Mechanical properties of various circumferential tendon suture techniques

Mechanical properties of various circumferential tendon suture techniques

MECHANICAL CIRCUMFERENTIAL P R O P E R T I E S OF V A R I O U S TENDON SUTURE TECHNIQUES H. KUBOTA, M. AOKI, D. L. PRUITT and P. R. MANSKE From the...

1MB Sizes 0 Downloads 31 Views

MECHANICAL CIRCUMFERENTIAL

P R O P E R T I E S OF V A R I O U S TENDON SUTURE TECHNIQUES

H. KUBOTA, M. AOKI, D. L. PRUITT and P. R. MANSKE

From the Department of Orthopaedic Surgery, Washington University School of Medicine, St Louis, Missouri, USA We evaluated the mechanical properties of six different circumferential tendon sutures with a variable number of suture strands. Seventy-two human cadaver flexor profundus tendons were cut and repaired using only a 6/0 polypropylene circumferential suture. The six running suture techniques were: Simple; Simple-locking; Lembert; Halsted; Cross-stitch; Lin-locking; using 10, 14 and 18 suture passes. The increased suture passes increased both tensile and gap strengths. The tensile strength of the Lin-locking technique (29 to 63 N) was significantly stronger than the others, followed by Cross-stitch (27 to 38 N), Halsted (21 to 27 N), Lembert (20 to 27 N), Simple (11 to 22 N) and Simple-locking (10 to 17 N). The gap strength values were between 3 to 14 N, with no significant differences between the techniques. The resistance to gliding, as measured by work of flexion, was not affected by the number of strands. However, the Lin-locking significantly increased the resistance to gliding (33-36%) compared to the other techniques (6-21%); there were no significant differences between these five techniques. Journal of Hand Surgery (British and European Volume, 1996) 21B: 4:474-480

aration and testing. The hand was fixed to a wooden frame using Kirschner wires (Fig 1 ). The tendon sheath was opened between the A2 and A4 pulleys and the flexor profundus tendons were surgically exposed. All intrinsic muscles and synovium attached to the profundus tendons proximal to the A1 pulley were removed. The carpal tunnel was opened and all tendons were identified and dissected free. All skin incisions were closed without tendon sheath repair. The hand with the wooden frame was mounted on a Scott Tensile Testing Machine (model JXL 101, Scott Tester, Providence, RI, USA). The work of flexion (WoF, see below) of each finger was measures as the control value for each finger. This value reflects the biomechanical work applied to the tendon during finger flexion (Carver et al, 1968; Lane et al, 1976; Peterson et al, 1986a, 1986b, 1986c and 1990; Dunlap et al, 1990; Aoki et al, 1995). The hand was maintained in the test frame during the subsequent surgical procedures. The skin incision of each finger was opened and the flexor profundus tendons were again exposed between the A2 and A4 pulleys. A sharp transverse cut was made in zone 2 and the tendons were repaired using one of the following six circumferential suture techniques according to described protocols (Fig 2):

Current flexor tendon suture techniques usually consist of a core suture and a peripheral circumferential suture (Kessler, 1973; Kleinert et al, 1975; Tsuge et al, 1977; Lister et al, 1977; Tajima, 1984; Strickland, 1985; Lee, 1990; Messina, 1992; Silverski61d and Andersson, 1993; Leddy, 1993). The term 'circumferential' is preferred to 'epitenon', since the suture technique includes more than the single cell layer of epitenon cells. Several biomechanical studies have evaluated these combined repair techniques (Urbaniak et al, 1975; Ketchum et al, 1977; Wray and Weeks, 1980, Wade et al, 1986 and 1989; Defino et al, 1986; Haddad et al, 1988; Pruitt et al, 1991; Trail et al, 1992; Gordon et al, 1992; Bhatia et al, 1992; Noguchi et al, 1993; Greenwald et al, 1994; Wagner et al, 1994). Other studies have shown the importance of the circumferential suture (Lister et al, 1977; Wade et al, 1986 and 1989; Lin et al, 1988; Pruitt et al, 199t; Mashadi and Amis, 1992). However, few studies have compared only the circumferential suture techniques with respect to tensile properties (Wade et al, 1986; Lin et al, 1988; Masha~ti and Amis, 1992) and there have been no studies of the gliding properties of flexor tendons following circumferential suture repair. The purpose of this study was to evaluate the mechanical properties of six different peripheral circumferential tendon suture techniques, as well as the number of suture strands, using tensile strength, gap strength and work of flexion, as measurement parameters.

