Cortisone and Corticosteroid Injections Info Evidence
Reviewed by Dr Kieran Hirpara, Specialist Orthopaedic Surgeon Last reviewed
What it is
A cortisone injection delivers a strong anti-inflammatory medicine directly into a joint, a tendon sheath or the space around an irritated tendon. Cortisone is one of a group of medicines called corticosteroids. They are not the same as the anabolic steroids used to build muscle, and they are not the same as the small amounts of steroid hormone your body makes for itself.
In the upper limb we use these injections for trigger finger, de Quervain's tenosynovitis, carpal tunnel syndrome, arthritis at the base of the thumb, frozen shoulder, rotator cuff pain and tennis elbow — though tennis elbow is a genuine exception, for reasons set out below.
The medicine works by damping down inflammation at one specific spot. In trigger finger it helps a swollen tendon glide through its pulley again. In carpal tunnel syndrome it reduces swelling around a compressed nerve. Most people who respond notice a change within days to a couple of weeks. What an injection does not do is repair worn cartilage or heal a degenerate tendon, so it is better thought of as a way of buying comfort and function rather than as a cure.
Does it work?
It depends a great deal on the condition, and the honest answer ranges from "usually" to "probably not worth it".
For trigger finger, corticosteroid injections settle the problem in about 57% of people, which is why an injection is normally offered before surgery is discussed. For carpal tunnel syndrome, an injection gives greater improvement than tablets for up to three months, but the benefit is temporary — most people who need something definitive still come to surgery. For frozen shoulder and thumb base arthritis, injections reliably reduce pain in the short to medium term. For rotator cuff pain, an injection can relieve pain and improve function enough to let rehabilitation proceed.
Tennis elbow is the exception, and it is worth explaining properly. A landmark randomised trial followed people for a full year. At six weeks the injection looked excellent: 51 of 65 people (78%) reported success, against 16 of 60 (27%) of those who simply waited. Then it reversed. Of those early successes, 47 of 65 (72%) relapsed. By 52 weeks the injection group was doing worse than the physiotherapy group on every outcome, and worse than the people who did nothing at all on two of the three main measures. The likely explanation is that the injection removes pain so effectively that the tendon gets loaded again before it has recovered. For this condition an injection is a considered short-term decision, not a default.
Platelet-rich plasma is sometimes raised as an alternative for tennis elbow. One review found it no better than steroid in the first month but better at three and six months — while cautioning that the underlying studies carry a high risk of bias and may not justify recommending either treatment over the other. It is not a settled question.
The choice of medicine also matters. For shoulder injections, triamcinolone gives better results and fewer flare reactions than methylprednisolone. For trigger finger, betamethasone tends to wear off sooner than triamcinolone.
What are the risks?
A flare of pain is the most common effect. A significant minority of people get a sharp increase in pain in the first day or two after the injection, before any benefit appears. It settles with ice, simple pain relief and rest, but it is unsettling if you were not warned it can happen.
Changes to the skin and the fat under it occur at the injection site in a small number of people. The fat can thin, leaving a visible dent, and the skin can lose its colour, leaving a pale patch. This is more likely with repeated injections into the same superficial spot, and it is reported around the elbow in particular. It is usually painless but it can be permanent, and it is more obvious on darker skin. Where it has occurred, a course of saline injections has been used to improve the appearance.
If you have diabetes, expect your blood sugar to rise. A shoulder injection produces a significant short-term rise in blood glucose in people with type 2 diabetes not treated with insulin. It settles within a few days. Test more often than usual over that period, and tell us you are diabetic before the injection rather than after.
Repeated injections into the same place are the real concern. People with rotator cuff tears who have had frequent injections show changes in the muscle itself, with genes and proteins shifting toward fat and muscle wasting. Injecting directly into a tendon rather than around it has been followed by tendon rupture, though this is rare and reported in single cases. Against that, repeated injections for carpal tunnel syndrome have been found to be safe and did not make later surgery more difficult. There is no single number of injections that is right for everyone — how many is reasonable depends on the site, your response to the first one and what else is planned, so it is a conversation to have each time rather than an entitlement to a set number.
Serious complications are rare. Events severe enough to require hospital admission are uncommon, and appear to be driven more by a person's other medical problems than by the injection itself. Having had an injection before carpal tunnel release or rotator cuff repair has not been shown to increase deep infection after those operations. After a rotator cuff repair we wait at least a month before considering any further injection, to protect the repair.
Is it right for you?
An injection suits you if you need pain controlled now — to sleep, to work, or to get far enough into a rehabilitation programme to benefit from it — and you understand that you are buying time rather than fixing the underlying problem. It is a reasonable first step for trigger finger, de Quervain's, thumb base arthritis and frozen shoulder.
