Development of a synthetic replace ent for the flexor tendon pulleys—An experimental study

Development of a synthetic replacement for the flexor tendon pulleys-An experimental study A method waS developed to reconstruct the fibro-osseous pulleys with Nitex, a synthetic material. Nitex is a closely woven fabric constructed from monofilament nylon fibers. Six adult monkeys (24 digits) had excision of the Al and A2 pulleys; this was followed by reconstruction of the A2. pulley with the Nitex synthetic material. The animals were killed, two at a time, at 4, 8, and 12 weeks to evaluate the effectiveness of the Nitex pulleys. Flexor tendon function was assessed by biomechanical methods with a tensile testing machine to measure the tendon excursion and the work of Hexion (the area under the force-excursion curve) necessary to fully Hex each digit; these parameters revealed that the Nitex pulleys were capable of preventing tendon bow-stringing and did not significantly impair tendon gliding. The breaking strength of the Nitex pulleys was comparable to that of normal A2 pulleys (for monkeys weighing less than 10 kg) and it was sufficient to allow immediate mobilization of the digits postoperatively without fear of pulley rupture. Histologic examination showed minimal foreign body reaction around the Nitex, and the gliding surface of a Nitex pulley was found to be covered with a smooth layer of fibrous tissue with minimal adhesions to the underlying Hexor tendon. The synthetic Nitex pulley appears to have the potential to function as an effective fibro-osseous pulley replacement. (J HAND SURG llA:403-9, 1986.)

WilHam W. Peterson, M.D., Paul R. Manske, M.D., Peggy A. Lesker, B.S., Christopher C. Kain, M.D., and Randall K. Schaefer, M.D., St. Louis, Mo.

T

he function of the fibro-osseous pulleys is to hold the flexor tendons close to the phalanges, thereby preventing them from bow-stringing when the digits are flexed.' These pulleys are essential for normal hand function. Previous experimental work examined the relative importance of the various pulleys. These studies found that the most important pulley is A2 (located at the base of the proximal phalanx), followed by A4 (located at the center of the middle phalanx).2-6 Thetefore, retention or reconstruction of at least these two pulleys has been advocated. 4. 7-9 However, preservation of these pulleys at the time of the operation is often not possible. During tenolysis From the Division of Orthopedic Surgery, Washington University School of Medicine, St. Louis, Mo. Supported in part by NIH Grant ROI AM 32111-02 and NIH GMO 7602-06. Received for publication May 10, 1985; accepted in revised form Sept. 10, 1985. Reprint requests: Paul R. Manske, M.D., Division of Orthopedic Surgery, Washington University School of Medicine, 4960 Audubon Ave. , St. Louis, MO 63110.

the pulleys are often incorporated into the fibrous scar surrounding the tendon and cannot be preserved. Similarly, it may not be possible to repair the pulleys after lacerations or crush injuries to the digits. In these situations, the reconstruction of the pulley mechanism becomes necessary. Many methods of pulley reconstruction have been developed. Most of these involve the use of biologic tissue, such as fascia lata,1O free tendon grafts,II-14 or portions of the extensor retinaculum. 15 Although these materials are capable of functioning as pulleys, they have limitations. Early postoperative mobilization may be delayed to allow for healing of the reconstructed pulley, which is sufficient to resist the forces that are generated during tendon flexion. Also, these biologic materials are not readily available when flexor tendon surgery is performed, without additional incisions or operative sites. Pulley reconstruction using free tendon grafts often results in a bulky replacement pulley because of the large diameter of the tendon grafts. Although extensor retinaculum is less bulky, it can be obtained in only small amounts; thus the number of pulleys that can be reconstructed with it is limited. THE JOURNAL OF HAND SURGERY

403

404

Peterson et al.

The Journal of HAND SURGERY

Fig. 1. a, Nitex pulley passed around the proximal phalanx in the A2 position. b, Ends of Nitex pulley sutured with a O.062-inch spacer between the pulley and underlying tendons to assure proper tension. c, Nitex pulley in its final position with sutures lateral to the phalanx.