1. 2. 3. 4. 5. 6.

MATERIALS AND METHODS Experimental design

Simple (Lister et al, 1977) Simple-locking (Leddy, 1993) Lembert mattress (Lister et al, 1977) Halsted mattress (Wade et al, 1989) Cross-stitch (Silverski61d and Andersson, 1993) Lin-locking (Lin et al, 1988).

Each stitch was inserted in the surface layer between 5 mm and 6 mm from each cut end at a depth of lmm to 1.5 mm using 4 x loop magnification. 6/0 polypropylene monofilament suture material (Prolene; Ethicon, Somerville, N J, USA) was used for all techniques in this study.

Seventy-two flexor profundus tendons from 18 adult human cadaver hands were used in this study. The hands were fresh frozen and stored at -20°C. The specimens were thawed to room temperature (25°C) and then kept moist with saline solution during prep474

475

TENDON SUTURE TECHNIQUES

a

b

k.

C

e Fig 2

Fig 1

Hand-mounted on a wooden frame attached to the tensile testing machine.

For each technique, 10, 14, or 18 suture strands were inserted per tendon. The number of strands was defined as the number of sutures crossing between the lacerated tendon ends; Fig 2 shows four strands in all techniques. To prevent loosening of the knot, the sutures were tied with four square surgical knots ( 1 = 1 - - 1 = 1 ; Tera and Aberg, 1976). The skin was closed by sutures. The post repair WoF of each specimen was again measured. All repaired tendon specimens were carefully dissected out of the hand and evaluated for tensile properties at the repair site. Measurement of work of flexion

The wooden frame was secured in the lower clamp and the proximal end of flexor profundus tendon was secured in the upper clamp on the tensile testing machine (Fig 1). A 25 g counter weight was attached to the finger distally to allow full finger extension. The cross-head was advanced at a constant speed of 5 cm/min. A force versus tendon excursion curve was plotted on an X-Y

/

k

d

f

),

The six circumferential suture techniques in this study. Note all techniques show four suture strands. (a) Simple; (b) Simplelocking; (c) Lembert; (d) Halsted; (e) Cross-stitch; (f) Linlocking.

chart recorder from full extension until the finger tip touched to the palm. The value of WoF in newtonmetres was calculated as the area under the force-excursion curve using an IBM personal computer. The force excursion curves initially showed a gradual elevation as finger flexion was initiated, then reached a plateau value throughout the mid-range of finger flexion, and increased in a logarithmic manner at terminal flexion. In order to standardize WoF measurements before and after tendon repair, we designated tendon excursion as the distance from the initiation of finger flexion until the point when the force measured twice the plateau value on the control (i.e. pre-repair) specimens. This defined excursion distance was then applied to the post repair specimens (Fig 3). Specimens were preconditioned by pulling them four times to full finger flexion before testing. To determine the increase in WoF due to the suture repair of the tendon, we compared the post repair values to the control values, utilizing the following formula: post repair WoF--control WoF x 100 control WoF This value can be considered as the increased mechanical drag caused by the tendon repair going through the fibroosseous canal. Load-to-failure testing of suture techniques

Following the measurement of WoF, each tendon specimen was harvested and secured in tendon clamps on

476

THE JOURNAL OF HAND SURGERY VOL. 21B No. 4 AUGUST 1996 Table 1--The mechanical properties of repaired flexor tendons

