Effectiveness of night splints for cubital tunnel syndrome

Abstract

Introduction

Cubital Tunnel Syndrome (CuTS) is a common condition of the elbow that is often treated with surgical decompression of the ulnar nerve but evidence for optimal management is unclear. A previous Cochrane review from 2016 identified very limited evidence to guide conservative management, in particular, night splints. The aim of this systematic review was to update the evidence related to the effectiveness of night splints in the treatment of CuTS.

Methods

We conducted an electronic search on January 15th 2025 of the MEDLINE, Embase, Emcare and CINAHL databases from the last 30 years, using a pre-defined protocol. Risk of bias was assessed using the RoB2 and ROBINS-I tools, with certainty assessed using GRADE.

Results

We identified only one randomised controlled trial (RCT), with high overall risk of bias, that compared night splints to a control arm of advice only. This under-powered trial with high loss to follow-up found no difference between groups. One additional RCT and three single-arm studies, all at high/serious/critical risk of bias, suggested the majority of patients with mild/moderate CuTS improve with night splinting but it is unclear whether the effect was due to treatment or time. Evidence certainty was very low.

Discussion

We identified a paucity of evidence, of low quality, regarding night splinting. The evidence identified by this review is currently insufficient to determine whether night splints should be recommended for the treatment of CuTS and there is a need for a high-quality research trial comparing night splints to a control intervention.

Keywords

Cubital tunnel syndrome ulnar neuritis splint orthosis systematic review

Introduction

Cubital Tunnel Syndrome (CuTS) is a common condition of the elbow, affecting 25 men and 19 women per 100,000 every year.¹ Surgery is commonplace when conservative treatments such as activity modification, physiotherapy or splints are not effective.² In 2023-4, according to Hospital Episodes Statistics data, 3751 cubital tunnel decompression surgeries were performed in England.³

The UK guidance from the National Institute of Health & Care Excellence (NICE) accredited ‘Commissioning guide: treatment of painful tingling fingers’ makes recommendations for the management of both CuTS and carpal tunnel syndrome.¹ The carpal tunnel syndrome recommendations were updated in 2017 but the CuTS guidance has not been updated since the original publication in 2013.¹ It recommends referral to secondary care for consideration of surgery for patients with moderate or severe symptoms and initial conservative management of patients with mild symptoms before referring for surgery. The use of night splints is advised against, though this is based upon a single under-powered low-quality randomised controlled trial (RCT), conducted by Svernlov et al in Sweden & Denmark.⁴ A subsequent Cochrane systematic review of treatments for CuTS, from 2016, only identified the same RCT.² It therefore remains unclear whether night splints should be recommended or not.

Similarly, the review found insufficient evidence to guide when to treat patients conservatively or surgically, in terms of clinical or symptom severity.² There is potential for night splints to reduce the numbers of patients undergoing surgery unnecessarily and therefore reduce waiting lists and healthcare costs. The aim of this systematic review was to update the evidence related to the effectiveness of night splints in the treatment of CuTS.

Methods

We conducted a systematic review using a protocol set a priori and registered on the PROSPERO database of systematic reviews (https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42024622986) in accordance with the PRISMA statement.⁵

Search strategys

We conducted an electronic search on January 15^th 2025 of the MEDLINE, Embase, Emcare and CINAHL databases from the last 30 years, using the search terms detailed in Table 1.

Table 1.

Search strategy.

Number	Search	Field	Limit
1	“Cubital tunnel” or (ulna* and neur) or (ulna and nerve)	Title and abstract	Jan 1995 to present
2	Splint* or orthos* or orthotic* or brace	Title and abstract	Jan 1995 to present
3	1 and 2	-	-

The search results were uploaded to the Rayyan online systematic review system (https://www.rayyan.ai/).⁶ Two reviewers (MB & HS) then independently screened titles and abstracts, whilst blinded, to select full-text reports based on the following pre-defined inclusion criteria:

• Adults (18+) diagnosed with Cubital Tunnel Syndrome (idiopathic).

• Treatment involving any form of night splint.

• Any form of comparator treatment or control including advice or physiotherapy.

• All study types (e.g. randomised controlled trials, case series, cohort studies, service evaluations).

