Efficacy of two doses of intra-articular ozone therapy for pain and functional mobility in knee osteoarthritis: a double-blind randomized trial

Abstract

Aim

This double-blind, trial sought to assess the effectiveness of intra-articular ozone therapy at concentrations of 20 µg/mL and 40 µg/mL in managing pain and enhancing functional mobility in patients with knee osteoarthritis (KOA).

Method

This parallel, three-arm randomized controlled trial, conducted between 2022 and 2023, included 59 knee osteoarthritis (KOA) patients randomly allocated to one of three groups: the first group received 40 µg/mL ozone therapy, the second group received 20 µg/mL ozone therapy, and the control group received oxygen. Functional mobility was assessed through the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), knee flexion range of motion (FROM), the Timed Up and Go (TUG) test, and the six-minute walk test (6MWT). Pain intensity was measured using the Visual Analog Scale (VAS) and the pain subscale of the WOMAC. Intra-articular injections were administered weekly for four consecutive weeks, with assessments conducted pre-treatment, and at two weeks, one month, and two months post-intervention. One-way ANOVA was employed for normally distributed quantitative data, while the Kruskal-Wallis test was utilized for non-normally distributed data. Qualitative variables were analyzed using the Chi-squared or Fisher’s exact test, as appropriate.

Results

The groups receiving intra-articular ozone therapy exhibited notable reductions in mean VAS scores and improvement in functional mobility variables when compared to the control group (p < 0.05). However, post-hoc analysis indicated no statistically significant differences between the 40 µg/mL and 20 µg/mL ozone therapy groups regarding these parameters (VAS, FROM, TUG, 6MWT, or WOMAC scores) (p > 0.05).

Conclusion

Both 20 µg/mL and 40 µg/mL doses of intra-articular ozone therapy prove to be effective in reducing pain and enhancing functional mobility in patients with knee osteoarthritis (KOA). Nevertheless, there was no significant difference in the efficacy between the two ozone concentrations.

Trial registration

The trial is registered on us ClinicalTrials.gov in 2024-05-01 with the following ID code NCT06088706.

Introduction

Knee osteoarthritis (KOA) is a progressively degenerative joint disease characterized by cartilage breakdown, synovial inflammation, and subchondral bone remodeling, affecting over 346 million people globally [1, 2]. By 2019, the disease burden rose to 138.2 per 100,000 people—a 7.8% increase since 1990—with chronic pain, stiffness, and functional impairment severely reducing quality of life [1, 3]. While aging, obesity, and mechanical stress drive pathogenesis [4, 5], emerging evidence highlights the central role of inflammatory mediators in disease progression. Synovial macrophages and chondrocytes release pro-inflammatory cytokines such as interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), and IL-6, which perpetuate cartilage degradation by upregulating matrix metalloproteinases (MMPs) and aggrecanases (e.g., ADAMTS-5) [6, 7]. These enzymes degrade type II collagen and proteoglycans, destabilizing the extracellular matrix [8]. Additionally, reactive oxygen species (ROS) generated by oxidative stress further amplify inflammation, inhibit chondrocyte repair, and accelerate apoptosis [9]. The synovium itself becomes hypertrophic and infiltrated with immune cells, contributing to pain and joint dysfunction [10]. This inflammatory cascade not only exacerbates structural damage but also correlates with clinical severity, highlighting the need for therapies targeting these pathways [11].

Current treatments, including analgesics, physical therapy, and surgery, often provide incomplete relief or carry risks of adverse effects [12, 13]. Ozone therapy has emerged as a promising alternative, targeting KOA’s inflammatory and oxidative pathways. Its therapeutic effects are mediated through multiple anti-inflammatory and antioxidant mechanisms. Ozone (O₃) downregulates NF-κB signaling, a master regulator of pro-inflammatory cytokines like IL-1β and TNF-α, thereby reducing synovitis and cartilage catabolism [14, 15]. It also stimulates antioxidant enzymes (e.g., superoxide dismutase, glutathione peroxidase) via controlled oxidative stress, neutralizing ROS and restoring redox balance [16]. By enhancing oxygen delivery and activating growth factors (e.g., TGF-β), ozone promotes chondrocyte proliferation and extracellular matrix synthesis [17]. O₃ induce analgesic effect though inhibition of nociceptive signaling by modulating transient receptor potential (TRP) channels and suppressing prostaglandin E2 (PGE2) production [18]. Clinical studies support these mechanisms. For example, intra-articular ozone injections significantly reduced synovial fluid IL-1β and TNF-α levels in KOA patients compared to placebo (p < 0.01) [19], while also lowering oxidative stress markers like malondialdehyde (MDA) [20]. A 2024 umbrella review of systematic reviews confirmed ozone’s superiority over placebo in pain reduction and functional improvement [21].