/

O h

Suture technique

Number of strands

Tensile strength (N)

Gap strength (N)

Increased work of flexion (%)

~'" (3) Lin-locking

(1)

I• Fig 3

Cross-stitch Tendon excursion

>

Schema demonstrates the force-excursion curve of control (pre-repair) condition. The measured area under the curve represented the work of flexion (i.e. the area of oblique lines); (1) the initiation of finger flexion; (2) plateau force; (3) twice plateau force. Note: apply the defined excursion distance of pre-repair specimen to the post-repair specimen measurement.

the same tensile testing machine, and loaded to failure at a constant cross-head speed of 5 cm/minute. The tensile forces in newtons (N) at which the gap formation was first observed was recorded as the gap strength, and the tensile force at which the repair failed was recorded as the tensile strength.

Data analysis Statistical analysis of data obtained from biomechanical testing was performed using a two-factor factorial analysis of variance (ANOVA) followed by the Fisher's protected least significant difference post-hoc test (Fisher's PLSD), with significance set at P=0.05 (Stat View IV, Abacus Concepts, Berkeley, CA, USA). RESULTS

Load-to-failure testing of suture techniques The results are noted numerically in Table 1 and graphically in Figs 4 and 5. All specimens failed at the repair site.

Halsted Lembert Simple Simple-locking

10 14 18 10 14 18 10 14 18 10 14 18 10 14 18 10 14 18

29.5 ±2.8 4.6_+3.0 40.5-t-2.0 8.7-t-1.9 63.4_+1.9 12.8_+2.1 27.8---2.3 6.1_+1.6 30.3-+1.1 6.3-+0.8 38.2_+3.7 9.9_+2.0 21.3_+0.9 4.2___1.3 23.8_+2.3 6.5+__1.0 27.9_+2.6 11.8+_2.8 20.8_+2.6 5.7+__0.7 22.3_+1.9 7.7_+2.7 27.7+3.3 10.1+_4.4 11.6__+2.0 6.6_+0.8 14.0_+1.9 9.3_+1.1 22.1-+4.2 14.0_+3.8 10.8 -+0.8 3.0 _+0.1 11.3:t_0.9 4.7 ±0.5 17.7-+1.0 8.7+1.2

36.8 ± 13.6 33.5±10.1 36.8_+11.6 12.6__3.1 18.0_+2.0 21.7_+3.0 16.7+_6.1 22.1±5.7 21.8+__6.1 12.6___4.1 12.0_+4.1 14.6,,,5.8 14.9_+9.1 14.7_+2.0 15.2_+6.9 6.2_+4.5 8.8 ± 5.5 13.1_+5.1

All groups n =4 (mean+SE).

among all techniques, except between the Halsted and the Lembert and between the Simple and the Simplelocking. The increased number of strands produced significantly (Fisher's PLSD; P < 0.015) increased tensile strength values.

Gap strength The Simple technique had the strongest gap strength (6.6-14.0 N) and the Simple-locking (3.0-8.7 N) had the lowest values at all numbers of strands; the other techniques had intermediate values. However, the factor of techniques did not significantly (ANOVA; P=0.17) affect gap strength values. On the other hand, the gap strength of all techniques was significantly (ANOVA; P < 0.001) increased by increasing the number of suture strands. Eighteen suture strands were significantly (Fisher's PLSD; P<0.015) stronger than 10 and 14 suture strands, but there was no significant differences between 10 and 14 suture strands.

