices – tape compared to placebo/sham and usual care or no treatment
  • Devices – slings compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment
  • Devices – braces compared to usual care or no treatment
  • Acupuncture/dry needling compared to placebo/sham and usual care or no treatment
  • Electroacupuncture compared to placebo/sham
  • Intra-articular medicine injections – Corticosteroids compared to placebo/sham
  • Nerve blocks (local anaesthetic) compared to transcutaneous electrical nerve stimulation (TENS) and placebo/sham

No relevant clinical studies including the following interventions were identified:

  • Devices – supports and other devices
  • Intra-articular medicine injections – saline
  • Injections into other sites (for example: bursae) – corticosteroids and saline

Studies were generally distributed evenly across the different interventions, with a limited number of studies reporting each comparison. Outcomes were of low or very low quality, with the majority being of very low quality. This was mainly due to risk of bias and imprecision. Risk of bias was a problem in a lot of studies and was mainly due to bias arising from the randomisation process, due to deviations from the intended intervention and due to missing outcome data, though all reasons for downgrading outcomes for risk of bias were present at least once during the analysis.

A large number of outcomes were downgraded due to imprecision. This was likely due to the studies included being, in general, small studies (with an average number of participants being 30 people) and that there were few studies to meta-analyse to improve the precision in the outcome.

Where meta-analysis was conducted, studies were generally downgraded for inconsistency due to heterogeneity that could not be resolved by the agreed sensitivity and subgroup analyses. Sensitivity and subgroup analyses often did not resolve the heterogeneity due to either there being an insufficient number of studies included in the results to allow for valid subgroups to be formed, or due to homogeneity in the subgroups present. The majority of studies investigated the effect of people with no previous shoulder pathology. There was a mixture of people in the subacute or chronic period after stroke. However, this did not resolve the heterogeneity when investigated.

A significant number of studies were excluded for not being reported in the English language. These studies primarily investigated the use of acupuncture. It is unclear whether these studies would be included if they were reported in English. However, there is a possibility of this influencing the results that were found from this review and so may introduce publication bias. This was highlighted to the committee during their deliberation.

These factors introduced additional uncertainty in the results. The effects on risk of bias did not appear to influence the direction of the effect in the trials. The committee took all of these factors into account when interpreting the evidence.

1.1.12.2.1. Transcutaneous electrical nerve stimulation (TENS)

Transcutaneous electrical nerve stimulation (TENS) was compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment.

  • When compared to neuromuscular electrical stimulation (NMES), 5 outcomes of very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended intervention, bias due to missing outcome data and bias in measurement of the outcome) and imprecision. The majority of outcomes included only one study, and at most two studies.
  • When compared to nerve blocks (local anaesthetic), 2 outcomes of very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process and bias in measurement of the outcome) and imprecision.
  • When compared to usual care or no treatment, 4 outcomes of very low quality were reported. This included the results from 1 trial with 54 participants. Outcomes were downgraded due to risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended intervention and bias due to missing outcome data) and imprecision.
1.1.12.2.2. Functional electrical stimulation

Functional electrical stimulation was compared to usual care or no treatment. This comparison included 4 outcomes of low or very low quality. This included results from 1 trial with 21 participants. Outcomes were generally downgraded due to risk of bias (due to bias arising from the randomisation process and bias in measurement of the outcome) and imprecision.

1.1.12.2.3. Neuromuscular electrical nerve stimulation (NMES)

Neuromuscular electrical nerve stimulation (NMES) was compared to transcutaneous electrical nerve stimulation (TENS), slings, placebo/sham and usual care or no treatment.

  • When compared to transcutaneous electrical nerve stimulation (TENS), 5 outcomes of very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended intervention, bias due to missing outcome data and bias in measurement of the outcome) and imprecision. The majority of outcomes included only one study, and at most two studies.
  • When compared to slings, 2 outcomes of low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process and bias due to missing outcome data).
  • When compared to placebo/sham, 4 outcomes of very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process and bias due to missing outcome data) and imprecision.
  • When compared to usual care or no treatment, 8 outcomes of low to very low quality were reported. These were downgraded due to risk of bias (due to a mixture of bias due to the randomisation process, bias due to deviations from the intended interventions, bias due to missing outcome data and bias in measurement of the outcome), inconsistency (in one outcome) and imprecision.
1.1.12.2.4. Devices - Tape

Tape was compared to placebo/sham and usual care or no treatment.

  • When compared to placebo/sham, 5 outcomes of low to very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended interventions and bias due to missing outcome data), inconsistency (in 1 outcome) and imprecision.
  • When compared to usual care or no treatment, 2 outcomes of low and very low quality were reported. These were downgrade due to either risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended interventions, bias due to missing outcome data and bias in measurement of the outcome) or risk of bias (due to deviations from the intended interventions) and imprecision.
1.1.12.2.5. Devices - Slings

Slings were compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment.

  • When compared to neuromuscular electrical stimulation (NMES), 2 outcomes of low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process and bias due to missing outcome data).
  • When compared to usual care or no treatment, 3 outcomes of very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended intervention, bias due to missing outcome data, bias in measurement of the outcome and bias in selection of the reported result), inconsistency (in 1 outcome) and imprecision.
1.1.12.2.6. Devices - Braces

Braces were compared to usual care or no treatment. This comparison included 2 outcomes of low and very low quality. This included the results from 1 trial. The outcomes were downgraded for either risk of bias (due to missing outcome data and bias in the measurement of outcome) or risk of bias (due to missing outcome data) and imprecision.

1.1.12.2.7. Acupuncture/dry needling

Acupuncture/dry needling was compared to placebo/sham and usual care or no treatment.

  • When compared to placebo/sham, 3 outcomes of very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process and bias due to missing outcome data) and imprecision.
  • When compared to usual care or no treatment, 5 outcomes of low or very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended interventions, bias due to missing outcome data and bias in measurement of the outcome), inconsistency (in 1 outcome) and imprecision.
1.1.12.2.8. Electroacupuncture

Electroacupuncture was compared to placebo/sham. This comparison included 1 outcome reported in 1 trial that was of very low quality. This was due to risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended interventions and bias in measurement of the outcome) and imprecision.

1.1.12.2.9. Intra-articular corticosteroids

Intra-articular corticosteroids were compared to placebo/sham. This comparison included 2 outcomes of very low quality. This was due to risk of bias (due to bias arising from the randomisation process, bias due to deviations from the intended interventions, bias due to missing outcome data and bias in measurement of the outcome), heterogeneity (in 1 outcome) and imprecision.

1.1.12.2.10. Nerve blocks (local anaesthetics)

Nerve blocks were compared to transcutaneous electrical nerve stimulation (TENS) and placebo/sham.

  • When compared to transcutaneous electrical nerve stimulation (TENS), 2 outcomes of very low quality were reported. These were downgraded due to risk of bias (due to bias arising from the randomisation process and bias in measurement of the outcome) and imprecision.
  • When compared to placebo/sham, 2 outcomes of low quality were reported. This was due to inconsistency (in 1 outcome) and imprecision.
1.1.12.3. Benefits and harms
1.1.12.3.1. Key uncertainties

The committee acknowledged the limited evidence available for all interventions in this review. Studies that were eligible for inclusion were often small and the quality of outcomes was often downgraded for risk of bias and imprecision. Based on this the committee agreed that more evidence investigating the effectiveness of all interventions would be important. Therefore, they agreed research recommendations to investigate this. The lack of certainty in the evidence made it difficult to determine the treatment that was most effective for shoulder pain after stroke. The committee decided that treatments where efficacy have been shown in this review should be considered as treatment options.

