Flexor tendon repair in zone 2 followed by early active mobilization

FLEXOR

TENDON

REPAIR IN ZONE 2 FOLLOWED ACTIVE MOBILIZATION

BY EARLY

A. BAKTIR, C. Y. T[SRK,~. KABAK,V. SAHIN and Y. KARDA$ From the Department of Orthopaedic Surgery, Division of Hand Surgery, Erciyes University, Kayseri, Turkey

The results following primary and delayed primary repair in zone 2 flexor tendon injuries were evaluated prospectively in 88 fingers of 71 patients using two different early postoperative mobilization programmes. In 33 patients, the Kleinert rubber band passive flexion method was used. In the remaining 38 patients, the early active mobilization programme was used. All patients were reviewed 1 year after operation and the results assessed by the Strickland criteria. During this evaluation maximum grip strength was also measured. The results were excellent or good in 78% of digits and mean grip strength was 84% of the uninjured hand in the Kleinert rubber band passive flexion group. In the early active mobilization group, excellent or good results were achieved in 85% of the digits and the mean grip strength was 90% of the uninjured hand. There were two early ruptures in each group.

Journal of Hand Surgery (British and European Volume, 1996) 21B." 5:624-628 71 patients with zone 2 flexor tendon injuries in 88 digits. In 32 digits (36%) there was complete division of the FDP only. In the other 56 digits (64%) there were lacerations of both the FDS and FDP tendons (Table 1). In 58 patients, primary repair was performed within 12 hours. In the other 13 patients, delayed primary repair was carried out within 2 weeks. In 33 patients, the Kleinert rubber band passive flexion/active extantion method was used (Group 1). In the remaining 38 patients, the early active motion regimen was used (Group 2).

Early controlled motion programs after flexor tendon repair are designed to minimize adhesion formation by promoting the excursion of repaired tendons. Several laboratory studies have also shown that controlled mobilization decreases adhesion formation and increases the strength of repairs (Burge and Brown, 1990; Gelberman et al, 1981; 1982; 1983; Matthews and Richards, 1976). The immediate controlled mobilization introduced by Kleinert et al (1967) is well accepted and has been subjected to several modifications. The most important change was the introduction of a palmar pulley by Slattery and McGrouther (1984). This improved the passive excursion of the flexor tendons within the flexor tendon sheath. Excellent clinical results were reported by Louisville group (Lister et al, 1977) using the original regimen, but results were inferior when the method was used in other centres (Earley and Milward, 1982; Ejeskfir, 1984; Gault, 1987; Singer and Maloon, 1988). More recent studies using the palmar pulley modification have reported better results (Burge and Brown, 1990; Chow et al, 1987; 1988; Karlander et al, 1993; Saldana et al, 1991; Werntz et al, 1989). Bainbridge et al (1994), Cullen et al (1989), Elliot et al (1994), Savage and Risitano (1989) and Small et al (1989) have all reported use of early active mobilization after zone 2 flexor tendon repair with encouraging results. The purpose of this study was to compare the Kleinert rubber band passive flexion/active extension method with the early active postoperative mobilization method in the treatment of zone 2 flexor tendon repairs.

Table 1--Details of patients and results*

Age: m e a n (range) Sex: males Injured hand (right) Isolated F D P injuries F D P + FDS injuries Multiple digit injuries** Digit injured Small Long Ring Index Grip strength [mean (SD)]t TAM~ (degrees) [mean (SD)] Rupture Extension deficit 0° 1 15 ° 16 30 ° >30 °

PATIENTS AND METHODS Between September 1991 and December 1993, we treated a total of 89 patients with 101 flexor tendon injuries in zone 2. We excluded fingers with phalangeal fractures, joint injuries, injury of more than one neurovascular bundle and major finger injuries. No thumbs are included in this series. After these exclusions we reviewed

Kleinert group (n=41)

Early active mobilization group (n=47)

33 (13-52) 26 27 14 27 15

34 (11 53) 33 30 18 29 18

10 9 8 14 84 (8) 130 (35) 2 11 20 10 0

11 7 14 15 90 (7)§ 137 (31) 2 23 19 5 0

* Apart from age and grip strength, all data refer to the number of digits. ** digits in patients with tendon injuries in more than one digit. ~"Expressed as percentage of uninjured hand. $ Total active motion. § Significantly more than in the Kleinert group (Student t-test for unpaired groups: P<0.01). 624