Tensile strength The tensile strength of the Lin-locking technique (29.5-63.4 N) was stronger than any other technique at all suture strand numbers. Lesser values were observed in the Cross-stitch (27.8-38.2N), followed by the Halsted (21.3-27.9 N), the Lembert 20.8-27.7 N), the Simple ( 11.6-22.1 N) and the Simple-locking (10.8-17.7N) techniques. The increased number of strands increased the values of tensile strength for all suture techniques. The two factors of techniques and suture strand numbers significantly (ANOVA; P < 0.0001) affected tensile strength values. There were significant (Fisher's PLSD; P<0.0002) differences

Increased work of flexion The increased WoF values for each group are summarized numerically in Table 1 and graphically in Fig 6. The factor of techniques significantly (ANOVA; P = 0.0004) affected these values, but the factor of suture strand number did not significantly (ANOVA; P > 0.6) affect them. The Lin-locking technique had significant (Fisher's PLSD; P<0.074) increased WoF values (33.5-36.8%) compared to the other five techniques. The Halsted technique had the second highest increased values (16.7-21.8%), followed by the Cross-stitch, the Simple, the Lembert and the Simple-locking. There were

TENDON SUTURE TECHNIQUES 477

Tensile strength

"---.,G-,---

Lin-|ocking

.-.-.¢,.__

Crots-t~tch

/-hdsced

60-

Simple

:

Simple-locking

20-

.k

:::::::::::::::::::::::::::::::::::: 0..J

...............................

L I !

10

14 Numbero f s u ~ strands

Fig 4

18

Tensile strength (mean values__+SE) of different suture techniques at each number of strands tested.

no significant differences among all suture techniques except the Lin-locking technique. DISCUSSION This study demonstrated that more suture strands crossing the repair site significantly increased tensile strengths of all suture techniques. This was true in particular for the Lin-locking technique, which more than doubled the tensile strength with an increase from I0 to 18 strands. The tensile strength results were grouped into four subgroups; (1) the Lin-locking was the strongest techtuque (29-63 N); (2) the Cross-stitch was the second (27-38 N); (3) the Halsted and the Lemb third (20-27 N ~ ---, . . . . . . . ert were the ,~. ~ .. ~auu~,UlOUowedo theSim ~lmple-lockmg (10_22 N. ~:;~ ~~ ~,Y . pie and the - , ~,~ -J). Jnese results probably reflect the different manner in which the sutures grasp the longitudinal collagen bundles. The Lin-locking is a grasping type suture, the Cross-stitch is grasping/mattress tvne sm . . . . . ~-. . . . a modified ....... , ~uc raalstea and the Lembert are mattress sutures, and the Simple and the Simple-locking are longitudinally oriented in line with the collagen bundles (Fig2). The observation that Simple-locking had weaker tensile properties than the Simple technique indicates that the locking configuration actually aligns the suture strands directly in line with the collagen bundles, thereby reducing the holding power

compared to the Simple suture configuration which passes ob/iquely across the collagen bundles. The gap strength values also increased according to the number of suture strands. However, there were no significant differences among the individual suture techniques, with values ranging between 3.0-6.6 N for 10 suture strand repairs, and between 8.7-14.0 N for 18 suture strand repairs. This data indicates that the circumferential tendon suture does more than 'tidy up' the repair site (Leddy, t993); it improves both tensile and gap strength of the tendon repair. The increased resistance to tendon gliding within the tendon sheath was noted by an increase in tile WoF. The WoF quantitatively represents the sum of all forces that resist finger flexion during tendon gliding. The parameter WoF has been studied in previous experiments (Carver et al, 1968; Lane et al, 1976; Peterson et al, 1986a, 1986b, 1986c and 1990; DuMa e . Aokietal, 19951 Ther .... ~. . . . . . . . p t al 1990; ,~ou,~ u~ tins study aemonstrated that the type of circumferential suture techniques affected the WoF values. The Lin-locking technique showed increased WoF values of 33% to 36%, significantly greater than the other techniques, which increased the WoF values 6% to 22%. This probably reflects the increased amount of exposed suture material in the Linlocking technique.

THE JOURNAL OF HAND SURGERY VOL. 21B No. 4 AUGUST 1996

478

Gap strength N

Lin-loeking

20-

Crou-stit~h

.... O----

llalmd

l

Lmml~,'t

-L

15-

10-

5-

0

I

I

I

10

14

18

N u m ,b a o f m t u ~

Fig 5

Gap strength (mean values _+SE) of different suture techniques at each number of strands tested.

