Teaigl

bedvosan

2

Predicting the Outcome of Surgery for theProximal Interphalangeal Joint in

Dupuytren’s DiseaseAlok Misra, MSc, Abhilash Jain, PhD, Reza Ghazanfar, MBBS,

Terrencia Johnston, BSc, Jagdeep Nanchahal, PhD

From the Imperial College School of Medicine, Department of Plastic and Reconstructive Surgery, Chelseaand Westminster Hospital, London, United Kingdom.

Purpose: We prospectively studied the outcome of limited Dupuytren’s fasciectomy, incombination with joint release if necessary, for disease involving 49 proximal interphalangealjoints (PIPJs) to identify factors that predispose to recurrent PIPJ contracture.Methods: Thirty-seven patients were treated over a 4-year period. The flexion contracture ofthe PIPJ was measured before surgery, immediately after surgery, and at more than 1 yearafter surgery.Results: A mean preoperative flexion contracture of 67° � 22° was corrected to 6° � 10° atthe time of surgery and 25° � 25° at the follow-up evaluation. There was a positivecorrelation between the severity of the preoperative flexion contracture and recurrent defor-mity, with a preoperative contracture greater than 60° leading to significantly worse outcome.Incomplete correction of PIPJ flexion contracture during surgery and poor postoperativecompliance with therapy were also associated with worse recurrent joint contractures. Thedigit involved and the necessity for joint release did not significantly affect outcome.Conclusions: In the absence of recurrent Dupuytren’s disease, severe preoperative deformity,incomplete correction at surgery, and noncompliance with therapy predispose patients toworse PIPJ contracture. (J Hand Surg 2007;32A:240–245. Copyright © 2007 by the AmericanSociety for Surgery of the Hand.)Type of study/level of evidence: Prognostic II.Key words: Dupuytren’s disease, outcome, proximal interphalangeal joint.

paPrflePfii

ofd

MP

he outcome after surgery to the proximal in-terphalangeal joint (PIPJ) in Dupuytren’s dis-ease is highly variable. Limited fasciectomy,

ven when combined with joint release may at bestchieve less than 60% improvement in extensor def-cit at 2 years.1 Despite incomplete correction, sur-ery for PIPJ contracture in Dupuytren’s diseaseeads to improved hand function.2,3

Flexion deformity after surgery may developecause of recurrent Dupuytren’s disease or postop-rative scarring. Factors associated with recurrentisease include knuckle pads, plantar disease, in-olvement of the radial side of the hand, early age ofnset, bilateral hand involvement, and surgery to themall finger.4 The features associated with postoper-tive flexion without recurrent disease, however, are

ot well defined. Legge and McFarlane5 found that o

40 The Journal of Hand Surgery

reoperative flexion deformity, small finger disease,nd multiple ray involvement correlated with worseIPJ contracture, whereas factors associated withecurrent disease were not predictive of repeat PIPJexion. Poor compliance with postoperative dynamicxtension splintage has also been associated withIPJ contracture,1 although these researchers did notnd a correlation with preoperative deformity, digit

nvolved, or PIPJ release.We sought to validate these factors and identify

thers that predict the outcome of surgery to the PIPJor Dupuytren’s disease in the absence of recurrentisease.

aterials and Methodsatients having surgery for Dupuytren’s contracture

ver a 4-year period (1998–2002) were evaluated

pgdaTadtflttppmtPPcawpcTwtrch2

SUeatpflitdlptolcp

flss

wfwfntfictbRtfwssfltcrt

SS(pacpuptpfaccSt

RTdfinyfcmst

Misra et al / Outcome of the PIPJ in Dupuytren’s Disease 241

rospectively. Patients with only metacarpophalan-eal joint (MCPJ) involvement or those requiringermofasciectomy were excluded. Those with PIPJnd MCPJ involvement were recruited for the study.he flexion contracture was measured before andfter surgery and the passive deficit was recordeduring surgery with a goniometer. Before surgery,he PIPJ measurement was performed with the MCPJexed, thereby correcting for any MCPJ flexion con-

racture. After surgery, the patient was asked to ac-ively extend the PIPJ. All surgical procedures wereerformed by the senior surgeon (J.N.). A total of 56atients were treated over the study period. Completeeasurements were obtained for 49 digits in 37 pa-