1.1.12.3.2. Transcutaneous electrical nerve stimulation (TENS)

Transcutaneous electrical nerve stimulation (TENS) was compared to neuromuscular electrical stimulation (NMES), nerve blocks (local anaesthetic) and usual care or no treatment. When compared to no treatment, no clinically important difference was seen in pain, physical function – upper limb, activities of daily living and stroke-specific Patient-Reported Outcome Measures at less than 6 months. When compared to neuromuscular electrical stimulation, a clinically important benefit in activities of daily living at less than 6 months was seen with transcutaneous electrical nerve stimulation in 1 study with 72 participants, while no clinically important difference was seen in physical function – upper limb, stroke-specific Patient-Reported Outcome Measures and withdrawal due to adverse events at less than 6 months. However, a clinically important benefit of neuromuscular electrical stimulation over TENS was seen in pain at less than 6 months in 2 studies with 110 participants. When compared to nerve blocks, nerve blocks (rather than TENS) showed clinically important benefits in reducing pain and improving stroke-specific Patient-Reported Outcome Measures at less than 6 months.

The committee acknowledged the limited evidence for benefit from TENS. The evidence primarily indicated that there was no clinically important benefit from the use of TENS and that other treatments (such as neuromuscular electrical stimulation and nerve blocks) were superior to TENS. Given this, they agreed that they would not recommend TENS for use for the management of shoulder pain after stroke.

1.1.12.3.3. Functional electrical stimulation

Functional electrical stimulation was compared to usual care or no treatment. A clinically important benefit of functional electrical stimulation was seen with pain at less than 6 months. No clinically important difference was seen in physical function – upper limb, activities of daily living and withdrawal due to adverse events at less than 6 months. These outcomes were all reported in 1 study with 21 participants.

The committee acknowledged the limited evidence discussing functional electrical stimulation. While this evidence did show a clinically important benefit in reducing pain, the outcomes came from 1 small study and given that other interventions had a more robust evidence base, the committee chose to recommend use of these treatments instead. However, the committee recommended for further research in the use of functional electrical stimulation in the research recommendations for this topic to allow for a more robust evaluation of the technique. In the meantime, the committee noted that functional electrical stimulation could be a treatment that may be effective to help reduce shoulder pain, but that the evidence was not sufficient to make a recommendation at this time.

1.1.12.3.4. Neuromuscular electrical nerve stimulation (NMES)

Neuromuscular electrical nerve stimulation (NMES) was compared to transcutaneous electrical nerve stimulation (TENS), slings, placebo/sham and usual care or no treatment. When compared to placebo/sham, clinically important benefits were seen in physical function – upper limb and activities of daily living at less than 6 months. However, no clinically important difference was seen in pain and withdrawal due to adverse events at less than 6 months. When compared to usual care or no treatment, clinically important benefits were seen in pain, activities of daily living and withdrawal due to adverse events. An unclear effect was seen in person/participant generic health-related quality of life at less than 6 months where a clinically important benefit was observed in the SF-36 mental component and no clinically important difference in the SF-36 physical component. An unclear effect was also seen for physical function – upper limb, where 1 outcome with 25 participants of low quality indicated a clinically important benefit while 1 outcome with 54 participants but of very low quality indicated no clinically important difference.

When compared to transcutaneous electrical nerve stimulation, a clinically important benefit in pain at less than 6 months was seen with neuromuscular electrical stimulation in 2 studies with 110 participants, while no clinically important difference was seen in physical function – upper limb, stroke-specific Patient-Reported Outcome Measures and withdrawal due to adverse events at less than 6 months. However, a clinically important benefit of transcutaneous electrical nerve stimulation was seen in activities of daily living at less than 6 months in 1 study with 72 participants. When compared to slings, a clinically important benefit of slings was seen in pain at less than 6 months and more than and equal to 6 months in outcomes from 1 study with 61 participants.

The committee acknowledged the inconsistency seen between the comparisons to placebo/sham and to usual care or no treatment. The comparison to placebo/sham indicated no clinically important difference of neuromuscular electrical nerve stimulation in reducing pain, while comparison to usual care or no treatment indicated a clinically important benefit. The committee agreed that the outcome showing no clinically important difference included inconsistency where 1 study showed a more beneficial effect and 1 study showed a more harmful effect, while the outcome showing benefit was based on 3 studies including 103 participants, with both outcomes being of very low quality. While they acknowledged the methodological concerns, the committee had greater certainty with the evidence of benefit. Based on the evidence of benefit when compared to usual care or no treatment, the committee agreed that neuromuscular electrical nerve stimulation should be considered for the treatment of post-stroke shoulder pain, and would also have benefits in other aspects of shoulder function, such as activities of daily living and upper limb motor function.

1.1.12.3.5. Devices - Tape

Tape was compared to placebo/sham and usual care or no treatment. When compared to placebo/sham, clinically important benefits were seen in pain and activities of daily living at less than 6 months. However, no clinically important difference was seen in physical function – upper limb, stroke-specific Patient-Reported Outcome Measures and withdrawal due to adverse events. When compared to usual care or no treatment, a clinically important difference was seen in pain at less than 6 months, while no clinically important difference was seen in withdrawal due to adverse events at less than 6 months.

The committee agreed that evidence of benefit for pain was seen when tape was compared to placebo/sham and to usual care or no treatment without showing any harms. The committee noted that taping may be useful for people with 1) hypotonic/unstable presentation of shoulder pain, 2) subacromial shoulder pain to optimise joint alignment. They acknowledged that this may not be the most common presentations of shoulder pain. A lay member on the committee discussed their experience, that tape was useful in reducing pain but would need replacing regularly and they would not be able to do that themselves. The practicalities of using tape for treatment needs to be considered by the stroke survivor and those supporting them when considering the treatment. The committee agreed that tape should be considered for the treatment of post-stroke shoulder pain.

1.1.12.3.6. Devices - Slings

Slings were compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment. When compared to usual care or no treatment, no clinically important difference was seen in pain and physical function – upper limb at less than 6 months. However, a clinically important harm of slings was seen in withdrawal due to adverse events in an outcome including 1 study with 32 participants with the outcome being of very low quality. When compared to neuromuscular electrical stimulation, a clinically important benefit of slings was seen in pain at less than 6 months and more than and equal to 6 months in outcomes from 1 study with 61 participants.

The committee agreed that there was no evidence of benefit with slings with potential evidence of harm in adverse events. However, they acknowledged that the harm in adverse events was due to 1 withdrawal in a small study, which made the applicability of this evidence limited. The committee reflected that in clinical practice there were people who would benefit from shoulder slings (including people with subluxed shoulders). Shoulder slings may be able to prevent future problems from taking place. However, they noted that sling use may lead to secondary stiffness, that can cause loss of range, pain and further disuse weakness. Taking this into account, the committee agreed that they would not make a recommendation on the use of slings as the evidence had not demonstrated convincingly that slings were effective at reducing shoulder pain after stroke. However, they noted that for some people after stroke a sling may be an effective treatment and did not make a recommendation that they should not be used. They recommended that slings should be considered as a part of further research in this area to investigate whether they could be an effective treatment for certain causes of shoulder pain.