Because of the disadvantages of pulley reconstruction with biologic materials, various attempts have been made to develop synthetic pulley replacements. Gonzales l6 used a thin sheet of Teflon to reproduce normal pulleys. However, the Teflon caused a massive foreign-body reaction and also the formation of adhesions around the tendon. Gonzales also tried to reconstruct the pulley using polyethylene tubes; although this caused less of a reaction than the Teflon, the reaction was still more than that observed with the tendon graft pUlleys. He therefore concluded that these materials were poor pulley substitutes. Baderl ? attempted pulley reconstruction using a thin sheet of silicone rubber. These synthetic pulleys were easily positioned and free from any foreign body reaction. However, additional experimental work has suggested that the silicone rubber (even when reinforced with Dacron) may have insufficient breaking strength t9 be a suitable pulley. 18 'Wray and Weeks l9 used knitted Dacron arterial grafts (5 rom in diameter), which were cut into sections, to reconstruct pulleys. However, the Dacron required ingrowth of fibrous tissue to create a smooth gliding surface. Therefore, they thought that this Dacron pulley should be used only in conjunction with a silicone rod and a second-stage tendon graft. Therefore, the need for a suitable synthetic pulley replacement remains. The requirements of such a synthetic pulley replacement are that (1) it is easy to handle and fashion into a pulley, (2) it maintains the normal biomechanics of flexor tendon gliding, (3) it possesses sufficient strength to allow immediate mobilization of

the digit postoperatively and ultimately resist the forces placed on it by flexor tendon action, and (4) it is made of a nomeactive material that does not cause adhesions to form to the tendon. Nitex is a synthetic material that appears to possess all of these characteristics. Nitex (Tetko, Inc., Elmsford, N. Y.) is a closely woven (660 mesh counts per inch) fabric constructed from small-diameter (35 micron) monofilament nylon fibers. It can easily be cut into strips that are similar in width to the pulley being reconstructed. Because of its smooth surface, pulleys made from Nitex should not significantly impede tendon gliding. The tensile strength of Nitex and its minimal elasticity should provide sufficient strength for its use as a pulley replacement. Finally, nylon has been shown to have low tissue reactivity; therefore, Nitex pulleys should not cause formation of adhesions to the underlying tendon. The purpose of this study was to develop a method of using Nitex as a pulley replacement in the hands of nonhuman primates. The effectiveness of the Nitex pulley was evaluated by the following: (1) Biomechanical testing with a tensile testing machine, of the flexor tendon function after replacement of the A2 pulley with a Nitex pulley. By measuring the tendon excursion necessary to fully flex the digit, the amount of tendon bowstringing could be quantitated. Similarly, by measuring the work necessary to fully flex the digit to touch the palm, (i.e., work of flexion) the forces that resist tendon gliding, adhesions, and the resistance of the Nitex pulley itself could be assessed. This biomechanical technique has been used previously to measure flexor tendon

Vol. IIA, No.3 May 1986

Synthetic replacement for flexor tendon pulleys

405

Fig. 2. Hand attached to mounting board on the tensile testing machine. a, Digit being tested fully extended. b, Digit flexed until the fingertip touched the palm .

function after various procedures. 2o, 21 (2) Determination of the baseline breaking strength of the Nitex pulley at the time of its placement and at various postoperative time intervals. These results could then be compared to the breaking strength of normal nonhuman primate A2 pulleys of a similar size. We have previously determined the breaking strength of normal human being pulleys IS with a similar technique. (3) Histologic examination of the Nitex pulley and the underlying tendon. Thus, the Nitex could be evaluated for any foreign body reaction, and the extent of the adhesions between the Nitex and the underlying tendon could be visualized. Materials and methods Operative technique. Six adult monkeys (24 digits) were initially tranquilized with Ketamine, 10 mg/kg that was injected intramuscularly. Adequate anesthesia was then maintained with intravenous Pentobarbatol, 5 mg/kg, as needed. The right hand and forearm were then shaved and prepared with Betadine. A pneumatic tourniquet and sterile surgical technique were used. A palmar Brunner zigzag incision was made from the distal palmar crease to the distal interphalangeal (DIP) joint of the index,'long, ring, and small fingers. The flexor sheath of ea,ch digit was carefully exposed, leaving the neurovascular bundles intact. The Al and A2 pulleys were identified and completely excised. The remainder of the pulley system was left undisturbed. The Nitex pulley was made by cutting 10 mm wide