ients. Of the 49 digits, 3 had combined MCPJ andIPJ contracture, with the remaining digits havingIPJ involvement only. Nineteen patients were ex-luded: 3 did not have complete preoperative data,nd 16 had no late follow-up measurements. Of thoseho could not be assessed for follow-up study, 1atient had died, 3 had moved out of the area, and 12ould not be traced because of an address change.he mean patient age was 63 years (range, 39–88 y),ith a male-to-female ratio of 3.6:1. Thirty-five pa-

ients presented with primary disease, and 2 hadecurrent Dupuytren’s disease. The latter were ex-luded from this study. Seventeen patients had right-and involvement, 18 had left-hand involvement, andhad bilateral disease.

urgical Techniquender tourniquet control, Dupuytren’s cords were

xcised via midline volar skin incisions in the digitsnd a transverse palmar incision. If full correction ofhe PIPJ was not achieved after fasciectomy, gentleassive manipulation of the joint, with the MCPJexed, was undertaken. If there was a residual flex-

on contracture of more than 20°, the following struc-ures were divided in order as described by An-rew6—accessory collateral ligaments, check reinigaments, and the flexor sheath just distal to the A2ulley—followed again by gentle passive manipula-ion. Any residual postoperative flexion contracturef the PIPJ was then recorded. After tourniquet re-ease and hemostasis, the digital incisions werelosed by Z-plasties and the palm left open. A well-added dressing was applied.The dressings were reduced at day 3, and active

exion of the digits commenced under the supervi-ion of the hand therapist. Static night extension

plintage was used for 3 months or until the scars 2

ere mature. The thermoplastic splint was initiallyabricated to hold the interphalangeal joints extendedith the MCPJs flexed, and the splint was adjusted to

ull extension at subsequent therapy visits. A dy-amic wire-based (Capner) splint was used in pa-ients who had PIPJ release and those who had dif-culty with active extension of the PIPJ because ofentral slip attenuation. Splintage was continued un-il the scars were mature, usually 3 months. Splintageeyond this period is unlikely to be of benefit, anditchie et al7 found little change in flexion contrac-

ion between 3 months and 3 years. All patients wereollowed up for a minimum of 1 year (range, 1–3 y),hen the final PIPJ flexion contracture was mea-

ured. The complications in each group were toomall for meaningful statistical comparison. The finalexion contracture of the PIPJ was assessed in rela-

ion to the preoperative deformity, digit involved,orrection at the time of surgery, necessity for PIPJelease, and postoperative patient compliance withherapy.

tatistical Analysistatistical analyses were performed using softwarePrism; GraphPad Software, San Diego, CA) for aersonal computer. The mean PIPJ flexion deformityt least 1 year after surgery, within groups, wasompared with preoperative measurements using theaired Student t test. Digits were compared using thenpaired Student t test by looking at the severity ofreoperative deformity, complete extension at theime of surgery, PIPJ release, digit involved, andostoperative patient compliance on PIPJ flexion de-ormity at 1 year or more after surgery. The associ-tions between preoperative flexion deformity andorrection at the time of surgery with the final flexionontracture were assessed using Pearson correlation.ignificance was achieved if p was less than or equal

o .05.

esultshe small finger was most commonly affected (30igits), followed by the ring (16 digits) and middlengers (3 digits). The index finger and thumb wereot involved. The mean follow-up period was 1.5ears (range, 1–3 y). All 3 involved MCPJs and wereully corrected. A mean preoperative PIPJ flexionontracture of 67° � 23° was corrected to a meaneasure of 6° � 10° at the time of surgery, and

ignificant improvement was maintained at morehan 1 year, with a mean flexion contracture of 25° �

5° (p � .001). Recurrent Dupuytren’s disease was

npPdna

htfw

Ppataiw.

Dlfi(

C

s1cdewcr(oawew

Pwawcl

Fpf

Ft

242 The Journal of Hand Surgery / Vol. 32A No. 2 February 2007

ot clinically detectable in any patient throughout theeriod of study. The final flexion contracture of theIPJ was assessed in relation to the preoperativeeformity, digit involved, intraoperative correction,ecessity for PIPJ release, and postoperative compli-nce with therapy.

