Plate Versus Tension-Band Wire Fixation for Olecranon ...files.artikeldatabas.webnode.se/200000027-a210da3093/Olecranonfrakt TBW... · Displaced fracture of the olecranon Minimal

$Page 1: Plate Versus Tension-Band Wire Fixation for Olecranon ...files.artikeldatabas.webnode.se/200000027-a210da3093/Olecranonfrakt TBW... · Displaced fracture of the olecranon Minimal$
Plate Versus Tension-Band Wire Fixation forOlecranon FracturesA Prospective Randomized Trial

Andrew D. Duckworth, BSc, MBChB, MSc, FRCSEd(Tr&Orth), PhD, Nicholas D. Clement, FRCSEd(Tr&Orth), PhD,Timothy O. White, MD, FRCSEd(Tr&Orth), Charles M. Court-Brown, MD, FRCSEd(Orth),

and Margaret M. McQueen, MD, FRCSEd(Orth)

Investigation performed at the Edinburgh Orthopaedic Trauma Unit, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom

Background: The aim of this single-center, single-blinded, prospective randomized trial was to compare the outcomesof tension-band wire (TBW) and plate fixation for simple isolated, displaced fractures of the olecranon.

Methods: We performed a prospective randomized trial involving 67 patients who were ‡16 to <75 years of age and hadan acute isolated, displaced fracture of the olecranon. Patientswere randomized to either TBW (n= 34) or plate fixation (n= 33)and were evaluated at 6 weeks, 3 months, 6 months, and 1 year following surgery. The primary outcomemeasure was theDisabilities of the Arm, Shoulder and Hand (DASH) score at 1 year.

Results: The baseline demographic and fracture characteristics of the 2 groups were comparable, except for age, whichwas lower in the TBWgroup. The 1-year follow-up rate was 85% (n= 57), with 84% (n= 56) completing the DASH. There wasa significant improvement in the DASH score over the 1-year period following surgery (p < 0.001). At 1 year, the DASHscore for the TBW group (12.8) did not differ significantly from that of the plate group (8.5) (p = 0.315). The groups also didnot differ significantly in terms of range of motion, the Broberg and Morrey score, the Mayo Elbow Score, or the DASH at allassessment points over the 1 year (all p ‡ 0.05). Complication rates were significantly higher in the TBW group (63%compared with 38%; p = 0.042), predominantly because of a significantly higher rate ofmetalwork removal in symptomaticpatients (50.0% compared with 22%; p = 0.021). Four infections occurred, all in the plate group (0% versus 13%; p =0.114), as did 3 revision surgeries (0% versus 9.4%; p = 0.238).

Conclusions: Among active patients with a simple isolated, displaced fracture of the olecranon, no difference wasfound between TBW and plate fixation in the patient-reported outcome at 1 year following surgery. The complication ratewas higher following TBW fixation and was due to a higher rate of implant removal in symptomatic patients. However, themore serious complications of infection and the need for revision surgery occurred exclusively following plate fixation inthis trial.

Level of Evidence: Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.

Tension-band wire (TBW) fixation is the most commonlyemployed technique for the treatment of simple isolated,displaced fractures of the olecranon. In contrast, plate

fixation is thought to provide superior fracture reduction andfixation for comminuted, unstable, distal, and/or oblique frac-tures1-9. Despite advocates for alternative surgical techniquesincluding intramedullary nailing10 and suture fixation11, TBWfixation remains the standard management for simple isolated,displaced fractures of the olecranon (Mayo type 2A)12-14.

Retrospective comparative studies of TBWand plate fixation,for both simple and comminuted displaced fractures of the olec-ranon, have noted comparable functional outcomes15-17. It remainsunclear, however, whether the initial higher cost of plate fixation isoffset by the cost associated with the higher rate of TBW constructremoval. A recent Cochrane meta-analysis of 244 surgically man-aged olecranon fractures in 6 randomized controlled trials con-cluded that further work is essential in order to determine theoptimal surgical management of simple isolated fractures of the

Disclosure: The Scottish Orthopaedic Research Trust into Trauma (SORT-IT) research charity supported the running of this trial through its research team;A.D.D. served as a fellow from 2010-2011. The Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article(http://links.lww.com/JBJS/D418).

1261

COPYRIGHT � 2017 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED

J Bone Joint Surg Am. 2017;99:1261-73 d http://dx.doi.org/10.2106/JBJS.16.00773

http://links.lww.com/JBJS/D418

olecranon18. Of the 6 trials included, 2 were quasi-randomizedand only 1 directly compared plate and TBW fixation and it usednonvalidated patient-reported outcome measures in a hetero-geneous group3. Regarding lower-demand elderly patients,retrospective case series have demonstrated good short-termand long-term patient-reported outcomes when employingprimary nonoperative management for simple isolated, dis-placed olecranon fractures19-22.

The aim of the current trial was to determine if any differ-ence exists between TBW and plate fixation with respect to the

primary outcome measure, the Disabilities of the Arm, ShoulderandHand (DASH) score, at 1 year post-surgery for simple isolated,displaced fractures of the olecranon among active patients. Thenull hypothesis was that there would be no difference in functionaloutcome, as measured by the DASH, at 1 year post-surgery.

Materials and Methods

This was a registered single-center, prospective randomized trial includingactive adult patients with an isolated, displaced fracture of the olecranon

(ClincalTrials.gov ID NCT01391936). The study center is a level-I trauma

Fig. 1

Consolidated Standards of Reporting Trials (CONSORT) diagram for the recruitment and flow of participants through the trial.

TABLE I Inclusion and Exclusion Criteria

Inclusion Criteria Exclusion Criteria

Age of ‡16 to <75 yr

Displaced fracture of the olecranon

Minimal or moderate fragmentation of the olecranon

Presentation within 2 weeks of olecranon fracture

Pregnant women with predetermined treatment

Patients unable to give informed consent or unable to comply with follow-up

Associated fractures of the coronoid, radial head, and/or distal aspect of thehumerus

Associated ligamentous injury, dislocation, or subluxation of the elbow

Open fracture of the olecranon

1262

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 99-A d NUMBER 15 d AUGUST 2, 2017PLATE VERSUS TENS ION-BAND WIRE FIXAT ION FOR

OLECRANON FRACTURES

center. Between October 2010 and October 2014, 67 patients who were ‡16to <75 years of age with an acute (within 2 weeks of injury), simple isolated,displaced Mayo type-2A fracture of the olecranon

13,14(OTA/AO fracture

type 21-B1.1)23

were recruited into the study (Fig. 1). The inclusion andexclusion criteria are listed in Table I. Displacement of >2 mm of the articular

surface on standard radiographs was used as the definition of displacement7,24

.The primary outcome measure was the DASH score at 1 year post-surgery

25,26.

