Musculoskeletal complaints account for up to 59% of new patient referrals to rheumatology practice1 and up to 15% of consultations in primary care.1 Tendinopathies represent a large proportion of these complaints with a high incidence of chronicity and recurrence.2 Clinically, pain and dysfunction are the main symptoms present in tendinopathy, whereas signs can include swelling or thickening of the tendon.
Patients presenting with chronic tendinopathy have a limited range of treatment options available once conservative treatment in the form of rehabilitation has failed. Injection of tendons with local steroid injection may increase the risk of tendon rupture, a particular concern in the management of athletes.3,4
Tendinopathy is considered to represent an incomplete healing response to repetitive microtrauma and interstitial tearing of the tendon. Autologous blood injection (ABI) is used to stimulate the healing process within the damaged tendon. The rationale for its therapeutic benefit is based on the fact that growth factors are contained within the injected blood.
The hypothesis of the mechanism of action is that the growth factors carried in the blood act as humeral mediators to induce the healing cascade.5–8 Once delivered, the platelets begin secretion of growth factors within 10 minutes with more than 95% of the presynthesized growth factors being released within an hour.9 Platelets are viable for 7 days and will continue to release growth factor into the tissue during this time.
Numerous studies support the use of autologous blood in the treatment of tendinopathies. A study injecting autologous blood into patellar tendon of rabbits found all tendons to have normal histology and a 15% increase in tensile strength compared with controls at 12 weeks.10
Edwards and Calandruccio prospectively examined the effect of ABI for refractory lateral epicondylitis.6 Twenty-eight patients were injected with 2 mL of autologous blood under the extensor carpi radialis brevis and followed up for at least 6 months (average 9.5 months). All patients had failed previous nonsurgical treatments. The average pain score decreased from 7.8 to 2.3 and Nirschl stage decreased from 6.5 to 2.0 with greater improvement in those who received more than 1 injection.
A prospective study by James et al evaluated the efficacy of ultrasound-guided ABI for the treatment of patellar tendinosis. In this study, 47 knees in 44 patients with refractory tendinosis (mean duration of symptoms 12.9 months) underwent ultrasound-guided ABI on 2 occasions 4 weeks apart. Results demonstrated a significant improvement in VISA scores (P < 0.001) at mean follow-up of 14.8 months. Follow-up sonographic assessment demonstrated a reduction in the overall tendon thickness, in the size of the identified hypoechoic area within the proximal patellar tendon, and in the size of the interstitial tears within the substance of the tendon.11
Our objective was to conduct a unique double-blind randomized controlled trial (RCT) to assess the effectiveness of ABI for chronic tendinopathy in the elbow (medial and lateral epicondylitis) and knee (patellar tendinopathy). Our primary hypothesis is that ABI would give superior results to placebo injections over a 12-month period. The secondary hypothesis is that we expect to observe an improvement in both groups due to the needling effect from the injections. In this study, we present our findings for patella tendinopathy (PT).
The study aims to enhance and advance the body of evidence to support the use of ABI as a treatment for chronic tendinopathies. NICE12,13 has advised future research should be in the context of RCTs that define chronicity of tendinopathy, adjunctive treatments (including physiotherapy and “dry needling”), use of ultrasound, and quality of life outcomes with a minimum follow-up of 1 year.
Our study population were all patients older than 18 years recruited from referrals received at the Homerton University Hospital Sports and Musculoskeletal Medicine Clinic and the Musculoskeletal Injection Clinic. Patients with a clinical diagnosis of PT underwent an ultrasound examination to confirm the diagnosis. Sonographic features of tendinosis are well established and include increased tendon size, focal alteration in tendon echotexture, interstitial clefts, and neovascularity (Supplemental Digital Content 1, https://links.lww.com/JSM/A60, modified Öhberg score).14–16
An ABI was offered only if a program of rehabilitation including eccentric exercise over a period of at least 3 months had failed to produce any significant response. If the patient had been given a previous steroid injection (not within 3 months), this did not preclude the option of an ABI.
Contraindications to injection included previous allergic reaction to local anesthetic or corticosteroid, possible local sepsis, concomitant anticoagulant therapy, peripheral vascular disease, and those younger than 18 years.
All patients completed the Short-Form—McGill Pain Questionnaire (SF-MPQ), visual analog scale (VAS) for pain, and the Victoria Institute of Sport Assessment for Patella Tendinopathy (VISA-P) scale.
