Abstract
Background:
Platelet-rich plasma (PRP) treatment for chronic lateral elbow tendinopathy (LET) has increased because of its potential for prolonged symptom relief and improved function. Limited studies have definitively documented long-term benefits.
Purpose:
To assess the efficacy of a single intratendinous PRP injection compared to a corticosteroid injection for the treatment of LET.
Study Design:
Randomized controlled trial; Level of evidence, 1.
Methods:
In total, 48 participants (n = 50 elbows), aged 18 to 65 years, were randomly assigned to ultrasound-guided PRP (n = 26) or corticosteroid (control, n = 24) injection. Patient-Rated Tennis Elbow Evaluation (PRTEE) and Quick version of the Disabilities of the Arm, Shoulder and Hand questionnaire (QuickDASH) were compared at baseline and 4, 8, 12, 16, 26, and 52 weeks. Secondary outcomes were assessed via grip strength, visual analog scale (VAS) scores, and overall satisfaction with treatment. Wilcoxon rank-sum tests and longitudinal analysis of covariance models were used to assess outcomes over time.
Results:
At 4 weeks, mean PRTEE scores were 47.6 ± 3.7 in the PRP group compared to 14.8 ± 3.9 in the CSI group (P < .001). At 8 weeks, PRTEE scores were 32.1 ± 3.7 for PRP and 15.2 ± 4.0 for CSI (P = .003). At 12 weeks, scores were 26.3 ± 3.9 for PRP versus 16.0 ± 4.1 for CSI (P = .07). By 26 weeks, mean scores favored PRP (17.7 ± 6.5 vs 35.3 ± 6.8; P = .07), and by 52 weeks, PRP scores remained lower (14.4 ± 6.3 vs 29.6 ± 6.3; P = .10). At 4 weeks, mean QuickDASH scores were 22.4 ± 1.1 in the PRP group versus 15.5 ± 1.1 in the CSI group (P < .001). At 8 weeks, PRP scores were 19.3 ± 1.1 compared to 15.8 ± 1.2 for CSI (P = .04). No significant differences were observed at 12 weeks (17.7 ± 1.1 vs 16.6 ± 1.2; P = .49) or 16 weeks (16.8 ± 1.1 vs 18.4 ± 1.2; P = .35). At 26 weeks, QuickDASH scores favored PRP (15.7 ± 1.6 vs 20.3 ± 1.7; P = .05), and this difference persisted at 52 weeks (14.0 ± 1.6 vs 18.6 ± 1.6; P = .05). However, VAS scores were on average 1.5 points lower in the PRP group across all time points.
Conclusion:
Our study demonstrated that corticosteroids resulted in greater short-term improvement, while PRP demonstrated superior longer-term outcomes at 6 and 12 months. PRP was associated with lower average VAS scores over time.
Tendinopathies result in significant individual and societal costs related to physician visits, lost wages, and disability. Lateral elbow tendinopathy (LET) is a common overuse tendinopathy encountered by clinicians, with an estimated 1% to 3% of the general population affected. 5 It most commonly affects the dominant arm and individuals between 35 and 60 years of age. 24 LET is frequently associated with repetitive occupational activities that contribute to overuse and result in degenerative changes at the origin of the common extensor tendon. 26
Histopathologic findings of chronic tendinopathy, described as “angiofibroblastic tendinosis,” include disruption and disorganization of collagen fibers, partial tearing, and increased neovascularity with immature fibroblastic elements.12,22,28 Imaging studies using ultrasound (US) have also shown increased tendon size, loss of fibrillar structure, calcifications, partial tearing with hypoechogenicity, and increased scattered Doppler activity. 15 Tendinosis changes are believed to arise from incomplete or failed tendon healing mechanisms; however, the exact pathophysiology continues to be investigated. LET is often refractory to conventional nonoperative treatments, including physical therapy, bracing, and nonsteroidal anti-inflammatory drugs (NSAIDs). 25 The absence of acute inflammatory cells in LET supports its classification as a degenerative pathology but may also explain why corticosteroid injection (CSI) is not consistently effective for long-term symptom relief. 16
In contrast, platelet-rich plasma (PRP) has emerged as a biological treatment aimed at stimulating tendon healing through the local delivery of growth factors and cytokines.3,9 PRP is an autologous concentrate of platelets and growth factors that is used as an injection therapy for recalcitrant tendon pathologies, including LET. 19 It is theorized that healing of the common extensor tendon origin may be limited by its relative hypovascularity and hypocellularity, which limit access to necessary regenerative growth factors. 6 PRP therapy delivers platelets directly to the site of pathology. Platelet activation leads to the release of various growth factors, which promote tissue regeneration (eg, platelet-derived growth factor, transforming growth factor β, and vascular-derived endothelial growth factor) from their α granules, as well as additional cytokines and clotting factors believed to stimulate tendon healing via tendon cell proliferation, differentiation, and angiogenesis.1,23 PRP may serve to enhance healing of degenerative tendinopathies by delivering supraphysiologic concentrations of growth factors that catalyze cellular chemotaxis, matrix synthesis, tissue proliferation, and remodeling.1,3,4,23,27
