CADTH is committed to supporting Canada’s health care decision-makers through this challenging and uncertain time.
For evidence, tools, and resources related to COVID-19, visit our COVID-19 Evidence Portal.


Begin main content

Orthotic Bracing or Splinting of Upper Extremities in Patients with Chronic, Non-Cancer Pain: A Review of Clinical Effectiveness

Last updated: March 2, 2020
Project Number: RC1252-000
Product Line: Rapid Response
Research Type: Devices and Systems
Report Type: Summary with Critical Appraisal
Result type: Report


  1.   What is the clinical effectiveness of orthotic bracing and splinting of the upper extremities in patients with chronic, non-cancer pain?

Key Message

​Evidence of limited quality from the two included systematic reviews suggested that compared to usual care or no intervention, splint use significantly reduced pain and improved functional performance in patients with osteoarthritis involving the thumb base (rhizarthrosis), carpometacarpal joints, or the distal interphalangeal joint in the long-term (13 to 52 weeks) but not in the short term.

Evidence of limited quality from four prospective, uncontrolled, before-and-after studies suggested that compared to baseline, the use of splints for four to six weeks resulted in statistically significant reduction in pain, improvement in functional performance, as well as range in motion, pinch strength, and hand strength in in patients with osteoarthritis involving the distal interphalangeal joint, and trapeziometacarpal joints, or those diagnosed with tennis elbow or carpal tunnel syndrome.

Evidence from the two included randomized controlled trials (RCTs) indicated that a single local injection of 20 mg methylprednisolone acetate was statistically significantly more effective than night splitting for reducing pain intensity and improved figure dexterity in patients with carpal tunnel syndrome after four to six weeks of treatment. However, the findings from these RCTs on functional performance were inconclusive as one reported a statistically significantly greater improvement in favor of the steroid injection, whereas the other did not find a significant between group difference.

Sources of uncertainty in the current report included a lack of definition of ‘‘usual care,’’ and a significant overlap of primary studies, all of which had high risk of bias between the two included systematic reviews. Also, the comparator data used in meta-analysis were from usual care or no intervention without reporting separate results for either of them compared to splint. Given that lack of treatment might not result in improved outcomes, it is unclear if splints might perform equally well in comparison with usual care alone as they did in analyses that considered data from usual care or no treatment together as comparator group.

Fundamental limitations of the primary studies included in this report were the open-label design of all which disposes them to biases and the fact that four of them were prospective, uncontrolled, before-and-after studies with inherently higher likelihood of systemic biases due to lack of the risk-diminishing property of randomization. Also, the before-and-after studies investigated different kinds of splints, with variations in design and materials used in construction, custom-made devices, and splints fabricated by patients for self-use. Thus, it was unclear if the finding of these studies could be replicated using generic splints.

There were no studies identified that compared splints to pharmacological interventions for pain relief such as non-steroidal anti-inflammatory drugs, acetaminophen, topical capsaicin, and topical salicylates, as well as less conventional medications such as pregabalin and duloxetine. Furthermore, none of the studies included in this report provided outcomes on health-related quality of life, disability level, or global impression of recovery after treatment with any of the studied interventions.