D(KaUtiFthA

a

KN

A Retrospective Comparative Study With Historical Control toDetermine the Effectiveness of Platelet-Rich Plasma as Part ofNonoperative Treatment of Acute Achilles Tendon Rupture

Nicole Kaniki, Ph.D., candidate, M.Sc., A.T.C., Kevin Willits, M.D., M.A., F.R.C.S.C.,Nicholas G. H. Mohtadi, M.D., M.Sc., F.R.C.S.C., Vincent Fung, P.T., A.T.C., and

Dianne Bryant, Ph.D., M.Sc.

Purpose: The aim of this study was to evaluate the effectiveness of platelet-rich plasma (PRP) in the nonoperativetreatment of acute Achilles tendon rupture. Methods: This was a comparative study that included a prospective cohortand a historical control group. The control group was formed from a randomized trial in which one arm of the trialunderwent nonoperative treatment, including accelerated functional rehabilitation after acute Achilles tendon ruptureidentical to that performed in the prospective treatment group. Patients in the prospective group were recruitedconsecutively and were administered 2 injections of PRP during the first 2 weeks after the injury. The primary outcomewas isokinetic plantar flexion strength at 1 and 2 years after injury. Secondary outcomes included range of motion(ROM), calf circumference, and Leppilahti score. The ankle-hindfoot scale (American Orthopedic Functional Ankle Scale[AOFAS]) was administered to patients who received the PRP injection in the prospective group but was not measuredfor the historical group. Results: A total of 73 patients participated in the prospective PRP study group and werecompared with a retrospective control group of 72 patients from a previous randomized controlled trial (RCT). The meandifference between groups in isokinetic plantar flexion strength (injured/uninjured) at 1 year after injury was �4.3%(95% confidence interval [CI], �15.9 to 7.3; P ¼ .5) and 2.4% (95% CI, �8.6 to 13.5; P ¼ .7) at 30�/s and 60�/s,respectively. Results at 2 years after injury were �3.1% (95% CI, �13.5 to 7.2; P ¼ .6) and 4.8% (95% CI, �3.5 to 13.1;P ¼ .3) at 30�/s and 60�/s, respectively. All secondary outcomes were also not statistically different. Conclusions: Theresults of this study suggest that there is no measurable clinical benefit to the addition of PRP to the treatment regimenfor nonoperatively treated acute Achilles tendon rupture. Level of Evidence: Level III, retrospective comparativestudy.

From Faculty of Health Sciences, Health and Rehabilitation Sciences (N.K.,.B.); Department of Surgery, Schulich School of Medicine and Dentistry.W., D.B.); Faculty of Health Sciences, School of Kinesiology (K.W., D.B.);nd Faculty of Health Sciences, School of Physical Therapy (D.B.), Westernniversity, London, Ontario; the Department of Epidemiology and Biosta-stics (D.B.), Faculty of Medicine, McMaster University, Hamilton, Ontario;owler Kennedy Sports Medicine Clinic (K.W., V.F.), London, Ontario; ande University of Calgary Sport Medicine Centre (N.G.H.M.), Calgary,lberta, Canada.The authors report that they have no conflicts of interest in the authorship

nd publication of this article.Received June 4, 2013; accepted April 10, 2014.Address correspondence to Kevin Willits, M.D., M.A., F.R.C.S.C., Fowlerennedy Sports Medicine Clinic, 3M Center, Western University, London, ON6A3K7, Canada. E-mail: kwillits2001@yahoo.com� 2014 by the Arthroscopy Association of North America0749-8063/13370/$36.00http://dx.doi.org/10.1016/j.arthro.2014.04.086

Arthroscopy: The Journal of Arthroscopic and Related

he use of platelet-rich plasma (PRP) in the treat-

Tment of orthopaedic injuries has been widely dis-cussed in recent literature.1-5 PRP is defined as aconcentration of platelets in plasma greater than thebaseline concentration.6 A platelet is a type of whiteblood cell that contains more than 30 bioactive pro-teins, many of which have a fundamental role in he-mostasis or tissue healing.7 The intended purpose ofPRP is to create an improved healing environment byisolating the plasma components from an individual’sblood and injecting them back into the same individualat the site of injury.With respect to studies that have evaluated the use of