It is a poorer fit if you are looking for a durable solution to tennis elbow, if you have already had several injections into the same spot without lasting benefit, or if the problem is mechanical rather than inflammatory — a tendon that has torn, or a joint whose cartilage has gone, will not respond for long. If you have diabetes, poorly controlled blood sugar is a reason to plan the timing rather than a reason to avoid the injection altogether.
Your surgeon will weigh your symptoms, what you have already tried and what you are aiming for. This is a shared decision, and the risks above are part of it rather than a footnote to it.
The bottom line
Cortisone injections give real but temporary relief for most upper-limb tendon and joint problems, and they let rehabilitation get started. Expect the benefit within days to weeks, and expect it to fade. The main trade-offs are a possible flare in the first days, skin and fat changes with repeated injections at the same site, a short-lived rise in blood sugar if you are diabetic, and — with repeated injections — changes in the tendon and muscle themselves. Tennis elbow is the one condition where an injection may leave you worse off at a year than doing nothing, and it should be approached differently.
- Tennis elbow — why an injection is treated with caution in this condition
- Trigger finger — where injection works well, and when surgery is discussed
- Carpal tunnel and nerve compression — the temporary role of injection before surgery
- De Quervain's tenosynovitis — a condition that often settles with a single well-placed injection
- Basal thumb arthritis — where injection sits among the options
- Frozen shoulder — how injection fits with a stretching programme
- Rotator cuff disorders — injection, rehabilitation and timing around surgery
Evidence & references
This is the clinical evidence summary written for health professionals. It is technical, and it lists the research this page was built from. You do not need to read it to understand your treatment or to make a decision about it.
Overview
- Response to intra-articular corticosteroid injection was not associated with differences in 10-year patient-reported outcomes, achievement of clinically significant outcomes, or reoperation rates for patients undergoing hip arthroscopy for femoroacetabular impingement syndrome [1].
- The combination of corticosteroid and hyaluronic acid injection is more effective than corticosteroid alone in relieving pain in ankle osteoarthritis [2].
- Preoperative corticosteroid injection at all studied timepoints was not associated with an increased risk of postoperative deep infection after carpal tunnel release [3].
- There is a statistically significant difference in pain during and shortly after injection when using a steroid with lidocaine versus steroid alone for trigger finger, but that difference may not be clinically relevant [4].
- Intraoperative corticosteroid pillar injection is not an effective option for preventing pillar pain symptoms after carpal tunnel release surgery [5].
- High-dose triamcinolone injections outperformed low-dose injections across most metrics including estimated time of relief, rate of repeat injection, and rate of surgery for soft tissue pathology of the hand [6].
- Timing of a single preoperative corticosteroid injection within 1 year of rotator cuff repair did not significantly affect failure rates, patient-reported outcomes, range of motion, or strength [7].
- Both surgical procedures are indicated for painful thumb carpometacarpal osteoarthritis after steroid injections [8].
- Repeated corticosteroid injections were found to be safe in the treatment of carpal tunnel syndrome and did not affect the morbidity of subsequent release [9].
How It Works
- Response to intra-articular corticosteroid injection is not associated with differences in 10-year patient-reported outcomes, achievement of clinically significant outcomes, or reoperation rates in patients undergoing hip arthroscopy for femoroacetabular impingement syndrome [1].
- Combination of corticosteroid and hyaluronic acid injection is more effective than corticosteroid alone in relieving pain in ankle osteoarthritis [2].
- Preoperative corticosteroid injection at all studied timepoints is not associated with an increased risk of postoperative deep infection after carpal tunnel release [3].
- There is a statistically significant difference in pain during and shortly after injection when using a steroid with lidocaine versus steroid alone, but that difference may not be clinically relevant [4].
- Repeated corticosteroid injections are safe in the treatment of carpal tunnel syndrome and do not affect the morbidity of subsequent release [9].
- Risks of intra-articular hip corticosteroid injections include rapidly progressive osteoarthritis, osteonecrosis, femoral head collapse, insufficiency fracture, and worsening osteoarthritis [11].
- Both corticosteroid injections and conservative treatments are effective in treating heel spurs, with corticosteroid injections recommended as the preferred option due to higher patient satisfaction [12].
- Patients who experienced temporary improvement after local corticosteroid injection had better clinical outcomes following endoscopic plantar fascia release [15].
- Posterior approach, women, and history of preoperative corticosteroid injection are identified as the strongest risk factors for postoperative greater trochanter bursitis injection or postoperative soft tissue injection in total hip arthroplasty [18].