strips of Nitex and folding them lengthwise, thus forming a 5 mm wide, two-layered synthetic pulley. The folded margin of the pulley was placed distally, thereby reducing the possibility of tendon fraying at the pulley edge. The Nitex pulley was gas sterilized before its use. At the base of the proximal phalanx (the site of the normal A2 pulley), the path for the Nitex pulley was created by dissection with scissors on each side of the phalanx onto the dorsum of the digit. The Nitex pulley was then passed around the proximal phalanx, palmar to the extensor hood. (Fig. 1, a). After the Nitex pulley had been passed around both the sublimis and the profundus flexor tendons, the ends were overlapped and sutured with three No. 5-0 nylon sutures. To adjust the tension of the Nitex pulley around the tendon, a 0.062 inch diameter Kirschner-wire was placed between the pulley and the tendon to avoid pulling the pulley too tight (Fig. 1, b). Previous experience showed that this spacer was needed between the Nitex pulley and the tendon to allow free tendon gliding without constriction, which could lead to rupture or severe scarring of the tendon. After the Nitex pulley had been sutured with the appropriate degree of tension, it was rotated around the finger so that the sutures were lateral to the phalanx and would not interfere with tendon gliding (Fig. 1, c). After assuring that the tendon was able to glide freely under the Nitex pulley, the skin was closed with No. 5-0 Dexon sutures. After the operation a bulky dressing was placed on

406

Peterson et

The Journal of

at.

.... a!:::::::::. ............ .:::!:::::::!::: . .................. ..:.:.:.:. :! :::::::!: ::::::::. :.:.:. :.:.: .. ...................... :::::::: ::::: i:i:::::: :::.

HAND SURGERY

:::::::.

Force (kgf)

::: :::: ::::::::::::: !:::: :!::::.

.

::~

_::: 1

::::: ::: :::::::: ::::::::::::::::::.. . .. :::!:' ::: ................. ~ ... ::::::: ::::: :::: ::::::' :::Work of FleXlon::::::::::::::::::::::l ::::::: ::::::::: ::::::: :::::: ::: :::: ::::: :::::: ~ ::::::::::::::::::: ::::: :::::::: ::::::::::::::: ~

:::::::::::::!::::: ::::::::!::::::::::::: !:::::

1

::::::::::::::::: ::::::: :.:.:.:.:.:.:.:.: ::::.!::::: :::::: ::::::: :.: ... :.:.:.:.:.:.:.:.:. .:.:.: ~

t.

t

Tendon Excursion (cm) full extension full flexion Fig. 3. Force-excursion curve showing the biomechanical parameters of tendon excursion (the excursion of the tendon required to fully flex the digit) and work of flexion (the area under the curve representing the forces that resist tendon gliding).

the hands with the digits in a slight amount of flexion. The dressing was removed after 48 to 72 hours and the monkeys were allowed unrestricted use of their hands. Since the dominant hand had been operated on, the monkeys gradually increased the use of this hand to feed themselves. In this manner, active motion of the digits was encouraged. At 4, 8, and 12 weeks, two of the monkeys were killed. The function of the Nitex pulleys was then evaluated: (1) biomechanically, (2) by determination of their breaking strength, and (3) histologically. Biomechanical assessment. Biomechanical assessment of flexor tendon function was carried out with a Scott tensile testing machine (Model JXL-lO I). Both the operated hand (with the A2 Nitex pulleys) and the unoperated, contralateral hand (control) were tested. Both hands were prepared by excising the palmar. skin to the level of the distal palmar flexion crease. The underlying flexor digitorum superficialis and lumbrical muscles were also excised, thus exposing the flexor digitorum profundus tendons. The profundus tendons were transected at the level of the distal forearm, and the tendons to the individual digits (index, long, ring, and small) were identified and separated. The transverse carpal ligament wasleft intact to prevent bow-stringing of the tendons at the wrist. The hands (with the fingers pointing down) were then placed on a mounting board that was attached to the lower clamp of the tensile testing machine (Fig. 2, a and b). A 15 gm counterweight was then suspended