PRP for Achilles tendon tears, Sanchez et al.5 publisheda case-control study in which 6 athletes with completeAchilles tendon tears underwent open suture repairaugmented with a preparation rich in platelet-rich

Surgery, Vol -, No - (Month), 2014: pp 1-7 1

Table 1. Inclusion and Exclusion Criteria

Inclusion Criteria Exclusion Criteria

� Complete primary Achillestendon rupture confirmed bya positive Thompson squeezetest and the presence of apalpable gap12,13

� Presentation within 14 d afterinjury

� Between 18 and 70 yr of age� Willing and able to comply

with and carry out theprescribed rehabilitationprotocol

� Provided informed consent� Ability to speak English

� Additional ipsilateral injury� Open injury� Fluoroquinolone-associated

rupture (i.e., rupture within2 wk after taking thismedication)

� Insulin-dependent diabetes� Achilles avulsion from the

calcaneus� Surgical contraindications� Neurologic or vascular disease

requiring medicationsrecognized to impair tendonhealing

2 N. KANIKI ET AL.

growth factor (PRGF). Range of motion, functional re-covery, complications, and cross-sectional area in these6 athletes were compared with a matched group of 6athletes who had the same injury and received thesame treatment with the exception of the PRGF injec-tion. Athletes receiving PRGF recovered their anklerange of motion (ROM) earlier (7 � 2 weeks versus 11� 3 weeks; P ¼ .03) (mean difference ¼ 4; 95% con-fidence interval [CI], 0.72 to 7.28; P ¼ .02), showed nowound complications, and were permitted by theirsurgeon to take up gentle running more quickly (11 �1 weeks versus 18 � 3 weeks; P ¼ .04) (meandifference ¼ 7; 95% CI, 4.12 to 9.88; P < .01), and toresume training activities (14 � 0.8 weeks versus 21 �3 weeks; P < .01) (mean difference ¼ 7.0; 95% CI, 4.18to 9.82; P < .01).Rupture of the Achilles tendon is one of the most

common tendon injuries in the adult population. Theincidence of Achilles tendon injury is increasing as ag-ing adults continue to participate in high-demandsports8-10 such as basketball, racquet sports, and soc-cer. There is debate about the most effective treatmentof Achilles tendon ruptures, some arguing for operativerepair while others argue for a nonoperative approach.In a randomized clinical trial of 144 patients, Willitset al.11 found that strength, ROM, and patient-reportedfunction in 72 patients treated nonoperatively weresimilar to outcomes reported by 72 patients who weretreated with operative repair. In a systematic review ofthe literature, they found similar conclusions wheneverthe nonoperative treatment entailed early mobiliza-tion.11 In addition, nonoperative management reducesthe rate of serious complications from postsurgicalinfection or anesthesia.The aim of this study was to evaluate the effectiveness

of PRP in the nonoperative treatment of acute Achillestendon rupture. We hypothesized that PRP wouldresult in improved isokinetic strength, ROM, calfcircumference measures, and Leppilahti scores at 1 and2 years after injury.

MethodsConsecutive patients were recruited from referrals

made to our clinic from local emergency rooms. Pa-tients were brought in for screening if the referralsuggested an Achilles tendon rupture with diagnosissupported by the presence of a palpable gap,12 a posi-tive Thompson squeeze test,13 and the use of ultraso-nography when available. Additionally, the injury musthave occurred within 14 days of presentation. Once atour clinic, patients were examined by one of us, whoconfirmed study eligibility using the criteria shown inTable 1. The Thompson squeeze test is a valid andreliable evaluation of an acute Achilles tendon rupture.The absence of plantar flexion at the ankle joint whenthe calf is squeezed is considered a positive indication of

rupture.13 This test was also performed for each patienton presentation to our clinic before the injection. Thestudy protocol was approved by the human researchethics board at our institution. All participants providedwritten consent.We compared the prospective PRP group to a group of

patients who participated in a randomized trial wepublished in 2010,11 in which patients who underwentoperative versus nonoperative treatment after acuteAchilles tendon rupture were compared. Because par-ticipants in the randomized controlled trial (RCT) wereassigned to groups in an unbiased fashion (i.e., con-cealed randomization) and the rehabilitation protocolused in the RCT was identical to that planned for thecurrent study, we elected to use the nonoperative groupfrom the RCT as a control group for the current study.Regarding the prospective treatment group, patients

received a PRP injection administered within 2 weeks ofinjury and a second injection 2 weeks later.