1.1.12.3.7. Devices - Braces

Braces were compared to usual care or no treatment. A clinically important benefit of braces was seen in pain at less than 6 months. However, a clinically important harm was observed in withdrawal due to adverse events at less than 6 months including 1 study with 41 participants with the outcome being of very low quality.

The committee acknowledge the inconsistent evidence of benefits in reducing pain but harms in withdrawal due to adverse events. The committee acknowledged that the harm in adverse events was due to 1 withdrawal in a small study, which made the applicability of this evidence limited. The committee reflected on their experience with lay members noting that they did not experience benefits from this, while healthcare professionals acknowledged that there may be some people where braces may be more helpful than others. The committee noted that this may be used for people with dense (severe) upper limb weakness and for people with subluxation. The committee weighed up these factors and agreed that due to the limited evidence compared to other interventions, they would not make a recommendation on the use of braces. However, they noted that braces may be effective for some people with shoulder pain and did not make a recommendation that braces should not be used. They included braces in a research recommendation to investigate whether they could be an effective treatment for certain causes of shoulder pain.

1.1.12.3.8. Acupuncture/dry needling

Acupuncture/dry needling was compared to placebo/sham and usual care or no treatment. When compared to placebo/sham, a clinically important benefit of acupuncture/dry needling was seen in pain at less than 6 months. No clinically important difference was seen in withdrawal due to adverse events. However, a clinically important benefit of placebo/sham was seen in activities of daily living. When compared to usual care or no treatment, clinically important benefits were seen in person/participant generic health-related quality of life and pain at less than 6 months. However, no clinically important difference was seen in physical function – upper limb and withdrawal due to adverse events at less than 6 months.

The committee acknowledged the evidence of benefits from acupuncture/dry needling. The committee acknowledged the limited evidence and how this may be further limited by a significant number of studies not being translated in English and so not being able to be checked for their relevance for inclusion in this review, which may have led to additional studies being added for this consideration. The committee agreed that acupuncture may be helpful for people with shoulder pain after stroke. The committee considered this evidence against the considerations of cost effectiveness and resource use.

Taking this into account, the committee acknowledged that there was insufficient evidence to recommend acupuncture at this time, but noted that the evidence appeared to be positive and recommended that acupuncture should have further research conducted, which included cost-effectiveness analysis, to investigate whether it could be a clinically and cost effective treatment to use for the management of shoulder pain after stroke in an NHS context.

1.1.12.3.9. Electroacupuncture

Electroacupuncture was compared to placebo/sham. 1 outcome was included for this comparison, pain at less than 6 months, where there was a clinically important benefit of electroacupuncture.

The committee acknowledged the evidence of benefits from electroacupuncture. The committee acknowledged the limited evidence and how this may be further limited by a significant number of studies not being translated in English and so not being able to be checked for their relevance for inclusion in this review, which may have led to additional studies being added for this consideration. The committee agreed that electroacupuncture may be helpful for people with shoulder pain after stroke. The committee considered this evidence against the considerations of cost effectiveness and resource use.

Taking this into account, the committee acknowledged that there was insufficient evidence to recommend electroacupuncture at this time, but noted that the evidence appeared to be positive and recommended that acupuncture should have further research conducted, which included cost-effectiveness analysis, to investigate whether it could be a clinically and cost effective treatment to use for the management of shoulder pain after stroke in an NHS context.

1.1.12.3.10. Intra-articular corticosteroids

Intra-articular corticosteroids were compared to placebo/sham. 2 outcomes were included for this comparison, pain and activities of daily living at less than 6 months, where there were clinically important benefits of intra-articular corticosteroids.

The committee acknowledged the evidence of benefits from intra-articular corticosteroids. On examining the studies, the committee noted that the identified studies only included one injection of intra-articular corticosteroids. The committee agreed that in their expert opinion there were benefits from use of intra-articular corticosteroids. The committee noted that these may be provided in primary care by general practitioners with a special interest, or in secondary care where it may be given under ultrasound guidance by radiologists, sports and exercise medicine clinicians or rehabilitation medicine physicians. Therefore, based on the limited evidence and committee’s expert opinion, they agreed that intra-articular corticosteroids should be considered for the treatment of post-stroke shoulder pain.

1.1.12.3.11. Nerve blocks (local anaesthetics)

Nerve blocks were compared to TENS and placebo/sham. When compared to TENS, clinically important benefits of nerve blocks were seen in reducing pain and improving stroke-specific Patient-Reported Outcome Measures at less than 6 months. When compared to placebo/sham, a clinically important benefit of nerve blocks was seen in pain at less than 6 months. No clinically important difference was seen in withdrawal due to adverse events at less than 6 months.

The committee acknowledged the benefits from nerve blocks. The nerve blocks included in this review included a combination of local anaesthetic and corticosteroids. The committee acknowledged their experiences that nerve blocks can be useful for some people with shoulder pain after stroke. The committee noted that providing nerve blocks required specialist input to provide them, which could include anaesthetists or another interventional clinician such as a sports and exercise medicine or rehabilitation medicine consultant. Taking into account these factors, the committee agreed that nerve blocks should be considered for the treatment of post-stroke shoulder pain.

1.1.12.4. Cost effectiveness and resource use

No relevant health economic analyses were identified for this review; therefore, unit costs were presented to aid committee consideration of cost-effectiveness.

1.1.12.4.1. Electrotherapies (FES, NMES, TENS)

The cost of electrotherapies relates primarily to the staff time to administer it and will depend on frequency and duration of therapy sessions, as well as the duration of treatment. There are also equipment costs, however, these were not presented to the committee as previous economic evaluations of electrotherapies did not include the costs of equipment as the per-use costs were expected to be small.

1.1.12.4.2. Transcutaneous electrical nerve stimulation (TENS)

The cost of a TENS machine varies (approximately £18-£50) and can be used at home which could incur less resource use relative to interventions that require staff supervision. However, the clinical evidence summarised in the section 1.1.12.2.1 indicated that there was no clinically important benefit from the use of transcutaneous electrical nerve stimulation compared to usual care and no treatment. The lack of clinical evidence and additional resource use required compared to usual care led the committee to agree to not recommend transcutaneous electrical nerve stimulation for the management of shoulder pain in post-stroke adults.

1.1.12.4.3. Functional electrical stimulation (FES)

Previous NHS reports on FES38 included an initial assessment appointment costing £140. The analysis also included a clinic model in which the costs of the FES device are incorporated in the ongoing clinical charges. Each ongoing clinical appointment was estimated at £300. The experience of some committee members suggested a less expensive alternative where FES can be delivered at home without staff supervision, although it was acknowledged that the number of FES devices available to take home varies across current practice. The clinical evidence (section 1.1.12.2.2) showed that when FES was compared to usual care or no treatment, a clinically important benefit of FES was seen with pain at less than 6 months. However, no clinically important difference was seen in physical function – upper limb, activities of daily living and withdrawal due to adverse events at less than 6 months. Given the limited clinical evidence and lack of cost-effective evidence the committee decided to not recommend FES for the treatment of post-stroke shoulder pain.