Study Design and Intervention
This was a double-blind RCT performed from March 2010 to March 2012.
Two practitioners (A and B) were present during the initial process of explaining the procedure and obtaining consent. Practitioner A prepared the area for injection with chlorhexidine spray. This was followed by an injection of 2-mL 1% lidocaine into the subcutaneous tissue around the tendon to be injected.
While the local anesthetic was given sufficient time to act, 2 mL of autologous blood was withdrawn from the contra lateral antecubital fossa using aseptic technique. Practitioner A then left the room with the syringe of autologous blood. The syringe was covered to disguise the contents and an identical syringe containing the same volume of control substance of normal saline was drawn up. Through a computer-generated randomization program, practitioner A could decipher whether this patient would receive autologous blood or normal saline.
Neither practitioner B nor the patient was aware of what the patient was receiving, hence double blinding the study. Practitioner A then returned to the room with the covered syringe containing either blood or normal saline ready for practitioner B to inject into the tendon under ultrasound guidance (Figure 1). The region of tendon demonstrated on ultrasound as being tendinopathic was repeatedly fenestrated with a 23-gauge needle. This took the form of 10 passes through the tendinopathic tissue, typically the deep aspect of the tendon at its attachment to the patella. This was followed by slow injection of the autologous blood or normal saline. Patient consent was obtained (Supplemental Digital Content 2, https://links.lww.com/JSM/A61, patient consent forms).
This study design process is summarized in the flowchart (Supplemental Digital Content 3, https://links.lww.com/JSM/A62, flowchart).
Patients were asked to stay in the department for 20 minutes to monitor for adverse reactions.
Patients were advised on relative rest with normal activities of daily living but no sport, activity modification, and a standardized rehabilitation program of eccentric exercise (Supplemental Digital Content 4, https://links.lww.com/JSM/A63, exercise program). Patients were advised that the only analgesia permitted was paracetamol, and that no form of nonsteroidal anti-inflammatory drug whether topical or parental was permitted so as not to confound results. Follow-up review appointments were made at 6 weeks during which pain scores were reassessed and a second injection considered. Follow-up was then at 3, 6, and 12 months. At each follow-up, VAS, SF-MPQ, and VISA-P questionnaires were filled.
Patients could withdraw from the study at any time (Supplemental Digital Content 5, https://links.lww.com/JSM/A64, patient information).
Unlike corticosteroid, the use of autologous blood is unlikely to increase risk of tendon damage or rupture. The risk of allergy or the introduction of exogenous infection is considered negligible.17 All patients were nevertheless warned of possible adverse reactions—postinjection flare, infection, and tendon rupture.
To calculate a sample size, it was necessary to establish what would be a clinically detectable change in outcome and be of clinical significance. Grafton et al18 examined the SF-MPQ for test–retest reliability and included calculation of the coefficient of repeatability (CoR) as a measure of the reliability of repeated measures. Grafton et al18 calculated a CoR for the sensory component of the SF-MPQ of 4.54. The sensory component of the SF-MPQ is a primary outcome measure that will be used in this study and progresses from 0 to 33, 4.54 therefore represents a change of 14%. In support of this, Todd et al determined the amount of change in pain severity, as measured by a VAS (0-100 mm scale), that which constitutes a minimum clinically significant difference as 13 mm (13%).19 Additionally a pilot study had demonstrated an SD in preinjection SF-MPQ scores of 4.6.
Given this information, a sample size calculation was performed using Minitab software. Using a power of 0.95 and significance of 0.05, a sample size of 28 was calculated for the equivalent nonparametric test.
Statistical analysis was performed using SPSS version 16 (SPSS Science, Chicago, Illinois) and significance relates to all P values less than 0.05. On completion of the study, there were data on the preinjection and postinjection SF-MPQ score, and VAS scores on all patients. In addition, data were analyzed for the patella tendon from the VISA-P scores. The data were considered to be nonparametric in nature and therefore analyzed using the Wilcoxon signed-rank test, looking for significant differences in the results at 1, 3, and 12 months. Data from the 2 groups were then compared using the independent-samples Mann–Whitney U tests.
Between March 2010 and March 2012, there were 22 cases of chronic PT who fulfilled the criteria of selection. These 22 patients were randomly assigned into either the saline arm of the study or the ABI group. Their baseline characteristics are shown in Tables 1 and 2.