PRP Injection for the Prospective GroupApproximately 12 mL of blood was drawn from the

patient’s arm and was then placed in an autologousconditioned plasma double-syringe system (ACP;Arthrex, Naples, FL) and spun in a table-top Rotafix32A centrifuge (Hettich, Tuttlingen, Germany) at1,500 rpm for 5 minutes. This standardized processproduced between 3 and 4 mL of plasma. A localanalgesic, lidocaine (2% concentration), was injectedsuperficially into the area of the palpable gap within theruptured Achilles tendon. This was followed immedi-ately by the PRP injection at the tendon gap deep to thearea of superficial analgesic. Two weeks later, anidentical protocol was followed to administer the sec-ond injection at the same location.

Accelerated Functional Rehabilitation ProtocolThe lower limbs of the patients were placed in a

removable below-knee orthosis (Aircast pneumaticwalking brace; DJO Global, Vista, CA) with a 2-cmheel lift providing approximately 20� of plantar

PLATELET-RICH PLASMA AND ACHILLES TENDON RUPTURE 3

flexion. Patients were instructed to maintain anoneweight-bearing status for the first 2 weeks andpractice protected weight bearing for the next 2 weeks.Patients were allowed to progress to weight bearing astolerated between 4 and 6 weeks. They were given acopy of the standardized rehabilitation protocol and aprescription written by the surgeon to the physio-therapist that outlined milestones and timelines.Therapists could progress through the protocol at theirdiscretion.

Outcome MeasuresThe primary outcome was isokinetic strength for

plantar flexion of the ankle.14 Strength was measuredusing a Biodex Multi-Joint System 3 dynamometer(Biodex Medical Systems, Shirley, NY). The patient waspositioned seated with the hip flexed at 45� and theknee flexed at 30� so that the tibia and fibula wereparallel to the ground. The thigh was supported with athigh rest and strapped into place to prevent its useduring testing. The medial malleolus was aligned withthe dynamometer’s axis of rotation and the foot wasfixed to the dynamometer footplate by 2 Velcro straps(Velcro USA, Manchester, NH) and a modified anklestrap for added ankle stability.For familiarization, patients performed practice ac-

tions at a submaximal strength before actual testing ateach velocity. Participants performed 3 repetitions ofeach test, which included 4 concentric isokinetic con-tractions of plantar flexion and dorsiflexion withmaximum effort at 30�/s and 60�/s and 10 concentricisokinetic contractions at 240�/s. One-minute rest pe-riods were given between each velocity. Testing wasdone bilaterally with the unaffected limb first for all 3velocities, followed by the affected limb. Peak plantarflexion and dorsiflexion torques (N$m) were calculatedby averaging peak torque values of the contractionscollected at each velocity.Secondary outcomes included ROM, calf circumfer-

ence, Leppilahti score, and the American OrthopedicFunctional Ankle Scale (AOFAS). The non-PRP his-torical control group did not complete the AOFASquestionnaire because the RCT began before publica-tion of the validation of this instrument.25 Thus, for thisinstrument we present scores over time for the pro-spective group only.Plantar flexion and dorsiflexion ROM was measured

using a standard long-arm goniometer14 with the pa-tient seated on a table with their legs relaxed over theedge of the examination table so that their knees wereflexed at 90�. The ankle was then placed in a neutralposition of approximately 90�. The axis of rotation wasthe point just inferior to the lateral malleolus, and thelong axis of the fibula was used as the fixed lever arm.The moving lever arm was lined up with the length ofthe fifth metatarsal. The patient was instructed to

actively move into plantar flexion or dorsiflexion, andthe lever arm was moved accordingly to record theamount of movement in degrees in relation to theneutral position.Calf circumference14 was measured with the patient

sitting at the edge of the table with the lower legsrelaxed. A standard tape measure was used to measurethe circumference of the calf in centimeters fromapproximately 15 cm distal to the inferior pole of thepatella, the area of largest muscle bulk in thegastrocnemius.The Leppilahti score16 is a disease-specific functional