Several prior randomized trials and systematic reviews have compared PRP and corticosteroid injections for LET, with early evidence suggesting that PRP may provide some benefit in pain and functional outcomes.2,19 More recent randomized trials and meta-analyses have strengthened this evidence base, demonstrating more consistent advantages of PRP over corticosteroids, particularly beyond 6 months of follow-up.18,29,30 Despite growing use, variability in PRP preparation methods and patient selection has led to mixed results across studies. 9 The purpose of this study is to evaluate the 1-year clinical outcomes of a single injection of leukocyte-rich PRP injection compared to corticosteroid injection for the treatment of chronic LET in a randomized, double-blind controlled trial. We hypothesized that a single PRP injection would lead to greater improvements in Patient-Rated Tennis Elbow Evaluation (PRTEE) and the Quick version of the Disabilities of the Arm, Shoulder and Hand questionnaire (QuickDASH) scores at 1 year compared with corticosteroid injection.
Methods
Study Design and Participants
This was a prospective, double-blinded, randomized 2-arm controlled clinical trial in 48 participants (50 elbows), aged 18 to 65 years, with clinical examination and US confirmation of LET. The protocol was approved by the institutional review board (IRB No. 2016-0086; ClinicalTrials.gov Identifier: NCT03072381), and all participants completed written informed consent.
Participants with clinically and US-confirmed LET, symptoms lasting >3 months, and at least 2 unsuccessful nonoperative treatments (eg, NSAIDs, bracing, activity modification, cryotherapy), including a supervised eccentric strengthening physical therapy program, were randomized into 2 groups. Each participant had completed a supervised eccentric strengthening–based physical therapy program before enrollment and was instructed to continue with their rehabilitation exercises after their treatment once procedural pain had subsided. Exclusion criteria included inability to comply with study follow-up requirements; history of bleeding disorders, low platelet counts, or other hematologic conditions; elbow pain due to an alternative cause; current use of anticoagulation or immunosuppressive therapy; known allergy to lidocaine or acetaminophen; self-reported pregnancy; a workers’ compensation injury; pending litigation; or concurrent opioid use for pain.
Interventions
Participants were treated with either a single US-guided corticosteroid injection (a 3-mL mixture of 1 mL 40 mg/mL triamcinolone and 2 mL 1% lidocaine) or a single US-guided injection of autologous PRP (3 mL), using the PEAK Platelet Rich Plasma system (DePuy Synthes Mitek Sports Medicine), a leukocyte-rich PRP system. Each participant also received 5 US-guided fenestrating injection passes of the common extensor tendon origin in a fan-shaped pattern across the tendon lateral epicondyle origin using a 22-gauge needle used to perform the procedure (Figure 1).

Ultrasound (US)–guided platelet-rich plasma injection technique of lateral elbow tendinopathy (LET) areas of tendinopathy seen as hypoechogenicity (short arrow). Five US-guided fenestrating injection passes (long arrows) into the common extensor tendon origin in a fan-shaped pattern across the tendon lateral epicondyle (LE) origin using a 22-gauge needle (arrowheads) were used to perform the procedure. Enthesophyte (pentagon) is commonly seen in LET. R, radial head.
Outcomes
The primary outcomes were the PRTEE and QuickDASH questionnaires. Both tools are validated measures with demonstrated content, construct, and criterion validity in populations with LET. For both instruments, higher scores indicate greater pain and disability.
Secondary outcome measures included a visual analog scale (VAS) score before grip strength testing, grip strength measured with a Jamar hand dynamometer (average of 3 maximum-effort tests), and overall satisfaction with treatment on a 7-point Likert scale. Because outcomes were expected to vary at different time frames between the 2 treatment groups, questionnaires were administered at baseline and 4, 8, 12, 16, 26, and 52 weeks following treatment. Grip strength was assessed in-person at baseline, 26 weeks, and 52 weeks.