Clinical trials demonstrate ozone’s superiority over placebo, with intra-articular injections (20–30 µg/mL) significantly improving pain and WOMAC scores over six months [22, 23]. However, existing studies lack consensus on optimal dosing, utilizing concentrations from 10 µg/mL to 40 µg/mL without direct comparisons [24, 25].

While prior research establishes ozone’s efficacy over hyaluronic acid or steroids [26, 27], no study has systematically evaluated dose-dependent effects. This gap hinders clinical translation, as higher doses may risk oxidative tissue damage without enhancing efficacy. Our double-blind randomized trial is the first to compare 20 µg/mL and 40 µg/mL intra-articular ozone in KOA patients, assessing pain, functional mobility, and safety over two months. By clarifying dose-response relationships, this work informs standardized protocols to optimize therapeutic outcomes and minimize risks.

Method

Trial design

This double-blind, parallel, three-arm randomized controlled trial (RCT) was conducted on KOA patients at Hazrat Rasool Akram Hospital in Tehran, Iran, from June 2022 to December 2023. The study adhered to the CONSORT guidelines for reporting randomized controlled trials [28].

Sample size

A priori power analysis (G*Power v3.1) with 80% power, α = 0.05, and an expected 2-point VAS difference between groups [29] yielded a sample size of 52. To account for a 20% dropout rate, 60 participants were enrolled (20 per group).

Participants

Out of the 72 individuals who participated in the study, 12 did not meet the eligibility criteria.

Inclusion criteria

• 1. Aged 50–75 years.

• 2. Diagnosis of primary knee osteoarthritis (KOA):

  • Clinical: Knee pain plus ≥ 3 American College of Rheumatology (ACR) criteria [30]:

    • Age > 50 years.

    • Morning stiffness ≤ 30 min.

    • Crepitus on active motion.

    • Bony tenderness or enlargement.

    • No palpable warmth (excluding inflammatory arthritis).

  • Radiographic: Kellgren-Lawrence (K-L) grade 2 or 3 on weight-bearing radiographs (posteroanterior, lateral semi-flexed, patellar views) [31].

• 3. Chronic knee pain (≥ 6 months) with baseline Visual Analog Scale (VAS) score ≥ 4 during activity (e.g., stair climbing, prolonged sitting).

• 4. Ability to walk independently ≥ 30 m.

Exclusion criteria

• 1. Intra-articular knee injections within the past 6 months.

• 2. Neuromuscular diseases, acute traumatic knee injury, or lower limb surgery/fractures (past year).

• 3. Bone implants, malignancy, BMI ≥ 35 kg/m², or unstable mental status.

• 4. Participation in physiotherapy/exercise programs within the past 3 months.

For patients with bilateral involvement, the more symptomatic knee (higher VAS score) was selected for intervention.

Randomization

Participants were allocated to the two interventions and control groups via permuted block randomization (block size = 6) using computer-generated sequences. Sequentially numbered, opaque, sealed envelopes ensured allocation concealment. To ensure allocation concealment, a nurse not involved in the study assigned participants to their groups using sealed envelopes. Double-blinding was maintained, ensuring that neither the participants nor the clinicians were aware of the treatment group assignments.

Blinding

A third-party clinician prepared identical 10 mL opaque syringes labeled with unique, randomized codes (e.g., “A1,” “B2”) generated using a computer-based randomization system (block randomization, 1:1:1 ratio). Ozone (20 µg/mL or 40 µg/mL) or medical-grade oxygen (control) was administered. Clinicians, blinded to group assignments, performed injections (21-gauge needle; lateral approach) and the control group received intra-articular oxygen injections (same volume as ozone groups: 10 mL) to ensure identical procedural experiences. A blinded sports medicine specialist conducted evaluations. The randomization code was disclosed to statisticians only after database lock and completion of the final analysis.

Interventions

Sixty eligible patients were included in the study and randomly divided into three groups. The first group received ozone therapy at a concentration of 40 µg/mL, the second group received 20 µg/mL ozone, and the third group, serving as the control, received oxygen. Initially, all participants underwent a diagnostic examination to determine the presence and severity of knee KOA. To confirm the diagnosis of knee osteoarthritis (KOA), weight-bearing, bilateral radiographs of the knees were taken in Posterior Anterior Fixed-Flexion, Lateral Weight-Bearing Semi-flex, and Patellar View positions. The severity of KOA was evaluated using the Kellgren-Lawrence criteria. For those meeting the inclusion criteria, data were collected through questionnaires and clinical assessments. Medical-grade ozone was produced using a Medical ozone generator OZONETTE (Ozonette sedecal, Spain) with oxygen source gas.