The saline group had 8 males and 3 female patients. They had an average age of 42 years with an average Öhberg neovascularity score pretreatment of 2.5. This group had a mean duration of symptoms of 19.2 months (range, 5-36). The ABI group had 10 males and 1 female patient. They had an average age of 39 years with an average Öhberg neovascularity score pretreatment of 2.9. The ABI group had a mean duration of symptoms of 16.7 months (range, 5-36).
Visual Analog Scale Scores
In the saline group, the mean VAS score reduced from 7.9 to 4.5 at 1-month (P = 0.003) follow-up. This was statistically significant and the VAS score continued to decrease to 4.0 by month 3 (P = 0.003) and 3.3 at then end of the study at 1 year (P = 0.005).
In the blood group, the mean VAS score reduced from a baseline of 7.1 to 4.5 (P = 0.005) by the end of month 1. This reduced further to 3.5 (P = 0.005) by the end of 3-month follow-up and at 1 year was the lower further with a mean of 3.1 (P = 0.003). These results were all statistically significant (Figure 2).
Victoria Institute of Sport Assessment for Patella Tendinopathy Scores
The participants in the saline group had a mean baseline VISA-P score of 19.6. By 1-month postinjection, this had improved to a mean of 39.2 (P = 0.012). This level was maintained by month 3, but at 1-year, postinjection had improved further to a mean of 48.6 (P = 0.005), which was significant.
In the ABI group, the mean starting VISA-P score was 34.1. After injection, there were significant improvements at each consultation. The mean score rose to 50.7 (P = 0.021) by month 1, and then increased further to 57.7 (P = 0.005) by month 3 and further increased at 1 year to a mean of 62.5 (P = 0.003) (Figure 3).
Short-Form—McGill Pain Questionnaire Scores
In the saline group, the mean preinjection SF-MPQ was 31.4. By 1 month postinjection, this had reduced to a mean of 22.4 (P = 0.020), decreasing further, to a mean of 17.5 (P = 0.005) by month 3, and this level was maintained at 1 year with a mean of 17.2, but this was still significant in comparison with the baseline scores (P = 0.003).
In the ABI group, the starting SF-MPQ average score was 22.5. There were statistically significant decreases in the scores compared with baseline at each stage. The average score by month 1 reduced to 12.6 (P = 0.003), by month 3 to 10.5 (P = 0.005) and finished with an average of 10.7 (P = 0.008) by the end of the study at 1 year (Figure 4).
Saline Versus Autologous Blood Injection
The results of the 2 arms of the study were compared statistically using independent-samples Mann–Whitney U test. Regarding the VAS score, there was no statistical difference at any stage between the 2 arms with a P value of 0.841 at 1 year. Similarly, the VISA-P scores in comparison between the 2 groups showed no statistical difference at 1 month (P = 0.292), 3 months (P = 0.094), or at 1 year (P = 0.276). With the SF-MPQ questionnaire, the findings followed a similar pattern with no statistical difference at any point. Examples of sonographic appearances after ABI are shown in Figures 5–7. These figures represented trends seen during our research project in both groups.
Tendinopathy is a common complaint among both the athletic and nonathletic populations. Patella tendinopathy primarily effects those involved with jumping activities affecting 14% to 16%20,21 of athletes. There has been evidence produced on rehabilitation regimes from rest to eccentrics22 on a decline board but often the management is difficult, protracted, and not satisfying, both for the athlete and the clinician. There has been a lot of interest in the recent years with platelet-rich plasma (PRP)11 and ABI13 for the treatment of tendinopathies. Although results have been variable, it is still widely used, especially for chronic-resistant cases in the elite sports population. Research has demonstrated benefits in patients treated with PRP or ABI, but most well-constructed RCTs have shown no real advantages. In this study, we postulated that ABI would have advantages compared with saline in those suffering with chronic PT after failure of other conservative measures, such as physiotherapy and alternative injections. Platelet-rich plasma was not investigated in this study.
Previous studies by James et al11 and Kon et al23 showed benefit with ABI and PRP, respectively, for PT with VISA-P scores, but both these studies were based on a case series with no control group. Filardo et al24 did a nonrandomized trial for PRP in PT with a control group only receiving exercise therapy. In this series, there was a significant improvement in both groups but no significant difference between the control or treatment group. However, they did conclude that those in the PRP group had a better return to sporting activity level.