The appropriate ethical and clinical trial committees authorized the study.Demographic data were collected at initial presentation (Table II). Pa-

tients were asked to complete a retrospective pre-injury DASH questionnaire

TABLE II Baseline Characteristics of Study Participants by Treatment Group

TBW, N = 34 Plate, N = 33 P Value

Age (yr) 0.028*

Mean and std. dev. 43 ± 16 52 ± 17

Range 19-73 18-74

95% CI 37-49 46-58

Sex (no. [%]) 0.397†

Male 21 (61.8) 17 (51.5)

Female 13 (38.2) 16 (48.5)

Dominant hand (no. [%]) 0.614‡

Left 1 (2.9) 2 (6.1)

Right 33 (97.1) 31 (93.9)

Side of injury (no. [%]) 0.889†

Left 19 (55.9) 19 (57.6)

Right 15 (44.1) 14 (42.4)

Associated injury (no. [%]) 9 (26.5) 7 (21.2) 0.614†

Smoker (no. [%]) 13 (38.2) 12 (36.4) 0.874†

Alcohol consumption§ (no. [%]) 0.708‡

£21 units/wk 30 (90.9) 28 (84.8)

>21 units/wk 3 (9.1) 5 (15.2)

‡1 comorbidity (no. [%]) 17 (50) 23 (69.7) 0.100†

SIMD quintile (no. [%]) 0.099†

1 8 (23.5) 4 (12.1)

2 7 (20.6) 2 (6.1)

3 7 (20.6) 5 (15.2)

4 4 (11.8) 10 (30.3)

5 8 (23.5) 12 (36.4)

ASA grade (no. [%]) 0.149†

1 19 (55.9) 11 (33.3)

2 11 (32.4) 18 (54.5)

3 4 (11.8) 4 (12.1)

Mechanism of injury (no. [%]) 0.609†

Fall from standing height 20 (58.8) 20 (60.6)

Fall from height 3 (8.8) 1 (3.0)

Other 0 (0) 2 (6.1)

Motor vehicle collision 7 (20.6) 6 (18.2)

Sports 2 (5.9) 3 (9.1)

Fight/assault 2 (5.9) 1 (3.0)

Pre-injury DASH§ 0.135#

Mean and std. dev. 1.1 ± 5.5 2.3 ± 6.2

Range 0-31.7 0-30

95% CI 0-3 0-4.5

*Student unpaired t test. †Chi-square test. ‡Fisher exact test. §N = 33 for the TBW group. #Mann-Whitney U-test.

1263



OLECRANON FRACTURES

to establish a baseline27,28

. The Scottish Index of Multiple Deprivation (SIMD)was used to assess socioeconomic deprivation

29.

RandomizationAfter providing informed consent, patients were randomized to undergoeither TBWor plate fixation. This was performed by block randomization(n = 4) using sequential closed opaque envelopes, which were prepared byour statistician and contained a card detailing to which of the 2 groups(TBWor plate) the patient had been randomized. Randomization was ona 1:1 basis.

Radiographic ClassificationAll injuries were assessed and classified at the time of presentation by 1 ofthe authors (A.D.D.) using standard anteroposterior and lateral elbow ra-diographs. The OTA/AO fracture classification and Mayo classification forolecranon fractures were used

13,14,23. Initial radiographs were reviewed to

confirm fracture displacement, the absence of comminution necessitatingplate fixation, and the absence of an associated fracture and/or subluxationor dislocation of the elbow. Fracture displacement was defined as the dis-tance or gap in the articular surface created by the fracture, using thepresentation lateral elbow radiograph

3. Measurements were carried out in a

standardized fashion with a calibrated radiograph (Kodak picture archivingcommunication system).

Management ProtocolThe median time to definitive surgery was 2 days (range, 0 to 14 days) and wascomparable between the 2 groups (p = 0.796). All fractures were operated onunder the supervision of a consultant orthopaedic trauma surgeon using well-established standardized techniques. In 19 (28.4%) of the cases, the consultantwas the primary surgeon; a trauma fellow or senior trainee was the primarysurgeon in the other 48 cases (71.6%). This was not significantly differentbetween the groups (p = 0.152). Because of the proximal nature of the fracture,1 patient in the plate group underwent TBW fixation, and because of theunexpected comminution of the fracture, 1 patient in the TBW group under-went plate fixation (Fig. 1).

Patients were placed in the lateral decubitus position with the arm overa bolster and a tourniquet on the arm. The median tourniquet time was 42minutes (range, 25 to 62 minutes; n = 61). The tourniquet time was not

significantly different between the groups (p = 0.116). Antibiotics (routinely,1.5 g of cefuroxime unless contraindicated) were administered intravenouslyprior to inflation. All operations were performed under image-intensifierguidance.

A posterior longitudinal direct midline skin incisionwas routinely used.Lateral andmedial fasciocutaneous flaps were raised to allow adequate exposureof the fracture site, with the length dependent on the type of fixation being usedand the fracture complexity. The ulnar nerve was not routinely dissected out ortransposed. The triceps tendon was identified proximally inserting into theproximal fracture segment. Subperiosteal dissection was performed in the in-terval between the flexor carpi ulnaris (FCU) and extensor carpi ulnaris (ECU) asnecessary to identify the fracture site and the proximal aspect of the ulna, with theFCU and anconeus elevated as required off the medial and lateral aspects of theulna to allow visualization of the joint and fracture fragments. The fracture wasprepared in a standard fashion and held reduced with a reduction clamp.

A standard TBWtechnique was used throughout the trial and employed2 parallel 1.6-mm Kirschner wires in a longitudinal direction going from the

Fig. 2 Fig. 3

Fig. 2 A lateral radiograph of an olecranon fracture managed with tension-band wire fixation. Fig. 3 A lateral radiograph of an olecranon fracture

managed with plate fixation.