Increased Work of Flexion

...........

Lin-I~king-18

...

All IlY

............. .................. .......... .............................~

-14 -10

m V

................................

C r o u - a t i t c h - 18

,.,.,.,.,.,.

m

-14 -10 ..-...................-.-.-.-...-...-......,.,.,.,.,.J

Halsted-18

,

-14 ..........

-10

2mi . ~ . * . i . * . i . ~ . o . o ~ * ~ * 2 ~

Lembc~-18 8

-14 -10

.-.-.-.-.-.-.-.-..-,-.-.-.-.-.-.

Simple-18 -14 -10

Simple-locklag-18 -14 -10 0

Fig 6

snm~

10

20

i

i

i

!

30

40

50

60

The bars show mean values_+SE for the increased work of flexion of six suture techniques at each number of strands tested.

Lin et al (1988) reported their circumferential locking suture had stronger tensile strength than the Lembert and the Simple circumferential suture techniques. Our results are compatible with their study. However, the Lin-locking technique has significantly increased WoF values compared to the other techniques. Lin-locking fixation provides strong tendon holding power, but produces an increased bulk at the repair site, causing increased resistance to tendon gliding. The Lin-locking and the Halsted techniques need two stitches in the tendon per strand, while the Cross-stitch and the Lembert need one stitch per strand, and the Simple and the Simple-locking need a half stitch per strand (Fig 2). Silverski61d and Andersson (1993) showed the Crossstitch has 62.8 N tensile strength in sheep flexor profundus tendons with 10 suture strands of 6/0 PDS. Our data at 10 suture strands of the Cross-stitch had 30.3 N strength, approximately half of their published results. Their repaired sheep tendons using the Kessler core suture plus the Simple circumferential suture had 47.8 N tensile strength, compared to our results of 24.9 N using the same technique in human cadaver tendons (unpublished data). This strongly suggests that the differences in tensile strength values of their study were due to the different species studied. The Cross-stitch had significantly less tensile strength than the Lin-locking technique, but also significantly less resistant to tendon gliding.

TENDON SUTURE TECHNIQUES

Mashadi and Amis (1992) showed that lacerated human cadaver flexor profundus tendons repaired with the Halsted technique using 5/0 multifilament stainless steel had more than 40 N rupture strength and were stronger than the Lembert and the Simple techniques. They did not indicate the number of suture strands used, but their reported values were 1.4 times stronger than our data using 6/0 Prolene (27.9 N at 18 strands). The difference is probably due to the different suture materials. Our data showed no significant difference between the Halsted and the Lembert; but the Halsted WoF values were the second highest of the six techniques, due to the increased bulk of suture at the repair site. Leddy (1993) introduced the Running Epitendinous (i.e. Simple) and the Locking Running Epitendinous (i.e. Simple-locking) as the ideal circumferential suture techniques, but he did not review their mechanical properties. Several studies (Lin et al, 1988; Wade et al, 1989, Mashadi and Amis, 1992) showed that the Simple technique had lower tensile strength compared to other techniques. Our results are compatible with these other studies. The number of suture strands has not been evaluated in previous studies. Only Lin et al (1988) and Wade et al (1989) standardized the number of suture strands to 10 in their studies. Silverskirld and Andersson (1993) calibrated the number of suture strands to circumferential tendon length, but did not compare different number of suture strands. Our data showed that increasing the number of suture strands increased both tensile and gap strength, but did not significantly increase the WoF values. These results may be helpful to surgeons in deciding which circumferential suture technique to use. The Linlocking has the best holding power, but also the greatest increase in gliding friction; it is the most complicated technique to place. The Cross-stitch, the Halsted and the Lembert have good tensile and gap strength values with lower WoF values noted for the Cross-stitch and the Lembert techniques. These would appear to be the recommended techniques. The Simple and the Simplelocking had the lowest tensile strengths; the locking configuration providing no advantage over the Simple technique. Acknowledgements The authors cordially acknowledge the technical assistance and useful suggestions of Mr Brian Larson, Mrs Annemarie Schmoeker, photographer Mr Ed Linn and Mr Fredrick Ludwig. This study has been supported in part by Grant #15953 from the Shriner's Hospital for Crippled Children. No benefits in any form have been received or will be received from a commerical party related directly or indirectly to the subject of this article.

