t

SCIENTIFIC ARTICLE

Pyrocarbon Proximal Interphalangeal Joint

Arthroplasty: Minimum Two-Year Follow-Up

A. C.Watts, MBBS, BSc, A. J. Hearnden, MBBS, BSc, I. A. Trail, MD, M. J. Hayton, BSc, D. Nuttall, PhD,J. K. Stanley, MBBS

Purpose To report the outcome and complications from pyrocarbon proximal interphalangeal(PIP) joint arthroplasty at a minimum of 2 years of follow-up.

Methods A retrospective case review was performed on 72 patients with an average age of 57years, and a total of 97 pyrocarbon PIP joint arthroplasties. Patient demographics, diagnosis,implant revisions, and other repeat surgeries were recorded. Subjective outcome was eval-uated at latest follow-up with the Disabilities of the Arm, Shoulder, and Hand score; PatientEvaluation Measure; and visual analog scores of pain, satisfaction, and appearance. Objec-tive outcomes included PIP joint range of motion, grip strength, and radiographic assessmentof alignment and loosening.

Results The principal diagnosis was primary osteoarthritis in 43 patients(60%), posttraumaticarthritis in 14 (19%), rheumatoid arthritis in 9 (13%), and psoriatic arthritis in 6 (8%). Theaverage follow-up was 60 months (range, 24–108 mo). Twenty-two of 97 digits (23%) hadrepeat surgery without revision, and 13 digits (13%) had revision at an average of 15 months.There were no significant differences in preoperative and postoperative range of motion. Theaverage Disabilities of the Arm, Shoulder, and Hand score was 22, and the average pain scorewas zero. Implant migration and loosening was observed but was not related to clinicaloutcome or revision.

Conclusions The survival of pyrocarbon PIP joint arthroplasty was 85% (83 of 97) at 5 yearsof follow-up, with high patient satisfaction. Patients should be advised that the procedureachieves good relief of pain but does not improve range of motion. (J Hand Surg 2012;37A:882–888. Copyright © 2012 by the American Society for Surgery of the Hand. All rightsreserved.)

Type of study/level of evidence Therapeutic IV.

Key words: Arthroplasty, pyrocarbon, proximal interphalangeal joint, survivorship.

t

ARTHRITIS OF THE proximal interphalangeal (PIP)joint can be treated with arthrodesis or ar-throplasty. Since the successful introduc-

ion of silicone spacers for the treatment of arthri-

From the Wrightington Hospital, Hall Lane, Appley Bridge, Wigan, UK.

Received for publication February 22, 2011; accepted in revised form February 6, 2012.

No benefits in any form have been received or will be received related directly or indirectly to thesubject of this article.

Corresponding author: Adam C. Watts, MBBS, BSc, Wrightington Hospital, Hall Lane, AppleyBridge, Wigan, WN6 9EP, UK; e-mail: adamcharleswatts@gmail.com.

0363-5023/12/37A05-0002$36.00/0

doi:10.1016/j.jhsa.2012.02.012

882 � © ASSH � Published by Elsevier, Inc. All rights reserved.

is by Swanson,1 there has been interest in smalljoint arthroplasty. Implant design has evolved, andcurrently a number of anatomical resurfacing de-signs are available for the PIP joint. These im-plants are designed to reproduce normal kinemat-ics of the joint by preserving soft tissueattachments and restoring balance. Pyrolytic car-bon is a biologically inert substance2 that has aYoung modulus close to that of cortical bone andhas a low coefficient of friction. The implants areinserted as a press fit, without cementing. Initially,it was thought that bony on-growth would occur,

but more recent evidence suggests that this is not

PYROCARBON PIP JOINT ARTHROPLASTY AT MINIMUM OF 2 YEARS 883

the case.3 The aim of this study is to report themedium-term outcome of pyrocarbon PIP jointarthroplasty at a minimum of 2 years of follow-upfrom a single center. The rate and causes of repeatsurgeries, revisions, and complications were exam-ined.

MATERIALS AND METHODSA retrospective review was performed of all patientshaving pyrocarbon PIP joint arthroplasty at a singleinstitution with a minimum follow-up of 2 years. Localethics committee approval was obtained. No externalfunding was received for this study. The procedureswere performed by 5 surgeons. Patients were identifiedfrom a prospective database, and medical records werereviewed. Patient demographics and data on surgicalapproach, repeat surgery, and revision were recorded.Revision was defined as removal of the implant for anyreason. Objective, subjective, and radiological datawere obtained at routine postoperative follow-up.Range of motion was measured with a goniometer, andgrip strength was measured with a Jamar dynamometer(Jamar, Preston, MI).