outcome measure that includes patient ratings of pain,stiffness, calf muscle weakness, footwear restrictions,ROM, and satisfaction with progress. The isokineticstrength measures used as our primary objective,including plantar flexion and dorsiflexion at speeds of30�/s, 60�/s, and 240�/s, are included in the traditionalscoring of the Leppilahti scale. The minimum score is0 and the maximum score is 100, where 100 is the bestpossible score. We used the Leppilahti scale as anoutcome score even though it is not a validatedoutcome measure for Achilles tendon rupture. How-ever, using this scale was necessary because it was usedin the randomized trial from which we obtained ourcontrol group data.The AOFAS ankle-hindfoot scale is a region-specific,

quality-of-life, and objective functional scale. It assessespain, functional ability during activities of daily living,ROM, stability, and ankle alignment. It is a combinationof subjective grading by the patient and objective eval-uation by the physician. The AOFAS has demonstratedconcurrent validity,17,18 criterion validity,18 adequateconstruct validity,15 and test-retest reliability.18

All measurements were conducted by a certified ath-letic trainer (N.K). Calf circumference, ROM, andAOFAS were measured at 6 weeks, 3 months,6 months, 1 year, 18 months, and 2 years after injection.The Leppilahti scale, which includes isokinetic strengthscores, was measured at 6 months, 1 year, 18 months,and 2 years after injection. Complications were assessedat all follow-up visits and included repeated rupture,superficial and deep infection, venous thrombosis,pulmonary embolus, and numbness.

Statistical AnalysisWe recruited 73 patients in the PRP group to arrive at

a sample size similar to the number of patients in thehistorical control group (n ¼ 72) who were part of theprevious RCT.5 With this sample size, we would have80% power to detect a moderate effect size (0.52) be-tween groups in strength (2-sided test; type I error of0.05).Independent-group t tests were used to make

between-group comparisons, in which the independentvariable was the treatment group and the dependent

Fig 1. Flow diagram of patient participation and availabledata.

Table 2. Demographics

Characteristics PRP (n ¼ 73) Non-PRP (n ¼ 72) P Value

Sex (male:female) 59:14 59:13 .52Age* (yr) 41.5 � 11.1 41.1 � 8.0 .72Height* (cm) 178.6 � 9.2 177.3 � 10.3 .20Weight* (kg) 88.5 � 18.3 88.2 � 17.6 .37Mechanism of Injury(no. of patients)

ADL, 11; sports, 62 ADL, 15; sports, 57 .78

ADL, activities of daily living; PRP, platelet-rich plasma.*Data are given as the mean and standard deviation.

4 N. KANIKI ET AL.

variable was the continuous outcome measure (i.e.,strength, Leppilahti score, ankle ROM, or calf circum-ference) at either 1 or 2 years after treatment. P ¼ .05was considered significant. Outcome data for each groupare presented as the mean and standard deviation withthe mean difference between groups and 95% CI.

Fig 2. (A) Plantar flexion strength comparison at 1 year afterinjury. (B) Plantar flexion strength comparison at 2 years afterinjury. (PRP, platelet-rich plasma.)

ResultsBetween 2008 and 2010, we screened 83 patients for

the prospective group. Of these, 10 were excluded; onerefused the injection and 9 were ineligible (3 werediabetic, 2 were older than 70 years of age, 2 had aninjury that occurred significantly longer than 2 weeksearlier, and 2 had an incomplete rupture). Therefore, atotal of 73 patients were eligible and gave consent(Fig 1).Of the 73 patients included in the prospective PRP

group, 53 had complete data at 1 year, and 59 hadcomplete data at 2 years for inclusion into our analysis.Available data from patients who withdrew because ofreasons unrelated to treatment (n ¼ 2), were lost tofollow-up (n ¼ 10), and experienced a repeated ruptureof the Achilles tendon (n ¼ 2) were also included in theanalysis up to the point of exit from the study (Fig 1).In the prospective sample, there was a total of 59 men

and 14 women with a mean age (SD) of 41.5 �11.1 years (range, 19.0 to 66.0 years) (Table 2). Theaverage number of days from date of injury to first PRPinjection was 8.3 days (range, 2 to 20 days).