Complications included 1 digital nerve injury, 1ematoma, and 1 superficial infection in the fasciec-omy alone group and 2 superficial infections in theasciectomy � PIPJ release group, which resolvedith oral antibiotic treatment.

reoperative flexion deformity. There was aositive correlation between the severity of preoper-tive contracture and final postoperative flexion con-racture (p � .019, r � 0.33) (Fig. 1). Subgroupnalyses found that patients with a preoperative flex-on contracture greater than 60° had significantlyorse flexion contracture at more than 1 year (p �

017) (Table 1).

igit involved. The digit involved did not corre-ate with the final outcome. In particular, the smallngers did not fare any worse than the other digitsp � .28).

orrection at the time of surgery. Full exten-

igure 1. The positive correlation between the severity ofreoperative contracture and final postoperative flexion de-ormity (p � .019, r � 0.33).

Table 1. Comparison of ProximalInterphalangeal Joints With Preoperative FlexionContracture of More Than 60° Evaluation

PIPJ Measurement (degrees) <60° >60°

Mean before surgery 43 79Mean after surgery 13* 31*

*p � .017. 0

ion was achieved at the time of surgery for 35 digits,6 by fasciectomy alone and the remaining 19 by aombined fasciectomy and PIPJ release. Fourteenigits did not straighten fully at the time of surgery,ven when the joint was released (Table 2). Thereas a positive correlation between the intraoperative

orrection and the final flexion contracture (p � .01,� .35) (Fig. 2). There was no significant difference

p � .22) between the mean preoperative deformityf the PIPJs for which complete correction waschieved at the time of surgery (64°) and those forhich this was not possible (73°). The final postop-

rative flexion contracture for the 2 groups, however,as significantly different (p � .04) (Table 3).

roximal interphalangeal joint release. Thereere no significant differences between the preoper-

tive and final postoperative flexion contractureshen those digits treated by fasciectomy alone were

ompared with those requiring additional PIPJ re-ease (Table 4). There was no difference in the flex-

Table 2. Surgical Correction of the ProximalInterphalangeal Joint by Fasciectomy and inCombination With Joint Release

Intraoperative PIPJMeasurement

FasciectomyAlone

Fasciectomy andJoint Release

Straight, n 16 19Not straight, n 0 14

Proximal interphalangeal joints that fully straightened and thosethat did not are shown.

igure 2. The positive correlation between correction at theime of surgery and the final flexion contracture (p � .01, r �

.35).

iab(aot.

Ccdt.

DTatefmwmofsp

(cceTwiccd

fiTFdcisslsaftteteb

ott

Misra et al / Outcome of the PIPJ in Dupuytren’s Disease 243

on contracture before and after surgery for digits thatchieved complete extension at the time of surgeryy fasciectomy alone or with additional PIPJ releaseTable 5). Furthermore, digits treated by fasciectomynd PIPJ release with complete extension at the timef surgery did significantly better in the long termhan those with a residual flexion contracture (p �04) (Table 3).

ompliance. Twenty-nine of the 37 patients (78%)omplied with postoperative hand therapy and 8 (22%)id not. The compliant patients did significantly bet-er in terms of the final PIPJ flexion contracture (p �003) (Table 6).

iscussionhe outcome of surgery for PIPJ contracture in thebsence of recurrent Dupuytren’s disease is difficulto predict. Legge and McFarlane5 developed a math-matic formula to assist with this process. Theyound that severe preoperative deformity was theost statistically significant variable in predictingorse PIPJ outcome, followed by small finger andultiple ray involvement. Nonetheless, the accuracy

f their predictive formula was limited by an erroractor, which they suggested may include variablesuch as patient compliance. Abe et al8 also found thatatients with a severe preoperative PIPJ deformity

Table 4. Proximal Interphalangeal Joint Releaseand Fasciectomy-Only Groups

PIPJ Measurement(degrees)

Fasciectomy andJoint Release

FasciectomyAlone

Before surgery 73 60During surgery 9 0After surgery 10 22

There were no significant differences between the groups: before

Table 3. Digits Treated by Fasciectomy andProximal Interphalangeal Joint Release ThatAchieved Complete Extension During SurgeryVersus Digits When Full Extension was NotAchieved

Time of PIPJ Measurement(degrees) Straight Not Straight

Before surgery 64 73After surgery 20* 36*

There was no significant difference in the preoperative measure-ments (p � .22).