from the tip of the digit to be tested to fully extend the metacarpophalangeal (MP) , proximal interphalangeal (PIP), and distal interphalangeal (DIP) joints. The proximal end of the profundus tendon was attached to a 5 kg load cell that was connected to the crosshead of the tensile testing machine. The crosshead was then advanced at a constant rate of 10 centimeters per minute until the fingertip touched the palm. The output of the load cell was recorded as force versus tendon excursion on a chart recorder (Fig. 3). The following values were measured: (1) Tendon excursion-the excursion expressed in centimeters (cm) of the tendon required to fully flex the digit, and (2) Work of flexion-the area under the force-excursion curve expressed in kilogram force-centimeters (kgf-cm) integrated with a polar planimeter. For the normal control hands, five repetitive runs were recorded and the average value of the tendon excursion and work of flexion was calculated for each digit. However, for the hands that had a Nitex pulley replacement, the forces needed to flex each digit were significantly higher during the first excursion of the tendon than during subsequent excursions. This was presumably due to the breakage of adhesions that occurred during the first excursion. 20 Therefore, for the experimental hands, only the data from the first run was used since this was considered to reflect the most accurate assessment of the extent of adhesion formation. Since the biomechanical parameters of tendon excursion and work of flexion are dependent upon the size and length of the digit, it was necessary to compare the values that were obtained in the experimental digit with those of the contralateral control digit (previously we showed that the biomechanical parameters of tendon excursion and work of flexion for an individual digit are not significantly different between digits of unoperated right and left hands).22 Results were then expressed as a percent difference between the experimental digit and the contralateral control: Percent difference = _E_x,,-pe:..:r:..:im:..:e:..:n:..:ta:..:l_v...:.:a:..:lu:..:e_-_C=-o:..:n...:.:tr:..:o_l_v:..:al-=.ue x 100. Control value

With this method, the mean value of tendon excursion and the work of flexion could be obtained for all the digits that had undergone A2 Nitex replacement at the different time intervals. Determination of breaking strength. Determination of breaking strength followed completion of the biomechanical studies. The palmar digital skin was carefully removed from the operated hands. The flexor sheath was then excised, with only the Nitex A2 pulley left intact. The proximal phalanx was secured in the

Vol. I lA, No.3 May 1986

Synthetic replacement for flexor tendon pulleys

407

Table I. Biomechanics of Nitex pulleys 4 Weeks (n = 7)

8 Weeks (n = 7)

3.00 ± 5.09 55.33 ± 12.48

8.12 ± 7.04 32.14 ± 14.86

Tendon excursion Work of flexion

12 Weeks (n

= 7)

Overall (n =21)

1.34 ± 3.86 22.56 ± 10.81

4.16 ± 3.07 36.67 ± 7.67

Results are the percent difference between the experimental and control digits expressed as mean :t SEM.

Table II. Effectiveness of A2 Nitex pulleys Normal digits Excision of A1 and A2 pulleys; A2 Nitex replacement (n = 21)

Excision of A1 (n = 16)

Excision of A1, A2 (n = 15)

4.16 ± 3.07

-0.64 ± 2.04

20.56 ± 3.76

Tendon excursion

Results are the percent difference between the experimental and control digits expressed as mean :t SEM.