Randomized controlled trials in this area are few and far between. Kiter et al25 performed an RCT on chronic heel pain looking at dry needling versus steroid versus ABI in 44 patients. All improved significantly, but there was no difference between the groups. There were similar findings in De Vos et al's26,27 group in Achilles tendinopathy who randomized patients to saline or PRP and found no statistical difference in results between the 2 groups. Dragoo et al28 recently found benefit for PRP over saline injections in PT at 12 weeks but no difference by 26 weeks. There does however seem to have been better results in an RCT in those with lateral epicondylitis comparing PRP and ABI (Creaney et al29) with benefits in up to 70% of patients reported but with no true control group in the study.
The patients in our study had PT that had failed at least 3 months of eccentric exercises. The shortest time period any athlete had been suffering with PT was 5 months with the longest at 3 years. Two subjects were playing sport at an elite level and the rest were all amateur athletes over a wide variety of sports from football, basketball, hockey, and rugby.
To the best of our knowledge, this is the first high-level study conducted as a double-blind RCT looking at ABI versus saline injections for those suffering with PT. There were 22 cases of recalcitrant PT that had failed conservative treatments. All participants were given advise on eccentrics to be performed but without any formal physiotherapy. The injection technique was the same for saline and blood and involved dry needling of the area. At follow-up, all patients had a repeat ultrasound scan looking at tendon integrity and neovascularity (undertaken by the same consultant throughout the trial with patients in a consistent position) in addition to the validated outcome measures (VISA-P, VAS, and SF-MPQ). Our results showed significant improvement in both groups in VAS, VISA-P, and SF-MPQ scores at each review including at 1-year postinjection. However, importantly, there was no statistical difference in results between the 2 groups in any of our scoring criteria or outcomes.
A limitation of our study was that despite our best attempts to keep this double blind, the experienced clinician injecting the patient may have been able to differentiate whether saline or blood was being injected due to the viscosity difference when injecting. Although, it should be stressed that subjects were unaware of this and were unable to see the ultrasound. The fenestration procedure may have resulted in some endogenously delivered blood, but this is highly unlikely given the poor vascularity of tendons and would be minimal in comparison with the 2 mL of blood in the ABI group. In addition, we were unable to recruit 28 subjects, as required for power calculations, in the allocated time period and thus have presented the data from 22 subjects.
In our subjects, there was a reduction in neovascularity over the year, which was a trend that we observed, although the exact role and significance of neovessels and the pathogenesis of tendinopathies is still widely debated.30 Our patients continued to improve throughout the follow-up period. Despite sound theory for ABI of introducing growth factors into injured tendons and our study supporting previous research showing a benefit of ABI in difficult cases of PT, there was a similar benefit with saline injections. This provides an argument that it is the tendon fenestration that provides the therapeutic effect rather than the substance being injected.
There were no side effects reported in our trial and no major side effects are reported from previous research on ABI injections. This is supported by the recent NICE interventional procedure guidance (January 2013),13 which concluded no major safety concerns with ABI injections but that evidence on efficacy remains inadequate. To conclude, our study showed that both ABI and saline injections had similar efficacy in those with recalcitrant PT. Further research is required to extend this to a multicentre study and in other tendinopathies to confirm whether the effect is similar on other tendons. None of our patients required surgical intervention.
The authors are grateful to the Sports and Musculoskeletal Department in Homerton University Hospital for their support.
1. Bamji AN, Dieppe PA, Haslock DI, et al.. What do rheumatologists do? A pilot audit study. Br J Rheumatol. 1990;29:295–298.
2. McCormick A, Fleming D, Charlton J. Morbidity Statistics from General Practice. Fourth National Study 1993 (1991-92). London, United Kingdom: HMSO. Series MB5 no.3. 1993.
3. Nygren A, Berglund A, Von Koch M. Neck and shoulder pain: an increasing problem. Strategies for using insurance material to follow trends. Scand J Rehabil Med Suppl. 1995;32:107–112.
4. Balasubramaniam P, Prathap K. The effect of injection of hydrocortisone into rabbit calcaneal tendons. J Bone Joint Surg Br. 1972;54:729–734.
5. Sanchez AR, Sheridan PJ, Kupp LI. Is platelet rich plasma the perfect enhancement factor? A current review. Int J Oral Maxillofac Implants. 2003;18:93–103.
6. Edwards SG, Calandruccio JH. Autologous blood injection for refractory lateral epicondylitis. J Hand Surg Am. 2003;28:272–278.
7. Connell DA, Ali KE, Ahmad M, et al.. Ultrasound guided autologous blood injection for tennis elbow. Skeletal Radiol. 2006;35:371–377.