Subjective outcome was measured using the quickDisabilities of the Arm, Shoulder, and Hand4 question-naire (in which a score of 0 means no disability, and ascore of 100 means complete disability) and the PatientEvaluation Measure (PEM).5 The PEM is a validated,6

subjective rating scale (from a best score of 1 to a worstscore of 7) that evaluates feeling in the hand, pain incold and damp weather, average pain, clumsiness, stiff-ness, grip strength, activities, work, appearance, andgeneral outcome. Ten-point visual analog scales wereused to record pain, patient satisfaction with the out-come of surgery, and patient satisfaction with the ap-pearance of the digit after surgery. For all 3 visualanalog scales, the best score possible was zero.

Radiological loosening was classified using a mod-ification of the system described by Herren7 (Table 1).Coronal and sagittal implant alignment of the proximaland distal components in relationship to the long axis ofthe phalanx were measured, respectively, from the an-teroposterior and lateral radiographs of the digit usingeither software within the radiology imaging system(Centricity PACS, GE Systems, Bucks, UK) or a gon-iometer for older, hard-copy radiographs. Radiographswere assessed by the same independent observer forinternal consistency, and the average of 3 measure-ments was recorded to ensure accuracy.

We identified 72 patients with a total of 97 primarypyrocarbon PIP joint arthroplasties. The average age at

the time of surgery was 57 years (range, 24 to 79 y).

JHS �Vol A,

Fifty-one patients (71%) were women. The dominanthand was involved in 41 of 97 implants (42%). Theprimary diagnosis was osteoarthritis in 43 of 72 patients(60%), rheumatoid arthritis in 9 patients (13%), psori-atic arthritis in 6 patients (8%), and trauma in 14 pa-tients (19%). The digit involved was the index finger in24 of 97 (25%), the middle finger in 31 (32%), the ringfinger in 32 (33%), and the little finger in 10 (10%). Themedian follow-up was 60 months, with a minimumfollow-up of 24 months (range, 24 to 108 mo).

Of the 72 patients having surgery, complete datawere available for 51 patients (72%) with 70 implants,at a minimum follow-up of 24 months.

Statistical analysis

Data that were normally distributed were describedusing means, minimum and maximum, and standarddeviation when appropriate. Nonparametric data weredescribed using median, minimum, and maximum val-ues, or inter-quartile range when appropriate. Compar-ative data were tested using nonparametric paired Wil-coxon signed ranks tests. Implant survival was analyzedusing Kaplan-Meier statistics. Cox regression analysiswas performed to examine factors responsible for im-plant revision. Results were considered to be statisti-cally significant if P values were � .05.

RESULTSOf the 97 arthroplasties, 46 (48%) were performed througha central split dorsal approach; 45 (46%) through a dorsalChamay approach8; and 6 (6%) through a lateral approach.

Thirteen of 97 implants (13%) were revised. Theaverage time to revision was 15 months following im-plantation (range, 0–60 mo). The reason for revisionwas joint instability in 6, aseptic implant loosening in 3,

TABLE 1. Revised Herren Grading ofRadiological Loosening Around Pyrocarbon PIPImplants7

Grade Description

0 No radiolucent line around the implant on plainradiograph

1 Radiolucent line present measuring up to 0.5 mmin width at any point around the implants

2 Radiolucent line greater than 0.5 mm in width atany point around the implant

3 Macroscopic migration of the implant*

*Macroscopic migration indicates increased angulation more than 5°.

stiffness in 2, and early infection in 1. Nine of 13 were

May

884 PYROCARBON PIP JOINT ARTHROPLASTY AT MINIMUM OF 2 YEARS

revised to silicone implants and 4 were revised to arth-rodesis. The patient with early infection had a 2-stagerevision to an arthrodesis. Using Kaplan-Meier survivalanalysis, the implant survival with revision of the im-plant as the outcome measure was 91% (� 2%) at theminimum follow-up of 2 years. At 5 years of follow-up,the survival was 85% (� 4%), with no revisions re-corded after this time point. Cox regression analysisidentified only one factor, surgery to the ring finger, thatwas related to implant revision, with 8 revision surger-ies in 31 digits and survival at 28 months of 75%.Patient age, sex, diagnosis, and immediate postopera-tive alignment were not related to implant revision atthe P � .05 significance level. Comparison of outcomeby diagnosis is given in Table 2.