1.1.12.4.4. Neuromuscular electrical nerve stimulation (NMES)

NMES was the most frequently evaluated of out the electrotherapy interventions (7 studies included in clinical review) and was compared to transcutaneous electrical nerve stimulation (TENS), slings, placebo/sham and usual care, or no treatment. It was challenging for the committee to determine resource use for NMES as the frequency and duration reported in the studies varied, with sessions ranging from 1–6-hours and were delivered between 3–7 days per week for 3–8 weeks.

Despite committee acknowledgement of the inconsistency seen between the comparisons to placebo/sham and to usual care or no treatment, it was agreed that there was more evidence of benefit than harm in the clinical evidence for NMES (see section 1.1.12.2.3) and agreed that and would also have benefits in other aspects of shoulder function, such as activities of daily living and upper limb motor function. For these reasons, alongside the lack of published health economic evidence, the committee agreed that NMES should be considered for the treatment of post-stroke shoulder pain.

1.1.12.4.5. Devices

The committee were informed that the following devices could take place at home which could incur lower resource use compared to other interventions in this review, with people tasked with wearing the devices all day or whenever the upper limb was unsupported.

1.1.12.4.6. Slings

The cost of the sling reported in the clinical studies was relatively low (£6.38) and staff time involved in the application and correction of the sling is less resource intensive compared to other shoulder-pain related interventions and can be incorporated into standard therapy. As previously summarised in section 1.1.12.2.5 above, 1 clinical study comparing slings to NMES found a clinically important benefit in pain at less than 6 months and more than and equal to 6 months in outcomes. No evidence of benefit was seen when slings were compared to usual care or no treatment, with limited evidence of a clinically important harm of slings for withdrawal due to adverse events. Despite limited clinical evidence, the committee’s experience in clinical practice had demonstrated the benefits of shoulder slings for some individuals (including people with subluxed shoulders), alongside the potential for slings to prevent future problems from taking place. However, without cost-effectiveness evidence and no clinical evidence of benefit when compared to usual care or no treatment, the committee agreed that slings were not recommended for the treatment of post-stroke shoulder pain.

1.1.12.4.7. Tape

Tape is relatively low cost (£2.14) compared to the other devices in this review. However, both the clinical evidence and a lay member’s experience of this intervention noted that a therapist is required to reapply the tape, resulting in frequent visits which could increase staff time costs. As described in section 1.1.12.2.4, studies comparing tape to usual care or no treatment placebo/sham found clinically important benefits in pain at less than 6 months. However, the comparison to placebo/sham indicated no clinically important difference in physical function – upper limb, stroke-specific Patient-Reported Outcome Measures and withdrawal due to adverse events.

The committee noted that taping would not be a practical treatment for all stroke survivors and that it may be useful for people with less common presentations of shoulder pain. This could possibly lower the impact on resource as less people would be ideal candidates for taping. Based on the limited clinical and economic evidence, the committee agreed that tape should be considered for the treatment of post-stroke shoulder pain, emphasising that stroke survivors and those supporting them should first consider the practicalities of using tape before beginning treatment.

1.1.12.4.8. Braces

One study reported the use of a shoulder brace which was significantly more costly than the other devices (£212). Although this specific device was not reported in the NHS supply chain catalogue, it was noted by a committee member this was one of the braces used in current practice and that a similar cost or higher (approximately £250) would apply for other shoulder braces typically used. It was also noted that shoulder braces are not customised to order but given the different sizes, some staff time is required for fitting the brace.

Committee members noted not everyone with post-stroke shoulder pain would be eligible for this treatment, as using a brace is thought to prevent future problems for some people while causing additional harm in others, particularly in instances where the shoulder is already very inflamed. There was limited clinical evidence (section 1.1.12.2.6) with the only included study reporting inconsistent evidence of benefits in reducing pain but harms in withdrawal due to adverse events. Given the lack of clinical evidence and economic evidence, alongside additional resource use requirements, the committee agreed to not recommend braces for the treatment of post-stroke shoulder pain.

1.1.12.4.9. Acupuncture/dry needling

The cost of acupuncture relates primarily to the staff time required to deliver treatment, with an outpatient procedure for acupuncture for pain management costing £141, although costs in the community might be lower. The frequency and duration for delivering acupuncture varied across studies, ranging from a one-off session with a 1-week follow-up to once daily for one month. Sessions typically lasted 30 minutes. Equipment costs for acupuncture are low as it mainly consists of the cost of needles (£0.06 per needle, with 10–14 needles used per session). The committee regarded acupuncture and electroacupuncture as one of the less frequently provided treatments for shoulder pain following stroke, meaning that staff training may be required to deliver these interventions.

The limited clinical evidence (reported in section 1.1.12.2.7) for acupuncture included a clinically important benefit for pain at less than 6 months for acupuncture when compared to both placebo/sham and usual care or no treatment. No clinically important difference was seen in withdrawal due to adverse events for either comparison, however a clinically important benefit was seen for placebo/sham in activities of daily living. The lack of clinical evidence for acupuncture may have been due to several studies that were not assessed because they were not published in English. Given the limited clinical evidence and lack of economic evidence, alongside additional resource use requirements, the committee agreed to not recommend acupuncture for the treatment of post-stroke shoulder pain.

1.1.12.4.10. Electroacupuncture

Aside from the staff time required to deliver electroacupuncture, example costs of electroacupuncture devices were presented to the committee, which ranged from £240-£534. Other costs associated with electrotherapy include clips, lead cables, batteries, needles, disinfectant swabs, and surgeons’ gloves. Clinical evidence for electroacupuncture was based on a single study that indicated a clinically important benefit for pain at less than 6 months when compared to placebo/sham (see section 1.1.12.2.8). As with standard acupuncture, the lack of clinical evidence for electroacupuncture may have been due to several studies that were not assessed because they were not published in English.

Given the limited clinical evidence and lack of economic evidence, alongside additional resource use requirements, the committee agreed to not recommend electroacupuncture for the treatment of post-stroke shoulder pain.

1.1.12.4.11. Intra-articular corticosteroids and Nerve blocks (local anaesthetics)

Participants in all clinical studies received a single injection and were followed up from between 3 to 12 weeks post-intervention. The committee agreed that 1–2 injections was typical of current practice.

The committee were informed that resource use relates to the drugs injected and the staff time to deliver the injections. Resource use between studies differed due to the cost of drugs, as the 4 included studies used different drug combinations and doses. The average drug cost per injection for each of the combinations and doses from the studies were created using drug costs from the BNF. The cost per injection was low £1.99-£11.92, with the most expensive being attributed to betamethasone, which the committee noted would not be used in an NHS clinical setting. The impact on resource use would also be dependent on the staff involved in the injection procedure, outpatient appointment costs associated with having an injection for pain management ranged between £752-£826 (for injection of therapeutic substance) to £529-£910 (for nerve block), with the higher ranges accounting for the use of ultrasonography. Training costs are another factor that could incur additional resource use, as one study reported the use of two rehabilitation doctors (physiatrists) and a radiologist who were required to attend a 2-day training course.

When nerve blocks were compared to placebo/sham, a clinically important benefit of nerve blocks was seen in pain at less than 6 months (see section 1.1.12.2.10). No clinically important difference was seen in withdrawal due to adverse events at less than 6 months. Committee experience of nerve blocks in clinical practice has shown benefits to some people with shoulder pain after stroke. However, the committee also acknowledged the resource use associated with nerve blocks, as specialist input is required to provide them, which could include anaesthetists or another interventional clinician such as a sports and exercise medicine or rehabilitation medicine consultant. In consideration of these factors, the committee decided that nerve blocks should be considered for the treatment of post-stroke shoulder pain.