Randomization and Blinding Procedures
Participants and assessors were blinded to the group allocation. Control participants also underwent phlebotomy to maintain blinding. Injection syringes were cloaked using opaque adhesive tape to ensure the injector and participants were unaware of treatment group allocation. Group assignments were placed in sequentially numbered, opaque, sealed envelopes, which were opened only after participant enrollment to ensure allocation concealment. A total of 48 participants (50 elbows) were enrolled, including 2 participants who had bilateral elbow involvement. Both elbows in these participants were randomized to the same treatment group to preserve group allocation consistency.
Power and Sample Size Estimation
As a result of no existing data on pain- and function-dependent, elbow-specific quality of life of LET posttreatment with PEAK PRP, we were unable to power this study based on potential differences in our primary and secondary outcome measures. Instead, we powered the study to ensure that the observed variance estimates did not grossly underestimate the actual values, so that eventual sample size calculations for a future confirmatory study are not overoptimistic. To restrict the probability that our sample variance (assumed to be 0.13) underestimated the true variance by 25% or more, we estimated that we would need 20 patients per group. After allowing for a 20% dropout rate, we aimed to recruit 25 participants per group.
Statistical Approach
Descriptive statistics were used to describe the study population and each treatment group. Frequencies and percentages were reported for categorical variables, and means (standard deviations) or medians [IQRs] were reported for continuous variables. Values in parentheses represent standard deviations unless otherwise specified. A Wilcoxon rank-sum test was used to compare satisfaction scores between treatment groups. Analysis of covariance, adjusted for baseline outcome scores and including a time × treatment interaction, was used to assess treatment effects across all follow-up time points. Outcomes analyzed included the PRTEE and QuickDASH scores, grip strength, and VAS. Least squares means and standard errors are provided for each treatment group at each time point. All statistical analyses were performed using SAS version 9.4 (SAS Institute).
Results
Of the 127 participants recruited, 79 were ineligible to participate (Figure 2). All participants suspected of having LET were confirmed to have LET at the time of US. A total of 48 participants were randomized to either PRP or corticosteroid treatment; 2 participants had bilateral elbows treated according to the same treatment group. At baseline, 56% (n = 28/50) of participants were female, and 92% were White. Mean (SD) age was 48.1 (8.2) years, and mean (SD) body mass index was 26.1 (4.3) kg/m2. The CSI group demonstrated slightly higher baseline grip strength and PRTEE scores compared to the PRP group (Table 1); however, these differences were not statistically significant (grip strength, P = .75; PRTEE, P = .45).

CONSORT diagram.
Demographics and Baseline Characteristics Overall and by Study Group a
BMI, body mass index; PRP, platelet-rich plasma; PRTEE, Patient-Rated Tennis Elbow Evaluation; QuickDASH, Quick version of the Disabilities of the Arm, Shoulder and Hand questionnaire; VAS = visual analog scale.
PRTEE
The association between study group and PRTEE scores differed across time points (P < .001; Table 2). At 4 weeks and 8 weeks postinjection, PRTEE scores in the PRP group were higher (worse) than those in the steroid group (P < .001 and P = .003, respectively). Although not statistically significant, evidence suggests that PRTEE scores were still higher among the PRP group at 12 weeks postinjection (PRP, 26.3 [3.9]; CSI, 16.0 [4.1]; P = .07); however, no difference was observed at 16 weeks and beyond. Although not statistically significant, by 26 and 52 weeks, the CSI group reported higher PRTEE scores (P = .07 and P = .10, respectively) than the PRP group (Figure 3).
Least Squares Mean (SE) PRTEE and QuickDASH Scores by Study Group Across Follow-up Time Period and Adjusted for Baseline Scores a
CSI, corticosteroid injection; QuickDASH, Quick version of the Disabilities of the Arm, Shoulder and Hand questionnaire; PRP, platelet-rich plasma; PRTEE, Patient-Rated Tennis Elbow Evaluation.
Overall P value represents group × time interaction.
Difference calculated as PRP – steroid.
Given significant interaction, the P value reflects differences in PRTEE scores by study group at each follow-up time point.

Patient-Rated Tennis Elbow Evaluation (PRTEE) and Quick version of the Disabilities of the Arm, Shoulder and Hand questionnaire (QuickDASH) mean scores and confidence interval bands for follow-up time points by study group.
QuickDASH
An association between study group and QuickDASH scores differed across time points (P < .001; Table 2). QuickDASH scores were higher (worse) for the PRP group at 4 weeks (PRP, 22.3 [1.1]; CSI, 15.5 [1.1]; P < .001) and 8 weeks (PRP, 19.3 [1.1]; CSI, 15.8 [1.2]; P = .04) postinjection; however, no differences were identified between study groups at 12 and 16 weeks. By 26 and 52 weeks, the CSI group reported higher (worse) QuickDASH scores than the PRP group (P = .05) (Figure 3).