• English language.

• Main outcomes: Patient reported outcome measures that may include and not limited to: pain, disability, nerve symptoms, quality of life, treatment satisfaction.

• Exclusions: studies of patients with cubital tunnel syndrome following trauma (e.g. fracture of the distal humerus).

The Rayyan software indicated when the process had been completed by both reviewers and would highlight any disagreements. A third reviewer (AT) was available for arbitration in the event of disagreement but in this case was not required as the two reviewers agreed.

Data extraction

One reviewer (MB) extracted data in relation to study characteristics, participant characteristics, interventions and results before a second reviewer (AT) independently verified the findings.

Risk of bias assessment

Risk of bias assessment was conducted by MB and subsequently reviewed by HS and AT. RCTs were assessed using the RoB2 tool⁷ and single arm studies were assessed using the ROBINS-I tool.⁸ The RoB2 tool rates each study in terms of high risk, unclear risk and low risk of bias within five domains (see Table 3), to reach an overall conclusion.⁷ The ROBINS-I tool rates each study in terms of low, moderate, serious and critical risk of bias within seven domains (see Table 4), to reach an overall conclusion.⁸ Any discrepancies between reviewers were then discussed and resolved.

Evidence synthesis

We had intended to perform a meta-analysis of RCTs using similar outcome measures but this was not possible due to the lack of RCTs identified by the search. Similarly, there was heterogeneity of outcome measures across the single-arm studies identified. A narrative evidence synthesis was therefore performed. Certainty of evidence from RCTs was assessed using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) process via the GRADEpro online tool (https://www.gradepro.org/).⁹ The tool assesses evidence in terms of risk of bias, inconsistency, indirectness, imprecision and treatment effect, to determine overall certainty.

Results

Initial searches identified 648 reports. After removal of duplicates and screening, six reports from five studies were included in the final analysis. One conference abstract was included in addition to the journal paper relating to the same case study, as it contained additional information (a second case study).^10,11 The study selection process is detailed in the PRISMA flow diagram (Figure 1).

Figure 1.

PRISMA flow diagram. *One study includes a journal paper report and a conference abstract report that contains additional data.

The extracted data and characteristics of the five studies are presented in Table 2.

Table 2.

Data extraction.