TABLE III Concomitant Injuries and Management (N = 16)

Injury Management

Ipsilateral proximal humeral fracture Nonoperative

Ipsilateral nondisplaced acetabularfracture

Nonoperative

Contralateral distal radial fracture Nonoperative

Contralateral 5th metacarpal fracture Nonoperative

Contralateral 3rd and 4th metatarsalfractures

Nonoperative

Back injury (no fracture) Nonoperative

Minor head injury (n = 7) Nonoperative

Ipsilateral femoral neck fracture Dynamic hip screw

Ipsilateral acromioclavicular jointdislocation

Open stabilization

Ipsilateral open tibial fracture Intramedullary nailing

1264



OLECRANON FRACTURES

proximal fragment of the olecranon into the ulna distally. Care was also taken toensure that the wires were extra-articular, and it was up to the discretion of thesurgeon whether these were placed in the anterior cortex or straight down theshaft of the ulna. If they were placed in the anterior cortex, care was taken toprevent them from penetrating too far anteriorly, with length allowed for thefinal burying of the trimmed wires in the proximal aspect of the ulna. Atransverse tunnel was then placed distally in the ulna using a drill at ;4 cmdistal to the fracture site. A 1.2-mm flexible cerclage wire was then passed pos-terior to the 2 Kirschner wires and through the triceps tendon proximally; the 2ends were then crossed, and 1 endwas placed through the distal tunnel in a figure-of-8 technique. The wire was then tensioned in the standard fashion, and all wireends were trimmed and buried at the conclusion of the procedure (Fig. 2).

For plate fixation, Kirschner wires were sometimes used initially tosupplement the fracture reduction and then removed once stability was

achieved. Once the fracture was reduced, a precontoured, nonlocking dorsalproximal ulnar plate (Zimmer) was applied in a standard fashion. Initially, along longitudinal screw entering the medullary canal distally was placed to holdfracture reduction. Once the construct was stable, distal screws were thenplaced to stabilize the construct (Fig. 3). The median number of screws usedper case was 5 (range, 3 to 7).

Following surgery, patients were routinely immobilized with the use ofeither an above-elbow backslab or a crepe bandage dressing with thick syntheticwool padding for 10 to 14 days. A physiotherapy exercise regime was initiatedfor each patient. The time frame for the initiation of active range of motion andfunctional activities varied among the patients and was individualized ac-cording to surgeon preference as well as patient and fracture characteristics.Although lifting and strengthening were typically allowed at 6 to 8 weeks fol-lowing surgery, this was also individualized to the patient. It is not routine in

TABLE IV Functional, Patient-Reported, and Surgeon-Reported Outcomes at Each Assessment Point After Injury by Treatment Group*

No. of PatientsTBW Plate

TimePoint TBW Plate Outcome

Mean andStd. Dev. Range 95% CI

Mean andStd. Dev. Range 95% CI P Value†

6 wk 29 30

Elbow flexionarc (�)

94 ± 30 45-150 83-106 97 ± 31 10-151 85-108 0.759

Forearm rotationarc (�)

174 ± 19 80-180 167-181 171 ± 20 85-180 164-178 0.511

B & M score 68 ± 13 34-86 63-73 67 ± 14 32-91 62-72 0.932

MES 74 ± 15 25-95 68-79 77 ± 16 30-100 71-83 0.430

DASH 35 ± 23 0-90 27-44 36 ± 19 0.8-89 29-44 0.811

12 wk 29 31


120 ± 24 58-150 111-129 117 ± 21 50-149 109-125 0.566


174 ± 19 100-180 167-182 178 ± 6.5 150-180 175-180 0.332

B & M score 79 ± 13 52-100 74-84 78 ± 15 34-100 73-84 0.902

MES 83 ± 11 65-100 79-87 84 ± 17 30-100 78-90 0.801

DASH 21 ± 22 0-80 13-30 20 ± 16 0-66 14-26 0.831

26 wk 28 28


131 ± 17 90-160 125-138 130 ± 19 80-158 123-138 0.809


178 ± 8.6 135-180 175-181 178 ± 8.6 135-180 175-181 0.951

B & M score 84 ± 13 54-100 79-89 86 ± 12 52-100 82-91 0.572

MES 86 ± 11 65-100 82-91 86 ± 11 65-100 82-91 0.906

28 27 DASH 19.7 ± 20 0-82 11-28 15.9 ± 15 0-54 9.8-22 0.464

52 wk 28 29


137 ± 15 84-160 131-143 131 ± 15 95-158 126-137 0.169


178 ± 10 130-180 174-181 180 ± 2.0 170-180 179-180 0.344

B & M score 89 ± 15 35-100 84-95 95 ± 6.7 78-100 92-97 0.072

MES 90 ± 14 40-100 84-95 96 ± 6.8 85-100 93-98 0.050

28 28 DASH 12.8 ± 20 0-79 5.3-20 8.5 ± 10 0-41 4.4-12.5 0.315

*B & M score = Broberg and Morrey score, MES = Mayo Elbow Score, and DASH = Disabilities of the Arm, Shoulder and Hand. †Student unpairedt test.

1265



OLECRANON FRACTURES

our unit to remove internal fixation following olecranon fracture fixation unlessthe patient is symptomatic.

Outcome AssessmentClinical and functional evaluations were performed prospectively at 2 weeks, 6weeks, 3 months, 6 months, and 1 year post-surgery. Radiographic evaluationswere performed at the same time points, although at 1 year only if clinicallyindicated. Patients were evaluated outside of these times as clinically indicated,with clinical and radiographic assessment performed as required, and this wasrecorded as part of the cost analysis. The development of complications andthe need for subsequent surgeries were also recorded.