F L E X O R T E N D O N REPAIRS IN Z O N E 2

Group 1 Thirty-three patients who had repair of 68 flexor tendons in 41 fingers in zone 2 were reviewed (Table 1). Bruner's incision was used to extend the original laceration. The F D P was repaired by a modified Kessler technique with 4/0 braided polyester and a 6/0 continuous epitendinous Prolene stitch. The FDS tendon was repaired with horizontal mattress sutures of 4/0 braided polyester. The injured sheath was opened by an "L"-shaped cut and was completely closed with interrupted 6/0 Prolene sutures. Postoperatively, the hand was immobilized in a dorsal plaster splint from the elbow to the fingertips with the wrist in 30 to 40 ° of palmar flexion, the MP joints at 70 to 90 ° of flexion and the IP joints straight. A rubber band was attached to the fingertip under appropriate tension, incorporating a distal palmar bar. The patients were instructed to perform active extension and passive flexion of the fingers for 4 weeks. This exercise was repeated ten times every hour during the day. After 4 weeks the rubber band was removed and active extension and flexion exercises commenced. The plaster was removed at 6 weeks. After this time, a slight load was permitted, increasing to full load within 10 to 12 weeks after operation.

Group 2 Thirty-eight patients who had repair of 76 flexor tendons in 47 fingers in zone 2 were reviewed (Table 1). Tendon repair was as in Group 1. The same type of splint was used as in Group 1 but the position of wrist was straight. Exercising was commenced on the first postoperative day. Under the supervision of the physiotherapist, gentle active and assisted active movements were begun. The exercises were performed at 4-hourly intervals throughout the day. The method described by Cullen et al (1989) was used. The programme consisted of two active flexion and extension movements of the fingers followed by two passive flexion and two active extension movements. The patients were warned not to use their hands in any other way. The dorsal splint was removed after 6 weeks in most cases. Some of the patients removed the plaster without our permission after 4 or 5 weeks. Then a slight load was permitted and during the following weeks it was increased, the aim being a return to full function by 12 weeks. All patients After careful training, patients were usually discharged from hospital on the fourth postoperative day, provided there were no difficulties with the hand therapy regimen. Some difficult cases were kept in the hospital until the end of the first postoperative week. Thereafter, the patients were reviewed twice a week by the surgeon and

625 Table2--Evaluation of results according ( S t r i c k l a n d ' s original classification system)

Grade Excellent Good Fair Poor

to

Strickland

criteria

% of return

Degrees

85 100 70-84 50 69 < 50

150+ 125 149 90-124 < 90

Active PIP + DIP flexion-extensor lag x 100 = % of normal P I P + D I P motion. 175°

hand therapist jointly. The tendency to develop flexion contractures at the PIP joint was easily detected and the patient was instructed to place a pencil dorsal to the proximal phalanx to act as a fulcrum for extension. All patients were warned to watch for any sudden decrease in active flexion and, if rupture had occurred, the patient was immediately readmitted and exploration and re-repair was undertaken. All of the patients were reviewed prospectively 1 year after the operation. Results were assessed using the original Strickland criteria (Strickland and Glogovac, 1980) (Table2). Maximum grip strength was also measured using a Jamar dynamometer 1 year after operation. Grip strength was measured three times in each hand with the elbow supported in 90 ° of flexion and the wrist held in a neutral position. The mean value in the injured hand was expressed as a percentage of the mean value in the uninjured hand. Statistical analysis of the results were undertaken using the Chi-square test and Student t-test for unpaired groups. A P value of 0.05 or less was considered to indicate a significant difference. RESULTS All patients were reviewed 1 year after injury. Of 41 fingers in the Kleinert group, 14 (34%) were classed as excellent. Eighteen fingers were classed as good and five as fair (44% and 12%, respectively). Poor results were found in four (10%) fingers. Of 47 fingers in the early active mobilization group, 19 (40%) were classed as excellent. 21 fingers were classed as good and 4 as fair (45% and 9%, respectively). Poor results were found in three fingers (6%). Overall, we achieved excellent or good results in 78% of 41 fingers in the Kleinert rubber band group and in 85% of 47 fingers in early active mobilization group. This difference is statistically unsignificant (Table 3). Flexor tendon injuries in two or three digits in the same hand did not affect the results in either group. Whether the profundus tendon alone was injured or both tendons were injured did not affect the result. One neurovascular bundle injury also did not affect the result. The incidence of early rupture was the same in both groups. There were two ruptures in each group (4.9% in the Kleinert group and 4.3% in the early active