Study characteristics	Participant characteristics	Interventions	Results
Randomised controlled trials
Hong et al 1996¹² Randomised controlled trial (a versus a+b design) Conducted in the USA	10 patients (12 elbows) All male ‘Average’ age 59.1 (range 37-70) Symptom duration 4-18 months Symptoms moderate or severe Clinical signs mild or moderate	Group 1: elbow splint only (n=5) Group 2: elbow splint + steroid injection (n=7) Splints were thermoplastic and immobilised the elbow and wrist. Splints were worn at night for 6 months.	Mean symptom severity (0-4 scale) Baseline: 3.4; 3.3 1 month: 1.6; 1.7 6 months: 1.8; 1.1 Mean clinical sign severity (0-3 scale) Baseline: 2.0; 1.6 1 month: 1.4; 1.2 6 months: 1.8; 1.1 Ulnar nerve conduction velocity % reduction Baseline: 31.4; 27.2 1 month: -5.0; 15.0 6 months: 9.8; 3.5 Displayed as group 1; group 2
Svernlov et al 2009⁴ Randomised controlled trial Conducted in Sweden and Denmark	70 patients but only 51 included in primary analysis at 6 months 27 female, 24 male Mean age 43 (range 17-72) Mean symptom duration 10.5 months (range 3-72) Mild or moderate CuTS (Classification: Dellon 1 or 2)	3 groups all received written information regarding anatomy, probable causes of symptoms and instructions on how to avoid provocative movements, in addition to: Group 1: Reinforced neoprene splint, restricting elbow flexion beyond 45°, worn at night for 3 months Group 2: Nerve gliding exercises 2-3 times per day for 3 months Group 3: Written information only	Patients in each group achieved statistically significant improvements in Canadian Occupational Performance Measure¹³ and diurnal/nocturnal pain VAS scores at 6 months. There were no between group differences. 6 patients from the original 70 underwent surgery after failing to improve (2 from each group) 9 out of 12 patients with abnormal neurophysiology tests prior to treatment, improved.
Single-arm studies
Cancio et al 2019¹¹ & 2021¹⁰ Case studies (n=2) Conducted in the USA Data obtained from conference abstract 2019, reporting two cases, and journal paper 2021, reporting only one of the cases.	2 patients (2 elbows) Case 1: Male, 33 years. 5-year history of persistent symptoms after ulnar nerve transposition surgery Case 2: Female, 56 years. 2-month history of CuTS.	Bespoke elbow orthosis, limiting elbow flexion, worn at night, plus activity modification and nerve gliding exercises for 2-3 months.	Case 1: at 3 months, 18.18% decrease in QuickDASH score,¹⁴ 6-point decrease in pain with activity to 0/10, increase of 24 lbs in grip strength. Case 2: at 8 weeks, 15.91% decrease in QuickDASH, 4-point decrease in pain with activity to 0/10, increase of 25 lbs in grip strength
Michell & Sesath 2020¹⁵ Single arm feasibility trial Conducted in the UK	15 patients (15 elbows) 11 female, 4 male. Median age 41.6 (range 21-84) Mild or moderate CuTS	Cambridge Ulnar Splint, worn day and night for 6-8 weeks. A clasp was fastened to prevent elbow flexion at night but allow free movement during the day	Follow-up median 55 days (range 42-147) 21-point Global Rating of Change score: median +4.5 points Mondelli Ulnar Neuropathy at the Elbow Questionnaire¹⁶: Pre: median 22 points Post: median 19 points Ulnar nerve motor conduction velocity: Pre: median 44m/s Post: median 50m/s
Shah et al 2013¹⁷ Single arm trial Conducted in the USA	18 patients (24 elbows) Demographics not reported Mild (n=20) or moderate (n=4) CuTS (Classification: Dellon 1 or 2)	Rigid splint at 45° elbow flexion, worn at night for 3 months, plus activity modification advice	21/24 (88%) were treated successfully. QuickDASH¹⁴: Baseline: 29 6 weeks: 16.4 3 months: 10.9 1 year: 8.7 Final: 10.5 SF-12¹⁸ Mental: Baseline: 54.3 6 weeks: 57.6 3 months: 60.3 1 year: 61.2 Final: 61.9 SF-12¹⁸ Physical: Baseline: 45 6 weeks: 48 3 months: 52.3 1 year: 54.8 Final: 53.6 Standard Deviation not reported

We identified only one RCT, by Svernlov et al, with high overall risk of bias (see Table 3), that included a control arm. Patients were randomised to one of three treatment arms: written advice only, written advice plus nerve gliding exercises and written advice plus night splinting.⁴ Two patients in each group underwent surgery due to lack of improvement. Per protocol analysis of the remaining patients at 6 months showed that patients in all groups achieved statistically significant improvements in pain, functional performance and satisfaction but there were no statistically significant differences between the groups.

Table 3.

RoB2 Risk of bias assessment.

D1 Randomisation process.

D2 Deviations from the intended interventions.

D3 Missing outcome data.

D4 Measurement of the outcome.

D5 Selection of the reported result.

Another RCT, by Hong et al, again at high risk of bias, compared night splinting to a combination of night splinting and steroid injection.¹² Patients in both groups achieved statistically significant improvements in signs and symptoms at 1 and 6 months but there were no differences between groups. Similarly, nerve motor conduction velocity results improved in both groups at 6 months, although there were greater improvements at 1 month in the splint-only group. The authors conclude that there is no benefit in adding a steroid injection to night splinting. However, without a control group no conclusion can be drawn regarding the true effect of night splinting alone.

We identified three single arm studies. One, by Cancio et al,^10,11 was at critical risk of selection bias (see Table 4) due to being a case study and is therefore excluded from evidence synthesis, as per the ROBINS-I tool guidance.⁸ The second, by Michell & Sesanth,¹⁵ was at serious risk of bias but demonstrated subjective improvement in symptoms in 11 out of 15 patients on a global rating of change scale and statistically significant improvements from baseline in ulnar nerve conduction velocity and Mondelli questionnaire scores from using a splint for 6-8 weeks. The third, by Shah et al,¹⁷ again at serious risk of bias, investigated the combination of activity modification advice and rigid night splinting for 3 months. Three out of 24 patients with mild or moderate CuTS subsequently underwent surgery, due to lack of improvement. The remainder demonstrated statistically significant improvements in QuickDASH and SF-12 questionnaires at 3 months and 1 year follow-up.