Repeated RuptureThere were 2 repeated ruptures in the PRP group and

3 repeated ruptures in the non-PRP group. All repeated

ruptures occurred within the first 3 months of treat-ment. From the PRP group, one patient elected to un-dergo surgical repair and another opted to continuewith the nonoperative protocol. From the non-PRPgroup, 2 patients elected to undergo surgical repairand the other opted to continue with the nonoperativeprotocol.

Isokinetic Plantar Flexion StrengthOn average, at all 3 test velocities (30�/s, 60�/s, and

240�/s), the affected limb in both groups achieved atleast 80% of the plantar flexion strength of the unaf-fected limb. There were no significant differences be-tween groups for isokinetic plantar flexion strength atany of the 3 speeds (Fig 2, A and B).

Table 3. Leppilahti Scores for Groups at 1 and 2 Years

PRP Non-PRP

n Mean � SD n Mean � SD

1 yr 53 81.4 (11.6) 40 79.2 (13.1)2 yr 59 84.2 (10.8) 41 82.2 (12.3)

PRP, platelet-rich plasma; SD, standard deviation.

Table 4. AOFAS Scores for PRP Group at 1 and 2 Years

Item Mean � SD 95% CI

Pain (40 points)1 yr 36.1 (5.3) 34.6-37.62 yr 38.1 (5.9) 36.5-39.8

Function (50 points)1 yr 48.8 (1.6) 48.4-49.32 yr 49.3 (1.5) 48.9-49.8

Alignment (10 points)1 yr 9.7 (1.4) 9.3-10.12 yr 10.0 (0) e

Total score (100 points)1 yr 94.6 (6.4) 92.9-96.42 yr 97.5 (6.3) 95.7-99.2

AOFAS, American Orthopedic Functional Ankle Scale; CI, confi-dence interval; SD, standard deviation.

PLATELET-RICH PLASMA AND ACHILLES TENDON RUPTURE 5

At 1 year, the mean isokinetic plantar flexion strengthratio at 30�/s for the PRP and non-PRP groups was87.9% (�19.6%) and 92.2% (�33.9%), respectively, at1 year and 87.8% (�15.7%) and 91.0% (�35.7%),respectively, at 2 years. The mean between-group dif-ference was �4.3% (95% CI, �15.9 to 7.3; P ¼ .5) at1 year and �3.1% (95% CI, �13.5 to 7.2; P ¼ .6) at2 years.The mean isokinetic plantar flexion strength ratio at

60�/s in the PRP and non-PRP groups was 90.0%(�23.3%) and 87.6% (�25.6%), respectively, at 1 yearand 89.6% (�21.3%) and 84.8% (�21.1%), respec-tively, at 2 years. The mean between-group differencewas 2.4% (95% CI, �8.6 to 13.5; P ¼ .7) at 1 year and4.8% (95% CI, �3.5 to 13.1; P ¼ .3) at 2 years.The mean isokinetic plantar flexion strength ratio at

240�/s in the PRP and non-PRP groups was 84.9%(�17.7%) and 83.7% (�27.1%), respectively, at 1 yearand 86.9% (�18.3%) and 83.3% (�24.3%) at 2 years.The mean between-group difference was 1.2% (95%CI, �8.7 to 11.2; P ¼ .8) at 1 year and 3.6% (95%CI, �4.8 to 12.0; P ¼ .4) at 2 years.

Range of MotionComparison of between-group side-to-side differ-

ences in plantar flexion ROM, in which the uninjuredside presented greater range in both groups, was �0.9�

(�9.3�) for the PRP group and �2.2� (�4.6�) for thehistorical control group at 1 year and 3.2� (�16.4�)and �1.0� (�5.6�), respectively, at 2 years. The meandifference between groups was 1.3� (95% CI, �1.8 to4.5; P ¼ .4) at 1 year, and 4.2� (95% CI, �0.7 to 9.1;P ¼ .1).Comparison of between-group side-to-side differ-

ences in dorsiflexion ROM, in which the uninjured sideagain presented greater range in both groups,was �0.7� (�5.2�) for the PRP group and �1.3� (�2.9�)for the historical control group at 1 year and �0.4�

(�5.6�) and �2.1� (�4.3�), respectively, at 2 years. Themean difference between groups was 0.6� (95%CI, �1.3 to 2.4; P ¼ .5) at 2 year, and 1.7� (95%CI, �0.2 to 3.7; P ¼ .09) at 2 years.