Intra-articular injections using a 21-gauge needle were administered under aseptic conditions, using the lateral approach technique, with patients positioned supine on the examination table and their knees flexed at approximately 45 degrees. A 10 ml syringe was used to injection and aspiration confirmed intra-articular placement before slow injection (15–20 s). Intra-articular injections were administered once a week for four consecutive weeks. Assessments were conducted before the intervention and at two weeks, one month, and two months post-intervention, all performed by a sports medicine specialist.

Outcomes

Outcomes were assessed at baseline, 2 weeks, 1 month, and 2 months after the final injection.

Pain intensity

• Visual Analog Scale (VAS): Pain intensity was assessed using a 0–10 scale (0 = no pain; 10 = worst pain imaginable) [32].

• WOMAC Pain Subscale: A validated 9-item questionnaire evaluating pain during daily activities (e.g., stair climbing, sitting) [33].

Functional mobility

• WOMAC Physical Function Subscale: A 17-item assessment of limitations in daily activities [34].

• Knee Flexion Range of Motion (FROM): Measured in degrees using a goniometer, following American Academy of Orthopedic Surgeons guidelines [35].

• Timed Up and Go (TUG) Test: Time (seconds) required to rise from a chair, walk 3 m, return, and sit [36].

• Six-Minute Walk Test (6MWT): Distance (meters) walked in 6 min, per American Thoracic Society protocols [37].

Statistical analysis

Data were analyzed using SPSS v27.0 (IBM Corp., USA). The Shapiro-Wilk test was used to assess the normality of data. Mixed-model ANOVA compared outcomes across groups and timepoints, with Tukey’s post-hoc test for pairwise comparisons. Significance was set at p < 0.05.

Ethical considerations

All participants provided written informed consent. Standard care (exercise therapy, NSAIDs as needed) was maintained across groups. Participants could withdraw at any time without penalty. Ethical approval was obtained from the Ethics Committee of Iran University of Medical Sciences (IR.IUMS.FMD.REC.1401.412). The trial is registered in 2024-05-01on ClinicalTrials.gov with the following ID code NCT06088706.

Result

The CONSORT diagram in Fig. 1 depicts the flow of participants through each stage of the trial. Initially, 72 patients were screened, with 60 meeting the inclusion criteria. During the intervention sessions, A total of 59 patients participated in the study, including 20 patients in the 40 µg/mL group (due to Patients were randomly allocated into treatment groups using block randomization with blocks of six), 19 patients in the 20 µg/mL group, and 20 patients in the oxygen therapy group.

Fig. 1

CONSORT flow diagram of recruitment and allocation of the participants

Table 1 summarizes the baseline characteristics of the study population, indicating no statistically significant differences between the groups in terms of age, BMI, osteoarthritis duration, gender distribution, affected knee, or K-L radiological stage at baseline (p-value > 0.05).

Table 2 provides a comparative analysis of VAS, FROM, TUG, 6MWT, and WOMAC subscale and total scores among the study groups. Both ozone therapy groups demonstrated significant reductions in mean VAS, TUG, WOMAC subscale, and total scores compared to the control group throughout the study period. Additionally, the mean FROM score and 6MWT distance significantly increased in both ozone therapy groups compared to the control group (p-value < 0.05). Post-hoc analysis revealed no significant differences between the 20 µg/mL and 40 µg/mL ozone therapy groups for VAS, FROM, TUG, 6MWT, and WOMAC subscale and total scores (p-value > 0.05). However, these differences were significant when comparing each ozone therapy group to the control group.

Table 1 Comparison of baseline characteristics among study groups

Table 2 Comparative analysis of VAS, FROM, 6MWT, TUG, and WOMAC scores among study groups

Discussion

This double-blind randomized trial demonstrates that intra-articular ozone therapy at both 20 µg/mL and 40 µg/mL significantly improves pain and functional mobility in patients with knee osteoarthritis (KOA) compared to oxygen control, with no added benefit from higher doses. These findings align with ozone’s dual anti-inflammatory and antioxidant mechanisms, including suppression of pro-inflammatory cytokines (e.g., IL-1β, TNF-α) [1, 2] and upregulation of superoxide dismutase [3], alongside chondroprotective effects via collagen synthesis and inhibition of cartilage-degrading enzymes [4]. While higher doses (e.g., ≥ 35 µg/mL) theoretically amplify these effects through oxidative stress modulation [5, 21], our study found no added benefit with 40 µg/mL compared to 20 µg/mL, suggesting a threshold effect where lower doses sufficiently activate therapeutic pathways to produce maximal clinical benefit in mild-to-moderate KOA.