Twelve early (occurring within 3 mo) and 36 latecomplications were recorded (Table 3). In addition,there were no implant fractures; squeaking occurred in3 digits in 1 patient but was well tolerated and did notrequire intervention. Twenty-two of 97 digits (23%)had repeat surgery without revision. The reasons forrepeat surgery are given in Table 4. One patient hadcollateral ligament reconstruction at 25 months, and thiswas revised to arthrodesis at 60 months because of

TABLE 2. Demographics and Outcome by Diagnosi

DiagnosisNo. ofCases

AverageAge, y

(Range)Sex

(M/F)

AverageFollow-Up,mo (Range)

Revisioo

Arthrop

OA 60 62 (36–79) 8:52 51 (24–105) 8 (1RA 12 47 (32–72) 0:12 53 (25–108) 1 (8PA 11 41 (35–64) 11:0 67 (32–90) 1 (9Trauma 14 50 (24–62) 8:6 60 (24–105) 3 (2Total 97 57 (24–79) 27:70 60 (24–108) 13 (1

OA, osteoarthritis; RA, rheumatoid arthritis; PA, psoriatic arthritis; T

TABLE 3. Early and Late PostoperativeComplications

Early (12 Digits) Late (36 Digits)

Phalanx fracture (3) Stiffness (19)Dislocation (2) Swan neck (7)Infection (1) Dislocation (4)Retained suture (1) Collateral rupture (3)Squeak (3) Aseptic loosening (3)Boutonniere (1)Heterotopic bone (1)

ongoing instability. The proportion of patients with late

JHS �Vol A,

complications was significantly higher in the patientswith a diagnosis of psoriatic arthritis than in the othergroups (Table 2; P � .02).

For all digits, the average preoperative extensor lagwas 21°, with an average flexion to 49°. The averagepreoperative arc of motion was 25° (range, 0° to 85°)(Table 2). After surgery, the average arc of motion was30° (range, 0° to 90°). On paired nonparametric analy-sis, there was no significant difference between preop-erative and postoperative arc of motion (P � .48),extension (P � .30) or flexion (P � .69). At follow-up,the average arc of motion was 50% (inter-quartilerange, 17% to 95%) of the contralateral digit, and gripstrength was, on average, 96% (inter-quartile range,78% to 109%) of the contralateral hand. Subgroupanalysis by diagnosis revealed that, in patients with adiagnosis of rheumatoid arthritis, the average arc ofmotion increased from 0° before surgery to 40° at finalfollow-up, a statistically significant difference on pairedanalysis (P � .04).

The average Disabilities of the Arm, Shoulder, andHand score at final follow-up was 22 (range, 10 to 48).The average scores for the 10 items of the PEM aregiven in Figure 1. On average, subjects reported near

s)

EarlyComplication

(% ofArthroplasties)

LateComplication

(% ofArthroplasties)

AveragePreoperative

Arc ofMotion, °(Range)

Average Arcof Motion at

FinalFollow-Up, °

(Range)

4 (7%) 21 (35%) 30 (0–85) 30 (0–90)1 (8%) 6 (50%) 0 (0–20) 40 (0–60)

0 9 (81%) 20 (0–35) 20 (0–75)2 (14%) 5 (36%) 23 (0–85) 15 (10–80)7 (7%) 41 (42%) 25 (0–85) 30 (0–90)

, posttraumatic arthritis.

TABLE 4. Reasons for Repeat Surgery (no.)

Stiffness Arthrolysis/tenolysis (9)

Percutaneous accessory collateral release (3)

Manipulation under anesthesia (3)

Deformity Flexor digitorum superficialis tenodesis forswan neck (4)

Central slip advancement for boutonniere (1)

Collateral ligament reconstruction (1)

Other Retained suture (1)

s

ns (%flastie

3%)%)%)1%)3%)

rauma

normal feeling in the hand, no pain most of the time,

May

the

PYROCARBON PIP JOINT ARTHROPLASTY AT MINIMUM OF 2 YEARS 885

and little pain in the cold or damp. The patient ratingindicated that, on average, the hand felt relativelyclumsy, stiff, and weak. Slightly better scores wereobserved for activities and work and for the appearanceof the hand. On average, the subjects felt unconcernedwhen thinking about the hand. At final follow-up, theaverage pain score, measured using a 10-point visualanalog scale (in which the best possible score was 0)was 0 (range, 0 to 7), the average satisfaction score was2 (range, 0 to 10), and the average satisfaction withappearance was 4 (range, 0 to 10).