References AOKI M, MANSKE P R, PRUITT D L and LARSON B J (1995). Work of flexion after tendon repair with various suture methods: a human cadaveric study. Journal of Hand Surgery, 20B: 3:310 313. BHATIA D, TANNER K E, BONFIELD W and CITRON N D (1992). Factors affecting the strength of flexor tendon repair. Journal of Hand Surgery, 17B: 550-552.

479 CRAVER J M, MADDEN J W and PEACOCK E E (1968). The effect of sutures, immobilization, and tenolysis on healing of tendons: a method for measuring work of digital flexion in a chicken's foot. Surgery, 64: 437-442. DEFINO H L A, BARBIERI C H, GONCALVES R P and PAULIN J B P (1986). Studies on tendon healing. A comparison between suturing techniques. Journal of Hand Surgery, 1IB: 444-450. DUNLAP J, MCCARTHY J A, JOYCE M E and MANSKE P R (1990). Biomechanical and histologic evaluations of pulley reconstructions in nonhuman primates. Journal of Hand Surgery, 15A: 1:57 63. GORDON L, GARRISON J L, CHENG J C et al. (1992). Biomechanical analysis of a step-cut technique for flexor tendon repair. Journal of Hand Surgery, 17B: 282-285. GREENWALD D P, HONG H-Z and MAY J W (1994). Mechanical analysis of tendon suture techniques. Journal of Hand Surgery, 19A: 4: 641-647. HADDAD R J, KESTER M A, McCLUSKEY G M, BRUNET M E and COOK S D (1988). Comparative mechanical analysis of a looped-suture tendon repair. Journal of Hand Surgery, 13A: 709-713. KESSLER I (1973). The 'grasping' technique for tendon repair. The Hand, 5: 253-255. KETCHUM L D, MARTIN N L and KAPPEL D A (1977). Experimental evaluation of factors affecting the strength of tendon repairs. Plastic and Reconstructive Surgery, 59:708 719. KLEINERT H E, KUTZ J E, COHEN M J. Primary Repair of Zone 2 Flexor Tendon Lacerations. In: American Academy of Orthopaedic Surgeons, Symposium on Tendon Surgery in the Hand, St Louis, Mosby, 1975: 91-104. LANE J M, BLACK J and BORA F W (1976). Gliding function following flexor-tendon injury. Journal of Bone and Joint Surgery, 58A: 985-990. LEDDY J P. Flexor Tendon-Acute Injuries. In: Green D P, (Ed.): Operative Hand Surgery, 3rd Edn. New York, Churchill Livingstone, 1993, Vol 2: 1823-1851. LEE H (1990). Double loop locking suture: a technique of tendon repair for early active mobilization. Part I: Evolution of technique and experimental study. Journal of Hand Surgery, 15A: 945-952. LIN G-T, AN K-N, AMADIO P C and COONEY W P (1988). Biomechanical studies of running suture for flexor tendon repair in dogs. Journal of Hand Surgery, 13A: 553-558. LISTER G D, KLEINERT H E, KUTZ J E and ATASOY E (1977). Primary flexor tendon repair followed by immediate controlled mobilization. Journal of Hand Surgery, 2: 441-451. MASHADI Z B and AMIS A A (1992). Strength of the suture in the epitenon and within the tendon fibres: development of stronger peripheral suture technique. Journal of Hand Surgery, 17B: 171-175. MESSINA A (1992). The double armed suture: tendon repair with immediate mobilization of the fingers. Journal of Hand Surgery, 17A: 137-142. NOGUCHI M, SELLER J G, GELBERMAN R H, SOFRANKO R A and WOO S L-Y (1993). In