The four main treatments given as knee injections are cortisone, hyaluronic acid, PRP and stem cells [7]. In knee osteoarthritis a cortisone injection can reduce pain for a short time [2]. The place of injections depends on the stage of the disease: European orthopaedic consensus statements consider PRP appropriate in the mild and moderate stages (grades 1–3) and not appropriate in end-stage (grade 4) knee osteoarthritis [8][9]. Hyaluronic acid has not shown a clinically meaningful benefit in large placebo-controlled trials, and for stem cells the certainty of the evidence is low; the American College of Rheumatology (ACR) guideline, without distinguishing stages, does not recommend hyaluronic acid, PRP or stem cells [3][6][7].
What is a knee injection and what kinds are there?
A knee injection, or joint injection, means putting a medicine, or a product prepared from blood or cells, into the joint through a needle [1][4][6]. It comes up most often in knee osteoarthritis, and the research below is also about knee osteoarthritis [2][3][5][6]. People ask about four kinds of injection:
- Cortisone injection (corticosteroid): an anti-inflammatory medicine [1]
- Hyaluronic acid injection: popularly called a "gel injection"; its medical name is viscosupplementation [3]
- PRP (platelet-rich plasma): prepared from the person's own blood [4]
- Stem cell injection: cells that can develop into cartilage, bone and fat tissue cells are put into the joint [6]
What does a cortisone injection do, and when does it start to work?
Cortisone injections are used for joint and soft tissue problems such as osteoarthritis, frozen shoulder, bursitis, trigger finger, carpal tunnel syndrome and plantar fasciitis [1]. They usually start to work within a few days, sometimes within hours; pain may get worse in the first 24 hours and is expected to ease over the next 2 to 3 days [1].
According to a Cochrane review of 27 trials (1767 patients) in knee osteoarthritis, a cortisone injection appears to reduce pain compared with a sham (placebo) injection; the average difference is 1 centimetre on a 10-centimetre pain scale [2]. The benefit is moderate in the first 1 to 2 weeks, small to moderate at 4 to 6 weeks and small at 13 weeks; there is no evidence that an effect remains at 26 weeks [2]. The review rated the quality of the evidence as low and reported that it remains unclear whether there is a clinically important benefit [2].
How far apart are cortisone injections given, and what are the side effects?
According to an NHS patient leaflet, there is usually a wait of at least 6 weeks between two injections, and more than 3 injections in the same area within 12 months are not usually recommended [1]. After the injection it is recommended to rest the area that day and the day after, and to restrict heavier activity and exercise for 2 weeks [1].
Before the injection the doctor should be told about diabetes, the use of blood thinners, an infection, pregnancy and any steroid injection given in the last few weeks [1].
The effect on cartilage is debated: the ACR guideline notes that one study raised the possibility that specific steroid preparations, or injections given at a certain frequency, may contribute to cartilage loss, but that the panel was uncertain of the clinical significance of this because the change in cartilage was not associated with worsening of pain or function [7]. In the Cochrane review no difference in side effects was shown between cortisone and sham injection [2].
The side effects listed in the leaflet are [1]:
- A flare of pain around the injection site: it can start within the first 24 hours and last up to 5 to 7 days
- Flushing of the face (redness and heat)
- Raised blood sugar for a few days in people with diabetes
- Paler skin at the injection site, or dimpling caused by loss of fat
- Infection: rare, but it can happen
- Damage to nearby tendons: a small risk
Hyaluronic acid injection ("gel injection")
This is the injection of hyaluronic acid or its derivatives into the joint; it has been used in knee osteoarthritis for more than 50 years. It is given as a single injection or as a course of two to six injections [3].
The main result of a review published in the BMJ in 2022 rests on 24 large, placebo-controlled trials (8997 patients): compared with placebo, hyaluronic acid reduced pain by an average of 2 millimetres on a 100-millimetre scale, which is below the smallest difference considered clinically meaningful [3].
Serious side effects, such as events requiring admission to hospital, occurred in 3.7% of those given hyaluronic acid and 2.5% of those given placebo [3]. According to the authors, the findings do not support broad use of this injection; even so, the possibility that some selected groups of patients could benefit cannot be excluded [3].
What is PRP treatment for the knee, and how many sessions?
For PRP, blood is taken from the person; the blood is spun in a centrifuge so that the platelets are separated and concentrated, and they are then injected into the affected area. Platelets contain growth factors that are important in healing [4].