626

THE JOURNAL OF HAND SURGERY VOL. 21B No. 5 OCTOBER 1996

Table 3--Results asseded by Strickland criteria

Kleinert group Fingers

ExceHent Good Fair Poor

14 18 5 4

Early active mobilization group %

34 44 12 10

Fingers

%

19 21 4 3

40 45 9 6

mobilization group). All were repaired immediately. The ruptures all occurred during the first 2 weeks after repair. These patients all attended for review. Only one of them was classified as having a successful result. One of them was rated fair and two of them were rated poor. The theoretical risk of early rupture did not prove to be a significant problem in the early active mobilization group. No patients required tenolysis. The degree of residual extension deficit in the two groups is shown in Table 1. There was a residual extension deficit greater than 15° in ten digits (24%) in the Kleinert group and in five digits (11%) in the early active mobilization group. This difference is statistically unsignificant. The mean grip strength was 84% of that of the uninjured hand in the Kleinert group and 90% in the early active mobilization group. This is a statistically significant difference (P<0.01). DISCUSSION The difficulty in achieving satisfactory functional results following the repair of divided flexor tendons in zone 2 is largely due either to the formation of severe adhesions or PIP joint flexion contraetures with extension deficits. Various techniques have been designed over the last 30 years in search of the optimal postoperative mobilization regime. Stegink Jansen and Watson (1993) compared four different system for measurement of active range of motion of the finger after flexor tendon repair in zone 2, and concluded that Strickland's original classification system (Strickland and Glogovac 1980) was the best for scientific and clinical purposes. So et al (1990) analysed five different evaluation methods for assessment of results after flexor tendon repair and suggested use of their modification of Buck-Gramcko's system. But they used the adjusted Strickland classification system (Strickland, 1985) not the original system for evaluation in their study. They thought that this system was not accurate enough and exclusion of MP joint measurement was the disadvantage of this system. Cullen et al (1989) and Stegink Jansen and Watson (1993) said that excluding measurement of the MP joint motion does not result in a loss of discriminative power of this system and Strickland's original system must be preferred over the

adjusted system because its stricter rating of the outcome tends to set higher clinical goals. We also preferred and found Strickland's original system more reliable than his adjusted classification system. May and Silverski61d (1993) carried out a prospective study of 145 digits to investigate the rate of recovery after flexor tendon repair in zone 2. They showed that considerable improvement in active IP joint range of motion occurs more than 3 months after tendon repair in zone 2, especially in the DIP joint. It is therefore worth waiting until 12 months after the operation for assessment of the results and before considering tenolysis. For this reason, we evaluated our patients 1 year after operation. The potential disadvantage of early active mobilization is that rupture may occur if the patient pulls too hard during active flexion (Savage and Risitano, 1989). The work of Savage (1988) suggests that a position of extension of the wrist and flexion of the MP joints produces the least tension on the repaired tendon, at least during attempts at active digital flexion. More extended wrist position was used by Elliot et al (1994) in a large series with reduction of the rupture rate to 4.8% from that of 9.4% originally reported by Small et al (1989). We also observed that it is easier to mobilize the fingers actively with the wrist straight. MP joints were flexed to between 70 ° and 90 ° to prevent full extension of fingers. Our incidence of rupture in the early active mobilization group was not higher than in the Kleinert group (4.3% and 4.9%, respectively). Bainbridge et al (1994) found that the rupture rate in the "controlled active motion" group, was twice that in the "'active extension-passive flexion" group (7.5% and 3.5% respectively). Chow et al (1987) reported 6.8% rupture rate and Saldana et al (1991) also reported 5% rupture rate with "'active extension/passive flexion" method. Silverski61d et al (1993a) modified the traditional Kleinert technique with rubber band traction to all four fingers, irrespective of how many were injured, and they achieved 81% excellent or good results. They had two ruptures in 43 zone 2 injuries. In a new study, Silverski61d and May (1994) used an early mobilization program combining passive and active flexion and obtained 100% excellent or good results except for the two ruptures in 55 zone 2 injuries. During the postoperative rehabilitation program, in their first study they applied a dorsal plaster splint with the wrist flexed from 30 to 45 °. In the later study they preferred to use a dorsal plaster splint with the wrist in a neutral position. We believe, like Silverski61d et al (1993b), that the rate of success will vary slightly in every programme, probably as a result of the psychological and biological characteristics of the patients and the experience of the hand therapist. The reported results of this and previous series indicate that gentle early active mobilization does not increase the possibility of early tendon rupture after zone 2 flexor tendon repair. With the "active extension/passive flexion" method, Singer and Maloon (1988) achieved 80% excellent or