Table 4.

ROBINS-I Risk of bias assessment.

D1 Risk of bias due to confounding.

D2 Risk of bias in selection of participants into the study (or into the analysis).

D3 Risk of bias in classification of interventions.

D4 Risk of bias due to deviations from intended interventions.

D5 Risk of bias due to missing data.

D6 Risk of bias arising from measurement of the outcome.

D7 Risk of bias in selection of the reported result.

The overall assessment of the certainty of evidence is displayed in Table 5. The evidence for the use of night splints for CuTS is very low, suggesting uncertainty whether they offer any benefit over advice alone.

Table 5.

GRADE assessment of evidence certainty.

CI: confidence interval; MD: mean difference.

Discussion

The current evidence, identified by this systematic review, is not sufficient to determine whether night splints are an effective treatment for CuTS. We identified only one RCT that compared night splinting to a control intervention (advice only).⁴ This was the same RCT referred-to in the NICE-accredited guidance and 2016 Cochrane review.^1,2 The trial was under-powered and had a high loss to follow-up, so the conclusions must be treated with caution. Of the 70 patients recruited, only 51 were included in the primary analysis. The analysis was conducted per protocol, rather than by intention-to-treat, so increases the risk of bias. There were also differences in baseline characteristics of participants, with a greater proportion of males in the splint group (57%) compared to the control group (33%). Participants in the splint group also had longer symptom duration by a mean of 4 months compared to controls. These disparities highlight the lack of power in the trial and are potential confounding factors for the effectiveness of the interventions.

The only other RCT used an A versus A + B design, where night splinting was itself the control intervention.¹² Again, this study was under-powered and was at high risk of bias due to having an unclear method of randomisation and concerns around bias of outcome assessment, whereby patient-reported outcome questionnaires were completed by the co-authors whilst interviewing the patients. The trial was designed to assess the effectiveness of steroid injection, using night splints as the control, so has limited value in answering the aim of this systematic review. It did provide some evidence that patients with mild or moderate clinical signs of CuTS do improve with night splinting over the course of 6 months. It is unclear, however, whether they would have simply improved with time without using a splint.

Similarly, the three single-arm studies provide further data to suggest that the majority of patients using night splints do improve over time. Of the 35 patients (41 elbows) included across those studies, only seven did not improve (20%).^10,11,15,17 It remains unclear whether patients would have improved without splinting and the lack of randomisation introduces a high risk of selection bias, so these findings must be treated with caution.

All of the studies included in this review selected patients with mild or moderate clinical signs of CuTS. Only two studies^12,17 reported the breakdown of patients by severity, with a total of 25 mild severity and 11 moderate severity, as defined by the Dellon classification.¹⁹ Shah et al reported results by severity, with 18/20 patients showing symptom improvement in the mild severity category and 3/4 in the moderate severity category.¹⁷ Data extracted from the Hong et al study, shows that nerve conduction velocity across the elbow improved at 6 months in 5/5 patients with mild severity and 6/7 with moderate severity.¹² Similarly, Svernlov et al, reported that prior to splinting, 5/17 had abnormal nerve conduction velocity (3 mild, 2 severe). After 6 months, this had reduced to 2/17 (2 mild).⁴ Graphical data from the Michell & Sesath suggests that 10/15 patients had improved nerve conduction velocity after 6-8 weeks of splinting.¹⁵ These data suggest that conservative management (possibly due to the effect of night splinting or natural resolution over time) may benefit those with either mild or moderate symptoms, both in terms of symptomatic and neurophysiological improvement.

Outcome assessment varied across the studies identified in this review. Many used functional measures, such as the QuickDASH, SF-12 and Canadian Occupational Performance Measure, rather than condition-specific outcomes.^13,14,18 It is unclear (beyond the scope of this review) whether these tools have been validated for CuTS and whether statistically significant changes correspond to meaningful clinical outcomes. Heterogeneity of outcome assessment is not conducive for future meta-analysis of studies, so there is a need for an agreed core outcome set for CuTS.