Table III. Breaking strength of Nitex pulleys

Breaking strength

Baseline (n = 5)

4 Weeks

8 Weeks

12 Weeks

Overall

(n = 8)

(n = 6)

(n = 6)

(n = 20)

4.01 ± 0.46

3.45 ± 0.22

4.74 ± l.l8

3.69 ± 0.85

3.91 ± 0.38

(kgf) Results are expressed as mean :t SEM.

lower clamp of the tensile testing machine. A length of tendon was then looped through the synthetic pulley and the ends were secured to a 100 kg load cell that was connected to the crosshead. The crosshead was again advanced at a constant rate of 10 centimeters per minute. The force required to rupture the synthetic A2 pulleys was measured in kgf and recorded on the chart recorder. The mean values for the breaking strength at 4, 8, and 12 weeks was thus obtained. To obtain baseline values for the breaking strength of the Nitex pulleys, a folded 5 mm Nitex pulley was placed around the proximal phalanx and sutured in the described manner. In a similar manner, a loop of tendon was passed through the Nitex pulley and the baseline breaking strength of the synthetic pulley was determined. Finally, with the same technique, the mean breaking strength of the normal A2 pulleys in the contralateral control hands was calculated. Histology. At 8 and 12 weeks, one digit was randomly selected for histologic examination and fixed in 10% buffered formalin. These digits were not biomechanically tested, so that any adhesions that may have developed between the ,Nitex pulley and the underlying tendons were not disrupted. The bone was decalcified with a solution of 50% formic acid and 20% sodium citrate in a 1: 1 ratio. The digit was then embedded in paraffin and serial 6-micron cross-sectional sections were made through the phalanx, tendon, and Nitex pul-

ley. The sections were then stained with van Gieson's solution by standard histologic techniques. Results Biomechanical assessment of tendon function. Twenty-two digits that had a A2 Nitex replacement were available for biomechanical testing (the remaining two digits were reserved for histologic study). However, one of the digits (from an animal killed at 4 weeks) could not be flexed because of scarring at the pulley site. This was believed to be caused by placing the Nitex pulley too tightly since the 0.062 inch spacer had not been used. The biomechanical determination of flexor tendon function for the remaining twenty-one digits compared to the contralateral control is summarized in Table I. After the replacement of the A2 pulley with a Nitex pulley, there is a minimal increase in tendon excursion at 4, 8, and 12 weeks. When the results of all of the monkeys' tests were combined, the tendon excursion of the Nitex pulley replacements is only 4% greater than that of the contralateral controls. Therefore, the synthetic pulley appears to be functioning well, since the amount of tendon bow-stringing is related to the increase in tendon excursion. Similarly, when the work of flexion is examined (Table I), at 4 weeks it is 55% higher with the Nitex pulley in place as compared to the control digits. However, in time, the work of flexion gradually diininishes,

408

The Journal of HAND SURGERY

Peterson et at.

Fig. 4. Cross-section photomicrograph of the Nitex pulley (P) and underlying flexor tendon (T). The Nitex pulley is covered with a smooth, thin layer of fibrous tissue without a significant inflammatory response. No adhesions are present between the Nitex pulley and the tendon (original magnification, X 200).

so that by 12 weeks it is increased by only 22%. Whether this downward trend is caused by the breakage and remodeling of adhesions or the ingrowth of fibrous tissue over the Nitex creating a smoother gliding surface cannot be determined biomechanically. Nonetheless, it is apparent that at 12 weeks the Nitex pulley replacement does not cause significant impairment of tendon gliding. Previously we determined the effect of excising either the Al pulley alone or both the Al and A2 pulleys in normal monkey cadaver digits with similar biomechanic parameters. 22 In Table II these results are compared with the overall results that were obtained in our present study. This demonstrates that the tendon excursion in the experimental digits (digits that had undergone excision of the Al and A2 pulleys with a Nitex A2 replacement) is not significantly different from that of normal cadaver digits that had undergone release of only the Al pulley. Therefore, it seems that the Nitex pulley replacement is able to function nearly as effec-