- A dose-dependent relationship exists between pre-operative corticosteroid injections and post-operative complications following total shoulder arthroplasty, with increasing numbers of injections correlated with higher risks of prosthetic loosening, stiffness, revision surgery, and new rotator cuff disease [19].
What the Evidence Shows
Preoperative Injections and Surgical Outcomes
- Response to preoperative intra-articular corticosteroid injection for hip arthroscopy in femoroacetabular impingement syndrome was not associated with differences in 10-year patient-reported outcomes, achievement of clinically significant outcomes, or reoperation rates [1].
- Timing of a single preoperative corticosteroid injection within 1 year prior to rotator cuff repair was not associated with increased risk of repair failure, nor did it significantly affect patient-reported outcomes, range of motion, or strength [7].
- Patients who experienced temporary improvement after local corticosteroid injection had better clinical outcomes following endoscopic plantar fascia release for chronic plantar fasciopathy [15].
Comparative Efficacy and Treatment Alternatives
- Platelet-rich plasma (PRP) has superior midterm efficacy compared to corticosteroids for improving pain and functional impairment in tendinopathy [20].
- In patients with adhesive capsulitis, suprascapular nerve blocks provide greater pain relief at 3-4, 6-7, and 12 weeks, greater improvements in shoulder function at 12 weeks, and greater active abduction at 12 weeks compared to intra-articular corticosteroid injections [21].
- Corticosteroid injections are recommended as the preferred option over conservative treatments for heel spurs due to higher patient satisfaction, although both are effective [12].
Injection Technique and Safety
- Repeated corticosteroid injections for carpal tunnel syndrome are safe and do not affect the morbidity of subsequent release [9].
- Postoperative corticosteroid injection is a viable treatment for postoperative stiffness after rotator cuff repair and may serve as a safer alternative to arthroscopic capsular release, particularly in high-risk patients such as those with osteoporosis or at risk of axillary nerve injury [16].
Risks and Adverse Events
- Risks of intra-articular hip corticosteroid injections include rapidly progressive osteoarthritis, osteonecrosis, femoral head collapse, insufficiency fracture, and worsening osteoarthritis, although the incidence rates of these outcomes vary notably [11].
Health Disparities
- Minority demographics have lower odds of receiving corticosteroid injections for the treatment of hand osteoarthritis [13].
- Minority demographics were less likely to receive a corticosteroid injection or undergo surgical repair for rotator cuff disease despite matching on medical comorbidities and smoking status [14].
Methodological Notes
- Methodological concerns regarding a trial comparing platelet-rich plasma to corticosteroid injections include the lack of ultrasound examination to assess inflammation and degeneration, the absence of Kellgren-Lawrence grade distribution data, and the potential confounding effect of lidocaine in corticosteroid preparations [10].
Practical Considerations
- Response to preoperative intra-articular corticosteroid injection is not associated with differences in 10-year patient-reported outcomes, achievement of clinically significant outcomes, or reoperation rates in patients undergoing hip arthroscopy for femoroacetabular impingement syndrome [1].
- Dual intra-articular injections of corticosteroid and hyaluronic acid are more effective than single corticosteroid injection alone in relieving pain in ankle osteoarthritis [2].
- High-dose triamcinolone injections outperform low-dose injections across most metrics, including estimated time of relief, rate of repeat injection, and rate of surgery, for soft tissue pathology of the hand [6].
- Timing of a single preoperative corticosteroid injection within 1 year of rotator cuff repair does not significantly affect failure rates, patient-reported outcomes, range of motion, or strength [7].
- Both corticosteroid injections and conservative treatments are effective in treating heel spurs, but corticosteroid injections are recommended as the preferred option due to higher patient satisfaction [12].
- Minority demographics are less likely to receive a corticosteroid injection or undergo surgical repair for rotator cuff disease despite matching on medical comorbidities and smoking status [14].
- Platelet-rich plasma injections are inferior to corticosteroid injections for short-term pain relief, suggesting that expectations regarding the clinical utility of PRP should be tempered [17].