8. Iwasaki M, Nakahara H, Nakata K, et al.. Regulation of proliferation and osteochondrogenic differentiation of periosteum-derived cells by transforming growth factor-B and basic fibroblast growth factor. J Bone Joint Surg Am. 1995;77:543–554.
9. Marx RE. Platelet-rich plasma: evidence to support its use. J Oral Maxillofac Surg. 2004;64:489–496.
10. Taylor MA, Norman TL, Clovis NB, et al.. The response to rabbit patella tendons after autologous blood injection. Med Sci Sports Exerc. 2002;34:70–73.
11. James SJ, Ali K, Pocock C, et al.. Ultrasound guided dry needling and autologous blood injection for patellar tendinosis. Br J Sports Med. 2007;41:518–522.
12. NICE (National Institute of Clinical Excellence). Autologous Blood Injection for Tendinopathy. Interventional Procedure Guidance (IPG 279). London, United Kingdom: National Institutes for Health and Care Excellence; 2009.
13. NICE (National Institute Clinical Excellence). Autologous blood injection for tendinopathy. Interventional procedure guidance IPG438. London, United Kingdom: National Institutes for Health and Care Excellence; 2013.
14. Weinberg EP, Adams MJ, Hollenberg GM. Color Doppler sonography of patellar tendinosis. AJR Am J Roentgenol. 1998;171:743–744.
15. Terslev L, Qvistgaard E, Torp-Pedersen S, et al.. Ultrasound and power Doppler findings in jumpers knee—preliminary observations. Eur J Ultrasound. 2001;13:183–189.
16. Peers KH, Lysens RJ. Patellar tendinopathy in athletes: current diagnostic and therapeutic recommendations. Sports Med. 2005;35:71–87.
17. Nirschl RP, Pettrone FA. Tennis elbow: the surgical treatment of lateral epicondylitis. J Bone Joint Surg Am. 1979;61:832–839.
18. Grafton KV, Foster NE, Wright C. Test-retest reliability of the short-form McGill pain questionnaire: assessment of intraclass correlation coefficients and limits of agreement in patients with osteoarthritis. Clin J Pain. 2005;21:73–82.
19. Todd KH, Funk KG, Funk JP, et al.. Clinical significance of reported changes in pain severity. Ann Emerg Med. 1996;27:485–489.
20. Jarvinen M. Epidemiology of tendon injuries in sports. Clin Sports Med. 1992;11:493–504.
21. Lian OB, Engebretsen L, Bahr R. Prevalence of jumper's knee among elite athletes from different sports. A cross sectional study. Am J Sports Med. 2005;33:561–567.
22. Stanish WD, Rubinovich RM, Curwin S. Eccentric exercise in chronic tendinitis. Clin Orthop Relat Res. 1986;208:65–68.
23. Kon E, Filardo G, Delcogliano M, et al.. Platelet-rich plasma: new clinical application: a pilot study for treatment of jumper's knee. Injury. 2009;40:598–603.
24. Filardo G, Kon E, Della Villa S, et al.. Use of platelet-rich plasma treatment of refractory jumper's knee. Int Orthop. 2010;34:909–915.
25. Kiter E, Celikbas E, Akkaya S, et al.. Comparison of injection modalities in the treatment of plantar heel pain: a randomized controlled trial
. J Am Podiatr Med Assoc. 2006;96:293–296.
26. De Vos RJ, Weir A, Tol JL, et al.. No effects of PRP on ultrasonographic tendon structure and neovascularisation in chronic midportion achilles tendinopathy. Br J Sports Med. 2011;45:387–392.
27. De Vos RJ, Weir A, van Schie HT, et al.. Platelet-rich plasma injection for chronic achilles tendinopathy: a randomized controlled trial
. JAMA. 2010;303:144–149.
28. Dragoo JL, Wasterlain AS, Braun HJ, et al.. Platelet-rich plasma as a treatment for patellar tendinopathy: a double-blind randomized controlled trial
. Am J Sports Med. 2014;42:610–618.
29. Creaney L, Wallace A, Curtis M, et al.. Growth factor-based therapies provide additional benefit beyond physical therapy in resistant elbow tendinopathy: a prospective, single blind, randomized trial of autologous blood injections versus platelet-rich plasma injections. Br J Sports Med. 2011;45:966–971.
30. Alvarez RH, Kantarjan HM, Cortes JE. Biology of platelet derived growth factor and its involvement in disease. May Clin Proc. 2006;81:1241–1257.