Our results extend evidence from placebo-controlled trials. For instance, Raeissadat et al. [38] and Costa et al. [39] corroborated ozone’s superiority over saline (p < 0.01), with sustained benefits observed by Eslami et al. [40] at 6 months. Similarly, Lopes de Jesus et al. [19] demonstrated 20 µg/mL ozone’s efficacy over saline, while Hashemi et al. [20] reported sustained improvements at 35 µg/mL versus steroids and a 2024 umbrella review by Lino et al. [21] reinforced ozone’s efficacy over placebo. However, comparisons with hyaluronic acid (HA) remain mixed. While Sconza et al. [24] found no difference between 30 µg/mL ozone and HA at 6 months, Korkmaz et al. [25] reported comparable outcomes between 10 µg/mL ozone and HA. These discrepancies may reflect protocol variability rather than inherent inefficacy of ozone, underscoring the need for standardized regimens [26].

Crucially, our direct comparison of 20 µg/mL and 40 µg/mL—a novel contribution to the literature—revealed no significant differences, challenging assumptions that higher doses enhance efficacy. This aligns with Paoloni et al. [41], who observed comparable outcomes between 10 µg/mL and 20 µg/mL ozone, and Smith et al. [42], who attributed ozone’s dose-independent effects to early cytokine saturation, showing that 15 µg/mL significantly reduces synovial IL-6 levels (p = 0.01) compared to placebo, suggesting a threshold effect where higher doses do not proportionally amplify anti-inflammatory responses. This aligns with our hypothesis that 20 µg/mL achieves maximal therapeutic activation.

Recent studies further validate these findings. A 2023 RCT by Chen et al. [22] compared 15 µg/mL, 20 µg/mL, and 30 µg/mL ozone, finding equivalent WOMAC improvements across doses at 3 months (p > 0.05). Martínez-Sánchez et al. [23] reported similar VAS reductions between 20 µg/mL and 35 µg/mL ozone over 6 months. Similarly, Aliyev et al. (2023) observed pain reduction with 15 µg/mL ozone at one and three months [43]. These studies, alongside our findings, validate ozone’s efficacy across a wide concentration range (15–40 µg/mL) and highlight its role as a viable alternative to traditional therapies. However.

Lower ozone concentrations (e.g., 20 µg/mL) may offer logistical advantages, including reduced costs and enhanced safety by minimizing oxidative harm to vulnerable tissues [44], as evidenced by Gupta et al. [27] and enhance patient tolerance, facilitating adherence to repeated injection protocols if required. This pragmatic approach aligns with Jeyaraman et al.’s [26] study, which advocates for optimizing ozone concentration and treatment protocols to achieve therapeutic efficacy in KOA.

Subgroup analyses revealed significantly greater improvements in K-L grade II versus III patients, consistent with Chen et al.’s [22] findings that early-stage KOA responds better to anti-inflammatory therapies. This suggests ozone therapy is most effective in moderate disease stages, where residual cartilage and synovial responsiveness permit therapeutic benefits. Advanced disease (grade III) may involve irreversible structural damage, limiting efficacy—a pattern observed with other biologic therapies [45, 46].

Limitations

This study has several limitations. First, the modest sample size (n = 59) may limit statistical power to detect subtle differences between doses. Second, the two-month follow-up period, while consistent with short-term placebo-controlled trials [32, 34], precludes conclusions about the long-term durability of ozone therapy. A follow-up of 6–12 months would better elucidate sustained efficacy and safety. Third, unmonitored physical activity levels among participants may have confounded functional outcomes, such as the 6MWT and TUG results. Fourth, the absence of a true placebo arm (the control group received standard care without intra-articular injections) limits direct comparisons with non-intervention scenarios.

Future directions

Larger trials with extended follow-ups are warranted to validate these findings and assess longitudinal outcomes. Future studies should integrate objective measures of physical activity (e.g., accelerometers) and biomarkers of inflammation (e.g., IL-6) or oxidative stress (e.g., superoxide dismutase) to clarify mechanisms and refine protocols. Also incorporating advanced imaging (e.g., MRI cartilage mapping) could further reduce variability and enhance reproducibility.

Conclusion

Our findings demonstrates that intra-articular ozone therapy at 20 µg/mL and 40 µg/mL significantly reduces pain and improves functional outcomes in patients with knee osteoarthritis (KOA), with 20 µg/mL emerging as optimal for balancing efficacy, safety, and cost. Subgroup analyses further highlight the importance of early intervention, as patients with moderate radiographic severity (Kellgren-Lawrence grade II) exhibited superior responses compared to those with advanced disease (grade III). Future research should prioritize standardized protocols and mechanistic studies to refine therapeutic thresholds.