A full clinical outcome assessment was completed by a blinded researchphysiotherapist or research fellow not involved in the patient’s management.Range of motion in the unaffected and affected elbows (flexion, exten-sion, supination, and pronation) was measured using a standard full-circlegoniometer. The primary outcome measure was the DASH score 1 year post-surgery

25,26. Secondary outcome measures included surgeon-reported measures,

pain, complications, radiographic assessment, and cost. The surgeon-reportedoutcome measures were the Mayo Elbow Score (MES)

30and the Broberg and

Morrey score31,32

. Because of logistical issues, 3 patients (all in the plate group)had their final 1-year outcome evaluation performed over the telephone. Thereare data to validate the verbal administration of the QuickDASH (an abbreviatedversion of the DASH questionnaire), with written scores correlating well withverbal scores

33. For these patients, the most recent range-of-motion measure-

ments were carried forward, as all had regained full elbow motion by 6 months.Complications were defined as the loss of fracture reduction, prominent

and problematic metalwork, further surgery including the removal of metal-

work, superficial or deep wound infections, and the onset of symptoms and/orsigns of neurological impairment following surgery. Removal of internal fixa-tion would routinely be due to pain or loss of function secondary to the device,with or without loss of position and/or prominence. Superficial infection wasdefined as a wound infection that was treated successfully with antibiotics andrequired no surgical intervention. Deep infection fulfilled the criteria as setout by Horan et al.

34.

Radiographic assessment (anteroposterior and lateral views of the el-bow) determined the quality of initial reduction, metalwork failure, and loss ofreduction as well as progression to union. The quality of reduction was con-sidered satisfactory if the articular surface was reduced to within 2 mm, with noevidence of a step-off or gap of >2 mm3. Fracture union was defined as en-dosteal healing, with ‡75% of organized trabecular bridging at the fracturesite

35,36.

Statistical AnalysisA power analysis was performed to determine the requirements for detecting aclinically relevant mean difference of 10 points (DASH) between the 2 groups at1 year after surgery

26,37-40. This indicated that a total sample size of 50 patients

(25 per group) would provide 80% power to detect a significant difference (a =

0.05) in the DASH score, assuming an effect size of 0.8 (mean difference [andstandard deviation] of 10 ± 12

26,37-40) using an unpaired t test. To account for

a possible loss to follow-up of 25%, the intention was to enroll 35 subjects ineach arm (total sample size of 70).

Data were analyzed using the intention-to-treat principle. Chi-square(all numbers in cell ‡5) or Fisher exact (1 cell <5) tests were used to analyzebinary variables. A Student t test (2 groups) or analysis of variance (ANOVA;

Fig. 4

Mean DASH score, and 95% confidence interval, at each assessment point according to randomization group.

1266



OLECRANON FRACTURES

multiple groups) was used to assess linear variables between groups forparametric data. A Mann-Whitney U-test (2 groups) or Kruskal-Wallis test(multiple groups) was used to assess linear variables between groups fornonparametric data. Linear regression was used to analyze the relationshipbetween 2 continuous variables. Multivariate regression analysis was used toassess the independent effect of the surgical group on the DASH whencontrolling for confounding variables. Two-tailed p values are reported, andsignificance was defined as p < 0.05, with 95% confidence intervals (CIs)presented.

Cost AnalysisStandardized costs were used with regard to the total number of days in thehospital, the cost of treatment, clinical evaluation appointments attended,and any complications including subsequent surgeries and antibiotics forinfection. Social and productivity costs were not evaluated. The standardizedcosts were obtained from the National Health Service Lothian. Antibioticcosts were calculated from the current edition of the British NationalFormulary

41.

Results

Sixty-seven patients were randomized to TBW (n = 34) orplate (n = 33) fixation (Fig. 1, Table II). The overall mean

age was 47 ± 17 years (range, 18 to 74 years). Thirty-eight(57%) of the patients were male and 29 (43%) were female(p = 0.272). One or more comorbidities were documented in

40 (60%) of the patients, and a majority of the patients wereclassified as ASA (American Society of Anesthesiologists)grade 1 (n = 30) or 2 (n = 29). The most frequent mechanismof injury was a fall from standing height (n = 40). All fractureswere classified radiographically as Mayo type 2A (OTA/AO21-B1.1). Three fractures had notable articular comminutionat the time of surgery: 2 randomized to plate fixation and 1 toTBW fixation. The mean fracture displacement was 13 mm(range, 4 to 32 mm). Sixteen (24%) of the patients had con-comitant injuries (Table III).

The baseline demographic and fracture characteristics ofthe 2 treatment groups are shown in Table II. Patients in theTBW group were younger (43 versus 52 years; p = 0.028), butall other characteristics were comparable.

Primary OutcomeFifty-seven (85%) of the patients were evaluated at 1 year,with 56 (84%) of the patients having completed a validDASH questionnaire. At 1 year following surgery, the overallmean DASH was 10.6 ± 16 (range, 0 to 79; n = 56). Overall,among patients who completed a valid DASH at both 6 weeksand 1 year (n = 53), scores improved significantly (mean score of35.2 at 6 weeks and 9.2 at 1 year; p < 0.001) (Table IV, Fig. 4).

Fig. 5

Mean elbow flexion arc (in degrees), and 95% confidence interval, at each assessment point according to randomization group.

1267



OLECRANON FRACTURES

There was no difference between the groups in terms of theDASH at any assessment point following surgery (p ‡ 0.05)(Table IV, Fig. 4). ThemeanDASH score at 1 year was 12.8 in theTBW group and 8.5 in the plate group (p = 0.315).

Secondary OutcomesSurgeon-Reported and Functional OutcomesAt 1 year following surgery, the overall mean Broberg andMorrey score was 92 (range, 35 to 100; n = 57), with 89% ofthe patients achieving an excellent (n = 30) or good (n = 21)outcome. Five patients had a fair outcome and 1, a pooroutcome. At 1 year, the overall mean MES was 93 (range, 40 to100; n = 57), with 93% achieving an excellent (n = 36) or good(n = 17) outcome. Three patients had a fair outcome and 1,poor. The mean elbow flexion arc at 1 year was 134� ± 15�(range, 84� to 160�), and the mean forearm rotation arc was179� ± 6.8� (range, 130� to 180�). The 2 treatment groups didnot differ significantly in terms of range of motion (Fig. 5), theBroberg and Morrey score (Fig. 6), or the MES (Fig. 7) at anyassessment point following surgery (all p ‡ 0.05) (Table IV).