The committee noted that the clinical evidence contained only one study for intra-articular corticosteroids, which found clinically important benefits for pain and activities of daily living at less than 6 months when compared to placebo/sham (see section 1.1.12.2.9). The committee agreed that in their expert opinion there were benefits from use of intra-articular corticosteroids. Disparity in resource use across current practice was also acknowledged, as intra-articular corticosteroids can be provided in secondary care involving specialist input, or in primary care by general practitioners (which would lower resource use); however, this is dependent on the clinician being comfortable with administering the injection. The limited clinical evidence and the committee’s expert opinion, paired with a lack of economic evidence lead the committee to agree that intra-articular corticosteroids should be considered for the treatment of post-stroke shoulder pain.

1.1.12.5. Other factors the committee took into account

The committee acknowledged the potential costs of treatments. Some treatments may be accessed outside of the NHS. Electrotherapy (including transcutaneous electrical nerve stimulation and functional electrical stimulation) and devices may be purchased without healthcare professional input, which can incur costs on stroke survivors. Acupuncture and electroacupuncture may be accessed more commonly by people with an Asian family background, which can lead to inequities in care where people may access this treatment privately instead of through the NHS.

The committee acknowledged that the treatment of shoulder pain after stroke should be dependent on the cause of the shoulder pain, which is often multifactorial but can include pain from glenohumeral subluxation, spasticity of shoulder muscles, impingement, soft tissue injury, rotator cuff tears, glenohumeral capsulitis, bicipital tendinitis and shoulder hand syndrome44. Therefore, it could be argued that treatment needs to be specific to the person after stroke. The committee acknowledged that the included studies did not investigate all of these causes and so it is difficult to conclude which treatments are more effective for each cause. The committee agreed that pain should be managed by the cause of the pain but noted that research was not currently designed in this manner, so made a research recommendation for research to be conducted to investigate whether assessing the cause of the shoulder pain and then treating accordingly is the best management strategy for post-stroke shoulder pain.

The committee noted that the majority of the evidence investigated people who did not have pre-existing shoulder conditions but acknowledged that, if present, such conditions would also have a role on the management required.

Furthermore, the committee agreed that it was often not apparent whether shoulder pain was acute or chronic in the studies they reviewed. The epidemiology of shoulder pain after stroke is unclear, with there being limited information about the proportion of cases that persisted beyond 3 months. The committee acknowledged that chronic pain could have a significant effect and may require different management to acute pain, including psychological therapy. The involvement of psychological services to support people with chronic secondary pain due to stroke-related shoulder pain should be considered if that is thought to be appropriate by the healthcare professionals involved in the person’s care.

1.1.13. Recommendations supported by this evidence review

This evidence review supports recommendations 1.14.2 to 1.14.4 and the research recommendations on the management of shoulder pain by cause and diagnostic assessment to inform management of shoulder pain.