tively as the normal A2 pulley. By contrast, there is a significant (p < 0.01) improvement in tendon excursion when these experimental digits are compared to normal cadaver digits that had undergone release of both Al and A2 (with no A2 replacement). Therefore, it is apparent that without the Nitex pulley there is a significant loss of flexor tendon function. Breaking strength. The breaking strength of the 5 mm folded Nitex pulleys is summarized in Table III. There was no significant difference in the breaking strength at any of the time intervals (baseline, 4, 8, or 12 weeks). The Nitex pulley, therefore, appeared to retain its initial strength for at least 3 months in vivo. The normal A2 pulley of primates that weighs less than 10 kg is approximately 5 mm wide. Its breaking strength was 4.75 ± 0.39 (n = 23). Although this is slightly more than the average breaking strength of the Nitex pulley (3.91 ± 0.38), none of the synthetic pulleys ruptured despite early active motion of the operated hands. We may conclude, therefore, that the Nitex replacement pulleys have sufficient strength to resist the forces placed on them by active flexion, even from the time of their initial placement. Histologic findings. The histologic appearance of the Nitex pulley and underlying tendon at 8 and 12 weeks is shown in Fig. 4. It was noted that the Nitex pulley was surrounded by a thin layer of fibrous tissue that was composed of collagen, fibroblasts, and an occasional multinucleated giant cell. This is comparable to the reaction seen after any foreign material has been implanted and represents the normal healing response. Of more importance, essentially there was no inflammatory response, either acute or chronic, associated with the Nitex pulley. In addition, the gliding surface of the Nitex pulley was covered with a smooth, thin layer of fibrous tissue and minimal adhesions were seen between the pulley and the underlying tendon. The flexor tendon itself appeared normal, except for occasional irregularities of the epitenon surface layer. Therefore it did not appear that the Nitex pulley would significantly impair tendon gliding. Discussion This study demonstrates that it is possible to effectively reconstruct the A2 pulley in primates with the synthetic material Nitex. The Nitex is easily fashioned into a pulley, and the technique of placing it around the digit is not difficult. After placement of the Nitex pulley, essentially normal flexor tendon excursion is restored as measured biomechanically. Thus, the Nitex pulley is able to effectively prevent tendon bowstringing.

Vol. l1A, No.3 May 1986

A major advantage of the Nitex pulley is that it does not interfere significantly with tendon gliding. This was shown both biomechanically and histologically. The moderate increase in work of flexion seen initially gradually decreased with time. Therefore, it was evident that the forces that resist tendon gliding (such as adhesions and the Nitex material itself) decreased. The reason for this decrease is suggested by the histologic appearance of the Nitex pulley and the flexor tendon. Since the Nitex gradually became covered with a smooth fibrous layer, it would be expected that the tendon would glide more freely. In a similar manner, since the Nitex elicits only a minimal inflammatory response, few adhesions developed which would restrict tendon gliding. Another advantage of the Nitex pulley is its strength. Even from .the time of its initial placement, the Nitex replacement pulley is nearly as strong as a normal A2 pulley (for nonhuman primates weighing less than 10 kg) and was able to resist the forces placed on it during active flexion. Therefore, use of the Nitex pulley would allow for immediate postoperative motion after flexor tendon surgery without fear of rupturing the reconstructed pulley. In addition, the Nitex appears to maintain its strength over time. Additional studies are required to fully assess the efficacy of Nitex pulley replacements in human subjects. The average length of the human A2 pulley is 17 mm with an average breaking strength of 14.0 kg,18 compared to width of 5 mm and a breaking strength of 4.75 kg for the monkeys that were examined in this study. Therefore, it is important that studies be carried out to determine the optimum size and strength of synthetic pulley replacements in human patients. Modifications of the Nitex pulleys may then be necessary to meet these functional requirements. Furthermore, the long-term effectiveness of Nitex pulleys in maintaining normal tendon biomechanics needs to be examined along.with a comparison of the proposed synthetic Nitex pulleys with the biologic pulleys currently being used. Nonetheless, the synthetic Nitex pulley appears to have the potential to function as an effective fibro-osseous pulley replacement. The authors appreciate the excellent biomechanical consultation provided by John L. Kardos, Ph.D ., Director, Biomaterials Laboratory, Department of Chemical Engineering, Washington University, and the technical assistance of AnnaMarie Schrnoeker, Qivision of Orthopedic Surgery.