ComplicationsComplications were assessed in 62 patients (Table V). Therewere 41 complications in 31 (50%) of the patients, including

removal of metalwork in symptomatic patients (n = 22,35.5%), loss of fracture reduction (n = 12, 19.4%), infection(n = 4, 6.5%), and the need for revision surgery (n = 3,4.8%).

Complications were significantly higher in the TBWgroup (63% compared with 38%; p = 0.042), predominantlybecause of a significantly higher rate of removal of metalworkin symptomatic patients (50% compared with 22%; p = 0.021).One patient in the TBW group underwent an early manipu-lation under anesthesia (MUA) for stiffness and subsequentlyunderwent implant removal following union. Loss of reductionwas twice as common in the TBW group (27% compared with13%), although this difference between the groups was notstatistically significant (p = 0.206).

Four infections occurred in the plate group (13% com-pared with 0% in the TBW group; p = 0.114), as did 3 revisionsurgeries (9.4% compared with 0%; p = 0.238). Of the 4 in-fections, 2 were superficial (resolvedwith oral antibiotic therapy)and 2 were deep. One deep infection was managed successfullywith antibiotics and plate removal once the fracture had healed,and the other resulted in revision to a TBW construct. In thatcase, the infection was successfully treated following a prolongedcourse of antibiotics and further surgery to remove the TBWconstruct, but a fibrous nonunion developed. Two other revision

Fig. 6

Mean Broberg and Morrey score, and 95% confidence interval, at each assessment point according to randomization group.

1268



OLECRANON FRACTURES

surgeries were in the plate group: 1 failed fixation that wassuccessfully converted to a TBW construct, and the other, anexchange of a long screw that was blocking forearm rotation.

Radiographic OutcomesIn the 57 of 62 patients who progressed to radiographic union,the median time to union was 12 weeks (range, 6 to 52 weeks).Three patients in the plate group and 2 patients in the TBWgroupprogressed to a functional fibrous nonunion (Fig. 8). Thesepatients had a loss of reduction, 2 associated with infection.

Cost AnalysisCosts were assessed for 62 patients. The median number ofdays in the hospital was 2 (range, 1 to 38). The overall mediancost per patient was £5,529 (UK) (range, £2,961 to £27,936;interquartile range [IQR], £3,487 to £6,224) or $8,349 (US)(range, $4,471 to $42,183; IQR, $5,265 to $9,398). No signifi-cant difference between the 2 groups was found in the cost perpatient (p = 0.131) (Table VI), despite a significantly highermedian cost for the primary intervention in the plate group(p < 0.001).

Fig. 7

Mean Mayo Elbow Score, and 95% confidence interval, at each assessment point according to randomization group.

TABLE V Complications within 1 Year Following Injury by Treatment Group

TBW, N = 30* Plate, N = 32* P Value

Total complications 19 (63.3) 12 (37.5) 0.042†

Infection 0 (0) 4 (12.5) 0.114‡

Loss of reduction 8 (26.7) 4 (12.5) 0.206‡

Subsequent surgeries

Removal of metalwork 15 (50) 7 (21.9) 0.021†

Revision 0 (0) 3 (9.4) 0.238‡

*The values are given as the number of patients, with the percentage of the group in parentheses. †Chi-square test. ‡Fisher exact test.

1269



OLECRANON FRACTURES

Predictors of Primary Outcome MeasureComorbidities (p = 0.023) and increasing ASA grade (p <0.001) were predictive of the 1-year DASH score. There wasalso an association with fracture nonunion (score of 34 forthose who developed a nonunion versus 9 for those demon-strating union; p = 0.002) and a moderate correlation with the

pre-injury DASH score (p < 0.001; r = 0.45). On multivari-ate linear regression analysis, controlling for age, sex, SIMDquintile, comorbidities, and ASA grade, the treatment armwas not predictive of the 1-year DASH, with increasing ASAgrade being the only independent predictor of a poorer out-come (Table VII). A post-hoc power analysis of the regressionmodel for predictors of outcome as assessed by the DASHscore at 1 year, using an R2 value of 0.273, a sample size of56, 6 predictors in the model, and an a value of 0.05, of-fered 91% power.

Discussion

To our knowledge, this is the first prospective randomizedtrial comparing TBW and plate fixation for isolated, dis-

placed fractures of the olecranon using location-specific plates,strict inclusion and exclusion criteria, and a validated patient-reported outcomemeasure. The data demonstrated that TBWandplate fixation provide comparable patient and surgeon-reportedoutcomes in the year following surgery. Although the overallcomplication rate was higher following TBW fixation, becauseof an increased rate of symptomatic implant removal, the moreserious complications of infection and revision surgery wereexclusive to the plate group.

Hume and Wiss performed the only other prospectiverandomized trial that we are aware of in the literature com-paring TBW (n = 19) and plate fixation (n = 22) for displacedolecranon fractures3. Comminuted and open fractures wereincluded, and no validated patient-reported outcome mea-sures were used. Despite this, the results of their study were

Fig. 8

Fibrous nonunion associated with infection following plate fixation.

TABLE VI Cost Analysis by Treatment Group*

TBW, N = 30 Plate, N = 32 P Value†

Median no. of days inhospital (range) [IQR]‡

2 (1-38) [1-3] 2 (1-30) [1-5] 0.446

Median cost of primaryintervention (range) [IQR]§ (£)

32 (—) [—] 552 (32-563) [547-558] <0.001

Median no. of clinicalevaluations (range) [IQR]#

6 (5-8) [5-6] 5 (5-12) [5-6] 0.610

Median cost of antibiotics(range) [IQR] (£)

0 0 (0-141) [0-0] 0.047

Total no. of extra trips tooperating room

16 10

Median cost of furtherimplants (range) [IQR] (£)

0 (—)[—] 0 (0-32) [0-0] 0.088

Overall median cost/patient(range) [IQR] (£)

5,546 (2,961-27,936) [2,961-5,654] 5,174 (3,476-23,056) [3,492-6,828] 0.131

Overall median cost/patient(range) [IQR] ($)

8,374 (4,471-42,183) [4,471-8,538] 7,812 (5,249-34,815) [5,273-10,310]

*Exchange rate of £1 = $1.51. †Mann-Whitney U-test. ‡Inpatient stay was £675/day ($1,021/day). §TBW construct: £12.24 ($18.53);plate: £505.60 ($765.48) and £5.36 ($8.12) per screw; and above-elbow backslab/cast: £20 ($30.28). Does not include a standardoperating room cost of £1,824 ($2,761), which was included in the overall cost per patient #Each surgical outpatient consultation was£86 ($130).