Grip Strength
No significant differences between study group and grip strength over time (eg, no significant interaction) were detected (P = .89; Table 3). However, averaged across all time points, the PRP group documented greater grip strength (37.32 [1.32] kg) compared to the CSI group (32.71 [1.40] kg), P = .02. No differences in grip strength were identified in follow-up time points after adjusting for study group and baseline grip strength (P = .38).
Least Squares Mean (SE) Grip Strength and VAS Scores, Adjusted for Baseline Scores by Study Group and Follow-up Time (Weeks) a
CSI, corticosteroid injection; PRP, platelet-rich plasma; VAS, visual analog scale.
VAS Scores
No significant differences between study group and VAS scores over time were detected (P = .86; Table 3). However, averaged across all time points, the PRP group reported lower VAS scores (1.34 [0.41]) compared to the CSI group (2.70 [0.43]; P = .03). After controlling for baseline VAS scores, average pain scores at 52 weeks were 1.09 points lower than they were at 26 weeks across both treatment groups (P = .008; Table 3).
Discussion
The major findings of our study showed that corticosteroid injection was associated with greater short-term improvement at 4 and 8 weeks, while PRP demonstrated superior outcomes at 26 and 52 weeks. Specifically, mean PRTEE scores were lower in the PRP group at long-term follow-up (26 weeks, 17.7 ± 6.5 vs 35.3 ± 6.8, P = .07; 52 weeks, 14.4 ± 6.3 vs 29.6 ± 6.3, P = .10). QuickDASH scores also favored PRP at 26 and 52 weeks (15.7 ± 1.6 vs 20.3 ± 1.7, P = .05; 14.0 ± 1.6 vs 18.6 ± 1.6, P = .05). Across all time points, grip strength was higher (37.3 ± 1.3 vs 32.7 ± 1.4, P = .02) and VAS scores were lower (1.34 ± 0.41 vs 2.70 ± 0.43, P = .03) in the PRP group. These findings highlight the potential durability of PRP compared with corticosteroid injection for the treatment of chronic LET. Our study findings align with recent studies demonstrating the long-term benefit of PRP over CSI for the treatment of LET. A 2025 meta-analysis by Maroun et al 18 confirmed that PRP is associated with greater reductions in pain and improved function beyond 6 months compared to corticosteroids. Similarly, Xu et al 29 reported that PRP provides more durable symptom relief, especially beyond the 6-month time point. These findings are consistent with the results of our trial, in which both PRTEE and QuickDASH scores improved more substantially in the PRP group by 26 and 52 weeks. Our findings are also consistent with other earlier studies, including Peerbooms et al, 23 who demonstrated a statistically significant improvement in DASH and VAS scores after a 1-year follow-up in the PRP group compared to the corticosteroid group. Although the CSI group initially showed short-term improvement, they also required a higher rate of reintervention. 23 These results are also in agreement with Mishra et al, 20 who showed that 71.5% of PRP participants had an improvement in pain scores versus 55% in the placebo group at 24 weeks. Finally, Mi et al 19 showed in a meta-analysis only short-term (2-4 weeks) improvement with CSI versus PRP, proving to be more effective for pain relief and functional improvement at intermediate-term (12 weeks) and long-term (6 months and 1 year) follow-up. Although pain and function did not improve synchronously at every follow-up visit, this pattern may reflect the hypothesized biological activity of PRP, in which early inflammatory responses are followed by progressive tendon remodeling and clinical improvement. 31
Our study adds to the existing literature by incorporating several unique methodological features that distinguish it from prior randomized controlled trials of PRP for LET. First, to our knowledge, this is the first randomized controlled trial to evaluate the PEAK leukocyte-rich PRP system, whereas earlier studies have examined other commercial preparations, including leukocyte-poor formulations. Second, we used a single ultrasound-guided intratendinous injection with standardized fenestration, in contrast to some prior protocols that employed multiple PRP injections or did not consistently use ultrasound guidance. Third, rigorous blinding procedures were implemented in our study, including sham phlebotomy for corticosteroid controls and masking of syringes to blind injectors, which reduced the risk of allocation bias. Taken together, these factors highlight important differences in study design and intervention compared with previously published trials, as well as support the contribution of our findings to the broader evidence base on PRP for chronic LET.
Importantly, we also found that the average improvement in pain and function approached but did not always exceed the minimal clinically important difference reported in prior studies. Franchignoni et al 7 suggested that a 16-point improvement in the QuickDASH reflects a clinically meaningful change, and our PRP cohort showed an average improvement of approximately 12 points. While this is below the minimal clinically important difference threshold, the trend toward lower VAS scores and higher patient satisfaction in the PRP group supports a perception of meaningful benefit.