We also identified heterogeneity regarding splint use. The duration of recommended use ranged from 6 weeks to 6 months, with some studies using custom-made devices, some using off-the-shelf devices, some rigid and some reinforced neoprene. It is unclear, if effectiveness is determined by the type of splint and treatment duration.

This systematic review may be limited by only including English language studies from the last 30 years. There may be additional studies outside of that time period or studies reported in other languages that might provide further evidence on night splints. The evidence identified by this review is currently insufficient to determine whether night splints should be recommended for the treatment of CuTS and there is a need for a high-quality research trial comparing night splints to a control intervention.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Marcus Bateman

Heather Swaile

References

Association BO . British Society for Surgery of the Hand, Royal College of Surgeons. Commissioning Guide: Treatment of Painful Tingling Fingers: BOA, BSSH. Cape Town: RCS, 2013, p. 2013.

Caliandro

La Torre

Padua

, et al. Treatment for ulnar neuropathy at the elbow. Cochrane Database Syst Rev 2016; 11: CD006839.

NHS-England . Hospital Admitted Patient Care Activity, 2023-24: Procedures and Interventions. Leeds: NHS England, 2024.

Svernlov

Larsson

Rehn

, et al. Conservative treatment of the cubital tunnel syndrome. J Hand Surg Eur Vol 2009; 34: 201–207.

Page

Mckenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021; 372: n71.

Ouzzani

Hammady

Fedorowicz

, et al. Rayyan—a web and mobile app for systematic reviews. Syst Rev 2016; 5: 210.

Sterne

JAC

Savović

Page

, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019; 366: l4898.

Sterne

Hernán

Reeves

, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016; 355: i4919.

Guyatt

Oxman

Akl

, et al. GRADE guidelines: 1. Introduction—GRADE evidence profiles and summary of findings tables. J Clin Epidemiol 2011; 64: 383–394.

10.

Cancio

Jones

Stanley

, et al. Design and fabrication of the cubital tunnel control orthosis (CTCO). J Hand Ther 2021; 34: 500–503.

11.

Cancio

Jones

Stanley

, et al. Cubital tunnel control orthosis (CTCO) for the non-operative management of cubital tunnel syndrome. J Hand Ther 2019; 32: 566–567.

12.

Hong

Long

Kanakamedala

, et al. Splinting and local steroid injection for the treatment of ulnar neuropathy at the elbow: clinical and electrophysiological evaluation. Arch Phys Med Rehabil 1996; 77: 573–577.

13.

Law

Baptiste

McColl

, et al. The Canadian occupational performance measure: an outcome measure for occupational therapy. Can J Occup Ther 1990; 57: 82–87.

14.

Beaton

Wright

Katz

, et al. Development of the QuickDASH: comparison of three item-reduction approaches. JBJS 2005; 87: 1038–1046.

15.

Michell

Sesath

HGR

. Feasibility trial of treatment of ulnar neuropathy at the elbow using a specifically designed splint. J Clin Rheumatol 2020; 26: 37–39.

16.

Mondelli

Padua

Giannini

, et al. A self-administered questionnaire of ulnar neuropathy at the elbow. Neurol Sci 2006; 27: 402–411.

17.

Shah

Calfee

Gelberman

, et al. Outcomes of rigid night splinting and activity modification in the treatment of cubital tunnel syndrome. J Hand Surg Am 2013; 38: 1125–1130.

18.

Ware

Kosinski

Keller

. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996; 34: 220–233.

19.

Dellon

Hament

Gittelshon

. Nonoperative management of cubital tunnel syndrome: an 8-year prospective study. Neurology 1993; 43: 1673–1677.

Effectiveness of night splints for cubital tunnel syndrome - A systematic review

Abstract

Introduction

Methods

Results

Discussion

Keywords

Introduction

Methods

Search strategys

Data extraction

Risk of bias assessment

Evidence synthesis

Results

Discussion

Footnotes

Declaration of conflicting interests

Funding

ORCID iDs

References