The evidence is conflicting. The American Academy of Orthopaedic Surgeons (AAOS) states that more and more publications show PRP to be effective in low- to moderate-grade knee osteoarthritis, but that research to back up the claims made in the media is lacking [4]. In the placebo-controlled RESTORE trial (288 patients with mild to moderate osteoarthritis), however, PRP was not significantly different from a saline injection for pain or cartilage volume at 12 months [5].
Stage is decisive here. According to the 2024 consensus of the European Society of Sports Traumatology, Knee Surgery and Arthroscopy (ESSKA), clinical evidence shows that PRP is effective in mild to moderate knee osteoarthritis (Kellgren-Lawrence grade 3 or lower); the consensus regards PRP as a valid option mainly for grades 1–3 [8]. In a randomised trial of 162 patients at different stages, three PRP injections gave better results at six months in early-stage disease, whereas in advanced disease (grade 4) giving more injections did not significantly improve the results [10].
These sources do not give an agreed schedule for the number of sessions: in the RESTORE trial three injections were given one week apart [5]; the ACR points out that there is no standard for how PRP is prepared or for the techniques used, which makes it difficult to identify exactly what is being injected [7].
Pain at the injection site may increase for the first week or two, and it may be several weeks before a benefit is felt; according to the AAOS, the incidence of problems such as infection, tissue damage and nerve injury appears to be no different from that with cortisone injections [4].
How is stem cell treatment given to the knee, and what does the evidence say?
In theory, stem cells are thought to be able to regrow damaged cartilage, which is why they are used in osteoarthritis [6]. The cells are injected into the knee joint; their source, the method of preparation and the dose vary from study to study, and the best method and dose are not known [6].
A Cochrane review dated 2025 examined 25 trials (1341 patients), most of them small. Compared with a placebo injection, a stem cell injection may slightly improve pain and function for up to six months: on a 10-point scale, pain was 4.5 points with placebo and 3.3 points with stem cells; the certainty of this evidence is low [6].
Whether it regrows cartilage is not known; progression on X-ray was not assessed in any trial [6]. Serious side effects were reported infrequently, but the possible risks of a treatment with the capacity to promote cell growth, and of cells taken from another person, are not known [6].
What do guidelines say? The place of injections by stage
Guidelines and consensus statements do not all say the same thing; some of them give their recommendation according to the stage of the disease. Knee osteoarthritis is staged on X-ray from 0 to 4 with the Kellgren-Lawrence classification; grade 4 is the advanced, that is the end, stage [5][9][10]. The joint consensus of ESSKA and the International Cartilage Regeneration and Joint Preservation Society (ICRS) considers PRP appropriate in patients aged 80 or younger, at grades 0–3, in whom non-surgical treatment or other injection treatments have not given an adequate result; it does not consider it appropriate as a first treatment or in end-stage disease (grade 4) [9]. In end-stage knee osteoarthritis, when pain cannot be controlled with other treatments, the treatment that comes up is knee replacement [6]; it is described in a separate article in this guide.
The ACR guideline (2019) does not distinguish stages. For knee and hip osteoarthritis it strongly recommends exercise and, for people who are overweight, weight loss; in England, NICE does not recommend hyaluronic acid injections, and the BMJ review also reports that most guidelines do not support this injection [3][7]. The ACR's recommendations on injections are as follows [7]:
- Cortisone injection: strongly recommended for knee and hip osteoarthritis and conditionally recommended for hand osteoarthritis; ultrasound guidance is strongly recommended for injections into the hip
- Hyaluronic acid: conditionally recommended against in the knee and the joint at the base of the thumb, and strongly recommended against in the hip; the conditional recommendation does not rule out its use, after discussion with the patient, when other options have proved inadequate
- PRP and stem cells: strongly recommended against in knee and hip osteoarthritis
- If an injection is to be given, cortisone is conditionally preferred to the other injections because the quality of the evidence for it is considerably higher
When not to wait
Every joint injection carries a risk, although small, of joint infection (septic arthritis) [6]. After an injection, contact the doctor who gave it or a healthcare facility without waiting in the following situations [1]:
- A significant increase in pain
- Redness or heat around the injection site
- A temperature above 38 degrees
- Feeling very unwell
Frequently asked questions
- How often can knee injections be given?
- With cortisone injections there is usually a wait of at least 6 weeks between two injections, and more than 3 injections in the same area in 12 months are not usually recommended [1]. Hyaluronic acid is given as a single injection or as a course of two to six injections [3]; for PRP there is no standard for preparation or technique [7].