Calf CircumferenceSide-to-side difference in calf circumference in the

PRP and non-PRP group was �1.7 cm (�1.2)and �1.7 cm (�1.9) at 1 year and �1.4 cm (�1.4)and �1.4 cm (�6.1), respectively, at 2 years. The un-injured side had greater calf circumference in bothgroups. The mean difference between groups was 0.03(95% CI, �0.6 to 0.7; P ¼ .9) at 1 year and �0.08 (95%CI, �1.7 to 1.6; P ¼ .9) at 2 years.

Quality-of-Life QuestionnairesThe final scores for the Leppilahti quality-of-life

questionnaire were not significant between groups at

1 and 2 years after injury (Table 3). The mean differ-ence between groups was 2.2 points (95% CI, �2.9 to7.2; P ¼ .4) at 1 year and 2.0 points (95% CI, �2.7 to6.6; P ¼ .4) at 2 years.The ankle-hindfoot AOFAS scores for the PRP group

are shown in Table 4. Higher scores indicate improvedresults. As expected, patients showed the greatestimprovement in the function domain over 1 year.

DiscussionWe found no statistically significant differences for

any of our comparisons between the prospective PRPand historical non-PRP groups for isokinetic strength(30�/sec, 60�/sec, and 240�/sec), calf circumference,ROM (plantar flexion and dorsiflexion), or Leppilahtiscores. These results suggest that there is unlikely to bea large positive effect from PRP in tendon healing andemphasizes the need for higher powered RCTs.The use of PRP in orthopaedic injuries is a relatively

new concept. There is limited literature on its use,specifically in tendon healing. Some in vivo and in vitrocell culture studies6,7,19,20 have shown PRP to be asignificant activator of circulation-derived cells forenhancement of the initial tendon healing process. Thisbenefit has not yet been demonstrated in the clinicalsetting. This study was the first large retrospective studyto compare the use of PRP in the treatment of patientswith acute Achilles tendon ruptures and with acceler-ated functional rehabilitation.The concentration of PRP obtained in our study

through the Arthrex double-syringe system is 2 to

6 N. KANIKI ET AL.

3 times the concentration of platelets in baseline blood.This concentration has been found to be effective inin vitro and in vivo studies in proliferation and angio-genesis of tissue,21,22 despite claims that higher con-centrations may have a greater effect.23,24

There were 2 repeated ruptures in the PRP group and 3in the historical non-PRP group (odds ratio, 0.65; 95%CI,0.11 to 4.00; P¼ .64). Our repeated rupture rate is similarto that observed in studies that report the repeatedrupture rate in operatively treated patients.11,20,25,26 Wedid not observe any major complications in either group.One patient in our prospective group had an unrelatedmyocardial infarction in the first 3 months of inclusionbut was able to complete the study.Similar functional results were found by Saleh et al.,26

Twaddle et al.,27 Gelberman et al.,28 McComis et al.,29

Maffulli et al.,30 and Costa et al.31 These studiesshowed evidence of spontaneous healing of tendonswithout immobilization. There is evidence suggestingthat early functional mobilization may offer patientsgreater functional outcome after Achilles tendonrupture.20,25,27,32

Suggestions for further studies include an RCT, the useof the validated Achilles tendon rupture score, eccentricmeasurement of isokinetic strength, and a more func-tional strength outcome measure such as tiptoe scoring.Other suggestions for future research include the use of aPRP system that yields a higher concentration of plate-lets as well as PRP injections administered closer to thedate of injury during the initial inflammation healingphase, as opposed to the proliferative phase, which mayhave occurred in some of our patients.