FLEXORTENDONREPAIRSIN ZONE2 good results using Lister's e v a l u a t i o n method. M c L e a n (1987) reported 66% excellent or good results with this regimen. Strickland (1987) also achieved 83% excellent or good results. We achieved 78% good a n d excellent results using the p a l m a r b a r m o d i f i c a t i o n o f the Kleinert regimen. C h o w et al (1987; 1988) in two different studies, using a considerably modified "active extension/ passive flexion" regimen, achieved results superior to a n y other published series. They used Strickland's form u l a a n d published 98% good or excellent results in each study, b u t their series were highly selective, consisting entirely o f military personnel. N o other g r o u p o f patients is in a situation where daily passive flexion/ extension exercises can be performed by a t r a i n e d h a n d therapist. W i t h a n "early active m o b i l i z a t i o n " p r o g r a m m e , Savage a n d R i s i t a n o (1989) achieved 69% excellent or good results in zone 2 t e n d o n lacerations. Cullen et al (1989), Small et al (1989) a n d Elliot et al (1994) also p u b l i s h e d 78%, 77% a n d 79% excellent or good results, respectively, with this p r o g r a m m e . Bainbridge et al (1994) published a comparative study o f the o u t c o m e of flexor t e n d o n repairs mobilized by either a "passive flexion/active e x t e n s i o n " or a " c o n t r o l l e d active m o t i o n " regimen. There was a clear difference in the results achieved in the two groups. The "passive flexion/active e x t e n s i o n " g r o u p achieved only 50% excellent or good results while the " c o n t r o l l e d active m o t i o n " g r o u p achieved 94% excellent or good results. However, these results for a K l e i n e r t regimen are p o o r a n d these a u t h o r s did n o t use the p a l m a r pulley modification. Elliot et al (1994) draw a t t e n t i o n to the similarity of the results in the different studies in which they achieved approximately 70% to 80% good a n d excellent results. We c a n say that all these series report a 4% to 6%, r u p t u r e rate whatever m e t h o d of m o b i l i z a t i o n is used postoperatively. Strickland (1987) p o i n t e d o u t that m o s t efforts to assess the recovery of f u n c t i o n following flexor t e n d o n injury have c o n c e n t r a t e d o n the r e t u r n o f interdigital flexion a n d extension, a n d that little emphasis has been placed o n the r e t u r n of strength. Silverski61d et al (1993b) f o u n d that the m e a n final m a x i m u m v o l u n t a r y grip strength of the injured h a n d was 85% of the u n i n j u r e d h a n d a n d h a n d s with m o r e t h a n one i n j u r e d digit were a p p r o x i m a t e l y , 9 % weaker t h a n h a n d s with a single digit injury. We also f o u n d that m a x i m u m grip strength was reduced after flexor t e n d o n repair. T h e patients in o u r early active m o b i l i z a t i o n g r o u p regained 90% of the strength of the u n i n j u r e d h a n d , which is slightly m o r e t h a n those of our Kleinert g r o u p a n d those reported b y Silverski61d. We f o u n d n o difference between the h a n d s with more t h a n one digit i n j u r e d a n d those with only one digit injured. P e r m a n e n t flexion contracture o f the P I P j o i n t is a c o m m o n c o m p l i c a t i o n following flexor t e n d o n repair. This w o u l d a p p e a r to be particularly likely following elastic b a n d t r a c t i o n because the resting p o s i t i o n of the