- Are cortisone injections harmful?
- A flare of pain at the injection site lasting a few days, flushing of the face and, in people with diabetes, a temporary rise in blood sugar can occur; infection is rare [1]. It has been suggested that injections given at a certain frequency may contribute to cartilage loss, but the clinical significance of this is uncertain [7].
- What does a hyaluronic acid knee injection do?
- In large placebo-controlled trials it reduced pain only very slightly compared with placebo, below the level considered clinically meaningful, while serious side effects were more frequent than with placebo [3]. NICE does not recommend this injection [3]; the ACR also conditionally recommends against it in the knee [7].
- Does PRP help the knees?
- It depends on the stage, and the evidence is conflicting. The ESSKA consensus statements consider PRP appropriate in the mild and moderate stages (grades 1–3) and not appropriate in end-stage disease (grade 4) [8][9]. In the placebo-controlled RESTORE trial PRP was not better than a saline injection at 12 months [5]; the ACR guideline, without distinguishing stages, also does not recommend PRP [7].
- Does stem cell treatment help the knees?
- According to low-certainty evidence it may slightly improve pain and function compared with a placebo injection; whether it regrows cartilage, and how safe it is, are not known [6]. The ACR guideline does not recommend stem cell injections for knee and hip osteoarthritis [7].
This article is a general summary; diagnosis and treatment decisions are made by the doctor who examines you.
Sources
- Steroid Injections: Integrated Musculoskeletal (iMSK) Service patient information (PI091). North Tees and Hartlepool NHS Foundation Trust, n.d.. nth.nhs.uk
- Intra-articular corticosteroid for knee osteoarthritis. Cochrane Database of Systematic Reviews (Jüni P, et al.), 2015. pubmed.ncbi.nlm.nih.gov
- Viscosupplementation for knee osteoarthritis: systematic review and meta-analysis. BMJ (Pereira TV, et al.), 2022. pubmed.ncbi.nlm.nih.gov
- Platelet-Rich Plasma (PRP). American Academy of Orthopaedic Surgeons (OrthoInfo), n.d.. orthoinfo.org
- Effect of Intra-articular Platelet-Rich Plasma vs Placebo Injection on Pain and Medial Tibial Cartilage Volume in Patients With Knee Osteoarthritis: The RESTORE Randomized Clinical Trial. JAMA (Bennell KL, et al.), 2021. pubmed.ncbi.nlm.nih.gov
- Stem cell injections for osteoarthritis of the knee. Cochrane Database of Systematic Reviews (Whittle SL, et al.), 2025. pubmed.ncbi.nlm.nih.gov
- 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. Arthritis & Rheumatology (Kolasinski SL, et al.), 2020. pubmed.ncbi.nlm.nih.gov
- The use of injectable orthobiologics for knee osteoarthritis: A European ESSKA-ORBIT consensus. Part 1—Blood-derived products (platelet-rich plasma). Knee Surgery, Sports Traumatology, Arthroscopy (Laver L, et al.), 2024. pubmed.ncbi.nlm.nih.gov
- Platelet-rich plasma injections for the management of knee osteoarthritis: The ESSKA-ICRS consensus. Recommendations using the RAND/UCLA appropriateness method for different clinical scenarios. Knee Surgery, Sports Traumatology, Arthroscopy (Kon E, et al.), 2024. pubmed.ncbi.nlm.nih.gov
- Multiple PRP injections are more effective than single injections and hyaluronic acid in knees with early osteoarthritis: a randomized, double-blind, placebo-controlled trial. Knee Surgery, Sports Traumatology, Arthroscopy (Görmeli G, et al.), 2017. pubmed.ncbi.nlm.nih.gov
This article is general information and does not replace an examination or imaging. If you have symptoms, see a doctor.
More in this area
- Anterior cruciate ligament tear: is surgery always needed?
- Baker's cyst: what causes a swelling behind the knee and how does it settle?
- Cartilage damage in the knee: does cartilage heal, and how is it treated?
- Knee osteoarthritis: exercise, weight, injections and replacement, what and when?
- Knee pain: what causes it, what helps and who should you see?
- Knee replacement surgery: who is it for and what is recovery like?
- Meniscus tear: exercise or arthroscopy?
- Kneecap pain and chondromalacia patella: what causes it and how does it settle?