1.1.14. References

1.
Ada L, Foongchomcheay A, Canning C. Supportive devices for preventing and treating subluxation of the shoulder after stroke. Cochrane Database of Systematic Reviews 2005, Issue 1. DOI: 10.1002/14651858.cd003863.pub2. [PMC free article: PMC6984447] [PubMed: 15674917] [CrossRef]
2.
Adey-Wakeling Z, Crotty M, Shanahan EM. Suprascapular nerve block for shoulder pain in the first year after stroke: a randomized controlled trial. Stroke. 2013; 44(11):3136–3141 [PubMed: 23970790]
3.
Allen ZA, Shanahan EM, Crotty M. Does suprascapular nerve block reduce shoulder pain following stroke: a double-blind randomised controlled trial with masked outcome assessment. BMC Neurology. 2010; 10:83 [PMC free article: PMC2954969] [PubMed: 20854696]
4.
Chae J, Ng A, Yu DT, Kirsteins A, Elovic EP, Flanagan SR et al Intramuscular electrical stimulation for shoulder pain in hemiplegia: does time from stroke onset predict treatment success? Neurorehabilitation and Neural Repair. 2007; 21(6):561–567 [PubMed: 17369520]
5.
Chae J, Yu DT, Walker ME, Kirsteins A, Elovic EP, Flanagan SR et al Intramuscular electrical stimulation for hemiplegic shoulder pain: a 12-month follow-up of a multiple-center, randomized clinical trial. American Journal of Physical Medicine and Rehabilitation. 2005; 84(11):832–842 [PubMed: 16244520]
6.
Chuang LL, Chen YL, Chen CC, Li YC, Wong AM, Hsu AL et al Effect of EMG-triggered neuromuscular electrical stimulation with bilateral arm training on hemiplegic shoulder pain and arm function after stroke: a randomized controlled trial. Journal of Neuroengineering and Rehabilitation. 2017; 14(1):122 [PMC free article: PMC5706163] [PubMed: 29183339]
7.
de Jong LD, Dijkstra PU, Gerritsen J, Geurts AC, Postema K. Combined arm stretch positioning and neuromuscular electrical stimulation during rehabilitation does not improve range of motion, shoulder pain or function in patients after stroke: a randomised trial. Journal of Physiotherapy. 2013; 59(4):245–254 [PubMed: 24287218]
8.
DiLorenzo L, Traballesi M, Morelli D, Pompa A, Brunelli S, Buzzi MG et al Hemiparetic shoulder pain syndrome treated with deep dry needling during early rehabilitation: A prospective, open-label, randomized investigation. Journal of Musculoskeletal Pain. 2004; 12(2):25–34
9.
Ersoy S, Paker N, Kesiktas FN, Bugdayci DS, Karakaya E, Cetin M. Comparison of transcutaneous electrical stimulation and suprascapular nerve blockage for the treatment of hemiplegic shoulder pain. Journal of Back and Musculoskeletal Rehabilitation. 2022; [PubMed: 36565102]
10.
Harmony Medical. AS-super 4 Cables. 2022. Available from: https://www​.harmonymedical​.co.uk/crocodile-clips-cable-for-as-super-4-digital-pack-of-4.html Last accessed: 01/02/2023.
11.
Harmony Medical. AS-super 4 electroacupuncture device 2022. Available from: https://www​.harmonymedical​.co.uk/as-super-4-digital-needle-stimulator.html Last accessed: 01/02/2023.
12.
Harmony Medical. ES-160 cables 2022. Available from: https://www​.harmonymedical​.co.uk/lead-suitable-for-es-130-units.html Last accessed: 01/02/2023.
13.
Harmony Medical. ES-160 electroacupuncture device. 2022. Available from: https://www​.harmonymedical​.co.uk/electro-acupuncture-machine-es-160.html Last accessed: 01/02/2023.
14.
Hartwig M, Gelbrich G, Griewing B. Functional orthosis in shoulder joint subluxation after ischaemic brain stroke to avoid post-hemiplegic shoulder-hand syndrome: a randomized clinical trial. Clinical Rehabilitation. 2012; 26(9):807–816 [PubMed: 22308558]
15.
Heo MY, Kim CY, Nam CW. Influence of the application of inelastic taping on shoulder subluxation and pain changes in acute stroke patients. Journal of Physical Therapy Science. 2015; 27(11):3393–3395 [PMC free article: PMC4681912] [PubMed: 26696705]
16.
Huang YC, Chang KH, Liou TH, Cheng CW, Lin LF, Huang SW. Effects of Kinesio taping for stroke patients with hemiplegic shoulder pain: A double-blind, randomized, placebo-controlled study. Journal of Rehabilitation Medicine. 2017; 49(3):208–215 [PubMed: 28233009]
17.
Huang YC, Leong CP, Wang L, Wang LY, Yang YC, Chuang CY et al Effect of kinesiology taping on hemiplegic shoulder pain and functional outcomes in subacute stroke patients: a randomized controlled study. European journal of physical & rehabilitation medicine. 2016; 52(6):774–781 [PubMed: 27575012]
18.
iDryNeedle. ES-160 crocodile clips. 2022. Available from: https://www​.idryneedle​.com/ito-es-160-replacement-clips.html Last accessed: 01/02/2023.
19.
Joint Formulary Committee. British National Formulary (online). 2022. Available from: https://bnf​.nice.org.uk/ Last accessed: 01/02/2023.
20.
Jones K, Burns A. Unit costs of health and social care 2021. Canterbury. Personal Social Services Research Unit University of Kent, 2021. Available from: https://www​.pssru.ac​.uk/project-pages/unit-costs​/unit-costs-of-health-and-social-care-2021/
21.
Karaahmet OZ, Gurcay E, Unal ZK, Cankurtaran D, Cakci A. Effects of functional electrical stimulation-cycling on shoulder pain and subluxation in patients with acute-subacute stroke: a pilot study. International Journal of Rehabilitation Research. 2019; 42(1):36–40 [PubMed: 30234733]
22.
Lakse E, Gunduz B, Erhan B, Celik EC. The effect of local injections in hemiplegic shoulder pain: a prospective, randomized, controlled study. American Journal of Physical Medicine and Rehabilitation. 2009; 88(10):805–811; quiz 812 [PubMed: 21119312]
23.
Lavi C, Elboim-Gabyzon M, Naveh Y, Kalichman L. A Combination of Long-Duration Electrical Stimulation with External Shoulder Support during Routine Daily Activities in Patients with Post-Hemiplegic Shoulder Subluxation: A Randomized Controlled Study. International Journal of Environmental Research and Public Health. 2022; 19(15) [PMC free article: PMC9368353] [PubMed: 35955118]
24.
Lee GE, Son C, Lee J, Lee SH, Lee HJ, Lee KJ et al Acupuncture for shoulder pain after stroke: A randomized controlled clinical trial. European Journal of Integrative Medicine. 2016; 8(4):373–383
25.
MacPherson H, Vickers A, Bland M, Torgerson D, Corbett M, Spackman E et al Acupuncture for chronic pain and depression in primary care: a programme of research. Programme Grants for Applied Research. Southampton (UK). 2017. [PubMed: 28121095]
26.
Mendigutia-Gomez A, Quintana-Garcia MT, Martin-Sevilla M, de Lorenzo-Barrientos D, Rodriguez-Jimenez J, Fernandez-de-Las-Penas C et al Post-needling soreness and trigger point dry needling for hemiplegic shoulder pain following stroke. Acupuncture in Medicine. 2020; 38(3):150–157 [PubMed: 32009422]
27.
Moghe DM, Kanase SB. Effect of therapeutic shoulder sling and proximal control exercises on shoulder subluxation in stroke survivors. Indian Journal of Forensic Medicine and Toxicology. 2020; 14(3):222–227
28.
National Institute for Health and Care Excellence. Chronic pain (primary and secondary) in over 16s: assessment of all chronic pain and management of chronic primary pain [NG193]. London. 2021. Available from: https://www​.nice.org.uk/guidance/ng193 [PubMed: 33939353]
29.
National Institute for Health and Care Excellence. Cost-utility analysis: Electroacupuncture for the management of osteoarthritis. 2022. Available from: https://www​.nice.org​.uk/guidance/ng226/evidence​/electroacupuncture-economic-analysis-pdf-11251847341
30.
National Institute for Health and Care Excellence. Developing NICE guidelines: the manual [updated January 2022]. London. National Institute for Health and Care Excellence, 2014. Available from: https://www​.nice.org.uk/process/pmg20
31.
NHS England and NHS Improvement. 2019/20 National Cost Collection Data Publication. 2022. Available from: https://www​.england.nhs​.uk/publication/2019-20-national-cost-collection-data-publication Last accessed: 01/02/2023.
32.
NHS Supply Chain. NHS Supply Chain Catalogue. 2022. Available from: https://my​.supplychain.nhs.uk/catalogue Last accessed: 01/02/2023.
33.
Pandian JD, Kaur P, Arora R, Vishwambaran DK, Toor G, Mathangi S et al Shoulder taping reduces injury and pain in stroke patients: randomized controlled trial. Neurology. 2013; 80(6):528–532 [PubMed: 23345636]
34.
Pillastrini P, Rocchi G, Deserri D, Foschi P, Mardegan M, Naldi MT et al Effectiveness of neuromuscular taping on painful hemiplegic shoulder: a randomised clinical trial. Disability and Rehabilitation. 2016; 38(16):1603–1609 [PubMed: 26678717]
35.
Price C, Pandyan A. Electrical stimulation for preventing and treating post?stroke shoulder pain. Cochrane Database of Systematic Reviews 2000, Issue 4. DOI: 10.1002/14651858.cd001698. [PMC free article: PMC8406756] [PubMed: 11034725] [CrossRef]
36.
Rah UW, Yoon SH, Moon DJ, Kwack KS, Hong JY, Lim YC et al Subacromial corticosteroid injection on poststroke hemiplegic shoulder pain: a randomized, triple-blind, placebo-controlled trial. Archives of Physical Medicine and Rehabilitation. 2012; 93(6):949–956 [PubMed: 22483593]
37.
Sui M, Jiang N, Yan L, Liu J, Luo B, Zhang C et al Effect of Electroacupuncture on Shoulder Subluxation in Poststroke Patients with Hemiplegic Shoulder Pain: A Sham-Controlled Study Using Multidimensional Musculoskeletal Ultrasound Assessment. Pain Research & Management. 2021; 2021:5329881 [PMC free article: PMC8626186] [PubMed: 34840636]
38.
Taft R, Lovell A, Cook R. Functional electrical stimulation for dropped foot of central neurological origin: economic report London. 2010. Available from: http://www​.cedar.wales​.nhs.uk/sitesplus/documents​/1091/CEP10012​%20FES%20econ%20report.pdf
39.
Terlemez R, Ciftci S, Topaloglu M, Dogu B, Yilmaz F, Kuran B. Suprascapular nerve block in hemiplegic shoulder pain: comparison of the effectiveness of placebo, local anesthetic, and corticosteroid injections-a randomized controlled study. Neurological Sciences. 2020; 41(11):3243–3247 [PubMed: 32388647]
40.
Turkkan C, Ozturk GT, Ugurlu FG, Ersoz M. Ultrasonographic assessment of neuromuscular electrical stimulation efficacy on glenohumeral subluxation in patients with hemiplegia: a randomized-controlled study. Turkish Journal of Physical Medicine and Rehabilitation. 2017; 63(4):287–292 [PMC free article: PMC6648069] [PubMed: 31453469]
41.
van Bladel A, Lambrecht G, Oostra KM, Vanderstraeten G, Cambier D. A randomized controlled trial on the immediate and long-term effects of arm slings on shoulder subluxation in stroke patients. European journal of physical & rehabilitation medicine. 2017; 53(3):400–409 [PubMed: 28145396]
42.
Vickers AJ, Vertosick EA, Lewith G, MacPherson H, Foster NE, Sherman KJ et al Acupuncture for Chronic Pain: Update of an Individual Patient Data Meta-Analysis. Journal of Pain. 2018; 19(5):455–474 [PMC free article: PMC5927830] [PubMed: 29198932]
43.
Vitego. Functional orthosis SPORLASTIC NEURO-LUX II right 8. 2022. Available from: https://www​.vitego-shop​.de/Bandagen-Schulter-SPORLASTIC-NEURO-LUX-II-black-right-8 Last accessed: 01/02/2023.
44.
Walsh K. Management of shoulder pain in patients with stroke. Postgraduate Medical Journal. 2001; 77(912):645–649 [PMC free article: PMC1742144] [PubMed: 11571371]
45.
Wilson RD, Gunzler DD, Bennett ME, Chae J. Peripheral nerve stimulation compared with usual care for pain relief of hemiplegic shoulder pain: a randomized controlled trial. American Journal of Physical Medicine and Rehabilitation. 2014; 93(1):17–28 [PMC free article: PMC4038163] [PubMed: 24355994]
46.
Wilson RD, Knutson JS, Bennett ME, Chae J. The Effect of Peripheral Nerve Stimulation on Shoulder Biomechanics: A Randomized Controlled Trial in Comparison to Physical Therapy. American Journal of Physical Medicine and Rehabilitation. 2017; 96(3):191–198 [PMC free article: PMC5321785] [PubMed: 28099193]
47.
Woods B, Manca A, Weatherly H, Saramago P, Sideris E, Giannopoulou C et al Cost-effectiveness of adjunct non-pharmacological interventions for osteoarthritis of the knee. PloS One. 2017; 12(3):e0172749 [PMC free article: PMC5340388] [PubMed: 28267751]
48.
Yang L, Yang J, He C. The Effect of Kinesiology Taping on the Hemiplegic Shoulder Pain: A Randomized Controlled Trial. Journal of Healthcare Engineering. 2018; 2018:8346432 [PMC free article: PMC6311752] [PubMed: 30651946]
49.
Yu DT, Chae J, Walker ME, Kirsteins A, Elovic EP, Flanagan SR et al Intramuscular neuromuscular electric stimulation for poststroke shoulder pain: a multicenter randomized clinical trial. Archives of Physical Medicine and Rehabilitation. 2004; 85(5):695–704 [PubMed: 15129391]
50.
Zhan J, Ai Y, Zhan L, Pan R, Wang Y, Dong C et al Effect of abdominal acupuncture combined with routine rehabilitation training on shoulder-hand syndrome after stroke: A randomized controlled trial. Integrative Medicine Research. 2022; 11(2):100805 [PMC free article: PMC8627967] [PubMed: 34877254]
51.
Zheng J, Wu Q, Wang L, Guo T. A clinical study on acupuncture in combination with routine rehabilitation therapy for early pain recovery of post-stroke shoulder-hand syndrome. Experimental and Therapeutic Medicine. 2018; 15(2):2049–2053 [PMC free article: PMC5776623] [PubMed: 29434804]
52.
Zhou M, Li F, Lu W, Wu J, Pei S. Efficiency of Neuromuscular Electrical Stimulation and Transcutaneous Nerve Stimulation on Hemiplegic Shoulder Pain: A Randomized Controlled Trial. Archives of Physical Medicine and Rehabilitation. 2018; 99(9):1730–1739 [PubMed: 29777714]