627 P I P j o i n t is in 60 to 90 ° of flexion (Small et al, 1989). G a u l t (1987) also c o m m e n t e d o n this problem. Bainbridge et al (1994) reported that 90% o f their patients had less t h a n 15 ° of extension deficit at the P I P j o i n t w h e n mobilized b y the " c o n t r o l l e d active m o t i o n " regimen, b u t only 60% of patients h a d less t h a n 15 ° extension deficit at the P I P j o i n t w h e n mobilized by the "active extension/passive flexion" regimen. Small et al (1989) reported 76% o f patients with less t h a n 15 ° of extension deficit using "early active m o b i l i z a t i o n " regimen. I n this study, there was a n 89% incidence of extension deficit less t h a n 15 ° in the "early active mobili z a t i o n " group a n d a 76% incidence of extension deficit less t h a n 15 ° in the K l e i n e r t group. O u r results c o m p a r e f a v o u r a b l y with previous studies a n d agree with the general impression that loss of full extension is less likely w h e n these injuries are mobilized b y a controlled active m o t i o n regimen. References BAINBRIDGEL C, ROBERTSONC, GILLIESD and ELLIOTD (1994). A comparison of postoperative mobilizationof flexor tendon repairs with "passive flexion-activeextension" and "controlled active motion" techniques.Journalof Hand Surgery, 19B:517 521. BURGE P D and BROWN M (1990). Elastic band mobilizationafler flexor tendonrepair;splintdesignand risk of flexioncontracture.Journalof Hand Surgery, 15B:443-448. CHOW J A, THOMESL J, DOVELLES, MILNORW H, SEYFERA E and SMITH A C (1987). A combinedregimenof controlledmotionfollowing flexortendonrepairin "no man'sland".Plasticand ReconstructiveSurgery, 79:447 453. CHOW J A, THOMES L J, DOVELLE S, MONSIVAISJ, MILNOR W H and JACKSON J P (1988). Controlledmotionrehabilitationafter flexor tendonrepairand grafting.A multi-centrestudy.Journalof Boneand Joint Surgery, 70B:591 595. CULLEN K W, TOLHURST P, LANG D and PAGE R E (1989). Flexor tendonrepair in zone2 followedby controlledactivemobilization.Journal of Hand Surgery, 14B:392 395. EARLEYM J and MILWARDT M (1982).The primaryrepairof digitalflexor tendons. BritishJournalof Plastic Surgery,35: 133-139. EJESKAR A (1984). Flexortendonrepair in no man'sland:resultsof primary repairwith controlledmobilization.Journalof Hand Surgery,gA: 171 177. ELLIOTD, MOIEMENN S, FLEMMINGA F S, HARRISS B and FOSTER A J (1994). The rupture rate of acute flexor tendonrepairs mobilizedby the controlled active motion regimen. Journal of Hand Surgery, 19B: 607 612. GAULT D T (1987). A review of repaired flexor tendons. Journal of Hand Surgery, 12B:321-325. GELBERMAN R H, AMIEL D, GONSALVESM, WOO S and AKESON W H (1981). The influenceof protectedpassivemobilizationon the healing of flexortendons: a biochemicaland microangiograpbicstudy. The Hand, 13: 120-128. GELBERMANR H, WOO S L-Y, LOTHRINGERK, AKESONW H and AMIEL D (1982). Effects of early intermittantpassive mobilizationon healingcanineflexor tendons.Journalof Hand Surgery,7:170 175. GELBERMANR H, BERG J S V, LUNDBORGG N and AKESONW H (1983). Flexortendonhealingand restorationof the glidingsurface.Journal of Boneand Joint Surgery,65A:70-80. KARLANDERL-E, BERGGRENM, LARSSONM, SCIDERBERGG and NYLANDER G (1993). Improvedresults in zone 2 flexortendoninjuries with a modifiedtechniqueof immediatecontrolledmobilization.Journalof Hand Surgery, 18B:26 30. KLEINERT H E, KUTZ J E, ASHBELL T S and MARTINEZE (1967). Primaryrepairof laceratedflexortendonsin "No man'sland". Proceedings, American Society for Surgery of the Hand. Journal of Bone and Joint Surgery, 49A: 577. LISTERG D, KLEINERTH E, KUTZJ E, ATASOYE (1977).Primaryflexor tendon repair followed by immediatecontrolledmobilisation.Journal of Hand Surgery,2: 441-451. MATTHEWSP and RICHARDSH (1976). Factors in the adherenceof flexor tendon after repair. An experimentalstudy in the rabbit. Journal of Bone and Joint Surgery, 58B:230-236.