*p � .04.

surgery, p � .09; at follow-up evaluation, p � .18.

�60° flexion contracture) had worse final PIPJ out-ome. Our data confirm that preoperative deformityorrelates with outcome, especially when the preop-rative PIPJ flexion contracture is greater than 60°.he relationship between preoperative deformity andorse final PIPJ outcome, however, is not borne out

n every study. Rives et al1 failed to show this asso-iation, but of the 23 PIPJs they studied, only 9 hadontractures of less than 60°, with a mean extensoreficit of 49°.Like Rives et al,1 we did not find that the small

nger did significantly worse than the other digits.his is at variance with the data of Legge and Mc-arlane,5 who looked at the MCPJ, the PIPJ, and theistal interphalangeal joints of each digit. It is diffi-ult to ascertain the relative number of small fingersn their study from the published data. Rives et al1

tudied only the PIPJs, and 70% of their digits weremall fingers; 62% of the digits in our study wereittle fingers. Abe et al4 also found that ring- andmall-finger PIPJs tended to do worse than index-nd middle-finger joints. Legge and McFarlane5

ound the small finger to be the only digit for whichhe postoperative outcome of the PIPJ depended onhe degree of MCPJ contracture.5 This is probablyxplained by the natural tendency of the MCPJ forhis finger to hyperextend with all the pull of thextensor mechanism expended at this level beforeeing transmitted to the PIPJ.We found that patients who complied with post-

perative hand therapy had less of a flexion contrac-ure (20°) than the noncompliant group (48°) at morehan 1 year. This confirms the findings of a previous

Table 5. Digits That Achieved CompleteExtension During Surgery

PIPJ Measurement [°]Fasciectomy

AloneFasciectomy and

Joint Release

Before surgery 68 60After surgery 18 22

No significant differences in the extensor deficit were seen beforeor after surgery in either group: before surgery, p � .3; aftersurgery, p � .58.

Table 6. Patient Compliances and Final ProximalInterphalangeal Joint Flexion Contracture

PIPJ Measurement [°] Compliant Not Compliant

Before surgery 60 68After surgery 20* 48*

*p � .003.

sosaiawLi

tPcasog

icpctlbpftfuaaganbgspdtciaact

reba

rdctcietdtw2lwrr

RN

aa

Me

R

244 The Journal of Hand Surgery / Vol. 32A No. 2 February 2007

tudy1 in which patients who complied with a post-perative dynamic extension splintage program didignificantly better at a 2-year follow-up evaluationfter fasciectomy with joint release. Noncompliances possible from the outset, but other patients who aret first well motivated may succumb from pain orhat they judge to be unsatisfactory progression.ong traveling distances for hand therapy and med-

cal illness can also cause reduced attendance.1

We have shown that residual flexion contracture athe time of surgery is a predictor of worse recurrentIPJ contracture. This was independent of whetheromplete correction was achieved by fasciectomylone or combined with PIPJ release. There was noignificant difference in the preoperative deformityf those achieving complete correction during sur-ery with those for whom this was not possible.In a study of amputated digits, Andrew6 found that

t was possible to achieve full PIPJ extension in allases provided a sufficiently radical joint release waserformed. This required the release of the accessoryollateral ligaments, check rein ligaments, and de-achment of the origin of the volar plate before re-ease of the transverse retinacular ligaments, lateralands, and proper collateral ligaments. After com-lete release of the volar plate the fingers snappedrom 45° of flexion into 10° of hyperextension. Weook great care to avoid this phenomenon. Afterasciectomy, the joints were gently passively manip-lated and, in the absence of complete extension, theccessory collateral ligaments, check rein ligaments,nd flexor sheath were divided, followed by furtherentle manipulation. If complete correction was notchieved at this stage, the position was accepted. It isot possible to quantify gentle passive manipulation,ut all patients were treated consistently by one sur-eon. Manipulation consisted of gentle, passivetretching of an extent that is sufficient to disrupt theeriarticular adhesions but not forceful enough toisrupt the volar plate. Other pathologic variableshat have been identified in long-standing flexionontracture at the PIPJ include central extensor slipncompetence and PIPJ articular changes.9,10 Thesere difficult to quantify, and we did not assess thems independent variables for postoperative PIPJ out-ome. When present, however, they are likely to leado a worse final outcome.