Key Evidence
- [L2] Response to intra-articular corticosteroid injection was not associated with differences in 10-year patient-reported outcomes, achievement of clinically significant outcomes, or reoperation rates. [1] (10.1016/j.arthro.2025.07.013)
- [L1] The combination of corticosteroid and HA injection is more effective than corticosteroid alone in relieving pain in ankle OA. [2] (10.1186/s12891-025-08488-0)
- [L3] Preoperative corticosteroid injection at all studied timepoints was not associated with an increased risk of postoperative deep infection, a divergence from existing literature. [3] (10.5435/jaaos-d-25-00317)
- [L2] There is a statistically significant difference in pain during and shortly after injection when using a steroid with lidocaine versus steroid alone, but that difference may not be clinically relevant. [4] (10.1016/j.jhsa.2024.05.016)
- [L1] Intraoperative corticosteroid pillar injection is not an effective option for preventing pillar pain symptoms. [5] (10.1186/s12891-025-09393-2)
- [L4] High-dose triamcinolone injections outperformed low-dose injections across most metrics including estimated time of relief, rate of repeat injection, and rate of surgery. [6] (10.1016/j.jhsa.2025.09.014)
- [L2] Timing of a single preoperative corticosteroid injection within 1 year of rotator cuff repair did not significantly affect failure rates, patient-reported outcomes, range of motion, or strength, suggesting that one injection before repair does not strongly influence outcomes. [7] (10.1016/j.jseint.2026.101632)
- [L4] Both procedures are indicated for painful thumb carpometacarpal osteoarthritis after steroid injections. [8] (10.1016/j.jhsa.2026.01.024)
- [L3] Repeated injections were found to be safe in the treatment of carpal tunnel syndrome and did not affect the morbidity of subsequent release. [9] (10.1177/17531934251396629)
- [L5] The letter highlights methodological concerns regarding the original trial, specifically the lack of ultrasound examination to assess inflammation and degeneration, the absence of Kellgren-Lawrence grade distribution data, and the potential confounding effect of lidocaine in corticosteroid preparations. [10] (10.1016/j.arth.2025.05.006)
- [L5] Risks of intra-articular hip corticosteroid injections include rapidly progressive osteoarthritis, osteonecrosis, femoral head collapse, insufficiency fracture, and worsening osteoarthritis, although the incidence rates of these outcomes vary notably. [11] (10.1016/j.asmr.2025.101169)
- [L3] Both corticosteroid injections and conservative treatments were effective in treating heel spurs; however, corticosteroid injections are recommended as the preferred option due to higher patient satisfaction. [12] (10.1186/s12891-025-08648-2)
- [L3] Minority demographics have lower odds of receiving corticosteroid injections for the treatment of hand osteoarthritis. [13] (10.1016/j.jhsg.2025.100837)
- [L3] Minority demographics were less likely to receive a corticosteroid injection or undergo surgical repair for rotator cuff disease despite matching on medical comorbidities and smoking status. [14] (10.1016/j.jse.2026.01.015)
- [L3] Patients who experienced temporary improvement after local corticosteroid injection had better clinical outcomes following endoscopic plantar fascia release. [15] (10.1186/s12891-025-08816-4)
- [L5] Current findings suggest that corticosteroid injections may serve as a safer alternative to arthroscopic capsular release, particularly in high-risk patients such as those with osteoporosis or at risk of axillary nerve injury. [16] (10.1016/j.arthro.2025.04.021)
- [L1] Based on these findings, expectations regarding the clinical utility of PRP should be tempered. [17] (10.1016/j.arth.2025.03.013)
- [L3] Posterior approach, women, and history of preoperative corticosteroid injection were identified as the strongest risk factors for postoperative greater trochanter bursitis injection or postoperative soft tissue injection. [18] (10.1016/j.arth.2025.03.045)
- [L2] A dose-dependent relationship exists between pre-operative corticosteroid injections and post-operative complications following total shoulder arthroplasty, with increasing numbers of injections correlated with higher risks of prosthetic loosening, stiffness, revision surgery, and new rotator cuff disease. [19] (10.1016/j.jse.2026.01.024)
- [L1] PRP can effectively improve pain and functional impairment in patients with tendinopathy, and its midterm efficacy is superior to that of corticosteroids. [20] (10.1186/s12891-025-08566-3)
- [L1] In patients with adhesive capsulitis, suprascapular nerve blocks provide greater pain relief at 3-4, 6-7, and 12 weeks, greater improvements in shoulder function at 12 weeks, and greater active abduction at 12 weeks, compared to intra-articular corticosteroid injections. [21] (10.1016/j.jse.2025.05.037)