LimitationsA limitation to this study is its nonconcurrent non-

randomized design. However, the historical controlgroup that we selected was part of a randomized studythat was led by the same authors. Thus, unlike mostretrospective designs, we are certain that eligibilitycriteria, protocols for measuring outcomes, and con-servative treatment methods were similar betweenstudies. In addition, we are certain that selection bias(perhaps the largest potential source of bias in non-randomized designs) did not occur because the histor-ical control group was part of a randomized trial andboth studies recruited consecutive patients. Theconsecutive recruitment of patients also gives us acertainty that our population is reflective of the type ofpatient with this injury.Another limitation to this study is the limited amount

of patient data available to compare at 1 and 2 yearsafter injury. Specifically, in our prospective treatmentgroup, 29% of our patients had incomplete data at1 year and 20% had incomplete data at 2 years. It ispossible that patients with missing data are differentfrom those with complete data, increasing the potential

for a biased estimate of the treatment effect. However,of the 14 patients with incomplete data at 2 years, 12were missing for seemingly random reasons (2 with-drew consent to participate and 10 were lost to follow-up), which in unlikely to bias the results. There were nostatistically significant differences in age, sex, andweight between patients included (n ¼ 59) and thosenot included (n ¼ 14) in the final analysis (P > .05).Patient adherence to the physiotherapy protocol was

not monitored, and patients were allowed to choosetheir place of treatment. Because the early functionalrehabilitation protocol played an important role in theprogression of the patient, it may have affected theiroutcomes. However, because we are certain that thestudy design, data collection protocols, and rehabilita-tion protocols were the same between our prospectivetreatment group and the historical control group, wehave no reason to believe that the rate or degree ofcompliance would differ between groups.

ConclusionsThe results of this study suggest that there is no

measurable clinical benefit to the addition of PRP to thetreatment regimen for nonoperatively treated acuteAchilles tendon rupture.

References1. Creaney L, Wallace A, Curtis M, Connell D. Growth

factor-based therapies provide additional benefit beyondphysical therapy in resistant elbow tendinopathy: A pro-spective, single-blind, randomised trial of autologousblood injections versus platelet-rich plasma injections. Br JSports Med 2011;45:966-971.

2. De Vos R, Weir A, Van Schie HT. Platelet-rich plasmainjection for chronic Achilles tendinopathy: A randomizedcontrolled trial. JAMA 2010;303:144-149.

3. Gosens T, Peerbooms JC, van Laar W, den Oudsten BL.Ongoing positive effect of platelet-rich plasma versuscorticosteroid injection in lateral epicondylitis: A double-blind randomized controlled trial with 2-year follow-up.Am J Sports Med 2011;39:1200-1208.

4. Mishra A, Pavelko T. Treatment of chronic elbow tendi-nosis with buffered platelet-rich plasma. Am J Sports Med2006;34:1774-1778.

5. Sanchez M, Anitua E, Azofra J, et al. Comparison ofsurgically repaired Achilles tendon tears using platelet-rich fibrin matrices. Am J Sports Med 2007;35:245-251.

6. Kajikawa Y, Morihara T, Sakamoto H, et al. Platelet-richplasma enhances the initial mobilization of circulation-derived cells for tendon healing. J Cell Physiol 2008;215:837-845.

7. Slater M, Patava J, Kingham K, Mason RS. Involvementof platelets in stimulating osteogenic activity. J Orthop Res1995;3:655-663.

8. Jozsa L, Kvist M, Balint BJ, et al. The role of recreationalsport activity in Achilles tendon rupture. A clinical,pathoanatomical, and sociological study of 292 cases. Am JSports Med 1989;17:338-343.

PLATELET-RICH PLASMA AND ACHILLES TENDON RUPTURE 7

9. Leppilahti J, Puranen J, Orava S. Incidence of Achillestendon rupture. Acta Orthop Scand 1996;67:277-279.

10. Moller A, Astron M, Westlin N. Increasing incidence ofAchilles tendon rupture. Acta Orthop Scand 1996;67:479-481.

11. Willits K, Amendola A, Bryant D, et al. Operative versusnonoperative treatment of acute Achilles tendon rup-tures: A multicenter randomized trial using acceleratedfunctional rehabilitation. J Bone Joint Surg Am 2010;92:2767-2775.

12. Cretnik A, Kosanovic M, Smrkolj V. Percutaneous versusopen repair of the ruptured Achilles tendon: A compara-tive study. Am J Sports Med 2005;33:1369-1379.