It is not clear from the literature whether PIPJelease affects long-term outcome. Some research-rs11 have suggested that joint release is not useful,ut in that study the patients were not randomized

nd may have been subject to selection bias. More

ecently, when studying patients with PIPJ flexioneformity of at least 60°, patients who achieved fullorrection at the time of surgery had an almost iden-ical final residual deformity (15°) at 6 months whenompared with those requiring joint release (16°) tomprove residual preoperative deformity after fasci-ctomy.12 By contrast, Ritchie et al7 found that digitsreated by PIPJ release had a greater mean residualeformity at 3 years (29°) compared with fasciec-omy alone (8°), but the preoperative contracturesere significantly different, with means of 75° and8°, respectively. We sought to determine the factorseading to recurrent PIPJ contractures, and our studyas not designed to compare the outcome of PIPJ

elease with fasciectomy alone. Hence, we did notandomize the patients to have joint release.

eceived for publication January 1, 2006; accepted in revised formovember 22, 2006.No benefits in any form have been received or will be received fromcommercial party related directly or indirectly to the subject of this

rticle.Corresponding author: Mr. Alok Misra, 32 Old Hall Gardens,onkspath, Solihull, Birmingham, B90 4NN United Kingdom;

-mail: alokmis@gmail.com.Copyright © 2007 by the American Society for Surgery of the Hand0363-5023/07/32A02-0016$32.00/0doi:10.1016/j.jhsa.2006.11.015

eferences1. Rives K, Gelberman R, Smith B, Carney K. Severe contrac-

tures of the proximal interphalangeal joint in Dupuytren’sdisease: results of a prospective trial of operative correctionand dynamic extension splinting. J Hand Surg 1992;17A:1153–1159.

2. Sinha R, Creswell TR, Mason R, Chakrabarti I. Functionalbenefit of Dupuytren’s surgery. J Hand Surg 2002;27B:378–381.

3. Draviaraj KP, Chakrabarti I. Functional outcome after sur-gery for Dupuytren’s contracture: a prospective study.J Hand Surg 2004;29A:804–808.

4. Abe Y, Rokkaku T, Ofuchi S, Tokunaga K, Takahashi,Moriya H. An objective method to evaluate the risk ofrecurrence and extension of Dupuytren’s disease. 2004;29B:427–430.

5. Legge JWH, McFarlane RM. Prediction of results of treat-ment of Dupuytren’s disease. J Hand Surg 1980;5:608–616.

6. Andrew JG. Contracture of the proximal interphalangealjoint in Dupuytren’s disease. J Hand Surg 1991;16B:446–448.

7. Ritchie JFS, Venu KM, Pillai K, Yanni DH. Proximal inter-phalangeal joint release in Dupuytren’s disease of the littlefinger. J Hand Surg 2004;29B:15–17.

8. Abe Y, Rokkaku T, Ofuchi S, Tokunaga K, Takahashi,Moriya H. Surgery for Dupuytren’s disease in Japanesepatients and a new preoperative classification. J Hand Surg2004;29B:233–237.

9. Field PL, Hueston JT. Articular cartilage loss in long-stand-

1

1

1

Misra et al / Outcome of the PIPJ in Dupuytren’s Disease 245

ing flexion deformity of the proximal interphalangeal joints.Aust N Z J Surg 1970;40:70–74.

0. Breed CM, Smith PJ. A comparison of treatment of PIP jointcontractures in Dupuytren’s disease. J Hand Surg 1996;21B:246–251.

1. Weinzweig N, Culver JE, Fleegler EJ. Severe contractures of

the proximal interphalangeal joint in Dupuytren’s disease:

combined fasciectomy with capsuloligamentous releaseversus fasciectomy alone. Plast Reconstr Surg 1996;97:560–566.

2. Beyermann K, Prommersberger KJ, Jacobs C, Lanz UB.Severe contracture of the proximal interphalangeal joint inDupuytren’s disease: does capsuloligamentous release im-

prove outcome? J Hand Surg 2004;29B:240–243.