vitro biomechanical analysis of suture methods for flexor tendon repair. Journal of Orthopedic Research, 11: 603-611. PETERSON W W, MANSKE P R, BOLLINGER B A, LESKER P A and MCCARTHY J A (1986a). Effect of pulley excision on flexor tendon biomechanics. Journal of Orthopedic Research, 4:96 101. PETERSON W W, MANSKE P R, KAIN C C and LESKER P A (1986b). Effect of flexor sheath integrity on tendon gliding: a biomechanical and histologic study. Journal of Orthopedic Research, 4: 458-465. PETERSON W W, MANSKE P R, LESKER P A, KAIN C C and SCHAEFER R K (1986c). Development of a synthetic replacement for the flexor tendon pulleys: an experimental study. Journal of Hand Surgery, l l A : 403 409. PETERSON W W, MANSKE P R, DUNLAP J, HORWITZ D S and KAHN B (1990). Effect of various methods of restoring flexor sheath integrity on the formation of adhesions after tendon injury. Journal of Hand Surgery, 15A: 48-56. PRUITT D L, MANSKE P R and FINK B (1991). Cyclic stress analysis of flexor tendon repair. Journal of Hand Surgery, 16A: 701-707. SILVERSKIOLD K L and ANDERSSON C H (1993). Two new methods of tendon repair: an in vitro evaluation of tensile strength and gap formation. Journal of Hand Surgery, 18A: 1:58 65. STRICKLAND J W (1985). Flexor tendon repair. Hand Clinics, i: 55-68. TAJIMA T (1984). History, current status, and aspects of hand surgery in Japan. Clinical Orthopaedics and Related Research, 184:41 49. TERA H andABERG C (1976). Tensile strengths of twelve types of knot employed in surgery, using different suture materials. Acta Chirurgica Scandinavica, 142: 1-7. TRAIL I A,.POWELL E S and NOBLE J (1992). The mechanical strength of various suture techniques. Journal of Hand Surgery, 17B: 89-91. TSUGE K, IKUTA Y and MATSUISHI Y (1977). Repair of flexor tendons by intratendinous tendon suture. Journal of Hand Surgery, 2: 436-440. URBANIAK J R, CAHILL J D and MORTENSON R A. Tendon Suturing Methods: Analysis of Tensile Strengths. In: American Academy of Orthopaedic Surgeons, Symposium on Tendon Surgery in the Hand, St Louis, cv. Mosby, 1975: 70-80.

480 WADE P M F, M U I R I F K and H U T C H E O N L L (1986). Prin'lary flexor tendon repair: the mechanical limitations of the modified Kessler technique. Journal of H a n d Surgery, 11B: 71-76. WADE P J F, W E T H E R E L L R G and AMIS A A (1989). Flexor tendon repair: significant gain in strength from the Halsted peripheral suture technique. Journal of H a n d Surgery, 14B: 232-235. WAGNER W F, C A R R O L L C, S T R I C K L A N D J W e t al. (1994). A biomechanical comparison of techniques of flexor tendon repair. Journal of Hand Surgery, 19A: 979-983.

T H E J O U R N A L OF H A N D SURGERY VOL. 21B No. 4 A U G U S T 1996 WRAY R C and WEEKS P M (1980). Experimental comparison of technics of tendon repair. Journal of H a n d Surgery, 5:144 148.

Accepted: 22 October 1995 Paul R. Manske, MD, Department of Orthopaedic Surgery, Washmgton Umversity School of Medichle, One Barnes Hospital Plaza, Suite 11300, St Louis, MO 63110, USA. © 1996 The Blitish Society for Surgery of the Hand