Lana et al 13 described a process of “inflammatory regeneration” in which platelets and leukocytes both play an important role in healing via roles in activating signaling cells to ignite a prohealing inflammatory cascade and cells responsible for tissue regeneration. It is this cascade that explains why PRP has improved long-term outcomes via targeting the true chronic degenerative changes in the tendon. Corticosteroids are known to have a potent anti-inflammatory effect; however, studies have shown that acute inflammatory cells are not present in chronic tendinopathy.2,9,13 In the study by Muto et al, 21 tendon degradation cells with apoptosis induction and increased matrix metalloproteinase activity were observed after CSI in tendon histological studies, resulting in weakened biomechanical properties of rat tendons. This is also consistent with prior studies identifying risks of CSI, including in expedited tendon degeneration with collagen destruction and increased risk of tendon rupture.8,11 In contrast, risks of PRP are minimal, with the most notable being higher postinjection pain, which resolves without additional treatment measures. 14
Our study is not without limitations. Our outcome measures are subjective in nature; however, Evans et al 5 confirmed the validity of these patient-centered instruments, with QuickDASH and PRTEE among the top-performing clinical outcome surveys supported by strong evidence and reviewer consensus. We also documented grip strength as an objective outcome measure, which helps mitigate this limitation. An additional limitation of the study is the lack of consensus agreement for the appropriate dose and type of PRP used among clinicians and researchers, resulting in difficulty in interpreting consistent results. Furthermore, we did not collect information regarding corticosteroid injections administered more than 3 months prior to enrollment. Remote prior corticosteroid exposure may have occurred but was not systematically captured, reflecting challenges with accurate patient recall and incomplete medical records from outside providers. This approach is consistent with prior randomized controlled trials, which focus on excluding recent corticosteroid use. Because different commercial PRP production machines are used for cell separation, a nonstandardized PRP product with varying cell-type concentrations results unless the same processing machines are used. However, advances in technology are allowing for more standardization, such as the ability to adjust leukocyte concentration. This study used leukocyte-rich concentrate, with this system preferentially retaining mononuclear leukocytes (eg, monocytes and lymphocytes, 84% recovery). 17 Mononuclear cells are long-lived and may supply additional growth factors, including vascular endothelial growth factor. As a result, preferentially concentrating mononuclear cells in a leukocyte-rich PRP may be desirable. Initial studies are mixed, however, with some showing increased histologic improvement with leukocyte-poor PRP. 10 With the rise in PRP treatment applications, more research and studies will need to be performed evaluating the merits of leukocyte-rich as compared to leukocyte-poor PRP.
This current study also showed no difference in grip strength scores at follow-up time frames when comparing both groups. VAS scores were, on average, 1.5 points lower in the PRP group versus CSI for all follow-up time points. While the differences in VAS scores between groups were not statistically significant at individual time points, the PRP group consistently reported lower average scores across follow-up. At 52 weeks, pain scores were approximately 1 point lower than at 26 weeks after adjusting for baseline values. Although a 1-point change on the VAS may seem modest, even small reductions in chronic pain can be meaningful to patients, especially when combined with improved function. Overall treatment satisfaction was higher in the PRP group than in the CSI group. These results suggest that PRP is a superior injection treatment option compared to CSI for the long-term treatment of LET and can be used for patients with chronic LET. Future research should be directed toward PRP effects on cellular and histologic changes in chronic LET. Further study should also investigate the tissue-regenerative effects on tendon morphology and biomechanical properties following PRP treatment.
Conclusion
Our study demonstrated that corticosteroid injection provided greater short-term improvement at 4 and 8 weeks, while PRP resulted in superior longer-term outcomes at 6 and 12 months. PRP was also associated with lower average VAS pain scores across all time points. PRP may offer superior long-term treatment outcomes compared with corticosteroid injection for chronic LET.
Footnotes
Final revision submitted July 17, 2025; accepted September 24, 2025.
One or more of the authors has declared the following potential conflict of interest or source of funding: This study was supported by an industry grant from Depuy Synthes Mitek Sports Medicine (Funding ID: MSN193096. Sponsor Reference Number: 201501). The sponsor had no role in study design, data collection, analysis, interpretation, or manuscript preparation. J.J.W. has received research support from Depuy Synthes Mitek Sports Medicine. K.S.L. has received research support from Mitek Depuy, NBA-GE Collaborative, and NFL and holds stock or stock options in Remedy Logic. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
The protocol was approved by the institutional review board The University of Wisconsin - Madison School of Medicine and Public Health (IRB No. 2016-0086).