1270



OLECRANON FRACTURES

comparable with the data presented here. Those authorsreported that elbow motion at 6 months was comparablebetween treatment groups, but with loss of fracture reductionand prominent metalwork in symptomatic patients morecommon following TBW, as our trial also reported. Humeand Wiss found that the overall clinical outcome was far su-perior in the plate-fixation group, with 86% obtaining a goodresult compared with 47% in the TBW group3. In contrast,our trial demonstrated no significant difference in outcomeat 1 year, but there was a trend toward superior outcomes inthe plating group for patient and surgeon-reported outcomemeasures.

The main perceived complication of plate fixation isprominent hardware, given the position of the plate on thedorsal aspect of the ulna42. However, the literature quotes lowerrates of plate removal (5% to 20%) when compared with re-moval following TBW3,5,7,16, which is consistent with our trial(22%). In the Hume and Wiss study 3, prominent or prob-lematic metalwork was seen in 42% of patients who underwentTBW compared with 5% in the plate group. Our trial found anequivalent rate for TBW, but a higher rate for plate fixation thatis more in keeping with current data3,5,7,16. Interestingly, allother complications occurred in the TBW group in the originaltrial by Hume and Wiss3, which is not consistent with our trial,in which infection and revision surgery occurred exclusivelyin the plate group.

Retrospective studies comparing TBW with plate fixa-tion15-17 have consistently reported comparable functionaloutcomes, with a higher rate of metalwork removal for TBWfixation and increased costs with plate fixation. We have dem-onstrated comparable overall costs for TBWand plate fixationdue to the much higher rate of implant removal in the TBWgroup. In contrast, Amini et al.15 retrospectively comparedTBW (n = 10) and plate (n = 10) fixation for simple isolated,transverse olecranon fractures and found that overall costswere significantly higher for plate fixation ($6,598.36 versus$14,333.46), despite a higher rate of metalwork removal forTBW (40% versus 10%). This discrepancy is likely due tosmaller patient numbers, no consideration of other costs (e.g.,length of stay), and most importantly, the higher implant cost

for plates compared with that in our trial ($6,688.52 versus$836.72 in our trial).

The rate of loss of reduction in the TBW group isconsistent with both the limited available clinical data,ranging from 33% to 53%3,43,44, as well as biomechanical ev-idence that questions the validity of this construct to maintainfracture stability and reduction. Wilson et al. performed abiomechanical comparison of TBW and plate fixation in 20models with identical transverse fractures of the olecranon45.They found that the modern precontoured location-specificplates were significantly better at providing fracture com-pression, particularly at the articular surface. However, wefound that the loss of reduction does not seem to influencepatient-reported outcome at 1 year following surgery, pro-vided the patient progresses to union. Nonunion was asso-ciated with an inferior outcome in our trial of younger activepatients. There is a growing body of evidence to support therole of nonoperative management for displaced olecranonfractures in low-demand elderly patients19-21. Nonunion is com-mon in these cases, but with a negligible re-intervention rateand no correlation found with the long-term functional out-comes reported22.

The primary strengths of this trial are the large numberof patients recruited, a good level of compliance with >90% ofpatients receiving their allocated treatment, and the highfollow-up rate at 1 year. The numbers recruited in each groupwere greater than required according to our initial powercalculation. Although multiple surgeons of different experi-ence levels were involved in the surgery for these patients, thisscenario is most representative of day-to-day clinical practice,i.e., pragmatic.

A primary limitation of the study was the lack ofblinding of both the surgeon and the patient to the allocatedtreatment arm. It is argued that this is pragmatic in that, inroutine clinical practice, patients would always be aware oftheir proposed treatment46,47. The study was also limited bythe fact that multiple surgeons were involved over the studyperiod and that patients with concomitant injuries were in-cluded, although again, this is pragmatic and reflective of day-to-day clinical practice. We acknowledge the subjective natureof the decision to remove an implant, and it could be arguedthat these findings, including in relation to cost, are mostapplicable to our center. However, it is not routine in our unitto remove implants following olecranon fracture fixationunless the patient is symptomatic, and the rate of metalworkremoval in our series is very consistent with that in the currentavailable literature.

A further limitation to acknowledge is the differencefound in age between the 2 groups despite randomization, withthe TBW group being 9 years younger, on average, than theplate group. A superior (lower) DASH score was found in theplate group at 1 year, and it is possible that a significant dif-ference would have been apparent if the age of the groups weremore comparable. However, when controlling for age andother confounding factors on multivariate linear regressionanalysis, no difference was found between the 2 groups in the

TABLE VII Multivariate Linear Regression Analysis*

VariableRegressionCoefficient 95% CI P Value

Age 20.09 20.39 to 0.20 0.528

Sex 2.25 26.61 to 11.12 0.612

SIMD quintile 20.68 23.83 to 2.47 0.667

Comorbidities 22.96 215.50 to 9.59 0.638

ASA grade 14.15 4.20 to 24.09 0.006

Management 26.27 214.40 to 1.86 0.128

*To identify significant predictors of the 1-year DASH score whencontrolling for confounding factors. R2 = 0.273.

1271



OLECRANON FRACTURES

primary outcome measure. It is important to note that thestudy was not originally powered for a multivariate regressionanalysis and that this was a post-hoc analysis performed due tothe unexpected age discrepancy found between the groups.

In conclusion, the findings of this trial demonstrate noclear patient or surgeon-reported benefit of plate fixation overthe current gold standard of TBW in active patients with asimple isolated, displaced fracture of the olecranon. The moreserious complications of infection and revision surgery occurredexclusively following plate fixation in this study. However, webelieve that patients should be counseled regarding the highrate of metalwork removal in symptomatic patients followingthe TBW technique. nNOTE: The authors acknowledge the Scottish Orthopaedic Research Trust into Trauma (SORT-IT) forassistance in performing this study. The authors also thank the trauma consultants and registrarsof the Edinburgh Orthopaedic Trauma Unit for their invaluable assistance and endless patience incarrying out this trial; Mr. Paul Jenkins and Mr. Iain Murray for their production assistance for some

of the figures in this manuscript; and Dr. Rob Elton for statistical assistance with setting up thestudy, including preparing the randomization sequence.