Radiographic outcome

Immediate postoperative radiographs and complete fol-low-up radiographs were available for 51 of 97 digits.No radiolucent line was visible on initial radiographstaken at the time of surgery in 26 of 51 (51%) proximalimplants and 21 of 51 (41%) distal implants (grade 0).A radiolucent line with a width measuring less than 0.5mm was present surrounding 24 (47%) proximal com-ponents and 28 (55%) distal components (grade 1). In 1proximal component and 2 distal components, thewidth of the radiolucent line was greater than 0.5 mm(grade 2). Two of these improved to grade 1 at finalfollow-up.

At final radiographic follow-up, 8 of 51 (16%) prox-imal implants and 10 of 51 (20%) distal implants hadmigrated macroscopically within the bone from theinitial radiographic position (grade 3). Eleven of 102(11%) components had a radiolucent line greater than

1

2

3

4

5

6

7

FIGURE 1: Average scores on

0.5 mm around the implant (grade 2). Thirty-eight

JHS �Vol A,

(75%) proximal implants and 35 (69%) distal implantshad a radiolucent line visible but with a width measur-ing less than 0.5 mm (grade 1). None of the postoper-ative radiographs were classified as grade 0. Sixty-fiveof 102 (64%) implants showed progression of radiolog-ical loosening on 2 temporally separate radiographs.Four implants were noted to be dislocated on radio-graphic follow-up but were asymptomatic.

Figure 2 shows a plot of the immediate postoperativealignment and alignment on the final follow-up radio-graph in the coronal and sagittal plane for the proximaland distal components. The difference in average cor-onal alignment of the proximal component immediatelyafter surgery and at final follow-up was not statisticallysignificant on paired analysis (P � .26). Coronal align-ment of the distal component and sagittal alignment ofboth proximal (P � .001) and distal components (P �.025) did change significantly from initial postoperativeradiographs to final follow-up.

DISCUSSIONArthritis affecting the PIP joint is frequentlytreated with either joint arthrodesis or silicone PIPjoint arthroplasty. Successful fusion provides re-lief of pain and a stable pinch in radial digits9;however, the abolition of movement might be as-sociated with notable loss of function, especiallywhen the ring or little finger is affected, becausePIP joint flexion is required to achieve power grip.Silicone arthroplasty can achieve high rates of pain

10-item PEM questionnaire.5

relief and patient satisfaction10; however, the prob-

May

he la

886 PYROCARBON PIP JOINT ARTHROPLASTY AT MINIMUM OF 2 YEARS

lem of implant fatigue and fracture remains.11,12

As in studies of silicone replacement of the metacarpo-phalangeal joint,13 the risk of revision increases in alinear fashion from the time of implantation, with re-ported implant survival at 9 years of 81% in patientswith rheumatoid arthritis.14 Surface replacement PIP

FIGURE 2: Difference in coronal and sagittal alignment of thphalanx on the initial postoperative radiograph (x axis) and at t

TABLE 5. Tabulated Outcomes from Previously Pu

Year Author NumberSexM:F

Follow-Up,mo (Range)

Age, y(Range

2006 Herren7 17 N/A 19 (12–27) 64 (55–82006 Nunley15 7 2:3 17 (12–23) 40 (28–52006 Tuttle14 18 0:8 13 (6–30) 62 (52–62007 Branam17 19 0:10 19 (6–36) 62 (52–62007 Bravo15 50 15:20 37 (27–46) 53 (21–72010 Wijk18 53 7:36 24 (12–60) 59 (40–82010 Present study 97 21:51 60 (24–109) 56 (24–7

N/A, not available.*Repeat surgery to the digit after the index episode without remov†Removal of the implant for any reason.

joint arthroplasty has been developed as a motion-

JHS �Vol A,

preserving alternative to arthrodesis, without the prob-lems of implant fatigue fracture observed with silicone.

The short-term outcome of pyrocarbon surface re-placement PIP joint arthroplasty has been reported by anumber of authors (Table 5).7,15–18 These studies havereported improvement in grip strength but no significant

oximal and distal component relative to the long axis of thest radiological review (y axis).

ed Studies

IncreaseArc of

Motion (°)Repeat

Surgery* Revision† SatisfactionAverage Final

Pain Score

8 6% 71% 1�2 28% 14% 4

0 11% 0 84% 01 0 81% 27 16% 12% 80% 1

�8 2% 11% 0.413 23% 13% 76% 0

xchange of the implant.

e pr

blish

)

1)6)9)9)3)5)9)

al or e

change in arc of motion.15–18 Arc of motion was better

May

PYROCARBON PIP JOINT ARTHROPLASTY AT MINIMUM OF 2 YEARS 887

in those who had a volar approach compared to thedorsal approach.7 Pain relief has been excellent in moststudies.16,18,19 The largest study published to date in-cluded 53 joints in 43 patients at an average of 23months.19 Pain improved significantly, but arc of mo-tion and grip strength were not considerably altered.