REFERENCES 1. Brand PW: Biomechanics of tendon transfer. Orthop Clin North Am, 5:205-30, 1974 2. Barton NJ: Experimental study of optimal location of

Synthetic replacement for flexor tendon pulleys

409

flexor tendon pulleys . Plast Reconstr Surg 43:125-9, 1969 3. Brand PW, Cranor KC, Ellis JC: Tendon and pulleys at the metacarpophalangeal joint of a finger. J Bone Joint Surg [Am] 57:779-84, 1975 4. Doyle JR, Blythe W: The finger flexor tendon sheath and pulleys: Anatomy and reconstruction. AAOS symposium on tendon surgery in the hand. St. Louis, 1975, The CV Mosby Co, pp 81-9 5. Manske PR, Lesker PA: Palmar aponeurosis pUlley. J HAND SURG 8A:259-63, 1983 6. Hunter JM: Anatomy of flexor tendons-pulley, vincular, synovia, and vascular structures. In Spenser M, editor: Kaplan's functional and surgical anatomy of the hand, ed 3. Philadelphia, 1984, JB Lippincott Co, pp 65-92 7. Rank BK, Wakefield AR, Hueston JT: In Surgery of repair as applied to hand injuries, ed 4. Baltimore, 1973, The Williams & Wilkins Co, pp 248-9 8. Hunter JM, Cook JF: The pulley system. Rationale for reconstruction. In Strickland JW, Steichen JB, editors: Difficult problems in hand surgery. St. Louis, 1982, The CV Mosby Co, pp 94-102 9. Schneider L: Staged tendon reconstruction. Hand Clin North Am 1:109-20, 1985 10. Cleveland M: Restoration of the digital portion of a flexor tendon and sheath in the hand. J Bone Joint Surg 15:76265, 1933 11. Weckesser E: On technique of tendon repair. In Flynn JE, editor: Hand surgery. Baltimore, 1966, The Williams & Wilkins Co, pp 184-94 12. Boyes JH: In Bunnell's surgery of the hand, ed 5. Philadelphia, 1970, JB Lippincott Co, 403-4 13. Kleinert HE, Bennett JB: Digital pulley reconstruction employing the always present rim of the previous pulley. J HAND SURG 3:297-8, 1978 14. Hunter JM: Staged flexor tendon reconstruction. J HAND SURG 8:789-93 , 1983 15 . Lister GD: Reconstruction of pulleys employing extensor retinaculum. J HAND SURG 4:461-4, 1979 16. Gonzalez RI: Experimental use of Teflon in tendon surgery. Plast Reconstr Surg 23:535-9, 1959 17. Bader KF, Seth G, Curtin JW: Silicone pulleys and underlays in tendon surgery. Plast Reconstr Surg 41 :15764, 1968 18. Manske PR, Lesker PA: Strength of human pulleys. Hand 9:147-52, 1977 19. Wray RC, Weeks PM: Reconstruction of digital pulleys. Plast Reconstr Surg 53:534-6, 1974 20. Lane 1M , Bora FW, Black J: cis-Hydroxyproline limits work necessary to flex a digit after tendon injury. Clin Orthop 109:193-200, 1975 21. Lane JM, Bora FW: Gliding function following flexortendon injury. J Bone Joint Surg [Am] 58:985-90, 1976 22. Peterson WW, Manske PR, Bollinger BA , Lesker PA , McCarthy JA: The effect of pulley excision on flexor tendon biomechanics . J Orthop Res (in press) 1986