Andrew D. Duckworth, BSc, MBChB, MSc, FRCSEd(Tr&Orth), PhD1

Nicholas D. Clement, FRCSEd(Tr&Orth), PhD1

Timothy O. White, MD, FRCSEd(Tr&Orth)1

Charles M. Court-Brown, MD, FRCSEd(Orth)1

Margaret M. McQueen, MD, FRCSEd(Orth)1

1Edinburgh Orthopaedic Trauma Unit, Royal Infirmary of Edinburgh,Edinburgh, United Kingdom

E-mail address for A.D. Duckworth: [email protected]

ORCID iD for A.D. Duckworth: 0000-0002-5317-1300

References

1. Horne JG, Tanzer TL. Olecranon fractures: a review of 100 cases. J Trauma. 1981Jun;21(6):469-72.2. Fyfe IS, Mossad MM, Holdsworth BJ. Methods of fixation of olecranon frac-tures. An experimental mechanical study. J Bone Joint Surg Br. 1985 May;67(3):367-72.3. Hume MC, Wiss DA. Olecranon fractures. A clinical and radiographic comparisonof tension band wiring and plate fixation. Clin Orthop Relat Res. 1992 Dec;(285):229-35.4. Hak DJ, Golladay GJ. Olecranon fractures: treatment options. J Am Acad OrthopSurg. 2000 Jul-Aug;8(4):266-75.5. Bailey CS, MacDermid J, Patterson SD, King GJ. Outcome of plate fixation ofolecranon fractures. J Orthop Trauma. 2001 Nov;15(8):542-8.6. Buijze G, Kloen P. Clinical evaluation of locking compression plate fixationfor comminuted olecranon fractures. J Bone Joint Surg Am. 2009 Oct;91(10):2416-20.7. Newman SD, Mauffrey C, Krikler S. Olecranon fractures. Injury. 2009 Jun;40(6):575-81. Epub 2009 Apr 23.8. Buijze GA, Blankevoort L, Tuijthof GJ, Sierevelt IN, Kloen P. Biomechanical eval-uation of fixation of comminuted olecranon fractures: one-third tubular versus lock-ing compression plating. Arch Orthop Trauma Surg. 2010 Apr;130(4):459-64. Epub2009 Oct 13.9. Erturer RE, Sever C, Sonmez MM, Ozcelik IB, Akman S, Ozturk I. Results of openreduction and plate osteosynthesis in comminuted fracture of the olecranon.J Shoulder Elbow Surg. 2011 Apr;20(3):449-54.10. Nowak TE, Burkhart KJ, Mueller LP, Mattyasovszky SG, Andres T, Sternstein W,Rommens PM. New intramedullary locking nail for olecranon fracture fixation—anin vitro biomechanical comparison with tension band wiring. J Trauma. 2010 Nov;69(5):E56-61.11. Bateman DK, Barlow JD, VanBeek C, Abboud JA. Suture anchor fixation ofdisplaced olecranon fractures in the elderly: a case series and surgical technique.J Shoulder Elbow Surg. 2015 Jul;24(7):1090-7. Epub 2015 Apr 1.12. Chalidis BE, Sachinis NC, Samoladas EP, Dimitriou CG, Pournaras JD. Is tensionband wiring technique the “gold standard” for the treatment of olecranon fractures?A long term functional outcome study. J Orthop Surg Res. 2008 Feb 22;3:9.13. Ring D. Elbow fractures and dislocations. In: Bucholz RW, Court-Brown CM,Heckman JD, Tornetta P III, editors. Rockwood and Green’s fractures in adults. 7thed. Philadelphia: Lippincott Williams & Wilkins; 2010. p 905-44.14. Morrey BF. Current concepts in the treatment of fractures of the radial head, theolecranon, and the coronoid. Instr Course Lect. 1995;44:175-85.15. Amini MH, Azar FM, Wilson BR, Smith RA, Mauck BM, Throckmorton TW.Comparison of outcomes and costs of tension-band and locking-plate osteosyn-thesis in transverse olecranon fractures: a matched-cohort study. Am J Orthop (BelleMead NJ). 2015 Jul;44(7):E211-5.16. Tarallo L, Mugnai R, Adani R, Capra F, Zambianchi F, Catani F. Simple andcomminuted displaced olecranon fractures: a clinical comparison between tensionband wiring and plate fixation techniques. Arch Orthop Trauma Surg. 2014 Aug;134(8):1107-14. Epub 2014 Jun 17.17. Schliemann B, Raschke MJ, Groene P, Weimann A, Wahnert D, Lenschow S,Kosters C. Comparison of tension band wiring and precontoured locking compres-sion plate fixation in Mayo type IIA olecranon fractures. Acta Orthop Belg. 2014Mar;80(1):106-11.