628 MAY E J and SILVERSKII3LD K L (1993), Rate of recovery after flexor tendon repair in zone 2. A prospective longitudinal study of 145 digits. Scandinavian Journal of Plastic and Reconstructive H a n d Surgery, 27: 89 94. M c L E A N N (1987). Some observations on controlled mobilization following flexor tendon injury. Journal of H a n d Surgery, 12B: 101-104. S A L D A N A J M, C H O W J A, G E R B I N O II P, WESTERBECK P and S C H A C H E R E R T G (1991 ). Further experience in rehabilitation of zone 2 flexor tendon repair with dynamic traction splinting. Plastic and Reconstructive Surgery, 87: 543-546. SAVAGE R (1988). The influence of wrist position on the minimum force required for active movement of the interphalangel joints, Journal of Hand Surgery, 13B: 262-268. SAVAGE R and RISITANO G (1989). Flexor tendon repair using a "six strand" method of repair and early active mobilization. Journal of H a n d Surgery, 14B: 396-399. SILVERSKIt3LD K L, MAY E J and T O R N V A L L A H (1993a) Tendon excursions after flexor tendon repair in zone 2: results with a new c o n t r o l l e ~ motion program. Journal of H a n d Surgery, 18A: 403-410. SILVERSKIOLD K L, MAY E J and O D E N A (1993b). Factors affecting results after flexor tendon repair in zone 2: a multivariate prospective analysis. Journal of H a n d Surgery, 18A: 654 662. SILVERSKII3LD K L and MAY E J (1994). Flexor tendon repair in zone 2 with a new suture technique and an early mobilization program combining passive and active flexion. Journal of H a n d Surgery, 19A: 53-60. SINGER M and M A L O O N S (1988). Flexor tendon injuries: the results of primary repair. Journal of H a n d Surgery, 13B: 269 272. SLATTERY P G and M c G R O U T H E R D A (1984). A modified Kleinert con-

T H E J O U R N A L O F H A N D SURGERY VOL. 21B No. 5 OCTOBER 1996 trolled mobilization splint following flexor tendon repair. Journal of H a n d Surgery, 9B: 217-218. SMALL J O, BRENNEN M D and COLVILLE J (1989). Early active mobilization following flexor tendon repair in zone 2. Journal of H a n d Surgery, 14B: 383-391. SO Y C, C H O W S P, P U N W K, L U K K D K, CROSBY C and N G C (1990). Evaluation of results in flexor tendon repair: a critical analysis of five methods in 95 digits. Journal of H a n d Surgery, 15A: 258 264. S T E G I N K JANSEN C W and WATSON M G (1993). Measurement of range of motion of the finger after flexor tendon repair in zone 2 of the hand. Journal of Hand Surgery, 18A: 411-417. S T R I C K L A N D J W (1985). Results of flexor tendon surgery in zone II. H a n d Clinics, 1: 167-180. S T R I C K L A N D J W (1987). Flexor tendon injuries. Part 5. Flexor tenolysis, rehabilitation and results. Orthopaedic Review, 16: 137-153. S T R I C K L A N D J W and G L O G O V A C S V (1980). Digital function following flexor tendon repair in zone 2: a comparison of immobilization and controlled passive motion techniques. Journal of Hand Surgery, 5: 537-543. W E R N T Z J R, C H E S H E R S P, B R E I D E N B A C H W C, K L E I N E R T H E and BISSONNETTE A (1989). A new dynamic splint for postoperative treatment of flexor tendon injury. Journal of Hand Surgery, 14A: 559 566.

Received: 21 November 1995 Accepted m%r revision: 1 April 1996 Ali Baktir MD, Erciyes Universitesi Tip Fak01tesi, Ortopedik Cerrahi Departmani, El Cerrahisi Bblttmti, 38039, Kayseri, Turkey. © 1996 The British Society for Surgery of the Hand