Appendices

Appendix B. Literature search strategies

B.1. Clinical search literature search strategy

Searches were constructed using a PICO framework where population (P) terms were combined with Intervention (I) and in some cases Comparison (C) terms. Outcomes (O) are rarely used in search strategies as these concepts may not be indexed or described in the title or abstract and are therefore difficult to retrieve. Search filters were applied to the search where appropriate.

Download PDF (215K)

B.2. Health Economics literature search strategy

Health economic evidence was identified by conducting searches using terms for a broad Stroke Rehabilitation population. The following databases were searched: NHS Economic Evaluation Database (NHS EED - this ceased to be updated after 31st March 2015), Health Technology Assessment database (HTA - this ceased to be updated from 31st March 2018) and The International Network of Agencies for Health Technology Assessment (INAHTA). Searches for recent evidence were run on Medline and Embase from 2014 onwards for health economics, and all years for quality-of-life studies. Additional searches were run in CINAHL and PsycInfo looking for health economic evidence.

Download PDF (202K)

Appendix C. Effectiveness evidence study selection

Download PDF (246K)

Appendix D. Effectiveness evidence

Adey-Wakeling, 2013 (PDF, 192K)

Allen, 2010 (PDF, 118K)

Chae, 2007 (PDF, 219K)

Chae, 2005 (PDF, 216K)

Chuang, 2017 (PDF, 205K)

de Jong, 2013 (PDF, 193K)

DiLorenzo, 2004 (PDF, 195K)

Ersoy, 2022 (PDF, 193K)

Hartwig, 2012 (PDF, 193K)

Heo, 2015 (PDF, 185K)

Huang, 2017 (PDF, 198K)

Huang, 2016 (PDF, 200K)

Karaahmet, 2019 (PDF, 206K)

Lakse, 2009 (PDF, 189K)

Lavi, 2022 (PDF, 203K)

Lee, 2016 (PDF, 201K)

Mendigutia-Gomez, 2020 (PDF, 197K)

Moghe, 2020 (PDF, 186K)

Pandian, 2013 (PDF, 194K)

Pillastrini, 2016 (PDF, 195K)

Rah, 2012 (PDF, 195K)

Sui, 2021 (PDF, 194K)

Terlemez, 2020 (PDF, 186K)

Turkkan, 2017 (PDF, 189K)

van Bladel, 2017 (PDF, 198K)

Wilson, 2014 (PDF, 221K)

Wilson, 2017 (PDF, 152K)

Yang, 2018 (PDF, 184K)

Yu, 2004 (PDF, 224K)

Zhan, 2022 (PDF, 203K)

Zheng, 2018 (PDF, 195K)

Zhou, 2018 (PDF, 203K)

Appendix E. Forest plots

E.1. Transcutaneous electrical nerve stimulation (TENS) compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment

E.1.1. Transcutaneous electrical nerve stimulation (TENS) compared to neuromuscular electrical stimulation (NMES)

Download PDF (103K)

E.1.2. Transcutaneous electrical nerve stimulation (TENS) compared to usual care or no treatment

Download PDF (102K)

E.2. Functional electrical stimulation compared to usual care or no treatment

E.2.1. Functional electrical stimulation compared to usual care or no treatment

Download PDF (100K)

E.3. Neuromuscular electrical stimulation (NMES) compared to placebo/sham and usual care or no treatment

E.3.1. Neuromuscular electrical stimulation (NMES) compared to placebo/sham

Download PDF (106K)

E.3.2. Neuromuscular electrical stimulation (NMES) compared to usual care or no treatment

Download PDF (112K)

E.4. Devices - tape compared to placebo/sham and usual care or no treatment

E.4.1. Devices - tape compared to placebo/sham

Download PDF (109K)

E.4.2. Devices - tape compared to usual care or no treatment

Download PDF (103K)

E.5. Devices - slings compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment

E.5.1. Devices - slings compared to neuromuscular electrical stimulation (NMES)

Download PDF (123K)

E.5.2. Devices - slings compared to usual care or no treatment

Download PDF (124K)

E.6. Devices - braces compared to usual care or no treatment

E.6.1. Devices - braces compared to usual care or no treatment

Download PDF (96K)

E.7. Acupuncture/dry needling compared to placebo/sham and usual care or no treatment

E.7.1. Acupuncture/dry needling compared to placebo/sham

Download PDF (102K)

E.7.2. Acupuncture/dry needling compared to usual care or no treatment

Download PDF (109K)

E.8. Electroacupuncture compared to placebo/sham

E.8.1. Electroacupuncture compared to placebo/sham

Download PDF (97K)

E.9. Intra-articular medicine injections - corticosteroids compared to placebo/sham

E.9.1. Intra-articular medicine injections - corticosteroids compared to placebo/sham

Download PDF (104K)

E.10. Nerve blocks (local anaesthetic) compared to transcutaneous electrical nerve stimulation (TENS) and placebo/sham

E.10.1. Nerve blocks (local anaesthetic) compared to transcutaneous electrical nerve stimulation (TENS)

Download PDF (103K)

E.10.2. Nerve blocks (local anaesthetic) compared to placebo/sham

Download PDF (101K)

Appendix F. GRADE tables

F.1. Transcutaneous electrical nerve stimulation (TENS) compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment

F.1.1. Transcutaneous electrical nerve stimulation (TENS) compared to neuromuscular electrical stimulation (NMES)

Download PDF (201K)

F.1.2. Transcutaneous electrical nerve stimulation (TENS) compared to usual care or no treatment

Download PDF (197K)

F.2. Functional electrical stimulation compared to usual care or no treatment

F.2.1. Functional electrical stimulation compared to usual care or no treatment

Download PDF (198K)

F.3. Neuromuscular electrical stimulation (NMES) compared to placebo/sham and usual care or no treatment

F.3.1. Neuromuscular electrical stimulation (NMES) compared to placebo/sham

Download PDF (197K)

F.3.2. Neuromuscular electrical stimulation (NMES) compared to usual care or no treatment

Download PDF (209K)

F.4. Devices - tape compared to placebo/sham and usual care or no treatment

F.4.1. Devices - tape compared to placebo/sham

Download PDF (201K)

F.4.2. Devices - tape compared to usual care or no treatment

Download PDF (191K)

F.5. Devices - slings compared to neuromuscular electrical stimulation (NMES) and usual care or no treatment

F.5.1. Devices - slings compared to neuromuscular electrical stimulation (NMES)

Download PDF (192K)

F.5.2. Devices - slings compared to usual care or no treatment

Download PDF (195K)

F.6. Devices - braces compared to usual care or no treatment

F.6.1. Devices - braces compared to usual care or no treatment

Download PDF (192K)

F.7. Acupuncture/dry needling compared to placebo/sham and usual care or no treatment

F.7.1. Acupuncture/dry needling compared to placebo/sham

Download PDF (197K)

F.7.2. Acupuncture/dry needling compared to usual care or no treatment

Download PDF (204K)

F.8. Electroacupuncture compared to placebo/sham

F.8.1. Electroacupuncture compared to placebo/sham

Download PDF (176K)

F.9. Intra-articular medicine injections - corticosteroids compared to placebo/sham

F.9.1. Intra-articular medicine injections - corticosteroids compared to placebo/sham

Download PDF (180K)

F.10. Nerve blocks (local anaesthetic) compared to transcutaneous electrical nerve stimulation (TENS) and placebo/sham

F.10.1. Nerve blocks (local anaesthetic) compared to transcutaneous electrical nerve stimulation (TENS)

Download PDF (179K)

F.10.2. Nerve blocks (local anaesthetic) compared to placebo/sham

Download PDF (179K)

Appendix G. Economic evidence study selection

Download PDF (182K)

Appendix H. Economic evidence tables

No health economic studies were included in this review.

Appendix I. Health economic model

New cost-effectiveness analysis was not conducted in this area.

Appendix J. Excluded studies

Clinical studies

Table 46. Studies excluded from the clinical review.

Table 46

Studies excluded from the clinical review.

Health Economic studies

Published health economic studies that met the inclusion criteria (relevant population, comparators, economic study design, published 2006 or later and not from non-OECD country or USA) but that were excluded following appraisal of applicability and methodological quality are listed below. See the health economic protocol for more details.

Table 47. Studies excluded from the health economic review.

Table 47

Studies excluded from the health economic review.

Appendix K. Research recommendations – full details

K.1. Research recommendation

What is the clinical and cost-effectiveness of diagnostic assessment to decide the choice of management for shoulder pain after stroke?

K.1.1. Why this is important

Shoulder pain is very common and disabling problem after a stroke. It can have a huge impact on a person’s health-related quality of life and ability to participate in rehabilitation. Post-stroke shoulder pain in is complex and multifactorial in aetiology, and different causes of post-stroke shoulder pain may impact the efficacy of various treatment options. A number of causes of post-stroke shoulder pain have been identified including: rotator cufftears, abnormal muscle tone, glenohumeral subluxation, impingement, tendinopathy and shoulder hand syndrome. This review has identified several treatments that may be effective in reducing post stroke shoulder pain including: taping, NMES, intra-articular corticosteroid injection and nerve blocks. However, the evidence base was limited in the amount of evidence and in linking the cause of the shoulder pain to the intervention. Some interventions may be more effective at managing certain types of shoulder pain than others.

In order to further assess the effectiveness of the interventions identified as clinically effective in the guideline, a research recommendation was made to investigate the effect of using a diagnostic assessment to assess the cause of the shoulder pain and then to use that knowledge to assess the correct treatment to use, compared to usual care. This would be useful as this would help to support the idea that people should have comprehensive assessments of the cause of shoulder pain. The trial would include an internal subgroup analysis as to which treatment was selected to treat which cause of pain to understand whether that treatment was effective for treating that cause of pain.

K.1.2. Rationale for research recommendation

Download PDF (186K)

K.1.3. Modified PICO table

Download PDF (190K)

K.2. Research recommendation

For people with different causes of shoulder pain after stroke, what is the clinical and cost-effectiveness of interventions in reducing pain?

K.2.1. Why this is important

Shoulder pain is a very common and disabling problem after a stroke. It can have a huge impact on a person’s health-related quality of life, activities of daily living and ability to participate in rehabilitation. Post-stroke shoulder pain in is complex and different causes of post-stroke shoulder pain may impact the efficacy of various treatment options. A number of causes of post stroke shoulder pain have been identified, including: rotator cuff tears, abnormal muscle tone, glenohumeral subluxation, impingement, tendinopathy, and shoulder hand syndrome. This review has identified several treatments that may be effective at reducing post-stroke shoulder pain, including: taping, NMES, intra-articular corticosteroid injection and nerve blocks. However, evidence supporting these was limited and there was no cost effectiveness evidence for the interventions. The evidence failed to identity the underlying causes of people’s shoulder pain which may have a large impact on the effectiveness of various treatments. Further research to determine which treatments are effective for different causes of shoulder pain is important to make treatment more targeted and person centred.

K.2.2. Rationale for research recommendation

Download PDF (62K)

K.2.3. Modified PICO table

Download PDF (187K)