The findings of the present study with a larger pop-ulation and longer follow-up are in agreement withthose outlined earlier. The present study is limited bythe retrospective nature of data collection, resulting in alack of preoperative measures; the variety of surgicalapproaches used; the heterogeneous case mix; the factthat radiographs were not standardized; and the numberof patients lost to follow-up. The data indicate a signif-icant change in arc of motion only in those with adiagnosis of rheumatoid arthritis, but with a small sub-ject group, there is a risk of type II error. We canconclude that PIP joint arthroplasty with pyrocarbonimplants can reliably relieve pain but does not increaseaverage arc of motion significantly from the preopera-tive range. This might account for the relatively poorsubjective patient rating for use of the hand, movement,and grip strength on the PEM score reported in thepresent study.

Nunley et al, in their cohort study, concluded thatpyrocarbon arthroplasty should not be used in posttrau-matic osteoarthritis.17 Collateral ligament integrity ismandatory for successful surface replacement PIP jointarthroplasty, and osteoarthritis is the major indication.In the present series, several procedures were per-formed in patients with inflammatory arthropathy orposttraumatic arthropathy after careful preoperative as-sessment of collateral ligament integrity. The propor-tion of patients with late complications was signifi-cantly greater in those with a diagnosis of psoriaticarthropathy, which was related to a high rate of dislo-cation. No other significant differences in outcome wereobserved when surgery was performed for these indi-cations when compared to those for a diagnosis ofprimary osteoarthritis, but with small numbers in eachgroup, caution must be exercised when interpretingthese data. A recently published study of 31 arthroplas-ties in 17 patients with osteoarthritis has suggested that,at an average follow-up of 55 months, early gains inmovement are lost, and implant migration is a notewor-thy problem.20

There remains concern about loosening of pyrocar-bon implants. Herren et al7 were the first authors tohighlight concerns about osseointegration of pyrocar-bon implants. A radiolucent line of less than 0.5 mm issaid to be normal and due to the radiolucent pyrocarbon

coating of the implant. The fact that it was not initially

JHS �Vol A,

visible in 51% of proximal and 41% of distal compo-nents in the present study casts doubt on this proposi-tion, although these results may simply reflect a radio-logical phenomenon due to the absence of boneapposition against the implant immediately after ream-ing the phalanges. Although implant loosening has beenidentified, there is currently no evidence that it affectsthe clinical outcome. The present study has reported arevision rate of 13% at an average follow-up of 5 years.This is comparable to revision rates reported to date ofbetween 0% and 28% at 10 to 37 months of follow-up.7,15–18 The data presented suggest that implant fail-ure requiring revision is a problem that occurs primarilywithin the first 2 years, and that after this period, the riskof revision decreases substantially. Early revisionsmight represent technical errors such as improper align-ment or early complications such as infection. Thispattern of failure differs from the experience with sili-cone arthroplasty14 and might be, in part, related toimplant settling that is believed to occur with pyrocar-bon arthroplasty,7 whereby the implant can initiallymigrate to a stable position, as observed in some of theimplants in the present series. The shape of the Kaplan-Meier curve is not consistent with progressive loosen-ing of an implant, in which an increase in revision ratewith time would be anticipated.

The overall repeat surgery rate reported in the pres-ent study is higher than those reported in previousstudies.15,16,19 Initially, with the ambition of restoringmovement to stiff digits, our practice was to treat post-operative stiffness aggressively with manipulation un-der anesthesia, percutaneous or open arthrolysis, ortenolysis. Our understanding that joint motion does notimprove to any degree after PIP joint arthroplasty hasled to an adjustment of expectations for ourselves andfor counseling patients. Other important lessons thathave been learned over the last 10 years include theliberal use of fluoroscopy at all stages of implantation toensure optimal alignment, greater attention to soft tissuebalance in both extension and flexion, and use of aside-cutting burr with copious saline lavage to removehard subcortical bone to ensure that the largest prosthe-sis possible is implanted with a good press fit. Althoughnot proven, it is our impression that prosthetic failure ismore common if the prosthesis implanted is too smallfor the digit and if the implant is not correctly aligned.The data presented indicate that pyrocarbon PIP jointarthroplasty is a reasonable alternative to PIP jointarthrodesis or silicone arthroplasty but that the compli-cation rate might be unacceptably high in patients with

psoriatic arthropathy.