18. Matar HE, Ali AA, Buckley S, Garlick NI, Atkinson HD. Surgical interventions fortreating fractures of the olecranon in adults. Cochrane Database Syst Rev. 2014 Nov26;11(11):CD010144.19. Veras Del Monte L, Sirera Vercher M, Busquets Net R, Castellanos Robles J,Carrera Calderer L, Mir Bullo X. Conservative treatment of displaced fractures of theolecranon in the elderly. Injury. 1999 Mar;30(2):105-10.20. Parker MJ, Richmond PW, Andrew TA, Bewes PC. A review of displacedolecranon fractures treated conservatively. J R Coll Surg Edinb. 1990 Dec;35(6):392-4.21. Gallucci GL, Piuzzi NS, Slullitel PA, Boretto JG, Alfie VA, Donndorff A, De Carli P.Non-surgical functional treatment for displaced olecranon fractures in the elderly.Bone Joint J. 2014 Apr;96-B(4):530-4.22. Duckworth AD, Bugler KE, Clement ND, Court-Brown CM, McQueen MM. Non-operative management of displaced olecranon fractures in low-demand elderly pa-tients. J Bone Joint Surg Am. 2014 Jan 1;96(1):67-72.23. Marsh JL, Slongo TF, Agel J, Broderick JS, Creevey W, DeCoster TA, ProkuskiL, Sirkin MS, Ziran B, Henley B, Audige L. Fracture and dislocation classificationcompendium - 2007: Orthopaedic Trauma Association classification,database and outcomes committee. J Orthop Trauma. 2007 Nov-Dec;21(10)(Suppl):S1-133.24. Veillette CJ, Steinmann SP. Olecranon fractures [vii.]. Orthop Clin North Am.2008 Apr;39(2):229-36: vii.25. Hudak PL, Amadio PC, Bombardier C; The Upper Extremity CollaborativeGroup (UECG). Development of an upper extremity outcome measure: the DASH(disabilities of the arm, shoulder and hand) [corrected]. Am J Ind Med. 1996 Jun;29(6):602-8.26. Gummesson C, Atroshi I, Ekdahl C. The disabilities of the arm, shoulder andhand (DASH) outcome questionnaire: longitudinal construct validity and measuringself-rated health change after surgery. BMC Musculoskelet Disord. 2003 Jun16;4:11. Epub 2003 Jun 16.27. Reynolds N, Thirkannad S. The recall DASH score—a novel research tool. HandSurg. 2013;18(1):11-4.28. Stepan JG, London DA, Boyer MI, Calfee RP. Accuracy of patient recall of handand elbow disability on the QuickDASH questionnaire over a two-year period. J BoneJoint Surg Am. 2013 Nov 20;95(22):e176.29. The Scottish Government. Scottish Index of Multiple Deprivation: 2009 generalreport. 2009 Oct. http://www.gov.scot/Resource/Doc/289599/0088642.pdf.Accessed 2017 Feb 22.30. Morrey BF, An K, Chao E. Functional evaluation of the elbow. In: Morrey BF, editor.The elbow and its disorders. 2nd ed. Philadelphia: WB Saunders; 1993. p 86-9.31. Broberg MA, Morrey BF. Results of delayed excision of the radial head afterfracture. J Bone Joint Surg Am. 1986 Jun;68(5):669-74.32. Broberg MA, Morrey BF. Results of treatment of fracture-dislocations of theelbow. Clin Orthop Relat Res. 1987 Mar;(216):109-19.33. London DA, Stepan JG, Boyer MI, Calfee RP. Performance characteristicsof the verbal QuickDASH. J Hand Surg Am. 2014 Jan;39(1):100-7. Epub 2013Nov 21.34. Horan TC, Gaynes RP, Martone WJ, Jarvis WR, Emori TG. CDC definitions of noso-comial surgical site infections, 1992: amodification of CDC definitions of surgical woundinfections. Infect Control Hosp Epidemiol. 1992 Oct;13(10):606-8.

1272



OLECRANON FRACTURES

http://orcid.org/0000-0002-5317-1300

http://www.gov.scot/Resource/Doc/289599/0088642.pdf

35. Uhthoff HK, Rahn BA. Healing patterns of metaphyseal fractures. Clin OrthopRelat Res. 1981 Oct;(160):295-303.36. Kristiansen TK, Ryaby JP, McCabe J, Frey JJ, Roe LR. Accelerated healing ofdistal radial fractures with the use of specific, low-intensity ultrasound. A multicen-ter, prospective, randomized, double-blind, placebo-controlled study. J Bone JointSurg Am. 1997 Jul;79(7):961-73.37. Hunsaker FG, Cioffi DA, Amadio PC, Wright JG, Caughlin B. The Americanacademy of orthopaedic surgeons outcomes instruments: normative values fromthe general population. J Bone Joint Surg Am. 2002 Feb;84-A(2):208-15.38. Smith MV, Calfee RP, Baumgarten KM, Brophy RH, Wright RW. Upper extremity-specific measures of disability and outcomes in orthopaedic surgery. J Bone JointSurg Am. 2012 Feb 1;94(3):277-85.39. Sorensen AA, Howard D, Tan WH, Ketchersid J, Calfee RP. Minimal clinicallyimportant differences of 3 patient-rated outcomes instruments. J Hand Surg Am.2013 Apr;38(4):641-9. Epub 2013 Mar 6.40. The DASH Outcome Measure. Frequently asked questions (FAQ). http://dash.iwh.on.ca/faq. Accessed 2017 Feb 22.41. British National Formulary. BNF Publications. http://www.bnf.org/. Accessed2017 Feb 22.

42. Gordon MJ, Budoff JE, Yeh ML, Luo ZP, Noble PC. Comminuted olecranonfractures: a comparison of plating methods. J Shoulder Elbow Surg. 2006 Jan-Feb;15(1):94-9.43. Karlsson MK, Hasserius R, Besjakov J, Karlsson C, Josefsson PO. Comparisonof tension-band and figure-of-eight wiring techniques for treatment of olecranonfractures. J Shoulder Elbow Surg. 2002 Jul-Aug;11(4):377-82.44. van der Linden SC, van Kampen A, Jaarsma RL. K-wire position in tension-bandwiring technique affects stability of wires and long-term outcome in surgical treat-ment of olecranon fractures. J Shoulder Elbow Surg. 2012 Mar;21(3):405-11. Epub2011 Oct 29.45. Wilson J, Bajwa A, Kamath V, Rangan A. Biomechanical comparison of inter-fragmentary compression in transverse fractures of the olecranon. J Bone Joint SurgBr. 2011 Feb;93(2):245-50.46. Costa ML, Achten J, Parsons NR, Rangan A, Griffin D, Tubeuf S, Lamb SE;DRAFFT Study Group. Percutaneous fixation with Kirschner wires versus volar lockingplate fixation in adults with dorsally displaced fracture of distal radius: randomisedcontrolled trial. BMJ. 2014 Aug 5;349:g4807.47. Schwartz D, Lellouch J. Explanatory and pragmatic attitudes in therapeuticaltrials. J Clin Epidemiol. 2009 May;62(5):499-505.

1273



OLECRANON FRACTURES

http://dash.iwh.on.ca/faq

http://dash.iwh.on.ca/faq

http://www.bnf.org/

Documents

Plate Versus Tension-Band Wire Fixation for Olecranon ...files.artikeldatabas.webnode.se/200000027-a210da3093/Olecranonfrakt TBW... · Displaced fracture of the olecranon Minimal