References
[1] No Difference in Responders and Nonresponders to Preoperative Intra-articular Corticosteroid Injection Undergoing Hip Arthroscopy for Femoroacetabular Impingement Syndrome at 10 Years: A Matched Analysis. Arthroscopy: The Journal of Arthroscopic & Related Surgery. 2025. DOI: 10.1016/j.arthro.2025.07.013
[2] Dual intra-articular injections of corticosteroid and hyaluronic acid versus single corticosteroid injection for ankle osteoarthritis: a randomized comparative trial. BMC Musculoskeletal Disorders. 2025. DOI: 10.1186/s12891-025-08488-0
[3] Ipsilateral Preoperative Corticosteroid Injection and Timing Not Associated With Postoperative Deep Infection After Carpal Tunnel Release. Journal of the American Academy of Orthopaedic Surgeons. 2025. DOI: 10.5435/jaaos-d-25-00317
[4] Corticosteroid Injection With and Without Local Anesthetic for the Treatment of Trigger Finger: A Randomized Clinical Trial. The Journal of Hand Surgery. 2025. DOI: 10.1016/j.jhsa.2024.05.016
[5] “Intraoperative pillar corticosteroid injection”: does it improve clinical outcomes after carpal tunnel release surgery? A double-blind, randomized controlled study. BMC Musculoskeletal Disorders. 2025. DOI: 10.1186/s12891-025-09393-2
[6] Efficacy of Low-Dose Versus High-Dose Corticosteroid Injections for Soft Tissue Pathology of the Hand. The Journal of Hand Surgery. 2025. DOI: 10.1016/j.jhsa.2025.09.014
[7] Timing of corticosteroid injection within 1 year prior to rotator cuff repair was not associated with increased risk of repair failure. JSES International. 2026. DOI: 10.1016/j.jseint.2026.101632
[8] Do Preoperative Intra-Articular Steroid Injections Affect the Choice of Surgical Procedures for Thumb Carpometacarpal Osteoarthritis?. The Journal of Hand Surgery. 2026. DOI: 10.1016/j.jhsa.2026.01.024
[9] The safety and cost of repeated corticosteroid injections for carpal tunnel syndrome. Journal of Hand Surgery (European Volume). 2025. DOI: 10.1177/17531934251396629
[10] Letter Regarding "Platelet-Rich Plasma Injections Are Inferior to Corticosteroid Injections for Short-Term Pain Relief: A Prospective, Double-Blinded, Randomized Controlled Trial". The Journal of Arthroplasty. 2025. DOI: 10.1016/j.arth.2025.05.006
[11] Risks of Intra‐articular Hip Corticosteroid Injections Include Rapidly Progressive Osteoarthritis and Femoral Head Collapse in Patients With and Without Pre‐existing Osteoarthritis: A Systematic Review. Arthroscopy, Sports Medicine, and Rehabilitation. 2025. DOI: 10.1016/j.asmr.2025.101169
[12] Comparison of corticosteroid injections and conservative treatments for heel spurs. BMC Musculoskeletal Disorders. 2025. DOI: 10.1186/s12891-025-08648-2
[13] Underutilization of Hand Corticosteroid Injections and Arthroplasty for Minority Demographics. Journal of Hand Surgery Global Online. 2025. DOI: 10.1016/j.jhsg.2025.100837
[14] Minority groups are less likely to undergo surgical fixation or receive a corticosteroid injection for rotator cuff disease: a large database study. Journal of Shoulder and Elbow Surgery. 2026. DOI: 10.1016/j.jse.2026.01.015
[15] Does local steroid injection have a prognostic value for endoscopic plantar fascia release in chronic plantar fasciopathy?. BMC Musculoskeletal Disorders. 2025. DOI: 10.1186/s12891-025-08816-4
[16] Editorial Commentary:
Postoperative Corticosteroid Injection Is a Viable Treatment for Postoperative Stiffness After Rotator Cuff Repair. *Arthroscopy*. 2025. DOI: 10.1016/j.arthro.2025.04.021
[17] Platelet-Rich Plasma Injections Are Inferior to Corticosteroid Injections for Short-Term Pain Relief: A Prospective, Double-Blinded, Randomized Controlled Trial. The Journal of Arthroplasty. 2025. DOI: 10.1016/j.arth.2025.03.013
[18] Does Surgical Approach in Total Hip Arthroplasty Affect Postoperative Corticosteroid Injection Requirements?. The Journal of Arthroplasty. 2025. DOI: 10.1016/j.arth.2025.03.045
[19] Pre-operative corticosteroid injections are associated with a dose-dependent risk for complications following anatomic and reverse total shoulderarthroplasty. Journal of Shoulder and Elbow Surgery. 2026. DOI: 10.1016/j.jse.2026.01.024
[20] Platelet-rich plasma and corticosteroid injection for tendinopathy: a systematic review and meta-analysis. BMC Musculoskeletal Disorders. 2025. DOI: 10.1186/s12891-025-08566-3
[21] Intra-articular corticosteroid injection vs. suprascapular nerve block for adhesive capsulitis: a systematic review and meta-analysis of level I randomized controlled trials. Journal of Shoulder and Elbow Surgery. 2026. DOI: 10.1016/j.jse.2025.05.037