May

888 PYROCARBON PIP JOINT ARTHROPLASTY AT MINIMUM OF 2 YEARS

REFERENCES

1. Swanson AB. Flexible implant arthroplasty of the proximal inter-phalangeal joint of the fingers. Ann Plast Surg 1979;3:346–354.

2. Haubold AD. On the durability of pyrolytic carbon in vivo. Med ProgTechnol 1994;20:201–208.

3. Daecke W, Veyel K, Wieloch P, Jung M, Lorenz H, Martini AK.Osseointegration and mechanical stability of pyrocarbon and tita-nium hand implants in a load-bearing in vivo model for small jointarthroplasty. J Hand Surg 2006;31A:90–97.

4. Beaton DE, Wright JG, Katz JN. Development of the QuickDASH:comparison of three item-reduction approaches. J Bone Joint Surg2005;87A:1038–1046.

5. Macey AC, Burke FD, Abbott K, Barton NJ, Bradbury E, Bradley A,et al. Outcomes of hand surgery. British Society for Surgery of theHand. J Hand Surg 1995;20B:841–855.

6. Dias JJ, Bhowal B, Wildin CJ, Thompson JR. Assessing the outcome ofdisorders of the hand. Is the patient evaluation measure reliable, valid,responsive and without bias? J Bone Joint Surg 2001;83B:235–240.

7. Herren DB, Schindele S, Goldhahn J, Simmen BR. Problematic bonefixation with pyrocarbon implants in proximal interphalangeal jointreplacement: short-term results. J Hand Surg 2006;31B:643–651.

8. Chamay A. A distally based dorsal and triangular tendinous flap fordirect access to the proximal interphalangeal joint. Ann Chir Main1988;7:179–183.

9. Leibovic SJ. Instructional Course Lecture. Arthrodesis of the inter-phalangeal joints with headless compression screws. J Hand Surg2007;32A:1113–1119.

10. Namdari S, Weiss AP. Anatomically neutral silicone small joint arthro-

plasty for osteoarthritis. J Hand Surg 2009;34A:292–300.

JHS �Vol A,

11. Lin HH, Wyrick JD, Stern PJ. Proximal interphalangeal joint siliconereplacement arthroplasty: clinical results using an anterior approach.J Hand Surg 1995;20A:123–132.

12. Iselin F, Conti E. Long-term results of proximal interphalangeal jointresection arthroplasties with a silicone implant. J Hand Surg 1995;20A:S95–97.

13. Trail IA, Martin JA, Nuttall D, Stanley JK. Seventeen-year survi-vorship analysis of Silastic metacarpophalangeal joint replacement.J Bone Joint Surg 2004;86B:1002–1006.

14. Ashworth CR, Hansraj KK, Todd AO, Dukhram KM, EbramzadehE, Boucree JB, et al. Swanson proximal interphalangeal joint arthro-plasty in patients with rheumatoid arthritis. Clin Orthop Relat Res1997;342:34–37.

15. Tuttle HG, Stern PJ. Pyrolytic carbon proximal interphalangeal jointresurfacing arthroplasty. J Hand Surg 2006;31A:930–939.

16. Bravo CJ, Rizzo M, Hormel KB, Beckenbaugh RD. Pyrolytic carbonproximal interphalangeal joint arthroplasty: results with minimumtwo-year follow-up evaluation. J Hand Surg 2007;32A:1–11.

17. Nunley RM, Boyer MI, Goldfarb CA. Pyrolytic carbon arthroplastyfor posttraumatic arthritis of the proximal interphalangeal joint.J Hand Surg 2006;31A:1468–1474.

18. Branam BR, Tuttle HG, Stern PJ, Levin L. Resurfacing arthroplastyversus silicone arthroplasty for proximal interphalangeal joint osteo-arthritis. J Hand Surg 2007;32A:775–788.

19. Wijk U, Wollmark M, Kopylov P, Tagil M. Outcomes of proximalinterphalangeal joint pyrocarbon implants. J Hand Surg 2010;35A:38–43.

20. Sweets TM, Stern PJ. Pyrolytic carbon resurfacing arthroplasty for

osteoarthritis of the proximal interphalangeal joint of the finger.J Bone Joint Surg 2011;93A:1417–1425.

May