13. Thompson TC, Doherty JH. Spontaneous rupture oftendon of Achilles: A new clinical diagnostic test. J Trauma1962;2:126-129.

14. Moller M, Lind K, Styf J, Karlsson J. The reliability ofisokinetic testing of the ankle joint and a heel-raise test forendurance. Knee Surg Sports Traumatol Arthrosc 2005;13:60-71.

15. SooHoo NF, Vyas R, Samimi D. Responsiveness of the footfunction index, AOFAS clinical rating systems, and SF-36after foot and ankle surgery. Foot Ankle Int 2006;27:930-934.

16. Leppilahti J, Forsman K, Puranen J, Orava S. Outcomeand prognostic factors of Achilles rupture repair using anew scoring method. Clin Orthop Relat Res 1998;346:152-161.

17. Goldstein CL, Schemitsch E, Bhandari M, et al. Compar-ison of different outcome instruments following foot andankle trauma. Foot Ankle Int 2010;31:1075-1080.

18. Ibrahim T, Beiri A, Azzabi M, et al. Reliability and validityof the subjective component of the American OrthopaedicFoot and Ankle Society clinical rating scales. J Foot AnkleSurg 2007;46:65-74.

19. Aspenberg P, Virchenko O. Platelet concentrate injectionimproves Achilles tendon repair in rats. Acta Orthop Scand2004;75:93-99.

20. Lo IK, Kirkley A, Nonweiler B, Kumbhare DA. Operativeversus nonoperative treatment of acute Achilles tendonruptures: A quantitative review. Clin J Sport Med 1997;7:207-211.

21. Choi BH, Zhu SJ, Kim BY, et al. Effect of platelet-richplasma (PRP) concentration on the viability and

proliferation of alveolar bone cells: An in vitro study. Int JOral Maxillofac Surg 2005;34:420-424.

22. Mastrangelo AN, Vavken P, Fleming BC, et al. Reducedplatelet concentration does not harm PRP effectiveness forACL repair in a porcine in vivo model. J Orthop Res2011;29:1002-1007.

23. Hall MP, Band PA, Meislin RJ, et al. Platelet-rich plasma:Current concepts and application in sports medicine. J AmAcad Orthop Surg 2009;17:602-608.

24. Weibrich G, Hansen T, Kleis W, et al. Effect of plateletconcentration in platelet-rich plasma on peri-implantbone regeneration. Bone 2004;34:665-671.

25. Metz R, Verleisdonk EJ, van der Heijden GJ, et al. AcuteAchilles tendon rupture: Minimally invasive surgeryversus nonoperative treatment with immediate fullweightbearingda randomized controlled trial. Am J SportsMed 2008;36:1688-1694.

26. Saleh M, Marshall PD, Senior R, MacFarlane A. TheSheffield splint for controlled early mobilisation afterrupture of the calcaneal tendon. A prospective, rando-mised comparison with plaster treatment. J Bone Joint SurgBr 1992;74:206-209.

27. Twaddle BC, Poon P. Early motion for Achilles tendonruptures: Is surgery important? A randomized, prospec-tive study. Am J Sports Med 2007;35:2033-2038.

28. Gelberman RH, Woo SL, Amiel D, et al. Influences offlexor sheath continuity and early motion on tendonhealing in dogs. J Hand Surg Am 1990;15:69-77.

29. McComis GP, Nawoczenski DA, DeHaven KE. Functionalbracing for rupture of the Achilles tendon. Clinical resultsand analysis of ground-reaction forces and temporal data.J Bone Joint Surg Am 1997;79:1799-1808.

30. Maffulli N, Tallon C, Wong J, et al. No adverse effect ofearly weight bearing following open repair of acute tearsof the Achilles tendon. J Sports Med Phys Fitness 2003;43:367-379.

31. Costa ML, MacMillan K, Halliday D, et al. Randomisedcontrolled trials of immediate weight-bearing mobilisationfor rupture of the tendo Achillis. J Bone Joint Surg Br2006;88:69-77.

32. Neumayer F, Mouhsine E, Arlettaz Y, et al. A newconservative-dynamic treatment for the acute rupturedAchilles tendon. Arch Orthop Trauma Surg 2010;130:363-368.